A new aubergine disease caused by a whitefly-borne strain of Tomato mild mottle virus (TomMMoV)


E-mail: aviv@volcani.agri.gov.il


The aims of the present study were to further characterize the causal agent of a new viral disease of aubergines in Israel, first observed in 2003 and tentatively named eggplant mild leaf mottle virus (EMLMV) in a previous work, and to identify the vector responsible for its spread. The disease could be transmitted mechanically from infected source plants to healthy aubergines or laboratory test plants. Transmission electron microscopy (TEM) analysis of purified virus preparations indicated the presence of viral particles with a flexible filamentous morphology (approximately 720 nm long). TEM analysis of ultrathin sections prepared from infected leaf tissue revealed the presence of cytoplasmic inclusion bodies with pinwheel and crystalline structures, typical of those induced by potyviral infection. The viral coat protein subunit was shown to have a molecular weight of 37·5 kDa by SDS-PAGE analysis. The viral particles reacted positively in western blot analysis with an antiserum against Tomato mild mottle virus (TomMMoV) from Yemen, described as a potyvirus, vectored by the aphid Myzus persicae. The current study describes some biological properties of EMLMV and presents evidence for its transmission by the whitefly Bemisia tabaci, but not by three aphid species. The taxonomic relationship between EMLMV and TomMMoV is discussed based on their biological characteristics and sequence analysis of their genomes. It is suggested that the Israeli EMLMV should be considered a distant strain of TomMMoV, designated TomMMoV-IL, according to the present rules of Potyviridae molecular taxonomy.


During the autumn of 2003, a previously unknown disease was observed on aubergine (Solanum melongena) growing in open fields and ‘walk-in’ tunnels in the Jordan Valley, Israel. Since then, the disease has spread to most parts of the country. The disease symptoms include mild mottling on leaves and different degrees of fruit distortion, sometimes accompanied by the formation of blisters on the surface of the fruit. In some cases, total yield loss has been observed. The causal agent of the new disease was determined and tentatively named eggplant mild leaf mottle virus (EMLMV) (Dombrovsky et al., 2012). Sequencing of the complete genome of EMLMV supported its taxonomic assignment to the genus Ipomovirus, one of the seven genera composing the Potyviridae family of plant viruses. The sequencing data showed a close relationship between EMLMV and tomato mild mottle virus (TomMMoV), a previously described aphidborne potyvirus found on tomatoes in Yemen (Walkey et al., 1994) that was subsequently classified as a candidate Ipomovirus based on the partial sequence of the NIb and the complete sequence of the coat protein (CP) (Monger et al., 2001). Recently, the complete genome sequence of an Ethiopian isolate of TomMMoV was determined and the possibility of whitefly transmission was suggested, based on a single transmission event (Abraham et al., 2012), leaving the identity of the insect vector vague.

The aim of the current study was to further study the biological characteristics of the new aubergine disease, to identify its insect vector and to clarify the taxonomic relationship between TomMMoV and EMLMV. The copublication of the genome organization of EMLMV and TomMMoV and their complete nucleotide sequences (Abraham et al., 2012; Dombrovsky et al., 2012) has paved the way for the use of molecular data to explain differences in their biological properties and for the accurate determination of their phylogenetic relationship.

Materials and methods

Host-range determination and plant maintenance

Infected aubergine plants were collected from a field located in the Jordan Valley, Israel. The virus was then transmitted from these plants to test plants by inoculation using sap prepared in 0·01 m phosphate buffer (pH 7·0) containing carborundum (silicon carbide) dust. The determination of the partial host range was carried out by a set of experiments repeated at least three times, using three to five plants for each plant species. The virus was maintained on systemically infected Nicotiana glutinosa, which served as a propagation host. All virus-infected plants were kept in an insect-proof greenhouse and were sprayed regularly with insecticides to prevent any infestation.

Virus purification and preparation of plant tissues and ultrathin sections

Virus purification was carried out essentially as described previously (Antignus et al., 1994). The presence of viral particles was confirmed by transmission electron microscopy (TEM) (Tecnai G2; FEI-Philips).

Infected N. glutinosa leaf tissue was fixed in 3·5% glutaraldehyde in phosphate-buffered saline (PBS), rinsed and fixed in 1% OsO4 in PBS. Following several washes in PBS, the tissue was stained en bloc with uranyl acetate. The samples were then dehydrated by serial washings in ethanol and acetone. After being dehydrated, the samples were embedded in epoxy resin Agar100 (Agar Scientific). Thin sections were mounted on formvar/carbon-coated grids, stained with uranyl acetate/lead citrate, and then examined by TEM. Micrographs were taken using a MegaView III camera operated by the Analysis software (Soft Imaging System).

