Plant Molecular Biology 6: 221-228, 1986 © 1986 Martinus Nijhoff Publishers, Dordrecht - Printed in the Netherlands
Potato spindle tuber viroid infections mediated by the Ti plasmid of
Agrobacterium
tumefaciens Richard C. Gardner, l Kim R. Chonoles & Robert A. Owens 2
Calgene Inc, 1920 5th Street, Davis, CA 95616, U.S.A. 1present address." Department of Cell Biology, University of Auckland, Private Bag, Auckland, New Zealand 2plant Virology Laboratory, US. Department of Agriculture, Beltsville, MD 20705, U.S.A.
Keywords: Agrobacterium tumefaciens, cauliflower mosaic virus, potato spindle tuber viroid, Ti plasmid, viroid replication
Summary Full length copies of potato spindle tuber viroid (PSTV) were introduced into plant cells using an
Agrobacterium tumefaciens vector. Crown galls containing the PSTV DNA were induced on tomato plants, and the plants analysed for systemic replication of the viroid. Two separately derived multimeric PSTV insertions in the T-DNA were infectious on every plant inoculated. However, monomeric PSTV gave rise to significant levels of infection only when an adjacent plant promoter could direct transcription of + strand PSTV RNA. Our results suggest that this experimental system will be useful for the analysis of viroid replication. A second application of the results may be the use of systemic viral infection as a sensitive assay for transient expression of transformed DNA.
Introduction
these cloned DNAs are infectious when inoculated to leaves of suitable host plants. Efficient infection was obtained when viroid DNA was inoculated as tandem multimers, with tetramers more efficient than trimers, and trimers more efficient than dimers. Monomeric DNA had very low infectivity unless excised from the vector (5, 14, 24, 16). Mixtures of subgenomic fragments were also infectious, presumably after ligation in vivo (24). Infectivity of multimers of single-stranded viroid DNA cloned in M13 vectors was similar to that of doublestranded viroid DNA (24). RNA transcripts that contain + strand viroid RNA were also found to be infectious, though - strand RNA was not (5, 11, 19). (In viroids, the + strand is defined as the sequence present in infectious, natural viroid RNA.) The mechanism or mechanisms by which circular + strand RNA's of the viroid are generated is not clear, although the ability of viroid + strand RNA to carry out self processing is probably involved (19).
Viroids are small (250-370 nucleotide) circular RNA molecules that replicate autonomously in plants. Although their presence often causes disease symptoms, there is no evidence that these small RNA's code for any proteins (reviewed in 6, 20). The nucleotide sequence has been determined for a number of viroids. The sequence data has been used to propose classification groupings (20) and the existence of five structural and functional domains on the genome (12). In infected cells, viroid RNA occurs primarily as a protein-nucleic acid complex located in the nucleolus (21, 26). Although a number of replicative forms have been identified and general schemes proposed for viroid replication (2, 10, 11), the precise details await clarification. Full-length cDNA clones have been constructed for PSTV (5, 24, 16), hop stunt viroid (14), and tomato apical stunt viroid (16). Certain forms of 221
222 In this report, we describe experiments in which DNA of PSTV is introduced into plant cells using Agrobacterium. Infectivity of the transferred PSTV DNA was analysed by symptom appearance and by hybridization analysis using a cDNA probe against leaf sap. Multimeric PSTV DNA was infectious, as expected. Monomeric PSTV DNA was infectious only when an adjacent plant promoter transcribed + strand RNA copies. Thus, when an adjacent plant promoter is present, RNA polymerase II transcripts are implicated in the generation of infectious viroid molecules from a cDNA template. molecules from a cDNA template.
meric BamHI clone of PSTV (16). pCGN157 is an 8.8 kb shuttle vector for Agrobacterium (V. Knauf and R. Gardner, unpublished). It consists of the origin of replication and the chloramphenicol resistance gene from pACYC184 (BamHI-HindlII, 3), the kanamycin gene from Tn5 (HindllI-XhoI), a TDNA fragment containing the tml polyadenylation signal (SmaI-BamHI, nucleotides 11207-9062, ref. 1, with an XhoI linker inserted at the SmaI site), and the kanamycin gene from Tn903 (a BamHI fragment from pUC4K, ref. 26). pCGN149a was constructed by the insertion of a 410 bp BamHIBgllI fragment containing the CaMV 35S promoter at the unique BgllI site at the 5' end of the Tn5 gene in pCGN157. The CaMV promoter fragment contains 293 bp upstream of the cap site (terminating at an AluI site at position 7144, ref. 8) and 112 bp of the 5' untranslated region (with a terminal BgllI linker). PSTV DNA was inserted at the unique BgllI sites in pCGN149a and pCGN157 to give the constructs shown in Fig. 1. The constructions were transferred
Materials and methods
DNA constructions DNA manipulations and cloning were performed using standard procedures (13). Cloned PSTV DNA was obtained from pST-B14, a di-
A. PSTV CONSTRUCTS Tn5 Kan C a M V p s T v Tn5 promoter 3 5 S Kan [ ~ ] ~ ~ ~ ; ~ ' ~
tml polyA ~~!~/~Z~;J 1 6 0 a 160b
[ ~ ] ~ { ~ ~ ~
[~'---'~--~
:" i~ i!~i i~/1161b
~"~'i"J'l~'i /i~1 6 2 a 'l"ll 1 62b ,i '. i , /.,?J
..... ,',1;"
~
t p r~N2t Vl~h omoter
}
wig:l~/t
promoter
)i
)l
) l~i'!'il
'~ '~ii ~!~ 1 6 3 a
B. TRANSCRIPTS AROUND INTEGRATED CONSTRUCT
tmr
6a
CaMV-driven PSTV transcript
Atmr
6a
tml
ocs
Cm R K~m ,YS~':v :~: ~:;'''• ~" " ;1
The structure of the six PSTV-containing constructs is shown in part A. The upper 1. Constructs of PSTV in A g r o b a c t e r i u m . three constructs are PSTV insertions in pCGN149a, which contains the 35S promoter, while the lower three are in pCGN157. The arrow in PSTV indicates the direction of transcripts that would give rise to + strand RNA. Part B shows the structure o f one of the constructs (pCGN 160a) following recombination into the T - D N A of pTiA6 (not to scale). Transcripts predicted in gall cells are calculated from the sequence and published maps (see 1). The duplicated segment of the T - D N A (stripes) corresponds to the fragment cloned in pCGN157, and consists of nucleotides 9062-11207 (1).
Fig.
223 into the T-DNA of the octopine plasmid pTiA6 of
Results
Agrobacterium A722, using a single crossover procedure (4) with selection for the kanamycin resistance gene of Tn903. The structure of the construct integrated into the T-DNA was confirmed by Southern analysis.
