Restriction Enzyme Analysis of Granulosis Viruses Isolated from Artogeia rapae and Pieris brassicae

Transcription

1 J. gen. Virol. (1986), 67, Printed in Great Britain 781 Key words: baculovirus/restriction profiles/ G V/host range Restriction Enzyme Analysis of Granulosis Viruses Isolated from Artogeia rapae and Pieris brassicae By NORMAN E. CROOK Glasshouse Crops Research Institute, Worthing Road, Litttehampton, West Sussex BN17 6LP, U.K. (Accepted 31 December 1985) SUMMARY Thirteen isolates of granulosis virus from Artogeia (= Pieris) rapae and two from Pieris brassicae were compared by restriction enzyme analysis. All the isolates gave very similar fragment profiles with XhoI, SmaI and BglI but at least 11 of them could be distinguished using EcoR], BstI and HindlII. Similarities and differences between profiles suggested that the isolates could be placed in three subtypes. This subtyping correlated closely with the geographical origin of the isolates, which came from Europe, North America, Asia and Australasia. All the isolates were highly infectious for with median lethal dose values for third instar larvae ranging from to 10 2'6. Only two of four isolates of one subtype had significant infectivity for third instar P. brassicae; thus, this broader host range did not correlate with grouping by restriction enzyme analysis. Artogeia (= Pier&) rapae, the small white butterfly, occurs across Europe, Asia, Australasia and North America, frequently causing severe damage to cruciferous crops. Granulosis virus (GV) has been isolated from larvae in all four continents and has been shown to be a successful means of controlling the pest (Tanada, 1956; Kelsey, 1958; Wilson, 1960; Akutsu, 1971; Hostetter et al., 1973; Ito & Sakikiyama, 197.5; Jaques & Laing, 1978; Sears et al., 1983; Tatchell & Payne, 1984). A GV (PbGV) isolated from Pieris brassicae, the large white butterfly, has also been shown to infect (Crook, 1981). Restriction enzyme analysis of the GVs from A. rapae and P. brassicae showed that they were distinct but closely related viruses (Crook, 1981). Restriction enzyme analysis of 12 more GV isolates from and one other isolate from P. brassicae has now shown that there are a large number of variants within this group of viruses. Similarities between restriction profiles from different isolates can be used to subtype variants into groupings which seem to reflect their geographical origins. The virus isolates are listed in Table 1. Although nos. 1 and 2 were isolated from P. brassicae, naming them 'P. brassicae GV' might imply that they were different viruses to the other isolates. Since all isolates are closely related variants of one virus, they will all be referred to as GV (ArGV) followed by the isolate number. Virus was propagated in and purified as previously described (Crook & Payne, 1980). The methods described by Crook et al. (1985) were used for viral DNA extraction and restriction enzyme analysis. Bioassays were performed as described by Crook (1981). An initial comparison of ArGV 6 to 8 failed to show any difference by restriction enzyme analysis with any of the enzymes used in this study. ArGV 7 and 8 were therefore not included in the comparison with the other isolates. DNAs from all the other isolates were compared by digestion with XhoI, SmaI, PstI, BglI, EcoRI, BstI and HindlII. The overall close relationship of all the isolates was most clearly demonstrated using XhoI. XhoI digests of all the isolates produced very similar electrophoretic profiles (Fig. 1). All contained five fragments with only slight size variations between different isolates; fragments XhoI-C and -D were very similar in SGM

