Tomato yellow leaf curl virus
(leaf curl)

The wide global distribution of tomato crops and the dramatic outbreaks of the populations of the TYLCV vector, the whitefly Bemisia tabaci, led to a pandemic of this devastating disease. The virus probably arose in the Middle East between the 1930s and 1950s. Its global invasion began in the 1980s after the emergence of two strains: TYLCV-IL and TYLCV-Mld. The long-distance transportation of viruliferous whiteflies contaminating commercial shipments of tomato seedlings and ornamentals is probably the major reason for the virus pandemic (Caciagli, 2007). Sequence analyses allowed Lefeuvre et al. (2010) to trace the history of TYLCV spread. For instance, TYLCV-IL has invaded the Americas at least twice, once from the Mediterranean basin in 1992-1994 and once from Asia (a descendant of imported Middle Eastern TYLCV) in 1999-2003. As a result the estimated losses caused by TYLCV  reached about 20% of tomato production in the USA, and 30-100% in the Caribbean Islands, Mexico, Central America and Venezuela. Therefore several countries (Australia, EU) have established severe quarantine measures to control the whitefly vector.

Europe

France: Reported in 1999 in a single field in the Camargue district. Surveys conducted in 2000 have indicated that the eradication measures taken have been successful (Dalmon et al., 2000; Lepoivre, 2001).

Greece: In late summer 2000, tomatoes grown in greenhouses in several locations in Crete, Attiki and southern Peloponnese showed severe TYLCV symptoms. All greenhouses with infected plants were infested with high populations of Bemisia tabaci. Partial sequencing indicated identity with the TYLCV strain from Israel (Avgelis et al., 2001).

Italy: First recorded in Sardinia in 1988 (Gallitelli et al., 1991), then in Sicily in 1989 (Credi et al., 1989) and in Calabria in 1991 (Polizzi and Areddia, 1992). The disease is associated with large populations of B. tabaci (Rapisarda, 1990). Two different, but related, isolates have been sequenced: from Sardinia (TYLCSV, Kheyr-Pour et al., 1991) and from Sicily (TYLCSV-Sic, Crespi et al., 1995).

Portugal: In late summer 1995, an epidemic outbreak of a disease associated with B. tabaci seriously affected tomato crops in the Algarve, a region in southern Portugal where tomatoes are cultivated year round (Louro et al., 1996). The disease occurred mainly in greenhouse crops where up to 100% of autumn crops were affected and yield was drastically reduced. So far the disease appears to be limited to the Algarve region. Sequencing has indicated that the virus (TYLCV-PT) is genetically related to TYLCV from Israel.

Spain: In 1992, a virus closely related to the Sardinian isolate of TYLCV was found in tomato fields in Almeria and Malaga (Moriones et al., 1993). A survey in 1997 showed that both the Israeli (TYLCV) and Sardinian isolates (TYLCSV) are widely distributed in southern Spain (Sanchez-Campos et al., 1999). Recombinants between the two viruses have been found (Monci et al., 2001). In 1999, TYLCV was reported in Phaseolus bean (Navas-Castillo et al., 1999). TYLCV has recently been diagnosed in the Canary Islands (Font et al., 2000).

Switzerland: Present, but identity has not been confirmed by sequencing (Pelet, 1992).

Asia

Bangladesh, Laos, Malaysia, Myanmar and Vietnam: Present, identified by sequencing. Five distinguished TYLCV isolates were identified, only that from Bangladesh had a bipartite genome (Green et al., 2001).

Bahrain: Present; identity has not been confirmed by sequencing (Traboulsi, 1994).

China: Increasingly prevalent and leading to serious yield losses especially in the south-western provinces of Yunan and Guangxi (Yin et al., 2001). Sequencing showed that TYLCV from China (TYLCCNV) is different from the known Mediterranean and Asian TYLCV isolates (Yongping et al., 2008).

Cyprus: Observed for the first time in 1974. Endemic in the southern coastal zone (Ioannou, 1987).

Iran: Prevalent in the southern provinces of Iran (Hajimorad et al., 1996). Sequencing indicated that TYLCV from Iran (TYLCV-IR) is related to the Middle Eastern TYLCV isolates (Bananej et al., 1998). An important centre of TYLCV diversity (Lefeuvre et al., 2010).

Iraq: Present; identity not confirmed by sequencing (Wilson et al., 1981).

Israel: Present since the early 1960s in the Jordan valley and the coastal plain (Cohen and Harpaz, 1964; Cohen and Nitzany, 1966; Czosnek et al., 1988). It is the most significant factor reducing yield in summer and autumn. The first TYLCV isolate to be sequenced (Navot et al., 1991).

Japan: Restricted distribution on tomato. Seven isolates have been sequenced (see Notes on taxonomy and Nomenclature; closely related to TYLCV from Israel (Kato et al., 1998).

Jordan: Present in the Jordan Valley; affects productivity in open field and greenhouse (Makkouk, 1978). Sequencing showed a complex of TYLCV-IL and TYLCSV-ES (Anfoka et al., 2008).

Korea: Rapidly spread to most regions of the Southern Korean peninsula after 2008. Sequencing of TYLCV and of mitochondrial cytochrome oxydase I from B. tabaci showed that TYLCV and its vector were introduced from Japan (Lee et al., 2010).