Plant growth and gall formation Tomato plants (Lycopersicon esculentum cv 'Rutgers') were grown in high light growth chambers at 23°C in a mixture of 3 : 2 : 1 : 1 peat moss:vermiculite:sand:perlite. Plants were fertilised every third day with a 50% dilution of MS salts (Gibco). Agrobacterium inoculations were made in the stem with a syringe (see ref. 22 for details) when the plants were just beginning to develop the first true leaves. For inoculations of DNA, cotyledons were lightly dusted with carborundum, and approximately 2/~g of DNA of plasmid pCGN163a was applied in 2 t~l of water using a glass rod.
Dot blot hybridization We followed the procedure of Owens and Diener (15) with slight modifications. We sampled about 0.1 g ( 1 - 2 cm 2) from the terminal leaflet of the uppermost expanded (or almost expanded) leaves of the tomato plants. Leaflets were placed in 1.5 ml microfuge tubes that contained 50/zl of extraction buffer (15) and a small quantity of sand. The tissue was ground by hand with 3 mm glass rods for about 30 seconds, and 50/~1 of phenol:chloroform (1:1) added. After mixing by brief additional grinding or vortexing, the slurry was centrifuged for 2 min in a microfuge, and 3/~1 of the clear supernatant spotted to a nitrocellulose filter that had been soaked in 20×SSC and dried. Filters were baked, pre-hybridized, hybridized with nicktranslated pST-B14, and washed in 0.1×SSC at 60°C (details in ref. 22). The addition of phenol was found to significantly decrease the background hybridization of non-infected samples, and to slightly improve the strength of signal from infected samples.
Lithium-soluble nucleic acM preparations The protocol of Pfannenstiel et al. (18) was followed without modification for both tomato leaves and turnip galls.
Infectivity of PSTV constructions Insertions of PSTV were made in the two related shuttle vectors, pCGN157 and pCGN149a (described in Materials and methods), pCGN157 contains a region of T-DNA homology, Tn903 as a selectable marker for Agrobacterium, and the Tn5 kanamycin resistance gene in a pACYC184 replicon. pCGN149a differs only by the insertion of a 400 bp BamHI-BglII fragment containing the CaMV 35S promoter in front of the Tn5 coding region. This insertion results in the creation of a chimaeric gene for kanamycin resistance which has been shown to function in plants (Gardner, Hiatt and Facciotti, unpublished). PSTV was inserted as BamHI linear DNA into the unique BglII site in front of the Tn5 coding region in each vector, to produce the six constructs shown in Fig. 1. PSTV monomers were obtained in both the + and - orientation in front of the CaMV promoter, along with the analogous constructs in the 'promoterless' pCGN157 vector. We also obtained a dimer of PSTV in the - orientation behind the CaMV promoter, and a trimer in the + orientation in pCG157. The ability of these six constructs to give rise to systemic PSTV infection was tested by inducing crown galls on tomato plants. For three of the constructs (pCGN160a, 161b, and 163a), gall induction was followed 10- 25 days later by the appearance of typical PSTV symptoms: stunting, epinasty, and rugosity (see Fig. 2A). To verify that these plants contained PSTV sequences, leaf extracts were spotted onto nitrocellulose and hybridized with a PSTV probe. Plants showing symptoms always gave a positive signal (for example see Fig. 2B). To further confirm the presence of autonomously replicating PSTV in plants showing symptoms, we extracted a crude nucleic acid fraction from leaves (see Materials and methods). Electrophoresis of such preparations (see Fig. 2C) showed a band with the same mobility as circular PSTV RNA. This band was present only in extracts from plants showing symptoms. Following electrotransfer of the RNA to nitrocellulose, this band hybridized to a PSTV probe (result not shown). We used the rapid dot blot hybridization assay to quantitate the infectivity of the six constructs (results shown in Table 1). The two multimeric
224
Fig. 2. Evidence for PSTV infection. A. Appearance of symptoms. Two plants inoculated with 161b and 163a (right) showed typical PSTV symptoms. The two healthy-looking plants (left) were among those inoculated with pCGN160a. B. Dot blot hybridization. The illustrated autoradiograph shows a 24 h exposure of a nitrocellulose filter containing 22 samples from day 27 of the time course experiment (Fig. 3). Seven of these samples were scored as healthy, while the fifteen with strong signals were scored as infected. C. Lithiumsoluble RNA. A crude preparation of nucleic acids that are soluble in 2 M lithium chloride were prepared and electrophoresed in 8°70 acrylamide. The gel shows a picture of such a gel stained with ethidium bromide. The arrow indicates the position of purified PSTV RNA (centre lane). Corresponding bands are present in the three samples from plants showing symptoms, but not from the healthy plant. P S T V constructs, p C G N I 6 1 b a n d 163a, were infectious o n all p l a n t s tested. O n e m o n o m e r i c insertion, p C G N 1 6 0 a , gave rise to s y m p t o m s in 18 o f 24 plants, while the o t h e r three m o n o m e r i c c o n s t r u c tions were always non-infectious. Direct a p p l i c a tion o f p C G N 1 6 3 a D N A to c o t y l e d o n s infected o n l y 1 o u t o f 10 plants, while m o c k i n o c u l a t i o n caused no infections. D u r i n g these experiments, we o b s e r v e d a difference in the rate o f a p p e a r a n c e o f s y m p t o m s ; the larger the o l i g o m e r o f PSTV, the faster the a p p e a r ance o f s y m p t o m s . To extend this o b s e r v a t i o n , we
p e r f o r m e d the time course e x p e r i m e n t shown in Fig. 3. T h e results c o n f i r m e d o u r initial observations, with s y m p t o m s a p p e a r i n g in the o r d e r 163a > 161b > 160a. In this experiment, one p l a n t ino c u l a t e d with 160b developed systemic s y m p t o m s a n d was positive in the d o t blot. It is u n c l e a r w h e t h e r this p l a n t b e c a m e infected as a result o f s o m e low level t r a n s c r i p t i o n events in the plant, as a result o f a t r a n s c r i p t f r o m A g r o b a c t e r i u m , or was a c c i d e n t a l l y infected f r o m an a d j a c e n t p l a n t d u r i n g the r e p e a t e d handling.