Characterization of the viral coat protein

The molecular weight (MW) of the viral coat protein (CP) subunit was determined by sodium dodecyl sulphate (SDS) 12% polyacrylamide gel electrophoresis (SDS-PAGE) analysis (Laemmli, 1970). The resulting protein bands were visualized by staining with Coomassie brilliant blue (Sigma-Aldrich).


Homologous EMLMV antiserum was prepared as follows: purified particles of the virus were separated by 12% SDS-PAGE. The putative viral CP band (37·5 kDa) was excised from the gel and crushed in PBS before being injected into rabbits. Pre-immune blood was taken from the rabbits before they were injected. Four sequential injections were given: the first included 1 mL 0·05 m phosphate buffer (pH 7·2) containing 1 mg viral antigen and 1 mL incomplete adjuvant. A second injection was given 4 weeks later. The third and fourth injections were performed later at 2-week intervals. For the last three injections, a 0·5-mg aliquot of the viral CP antigen was resuspended in 1 mL PBS and 1 mL incomplete adjuvant. Blood was collected from immunized rabbits 1 week after the third injection (10 mL) and 1 week after the fourth injection (25 mL). Serum was separated from blood cells and the specificity of the immunogenic reaction was determined by western blots of the purified viral CP or crude preparations of total proteins extracted from infected or uninfected plants, as described below.

The following antisera were used for the serological tests: monoclonal antibodies for general detection of potyviruses (Agdia); general polyclonal antibodies against potyviruses; two polyclonal antibodies developed against two carlaviruses (Carnation latent virus (CLV) and Passiflora latent virus (PLV) (kindly provided by A. Gera, ARO, The Volcani Center)); polyclonal antibodies specific to Potato virus Y (PVY) (BIOREBA); and Turnip mosaic virus (TuMV) antiserum (prepared in the authors’ laboratory). A polyclonal antiserum specific to TomMMoV (1:1500) was kindly provided by H. J. Vetten (Braunschweig, Germany).

Enzyme-linked immunosorbent assay (ELISA)

The serological identity of the virus was determined by indirect enzyme-linked immunosorbent assays (indirect ELISA; Koenig, 1981), as well as by double-antibody sandwich (DAS-ELISA; Clark & Adams, 1977). In brief, indirect ELISA was performed on crude sap from infected plants, or purified virus preparations, loaded onto polystyrene plates. Following washings, plates were treated with the specific antiserum and then with alkaline phosphatase (AP) conjugated to goat antirabbit IgG (secondary antibody). In the case of DAS-ELISA, plates were coated with the specific antiserum prior to sample loading and then treated with AP conjugated to the specific antiserum. The optical densities of the hydrolysed substrate (p-nitrophenylphosphate) were recorded at 405 nm by an ELISA plate reader (Anthos Labtec Instruments).

Western blot analysis

Western blots were used to study the serological relationships between EMLMV and other viruses. Total plant protein was extracted from leaf discs (two leaf discs, total approximately 25 mg) harvested from infected and uninfected plants. Each sample was ground in 170 μL urea–SD–β-mercaptoethanol (USB) buffer (75 mm Tris–HCl (pH 6·8), 9 m urea, 4·5% (v/v) SDS and 7·5% (v/v) β-mercaptoethanol). The extracted samples were then boiled for 5 min and cooled on ice. The cooled homogenates were centrifuged for 10 min at 13 000 g and 100-μL aliquots of the resulting supernatant for each homogenate, containing the extracted proteins, were mixed with 35 μL of 4 × loading buffer (Laemmli, 1970). Later, 10-μL aliquots of the denatured mixture were separated by SDS-PAGE in 12·5% gels. The resulting gels were electroblotted onto nitrocellulose membranes (OPTITRAN BA-S85, Schleicher & Schuell) for 40 min at 240 mA by a semidry transfer blot apparatus (Trans-Blot SD, Bio-Rad). Blotted membranes were stained with Ponceau S (Sigma) for 2 min, washed with double-distilled water and photographed. The membranes were then blocked with PBS (pH 7·4) containing 3% non-fat milk powder (Sigma) for 3 h at room temperature and then probed with the tested antisera.