Kuwait: Widespread, causing a devastating disease of field-grown tomatoes since 1993 (Montasser et al., 1999).

Lebanon: Widespread in the coastal plains (Makkouk et al., 1979). The virus is closely related to other Middle Eastern TYLCV isolates (Abou-Jawdah et al., 1999).

Oman: Present; identity not confirmed by sequencing (Zouba et al., 1993; Azam et al., 1997).

Saudi Arabia: Present in the Al-Kharj and Qasim regions where it reaches epidemic proportions in summer plantations (Mazyad et al., 1979). Two different viruses have been isolated and partially sequenced: a Northern isolate (TYLCV-NSA) related to TYLCV from Israel, and a Southern isolate (TYLCSAV-SSA) (Hong and Harrison, 1995).

Thailand: A TYLCV disease broke out in 1978 associated with large populations of B. tabaci (Thanapase et al., 1983; Thongrit et al., 1986). TYLCV from Thailand (TYLCTHV) possesses a bipartite genome (Attathom et al., 1994; Rochester et al., 1994) whereas the other known TYLCV isolates are monopartite (Navot et al., 1991).

Turkey: Observed first in the Cukurova region, Adana province (Yilmaz et al., 1980) and more recently (1993) in the Aegean region. An isolate was found to be almost identical to the TYLCV from Israel (Morris, 1997).

Yemen: Increasing whitefly-related virus problems have been observed in tomato-growing regions since the 1970s. The virus is present in the Abayan and Hadramauvat regions. Partial sequencing has indicated that TYLCV from Yemen (TYLCYV) is distinct from the other TYLCV isolates (Bedford et al., 1994).

Africa

Algeria: Found in glasshouses on tomato and green capsicums, in association with large whitefly populations (Kerkadi et al., 1998). Identity not confirmed by sequencing.

Burkina Faso: The virus is found in many parts of the country but the incidence of disease varies from year to year; identity not confirmed by sequencing (Konate et al., 1995).

Cape Verde: The most important disease affecting tomato (Defrancq D'hondt and Russo, 1985). Omnipresent on Santiago Island (Czosnek and Laterrot, 1997); its identity has not been confirmed by sequencing.

Côte d’Ivoire: Virus present in the Bouake region (Czosnek and Laterrot, 1997); identification not confirmed by sequencing.

Egypt: TYLCV has invaded tomato plantations throughout lower and middle Egypt since 1989. Widespread in Fayoum, Ismailia and Giza (Mazyad et al., 1986), the virus is quasi-identical to the TYLCV isolate from Israel (Nakhla et al., 1993).

Libya: Present (Traboulsi, 1994); identity not confirmed by sequencing.

Mali: Present in the Bamako region (Czosnek and Laterrot, 1997); identity confirmed by sequencing (Zhou et al., 2008).

Morocco: First found in 1998 on tomato crops in the coastal region near Casablanca, TYLCV was identified by sequencing (Peterschmit et al., 1999). At the same time, symptoms were also seen in tomato crops in the north-eastern region of Morocco, causing crop losses of between 20 and 100%. Probably imported into Morocco on grafted tomato plants from the Netherlands. It constitutes a major problem today (Monci et al., 2000).

Nigeria: Present in the Zaria region. Identification confirmed by sequencing (Hong and Harrison, 1995). TYLCV from Nigeria (TYLCNV) is not closely related to other TYLCV isolates.

Réunion Island: TYLCV-Mld was introduced in 1997 and TYLCV-IL in 2004. A 5-year survey (2004-2008) of the viral population implicated in the tomato yellow leaf curl disease showed that TYLCV-IL was found to rapidly displace TYLCV-Mld. In 2008, TYLCV-Mld was only found in co-infections with TYLCV-IL (Delatte et al., 2007).

Senegal: Widespread along the Senegal River, Casamance and Dakar regions where many tomato fields have been abandoned (Defrancq D'hondt and Russo, 1985).

Sudan: Identified in the early 1960s (Yassin and Nour, 1965); widespread in tomato-growing regions (Yassin, 1989). Two distinct viral genotypes were identified in the same tomato plant collected from Gezira, Sudan; a third genotype was identified in tomato samples collected in Shambat (Idris and Brown, 2005).

Tanzania: Widespread and economically important. A virus species distinct from the Mediterranean isolates has been sequenced (Chiang et al., 1997).

Tunisia: Present in the regions of Tunis, Sousse and Southern oasis (Cherif and Russo, 1983; Czosnek and Laterrot, 1997). Identity was confirmed by sequencing (Chouchane et al., 2007).

Western Hemisphere

Bahamas: A virus with a sequence similar to the TYLCV isolate from Israel has been identified (Sinisterra et al., 2000).

Cuba: TYLCV was first detected in the early 1990s and is now widespread. Sequencing indicated a close relationship with the TYLCV isolate from Israel (Martinez-Zubiaur et al., 1996; Ramos et al., 1996). It has also been reported in pepper (Quiñones et al., 2002).

Dominican Republic: The virus was first identified as TYLCV in 1992. Sequencing has indicated that the virus is closely related to TYLCV from Israel (Nakhla et al., 1994; Polston et al., 1994). Widespread.