225 Table 1. Infectivity of PSTV constructs in Agrobacterium for tomato plants.
Construct
Description*
Infectivity (plants infected/plants inoculated**) Experiment n u m b e r 1
160a 160b 161b 162a 162b 163a
CaMV + CaMV CaMV - + + + +
2
2/5 0/4 5/5
3
4
5
Total
4/4 0/5
3/3 0/3 3/3 0/3 0/3 3/3
8/12 0/4 3/3 0/2 0/3 3/3
17/24 0/16 11/11 0/ 9 0/10 10/10
0/4 0/4 4/4
* C a M V denotes the presence of the CaMV promoter. The designation + / - refers to the orientation of PSTV R N A that would be transcribed with the CaMV or Tn5 promoters (see Fig. 1). ** Infectivity was measured by dot blot hybridization of leaf samples. For the five experiments, the samples were taken at slightly differing intervals after inoculation of the plants with Agrobacterium. The intervals were as follows: experiment 1, 21 days; experiments 2 and 3, 22 days; experiment 4, 27 days; experiment 5, 28 days. These time differences m a y have slightly affected the results, particularly for 160a (see Fig. 3).
Infectivity of P S T V constructs in turnip galls PSTV is not propagated in several Brassica species, including turnip (23). We were interested to see whether it might be possible to see replication of PSTV in turnip galls, in which a large number of cells would contain the infectious DNA. We therefore inoculated plants with Agrobacterium con-
10
161b
,.=,
|
taining the two most infectious constructs, 163a and 16lb. Crude preparations of lithium-soluble nucleic acids were extracted from the resulting galls. Electrophoresis of these fractions on acrylamide gels showed no viroid band visible by inspection (data not shown). Northern hybridizations against these fractions showed some material of high MW that hybridized with PSTV (presumably DNA), but there was no hybridizing material at the mobility of circular PSTV RNA (data not shown). We also tested these preparations for the presence of infectious PSTV by inoculating them to tomato plants. The 163a preparation did not infect any of 6 plants inoculated, while the 161b turnip preparation infected one out of 6. In contrast, a preparation from PSTV-infected tomato leaf infected 6 of 6 plants at a 100-fold dilution. In summary, our results provided no suggestion that replication of PSTV occurs in turnip when the viroid is introduced to the chromosome in the form of multimeric DNA.
.
Discussion 0'
'
6'
'
12'
~
1'8
'
2~4
'
3'0
DAYS AFTER INOCULATION
Fig. 3. Time course of infectivity. Four groups of ten plants were inoculated with Agrobacterium containing the indicated PSTV constructs. Infectivity was scored by dot blot hybridizations.
Full-length cDNA copies of PSTV introduced into crown gall cells were able to initiate systemic infection of tomato plants. Our infectivity results using this system are consistent with results reported from inoculation of leaves with purified DNA
226 (5, 14, 16, 24) or RNA (11, 16, 19). First, higher multimers of viroid DNA showed increasing infectivity, with trimeric DNA more infectious than dimeric DNA, and dimeric more infectious than monomers. Second, RNA transcripts containing the + strand of PSTV were infectious while those containing the - strand were noninfectious. Out of the four monomeric PSTV constructs we tested, only 160a was infectious. This construct has a CaMV promoter which would make + strand copies of PSTV inside the plant cell. The construct making - strand copies of PSTV in the plant (160b) was generally uninfectious, as were the constructs lacking the CaMV promoter (162a and 162b). These experiments provide genetic evidence that the mechanism of infection by these constructs is via excision of viroid RNA from a longer RNA polymerase II-derived transcript. By contrast, the infectivity of the two multimeric constructs is probably independent of the CaMV promoter. The trimer (163a) lacks the CaMV promoter altogether, while in the dimer (161b) the CaMV promoter would transcribe - strand viroid RNA, which has been shown to be non-infectious (11). The most likely explanation for the infectious nature of the multimeric cDNA's is that PSTV RNA is derived from a + strand transcript, possibly arising from 'background' transcription of adjacent T-DNA sequences in the plant, or perhaps from a functional promoter in PSTV itself (see 24). The high degree of infectivity of the multimeric PSTV constructs presumably reflects a greatly increased ability of multimeric PSTV RNA (compared to monomeric RNA) to undergo cleavage and ligation and thereby produce a circular viroid RNA. Thus two factors may be affecting the infectivity observed in our experiments: first, the amount of + strand viroid transcript produced; and second, the ability of the transcript to give rise to infectious circular viroid molecules. Previous demonstrations of the infectivity of bacterial RNA containing PSTV (5, 14, 24, 19) suggest the possibility that the symptoms we observed might have resulted from Agrobacterium transcripts. Two observations appear to rule out this possibility. First, 160a was infectious while 162a was not, despite the fact that both constructs have the bacterial Tn5 promoter upstream of PSTV making + strand RNA copies of PSTV in Agrobacterium (Fig. 1). The requirement for the
CaMV promoter suggests the involvement of plant transcripts in the infection process. Second, the highly infectious trimeric construct, 163a, has been transferred into a number of Agrobacterium strains containing virulence mutations, or into a plasmid lacking borders. With the exception of virE mutants, symptoms never resulted from inoculation of plants with these strains (7). Thus systemic PSTV infection depends on bona fide transfer of the TDNA from Agrobacterium into the plant cell. The non-infectivity of the 160b construct, which would transcribe - strand PSTV RNA in the plant cell, might be attributed to the phenomenon of 'cross-protection'. Palukaitis and Zaitlin (i7) have suggested that hybridization of the - strand to the + strand of viroids (and viruses) could be a mechanism that limits replication and results in crossprotection. If such a mechanism were operative, then 161b should also be non-infectious since it would also produce - strand PSTV RNA from the CaMV promoter. Northern analysis of tomato gall RNA demonstrated that the transcripts present in 160a, 160b, and 161b did indeed contain copies of the viroid, and were present at concentrations similar to the parental 149a construct (Chonoles and Gardner, unpublished). Thus the noninfectivity of the 160b construct is unlikely to result from cross-protection. The demonstration that PSTV, and presumably other viroids and viruses, can initiate a systemic infection following introduction into plants via Agrobacterium has a number of consequences. First, it has several implications for the design of vectors, which have been outlined elsewhere (9). Second, the experimental system we have described may be useful in the analysis of viroid replication. Preliminary results presented here suggest that PSTV replication is blocked in turnips. Further analysis of galls or transformed plants might allow identification of the step at which replication is blocked, since large numbers of transformed cells containing RNA replicative intermediates would be available. Similar analysis could be performed with non-viable mutants of PSTV introduced into a host that is permissive for replication. Together these approaches may help to elucidate steps in the infection cycle of viroids. The use of this experimental system in structure-function analysis of viroid replication is discussed more fully in the accompanying paper (16). Systemic PSTV infection provides a novel assay
227 for D N A transfer. Using P S T V i n f e c t i o n as an assay, we have b e e n able to subdivide the process o f t r a n s f o r m a t i o n by Agrobacterium into two stages, which we suggest c o r r e s p o n d to D N A transfer and D N A i n t e g r a t i o n (7). T h e basis for this subdivision m a y be t h a t systemic infection by P S T V requires D N A transfer and transcription, but n o t D N A integration. T h u s viral i n f e c t i o n m a y provide a useful assay for transient expression o f t r a n s f o r m e d DNA. Two results suggest t h a t this i n f e c t i o n assay is quite sensitive: revertants were recovered after Agrobacterium infection but not after D N A inoculations with a m u t a n t P S T V (16), and d i s a r m e d strains o f Agrobacterium that p r e s u m a b l y infect o n l y a small n u m b e r o f cells are consistently able to induce P S T V s y m p t o m s (7). A viral infection assay might be useful in situations where o t h e r transf o r m a t i o n assays are i n a p p r o p r i a t e ; for example, analysing the host range o f Agrobacterium in species or cell types which do not readily f o r m galls or u n d e r g o cell division in culture.
