Content uploaded by Vaneson Moodley
Author content
All content in this area was uploaded by Vaneson Moodley on May 26, 2019
Content may be subject to copyright.
Title: A Survey of Whitefly-Transmitted Viruses on Tomato Crops in
South Africa
Vaneson Moodleya, Augustine Gubbaa and Paramu L. Mafongoyab
aDiscipline of Plant Pathology, School of Agricultural, Earth and Environmental Sciences,
University of KwaZulu-Natal; bRural Agronomy and Development, School of Agricultural,
Earth and Environmental Sciences, University of KwaZulu-Natal, Scottsville,
Pietermaritzburg, Republic of South Africa
Corresponding Author: Vaneson Moodley, email: vaneson.moodley@gmail.com
Co-author email addresses: Augustine Gubba (GubbaA@ukzn.ac.za); Paramu L. Mafongoya
(mafongoya@ukzn.ac.za)
Abstract
Whitefly-transmitted viruses are a growing threat to modern day agriculture. Their impact on
South Africa’s vegetable industry results in unprecedented economic losses. Tomatoes are
important vegetable crops that are highly susceptible to whiteflies and an almost indispensable
part of meal preparation in many South African homes. Against this background, tomato crops
and nearby weed species in South Africa were surveyed for whitefly-transmitted viruses. In
addition, some pepper crops in adjacent tomato fields were also included in this study. Field
and greenhouse crops were inspected for the development of virus-like symptoms in the
presence of whitefly infestations. Leaf samples exhibiting virus-like symptoms were collected
and analyzed for whitefly-transmitted crini-, torrado-, begomo-, ipomo- and carla viruses using
molecular assays. The identity of each virus positive sample was confirmed by Sanger
sequencing and used in subsequent phylogenetic studies. Tomato chlorosis crinivirus (ToCV),
Tomato torrado virus (ToTV) and Tomato curly stunt begomovirus (ToCSV) were three major
viruses identified in the study. ToCV was the most abundant whitefly-transmitted virus in
South Africa with an overall prevalence of 47.1% (tomatoes) and 21% (weeds). ToCV isolates
from South Africa matched ≥ 97% to isolates from Spain and Sudan. ToTV outbreaks emerged
on tomato crops in the northern parts of South Africa in the presence of abnormally high
whitefly populations. ToTV infections of tomatoes were restricted to the Limpopo province,
however, a second isolate was identified on an unknown arable weed that did not infect nearby
tomato crops. ToTV isolates from South Africa matched 99% and 92.8% to Polish and Italian
isolates and had an overall incidence of 7.5% (tomatoes) and 11% (weeds). ToCSV isolates
from South Africa matched 100% with Mozambican isolates. Phylogenetic analysis showed
that current ToCSV isolates in South Africa were distantly related to a previously identified
South African ToCSV isolate. The disease was identified on tomatoes in three provinces and
had an overall incidence of 9.4%. On the contrary, whitefly-transmitted viruses were not
identified on pepper crops exhibiting virus-like symptoms. In this study, whitefly-transmitted
viruses infecting tomatoes, and some weed species in South Africa were elucidated. These
findings are intended to raise awareness on the impact of whitefly-transmitted viruses in South
Africa’s tomato industry.
Keywords: whitefly-transmitted viruses; epidemiology; survey; tomatoes; prevalence
1.1 Introduction
Pests and disease are constant threats to global agriculture and require the need for new and
adaptive strategies to lessen the economic impact on cropping systems. It is therefore important
to regularly scout fields for the presence of insect pests and the onset of disease. Crop
surveillance provides valuable information on abnormalities that may arise during the growing
season. According to the Food and Agricultural Organization (FAO, 2016), reduced resilience
resulting from decades of agricultural intensification together with globalization, trade and
climate change has significantly contributed to the increase in transboundary plant pests and
diseases. Exotic pests and pathogens that are introduced in other countries easily spread and
are the cause or rising global epidemics. The FAO (2016) further outlines that outbreaks result
in huge crop and pasture losses which threaten the livelihoods of farmers and global food and
nutritional security.
Whiteflies (Hemiptera: Aleyrodidae) are polyphagous sap-sucking pests that reduce plant vigor
by tapping into the phloem tissue and subsequently infecting crops with destructive viral
pathogens. Species such as Bemisia tabaci (Gennadius) and Trialeurodes vaporariorum
(Westwood) are notorious pests of vegetable crops grown in warmer climates and greenhouses
(Martin et al. 2000). De Barro et al. (2011), lists B. tabaci (a cryptic species complex) as one
of the world’s most destructive agricultural pests which has shown resistance to 56 active
ingredients (AIs) to date (Whalon et al. 2017). T. vaporariorum commonly known as the
greenhouse whitefly has been identified as one of the most damaging arthropod pests in
greenhouse production; particularly on tomatoes (McDaniel et al. 2016). The cosmopolitan
greenhouse whitefly is currently resistant to 27 active ingredients according to the Arthropod
Resistance Pesticide Database (Whalon et al. 2017).
In South Africa, whiteflies are a serious threat to vegetable crop production in commercial and
smallholder agribusiness. According to the Agricultural Research Council (ARC)
(http://www.arc.agric.za/arc-ppri/Pages/Insect%20Ecology/General-information-on-
whiteflies-.aspx), B. tabaci (Middle East-Asia Minor 1, formerly biotype B) and T.
vaporariorum are the two major whitefly pests affecting South Africa’s vegetable crop
production. Between 2002 and 2009 Esterhuizen et al. (2013) elucidated the B. tabaci
haplotype diversity in eight of the nine provinces of South Africa. They established the
presence of an exotic, highly invasive Mediterranean species (formerly known as the biotype
Q) on tomatoes in the south-eastern parts of the country, and the widespread prevalence of the
Middle East-Asia Minor 1 species in five of the eight provinces. Subsequently, both were
linked to outbreaks of the Tomato curly stunt begomovirus (ToCSV) in South Africa. The only
published study of whitefly-transmitted viruses in South Africa dates back to 2008 by Pietersen
and colleagues in which they characterized ToCSV as a new whitefly-transmitted begomovirus
infecting tomato crops.