Grenada: In 2007, severe symptoms of tomato yellow leaf curl disease were observed on tomato in six sites on Grenada Island. Sequencing confirmed the presence of TYLCV-IL (Lett et al., 2011).

Jamaica: The most widespread tomato disease, often resulting in 100% crop loss. Since 1991, farmers in south St. Elizabeth have reported losses in their tomato crops to 'jherri curl' disease, which is caused by TYLCV. Sequencing has indicated that this virus is similar to the Eastern Mediterranean TYLCV isolates (Wernecke et al., 1997).

Mexico: In 1999, samples collected in Yucatan were found to be infected by TYLCV (Ascencio-Ibanez et al., 1999). In 2005, TYLCV was identified in Sinaloa (Brown and Idris, 2006).The current incidence of the disease is low and the virus is spreading slowly. The TYLCV isolates from Mexico are closely related to the Eastern Mediterranean TYLCV (Duffy and Holmes, 2007).

Puerto Rico: Found in 2001 in one tomato field on the southern coast of the island (Bird et al., 2001).

Trinidad and Tobago: Widespread, causes a serious threat to the tomato industry (Umaharan et al., 1998). Identity not confirmed by sequencing.

USA

Florida: The first report of TYLCV in the USA came from commercial plantings from Virginia. The tomato plants had been produced in a screenhouse in Manatee county, Florida (Polston, 1998). In July 1997, symptoms characteristic of TYLCV were observed on one tomato plant in a field in Collier County and on several tomato plants in a garden centre in Sarasota. Sequencing indicated that the virus was closely related to TYLCV-IL (Polston et al., 1999a, b). In October 1998, the virus was found in Gadsen County, northern Florida. Spreading, now present in almost all tomato-growing counties (Polston et al., 1999a).

Georgia: First found in Decatur County (south Georgia), with few occurrences since. Identity has been confirmed by sequencing (Momol et al., 1999).

Louisiana: Present, confirmed by sequencing (Valverde et al., 2001).

North Carolina: Symptoms observed in the summers of 2000 and 2001 in Henderson County. The sequence was closely related to the Eastern Mediterranean TYLCV isolates (Polston et al., 2002).

Mississipi: In January 2001, mild symptoms were observed in a greenhouse tomato production operation in east-central Mississippi. Whiteflies were present in the greenhouse during the previous month, but in relatively low numbers. PCR indicated the presence of TYLCV but the strain was not determined (Ingram and Henn, 2001).

Arizona: TYLCV found in the region of Phoenix in 2006, close to TYLCV-IL (Idris et al., 2007).

Venezuela: During 2004, tomato plants showing symptoms similar to those of TYLCV were observed in commercial fields in Zulia state. Sequencing of PCR amplicons from tomato plants sampled in the field showed the presence of TYLCV homologous to TYLCV-Mld from Spain and Portugal and TYLCV from Israel and Mexico (Zambrano et al., 2007).

The domesticated tomato (Solanum lycopersicum) is the primary host of TYLCV. Most of the wild tomato species such as S. chilense, S. habrochaites, S. peruvianum and S. pimpinellifolium include accessions that are symptomless carriers (Zakay et al., 1991) or are immune (Vidavsky and Czosnek, 1998) to the virus. Several cultivated plants are hosts of TYLCV and present severe symptoms upon whitefly-mediated inoculation: bean (Phaseolus vulgaris), petunia (Petunia hybrida) and lisianthus (Eustoma grandiflorum). Weeds such as Datura stramonium and Cynanchum acutum present distinct symptoms whereas others such as Malva parviflora are symptomless carriers. A recent survey in Cyprus showed that 49 different species belonging to the families Amaranthaceae, Chenopodiaceae, Compositae, Convolvulaceae, Cruciferae, Euphorbiaceae, Geraniaceae, Leguminosae, Malvaceae, Orobanchaceae, Plantaginaceae, Primulaceae, Solanaceae, Umbelliferae and Urticaceae were TYLCV hosts (Papayiannis et al., 2011). Plants used to rear whiteflies, such as cotton and aubergine, are immune to the virus. Experimental hosts of the virus include tomato (S. lycopersicum) and jimsonweed (D. stramonium).

Transmission

TYLCV is transmitted by the whitefly, Bemisia tabaci (Cohen and Harpaz, 1964), which is commonly found in tropical and subtropical countries. B. tabaci has a very wide host range of at least six plant families (Cohen and Antignus, 1994). Colonies of B. tabaci are usually established under laboratory conditions on TYLCV non-host plants such as cotton, aubergine or jimsonweed (Datura stramonium) in a greenhouse or in a growth chamber.

Virus-Vector Relationship

TYLCV is transmitted naturally by B. tabaci in a persistent manner. The efficient acquisition access period is 15-30 minutes, the latent period is 8-24 hours, and the efficient inoculation access period is at least 15 minutes (Cohen and Nitzany, 1966; Mansour and Al-Musa, 1992; Mehta et al., 1994; Ghanim et al., 2001). The virus can be detected in every stage of vector development (Ghanim et al., 1998). Female B. tabaci are more efficient vectors than males; transmission capacities decrease as the insect ages (Czosnek et al., 2001). Following a 48 h acquisition access period on infected tomato, TYLCV DNA remains associated with B. tabaci for the entire adult life of the insect; infectivity decreases with time, but remains significant (Rubinstein and Czosnek, 1997). Symptoms develop on inoculated tomato seedlings 2 to 3 weeks after insect-mediated inoculation. The presence of TYLCV in the egg of the whitefly vector suggests transovarial passage (Ghanim et al., 1998).