Acknowledgements We t h a n k our colleagues at Calgene, p a r t i c u l a r l y Bob G o o d m a n , Luca C o m a i , Bill H i a t t , and Christine Shewmaker, for their advice and help, and Vic K n a u f , Bill H i a t t and D a n i e l Facciotti for allowing us to r e p o r t u n p u b l i s h e d results. Prelimin a r y studies in R.A.O.'s l a b o r a t o r y were partially s u p p o r t e d by G r a n t No. 81-CRCR-1-0719 o f the C o m p e t i t i v e Research Grants p r o g r a m o f the U S D e p a r t m e n t o f A g r i c u l t u r e (R.A. Owens and D.E. Cress, c o p r i n c i p a l investigators). These experiments were reviewed by an institut i o n a l b i o s a f e t y c o m m i t t e e p r i o r to initiation. All plant m a t e r i a l was c o n t a i n e d within a single c o m m i t t e d g r o w t h c h a m b e r and a u t o c l a v e d after analysis.
References 1. Barker RF, Idler KB, Thompson DV, Kemp JD: Nucleotide sequence of the T-DNA region from the Agrobacterium tumefaciens octopine Ti plasmid pTi15955. Plant Mol Biol 2:335 - 350, 1983. 2. Branch AD, Robertson HD: A replication cycle for viroids and other small infectious RNA's. Science 223:450-455, 1984.
3. Chang ACY, Cohen SN: Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the P15A cryptic miniplasmid. J Bacteriol 134:1141 - 1156, 1978. 4. Comai L, Schilling-Cordaro C, Mergia A, Houck CM: A new technique for genetic engineering of Agrobacterium Ti plasmid. Plasmid 10:21- 30, 1983. 5. Cress DE, Kiefer MC, Owens RA: Construction of infectious potato spindle tuber viroid cDNA clones. Nucl Acids Res 11:6821-6835, 1983. 6. Diener TO: Viroids and their interactions with host cells. Ann Rev Microbiol 36:239-258, 1982. 7. Gardner RC, Knauf VC: A T-DNA border but not the virE gene product is required for transfer of DNA from Agrobacterium to plant cells. Science (in press), 1986. 8. Gardner RC, Howarth A J, Hahn P, Brown-Luedi M, Shepherd RJ, Messing J: Complete nucleotide sequence of an infectious clone of cauliflower mosaic virus by M13mp7 shotgun sequencing. Nucl Acids Res 9:2871- 2888, 1981. 9. Gardner RC, Hiatt WR, Facciotti D, Shewmaker CK: The use of integrated copies of plant viruses. Plant Mol Biol Reporter 2:3- 8, 1984. 10. Hutchins C J, Keese P, Visvader JE, Rathjen PD, Mclnnes JL, Symons RH: Comparison of multimeric plus and minus forms of viroids and virusoids. Plant Mol Biol 4:293 - 304, 1985. 11. Ishikawa M, Meshi T, Ohno T, Okada Y, Sano T, Ueda I, Shikata E: A revised replication cycle for viroids: The role of longer than unit length RNA in viroid replication. Mol Gen Genet 196:421-428, 1984. 12. Keese P, Symons RH: Domains in viroids: Evidence of intermolecular RNA rearrangements and their contribution to viroid evolution. Proc Natl Acad Sci USA 82:4582- 4586, 1985. 13. Maniatis T, Fritsch EF, S~tmbrook J: Molecular cloning A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, 1982. 14. Meshi T, Ishikawa M, Ohno T, Okada Y, Sano T, Ueda I, Shikada E: Double-stranded cDNAs of hop stunt viroid are infectious. J Biochem 95:1521 - 1524, 1984. 15. Owens RA, Diener TO: Rapid and sensitive diagnosis of potato spindle tuber viroid disease by nucleic acid hybridization. Science 213:670- 672, 1981. 16. Owens RA, Hammond RW, Gardner RC, Kiefer MC, Thompson SM, Cress DE: Site-specific mutagenesis of potato spindle tuber viroid cDNA: alterations within premelting region 2 that abolish infectivity. Plant Mol Biol 6:179- 192, 1986. 17. Palukaitis P, Zaitlin M: A model to explain the 'crossprotection' phenomenon shown by plant viruses and viroids. In: Kosuge T, Nester EW (eds) Plant Microbe Interactions - Molecular and Genetic Perspectives Volume 1. Macmillan, New York, 1984, pp 420-429. 18. Pfannenstiel MA, Slack SA, Lane LC: Detection of potato spindle tuber viroid in field-grown potatoes by an improved electrophoretic assay. Phytopathology 70:1015- 1018, 1980. 19. Robertson HD, Rosen AL, Branch AD: Cell-free synthesis and processing of an infectious dimeric transcript of potato spindle tuber viroid RNA. Virology 142:441- 447, 1985. 20. Sanger HL: Biology, structure, functions, and possible ori-
228
21.
22.
23. 24.
gin of viroids. In: Parthier B, Boulter D (eds) Encyclopaedia of Plant Physiology, New Series Vol. 14B. Springer° Verlag, Berlin, 1982, pp 368-454. Schumacher J, Sanger HL, Riesner D: Subcellular localization of viroids in highly purified nuclei from tomato leaf tissue. EMBO J 2:1549- 1555, 1983. Shewmaker CK, Caton JR, Houck CM, Gardner RC: Transcription of cauliflower mosaic virus integrated into plant genomes. Virology 140:281-288, 1985. Singh RP: Host range of potato spindle tuber virus. Am Potato J 50:111- 123, 1973. Tablet M, Sanger HL: Cloned single- and double-stranded DNA copies of potato spindle tuber viroid (PSTV) RNA
and coinoculated subgenomic DNA fragments are infectious. EMBO J 3:3055-3062, 1984. 25. Vieira J, Messing J: The pUC plasmids, an M13mp7-derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene 19:259- 268, 1982. 26. Wolff P, Gilz R, Schumacher J, Riesner D: Complexes of viroids with histones and other proteins. Nucl Acids Res 13:355 - 367, 1985.
Received 17 September 1985; in revised form 3 December 1985; accepted 10 December 1985.
