VIRUSES OF BANANA IN EAST AFRICA
by
Anthony Peter James
Bachelor of Applied Science (Honours)
Master of Plant Protection
Centre for Tropical Crops and Biocommodities
Faculty of Science and Technology
A thesis submitted for the degree of Doctor of Philosophy
Queensland University of Technology
2011
I
ABSTRACT
Bananas are one of the world’s most important food crops, providing sustenance
and income for millions of people in developing countries and supporting large
export industries. Viruses are considered major constraints to banana production,
germplasm multiplication and exchange, and to genetic improvement of banana
through traditional breeding. In Africa, the two most important virus diseases are
bunchy top, caused by Banana bunchy top virus (BBTV), and banana streak disease,
caused by Banana streak virus (BSV). BBTV is a serious production constraint in a
number of countries within/bordering East Africa, such as Burundi, Democratic
Republic of Congo, Malawi, Mozambique, Rwanda and Zambia, but is not present
in Kenya, Tanzania and Uganda. Additionally, epidemics of banana streak disease
are occurring in Kenya and Uganda.
The rapidly growing tissue culture (TC) industry within East Africa, aiming to
provide planting material to banana farmers, has stimulated discussion about the
need for virus indexing to certify planting material as virus‐free. Diagnostic
methods for BBTV and BSV have been reported and, for BBTV, PCR‐based assays
are reliable and relatively straightforward. However for BSV, high levels of
serological and genetic variability and the presence of endogenous virus sequences
within the banana genome complicate diagnosis. Uganda has been shown to
contain the greatest diversity in BSV isolates found anywhere in the world. A
broad‐spectrum diagnostic test for BSV detection, which can discriminate between
endogenous and episomal BSV sequences, is a priority. This PhD project aimed to
establish diagnostic methods for banana viruses, with a particular focus on the
development of novel methods for BSV detection, and to use these diagnostic
methods for the detection and characterisation of banana viruses in East Africa.
A novel rolling‐circle amplification (RCA) method was developed for the
detection of BSV. Using samples of Banana streak MY virus (BSMYV) and Banana
streak OL virus (BSOLV) from Australia, this method was shown to distinguish
between endogenous and episomal BSV sequences in banana plants. The RCA
II
assay was used to screen a collection of 56 banana samples from south‐west
Uganda for BSV. RCA detected at least five distinct BSV isolates in these samples,
including BSOLV and Banana streak GF virus (BSGFV) as well as three BSV isolates
(Banana streak Uganda‐I, ‐L and –M virus) for which only partial sequences had
been previously reported. These latter three BSV had only been detected using
immuno‐capture (IC)‐PCR and thus were possible endogenous sequences. In
addition to its ability to detect BSV, the RCA protocol was also demonstrated to
detect other viruses within the family Caulimoviridae, including Sugar cane
bacilliform virus, and Cauliflower mosaic virus.
Using the novel RCA method, three distinct BSV isolates from both Kenya
and Uganda were identified and characterised. The complete genome of these
isolates was sequenced and annotated. All six isolates were shown to have a
characteristic badnavirus genome organisation with three open reading frames
(ORFs) and the large polyprotein encoded by ORF 3 was shown to contain
conserved amino acid motifs for movement, aspartic protease, reverse
transcriptase and ribonuclease H activities. As well, several sequences important
for expression and replication of the virus genome were identified including the
conserved tRNAmet primer binding site present in the intergenic region of all
badnaviruses. Based on the International Committee on Taxonomy of Viruses
(ICTV) guidelines for species demarcation in the genus Badnavirus, these six
isolates were proposed as distinct species, and named Banana streak UA virus
(BSUAV), Banana streak UI virus (BSUIV), Banana streak UL virus (BSULV), Banana
streak UM virus (BSUMV), Banana streak CA virus (BSCAV) and Banana streak IM
virus (BSIMV). Using PCR with species‐specific primers designed to each isolate, a
genotypically diverse collection of 12 virus‐free banana cultivars were tested for
the presence of endogenous sequences. For five of the BSV no amplification was
observed in any cultivar tested, while for BSIMV, four positive samples were
identified in cultivars with a B‐genome component.
During field visits to Kenya, Tanzania and Uganda, 143 samples were
collected and assayed for BSV. PCR using nine sets of species‐specific primers, and
RCA, were compared for BSV detection. For five BSV species with no known
III
endogenous counterpart (namely BSCAV, BSUAV, BSUIV, BSULV and BSUMV), PCR
was used to detect 30 infections from the 143 samples. Using RCA, 96.4% of these
samples were considered positive, with one additional sample detected using RCA
which was not positive using PCR. For these five BSV, PCR and RCA were both
useful for identifying infected samples, irrespective of the host cultivar genotype
(Musa A‐ or B‐genome components). For four additional BSV with known
endogenous counterparts in the M. balbisiana genome (BSOLV, BSGFV, BSMYV and
BSIMV), PCR was shown to detect 75 infections from the 143 samples. In 30
samples from cultivars with an A‐only genome component there was 96.3%
agreement between PCR positive samples and detection using RCA, again
demonstrating either PCR or RCA are suitable methods for detection. However, in
45 samples from cultivars with some B‐genome component, the level of agreement
between PCR positive samples and RCA positive samples was 70.5%. This suggests
that, in cultivars with some B‐genome component, many infections were detected
using PCR which were the result of amplification of endogenous sequences. In
these latter cases, RCA or another method which discriminates between
endogenous and episomal sequences, such as immuno‐capture PCR, is needed to
diagnose episomal BSV infection.
Field visits were made to Malawi and Rwanda to collect local isolates of
BBTV for validation of a PCR‐based diagnostic assay. The presence of BBTV in
samples of bananas with bunchy top disease was confirmed in 28 out of 39
samples from Malawi and all nine samples collected in Rwanda, using PCR and RCA.
For three isolates, one from Malawi and two from Rwanda, the complete
nucleotide sequences were determined and shown to have a similar genome
organisation to previously published BBTV isolates. The two isolates from Rwanda
had at least 98.1% nucleotide sequence identity between each of the six DNA
components, while the similarity between isolates from Rwanda and Malawi was
between 96.2% and 99.4% depending on the DNA component. At the amino acid
level, similarities in the putative proteins encoded by DNA‐R, ‐S, ‐M, ‐ C and –N
were found to range between 98.8% to 100%. In a phylogenetic analysis, the three
East African isolates clustered together within the South Pacific subgroup of BBTV
IV
isolates. Nucleotide sequence comparison to isolates of BBTV from outside Africa
identified India as the possible origin of East African isolates of BBTV.
Keywords
Banana, virus, BSV, Badnavirus, diagnostic, rolling‐circle amplification, BBTV, PCR,
East Africa
V
PUBLICATIONS
Peer reviewed publications related to this PhD thesis
1. James, A.P., Geijskes, R.J., Dale, J.L., and Harding, R.M. 2011a. Development of
a novel rolling‐circle amplification technique to detect Banana streak virus
which also discriminates between integrated and episomal virus sequences.
Plant Dis. 95:57‐62.
2. James, A.P., Geijskes, R.J., Dale, J.L., and Harding, R.M. 2011b. Molecular
characterisation of six badnavirus species associated with leaf streak disease of
banana in East Africa. Ann. Appl. Biol. 158:346‐353.
VI
TABLE OF CONTENTS
ABSTRACT .............................................................................................................. I
PUBLICATIONS ...................................................................................................... V
TABLE OF CONTENTS ........................................................................................... VI
LIST OF FIGURES .................................................................................................. IX
LIST OF TABLES ..................................................................................................... X
LIST OF ABBREVIATIONS ...................................................................................... XI
STATEMENT OF ORIGINAL AUTHORSHIP ............................................................ XIII
ACKNOWLEDGEMENTS ...................................................................................... XIV
CHAPTER 1 INTRODUCTION ................................................................................. 1
DESCRIPTION OF SCIENTIFIC PROBLEM INVESTIGATED ................................ 1
OVERALL OBJECTIVES OF THE STUDY ............................................................. 2
SPECIFIC AIMS OF THE STUDY ......................................................................... 2
ACCOUNT OF SCIENTIFIC PROGRESS LINKING THE SCIENTIFIC PAPERS ......... 3
CHAPTER 2 LITERATURE REVIEW .......................................................................... 5
2.1 BANANAS – IMPORTANCE AND HISTORY ................................................. 5
2.2 BANANAS IN EAST AFRICA ........................................................................ 7
2.3 VIRUS DISEASES OF BANANAS ................................................................ 11
2.3.1 DNA viruses ............................................................................. 11
2.3.1.1 Streak disease ........................................................................... 11
2.3.1.2 Bunchy top disease ................................................................... 23
2.3.2 RNA viruses .............................................................................. 28
2.3.2.1 Bract mosaic disease................................................................. 28
2.3.2.2 Mosaic disease .......................................................................... 30
2.3.2.3 Mild mosaic disease and BVX ................................................... 33
2.4 VIRUS INDEXING ...................................................................................... 34
2.4.1 Rolling‐circle amplification ..................................................... 36
2.5 TISSUE CULTURE OF BANANA ................................................................. 36
2.6 REFERENCES ............................................................................................ 39
VII
CHAPTER 3 DEVELOPMENT OF A NOVEL ROLLING‐CIRCLE AMPLIFICATION
TECHNIQUE TO DETECT BANANA STREAK VIRUS WHICH ALSO DISCRIMINATES
BETWEEN INTEGRATED AND EPISOMAL VIRUS SEQUENCES ............................... 57
ABSTRACT ...................................................................................................... 59
INTRODUCTION ............................................................................................. 60
MATERIALS AND METHODS .......................................................................... 62
RESULTS ......................................................................................................... 69
DISCUSSION ................................................................................................... 78
ACKNOWLEDGEMENTS ................................................................................. 81
REFERENCES ................................................................................................. 82
CHAPTER 4 MOLECULAR CHARACTERISATION OF SIX BADNAVIRUS SPECIES ASSOCIATED WITH LEAF STREAK DISEASE OF BANANA IN EAST AFRICA ............. 87
ABSTRACT ...................................................................................................... 89
INTRODUCTION ............................................................................................. 90
MATERIALS AND METHODS .......................................................................... 92
RESULTS ......................................................................................................... 98
DISCUSSION ................................................................................................. 104
ACKNOWLEDGEMENTS ............................................................................... 108
REFERENCES ................................................................................................ 110
CHAPTER 5 COMPARISON OF ROLLING‐CIRCLE AMPLIFICATION AND DIRECT‐PCR BASED METHODS FOR DIAGNOSIS OF BSV INFECTION IN EAST AFRICA ............. 115
ABSTRACT .................................................................................................... 118
INTRODUCTION ........................................................................................... 119
MATERIALS AND METHODS ........................................................................ 121
RESULTS ....................................................................................................... 127
DISCUSSION ................................................................................................. 134
ACKNOWLEDGEMENTS ............................................................................... 138
REFERENCES ................................................................................................ 139
VIII
CHAPTER 6 MOLECULAR CHARACTERISATION OF BANANA BUNCHY TOP VIRUS
ISOLATES FROM MALAWI AND RWANDA ........................................................ 147
ABSTRACT .................................................................................................... 149
INTRODUCTION ........................................................................................... 150
MATERIALS AND METHODS ........................................................................ 152
RESULTS....................................................................................................... 156
DISCUSSION ................................................................................................. 161
ACKNOWLEDGEMENTS ............................................................................... 165
REFERENCES ................................................................................................ 166
CHAPTER 7 GENERAL DISCUSSION ................................................................... 169
IX
LIST OF FIGURES
Chapter 2
Fig. 1: Symptoms of streak disease in leaf tissue ...................................................... 12
Fig. 2: Virions of banana streak virus and genome organisation of Badnavirus genus
members ................................................................................................................... 14
Fig. 3: Symptoms of bunchy top disease ................................................................... 24
Fig. 4: Nanovirus genome organisation. ................................................................... 26
Fig. 5: Symptoms of BBrMV in cultivar Cavendish .................................................... 29
Fig. 6: Symptoms of CMV infection in banana leaf tissue ......................................... 31
Fig. 7: Schematic of rolling circle amplification ........................................................ 37
Chapter 3
Fig. 1: Agarose gel analysis of RCA‐amplified DNA ................................................... 71
Fig. 2: Comparison of PCR and RCA for the differential detection of episomal‐ and
eaBSVS ....................................................................................................................... 77
Chapter 4
Fig. 1: Phylogenetic tree using neighbour‐joining method (Kimura 2‐parameter
model with bootstrapping (1000 replicates)) of the RT/RNaseH region of selected
badnaviruses ........................................................................................................... 102
Chapter 5
Fig. 1: Maps of Uganda, Tanzania and Kenya indicating regions surveyed ............ 123
Chapter 6
Fig. 1: Neighbour‐joining phylogram of BBTV DNA‐R sequences based on CLUSTAL
W alignment ............................................................................................................ 162
Chapter 7
Fig. 1: Phylogenetic tree using neighbour‐joining method (Kimura 2‐parameter
model with bootstrapping (1000 replicates)) of partial RT/RNaseH region of
selected badnaviruses ............................................................................................. 174
X
LIST OF TABLES
Chapter 3
TABLE 1 Selected plant samples used in this study ................................................. 64
TABLE 2 Degenerate primer sequences included in the RCA reaction mixes ......... 66
TABLE 3 Predicted restriction profiles of genomic DNA from characterised BSV
species ....................................................................................................................... 67
Chapter 4
TABLE 1 Plant samples for RCA analysis and PCR primers used for detection of
integrated sequences ................................................................................................ 94
TABLE 2 BSV‐indexed leaf samples used for PCR analysis ........................................ 95
TABLE 3 Genome features of East African BSV species .......................................... 100
TABLE 4 Pair‐wise distance matrix using core RT/RNaseH sequences of selected
badnaviruses ........................................................................................................... 105
Chapter 5
TABLE 1 Summary of plant samples collected in East Africa .................................. 122
TABLE 2 PCR primers used for detection of BSV species ........................................ 125
TABLE 3 Predicted restriction profiles of genomic DNA from characterised BSV
using PstI and StuI ................................................................................................... 126
TABLE 4 PCR and RCA detection of BSV species with no confirmed eBSV ............. 129
TABLE 5 PCR and RCA detection of BSV species with confirmed eBSV .................. 133
TABLE 6 Advantages and disadvantages of RCA and PCR for BSV detection ......... 137
TABLE 7 Supplementary data: Plant samples and analysis for BSV using PCR and
RCA .......................................................................................................................... 142
Chapter 6
TABLE 1 Leaf samples collected in Malawi and Rwanda, and results of PCR and RCA
assays for detection of BBTV .................................................................................. 153
TABLE 2 Characteristics of DNA components of BBTV isolates from Malawi and
Rwanda ................................................................................................................... 158
TABLE 3 Sequence similarity between Malawi and Rwanda BBTV DNA components
................................................................................................................................. 160
XI
LIST OF ABBREVIATIONS
°C degrees Celsius
μl microlitre/s
μM micromolar
aa amino acid/s
BLAST basic local alignment search tool
bp base pair/s
CTAB cetyl trimethyl ammonium bromide
DNA deoxyribonucleic acid
dNTP deoxyribonucleoside triphosphate
ds double‐stranded
ELISA enzyme linked immunosorbent assay
g gram/s
g relative centrifugal force in units of gravity
h hour/s
ha hectare
IC immunocapture
kbp kilobase pair/s
kDa kilodalton/s
min minute/s
mM millimolar
mol mole
nm nanometre/s
no. number
nt nucleotide/s
XII
ORF open reading frame
PCR polymerase chain reaction
pH ‐ log (proton concentration)
ρmol picomole/s
RCA rolling‐circle amplification
RNA ribonucleic acid
RNaseH ribonuclease H
RT reverse transcriptase
RT‐PCR reverse transcription PCR
spp. species
s second/s
ss single‐stranded
t tonne/s
Tris tris(hydroxymethyl)aminomethane
UTR untranslated region
V volt/s
XIII
STATEMENT OF ORIGINAL AUTHORSHIP
The work contained in this thesis has not been previously submitted to meet
requirements for an award at this or any other higher education institution. To the
best of my knowledge and belief, the thesis contains no material previously
published or written by another person except where due reference is made.
Signature
Date
XIV
ACKNOWLEDGEMENTS
There are many people I would like to thank. Firstly, I am most grateful to my wife
Lynda for her encouragement, patience, understanding and support. Along with
our two beautiful children Alec and Jenna, your support is what made this possible.
I would like to express my gratitude to my principal supervisor James for
offering me such a great opportunity to work within the CTCB and particularly on
the Bill and Melinda Gates Foundation project based in East Africa. The offer to
travel to and work with scientists in East Africa was too good to refuse and I have
thoroughly enjoyed the challenges of the science, the visits to amazing countries I
have not visited previously and the wonderful people involved in this work at both
QUT and our partner organisations overseas. To Rob, I am grateful for your
unending patience and hard work on the writing front. Although frustratingly
pedantic at times, your efforts have crafted my writing into publication‐worthy
prose. I have found great solace in our discussions over a few quiet ones. To my
buddy Jas, my thanks too, for your help with all problems technical or otherwise
and for keeping my caffeine and sugar levels pumped daily for the duration of all
this I am very grateful.
To my wonderful colleagues in virology at our partner organisations in East
Africa – Charles and Jerome in Uganda, Laura in Kenya, and of course Julius in
Tanzania. I can’t express my thanks enough for your wonderful hospitality during
our visits and for sharing the hard work with us during the project. Every time
some new problem arises I simply recall the phrase ‘hakuna matata’ and laugh out
loud thinking about roadside waits with flat tyres that drive off on moto‐taxis, a
few quiet Nile Specials at the end of a long day in the field (and some noisy ones
later on) and the small successes along the way that got the project finished
(mostly) in the end. I wish all of you great happiness and hope we can continue our
wonderful collaboration!
To my colleagues at QUT in the CTCB – thank you all for your help,
suggestions, comforting words, encouragement, shared frustrations and shared
celebrations. Especially to my good friends who have been down this road – Scotty,
Andrew, Glenn, JY, Suze, and Mumbi – your encouragement has been uplifting and
your patient ears have borne out my frustrations many times, thank you all!
1
CHAPTER 1
INTRODUCTION
This thesis is being presented in the style of ‘Thesis by Publication’. The following
chapters contain a detailed literature review (Chapter 2), followed by four chapters
of results (Chapters 3 to 6) and a general conclusions chapter (Chapter 7). Each of
the results chapters are presented in a style formatted for publication, with the
style of the target journal used for formatting of each individual chapter. For this
reason, consistency in formatting between each of the chapters varies.
Description of scientific problem investigated
In 2005, a collaborative project between QUT and National Agricultural Research
Organisation (NARO), Uganda, funded by the Bill and Melinda Gates Foundation
(BMGF), was initiated with the aim to develop bananas with enhanced levels of
pro‐vitamin A and iron. Bananas (Musa spp.) are an important staple food and cash
crop for people in East Africa, with East African Highland bananas the most
important group of cultivars in this region. However, these bananas are low in pro‐
vitamin A and iron, and hence banana‐based diets are low in these key
micronutrients. The aim of the project, therefore, was to address the chronic
micronutrient malnutrition problem in Uganda through transgenic biofortification
of East African Highland bananas. During the course of this project, it was
recognised that a major hurdle to releasing biofortified bananas in East Africa was
that there was no testing of bananas for viruses done anywhere in East Africa nor
was there a capacity for virus‐indexing. The capacity to index bananas for viruses
within the region is critical for disease control and the provision of disease‐free
planting material to farmers. In addition, knowledge of the viruses present within
East Africa and their distribution is useful for local disease control efforts and
regional quarantine programs. In response to these concerns, a supplementary
2
project was funded by the BMGF to develop banana virus diagnostics, use the
diagnostics to survey Uganda and neighbouring East African countries for viruses
and transfer the indexing technologies to laboratories in East Africa.
The two most important viruses of banana in East Africa are Banana bunchy
top virus (BBTV) and Banana streak virus (BSV). Diagnostic tests for BBTV using PCR
are relatively straight‐forward, reliable, sensitive and cost‐effective. However,
diagnostic methods for BSV are inadequate because of the considerable genetic
and serological diversity amongst BSV isolates and the presence of endogenous
BSV sequences in some banana cultivars that leads to false positives. The
development of a diagnostic method that specifically detects the episomal form of
BSV in bananas, which can also detect the wide sequence variability reported,
would significantly enhance capacity to undertake BSV indexing.
Overall objectives of the study
The overall objective of this study was to establish diagnostic methods for banana
viruses, with a particular focus on the problem of BSV detection, and use these
diagnostic methods for the detection and characterisation of banana viruses in East
Africa.
Specific aims of the study
The specific aims of the project were to (i) develop diagnostic protocols for the
detection of viruses which infect banana with a particular focus on the
development of a new diagnostic method for detection of BSV, (ii) conduct field
surveys in Uganda, Kenya and Tanzania to determine the occurrence of banana
viruses, and (iii) characterise any new viruses detected.
3
Account of scientific progress linking the scientific papers
The first scientific paper (Chapter 3) describes the development of a novel method
for the detection of BSV infection in bananas. Diagnosis of BSV infection in banana
is difficult because there is significant sequence variability reported in BSV isolates
and because integrated sequences lead to false positives when using approaches
based on polymerase chain reaction (PCR). A novel rolling‐circle amplification
(RCA)‐based assay which is sequence non‐specific was developed and shown to
specifically detect the episomal form of BSV DNA and avoid detection of integrated
sequences. Further, the assay was shown to be capable of detecting at least nine
species of BSV in samples from Australia, Uganda and Kenya.
The second scientific paper describes the characterisation of six distinct BSV
species for which only partial sequences were previously reported. Using samples
collected in Uganda and Kenya, RCA was used to amplify the complete genomes of
these viruses, which were subsequently cloned and sequenced. Sequence analyses
supported the proposal of these isolates as new BSV species (designated as BSUAV,
BSUIV, BSULV, BSUMV, BSCAV and BSIMV). The occurrence of integrated
sequences with similarity to each of the six BSV was also investigated using PCR
and species‐specific primers, and these were detected for BSIMV only.
Despite the occurrence of integrated sequences specific to several BSV
species within the banana genome, there are circumstances where PCR could still
be considered a useful method for BSV detection. For example, no integrated
sequences identical to episomal BSV occur in the genome of Musa acuminata, the
progenitor of many important banana cultivars. Additionally, for five species of BSV
characterised in this study PCR was demonstrated to not detect integrated
sequences in virus‐free samples from cultivars with both M. acuminata and M.
balbisiana genetic backgrounds. The third scientific paper reports a comparison of
RCA and PCR for the detection of BSV in field samples from East Africa. A large
collection of samples was obtained during field visits to Kenya, Tanzania and
Uganda as part of an effort to survey for banana viruses. Samples were assayed for
the presence of BSV using PCR with nine sets of BSV species‐specific primers, and
4
by RCA. The results of this screening demonstrated that for five of the
characterised species of BSV which do not have an integrated counterpart (BSCAV,
BSUAV, BSUIV, BSULV and BSUMV), either PCR or RCA are suitable methods to
detect infections. For four additional BSV species known to have integrated
counterparts in the M. balbisiana genome (BSIMV, BSGFV, BSMYV and BSOLV), PCR
and RCA were equivalent methods to detect BSV infection in cultivars with a pure
M. acuminata genetic background. However, in cultivars with some M. balbisiana
genome component, many more samples were positive by PCR compared to RCA,
suggesting detection of integrated BSV sequences as demonstrated in Chapter 3. In
these samples PCR was not considered suitable for episomal BSV detection.
Banana bunchy top disease was first reported in Africa in Egypt in 1901, and
since the 1950s has spread to many countries in sub‐Saharan Africa. Bunchy top is
not present in Uganda, Kenya and Tanzania, but presents the greatest threat from
a virus to banana production in these countries. Although diagnosis of BBTV is
relatively straightforward using PCR, demonstration of the ability for a diagnostic
test to detect local isolates is paramount. Additionally, BBTV has been present in
Africa for more than a century, and only a small amount of sequence information is
available for African isolates. The fourth results chapter focussed on the
identification and characterisation of Banana bunchy top virus (BBTV) in the East
African countries of Rwanda and Malawi. In Chapter 6, BBTV isolates from Malawi
and Rwanda were completely sequenced and characterised. Analyses confirmed
previous reports which suggested that African isolates belonged in the South
Pacific subgroup of BBTV sequences. Samples were provided to laboratories in
Uganda, Kenya and Tanzania to use as positive controls for BBTV indexing.
5
CHAPTER 2
LITERATURE REVIEW
2.1 Bananas – importance and history
Bananas are among the world’s most important food crops. In terms of
gross value of production, bananas are the developing world’s fourth most
important crop after rice, wheat and maize (CGIAR, 2011), and constitute an
important staple nutrition source for millions of people in tropical and subtropical
countries. Bananas are the world’s highest volume export fruit crop as well as
highest value export crop (FAO, 2011). Approximately 100 million tonnes of
bananas are produced annually worldwide, with 15% of production traded on the
world market. The majority of banana fruit produced globally is consumed locally.
In fact, consumption in some countries may be as high as 1 kg/person/day
(Bioversity International, 2007). In tropical countries, bananas may be harvested
year‐round, providing a secure, staple food source high in carbohydrates, vitamin
C, potassium and fibre, and free of fat, cholesterol and sodium (International
Banana Association, 2007).
Most modern cultivated bananas are derived from two wild species, Musa
acuminata and M. balbisiana, contributing the ‘A’ and ‘B’ genomes, respectively.
Bananas can have diploid, triploid or tetraploid genomes, and cultivated bananas
include hybrids within/between the two aforementioned Musa species. Originating
in the Asia‐Pacific region, M. acuminata was domesticated as early as 8000 BC.
With movement of germplasm westwards, hybridisation with M. balbisiana
probably occurred in India where the latter species originates (Price, 1995). The
oldest written records of banana cultivation date from the Indus Valley in
approximately 4000 B.C. (Trager, 1997). Banana germplasm was moved as
populations migrated throughout the Asia‐Pacific region, followed by movement
westward into the Middle East and Africa. The original introductions of bananas
into sub‐Saharan Africa occurred during the first millennium AD (Price, 1995) and
subsequent introductions occurred several times as trade developed with the
6
Middle East and Asia. Many varieties of bananas unique to East Africa probably
arose through mutation and subsequent selective cultivation within the region,
leading the region to be considered a secondary centre of diversity (Davies, 1995).
Portuguese and Spanish traders were instrumental in the dissemination of
banana and plantain germplasm across their trading routes. The discovery of the
Americas prompted the last great stage in banana dissemination, with movement
into the New World, where bananas were considered the most important fruit
within 100 years of their arrival (Price, 1995).
The International Network for the Improvement of Banana and Plantain
(INIBAP) maintain a germplasm resource collection in Leuven, Belgium. The
collection aims to conserve the available genetic variability in banana and plantain
for future use, under the auspices of the Food and Agriculture Organization of the
United Nations (FAO). The international collection currently holds 1,168 accessions
among which 15% are wild relatives and 75% are landraces covering most of the
genetic diversity within the genus Musa (Bioversity International, 2007). A
duplicate of the Asia‐Pacific accessions is maintained in Taiwan, with approximately
500 accessions in the collection there. In recent years, banana germplasm
conservation has also generated considerable interest at a regional level. Collection
expeditions undertaken in centres of origin, including north east India, southern
China, Vietnam, Philippines and Indonesia, have resulted in the establishment of
additional germplasm resources.
Most cultivated bananas are sterile, parthenocarpic triploids, and fall into
the Eumusa division of the genus Musa (family Musaceae). Low levels of female
fertility, the inability to set seed and a narrow genetic base are all constraints on
genetic improvement (Ortiz et al., 1995). Consumer and producer preferences also
complicate progress in releasing improved cultivars. Consumers have a narrow
preference of fruit presentation and taste, while producers require equivalent
agronomic performance and disease resistance to current market standards. These
difficulties explain why only a small group of commercial cultivars for the export
market have remained in production for a relatively long period.
7
Natural landraces are still the predominant cultivated bananas, despite the
efforts of plant breeders over the past 90 years. Some progress in plant
improvement has been made, mostly in the area of disease resistance. Breeders in
the Caribbean and Honduras, for example, have incorporated resistance to Black
Sigatoka (Mycosphaerella fijiensis) and Fusarium wilt (Fusarium oxysporum fsp.
cubense) into cultivated triploid backgrounds, with several agronomically
promising cultivars being produced (Silva et al., 2001). Modern efforts remain
focussed on the development of disease resistance through crossing to wild
relatives, and the selection of acceptable cultivars from wild germplasm
collections, under the guise of two international groups, Promusa and the
International Musa Testing Program (Bioversity International, 2007).
Laboratory based genetic modification offers an opportunity to circumvent
the traditional roadblocks to genetic improvement by conventional breeding.
Characterised genes from banana, or other organisms, can be introduced using
biolistics or Agrobacterium‐mediated transformation processes (Becker et al.,
2000; May et al., 1995; Sagi et al., 1995). Stably transformed plants created using
these processes are attractive alternatives for pest and disease control in resource
poor areas, where farmers cannot afford to purchase chemical controls
.
2.2 Bananas in East Africa
East Africa (EA) is generally considered to include Kenya, Uganda, Tanzania,
Rwanda and Burundi. The eastern region of Democratic Republic of the Congo is
also included when describing the area known as the East African Highlands (EAH).
This region is located along the western border of the Rift Valley and comprises a
system of plateaus, mountain valleys and lakes. Around 20% of the worlds banana
production occurs in the EAH, principally for local consumption and trade (Davies,
1995). Banana is seen as the major food staple in the EAH, where consumption is
as high as 1kg/person/day (Bioversity International, 2007).
8
In contrast to plantation style production in Central and South America and
the Philippines where large volumes of dessert bananas are produced for export,
banana cultivation in EA includes a broad range of cooking, dessert and beverage
(i.e. beer) banana varieties, predominantly grown by small landholders and
subsistence farmers. Personal varietal preference and cultural associations have
driven a highly diverse collection of introduced and locally selected banana
varieties collectively known as East African Highland Bananas (EAHB). These local
varieties are still grown preferentially, although in more recent times there has
been a slight shift towards cultivation of Cavendish type banana varieties.
As a staple food crop in the EAH, fruit production is required year‐round. In
the highland regions production is limited in elevation to 900‐1800m above sea
level, with frosts occurring at higher altitudes. The majority of production is rain‐
fed, with the area receiving reliable rainfall, peaking in March‐May and October‐
November. While soils are moderately fertile, supplementary nutrition is required
by application of household refuse, mulched banana trash and animal manure.
Yields can vary widely between cultivars (often dependent on cultivation practices)
and peaks around 4‐5 months after the high rainfall periods for each local region.
Average banana yields are 6‐17 t/ha‐1 (Lufafa et al., 2002). Major constraints on
production historically include the banana weevil, Cosmopolites sordidus, present
in all EAH growing regions, several species of nematodes and a number of diseases,
the most important of which include Fusarium wilt, Xanthomonas wilt, bunchy top,
leaf steak (viral) and the black and yellow leaf streak (Sigatoka) diseases (Davies,
1995).
Uganda is East Africa’s largest banana producer, with annual production
estimated at over 10 million tonnes, comprised mostly of EAH cooking (70%) and
beer (20%) cultivars, with the remaining 10% divided between dessert bananas and
plantains (Lescot and Ganry, 2010; van Asten, et al., 2010). EAH cooking bananas
are considered the most important staple crop in the EA great lakes region
(Bagamba et al., 2010). Historically, bananas have been produced for home
consumption under a low‐input system and have been seen as a woman’s‐crop
produced for household food self sufficiency. Changing socio‐economic factors
9
over recent decades, related to declining income from traditional cash crops such
as coffee and cotton and increased urbanisation of the population in Uganda, have
seen an increased demand for banana production, resulting in a concomitant
increase in areas planted to banana, particularly in the south west of the country
(Bagamba et al., 2010).
In Kenya, bananas account for 7.4% (approx. 74 000 ha) of the gross
cropped area and 55% of the total area under fruit production, with production
approximately 1.2 million tonnes per year (Lescot and Ganry, 2010). In western
Kenya, banana contributes 40‐50% of household income (Mwangi et al., 2010).