Although whitefly species in the genera Bemisia and Trialeurodes directly damage crops
through feeding it is their ability to acquire and spread viruses that result in the highest
economic crop losses to global agriculture. Navas-Castillo et al. (2011), explain that many of
the emerging viruses in the past two decades limiting vegetable crop production in the tropics,
sub-tropics and temperate parts of the world are transmitted by whiteflies. Additionally, most
whitefly-transmitted viruses belong to the Begomovirus genus in the family Geminiviridae,
although they also vector ipomoviruses (Potyviridae), criniviruses (Closteroviridae),
torradoviruses (Secoviridae) and to a lesser extent, carlaviruses (Betaflexiviridae). B. tabaci
has also been implicated in the emergence of criniviruses (Wisler et al. 1998), ipomoviruses
(Adkins et al. 2009) and torradoviruses (Verbeek et al. 2007) whereas T. vaporariorum is
associated with crinivirus and torradovirus outbreaks. Wintermantel (2004) reports the
increasing incidence of T. vaporariorum from its greenhouse limited habitat to open field
vegetable production.
Rojas and Gilbertson (2008) emphasize that the majority of emerging plant viruses are existing
viruses (identified/unidentified) that have spread to new geographical locations and have
acquired the ability to infect new hosts. Emerging viruses that are newly discovered species
are rare. Novas-Castillo et al. (2011) highlight factors that contribute to emerging plant viruses
which include genetic changes i.e. mutations, recombination or reassortment; changes in vector
populations, and trade/trafficking of infected plant material across geographical locations. In
addition, the density and diversity of vector populations will be influenced by changes in their
environment such as climate variation and human activity.
Whiteflies continue to damage crops and spread diseases in cropping systems throughout the
world. Their impact results in losses that exceed hundreds of millions of dollars to the
agricultural industry. South Africa is experiencing a rising trend in whitefly populations that
very likely led to the emergence of Tomato torrado virus (ToTV), a higher prevalence of
Tomato chlorosis virus (ToCV) and the increased incidence of Tomato curly stunt virus
(ToCSV). Against this background, a national survey was undertaken to identify and determine
the prevalence and distribution of whitefly-transmitted viruses in South Africa’s tomato and
pepper industry.
1.2 Materials and Methods
1.2.1 Virus Surveys and Sample Collection
During the 2015 and 2016 growing season, tomato and pepper farms were surveyed for
whitefly-transmitted viruses. Commercial, smallholder, and some rural farms located
throughout South Africa’s nine provinces were inspected for virus-like symptoms in the
presence of whiteflies. Farms were selected based on information provided by internet searches
and government extension services situated at various locations in each province. Three or four
apical leaves were collected from symptomatic and some asymptomatic crops and nearby
weeds in plastic zip-lock bags and placed on ice. GPS (global positioning system) coordinates
and the level of infestation were recorded at each field and greenhouse site and a subsequent
map was generated using ArcGIS 10.4 mapping software (Figure 1).
Figure 1: Map of sampling sites and whitefly infestation levels in South Africa. The elevation
is represented by areas shaded in light (lower) and dark (higher) areas of grey.
A total of 135 sites were assessed for the presence or absence of whiteflies and assigned a
rating according to the level of infestation i.e. absent (no adults/nymphs), low (an average of <
50 adults/leaf), medium (> 50 but < 150 adults/leaf) or high (> 150 adults/leaf). The average
was determined from whitefly counts of ten randomly sampled tomato leaves per plant within
a 50m×50m plot. A total of 50 plants per plot were used in the screening process. Whiteflies
were subsequently aspirated from the underside of symptomatic leaves and stored in sterile
15ml plastic tubes containing 95% ethanol for basic phenotypic analysis. Each sample was
labeled with a description (of the symptoms) and assigned the corresponding GPS coordinate.
Samples were then transported to the laboratory on dry ice and stored at -80°C pending
analysis.
1.2.2 Nucleic Acid Extraction
Total RNA was extracted from frozen leaf material using a Quick-RNA™ MiniPrep kit (Zymo
Research, USA) according to the manufacturers’ guidelines. Samples were homogenized in
tissue lysis buffer using a bead beater and eluted in 30µl volumes to concentrate the RNA.
Similarly, total DNA was extracted from frozen leaf tissue using Universal Quick-DNA
Miniprep kit (Zymo Research, USA) following the manufacturers’ protocol and eluted in 35µl
volumes. The quality and quantity of each extraction were analyzed using a nanodrop 1000
spectrophotometer (Thermo Fisher Scientific, Massachusetts, U.S.).
1.2.3 Virus Detection
Samples were tested for the presence of begomo-, crini-, torrado-, ipomo- and carla viruses
using reverse transcription-polymerase chain reaction (RT-PCR) for the detection of RNA
viruses and polymerase chain reaction (PCR) to detect DNA viruses. Asymptomatic plants
were also tested for the possibility of latent infections or whereby the viral load may not have
reached significant concentrations to produce distinct symptoms.
The primer sets, annealing temperatures, and expected amplicon sizes for each virus group are
listed in Table 1. Since more than one crinivirus species is capable of infecting different
solanaceous hosts, a multiplex RT-PCR approach was used. Similarly, generic torradovirus
primers were used to detect tomato-infecting torradoviruses using RT-PCR. Ipomoviruses and
carlaviruses were tested using an anchored primer and a subsequent set of universal Potyviridae
and carlavirus primers respectively. The GN47, GN54 and GN55 carlarvirus anchored primers
(Table 1) were used in combination at an equal concentration (5µM) for RT and PCR reactions.
Molecular assays to detect for Tomato yellow leaf curl begomovirus (TYLCV) and Tomato
curly stunt begomovirus (ToCSV) were carried out using a pair of primers (Ty1/Ty2) designed
by Accotto et al. (2000).
Table 1: Primers used to screen for whitefly-transmitted viruses in South Africa.
(+) – Forward primer; (-) – Reverse primer; T(A)°C – Annealing temperature; P(Conc.) – Primer concentration
Anchored primers – M4T (Ipomovirus); GN47, GN54, GN55 (Carlavirus)
cDNA synthesis for crini-, ipomo-, torado- and carla viruses was carried out in 20µl volumes
using a Revert Aid premium reverse transcriptase kit (Thermo Fisher Scientific Inc. USA).
