Journal of Virological Methods, 32 (1991) 5746 0 1991 Elsevier Science Publishers B.V.

AD0/V1~0168851091001101

57

VIRMET 01142

The use of PCR for cloning of large cDNA fragments of turnip mosaic potyvirus

Olivier Nicolas and Jean-Franqois LalibertC

CRESALA. Institut Armand-Frappier, Ville de Laval. Que’hec, Canada

(Accepted 29 October 1990)

Summary

A method is described whereby turnip mosaic virus RNA (TuMV RNA) was reverse transcribed and the resulting cDNA amplified enzymatically using the Taq DNA polymerase and degenerate oligonucleotide primers. Two degenerate oligo- nucleotide primers based on regions of homology in the amino acid sequence of the cytoplasmic inclusion protein and the nuclear inclusion b protein from five potyvi- ruses were synthesized. Polymerase chain reactions utilizing these degenerate pri- mers in association with specific primers amplified a 1.2 kb and a 3.3 kb fragment. These amplified fragments were dC-tailed and cloned into pUC9. Their partial sequence, when compared to potyvirus sequences, showed that they were derived from TuMV RNA and approximately 4.4 kb of viral genome was cloned.

TuMV, Turnip mosaic virus; Potyvirus; PCR; Degenerate primer

Introduction

Turnip mosaic virus (TuMV) is a member of the plant potyvirus group. Its genome is a single positive-sense RNA of approximately 10 kb in length. This RNA is polyad- enylated and codes for a polyprotein precursor proteolytically processed into possibly eight polypeptides (Dougherty and Carrington, 1988). The genomes of five potyvi- ruses were completely sequenced: tobacco etch virus (TEV) (Allison et al., 1986); tobacco vein mottling virus (TVMV) (Domier et al., 1986); two strains of plum pox

Correspondence to: J.-F. LalibertC, CRESALA, Institut Armand-Frappier, 531 boulevard des Prairies, Ville de Laval, QC, H7N 423 Canada.

0168.8510/91/$03.50

virus (PPV) (Lain et al., 1989; Maiss et al., 1989), as well as potato virus Y (PVY) (Robaglia et al., 1989). These sequences share many regions of homology. Recently, we reported the cloning and the sequencing of 1801 nucleotides from 3’ terminal region of TuMV and of the encoded capsid protein (Tremblay et al., 1990).

The ability to clone large segments of plant viral RNA contributes to the under- standing of the function of viral-coded proteins. Unfortunately, it is still difficult to obtain large cDNA fragments of viral genomes. Full-length clones are produced after lengthy manipulations involving the screening of a large number of recombinants and the joining together of partial cDNAs. A way of obtaining full-length, or at least rea- sonably large clones, is to utilize a large amount of RNA and to select for long cDNAs (Le Gall et al., 1988). When the amount of viral RNA is limited, as in the case of TuMV, the above method is quite difficult to accomplish.

Instead, we decided to use the polymerase chain reaction (PCR) (Saiki et al., 1985, 1988) for the generation of TuMV cDNAs because it requires small amounts of RNA and has the potential of amplifying large fragments (Saiki et al., 1988; Grady and Wayne, 1989). However, the further cloning of the TuMV genome by PCR requires sequence information towards the 5’ terminal region of the virus in order to synthesize the primers. Different strategies have been developed for the amplification of uncharacterized regions: the restriction digestion and subsequent circularization of genomic DNA (inverse polymerase chain reaction, IPCR) (Ochman et al., 1988); the tailing of cDNA (RACE) (Frohman et al., 1988), or the addition of an adaptor to the cDNA and using a primer complementary to one strand of the adaptor (Akowitz and Manuelidis, 1989). In addition, a degenerate primer derived from conserved amino acid sequences in association with a specific primer were used (Ohara et al., 1989; Hyypia et al., 1989; Donehower et al., 1990).

We describe an efficient method for the amplification of large portions of TuMV RNA by the PCR using degenerate primers based on regions of homology in the amino acid sequence of the polyprotein of five potyviruses. This method required a small amount of RNA and large cDNA fragments were cloned easily.

Materials and Methods

Virus and viral RNA purification

Turnip mosaic virus was isolated from a naturally infected rutabaga plant from a field near l’Assomption, Quebec, Canada and was maintained and propagated in Brassica napus L. ssp. napobrassica cv. Laurentian. The virus was purified using basically the procedure described by Choi et al. (1977). Based on the fact that the com- mon extinction coefficient 2.7 ODZm unit is equivalent to 1 mg/ml of virus for the potyvirus group (State-Smith and Tremaine, 1970), the estimated yield of purified TuMV was 1.5-2 mg/lOO g of fresh leaves. RNA was extracted from virions as described by Maiss et al. (1988), resuspended in RNase-free water and conserved at -70°C until further manipulation.

