The detection of beet western yellows virus and beet mild yellowing virus in crop plants using the polymerase chain reaction

Journal of Virological Methoak, 35 (1991) 287-296 0 1991 Elsevier Science Publishers B.V. / All rights reserved / 0166-0934/91/%03.50

287

VIRMET 01262

The detection of beet western yellows virus and beet mild yellowing virus in crop plants using the polymerase chain reaction T.D. Jones’, K.W. Buck’ and R.T. Plumb* ‘Department of Biology, Imperial College of Science, Technology and Medicine, London, U.K. and ‘Plant Pathology Department, Institute of Arable Crops Research, Rothamsted Experimental Station, Harpenden, Herts, U.K.

(Accepted 26 July 1991)

Summary

Oligonucleotide primers were synthesised corresponding to conserved sequences between three isolates of beet western yellows virus (BWYV), flanking a 913 base fragment of BWYV genomic RNA. Using the polymerase chain reaction (PCR), these primers successfully amplified the target fragment in total. RNA extracts from two oilseed rape plants infected with different isolates of BWYV. The PCR products were readily detected by staining with ethidium bromide following agarose gel electrophoresis, but the limit of detection could be increased further by Southern blotting. However, three isolates of beet mild yellowing virus (BMYV) in sugar beet did not give a signal which could be detected by ethidium bromide staining, although the target fragment could be detected by Southern blotting. The primers used have the potential to detect BWYV in crops with far greater sensitivity than enzymelinked immunosorbent assay or nucleic acid hybridisation (dot-blotting) and may be capable of distinguishing between BWYV and BMYV. The apphcation of PCR to detection and distinction of luteoviruses in general is discussed. Beet western yellows virus; Beet mild yellowing virus; Polymerase chain reaction

Correspondence to: T.D. Jones, Department of Biology, Imperial College of Science, Technology and Medicine, London SW7 2BB, U.K.

288

Introduction Beet western yellows virus (BWYV), first described by Duffus (1960) is probably the most widespread member of the luteovirus group of plant viruses (Waterhouse et al., 1988). It infects a wide range of mono- and dicotyledonous plants world-wide, causing serious economic damage. A virus with similar properties to BWYV, beet mild yellowing virus (BMYV), first reported by Russell (1958), has been very difficult to distinguish from BWYV by serological, methods and it has generally been accepted that they are different strains of the same virus (Govier, 1985). Enzyme-linked immunosorbent assay (ELISA) has for many years been the method of choice for detecting and characterising luteoviruses (e.g. Hewings and D’Arcy, 1984) but it is only recently that antisera have become available which can distinguish between these two strains (D’Arcy et al., 1989) and also another closely related luteovirus, the RPV isolate of barley yellow dwarf virus (BYDV). Another method for detecting and distinguishing between luteoviruses is based upon the hybridisation of viral RNA to DNA probes generated from selected cDNA clones (Waterhouse et al., 1986; Skotnicki et al., 1987). This technique has comparable sensitivity to ELISA and probes can be selected to hybridise to only one virus serotype or to encompass a range of serotypes. The potential of this technique has been increased by the range of sequence data now available for different luteoviruses (Miller et al., 1988; Veidt et al., 1988; van der Wilk et al., 1989; Mayo et al., 1989), making it possible to target probes more effectively to conserved or varying sequences. We have investigated the possibility of increasing the sensitivity of detection of BWYV and BMYV by using the polymerase chain reaction (PCR) (Saiki et al., 1988) to amplify selectively a specific region of the viral genome. PCR has been used to detect a number of animal viruses and a DNA plant virus (Olive et al., 1990; Rybicki and Hughes, 1990). In a recent preliminary communication, we demonstrated amplification of cDNA synthesised from BWYV RNA from purified virus particles and from total RNA extracts of infected indicator plants (Montiu perfoliata) (Jones et al., 1989), but did not examine its potential for detecting the virus in crop plants and field samples. Here we report the detection of two isolates of BWYV in oilseed rape and three isolates of BMYV in sugar beet via PCR. We have also explored the limits of detection of this technique on its own and in conjunction with Southern blotting.