2 782 Short communication Table 1. Isolate no. Origin 1 Cambridge, UK. 2 Reading, U.K. 3 Slindon, Sussex, U.K. 4 New Zealand 5 Canada 6 Brisbane, Australia 7 Brisbane, Australia 8 Queensland, Australia 9 Taiwan 10 Hunan, China 11 Beijing, China 12 Shanghai, China 13 Wuhan, China 14 Wuhan, China 15 Ichang, China Granulosis virus isolates from P. brassicae and Year of isolation Isolated if known from 1955 P. brassicae* 1983 P. brassicae * Occurred in P. brassicae and (Smith & Rivers, 1956). Supplier and comments See Crook (1981), referred to as PbGV. Mr M. A. Chaudrey. From a laboratory colony at Tate & Lyle Ltd, Reading, U.K. Miss C. Williams. From a single diseased larva. Dr J. M. Kelsey, Entomology Division, DSIR, Ashburton, N.Z. in See Kelsey (1957). See Crook (1981), referred to as PrGV. Isolate A93 "~ Dr R. Teakle, Department of Primary Industries, Isolate B64 Entomology Branch, Indooroopilly, Queensland, lsolate B72 Australia. Dr R. I. Rose, Asian Vegetable Research & Development Centre, Shanhua, Taiwan. Prof. Gao Shang Yin and Prof. N. C. Liu, Wuhan University, Wuchang, Hubei, P.R.C. Dr J. Adams, Beltsville Agricultural Research Center, Maryland, U.S.A. Described as isolate US/PRC-79-2 obtained from Institute of Ecology, China Academy of Sciences, Beijing, P.R.C. size and for some isolates these two fragments co-migrated. A comparison of the profiles obtained following digestion with HindIII, BstI and EcoRI (Fig. 2) demonstrated differences between at least 12 of the isolates; ArGV 6, 7 and 8 gave indistinguishable profiles with these enzymes and ArGV 13 and t4 differed only in their submolar bands (e.g. the largest BstI fragment is submolar but there is a higher proportion in ArGV 14 than 13), indicating that they contained the same major variant but probably differed in the proportion of minor variants present. All other isolates could be distinguished by at least one of the enzymes and some showed a number of differences with all three of these enzymes. Similarities and differences between the profiles indicated that there were clearly defined subtypes of ArGV. For instance, H&dIII profiles of ArGV 10 to 15 were virtually indistinguishable and those of ArGV 6 and 9 were very similar to these six; ArGV 6 had an additional H&dIII site in the largest fragment and ArGV 9 showed slight size differences in two of the fragments. On the other hand, HindIII profiles of the other isolates had less than half of their fragments co-migrating with those of ArGV 6 to 15. ArGV 1 to 4, however, also showed their own distinct pattern of fragments with only minor variation within this group. It appeared therefore that isolates could be subtyped into three distinct groupings: subtype A, ArGV 1 to 4;

3 Short communication A --B --C,D --E Fig. 1. Electrophoresis on 0'7~o agarose gel of DNAs of ArGV isolates! to 6 and 9-to 15, dlgestect with XhoI. The left-hand lane contains 2 DNA digested with HindIlI for reference. Gels were run in 40 mm- Tris-acetate, 20 mm-sodium acetate, 10 mm-edta, ph 7.8 at 50 V for 7.5 h in a Bethesda Research Laboratories H5 gel apparatus. subtype B, ArGV 5; and subtype C, ArGV 6 to 15. Examination of the EcoRI and BstI restriction profiles clearly supports this grouping. All isolates lacked any ApaI sites and contained a single Sinai site, two BglI sites and one or two PstI sites. These are shown most clearly in double digests with XhoI and Fig. 3 shows profiles obtained with one isolate from each subtype. Double digests with ApaI and XhoI or EcoRI gave profiles identical to single digests with the second enzyme (results not shown). The SmaI site occurred on the XhoI-B fragment in all the variants. One PstI site occurred on the XhoI-C fragment in all isolates and subtype A isolates contained a second PstI site on the XhoI-A fragment. BglI sites occurred on the XhoI-A and -B fragments; the smallest double digest fragment co-migrated with the XhoI-E fragment from ArGV 1 and 11. The small number of sites for these enzymes is unusual since, although many baculovirus DNAs have been analysed using a wide range of enzymes, no profiles have been published demonstrating unique sites or a complete lack of sites for any enzyme. Although this is unusual, it is presumably partly a consequence of the low GC content of these DNAs and the high GC content in the recognition sequence of these enzymes (Sinai, ApaI and BglI, 100~; PstI, 67~). Nevertheless, these low frequency sites could be useful in facilitating genetic manipulation of these large genomes and also in allowing the production of mutations at specific sites on the molecule. Profiles from a number of isolates contained submolar bands, indicating that more than one genotypic variant was present in the samples. In some of these cases, different variants have been obtained from a single isolate by infecting neonate with limiting dilutions of virus (approx. 0.1 capsules/larva) such that only about 5 ~ of larvae became infected. Under these conditions it is very likely that infected larvae receive only a single infectious virus particle and therefore the progeny virus is a clone of a single variant. Previous results (Crook, 1981) had shown that BstI digests of ArGV 1 (PbGV) contained a submolar band. This band was not

4 (a) (b) (c) Fig. 2. Electrophoresis on 0-7 ~ agarose gel o f A r G V D N A s digested with (a) HindlII, (b) Bstl and (c) EcoRI. Conditions of electrophoresis and designation of a n e s were the same as for Fig P~