59

Reverse transcriptase reaction

Single-stranded (ss) cDNA was synthesized from 500 ng of TuMV RNA using oli- go(dT) 12-l 8 and the moloney murine leukemia virus reverse transcriptase (Bethesda Research Laboratories, BRL) (Gubler and Hoffman, 1983). Briefly, after an incu- bation of 60 min at 37OC, the ss-cDNA was extracted with phenol-chloroform and pre- cipitated with 0.1 volume of 3 M sodium acetate, pH 5.5, and two volumes of ethanol. The solution was centrifuged and the pellet resuspended in 30 ~1 of TE buffer and con- served at -70°C until further manipulation.

Amplification of viral samples

Four different oligonucleotides (FT2, JF3, FT8 and JF9) were used, varying from 19 to 25 nucleotides in length. They were synthesized using the Gene Assembler from Pharmacia. Two different amplifications were achieved, one with FT2 and JF3 pri- mers using 0.2 ng of ss-cDNA, the second with FI’8 and JF9 primers using 3.5 ng of ss-cDNA. The appropriate primers (50 pmol) were added and the reaction was con- ducted in a volume of 100 ~1 of PCR buffer (20 mM Tris-HCl pH 8.3,1.5 mM MgClz, 25 mM KCl, 100 clg/ml bovine serum albumin) (Innis and Gelfand, 1990) containing 50 PM each of dATP, dCTP, dGTP and TTP and 2.5 units of Taq DNA polymerase (Perkin Elmer Cetus Corporation). The reaction was performed in a DNA Thermal Cycler from Perkin-Elmer Cetus Corp. Two sets of conditions were used, depending on the length of the expected PCR product: condition for n2-JF3 primers: cycles l-2: denaturation at 94°C for 30 s, annealing at 37°C for 30 s, extension at 72°C for 2 min; cycles 3-34: same conditions as above except an increase at 45°C for the annealing temperature; last cycle: same conditions as cycles 3-34 except the exten- sion time was 4 min. Condition for FT8-JF9 primers: cycles l-2: denaturation at 94°C for 1 min, annealing at 37°C for 30 s, extension at 72°C for 5 min; cycles 3-34: same conditions as above except temperature of annealing was raised to 45°C; last cycle: same conditions as cycles 3-34 except the extension time was raised to 7 min. A nega- tive control consisted of the same reaction conditions without adding any ss-cDNA. After completion, the PCR solution was extracted with chloroform and one tenth or more, if necessary, was subjected to electrophoresis in a 0.7% agarose gel in the pres- ence of ethidium bromide (0.5 pg/ml) and Tris-acetate, EDTA buffer (Maniatis et al., 1989). DNA was visualised on a UV transilluminator.

Molecular cloning and sequencing of amplifiedfragments

Nine tenths of the PCR (90 ~1) containing at least 50 ng of amplified DNA was loaded in a 0.7% agarose gel and submitted to electrophoresis. The band of the expected length was excised and electroeluted using the Bio-Rad Electra-Eluter. The electroeluted DNA was dC-tailed and annealed to dG-tailed pUC9 (Pharmacia). 200 ~1 of E. coli strain XL 1 -blue (Stratagene) competent cells, prepared as described by Hanahan (1983), were transformed with 5 to 10 ng of pUC9 vector from the annealing reaction. Recombinant plasmids were monitored for appropriate size inserts by cleav-

age with Psrl or by two enzymes on either side ofthe PstI restriction site in pUC9 vec- tor. Each fragment was then subcloned into M 13mp 18 and M 13mp 19 and sequenced by the dideoxy chain termination method (Sanger et al., 1977; Tabor et al., 1987).

Results

RNA isolation and cDNA synthesis

Viral RNA was isolated from purified TuMV preparations and visualised after electrophoresis on a formaldehyde agarose gel. A single band of approximately 10000 nucleotides was observed with no apparent degradation products and used without any further purification. A major problem with the reverse transctiptase reac- tion is the premature stop along the RNA template which leads to the production of incomplete cDNA molecules. A cDNA library from TuMV RNA was previously cloned in LambdaZap (Tremblay et al., 1990) but the longest inserts obtained by this method were approximately 1200 nucleotides in length. Therefore, it was important to verify, before attempting the PCR, that some of the cDNA molecules produced from the reverse transcription of the TuMV RNA were full size. A pilot cDNA syn- thesis reaction, labeled with [32P]dCTP, was carried out and analysed by electrophore- sis on an alkaline agarose gel. As expected, a population of molecules of different sizes was produced, some of them being apparently full length (data not shown).