Template RNA was heated at 65°C for 5min and immediately chilled on ice prior to the
addition of a master mix containing 4µl (5X) reaction buffer, a gene-specific primer (10µM)
Group
Primer
5ʹ-3ʹ sequence
T(A)°C
P(Conc.)
Amplicon
Reference
Begomovirus
(TYLCV/ToCSV)
Ty1 (+)
GCCCATGTA(T/C)-CG(A/G)AAGCC
50
10 µM
580 bp
Accotto et al. 2000
Ty2 (-)
GG(A/G)TTAGA(A/G)GCATG(A/C)GTAC
10 µM
Crinivirus
Solanaceous R (-)
GTGTTBGAYAACCAWGTGTT
52
10 µM
Wintermantel and Hladky,
2010
ToCV (+)
GCACCCTGATTGGTTCTAAAC
50 nM
265 bp
TICV (+)
AAGAATGGACCTACCCAG
400 nM
995 bp
BPVY (+)
TGATGTCTGGTTTGATGACGGG
240 nM
643 bp
PYVV (+)
ATCGTTCGTTCTCAACCG
600 nM
514 bp
Torradovirus
Torrado 2F (+)
TGGGATGARTGYAATGTKCT
52
10 µM
515 bp
Verbeek et al. 2012
Torrado 2R (-)
CCWGTCCACCAYTTGCAATT
10 µM
Ipomovirus
M4T
GTTTTCCCAGTCACGAC(T)15
47
10 µM
1700 bp
Chen et al. 2001
S-Primer (+)
GGXAAYAAYAGYGGXCAZCC
10 µM
M4 (-)
GTTTTCCCAGTCACGAC
10 µM
Carlavirus
Carla-Uni Primer (+)
GGAGTAACC (or T) GAGGTGATACC
50
10 µM
120 bp
Badge et al. 1996
GN47
T21A
5 µM
Pappu et al. 1993
GN54
T21C
5 µM
GN55
T21G
5 µM
(reverse primer/anchored primer) (Table 1), 0.5µl ribolock RNase inhibitor (20 U/µL) and 1µl
of reverse transcriptase enzyme (200 U/µL). Reaction volumes were made up to 20µl using
nuclease-free water. Conditions for RT were 42°C for 60min and 70°C for 10min. PCR was
performed using a KAPA2G Fast HotStart ReadyMixPCR kit (KAPA Biosystems, USA). Each
20µl PCR reaction contained 10µl KAPA, the respective primer concentrations listed in Table
1, 30ng cDNA and the required amount of nuclease-free water. Conditions for PCR were: 95°C
for 1min followed by 35 cycles of 95°C for 30s; 47°C – 52°C for 35s (depending on the primer
pair listed in Table 1) and 72°C for 45s following a final elongation of 72°C for 10min. All
PCR products were resolved on a 1.5% agarose gel stained with SYBR Safe DNA gel stain
(Invitrogen, USA).
1.2.4 Cloning, Sequencing and Phylogenetic Analysis
PCR positive samples were selectively excised from the gel and purified using a Zymoclean™
Gel DNA Recovery Kit (Zymo Research, USA). Cloning of DNA fragments was performed
using a TA cloning® kit (Invitrogen, USA) as recommended by the manufacturer. DNA
fragments were ligated to a PCR™2.1 cloning vector and transformed into TOPO 10
chemically competent cells using a heat shock technique. White colony formation on Luria-
Bertani (LB) agar plates containing 40µg/µl X-Gal and 50µg/ml kanamycin were individually
selected and cultured overnight in LB broth (37°C/200rpm). Plasmid extractions were carried
out using a Zyppy™ Plasmid Miniprep Kit (Zymo Research, USA) and two positive clones per
sample were sequenced in both directions at Inqaba Biotec (Pretoria, South Africa) using a
3500xL Genetic Analyzer (Applied Biosystems, USA). Sequence alignments, Blast, and
phylogenetic analysis were carried out using tools in MEGA 7 software (Kumar et al. 2016).
Raw data (forward and reverse sequences) were aligned with MUSCLE. The curated sequences
were matched to known sequences available on the NCBI database using BLAST analysis. A
best-fit model was used to generate optimal parameters for the phylogenetic analysis of each
positively matched virus identified in this study.
1.3 Results
1.3.1 Survey Data
The density of whiteflies varied throughout sampling points across South Africa. Higher
whitefly populations occurred at sites located in the northern parts of South Africa (Figure 1).
Trialeurodes sp. and Bemisia sp. were identified from nymph and adult phenotypic screening.
Limpopo had the highest density of whiteflies and is also the largest producer of tomatoes in
South Africa contributing up to 600 000 tons annually. Whiteflies were not identified at
sampling sites located in the Northern Cape Province (a dry semi-arid part of South Africa
which does not produce a lot of vegetables).
The yield and quality of tomato crops were affected by the overwhelming presence of
whiteflies in most of South Africa’s major cultivation sites (Figure 2). The damage to crops
through feeding was exacerbated by black layers of sooty mold (Figure 2B) which is known to
impede photosynthesis and reduce yields. Tomato crops and nearby weeds infested with dense
whitefly populations (Figure 2A), displayed an array of virus-like symptoms (Figure 2C-2H).
Necrotic spots (Figure 2C), interveinal leaf chlorosis (Figure 2D-2E), stunting/leaf curling
(Figure 2F), and leaf deformation (Figure 2G) were some of the common symptoms observed
on tomato crops and weeds infested with whiteflies. These symptoms were more prevalent in
the northern parts of the country where whitefly populations were generally higher.
Necrotic ‘burnt-like’ symptoms resembling torradovirus disease (Figure 2C) emerged on
tomato crops in the Limpopo province of South Africa in the presence of dense whitefly
populations. Interveinal leaf chlorosis symptoms (Figure 2D; Figure 2E) typically associated
with crinivirus infections was widespread in field and greenhouse-cultivated tomato crops.