Historically production has been constrained by pests and diseases (particularly
banana weevils, nematodes and Fusarium wilt), non‐availability of disease‐free
planting materials and poor agronomic practices (Njuguna et al., 2010). Production
is mainly smallholder farming, with 80% of banana producers having an average
production area of 0.12 ha, and only 2% of farms >1.98 ha.
In Tanzania, annual production of banana is approximately 750 000 tonnes
(Lescot and Ganry, 2010) produced on about 300 000 ha divided amongst five
banana producing areas, with only maize and cassava produced in greater
quantities. Utilisation is mainly local with approximately 60% cooked/fresh
consumption, 30% brewing and the remainder divided amongst other products
(wine/dried fruit/flour) (Mgenzi et al., 2010).
There are few reports of viruses affecting banana production in East Africa.
Banana bunchy top virus (BBTV) has been reported from Burundi and Rwanda, but
not further east (Dale, 1994, Thomas et al., 1994). However, bunchy top is now
widespread across the equatorial region of the African continent and is present in a
number of countries bordering EA to the west and south including Mozambique,
Malawi, Zambia and the Democratic Republic of the Congo (Kumar and Hanna,
2008) and presents the greatest risk of the virus diseases to banana production in
the region. Streak disease is widespread in Uganda (Harper et al., 2002, 2004 &
2005), and is also present in Kenya (Karanja et al., 2008). It is likely that streak
disease is present in the other countries in EA although its occurrence is
10
unconfirmed at present. Cucumber mosaic virus (CMV) has been recorded in
banana in Kenya (Anne Wangai, personal communication), although not from the
other countries, however the widespread distribution of CMV in other hosts means
that CMV is likely to be present in most banana growing areas. Bract mosaic, mild
mosaic and Banana virus X have not been reported from EA.
Of the banana diseases caused by viruses, only streak disease, caused by
Banana streak virus (BSV), has been researched to a reasonable extent in EA.
Kubiriba et al. (2001a) studied the distribution of streak in infected fields and
showed that distinct clusters of diseased plants were found amongst healthy
plants, and that a slow moving vector was likely associated with plant‐to‐plant
transmission. Transmission studies showed that both Dysmicoccus brevipes
(pineapple mealybug) and Saccharicoccus sacchari (sugarcane mealybug) could
vector the causal virus (one of a number of BSV species which infect banana) in a
semi‐persistent manner, while the banana aphid (Pentalonia nigronervosa) did not
(Kubiriba et al., 2001b). Harper et al. (2002) used PCR and ELISA to study field
samples of banana showing streak‐like symptoms from 14 regions in Uganda. Fifty
one of 59 samples collected tested positive for badnavirus infection, and PCR with
degenerate primers proved a better detection method than ELISA. These authors
later demonstrated approximately 16 different BSV species associated with streak
disease in Uganda (Harper et al., 2005). In Kenya, Karanja et al. (2008) identified
four species of BSV from field samples collected in central/western districts.
11
2.3 Virus diseases of bananas
2.3.1 DNA viruses
2.3.1.1 Streak disease
Introduction
Banana streak disease was first reported in Morocco in 1974 (Lassoundiere,
1974, in Lockhart, 1986) but an earlier report from the Ivory Coast, circa 1958,
probably describing the symptoms of streak disease as mosaic, was noted by
Lockhart and Jones (2000b). The disease has since been reported from numerous
countries in Africa, South America and the Pacific, and probably occurs worldwide
where bananas are grown (Lockhart and Jones, 2000b; Lockhart and Olszewski,
1993). Symptoms of streak disease vary widely and are influenced by the cultivar,
virus species and environmental conditions. The most common symptoms include
narrow, discontinuous (sometimes continuous) chlorotic and/or necrotic streaks
which run parallel to the veins of the leaf lamina (Lockhart and Jones, 2000b) (Fig.
1). The leaf streaks become darker as the leaves age, eventually turning dark
brown‐black. Interestingly, symptoms can regress with periods of symptomless
growth occurring before the disease re‐emerges at later stages of growth, with
temperature considered a factor in disease expression (Dahal et al., 2000). Other
symptoms associated with streak disease include stunting, choking (constriction of
the bunch on emergence), cigar‐leaf necrosis, internal necrosis of the pseudostem,
and leaf/fruit distortion (Lockhart and Jones, 2000b).
Banana streak disease can cause reduced plant growth and vigour and a reduction
in bunch size and yield (Lockhart and Jones, 2000b). In Australia, Cavendish cultivar
‘Williams’ showed a 7% reduction in bunch weight under disease conditions, and a
delay in harvest interval compared with healthy plants (Daniells et al., 2001).
Additionally, a 5% reduction in fruit length was observed in diseased plants. In a
separate study conducted in Nigeria, Dahal et al. (2000) reported losses of up to
15% in some banana cultivars with banana streak.
12
Fig. 1. Symptoms of streak disease in leaf tissue. A) chlorotic flecking and B)
chlorotic and necrotic flecks.
A)
B)
13
Causal agent and Taxonomy
Banana streak disease is caused by a collection of BSV species (genus
Badnavirus, family Caulimoviridae). Badnaviruses are characterised by non‐
enveloped bacilliform particles of 130‐150 nm x 30 nm (Fig. 2) and have a circular,
non‐covalently closed, double‐stranded DNA genome of 7.4 kbp (Lockhart, 1990;
Hull et al., 2005). All badnaviruses typically contain three open reading frames
(ORFs) on the virus‐sense coding strand (Fig. 2). ORF 1 encodes a small protein of
unknown function which has been associated with virions (Cheng et al., 1996). ORF
2 encodes a protein of 14 kDa with a non‐specific DNA‐ and RNA‐binding activity
(Jacquot et al., 1996). This protein may function in virion assembly due to the
presence of a predicted N‐terminal coiled‐coil domain which supports self‐
interaction to form a tetramer (Leclerc et al., 1998). The large polyprotein encoded
by ORF 3 contains domains associated with movement, the virus capsid, aspartic
protease, reverse transcriptase (RT) and ribonuclease H (RNaseH) functions.
Additional conserved genomic features include the presence of a short sequence
(16‐18 nucleotides) complementary to plant transfer‐RNA methionine (which binds
to the RNA‐intermediate and primes plus‐strand DNA synthesis), several
discontinuities in the DNA genome which are involved in replication, and
regulatory sequences associated with expression of the genome (as‐1, TATA and
PolyA signals) (Hull, 2002).
Based on sequence information, a number of distinct BSV sequences have
been reported from banana. Originally isolates were referred to as Banana streak
virus (BSV, Lockhart, 1986) with isolates differentiated based on their location or
their host cultivar. For example, Banana streak virus–Onne reported by Harper and
Hull (1998) was originally collected at the International Institute for Tropical
Agriculture (IITA) Onne field station in Nigeria. The sequence reported in this paper
later came to be known as Banana streak virus strain OL (BSV‐OL), after a sequence
identical to the episomal form was characterised from
14
Fig. 2. (A) Electron micrograph showing virions of Banana streak virus and
(B) typical genome organisation of Badnavirus genus members showing
the three open reading frames (ORFs) and functional domains of ORF3.
A)
B)
15
the nuclear genome of banana cultivar Obino l’Ewai (Geering et al., 2001) and was
nearly identical to an isolate from Australia named BSV‐RD, or ‘Red Dacca’ after the
host cultivar studied (Geering et al., 2000). Geering et al. (2000) also reported
several other sequences named BSV‐Cav (for ‘Cavendish’), BSV‐Mys (for ‘Mysore’)
and BSV‐GF (for ‘Gold Finger’). The BSV‐Mys isolate was later completely
characterised and renamed Banana streak Mysore virus (BSMysV; Geering et al.,
2005b).
Around this time two important studies were undertaken with regards to
badnavirus infection in bananas. In the first, a large field collection of bananas from
Uganda was demonstrated, using a PCR‐based strategy, to contain widely diverse
BSV sequences and thirteen new isolates were sequentially designated Banana
streak Uganda A‐M virus respectively (BSUgA‐MV; Harper et al., 2004 & 2005).
Furthermore, Geering et al. (2005a) reported 36 groups of badnavirus‐like
sequences present in the genome of selected Musa accessions. Evidence was
presented that both BSMysV and the previously described BSV‐GF had integrated
counterparts, as well as another partially sequenced isolate from Australia named
BSV‐Imové (BSImV), named again for its host cultivar (Gayral et al., 2008; Geering
et al., 2005a & b; Geering et al., 2011). Endogenous virus sequences were
described as ‘Banana endogenous virus’ (BEV) with a numerical assignment (1‐33),
with the episomal counterpart taking precedence where identified (eg. BSMysV,
BSImV and BSV‐OL) (Geering et al., 2005a). More complication in naming has arisen
with the publication of an additional isolate named Banana streak Acuminata
Vietnam virus (BSAcVnV) (Lheureux et al., 2007) and an as‐yet unpublished
badnavirus isolate which is completely sequenced, named Banana streak Yunnan
virus (GenBank accession no. DQ092436).
Classification of Badnavirus species is based on a measure of sequence
similarity between isolates, with nucleotide sequences which differ by more than
20% within the conserved reverse transcriptase‐ribonuclease H (RT/RNaseH)
coding region of ORF 3 considered distinct species (Hull et al., 2005). For naming
purposes however, the historical account can often be confusing, where no
established convention exists when a plant species plays host to many similar
16
viruses and this is further complicated where endogenous counterparts exist, some
of which are infectious and others not. A resolution has recently been
implemented by the International Committee for Taxonomy of Viruses (ICTV)
Caulimoviridae study group with a two letter abbreviated naming system
implemented and the retention of historical host/symptom descriptors (Geering,
2010). As such, the isolates previously recognised as Obino l’Ewai, Mysore, Gold
Finger and Acuminata Vietnam (the only banana‐infecting badnavirus species
which are officially recognised by the ICTV) are presently known as Banana streak
OL virus (BSOLV), Banana streak MY virus (BSMYV), Banana streak GF virus (BSGFV)
and Banana streak VN virus (BSVNV), respectively. This recently established
naming convention will be used throughout the remaining sections for clarity,
although throughout the results chapters which follow the usage has evolved to
this form over time. Where the BSV species is not explicitly stated in the literature,
the term ‘BSV’ has been retained.
Integrated virus sequences have been reported in a number of plant
species, including banana, rice, petunia, potato, tomato and Nicotiana spp (see
references in Geering et al., 2010 and Staginus et al., 2009). For purposes of
classification, these sequences are usually identified with episomal counterparts if
they exist or, if not, with the most similar sequence previously reported. This
allows a taxonomic placement of these sequences whether they are determined to
be putatively functional (based on the integrity of putative ORFs) or not functional
due to the absence of ORFs or a mutation in putative ORFs. However, several
authors have proposed naming conventions for the distinction of episomal forms
from integrated forms (both functional and non‐functional).
Where an episomal counterpart has previously been reported the use of
the prefix ‘e’ (endogenous) has been proposed (Staginus et al., 2009; Geering et al.,
2010). In cases where the integrated sequence is functional and can be expressed
to give rise to episomal infection the prefex ‘ea’ (endogenous activatable) is
suggested (Staginus et al., 2009). Where no episomal form is recognised, Staginus
et al. (2009) proposed that the naming system comprise the host plant initials plus
the suffix ‘EPRS’ (endogenous pararetroviral sequence), for example ‘SotuEPRS’ for
17
Solanum tuberosum endogenous pararetroviral sequence. Geering et al. (2010)
disagree with the proposal to include the term pararetrovirus as a taxonomic
descriptor as this is not a recognised taxon name. Additionally, these authors
dispute the differentiation applied to functional and non‐functional sequences, as a
taxonomic framework must cope with both living and dead organisms. As a
functional taxonomic framework exists at present (under the ICTV system),
integrated virus sequences with sufficient information should be treated within
this framework (Geering et al., 2010). Furthermore, these authors also
recommend against the use of the suffix ‘a’ or ‘d’ for the designation of
‘activatable’ or ‘dead’ as proposed by Staginus et al. (2009) as many factors
contribute to the expression of such integrated sequences, and this information
can be handled in genus/species descriptions without inclusion in a naming
convention.
Characterisation
Full‐length sequencing of a Nigerian BSV isolate, BSOLV, confirmed earlier
reports based on partial DNA sequences, that BSD was caused by a Badnavirus
(Harper and Hull, 1998). BSOLV was most similar to Sugar cane bacilliform MO virus
(SCBMOV) (51.6%) at the nucleotide level, however at the amino acid level BSOLV
was most similar to Commelina yellow mottle virus (CoYMV). In Australia, four
isolates of BSV were cloned and sequenced from four distinct cultivars (Geering et
al., 2000). One isolate (BSV‐RD) was identical to the BSOLV sequence previously
reported by Harper and Hull (1998). While the four isolates from Australia were
more similar to each other than other reported badnaviruses, sequence similarity
was only 66.4 to 78.2%, low enough to be considered distinct species. Two of these
isolates, BSGFV and BSMYV, were later completely sequenced (GenBank accession
no. AY493509; Geering et al., 2005b). Harper et al. (2004) reported that variability
of BSV sequences in Ugandan banana was extremely high based on DNA
sequencing, and later classified 13 new species (Harper et al., 2005). However this
study only reported partial sequences based on PCR‐amplification of a small region
18
of ORF 3 (encoding a part of the RT/RNaseH domain) and no further evidence has
been presented to confirm the episomal origin of these amplicons. Three further
distinct full‐length BSV sequences, namely BSVNV (Lheureux et al., 2007), Banana
streak Yunnan virus (unpublished) and Banana streak IM virus (previously BSImV)
have also recently been characterised (Geering et al., 2011).
Integrated sequences and their expression
Since initial reports describing the molecular characterisation of BSOLV, and
the subsequent use of PCR for detection of the viral DNA, researchers have
suggested that sequences homologous to the BSV genome are present in the
genome of bananas (Harper and Hull, 1998). This was supported by studies
showing that viral DNA hybridized to genomic DNA of banana and in‐situ
hybridization studies which revealed two distinct integrants of BSOLV in banana
cultivar Obino l’Ewai (Harper et al., 1999b; Ndowora et al., 1999). The observation
that tissue culture plantlets of improved hybrids expressed high levels of banana
streak disease, despite parent plants being apparently virus free, prompted
researchers to investigate whether infection could arise from integrated
sequences. Evidence was presented to support this hypothesis, with the virus
isolates obtained from tissue‐cultured plants displaying genetic uniformity
(Ndowora et al., 1999), in stark contrast to isolates arising from field infections
which are generally genetically different (based on studies using RFLPs). This
suggested for the first time that viral sequences integrated into a plant genome
could become activated to produce virus infection (Harper et al., 1999b; Ndowora
et al., 1999). Characterisation of integrated BSOLV sequences showed that the viral
sequences were flanked by direct repeats, suggesting a model for activation by
homologous recombination. Additionally, BSOLV was shown to be integrated
within a range of Musa genotypes, but linked to the B‐genome of Musa spp.
(Geering et al., 2001).
Geering et al. (2005a) characterised badnavirus sequences integrated into a
range of Musa spp. and identified 36 distinct groups of sequences from 103 PCR
19
clones, with less than 85% similarity between sequence groups. Based on the
detection of many groups of integrated badnavirus sequences in the genomes of
the two progenitor species of edible bananas Musa acuminata and M. balbisiana,
with distinct sequence groups detected in each species, the authors concluded that
badnaviruses had integrated into the Musa genome independently many times,
and that the integration probably occurred after speciation of M. acuminata and
M. balbisiana. Interestingly, many of these integrated sequences were not
conceptually translatable due to mutations within the sequences. Several of these
sequence groups were identical to episomal badnavirus sequences previously
identified from banana and many were similar to putative episomal sequences
reported in Ugandan bananas (Gayral and Iskra‐Caruana, 2009). Since this time,
studies have confirmed the presence of BSMYV in the B‐genome of banana
(Geering et al., 2005b) and the presence of BSGFV sequences which are integrated
and can be expressed to give rise to episomal infection (Gayral et al., 2008). To
date, no activatable, or ‘live’ sequences have been characterised from the A‐
genome.
Dallot et al. (2001) clearly demonstrated that activation of integrated
sequences occurs in the synthetic (i.e. bred) cultivar FHIA‐21 during tissue culture
multiplication. Activation was sporadic and was greatest during the cellular
proliferation stage of culture (compared to the rooting or acclimatisation stages),
with increased activation seen with increased cycles of multiplication. Côte et al.
(2010) later demonstrated activation of integrated BSOLV from both natural and
synthetic Musa hybrids, while activation of integrated BSGFV only occurred in the
synthetic hybrid studied. Meyer et al. (2008) were able to transmit episomal
BSOLV, originating from integrated sequences, using several species of mealybug.
Diagnostic testing
An antiserum was first prepared for detection of ‘BSV’ by Lockhart (1986).
Initial studies in ‘BSV’ detection using serological methods were encouraging,
however a high level of serological heterogeneity was later revealed when
20
screening a wider range of isolates (Lockhart and Olszewski, 1993). One approach
to help overcome the large serological variability was to produce an antiserum
from a mixture of isolates. An antiserum raised against 30 isolates of ‘BSV’ and a
serologically related badnavirus which naturally infected sugarcane and could
infect banana experimentally (historically named Sugar cane bacilliform virus
(ScBV)) was used with some success to detect a range of ‘BSV’ isolates. One major
concern with reliability of this approach is that availability of the antisera will
fluctuate over time and there is some difficulty in reproducing the range of
detection when subsequent antisera are prepared. As such, there is a greater
attraction in developing a diagnostic assay based on the detection of the viral
nucleic acid.
The high level of serological variability observed in ‘BSV’ isolates is reflected
in the high levels of diversity seen in the viral DNA. DNA probes were generated
that would detect individual isolates of ‘BSV’, but no cross‐hybridisation was
observed amongst probes and differing virus isolates (Lockhart and Olszewski,
1993). This high level of genetic variability is characteristic of badnaviruses and
other members of the Caulimoviridae and is due to their replication strategy. All
Caulimoviridae members are pararetroviruses which replicate by producing an RNA
intermediate from the dsDNA genome. This RNA molecule is used as template for
translation of viral proteins, as well as template for the production of genomic DNA
by a virus‐encoded reverse transcriptase (RT) (Harper and Hull, 1998). The viral
encoded RT lacks proof‐reading function and so errors in replication lead to high
levels of genetic diversity.
Development of PCR‐based diagnostic tests for several badnaviruses,
including ScBV, Taro bacilliform virus (TaBV) and Pineapple bacilliform virus (PBV),
have targeted conserved regions of the genome in ORF 3, where the RT/RNaseH
domains show the highest levels of sequence conservation (Braithwaite et al.,
1995; Thomson et al., 1996; Yang et al., 2003). The PCR primers reported by Yang
et al. (2003) have been used for the successful detection of a range of BSV isolates
(Cherie Gambley, personal communication).
21
Harper and Hull (1998) developed a PCR assay for BSOLV based on the
sequence of ORF 3. Screening of banana accessions in this study revealed an
unusually high level of infection compared to results with ELISA and symptom
expression, and it was suggested that integration of the virus occurs (discussed
previously). Geering et al. (2000) also reported several species‐specific primer sets
for the detection of different BSV characterised from banana in Australia, including
BSGFV, BSV‐Cav and BSMYV. However, these primers are not suitable for detection
of BSMYV and BSGFV from banana cultivars with a B‐genome component due to
the presence of eBSMYV and eBSGFV.
Integrated badnavirus sequences pose a serious challenge to diagnostic
testing in Musa spp. An assay that will detect episomal (non‐integrated) virus
sequences may also detect integrated sequences due to similarity in the sequences
used to design PCR primers. To avoid detecting integrated sequences, diagnostic
assays need to exclude host plant DNA from detection. To allow detection of
episomal virus DNA only, immuno‐capture (IC)‐PCR is used (Geering et al., 2000;
Harper et al., 1999a; Le Provost et al., 2006). A broad spectrum antiserum is used
to trap virus particles from crude plant extracts in PCR tubes. Washing removes
unbound materials including genomic DNA and the viral particles are heat
denatured, allowing PCR on DNA within the trapped virus particles only. Several
sets of PCR primers, including degenerate and specific, have been reported for
detection of BSV in banana using this procedure (Dallot et al., 2001; Harper et al.,
2002; Le Provost et al., 2006).
Unfortunately, IC‐PCR is complicated by the non‐specific binding of host
plant DNA to the capture vessel (Iskra‐Caruana et al., 2009; Le Provost et al., 2006),
so the method was improved further by the addition of PCR primers for host
genome DNA in a multiplex IC‐PCR format (Le Provost et al., 2006). IC‐PCR was
shown to be significantly more sensitive for BSV detection than immuno‐sorbent
electron microscopy (ISEM) or triple‐antibody sandwich ELISA (TAS‐ELISA)
(Agindotan et al., 2006). Unfortunately, this assay format is still confounded by the
high serological variability and serum supply problems associated with ELISA.
22
Epidemiology of BSV
Transmission of BSV occurs both vegetatively (in suckers taken from
infected mother plants and through tissue‐culture propagation of infected
explants) and by several mealybug (Hemiptera: Coccoidea: Pseudococcidae)
species, but not mechanically (Lockhart and Jones, 2000b). Mealybug species which
have been confirmed as vectors include the citrus mealybug, Planococcus citri
Russo (Lockhart and Olszewski, 1993), Pseudococcus comstiki Kuwana (Su, 1998),
both the pineapple mealybug Dysmicoccus brevipes Cockerell and the sugarcane
mealybug Sacchirococcus sacchari Cockerell (Kubiriba et al., 2001b), and the vine
mealybug Planococcus ficus Signoret (Meyer et al., 2008). Additionally, isolates of
‘ScBV’ can be transmitted from sugarcane to banana by the sugarcane mealybug
(Lockhart and Olszewski, 1993).
The in‐field spread of BSV by vectors is relatively slow with most
transmission occurring vegetatively via the movement of infected planting material
during propagation. As supply of planting material increasingly involves tissue
culture, there is an opportunity to spread ‘BSV’ (and the other banana viruses
discussed) by propagating diseased mother plants. Tissue culture has been shown
to naturally eliminate up to 52% of ‘BSV’ infections, and this can be increased to
90% using cryopreservation and chemical treatments (Helliot et al., 2002; Helliot et
al., 2003). These methods may be desirable to recover elite planting material from
infection, but virus free source plants are essential for the supply of clean planting
material. In the case of several BSV species there are complications with the
development of an effective assay for source plant screening that will not falsely
detect integrated forms of the virus. In addition, plantlets will need to be assayed
following tissue culture to ensure that virus infection has not resulted from the
activation of integrated sequences in the banana genome, particularly in the case
of those cultivars containing the B genome.
23
2.3.1.2 Bunchy top disease
Introduction
Banana bunchy top disease is the most important viral disease of banana
(Dale, 1987). Bunchy top was first recorded in Fiji in 1879 and has since spread to a
number of countries in the south Pacific, Asia and Africa (Magee, 1927; Thomas
and Iskra‐Caruana, 2000; Geering 2009a). Symptoms of bunchy top include a
narrow, upright appearance of the plant apex and dark green streaks on the
petioles, midribs and leaf veins (Fig. 3). Infected plants produce no fruit, or a
reduced bunch with no market value. Bunchy top is caused by Banana bunchy top
virus (BBTV) (Harding et al., 1991; Thomas and Dietzgen, 1991). The virus is
characterised by isometric virions of 18‐20 nm (Fig. 3) with a ssDNA genome
consisting of at least six components (Burns et al., 1995). Plant to plant
transmission is by the black banana aphid, Pentalonia nigronervosa (Magee, 1927)
(Fig. 3) in a persistent manner, although the virus does not replicate in the vector
(Hafner et al., 1995). Vegetative transmission occurs through rhizomes, suckers
and tissue cultured plants and is the major factor for long distance movement of
the virus (Thomas et al., 1994).
Causal agent and Taxonomy
BBTV is the type member of the genus Babuvirus in the family Nanoviridae.
Viruses in this group contain at least six circular, single‐stranded (ss) DNA
molecules all encoding at least one protein (Gronenborn, 2004; Vetten et al., 2005)
(Fig. 4). Each DNA component has two common regions; a common stem‐loop
region (CR‐SL) of 69 nucleotides that is 62% conserved between viral components
and a major common region (CR‐M) located 5’ to the CR‐SL, of 92 nucleotides, with
76% conservation between components (Burns et al., 1995). Beetham et al. (1997)
showed that BBTV DNA‐R encodes two ORFs and showed that both ORFs are
24
Fig 3. Symptoms of bunchy top disease including (A) dark green streaking
on petiole; B) yellowing of the margins of leaf lamina; and C) upright
growth habit and stunting symptoms. Virions of BBTV are shown in (D);
while the aphid vector of BBTV is shown in (E).
A
C)
B
D
E)
25
transcribed. The major ORF of BBTV DNA‐R (labelled DNA‐1 in Fig. 4) encodes the
viral replication initiation protein, also known as the master Rep (Harding et al.,
1993; Horser et al., 2001), while DNA‐S, ‐M, ‐C and ‐N (labelled DNA‐3, ‐4, ‐X1 and ‐
5 in Fig. 4) encode the viral coat protein (Wanitchakorn et al., 1997), intracellular
movement protein, retinoblastoma like binding protein (Clink) and nuclear shuttle
protein (Wanitchakorn et al., 2000a), respectively. No function has been ascribed
to DNA‐U3 at present, or the small internal gene of DNA‐R. Additional Rep‐
encoding components, named satellite DNAs, are also known to occur in BBTV‐
infected bananas. These satellite DNAs also encode Rep proteins, but in contrast to
the master Rep protein which can initiate replication of all other genomic
component, these Reps are only capable of self‐replication (Gronenborn, 2004).
Interestingly, satellite DNAs are only associated with Asian isolates of BBTV
(discussed below) (Bell et al., 2002; Horser et al., 2001).
Studies of geographic isolates of BBTV have identified two distinct
subgroups based on sequence analysis of DNA‐R (Karan et al., 1994), ‐S
(Wanitchakorn et al., 2000b) and ‐N (Karan et al., 1997). The ‘South Pacific’
subgroup includes isolates from Australia, Burundi, Egypt, Fiji, India, Tonga and
Western Samoa, while the ‘Asian’ subgroup includes isolates from the Philippines,
Taiwan and Vietnam. Sequence variation of DNA‐R within isolates of each
subgroup ranged from 1.9% (South Pacific) to 3.0% (Asian), while variation
between isolates of the two subgroups was as high as 10% (Karan et al., 1994).
Greater variability was also observed in sequences of DNA‐N in Asian subgroup
members (9.9%) compared to South‐Pacific subgroup members (3.2%), with
sequences of this DNA component showing greater variability overall (14.5%)
compared with DNA‐R sequences (Karan et al., 1997). Using coat protein gene
sequences (DNA‐S ORF), up to 13% nucleotide variability was observed between
isolates of the two subgroups, with sequence variability within the three
26
Fig. 4. Nanovirus genome organisation. Individual DNAs are grouped
according to encoded proteins. M‐Rep: master Rep protein; CP: capsid
protein; Clink: cell cycle link protein; MP: movement protein. X1‐X4:
proteins of unknown function; grouping is according to sequence
similarity. Genome component designations used for each virus are given
below each DNA. A dash (‐) behind a virus name denotes that a
corresponding DNA has not been identified from this virus. Arrows
represent open reading frames; (><): inverted repeat sequence at the
replication origins (ori); (*): TATA‐boxes; (~): polyadenylation signals
(figure from Gronenborn, 2004).
27
Asian subgroup isolates again greater than the South Pacific subgroup isolates
studied (Wanitchakorn et al., 2000b).
Further isolates have been sequenced since these reports, although in
many cases the sequences of any or all of the viral DNAs are not complete. In spite
of this, additional South Pacific subgroup isolates have been reported from
Cameroon (Oben et al., 2009), Hawaii (Xie and Hu, 1995), Pakistan (Amin et al.,
2008) and Angola (Kumar et al., 2008) while Asian subgroup isolates have been
reported from Indonesia (Furuya et al., 2004) and Japan (Furuya et al., 2005).
Additional unpublished sequences present in GenBank include Myanmar (South
Pacific subgroup) and China (Asian subgroup).
Diagnostic testing
Geering and Thomas (1996) compared four serological tests for the
detection of BBTV. A triple antibody sandwich ELISA using BBTV monoclonal
antibodies was shown to be the most effective of the methods trialled when
labour, cost and sensitivity were considered. PCR‐based methods have
subsequently been developed for detection of BBTV. Using primers targeting DNA‐
R, BBTV could be detected in virus infected banana leaves, tissue cultured plantlets
and viruliferous aphids (Shamloul et al., 1999). Similarly, primers to DNA‐R were
combined with primers to a housekeeping gene to detect BBTV isolates from major
banana growing regions of Pakistan (Mansoor et al., 2005). DNA‐R appears to
represent a suitable target for PCR diagnosis and sequence information for a
number of isolates is now available in GenBank allowing primers to be designed
which can identify virus isolates from both molecular subgroups. Sequence
information for other BBTV DNAs are also available in public databases, making it
feasible to develop PCR‐based diagnostics for all of the virus‐specific DNAs.
28
Epidemiology
Infected planting material and aphids are both important means of in‐field
spread of bunchy top (Magee, 1927). Epidemics can be attributed to several
combining factors, which include the presence of large numbers of aphid vectors,
planting of diseased vegetative materials providing an inoculum source and the
failure of farmers to identify and remove diseased plants in a timely fashion. In
Australia, bunchy top has been controlled through strict programs of quarantine
restrictions on plant movement, inspection and removal of infected plants and
resupply of virus‐free planting material. Where these measures are able to be
implemented and enforced, successful control of bunchy top can be accomplished,
although eradication is not usually considered feasible.
Thomas et al. (1995) showed that BBTV could be transmitted through tissue
culture and reported that virus‐infected and healthy plants in tissue culture were
difficult to distinguish by visual means. However, transmission was inconsistent
and in some cases BBTV was not detected from tissue culture banana plants for up
to 16 months of monitoring. Similarly, Drew et al. (1992) demonstrated that 27% of
explants of cultivar Cavendish were freed from virus symptoms following tissue
culture. These studies suggest that tissue culture could be a useful means to rescue
BBTV‐infected plants of high value, although a prolonged period of virus indexing
and observation is required to confirm that explants are virus‐free.
2.3.2 RNA viruses
2.3.2.1 Bract mosaic disease
Banana bract mosaic disease is characterised by streaking on the bracts of
the banana inflorescence, spindle‐shaped streaks on the petiole and mottling on
the pseudostem of infected plants (Thomas et al., 2000) (Fig. 5). Mosaic patterns
on the bracts are diagnostic for this disease. Losses in the field as high as 40% have
been reported (Ploetz, 1994). The causal agent is Banana bract mosaic virus
(BBrMV) (Bateson and Dale, 1995; Thomas et al., 1997), which has been reported
from the Philippines, India, Sri Lanka, Thailand, Vietnam and Samoa (Rodoni et al.,
29
Fig. 5. Symptoms of BBrMV in cultivar Cavendish. (A) Light
white mosaic in the lamina of the youngest emerging leaf;
(B) & (C) typical dark red streaks in the bracts of the
inflorescence (marked by arrows); and (D) electron
micrograph of purified BBrMV (from Iskra‐Caruana et al.,
2008).
D
C
B
A
30
1997 & 1999; Sharman et al., 2000a; Thomas et al., 1997) where the virus naturally
infects both banana and abaca (M. textilis). BBrMV is transmitted vegetatively,
mechanically from banana to banana, and is vectored by several aphid species,
including P. nigronervosa (Thomas et al., 2000).
Isolates of BBrMV show low sequence variability in the coat protein‐coding
and 3’ untranslated (UTR) regions, with levels of 0.3‐5.6% and 0.3‐4.3% at the
nucleotide and amino acid level, respectively (Rodoni et al., 1999). The limited
distribution, combined with vegetative transmission, makes this virus an important
quarantine concern for banana germplasm exchange. ELISA, RT‐PCR and IC‐RTPCR
have been used for detection of BBrMV (Iskra‐Caruana et al., 2008; Rodoni et al.,
1999; Sharman et al., 2000b) with RT‐PCR the most sensitive method.
