M Karan, RM Harding, JL Dale, 1997. Association of Banana Bunchy Top Virus DNA Components 2 to 6 with Bunchy Top Disease. Molecular Plant Pathology On-Line 1997/0624karan

MPPOL

Introduction
Materials & Methods
Results
Discussion
References

Molecular Plant Pathology On-line [http://www.bspp.org.uk/mppol/] 1997/0624karan

Association of Banana Bunchy Top Virus DNA Components 2 to 6 with Bunchy Top Disease.

Mirko Karan¹ , Robert M Harding, James L Dale*

Centre for Molecular Biotechnology, School of Life Sciences, Queensland University of Technology, GPO Box 2434, Brisbane 4001, Queensland, Australia.

Current address: ¹ Biochemistry Department, School of Molecular Sciences, James Cook University of North Queensland, Townsville 4811, Queensland, Australia.
Fax: +61 77 251 294, Email: Mirko.Karan@jcu.edu.au

Corresponding
author

Professor James L Dale, Centre for Molecular Biotechnology, School of Life Sciences, Queensland University of Technology, GPO Box 2434, Brisbane 4001, Queensland, Australia.
Telephone: +61 7 3864 2819 Fax: +61 7 3864 1534 email: j.dale@qut.edu.au

ABSTRACT

Banana bunchy top virus (BBTV) has a genome consisting of six different ssDNA components that have been identified in an Australian BBTV isolate. Further, BBTV DNA-1 has previously been shown to be present in all BBTV isolates tested from ten countries. The sequences of BBTV DNA-1 from the different isolates divided into two groups, the South Pacific group and the Asian group. We have now demonstrated, by Southern analysis, PCR and sequencing, that BBTV DNA-2, 3, 4, 5 and 6 are present in all isolates tested from eight countries suggesting that BBTV DNA-1 to 6 are integral components of the BBTV genome. The sequences of BBTV DNA-6 from the different isolates also divided into the two groups, the South Pacific group and the Asian group. Interestingly, there was far greater sequence variability between the sequences within the Asian group than those within the South Pacific group of isolates suggesting that either BBTV has been present in bananas in this region for a longer period or there has been more than one introduction of BBTV into bananas from another host in this region.

INTRODUCTION

Banana bunchy top virus (BBTV) is the most economically important virus infecting bananas (Dale, 1987). The virus is transmitted persistently by the aphid Pentalonia nigronervosa but does not appear to replicate in its vector (Hafner et al., 1995), is not mechanically transmissible and is apparently phloem limited. BBTV has only recently been isolated and characterised (Wu and Su, 1990; Harding et al., 1991). The virus has isometric virions 18-20 nm in diameter and a genome that consists of multiple circular single-stranded DNA components each about 1.1 kb. A single protein of 19.6 kDa is associated with the virions and is assumed to be the coat protein (Harding et al., 1991; Burns et al., 1994). A number of BBTV ssDNA components have been identified and sequenced. Harding et al.(1993) reported the sequence of BBTV DNA component 1 (BBTV DNA-1) from an Australian isolate of BBTV. This component is 1.1kb and contains one large open reading frame (ORF) in the virion sense. This ORF encodes a putative replication protein (Rep) based on the presence of the dNTP binding motif GGEGKT. Associated with this ORF are a potential TATA box and polyadenylation signals. BBTV DNA-1 also contains a sequence capable of forming a stable stem-loop structure; 9 nucleotides of the loop sequence are very similar to the highly conserved nonanucleotide loop sequence present in geminivirus stem-loop structures (Lazarowitz, 1992). Yeh et al. (1994) and Wu et al. (1994) have reported the sequences of three additional components from Taiwanese BBTV isolates. These three components have a similar structure to BBTV DNA-1 in that each contains a sequence capable of forming a stable stem-loop structure and includes the highly conserved nonanucleotide loop sequence and the two components of Wu et al. (1994) each contained one large ORF which encode putative Reps. These Reps are different to each other and different to that encoded by BBTV DNA-1. The component of Yeh et al. (1994) could potentially encode a putative Rep if the sequence is slightly altered and this potential Rep would be different to that encoded by BBTV DNA-1 and almost identical to that for DNA-1 of Wu et al. (1994).