Potato spindle tuber viroid infections mediated by the Ti plasmid of
Agrobacterium
tumefaciens Richard C. Gardner, l Kim R. Chonoles & Robert A. Owens 2
Calgene Inc, 1920 5th Street, Davis, CA 95616, U.S.A. 1present address." Department of Cell Biology, University of Auckland, Private Bag, Auckland, New Zealand 2plant Virology Laboratory, US. Department of Agriculture, Beltsville, MD 20705, U.S.A.
Keywords: Agrobacterium tumefaciens, cauliflower mosaic virus, potato spindle tuber viroid, Ti plasmid, viroid replication
Summary Full length copies of potato spindle tuber viroid (PSTV) were introduced into plant cells using an
Agrobacterium tumefaciens vector. Crown galls containing the PSTV DNA were induced on tomato plants, and the plants analysed for systemic replication of the viroid. Two separately derived multimeric PSTV insertions in the T-DNA were infectious on every plant inoculated. However, monomeric PSTV gave rise to significant levels of infection only when an adjacent plant promoter could direct transcription of + strand PSTV RNA. Our results suggest that this experimental system will be useful for the analysis of viroid replication. A second application of the results may be the use of systemic viral infection as a sensitive assay for transient expression of transformed DNA.
Introduction
these cloned DNAs are infectious when inoculated to leaves of suitable host plants. Efficient infection was obtained when viroid DNA was inoculated as tandem multimers, with tetramers more efficient than trimers, and trimers more efficient than dimers. Monomeric DNA had very low infectivity unless excised from the vector (5, 14, 24, 16). Mixtures of subgenomic fragments were also infectious, presumably after ligation in vivo (24). Infectivity of multimers of single-stranded viroid DNA cloned in M13 vectors was similar to that of doublestranded viroid DNA (24). RNA transcripts that contain + strand viroid RNA were also found to be infectious, though - strand RNA was not (5, 11, 19). (In viroids, the + strand is defined as the sequence present in infectious, natural viroid RNA.) The mechanism or mechanisms by which circular + strand RNA's of the viroid are generated is not clear, although the ability of viroid + strand RNA to carry out self processing is probably involved (19).
Viroids are small (250-370 nucleotide) circular RNA molecules that replicate autonomously in plants. Although their presence often causes disease symptoms, there is no evidence that these small RNA's code for any proteins (reviewed in 6, 20). The nucleotide sequence has been determined for a number of viroids. The sequence data has been used to propose classification groupings (20) and the existence of five structural and functional domains on the genome (12). In infected cells, viroid RNA occurs primarily as a protein-nucleic acid complex located in the nucleolus (21, 26). Although a number of replicative forms have been identified and general schemes proposed for viroid replication (2, 10, 11), the precise details await clarification. Full-length cDNA clones have been constructed for PSTV (5, 24, 16), hop stunt viroid (14), and tomato apical stunt viroid (16). Certain forms of 221
222 In this report, we describe experiments in which DNA of PSTV is introduced into plant cells using Agrobacterium. Infectivity of the transferred PSTV DNA was analysed by symptom appearance and by hybridization analysis using a cDNA probe against leaf sap. Multimeric PSTV DNA was infectious, as expected. Monomeric PSTV DNA was infectious only when an adjacent plant promoter transcribed + strand RNA copies. Thus, when an adjacent plant promoter is present, RNA polymerase II transcripts are implicated in the generation of infectious viroid molecules from a cDNA template. molecules from a cDNA template.
meric BamHI clone of PSTV (16). pCGN157 is an 8.8 kb shuttle vector for Agrobacterium (V. Knauf and R. Gardner, unpublished). It consists of the origin of replication and the chloramphenicol resistance gene from pACYC184 (BamHI-HindlII, 3), the kanamycin gene from Tn5 (HindllI-XhoI), a TDNA fragment containing the tml polyadenylation signal (SmaI-BamHI, nucleotides 11207-9062, ref. 1, with an XhoI linker inserted at the SmaI site), and the kanamycin gene from Tn903 (a BamHI fragment from pUC4K, ref. 26). pCGN149a was constructed by the insertion of a 410 bp BamHIBgllI fragment containing the CaMV 35S promoter at the unique BgllI site at the 5' end of the Tn5 gene in pCGN157. The CaMV promoter fragment contains 293 bp upstream of the cap site (terminating at an AluI site at position 7144, ref. 8) and 112 bp of the 5' untranslated region (with a terminal BgllI linker). PSTV DNA was inserted at the unique BgllI sites in pCGN149a and pCGN157 to give the constructs shown in Fig. 1. The constructions were transferred
Materials and methods
DNA constructions DNA manipulations and cloning were performed using standard procedures (13). Cloned PSTV DNA was obtained from pST-B14, a di-
A. PSTV CONSTRUCTS Tn5 Kan C a M V p s T v Tn5 promoter 3 5 S Kan [ ~ ] ~ ~ ~ ; ~ ' ~
tml polyA ~~!~/~Z~;J 1 6 0 a 160b
[ ~ ] ~ { ~ ~ ~
[~'---'~--~
:" i~ i!~i i~/1161b
~"~'i"J'l~'i /i~1 6 2 a 'l"ll 1 62b ,i '. i , /.,?J
..... ,',1;"
~
t p r~N2t Vl~h omoter
}
wig:l~/t
promoter
)i
)l
) l~i'!'il
'~ '~ii ~!~ 1 6 3 a
B. TRANSCRIPTS AROUND INTEGRATED CONSTRUCT
tmr
6a
CaMV-driven PSTV transcript
Atmr
6a
tml
ocs
Cm R K~m ,YS~':v :~: ~:;'''• ~" " ;1
The structure of the six PSTV-containing constructs is shown in part A. The upper 1. Constructs of PSTV in A g r o b a c t e r i u m . three constructs are PSTV insertions in pCGN149a, which contains the 35S promoter, while the lower three are in pCGN157. The arrow in PSTV indicates the direction of transcripts that would give rise to + strand RNA. Part B shows the structure o f one of the constructs (pCGN 160a) following recombination into the T - D N A of pTiA6 (not to scale). Transcripts predicted in gall cells are calculated from the sequence and published maps (see 1). The duplicated segment of the T - D N A (stripes) corresponds to the fragment cloned in pCGN157, and consists of nucleotides 9062-11207 (1).
Fig.
223 into the T-DNA of the octopine plasmid pTiA6 of
Results
Agrobacterium A722, using a single crossover procedure (4) with selection for the kanamycin resistance gene of Tn903. The structure of the construct integrated into the T-DNA was confirmed by Southern analysis.