A multiplex IC‐PCR method that detects BBTV, CMV and BBrMV has also
been developed (Sharman et al., 2000b). This assay was used to confirm the PCR
results reported by Rodoni et al. (1999) although in the multiplex format a second
reverse primer was required to detect all six BBrMV isolates tested. However, the
multiplex RT‐PCR format was plagued with difficulties regarding primer selection
and cross reaction to the other viruses of interest, making this a difficult format to
work with.
2.3.2.2 Mosaic disease
Mosaic disease in banana is caused by strains of Cucumber mosaic virus
(CMV) and is characterised by symptoms including infectious chlorosis, mosaic and
heart rot. First reported as a pathogen of banana in NSW in 1929, CMV is now
recognised in most banana growing regions of the world (Lockhart and Jones,
2000a; Niblett et al., 1994). Symptom expression is strain dependent and relies
also on host cultivar and growing conditions. Mosaic and chlorotic streaking along
the veins of the leaves (Fig. 6), leading to necrosis, are common. Fruit may show
disease symptoms and bunches may be bear malformed fruit or no fruit. In very
severe cases plants may die.
31
Fig. 6. Symptoms of CMV infection on banana leaf tissue.
32
Although common, mosaic does not have a major impact on banana
production and simple cultural practices are usually sufficient to control the
disease. These include selection of virus free planting material, avoidance of
intercropping with susceptible alternative hosts (especially cucurbits) and removal
of weeds that act as a reservoir for CMV. The roguing of infected banana plants is
practiced although little evidence is available to suggest that CMV can spread from
banana to banana by aphid transmission. Most infections are thought to arise
through short probes from various aphid species while moving from alternate
hosts into banana (Jeger et al., 1995). CMV transmission can also occur through
meristem tissue culture, although in one study approximately 68% of plants
regenerated from infected mother plants were found to be virus‐free (Surga‐Rivas,
1988).
CMV has a host range of over 1000 plant species, the largest of any plant
virus, and is vectored by at least 60 aphid species including the banana aphid. Virus
isolates are grouped into two major subgroups (named 1 and 2) based on
differences in serological and molecular characters (Eiras et al., 2004).
Approximately 75% similarity exists between the genome of subgroup 1 and
subgroup 2 members. CMV is the type member of the genus Cucumovirus (family
Bromoviridae), has 29 nm icosahedral particles and a genome consisting of three
segments of linear positive sense ssRNA with a total length of 8621 nucleotides
(Fauquet et al., 2005). The genome encodes five proteins involved in particle
formation, replication, post‐transcriptional gene silencing, aphid transmission, and
cell‐to‐cell and long‐distance movement. The virus is present in high levels in
infected plants making diagnostic testing relatively straightforward. A number of
diagnostic methods have been reported for detection of CMV in banana including
DNA probes (Kiranmai et al., 1998), ELISA and RT‐PCR (Hu et al., 1995; Singh et al.,
1995). Strains from various hosts have been detected using a single primer set.
33
2.3.2.3 Mild mosaic disease and BVX
Banana mild mosaic virus (BanMMV) and Banana virus X (BVX) are two
recently characterised banana viruses belonging to the Flexiviridae family.
BanMMV was first reported in 2001 (Gambley and Thomas, 2001) following the
discovery of flexuous virus particles in banana accessions that failed to react to
general potyvirus antisera or potyvirus degenerate PCR primers. BanMMV has a
single‐stranded RNA genome of 7352 nucleotides with five ORFs and is most similar
to fovea‐ and carla‐viruses, based on phylogenetic analysis (Gambley and Thomas,
2001).
A study on the diversity of BanMMV in a worldwide germplasm collection
identified 68 infected banana accessions. Analysis of 154 sequences of the viral
RNA dependent RNA polymerase (RdRp) from these plants showed a mean pair‐
wise nucleotide sequence divergence of 20.4%, and that mixed infections occur
naturally, suggesting a vector is able to transmit the virus from plant to plant
(Teycheney et al., 2005a). During this study the authors identified a novel nucleic
acid sequence in bananas the sequence of which differed significantly from
BanMMV. Sequence analysis of a 2917 nucleotides region of the 3’ end of the viral
RNA revealed a new virus species (Teycheney et al., 2005b). The virus was named
‘Banana virus X’ and was shown to be only distantly related to other members in
the Flexiviridae.
The Flexiviridae is a relatively recently created virus taxon (Fauquet et al.,
2005). All members have flexuous particles, a monopartite positive sense ssRNA
genome with a 3’ polyA tail and up to six ORFs. Eight genera are currently identified
based on differences in genome organisation. Distinct species have less than 72%
identical nucleotides or 80% identical amino acids between their entire coat
protein or replication protein genes. A number of unclassified members are
included in the family, including BanMMV, and Teycheney et al. (2005b) proposed
that BVX is an additional member of this family.
BanMMV causes mild chlorotic mosaic and streak symptoms on highly
susceptible cultivars, but otherwise infection is asymptomatic. BVX has no ascribed
34
symptoms and has only been detected in a handful of individual plants. Both
viruses are vegetatively transmitted and may be problematic to banana germplasm
transfer and potentially in banana micropropagation. RT‐PCR, using a method
called polyvalent degenerate oligonucleotide RT‐PCR (PDO‐RT‐PCR), was recently
reported for detection of both viruses (Teycheney et al., 2007).
2.4 Virus Indexing
Historically, little thought was probably given to selection of disease free
germplasm for movement into new areas. Before the modern science of plant
pathology was established in the mid 19th century, very little was understood about
the cause of plant disease. Because of this it was commonplace to move pests and
diseases of plants when germplasm was transferred, particularly in the case of
vegetatively propagated crops, like bananas. In modern times, the exchange of
banana germplasm is commonplace in order to facilitate efforts in plant breeding
and research activities. Guidelines for the movement of banana germplasm are in
place to ensure that only disease free plants are exchanged (Diekman and Putter,
1996). Plants selected for transfer should be free from visible signs of pests and
diseases, and preferably obtained from disease indexed banana germplasm
collections. Tissue cultured plants should be used where possible, and these should
be virus indexed at an approved virus indexing centre (VIC).
Virus indexing is the process of establishing the presence of a known virus
in a plant, particularly in the absence of disease symptoms. The key parameters for
choice of a diagnostic technique are specificity, sensitivity, cost and speed.
Historically, the predominant methods used for virus indexing were electron
microscopy and host plant assays, both time‐consuming methods that were not
suitable for large scale indexing programs. In regards to cost, sensitivity and
throughput, ELISA is the most historically important development in plant virus
indexing to date (Lopez et al., 2003). ELISA is a highly routine format to assay for
plant viruses and commercial kits for a number of important viruses are widely
available. More recently, PCR has been applied to detection of plant viruses. While
35
the sensitivity and speed are major improvements over ELISA, traditionally PCR has
been more expensive. In addition, the method suffers from a lack of robustness in
some instances, with carryover of product leading to false positives and inhibitors
in plant extracts leading to false negatives. Still, PCR is the most widely used
molecular method for virus indexing and protocols for many important plant
viruses have been reported (Lopez et al., 2003 & 2009).
PCR requires some prior knowledge of the virus genomic sequence. From
this information, primers (oligonucleotides of around 18‐24 bp) are designed to
specifically amplify the genome of the virus of interest. In the case of viruses with
an RNA genome, an initial reverse transcription step, using the enzyme reverse
transcriptase, is necessary prior to the PCR. With the available sequence data in
online databases, it is possible to design primers with several levels of detection.
Variable regions in the genome can be used to design species/strain specific assays,
while conserved regions can be targeted to detect groups of viruses (Shaad, et al.,
2003). Degenerate PCR primers allow groups of viruses to be detected, or to detect
new viruses based on sequence information for related, previously characterised
viruses (Foissac et al., 2005; Yang et al., 2003).
The use of ‘housekeeping’ primers, targeting an endogenous plant gene co‐
extracted with the target, allows confidence in results where no virus is detected
(Iskandar et al., 2004; Kim et al., 2003; Nicot et al., 2005). Housekeeping assays
may be in a multiplex format with the virus specific assay (as an ‘internal’ control),
or a separate test on the same extract. If the housekeeping assay fails, then nucleic
acid extraction from the plant needs to be repeated. Major savings in the cost of
PCR are made by multiplexing several virus assays in one reaction, with up to nine
viruses able to be detected in a single assay (Gambino and Gribaudo, 2006). This
method can, however, be complicated by competition between primers and non‐
target interactions of primer and template (Sharman et al., 2000b).
Major advances in PCR were achieved with the development of real‐time
(rt) PCR. By removing the need for gel electrophoresis and staining, results were
obtained faster and there was no requirement for the extremely toxic DNA stain,
36
ethidium bromide. Additionally, rtPCR allows a quantitative measure of the
amount of the target virus to be made. Although such information is not essential
for indexing, where a yes/no result will suffice, it may provide useful information in
other research areas. While a number of different formats of rtPCR are available
(reviewed in Zhang and Fang, 2006) they involve greater setup costs and time in
development than conventional PCR formats, and have higher ongoing costs.
2.4.1 Rolling‐circle amplification
Rolling‐circle amplification (RCA) (Fig. 7), using bacteriophage Phi29 DNA
polymerase, is a sequence‐independent protocol which has been used for the
amplification and characterisation of circular DNA molecules, including plasmids
(Dean et al., 2001; Reagin et al., 2003) and several groups of DNA‐viruses infecting
humans, animals and plants (reviewed in Johne et al., 2009). To date, the
application of RCA technology to plant‐infecting viruses has been limited to the
small, single‐stranded DNA genomes of viruses in the families Geminiviridae and
Nanoviridae. RCA has been used for the detection and cloning of full‐length
genomes of geminiviruses (Haible et al., 2006; Inoue‐Nagata et al., 2004; Shepherd
et al., 2008) and RCA products have been used as a convenient way to develop
infectious clones of both nanoviruses and geminiviruses (Grigoras et al., 2009;
Knierim and Maiss, 2007; Wu et al., 2008).
2.5 Tissue culture of banana
Tissue culture (TC) plantlets are the internationally accepted form for the
movement of banana germplasm (Diekmann and Putter, 1996). Additionally,
commercial scale banana plantings are now routinely established in many
countries using TC plants to ensure planting material is disease free. Healthy field
plants are selected and used for the in‐vitro propagation of clones, with indexing to
confirm virus freedom in areas where they are known to occur. South Africa’s Du
Roi laboratory produces over six million virus indexed TC banana plants annually,
37
Fig. 7. Schematic of rolling circle amplification. Random hexamer primers anneal to
the circular template DNA at multiple sites. Phi29 DNA polymerase extends each of
these primers. When the DNA polymerase reaches a downstream extended primer,
strand displacement synthesis occurs. The displaced strand is rendered single‐
stranded and available to be primed by more hexamer primer. The process
continues, resulting in exponential, isothermal amplification.
38
exporting 70% of these to growers in Africa, the Middle East, and South and
Central America (Du Roi, 2007).
In East Africa, a number of laboratories have been established to provide TC
banana plants for commercial planting including Kenya Agricultural Research
Institute (KARI) and Jomo Kenyatta University of Agriculture and Technology
(JKUAT), both in Kenya. JKUAT can produce approximately one million TC banana
plants per annum. A number of other laboratories have also been established in
Kenya and Uganda for production of TC banana, although at present all produce
bananas without indexing for the presence of viruses (Florence Wambugu,
personal communication). The consequences of this are the high potential for virus
infected TC banana plants to be disseminated to growers.
39
2.6 References
Agindotan, B., Winter, S., Lesemann, D., Uwaifo, A., Mignouna, J., Hughes, J. and
Thottappilly, G. (2006). Diversity of banana streak‐inducing viruses in Nigeria
and Ghana: Twice as many sources detected by immunoelectron microscopy
(IEM) than by TAS‐ELISA or IC‐PCR. Afr. J. Biotechnol. 5:1194–1203.
Amin, I., Qazi, J., Mansoor, S., Ilyas, M. and Briddon, R.W. (2008). Molecular
characterisation of Banana bunchy top virus (BBTV) from Pakistan. Virus
Genes 36:191–198.
Bagamba, F., Burger, K. and Tushemereirwe, W.K. (2010). Banana (Musa spp.)
production characteristics and performance in Uganda. In Proc. IC on Banana
& Plantain in Africa pp187–198. Edited by T. Dubois, S. Hauser, C. Staver, D.
Coyne. Acta Hort. 879, ISHS.
Bateson, M.F. and Dale, J.L. (1995). Banana bract mosaic virus: Characterisation
using potyvirus specific degenerate primers. Arch. Virol. 140: 515–527.
Becker, D.K., Dugdale, B., Smith, M.K., Harding, R.M. and Dale, J.L. (2000). Genetic
transformation of Cavendish banana (Musa spp. AAA group) cv ‘Grand Nain’
via microprojectile bombardment. Plant Cell Rep. 19:229–234.
Beetham, P. R., Hafner, G. J., Harding, R. M. and Dale, J. L. (1997). Two mRNAs are
transcribed from banana bunchy top virus DNA‐1. J. Gen. Virol. 78:229–236.
Bell, K.E., Dale, J.L., Ha, C.V., Vu, M.T. and Revill, P.A. (2002). Characterisation of
Rep‐encoding components associated with banana bunchy top nanovirus in
Vietnam. Arch. Virol. 147:695–707.
Bioversity International (2007). Why bananas matter. Accessed 18/04/2007.
http://bananas.bioversityinternational.org/.
Braithwaite, K.S., Egeskov, N.M. and Smith, G.R. (1995). Detection of sugarcane
bacilliform virus using the polymerase chain reaction. Plant Dis. 79:792–796.
40
Burns, T. M., Harding, R.M. and Dale, J.L. (1995). The genome organization of
banana bunchy top virus: analysis of six ssDNA components. J. Gen. Virol.
76:1471–1482.
CGIAR. (2011). Research and Impact: Areas of research ‐ Banana. Accessed
22/09/2011. www.cgiar.org/impact/research/banana.html.
Cheng, C.P., Lockhart, B.E.L. and Olszewski, N.E. (1996). The ORFI and II proteins of
Commelina yellow mottle virus are virion‐associated. J. Gen. Virol. 223:263–
271.
Côte, F.X., Galzi, S., Folliot, M., Lamagnère, Y., Teycheney, P‐Y. and Iskra‐Caruana, M‐
L. (2010). Micropropagation by tissue culture triggers differential expression of
infectious endogenous banana streak virus sequences (eBSV) present in the B
genome of natural and synthetic interspecific banana plantains. Mol. Plant
Pathol. 11:137–144.
Dahal, G., Ortiz, R., Tenkouano, A., Hughes, J. d’A., Thottappilly, G., Vuylsteke, D. and
Lockhart, B.E.L. (2000). Relationship between natural occurrence of banana
streak badnavirus and symptom expression, relative concentration of viral
antigen, and yield characteristics of some micropropagated Musa spp. Plant
Pathol. 49:68–79.
Dale, J.L. (1987). Banana bunchy top: An economically important tropical plant virus
disease. Adv. Virus Res. 33:301–325.
Dale, J.L. (1994). Banana bunchy top. In Compendium of tropical fruit diseases
pp17–18. Edited by R.C. Ploetz, G.A. Zentmyer, W.T. Nishijima, K.G. Rohrbach
and H.D. Ohr. APS.
Dallot, S., Acuna, P., Rivera, C., Ramirez, P., Côte, F., Lockhart, B.E.L. and Caruana,
M.L. (2001). Evidence that the proliferation stage of micropropagation
procedure is determinant in the expression of Banana streak virus integrated
into the genome of FHIA 21 hybrid (Musa AAAB). Arch. Virol. 146:2179–2190.
41
Daniells, J.W., Geering, A.D.W., Bryde, N.J. and Thomas, J.E. (2001). The effect of
Banana streak virus on the growth and yield of dessert bananas in tropical
Australia. Ann. Appl. Biol. 139:51–60.
Davies, G. (1995). Banana and plantain in the East African highlands. In Banana and
Plantains pp493–508. Edited by S. Gowen. Chapman & Hall.
Dean, F.B., Nelson, J.R., Giesler, T.L. and Lasken, R.S. (2001). Rapid amplification of
plasmid and phage DNA using Phi29 DNA polymerase and multiply‐primed
rolling‐circle amplification. Genome Res. 11:1095–1099.
Diekmann, M. and Putter, C.A.J. (1996). FAO/IPGRI Technical Guidelines for the Safe
Movement of Germplasm. No. 15. Musa. 2nd edition. FAO/IPGRI, Rome.
Drew, R. A., Smith, M. K. and Andersen, D. W. (1992). Field evaluation of
micropropagated bananas derived from plants containing banana bunchy top
virus. Plant Cell, Tiss. Organ Cult. 28:203–205.
Du Roi (2007). Du Roi Laboratory Home Page. Accessed 25/04/2007.
http://www.duroilab.co.za/.
Eiras, M., Boari, A.J., Colariccio, A., Chaves, A.L.R., Briones, M.R.S., Figueira, A.R. and
Harakava. R. (2004). Characterization of isolates of the Cucumovirus
Cucumber mosaic virus present in Brazil. J. of Plant Pathology 86:61–69.
FAO. 2011. The World Banana Forum (WBF): Working together for sustainable
banana production and trade. Introductory note. Accessed 23/02/2011
www.fao.org/economic/worldbananaforum.
Fauquet, C.M., Mayo, M.A., Maniloff, J., Desselberger, U. and Ball, L.A. (2005). Virus
Taxonomy: Eighth Report of the International Committee on Taxonomy of
Viruses. Elseiver Academic Press.
Foissac, X., Svanella‐Dumas, L., Gentit, P., Dulucq, M‐J., Marais, A. and Candresse, T.
(2005). Polyvalent degenerate oligonucleotide reverse transcription
42
polymerase chain reaction: a polyvalent detection and characterization tool
for trichoviruses, capilloviruses and foveaviruses. Phytopathology 95:617–625.
Furuya, N., Kawano, S. and Natsuaki, K.T. (2005). Characterisation and genetic status
of banana bunchy top virus isolated from Okinawa, Japan. J. Gen. Plant Pathol.
71:68–73.
Furuya, N., Somowiyarjo, S. and Natsuaki, K.T. (2004). Virus detection from local
banana cultivars and the first molecular characterisation of banana bunchy
top virus in Indonesia. Jour. Agri. Sci. Tokyo Univ. of Agric. 49:75–81.
Gambino, G. and Gribaudo, I. (2006). Simultaneous detection of nine grapevine
viruses by multiplex reverse transcription‐polymerase chain reaction with
coamplification of a plant RNA as internal control. Phytopathology 96:1223–
1229.
Gambley, C. F. and Thomas, J. E. (2001). Molecular characterisation of Banana mild
mosaic virus, a new filamentous virus in Musa spp. Arch. Virol. 146:1369–
1379.
Gayral, P. and Iskra‐Caruana, M.L. (2009). Phylogeny of Banana streak virus reveals
recent and repetitive endogenization in the genome of its banana host (Musa
sp.). J. Mol. Evol. 69:65–80.
Gayral, P., Noa‐Carrazana, J‐C., Lescot, M., Lheureux, F., Lockhart, B.E.L.,
Matsumoto, T., Piffanelli, P. and Iskra‐Caruana, M‐L. (2008). A single Banana
streak virus integration event in the banana genome as the origin of infectious
endogenous pararetrovirus. J. Virol. 82:6697–6710.
Geering, A.D.W. (2009a). Viral Pathogens of Banana: Outstanding Questions and
Options for Control. In Proc. IS on Banana Crop Prot., Sust. Prod. & Impr.
Livelihoods pp39‐50. Edited by D. Jones and I. Van den Bergh. ISHS.
Geering, A.D.W. (2009b). The distribution of badnavirus‐infected Musa germplasm
from the international transit centre, Katholieke Universiteit Leuven, Belgium.
43
In Proc. IS on Banana Crop Prot., Sust. Prod. & Impr. Livelihoods pp85‐88.
Edited by D. Jones and I. Van den Bergh. ISHS.
Geering, A.D.W. (2010). Five new species and four nomenclature changes in the
genus Badnavirus. [online] ICTV Taxonomy Proposals ‐ Plant. Available at
http://talk.ictvonline.org/files/proposals/taxonomy_proposals_plant1/m/plan
t01/2216.asp [Accessed 14 December 2010].
Geering, A.W.D and Thomas, J.E. (1996). A comparison of four serological tests for
the detection of banana bunchy top virus in banana. Aust. J. Agric. Res.
47:403–412.
Geering, A.D.W., McMichael, L.A., Dietzgen, R.G. and Thomas, J.E. (2000). Genetic
diversity among Banana streak virus isolates from Australia. Phytopathology
90:921–927.
Geering, A.D.W., Olszewski, N.E., Dahal, G., Thomas, J.E. and Lockhart, B.E.L. (2001).
Analysis of the distribution and structure of integrated banana streak virus
DNA in a range of Musa cultivars. Mol. Plant Pathol. 2:207–213.
Geering, A.D.W., Olszewski, N.E., Harper, G., Lockhart, B.E., Hull, R. and Thomas, J.E.
(2005a). Banana contains a diverse array of endogenous badnaviruses. J. Gen.
Virol. 86:511–520.
Geering, A.D.W., Parry, J.N. and Thomas, J.E. (2011). Complete genome sequence of
a novel badnavirus, banana streak IM virus. Arch. Virol. 156:733–737.
Geering, A.D.W., Pooggin, M., Olszewski, N.E., Lockhart, B.E.L. and Hohn, T. (2005b).
Characterization of Banana streak mysore virus and evidence that its DNA is
integrated in the B genome of cultivated Musa. Arch. Virol. 150:787–796.
Geering, A.W.D., Scharaschkin, T. and Teycheney, P.Y. (2010). The classification and
nomenclature of endogenous viruses of the family Caulimoviridae. Arch. Virol.
155:123–131.
44
Grigoras, I., Timchenko, T., Katul, L., Grande‐Pérez, A., Vetten, H‐J. and Gronenborn,
B. (2009). Reconstitution of authentic nanovirus from multiple cloned DNAs. J.
Virol. 83:10778–10787.
Gronenborn, B. (2004). Nanoviruses: genome organisation and protein function.
Vet. Microbiol. 98:103–109.
Hafner, G.J., Harding, R.M. and Dale, J.L. (1995). Movement and transmission of
banana bunchy top virus DNA component one in bananas. J. Gen. Virol.
76:2279–2285.
Haible, D., Kober, S. and Jeske, H. (2006). Rolling‐circle amplification revolutionizes
diagnosis and genomics of geminiviruses. J. Virol. Methods 135:9–16.
Harding, R.M., Burns, T.M. and Dale, J.L. (1991). Virus‐like particles associated with
banana bunchy top disease contain small single‐stranded DNA. J. Gen. Virol.
72:225–230.
Harding, R.M., Burns, T.M., Hafner, G., Dietzgen, R.G. and Dale, J.L. (1993).
Nucleotide sequence of one component of the banana bunchy top virus
genome contains a putative replicase gene. J. Gen. Virol. 74:323–328.
Harper, G., Ganesh, D., Thottappilly, G. and Hull, R. (1999a). Detection of episomal
Banana streak virus by IC‐PCR. J. Virol. Methods 79:1–8.
Harper, G. and Hull, R. (1998). Cloning and sequence analysis of banana streak virus
DNA. Virus Genes 17:271–278.
Harper, G., Hart, D., Moult, S. and Hull, R. (2002). Detection of Banana streak virus in
field samples of banana from Uganda. Ann. Appl. Biol. 141:247–257.
Harper, G., Hart, D., Moult, S. and Hull, R. (2004). Banana streak virus is very diverse
in Uganda. Virus Res. 100:51–56.
45
Harper, G., Hart, D., Moult, S., Hull, R., Geering, A. and Thomas, J. (2005). The
diversity of Banana streak virus isolates in Uganda. Arch. Virol. 150:2407–
2420.
Harper, G., Osuji, J., Heslop‐Harrison, J.S. and Hull, R. (1999b). Integration of banana
streak badnavirus into the Musa genome: molecular and cytogenetic
evidence. Virology, 255:207–213.
Helliot, B., Panis, B., Frison, E., De Clercq, E., Swennen, R., Lepoivre, P. and Neytsd, J.
(2003). The acyclic nucleoside phosphonate analogues, adefovir, tenofovir and
PMEDAP, efficiently eliminate banana streak virus from banana (Musa spp.).
Antiviral Res. 59:121–126.
Helliot, B., Panis, B., Poumay, Y., Swennen, R., Lepoivre, P. and Frison, E. (2002).
Cryopreservation for the elimination of cucumber mosaic and Banana streak
viruses from Banana (Musa spp.). Plant Cell Rep. 20:1117–1122.
Horser, C.L., Harding, R.M. and Dale, J.L. (2001). Banana bunchy top nanovirus DNA‐
1 encodes the “master” replication initiation protein. J. Gen. Virol. 82:459–
464.
Hu, J.S., Li, H.P., Barry, K., Wang, M. and Jordan, R. (1995). Comparison of dot blot,
ELISA, and RTPCR assays for detection of two cucumber mosaic virus isolates
infecting banana in Hawaii. Plant Dis. 79:902–906.
Hull, R. (2002). Plant Virology 4th Ed. Academic Press.
Hull, R., Geering, A., Harper, G., Lockhart, B.E. and Schoelz, J.E. (2005). Family
Caulimoviridae. In Virus Taxonomy: Eighth Report of the International
Committee on Taxonomy of Viruses pp.385–396. Edited by C.M. Fauquet, M.A.
Mayo, J. Maniloff, U. Desselberger and L.A. Ball. Elsevier.
Inoue‐Nagata, A.K., Albuquerque, L.C., Rocha, W.B. and Nagata, T. (2004). A simple
method for cloning the complete begomovirus genome using the
bacteriophage φ29 DNA polymerase. J. Virol. Methods 116:209–211.
46
International Banana Association (2007). Banana facts: nutrition. Accessed
18/04/2007. http://www.eatmorebananas.com/facts/nutrition.htm.
Iskandar, H.M., Simpson, R.S., Casu, R.E., Bonnett, G.D., Maclean, D.J. and Manners,
J.M. (2004). Comparison of reference genes for quantitative real‐time
polymerase chain reaction analysis of gene expression in sugarcane. Plant
Mol. Biol. Reptr. 22:325–337.
Iskra‐Caruana, M.L., Galzi, S. and Laboureau, N. (2008). A reliable IC one‐step RT‐PCR
method for the detection of BBrMV to ensure safe exchange of Musa
germplasm. J. Virol. Methods 153:223–231.
Iskra‐Caruana, M.L., Gayral, P., Galzi, S. and Laboureau, N. (2009). How to control
and prevent the spread of banana streak disease when the origin could be
viral sequences integrated in the banana genome? In Proc. IS on Banana Crop
Prot., Sust. Prod. & Impr. Livelihoods pp77–84. Edited by D. Jones and I. Van
den Bergh. ISHS.
Jacquot, E., Hagen, L.S., Jacquemond, M. and Yot, P. (1996). The open reading frame
2 product of Cacao swollen shoot badnavirus is a nucleic acid‐binding protein.
Virology 225:191–195.
Jeger, M.J., Eden‐Green, S., Thresh, J.M., Johanson, A., Waller, J.M. and Brown, A.E.
(1995). Banana diseases. In Banana and Plantains pp317–381. Edited by S.
Gowen. Chapman & Hall.
Johne, R., Müller, H., Rector, A., van Ranst, M. and Stevens, H. (2009). Rolling‐circle
amplification of viral DNA genomes using phi29 polymerase. Trends Microbiol.
17:205–211.
Karan, M., Harding, R.M. and Dale, J.L. (1994). Evidence for two groups of banana
bunchy top virus isolates. J. Gen. Virol. 75:3541‐3546.
47
Karan, M., Harding, R.M. and Dale, J.L. (1997). Association of banana bunchy top
virus DNA components 2 to 6 with bunchy top disease. Molecular Plant
Pathology On‐line: [http://www.bspp.org.uk/mppol/] 1997/0624karan].
Karanja, L., Wangai, A., Harper, G. and Pathak, R.S. (2008). Molecular identification
of Banana streak virus isolates in Kenya. J. Phytopath. 156:678–686.
Kim, B.R., Nam, H.Y., Kim, S.U., Kim, S.I. and Chang, Y.J. (2003). Normalization of
reverse transcription quantitative‐PCR with housekeeping genes in rice.
Biotechnol. Lett. 25:1869–1872.
Kiranmai, G., Satyanaranaya, T. and Sreenivasulu, P. (1998). Molecular cloning and
detection of cucumber mosaic cucumovirus causing infectious chlorosis
disease of banana using DNA probes. Curr. Sci. 74:356–359.
Knierim, D and Maiss, E. (2007). Application of Phi29 DNA polymerase in
identification and full‐length clone inoculation of tomato yellow leaf curl
Thailand virus and tobacco leaf curl Thailand virus. Arch. Virol. 152:941–954.
Kubiriba, J., Legg, J.P., Tushemereirwe, W. and Adipala, E. (2001a). Disease spread
patterns of Banana streak virus in farmers’ fields in Uganda. Ann. Appl. Biol.
139:31–36.
Kubiriba, J., Legg J.P., Tushemereirwe, W. and Adipala, E. (2001b). Vector
transmission of Banana streak virus in the screenhouse in Uganda. Ann. Appl.
Biol. 139:37–43.
Kumar, L and Hanna, R. (2008). Banana bunchy top virus in sub‐Saharan Africa:
established or emerging problem? p95–96. In Banana and Plantain in Africa:
Harnessing international partnerships to increase research impact. Eds.: T.
Dubois, S. Hauser, C. Staver, D. Coyne. ISHS.
Kumar, P.L., Ayodele, M., Oben, T.T., Mahungu, N.M., Beed, F., Coyne, D., Londa, L.,
Mutunda, M.P. Kiala, D. and Maruthi, M.N. (2008). First report of banana
48
bunchy top virus in banana and plantain (Musa spp.) in Angola. Plant
Pathology. 58:402.
Lassoundiere, A. (1974). La mosaique dite “a tirets” du bananier ‘Poyo’ en Cote
d’Ivore. Fruits 29:349–357.
Leclerc, D., Burri, L., Kajava, A.V., Mougeot, J.L., Hess, D., Lustig, A., Kleeman, G. and
Hohn, T. (1998). The open reading frame III product of Cauliflower mosaic
virus forms a tetramer through a N‐terminal coiled‐coil. J. Biol. Chem.
237:29015–29021.
Le Provost, G., Iskra‐Caruana, M., Acina, I. and Teycheney, P‐Y. (2006). Improved
detection of episomal Banana streak viruses by multiplex immunocapture PCR.
J. Virol. Methods 137:7–13.
Lescot, T. and Ganry, J. (2010). Plantain (Musa spp.) cultivation in Africa: a brief
summary of developments over the previous two decades. In Proc. IC on
Banana & Plantain in Africa p445–456. Edited by T. Dubois, S. Hauser, C.
Staver, D. Coyne. Acta Hort. 879, ISHS.
Lheureux, F., Laboureau, N., Muller, E., Lockhart, B.E.L. and Iskra‐Caruana, M.L.
(2007). Molecular characterisation of banana streak acuminata Vietnam virus
isolated from Musa acuminata siamea (banana cultivar). Arch Virol. 152:1409–
1416.
Lockhart, B. E. L. (1986). Purification and serology of a bacilliform virus associated
with banana streak disease. Phytopathology 76:995–999.
Lockhart, B.E.L. (1990). Evidence for a double‐stranded circular DNA genome in a
second group of plant viruses. Phytopathology 80:127–130.
Lockhart, B.E.L. and Jones, D.R. (2000a). Banana mosaic. In Diseases of banana,
abaca and enset pp256–263. Edited by D.R. Jones. CABI.
Lockhart, B.E.L. and Jones, D.R. (2000b). Banana streak. In Diseases of banana, abaca
and enset pp263–274. Edited by D.R. Jones. CABI.