Burns et al. (1995) have recently reported the sequence and structure of five BBTV components called BBTV DNA-2, 3, 4, 5 and 6; none of these components encode Reps. Each component has a structure similar to that of BBTV DNA-1 in that each component has one large ORF in the virion sense (except for BBTV DNA-2) with an associated potential TATA box and polyadenylation signal(s) and a sequence 5' of the ORF capable of forming a stable stem-loop structure; the loop sequence in each component contains the highly conserved nonanucleotide sequence. Further, Burns et al. (1995) identified two regions that are highly conserved between all six components (BBTV DNA-1 to 6) identified from Australian BBTV isolates. The stem-loop common region (CR-SL) extends up to 25 nucleotides 5' of the stem-loop sequence and up to 13 nucleotides 3' of the stem-loop sequence. The major common region (CR-M) is located 5' of the CR-SL and is between 65 and 92 nucleotides in length. The functions of the putative proteins encoded by the large virion sense ORFs in BBTV DNA-3, 4, 5 and 6 have not been determined.

A further four BBTV-like viruses have been identified. All these viruses have isometric virions about 20 nm in diameter, genomes of ssDNA about 1 kb and three of them are persistently transmitted by aphids. These four viruses, together with BBTV, are probably members of a new plant virus group. Boevink et al. (1995) have identified and sequenced seven ssDNA components of subterranean clover stunt virus (SCSV). All have one large ORF in the virion sense, a sequence 5' of the ORF capable of forming a stable stem-loop structure and the loop sequence includes the highly conserved nonanucleotide sequence. Five of the components share a conserved common region. SCSV-5 has been shown to encode the SCSV coat protein (Chu et al., 1993; Boevink et al., 1995) and two components encode putative Reps. Interestingly, the two Rep encoding components do not contain the common region. One component has been identified and sequenced from coconut foliar decay virus (CFDV) (Rohde et al., 1990) and faba bean necrotic yellows virus (FBNYV) (Katul et al., 1995). Both these components have a similar structure to the components of BBTV and SCSV with one large ORF 3' of a stem-loop sequence. Both these components encode putative Reps. No sequence information is available for milk vetch dwarf virus (MVDV) (Sano et al., 1993).

None of the BBTV-like viruses have been shown to be infectious even though Chu et al.(1993) demonstrated that SCSV can replicate in pea protoplasts. Therefore, it is not known which components constitute the essential genome of these viruses and whether, particularly for BBTV and SCSV, all components have been identified and whether these viruses require more than one Rep to replicate.

Recently, we demonstrated that BBTV DNA-1 is associated with all BBTV isolates tested from ten countries (Karan et al., 1994). This provides some evidence that BBTV DNA-1 is an integral component of the BBTV genome. Interestingly, when the sequences of BBTV DNA-1 from the different isolates were compared, it was evident that there are two distinct groups of isolates, the South Pacific group (isolates from Australia, Burundi, Egypt, Fiji, India, Tonga and Western Samoa) and the Asian group (isolates from the Philippines, Taiwan and Vietnam). Previously, BBTV DNA-2 and 5 had only been identified in Australian and Hawaiin isolates, BBTV DNA-3, 4 and 6 in Australian isolates and the Rep encoding components of Yeh et al. (1994) and Wu et al. (1994) in Taiwanese isolates. In this paper, we report the consistent association of BBTV DNA-2, 3, 4, 5 and 6 with BBTV isolates from eight countries and further evidence for two distinct groups or strains of BBTV isolates.

MATERIALS
AND METHODS

Origin of isolates and extraction of total nucleic acid

The BBTV isolates used in this study were from Australia, Burundi, Fiji, India, Western Samoa, the Philippines, Taiwan and Vietnam and were the same isolates used by Karan et al. (1994). Total nucleic acid was extracted from these samples as described by Karan et al. (1994).

Detection of BBTV components by Southern hybridisation
Total nucleic acid extracts from BBTD isolates (10 or 20 ml) were electrophoresed in 1 % agarose gels and blotted onto Hybond-N+ (Amersham) using positive pressure (Posiblot; Stratagene). Random priming labelling kits (Ready-To-Go; Pharmacia or 'OLK-C, oligo-labelling kit'; Bresatec) were used to prepare ³²P-labelled probes. Membranes were prehybridised in Rapid-hybe (Amersham) for 1 h at 65蚓 . Membranes were then hybridised for 16 h at 65蚓 in Rapid-hybe (Amersham). The membranes were washed twice for 10 min with 2 x SSC, 0.5 % SDS, followed by two 10 min washes in 1 x SSC, 0.1% SDS, two 10 min washes in 0.1 x SSC, 0.1 % SDS all at 65蚓 and a 5 min wash in 0.1 x SSC at room temperature. The membranes were then exposed to Agfa Curix RP1 film at -80蚓 using intensifying screens.

