Molecular cloning of the complementary DNA copies of the common and cowpea strains of tobacco mosaic virus RNA

VIROLOGY

118, 64-75 (1982)

Molecular

Cloning of the Complementary DNA Copies of the Common and Cowpea Strains of Tobacco Mosaic Virus RNA

TETSUO MESHI,’ NOBUHIKO TAKAMATSU, AND YOSHIMI OKADA Departm.entof Biophysics and Biochemistry,

Faculty

TAKESHI OHNO,

of Science, University

Received August 12, 1981; accepted November

of Tokyo, Tokyo 11.9,Japan

19, 1981

Complementary DNA (cDNA) copies of the common (OM) and cowpea (Cc) strains of tobacco mosaic virus (TMV) RNA polyadenylated in vitro have been synthesized with AMV reverse transcriptase and oligo (dT),, as primer. The cDNA was converted to a double-stranded form by .E. coZi DNA polymerase I followed by Sl nuclease digestion. The double-stranded cDNA copies were cloned in pBR322 at the PstI site using the oligo(dC)-oligo(dG) tailing method. With the common strain, several clones containing inserts covering different parts of genomic RNA were selected using partially reconstituted RNA and partially stripped virus RNA as hybridization probes. Restriction mapping of three clones and their overlaps showed that cloned sequences covered about 4000 nucleotides of the common strain RNA from the 3’ end. With the Cc strain, clones containing the 3’ portion were selected by Southern hybridization using coat protein mRNA isolated from short particles as a probe. One cloned recombinant plasmid was found by restriction analysis and R-loop mapping to carry about a 1700 nucleotide sequence of Cc strain RNA from the 3’ end. A restriction map of the OM strain was very similar to the map of the vulgare strain, as predicted from its nucleotide sequence, but completely different from that of Cc strain.

RNA and of other viruses of the tobamovirus group. The RNA genome of tobacco mosaic viTMV particles may be reconstituted rus (TMV) is about 6400 nucleotides long from viral RNA and coat protein in titro. and carries the information at least for The reaction starts by the binding of a 20 four polypeptides, 130K (sometimes re- S coat protein aggregate at a specific referred to as 1lOK or 140K), 165K, 30K, and gion (assembly origin) on the RNA (Butler coat protein (Beachy et al., 1976; Hunter and Klug, 1971; Okada and Ohno, 1972; et al., 19’76;Pelham, 1978). The coat protein Richards and Williams, 1972). Several cistron is known to be present at the 3’ end strains of TMV have been divided into two of the genome. The coat protein is trans- subgroups on the basis of the location of lated from a subgenomic RNA of about 700 the assembly origin (Fukuda et al., 1981). nucleotides which is produced from the The common strain (vulgare) and the regenomic RNA by an unknown mechanism lated OM and tomato strains (T) belong (Beachy et al., 1976; Hunter et al., 1976; to subgroup 1, in which the assembly orSiegel et al., 1976). The 1006 nucleotide se- igin is 800-1006 nucleotides away from the quence at the 3’ end of vulgare strain RNA 3’ end of RNA, that is, outside the coat has been determined (Guilley et al., 1979), protein mRNA region (Lebeurier et al., but we do not have enough sequence in- 1977; Otsuki et al., 1977; Fukuda et al., formation to describe the precise genome 1980). From the results of nucleotide sestructure of the remaining parts of that quencing of vulgare strain RNA, a highly base-paired hairpin loop structure is pos‘To whom requests for reprints should be ad- tulated at the assembly origin (Jonard et al., 1977; Zimmern, 1977). dressed. INTRODUCTION

0042-6822/82/050064-12$02.00/O Copyright All rights

0 1982 by Academic Prew, Inc. of reproduction in any form reserved.

64

CLONING

OF cDNA COPIES OF TMV RNA

The cowpea strain (Cc) of TMV has the significant feature of particle mutliplicity (Whitfeld and Higgins, 1976), producing short rods of encapsidated subgenomic mRNA for the coat protein in addition to the normal sized particles (Bruening et al., 1976; Higgins et al., 1976). The Cc strain is known to have a distant relationship with the common strain (Zaitlin et aZ., 1977), and belongs to subgroup 2 TMV, together with cucumber green mottle mosaic virus, in which the assembly origin is about 320 nucleotides away from the 3’ end, within the coat protein cistron (Fukuda et al., 1980, 1981). Determination of the nucleotide sequences of several strains of TMV and their comparison should lead to better understanding of the assembly mechanism, genome structure, and regulation of gene expression of TMV, and also to establishing a evolutionary relationship among viruses in the tobamovirus group. Synthesis and molecular cloning of complementary DNA to TMV RNA makes sequencing much easier and also provides a useful probe for analyzing the process of replication and gene expression of the TMV genome in vivo. In this paper we report the in vitro synthesis of double-stranded cDNA copies of the common and cowpea strains of TMV RNA and their cloning in E. coli, using pBR322 as a vector. MATERIALS