49
Lockhart B.E.L. and Olszewski N.E. (1993). Serological and genomic heterogeneity of
banana streak badnavirus: Implications for virus detection in Musa
germplasm. In: Breeding Banana and Plantain for Resistance to Diseases and
Pests pp105–113. Edited by J. Ganry. CIRAD/INIBAP.
Lopez, M.M., Bertolini, E., Olmos, A., Caruso, P., Gorris, M.T., Llop, P., Penyalver, R.
and Cambra, M. (2003). Innovative tools for detection of plant pathogenic
viruses and bacteria. Int. Microbiol. 6:233–243.
Lopez, M.M., Llop, P., Olmos, A., Marco‐Noales, E., Cambra, M. and Bertolini, E.
(2009). Are molecular tools solving the challenges posed by detection of plant
pathogenic bacteria and viruses? Curr. Issues Mol. Biol. 11:13–46.
Lufafa, A., Tenywa, M.M., Isabirye, M., Majaliwa M.J.G. and Woomer, P.L. (2002).
Prediction of soil erosion in a Lake Victoria basin catchment using a GIS‐based
Universal Soil Loss model. Agr. Syst. 76:883–894.
Magee, C.J.P. (1927). Investigation on the bunchy top disease of the banana. Council
for scientific and industrial research. Bulletin No. 30. Melbourne.
Mansoor, S., Qazi, J., Amin, I., Khati, A., Khan, I.A., Raza, S., Zafar, Y. and Briddon,
R.W. (2005). A PCR‐based method, with internal control, for the detection of
Banana bunchy top virus in banana. Mol. Biotech. 30:167–169.
May, G.D., Afza, R., Mason, H.S., Wiecko, A., Novak, F.J. and Arntzen, C.J. (1995).
Generation of transgenic banana (Musa acuminata) plants via Agrobacterium‐
mediated transformation. Bio‐Technology 13:486–492.
Meyer, J. B., Kasdorf, G. G. F., Nel, L. H. and Pietersen, G. (2008). Transmission of
activated episomal Banana streak OL (badna)virus (BSOLV) to cv. Williams
banana (Musa sp.) by three mealybug species. Plant Dis. 92:1158–1163.
Mgenzi, S.R.B., Mshaghuley, I.M., Staver, C. and Nkuba, J.M. (2010). Banana (Musa
spp.) processing businesses: support environment and role in poverty
50
reduction in rural Tanzania. In Proc. IC on Banana & Plantain in Africa pp249–
256. Edited by T. Dubois, S. Hauser, C. Staver, D. Coyne. Acta Hort. 879, ISHS.
Mwangi, T., Otipa, M., Wangai, A., Karanja, L., Nyaboga, E., Mwaura, S., Bett, B.,
Irungu, J., Odoyo, E., Muchira, D. and Muchira, C. (2010). Indexing tissue
culture banana (Musa spp.) seedlings for banana streak virus with the aim of
establishing a clean mother block. In Proc. IC on Banana & Plantain in Africa
pp357–360. Edited by T. Dubois, S. Hauser, C. Staver, D. Coyne. Acta Hort. 879,
ISHS.
Ndowora, T., Dahal, G., LaFleur, D., Harper, G., Hull, R., Olszewski, N.E. and Lockhart,
B. (1999). Evidence that badnavirus infection in Musa can originate from
integrated pararetroviral sequences. Virology 255:214–220.
Niblett, C.L., Pappu, S.S., Bird, J. and Lastra, R. (1995). Infectious chlorosis, mosaic
and heart rot. In Compendium of tropical fruit diseases pp18‐19. Edited by
R.C. Ploetz, G.A. Zentmyer, W.T. Nishijima, K.G. Rohrbach and H.D. Ohr. APS.
Nicot, N., Hausman, J‐F., Hoffmann, L. and Evers, D. (2005). Housekeeping gene
selection for real‐time RT‐PCR normalization in potato during biotic and
abiotic stress. J. Exp. Bot. 56:2907–2914.
Njuguna, M.M., Wambugu, F.M., Acharya, S.S. and Mackey, M.A. (2010). Socio‐
economic impact of tissue culture banana (Musa spp.) in Kenya through whole
value chain approach. In Proc. IC on Banana & Plantain in Africa pp77–86.
Edited by T. Dubois, S. Hauser, C. Staver, D. Coyne. Acta Hort. 879, ISHS.
Oben, T.T., Hanna, R., Ngeve, J., Alabi, O.J., Naidu, R.A. and Kumar, P.L. (2009).
Occurrence of banana bunchy top disease caused by Bananan bunchy top
virus on banana and plantain (Musa sp.) in Cameroon. Plant Dis. 93:1076.
Ortiz, R., Ferris, R.S.B. and Vuylsteke, D.R. (1995). Banana and plantain breeding. In
Banana and Plantains pp110–146. Edited by S. Gowen. Chapman & Hall.
51
Ploetz, R.C. (1994). Bract Mosaic. In Compendium of tropical fruit diseases pp20.
Edited by R.C. Ploetz, G.A. Zentmyer, W.T. Nishijima, K.G. Rohrbach and H.D.
Ohr. APS.
Price, N.S. (1995). The origin and development of banana and plantain cultivation. In
Banana and Plantains pp1‐14. Edited by S. Gowen. Chapman & Hall.
Reagin, M.J., Giesler, T.L., Merla, A.L., Resetar‐Gerke, J.M., Kapolka, K.M. and
Mamone, J.A. (2003). TempliPhi: A sequencing template preparation
procedure that eliminates overnight cultures and DNA purification. J. Biomol.
Tech. 14:143–148.
Rodoni, B.C., Ahlawat, Y.S., Varma, A., Dale, J.L., and Harding, R.M. (1997).
Identification and characterization of banana bract mosaic virus in India. Plant
Dis. 81:669–672.
Rodoni, B.C., Dale, J.L. and Harding, R.M. (1999). Characterization and expression of
the coat protein‐coding region of banana bract mosaic potyvirus,
development of diagnostic assays and detection of the virus in banana plants
from five countries in southeast Asia. Arch. Virol. 144:1725–1737.
Sagi, L., Panis, B., Remy, S., Schoofs, H., De Smet, K., Swennen, R. and Cammue,
B.P.A. (1995). Genetic transformation of banana and plantain (Musa spp.) via
particle bombardment. Bio‐Technology 13:481–485.
Schaad, N.W., Frederick, R.D., Shaw, J., Schneider, W.L., Hickson, R., Petrillo, M.D.
and Luster, D.G. (2003). Advances in molecular‐based diagnostics in meeting
crop biosecurity and phytosanitary issues. Ann. Rev. Phytopathol. 41:305–24.
Shamloul, A.M., Hadidi, A., Madkour, M.A. and Makkouk, K.M. (1999). Sensitive
detection of Banana bunchy top and faba bean necrotic yellow viruses from
infected leaves, in‐vitro tissue cultures and viruliferous aphids using
polymerase chain reaction. Can. J. Plant. Pathol. 21:326–337.
52
Sharman, M., Gambley, C.F., Oloteo, E.O., Abgona, R.V.J. and Thomas, J.E. (2000a).
First record of natural infection of abaca (Musa textilis) with banana bract
mosaic potyvirus in the Philippines. Australasian Plant Pathol. 29:69.
Sharman, M., Thomas, J.E. and Dietzgen, R.G. (2000b). Development of a multiplex
immunocapture PCR with colourimetric detection for viruses of banana. J.
Virol. Methods 89:75–88.
Shepherd, D.N., Martin, D.P., Lefeuvre, P., Monjane, A.L., Owor, B.E., Rybicki, E.P.
and Varsani, A. (2008). A protocol for the rapid isolation of full geminivirus
genomes from dried plant tissue. J. Virol. Methods 149:97–102.
Silva, S.O., Souza, M.T., Alves, E.J., Silveira, J.R.S. and Lima, M.B. (2001). Banana
breeding program at Embrapa. Cop Breeding and Applied Biotechnology
1:399–436.
Singh, Z., Jones, R.A.C. and Jones, M.G.K. (1995). Identification of cucumber mosaic
virus subgroup 1 isolates from banana plants affected by infectious chlorosis
disease using RTPCR. Plant Dis. 79:713–716.
Staginnus, C., Iskra‐Caruana, M‐L., Lockhart, B., Hohn, T. and Richert‐Poggeler, K.R.
(2009). Suggestions for a nomenclature of endogenous pararetroviral
sequences in plants. Arch. Virol. 154:1189–1193.
Su, H.J. 1998. First occurrence of Banana streak badnavirus and studies on its
vectorship in Taiwan. In: Banana streak virus: A Unique Virus–Musa
Interaction? Proc. Workshop Promusa Virology Working Group pp20–25.
Edited by E.A. Frison and S.L. Sharrock, INIBAP.
Surga‐Rivas, J.G. (1988). Production of plants of two banana cultivars free from
cucumber mosaic virus by culture of apical meristems isolated in‐vitro.
Fitopatologia Venezolana 1:69–72.
53
Teycheney, P‐Y., Acina, I., Lockhart, B.E.L. and Candresse, T. (2007). Detection of
Banana mild mosaic virus and Banana virus X by polyvalent degenerate
oligonucleotide RT‐PCR (PDO‐RT‐PCR). J. Virol. Methods 142:41–49.
Teycheney, P‐Y., Laboureau, N., Iskra‐Caruana, M‐L. and Candresse, T. (2005a). High
genetic variability and evidence for plant‐to‐plant transfer of Banana mild
mosaic virus. J. Gen. Virol. 86:3179–3187.
Teycheney, P‐Y., Marais, A., Svanella‐Dumas, L. and Candresse, T. (2005b).
Molecular characterization of banana virus X (BVX), a novel member of the
Flexiviridae. Arch. Virol. 150:1715–1727.
Thomas, J.E. and Dietzgen, R.G. (1991) Purification, characterization and serological
detection of virus‐like particles associated with banana bunchy top virus in
Australia. J. Gen. Virol. 72:217–224.
Thomas, J. E., Geering, A. D. W., Gambley, C. F., Kessling, A. F. and White, M. (1997).
Purification, properties, and diagnosis of banana bract mosaic potyvirus and
its distinction from abaca mosaic potyvirus. Phytopathology 87:698–705.
Thomas, J.E., Iskra‐Caruana, M‐L. and Jones, D.R. (1994). Musa Disease Fact Sheet N°
4: Banana Bunchy Top Disease. INIBAP.
Thomas, J.E. and Iskra‐Caruana, M.L. (2000). Bunchy top. In Diseases of banana,
abaca and enset pp241–253. Edited by D.R. Jones. CABI.
Thomas, J.E., Iskra‐Caruana, M.L., Magnaye, L.V. and Jones, D.R. (2000). Bract
mosaic. In Diseases of banana, abaca and enset pp253–256. Edited by D.R.
Jones. CABI.
Thomas, J.E., Smith, M.K., Kessling, A.F. and Hamill, S.D. (1995). Inconsistent
transmission of Banana bunchy top virus in micropropagated bananas and its
implication for germplasm screening. Aust. J. Agric. Res. 46:663–671.
54
Thomson, K.G., Dietzgen, R.G., Thomas, J.E. and Teakle, D.S. (1996). Detection of
pineapple bacilliform virus using the polymerase chain reaction. Ann. Appl.
Biol. 129:57–59.
Trager, J. (1997). The Food Chronology: A food lovers companion of events and
anecdotes from prehistory to present. Owl Books.
van Asten, P.J.A., Florent, D. and Apio, M.S. (2010). Opportunities and constraints
for dried dessert banana (Musa spp.) export in Uganda. In Proc. IC on Banana
& Plantain in Africa pp105–112. Edited by T. Dubois, S. Hauser, C. Staver, D.
Coyne. Acta Hort. 879, ISHS.
Vetten, H. J., Chu, P. W. G., Dale, J. L. & 7 other authors (2005). Nanoviridae. In Virus
Taxonomy: Eighth Report of the International Committee on Taxonomy of
Viruses, pp. 343–352. Edited by C.M. Fauquet, M.A. Mayo, J. Maniloff, U.
Desselberger & L.A. Ball. Elsevier Academic Press.
Wanitchakorn, R., Harding, R.M. and Dale, J.L. (1997). Banana bunchy top virus DNA‐
3 encodes the viral coat protein. Arch. Virol. 142:1673–1680.
Wanitchakorn, R., Hafner, G.J., Harding, R.M. and Dale, J.L. (2000a). Functional
analysis of proteins encoded by banana bunchy top virus DNA‐4 to ‐6. J. Gen.
Virol. 81:299–306.
Wanitchakorn, R., Harding, R.M. and Dale, J.L. (2000b). Sequence variability in the
coat protein gene of two groups of banana bunchy top isolates. Arch. Virol.
145:593‐602.
Wu, C‐Y., Lai, Y‐C., Lin, N‐S., Hsu, Y‐H., Tsai, H‐T., Liao, J‐Y. and Hu, C‐C. (2008). A
simplified method of constructing infectious clones of begomovirus employing
limited restriction enzyme digestion of products of rolling‐circle amplification.
J. Virol. Methods. 147:355–359.
Xie, W.S. and Hu, J.S. (1995). Molecular cloning, sequence analysis, and detection of
banana bunchy top virus in Hawaii. Phytopathology 85:339–347.
55
Yang, I.C., Hafner, G.J., Revill, P.A., Dale, J.L. and Harding, R.M. (2003). Sequence
diversity of South Pacific isolates of Taro bacilliform virus and the
development of a PCR‐based diagnostic test. Arch. Virol. 148:1957–1968.
Zhang. T. and Fang, H.H.P. (2006). Applications of real‐time polymerase chain
reaction for quantification of microorganisms in environmental samples. Appl.
Microbial Biotechnol. 70:281–289.
56
57
CHAPTER 3
DEVELOPMENT OF A NOVEL ROLLING‐CIRCLE AMPLIFICATION TECHNIQUE TO
DETECT BANANA STREAK VIRUS WHICH ALSO DISCRIMINATES BETWEEN
INTEGRATED AND EPISOMAL VIRUS SEQUENCES
A. P. James, R. J. Geijskes, J. L. Dale and R. M. Harding
Centre for Tropical Crops and Biocommodities, Queensland University of
Technology, GPO Box 2434, Brisbane, Qld, 4001, Australia.
Plant Disease 95: 57‐62
58
STATEMENT OF AUTHORSHIP
Anthony James (principal author): Contributed to project concepts, executed the
work (collected samples, designed and conducted laboratory experiments,
analysed and interpreted results) and prepared initial manuscript.
Signed……………………………………………… Date…………………………………………………
Jason Geijskes: Supervised execution of the work, critically interpreted data and
contributed to final manuscript.
Signed……………………………………………… Date…………………………………………………
James Dale: Conceived project idea, collected samples, supervised execution of the
work, critically interpreted data and contributed to final manuscript.
Signed……………………………………………… Date…………………………………………………
Rob Harding: Conceived project idea, supervised execution of the work, critically
interpreted data and contributed to final manuscript.
Signed……………………………………………… Date…………………………………………………
87
CHAPTER 4
MOLECULAR CHARACTERISATION OF SIX BADNAVIRUS SPECIES ASSOCIATED
WITH LEAF STREAK DISEASE OF BANANA IN EAST AFRICA
A. P. James, R. J. Geijskes, J. L. Dale and R. M. Harding
Centre for Tropical Crops and Biocommodities, Queensland University of
Technology, GPO Box 2434, Brisbane, Qld, 4001, Australia.
Annals of Applied Biology 158: 346‐353
88
STATEMENT OF AUTHORSHIP
Anthony James (principal author): Contributed to project concepts, executed the
work (collected samples, designed and conducted laboratory experiments,
analysed and interpreted results) and prepared initial manuscript.
Signed………………………………………………………………..Date……………………………………………
Jason Geijskes: Supervised execution of the work, critically interpreted data and
contributed to final manuscript.
Signed………………………………………………………………..Date……………………………………………
James Dale: Conceived project idea, collected samples, supervised execution of the
work, critically interpreted data and contributed to final manuscript.
Signed………………………………………………………………..Date……………………………………………
Rob Harding: Conceived project idea, supervised execution of the work, critically
interpreted data and contributed to final manuscript.
Signed………………………………………………………………..Date……………………………………………
89
ABSTRACT
Banana leaf streak disease, caused by several species of Banana streak virus (BSV),
is widespread in East Africa. We surveyed for this disease in Uganda and Kenya,
and used rolling‐circle amplification (RCA) to detect the presence of BSV in banana.
Six distinct badnavirus sequences, three from Uganda and three from Kenya, were
amplified for which only partial sequences were previously available. The complete
genomes were sequenced and characterised. The size and organisation of all six
sequences was characteristic of other badnaviruses, including conserved functional
domains present in the putative polyprotein encoded by open reading frame (ORF)
3. Based on nucleotide sequence analysis within the reverse
transcriptase/ribonuclease H‐coding region of open reading frame 3, we propose
that these sequences be recognised as six new species and be designated as
Banana streak UA virus, Banana streak UI virus, Banana streak UL virus, Banana
streak UM virus, Banana streak CA virus and Banana streak IM virus. Using PCR and
species‐specific primers to test for the presence of integrated sequences, we
demonstrated that sequences with high similarity to BSIMV only were present in
several banana cultivars which had tested negative for episomal BSV sequences.
90
INTRODUCTION
Bananas (Musa spp) are hosts to several badnaviruses collectively named banana
streak virus (BSV, genus Badnavirus, family Caulimoviridae). BSV infection causes
leaf streak disease which is characterised by distinct chlorotic and necrotic flecking
on leaves, as well as a range of other symptoms including distortion of leaves and
petioles, stem cracking, abnormal bunch development and death of the growing
point (Dahal et al., 2000). Although found in most banana growing regions, leaf
streak is the most frequently observed viral disease of banana in the Americas and
most of Africa (Geering, 2009; and references therein). The disease is particularly
widespread in both Uganda (Harper et al., 2004, 2005) and Kenya (Karanja et al.,
2008).
Badnaviruses are plant pararetroviruses with non‐enveloped, bacilliform‐
shaped virions of approximately 30 x 130‐150 nm, and circular double‐stranded
DNA genomes of 7‐8 kbp (Hull et al., 2005). All badnaviruses typically encode three
open reading frames (ORFs) on the virus‐sense coding strand. ORF 1 encodes a
small protein of unknown function which has been associated with virions (Cheng
et al., 1996). ORF 2 encodes a protein of 14 kDa with a non‐specific DNA‐ and
RNA‐binding activity (Jacquot et al., 1996). This protein may function in virion
assembly due to the presence of a predicted N‐terminal coiled‐coil domain which
supports self‐interaction to form a tetramer (Leclerc, 1998). The large polyprotein
encoded by ORF 3 contains domains associated with movement, the virus capsid,
aspartic protease, reverse transcriptase (RT) and ribonuclease H (RNaseH)
functions.
91
Three distinct species of BSV, namely Banana streak OL virus (BSOLV),
Banana streak GF virus (BSGFV) and Banana streak MY virus (BSMYV), previously
Banana streak Mysore virus, are now recognised by the International Committee
on the Taxonomy of Viruses (ICTV) (Geering, 2010), while a fourth species, Banana
streak VN virus (BSVNV, previously Banana streak acuminata Vietnam virus) has
also recently been proposed based on full‐length sequence analyses (Lheureux et
al., 2007). However, many other BSV species are thought to exist based on the
analyses of numerous reported partial sequences. Geering et al., (2000; 2005A)
reported partial sequences of two BSV isolates from Australia (named Banana
streak Cavendish virus (BSV‐Cav) and Banana streak Imové virus (BSImV)), while
Harper et al., (2005) reported the presence of 13 distinct BSV sequence groups
from Uganda, named consecutively as Banana streak Uganda A virus to Banana
streak Uganda M virus. Further, a complete but as yet unpublished BSV sequence,
named Banana streak acuminata Yunnan virus, has been deposited in the NCBI
database (GenBank accession no. DQ092436). This isolate is phylogenetically most
closely related to BSVNV (Gayral and Iskra‐Caruana, 2009).
Several episomal BSV sequences, namely BSOLV, BSMYV, BSGFV and
BSIMV, have been shown to have integrated counterparts, termed endogenous
BSV (eBSV), in the Musa genome (Harper et al., 1999; Geering et al., 2005A & B;
Gayral et al., 2008). Under certain stress conditions, these sequences can be
activated to cause episomal infections (Ndowora et al., 1999; Dallot et al., 2001;
Côte et al., 2010). Although many other endogenous badnavirus sequences occur
in the banana genome, these have no known episomal counterparts and are not
92
known to give rise to episomal infections (Geering et al., 2005A). While
endogenous badnaviruses occur in genetic backgrounds which include both Musa
acuminata (A‐genome) and M. balbisiana (B‐genome) and their hybrids, eBSVs are
only known to occur in some Musa accessions which contain a B‐genome.
As part of a Grand Challenges in Global Health initiative funded by the Bill
and Melinda Gates Foundation, we have been developing a diagnostic capacity for
banana viruses in East Africa with a specific focus on BSV. Diagnostic tests for BSV
have been complicated by the extensive genetic and serological diversity that
exists amongst BSV isolates and the presence of integrated BSV sequences in some
banana cultivars which leads to false positives. Recently, however, we have
developed a rolling‐circle amplification (RCA)‐based assay that specifically detects
episomal, and not integrated, BSV sequences (James et al., 2011). To determine
the prevalence of BSV in East Africa, we conducted disease surveys of bananas in
Uganda and Kenya and tested samples by RCA. In this paper, we report for the first
time the complete nucleotide sequence and molecular characterisation of six
distinct BSV species from these two countries.
MATERIALS AND METHODS
Virus nomenclature
For consistency throughout the remainder of this manuscript, we have elected to
use the nomenclature suggested by Geering (2010) to describe BSV species.
Banana streak Imové virus (BSImV; Geering et al., 2005A) will be referred to as
Banana streak IM virus (BSIMV); Banana streak Cavendish virus (BSV‐Cav; Geering
93
et al., 2000) will be referred to as Banana streak CA virus (BSCAV); Banana streak
Uganda A virus (BSUgAV; Harper et al., 2005) will be referred to as Banana streak
UA virus (BSUAV); Banana streak Uganda I virus (BSUgIV; Harper et al., 2005) will
be referred to as Banana streak UI virus (BSUIV); Banana streak Uganda L virus
(BSUgLV; Harper et al., 2005) will be referred to as Banana streak UL virus (BSULV);
and Banana streak Uganda M virus (BSUgMV; Harper et al., 2005) will be referred
to as Banana streak UM virus (BSUMV).
Plant samples and amplification of viral DNA
Leaf samples were collected from banana plants displaying the chlorotic and
necrotic flecking symptoms typically associated with leaf streak disease (Table 1).
Three samples (Ug1, Ug8 & Ug12) were collected in south‐west Uganda during a
survey conducted in April 2008, and were previously shown to contain BSV‐like
sequences with homology to BSUIV, BSULV and BSUMV, respectively (James et al.,
2011). One sample (Ke171) was collected in western Kenya during a survey in April
2009, while the remaining two samples (Ke8 & Ke10) were obtained from the
Kenyan Agricultural Research Institute research station at Njoro, Kenya. Total
nucleic acid (TNA) extracts were prepared and virus DNA amplified using the
Illustra TempliPhi 100 Amplification Kit (GE Healthcare, Buckinghamshire, United
Kingdom) as described previously (James et al., 2011).
Leaf samples were also obtained from 12 genotypically diverse Musa
cultivars (Table 2) growing in tissue culture at DEEDI, Agri‐Science Queensland,
Nambour, Australia. These plants had previously been certified as BSV negative
94
Table 1 Plant samples for RCA analysis and PCR primers used for detection of integrated sequences
Sample Primer 1 (5´ ‐ 3´) Primer 2 ( 5´ ‐ 3´) Amplicon size (bp)
Annealing temp (°C)
Region of genome amplified
Ug1 GAACTGACAGTAGCGCAATCG GACTTGGCTTGCCTGAGTATCG 943 60 6282‐7224Ug8 GAATCCTCAAAGGTACCCC CATGAGGTCAAGCATATGC 619 50 435‐1053 Ug12 GACGAGCTGCAAGCTCTCAGG TGTGCCTATTCTGAGGTTGG 467 50 973‐1439Ke8 CTCAGCGGCAAGATTAGGAAGG TCCCCATTGGTCGTCATTGC 517 60 6513‐7029 Ke10 GCTAGGAAGAAAAGTCTGGG TGCAAGTCTACTTACACAGC 475 50 7417‐122Ke171 AGGATTGGATGTGAAGTTTGAGC ACCAATAATGCAAGGGACGC 783 57 6425‐7207
95
Table 2 BSV‐indexed leaf samples used for PCR analysis
Cultivar Genotype Virus specific PCR test
BSUAV BSUIV BSULV BSUMV BSCAV BSIMV
Calcutta 4 AA ‐ ‐ ‐ ‐ ‐ ‐Pisang Oli AA ‐ ‐ ‐ ‐ ‐ ‐ Yangambi km5 AAA ‐ ‐ ‐ ‐ ‐ ‐ NC‐301 AAA ‐ ‐ ‐ ‐ ‐ ‐FHIA‐17 AAAA ‐ ‐ ‐ ‐ ‐ ‐ Da Jiao ABB ‐ ‐ ‐ ‐ ‐ ‐Ainu AAB ‐ ‐ ‐ ‐ ‐ + SH‐3460.10 AAAB ‐ ‐ ‐ ‐ ‐ ‐FHIA‐03 AABB ‐ ‐ ‐ ‐ ‐ + Balonkawe ABB ‐ ‐ ‐ ‐ ‐ +Goly Goly Pot Pot ABB ‐ ‐ ‐ ‐ ‐ + Lal Velchi BB ‐ ‐ ‐ ‐ ‐ ‐
96
using immuno‐sorbent electron microscopy (ISEM, Geering et al., 2000) as well as
by RCA (James et al., 2011).
Cloning and sequencing of virus DNA
RCA‐amplified virus DNA was digested using either StuI or PstI and products were
separated by agarose gel electrophoresis. DNA was subsequently cloned into
pUC19 and sequenced as described previously (James et al., 2011). In all cases, at
least three independent clones were sequenced in both directions to determine a
consensus sequence. Putative identification of cloned fragments was made by
comparison to sequences in NCBI database (http://www.ncbi.nlm.nih.gov) using
the Basic local alignment search tool (BLAST) programs. According to the ICTV
criteria (Hull et al., 2005) for species demarcation within the genus Badnavirus,
sequence differences within the RT/RNaseH‐coding region of more than 20% are
considered distinct badnavirus species. Sequence comparisons were based on a
529 bp region of the RT/RNaseH‐coding region delimited by the BadnaFP/RP
primers reported in Yang et al. (2003). In cases where BSV isolates were identified
for which full‐length sequences had not been previously reported, the complete
sequence was obtained by primer‐walking. The sequence spanning putative
restriction sites was confirmed by sequencing of PCR products generated using
sequence‐specific primers for each putative site present in each virus isolate. PCR
mixes (20 µL) contained 10 µl 2x GoTaq Green Master Mix (Promega Corp,
Madison, WI), 5 ρmol of each primer, 1 µl of nucleic acid extract and water to final
volume. PCR cycling conditions were an initial denaturation of 94°C for 2 min
97
followed by 35 cycles of 94°C for 20 s, 50°C for 20 s, and 72°C for 30 s, with a final
extension at 72°C for 2 min. Reactions products were electrophoresed through
1.5% agarose gels, stained using 0.25x SYBR® Safe DNA Gel Stain (Invitrogen Corp,
Carlsbad, CA) and DNA fragments visualised on a Safe imager blue‐light
transilluminator. Amplified fragments were cloned into pGEM‐T easy (Promega
Corp, Madison, WI) and three clones were sequenced in both directions using
universal M13 primers.
PCR for detection of integrated sequences
PCR primers for detection of each virus were designed from sequences obtained in
this study (Table 1), with the exception of BSCAV, for which previously published
primers were used (Geering et al., 2000). PCR was carried out as described above,
with annealing temperatures of 50°C, 57°C or 60°C (Table 1).
Phylogenetic analysis
Full‐length sequences were assembled in VectorNTI Advance v.11 (Invitrogen Corp,
Carlsbad, CA), which was subsequently used for the identification of putative ORFs
and other conserved genomic features of badnaviruses, as well as translation of
putative ORFs for analysis of conserved domains present in badnavirus proteins.
Conserved features of genome sequences were identified as previously described
(Geering et al., 2010; Lheureux et al., 2007). For phylogenetic analysis and
sequence comparison to published badnavirus sequences, RT/RNaseH core
98
sequences were identified using the method of Geering et al., (2010). RT/RNaseH
sequences were aligned using the CLUSTAL W algorithm in MEGA 4.0 (Tamura et
al., 2007). Phylogenetic trees were constructed using the neighbour‐joining
method, following pair‐wise sequence comparison using the Kimura 2‐parameter
model in MEGA 4.0 (Kimura, 1980).
RESULTS
Cloning and sequencing
Nucleic acid was extracted from the three Ugandan leaf samples (Ug 1, 8 and 12)
and three Kenyan leaf samples (Ke8, 10 and 171) and was subjected to RCA. The
products were digested with StuI (or PstI in the case of sample Ke171) and
analysed by agarose gel electrophoresis. A single band of 7.5 kbp, presumably
representing full‐length BSV genomic DNA, was observed in the Ug1, Ke8 and
Ke171 extracts, while two bands (5.5 and 1.8 kbp), three bands (3.7, 2.2 and 1.6
kbp) and four bands (7.5, 4.4, 2.1 and 1 kbp) were observed in the Ug8, Ug12 and
Ke10 extracts, respectively.
When the restriction fragment/s from Ug1, Ug12, Ke8 and Ke171 were
cloned and analysed, contiguous full‐length sequences were obtained. Analysis of
the RT/RNaseH‐coding region of these sequences revealed 82‐86%, 90‐95%, 94%
and 94‐95% similarity to BSULV, BSUMV, BSUAV and BSCAV isolates, respectively.
Analysis of the 3’ and 5’ terminal sequences of the 7.5 kbp band derived from
Ke10 revealed it was identical to that of Ke8. When the remaining three restriction
99
fragments from this isolate were cloned and analysed, a contiguous full‐length
sequence was obtained. The sequence of the RT/RNaseH‐coding region showed 97‐
99% similarity to BSIMV.
Although analysis of the 5.5 kbp StuI‐digested fragment from Ug8 yielded
a consensus sequence, the sequences obtained from the 1.8 kbp fragment varied
in length. To resolve this problem and obtain the complete sequence, the RCA
product derived from Ug8 was digested with PstI to yield a major fragment of 7
kbp. The entire sequence of the Ug8 isolate was subsequently obtained from
alignment of sequences obtained from the StuI‐ and PstI‐derived fragments and
from PCR‐derived sequences used to confirm the presence of the PstI restriction
site. Analysis of the RT/RNaseH‐coding sequences of Ug8 revealed 92‐98%
similarity to BSUIV.