Insect propagation

All aphid species were collected outdoors and maintained separately in growth chambers at 25 ± 3°C under continuous lighting. The green peach aphid Myzus persicae was raised on mustard (Brassica perviridis cv. Tendergreen), whilst the melon aphid, Aphis gossypii, and the potato aphid (the pink morph), Macrosiphum euphorbiae, were both raised on aubergine in order to minimize host-plant effects during the acquisition access feeding (AAF) and inoculation access feeding (IAF) periods. The whitefly colony of Bemisia tabaci that was used in this study was verified to be the B biotype using the procedure described by De Barro et al. (2003). The whiteflies were maintained on cotton plants (Gossypium hirsutum cv. Acala) in 50-mesh screen-covered cages kept in an air-conditioned greenhouse.

Aphid transmission

For plant-to-plant transmission experiments, aphids were collected from healthy plants and placed in a glass vial for a starvation period of 1 h, which was followed by an AAF period. The starved aphids were allowed a 12-h AAF period on infected source plants. Groups of 20 aphids each were transferred to each of the test plants (Nicotiana clevelandii, N. glutinosa, Solanum nigrum and aubergine cv. Classic) for an IAF. The potyvirus Zucchini yellow mosaic virus (ZYMV), maintained on Cucurbita pepo cv. Ma'ayan, and the aphid M. persicae were used for ZYMV transmission to cucumber (Cucumis sativus) cv. Beit Alpha as a positive control. In all experiments, the IAF lasted 48 h and was followed by treatment with the insecticide imidacloprid (Confidor, Bayer) before plants were transferred to a growth chamber to allow symptom development.

Whitefly transmission

Five intact aubergine plants showing typical disease symptoms were collected from fields located in the Jordan Valley, Israel and transplanted into 10-L pots after being sprayed against insects. These plants served as source plants in some of the transmission experiments. Alternatively, in another set of experiments, mechanically inoculated tobacco (Nicotiana tabacum) cv. Samsun and N. glutinosa served as source plants for virus acquisition and transmission. The presence of the virus in source plants and inoculated test plants was tested by mechanical inoculation of N. glutinosa, TEM dip analysis or western blots, using an EMLMV-specific antiserum. For the transmission experiments, infected plants were placed in large cages (1 × 1 × 1 m) covered with a 50-mesh screen. Large groups of whiteflies (a few hundred in each group) were given a 48-h AAF before being released onto healthy test plants (aubergine and tobacco) for a 48-h IAF. Virus presence in the whitefly vector was determined postacquisition by RT-PCR on total RNA extracted from viruliferous whiteflies starved for 3 h. To study the retention of the virus in the vector, serial transfers of the same group of viruliferous whiteflies were performed at 24- to 48-h intervals. Inoculation access was given on two to four healthy aubergine seedlings. At each step of the serial transmission, the whitefly-infested ‘source’ plants were carefully shaken over the test plants to avoid mechanical contact between source plants and test plants. Following IAF, plants were treated with the insecticide imidacloprid and the treated plants were moved to an insect-proof greenhouse for symptom appearance. The presence of the virus in the inoculated test plants was verified by mechanical inoculation to N. glutinosa, by electron microscopy examination and by western blotting or by dot-spot hybridization with a specific molecular probe, as described below.

Identification of EMLMV RNA in viruliferous Bemisia tabaci

Using the EZ-RNA Total RNA Isolation Kit (Biological Industries-Beit Haemek), total RNA was extracted from 20 mg B. tabaci collected after 3 h of starvation and 48 h of AAF on Cucumber vein yellowing virus (CVYV)-infected cucumber, EMLMV-infected tobacco or EMLMV-infected aubergine, and from B. tabaci from a laboratory culture maintained on cotton. The extracted total RNA was treated with RQ RNase-free DNase I for 30 min at 37°C before being used as a template in a first strand synthesis in the presence of 15 pmol of the primer R-EM-8852 5′-TGGTAGTGAGGGGGAAACTG-3′ (positions 8958–8977 in the EMLMV genome) and 10 pmol of the R-B.t-Actin gene primer 5′-CCAGCCAAGTCCAAACGAAG-3′. cDNA was amplified in a PCR using 15 pmol of the EMLMV forward primer F-CP-8470 5′-CAGAGTTTTTGGGTTCCTCCAGCA-3′ (positions 8470–8493 in the EMLMV genome) and 10 pmol of the whitefly gene primer F-B.t-Actin 5′-TGGAGATGGTGTTTCCCACAC-3′. The following complementary primers were used in the amplification reaction: 15 pmol R-EM-8852 and 10 pmol R-B.t-Actin. The B. tabaci actin gene (EST accession number EE597333) was used as an internal control (Mahadav et al., 2008).