Figure 2: The impact of whiteflies on tomato production in South Africa. A: High whitefly
pest pressure on greenhouse-produced tomatoes in the Western Cape province reduces plant
vigor. B: The development of sooty mold (complexes of ascomycetes and fungi imperfecti)
resulting from the honeydew excrement of sucking insects such as whiteflies which impedes
photosynthesis. C: Necrotic lesions surrounded by mild chlorosis at the base of immature
leaves observed on field-grown tomatoes in the Limpopo province typically associated with
whitefly-transmitted torrado viruses. D: Severe interveinal leaf chlorosis, and upward leaf
curling symptoms that generally develop from crinivirus infection. E: Interveinal leaf chlorosis
symptoms on the solanaceous weed Datura stramonium growing among tomato crops and
nearby tomato fields. F: Severe stunting and leaf curling symptoms on tomato crops in the
northern KwaZulu Natal province which typically resemble those associated with begomoviral
infections. G: An unknown arable weed growing among tomato crops in the northern KwaZulu
Natal province. Symptoms included stunting, chlorosis, and leaf deformation. H: Total crop
loss of tomatoes infested with whiteflies at a smallholder farm in the North West Province.
Plants exhibited severe stunting, leaf curling, and interveinal leaf chlorosis symptoms.
These were among the most commonly occurring symptoms throughout most of South Africa’s
major tomato growing areas; however, these symptoms were concentrated in the northern parts
of the country.
Severe stunting and upward leaf curling symptoms typically associated with begomoviral
infections were observed on tomato crops in the north-eastern parts of the KwaZulu Natal
province (Figure 2F) and in the North West province (Figure 2H). The infection rate of field
and greenhouse cultivated tomatoes exhibiting curly stunt symptoms were > 90% at some sites,
which resulted in near total crop loss to some farms in the northern KwaZulu Natal province
and parts of the North West province. To a lesser extent, these symptoms were observed at a
site in the Mpumalanga province situated close to the Swaziland border in the north-eastern
parts of South Africa.
1.3.2 Molecular Identification
The use of a multiplex PCR to identify crinivirus infection on symptomatic tomato crops
produced a 265 bp amplicon (positive for ToCV). ToCV was identified on tomato crops in the
Limpopo, Mpumalanga, Gauteng, Eastern Cape, Free State, and North West provinces.
Limpopo had the highest prevalence of ToCV followed by the Free State, North West, Eastern
Cape and Mpumalanga provinces (based on RT-PCR assays).
Begomoviral infections were positively identified in the northern parts of KwaZulu Natal,
Mpumalanga, and the North West provinces. The outbreak was concentrated in the northern
parts of KwaZulu Natal and parts of the North West province. Crops were tested regardless of
symptom expression to establish disease incidence.
Torradoviruses were identified on field and greenhouse cultivated tomatoes crops and Datura
stramonium in parts of the Limpopo province, and on an unknown arable weed species (Figure
2G) growing among tomato field crops in the northern parts of KwaZulu Natal. A 515 bp PCR
product was expected for samples that tested positive for ToTV infections.
1.3.3 Sequencing
Two PCR positive amplicons for each virus species that was identified at a particular sampling
site were validated by Sanger sequencing. ToCV isolates collected from the northern provinces
i.e. Limpopo, Mpumalanga and the North West province shared a high level of nucleotide
sequence similarity. The subsequent consensus sequence Nel-186Cr (KT989862) from South
Africa (Moodley et al. 2016b) matched 98% to ToCV isolates Tenerife (KJ175084) and Sudan
3 (JN411686) from Spain and Sudan respectively. The consensus sequence ToC1-EL186
(KY593325) from ToCV isolates identified in the Eastern Cape province were similar to those
collected in the northern provinces and matched 97% to Tenerife [(KJ175084); Spain] and
Sudan 3 [(JN411686); Sudan].
The nucleotide sequences of ToTV isolates infecting tomato crops in the Limpopo province
were similar to those identified on the weed D. stramonium. The subsequent consensus
sequence Lim-186 [(KP890356; Moodley et al. 2016a)] was deposited into the NCBI GenBank
database. Lim-186 from the Limpopo province matched 99% to the Polish isolate Wal’03
(EU563947). On the other hand, the ToTV isolate infecting the unknown arable weed species
in northern KwaZulu Natal [(TorKZN-186) KY581570] matched 92.8% to the isolates T795
(KX132809) and Lim-186 from Italy and South Africa respectively.
Sequences of ToCSV from the northern KwaZulu Natal, Mpumalanga and parts of the North
West provinces also shared a high level of nucleotide sequence similarity. The subsequent
consensus sequence from South Africa [(Joz-Beg186) KY290392] matched 100% to
Mozambican isolates Namaacha1111 (KM438746), Chokwe1111 (KM438740), and Baone_9
(KM438741).
1.3.4 Virus Prevalence
A total of 787 tomato, 269 bell pepper and 182 weed samples collected from fifteen species
belonging to six botanical families (Amaranthaceae, Asteraceae, Brassicaceae, Euphorbiaceae,
Malvaceae, and Solanaceae), were tested for whitefly-transmitted viruses (Table 2). ToCV had
the highest overall prevalence (47.1%) in South Africa [identified in six of the nine provinces
with a higher prevalence in some of the major tomato growing regions such as Limpopo
(81.9%)]. In addition, ToCV was present on two commonly occurring weed species i.e. Datura
stramonium and Solanum nigrum in four provinces. ToCV infected S. nigrum was restricted to
the Mpumalanga province with a prevalence of (27.8%), whereas D. stramonium infection was
more widely distributed with a higher prevalence in the Limpopo province [79.2% (Table 2.2)].
Table 2: Prevalence of Tomato chlorosis virus (ToCV), Tomato torrado virus (ToTV), and
Tomato curly stunt virus (ToCSV) on tomato, pepper and weed species in South Africa.