Plant growth and gall formation Tomato plants (Lycopersicon esculentum cv 'Rutgers') were grown in high light growth chambers at 23°C in a mixture of 3 : 2 : 1 : 1 peat moss:vermiculite:sand:perlite. Plants were fertilised every third day with a 50% dilution of MS salts (Gibco). Agrobacterium inoculations were made in the stem with a syringe (see ref. 22 for details) when the plants were just beginning to develop the first true leaves. For inoculations of DNA, cotyledons were lightly dusted with carborundum, and approximately 2/~g of DNA of plasmid pCGN163a was applied in 2 t~l of water using a glass rod.
Dot blot hybridization We followed the procedure of Owens and Diener (15) with slight modifications. We sampled about 0.1 g ( 1 - 2 cm 2) from the terminal leaflet of the uppermost expanded (or almost expanded) leaves of the tomato plants. Leaflets were placed in 1.5 ml microfuge tubes that contained 50/zl of extraction buffer (15) and a small quantity of sand. The tissue was ground by hand with 3 mm glass rods for about 30 seconds, and 50/~1 of phenol:chloroform (1:1) added. After mixing by brief additional grinding or vortexing, the slurry was centrifuged for 2 min in a microfuge, and 3/~1 of the clear supernatant spotted to a nitrocellulose filter that had been soaked in 20×SSC and dried. Filters were baked, pre-hybridized, hybridized with nicktranslated pST-B14, and washed in 0.1×SSC at 60°C (details in ref. 22). The addition of phenol was found to significantly decrease the background hybridization of non-infected samples, and to slightly improve the strength of signal from infected samples.
Lithium-soluble nucleic acM preparations The protocol of Pfannenstiel et al. (18) was followed without modification for both tomato leaves and turnip galls.
Infectivity of PSTV constructions Insertions of PSTV were made in the two related shuttle vectors, pCGN157 and pCGN149a (described in Materials and methods), pCGN157 contains a region of T-DNA homology, Tn903 as a selectable marker for Agrobacterium, and the Tn5 kanamycin resistance gene in a pACYC184 replicon. pCGN149a differs only by the insertion of a 400 bp BamHI-BglII fragment containing the CaMV 35S promoter in front of the Tn5 coding region. This insertion results in the creation of a chimaeric gene for kanamycin resistance which has been shown to function in plants (Gardner, Hiatt and Facciotti, unpublished). PSTV was inserted as BamHI linear DNA into the unique BglII site in front of the Tn5 coding region in each vector, to produce the six constructs shown in Fig. 1. PSTV monomers were obtained in both the + and - orientation in front of the CaMV promoter, along with the analogous constructs in the 'promoterless' pCGN157 vector. We also obtained a dimer of PSTV in the - orientation behind the CaMV promoter, and a trimer in the + orientation in pCG157. The ability of these six constructs to give rise to systemic PSTV infection was tested by inducing crown galls on tomato plants. For three of the constructs (pCGN160a, 161b, and 163a), gall induction was followed 10- 25 days later by the appearance of typical PSTV symptoms: stunting, epinasty, and rugosity (see Fig. 2A). To verify that these plants contained PSTV sequences, leaf extracts were spotted onto nitrocellulose and hybridized with a PSTV probe. Plants showing symptoms always gave a positive signal (for example see Fig. 2B). To further confirm the presence of autonomously replicating PSTV in plants showing symptoms, we extracted a crude nucleic acid fraction from leaves (see Materials and methods). Electrophoresis of such preparations (see Fig. 2C) showed a band with the same mobility as circular PSTV RNA. This band was present only in extracts from plants showing symptoms. Following electrotransfer of the RNA to nitrocellulose, this band hybridized to a PSTV probe (result not shown). We used the rapid dot blot hybridization assay to quantitate the infectivity of the six constructs (results shown in Table 1). The two multimeric
224
Fig. 2. Evidence for PSTV infection. A. Appearance of symptoms. Two plants inoculated with 161b and 163a (right) showed typical PSTV symptoms. The two healthy-looking plants (left) were among those inoculated with pCGN160a. B. Dot blot hybridization. The illustrated autoradiograph shows a 24 h exposure of a nitrocellulose filter containing 22 samples from day 27 of the time course experiment (Fig. 3). Seven of these samples were scored as healthy, while the fifteen with strong signals were scored as infected. C. Lithiumsoluble RNA. A crude preparation of nucleic acids that are soluble in 2 M lithium chloride were prepared and electrophoresed in 8°70 acrylamide. The gel shows a picture of such a gel stained with ethidium bromide. The arrow indicates the position of purified PSTV RNA (centre lane). Corresponding bands are present in the three samples from plants showing symptoms, but not from the healthy plant. P S T V constructs, p C G N I 6 1 b a n d 163a, were infectious o n all p l a n t s tested. O n e m o n o m e r i c insertion, p C G N 1 6 0 a , gave rise to s y m p t o m s in 18 o f 24 plants, while the o t h e r three m o n o m e r i c c o n s t r u c tions were always non-infectious. Direct a p p l i c a tion o f p C G N 1 6 3 a D N A to c o t y l e d o n s infected o n l y 1 o u t o f 10 plants, while m o c k i n o c u l a t i o n caused no infections. D u r i n g these experiments, we o b s e r v e d a difference in the rate o f a p p e a r a n c e o f s y m p t o m s ; the larger the o l i g o m e r o f PSTV, the faster the a p p e a r ance o f s y m p t o m s . To extend this o b s e r v a t i o n , we
p e r f o r m e d the time course e x p e r i m e n t shown in Fig. 3. T h e results c o n f i r m e d o u r initial observations, with s y m p t o m s a p p e a r i n g in the o r d e r 163a > 161b > 160a. In this experiment, one p l a n t ino c u l a t e d with 160b developed systemic s y m p t o m s a n d was positive in the d o t blot. It is u n c l e a r w h e t h e r this p l a n t b e c a m e infected as a result o f s o m e low level t r a n s c r i p t i o n events in the plant, as a result o f a t r a n s c r i p t f r o m A g r o b a c t e r i u m , or was a c c i d e n t a l l y infected f r o m an a d j a c e n t p l a n t d u r i n g the r e p e a t e d handling.
225 Table 1. Infectivity of PSTV constructs in Agrobacterium for tomato plants.
Construct
Description*
Infectivity (plants infected/plants inoculated**) Experiment n u m b e r 1
160a 160b 161b 162a 162b 163a
CaMV + CaMV CaMV - + + + +
2
2/5 0/4 5/5
3
4
5
Total
4/4 0/5
3/3 0/3 3/3 0/3 0/3 3/3
8/12 0/4 3/3 0/2 0/3 3/3
17/24 0/16 11/11 0/ 9 0/10 10/10
0/4 0/4 4/4
* C a M V denotes the presence of the CaMV promoter. The designation + / - refers to the orientation of PSTV R N A that would be transcribed with the CaMV or Tn5 promoters (see Fig. 1). ** Infectivity was measured by dot blot hybridization of leaf samples. For the five experiments, the samples were taken at slightly differing intervals after inoculation of the plants with Agrobacterium. The intervals were as follows: experiment 1, 21 days; experiments 2 and 3, 22 days; experiment 4, 27 days; experiment 5, 28 days. These time differences m a y have slightly affected the results, particularly for 160a (see Fig. 3).