Sequence analysis
A summary of the characteristics of the complete genomes of each of the six BSV
species is presented in Table 3. The complete genomic sequences of BSUAV
(sample Ke8, Genbank accession number HQ593107), BSUIV (Ug8, accession
number HQ593108), BSULV (Ug1, accession number HQ593109), BSUMV (Ug12,
accession number HQ593110), BSIMV (Ke10, accession number HQ593112) and
BSCAV (Ke171, accession number HQ593111) comprised 7519, 7458, 7401, 7532,
7769 and 7408 bp, respectively. For consistency with previous conventions,
numbering of each of the genomes begins with the 5’ nucleotide of the putative
tRNAmet priming site. Each of the six genomes contained three ORFs and, with the
100
Table 3 Genome features of East African BSV species
ORF1 ORF2 ORF3 Transcriptional elements
Badnavirus species length total
length start‐stop
protein size
length start‐stop protein size
length start‐stop
protein size
TATA <gap> polyA
(nt) (nt) (codon use) (kDa) (nt) (codon use) (kDa) (nt) (codon use) (kDa)
Banana streak UA virus 7519 534 483‐1016 20.9 390 1013‐1402 14.3 5637 1402‐7038 216 ctcTATATAAgga <56> aataag
(BSUAV) (atg‐tga) (atg‐taa) (atg‐taa)
Banana streak UI virus 7458 561 502‐1062 21.6 336 1059‐1394 12.3 5514 1395‐6908 211 ctcTATATAAgga <66> aataaa
(BSUIV) (atg‐tga) (atg‐taa) (atg‐taa)
Banana streak UL virus 7401 561 532‐1092 21.8 339 1089‐1427 12.3 5502 1430‐6931 211 ctcTATATAAgga <64> gataag
(BSULV) (atg‐tga) (atg‐taa) (atg‐tga)
Banana streak UM virus 7532 564 622‐1185 21.7 312 1182‐1493 11.7 5547 1497‐7043 213 ggcTATATATAggt <45> aataaa
(BSUMV) (atg‐tga) (atg‐tag) (atg‐taa)
Banana streak IM virus 7769 531 668‐1198 20.9 393 1195‐1587 14.3 5613 1590‐7202 215 atcTATAA‐‐gag <74> aataaa
(BSIMV) (ctg‐tga) (atg‐taa) (atg‐taa)
Banana streak CA virus 7408 531 515‐1045 21.1 405 1042‐1446 14.7 5511 1446‐6956 212 ctcTATAAATAgga <55> aataag
(BSCAV) (atg‐tga) (atg‐taa) (atg‐taa)
101
exception of BSIMV ORF1, a conventional ATG initiation codon was present for all
ORFs. A non‐conventional CTG initiation codon was predicted for BSIMV ORF1
based on sequence comparisons with both BSMYV and BSVNV. The predicted size
of proteins encoded by the putative ORFs identified for each sequence is included
in Table 3. The size of the intergenic region in all six virus genomes ranged from
963 nt (BSUAV) to 1234 nt (BSIMV) and contained a region with between 16‐18
nucleotides complementary to the consensus sequence of plant tRNAmet. Putative
TATA boxes and polyadenylation signals, were also identified in the intergenic
region of all sequences, 5’ of the putative tRNAmet primer binding site (Table 3).
Analysis of the putative protein encoded by ORF 3 of each species revealed the
presence of several motifs that are highly conserved in badnavirus proteins
including movement, RNA‐binding (zinc‐finger motif), aspartyl proteinase, reverse
transcriptase and RNaseH.
Phylogenetic analysis of the RT/RNaseH region from full‐length known
episomal badnavirus sequences showed that the three Ugandan BSVs (BSUIV,
BSULV & BSUMV) clustered together and that these viruses were more closely
related to the sugarcane‐infecting badnaviruses than to other banana‐infecting
badnaviruses (Fig. 1). Two Kenyan BSVs, BSUAV and BSCAV, were shown to be
closely related and these formed a separate cluster with BSOLV (Fig. 1). The Kenyan
BSIMV did not cluster according to provenance but instead was found to be most
closely related to BSVNV. Pair‐wise nucleotide similarities within the RT/RNaseH‐
coding region of full‐length sequences derived from the six BSV sequences
reported in this study together with their phylogenetically most closely‐related
102
BSCAV
BSUAV
BSOLV
BSMYV
KTSV
BSIMV
BSVNV
BSGFV
ComYMV
CSSV
CiYMV
DBSNV
TaBV
BCVBV
SCBIMV
SCBMOV
BSUIV
BSULV
BSUMV
RTBV
CaMV
100
97
94
97
79
62
68
71
100
92
98
100
100
0.05
103
Figure 1 Phylogenetic tree using neighbour‐joining method (Kimura 2‐parameter
model with bootstrapping (1000 replicates)) of the RT/RNaseH region of selected
badnaviruses. Rice tungro bacilliform virus (RTBV; genus Tungrovirus) and
Cauliflower mosaic virus (CaMV; genus Caulimovirus) were used as out‐groups to
the genus Badnavirus. GenBank accession numbers are: Banana streak OL virus
(BSOLV; GenBank accession NC_003381), Banana streak MY virus (BSMYV;
GenBank accession NC_006955), Kalanchoe top‐spotting virus (KTSV; GenBank
accession NC_004540), Banana streak VN virus (BSVNV; Genbank accession
AY750155), Banana streak GF virus (BSGFV; GenBank accession NC_007002),
Commelina yellow mottle virus (ComYMV; GenBank accession NC_001343), Cacao
swollen shoot virus (CSSV; GenBank accession NC_001574), Citrus yellow mosaic
virus (CiYMV; GenBank accession NC_003382), Dioscorea bacilliform SN virus (DBV;
GenBank accession DQ822073), Taro bacilliform virus (TaBV; Genbank accession
AF357836), Bougainvillea chlorotic vein banding virus (BCVBV; GenBank accession
EU034539), Sugarcane bacilliform IM virus (SCBIMV; GenBank accession
NC_003031), Sugarcane bacilliform MO virus (SCBMOV; GenBank accession
NC_008017), RTBV (GenBank accession NC_001914), and CaMV (GenBank
accession NC_001497). Species whose genome was fully sequenced for the first
time are shown in bold.
104
counterparts revealed at least a 20% nucleotide difference between the BSVs reported here
and other recognised badnaviruses (Table 4).
Detection of integrated sequences
To investigate the possible presence of integrated DNA of the six new BSV species in banana
genomic DNA, nucleic acid was extracted from a diverse collection of banana cultivars (Table
2) which had previously been certified as negative for episomal BSV sequences by ISEM and
RCA. The nucleic acid extracts were subsequently used in PCR with primers designed to
specifically amplify a selected fragment of each of the six BSVs. Whereas no amplicons were
detected in extracts from any of the 12 banana cultivars tested for BSUAV, BSUIV, BSULV,
BSUMV or BSCAV, amplicons of the expected size were detected in extracts derived from
the four banana cultivars Ainu, FHIA‐3, Balonkawe and Goly Goly Pot Pot tested for BSIMV
(Table 2). These results indicated that, for BSUAV, BSUIV, BSULV, BSUMV or BSCAV, the
sequences delimited by these primers were not present as either episomal or integrated
DNA in any of the 12 cultivars tested. In contrast, the genomic DNA of four of the 12
cultivars tested appeared to contain an integrated BSIMV sequence.
DISCUSSION
We have completely sequenced the genomes of six BSV isolates collected in East
Africa, for which only partial sequences have been previously available. All six
105
aPercent difference shown was calculated using the Kimura 2‐parameter model following ClustalW alignment in MEGA 4.0.
Table 4 Pair‐wise distance matrix using core RT/RNaseH sequences of selected badnavirusesa
BSCAV BSUAV BSOLV BSGFV BSUIV BSULV BSUMV SCBIMV SCBMOV BSMYV BSIMV BSVNV KTSV
BSCAV 23.2 26.1 38.3 45.8 43.1 47.1 50.8 47.9 38.9 33.5 34.8 37.4BSUAV 30.6 40.0 45.6 46.3 48.6 51.2 52.8 37.3 35.7 34.4 37.0BSOLV 43.4 44.4 48.2 46.4 53.6 50.8 38.0 28.0 38.2 34.6BSGFV 50.3 47.8 47.9 58.1 53.9 45.2 42.1 40.2 41.9BSUIV 23.2 27.9 38.1 34.2 47.8 52.7 49.9 50.4BSULV 26.1 39.2 35.6 48.2 50.3 51.2 53.0BSUMV 42.3 35.8 48.8 48.9 47.6 50.4SCBIMV 25.0 59.0 57.4 60.3 60.0SCBMOV 52.7 54.5 54.2 56.3BSMYV 43.3 38.6 39.5BSIMV 27.0 37.8BSVNV 36.8KTSV
106
sequences had a typical badnavirus genome organisation and contained the
conserved motifs characteristically found in the putative ORF 3 polyprotein of
badnaviruses. Based on the criteria for recognition of distinct species in the genus
Badnavirus (a difference in the nucleotide sequence of the RT/RNaseH‐coding
region of more than 20% (Hull et al., 2005)), we propose that the six isolates should
be recognised as new BSV species and be designated as BSUAV, BSUIV, BSULV,
BSUMV, BSIMV and BSCAV based on a recent amendment to the naming
convention for badnaviruses (Geering, 2010). The results from this study confirm
the presence, in Uganda, of three of the 13 putative BSV species reported
previously (Harper et al., 2005). Further, this is the first report of BSCAV from
Kenya, and confirms a previous report of BSUAV and BSIMV in Kenya using IC‐PCR
(Karanja et al., 2008). Importantly, the detection of BSUAV, BSUIV, BSULV and
BSUMV in this study is based on episomal DNA amplified by RCA, and not IC‐PCR
which may detect integrated sequences (Le Provost et al., 2006, Iskra‐Caruana et
al., 2009), as in previous studies.
Previous phylogenetic analyses of full‐length and partial sequences have
consistently identified three distinct clades of banana‐infecting badnaviruses
(Harper et al., 2005; Bousalem et al., 2008; Gayral and Iskra‐Caruana, 2009). The
isolates reported in this study grouped within two of the three clades, consistent
with previous reports. BSUIV, BSULV and BSUMV grouped within clade 3 which also
includes badnavirus species characterised from sugarcane while BSUAV, BSCAV and
BSIMV grouped within clade 1, which contained only badnavirus species originating
from banana. Interestingly, however, the sequence of our BSCAV isolate also
107
showed 93‐96% homology at the nucleotide level to six unpublished Sugarcane
bacilliform virus (ScBV) isolates in the GenBank database. As such, ScBV isolates
now appear to group within two different clades (clades 1 and 3), suggesting that
the movement of badnaviruses across the host‐plant boundary between sugarcane
and banana has likely occurred on more than one occasion.
Several authors (Bousalem et al., 2008; Gayral and Iskra‐Caruana, 2009)
have reported a close phylogenetic relationship between BSULV and the proposed
BSUgKV identified by Harper et al (2005). A comparison of the RT/RNaseH‐coding
sequence of our BSULV isolate to other badnavirus sequences revealed 82%
homology to four sequences described as BSUgKV (Harper et al., 2005). This finding
suggests that BSULV and BSUgKV are a single species.
The sequences of several BSV species are known to occur within the M.
balbisiana genome, the presence of which has provided a challenge for the
diagnosis of episomal BSV infection using PCR‐based methodologies. As a
preliminary study to determine whether sequences related to the six viruses
reported in this study had integrated counterparts, 12 banana genotypes known to
be free of episomal BSV were tested for each of the six viruses by PCR using virus‐
specific primers. None of the samples tested positive for BSUAV, BSUIV, BSULV,
BSUMV or BSCAV, while four plants, all containing B‐genomes, tested positive for
BSIMV. The detection of integrated BSIMV sequences in banana accessions with a
M. balbisiana genome component is consistent with previous reports (Geering et
al., 2005A, Gayral et al., 2010). Further, when the sequences of the core
RT/RNaseH‐coding region of the six BSV species described in this manuscript were
108
used to search for homologous endogenous sequences (both endogenous
badnavirus and Musa genomic BAC sequences) in GenBank, BSIMV showed 99%
homology to several endogenous badnavirus sequences and 94% homology to a
Musa balbisiana BAC sequence (GenBank accession AP009334). In contrast, the
remaining five species did not produce a significant match (i.e. 80% similarity or
greater) to either endogenous badnavirus or Musa BAC sequences. While our
findings do not conclusively exclude the presence of integrated sequences for each
of the other five BSV species in the genotypes tested, the results suggest that PCR
might be a suitable tool for diagnosis of these species in bananas with selected
genetic backgrounds.
Diagnosis of BSV using the non‐sequence specific RCA assay should
dramatically improve the scope of detection of heterogeneous mixture of viruses
comprising the BSV complex. Further, the full‐length sequences presented here will
improve the opportunity to diagnose BSV infections using restriction‐digest based
RCA assays. Additional work is still required, however, to confirm the episomal
nature of the additional Ugandan BSV isolates reported by Harper et al., (2005).
Biological information pertaining to the BSV species reported in this work will be
useful to understand the BSV‐banana system with more clarity, and allow the
development of improved diagnostic tests.
ACKNOWLEDGEMENTS
This research was funded by the Bill and Melinda Gates Foundation Grand
Challenges in Global Health Program. We are grateful to our colleagues Dr. Charles
109
Changa and Dr. Jerome Kubiriba at the National Agricultural Research Organisation
(NARO), Uganda, as well as Dr. Laura Karanja at the Kenyan Agricultural Research
Institute (KARI), Njoro, Kenya, for assisting with field collections and leaf samples of
banana.
110
REFERENCES
Bousalem, M., Douzery, E.J.P., Seal, S.E. (2008) Taxonomy, molecular phylogeny
and evolution of plant reverse transcribing viruses (family Caulimoviridae)
inferred from full‐length genome and reverse transcriptase sequences.
Archives of Virology, 153, 1085‐1102.
Cheng, C.P., Lockhart, B.E.L., Olszewski, N.E. (1996) The ORFI and II proteins of
Commelina yellow mottle virus are virion‐associated. Journal of General
Virology, 223, 263‐271.
Côte, F.X., Galzi, S., Folliot, M., Lamagnére, Y., Teycheney, P‐Y., Iskra‐Caruana, M‐L.
(2010) Micropropagation by tissue culture triggers differential expression of
infectious endogenous Banana streak virus sequences (eBSV) present in the B
genome of natural and synthetic interspecific banana plantains. Molecular
Plant Pathology, 11, 137‐144.
Dahal, G., Hughes, J., Gauhl, F., Pasberg‐Gauhl, C., Nokoe, K.S. (2000)
Symptomatology and development of banana streak, a disease caused by
banana streak badnavirus, under natural conditions in Ibadan, Nigeria. Acta
Horticulturae (ISHS), 540, 361‐375.
Dallot, S., Acuña, P., Rivera, C., Ramírez, P., Côte, F., Lockhart, B.E.L., Caruana, M.L.
(2001) Evidence that the proliferation stage of micropropagation procedure
is determinant in the expression of banana streak virus integrated into the
genome of the FHIA 21 hybrid (Musa AAAB). Archives of Virology, 146, 2179‐
2190.
Gayral, P., Blondin, L., Guidolin, O., Carreel, F., Hippolyte, I., Perrier, X., Iskra‐
Caruana, M‐L., (2010) Evolution of endogenous sequences of Banana streak
virus: What can we learn from banana (Musa sp.) evolution? Journal of
Virology, 84, 7346 – 7359.
Gayral, P., Iskra‐Caruana, M.L. (2009) Phylogeny of Banana streak virus reveals
recent and repetitive endogenization in the genome of its banana host (Musa
sp.). Journal of Molecular Evolution, 69, 65‐80.
111
Gayral, P., Noa‐Carrazana, J‐C., Lescot, M., Lheureux, F., Lockhart, B.E.L.,
Matsumoto, T., Piffanelli, P., Iskra‐Caruana, M‐L. (2008) A single Banana
streak virus integration event in the banana genome as the origin of
infectious endogenous pararetrovirus. Journal of Virology, 82, 6697‐6710.
Geering, A.D.W. (2009) The distribution of badnavirus‐infected Musa germplasm
from the international transit centre, Katholieke Universiteit Leuven,
Belgium. In International Symposium on Recent Advances in Banana Crop
Protection for Sustainable Production and Improved Livelihoods, pp85‐88. Eds
D. Jones and I. Van den Bergh. ISHS.
Geering, A.D.W. (2010) Five new species and four nomenclature changes in the
genus Badnavirus. [online] ICTV Taxonomy Proposals ‐ Plant. Available at
http://talk.ictvonline.org/files/proposals/taxonomy_proposals_plant1/m/pla
nt01/2216.asp [Accessed 14 December 2010].
Geering, A.W.D., McMichael, L.A., Dietzgen, R.G., Thomas, J.E. (2000) Genetic
diversity among banana streak virus isolates from Australia. Phytopathology,
90, 921‐927.
Geering, A.D.W., Olszewski, N.E., Harper, G., Lockhart, B.E.L., Hull, R., Thomas, J.E.
(2005A) Banana contains a diverse array of endogenous badnaviruses.
Journal of General Virology, 86, 511‐520.
Geering, A. D. W., Pooggin, M.M., Olszewski, N.E., Lockhart, B.E.L., Thomas, J.E.
(2005B) Characterisation of banana streak Mysore virus and evidence that its
DNA is integrated in the B genome of cultivated Musa. Archives of Virology,
150, 787‐796.
Geering, A.W.D., Scharaschkin, T., Teycheney, P.Y. (2010) The classification and
nomenclature of endogenous viruses of the family Caulimoviridae. Archives
of Virology, 155, 123‐131.
Harper, G., Hart, D., Moult, S., Hull, R. (2004) Banana streak virus is very diverse in
Uganda. Virus Research, 100, 51‐56.
112
Harper, G., Hart, D., Moult, S., Hull, R., Geering, A., Thomas, J. (2005) The diversity
of banana streak virus isolates in Uganda. Archives of Virology, 150, 2407‐
2420.
Harper, G., Osuji, J.O., Heslop‐Harrison, J.S., Hull, R. (1999) Integration of banana
streak badnavirus into the Musa genome: molecular and cytogenetic
evidence. Virology, 255, 207‐213.
Hull, R., Geering, A., Harper, G., Lockhart, B.E., Schoelz, J.E. (2005) Family
Caulimoviridae. In Virus Taxonomy: Eighth Report of the International
Committee on Taxonomy of Viruses p. 385‐396. Eds C.M. Fauquet, M.A.
Mayo, J. Maniloff, U. Desselberger and L.A. Ball. London: Elsevier.
Iskra‐Caruana, M.L., Gayral, P., Galzi, S., Laboureau, N. (2009) How to control and
prevent the spread of banana streak disease when the origin could be viral
sequences integrated in the banana genome? In International Symposium on
Recent Advances in Banana Crop Protection for Sustainable Production and
Improved Livelihoods, pp77‐84. Eds D. Jones and I. Van den Bergh. ISHS.
Jacquot, E., Hagen, L.S., Jacquemond, M., Yot, P. (1996) The open reading frame 2
product of Cacao swollen shoot badnavirus is a nucleic acid‐binding protein.
Virology, 225, 191‐195.
James, A.P., Geijskes, R.J., Dale, J.L., Harding, R.M. (2011) Development of a novel
rolling‐circle amplification technique to detect banana streak virus which also
discriminates between integrated and episomal virus sequences. Plant
Disease, 95, 57‐62.
Karanja, L., Wangai, A., Harper, G., Pathak, R.S. (2008) Molecular identification of
Banana streak virus isolates in Kenya. Journal of Phytopathology, 156, 678‐
686.
Kimura, M. (1980) A simple method for estimating evolutionary rate of base
substitutions through comparative studies of nucleotide sequences. Journal
of Molecular Evolution, 16, 111‐120.
113
Le Provost, G., Iskra‐Caruana, M‐L., Acina, I., Teycheney, P‐Y. (2006) Improved
detection of episomal Banana streak viruses by multiplex immunocapture
PCR. Journal of Virological Methods, 137, 7–13.
Leclerc, D., Burri, L., Kajava, A.V., Mougeot, J.L., Hess, D., Lustig, A., Kleeman, G.,
Hohn, T. (1998) The open reading frame III product of Cauliflower mosaic
virus forms a tetramer through a N‐terminal coiled‐coil. The Journal of
Biological Chemistry, 237, 29015‐29021.
Lheureux, F., Laboureau, N., Muller, E., Lockhart, B.E.L., Iskra‐Caruana, M.L. (2007)
Molecular characterisation of banana streak acuminata Vietnam virus
isolated from Musa acuminata siamea (banana cultivar). Archives of Virology,
152, 1409‐1416.
Ndowora, T., Dahal, G., LaFleur, D., Harper, H., Hull, R., Olszewski, N.E., Lockhart, B.
(1999) Evidence that badnavirus infection in Musa can originate from
integrated pararetroviral sequences. Virology, 255, 214‐220.
Tamura, K., Dudley, J., Nei, M., Kumar, S. (2007) MEGA4: Molecular Evolutionary
Genetics Analysis (MEGA) software version 4.0. Molecular Biology and
Evolution, 24, 1596‐1599.
Yang, I.C., Hafner, G.J., Revill, P.A., Dale, J.L., and Harding, R.M. (2003) Sequence
diversity of South Pacific isolates of taro bacilliform virus and the
development of a PCR‐based diagnostic test. Archives of Virology, 148, 1957–
1968.
114
115
CHAPTER 5
COMPARISON OF ROLLING‐CIRCLE AMPLIFICATION AND DIRECT‐PCR BASED
METHODS FOR DIAGNOSIS OF BSV INFECTION IN EAST AFRICA
A. P. James1, L. Karanja2, J. Kubiriba3, C. M. Changa3, J. A. Mugini4, R. J. Geijskes1,
W. K. Tushemereirwe3, J. L. Dale1 and R. M. Harding1
1Centre for Tropical Crops and Biocommodities, Queensland University of
Technology, P.O. Box 2434, Brisbane, Queensland, 4001, Australia
2Kenya Agricultural Research Institute, Njoro, P.O. Njoro, Kenya
3National Agricultural Research Laboratories, Kawanda, National Agricultural
Research Organisation, P.O. Box 7065, Kampala, Uganda
4Mikocheni Agricultural Research Institute, P.O. Box 6226, Dar Es Salaam, Tanzania
[Formatted for submission to Journal of Virological Methods]
116
STATEMENT OF AUTHORSHIP
Anthony James (principal author): Contributed to project concepts, executed the
work (collected samples, designed and conducted laboratory experiments,
analysed and interpreted results) and prepared initial manuscript.
Signed………………………………………………………………..Date…………………………………………
Laura Karanja: Collected samples, conducted laboratory experiments, analysed and
interpreted results, critically interpreted data and contributed to final manuscript.
Signed………………………………………………………………..Date…………………………………………
Jerome Kubiriba: Collected samples, conducted laboratory experiments, analysed
and interpreted results, critically interpreted data and contributed to final
manuscript.
Signed………………………………………………………………..Date…………………………………………
Charles Changa: Collected samples, conducted laboratory experiments, analysed
and interpreted results, critically interpreted data and contributed to final
manuscript.
Signed………………………………………………………………..Date…………………………………………
Julius Mugini: Collected samples, conducted laboratory experiments, analysed and
interpreted results, critically interpreted data and contributed to final manuscript.
Signed………………………………………………………………..Date…………………………………………
117
Jason Geijskes: Supervised execution of the work, critically interpreted data and
contributed to final manuscript.
Signed………………………………………………………………..Date……………………………………………
Wilberforce Tushemereirwe: Conceived project idea, supervised execution of the
work, critically interpreted data and contributed to final manuscript.
Signed………………………………………………………………..Date……………………………………………
James Dale: Conceived project idea, collected samples, supervised execution of the
work, critically interpreted data and contributed to final manuscript.
Signed………………………………………………………………..Date……………………………………………
Rob Harding: Conceived project idea, supervised execution of the work, critically
interpreted data and contributed to final manuscript.
Signed………………………………………………………………..Date……………………………………………
118
Abstract
BSV diagnosis is complicated by significant serological and genomic variability
between isolates and the presence of endogenous virus sequences in the Musa
genome. We tested samples of banana from Uganda, Kenya and Tanzania for BSV
and compared the results obtained using PCR with species‐specific primers, and
rolling‐circle amplification. For BSV with no known endogenous counterpart, both
PCR and RCA were suitable methods for BSV detection, irrespective of the host
cultivar genotype. For four BSV with endogenous counterparts, namely Banana
streak OL virus (BSOLV), Banana streak MY virus (BSMYV), Banana streak GF virus
(BSGFV) and Banana streak IM virus (BSIMV), PCR and RCA were suitable in
cultivars with a pure Musa acuminata genetic background. However, in bananas
with a M. balbisiana (B‐genome) genetic background, many more samples were
positive using PCR compared to RCA, suggestive of positive results due to
integrated BSV sequences and not episomal virus infection. In these cases, RCA was
needed to diagnose BSV infection. For generic BSV indexing, RCA is the most
suitable method as it is able to detect a broad range of BSV species while avoiding
detection of integrated sequences. However, the faster and cheaper PCR serves as
a suitable alternative for BSV diagnostics in specific combinations of host plant
cultivar and BSV species.
119
Introduction
Bananas (Musa spp. (L)) are an important crop worldwide and are the largest fresh
fruit crop traded internationally in terms of both volume and value (FAO, 2011). In
many developing countries, including Uganda, Kenya and Tanzania, bananas also
serve as an important staple food and cash crop. Sustainable production, however,
is hindered by a range of different pathogens of which viruses are considered some
of the most important.
Banana streak disease (BSD), caused by a complex of Banana streak virus
(BSV) species in the genus Badnavirus, family Caulimoviridae, has been reported
from most banana‐producing countries in Africa, Latin America‐Caribbean and the
Asia‐Pacific region (Lockhart and Jones, 2000). Although yield losses of 6‐15% have
been associated with BSD (Dahal et al., 2000; Daniells et al., 2001), of greater
significance is the effect of BSV on Musa breeding programmes and the
international exchange of Musa germplasm (Geering, 2009; Iskra‐Caruana et al.,
2010).
There are currently four characterised BSV which are considered distinct
species, namely Banana streak OL virus (BSOLV), Banana streak GF virus (BSGFV),
Banana streak MY virus (BSMYV) and Banana streak VN virus (BSVNV, previously
Banana streak acuminata Vietnam virus) (Geering, 2010). However, six additional
distinct BSV species have recently been proposed based on analyses of full‐length
sequences (Geering et al., 2011; James et al., 2011b). Harper et al. (2005) reported
an additional nine distinct groups of BSV partial sequences from bananas in
Uganda, suggesting further diversity of BSV exists in this region.
The banana‐BSV pathosystem is complicated by the presence of
endogenous badnavirus sequences present in the Musa genome (Iskra‐Caruana et
al., 2010). The complete genome of several BSV, including BSOLV, BSGFV, BSMYV
and Banana streak IM virus (BSIMV), are present in Musa accessions with a genetic
background that includes a Musa balbisiana (B)‐genome. Further, the integrated
sequences of these four species, which are referred to as endogenous Banana
streak OL virus (eBSOLV), eBSMYV, eBSGFV and eBSIMV, can be activated under
120
stress conditions (eg. tissue culture) to cause episomal infections (Côte et al., 2010;
Dallot et al., 2001; Gayral et al., 2010; Geering et al., 2011; Ndowora et al., 1999).
In addition to these four integrants, Geering et al. (2005a) identified partial
sequences of 33 distinct badnavirus sequence groups in bananas free of episomal
virus, which shared less than 85% nucleotide identity to each other and many of
which were non‐functional.
Reliable diagnosis of BSV is complicated by the high genomic sequence
variability of individual isolates and the presence of eBSV sequences within some
Musa genomes. For example, the use of PCR to detect BSOLV, BSMYV and BSGFV in
Musa accessions with a B‐genome can result in false positives due to amplification
of eBSVs (Harper et al., 1999; James et al., 2011a; Le Provost et al., 2006). Although
PCR‐based methods which aim to avoid detection of eBSVs, such as immuno‐
capture PCR (IC‐PCR), have been reported (Harper et al., 1999; Le Provost et al.,
2006; Iskra‐Caruana et al., 2009), these methods suffer from the inability of
antiserum to capture all BSV isolates (Harper et al., 2002), the use of BSV species‐
specific primer sets which are unlikely to detect the entirety of the BSV sequence
diversity and the presence of contaminating, carry‐over nucleic acid remaining in
capture tubes leading to false positives (Le Provost et al., 2006; Iskra‐Caruana et
al., 2009). Further, despite attempts to improve the breadth of detection by using
degenerate primers a loss in sensitivity of detection and a failure to detect all
isolates was reported (Iskra‐Caruana et al., 2009).
Recently, we developed a sequence‐independent BSV detection method
based on rolling‐circle amplification (RCA) which was shown to detect a range of
episomal BSV species and not detect eBSVs (James et al., 2011a). Despite the
development of this improved diagnostic protocol, the expense of the RCA test
may preclude its widespread utility, particularly in resource‐poor African countries.
As such, in particular host cultivar‐virus species combinations, PCR may remain a
sensitive, specific and less‐costly option for diagnosing infection. The feasibility of
such an approach was recently demonstrated in a study involving the identification
and characterisation of BSV in East Africa (James et al., 2011b). In this study, six
BSV were characterised and it was shown, using virus‐free samples, that PCR could
121
be reliably used to detect five of these proposed BSV species. Furthermore, for
three of the six BSV, namely Banana streak UI virus (BSUIV), Banana streak UL virus
(BSULV) and Banana streak UM virus (BSUMV), no similar endogenous sequences
are known to occur in banana (Geering et al., 2005a; Gayral and Iskra‐Caruana,
2009), thus reinforcing the notion that a direct PCR approach may provide a
sensitive, low cost approach to diagnosis in some instances.
To further investigate the most appropriate and reliable diagnostic tests for
BSV detection, we assessed both RCA and PCR for their ability to detect BSV in field
samples of banana from Uganda, Kenya and Tanzania. In this paper, we show that
for several species of BSV, direct PCR is a suitable method to screen for virus
infection, irrespective of the host cultivar genotype while, as previously
demonstrated, detection of those species of BSV with endogenous counterparts in
cultivars with a B‐genome component requires a method such as RCA which can
distinguish between integrated and episomal virus DNA. The advantages and
limitations of PCR and RCA as molecular tests for BSV detection are also discussed.
Materials and Methods
Plant samples
Leaf samples used for this study were collected from banana plants during field
visits in Uganda (April 2008 and February 2010), Tanzania (October 2008) and
Kenya (March 2009) (Table 1 and Fig. 1). Fresh leaf tissue (10 cm2) was cut into
small pieces (0.5 cm2) and stored dried over silica gel until analysis. The identity
and genotype of banana cultivars (see supplementary Table 6) was determined
based on morphological characteristics by local agricultural services staff and
through discussion with farmers/agricultural officers at each site. Verification was
also attempted using the Musa genome information service (MGIS) website
(http://www.crop‐diversity.org/banana/#AccessionSearch). As a positive control
for BSV detection, a BSMYV‐infected banana plant was used (James et al., 2011a).
122
Table 1 Summary of plant samples collected in East Africa
Country District No. of field sites No. of samples
Uganda Rakai 2 8 Masaka 2 7 Ntungamo 1 5 Kyangara 1 5 Kamunyiga 1 4 Ibanda 3 11 Bushenyi 2 9 Rwimi 1 3 Bulera 1 4 Mbarara 1 20 Tanzania Kyela 11 10 Rungwe 13 16Kenya Nyamira 1 2 Kisii 7 15 Vihiga 3 3 Ibonda 1 3 Kakamega 6 10 Mumias 3 3 Bungoma West 4 4 Bungoma 1 1
123
Fig. 1 Maps of A) Uganda, B) Tanzania and C) Kenya indicating regions surveyed. Open circles indicate the location of districts in each country
where field visits were made. Numbers within circles indicate the number of samples collected in each district.
124
PCR and RT‐PCR assays
Total nucleic acid (TNA) was extracted from dried leaf tissue as described by James et al.
(2011a). Virus‐specific PCR assays were carried out for the detection of BSMYV, BSOLV,
BSGFV, BSIMV, Banana streak CA virus (BSCAV), Banana streak UA virus (BSUAV), BSUIV,
BSULV and BSUMV using species‐specific primers (Table 2). As an extraction control, a PCR
assay was initially done on TNA extracts using primers specific for the banana actin gene
(James et al., 2011a) or 18S ribosomal RNA gene using primers Musa18Sf (5’
CATCACAGGATTTCGGTCCT 3’) and Musa18Sr (5’ AGACAAATCGCTCCACCAAC 3’) which
amplify a 500 bp product. PCR mixes (20 µL) contained 10 µL 2x GoTaq Green Master Mix
(Promega Corp, Madison, WI), 5 ρmol of each primer, 1 µL of nucleic acid extract (diluted to
10 ng/µL) and water to final volume. PCR cycling conditions were an initial denaturation of
94°C for 2 min followed by 35 cycles of 94°C for 20 s, annealing for 20 s and 72°C for 30 s,
with a final extension at 72°C for 2 min. Primer annealing temperatures were 57°C (Actin) or
50°C (18S rRNA) while those for BSV were as listed in Table 2.