Amplification of BBTV component sequences
Oligonucleotide primers for PCR were derived from the published sequences of BBTV DNA-2, 3, 4, 5 and 6 (Burns et al., 1995) (Table 1). All PCR reactions (50 痞) were heated to 94蚓 for 4 min; then subjected to 30 cycles of 94蚓 for 45 s, 50蚓 for 45 s, and 72蚓 for 1 min; and finally 1 cycle of 72蚓 for 10 min.

Restriction enzyme analysis of PCR products
PCR products in 40 痞 aliquots were purified on columns (Wizard PCR Preps; Promega) and eluted in water (25 痞). Purified PCR product (10 痞) was digested with Rsa I (10 U; Gibco BRL) in final volumes of 15 痞 at 37蚓 for 1.5 h. A 7.5 痞 aliquot of each digest was electrophoresed in a 2 % agarose gel and stained with ethidium bromide. DNA fragment sizes were estimated by comparison with a 傌174 DNA - Hae III digest marker.

Cloning and sequencing of BBTV DNA-6
Full-length PCR products of BBTV DNA-6 were cloned into the T-tailed vector pGEM-T (Promega). Cloned DNA was sequenced using [³⁵S]dATP and a Sequenase kit (US Biochemical) as recommended by the manufacturer. Reaction products were analysed by electrophoresis in a 6% (w/v) polyacrylamide gel containing 7 M urea. Gels were fixed, dried and exposed to Agfa Curix RP1 film. The primers used for sequencing were either universal sequencing primers (US Biochemical) or 17 to 30 nucleotide primers complementary to appropriate regions of the cloned DNA.

Computer analysis
The computer programs used for sequence analysis were accessed from the Australian National Genomic Information Service (ANGIS), University of Sydney. CLUSTAL V (Higgins et al., 1992) was used to align nucleotide and amino acid sequences and to construct Neighbor-Joining trees. The Wisconsin Genetics Computer Group (GCG) package of programs version 7.3 (Devereux et al., 1984) was used to translate nucleotide sequences (TRANSLATE) and to create similarity matrices (DISTANCES). Trees were drawn using the DRAWTREE program (PHYLIP package version 3.5c; Felstenstein, 1993).

RESULTS

A range of available BBTV component specific primers pairs were tested in an attempt to detect BBTV DNA-2 to 6 in BBTV isolates from eight countries. Initially, we attempted to amplify full length copies of each component from each isolate using adjacent, outward extending primer pairs as described by Burns et al. (1994). Where products were not amplified from all isolates tested, other component specific primers were investigated that were designed to amplify smaller products. Further, for two components, no primer pairs were identified that amplified products from all isolates tested. Therefore, component specific DNA probes were generated and used in Southern hybridisations in an attempt to detect these components in the various isolates.

BBTV DNA-1
BBTV DNA-1 has previously been shown to be present in all BBTV isolates from 10 countries using BBTV DNA-1 specific primers and PCR (Karan et al., 1994).

Figure 1

Figure 2

BBTV DNA-2 and 3

A number of component specific primer pairs were tested in attempts to amplify either BBTV DNA-2 or 3. No primer pairs were identified that would amplify all or part of either BBTV DNA-2 or 3 from all isolates. For instance, primers BT80-F1 and BT80-R3 amplified the expected BBTV DNA-2 product from the Australian, Fijian and Western Samoan isolates but not from the Philippines, Taiwanese or Vietnamese isolates and primers BT3V1-EXP and BT3C1-EXP amplified the expected BBTV DNA-3 product from the Australian and Indian isolates but not from the Philippines, Taiwanese or Vietnamese isolates. Therefore, two component specific DNA probes were designed and generated from cloned BBTV DNA-2 and 3 of the Australian isolate (S. Prasad, R. Harding and J. Dale, unpublished results; Burns et al., 1995). The BBTV DNA-2 specific probe corresponded to nucleotides 266-574 and hybridised with cloned BBTV DNA-2 but not with cloned BBTV DNA-1, 3, 4 ,5 and 6 whereas the BBTV DNA-3 specific probe corresponded to nucleotides 213-737 and hybridised with cloned BBTV DNA-3 but not with cloned BBTV DNA-1, 2, 4, 5 and 6. These two probes used separately, hybridised in Southern blots with nucleic acid extracts from all BBTV isolates tested but not with extracts from healthy bananas (Figure 1 and 2). Again, the intensity of hybridisation varied between extracts and the probe hybridised with bands of different sizes.