Design of the PCR primers

For the cloning of the TuMV RNA genome, advantage was taken of the fact that a degenerate primer derived from conserved sequences can be used in conjunction with a specific primer for gene amplification (Ohara et al., 1989). The approach consisted of performing the amplification on ss-cDNAs of TuMV, with one specific primer complementary to TuMV RNA and one degenerate primer based on potyvirus publi- shed sequences. The amino acid sequences of the polyproteins derived from the nucleotide sequences of five potyviruses were compared and different regions pre- senting a high percentage of homology were selected. Fig. 1 shows two conserved regions which were chosen to derive the degenerate primers. The first region, from which JF3 was based, is located in the nuclear inclusion b (Nib) protein, approxi- mately 2500 amino acids downstream from the NH2-terminal amino acid of the poly- protein. This homology is also shared at the nucleotide level and the only differences in the sequences are located in the third nucleotide of the codons. The second region from which JF9 was deduced, is situated in the cytoplasmic inclusion (CI) protein, approximately 1450 amino acids downstream from the NH2-terminal amino acid of the polyprotein. Accordingly, degenerate primers that represented all possible nucleotides found in the respective potyvirus sequences were synthesized. JF3 was a 20-mer oligonucleotide with a 32-fold degeneracy and was entirely deduced from the Nib homology region. JF9 was a 25-mer oligonucleotide with a 24-fold degeneracy

61

TVMV ‘A- -c; ;:,- G,T ‘3/+ 3_.;- ‘T::

’ TEV __ _.. __. .___. ‘A- .(J.- c/.c GI; s,\- y- <;

G c- s,,- c;’ r, ’ JF3L-$ -G; “‘= ,: c

--- -. - ._._?_

Fig. 1. Genetic map based on five potyviruses showing the relative positions of two highly conserved regions from which were derived the degenetate PCR primers. The amino acid andnucleotide sequences for these two regions are given for five potyviruses: potato virus Y (PVY). two straim of plum pox virus (PPV), tobacco vein mottling virus (TVMV) and tobacco etch vitua (TEV). The two&rived&generate primers are boxed. Primer sequences are oriented 5’ to 3’. Lower case letters repnsent nuckotkkanot derived from the viral sequences. H.C.: helper c omponent; C.I.: cytoplasmic inclusion; VPg: ~-linked protein; NIa:

nuclear inclusion a; Nib nuclear inclusion b; C.P.: capsid pfoktL

and was composed of 17 nucleotides derived from the CI homology region to which the sequence 5’-GCGGATCC-3’ was added at* 5’4, TQix sequence contained the BumHI restriction site that allowed the eliminatton ’ “~&&&!%&during subcloning into M13mp19.

The specific PCR primers were synthesized from TuMV sequence data. The sequence for the primer FT2 was S’CGTTCTCAATGCACCAGAC-3’ and was complementary to TuMV RNA at position 435 to 456 upstream fromthe first nucleo- tide of the stop codon in the Tuh4V sequence. This primer was used in association with the JF3 primer for the amplification of an expected 1.2 kb fragment, The second speci- fic primer, FT8, was synthesized after the entire FE!-IF3 ampii&d fragment had been sequenced. Its sequence was 5’-ATACACGAATTCGGTGTAC-3’ and was complementary to TuMV RNA at position 1542 to 1561 from the first nucleotide of the stop codon, except for the ninth nucleotide from the 5’ end which originally was a T but was replaced by an A to create an EcoRI site. This primer was used in conjunc- tion with JF9 for the amplification of an expected 3.3 kb fragment. This fragment overlapped the 1.2 kb fragment by 102 bases and contained a BcA restriction site that allowed us to join the two fragments together after amplification and cloning.