AND METHODS

Preparation of virus, RNA, and coat pre tin TMV OM particles and Cc long and short particles were purified from infected Nicotianu tabacum L. cv. Xanthi and infected Phaseolus vulgaris L. cv. Gintebo, respectively, as described previously (Otsuki et al., 1977; Fukuda et al., 1980). RNA was extracted from purified virus with phenol and SDS (Gierer and Schramm, 1956; Fraenkel-Conrat et al., 1957). Coat protein was isolated by the acetic acid method (Fraenkel-Conrat, 1957). Polyadenylation of TMV RNA. Poly(A) polymerase was purified from E. coli B/r as described by Sippel (1973). Polyadenylation of RNA was done by the method of

65

Devos et al. (1976) with following modifications. One milliliter of reaction mixture containing 50 mM Tris-HCl pH 8.0, 0.2 M NaCl, 10 mM MgClz, 2 mM MnClz, 0.5 mM spermidine, 0.4 mMEDTA, 1 mMDTT, 150 MgRNA, 0.2 mM vH]ATP (15 Ci/mol), and l/10 vol of concentrated poly(A) polymerase was incubated at 37’ for 5 min. The reaction was stopped by phenol extraction and RNA was recovered by ethanol precipitation. Synthesis of double-stranded cDNA to polycwknylated TMVRNA. Single-stranded cDNA was synthesized essentially as described by Ohno et al. (1980) in a l-ml reaction mixture containing 50 mM TrisHCl pH 7.9 at 42”, 12.5 mM MgClz, 10 mM DTT, 30 mM KCl, 100 pg actinomycin D, 1 mM each dATP, dGTP, dTTP, 1 mM [3H]dCTP (0.4 Ci/mmol), 5 pg oligo(dT)l,,, 50 pg polyacdenylated TMV RNA, 100 U AMV reverse transcriptase (kindly supplied by Dr. J. W. Beard) for 1 hr at 42”. After the reaction was stopped by phenol extraction, nucleic acids were recovered by ethanol precipitation. RNA was hydrolyzed in 0.1 N NaOH for 1 hr at 60”. After neutralization, the mixture was passed through a Sephadex G-100 column in TES (10 mM Tris-HCl pH 7.5, 0.1 M NaCl, 1 mM EDTA), and cDNA was precipitated with ethanol. Second strand was synthesized in a 0.3ml reaction mixture containing 67 mM Tris-HCl pH 7.4, 6.7 mM MgClz, 6.7 mM DTT, 1 mM each dATP, dGTP, dCTP, dTTP, 2-10 &ml rH]cDNA, 125 U/ml E. coli DNA polymerase I (large fragment, New England BioLabs), which was incubated for 6 hr at 20”. After gel filtration and ethanol precipitation, the non-basepaired regions of double-stranded cDNA were digested with Sl nuclease (Sankyo Co.) as described by Ohno et al. (1980) except using 4 U/ml of the enzyme. Sl nuclease-treated double-stranded cDNA was fractionated by sucrose gradient centrifugation and DNA longer than 1000 bp was pooled for the next step. Cmtructim of hybrid plasmids and tramformation, Double-stranded cDNA was inserted into the pBR322 vector at the

66

MESH1 ET AL.

PstI site, using the oligo(dG)-oligo(dC) tailing method (Bolivar et al., 1977; VillaKomaroff et al., 1977). Sl digested double-stranded cDNA was tailed with about 20 dC residues in a 30~1 reaction mixture containing 0.14 M Kcacodylate, 30 mM Tris base pH 6.95 (final), 1 mM CoCIZ, 0.1 mM DTT, 0.2 mM dCTP, 830 U/ml terminal deoxynucleotidy1 transferase (Bethesda Research Laboratories) (Roychoudhury et al., 1976). After 4 min at 30”, the reaction was stopped by adding 200 ~1 of 1 mM EDTA and the DNA was concentrated by ethanol precipitation. pBR322 DNA (4 pg) linearized with PstI was tailed in a similar reaction mixture (75 ~1) containing 0.2 mM [3H]dGTP (4.4 Ci/mmol) instead of dCTP, 600 U/ml terminal deoxynucleotidyl transferase. About 15 residues were added per end after incubation at 30” for 5 min. Tailed, double-stranded cDNA (l-2 X 10m2pmol) was annealed with 16 ng (about equal molar to the double-stranded cDNA) of vector in 30 ~1 of TES. The mixture was heated to 65” for 10 min, then held at 42” for 2.5 hr, and finally cooled slowly to 25” (Villa-Komaroff et al., 1978). Preparation of Ca2+-treated E. coli HBlOl was carried out by the method of Mandel and Higa (1970) with partial modification. Cells grown to 5-10 X lo8 cells/ ml in a 40 ml L-broth were collected and washed, they were then suspended in 20 ml of 10 mM MgS04, and kept for 20 min at 0”, recollected and suspended in 20 ml of 50 mM CaC12and settled as above, and finally suspended in 4 ml of 50 mM CaC12. Ten microliters of the solution containing hybrid plasmid and 90 ~1 of 10 mM TrisHCl pH 7.5, 10 mM MgC12, 10 mM CaC12 were mixed, and then 0.2 ml of Ca2+treated cells was added at 0”. After 15 min, the mixture was heated at 42” for 2 min and added to 1 ml of L-broth containing 12.5 rg/ml of tetracycline. The mixture was incubated for 1 hr at 37” and plated on L-agar containing tetracycline, followed by incubation at 37” for 20 hr. Bacterial transformation were carried out under P2 physical containment in accordance with the guidelines for recombinant DNA experiments (Japan).