Reaction products were electrophoresed through 1.5% agarose gels, stained using
0.25x SYBR® Safe DNA Gel Stain (Invitrogen Corp, Carlsbad, CA) and DNA fragments
visualised on a Safe imager blue‐light transilluminator.
Rolling‐circle amplification (RCA)
RCA was performed using the protocol described in James et al. (2011a). RCA reactions were
digested with the restriction enzymes StuI or PstI and the products visualised on 1% agarose
gels. RFLP‐based identification of BSV from RCA digests was based on the characteristic
patterns listed in Table 3.
125
Table 2 PCR primers used for detection of BSV speciesa
BSV Primer 1 (coding strand 5´ ‐ 3´) Primer 2 (complementary strand 5´ ‐ 3´)
Amplicon size (bp)
Annealing temp (°C)
BSOLV ATCTGAAGGTGTGTTGATCAATGC GCTCACTCCGCATCTTATCAGTC 522 57 BSGFV ACGAACTATCACGACTTGTTCAAGC TCGGTGGAATAGTCCTGAGTCTTC 476 57BSMYV TAAAAGCACAGCTCAGAACAAACC CTCCGTGATTTCTTCGTGGTC 589 57BSCAV AGGATTGGATGTGAAGTTTGAGC ACCAATAATGCAAGGGACGC 782 57BSULV GAACTGACAGTAGCGCAATCG GACTTGGCTTGCCTGAGTATCG 943 60BSUIV GAATCCTCAAAGGTACCCC CATGAGGTCAAGCATATGC 619 50BSUMV GACGAGCTGCAAGCTCTCAGG TGTGCCTATTCTGAGGTTGG 467 50BSUAV CTCAGCGGCAAGATTAGGAAGG TCCCCATTGGTCGTCATTGC 517 60BSIMV GCTAGGAAGAAAAGTCTGGG TGCAAGTCTACTTACACAGC 475 50a Primer sequences derived from Geering et al. (2000) and James et al. (2011b)
126
Table 3 Predicted restriction profiles of genomic DNA from characterised BSV using PstI and StuI
Restriction enzyme
Predicted sizes (bp) of restriction fragments
BSMYV BSOLV BSGFV BSIMV BSCAV BSUAV BSUIV BSULV BSUMV
PstI
6881, 769
2635, 2442, 2312
7218, 45
7769 Does not cut
4490, 3029
6989, 407, 62
4746, 2655
5005, 2527
StuI 7650 3111, 2436, 1842
6034, 1229
4389, 2407, 973
7408 7519 5532, 1936
7401 3726, 2239, 1567
127
Results
A total of 143 samples were collected (Table 1) and tested for BSV using PCR and
RCA (see supplementary Table 6). Of these, 95 were recorded as having only an A‐
genome component, while 46 were recorded as possessing some B‐genome
component, with two samples (Ug233/234) collected from genotypes not derived
from either Musa acuminata or M. balbisiana.
Using the nine sets of BSV species‐specific primers, 44 samples were not
considered positive for any BSV tested by PCR. Of these, 43 samples were recorded
as having no B‐genome component. One sample containing a B‐genome
component was PCR negative for all nine primer sets (Ke166); given this cultivar
was recorded as Sukali Ndizi (AAB genome) and all other samples from this cultivar
(n=14) were PCR positive it is likely this sample was misidentified.
Detection of BSV with no known endogenous equivalents
Of the nine BSV species assayed using PCR, five (BSCAV, BSUAV, BSUIV, BSULV and
BSUMV) do not have a confirmed endogenous equivalent. When the 143 samples
were tested for these five viruses by PCR, two plants with AAA genotypes
(Ke171/172) were positive for BSCAV. Both of these samples, and no others, were
confirmed to contain BSCAV by RCA/sequencing. When tested for BSUAV, four
samples were PCR positive, including three (Ke155/156/177) from cultivars with an
A‐only genotype, and one (Ke169) from cultivar Sukali Ndizi (AAB genotype).
RCA/sequencing confirmed BSUAV in the three A‐only cultivar samples but sample
Ke169 was considered RCA negative.
Six samples, all from cultivars with AAA genotypes, tested positive for
BSUIV by PCR. Five of these six samples also tested positive for BSV by RCA, with
the virus status of one sample (Ug42) inconclusive. Two of the RCA positive
samples (Ug8/11) had the predicted BSUIV digest pattern (Table 3) and the
presence of this virus was confirmed by sequencing. Although the restriction
profiles of the RCA products from the remaining three samples were not typical of
128
BSUIV, both BSULV and BSUMV were detected in samples Ug5 and Ug12 (by PCR
and RCA/RFLP), respectively, while Ug15 had a full‐length RCA product and was
also PCR positive for BSOLV.
PCR screening of the 143 samples identified BSULV in five samples
(Ug1/2/5/7/10), all of which possessed AAA genotypes. Four of these five samples
were confirmed to contain BSULV by RCA using RFLP analysis and/or sequencing,
while the virus status in sample Ug7 was inconclusive.
When the 143 samples were screened for BSUMV using PCR, 15 positive
samples were identified all of which were AAA genotypes. All samples were
confirmed to be infected with BSUMV by RCA based on the distinct RFLP profile
predicted for this virus (Table 3). One additional sample (Ug28; AAB genome) also
tested positive for BSUMV by RCA.
Of the 30 samples which where PCR positive for one or more of the five BSV
with no integrated counterparts, 96.7% (or 29 samples) were from cultivars with
no B‐genome component (Table 4). One sample (Ug28) with a B‐genome
component was RCA positive for BSUMV, despite not testing positive by PCR, giving
a total of 31 samples testing positive by both PCR and RCA. When the results of
BSV detection by PCR and RCA were compared, there was inconsistency in only
two samples (Ke169 and Ug28). Interestingly, sample Ke169 was asymptomatic
(PCR positive/RCA negative), while Ug28 had a mild mosaic symptom (PCR
negative/RCA positive). RCA confirmed a positive diagnosis by PCR in 96.4% of
samples (Table 4), with at least one BSV species identified by PCR confirmed using
RCA, except sample Ug15 where sequencing was not completed.
129
Table 4 PCR and RCA detection of BSV species with no confirmed eBSV
BSV sp. No. of samples positive by PCR No. of PCR positive samples also positive by RCA (% of PCR positive samples)a
Country where detected
No. of sites where detected
A‐only genome (% of total positives)
Some B‐genome(%of total positives)
Total
BSCAV 2 (100%) 0 2 2 (100%) Kenya 1 BSUAV 3 (75%) 1 (25%) 4 3 (75%) Kenya 3 BSUIV 6 (100%) 0 6 2 (40%)
5 (83.3%) Uganda 3
BSULV 5 (100%) 0 5 4 (100%) Uganda 1 BSUMV 15 (100%) 0 15 15b (100%) Uganda 6 TOTAL 29 (96.7%) 1 (3.3%) 30c 27b (96.4%) n/a n/a a Percent detection does not include samples where the result was “inconclusive”. Text formatted in italics indicates all samples
considered RCA positive from the specific subset of PCR positive samples, despite the specific BSV not identified based on results of
RFLP analysis or sequencing. b One additional RCA positive sample was not detected using PCR c Multiple infections detected in two samples
130
Detection of BSV with known endogenous equivalents
Four of the nine BSV species assayed using PCR, namely BSIMV, BSGFV, BSOLV and
BSMYV, have integrated counterparts (eBSV) in bananas with a B‐genome
component. When the 143 samples were tested for these four viruses by PCR,
BSIMV was detected in eight samples. Six of these were in cultivars which
contained a B‐genome, while the remaining two samples (Ug29/42) were from
bananas with AAA genotypes. When the two A‐only genome samples were tested
by RCA, BSIMV was not clearly identified in either sample following RFLP analysis.
However sample Ug29 was shown to be infected with BSUMV (by RCA/RFLP and
PCR) while the virus status of sample Ug42 was inconclusive. Of the six B‐genome
containing samples tested by RCA, the presence of both BSIMV and BSGFV was
confirmed in Ug250 by sequencing, BSGFV only was identified in Ug244 by
sequencing, while sample Ug32 had an RFLP profile consistent with BSIMV. The
remaining three samples (Ug51/241 and Tz134) were considered RCA negative.
BSMYV was detected in 44/143 samples using PCR, only one of which
(Ke176) was an A‐only genome cultivar. BSOLV and BSGFV were also detected in
this sample by PCR, and the presence of both BSMYV and BSGFV was confirmed by
RCA/sequencing. In the remaining 43 samples from cultivars which contained a B‐
genome component, eight samples were found to be solely infected with BSMYV
using PCR and, in all cases, this result was confirmed by RCA/sequencing The
remaining 35 samples tested PCR positive for at least one other BSV with 34, 20,
five and one samples testing positive for BSOLV, BSGFV, BSIMV and BSUAV,
respectively. When these 35 samples were tested by RCA, BSV was detected in 22
samples, with one failed reaction (Tz114) and the remaining 12 samples considered
RCA negative. Interestingly, BSMYV was not identified by either RFLP or sequencing
in any of these 22 B‐genome samples, despite dual infections confirmed in six
samples and single infections confirmed by RFLP and/or sequencing in 15 samples
(with sample Ug32 not sequenced).
A total of 27/143 samples tested positive for BSGFV by PCR, of which six
were in cultivars with an A‐only genome. Of these six samples, five were also PCR
131
positive for BSOLV and one of these five (Ke176) was also positive for BSMYV.
When tested by RCA, the presence of BSGFV was confirmed in all six samples by
sequencing, while BSMYV was identified in sample Ke176. Based on RFLP profiles
and sequencing, BSOLV was only identified in two of the six samples. Of the 21
samples containing a B‐genome component that were PCR positive for BSGFV, 20
were also PCR positive for both BSOLV and BSMYV, suggesting that multiple PCR
positives are common to field samples from both A and B‐genome cultivars using
these three primer sets. However, in contrast to A‐only genome cultivars, only 13
B‐genome samples with BSGFV were confirmed using RCA (by RFLP and/or
sequencing), with the remaining eight samples either negative by RCA (six samples)
or RCA positive for BSOLV only (Ug243/245). No samples (either A or B genome
groups) which tested negative for BSGFV by PCR were positive for BSGFV by RCA
analysis.
PCR screening for BSOLV identified 63 positive samples (27 with an A‐only
genotype and 36 with some B‐genome component). When the 27 samples from
the cultivars with A‐only genotypes were tested by RCA, 24 were considered
positive for BSV with BSOLV identified in eight of these by RFLP (seven samples) or
sequencing (one sample). Of the three samples which were not considered positive
by RCA, one failed (Tz112), one was RCA negative (Ug9) and the digest results in
one sample were inconclusive (Ug30). In contrast to PCR results for BSGFV, of the
27 A‐only samples which were BSOLV positive, only five were also PCR‐positive for
BSGFV and the same single co‐positive BSMYV sample (Ke176) was identified. Of
the 36 B‐genome samples which were PCR positive for BSOLV, 34 were also PCR
positive for BSMYV and 19 were PCR positive for BSGFV. Interestingly, of the 16
samples which tested PCR positive for BSOLV but not BSGFV, 15 were of the apple
banana cultivar (Sukali Ndizi). Of these 16 samples, 15 were also PCR positive for
BSMYV. When the 36 samples were tested using RCA, 21 were considered BSV
positive, with BSOLV identified in 12 samples by RFLP. Of the 15 samples not
considered BSV positive, 14 were considered RCA negative while one reaction
failed (Tz114).
132
PCR diagnosis identified 75 samples which were positive for one or more of
the four BSV species with integrated counterparts. Of these, 60% were from
cultivars with some B‐genome component (Table 5). Interestingly, PCR detection of
BSMYV was almost exclusively in B‐genome cultivars (97.8%), with detection of
BSGFV (77.8%), BSIMV (75%), and BSOLV (57.1%) all considerably higher in B‐
genome cultivars compared with BSV with no endogenous equivalent (Table 4 & 5).
RCA based screening identified 57 positives out of the 75 (or 76%) which tested
positive by PCR, although two samples (Tz112/114) failed and the results for two
samples (Ug30/42) were inconclusive. The remaining 14 samples were considered
negative by RCA. Of the 30 samples from A‐genome cultivars, only one sample
(Ug9) was considered positive by PCR but not by RCA. This detection rate of 96.3%
(Table 5) is consistent with those samples infected with BSV only known to have
episomal forms. Similarly, a single sample (Tz119) was RCA/RFLP positive for
BSOLV, but PCR negative, maintaining consistency in “misses” between assay
formats. In contrast, in samples from cultivars with a B‐genome, only 70.5% of PCR
positive samples were diagnosed positive using RCA (Table 5).
Furthermore, in A‐genome samples positive for one or more of these four
species 11 had dual infections and a single sample contained three viruses, based
on PCR analysis. In contrast, when B‐genome samples were tested by PCR, 2 had
quadruple infections, 22 had triple infections and 12 had dual infections, with only
nine samples a single infection. This high level of detection of multiple infections in
B‐genome cultivar samples probably reflects the detection of eBSV in some, if not
all, of these samples when using PCR.
Samples with unidentified BSV
Based on field observations (see supplementary Table 6), 15 samples which tested
negative by PCR displayed characteristic streak symptoms. Four of these
(Ug33/34/38/Ke164) had full‐length digest profiles in RCA (none of which were
133
Table 5 PCR and RCA detection of BSV species with confirmed eBSV
BSV sp. No. of samples positive by PCR No. of PCR positive samples also positive by RCA (% of PCR positive samples)a, b
Country where detectedc
No. of districts where detectedc
A‐only genome (% of total positives)
Some B‐genome (% of total positives) Total
A‐only genome
Some B‐genome Total
BSIMV 2 (25%) 6 (75%) 8 0
1 (100%)a 2 (33.3%)3 (50%)
2 (33.3%)4 (57%)a
Uganda 3
BSGFV 6 (22.2%) 21 (77.8%) 27 6 (100%) 13 (62%) 19 (70%) Kenya
Uganda 4 3
BSMYV 1 (2.2%) 43 (97.8%) 44 1 (100%) 8 (19%)
30 (71.4%) 9 (21%) 31 (72%)
Kenya Uganda
5 1
BSOLV 27 (42.9%) 36 (57.1%) 63 8 (32%)
24 (96%) 12(34.3%)21 (60%)
20/60 (33.3%) 45/60 (75%)
KenyaTanzania Uganda
22 4
TOTAL 30 (40%) 45 (60%) 75 26d (96.3%) 31 (70.5%) 57 (76%) n/a n/aaPercent detection does not include samples where the result was “inconclusive”. b Text formatted in italics indicates all samples considered RCA positive from the specific subset of PCR positive samples, despite the
specific BSV not identified based on results of RFLP analysis or sequencing. cConfirmed by RFLP or sequencing in B‐genome samples dOne additional RCA positive sample was not detected using PCR
134
cloned), while for three samples (Ug37/39/40) the RCA digest was inconclusive.
However, the remaining eight samples with streak symptoms were all RCA
negative. One remaining sample (Ug44), with abnormal bunch symptoms, was also
considered RCA positive with a distinct RFLP pattern.
Discussion
In this study, we collected leaf samples from bananas from several districts in
Uganda, Tanzania and Kenya and tested them for the presence of BSV using PCR
and RCA. Symptoms of streak disease were widespread in south‐west Uganda and
western Kenya, consistent with previous studies (Harper et al., 2004, 2005; Karanja
et al., 2008). In Tanzania, BSD was also observed at several locations although the
incidence was considerably lower than that seen in Uganda and Kenya.
For five of the nine BSV studied, no known eBSV are reported. When
screening for these five BSV, PCR and RCA were equally efficacious at detecting BSV
(Table 4). In some cases, multiple infections comprising these five BSV were
identified using PCR but not by RCA (samples Ug5/12). Similarly, PCR identified a
further eight samples with multiple infections comprising one of these five BSV
with one of the four BSV having an integrated counterpart. In contrast, RCA only
identified one of these samples (Ug17) to contain a dual infection based on RFLP.
In light of these results, PCR could be considered a better method to diagnose
infection.
The remaining four BSV studied (BSIMV, BSOLV, BSGFV and BSMYV) all have
characterised eBSV present in bananas with a B‐genome component, which are
identical to episomal sequences which infect banana (Geering et al., 2001 &
2005a/b; Gayral et al., 2008; Iskra‐Caruana et al., 2010). Diagnosis of these four
BSV species using PCR is complicated by eBSV sequences which give rise to positive
results which do not reflect the presence of episomal virus infection (Harper et al.,
1999; James et al., 2011a; Le Provost et al., 2006). When samples from A‐only
genome cultivars were tested for BSIMV, BSOLV, BSGFV and BSMYV by PCR, there
135
was 96.3% agreement in the detection of infected samples between PCR and RCA
(Table 5). However, in samples collected from cultivars with some B‐genome
component, only 70.5% of samples which tested positive for one or more of these
four BSV using PCR were also considered positive using RCA. This level of
agreement is considerably lower than that for these four BSV in A‐only cultivars, or
for the five BSV without integrated counterparts in bananas of any genetic
background, where consistency in detection levels between PCR and RCA was in
the order of 96% (Table 4/5).
In the present study, a distinct cultivar preference of BSV detection was
evident. For the five viruses with no identified eBSV, 96.7% of PCR positive samples
were from cultivars with an A‐only genotype and all but one of these were also
positive using RCA (Table 4). Four of the five BSV in this group were exclusively
detected in A‐only genome cultivars. Whether this is a reflection on farmer cultivar
preference in East Africa, or indicative of the slow nature of field spread of BSV
remains to be confirmed. In contrast, 40% of PCR positive samples for the four BSV
with integrated counterparts were from A‐only genome cultivars, again all but one
were positive by RCA. Of the nine BSV studied, BSOLV was the most prevalent virus
detected with 27 PCR‐positive in A‐only genome samples and 12 confirmed
positive in B‐genome samples. The next most prevalent BSV species’ were BSGFV
(19 confirmed), BSUMV (15 confirmed), BSMYV (10 confirmed) and smaller
numbers of the other species.
In a number of samples collected from all three countries, symptoms of
streak disease were recorded but either no virus was detected using either PCR or
RCA (Ug36, Tz129/130, Ke153/160/161/168/174), or samples were considered RCA
positive and PCR negative but for which cloning was unsuccessful (Ug33/34/38,
Ke164). These samples represent an interesting finding and suggest that further
BSV species might be present in the field in East Africa, such as BSVNV or one of
the nine additional Ugandan isolates reported by Harper et al. (2005).
For several samples, the presence of multiple bands or faint bands in RFLP
analyses of RCA products hindered the interpretation. Although repeat testing is
136
necessary in these cases in order to make a diagnosis, this is not always possible
when small samples are collected at distant field sites on surveys. Optimisation of
reaction parameters, including template quality and quantity, as well as
modifications to the reagents present in the reaction mixture (degenerate primer
mixture/random hexamer primers) may improve the robustness of the RCA assay.
Alternatively, the choice of restriction enzyme may prove critical in reducing the
number of fragments produced in digests. Failure of RCA to detect BSV in
asymptomatic samples where PCR was successful may reflect the difference in
sensitivity of the two methods, or the detection of eBSV in cultivars with a B‐
genome.
Both PCR and RCA are useful techniques for BSV detection, each with
distinct advantages and limitations (Table 6). The choice of test, however, should
be informed by the intended purpose of the testing. In virus‐indexing schemes, for
example, where virus detection, and not characterisation, is of paramount
importance, RCA should be the preferred choice based on the broad range of
detection, ability to specifically amplify episomal DNA, and capacity to detect
uncharacterised BSV species (James et al., 2011a). In contrast, in situations such as
disease surveys, the identification of all BSVs present in a sample is required. For A‐
genome cultivars or BSV species without integrated equivalents, PCR can satisfy
this requirement quickly and cost‐effectively. In all other cases, it would be
necessary to use RCA to identify and characterise BSV either by their distinct RFLP
patterns, or by cloning and sequencing of digest fragments.
Acknowledgements
This research was funded by the Bill and Melinda Gates Foundation Grand
Challenges in Global Health Program. We are grateful to project staff that assisted
with field visits as well as farmers in all three countries for their permission to visit
and collect samples.
137
Table 6 Advantages and disadvantages of RCA and PCR for BSV detection
Advantages Disadvantages
RCA Broad spectrum Avoids integrated sequences Uncharacterised BSV
Cost Inconclusive results
PCR Cost Identification
Narrow spectrum Detects integrated sequences Sequence variants Uncharacterised BSV
138
References
Côte, F.X., Galzi, S., Folliot, M., Lamagnère, Y., Teycheney, P‐Y., Iskra‐Caruana, M‐L.
2010. Micropropagation by tissue culture triggers differential expression of
infectious endogenous Banana streak virus sequences (eBSV) present in the
B genome of natural and synthetic interspecific banana plantains. Mol.
Plant Pathol. 11, 137–144.
Dahal, G., Ortiz, R., Tenkouano, A., Hughes, J. d’A., Thottappilly, G., Vuylsteke, D.,
Lockhart, B.E.L. 2000. Relationship between natural occurrence of banana
streak badnavirus and symptom expression, relative concentration of viral
antigen, and yield characteristics of some micropropagated Musa spp. Plant
Pathol. 49, 68–79.
Dallot, S., Acuña, P., Rivera, C., Ramírez, P., Côte, F., Lockhart, B.E.L., Caruana, M‐L.
2001. Evidence that the proliferation stage of micropropagation procedure
is determinant in the expression of Banana streak virus integrated into the
genome of the FHIA 21 hybrid (Musa AAAB). Arch. Virol. 146, 2179–2190.
Daniells, J.W., Geering, A.D.W., Bryde, N.J., and Thomas, J.E. 2001. The effect of
Banana streak virus on the growth and yield of dessert bananas in tropical
Australia. Ann. Appl. Biol. 139, 51–60.
FAO. 2011. The World Banana Forum (WBF): Working together for sustainable
banana production and trade. Introductory note. Accessed 23/2/2011
www.fao.org/economic/worldbananaforum.
Gayral, P., Blondin, L., Guidolin, O., Carreel, F., Hippolyte, I., Perrier, X., Iskra‐
Caruana, M‐L 2010. Evolution of endogenous sequences of Banana streak
virus: What can we learn from banana (Musa sp.) evolution? J. Virol. 84,
7346–7359.
Gayral, P., Iskra‐Caruana, M‐L. 2009. Phylogeny of Banana streak virus reveals
recent and repetitive endogenization in the genome of its banana host (Musa
sp.). J. Mol. Evol. 69, 65–80.
139
Gayral, P., Noa‐Carrazana, J‐C., Lescot, M., Lheureux, F., Lockhart, B.E.L.,
Matsumoto, T., Piffanelli, P., Iskra‐Caruana, M‐L. 2008. A single Banana
streak virus integration event in the banana genome as the origin of
infectious endogenous pararetrovirus. J.Virol. 82, 6697–6710.
Geering, A.D.W. 2009. The distribution of badnavirus‐infected Musa germplasm
from the international transit centre, Katholieke Universiteit Leuven,
Belgium. In International Symposium on Recent Advances in Banana Crop
Protection for Sustainable Production and Improved Livelihoods, pp85–88.
Eds D. Jones and I. Van den Bergh: ISHS.
Geering A.D.W. 2010. Five new species and four nomenclature changes in the
genus Badnavirus. [online] ICTV Taxonomy Proposals – Plant. URL
http://talk.
ictvonline.org/files/proposals/taxonomy_proposals_plant1/m/plant01/221
6.asp [accessed 14 December 2010].
Geering, A.W.D., McMichael, L.A., Dietzgen, R.G., and Thomas, J.E. 2000. Genetic
diversity among Banana streak virus isolates from Australia. Phytopathology
90, 921–927.
Geering, A.D.W., Olszewski, N.E., Dahal, G., Thomas, J.E., Lockhart, B.E.L. 2001.
Analysis of the distribution and structure of integrated Banana streak virus
DNA in a range of Musa cultivars. Mol. Plant Pathol. 2, 207–213.
Geering, A.D.W., Olszewski, N.E., Harper, G., Lockhart, B.E.L., Hull, R., Thomas, J.E.
2005a. Banana contains a diverse array of endogenous badnaviruses. J.
Gen. Virol. 86, 511–520.
Geering, A.D.W., Parry, J.N., Thomas, J.E. 2011. Complete genome sequence of a
novel badnavirus, banana streak IM virus. Arch. Virol. 156, 733–737.
Geering, A. D. W., Pooggin, M.M., Olszewski, N.E., Lockhart, B.E.L., Thomas, J.E.
2005b. Characterisation of banana streak Mysore virus and evidence that
140
its DNA is integrated in the B genome of cultivated Musa. Arch. Virol. 150,
787–796.
Harper, G., Ganesh, D., Thottappilly, G., Hull, R. 1999. Detection of episomal Banana
streak virus by IC‐PCR. J. Virol. Methods 79, 1–8.
Harper, G., Hart, D., Moult, S., Hull, R. 2002. Detection of Banana streak virus in field
samples of banana from Uganda. Ann. Appl. Biol. 141, 247–257.
Harper, G., Hart, D., Moult, S., Hull, R. 2004. Banana streak virus is very diverse in
Uganda. Virus Res. 100, 51–56.
Harper, G., Hart, D., Moult, S., Hull, R., Geering, A., Thomas, J. 2005. The diversity
of Banana streak virus isolates in Uganda. Arch. Virol. 150, 2407–2420.
Iskra‐Caruana, M.L., Gayral, P., Galzi, S., Laboureau, N. 2009. How to control and
prevent the spread of banana streak disease when the origin could be viral
sequences integrated in the banana genome? In International Symposium
on Recent Advances in Banana Crop Protection for Sustainable Production
and Improved Livelihoods, pp. 77–84. Eds D. Jones and I. Van den Bergh:
ISHS.
Iskra‐Caruana, M‐L., Baurens, F‐C., Gayral, P., Chabannes, M. 2010. A four‐partner
plant‐virus interaction: Enemies can also come from within. MPMI 23,
1394–1402.
James, A.P., Geijskes, R.J., Dale, J.L., Harding, R.M. 2011a. Development of a novel
rolling‐circle amplification technique to detect Banana streak virus which
also discriminates between integrated and episomal virus sequences. Plant
Dis. 95, 57–62.
James, A.P., Geijskes, R.J., Dale, J.L., Harding, R.M. 2011b. Molecular
characterisation of six badnavirus species associated with leaf streak
disease of banana in East Africa. Ann. Appl. Biol. 158, 346–353.
141
Karanja, L., Wangai, A., Harper, G., Pathak, R.S. 2008. Molecular identification of
Banana streak virus isolates in Kenya. J. Phytopathol. 156, 678–686.
Le Provost, G., Iskra‐Caruana, M‐L., Acina, I., Teycheney, P‐Y. 2006. Improved
detection of episomal Banana streak viruses by multiplex immunocapture
PCR. J. Virol. Methods 137, 7–13.
Lockhart, B.E.L., Jones, D.R. 2000. Banana streak. In Diseases of banana, abaca and
enset, pp. 263–274. Ed D.R. Jones. CABI: United Kingdom.
Ndowora, T., Dahal, G., LaFleur, D., Harper, H., Hull, R., Olszewski, N.E., Lockhart, B.
1999. Evidence that badnavirus infection in Musa can originate from
integrated pararetroviral sequences. Virology 255, 214–220.
142
Supplementary data ‐ Table 7Plant samples and analysis for BSV using PCR and RCA. Sample District Cultivar Genome
typeaSymptoms BSV PCR assaysb RCAc,d Sequence
ID
MY OL GF CA IM UA UI UL UM
Uganda, April 2008 Ug1 Rakai Kibuzi AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ + ‐ UL ULUg2 Rakai Kibuzi AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ + ‐ UL ULUg5 Rakai Mbwazirume AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ + + ‐ ULUg7 Rakai Nakinyika AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ + ‐ ?Ug8 Rakai Kisansa AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ + ‐ ‐ UI UIUg9 Rakai Kisansa AAA (EAHB) Asymptomatic ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ug10 Rakai Kisansa AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ ‐ + ‐ UL ULUg11 Rakai Nakitembe AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ + ‐ ‐ UI UIUg12 Masaka Mbwazirume AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ + ‐ + UM UMUg13 Masaka Mbwazirume AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ + UM UMUg14 Masaka Sukali Ndizi AAB Streak + + ‐ ‐ ‐ ‐ ‐ ‐ ‐ OL OLUg15 Masaka Nakinyika AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ + ‐ ‐ FLUg16 Masaka Gonja AAB Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ GF GFUg17 Masaka Kibuzi AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ + OL/UMUg18 Masaka Uncertain
Plantain AAB Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ GF
Ug19 Ntungamo Enyeru AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ + UMUg20 Ntungamo Mbwazirume AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ + UMUg21 Ntungamo Mbwazirume AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ + UMUg22 Ntungamo Kibuzi AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ + UMUg23 Ntungamo Enzirabushera AAA (EAHB) Mild flecks ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ug24 Kyangara Entaragaze AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ + UMUg25 Kyangara Enyeru AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ + UM Ug26 Kyangara Enyeru AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ + UMUg27 Kyangara Embire AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ + UMUg28 Kyangara Sukali Ndizi AAB Mild mosaic + + ‐ ‐ ‐ ‐ ‐ ‐ ‐ UMUg29 Kamunyiga Enyeru AAA (EAHB) Streak ‐ ‐ ‐ ‐ + ‐ ‐ ‐ + UMUg30 Kamunyiga Kibuzi AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ? Ug31 Kamunyiga Enyeru AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ + UMUg32 Kamunyiga Sukali Ndizi AAB Streak + + ‐ ‐ + ‐ ‐ ‐ ‐ + Ug33 Ibanda Enjuma AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ FLUg34 Ibanda Enkonga AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ FLUg35 Ibanda Sukali Ndizi AAB Streak + + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ug36 Ibanda Enjuma AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ug37 Ibanda Embururu AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ? Ug38 Ibanda Enjuma AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ FLUg39 Ibanda Gros Michel AAA Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ? Ug40 Ibanda Kibuzi AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ?Ug41 Ibanda Enjuma AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ + UMUg42 Ibanda Cavendish AAA Streak ‐ ‐ ‐ ‐ + ‐ + ‐ ‐ ?