Epidemiology

Tomato yellow leaf curl disease can be observed in tomato fields throughout the affected regions. The virus is transmitted to tomato plants after vector feeding on infected tomato plants or alternative hosts. The disease is first observed on tomato seedlings about 3 weeks after transplanting. Disease incidence increases rapidly and can reach 100% infection at harvest. In nature, the virus is only transmitted by the whitefly B. tabaci. In affected regions, the incidence of the disease is directly correlated with the pressure of the whitefly population (Mazyad et al., 1979).

TYLCV epidemiology has been studied thoroughly in the Jordan Valley, Israel (Cohen, 1990). B. tabaci populations start to increase in May-June and peak in September-October and decrease to minimum levels in December. The gender prevalent in the whitefly population varies from predominantly males in the winter to predominantly females in the summer. The weed Cynanchum acutum seems to be the overwintering reservoir of the virus, although it is a poor host for B. tabaci. There is a positive correlation between B. tabaci population size and the spread of TYLCV. Winds influence the spread of the virus. The main flight activity takes place during the morning hours, sometimes with a short peak in the late afternoon. The longest flight distance measured was 7 km. During the peak whitefly season, which corresponds to the time of tomato planting (August-September), 2000 to 20,000 whiteflies land weekly on each square metre, thus ensuring high levels of infection within a short time. At this time, the maximum level of whiteflies able to transmit TYLCV to test plants in the laboratory is 4-5% of the population present in the field.

Using DNA probes, the time course of infection has been followed recently in the coastal plain, Israel, where TYLCV infection and whitefly populations are at their peak during August to October (Vidavski et al., 1998). Samples were taken every 7-10 days from the shoot apex of 122 plants up to 89 days after transplanting and tested for the presence of viral DNA using the squash blot hybridization procedure (Navot et al., 1989). Inoculation in the field seems to be at random, as the location of plants free of viral DNA does not point to 'hot spots' of escapes. Although the field was swarming with whiteflies from the first day of planting, only 32% of plants exhibited detectable amounts of viral DNA 17 days after transplanting. This value increased to 64% 39 days after transplanting and to 90% 60 days thereafter. Thus, even 3 months after planting, about 10% of the plants did not contain detectable amounts of viral DNA. Such a high level of escapes may interfere in the selection of resistant/tolerant individuals and makes progress in the breeding programme problematic.

Accurate diagnosis of TYLCV is best achieved using methods based on virus-specific DNA probes and PCR primers derived from the sequence of the viral genome. TYLCV can be detected in tissues of infected plants as well as in insect vectors. Monoclonal as well as polyclonal antibodies do not have such a level of specificity and may detect several begomoviruses.

1. Molecular DNA-DNA Hybridization

1.1. Southern blot hybridization

Plant DNA may be prepared according to the CTAB-based method (Taylor and Powell, 1982) and whitefly DNA is extracted with SDS-Proteinase K (Zeidan and Czosnek, 1991). DNA is subjected to agarose gel electrophoresis, blotted, pre-hybridized and hybridized with a virus-specific DNA probe as described (Ber et al., 1990). The probe may consist of the full-length viral genome or virus species-specific sequences such as the intergenic region. The probe is labelled either with a radioactive nucleotide (e.g. 32P-adCTP) or with a non-radioactive nucleotide (e.g. digoxygenin-11-dUTP). The blots are then washed at 65°C for 30 min (twice) in 150 mM NaCl and 15 mM trisodium citrate (1 x SSC). Washing the blot at 70°C in 0.1 x SSC allows discrimination between closely related TYLCV isolates, such as viruses from Israel and from Italy (Czosnek et al., 2001). If the probe is radiolabelled, the blot is exposed to an X-ray film. For non-radioactive probing, the blot is subjected to immunological detection. After blocking, the filter is incubated with an anti digoxygenin alkaline phosphatase conjugate (diluted 1:5000). After washing, the filter may be incubated with the alkaline phosphatase substrates Nitro Blue Tetrazolium (NBT) and 5-Bromo-4-chloro-3-indolyl phosphate (BCIP), until a dark blue colour is obtained. Alternatively the virus-probe complex can be detected by chemiluminescence (Caciagli and Bosco, 1996). Non-radioactive digoxygenin-labelled DNA probes can be almost as sensitive as radioactive probes.