Preparation of RNA from partially stripped vimcs (PSV) and partially reconstituted RNA (PRR). TMV OM at a concentration of 5 mg/ml was incubated for 18 hr at 0” in 20 mMNaC03 pH 10.5. After neutralization and fractionation by sucrose gradient centrifugation, nucleoprotein particles corresponding to PSV 5 and 6 (Pelcher and Halasa, 1979) were collected. Protruding RNA was digested with 15 U/ml RNase Tl at 37” for 30 min, followed by incubation at 4” for 18 hr. After RNase-treated PSV was collected by centrifugation, PSV RNA was extracted by the phenol-bentonite method (Fraenkel-Conrat et al., 1961). PSV RNA was estimated to be mainly about 1500 nucleotides long by electrophoretic analysis on a 2.4% polyacrylamide-8 iWurea gel, using E. co.5 16 S, 23 S, and 5 S rRNA as size markers. PRR was prepared essentially as described by Otsuki et al. (1977). The weight ratio of protein/RNA was 3. After fractionation, unprotected RNA was digested by RNase Tl to get PRR-RNA. PRR-RNA was mainly about 2100 nucleotides long, with several bands ranging from 750-2500 nucleotides as determined by polyacrylamide gel electrophoresis as above. Partial degradation of TMV RNA and its 5’ end labeling. RNA was fragmented by Mgz+ at high pH (Stinger et al., 1979). The reaction mixture containing 25 mM glytine-NaOH pH 9.0, 5 mM MgC12,200 pg/ ml genomic RNA (80 pg/ml for Cc coat protein mRNA, PSV-RNA, or PRR-RNA) was incubated for 3 hr at 45” and the reaction was terminated with EDTA (final 7 mM) at 0”. The resultant fragments were mainly ZOO-600nucleotides long. To prepare probes for hyrbidization, the 5’ ends of the RNA fragments were labeled in a reaction mixture containing 50 miW Tris-HCl pH 8.0 at 37”, 10 m2M MgC12, 5 mM DTT, 5.3 pM [T-~P]ATP (160 Ci/ mmol), 100 U/ml T4 polynucleotide kinase (Miles). The specific radioactivities of the labeled RNA fragments were 0.75-3.3 X lo6 cpm/pg. After incubation at 37” for 30 min, the mixture was passed through a Sephadex G-100 column and the labeled RNA fragments which were excluded in

CLONING

67

OF cDNA COPIES OF TMV RNA

the void volume (longer than about 100 nucleotides) were pooled. Screening of recombinant clones. Colony hybridization was carried out essentially according to the method of Grunstein and Hogness (1975). Tetracycline-resistant colonies were grown on nitrocellulose filters (Schleicher and Schuell, BA 85) and lysed, whereupon the DNA was fixed to the filters. Hybridization was carried out at 42” for 16 hr using about lo5 cpm of =Plabeled RNA fragments per filter by the method of Sugiura and Kusuda (1979). For rapid screening, plasmid DNA was extracted from 0.5 ml cultures by the rapid alkaline extraction procedure’ (Birnboim and Doly, 1979). After gel electrophoresis, DNA was transferred from gel to nitrocellulose filter by the method of Southern (1975). Following the fixation of DNA to the filter, hybridization was done as above. Large-scale purzfication of plasmid DNA and restriction digesticm Plasmid DNA was purified by the method described by Katz et al. (1973) with slight modification. After CsCl centrifugation and removal of ethidium bromide with isoamyl alcohol, DNA was recovered directly by ethanol precipitation as described by Gorecki and Rozenblatt (1980). Contaminating tRNA was removed by passing the DNA through the Sephadex G-100 column, if necessary. Restriction enzymes were purchased from New England BioLabs and Takara Shuzo Co. Complete restriction digestion was carried out for 4 hr at 37” (with P&I, BamHI at 30”, and BstNI at SO’) in solution containing 0.3-1.5 pg of DNA with 24 U of enzyme. The standard buffer for enzymes requiring salt was 10 mM TrisHCl pH 7.5, 7 mi+f MgClz, 5 mlM DTT, 50 mlM (100 mM for SaZI) NaCl. For other enzymes which have no salt requirement, for example, HpaII, HhaI, and HaeIII, NaCl was omitted. Digested fragments were analyzed by l1.5% agarose gel electrophoresis in Loening buffer (Loening, 1969) or on 4-5s polyacrylamide gels in 50 mM Tris-borate pH 8.3,l mM EDTA (Maxam and Gilbert, 1980). R loop formation for electron microscopy. Formation of R loop between pCc8C5 li-