62

Amplification and cloning cfrhe PCR products

The amplifications were carried out as described above in Materials and Methods. Fig. 2 shows an agarose gel in which the two amplification products were separated by electrophoresis. Note that one tenth of the 1.2 kb amplification was applied while the entire 3.3 kb PCR was loaded on the gel due to a very low yield. The amplification products were electroeluted and cloned into pUC9 plasmids. Following transforma- tion, small-scale plasmid preparations of approximately 20 white colonies on X-gal selection plates were prepared. These plasmids were digested with PstI and analysed by agarose gel electrophoresis for the presence of the appropriate insert. Fig. 2 shows restriction analysis of two clones, pTUM 1200 from the 1.2 kb amplification and pTUM3300 from the 3.3 kb amplification. The digestion with PstI of the clone pTUM1200 revealed that there was no such site in the insert. The insert was slightly longer than its corresponding amplification product due to the addition of a CC-tail. On the other hand, clone pTUM3300 had one Pstl site located in the insert but was cleaved out entirely from the plasmid with BamHI and Hind111 enzymes.

-5.0 -4.0 -3.0

-0.5

1234567

Fig. 2. Agarose gel stained withethidium bromide showing the results of amplification products and ampli- tied fragments cloned in pUC9 vector. Lane I : l/l0 of the I .2 kb amplification reaction; lane 2: the entire reaction of the 3.3 kb amplification; lane 3: pUC9 vector digested by !%I; lane 4: pTUMl200 digested by PsrI: lane 5: pTUM3300 digested by PSI; lane 6: pTUM33OOdigested by BumHI-HindIII; lane 7: 1 kb lad-

der (BRL) with sizes indicated on the right.

63

1 MC ATA ATT GAG AAT GGT GTC ACC CTA GAC ATT GAT GTG GTC GCC N I I E N G V T L D I D V V A

46 GAC 'S-M' GGA ACG AAA GTA CTT CCA TAT CTT GAC ACA GAC AAC AGG D F G T K V L P Y L D T D N R

91 ATG CM' AGC ACA ACA AAG ACA AGC ATT AAC TAC GGG GAG CGG ATT ?I L S T T K T S I N Y G E R I

136 CAA AGA CTT GGA AGA GTT GGG AGG CAC AAG CCA GGT CAT GCA CTA Q R L G R V G R H K P G ii A L

181 CGA ATT GGC CAC ACA GAG AGA GGA TTG AGC GGA GTC CCA AGT TGC R I G H T E R G L S E V P S C

226 ATT CCC ACA GM GCA GCT TTG MC TGC TTC ACT TAT GGA CTC CCA I A T E A A L K C F T Y G L P

271 GTA ATC ACC AAC AAC V I T N N

Fig. 3. Sequence of 285 nucleotides from the pTUM3300 insert. The predicted amino acid sequence of the single ORF is presented under the nucleotide sequence.

Sequence of the PCR products

The 1.2-kb fragment was sequenced entirely (data not shown). The sequence was similar to other potyvirus sequences and spanned part of the TuMV Nib and capsid protein cistrons (see Fig. 1 for the location of the proteins). For the 3.3 kb amplifi- cation product, 285 nucleotides from the 5’ end were sequenced (Fig. 3). That sequence had a unique open reading frame and coded for 95 amino acids. This amino acid sequence was compared with the five other potyvitus amino acid sequences (Fig. 4). This figure shows that the TuMV sequence was very similar to the potyvirus CI sequences and confirmed that the 3.3 kb fragment that was cloned was indeed derived from the TuMV genome. The TuMV sequence obtained from pTUM33~ shared respectively 70,75,70 and 65% of homology with PVY, PPV, TVMV and TEV. That amplified fragment would then cover part of the CI, the entire VPg and NIa, and part of the Nib cistrons.

Discussion

The use of degenerate primers with the 7’aq polymerase has an advantage over the other available methods for amplifying exceptionally long, uncharacterized cDNA because a degenerate primer along with a specific primer produce only one fragment of a predetermined size which can easily be detected and electroeluted after agarose gel efectrophoresis. cDNAs which are not of the correct size will not be amplified and will not be cloned, reducing the number of recombin~ts to screen. On the other hand, tailing or addition of adaptors to the cDNA generates fragments of different lengths

64

Tut-IV N I I E--i--C= D I D "%rbr

PVY N I IENGVTLDIDVVVDFGL

PPV N I IENGVTLDIDVVVDFGL

TVMV N I IENGVTLDVDVVVDFGL 0

TEV NIIENGVT DIDVVVDFG

TUW

PVY

PPV

TV?dV

TEV

TUMV

PVY

PPV I s

TVMV I s

TEV

TUMV

PVY

PPV

TVMV

TEV

TUMV

PVY

PPV

TVMV

TEV

I

YGERIQRLGRVGRHKPG

u YGERIQRLGRVGRFKKG

0

YGERIQRLGRVGRNKPG

LGERIQRFGRVGRNKPG cl

TGERIQ LGRVGRNKEG L

VA L R

V A L R I GjQ

ATEAALK

A T E A AL A L ATEAAFL

ATEAAFL

A T E A A FL

Fig. 4. Alignment of five potyviral amino acid sequences in the cytoplasmic inclusion protein correspond- ing to the region of TuMV which was sequenced in this work (see Fig. 3). Amino acids identical in at least three of the sequences are boxed. The first amino acid of the sequence lies 1463.1474, 1418, 1468 amino

acids from the NH*-terminal of the polyprotein for PVY, PPV, TVMV, TEV, respectively.