5 Short communication 785 x x+s x+p X+B ' ~ '1 5 11' '1 5 11' '1 5 11' Fig. 3. Electrophoresis on 0.7 ~ agarose gel of ArGV DNA from isolates 1,5 and 11 digested with XhoI (X), and double digests with XhoI and SmaI (X + S), XhoI and PstI (X + P), and XhoI and Bgll (X + B). Conditions of electrophoresis were the same as for Fig. 1. present in clones of virus obtained from ArGV 1. Although no clone was obtained from ArGV 1 containing the submolar band, ArGV 2 does contain a BstI fragment of this size and is otherwise very similar to ArGV 1 (there are very small size changes in two EcoRI fragments). It is therefore likely that ArGV 2 is the same genotype that is the minor component in isolate 1. The relatedness of the ArGV isolates as indicated by the restriction profiles of their DNA correlates closely with their geographical origins. All the Chinese isolates are very similar whilst the other subtype C isolates come from Taiwan and Queensland. The single North American isolate is not closely related to any other isolate and is in subtype B on its own. Three of the four subtype A isolates were found in England whilst the fourth was from New Zealand. The New Zealand virus appears to have been isolated prior to ArGV being sent from England (Kelsey, 1958) but since was introduced into New Zealand from Britain, it is likely that virus was introduced by the same route. is also an introduced species in Australia and America. The Australian virus appears to have come from Asia but there is no indication from these results of the origin of the Canadian isolate. Possibly it is an endemic strain from native species of Pieridae. The fact that all the isolates of ArGV that have been obtained from different locations are distinct, indicates that there are probably many more different variants since A. rapae occurs in many areas from which no isolates have been obtained. The virus isolates were bioassayed in third instar larvae of both and P. brassicae. Preliminary data indicated that all the isolates were highly infectious for. Five isolates (ArGV 1, 4, 5, 9 and 11) representing all the subtypes were accurately assayed using five doses of

6 786 Short communication virus at 0.5 log intervals and infecting 50 larvae with each dose. There were no significant differences between the LDs0 values obtained for these isolates which were all in the range of to capsules. Previous results (Payne et al., 1981) had shown that ArGV 1 was moderately infectious for P. brassicae with an LDs0 value for third instar larvae of 105"8. An assay of the closely related ArGV 2 showed that this was similarly infectious with an LDs0 value of 105'9. In both these cases the EcoRI restriction enzyme profile of the progeny virus was identical to that of the inoculum virus. It was extremely difficult to cause infection of third instar P. brassicae larvae with any of the other isolates. Ability to cause infection was measured by feeding groups of 50 third instar larvae the massive dose of 109 capsules/larva. Ten of the isolates failed to cause any deaths even at this dosage. The other three isolates, ArGV 5, 11 and 14, caused a few deaths (~< 10~) but the amount of progeny virus was insufficient to check its identity by restriction enzyme analysis. There is therefore a major difference in the host range of ArGV 1 and 2 compared to all the other isolates. This difference is not reflected in the restriction profiles of the isolates. ArGV 3 and 4 are genetically more closely related to ArGV 1 and 2 than they are to ArGV 5 to 15 yet ArGV 3 and 4 caused no deaths in third instar P. brassicae and in this respect are more similar to ArGV subtypes B and C. Thus, a considerable amount of variation occurs within the ArGV genome, as shown in isolates 3 to 15, without causing any significant change in its virulence for or avirulence for P. brassicae. Yet the relatively minor changes between ArGV 1 and 2, on the one hand, and ArGV 3 and 4, on the other, cause a major change (> 104-fold) in virulence for P. brassicae although not affecting the virulence for. Although a number of variants have been identified for other baculoviruses such as Autographa californica MNPV (Smith & Summers, 1979), Mamestra brassicae MNPV (Vlak & Gr6ner, 1980; Wiegers & Vlak, 1984), Lacanobia oleracea GV (Crook et al., 1982), Heliothis SNPV and MNPV (Gettig & McCarthy, 1982; Williams & Payne, 1984), Spodopterafrugiperda MNPV (Maruniak et al., 1984) and Cydia pomonelta GV (Crook et al., 1985) and small differences in the infectivities of different isolates have been found (Williams & Payne, 1984), there are no reports of very large changes in the infectivity for a host insect especially where the infectivity for another host remains unchanged. The ArGV isolates therefore provide a unique system for studying the effect of genotypic differences on virulence and host range. REFERENCES AKUTSU, K. (1971). Control of the common cabbagevcorm, Pieris rapae erucivora Boisduval by a granulosis virus. Japanese Journal o[ Applied Entomology and Zoology 15, CROOK, N. E. (1981). A comparison of the granulosis viruses from Pieris brassieae and Pieris rapae. Virology 115, CROOK, N. ~. & PAYNE, C. C. (1980). Comparison of three methods of ELISA for baculoviruses. Journal of General Virology 46, CROOK, N. E., BROWN, J. D. & FOSTER, G. N. (1982). Isolation and characterization of a granulosis virus from the tomato moth, Lacanobia oleracea, and its potential as a control agent. Journal of Invertebrate Pathology 40, CROOK, N. E., Sl'ENCBR, g. A., PAVNE, C. C. & LEISV, D. J. (1985). Variation in Cydiapomonella granulosis virus isolates and physical maps of the DNA from three variants. Journal of General Virology 66, GETTIG, R. R. & McCARTHY, W. J. (1982). Genotypic variation among wild isolates of Heliothis spp nuclear polyhedrosis viruses from different geographical regions. Virology 117, HOSTETTER, D. L., PINNELL, R. E., GREER, P. A. & IGNOFFO, C. M. (1973). A granulosis virus of Pieris rapae as a microbial control agent on cabbage in Missouri. Em, ironmental Entomology 2, ITO, Y. & SAKIKIYAMA, M. (1975). Population dynamics of Pieris rapae crucivora Boisduval (Lepidoptera, Pieridae) an introduced pest in Okinawa. III. Results of the introduction of Apanteles glomeratus and a cabbage butterfly granulosis virus, Japanese Journal of Applied Entomology and Zoology 19, JAQUES, R. P. & LAING, D. R. (1978). Efficacy of mixtures of Bacillus thuringiensis, viruses and chlordimeform against insects on cabbage. Canadian Entomologist 110, KELSEY, J. M. (1957). Virus sprays for the control of Pieris rapae (L). New Zealand Journal of Science and Technology 38, KELSEY, J. M. (1958). Control of Pieris rapae by granulosis viruses. New Zealand Journal of Agricultural Research 1, MARUNIAK, J. E., BROWN, S. E. & KNUDSON, D. L. (1984). Physical maps of SfMNPV baculovirus DNA and its genomic variants. Virology 136,