Infectivity of P S T V constructs in turnip galls PSTV is not propagated in several Brassica species, including turnip (23). We were interested to see whether it might be possible to see replication of PSTV in turnip galls, in which a large number of cells would contain the infectious DNA. We therefore inoculated plants with Agrobacterium con-
10
161b
,.=,
|
taining the two most infectious constructs, 163a and 16lb. Crude preparations of lithium-soluble nucleic acids were extracted from the resulting galls. Electrophoresis of these fractions on acrylamide gels showed no viroid band visible by inspection (data not shown). Northern hybridizations against these fractions showed some material of high MW that hybridized with PSTV (presumably DNA), but there was no hybridizing material at the mobility of circular PSTV RNA (data not shown). We also tested these preparations for the presence of infectious PSTV by inoculating them to tomato plants. The 163a preparation did not infect any of 6 plants inoculated, while the 161b turnip preparation infected one out of 6. In contrast, a preparation from PSTV-infected tomato leaf infected 6 of 6 plants at a 100-fold dilution. In summary, our results provided no suggestion that replication of PSTV occurs in turnip when the viroid is introduced to the chromosome in the form of multimeric DNA.
.
Discussion 0'
'
6'
'
12'
~
1'8
'
2~4
'
3'0
DAYS AFTER INOCULATION
Fig. 3. Time course of infectivity. Four groups of ten plants were inoculated with Agrobacterium containing the indicated PSTV constructs. Infectivity was scored by dot blot hybridizations.
Full-length cDNA copies of PSTV introduced into crown gall cells were able to initiate systemic infection of tomato plants. Our infectivity results using this system are consistent with results reported from inoculation of leaves with purified DNA
226 (5, 14, 16, 24) or RNA (11, 16, 19). First, higher multimers of viroid DNA showed increasing infectivity, with trimeric DNA more infectious than dimeric DNA, and dimeric more infectious than monomers. Second, RNA transcripts containing the + strand of PSTV were infectious while those containing the - strand were noninfectious. Out of the four monomeric PSTV constructs we tested, only 160a was infectious. This construct has a CaMV promoter which would make + strand copies of PSTV inside the plant cell. The construct making - strand copies of PSTV in the plant (160b) was generally uninfectious, as were the constructs lacking the CaMV promoter (162a and 162b). These experiments provide genetic evidence that the mechanism of infection by these constructs is via excision of viroid RNA from a longer RNA polymerase II-derived transcript. By contrast, the infectivity of the two multimeric constructs is probably independent of the CaMV promoter. The trimer (163a) lacks the CaMV promoter altogether, while in the dimer (161b) the CaMV promoter would transcribe - strand viroid RNA, which has been shown to be non-infectious (11). The most likely explanation for the infectious nature of the multimeric cDNA's is that PSTV RNA is derived from a + strand transcript, possibly arising from 'background' transcription of adjacent T-DNA sequences in the plant, or perhaps from a functional promoter in PSTV itself (see 24). The high degree of infectivity of the multimeric PSTV constructs presumably reflects a greatly increased ability of multimeric PSTV RNA (compared to monomeric RNA) to undergo cleavage and ligation and thereby produce a circular viroid RNA. Thus two factors may be affecting the infectivity observed in our experiments: first, the amount of + strand viroid transcript produced; and second, the ability of the transcript to give rise to infectious circular viroid molecules. Previous demonstrations of the infectivity of bacterial RNA containing PSTV (5, 14, 24, 19) suggest the possibility that the symptoms we observed might have resulted from Agrobacterium transcripts. Two observations appear to rule out this possibility. First, 160a was infectious while 162a was not, despite the fact that both constructs have the bacterial Tn5 promoter upstream of PSTV making + strand RNA copies of PSTV in Agrobacterium (Fig. 1). The requirement for the
CaMV promoter suggests the involvement of plant transcripts in the infection process. Second, the highly infectious trimeric construct, 163a, has been transferred into a number of Agrobacterium strains containing virulence mutations, or into a plasmid lacking borders. With the exception of virE mutants, symptoms never resulted from inoculation of plants with these strains (7). Thus systemic PSTV infection depends on bona fide transfer of the TDNA from Agrobacterium into the plant cell. The non-infectivity of the 160b construct, which would transcribe - strand PSTV RNA in the plant cell, might be attributed to the phenomenon of 'cross-protection'. Palukaitis and Zaitlin (i7) have suggested that hybridization of the - strand to the + strand of viroids (and viruses) could be a mechanism that limits replication and results in crossprotection. If such a mechanism were operative, then 161b should also be non-infectious since it would also produce - strand PSTV RNA from the CaMV promoter. Northern analysis of tomato gall RNA demonstrated that the transcripts present in 160a, 160b, and 161b did indeed contain copies of the viroid, and were present at concentrations similar to the parental 149a construct (Chonoles and Gardner, unpublished). Thus the noninfectivity of the 160b construct is unlikely to result from cross-protection. The demonstration that PSTV, and presumably other viroids and viruses, can initiate a systemic infection following introduction into plants via Agrobacterium has a number of consequences. First, it has several implications for the design of vectors, which have been outlined elsewhere (9). Second, the experimental system we have described may be useful in the analysis of viroid replication. Preliminary results presented here suggest that PSTV replication is blocked in turnips. Further analysis of galls or transformed plants might allow identification of the step at which replication is blocked, since large numbers of transformed cells containing RNA replicative intermediates would be available. Similar analysis could be performed with non-viable mutants of PSTV introduced into a host that is permissive for replication. Together these approaches may help to elucidate steps in the infection cycle of viroids. The use of this experimental system in structure-function analysis of viroid replication is discussed more fully in the accompanying paper (16). Systemic PSTV infection provides a novel assay
227 for D N A transfer. Using P S T V i n f e c t i o n as an assay, we have b e e n able to subdivide the process o f t r a n s f o r m a t i o n by Agrobacterium into two stages, which we suggest c o r r e s p o n d to D N A transfer and D N A i n t e g r a t i o n (7). T h e basis for this subdivision m a y be t h a t systemic infection by P S T V requires D N A transfer and transcription, but n o t D N A integration. T h u s viral i n f e c t i o n m a y provide a useful assay for transient expression o f t r a n s f o r m e d DNA. Two results suggest t h a t this i n f e c t i o n assay is quite sensitive: revertants were recovered after Agrobacterium infection but not after D N A inoculations with a m u t a n t P S T V (16), and d i s a r m e d strains o f Agrobacterium that p r e s u m a b l y infect o n l y a small n u m b e r o f cells are consistently able to induce P S T V s y m p t o m s (7). A viral infection assay might be useful in situations where o t h e r transf o r m a t i o n assays are i n a p p r o p r i a t e ; for example, analysing the host range o f Agrobacterium in species or cell types which do not readily f o r m galls or u n d e r g o cell division in culture.