T – Tomato; W – Weed; P – Pepper
ToTV infection of tomato crops and D. stramonium was restricted to the Limpopo province
with a prevalence of 71.1% and 66.7% respectively. To a lesser extent, a ToTV isolate was
identified on an unknown arable weed species (Figure 2G) in the northern parts of KwaZulu
Natal with a prevalence of 13.8% (Table 2). The low overall prevalence of ToTV on tomatoes
Number of Samples
ToCV (+)
ToTV (+)
ToCSV (+)
T
W
P
T
W
P
T
W
P
T
W
P
Limpopo
83
24
0
68 (81.9%)
12 (79.2%)
0
59 (71.1%)
16 (66.7%)
0
0
0
0
Mpumalanga
55
18
47
19 (34.5%)
5 (27.8%)
0
0
0
0
18 (32.7%)
0
0
KwaZulu Natal
147
29
68
0
0
0
0
4 (13.8%)
0
41 (27.9%)
0
0
Eastern Cape
155
38
89
59 (38.1 %)
8 (31.6%)
0
0
0
0
0
0
0
Western Cape
38
8
11
0
0
0
0
0
0
0
0
0
Gauteng
97
33
12
74 (76.3%)
13 (39.4%)
0
0
0
0
0
0
0
North West
132
15
42
87 (65.9%)
0
0
0
0
0
9 (6.8%)
0
0
Free State
80
17
0
64 (80%)
0
0
0
0
0
0
0
0
Northern Cape
0
0
0
0
0
0
0
0
0
0
0
0
Total
787
182
269
371 (47.1%)
38 (21%)
0
59 (7.5%)
20 (11%)
0
68 (9.4%)
0
0
(7.5%) and weeds (11%) in South Africa is a consequence of the limited distribution of the
virus.
ToCSV occurred sporadically in South Africa with an overall prevalence of (9.4%). Tomato
crops infected with ToCSV were confined to the northern parts of the country. In the North
West province, ToCSV was only identified from tomato samples collected at one greenhouse
site, hence the low prevalence (6.8%). On the other hand, some tomato farms located in the
northern KwaZulu Natal province were nearly destroyed as a result of ToCSV infection (> 90%
incidence), however, ToCSV was not identified in other parts of the KwaZulu Natal province
which had an overall prevalence of (27.9%) (Table 2). Field-grown tomato crops in the
Mpumalanga province had the highest overall prevalence of ToCSV (32.7%), however, only
55 samples were tested relative to 144 samples in KwaZulu Natal; indicating that these results
are biased.
1.3.5 Phylogeny
Phylogenetic analysis of ToCV isolates Nel-186Cr and ToC1-EL186 from South Africa
grouped exclusively with Spanish isolates and a Sudanese isolate but did not cluster (Figure
3). The tree topology and branch lengths indicate that both isolates from South Africa were
closely related. The Spanish isolate 2.5, however, did not group with other Spanish or South
African isolates.
All ToTV isolates analyzed in this study diverged from the South African isolate TorKZN-186
(Figure 4). This isolate of ToTV was only identified on an unknown arable weed species
(Figure 2G) whereas Lim-186 was identified on tomatoes and D. stramonium. Lim-186 did not
group with any of the other ToTV isolates analyzed in this study, however, it diverged from
Spanish and Columbian isolates.
ToCSV isolates from Mozambique expressed a wide range of diversity (Figure 5). The South
African ToCSV isolate (Joz-Beg186) from this study clustered with Mozambican isolate
Namaacha1111, however, it was distantly related to the previously identified South African
ToCSV isolate (AF261885). The relationship of Joz-Beg186 with Namaacha1111 was
supported by a bootstrap value of 67%.
Figure 3: Phylogenetic analysis of the partial RNA dependent RNA polymerase (RdRP) gene
located on the RNA 1 molecule of ToCV. South African isolates were studied in conjunction
with all other ToCV isolates currently available on the NCBI database. The evolutionary
history was inferred using the Neighbor-Joining method based on the number of differences
method and a bootstrap of 1000 replicates. The tree was rooted using Tomato infectious
chlorosis virus (isolate TICV-SP5652-8), a closely related species to ToCV.
ToC-Br2 Brazil (JQ952600)
Gr-535 Greece (EU284745)
Merkez Turkey (KY419525)
Kas Turkey (KY419526)
Kebbi Nigeria (MF459657)
Jigawa Nigeria (MF459658)
Isolate 2 Potato Tunisia (KC156621)
Isolate 4 Pepper Tunisia (KC156623)
Isolate 1 Weed Tunisia_ KC156620)
ToCV USA (AY903447)
SDSG China (KC709509)
JJ South Korea (KP137100)
HS South Korea (KP137098)
HP South Korea (KP114530)
JJ5 South Korea (KP114527)
ToCV-BJ China (KC887998)
NJ China (KF018280)
IS29 South Korea (KP114538)
IS17 South Korea (KP114535)
YG South Korea (KP114526)
FERA United Kingdon((KY810786)
2.5 Spain (KJ200304)
Nel-186Cr South Africa (KT989862)
ToC1-EL186 South Africa (KY593325)
Sudan 3 Sudan (JN411686)
Tenerife Spain (KJ175084)
Pl-1-2 Spain (KJ200308)
MM8 Spain (KJ200306)
AT80/99-IC Spain (KJ740256)
AT80/99 Spain (DQ983480)
Tomato infectious chlorosis virus (JQ973707)
Figure 4: Phylogenetic analysis of a region overlapping the first two coat protein genes on the
RNA 2 molecule of ToTV. Two isolates from South Africa were studied in conjunction with
all ToTV isolates currently available on the NCBI database. The evolutionary history was
inferred using the Neighbor-Joining method based on the number of differences method and a
bootstrap of 1000 replicates. The tree was rooted with Tomato marchitez virus (isolate pJL89-M-
R2), a related species of the torradovirus genus.
Figure 5: Phylogenetic analysis of the partial coat protein gene of ToCSV. The recently
identified ToCSV isolate (Joz-Beg186) from South Africa was analyzed together with all
ToCSV isolates currently available on the NCBI database. The evolutionary history was
inferred using the Neighbor-Joining method based on the Jukes-Cantor model and a bootstrap
of 1000 replicates. The tree was rooted using Tomato yellow leaf curl virus (isolate Tom-142),
a closely related species of ToCSV.