nearlized by EcoRI and the Cc genomic RNA was carried out by the method of Thomas et a2. (1976). After incubating for 3 hr at 47”, the R-loop solution was diluted and spread according to the method of Wahli et al. (1978). The surface film was picked up on a Parlodion coated grid, stained with uranylacetate, and shadowed with PtPd. RESULTS

Cloning of Double-Stranded cDNA Copies of TMV RNA Since the 3’ end of TMV RNAs have no poly(A) sequence, the RNA was first polyadenylated at the 3’ end using E. coli poly(A) polymerase. The average length of the poly(A) sequence was calculated to be about 50 residues, based on the radioactivity incorporated. The polyadenylated TMV RNA was used as a template for the cDNA synthesis with AMV reverse transcriptase. The size of the cDNA ranged from 350 to 6500 nucleotides with two peaks at 500 and 1500 nucleotides (Fig. 1A). There was only a small amount of full-sized cDNA, corresponding to about 6400 nucleotides. The second strand was synthesized with E. coli DNA polymerase I using the selfpriming ability of single-stranded cDNA. Figure 1B shows the size distribution of double-stranded cDNA after digestion with Sl nuclease. To obtain recombinant plasmids with longer inserts, the doublestranded cDNAs longer than about 1000 bp were collected (Fig. 1B). Tailed doublestranded cDNA was inserted into the pBR322 vector at the PstI site. The hybrid plasmid was used to transform E. coli HBlOl, and tetracycline-resistant colonies were selected. Selection of the Plasmids Coveri~ about the @OONucleotide Sequencefrom the 3’ End of TMV OM RNA From the preliminary observation that transformants carrying plasmids containing longer inserts gave a more intense hy-

MESH1 ET AL.

1979). We selected several clones which were expected to contain more than 2000 bp inserts as candidates for further analysis. Since all the inserted sequences were found not to be derived from the 3’ end of the RNA by preliminary experiment, we used two more probes to estimate which region on the RNA genome was cloned in each transformant. One is PRR-RNA that was made by partial reconstitution of genomic RNA with limited amounts of coat protein (Otsuki et al., 1977) and subsequent RNase digestion of RNA not encap20 15 5 10 Fraction No. sidated. PRR-RNA contains generally the assembly origin and its 5’ flanking region (Otsuki et al., 1977). Our PRR preparation B in this experiment was mainly about 2100 nucleotides long and corresponded to nucleotide sequences from 800 to 3000 nuz 300 cleotides from the 3’ end of the genomic E RNA (Fig. 2D). Colonies hybridizing intensely with =P-labeled PRR-RNA frag2 200 ments should be carrying plasmids that Y contained at least a part of PRR-RNA se4 quence. Another probe is PSV-RNA. Our + 100 preparation was mainly PSV 5 (Pelcher and Halasa, 1979), which corresponds to the 1500 nucleotide sequence from the 3’ end of the genomic RNA (Fig. 2D). Those 5 10 15 20 2!5 hybridizing intensely with %P-labeled Fraction No. PSV-RNA fragments will contain the 3 FIG. 1. Sedimentation profile of TMV cDNA and portion of the genomic RNA. Figures 2B double-stranded cDNA. (A) TMV OM cDNA was sed- and 2C show one of the results of colony imented through an alkaline sucrose gradient (5-20% hybridization using these probes. Clones sucrose, 0.1 N NaOH, 0.9 M NaCl, 5 mM EDTA, at 5H2 and 5C6 hybridized intensely with 150,009g for 11.5 hr). (B) Sl nuclease treated doubleboth probes. Plasmids extracted from these stranded OM cDNA was fractionated on a neutral transformants, pOM5H2 and pOM5C6, sucrose gradient (5-20% sucrose, TES, at 136,000g contained about 2000 and 2800 bp inserts, for 10.7 hr). Larger double-stranded cDNA (L fraction) was pooled for transformation. The sizes of respectively. From similar experiments cDNA and double-stranded cDNA were estimated we selected for further analysis pOM4C5, using q-labeled pBR 322, linearlized by EcoRI (4.36 which hybridized intensely only with PRR kb), double-digested by EcoRI and PstI (3.61 kb fragments (data not shown). + 0.75 kb) as size markers. The relationship of inserts in these three plasmids with respect to a region on bridization signal, some hundred clones the genomic RNA was clarified by precise were selected from the result of colony restriction analysis. pOM5C6 had the hybridization using q-labeled OM geno- longest insert among characterized plasmic RNA fragments (Fig. 2A). The lengths mids and overlapped with both pOM5H2 of the plasmids in these transformants and pOM4C5. pOM5H2 was deduced to were determined by a rapid alkaline carry the sequence of the 3’ end region of screening procedure (Birnboim and Doly, the genomic RNA, as described below. It A