1.2. Tissue print hybridization

TYLCV DNA sequences can be detected specifically and sensitively by hybridization of infected plant tissues squashed onto a nylon membrane (squash-blot) with a radiolabelled specific DNA probe (Navot et al., 1989). No treatment of the sample is necessary prior to squashing and hybridization. TYLCV DNA could be detected in squash-blots of tomato leaves, roots, stems, flowers and fruits. Viral sequences can also be detected in squashes of a single whitefly that has fed on infected tomato plants. An assay that can be used in the field for the detection of TYLCV has been developed (Zilberstein et al., 1989). Plant and insect tissue squashes are hybridized with sulfonated virus complementary (-) strand DNA produced from a full-length DNA clone using the M13K07 helper phage. A mouse monoclonal antibody binds to the sulfone groups of the DNA hybrid; this complex is recognized by a goat anti-mouse immunoglobulin antibody conjugated to alkaline phosphatase which enzyme transforms a colourless substrate into a coloured product, indicating the presence of viral nucleic acids. Virus can be detected in stems, leaves, roots, flowers and fruits. It can also be detected in the whitefly vector, at the individual level.

2. Polymerase Chain Reaction (PCR)

PCR is widely used for the diagnosis of geminiviral diseases, allowing the detection of very small amounts of the disease agent in the infected plant and vectors, and also the cloning of genomic fragments of the pathogen. The following cycling protocol can be used to amplify the full-length TYLCV genome or viral DNA fragments from infected plants (50 ng DNA) or from viruliferous whiteflies (10 ng): initial denaturation for 3 min at 95°C, annealing of primers (0.2 mM each) for 1 min at 55°C, extension for 2 min at 72°C, and denaturation for 1 min at 94°C; subsequent cycles are: 1 min at 55°C, 2 min at 72°C and 1 min at 94°C; after 30 cycles, the reaction is terminated by a 10 min incubation at 72°C (Navot et al., 1992). The PCR products are subjected to agarose gel electrophoresis, stained with ethidium bromide and photographed. The primers are deduced from the sequence of the virus genome. Amplification of the full-length TYLCV genome can be achieved using primers V41 (nucleotides 61-80, from 5' to 3': ATACTTGGACACCTAATGGC) and C60 (nucleotides 41-60: AAGTAAGACACCGATACACC). Specific primers can be designed to discriminate between TYLCV species, e.g. from Israel (TYLCV) and from Sardinia, Italy (TYLCSV). For example, the TYLCV primers V2325 (nt 2325-2350, 5'CGTAGGTCTTGACATCTGTTGAGCTC3') and C2714 (nt 2714-2690, 5'CAAATAGCCATTAGGTGTCCAAGTA3') allow amplification of a 385 bp DNA from TYLCV from Israel (but not from TYLCSV). Conversely, the TYLCSV primers VS2308 (nt 2308-2333 viral strand, 5'TATAGGA CTTGACGTCGGAGCTCGAT3') and CS2698 (nt 2698-2670, complementary strand, 5'GGGG GCATCATATATATTGCCCCCCAATT3') allowing amplification of a 390 bp DNA fragment of TYLSCV (but not from TYLCV) (Goldman and Czosnek, 2002). Multiplex PCR with virus-specific primers was used to detect TYLCV-IL, TYLCV-Mld and TYLCSV in tomato samples from Egypt, Israel, Jordan and Lebanon (Anfoka et al., 2008)

TYLCV DNA can be amplified by PCR using plant and insect tissues squashed on a membrane as a template (Atzmon et al., 1998; Navas-Castillo et al., 1998). Samples are used as is. They may be stored in a dry state for several weeks without using their capacity to serve as template for PCR. Practically, a strip of 1 x 2 mm containing the tissue print is cut and immersed into the 25 ml PCR mix contained in a vial before initiation of amplification. Cycling and analysis of products is as described above.

3. Rolling-circle amplification

The rolling circle mode of geminivirus replication can be used for their diagnosis (Jeske, 2007). TYLCV, like any other geminivirus, can be amplified from total DNA of infected plants using the bacteriophage phi29 DNA polymerase. The amplified viral DNA induced typical symptoms after biolistic inoculation of test plants. Infectious DNA was obtained successfully from fresh, freeze-dried or desiccated plant material, from squashes of plant leaves on FTA cards, as well as from the insect vector. Plant material collected and dried as long as 25 years ago yielded infectious DNA by this method (Guenoune-Gelbart et al., 2010).

4. Microarrays

Microarray-based platforms constitute a new tool in the ever challenging world of plant virus diagnosis. They have the potential to simultaneously detect a number of viruses in a single reaction. Oligonucleotides (40-70 nt-long) with sequences based on the viral genomes that are the object of potential diagnosis) are printed on a slide. In general the RNA or the DNA of the suspected plant is labelled with a fluorescent dye (during or after reverse transcription or PCR) and hybridized with the printed oligo targets on the slide array. The fluorescent spots are localized and the pathogen is identified (Boonham et al., 2007). This technology has been used to identify TYLCV in infected tomato plants (Tiberini et al., 2010).

5. Immunological Methods

5.1. ELISA

TYLCV can be detected by standard ELISA procedures; however, as for all immunological methods, the antibodies available are usually unable to distinguish TYLCV from closely-related begomoviruses, and unable to discriminate between TYLCV species and strains.