Ros Poland (KM114266)
Ros Poland (EU652401)
sec2 Colombia (KJ571199)
Wal03 Poland (EU563947)
T795 Italy (KX132809)
ToTV Spain (DQ388880)
Kra Poland (KJ940974)
ToTV-CE Spain (EU476182)
sec3 Colombia (KJ571200)
sec1 Colombia (KJ571198)
Lim-186 South Africa (KP890356)
TorKZN-186 South Africa (KY581570)
Tomato marchitez virus (KT756877)
Chokwe1121 Mozambique (KM438737)
Chokwe1122 Mozambique (KM438738)
Namaacha 121 Mozambique (KM438739)
Chokwe 522 Mozambique (KM438747)
Chokwe511 Mozambique (KM438745)
Chokwe421 Mozambique (KM438743)
Chokwe 21 Mozambique (KM438744)
Chokwe1111 Mozambique (KM438740)
Boane 9 Mozambique (KM438741)
Boane 211 Mozambique (KM438742)
Boane 162 Mozambique (KM438748)
Joz-Beg186 South Africa (KY290392)
Namaacha1111 Mozambique (KM438746)
Boane 21 Mozambique (KM438735)
Boane 91 Mozambique (KM438736)
Segment A V2 South Africa (AF261885)
Chokwe 91 Mozambique (KM438733)
Chokwe 92 Mozambique (KM438734)
Tomato yellow leaf curl virus (LM651401)
1.4 Discussion
Pests and disease compromise the yield and quality of tomato crops throughout the world. In
South Africa, tomatoes are the second most important vegetable crop that constitutes nearly a
quarter of the total vegetable crop production. The whitefly epidemic in South Africa is a
constant threat to vegetable production and a major limiting factor for tomato growers
throughout the country. Although whiteflies cause mechanical damage to crops during feeding,
it is their role as viral vectors that severely reduce profit margins and threaten global food
security. According to Navas-Castillo et al. (2011), whiteflies typically transmit
begomoviruses, criniviruses, carlaviruses, ipomoviruses, and some carlaviruses. ToCV, ToTV,
and ToCSV were the only three whitefly-transmitted viruses identified on tomato crops and
some weed species in South Africa. B. tabaci was the predominant vector (from adult and
nymph phenotypic screening) throughout most of South Africa’s tomato production areas,
although T. vaporariorum was more abundant in greenhouse-produced tomatoes. The majority
of whitefly-transmitted viruses (confirmed with molecular assays) in South Africa were often
accompanied by the presence of B. tabaci populations that were collected and phenotyped from
the abaxial leaf surface of plants displaying virus-like symptoms.
Our results showed that ToCV was the predominant whitefly-transmitted virus infecting
tomato crops (47.1%) and solanaceous weeds (21%) in South Africa (identified in six of South
Africa’s nine provinces). Although many of these sampling sites were located within major
tomato growing regions in the country, ToCV infections were concentrated in the northern
parts of South Africa where conditions are significantly warmer and dryer. Phylogenetic
analysis showed that both ToCV isolates from South Africa clustered within a clade of Spanish
and Sudanese isolates. The branch lengths indicated minimal genetic change between South
African ToCV isolates in this group. The tree topology suggests that the ToCV isolates from
Sudan and South Africa may be of Spanish origin. Chinese and South Korean ToCV isolates
expressed a higher degree of diversity among one another, whereas some isolates from the UK,
USA, Turkey, Spain, and China did not group with other ToCV isolates. This diversity is often
a consequence of mutations, vector dynamics, environmental stress, and selection pressure.
Sanjuán et al. (2010) explain that the fast error-prone replicative nature of RNA viruses may
be linked to their rapid evolutionary potential. On the other hand, ToCV isolates from Turkey,
Brazil, Greece Nigeria and Tunisia grouped within a separate clade. ToCV isolates in West
Africa and Tunisia are highly diverse to those from Spain, East Africa, and South Africa. These
variations are possibly linked to international trade relations. ToCV may have spread from
Greece and Brazil to Nigeria, Turkey, and Tunisia, gaining diversity/complexity. In South
Africa, ToCV may have spread from Spain, the Mediterranean, and Sudan.
ToTV was only identified on tomato crops growing in the Limpopo province (located north of
South Africa). The typical burnt-like symptoms exhibited by tomato crops that are infected
with a torradovirus [described by Verbeek et al. (2007)] emerged in the presence of abnormally
dense whitefly populations. The warm, low rainfall climate in the Limpopo province seemed
to favor the whitefly epidemic which likely led to the emergence of ToTV in South Africa.
ToTV infections were not limited to field or greenhouse cultivation and the same isolate of
ToTV (Lim-186) infecting tomato crops was subsequently identified on D. stramonium weeds
growing alongside rows of tomatoes. In addition, ToTV was identified on an unknown arable
weed species (TorKZN-186) growing among tomato field crops in the northern parts of
KwaZulu Natal. This isolate of ToTV was not identified on any of the tomato crops in this
province and may likely be a non-tomato infecting isolate of ToTV (requires further biological
assays for validation). The low overall prevalence of ToTV on tomatoes (7.5%) and weeds
(11%) in South Africa results from the restricted distribution of the disease to the Limpopo
province.
On the contrary, TorKZN-186 identified on the unknown arable weed species did not share any
relation with other ToTV isolates including Lim-186 which was also identified in South Africa.
Interestingly, all previously identified ToTV isolates from around the world diverged from
TorKZN-186. The tree topology shows the evolutionary trends and diversity among all of the
presently identified ToTV isolates throughout the world. Although TorKZN-186 matched
92.8% to Italian isolate T795 and South African isolate Lim-186, it is illustrated as an
‘outgroup’ or ‘root’ on the phylogenetic tree suggesting that it may be an ancestral isolate of
ToTV. The branch lengths indicate the evolutionary diversity of TorKZN-186 in relation to
other ToTV isolates in this study. Our results suggest that a weed may have been the primary
host of torradoviruses. According to Elena and Lalić (2013), the frequency of host range
mutations (not considering the effects of genetic drift) within the primary host is dependent on
a balance between the rate at which mutations and recombination are produced, and the fitness
advantage or disadvantages they may have in the primary host. In other words, if the host range
mutations are disadvantageous, the frequency of mutations is lower and hence the likelihood
of emergence is low. On the contrary, if the host range mutations are impartial or beneficial,
the frequency of mutations will increase and so will the likelihood of emergence. Hence, in
theory, the emergence of ToTV on tomato crops in various parts of the world may be the result
of standing genetic variation in the primary host i.e. a weed species that facilitated successful
replication in a new host i.e. tomatoes after infrequent spillovers (Elena et al. 2011). These are
a few of the reasons why some viruses such as ToTV cause outbreaks and epidemics that are
limited in time and space versus those such as ToCV that cause pandemics.