Dot-spot hybridization

Total RNA preparations from infected and healthy plants were obtained using the EZ-RNA Total RNA Isolation Kit. Each RNA sample was mixed with an equal volume of a denaturing solution (8 × SSC, 10% formaldehyde, 5% formamide) and incubated at 65°C for 15 min. Five microlitres of the denatured viral RNA were spotted on a nylon membrane (Zeta-Probe, Bio-Rad) and baked at 80°C for 2 h. Hybridization was performed at 48°C for 16 h with a riboprobe prepared from a pGEM-T Easy plasmid carrying a viral cDNA, encoding the viral-NIb-CP. The plasmid was linearized by ApaI digestion and the digested product served as a template for an in vitro transcription reaction with SP6 RNA polymerase, carried out using a riboprobe kit (Epicentre) in the presence of 32P-labelled uridine (Izotop). After hybridization, the membranes were washed twice with 2 × SSC containing 0·2% SDS for 5 min each at 37°C and two additional washes with 0·2 × SSC containing 0·2% SDS for 5 min each at 60°C. Signals were recorded by exposing the membrane to an amplifying screen and were then analysed in a Bio-Imaging Analyser, Bas 1500 IP Reader (Fujifilm).

Virus sequencing

Virion RNA was extracted from EMLMV-infected aubergine plants that were collected in the field and served as source plants in transmission experiments. To eliminate the possibility of coinfection with other viruses the extracted RNA was subjected to next-generation sequencing (NGS) conducted using the SOLiD v. 3 instrument in accordance with the protocols published by Applied Biosystems at the Centre for Genomic Technologies at the Hebrew University of Jerusalem, Israel as formerly described (Dombrovsky et al., 2012). The bioinformatic analysis was carried out at the Goldyne Savad Institute of Gene Therapy at the Hadassah Medical Center at Jerusalem, Israel.

Sequence analysis

Nucleotide sequence analysis was carried out using dnaman (Lynnon Biosoft) and software from the NCBI database. Sequence identity was determined using the Basic Local Alignment Search Tool (blast; http://blast.ncbi.nlm.nih.gov/Blast.cgi) or the blastx, tblastx or blastp algorithms. Multiple sequence alignments were produced using the BioEdit software (Hall, 1999) and clustalW algorithms (Thompson et al., 1994). Phylogenetic tree prediction was carried out using the deduced amino acid sequences of the viral polyproteins of virus species in the Potyviridae. First, the muscle program was used for sequence alignments (Edgar, 2004). Then, a phylogenetic tree was constructed based on maximum likelihood (ML), using the phyml software with 100 bootstrap replicates (Guindon & Gascuel, 2003).


Virus culture, symptomatology and host range

Intact aubergine plants showing leaf mild mottle symptoms were brought from the field and served as source plants to establish a virus culture. The virus was easily transmitted to test plants by mechanical sap inoculation. As shown in Table 1, no symptoms were observed on pepper (Capsicum annuum) or tomato (Solanum lycopersicum), which are economically important crops in Israel. Serological tests, as well as back-inoculation from inoculated tomato to N. glutinosa test plants, confirmed the symptomless infection of tomato by the virus. In contrast, these same biological and serological tests showed that pepper is immune to EMLMV infection. The results indicate that EMLMV can infect Datura stramonium and S. nigrum, which are common weeds throughout Israel. The symptoms observed on Nicotiana species included apical leaf distortion, vein clearing and leaf bronzing, as well as the formation of widely diffused local lesions on the lower leaves (not shown). Aubergine plants reacted to virus infection with very mild green mottling on the leaves (Fig. 1a). However, in some cases, recovery from these symptoms was observed. In sharp contrast to the symptoms observed on aubergine leaves, the effect of infection on fruits was devastating (Fig. 1a–d). Fruits of infected plants showed different extents of deformation, accompanied by a hardening of their flesh which made them unmarketable. In cases of widespread epidemics, total yield loss was observed (Fig. 1e,f).

Figure 1.

(a) Mild green mottling on the apical leaf of an aubergine (Solanum melongena) plant naturally infected with Tomato mild mottle virus (TomMMoV)-IL. (b,c,d) Blistering and deformation of fruits collected from naturally infected plants grown in the Jordan Valley. (e) A heap of unmarketable TomMMoV-IL-infected aubergine fruits. (f) An aubergine field in the Jordan Valley heavily infected with TomMMoV-IL (March 2009) and the disposal of unmarketable fruits at harvest. (g) Electron micrograph of purified particles from infected Nicotiana glutinosa, with a modal size of 720–750 nm (bar represents 500 nm).