Materials and Methods Virus isolates Two BWYV isolates maintained in oilseed rape, one causing normal symptoms (BWYV-N) and one causing more severe symptoms (BWYV-S, and two BMYV isolates in sugar beet, one from the greenhouse (BMYV-BG) and

289

one from the field (BMYV-BF), were provided by Dr. Helen Smith (Brooms Barn Experimental Station, Suffolk). A further BMYV isolate in sugar beet (BMYV-R) was provided by Mrs E. Lennon (IACR Rothamsted, Harpenden, Herts). Preparation

of plant total RNA

Total plant RNA was prepared by modification of the method of Lichtenstein et al. (1986), 200 mg of leaf material (either fresh or frozen) was placed in a 1.5-ml Eppendorf tube and frozen in liquid nitrogen. The material was then ground to a coarse powder in the tube, using a 4-mm metal or glass round-ended stirring rod. Before use, the rod was flamed to red heat to remove contamination from previous preparations, to destroy ribonucleases and to sterilise. 400 ~1 of homogenisation buffer (0.2 M Tris-HCl pH 8.5, 0.2 M sucrose, 30 mM magnesium acetate, 60 mM potassium chloride, 1% polyvinylpyrollidone, 0.31% 2-mercaptoethanol) were added and mixed thoroughly as the leaf powder thawed, 30 ~1 10% w/v SDS was added, followed by 400 ~1 watersaturated phenol and 400 ~1 chloroform: isoamyl alcohol (24: 1 by volume). The phases were mixed thoroughly for 10 min and then separated by centrifugation for 5 min. The aqueous phase was removed and re-extracted with 400 ~1 chloroform as above. At this stage there was usually about 400 ,a1of clarified plant extract. To this was added 20 ,~l 3 M sodium acetate followed by 1 ml ice-cold ethanol and the mixture was stored at -70°C for 15 min to precipitate the nucleic acids. The precipitate was collected by centrifugation for 15 min at 4°C. The pellet was resuspended in 100 ~13 M sodium acetate and the suspension was incubated on ice for 15 min and recentrifuged as above. The RNA pellet was resuspended in 200 ~1 TE (10 mM Tris-HCl pH 8.0, 1 mM EDTA, 20 ~1 3 M sodium acetate were added and the solution was extracted with 200 ~1 phenol/chloroform. 500 ,~lethanol was added to the aqueous phase and the mixture was kept at - 70°C for 15 min. The precipitate was collected by centrifugation for 15 min at 4”C, washed with 70% ethanol, dried and resuspended in 50 ~1 sterile deionised diethyl pyrocarbonate treated water. This procedure gave yields of 50 to 100 ,ug of plant RNA. Primers

The primers were based on conserved sequences between three isolates of BWYV: an isolate from oiiseed rape (Jones et al., 1989), an isolate from lettuce (BWYV-L), sequenced by Veidt et al. (1988), and an isolate from sugar beet, partially sequenced by Veidt et al. (1988). Primers Pl and P2 had the sequences GTTGAACTTCTTTACTCGT and AGGGAGAAGGCCCTGGGCT, corresponding to nucleotides 3218-3236 and 4131-4113 respectively of BWYV-L (Veidt et al., 1988). These sites flank the coat protein gene and the intergenic

290

region between the coat protein gene and the polymerase 913 bases long. First

strand

gene. This fragment

is

cDNA synthesis

0.5 pg total plant RNA was diluted to 10 ~1 with sterile deionised water containing 50 pmols primer P2, heated to 65°C for 5 min and chilled on ice. 10 ~1 2 x reverse transcriptase (RT reaction mix (4 ~1 5 x M-MLV RT buffer (BRL), 1 ~10.1 M DTT, 1 ~12 mg/ml BSA, 2 ~1 10 mM dNTPs, 1 ,u1(200 units) M-MLV RT (BRL), 1 ~1 Inhibit-Ace (5’-3’)) were added and the mixture was incubated at 37°C for 1 h. Ampllyication

of first strand cDNA

2 ~1 of first-strand cDNA was added to 98 ~1 PCR reaction mixture (10 ~1 10 x PCR buffer (Promega), 1 ~150 mM primer Pl, 1 ~150 mM primer P2,2 ~1 10 mM dNTPs, 1 ,ul 5 mM tetramethylammonium chloride (Aldrich), 83 ~1 sterile deionised water) and heated to 95°C for 5 min, 2 units of Taq polymerase (Promega) were added and the reaction incubated in a Techne PHC-1 Programmable Dri-Block as follows : 30 s at 94°C 30 s at 40°C 18 s at 72°C 40 cycles. The PCR products were separated by electrophoresis in 1% agarose gels and the gels were stained with 0.5 pg/ml ethidium bromide. Prevention of cross contamination

between samples to be amplified

Cross-contamination between samples can be a considerable problem when using PCR, particularly when used as a diagnostic test. We have found that the greatest source of this is the barrels of automatic pipettes. For this reason all pipette tips used for the above procedures were plugged with small balls of nonabsorbent cotton wool. The avoidance of contamination is also the reason for grinding plant material in 1.5-ml Eppendorf tubes. Pestle and mortars when re-used have been shown to be a source of contamination even after washing and autoclaving. Probe preparation