ToCSV was identified in northern KwaZulu Natal, Mpumalanga and the North West provinces.
In the North West province, ToCSV was limited to greenhouse-cultivated tomato crops,
whereas in Mpumalanga and KwaZulu Natal, ToCSV was predominant in open field
production. Although ToCSV was not highly prevalent among tomato crops in South Africa
(9.4%), a high incidence of the disease occurred in affected areas. In both cases, ≥ 90%
infection rate and crop loss were observed. The phylogram showed that Joz-Beg186 grouped
with various Mozambican isolates but clustered with Namaacha1111. Other Mozambican
isolates diverge from this group including a previously identified ToCSV isolate from South
Africa (AF261885) which dates back to the early 2000s in which ToCSV was first reported as
a new tomato-infecting begomovirus species (Pietersen et al. 2000). In a follow-up study,
Pietersen and colleagues (2008) described ToCSV disease on tomato crops that emerged in the
north-eastern parts of South Africa (Mpumalanga province) as early as 1997.
ToCSV disease spread north and south from the epicenter in the Onderberg area which is
situated close to the Mozambican border. The outbreak of ToCSV in South Africa may have
originated in Mozambique based on the evolutionary diversity of all presently known ToCSV
isolates. The level of genetic diversity among ToCSV isolates is likely the result of mutations.
In general, DNA viruses such as ToCSV mutates at substantially lower rates compared to that
of RNA viruses. Sanjuán and Domingo-Calap (2016), explain that the genetic diversity of
viruses is associated with multiple virus and host dependent processes in addition to specific
selective pressures. They concluded that clarification into the mutational mechanisms of small
DNA viruses is currently a challenge in virus molecular biology and evolution. The spread of
ToCSV between South Africa and Mozambique is likely facilitated at the Maputo boarder
(located west of Mozambique) into Mpumalanga and the Kosi Bay border (located south of
Mozambique) into the northern parts of KwaZulu Natal where it recurred over the past twenty
years. As with ToTV, the occurrence of ToCSV was restricted to the northern regions of South
Africa. Interestingly, ToCSV has never spread from Mozambique and South Africa to other
parts of the world.
1.5 Conclusion
The tomato industry in South Africa is severely impacted by the progression of insect vectors.
Whiteflies are a key constraint to vegetable crop production and are especially challenging for
tomato growers as they are highly susceptible to whitefly infestation and whitefly-transmitted
viruses. Although farming practices have greatly evolved to cope with the onset of pests and
diseases, their persistence in South Africa’s agricultural industry has spawned a playground for
vector-borne viruses. Weeds such as D. stramonium and S. nigrum are alternative hosts and
therefore play a significant role in the epidemiology of whitefly-transmitted viruses. Their
abundance throughout South Africa’s agro-ecosystems raises concern as potential sources of
inoculum.
It is recommended that whiteflies are managed using a revised combination of cultural,
biological, and chemical methods. The rapidly evolving nature of whiteflies and their
associated pathogens require a collaborative multidisciplinary effort to effectively manage their
impact on global agriculture. It is the responsibility of policymakers to stringently regulate the
movement of plant and seed material into South Africa. In addition, farmers should refrain
from the exchange or purchase of seed and vegetative propagules that are not certified
vector/disease free. The information generated from this study is intended to provide a
foundation for follow-up research that is targeted at reducing the effects of whitefly-transmitted
viruses in South Africa.
Acknowledgments
The authors would like to express their gratitude to the rural and commercial farming
communities, and government extension workers throughout South Africa for their courtesy,
time, and assistance during this study. We are also grateful to the National Research Foundation
(NRF) of South Africa (Grant number 86893) for their financial support throughout this study
(2014 – 2017).
Funding: This work was supported by the National Research Foundation (NRF) of South Africa
(grant number 86893).
References:
Accotto, G.P., Navas-Castillo, J., Noris, E., Moriones, E., Louro, D., 2000. Typing of Tomato
yellow leaf curl viruses in Europe. Eur. J. Plant Pathol. 106, 179-186.
https://doi.org/10.1023/A:100873602.
Adkins S., Webb, S.E., Roberts, P.D., Kousik, C.S., Stansly, P.A., Bruton, B.D., Achor, D.,
Muchovej, R.M., Baker, C.A., 2009. A Review of Ipomoviruses and Watermelon Vine Decline
in Florida, in: Stansly, P., Naranjo, S. (Eds.), Bemisia: Bionomics and Management of a Global
Pest. Springer, Dordrecht.
Badge, J., Brunt, A., Carson, R., Dagless, E., Karamagioli. M., Phillips, S., Seal, S., Turner,
R., Foster, G.D., 1996. A carlavirus-specific PCR primer and partial nucleotide sequence
provides further evidence for the recognition of cowpea mild mottle virus as a whitefly
transmitted carlavirus. Eur. J. Plant Pathol. 102, 305-310.
https://doi.org/10.1007/BF01877970.
Chen, J., Chen, J., Adams, M.J., 2001. A universal PCR primer to detect members of the
Potyviridae and its use to examine the taxonomic status of several members of the family. Arch.
Virol. 146, 757-766. https://doi: 10.1007/s11262-008-0284-3.
De Barro, P.J., Liu, S-S., Boykin, L.M., Dinsdale, A.B., 2011. Bemisia tabaci: a statement of
species status. Annu. Rev. Entomol. 56, 1-19. https://doi.org/10.1146/annurev-ento-112408-
085504
Elena, S.F., Bedhomme, S., Carrasco, P., Cuevas, J.M., de la Iglesia, F., Lafforgue, G., Lalić,
J., Pròsper, À., Tromas, N., Zwart, M.P., 2011. The evolutionary genetics of emerging plant
RNA viruses. Mol. Plant Microbe Interact. 24, 287-93. https://doi: 10.1094/MPMI-09-10-
0214.