Table 1. Partial host range of the Israeli strain of Tomato mild mottle virus (TomMMoV-IL)
FamilyHost plantCommon nameSymptomsaSerologyb
  1. a

    NS, no symptoms; R-LL-pc, red local lesions with pale centers; B-LL-yb, brown local lesions with yellow borders; VC, vein clearing; DGT, distortion of growth tips; DAL, distortion of apical leaves; SG, stopped growth; ML, mottling on leaves; MM-L, mild mottling on leaves; FD, fruit distortion.

  2. b

    Crude plant sap samples were analysed by indirect ELISA using a polyclonal antiserum specific to TomMMoV-IL (1: 1500 dilution).

Amaranthaceae Gomphrena globosa Globe amaranthR-LL-pc
Chenopodiaceae Chenopodium quinoa LambsquartersNS
Ch. amaranticolor B-LL-yb
Cucurbitaceae Cucumis sativus CucumberNS
Cu. melo MelonNS
Citrullus lanatus WatermelonNS
Fabaceae Vigna unguiculata CowpeaNS
Solanaceae Capsicum annuum PepperNS
Datura stramonium JimsonweedVC, DGT+
Nicotiana glutinosa  VC, DAL+
N. tabacum cv. SamsunMM-L, VC+
N. tabacum cv. Samsun NNMM-L, VC+
N. tabacum cv. XanthiVC+
N. tabacum cv. White BurleyML, VC+
N. rustica DAL+
N. clevelandii DAL+
N. benthamiana DAL+
N. sylvestris DAL+
Physalis floridana DGT, SG+
Petunia sp.NS+
N. occidentalis NS+
Solanum lycopersicum cvs Marmande and LindaTomatoNS+
S. melongena AubergineMM-L, FD+
S. nigrum Black nightshadeVC, DAL+

Virus purification, particle morphology and testing of Koch's postulates

The purification procedure that was developed for virion isolation yielded highly purified preparations. Both the TEM analysis and electrophoresis of the viral CP confirmed the absence of any foreign viral or plant protein contaminants in the final preparation (Figs 1g and 2a). A TEM analysis of the purified preparations indicated their flexible filamentous morphology, with a modal length of approximately 720 nm (Fig. 1g). The viral CP subunit migrated on 12·5% polyacrylamide gels as a single band of approximately 37·5 kDa (Fig. 2a–e). The purified viral particles from three independent viral purifications were used for three mechanical inoculation experiments, each including two aubergine plants, three N. glutinosa plants and four tobacco plants (27 plants in total). All the inoculated plants developed typical symptoms 14–16 days postinoculation. Virus purification from the inoculated test plants showed the presence of virus particles identical in their morphological and serological properties to those extracted from the source plants. Reinoculation of the purified virus preparation to aubergine plants resulted in symptoms identical to those found in infected fields.

Figure 2.

Serological relationship between the Israeli and the Yemeni strains of Tomato mild mottle virus (TomMMoV). (a) M, protein size ladder; V, electrophoretic separation of purified virions extracted from TomMMoV-IL-infected plants. (b) Western blot analysis using an antiserum prepared against the Yemeni strain of TomMMoV. Lanes: 1, leaf extract of TomMMoV-IL infected aubergine; 2, leaf extract of uninfected aubergine. (c) Unspecific Ponceau S staining of blotted crude leaf extracts from: TomMMoV-IL-infected (lanes 3–5) and uninfected plants (lanes 6–8) of Nicotiana glutinosa (lane 3), Nicotiana tabacum cv. White Burley (lane 4) and Solanum melongena (lane 5); Capsicum annuum cv. Maor infected with Potato virus Y (PVY). (d) Western blot analysis using TomMMoV-IL homologous antiserum; lanes as in (c). (e) Transmission of TomMMoV-IL by Bemisia tabaci. RNA preparations from: Eg-I aubergine plants naturally infected with TomMMoV-IL; Eg-H, uninfected aubergine plants; 1–6, aubergine plants inoculated with a laboratory culture of Bemisia tabaci; and 7–12, tobacco plants inoculated with B. tabaci, spotted on nylon membranes hybridized using a riboprobe produced by in vitro transcription of a cloned fragment for the TomMMoV-IL NIb and CP genes. (f) Acquisition of TomMMoV-IL by the whitefly vector: total RNA was extracted from insects following 48 h of acquisition access on (1) TomMMoV-IL-infected aubergine plants; (2) TomMMoV-IL-infected tobacco plants; (3) Cucumber vein yellowing virus (CVYV)-infected cucumber plants; (4) non-viruliferous B. tabaci from a laboratory culture raised on cotton. The insect RNA served as a template in an RT-PCR amplification using specific primers for the TomMMoV-IL CP gene and specific primers for the B. tabaci actin gene, which served as an internal control.