and Southern blotting

[a-32P]dCTP-labelled probe was prepared by the random hexanucleotide labelling method of Feinberg and Vogelstein (1984). The template was a BstNl restriction fragment of a pUC19 clone of the region being amplified (corresponding to nucleotides 32694116 of BWYV-L RNA, Veidt et al., 1988). This fragment lacks both priming sites and hence is unlikely to hybridise to non-specifically amplified PCR products. The probe was designated P3. DNA was transferred from agarose gels to Hybond-N+ nylon membranes,

291

fixed and hybridised as described by the manufacturer (Amersham).

of the membrane

Northern blotting 1 ,ug total plant RNA, denatured with formamide/formaldehyde, was electrophoresed through 1% agarose gels containing formaldehyde, transferred to Hybond-N+ nylon membranes, fixed and hybridised as described by the manufacturer of the membrane (Amersham).

Results Northern blotting of the total preparations (Fig. 1) demonstrated that the probe P3 was able to detect all the virus isolates. The strength of the signal may depend more on the amount of viral RNA present in the plants than on the degree of homology with the probe. The strongest signal was given by BMYVBF, whereas the signal obtained from isolate BMYV-BG was the least strong. Amplification of the RNA preparations using primers Pl and P2 gave clearly different results between the BWYV and BMYV isolates (Fig. 2a). Both BWYV isolates gave a clear band of the expected size (913 bp, lanes 1 and 2) although the BWYV-N isolate gave a stronger signal. A band of this size was not detected in RNA from healthy beet or rape plants (Fig. 3a, lanes 15-18).

6.05.04.03.02.01.6l.O-

0.5-

Fig. 1. 1.O pg of total RNA preparations from infected or healthy plants, electrophoresed through a 1% denaturing agarose gel and Northern blotted. Lane 1 is BMYV-BG, lane 2 is BMYV-BF, lane 3 is BMYVR, lane 4 is BWYV-S, lane 5 is BWYV-N, lane 6 is healthy beet, lane 7 is healthy rape.

292

b

a 1

2

3

4

5

1

6

2

3

4

5

6

,6.0 ,5.0 ‘4.0 .3.0 2.0 ,I.6

Fig. 2. (a) PCRs of total RNA from infected plants electrophoresed through a 1% agarose gel and stained with 0.5 pg/ml ethidium bromide. The amount loaded in each lane is equivalent to 5 ng total plant RNA. Lane 1 is BWYV-N, lane 2 is BWYV-S, lane 3 is BMYV-R, lane 4 is BMYV-BF, lane 5 is BMYV-BG, lane 6 is BRL kb. Ladder marker, sizes as indicated in kbp. (b) Southern blot of the gel described in Fig. 2a. Lanes as indicated previously.

However, a non-specific band of about 500 bp was also amplified. A band in this position, but of much lower intensity, was sometimes seen in PCRs of total RNA from healthy plants (Fig. 3a, lane 16). Southern blotting of the gel in Fig. 2a (Fig. 2b) with probe P3 indicated that the longer of the two bands in lanes 1 and 2 was viral in origin. Virus-specific bands were also detected in the PCR products of the three BMYV isolates (lanes 3, 4 and 5) but the bands were not visualised with ethidium bromide staining. The signal obtained from the BMYV isolates is no stronger than that achieved by Northern blotting (Fig. 1). In Southern blots of BWYV-N (Fig. 3b, lanes 12-14) a larger virus-specific band of about 1.7 kb was also detected. The

Fig. 3. (a) PCRs of a dilution series of BWYV-N and BWYV-S. Total RNA from infected plants was diluted serially, 1 in 5, with total RNA from healthy plants. Each sample was subjected to the PCR protocol and the products (each equivalent to 5 ng of total plant RNA) were electrophoresed through a 1% agarose gel and stained with 0.5 pg/ml ethidium bromide. Lanes l-7, BWYN-S; lanes S-14, BWYNN; lanes 1 and 14, undiluted; lanes 2 and 13, 1:5 dilution; lanes 3 and 12, I:25 dilution; lanes 4 and 11, 1:125 dilution; lanes 5 and 10, I:625 dilution; lanes 6 and 9, 1:3125 dilution; lanes 7 and 8, I:15625 dilution. Lane M is BRL kb ladder marker, sizes as indicated in Fig. 2a. Lane 15 is PCR of healthy oilseed rape total RNA done with primers present in the RT and PCR reactions. Lane 16 is a PCR of healthy sugar beet total RNA done with primers present in the RT and PCR reactions. Lanes 17 and 18, respectively, are PCRs of healthy oilseed rape and healthy sugar beet total RNA done with no primers in the RT and PCR reactions. (b) Southern blot of the gel described in Fig. 3a. Lanes as indicated previously.