143
Ug43 Ibanda FHIA‐17 AAAA Mild flecks ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ?Ug44 Bushenyi Enyeru AAA (EAHB) Abnormal
bunch‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ +
Ug45 Bushenyi Entaziduka AAA (EAHB) Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ug46 Bushenyi Enyambururu AAA (EAHB) Streak ‐ + + ‐ ‐ ‐ ‐ ‐ ‐ FL GFUg47 Bushenyi Endyabwali AAA (EAHB) Streak ‐ + + ‐ ‐ ‐ ‐ ‐ ‐ OL/FL GFUg48 Bushenyi Kisubi AB Asymptomatic + + + ‐ ‐ ‐ ‐ ‐ ‐ ‐Ug49 Bushenyi Enyeru AAA (EAHB) Streak ‐ + + ‐ ‐ ‐ ‐ ‐ ‐ OL/FL GFUg50 Bushenyi Enkara AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ +Ug51 Bushenyi Sukali Ndizi AAB Streak + + ‐ ‐ + ‐ ‐ ‐ ‐ ‐Ug52 Bushenyi Enjagata AAA (EAHB) Streak ‐ + + ‐ ‐ ‐ ‐ ‐ ‐ + GFUg53 Rwimi Nakyetengu AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ FLUg54 Rwimi Nakyetengu AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ FLUg55 Rwimi Mutule AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ + UMUg57 Bulera Kibuzi AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ FLUg58 Bulera Mbwazirume AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ FLUg59 Bulera Mbwazirume AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ FLUg60 Bulera Sukali Ndizi AAB Streak + + ‐ ‐ ‐ ‐ ‐ ‐ ‐ FLTanzania, October 2008 Tz105 Kyela Cavendish AAA Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ OLTz106 Kyela Cavendish AAA Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ +Tz107 Kyela Cavendish AAA Leaf chlorosis ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Tz108 Kyela Cavendish AAA Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ +Tz111 Kyela Mzuzu AAB Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ ‐Tz112 Kyela Cavendish AAA Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ failedTz113 Kyela Cavendish AAA Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ failed Tz114 Kyela Zambia AAB Leaf chlorosis + + ‐ ‐ ‐ ‐ ‐ ‐ ‐ failedTz117 Kyela Sukali Ndizi AAB Leaf chlorosis + + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Tz118 Kyela Cavendish AAA Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ FLTz119 Rungwe Cavendish AAA Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ OLTz120 Rungwe Cavendish AAA Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ Tz121 Rungwe Grand Naine AAA Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Tz122 Rungwe Grand Naine AAA Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ Tz123 Rungwe Chinese
Cavendish AAA Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ failed
Tz124 Rungwe Chinese Cavendish
AAA Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐
Tz125 Rungwe Sukali Ndizi AAB Streak + + ‐ ‐ ‐ ‐ ‐ ‐ ‐ OL Tz126 Rungwe FHIA‐23 AAAA Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Tz127 Rungwe FHIA‐17 AAAA Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ Tz128 Rungwe Yangambi Km5 AAA Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ?Tz129 Rungwe Cavendish AAA Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Tz130 Rungwe Cavendish AAA Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ Tz131 Rungwe Sukali Ndizi AAB Leaf chlorosis + + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Tz134 Rungwe Mitalula AAB Mild
mosaic/mottle+ + + ‐ + ‐ ‐ ‐ ‐ ‐
Tz136 Rungwe Uganda AAA (EAHB) Streak or ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐
144
SigatokaTz137 Rungwe Mzuzu AAB Chlorotic flecks + + + ‐ ‐ ‐ ‐ ‐ ‐ ‐Kenya, April 2009 Ke146 Nyamira Sukali Ndizi AAB Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ FL GFKe147 Nyamira Uganda Green AAA (EAHB) Veinal chlorosis ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ke148 Kisii Sukali Ndizi AAB Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ FL GFKe149 Kisii Mysore AAB Streak + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ MY MYKe150 Kisii Mysore AAB Streak + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ MY MYKe151 Kisii Muraru AA Die‐back ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ke152 Kisii Kaburut AAA (EAHB) Mild mosaic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ke153 Kisii Jamaga AAA Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ke154 Kisii Lisulya AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ + OLKe155 Kisii Uganda green AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ + ‐ ‐ ‐ + UAKe156 Kisii PAZ‐Giant AAA Streak ‐ + ‐ ‐ ‐ + ‐ ‐ ‐ + UAKe158 Kisii Chinese
Cavendish AAA Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐
Ke159 Kisii Sukali Ndizi AAB Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ OLKe160 Kisii Uganda green AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ke161 Kisii Uganda green AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ Ke162 Kisii Valary AAA Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ke163 Kisii Uganda green AAA (EAHB) Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ OLKe164 Vihiga Uganda green AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ FLKe165 Vihiga Cavendish AAA Chlorotic
blotches‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐
Ke166 Vihiga Sukali Ndizi AAB Chlorotic blotches
‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐
Ke167 Ibonda Sukali Ndizi AAB Streak + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ MY MYKe168 Kakamega Uganda green AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ke169 Kakamega Sukali Ndizi AAB Asymptomatic + + ‐ ‐ ‐ + ‐ ‐ ‐ ‐Ke170 Kakamega Shisikame AAA (EAHB) Mosaic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ke171 Kakamega Uncertain EAHB AAA (EAHB) Streak ‐ ‐ ‐ + ‐ ‐ ‐ ‐ ‐ FL CA Ke172 Kakamega Esianamuni AAA (EAHB) Streak ‐ ‐ ‐ + ‐ ‐ ‐ ‐ ‐ FL CAKe173 Kakamega Mysore AAB Streak + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ MY MY Ke174 Kakamega Yangambi Km5 AAA Streak ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ke175 Kakamega Sianamule ABB Streak + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ MY MYKe176 Kakamega Cavendish AAA Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ MY MY&GFKe177 Kakamega Uganda Red AAA (EAHB) Streak ‐ ‐ ‐ ‐ ‐ + ‐ ‐ ‐ FL UAKe178 Kakamega Sukali Ndizi AAB Streak + + ‐ ‐ ‐ ‐ ‐ ‐ ‐ OL OL Ke179 Kakamega Sukali Ndizi AAB Streak + + ‐ ‐ ‐ ‐ ‐ ‐ ‐ OLKe180 Mumias Sukali Ndizi AAB Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ FL GF Ke181 Mumias Sukali Ndizi AAB Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ FL GFKe182 Mumias Sukali Ndizi AAB Streak + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ MY MYKe183 Bungoma
West Sukali Ndizi AAB Mild mottle + + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐
Ke184 Bungoma West
Sukali Ndizi AAB Streak + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ MY MY
Ke185 Bungoma Uganda Green AAA (EAHB) Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐
145
West Ke186 Bungoma
West Sukali Ndizi AAB Mild mottle + + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐
Ke187 Bungoma Uncertain EAHB AAA (EAHB) Mild mottle ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Uganda February 2010 Ug233 Mbarara Musa
shizocarpa n/a Leaf chlorosis ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐
Ug234 Mbarara Musa textilis n/a Asymptomatic ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐Ug235 Mbarara Suu AAA Leaf
malformation‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐
Ug236 Mbarara To'o AA Mosaic or streak
‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐
Ug237 Mbarara Monyet (M. acum. zebrina)
AA Mild chlorotic flecking
‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ OL
Ug238 Mbarara Dwarf Cavendish
AAA Streak ‐ + ‐ ‐ ‐ ‐ ‐ ‐ ‐ OL
Ug239 Mbarara TMPx 548‐4 unsuree Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ OL&FL GFUg240 Mbarara Red Gonja AAB Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ OL&FL GFUg241 Mbarara Namwa Khom ABB Mild flecks ‐ + + ‐ + ‐ ‐ ‐ ‐ ‐ Ug242 Mbarara FHIA2 AAAB Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ OL&FL GFUg243 Mbarara TMPx 7002 AAAB Mild flecks + + + ‐ ‐ ‐ ‐ ‐ ‐ OLUg244 Mbarara TMPx 548‐9 AAAB Streak + + + ‐ + ‐ ‐ ‐ ‐ FL GFUg245 Mbarara TMBx 5295‐1 AAB Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ OLUg246 Mbarara Gros Michel AAA Streak ‐ ‐ + ‐ ‐ ‐ ‐ ‐ ‐ FL GFUg247 Mbarara TMPx 5511‐2 AAAB Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ OL&FL GFUg248 Mbarara SH 2796‐5 unsuree Streak + + + ‐ ‐ ‐ ‐ ‐ ‐ OL&FL GF Ug249 Mbarara Mysore AAB Streak + ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ FL MYUg250 Mbarara Muracho AAB Mild flecks + ‐ + ‐ + ‐ ‐ ‐ ‐ FL IM&GFUg251 Mbarara Obino l'Ewai AAB Asymptomatic + + + ‐ ‐ ‐ ‐ ‐ ‐ ‐Ug252 Mbarara Kirun AA Leaf
malformation/ mild flecks
‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐
a EAHB indicates East African Highland Banana Musa subgroup. b Grey shaded boxes indicate likely results due to detection of eBSV. c Where a distinct digest pattern was observed which is associated with a particular BSV species this is recorded using the abbreviations in footnote 4. Additionally, full‐length digest results are denoted by ‘FL’ and a number of samples were considered positive (‘+’) where neither a full‐length digest fragment was present, and the observed profile was not associated with any predicted profile. Negative results (i.e an absence of bands in digests) are denoted by ‘‐’. For samples recorded with ‘?’ either faint bands were observed in digests, or too many fragments were observed for a clear result to be determined. d Badnavirus names are abbreviated as follows: Banana streak UL virus = UL, Banana streak UI virus = UI, Banana streak UM virus = UM, Banana streak OL virus = OL, Banana streak GF virus = GF, Banana streak MY virus = MY, Banana streak UA virus = UA, Banana streak CA virus = CA and Banana streak IM virus = IM. e Both Ug239 and Ug248 are hybrid banana cultivars with one parent containing a B‐genome component; as such, these samples are considered to contain a B‐genome.
146
147
CHAPTER 6
MOLECULAR CHARACTERISATION OF BANANA BUNCHY TOP VIRUS
ISOLATES FROM MALAWI AND RWANDA
A.P. James1, R.J. Geijskes1, M. Soko2, J.A. Mugini3, J.L. Dale1 & R.M. Harding1
1Centre for Tropical Crops and Biocommodities, Queensland University of
Technology, P.O. Box 2434, Brisbane, Queensland, Australia. 4001
2Bvumbe Agricultural Research Station, P.O. Box 5748, Limbe, Malawi.
3Mikocheni Agricultural Research Institute, P.O. Box 6226, Dar Es Salaam, Tanzania.
[Formatted for submission to Annals of Applied Biology]
148
STATEMENT OF AUTHORSHIP
Anthony James (principal author): Contributed to project concepts, executed the
work (collected samples, designed and conducted laboratory experiments,
analysed and interpreted results) and prepared initial manuscript.
Signed………………………………………………………………..Date……………………………………………
Jason Geijskes: Supervised execution of the work, critically interpreted data and
contributed to final manuscript.
Signed………………………………………………………………..Date……………………………………………
Mishek Soko: Collected samples, supervised execution of the work, critically
interpreted data and contributed to final manuscript.
Signed………………………………………………………………..Date……………………………………………
Julius Mugini: Collected samples, supervised execution of the work, critically
interpreted data and contributed to final manuscript.
Signed………………………………………………………………..Date……………………………………………
James Dale: Conceived project idea, collected samples, supervised execution of the
work, critically interpreted data and contributed to final manuscript.
Signed………………………………………………………………..Date……………………………………………
Rob Harding: Conceived project idea, supervised execution of the work, critically
interpreted data and contributed to final manuscript.
Signed………………………………………………………………..Date……………………………………………
149
Abstract
Field samples of banana bunchy top disease were collected from bananas growing
in Malawi and Rwanda. The presence of BBTV was confirmed using PCR with virus‐
specific primers and by rolling‐circle amplification. The six virus‐specific DNA
components from one isolate from Malawi and two isolates from Rwanda were
completely sequenced. The BBTV isolates from Malawi and Rwanda had a similar
genome organisation to previously published BBTV sequences including conserved
features such as the major common region (CR‐M), stem‐loop common region (CR‐
SL) and iterons. The two isolates from Rwanda had at least 98.1% nucleotide
sequence identity between each of the six DNA components, while the similarity
between isolates from Rwanda and Malawi was between 96.2% and 99.4%
depending on the DNA component. At the amino acid level, similarities in the
putative proteins encoded by DNA‐R, ‐S, ‐M, ‐ C and –N were found to range
between 98.8% and 100%. Phylogenetic analysis using DNA‐R and ‐S supported
placement of these sequences in the South Pacific subgroup of BBTV isolates.
Sequence comparisons using full‐length DNA‐R sequences identified Indian isolates
as the most similar to East African BBTV isolates.
150
Introduction
Bunchy top is the most important virus disease of bananas worldwide. The disease
occurs in most banana‐growing areas of the Asia‐Pacific region as well as a number
of countries in Africa (Thomas and Iskra‐Caruana, 2000), but has not been reported
in the Latin American‐Carribean region where there is a large export‐oriented
banana industry. The first report of bunchy top in Africa was in Egypt (Fahmy, 1924
in Magee 1927) with subsequent reports from Burundi, Central African Republic,
Republic of Congo, Democratic Republic of Congo, Gabon, Malawi and Rwanda (see
Thomas and Iskra‐Caruana, 2000). More recently, bunchy top has been reported in
Angola, Cameroon, Equatorial Guinea, Mozambique and Zambia (Kumar et al.,
2009; Kumar and Hanna, 2008; Oben et al., 2009).
Characteristic symptoms of bunchy top include a rosetting effect caused by
narrowing of the leaf lamina and shortening of the petioles giving a distinct upright
appearance of the plant apex. Leaves develop a chlorotic, wavy margin while dark
green discontinuous streaks (known as the morse‐code pattern) occur on the
petioles, midribs and leaf veins (Magee, 1927). The disease often leads to a
complete loss in production since infected plants generally produce no fruit due to
the severe stunting and choking of the pseudostem.
Bunchy top is caused by Banana bunchy top virus (Harding et al., 1991;
Thomas and Dietzgen, 1991), the type member of the genus Babuvirus in the
family Nanoviridae. The virus is characterised by isometric virions of 18‐20 nm with
a circular, single‐stranded (ss) DNA genome consisting of at least six components
(Burns et al., 1995) named DNA‐R, ‐U3, ‐S, ‐M, ‐C and ‐N (Vetten et al., 2005).
Plant‐to‐plant transmission occurs via the banana aphid, Pentalonia nigronervosa
(Magee, 1927) in a persistent, non‐propagative manner, while long distance
movement of the virus occurs by vegetative transmission through rhizomes,
suckers and tissue‐cultured plants (Thomas et al., 1994). In Australia, bunchy top
has been effectively controlled through a variety of phytosanitary measures
including regular inspections of plantations for disease and removal of infected
plants which act as virus reservoirs. These measures, combined with strict
151
quarantine control over plant movement and regulations on the use of disease‐
free planting material, have been used successfully to manage disease outbreaks
(Magee, 1967). Despite proving an effective strategy in Australia, the control of the
disease in other less developed countries has been less effective primarily due to
their unregulated banana industries and lack of education amongst growers.
Two groups of BBTV isolates, termed the South Pacific and Asian groups,
have been distinguished based on sequence analyses of DNA‐R, ‐S and –N. The
South Pacific subgroup includes isolates from Angola, Australia, Burundi,
Cameroon, Democratic Republic of Congo, Egypt, Fiji, Gabon, Hawaii, India,
Malawi, Pakistan, Tonga and Samoa, while the Asian subgroup includes isolates
from China, Indonesia, Japan, the Philippines, Taiwan and Vietnam (Amin et al.,
2008; Furuya et al., 2004 & 2005; Karan et al., 1994 & 1997; Kumar et al., 2011;
Wanitchakorn et al., 2000; Xie and Hu, 1995). Between isolates of each subgroup,
the sequences of the major gene of DNA‐R, ‐N and ‐S differ by 7.5, 8.6 and 6.3%
which translates to mean differences of 5.6, 6.7 and 1.4%, respectively, at the
amino acid level. Further, there is considerably greater variability between the
sequences within the Asian group than those within the South pacific group of
isolates. In spite of this variability, diagnostic tests using polymerase chain reaction
(PCR) can reliably detect both subgroups of BBTV isolates (Mansoor et al., 2005).
Bunchy top is an important quarantine disease within East Africa. Despite
the presence of this disease in Burundi, D.R. Congo, Malawi, Mozambique, Rwanda
and Zambia, it has not been detected in Uganda, Kenya and Tanzania. We have
been developing a diagnostic capacity for BBTV, and other important banana
viruses, in Uganda, Kenya and Tanzania to enable ongoing surveillance and allow
the local provision of virus‐indexed banana planting material. As part of this work,
surveys for banana viruses have been conducted in several East African countries.
In this paper, we report the molecular characterisation of BBTV isolates from
Malawi and Rwanda. This report presents an important sequence record for the
region and will contribute to regional efforts towards the control of this important
disease.
152
Materials and Methods
Plant samples
Leaf samples (Table 1) were collected from field grown banana plants in Malawi in
October, 2008 and Rwanda in March, 2009. Samples were collected at random
field sites in each country and while most samples were collected based upon the
presence of characteristic symptoms, a number of samples without symptoms of
bunchy top were also collected. Fresh leaf tissue was dried over silica gel for
preservation until analysis.
Amplification and cloning of BBTV
Total nucleic acids were extracted from dried leaf tissue as described by James et
al., (2011a). The presence of BBTV was confirmed in samples using primers BBTV‐
DNARf (5’ TGGTATATCAAGTGGAGAGGGG 3’) and BBTV‐DNARr (5’
CCAGCTATTCATCGCCTTCG 3’) designed to amplify a 373 bp region of the BBTV
DNA‐R major ORF. As a control for quality of DNA extracts, a PCR for the banana
actin gene was performed using gene specific primers (James et al., 2011a). PCR
mixes (20 µl) contained 10 µl 2x GoTaq Green Master Mix (Promega Corp,
Madison, WI), 5 ρmol of each primer, 1 µl of nucleic acid extract and water to final
volume. PCR cycling conditions were an initial denaturation of 94°C for 2 min
followed by 35 cycles of 94°C for 20 s, 57°C for 20 s, and 72°C for 30 s, with a final
extension at 72°C for 2 min. Reactions products were electrophoresed through
1.5% agarose gels, stained using 0.25x SYBR® Safe DNA Gel Stain (Invitrogen Corp,
Carlsbad, CA) and DNA fragments visualised on a Safe imager blue‐light
transilluminator (Invitrogen Corp, Carlsbad, CA).
Following the initial PCR screening, the complete sequences of each of the
six integral DNA components of two BBTV isolates from Rwanda and one isolate
from Malawi were obtained. Initially, the ORFs of the DNA components of the
153
Table 1 Leaf samples collected in Malawi and Rwanda, and results of PCR and RCA assays for detection of BBTV
Sample number
District ‘Cultivar name’ Genome type
Symptoms BBTV PCR
RCA digest result
Malawi, October 2008
Ma64 Lilongwe Cavendish AAA bunchy top + 1kb
Ma65 Lilongwe Cavendish AAA bunchy top + 1kb
Ma66 Lilongwe Cavendish AAA bunchy top + 1kb
Ma67 Lilongwe Cavendish AAA bunchy top + 1kb
Ma68 Lilongwe Cavendish AAA bunchy top + 1kb
Ma69 Lilongwe Sukali AAB asymptomatic ‐ ‐
Ma71 Lilongwe Cavendish AAA bunchy top + 1kb
Ma72 Nkhotakota Cavendish AAA bunchy top + 1kb
Ma73 Nkhotakota Nzeru AAA bunchy top + 1kb
Ma74 Nkhotakota Sukali AAB bunchy top + 1kb
Ma75 Nkhotakota Harare ABB bunchy top/mosaic + 1kb
Ma76 Nkhotakota Harare ABB bunchy top/mosaic + 1kb
Ma77 Nkhotakota Harare ABB bunchy top + 1kb
Ma78 Nkhotakota Harare ABB bunchy top + 1kb
Ma79 Dwambadzi Pisang Awak ABB asymptomatic ‐ 1kb
Ma80 Dwambadzi Pisang Awak ABB bunchy top + 1kb
Ma81 Dwambadzi Pisang Awak ABB asymptomatic sucker from same stool as sample Ma82
+ 1kb
Ma82 Dwambadzi Pisang Awak ABB bunchy top/mosaic + 1kb
Ma83 Dwambadzi Zomba Red AAA bunchy top + 1kb
Ma84 Dwambadzi unknown Plantain AAB bunchy top + 1kb
Ma85 Dwambadzi FHIA‐21 AAAB bunchy top + 1kb
Ma86 Dwambadzi Cavendish AAA bunchy top + 1kb
Ma87 Dwambadzi TMBx 5295‐1 AABB bunchy top + 1kb
Ma88 Dwambadzi FHIA‐25 AAAA bunchy top + 1kb
Ma89 Dwambadzi SH‐3436 AAAA bunchy top + 1kb
Ma90 Dwambadzi FHIA‐23 AAAA bunchy top + 1kb
Ma91 Dwambadzi Yangambi Km5 AAA asymptomatic ‐ ‐
Ma92 Dwambadzi FHIA‐17 AAAA bunchy top + 1kb
Ma93 Dwambadzi Bluggoe ABB bunchy top + failed
Ma95 Nkhata bay Pelipita ABB leaf mottling ‐ ‐
Ma96 Nkhata bay Calcutta4 AA asymptomatic ‐ ‐
Ma97 Nkhata bay M.balbisiana BB asymptomatic ‐ ‐
Ma98 Nkhata bay FHIA‐2 AAAB bunchy top + 1kb
Ma99 Nkhata bay Calcutta4 AA mild mottle ‐ ‐
Ma100 Nkhata bay Pisang Awak ABB bunchy top/mosaic + 1kb
Ma101 Karonga FHIA‐3 AABB leaf chlorosis ‐ ‐
Ma102 Karonga FHIA‐3 AABB leaf chlorosis ‐ ‐
Ma103 Karonga TMBx 5295‐1 AABB leaf chlorosis ‐ ‐
Ma104 Karonga Cavendish AAA leaf chlorosis ‐ ‐
154
Table 1 continued
Rwanda, March 2009
Rw137 Rusizi (highlands) Barabeshya EAH‐AAA bunchy top + 1kb
Rw138 Rusizi (rift valley) Barabeshya EAH‐AAA bunchy top + 1kb
Rw139 Rusizi (rift valley) Barabeshya EAH‐AAA heart leaf chlorosis/ necrosis + 1kb
Rw140 Rusizi (rift valley) Barabeshya EAH‐AAA bunchy top/heart leaf chlorosis + 1kb
Rw141 Rusizi (rift valley) Yangambi Km5 AAA bunchy top + 1kb
Rw142 Rusizi (highlands) Barabeshya EAH‐AAA bunchy top + 1kb
Rw143 Rusizi (highlands) Red Gros Michel AAA bunchy top + 1kb
Rw144 Rusizi (highlands) Kisubi AAB bunchy top + 1kb
Rw145 Rusizi (highlands) Barabeshya EAH‐AAA bunchy top + 1kb
155
Malawi isolate were amplified by PCR using primers designed on the sequence of
an Australian isolate of BBTV (Harding et al., 1993; Burns et al., 1995). PCR was
carried out using Pwo DNA polymerase (Roche Applied Science, Australia) in 25 µl
reactions as follows: 1 x supplied PCR buffer, 0.5 U of polymerase, 0.2 mM dNTPs,
5 ρmol of each primer, 1 µl of nucleic acid extract and water to final volume. PCR
products were subsequently A‐tailed using Taq polymerase, gel purified and cloned
into pGEM®‐T Easy (Promega Corp, Madison, WI). Three independent clones were
sequenced in both the forward and reverse directions with universal M13 primers,
using the BigDye® Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems,
Foster City, CA), followed by product separation on an Applied Biosystems 3130xl
Genetic Analyser (Griffith University DNA‐Sequencing Facility, Griffith University,
Nathan Campus, Qld, Australia).
Using the sequences from cloned ORF fragments, sequence‐specific primers
were designed within the ORF of each BBTV DNA to amplify the intergenic region
of each virus DNA, including a region of the ORF at each end which overlapped with
the ORF PCR product. These primers were subsequently used in a PCR as described
above to amplify the intergenic region of each of the six BBTV DNAs from the
Malawi isolate. The sequences obtained from both the ORF PCRs and intergenic
region PCRs were compared, primer sequences were removed from each sequence
and a consensus sequence determined for each DNA component. Full‐length
sequences were assembled in Vector NTI Advance v.11 (Invitrogen Corp, Carlsbad,
CA), which was subsequently used for the identification of putative ORFs and other
conserved features of BBTV DNAs.
Cloning and sequencing of the two Rwandan isolates was essentially as
described above with minor variation. The ORF and intergenic region primers
described previously were used to amplify BBTV DNA sequences from all
components, except for three ORF fragments which failed to amplify. In order to
complete the sequence in these cases, sequence‐specific ORF primers were
designed for the Rwandan isolates (based on the intergenic region sequences
generated with primers described previously for the Malawi isolates), and these
were used in a PCR with GoTaq green as described above.
156
Rolling‐circle amplification
Rolling‐circle amplification (RCA) was performed using the protocol described in
James et al. (2011a). RCA reactions were digested using 2 U of restriction enzyme
StuI (Roche Applied Science, Australia), and the presence of a distinct 1 kb band in
digests was considered indicative of the presence of BBTV.
Phylogenetic analysis
For phylogenetic analysis, sequences were aligned using the CLUSTAL W algorithm
of MEGA4 (Tamura et al., 2007). A phylogenetic tree was constructed using the
neighbour‐joining method, following pair‐wise sequence comparison using the
Kimura 2‐parameter model in MEGA4 (Kimura, 1980). For comparison purposes,
only selected isolates with full‐length DNA‐R sequences were included.
Results
Initial screening for BBTV
Leaf samples were collected from 39 banana plants growing in five districts in
Malawi and from nine banana plants growing in one district in south‐west Rwanda
(Table 1). Most samples were taken from plants showing characteristic symptoms
of bunchy top, although leaves from symptomless plants were also collected in
some cases. Nucleic acid extracts prepared from the leaf samples were screened
for BBTV using PCR and RCA.
Using PCR, a product of the expected size of 400 bp was amplified from 28
out of 39 samples from Malawi and all nine samples collected from Rwanda (Table
1). These results were supported by the presence of a distinct 1 kb band in digests
of RCA reactions from all PCR positive samples except for one Malawi sample
(Ma93) for which the RCA reaction failed. A faint 1 kb band was also observed in
157
the digested RCA product from sample Ma79, which was taken from an
asymptomatic leaf and which tested negative for BBTV using PCR. Whether this
indicates the presence of BBTV in a sample with no symptoms, a very early
infection or represents another DNA target detected using RCA is unknown. In all
other cases, the detection of BBTV in samples was consistent with the observation
of typical bunchy top symptoms.
Characterisation of Malawi and Rwanda BBTV isolates
Two BBTV isolates from Rwanda (Rw138 & 142) and one isolate from Malawi
(Ma73) were chosen for further characterisation. The two Rwandan isolates were
selected to enable comparisons between BBTV isolates occurring at different
altitudes; sample Rw138 was taken from a plant growing in the Rift Valley (altitude
970 m) while sample Rw142 was taken from a plant growing in the highlands
(altitude 1648‐1687m). Using a PCR‐based strategy, the six integral DNA
components of all three BBTV isolates were amplified, and these were
subsequently cloned and sequenced. A summary of the characteristics of each of
the six DNAs from the three samples is presented in Table 2.
The BBTV isolates from Malawi and Rwanda had a similar genome
organisation to previously published BBTV sequences. Each DNA component
encoded a single large open reading frame (ORF), while DNA‐R also encoded a
smaller ORF internal to the major ORF. The size and relative position of the ORFs
on each component were analogous to a previously published isolate from
Australia with the exception of the ORF from DNA‐U3 which was shorter in the
three isolates studied. The predicted major ORF of DNA‐U3 in isolates Ma73 and
Rw142 comprised 231 nt; while in sample Rw138 the predicted ORF length was 177
nt due to a two bp insertion in the sequence of this isolate creating a stop codon
earlier in the coding sequence compared with the other isolates. Although shorter
than the predicted ORF length in an Australian isolate (261 nt), the 231 nt ORF
length in samples Ma73 and Rw142 was consistent with sequences of this
component from India and Pakistan.
158
Table 2 Characteristics of DNA components of BBTV isolates from Malawi and Rwanda
Malawi 73 Rwanda 138 Rwanda 142
BBTV DNA Total length (nt)
Length of major ORF
Total length (nt)
Length of major ORF
Total length (nt)
Length of major ORF
DNA‐Ra 1111 858 1111 858 1110 858 DNA‐U3 1063 231 1062 177 1062 231 DNA‐S 1075 510 1075 510 1074 510 DNA‐M 1046 351 1047 351 1046 351 DNA‐C 1018 483 1018 483 1018 483 DNA‐N 1086 462 1089 462 1088 462 aDNA‐R also contains a conserved small ORF internal to the major ORF
159
Additional conserved features of BBTV genomes, including the major
common region (CR‐M), stem‐loop common region (CR‐SL) and iteron sequences
were also identified in the three isolates sequenced. Analysis of the six DNA
components for each isolate revealed a 89 bp common region located 3´ of the
major ORF on each component. Sequence conservation within this region for each
isolate was in the order of 63% (Ma73), 66% (Rw138) and 65% (Ma142), while
sequence conservation when all DNAs of the three isolates were considered
together was 63%. When the three DNA‐R sequences were omitted from the
analysis, CR‐M sequence conservation across the three isolates increased to 72%. A
31 nucleotide CR‐SL was identified which comprised a 11 nucleotide loop section
flanked by 10 nucleotide stem sequences. This 31 nucleotide region was 81%
conserved amongst the 18 sequences comprising the three isolates. The F1F2
iteron sequence (GGGACGGGAC) was conserved on 10 of the 18 DNAs while on the
remaining eight DNAs a single G to A change was present (AGGGACGGGAC). This
single base change was limited to the three DNA‐C and DNA‐M sequences, as well
as the two DNA‐U3 sequences from isolates Rw138 and Rw142. The R1 iteron
sequence (GGGAC) was completely conserved across all DNAs sequenced.
Sequence comparisons between the three isolates were done using the full‐
length nucleotide sequences of each DNA component, as well as the amino acid
sequences of the major ORF of each component (except DNA‐U3), and internal ORF
of DNA‐R (Table 3). At the nucleotide level, the two isolates from Rwanda were the
most similar, with at least 99.4% similarity across all components except DNA‐U3
(98.1%). When the nucleotide sequences of isolate Rw138 were compared to
isolate Ma73, the similarity varied from 96.2% (DNA‐U3) to 99.2% (DNA‐S, ‐M & ‐
C). When isolate Rw142 was compared to Ma73, the similarity varied from 97.6%
(DNA‐U3) to 99.4% (DNA‐C). At the amino acid level, very high similarities (98.8‐
100%) were seen across the putative proteins encoded by DNA‐R, ‐S, ‐M, ‐ C and ‐N
of all three isolates (Table 3). When the amino acid sequences of the internal ORF
of DNA‐R were compared, the sequences of the two isolates from Rwanda were
identical, while the Malawi isolate had a single amino acid change.
160
Table 3 Sequence similarity between Malawi and Rwanda BBTV DNA components
BBTV DNA % nt similarity % aa similaritya, b Rwanda 138 Rwanda 142 Rwanda 138 Rwanda 142
DNA‐R (REP) 98.9 98.9 99.6 100 Malawi 73 99.4 99.6 Rwanda 138 DNA‐R (intORF) 97.6 97.6 Malawi 73 100 Rwanda 138 DNA‐U3 96.2 97.6 Malawi 73 98.1 Rwanda 138 DNA‐S 99.2 99.3 100 100 Malawi 73 99.5 100 Rwanda 138 DNA‐M 99.2 98.9 100 100 Malawi 73 99.5 100 Rwanda 138 DNA‐C 99.2 99.4 99.4 99.4 Malawi 73 99.6 98.8 Rwanda 138 DNA‐N 98.8 98.9 100 100 Malawi 73 99.7 100 Rwanda 138 ain putative protein sequence of translated ORF nucleotide sequence bdue to differing length of the putative ORFs for DNA‐U3 a comparison was not done
161
When the nucleotide sequences of each of the six DNA components of the
three isolates were compared to published sequences, DNA‐R, ‐U3, ‐M, ‐C and ‐N
showed highest similarities to Indian isolates of BBTV with sequence identities of
99%, 97%, 99%, 99% and 98%, respectively. The nucleotide sequence of DNA‐S was
equally similar (99%) to isolates from Australia, Burundi, Cameroon, Fiji, India and
Myanmar.
Phylogenetic analysis
Phylogenetic analysis based on the complete nucleotide sequences of DNA‐R
revealed that the three East African BBTV isolates grouped within the South Pacific
cluster of BBTV isolates and were most closely related to the representative Indian
BBTV isolate included in the analysis (Fig. 1). When the analysis was done using
nucleotide sequences of DNA‐S, the three isolates again clustered within the South
Pacific subgroup of BBTV isolates, but were most closely related to an Australian
isolate (result not shown).
Discussion
In this study, samples of bananas were collected from several districts in central
and northern Malawi, as well as one district in south‐west Rwanda. Bunchy top
disease symptoms were observed at all but one of the locations visited (Karonga,
Malawi) and the presence of BBTV was confirmed using PCR and RCA. Further, full‐
length sequences of all six DNA components from two BBTV isolates from Rwanda
and one isolate from Malawi were cloned, sequenced and analysed. To the best of
our knowledge, this is the first report of full‐length sequences of African BBTV
isolates. Phylogenetic analysis of the full‐length sequence of BBTV DNA‐R and ‐S
confirmed previous studies which suggested that BBTV isolates from Africa were
more similar to the South Pacific subgroup of BBTV isolates than to the Asian
subgroup of isolates (Karan et al., 1994 & 1997; Kumar et al., 2011; Wanitchakorn
et al., 2000).