The NGS procedure was used twice in order to confirm the absence of mixed infections in both the original aubergine source plants and the virus culture on N. glutinosa. In both cases the NGS results indicated clearly that all the identified viral contigs belonged to the EMLMV genome (Dombrovsky et al., 2012).


None of the tested antisera against potyviruses or carlaviruses reacted with the purified antigen of EMLMV in ELISA (not shown). However, a positive reaction was obtained in western blots, when the viral antigen reacted with an antiserum that was prepared against the CP of TomMMoV (Walkey et al., 1994; Fig. 2b). An antiserum prepared against the EMLMV CP subunit, reacted specifically with its homologous antigen in western blots (Fig. 2c,d), but failed to react with either PVY or plant proteins; by contrast, TomMMoV antiserum reacted unspecifically with the plant's ribulose-1,5-bisphosphate carboxylase oxygenase (RuBisCO) (Fig. 2b).

Electron microscopy of virus-infected tissues

Examination of thin sections prepared from EMLMV-infected N. glutinosa leaves revealed the presence of dichotomously branched pinwheel-like inclusion bodies typical of those induced by members of the Potyviridae (Fig. 3a–c). Moreover, in some of the sections, cylindrical inclusion bodies and amorphous material containing diamond-shaped crystals were also observed (Fig. 3a,b).

Figure 3.

Transmission electron microscopy images of ultrathin tissue sections from a Nicotiana glutinosa plant infected by the aubergine strain of Tomato mild mottle virus (TomMMoV-IL). (a) Cytoplasmic inclusion bodies: Pi, pinwheel inclusions; Cr, crystalline structure. (b) Infected cell showing bundles of cylindrical inclusions (CI). (c) Pinwheel inclusion.

Identification of the insect vector

The results of transmission experiments, in which three aphid species known as vectors for many plant viruses were tested, are summarized in Table 2. Based on symptomatology, TEM analyses and dot-blot hybridization analyses, none of the tested aphids was able to transmit the EMLMV.

Table 2. Transmission of the Israeli strain of Tomato mild mottle virus (TomMMoV-IL) by three aphid species
Vector speciesPropagation plantSource plants for AAFaInoculation plants for IAFbTransmission rate (%)Number of transmission experiments ± SDTotal number of plants
  1. a

    AAF, acquisition access feeding period.

  2. b

    IAF, inoculation access feeding period.

  3. c

    Zucchini yellow mosaic virus (ZYMV) was used as a positive control.

Myzus persicae Brassica perviridis TomMMoV-IL-infected Nicotiana glutinosa N. glutinosa 0321
  S. melongena 0322
M. persicae B. perviridis TomMMoV-IL-infected N. tabacum N. tabacum 0428
  S. melongena 0321
M. persicae B. perviridis TomMMoV-IL-infected S. melongena S. melongena 0323
M. persicae B. perviridis ZYMV-infected Cucurbita pepo cv. Ma'ayancCucumis sativus cv. Beit Alpha80·44 ± 7·331
Macrosiphum euphorbiae Solanum melongena TomMMoV-IL-infected S. melongena S. melongena 0320
Aphis gossypii S. melongena TomMMoV-IL-infected S. melongena S. melongena 0321

The acquisition of the virus by B. tabaci was confirmed by RT-PCR. The presence of the virus in total RNA extracted from whiteflies following 3 h of starvation and 48 h of AAF is shown in Fig. 2f. The amplified fragments shown on the gel represent the viral CP gene with the expected size. The viral origin of the fragment was further verified by dot-spot hybridization with EMLMV-riboprobe (data not shown).

In whitefly transmission experiments with B. tabaci, EMLMV was successfully transmitted from infected to healthy aubergine plants, or from infected to healthy tobacco cv. Samsun plants. However, the transmission efficiency of B. tabaci was poor and a large number of B. tabaci individuals (a few hundred per plant) were required for successful transmission. The transmission of EMLMV by whiteflies occurred in five out of seven independent transmission experiments, yielding the following transmission rates (infected/inoculated test plants): 6/6, 6/8, 0/7, 5/7, 0/6, 8/8 and 6/8. Independent serial transmission experiments on aubergine revealed that the virus was able to persist in the whitefly vector for at least 5 days (Table 3). Transmission was confirmed by dot-spot hybridization of total RNA extracted from the whitefly-inoculated test plants using an EMLMV-specific riboprobe (Fig. 2e), or by western blot analysis with the specific antiserum against EMLMV (not shown). Transmission was further validated by TEM examination of dip preparations from the whitefly-inoculated aubergine plants and by sap inoculation from the whitefly-inoculated plants to N. glutinosa test plants (not shown).