which require size selection and further amplifications. The success of the amplification depended greatly on the quality of the RNA. We

used RNA from purified viral preparations and ensured that it was not degraded. We did not attempt to use RNA from crude viral preparations but it was reported that a cer-

65

tain degree of purification of animal viruses was desirable before attempting the PCR (Donehower et al., 1990; Carman et al., 1989). Equally, we considered it to be safer to use purified RNA because the degenerate primers might hybridize non-specifically to contaminating nucleic acid materials. However, this requirement is balanced by the need for very low amounts of RNA. It was observed that if the region to be amplified laid closer to the 3’ end of the genome, the yield after PCR was greater than for regions located further upstream. This probably reflects the initial abundance of each region in the cDNA mix and the cDNA must be as long as possible to be sure that the degener- ate primer will anneal to it.

This approach showed that it was possible to amplify large fragments but it should be noted that the fidelity of the Taq polymerase is low, which may lead to amplifi- cation errors. Saiki et al. (1988) estimated that the rate of misincorporation is 0.25% in a 30-cycle amplification. Consequently, a sequence such as that shown in Fig. 3 will have to be determined from at least two other independent clones in order to confii the result. Recently, a new thermostable bacterium, Thermococcus lirorulis, was reported (Neuner et al., 1990) and a DNA polymerase with editing functions was puri- fied and is now commercially available. Its use would overcome the limitations impo- sed by the Tuq polymerase and would be more appropriate for cDNA cloning by PCR.

This approach was found to be simple and was considered to be applicable for clon- ing of large cDNA fragments of other plant viruses. Sequences of many plant viruses are available and viruses of the same group have regions of homology. Since seg- ments in the range of 1 to 3 kb can be amplified from a small amount of purified RNA, most plant viral genomes could be cloned entirely by carrying on only a few amplifi- cations. By choosing judiciously the position of the specific primer for the next PCR, the amplified fragments can be joined together to produce a full-size cDNA clone. It might be argued that traditional cDNA library screening for a full-length clone is more rapid but it requires a large amount of RNA for the cDNA synthesis and experi- ence of the technique. Here, the amplification takes only a few hours and the outcome is easily monitored.

Acknowledgements

The authors would like to acknowledge the technical expertise of Karl Whissel. 0. Nicolas is supported by a studentship from Le fond concert6 pour l’aide a la r6cherche (FCAR) du Quebec. This work was supported by L’Entente auxilliaire Canada-QuC- bet sur le developpement agro-alimentaire.

References

Akowitz, A. and Manuelidis, L. (1989) A novel cDNA/PCR strategy for efficient cloning of small amounts of undefined RNA. Gene 8 I. 295-306.

Allison, R., Johnston, R.E. and Dougherty, W.G. (1986) The nucleotide sequence of the coding region of tobacco etch virus genomic RNA: evidence for the synthesis of a single polyproteirt. Virology 154,9- 20.

Carman, W.F., Williamson, C., Cunliffe, B.A. and Kidd, A.H. (1989) Reverse transcription and subsequent

66

DNA amplification of rubella virus RNA. J. Virol. Methods 25.21-30. Choi, J.K., Maeda, T. and Wakimoto, S. (1977) An improved method forpurification of turnip mosaic virus.

Ann. Phytopathol. Sot. Japan 43,440-448. Domier, L.L., Franklin, K.M., Shahabuddin, M., Hellmann, G.M., Overmeyer, J.M., Hiremath, S.T., Siaw,

M.F.E., Lomonossoff, G.P., Shaw, J.G. and Rhoads, R.E. (1986) The nucleotide sequence of tobacco vein mottling virus RNA. Nucleic Acids Res. 14,5417-5430.

Donehower, L.A., Bohannon, R.C., Ford, R.J. and Gibbs, R.A. (1990) The use of primers from highly con- served pol regions to identify uncharacterized retroviruses by the polymerase chain reaction. J. Virol. Methods 28,33-46.