162
Figure 1 Neighbour‐joining phylogram of BBTV DNA‐R sequences based on
CLUSTAL W alignment (Bootstrap consensus tree shown following 1000 replicates).
GenBank accession numbers for BBTV isolates are India (GenBank accession
AF416470), Australia (GenBank accession NC_003479), Myanmar (GenBank
accession AB252639), USA (GenBank accession BBU18077), Tonga (GenBank
accession AF416467), Fiji (GenBank accession AF416466), Pakistan (GenBank
accession AM418536), Egypt (GenBank accession AF416465), Vietnam (GenBank
accession AF416464), China (GenBank accession AF246123), Indonesia (GenBank
accession AB186924), Philippines (GenBank accession AF416469), Japan (GenBank
accession AB108453), Taiwan (GenBank accession AF416468), Abaca bunchy top
virus (GenBank accession EF546807).
GC138 Rwanda
GC142 Rwanda
GC73 Malawi
India
Australia
Myanmar
USA (Hawaii)
Tonga
Fiji
Pakistan
Egypt
Vietnam
China
Indonesia
Philippines
Japan
Taiwan
Abaca bunchy top virus DNA-R
87
64
100
100
99
99
61
0.01
South
Pacific
subgroup
Asian
subgroup
163
Bunchy top symptoms were first observed in Malawi in the early 1990s
(Mshani et al., 2008) although the disease was not officially reported until 1997
(Kenyon et al., 1997). The early reports were from the central‐northern districts of
Nkhata‐bay and Nkota‐kota, but bunchy top is now widespread and occurs in
banana growing regions in the south, west, central and north districts of Malawi,
with the exception of Karonga and Chitipa in the north, and Thyolo and Mulanje in
the south (this study; Kumar et al., 2011). One possibility for this in the north is the
geographic separation of Karonga/Chitipa from the nearest banana growing areas.
Nkata bay is a distance of greater than 200 km, and is separated from Karonga by
the northern limits of the Viphya Mountains and the Nyika Plateau, both of which
may serve to limit banana cultivation, as well as movement of banana plants and
the aphid vector. It is more likely, however, that the absence of bunchy top is
related to the limited cultivation of highly susceptible Cavendish banana types at
this location (authors’ observations; Kumar et al., 2011).
A very high incidence of bunchy top disease was observed at many sites
visited in Malawi during the present study, with plants showing severe stunting
and suckers exhibiting strong bunchy top symptoms. Further, large populations of
the black banana aphid vector were observed at almost all locations. As a
consequence of the high disease pressure and poor management by growers,
banana production in some areas has become unprofitable with farmers
abandoning their plots. In an effort to control the disease, Mshani et al. (2008)
reported that 800 ha of severely infected banana plantations in the Nkhata‐bay
and Nkota‐kota districts were rogued to remove infected plants. However, without
farmer education, and available sources of clean planting material, this control
strategy is unlikely to be effective in the long term.
In contrast to Malawi, both the disease incidence and symptom severity in
Rwanda were not as high. In most cases, symptoms were observed on older plants
although, upon closer inspection, characteristic symptoms (i.e. morse code
pattern) could also be detected on suckers. These observations were consistent
amongst banana plants growing at different altitudes in both the Rift Valley and
highland regions of Rwanda. Sequence analysis of BBTV isolates obtained from
164
two plants growing at altitudes of 970m and 1648‐1687m showed that the two
isolates were essentially identical, and were also most similar to the isolate from
Malawi, suggesting a common origin for these isolates. As bunchy top has
reportedly been confined to the Rift Valley region in Rwanda, this suggests that
movement from the Rift Valley into the highlands is most likely from a local source
through infected planting material.
Although bunchy top has been observed on the African continent for more
than a century and in central Africa for several decades, it is only recently that
serious disease outbreaks have been reported (Mshani et al., 2008; Kumar and
Hanna, 2008; Kumar et al., 2011). Although the reason for this is unknown, it is
likely to be associated with a range of different factors including host cultivar,
temperature and vector. Cavendish type banana cultivars, for example, are
extremely susceptible to BBTV compared to other cultivars but Cavendish types are
becoming increasingly popular in many regions of Africa. Bunchy top disease is also
more prevalent in locations where warmer temperatures are more favourable for
both the aphid vector (both growth and virus transmission) and rapid
growth/expression of symptoms (Anhalt and Almeida, 2008; Magee, 1927). This
may explain the lower incidence and slower spread of bunchy top disease in
regions such as Rwanda where vector prevalence is low, temperatures are cooler
throughout the year, and East African Highland bananas (and not Cavendish) are
the dominant cultivars grown. In contrast, in countries like Malawi, the
combination of drier climate, warmer temperatures, large vector populations and
relatively recent shift to growing the highly susceptible cultivars in the Cavendish
complex may account for the increasing disease incidence. An important
observation made during these disease surveys was that, whereas aphids were
difficult to find at the locations visited in the East African Highlands, mealybugs
were far more prevalent as was the incidence of BSV infection (A. James,
unpublished).
The availability of sequences of East African isolates of BBTV will improve
confidence in PCR‐based diagnosis with detection of local isolates validated by
diagnostic laboratories based in the region. The availability of positive controls to
165
indexing laboratories in the region is critical for preparedness, as BBTV is the most
important virus threat to banana production in East Africa, particularly in Uganda,
Kenya and Tanzania where the virus has not yet been reported. Additionally,
training of local agricultural and extension program staff in recognising and
controlling the disease is a critical activity, Finally, with transgenic BBTV‐resistant
bananas considered the most suitable long‐term approach for BBTV control, efforts
will be bolstered by knowledge that virus isolates present in the East Africa share
significant sequence similarity with previously characterised isolates, improving the
chances of developing durable, transgene mediated resistance to BBTV.
Acknowledgements
We are grateful to farmers in both Malawi and Rwanda for allowing us to collect
samples. This research was funded by the Bill and Melinda Gates Foundation
Grand Challenges in Global Health Program.
166
References
Amin, I., Qazi, J., Mansoor, S., Ilyas, M., Briddon, R.W. (2008) Molecular
characterisation of Banana bunchy top virus (BBTV) from Pakistan. Virus
Genes, 36, 191–198.
Anhalt, M.D., Almeida, R.P.P. (2008) Effect of temperature, vector life stage, and
plant access period on transmission of Banana bunchy top virus to banana.
Phytopathology, 98, 743–748.
Burns, T. M., R. M. Harding, J. L. Dale (1995) The genome organization of banana
bunchy top virus: analysis of six ssDNA components. Journal of General
Virology, 76, 1471–1482.
Fahmy, T. (1924) A banana disease caused by a species of Heteroda. Ministry of
Agriculture (Bot. Section), Egypt. Bulletin No. 30.
Furuya, N., Kawano, S., Natsuaki, K.T. (2005) Characterisation and genetic status of
banana bunchy top virus isolated from Okinawa, Japan. Journal of General
Plant Pathology, 71, 68–73.
Furuya, N., Somowiyarjo, S., Natsuaki, K.T. (2004) Virus detection from local banana
cultivars and the first molecular characterisation of banana bunchy top virus
in Indonesia. Journal of Agricultural Science ‐ Tokyo University of Agriculture,
49, 75–81.
Harding, R.M., Burns, T.M., Dale, J.L. (1991) Virus‐like particles associated with
banana bunchy top disease contain small single‐stranded DNA. Journal of
General Virology, 72, 225–230.
Harding, R. M., Burns, T. M., Hafner, G., Dietzgen, R. G., Dale, J. L. (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.
James, A.P., Geijskes, R.J., Dale, J.L., Harding, R.M. (2011a) Development of a novel
rolling‐circle amplification technique to detect banana streak virus which
also discriminates between integrated and episomal virus sequences. Plant
Disease, 95, 57–62.
167
Karan, M., Harding, R.M., Dale, J.L. (1994) Evidence for two groups of banana
bunchy top virus isolates. Journal of General Virology, 75, 3541–3546.
Karan, M., Harding, R.M., Dale, J.L. (1997) Association of banana bunchy top virus
DNA components 2 to 6 with bunchy top disease. Molecular Plant Pathology
On‐line: [http://www.bspp.org.uk/mppol/] 1997/0624karan].
Kenyon, L., Brown, M., Khonje, P. (1997) First report of banana bunchy top virus in
Malawi. Plant Disease, 81, 1096.
Kimura, M. (1980) A simple method for estimating evolutionary rate of base
substitutions through comparative studies of nucleotide sequences.
Journal of Molecular Evolution, 16, 111–120.
Kumar, L., Hanna, R. (2008) Banana bunchy top virus in sub‐Saharan Africa:
established or emerging problem? In Banana and Plantain in Africa:
Harnessing international partnerships to increase research impact, pp. 95–
96. Eds T. Dubois, S. Hauser, C. Staver, and D. Coyne: ISHS.
Kumar, P.L., Ayodele, M., Oben, T.T., Mahungu, N.M., Beed, F., Coyne, D., Londa, L.,
Mutunda, M.P., Kiala, D., Maruthi, M.N. (2009) First report of Banana
bunchy top virus in banana and plantain in Angola. Plant Pathology, 58, 402.
Kumar, P.L., Hanna, R., Alabi, O.J., Soko, M.M., Oben, T.T., Vangu, G.H.P., Naidu,
R.A. (2011) Banana bunchy top virus in sub‐Saharan Africa: Investigations
on virus distribution and diversity. Virus Research, 159, 171–182.
Magee, C.J.P. (1927) Investigation on the bunchy top disease of the banana. Council
for scientific and industrial research. Bulletin No. 30.
Magee, C.J.P. (1967) The control of banana bunchy top. South Pacific Commission
Technical paper No. 150. Noumea: SPC.
Mansoor, S., Qazi, J., Amin, I., Khati, A., Khan, I.A., Raza, S., Zafar, Y., Briddon, R.W.
(2005) A PCR‐based method, with internal control, for the detection of
Banana bunchy top virus in banana. Molecular Biotechnology, 30, 167–169.
Mshani, V. E., Soko, M. M., Banda, D. L. N., Mwenebanda, B. M. L., Msosa, T. K.,
Mkankhama, Mwamlima, L. H. (2008) Evaluation of banana and plantain
168
hybrid cultivars in banana bunchy top affected areas. In Plant protection
progress report for the 2007/2008 season, pp. 23–31. Lilongwe : Ministry of
Agriculture and Food Security.
Oben, T.T., Hanna, R., Ngeve, J., Alabi, O.J., Naidu, R.A., Kumar, P.L. (2009)
Occurrence of banana bunchy top disease caused by Bananan bunchy top
virus on banana and plantain (Musa sp.) in Cameroon. Plant Disease, 93, 1076.
Tamura, K., Dudley, J., Nei, M., Kumar, S. (2007) MEGA4: Molecular Evolutionary
Genetics Analysis (MEGA) software version 4.0. Molecular Biology and
Evolution, 24, 1596–1599.
Thomas, J. E., Dietzgen, R. G. (1991) Purification, characterization and serological
detection of virus‐like particles associated with banana bunchy top virus in
Australia. Journal of General Virology, 72, 217–224.
Thomas, J.E., Iskra‐Caruana, M.L. (2000) Bunchy top. In Diseases of banana, abaca
and enset, pp. 241–253. Eds D.R.Jones. United Kingdom: CABI.
Thomas, J.E., Iskra‐Caruana, M‐L., Jones, D.R. (1994) Musa Disease Fact Sheet N° 4:
Banana Bunchy Top Disease. INIBAP.
Vetten, H. J., Chu, P. W. G., Dale, J. L. & 7 other authors (2005) Family Nanoviridae.
In Virus Taxonomy: Eighth Report of the International Committee on
Taxonomy of Viruses, pp. 343–352. Eds C.M. Fauquet, M.A. Mayo, J.
Maniloff, U. Desselberger and L.A. Ball. London: Elsevier.
Wanitchakorn, R., Harding, R.M., Dale, J.L. (2000) Sequence variability in the coat
protein gene of two groups of banana bunchy top isolates. Archives of
Virology, 145, 593–602.
Xie, W.S., Hu, J.S. (1995) Molecular cloning, sequence analysis, and detection of
banana bunchy top virus in Hawaii. Phytopathology, 85, 339–347.
169
CHAPTER 7
GENERAL DISCUSSION
Viruses are important constraints for banana production in many countries.
Historically, Banana bunchy top virus (BBTV) has been the most important virus
disease of bananas worldwide. In the past 20 years BBTV has emerged as a
significant production constraint in many countries of sub‐Saharan Africa. Although
not present in several countries in East Africa, bunchy top represents the greatest
threat to production in these areas from a virus disease. In contrast, Banana streak
disease, caused by a complex of Banana streak virus (BSV) species, is widespread in
Kenya and Uganda. Although the effects on production are not as limiting as
bunchy top, the prevalence of streak disease in these countries makes it regionally
important. Diagnosis of BSV is difficult due to considerable nucleotide sequence
variability in characterised BSV isolates and the presence of endogenous BSV
(eBSV) sequences present in the Musa genome, some of which can become
activated to cause episomal infections. Because of these factors BSV is considered
a serious constraint for Musa breeding programmes and the international
exchange of Musa germplasm. The research contained in this thesis, therefore, was
directed towards the development of diagnostics for banana viruses, with a
particular focus on BSV detection, and the characterisation of viruses which infect
banana in East Africa.
Banana streak virus
Banana streak disease does not cause the same spectacular yield losses as bunchy
top disease. In fact, based on field observations during this study, yield losses
appeared negligible in many instances despite clear symptoms being observed on
banana plants. It was not uncommon to see large bunches developing on full‐
grown bananas despite many plants in a field showing symptoms consistent with
streak disease, particularly in the south west of Uganda. The severity of banana
streak is related to the combined effects of host‐plant cultivar, virus species
present and the prevailing environmental and cultivation conditions.
170
The RCA method developed in this study is an important step forward in the
detection and characterisation of badnaviruses, and other members of the family
Caulimoviridae. With endogenous badnavirus sequences prevalent in both the
Musa acuminata and M. balbisiana genome (Geering et al., 2005), a method which
specifically detects episomal forms of the virus DNA is advantageous. Furthermore,
with endogenous virus sequences reported for virus species from five of the six
genera in the family Caulimoviridae, it is highly likely that RCA detection of these
large DNA viruses will find application in a wide range of host‐virus combinations. A
recent report by Geering et al. (2010) described the occurrence of a novel group of
endogenous virus sequences, which have similarity to members of the
Caulimoviridae, in species from a number of plant families. RCA would undoubtedly
be a useful method to investigate the possible occurrence of episomal forms of
these sequences, which may or may not infect the same host plants as their
endogenous counterparts.
This study used the novel RCA method to characterise six new full‐length
BSV isolates which represent distinct species based on ICTV guidelines. Geering et
al. (2011) recently reported the full‐length sequence of one of these putative new
species, and proposed the name Banana streak IM virus (BSIMV). The sequence of
BSIMV reported by Geering et al. (2011) was one nucleotide shorter than that of
the Kenyan isolate of BSIMV identified in the present study while the two
sequences shared 99.2% nucleotide identity over their complete genomes (99.6%
in the RT/RNaseH‐coding region). For the six putative new BSV species, biological
information on their transmission, cultivar reactions and yield effects would be
useful to determine if control is warranted. Interestingly, the results from surveys
revealed that of these six species, only BSIMV had a significant number (75%) of
infections in B‐genome banana cultivars, while the remaining five species were
almost exclusively detected in AAA‐genome cultivars (one RCA positive sample of
BSUMV was detected in a AAB‐genome cultivar). This result is consistent with the
findings of Harper et al. (2005), which showed that of the 19 sequences reported
for BSUAV, BSUIV, BSULV and BSUMV, all but one (sample 30 therein) were present
in AAA‐genome cultivars. This finding may simply highlight the cultivar preference
171
of farmers in the East African highlands (predominantly AAA‐genome bananas from
the East African highland banana subgroup) or reinforce previous observations of
the very slow rate of mealy bug transmission of BSV between plants (Kubiriba et
al., 2001), and thus confirming that the primary means of BSV dissemination for
these species is through vegetative propagation.
Harper et al. (2005) identified 13 distinct BSV sequence groups (named
banana streak Uganda‐A to ‐M virus) in samples of banana with streak disease in
Uganda. In the present study, the episomal nature of four of these species was
confirmed and one isolate of each was completely characterised. However, a
number of potential BSV species detected by Harper et al. (2005) were not
identified in this study, although several samples were collected from bananas in
Uganda with streak symptoms which were positive for BSV infection using RCA and
the BSV isolate was not identified. This result could indicate that further isolates of
episomal BSV are present in Uganda awaiting characterisation.
In banana, the opportunity exists to identify and characterise further BSV
species. With many endogenous sequences identified in banana for which
episomal counterparts have not yet been identified (Geering et al., 2005), as well
as field isolates which are only partially characterised (Harper et al., 2005), the
search to identify further episomal BSV isolates will continue. Phylogenetic
analyses have demonstrated three distinct subgroups of badnavirus sequences
(Gayral and Iskra‐Caruana, 2009; Harper et al., 2005). BSV isolates separate into all
three of these subgroups, with episomal BSV only identified in subgroup 1 (BSGFV,
BSMYV, BSOLV, BSVNV, BSIMV, BSCAV and BSUAV) and subgroup 3 (BSUIV, BSULV
and BSUMV). Seven of the Ugandan BSV sequences (Banana streak Uganda‐B to ‐H
virus) reported by Harper et al. (2005) cluster in subgroup 2. Confirmation of the
episomal nature of BSV isolates from subgroup 2 is of interest. In the present
study, a putative new episomal BSV sequence was identified from an Australian
germplasm collection using the novel RCA method. This BSV sequence, of
approximately 7 kbp, appears to be complete based on comparison with other
members of badnavirus subgroup 2. Based on nucleotide similarity in the 543 bp
region of the polymerase gene delineated by the Badna‐FP/RP primers reported in
172
Yang et al. (2003), this sequence is most similar to the BSUgBV/BSUgEV subgroup
with approximately 78% nucleotide similarity. This isolate is also phylogenetically
most closely related to the putative subgroup 2 BSV sequences infecting banana in
Uganda (Fig. 1). This recent work provides evidence that a banana‐infecting
‘subgroup 2’ isolate of BSV exists as an episomal infection. Further, using PCR
primers designed to amplify a 2.5 kbp region of the genome, a fragment of the
expected size was amplified from a diverse genetic group of banana cultivars
suggesting this episomal sequence may have a highly similar endogenous
counterpart.
The most significant effects of BSV at present are not on the field
production of banana but on the propagation of banana by tissue‐culture,
international movement of banana germplasm (as tissue cultures) and on efforts at
genetic improvement of banana by breeding activities. This is mostly as a
consequence of the activation of eBSV within the genome of M. balbisiana (Iskra‐
Caruana et al., 2010), but is coupled to the lack of a robust, sensitive method to
index for BSV. The RCA method developed in this study should detect all described
BSV species. Importantly, the four species of BSV with endogenous equivalents
were shown to be detected using the RCA method. This method could be used to
index tissue culture accessions which are received by importing organisations in
countries interested in exchanging Musa germplasm, particularly those accessions
which contain a M. balbisiana genome component. As activation of eBSV appears
limited to genotypes which possess only a single B‐genome component, such as
tetraploid hybrids (AAAB) and plantains (AAB), monitoring for activation need not
be conducted on every Musa accession exchanged. Additionally, as activation is
limited to less than 25% of progeny plants produced in tissue culture (Iskra‐
Caruana et al., 2010), there is an almost certain chance that plants free of infection
would be identified.
173
Banana bunchy top virus
Banana bunchy top virus (BBTV) is the most important virus which infects
banana. As Charles McGee wrote in 1927:‐ “It would be difficult for anyone who
has not visited these devastated areas to visualize the completeness of the
destruction wrought in such a short time by a plant disease”. McGee was writing
about the devastation to banana production in eastern Australia, specifically
northern New South Wales and south east Queensland, where bunchy top disease
nearly destroyed the banana industry. At this time bunchy top incidence in
plantations reached 100% is some districts, and many banana plantations were
being deserted due to lack of fruit production. Almost 90 years later a profitable
banana industry now exists in these areas due to controls devised and
implemented following McGee’s early work to understand the epidemiology of
bunchy top. Strict controls over the movement and selection of planting material
combined with inspections of plantations where the disease occurs and strict
roguing of infected plants have allowed banana production to continue in areas
where the disease occurs, without eradication being achieved.
Despite the lack of any known sources of host plant resistance in cultivated
banana varieties, long term control of bunchy top can be still achieved. However,
these controls rely on strong legislative regulation and strict enforcement.
Epidemics of bunchy top have been recorded in many places, more recently in
Pakistan in the 1990’s (Amin et al., 2008) and in modern times across the African
continent (Kumar and Hanna, 2008; Kumar et al., 2011; Mshani et al., 2008). The
devastation that Charles McGee described in Australia is the scenario in present
day Malawi, where fields of abandoned bananas showing the most severe of
bunchy top symptoms provide a reservoir of infection which, when combined with
large numbers of the aphid vector, enable the disease to thrive. Grower education,
legislative controls and clean planting material are paramount for banana
production to continue under these conditions. Even where fields have a lower
incidence of bunchy top, diseased plants which never produce fruit are not
removed by growers. Without education on the factors which contribute to disease
174
AJ968450 BSIMV Ug49-2
HQ659760 BSIMV-Au
HQ593112 BSIMV-Ke
AJ968447 BSIMV\21-6
AY750155 BSVNV
NC_004540 KTSV
NC_003381 BSOLV
HQ593111 BSCAV
HQ593107 BSUAV
AJ968452 BSUgAV Ug44-3
AJ968455 BSUgAV Ug45-2
NC_006955 BSMYV
GU121676 PBCoV
NC_003382 CiYMV
NC_001574 CSSV
GQ428155 PVBV
AJ968464 BSUgCV Ug11-4
AJ968470 BSUgGV Ug53-1
AJ968472 BSUgHV Ug22-1
Novel BSV sequence
AJ968465 BSUgDV Ug52-1
AJ968469 BSUgFV Ug11-3
AJ968463 BSUgBV Ug11-5
AJ968466 BSUgEV Ug11-2
NC_007002 BSGFV
NC_001343 ComYMV
NC_003031 SCBIMV
NC_008017 SCBMOV
AJ968539 BSUgMV Ug21-5
AJ968540 BSUgMV Ug26-3
HQ593110 BSUMV
AJ968501 BSUgJV Ug29-1
AJ968481 BSUgIV Ug2-10
HQ593108 BSUIV
AJ968473 BSUgIV Ug1-2
AJ968504 BSUgKV Ug8-1
AJ968506 BSUgKV Ug9-2
HQ593109 BSULV
AJ968510 BSUgLV Ug6-10
AJ968513 BSUgLV Ug7-4
AF357836 TaBV
NC_001914 RTBV
NC_001497 CaMV
86
100
100
74
100
100
100
77
98
45
40
62
100
100
99
100
52
78
100
99
81
72
42
95
93
91
82
65
39
40
54
39
43
18
27
73
85
99
0.05
175
Fig. 1. Phylogenetic tree, using neighbour‐joining method (Kimura 2‐parameter model with
bootstrapping (1000 replicates)), of partial RT/RNaseH‐coding region sequences of
selected badnaviruses. Rice tungro bacilliform virus (RTBV; genus Tungrovirus) and
Cauliflower mosaic virus (CaMV; genus Caulimovirus) were used as out‐groups to the
genus Badnavirus. GenBank accession numbers are listed with virus names in parentheses.
Virus name abbreviations are Banana streak IM virus (BSIMV), Banana streak OL virus
(BSOLV), Banana streak MY virus (BSMYV), Kalanchoe top‐spotting virus (KTSV), Banana
streak VN virus (BSVNV), Banana streak CA virus (BSCAV), Banana streak UA virus (BSUAV),
Banana streak GF virus (BSGFV), Pineapple bacilliform cosmosus virus (PBCoV), Commelina
yellow mottle virus (ComYMV), Cacao swollen shoot virus (CSSV), Citrus yellow mosaic virus
(CiYMV), Pelargonium vein banding virus (PVBV), Banana streak Uganda A‐M virus (BSUgA‐
MV), Dioscorea bacilliform SN virus (DBV), Taro bacilliform virus (TaBV), Sugarcane
bacilliform IM virus (SCBIMV), Sugarcane bacilliform MO virus (SCBMOV), Banana streak UI
virus (BSUIV), Banana streak UL virus (BAULV) and Banana streak UM virus (BSUMV). New
group 2 BSV identified in bold font.
development and spread, this situation will continue and control will be impossible
to achieve.
Host‐plant resistance is the most promising way of controlling an endemic
plant disease, especially in countries where legislative controls are weak and
education on managing virus diseases is poor. For BBTV, transgenic resistance to
the virus is the most promising option for the development of resistant banana.
This is because no known natural resistance to BBTV has been reported and, even
if these were available, the low levels of fertility which complicate genetic
improvement of banana by traditional breeding approaches would make it nearly
impossible to reconstitute the diverse range of dessert and cooking bananas which
are in use by growers across such a diverse continent as Africa. Sequencing of East
African isolates of BBTV in this work has confirmed their identity within the south
Pacific subgroup of BBTV isolates and identified very high levels of sequence
conservation with other isolates from this subgroup. As a consequence, transgenic
resistance against isolates of BBTV in East Africa might be possible using sequences
derived from other south Pacific isolates, for example an isolate from Australia,
Hawaii or Pakistan. Several authors have recently suggested the use of hairpin‐
mediated RNA silencing (RNAi) as an option to generate transgenic resistance to
BBTV in banana (Amin et al., 2008; Borth et al., 2011). Once proven, this
technology could be applied into selected cultivars which are regionally important
within East Africa and elsewhere to achieve control of bunchy top disease.
PCR is routinely used to detect BBTV as this method is extremely sensitive,
rapid, robust and cost‐effective. For countries in East Africa where BBTV is not
present, such as Kenya, Tanzania and Uganda, confirmation that the primers used
in PCR can detect local isolates of BBTV is paramount. The work described in this
thesis has enabled the establishment of virus indexing capabilities for BBTV at
laboratories in these three countries, with BBTV isolates from Rwanda and Malawi
made available as positive control samples and primers for indexing validated on
these samples. Additionally, the training of project staff from Uganda, Kenya and
Tanzania in symptom recognition should improve regional efforts to monitor for
BBTV. The survey work described in this thesis will form a useful basis for further
177
efforts to monitor border areas for bunchy top incursions and hopefully assist with
early detection of the disease if it does move into these countries.
Conclusion
Internationally, diagnostic testing for banana viruses has historically been confined
to a very few groups of laboratories and not located within the major banana
growing regions. This has placed severe limitations on (i) the international
movement of banana germplasm, (ii) the within country production and
distribution of virus tested banana tissue culture and clean field planting material
and (iii) the general availability of, and access to, diagnostics for banana viruses for
quarantine and survey purposes. Prior to the research presented here, detection of
BSV was the major factor limiting the development of banana virus diagnostic
capacity because the accepted technique, IC‐PCR, was dependent on the
availability of a polyclonal antibody that was not commercially available. The
diagnostics for the other significant banana viruses were or could be made
available using commercially and readily available components. Importantly, the
work described in this thesis has made a significant contribution to the availability
of generic DNA‐based diagnostic tests for BSV. The ability to diagnose infection of
BSV using straight‐forward, commercially prepared reagents which avoid the
requirement for the use of antisera, will make BSV diagnostics universally available.
Methods which utilise antisera for BSV detection suffer from the serological
variability of BSV isolates, a likely lack of continuity of supply, and when using IC‐
PCR the narrow scope of detection when using virus‐specific primers sets and the
occurrence of false positives arising from integrated virus sequences. In contrast,
RCA negates the use of antisera, detects a broad range of isolates and specifically
detects episomal virus sequences. The availability of improved diagnostics for BSV
together with DNA based diagnostics for the other significant banana viruses will
enhance international efforts to exchange germplasm and utilise tissue culture for
multiplication of bananas in which activation of endogenous sequences occurs. As
an example, three laboratories in East Africa have developed capacity for banana
virus diagnostics as part of the broader collaboration in the Bill and Melinda Gates
178
Foundation funded project and are already implementing this capacity for the
practical benefit of farmers in those countries.
179
References
Amin, I., Qazi, J., Mansoor, S., Ilyas, M. and Briddon, R.W. (2008). Molecular
characterisation of Banana bunchy top virus (BBTV) from Pakistan. Virus
Genes 36:191–198.
Borth, W., Perez, E., Cheah, K., Chen, Y., Xie, W.S., Gaskill, D., Khalil, S., Sether, D.,
Melzer, M., Wang, M., Manshardt, R., Gonsalves, D. and Hu, J.S. (2011). In
Proc. Int’l ISHS‐Promusa Symp. on Global Perspectives on Asian Challenges
pp449–458. Edited by I. Van den Bergh, M. Smith, R. Swennen, C.
Hermanto. Acta Hort. 897, ISHS.
Gayral, P. and Iskra‐Caruana, M.L. (2009). Phylogeny of Banana streak virus reveals
recent and repetitive endogenization in the genome of its banana host
(Musa sp.). J. Mol. Evol. 69:65–80.
Geering, A.D.W., Choisne, N., Vezzuli, S., Scalabrin, S., Quesneville, H., Velasco, R.
and Teycheney, P–Y. (2010). Viruses for breakfast, lunch and dinner. Proc. of
the 9th Australasian Plant Virology Workshop pp38. Australasian Plant
Pathology Society.
Geering, A.D.W., Olszewski, N.E., Harper, G., Lockhart, B.E., Hull, R. and Thomas, J.E.
(2005). Banana contains a diverse array of endogenous badnaviruses. J. Gen.
Virol. 86:511–520.
Geering, A.D.W., Parry, J.N. and Thomas, J.E. (2011). Complete genome sequence of
a novel badnavirus, banana streak IM virus. Arch. Virol. 156:733–737.
Harper, G., Hart, D., Moult, S., Hull, R., Geering, A. and Thomas, J., (2005). The
diversity of Banana streak virus isolates in Uganda. Arch. Virol. 150:2407–
2420.
Iskra‐Caruana, M‐L., Baurens, F‐C., Gayral, P. and Chabannes, M. (2010). A four–
partner plant–virus interaction: enemies can also come from within. MPMI.
23:1394–1402.
180
Kubiriba, J., Legg, J.P., Tushemereirwe, W. and Adipala, E. (2001). Disease spread
patterns of Banana streak virus in farmers’ fields in Uganda. Ann. Appl. Biol.
139:31–36.
Kumar, L. and Hanna, R. (2008). Banana bunchy top virus in sub‐Saharan Africa:
established or emerging problem? In Banana and Plantain in Africa:
Harnessing international partnerships to increase research impact pp95–96.
Edited by T. Dubois, S. Hauser, C. Staver, and D. Coyne. ISHS.
Kumar, P.L., Hanna, R., Alabi, O.J., Soko, M.M., Oben, T.T., Vangu, G.H.P. and Naidu,
R.A. (2011). Banana bunchy top virus in sub‐Saharan Africa: Investigations
on virus distribution and diversity. Virus Res. 159:171–182.
Magee, C.J.P. (1927). Investigation on the bunchy top disease of the banana.
Council for scientific and industrial research. Bulletin No. 30. Melbourne.
Mshani, V.E., Soko, M.M., Banda, D.L.N., Mwenebanda, B.M.L., Msosa, T.K.,
Mkankhama and Mwamlima, L. H. (2008). Evaluation of banana and plantain
hybrid cultivars in banana bunchy top affected areas. In Plant protection
progress report for the 2007/2008 season pp23–31. Ministry of Agriculture
and Food Security, Malawi.
Yang, I.C., Hafner, G.J., Revill, P.A., Dale, J.L. and Harding, R.M. (2003). Sequence
diversity of South Pacific isolates of Taro bacilliform virus and the
development of a PCR‐based diagnostic test. Arch. Virol. 148:1957–1968.