CLONING OF cDNA COPIES OF TMV RNA

69

0 A C

6 D.

5

I

3

1 (kb) .

2

5’

3’ Genomic f?NA I+ 1-4

Psv-RNA PR?-RNA

FIG. 2. Selection of OM clones by colony hybridization. (A) With [q]OM genomic RNA fragments as the probe, (B) with p]PRR-RNA fragments, and (C) p2P]PSV-RNA fragments. (B) and (C) show the same area. Well-characterized transformants were indicated by arrows: 1 for 4C5,2 for 5C6, and 3 for 5H2. (D) Tbe relative positions of PRR- and PSV-RNA on the genomic RNA is shown schematically. Arrows indicate the assembly origin.

overlapped pOM4C5 in a short region (about 100 nucleotides). Figure 3 shows the restriction map of about a 4000 nucleotide sequence of OM strain RNA constructed by overlapping of the three cloned sequences and also the inserted region in each cloned plasmid. The 3’ terminal 1000 nucleotides of vul-

gare RNA, very closely related to the OM strain, has been determined (Guilley et al., 1979). The restriction map predicted from the result of sequencing of vulgare RNA with that of OM strain (Fig. 3) coincides very well except at two sites. From the vulgare sequence, a HpaII site is predicted at residue 218-221 from the 3’ end, but the

FIG. 3. Restriction sites of eight restriction enzymes in about a 4900 nucleotide sequence of OM RNA, from the 3’ end. Comparison of the map predicted from the sequencing results of vulgare RNA (Guilley et al., 1979) with the scale-expanded map of the OM-cloned sequence of 1096 nucleotides at the 3’ end is also shown. Closed thick bars indicate the cloned region in each plasmid. Scale in kilobases is indicated as the distance from the 3’ terminal Hue111 site. Abbreviations used: 0, Hoe111 site; 0, HpaII site; 7, HinfI site; V, HhaI site.

CLONING OF cDNA COPIES OF TMV RNA

(6)

(A)

71

Genomic RNA

12345676

CP-mRNA FIG. 4. Restriction analysis of plasmids carrying a TMV Cc RNA sequence. (A) Plasmids extracted by the rapid alkaline method (Birnboim and Doly, 1979) were digested by PstI and analyzed by 1.5% agarose gel electrophoresis. hDNA fragments generated by a double digestion with EcoRI and Hind111 were run at the right side as size markers. (B) Autoradiograms of Southern hybridization of the region containing fragments derived from the insert with 32P-labeled Cc genomic RNA fragments (genomic RNA) or coat protein mRNA fragments (CP-mRNA). The plasmid in slot 8 does not contain the coat protein mRNA sequence.

long (Fig. 1). Thus, it is most probable that the reaction by DNA polymerase I to convert cDNA into a double-stranded form did not go to completion, and thus internal sequences of the genomic RNA might be cloned. From the overlap of restriction maps of OM clones, cloned sequences covered about 4000 nucleotides from the 3’ end of OM RNA, which would contain the coat protein cistron, the assembly origin, the 30K protein cistron, and the C-terminal portion of the 130K (or 1lOK or 140K) and 165K protein cistrons, including the leaky UAG codon at the read-through site (Pelham, 1978). The precise location of the coat protein cistron of the Cc strain could be deduced from the comparison of the amino acid sequence of Cc coat protein previously reported (Rees and Short, 1975) with the restriction map of the cloned sequence

(Fig. 5). A recognition sequence of a restriction enzyme that recognizes five or six bases can code for several possible sequences comprising two or three amino acids, by changing reading phases. Of possible amino acid sequences to be coded by the BumHI recognition sequence (-GGATCC-), only one sequence, Arg-AspPro, was present at residue 73-75 from the N-terminus of the Cc coat protein. For the PstI sequence (CTGCAG-), three possible amino acid sequences can be found, ProAla-Glu at 63-65, Leu-Gln at 85-86, and Thr-Ala-Glu at 105-107. From the length between the BumHI and PstI sites on the restriction map, the PstI sequence was located at residue 105-107 of the protein. By such a process, observed restriction sites could be assigned on the coat protein sequence without any ambiguity. As the result, the coat protein cistron of the Cc strain was deduced to be located approx-

MESH1 ET AL.