Acknowledgements We t h a n k our colleagues at Calgene, p a r t i c u l a r l y Bob G o o d m a n , Luca C o m a i , Bill H i a t t , and Christine Shewmaker, for their advice and help, and Vic K n a u f , Bill H i a t t and D a n i e l Facciotti for allowing us to r e p o r t u n p u b l i s h e d results. Prelimin a r y studies in R.A.O.'s l a b o r a t o r y were partially s u p p o r t e d by G r a n t No. 81-CRCR-1-0719 o f the C o m p e t i t i v e Research Grants p r o g r a m o f the U S D e p a r t m e n t o f A g r i c u l t u r e (R.A. Owens and D.E. Cress, c o p r i n c i p a l investigators). These experiments were reviewed by an institut i o n a l b i o s a f e t y c o m m i t t e e p r i o r to initiation. All plant m a t e r i a l was c o n t a i n e d within a single c o m m i t t e d g r o w t h c h a m b e r and a u t o c l a v e d after analysis.
References 1. Barker RF, Idler KB, Thompson DV, Kemp JD: Nucleotide sequence of the T-DNA region from the Agrobacterium tumefaciens octopine Ti plasmid pTi15955. Plant Mol Biol 2:335 - 350, 1983. 2. Branch AD, Robertson HD: A replication cycle for viroids and other small infectious RNA's. Science 223:450-455, 1984.
3. Chang ACY, Cohen SN: Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the P15A cryptic miniplasmid. J Bacteriol 134:1141 - 1156, 1978. 4. Comai L, Schilling-Cordaro C, Mergia A, Houck CM: A new technique for genetic engineering of Agrobacterium Ti plasmid. Plasmid 10:21- 30, 1983. 5. Cress DE, Kiefer MC, Owens RA: Construction of infectious potato spindle tuber viroid cDNA clones. Nucl Acids Res 11:6821-6835, 1983. 6. Diener TO: Viroids and their interactions with host cells. Ann Rev Microbiol 36:239-258, 1982. 7. Gardner RC, Knauf VC: A T-DNA border but not the virE gene product is required for transfer of DNA from Agrobacterium to plant cells. Science (in press), 1986. 8. Gardner RC, Howarth A J, Hahn P, Brown-Luedi M, Shepherd RJ, Messing J: Complete nucleotide sequence of an infectious clone of cauliflower mosaic virus by M13mp7 shotgun sequencing. Nucl Acids Res 9:2871- 2888, 1981. 9. Gardner RC, Hiatt WR, Facciotti D, Shewmaker CK: The use of integrated copies of plant viruses. Plant Mol Biol Reporter 2:3- 8, 1984. 10. Hutchins C J, Keese P, Visvader JE, Rathjen PD, Mclnnes JL, Symons RH: Comparison of multimeric plus and minus forms of viroids and virusoids. Plant Mol Biol 4:293 - 304, 1985. 11. Ishikawa M, Meshi T, Ohno T, Okada Y, Sano T, Ueda I, Shikata E: A revised replication cycle for viroids: The role of longer than unit length RNA in viroid replication. Mol Gen Genet 196:421-428, 1984. 12. Keese P, Symons RH: Domains in viroids: Evidence of intermolecular RNA rearrangements and their contribution to viroid evolution. Proc Natl Acad Sci USA 82:4582- 4586, 1985. 13. Maniatis T, Fritsch EF, S~tmbrook J: Molecular cloning A laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, 1982. 14. Meshi T, Ishikawa M, Ohno T, Okada Y, Sano T, Ueda I, Shikada E: Double-stranded cDNAs of hop stunt viroid are infectious. J Biochem 95:1521 - 1524, 1984. 15. Owens RA, Diener TO: Rapid and sensitive diagnosis of potato spindle tuber viroid disease by nucleic acid hybridization. Science 213:670- 672, 1981. 16. Owens RA, Hammond RW, Gardner RC, Kiefer MC, Thompson SM, Cress DE: Site-specific mutagenesis of potato spindle tuber viroid cDNA: alterations within premelting region 2 that abolish infectivity. Plant Mol Biol 6:179- 192, 1986. 17. Palukaitis P, Zaitlin M: A model to explain the 'crossprotection' phenomenon shown by plant viruses and viroids. In: Kosuge T, Nester EW (eds) Plant Microbe Interactions - Molecular and Genetic Perspectives Volume 1. Macmillan, New York, 1984, pp 420-429. 18. Pfannenstiel MA, Slack SA, Lane LC: Detection of potato spindle tuber viroid in field-grown potatoes by an improved electrophoretic assay. Phytopathology 70:1015- 1018, 1980. 19. Robertson HD, Rosen AL, Branch AD: Cell-free synthesis and processing of an infectious dimeric transcript of potato spindle tuber viroid RNA. Virology 142:441- 447, 1985. 20. Sanger HL: Biology, structure, functions, and possible ori-
228
21.
22.
23. 24.
gin of viroids. In: Parthier B, Boulter D (eds) Encyclopaedia of Plant Physiology, New Series Vol. 14B. Springer° Verlag, Berlin, 1982, pp 368-454. Schumacher J, Sanger HL, Riesner D: Subcellular localization of viroids in highly purified nuclei from tomato leaf tissue. EMBO J 2:1549- 1555, 1983. Shewmaker CK, Caton JR, Houck CM, Gardner RC: Transcription of cauliflower mosaic virus integrated into plant genomes. Virology 140:281-288, 1985. Singh RP: Host range of potato spindle tuber virus. Am Potato J 50:111- 123, 1973. Tablet M, Sanger HL: Cloned single- and double-stranded DNA copies of potato spindle tuber viroid (PSTV) RNA
and coinoculated subgenomic DNA fragments are infectious. EMBO J 3:3055-3062, 1984. 25. Vieira J, Messing J: The pUC plasmids, an M13mp7-derived system for insertion mutagenesis and sequencing with synthetic universal primers. Gene 19:259- 268, 1982. 26. Wolff P, Gilz R, Schumacher J, Riesner D: Complexes of viroids with histones and other proteins. Nucl Acids Res 13:355 - 367, 1985.
Received 17 September 1985; in revised form 3 December 1985; accepted 10 December 1985.