BBTV DNA-4 and 5

The primer pair BT4V3.17 and BT4C2.17 amplified the expected 350bp product from nucleic acid extracts of all isolates tested but did not amplify a product from healthy banana extracts. The sequences of these two primers were derived from within the major ORF in BBTV DNA-4 (Burns et al., 1995) and had no significant sequence homology with any other DNA component of BBTV. However, to confirm that the amplified products were derived from BBTV DNA-4, they were digested with Rsa I. Sequence analysis of BBTV DNA-4 (Australian isolate; Burns et al., 1995), indicated that Rsa I digestion of the expected PCR products from this component should result in the generation of two fragments of 112 nts and 238 nts. When the PCR products were digested with Rsa I and analysed, the two predicted restriction fragments were observed from the PCR products from Australia, Fiji, India and Western Samoa (Table 2). The 238 nt predicted fragment was present in the Rsa I digest of the Burundi isolate; however the 112 nt product was not detected. Interestingly, PCR products from the Taiwan and Philippines isolates were not digested with Rsa I and the PCR product from the Vietnam isolate was only partially digested to yield a band of approximately 283 nt.

The primer pair BT129V3.17 and BT129C1.17 amplified the expected 355 bp product from nucleic acid extracts of all isolates tested but did not amplify a product from healthy banana extracts. The sequence of these two primers was derived from within the major ORF in BBTV DNA-5 (Burns et al., 1995). The second primer, BT129 C1.17, was located 5' of the stem-loop sequence. Neither primer had significant sequence homology with any other DNA component of BBTV. However, to confirm that the amplified products were derived from BBTV DNA-5, they were digested with Rsa I. Sequence analysis of BBTV DNA-5 (Australian isolate; Burns et al., 1995), indicated that Rsa I digestion of the expected PCR products from this component should result in the generation of two fragments of 129 nt and 226 nt. When the PCR products were digested with Rsa I, the two predicted fragments were observed from the PCR products from Australia, India, Fiji, Western Samoa and Burundi (Table 2). In contrast, the Rsa I digested PCR product from Taiwan, Philippines and Vietnam isolates gave different banding patterns. Fragments of approximately 166 nt and 141 nt were present in the Taiwan and Philippines isolates, with the Taiwan isolate also containing a minor fragment similar in size to the 226 nt band of the South Pacific group. The Vietnam PCR product was not digested with Rsa I.

BBTV DNA-6
Two adjacent, outwardly extending primers, BTP2F1.17 and BTP2R1.17, derived from within the major ORF of BBTV DNA-6 (Burns et al., 1995) were used to amplify an apparent full component length product from nucleic acid extracts of all BBTV isolates tested but not from nucleic acid extracts from healthy bananas. Again, these products were digested with Rsa I and analysed. Sequence analysis of BBTV DNA-6 (Australian isolate; Burns et al., 1995), indicated that Rsa I digestion of the expected PCR products from this component should result in the generation of eight fragments of 9 nt, 45 nt, 59 nt, 114 nt, 126 nt, 136 nt, 181 nt and 419 nt. Sequence analysis of BBTV DNA-6 from Burundi and Fiji indicated that Rsa I digestion of the PCR products from these components would result in the generation of 416 nt (Burundi), 419 nt (Fiji), 181 nt (both), 136 nt (both), 126 nt (both), 114 nt (both), 104 nt (Fiji), 59 nt (Burundi), 45 nt (Burundi) and 9 nt (both) fragments. Sequence analysis of BBTV DNA-6 from Western Samoa indicated that Rsa I digestion of the PCR product from this component would result in the generation of a 420 nt, 316 nt, 126 nt, 123 nt and 104 nt fragments.

The 181 nt and approximately 419 nt fragments were present in the Rsa I digested PCR product from Australia, Burundi, Fiji and India. The Rsa I digested PCR product from Western Samoa contained no 181 nt fragment but, in addition to a 419 nt fragment, a band of approximately 316 nt that was not detected in any other isolate. Multiple fragments ranging from 95 to 136 nt were observed in the Rsa I digest of each PCR product but were not in sufficient concentration to allow a reliable comparison.