5.2. Dot immuno-binding assay (DIBA)

This detection procedure is a modification of that described by Hibi and Saito (1985). Sample preparation has been described by Poolpol (1986). The antiserum raised in rabbits against TYLCV is cross-absorbed with dried leaf powder of non-infected tomato tissue. One µl of crude sap obtained after crushing tissues from infected plants is spotted onto a nitrocellulose membrane. After blocking, the membrane is covered with the anti-TLYCV antibody, washed and incubated with alkaline phosphatase conjugated-goat anti-rabbit IgG, washed and incubated with the substrate solution (0.1% napthol AS-MX phosphate mixed with an equal volume of 6 mg/ml Fast Red TR salt in 0.2 M Tris-HCl, pH 8.2). The results are evaluated by colour development; a pink coloration on the sample spot is a positive reaction. The green colour of concentrated crude sap could interfere with the development of colour in a positive reaction.

5.3. Immune Electron Microscopy (IEM)

The trap-decoration method (Milne and Luisoni, 1977) can be used to detect TYLCV but it is not practical to use for the detection of the virus from large numbers of plant samples. Membrane-coated grids are floated on drops of 1:10 diluted antiserum, washed and floated on sap of infected plant. Washed grids are then floated on diluted (1:100) absorbed-antiserum to allow decoration of virus particles, washed and stained with 2% uranyl acetate. Virus particles are observed by transmission electron microscopy.

Symptom severity depends on the date of inoculation, tomato variety and whitefly pressure. Typical symptoms include curling and yellowing of young leaves and severe stunting. TYLCV and the strain involved are best identified by sequencing either full-length clones or PCR-amplified genomic DNA fragments using specific primers. Squash blot hybridization with a virus-specific DNA probe, followed by high stringency washes, also allows discrimination between viruses and their strains (see Diagnosis).

The DNA sequence of the genome (full-length or partial) of several tens of begomovirus species and strains infecting tomato available in public databases such as Genbank show that the many different begomoviruses that infect tomato are characterized by considerable genetic diversity. Some possess a monopartite genome (e.g. TYLCV, TYLCSV, TYLCCV, TLCV from southern India, Australia and Taiwan), while others have their genome split between two DNA molecules (e.g. TYLCTHV, TLCV from northern India, TGMV, TMoV, ToLCrV).

Nucleotide comparisons of the full-length genome or of selected areas (e.g. intergenic region, coat protein gene, Rep gene) have provided a glimpse into the phylogenetic relatedness of these viruses (Rybicki, 1994; Padidam et al., 1996; Nakhla and Maxwell, 1998; Brown et al., 2001; Abhary et al., 2007; Lefeuvre et al., 2010). Sequence comparisons of begomovirus genomes and open reading frames (ORF) has allowed these viruses to be grouped according to their geographical origin: 1) Middle East and Central Asia, 2) Western Mediterranean basin, 3) Indian subcontinent, 4) North and Sub-Saharan Africa, 5) East and South-East Asia and Australia, and 6) New World with subgroups including USA, Central and South America and the Caribbean Islands (except the newly introduced Middle Eastern TYLCV). Similarly, the whitefly Bemisia tabaci complex could be resolved into seven major groups on the basis of mitochondrial DNA markers, essentially overlapping the geographical distribution of begomoviruses (Brown, 2010; De Barro et al., 2011).

Based upon their sequence, TYLCV from Israel and from Egypt can be considered to be isolates of the same strain. The same is valid for TYLCV from Israel and from the Dominican Republic and Jamaica, indicating that TYLCV was recently introduced to the Caribbean from the Middle East. Similarly, the Sardinian and the Spanish isolates of TYLCV are almost identical and are therefore strains of the same species. On the other hand, the Italian and Israeli TYLCV isolates are different enough to be considered as two virus species. The Thai TYLCV isolate constitutes a separate species. The nucleotide sequence of TYLCTHV component B shows low homology with other bipartite begomoviruses (e.g. about 55% with TGMV and Squash leaf curl virus).

The geographically associated variation of tomato begomoviruses is best illustrated by the comparison of their coat protein (CP). Differences associated with geography are best exemplified when comparing the isolates from north and south Saudi Arabia, two regions separated by a large expanse of desert. Their capsid proteins share only 78% homology. The northern isolate is closely related to TYLCV from Israel (95% homology), while the southern isolate is not (81% homology). On the other hand, the Sicilian and Sardinian isolates of TYLCV are found on two islands separated by 300 km; nevertheless their capsid protein share 95% homology. In several cases, the coat protein of TYLCV is more homologous to the coat protein of other begomoviruses occurring in the same region than to TYLCV isolates from other regions. For example, the coat protein of TYLCV from Nigeria is more related to the local African cassava mosaic virus (ACMV) than to TYLCV from Thailand (80% vs. 76%), a fact that may point to adaptation of the geminivirus to its vector. Moreover, viruses in different geographical regions have different epitope profiles, whereas those from the same region have similar profiles (Macintosh et al., 1992). Differences in the sequence of the capsid protein may lead to differences in virus transmissibility by B. tabaci (McGrath and Harrison, 1995).