293

11

12

13

14

17

13

lb

M

15

16

17

18

b 173A5677

9

10

11

M

15

16

17

16

294

origin of this band is not known, but it could be the result of one of the primers binding to partially homologous sequences elsewhere in the genome of this virus. A dilution series was done to test the limits of detection of this technique when applied to the two BWYV isolates (Fig. 3a,b). Five-fold serial dilutions of total RNA from infected plants diluted in total RNA from healthy plants were made. The total amount of RNA present in each reverse transcriptase reaction was 0.5 ,ug. Both BWYV-N and BWYV-S can be detected visually with an unambiguous positive result, when compared to the healthy controls, at a dilution of 5-4 (625) (Fig. 3a). When the gel is Southern blotted and hybridised with probe P3 (Fig. 3b) the BWYV-N isolate can be detected at a dilution of 5-6, (15625), but the BWYV-S isolate can only be detected at a dilution of 5-5 (3125). This represents 32 pg and 160 pg of total RNA from infected plants in a total of 0.5 pg RNA, respectively. However, only 10% of the reverse transcriptase reaction was amplified and only 10% of the PCR products were analysed on the agarose gel. Therefore the equivalent amount of total RNA from infected plants loaded on the gel is only 0.32 pg and 1.6 pg respectively. The PCRs of healthy controls (Fig. 3a,b) gave no signal via ethidium bromide staining or Southern blotting. There is no ambiguity between the results from infected and healthy plants.

Discussion We have shown that the polymerase chain reaction is a highly sensitive method for the detection of BWYV in crop plants. The primers we used are based on the known sequence of three strains of BWYV and have been used to amplify two further strains in this work. The latter were detected in as little as 0.32 pg of total RNA from infected plants. The proportion of this which is viral RNA is not known. However, luteoviruses are present at very low concentrations and only in phloem and the immediately surrounding tissue, and it is likely that viral RNA comprises less than 1% of the total RNA in an infected plant (Jensen, 1969). The limit of sensitivity of PCR detection is therefore probably less than 3 fg of viral RNA. The limits of detection of dot-blotting and ELISA are approximately the same (Hewings and D’Arcy, 1984; Waterhouse et al., 1986; Skotnicki et al., 1987) being the equivalent of about 100 pg viral RNA. Therefore PCR is at least 3 x lo4 times more sensitive. This should enable BWYV infections in crops to be detected at a much earlier stage than has been possible hitherto. It should also provide a sensitive method for detection of the virus in its aphid vectors. The three BMVY isolates did not give as strong a signal as the BWYV isolates, even though the strongest signal in the Northern blot (Fig. 1) was given by BMYV-BF. This suggests that the efficiency of binding of the primers, and/or the ability of the polymerases (both RT and Taq) to initiate from the

295

primers, is greatly reduced in these isolates. Therefore there must be a significant variation in the sequences of the BMYV isolates at one or both of the priming sites, probably at the 3’ end which must be homologous for efficient priming to occur. It is noteworthy that four out of seven RNA probes, corresponding to different regions of the BWYV genome, described by Herbach et al. (1991) were able to detect BWYV but not BMYV by hybridisation. One of these probes, corresponding to nucleotides 3041 to 3423 includes the sequence of primer 1. These results raise the possibility that different strains of luteoviruses could be characterised on the ability to be amplified with a specific set of primers. It may be possible to select priming sites which amplify specific isolates with high efficiency, and to use degenerate primers to amplify a broad spectrum of isolates. It may therefore be possible to test a plant for the presence of BWYV and closely related luteoviruses with one set of primers, and then to identify the strain using a variety of other primers. Construction of primers to identify different luteoviruses should also be feasible. After this manuscript was complete, Robertson et al. (1991) described two primers (one of which was degenerate at one position) which allowed amplification of a 530 bp sequence of three luteoviruses, barley yellow dwarf virus, potato leafroll virus and BWYV. However, the applicability of the method to detection of BWYV in crop plants or field samples was not determined~ Moreover the sensitivity of detection was not reported. In addition to the diagnostic result of the presence of an amplified fragment, this fragment can then be probed for sequence changes between the priming sites via restriction endonuclease digestion, hybridisation to radiolabelled probe, or by sequencing. This would further increase the possibility of distinguishing between closely related viruses. Dot-blotting has been shown to identify, or differentiate between a number of luteoviruses (Waterhouse et al., 1986) and ELISA is now capable of distinguishing between BWYV and BMYV (D’Arcy et al., 1989). However, both techniques have disadvantages. ELISA can only detect alterations in the coat protein gene. It has been suggested (Mayo et al., 1989) that it is ORF 1 which determines host-range and that alterations in ORF 6 and the 3’ noncoding region represent adaptations to different hosts (Veidt et al., 1988). ELISA cannot detect these differences. Dot-blotting can be targeted to particular regions of the genome. However the signal will only be reduced by gross alterations to the RNA sequence. PCR has the advantage of targeting, such that alterations in only short stretches of sequence (the priming sites) can be detected, or it can be used to amplify a target sequence which can then be analysed further for more subtle changes. Hence it is likely that PCR will prove to be a versatile tool in detecting and distinguishing between luteoviruses.