Table 3. Detection of the Israeli strain of Tomato mild mottle virus (TomMMoV-IL) in aubergine plants inoculated via Bemisia tabaci in serial transmission experiments
Number of days post-IAFaSerial whitefly transmission on N. glutinosa (infected plants/total inoculated plants)
  1. a

    IAF, inoculation access feeding.

  2. b

    Dot-spot hybridization confirmation.

  3. c

    First replica.

  4. d

    Second replica.

Start 24 h+b (4/4)c; (4/4)d
2 days (48 h)+ (4/4)c; (4/4)d
3 days (72 h)+ (4/4)c; (3/4)d
5 days (120 h)+ (2/3)c; (3/3)d
9 days (216 h) (0/2)c; (0/3)d


The tentatively named EMLMV was found to be the causal agent of a new disease of aubergine that has spread in Israel in recent years.

Serological analysis clearly showed that EMLMV has a close relationship with TomMMoV (Fig. 2b), which was first described in 1994 as an aphidborne ‘unusual potyvirus’ isolated from tomato, D. stramonium and S. nigrum in the Republic of Yemen (Walkey et al., 1994). Later, this virus was detected in tomato, D. stramonium and Nicandra physalodes during a field survey in Ethiopia (Hiskias et al., 1999). No published information is available on the ability of TomMMoV to infect aubergine. The determined host range of EMLMV is quite similar to that described for TomMMoV. However, EMLMV did not elicit any disease symptoms in tomato (cvs Marmande and Linda), whilst the Yemeni isolate of TomMMoV induced visible symptoms in tomato cvs Roma (Walkey et al., 1994) and Linda (Abraham et al., 2012). Interestingly, both the Yemeni and the Ethiopian isolates of TomMMoV were collected in tomato plots, while EMLMV was not identified in tomatoes grown in the vicinity of infected aubergine in Israel. The differences in host range and host reaction between EMLMV and TomMMoV may be explained by the significant differences in the amino acid sequences of the viral P1 and PIPO proteins of the two viruses. The sequence identities between the P1 genes of TomMMoV and EMLMV, at both the nucleotide and the amino acid levels, are 79 and 86%, respectively, while the PIPO ORFs of the two viruses shares 80 and 61% identity at the nucleotide and amino acid levels, respectively (Table 4). The viral P1 and PIPO proteins are putatively involved in host-range determination and long-distance movement, respectively (Salvador et al., 2008; Wen & Hajimorad, 2010). Both EMLMV and TomMMoV have filamentous particles with a size typical of Potyviridae. Both induce the formation of unique, dichotomously branched pinwheel inclusion bodies in their hosts (Fig. 3; Walkey et al., 1994). The putative EMLMV CP is composed of 304 amino acid residues, with a calculated MW of 34·11 kDa (Dombrovsky et al., 2012). However, a slightly higher value (37·5 kDa) was estimated from the SDS-PAGE analysis (Fig. 2a–d).

Table 4. Percentages of nucleotide and deduced amino acid sequence identities between the complete genomes of the Israeli strain of Tomato mild mottle virus (TomMMoV-IL) and the Ethiopian isolate of the virus (TomMMoV)
TomMMoV-ILTomMMoV amino acidsa (%)TomMMoV nucleotides (%)
  1. a

    Deduced amino acid sequences.

Whole-genome identities 7548/9288 (81)
Polyprotein identities2780/3011 (92) 
P1282/327 (86)774/981 (79)
HC-Pro354/390 (91)942/1170 (81)
P3286/319 (90)794/957 (83)
6K149/53 (92)123/152 (81)
CI616/642 (96)1588/1926 (82)
6K248/53 (91)127/159 (80)
NIa396/417 (95)1031/1251 (82)
NIb468/506 (92)1205/1493 (81)
CP279/304 (92)735/906 (81)
PIPO23/38 (61)146/182 (80)