72

l 3’

5’*

d z .-C I

iATTC

GTCAAC

GGATCC

GTCTAC

CP cistron z E’3Z:;; izY $2

GAGTC CTGCAG

GTCTAC

GAGTC

---SerIleP~O---Se=GInLeu---ArgRspPro---~~”~~=~~~---*~~“~~---~~~~~=~~“---~~~~~~~~=---~~“~~~~~”--3 4 5 8 9 10 73 74 75 86 87 88 93 94 105106107 118 119120 130131132

FIG. 5. Restriction map of pC&C5 corresponding to the 3’ terminal 1700 nucleotides of Cc RNA. Correspondence between the restriction map and the amino acid sequence of Cc coat protein is also shown. The location of the coat protein cistron (CP cistron) including the assembly origin (Oa) is shown above the map. SaZI site is also H&II and Ace1 sites, although they are not indicated in the figure. Abbreviations used: 0, Hue111 site; 0, HpuII site; v, i%nfI site; V, HhaI site.

FIG. 6. Electron microscopic observation of an R-loop formed between p&W5 lineal&d by EcoRI and TMV Cc RNA. Scale bar indicates 300 nm. Plasmid DNA and Cc RNA are indicated by solid line and dashed line, respectively, in the drawing. An arrow indicates the 3’ end of the Cc RNA that corresponds to an end of the loop.

CLONING OF cDNA COPIES OF TMV RNA

imately 210-700 nucleotides from the Hue111 site, near the Pat1 site at the end of the insert (Fig. 5). This is very similar to the case of vulgare strain whose coat protein cistron is located in residues 205634 from the 3’ end (Guilley et al., 1979). It is known that RNA fragments of vulgare RNA can be encapsidated with coat protein (Guilley et al., 1975a, b). These fragments called SERF (specifically encapsidated RNA fragments) are located at residue 297-422 from the 3’ end and contain the sequence similar to the assembly origin of vulgare strain (Zimmern, 1977). The assembly origin of the Cc strain is located at about 320 nucleotides away from the 3’ end (Fukuda et al., 1980). From the coincidence of the positions of the SERF and the assembly origin of Cc strain, it has been proposed that the assembly origin of Cc strain would correspond to the SERF region on the vulgare strain RNA and that assembly origins of subgroup 1 and 2 TMV and SERF would be evolutionally related (Zimmern, 1977; Fukuda et al., 1980). Although it is known that the coat protein cistron of the cowpea strain is situated at the 3’ end of the genomic RNA, (Beachy et al., 1976), the precise position was not determined. The above results which show that the coat protein cistron of both strains are located nearly at the same position from the 3’ end of their genomic RNAs would substantiate this hypothesis. Figure 5 also shows the location of the assembly origin of Cc strain. Restriction maps of the two common strains, OM and vulgare, coincide very well within experimental error (Fig. 3), which suggests that there are very few differences in nucleotide sequence between them. On the other hand, we could find little homology in the restriction map between the Cc and common strains in 1.5 kb region from the 3’ end. Moreover, even in the region where several consecutive amino acid residues were conserved, we could find no common restriction site. These results suggest that their nucleotide sequences are very dissimilar. This agrees with the results of hybridization experi-