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

Cultural Control

Growing tomato in open fields is much more constraining than growing under cover. In all regions where TYLCV is endemic, seedlings need to be grown and kept in a glasshouse or a screenhouse in order to prevent early infection by whitefly feeding. In Thailand, farmers grow tomato plants in isolated paddy fields after rice harvesting. In Egypt the spread of TYLCV in open fields was successfully controlled by roguing infected plants and removing all overwintered tomato crops before the emergence of whitefly populations (Mazyad et al., 1986). Similar measures were undertaken in Cyprus over 3 consecutive years resulting in a decrease in the incidence of the disease from 50 to 5% (Ioannou, 1987). In the Dominican Republic, a mandatory 3 months host-free growing period which included bean, cucurbits, aubergine, melon, okra, pepper and tomato, significantly reduced the inoculum and lowered to a minimum the incidence of TYLCV (Gilbertson et al., 2007). Early planting when the incidence of B. tabaci is low, effectively reduced TYLCV infection in Egypt, Israel and Cyprus. Interplanting with rows of TYLCV non-host trap plants such as squash and cucumber have been used to divert whiteflies from tomato in the Jordan Valley, delaying TYLCV infection by 2 months (Al-Musa, 1982). Sticky yellow plastic polyethylene sheets have been used, with limited success, to decrease the pressure of the whitefly population (Cohen et al., 1974). When economically feasible, growing tomatoes in greenhouses covered with insect-proof 50-mesh screens and double-doors constitutes the optimal solution for crop protection against the TYLCV whitefly vector (Cohen and Berlinger, 1986; Ioannou, 1987). UV-absorbing plastic sheets or UV-absorbing 50-mesh screens, used as tunnels or as screenhouse covers, greatly reduced the infestation of tomato plants by B. tabaci and the incidence of TYLCV (Antignus et al., 1998). Whitefly natural enemies (Gerling, 1990) have not been used on a large scale in open fields.

Chemical Control

Applying systemic insecticides as soil drenches or regular spraying during the seedling stage can reduce the population of the B. tabaci vector and the incidence of TYLCV. Many insecticides such as organophosphates, carbamates and pyrethroids effectively reduce the whitefly population, whether sprayed one insecticide at a time, alternately, or together with oil and emulsifiers (Makkouk and Laterrot, 1983). However, these insecticides provided only partial TYLCV control, even when sprayed daily (Cohen and Antignus, 1994). Continued extensive use of insecticides is not only detrimental to the environment but it promotes the development of resistant whitefly populations (Cahill et al., 1995). More potent insecticides with novel modes of action such as imidacloprid were introduced in the 1990s (Elbert et al., 1990). Over the past few years imidacloprid has assumed major importance for control of B. tabaci in the field and in greenhouses. To be efficient, the insecticide has to induce the death of all whiteflies within 35-40 min before it succeeds in infecting the plant, thereby preventing inoculation of TYLCV, which is not always the case in the field (Rubinstein et al., 1999). As for other commonly used insecticides, whitefly populations resistant to imidacloprid have been emerging in many places (Castle et al., 2010).

Host-Plant Resistance

Breeding programmes aimed at producing tomato cultivars resistant to TYLCV started in the late 1960s and have expanded since. These programmes are based on the introgression into the domesticated tomato Solanum lycopersicum of resistance or tolerance found in some accessions of wild tomato species such as S. cheesmaniae, S. chilense, S. habrochaites, S. peruvianum and S. pimpinellifolium. Depending on the plant source, resistance was reported to be controlled by one to five loci, either recessive or dominant (Zakay et al., 1991; Picó et al., 1996; Nakhla and Maxwell, 1998). The first commercial tolerant cultivar, TY20 (Hazera) carrying tolerance from S. peruvianum, showed delayed symptoms and the accumulation of viral DNA (Pilowski and Cohen, 1990; Lapidot et al., 1997). More advanced cultivars derived from TY-20 include 8484 and 8472. Additional commercial hybrids tolerant to TYLCV include Anastasia (Brunsma), Silver, Beludo and Ulyses (Seminis), Fiona and Jackal (Sluis and Groot), TOP 21 (Clause), Saria and Gemstar (Petoseed), Hilario, Rex, Tycoon and Tyking, (Royal Sluis), Pitenza-TY and Pal-TY (Enza-Zaden). In heavily affected areas, these hybrids need to be protected with insecticides during the first weeks after planting to produce acceptable yields. Many experimental lines are being developed which show good levels of resistance against TYLCV and other begomoviruses affecting tomato worldwide. The breeding line TY172 originating from S. peruvianum is a symptomless carrier of TYLCV whether infected in the greenhouse or in the field; at least three loci may account for the resistance (Friedmann et al., 1998). The breeding line 902 originating from S. habrochaites support little virus accumulation; another line derived from the same wild tomato progenitor, line 908, is tolerant to the virus. Tolerance is controlled by a dominant major locus, and resistance by two to three additive recessive loci (Vidavski and Czosnek, 1998). Nine hybrids based on these lines have been successfully tested in infested fields in Israel, Egypt, Morocco and Guatemala.