296

References Caspar, R. (1988) Luteoviruses. In: R. Koenig (Ed), The Plant Viruses: Polyhedral Virions with Monopartite RNA genomes, Vol. 3, Plenum, New York. pp 2355258. D’Arcy, C.J., Torrance, L. and Martin, R.R. (1989) Discrimination among luteoviruses and their strains by monoclonal antibodies and identification of common epitopes. Phytopathol. 79, 8699 873. Duffus, J.E. (1961) Radish yellows, a disease of radish, sugarbeet and other crops. Phytopathol. 50, 389-394. Feinberg, A.P. and Vogelstein, B. (1983) A technique for radiolabelling DNA restriction endonuclease fragments to a high specific activity. Anal. Biochem. 137, 266-267. Govier, D.A. (1985) Purification and partial characterisation of beet mild yellowing virus and its serological detection in plants and aphids. Ann. App. Biol. 107. 439447. Herrbach, E., Lemaire, O., Ziegler-Graff, V., Lot, H., Rabenstein, F. and Bouchery, Y. (1991) Detection of BMYV and BWYV isolates using monoclonal antibodies and radioactive RNA probes, and relationships among luteoviruses. Ann. App. Biol. 118, 127-138. Hewings, A.D. and D’Arcy, C.J. (1984) Maximizing the detection capability of a beet western yellows virus ELISA system. J. Virol. Methods 9, 131-142. Jensen, S.G. (1969) Occurrence of virus particles in the phloem tissue of BYDV infected barley. Virology 38, 83-91. Jones, T.D., Buck, K.W. and Plumb, R.T. (1989) The detection and analysis of sequence variation of beet western yellows virus. In: Production and protection of oilseed rape and other brassica crops. Aspects of Applied Biology vol. 23, Association of Applied Biologists, Warwick, pp. 319327. Lichtenstein, C.P. and Draper, J. (1985) Genetic engineering of plants. In: D.M. Glover (Ed), DNA Cloning, Vol. 2, IRL Press, Oxford, pp. 67-119. Mayo, M.A., Robinson, D.J., Jolly, C.A. and Hyman, L. (1989) Nucleotide sequence of potato leafroll luteovirus RNA. J. Gen. Virol. 70, 1037-1051. Miller, W.A., Waterhouse, P.M. and Gerlach. W.L. (1988) Sequence and organisation of barley yellow dwarf virus genomic RNA. Nucleic Acids Res. 16, 6097-611 I. Olive, D.M., Al-Mufti, S., Al-Mulla, W., Khan, M.A., Pasta, A., Stanway, G. and Al-Nakib, W. (1990) Detection and differentiation of picornaviruses in clinical samples following genomic amplification. J. Gen. Viral 71, 2141-2147. Robertson, N.L., French, R. and Gray, S.M. (1991) Use of group-specific primers and the polymerase chain reaction for the detection and identification of luteoviruses. J. Gen. Viral. 72, 147331477. Russell, G.E. (1958) Sugar beet yellows: a preliminary study of the distribution and interrelationships of viruses and virus strains found in East Angha, 1955-1957. Ann. App. Biol. 46. 3933398. Rybicki. E.P. and Hughes, F.L. (1990) Detection and typing of maize streak virus and other distantly related geminiviruses of grasses by polymerase chain reaction amplification of a conserved viral sequence. J. Gen. Viral. 71. 251992526. Waterhouse, P.M., Gildow, F.E. and Johnstone, G.R. (1988) Luteovirus group. AAB Descriptions of Plant Viruses, No. 339.