The restriction pattern of the Rsa I digested PCR products from the Asian isolates was very different to that of the South Pacific group. The digested PCR products from Taiwan, the Philippines and Vietnam all contained a fragment of approximately 870 nt, which was not found in the digested PCR products of the South Pacific group. Sequence analysis of BBTV DNA-6 from Taiwan and Vietnam indicated that Rsa I digestion of the expected PCR products from these components should result in the generation of 9 nt (Taiwan), 45 nt (both), 59 nt (both), 114 nt (both), 126 nt (Taiwan), 135 nt (Vietnam), 351 nt (Vietnam), 374 nt (Vietnam) and 730 nt (Taiwan) fragments. In some cases the PCR products appeared only partially digested. This would explain the differences between the expected range of Rsa I digest fragments and those obtained and may have been due to (i) inefficient cleavage by the enzyme or (ii) a DNA sub-population that lacked Rsa I restriction sites.

Figure 3

Figure 4

The apparent full component length amplified products from the Burundi, Fiji, Taiwan, Vietnam and Western Samoa isolates were cloned into pGEM-T and sequenced in both directions. These sequences were aligned with BBTV DNA-6 from an Australian isolate (Burns et al., 1995) and compared (Figure 3, Table 3). The sequences varied in length from 1078 nts to 1089 nts but importantly all contained a potential stem-loop structure, the CR-SL, the CR-M, a potential TATA box, one major ORF in the virion sense and one or two potential polyadenylation signals. All these sequences had been previously identified in other BBTV components as well as BBTV DNA-6 (Burns et al., 1995; Harding et al., 1993). When the BBTV DNA-6 sequences from all six isolates were compared, two groups of isolates were evident, the South Pacific group (Australia, Burundi, Fiji and Western Samoa) and the Asian group (Taiwan and Vietnam) (Figure 4). The mean difference over the full component sequence was 3.2% within the South Pacific group, 9.9% within the Asian group and 14.5% between the two groups (Table 3; Figure 4). The major ORF was highly conserved within the South Pacific group both at the nucleotide and amino acid level; however, the difference between the ORF sequences of the two Asian isolates was nearly as great as the difference between the two groups of isolates (Table 3; Figure 4).

Figure 3

The potential stem-loop sequence was highly conserved between all isolates with only three nucleotide differences in the Vietnam isolate and three nucleotide differences in the Western Samoa isolate sequence compared to all other isolates. The potential stem sequence of the Western Samoa isolate was one base pair longer than the other isolates. The invariant nonanucleotide loop sequence was fully conserved between all isolates.

The potential TATA box (CTATTAATA) was also fully conserved in all the BBTV DNA-6 sequences. One potential polyadenylation signal (nt 772 - nt 778) was identical in all isolates. A second potential polyadenylation signal (nt 745 - nt 751) was present and identical in the sequences from Australia, Burundi, Fiji and Western Samoa but not present in the sequences from Taiwan and Vietnam (Figure 3).

The CR-M of BBTV DNA-6 was highly conserved within the South Pacific group of isolates (1.1% mean difference) and was reasonably well conserved between the two isolates in the Asian group (5.8% mean difference). However, the mean difference between the groups was 27.0%. The CR-M sequences of BBTV DNA-1 and BBTV DNA-6 for each isolate were compared. The CR-M sequences (BBTV DNA-1 and 6) from Australia had 75.0 % similarity, from Burundi had 74.4 % similarity, from Fiji had 73.9 % similarity and from Western Samoa had 73.9 % similarity. The mean sequence similarity between the CR-M regions from BBTV DNA-1 and BBTV DNA-6 of the South Pacific group isolates was 74.3 %. The CR-M sequence from Taiwan (BBTV DNA-1 and 6) had 83.1 % similarity and the CR-M sequences from Vietnam BBTV DNA-1 and 6 had 84.6 % similarity. The mean sequence similarity between the CR-M regions from BBTV DNA-1 and BBTV DNA-6 of Asia group isolates was 83.9 %.

Between the ORF amino acid sequence of BBTV DNA-6 from the four South Pacific isolates, there was only one amino acid difference compared to the Australian isolate (amino acid 17, W to C in Fiji). There were 8 amino acid differences between the Taiwan and Vietnam isolates and 14 differences between the South Pacific group and the Asian group.

DISCUSSION

We have previously demonstrated that BBTV DNA-1, which encodes a putative Rep, was associated with all BBTV isolates tested from 10 countries. Interestingly, when the sequences of BBTV DNA-1 from these different isolates were compared, it was evident that there were two groups of isolates, the South Pacific group and the Asian group (Karan et al., 1994). We have now demonstrated using either component specific PCR primers or DNA probes that BBTV DNA-2, 3, 4, 5 and 6 were also present in all isolates tested from seven countries (Table 4) and therefore it is probable that BBTV DNA-1 to 6 are all essential components of the BBTV genome. However, it is not possible to predict whether BBTV requires only the Rep encoded by BBTV DNA-1 or one or more of the additional Reps for replication.