Elena, S.F., Lalić, J., 2013. Plant RNA virus fitness predictability: contribution of genetic and
environmental factors. Plant Pathol. 62, 10-18. https://doi.org/10.1111/ppa.12102
Esterhuizen, L., Mabasa, K., van Heerden, S., Czosnek, H., Brown, J., van Heerden, H., Rey,
M.E.C., 2013. Genetic identification of members of the cryptic species complex from South
Africa reveals native and introduced haplotypes. J. Appl. Entomol. 137(1-2), 122-135.
https://doi.org/10.1111/j.1439-0418.2012.01720.x.
FAO (Food and Agriculture Organization), 2016. http://www.fao.org/emergencies/emergency-
types/plant-pests-and-diseases/en/ (accessed 30 January 2017).
Kumar, S., Stecher, G., Tamura, K., 2016. MEGA7: Molecular Evolutionary Genetics Analysis
version 7.0 for bigger datasets. Mol. Biol. Evol. 33, 1870-1874. https://doi
10.1093/molbev/msw054.
Martin, J.H., Mifsud, D., Rapisarda, C., 2000. The whiteflies (Hemiptera: Aleyrodidae) of
Europe and the Mediterranean Basin. Bull. Entomol. Res. 90, 407-448.
https://doi.org/10.1017/S0007485300000547
McDaniel, T., Tosh, C.R., Gatehouse, A.M.R., George, D., Robson, M., Brogan, B., 2016.
Novel resistance mechanisms of a wild tomato against the glasshouse whitefly. Agron. Sustain.
Dev. 36, 14. https://doi.org/10.1007/s13593-016-0351-4.
Moodley, V. Gubba, A., Mafongoya, P.L., 2016a. First Report of Tomato torrado virus (ToTV)
in South Africa. Plant Dis. 100(1), 231. https://doi.org/10.1094/PDIS-06-15-0683-PDN.
Moodley, V. Gubba, A., Mafongoya, P.L., 2016b. Occurrence of Tomato chlorosis virus
(ToCV) on Datura stramonium near tomato crops in South Africa. Plant Dis. 100(7), 1512.
https://doi.org/10.1094/PDIS-01-16-0131-PDN.
Navas-Castillo, J., Fiallo-Olive, E., Sanchez-Campos, S., 2011. Emerging virus diseases
transmitted by whiteflies. Annu. Rev. Phytopathol. 49, 219-48.
https://doi.org/10.1146/annurev-phyto-072910-095235
Pappu, S.S., Brand, R., Pappu, H.R., Rybicki, E.P., Gough, K.H., Frenkel, M.J., Niblett, C.L.,
1993. A polymerase chain reaction method adapted for selected amplification and cloning of 3
p sequences of potyviral genomes: application to Dasheen mosaic virus. J. Virol. Methods. 41,
9-20. https://doi.org/10.1016/0166-0934(93)90158-N.
Pietersen, G., Idris, A.M., Krüger, K., Brown, J.K., 2000. Tomato curly stunt virus, a new
begomovirus of tomato within the Tomato yellow leaf curl virus-IS cluster in South Africa.
Plant Dis. 84(7), 810. http://dx.doi.org/10.1094/PDIS.2000.84.7.810B.
Pietersen, G., Idris, A.M., Krüger, K., Brown, J.K., 2008. Characterization of Tomato curly
stunt virus: a new tomato-infecting begomovirus from South Africa. Plant Pathol. 57(5), 809-
818. . https://doi.org/10.1111/j.1365-3059.2008.01882.x.
Rojas, M.R., Gilbertson, R.L., 2008. Emerging plant viruses: a diversity of mechanisms and
opportunities, in: Roossinck, M.J. (Eds.), Plant Virus Evolution, Berlin, Heidelberg: Springer-
Verlag, pp. 27-51.
Sanjuán, R., Nebot, M.R., Chirico, N., Mansky, L.M., Belshaw, R., 2010. Viral mutation rates.
J. Virol. 84, 9733-48. https://doi.org/10.1128/JVI.00694-10.
Sanjuán, R., Domingo-Calap, P., 2016. Mechanisms of viral mutation. Cell. Mol. Life Sci.
73(23), 4433-4448.
Verbeek, M., Dullemans, A.M. van den Heuvel, J.F.J.M., Maris, P.C., van der Vlugt, R.A.A.,
2007. Identification and characterization of tomato torrado virus, a new plant picorna-like virus
from tomato. Arch. Virol. 152, 881-890. https://doi.org/10.1007/s00705-006-0917-6.
Verbeek, M., Tang, J., Ward, L.I., 2012. Two generic PCR primer sets for the detection of
members of the genus Torradovirus. J. Virol. Methods. 185, 184-188.
https://doi.org/10.1016/j.jviromet.2012.06.020
Whalon, M.E., Mota-Sanchez, D., Hollingworth, R.M. 2017. Michigan State University,
Arthropod Pesticide Resistance Database. http://www.pesticideresistance.com/ (accessed 31
January 2017).
Wintermantel, W.M., 2004. Emergence of Greenhouse Whitefly (Trialeurodes vaporariorum)
Transmitted Criniviruses as Threats to Vegetable and Fruit Production in North America.
APSnet Features. Online. https://doi.org/10.1094/APSnetFeature-2004-0604.
Wintermantel, W.M., Hladky, L.L., 2010. Methods for detection and differentiation of existing
and new crinivirus species through multiplex and degenerate primer RT-PCR. J. Virol.
Methods. 170, 106-114. https://doi.org/10.1016/j.jviromet.2010.09.008.
Wisler, G.C., Li, R.H., Liu, H-Y., Lowry, D.S. and Duffus, J.E. (1998). Tomato chlorosis virus:
A new whitefly-transmitted, phloem-limited, bipartite closterovirus of tomato.
Phytopathology. 88, 402-409. https://doi.org/10.1094/PHYTO.1998.88.5.402.