73

ments (Zaitlin et al., 1977; Palukaitis and Symons, 1980). ACKNOmEDGMENTS

We thank Dr. Y. Otauki for propagation of TMV and Dr. A. Hirashima for advice on polyadenylation. Reverse transcriptase was kindly supplied by Dr. J. W. Beard of Life Science Laboratory through the Office of Program Resources, National Cancer Institute. This work was supported in part by a grant from the Ministry of Education, Science and Culture, Japan. REFERENCES BEACHY,R.N., ZAITLIN, M., BRUENING,G., and ISRAEL, H. W. (1976). A genetic map for the cowpea strain of TMV. virdogy 73,498~597. BIRNBOIM,H. C., and DOLY, J. (1979). A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Ran 7.1513-1523. BOLIVAR, F., RODRIGUEZ,R. L., GREENE,P. J., BETLACH, M. C., HEYNEKER, H. L., and BOYER,H. W. (19’77). Construction and characterization of new cloning vehicles. II. A multipurpose cloning system. Gene 2, 95-113. BRUENING, G., BEACHY, R. N., SCALLA, R., and ZAITLIN, M. (1976). In vitro and in tiwo translation of the ribonucleic acids of a cowpea strain of tobacco mosiac virus. fiti 71,498-517. BUTLER, P. J. G., and ELUG, A. (1971). Assembly of the particle of tobacco mosaic virus from RNA and disks of protein. Nature New Bid 229,47-k%. DEVOS,R., VAN EMMJXU&J.. SEURINCK-OPSOMER, C., GILLIS, E., and FIERS, W. (1976). Addition by ATP: RNA adenyltransferase from Eschmichia edi of 3’linked oligo(A) to bacteriophage &s RNA and its effect on RNA replication. B&him Biophys. Acta 447.319-32-T. FRAENKEL-CONRAT,H. (1957). Degradation of tobacco mosaic virus with acetic acid. l%oh 4, l4. FRAENKEL-CONRAT,H., SINGER, B., and WILLIAMS, R. C. (1957). Infectivity of viral nucleic acid. Biochim Biuphgs Acta 25,87-96. FRAENKEL-CONRAT,H., SINGER,B., and TSUGITA, A. (1961). Purification of viral RNA by means of bentonite. virdogy 14, 54-58. FUKUDA, M., OKADA, Y., OTSUKI, Y., and TAKEBE, I. (1980). The site of initiation of rod assembly on the RNA of a tomato and a cowpea strain of tobacco mosaic virus. V&o&y 101,493~592. FUKUDA, M., MESHI, T., OKADA, Y., OTSUKI, Y., and TAKEBE, I. (1981). Correlation between particle multiplicity and the location on the virion RNA of the assembly initiation site for viruses of tobacco

74

MESH1 ET AL.

mosaic virus group. Proc Nat. Ad Sci. USA 73, 4231-4235. GIERER, A., and SCHRAMM,G. (1956). Infectivity of ribonucleic acid from tobacco mosaic virus. Nature (London) 177,702-703. GORECKI,M., and ROZENBLATT,S. (1980). Cloning of DNA complementary to the measles virus mRNA encoding nucleocapsid protein. Proc. Nat. Ad sci USA 77,3686-3690. GRUNSTEIN,M., and HOGNESS,D. S. (1975). Colony hybridization: A method for the isolation of cloned DNAs that contain a specific gene. Proc. Nat. Acad Sci USA 72.3961-3965. GUILLEY, H., JONARD, G., RICHARDS, K. E., and HIRTH, L. (19’75a). Sequence of a specifically encapsidated RNA fragment originating from the tobacco-mosaic-virus coat-protein cistron. Eur. J. Biochem 54,135-144. GUILLEY, H., JONARD, G., RICHARDS, K. E., and HIRTH, L. (1975b). Observations concerning the sequence of two additional specifically encapsidated RNA fragments originating from the tobacco-mosaic-virus coat-protein cistron. Eur. J. Biochem. 54.145-153. GUILLEY, H., JONARD,G., KUKLA, B., and RICHARDS, K. E. (1979). Sequence of 1,CHXlnucleotides at the 3’ end of tobacco mosaic virus RNA. Nucleic Acida Res. 6,1287-1398. HIGGINS, T. J. V., GOODWIN,P. B., and WHITFELD, P. R. (1976). Occurrence of short particles in beans infected with the cowpea strain of TMV. II. Evidence that short particles contain the cistron for coat-protein. Vi71,486-497. HUNTER, T. R., HUNT, T., KNOWLAND,J., and ZIMMERN,D. (1976). Messenger RNA for the coat protein of tobacco mosaic virus. Nature (London) 260, 759-764. JONARD, G., RICHARDS, K. E., GUILLEY, H., and HIRTH, L. (1977). Sequence from the assembly nucleation region of TMV RNA. CeU 11.483-493. KATZ, L., KINGSBURY,D. T., and HELINSKI, D. R. (1973). Stimulation by cyclic adenosine monophosphate of plasmid deoxyribonucleic acid replication and catabolite repression of the plasmid deoxyribonucleic acid-protein relaxation complex. J. Bactemid 114,577-591. LEBEURIER,G., NICOLAIEFF,A., and RICHARDS,K. E. (1977). Inside-out model for self-assembly of tobacco mosaic virus. Proc Nat. Acad Sci USA 74, 149-153. LOENING,U. E. (1969). The determination of the molecular weight of ribonucleic acid by polyacrylamide-gel electrophoresis. The effects of changes in conformation. Bioch.em J. 113,131-138. MANDEL,M., and HIGA, A. (1970). Calcium-dependent bacteriophage DNA infection. J. Mel Bid 53,159162.