There was obvious variation within the individual component sequences between different isolates. For instance, none of the BBTV DNA-2 or 3 specific primer pairs designed from the Australian isolate amplified those components from the Asian isolates. The restriction fragment length polymorphism patterns generated by the digestion of the BBTV DNA-4, 5 and 6 PCR products confirmed earlier evidence that there were two groups of BBTV isolates. The RFLP patterns of the isolates from Australia, Burundi, Fiji, India and Western Samoa were almost identical (Table 2) and these five isolates had previously been included in the South Pacific group based on the sequence of their BBTV DNA-1 (Karan et al., 1994). However, the RFLP patterns for the three Asian isolates (the Asian group) were more similar to each other than they were to those of the South Pacific group of isolates and within these isolates, the Philippines and Taiwan isolates were more similar to each other than they were to the Vietnam isolate (Table 2). These results confirm the previously determined relationship of these three isolates to each other and the South Pacific group of isolates based on the sequences of their BBTV DNA-1 where the Philippines and Taiwan isolates were more closely related to one another than they were to the Vietnam isolate and all three were more closely related to one another than they were to the South Pacific group of isolates (Karan et al., 1994).

Figure 3

The comparisons of the full sequences of BBTV DNA-6 from six isolates again confirmed the division of BBTV isolates into the two groups (Figure 3, Table 3). There was greater variability within BBTV DNA-6 compared with BBTV DNA-1 between the two Asian isolates and between the two groups whereas within the South Pacific group of isolates, the CR-M and the ORF were more strongly conserved than the equivalent regions within BBTV DNA-1 (Table 3). The CR-M appears to be an important marker for the division of the two groups as it is strongly conserved within groups of isolates both within and between components but there is between 27% and 32% sequence difference between the two groups of isolates.

These results suggest that, while the Asian isolates are clearly more closely related to each other than they are to the South Pacific group of isolates, they have undergone considerably more divergence than the South Pacific group. This phenomenon could reflect either the period of time in which these isolates have been evolving in bananas in the Asian region or that there has been more than one introduction of BBTV into bananas in the south east Asian region. The South Pacific group of isolates probably have a much more recent common progenitor strengthening the probability that the movement of BBTV through the South Pacific, Australia, South Asia and Africa has been recent and derived from a single source.

It is now important to determine whether the other Rep encoding components identified from Taiwanese isolates (Yeh et al., 1994; Wu et al., 1994) are present in all isolates of BBTV and to develop an infectivity assay for BBTV to enable the identification of the essential components of the BBTV genome.

REFERENCES

Boevink P, Chu PWG, Keese P, 1995. Sequence of subterranean clover stunt virus DNA: Affinities with the geminiviruses. Virology 207, 354-361.

Burns TM, Harding RM, Dale JL, 1994. Evidence that banana bunchy top virus has a multiple component genome. Archives of Virology 137, 371-380.

Burns TM, Harding RM, Dale JL, 1995. Genome organisation of banana bunchy top virus. Journal of General Virology 76, 1471-1482.

Chu PWG, Qui B, Li Z, Larkin P, 1993. Replication of subterranean clover stunt virus in pea and subterranean clover protoplasts. Virus Research 27, 173-183.

Dale JL, 1987. Banana bunchy top: an economically important tropical plant virus disease. Advances in Virus Research 33, 301-325.

Deveraux J, Haeberli P, Smithies O, 1984. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Research 12, 387-395.

Felsenstein J, 1993. PHYLIP-Phylogeny Inference Package version 32 Cladistics 5, 164-166.

Hafner G, Harding RM, Dale, JL, 1995. Movement and transmission of banana bunchy top virus DNA component one in bananas. Journal of General Virology 76: 2279-2285.

Harding RM, Burns TM, Dale JL, 1991. Virus-like particles associated with banana bunchy top disease contain small single-stranded DNA. Journal of General Virology 72, 225-230.

Harding RM, Burns TM, Hafner G, Dietzgen RG, Dale JL, 1993. Nucleotide sequence of one component of the banana bunchy top virus genome contains a putative replicase gene. Journal of General Virology 74, 323-328.

Higgins DG, Bleasby A, Fuchs R, 1992. CLUSTAL V: improved software for multiple sequence alignment. CABIOS 8, 189-191.