Marker-assisted breeding has allowed localization of TYLCV-resistance loci on the tomato genetic map and has been instrumental in developing new tomato varieties. Using polymorphic DNA markers, a TYLCV-tolerance gene originating from S. chilense LA1969, Ty-1, has been mapped to tomato chromosome 6, close to the nematode resistance Mi gene (Zamir et al., 1994). Another locus associated with tolerance originating from S. pimpinellifolium has been mapped to chromosome 6 but to a locus different from Ty-1 (Chague et al., 1997). Polymorphic markers have also been used to map a loci coinded Ty-2 conferring resistance to ToLCV, originating from S. habrochaites to tomato chromosome 11 (Hanson et al., 2000). A major partially dominant locus, termed Ty-3, was mapped to chromosome 6, near the Ty-1 locus (Ji et al., 2007). Another TYLCV resistance locus, termed Ty-4, was mapped on the long arm of Chromosome 3 (Ji et al., 2008). Another locus, termed Ty-5, was mapped to chromosome 4 using the breeding highly resistant line TY172, which originating from S. peruvianum (Anbinder et al., 2009). The different sources of resistance were piled up to provide tomato plants with better resistance to TYLCV (Vidavski et al., 2008). The genes conferring resistance have not yet been isolated. An RNAi-based genome-wide screening is under way to uncover the gene network sustaining TYLCV resistance in tomato lines issued from S. habrochaites (Czosnek et al., 2011b).

Genetic Engineering

1. Pathogen-derived resistance

Virus-resistant transgenics have been developed in many crops by introducing either viral capsid protein or replicase gene encoding sequences. This concept has been called pathogen-derived resistance (PDR) (Lomonossoff, 1995; Baulcombe, 1996). PDR has been the most common approach used to obtain resistance to TYLCV, including viral sequences that generate antisense RNA as well as the expression of full-length and truncated viral genes (Polston and Hiebert, 2007). The CP gene was one of the first TYLCV genes evaluated for the ability to generate pathogen-derived resistance. Tomato plants expressing the TYLCV CP showed a delay in symptoms, a recovery from infection, and resistance upon repeated inoculations (Kunik et al., 1994). Expression antisense RNA of the TYLCSV Rep gene provided a good level of resistance in Nicotiana benthamiana (Bendahmane and Gronenborn, 1997). The expression of sense and antisense constructs of truncated TYLCSV Rep in N. benthamiana also showed resistance (Noris et al., 1996). When a truncated TYLCV-Mld Rep gene was expressed in tomato, the transgenic plants were resistant to TYLCV-Mld, but were susceptible to TYLCV-IL (Antignus et al., 2004). A 2/5 TYLCV Rep gene construct conferred high levels of resistance and often immunity in both transformed N. tabacum and S. lycopersicum (Yang et al., 2004).

2. Virus gene silencing

 Viruses trigger plant antiviral mechanisms leading to the degradation of viral RNA via the formation of dsRNA derived from viral sequences, initiating post-transcriptional gene silencing (PTGS). This phenomenom was named RNA-mediated gene silencing. Some viruses establish themselves by expressing silencing supressor genes, which interfere with plant host silencing mechanisms (Voinnet et al., 1999; Roth et al., 2004). RNA-mediated virus resistance proved to be more efficient than protein-mediated resistance but was highly-sequence dependent. Transgenic plants have been developed which exploited the mechanism of silencing via dsRNA (Smith et al., 2000). Using a transformation cassette consisting of the 3’-end of the Rep (sense and antisense orientation) as the arms of the hairpin, and a castor bean catalase as the intron, the treated tomato plants showed immunity to TYLCV upon whitefly-mediated inoculation (Fuentes et al., 2006). A similar approach was used with the 5’-end of the TYLCV CP gene and a maize ubiquitin intron, inducing resistance to TYLCV in tomato (Zrachya et al., 2007). The intron-hairpin approach was used to express multiple TYLCV sequences in tobacco and tomato plants (Abhary et al., 2006). The treated plants were symptomless and TYLCV DNA was not detected. In general, silencing was specific to the TYLCV strain; silencing using TYLCV sequences conferred resistance to TYLCV but not to other TYLCV species or strains, such as TYLCSV (Noris et al., 2004).

3. GroEL

This novel approach is based on the observation that GroEL produced by whitefly endosymbiotic bacteria binds to the CP of TYLCV in the insect hemolymph, thereby protecting the virion from destruction and ensuring transmission (Morin et al., 1999). TYLCV-GroEL binding has been exploited to generate TYLCV resistant tomato plants. Tomatoes transformed with GroEL under a phloem-specific promoter showed milder to no symptoms in plants in the R₀ through R₂ generation. GroEL/TYLCV complexes were readily detected in resistant plants. It was hypothesized that GroEL/TYLCV complexes formed in transformed plants and that these complexes were interfering in virus movement (Akad et al., 2007).

Israel: Agricultural Research Organization, Bet Dagen, www.volcani.gov.il

Israel: The Hebrew University of Jerusalem, Faculty of Agriculture, Rehovot, www.agri.huji.ac.il

Taiwan: Asian Vegetable Research and Development Center, Tainan, www.avrdc.org

Spain: Instituto de Hortofruticultura Subtropical y Meditarránea, Malaga, www.eelm.csic.es

USA: University of Arizona, Tucson, Arizona, www.arizona.edu

USA: University of California at Davis, Davis, California, www.ucdavis.edu

USA: University of Florida - Gainesville, Gainesville, Florida, www.ufl.edu

01/01/12 Review by:

Henryk Czosnek, The Haim Gvati Professor of Agriculture, Institute of Plant Sciences and Genetics in Agriculture, Faculty of Agriculture, Food and Environment, The Hebrew University of Jerusalem, Rehovot 76100, Israel.