MAXAM, A. M., and GILBERT, W. (1980). Sequencing end-labeled DNA with base-specific chemical cleavages. In “Methods in Enzymology” (L. Grossman, ed.), Vol. 65, pp. 499-569. Academic Press, New York. OHNO, T., COZENS,P. J., CATO, A. C. B., and JOST, J-P. (1989). Recombinant plasmids containing avian vitellogenin structural gene sequences derived from complementary DNA. Biochim Biophys. Acta 606, 34-46. OTSUKI, Y., TAKEBE, I., OHNO, T., FUKUDA, M., and OKADA, Y. (1977). Reconstitution of tobacco mosaic virus rods occurs bidirectionally from an internal initiation region: Demonstration by electron microscopic serology. Proc Nat. Ad Sci USA 76, 1727-1730. OKADA, Y., and OHNO, T. (1972). Assembly mechanism of tobacco mosaic virus particle from its rebonucleic acid and protein. Molec. Gen Gen& 114, 205-213. PALUKAITIS, P., and SYMONS,R. H. (1989). Nucleotide sequence homology of thirteen tobamovirus RNAs as determined by hybridization analysis with complementary DNA. Vi197.354-361. PELCHER,L. E., and HALASA, M. C. (1979). Reassembly or particles using the RNA from partially disassembled tobacco mosaic virus. Vim&q 97, 48% 491. PELHAM, H. R. B. (1978). Leaky UAG termination codon in tobacco mosaic virus RNA. Nature (Lmdon) 272, 469-471. REES, M. W., and SHORT,M. N. (1975). The amino acid sequence of the cowpea strain of tobacco mosaic virus protein. Biochim Biophys. Actn 393,1523. RICHARDS,K., and WILLIAMS, R. C. (1972). Assembly of tobacco mosaic virus in vitro: Effect of state of polymerization of the protein component. Proc. Nat. Acud Sci. USA 69, 1121-1124. ROYCHOUDHURY, R., JAY, E., and WV, R. (1976). Terminal labeling and addition of homopolymer tracts to duplex DNA fragments by terminal deoxynucleotidyl transferase. Nucleic Acids Res. 3,101-116. SAGER, H. L., RAMM, K., DOMDEY,H., GROSS,H. J., HENCO,K., and RIESNER,D. (1979). Conversion of circular viroid molecules to linear strands. FEBS Lett. 99, 117-122. SIEGEL, A., HARI, V., MONTGOMERY,I., and KOLACZ, K. (1976). A messenger RNA for capsid protein isolated from tobacco mosaic virus-infected tissue. Vi* 73,363-371. SIPPEL, A. E. (1973). Purification and characterization of adenosine triphosphate: Ribonucleic acid adenyltransferase from Eschmiehin coli EUT. J. Biochem 37,31-40. SOUTHERN,E. M. (1975). Detection of specific se-

CLONING

OF cDNA COPIES OF TMV RNA

quences among DNA fragments separated by gel electrophoresis. J. Mel Bid 98, 503-517. SUGIURA,M., and KUSUDA, J. (1979). Molecular cloning of tobacco chloroplast rihosomal RNA genes. Mokc Gen Genet. 172,137-141. THOMAS,M., WHITE, R. L., and DAVIS, R. W. (1976). Hybridization of RNA to double-stranded DNA: Formation of R-loops. Proc. Nat. Ad Sci USA 73,2294-2298. VILLA-K• MAROFF, L., EFSTRATIADIS,A., BROOME,S., LOMEDICO,R., TIZARDS, R., NABER, S. P., CHICK, W. L., and GILBERT, W. (1978). A bacterial clone synthesizing proinsulin. Proc. Nat. Acad sci. USA 75,3727-3731. WAHLI, W., RYFFEL, G. U., WYLER, T., JAGGI, R. B.,

75

WEBER, R., and DAVID, I. B. (1978). Cloning and characterization of synthetic sequences from the Xenqpus laevis vitellogenin structural gene. Den. Bid

67,371-383.

WHITFELD, P. R., and HIGGINS, T. J. V. (1976). Occurrence of short particles in beans infected with the cowpea strain of TMV. I. Purification and characterization of short particles. virology 71,471-485. ZAITLIN, M., BEACHY,R. N., and BRUENING,G. (1977). Lack of molecular hybridization between RNAs of two strains of TMV: A reconsideration of the criteria for strain relationships. firology 82,237~241. ZIMMERN,D. (1977). The nucleotide sequence at the origin for assembly on tobacco mosaic virus RNA. Cell 11,463-482.