Dougherty, W.G. and Carrington, J.C. (1988) Expression and function of potyviral gene product. Ann. Rev. Phytopathol. 26,123-143.

Frohman, M.A., Dush, M.K. and Martin, G.R. (1988) Rapid production of full-length cDNAs from rare transcripts: amplification using a single gene-specific oligonucleotide primer. Proc. Natl. Acad. Sci. USA 85,8998-9002.

Grady, L.J. and Wayne, P. (1989) Amplification of large RNAs (> 1.5 kb) by polymerase chain reaction. Biotechniques 7.798-800.

Gubler, U. and Hoffman, B.J. (1983) A simple and very efficient method for generating cDNA libraries. Gene 25,263-269.

Ham&an, D. (1983) Studies on transformation ofB. coli with plasmids. J. Mol. Biol. 166.557-580. Hyypia, T., Auvinen, P. and Maaronen, M. (1989) Polymerase chain reaction for human picomavirus. J.

Gen. Virol. 70,3261-3268. Innis, M.A. and Gelfand, D.H. (1990) Optimization ofPCRs. In: M.A. Innis, D.H. Gelfand, J.J. Sninsky and

T.J. White (Eds), PCR Protocols, a Guide to Methods and Applications. Academic Press Inc., New York, pp. 3-12.

Lain, S., Riechmann, J.L. and Garcia, J.A. (1989) The complete nucleotide sequence of plum pox potyvirus RNA. VirusRes. 13,157-172.

Le Gall, O., Candresse, T., Brault, V., Bretout, C., Hibrand, L. and Dunez, J. (1988) Cloning full-length cDNA of grapevine chrome mosaic nepovirus. Gene 73,67-75.

Maiss, E., Breyel, E., Brisske, A. and Casper, R. (1988) Molecular cloning of DNA complementary to the RNA-genome of plum pox virus (PPV). J. Phytopathol. 122,222-23 1.

Maiss, E., Timpe, U., Brisske, A., Jelkmann, W., Casper, R., Himmler, G., Mattanovich, D. and Katinger, H. W.D. ( 1989) The complete nucleotide sequence of plum pox virus RNA. J. Gen. Virol. 70,5 13-524.

Maniatis, T., Fritsch, E.F. and Sambrook, J. (1989) Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, New York.

Neuner, A., Jannasch, H.W., Belkin, S. and Stetter, K.O. (1990) Thermococcus litorulis sp. nov.: a new species of extremely thermophylic marine archaebacteria. Arch. Microbial. 153,205-207.

O&man, H., Gerber, AS. and Hartl, D.L. (1988) Genetic applications of an inverse polymerase chain reac- tion. Genetics 120,621-623.

Ohara, O., Dorit, R.L. and Gilbert, W. (1989) One-sided polymerase chain reaction: the amplification of cDNA. Proc. Natl. Acad. Sci. USA 86,5673-5677.

Robaglia, C., Durand-Tardif, M., Tronchet, M., Boudazin, G., Astier-Manifacier, S. and Casse-Delbart, F.(1989) Nucleotide sequence of potato virus (N strain) genomic RNA. J. Gen. Virol. 70,935-947.

Saiki, R.K., Scharf, S.J., Faloona, F., Mullis, K.B., Horn, G.T., Erlich, H.A. and Amheim, N. (1985) Enzy- matic amplification of globin genomic sequence and restriction site analysis for diagnosis of sickle cell anemia. Science 230,1350-1354.

Saiki, R.K., Gelfand, D.H., Stoffel, S., Scharf, S.J., Higuchi, R., Horn, G.T., Mullis, K.B. and Erlich, H.A. (1988) Primer-directed enzymatic amplification of DNA with a thermostable DNA polymerase. Science 239,487491.

Sanger, F., Nicklen, S. and Coulson, A.R. (1977) DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74,5463-5467.

State-Smith, R. and Tremaine, J.H. (1970) Purification and composition of potato virus Y. Phytopatology 60,1785-1789.

Tabor, S. and Richardson, C.C. (1987) DNA sequence analysis with a modified bacteriophage T7 DNA polymerase. Proc. Natl. Acad. Sci. USA 84,4767-477 1.

Tremblay, M.-F., Nicolas, O., Lazure, C., Sinha, R.C. and Laliberte, J.-F. (1990) Sequence of the 3’-termi- nal region of turnip mosaic virus RNA and the capsid protein gene. J. Gen. Viral., in press.