7 Short communication 787 PAYNE, C. C., TATCHELL, G. M. & WILLIAMS, C. F. (1981). The comparative susceptibilities ofpieris brassicae and P. rapae to a granulosis virus from P. brassicae. Journal of Invertebrate Pathology 38, SEARS, K. M., JAQUES, R. P. & LAING, J. E. (1983). Utilization of action thresholds for microbial and chemical control of lepidopterous pests (Lepidoptera: Noctuidae, Pieridae) on cabbage. Journal of Economic Entomology 76, SMITH, G. E. & SUMMERS, M. D. (1979). Restriction maps of five Autographa californica MNPV variants, Trichoplusia ni MNPV, and Galleria mellonella MNPV DNAs with endonucleases SmaI, KpnI, BstI, SacI, XhoI, and EcoRI. Journal of Virology 30, SMITH, K. M. & RIVERS, C. F. (1956). Some viruses affecting insects of economic importance. Parasitology 46, TANADA, Y. (1956). Microbial control of some lepidopterous pests of crucifers. Journal of Economic Entomology 49, TATCHELL, G. M. & PAYNE, C. C. (1984). Field evaluation of a granulosis virus for control of Pieris rapae (Lep. : Pieridae) in the United Kingdom. Entomophaga 29, VLAK, J. M. & GRONER, A. (1980). Identification of two nuclear polyhedrosis viruses from the cabbage moth, Mamestra brassicae (Lepidoptera: Noctuidae). Journal of Invertebrate Pathology 35, WIEGERS, F. P. & VLAK, J. M. ([984). Physical map of the DNA ofa Mamestra brassicae nuclear polyhedrosis virus variant isolated from Spodoptera exigua. Journal of General Virology 65, WILLIAMS, C. F. & PAYNE, C. C. (1984). The susceptibility of Heliothis armigera larvae to three nuclear polyhedrosis viruses. Annals of Applied Biology 104, WILSON, F. (1960). The effectiveness of a granulosis virus applied to field populations of Pieris rapae (Lepidoptera). Australian Journal oj Agricultural Research 11, (Received 6 August 1985)