The Yemeni isolate of TomMMoV was reported as a non-persistent aphidborne potyvirus (Walkey et al., 1994), while recent transmission results with both the Yemeni and the Ethiopian isolates were essentially negative, showing a single transmission event by B. tabaci, which leaves vector identity vague (Abraham et al., 2012). By contrast, EMLMV is unequivocally a whitefly-transmitted virus (Tables 2 and 3; Fig. 2e). No transmission was recorded when three aphid species were tested (Table 2). Transmission efficiency of EMLMV in the laboratory was relatively poor, contradictory to its rapid spread under field conditions, which is probably enhanced by large field populations of B. tabaci (mostly the B biotype) during autumn. Similar results were observed in epidemiological and transmission studies of other members of the genus Ipomovirus. Hollings et al. (1976) observed a low rate of transmission of Sweet potato mild mottle virus (SPMMV), the type member of the genus Ipomovirus, by whiteflies. Large numbers of viruliferous whiteflies were required to obtain high CVYV transmission rates in cucumbers (Harpaz & Cohen, 1965). Furthermore, extremely low transmission rates and inconsistency in transmission rates were also observed for Cassava brown streak virus (CBSV) (Maruthi et al., 2005). A large number of whiteflies were needed for the successful transmission of Squash vein yellowing virus (SqVYV) (Adkins et al., 2007). Alignment analysis of the amino acid sequence of the EMLMV, TomMMoV and SPMMV CPs revealed a deletion of 50–60 amino acids located between positions approximately 50–105 in the CP N-terminus, which is absent from the CPs of SqVYV, CVYV and CBSV isolates (Fig. 4). Future studies may reveal if this difference is significant in the transmission mechanism of ipomoviruses by their whitefly vectors. In both TomMMoV and EMLMV the CP lacks the highly conserved DAG motif located at the potyvirus CP N-terminus which is involved in the interaction with the potyviral HC-Pro during aphid transmission (Atreya et al., 1995). The absence of the DAG motif is characteristic of all described ipomoviruses (Colinet et al., 1996; Janssen et al., 2005; Adkins et al., 2007; Mbanzibwa et al., 2009). Interestingly SqVYV, CVYV and CBSV do not have an HC-Pro gene, while SPMMV, TomMMoV and EMLMV do; however, its involvement in the transmission mechanism is unknown as yet.

Figure 4.

Alignment of the deduced amino acid sequences of the coat proteins (CP) of the Ethiopian and Israeli strains of Tomato mild mottle virus with the deduced amino acid sequences of members of the genus Ipomovirus. Line 1: aubergine strain of Tomato mild mottle virus (TomMMoV-IL, current study; HQ840786); line 2: TomMMoV (AF359575); line 3: Sweet potato mild mottle virus (SPMMV; NP_620656); line 4: Squash vein yellowing virus (SqVYV; YP_001788991); line 5: Cucumber vein yellowing virus (CVYV; NC_006941); line 6: Cassava brown streak virus (CBSV; ACM48176); line 7: Ugandan cassava brown streak virus (UCBSV; YP_004063681). SPMMV, TomMMoV and TomMMoV-IL have a deletion of approximately 60 amino acids at the N-terminus of their CP which is absent in the CPs of SqVYV, CVYV and CBSV.

Obtaining the full genome sequence and genome organization of EMLMV and TomMMoV (Abraham et al., 2012; Dombrovsky et al., 2012) has paved the way for elucidating the phylogenetic relationship between these viruses and providing explanations for the differences in their biological properties.

The Ninth Report of the International Committee on Taxonomy of Viruses (ICTV) provides the updated criteria for species demarcation within the family Potyviridae (King et al., 2012). According to this report, different species have CP amino acid sequence identity less than 80%; and nucleotide sequence identity less than 76%, either in the CP or over the whole-genome.

A blast search and sequence comparison of the complete genomes of EMLMV and TomMMoV reveal identities of 81% at the nucleotide level and 92% at the amino acid level. Similar identity levels were obtained for the nucleotide and amino acid sequences of the CP of these viruses (Table 4). These identity levels are above the threshold that would classify these viruses as members of the same species.

According to the criteria described above, EMLMV and TomMMoV are isolates of the same virus and both are putative members of the genus Ipomovirus in the family Potyviridae (Abraham et al., 2012; Dombrovsky et al., 2012). High bootstrap values obtained from the phylogenetic analysis of the deduced amino acid sequences of the polyproteins of virus species in the Potyviridae emphasize the intra- and intergenus diversity levels. Intragenus diversity within the genus Ipomovirus is demonstrated by the phylogenetic tree (Fig. 5), showing the grouping of four putative subclusters: subcluster I includes CVYV and SqVYV, subcluster II includes CBSV and Ugandan cassava brown streak virus (UCBSV), subcluster III contains only the Ipomovirus type member SPMMV, and subcluster IV comprises TomMMoV and TomMMoV-IL.

Figure 5.

Phylogenetic tree derived from the deduced amino acid sequences of the polyproteins of the Israeli strain of Tomato mild mottle virus (TomMMoV-IL; HQ840786) and those of several other viruses belonging to the family Potyviridae: TomMMoV (HE600072), Sweet potato mild mottle virus (SPMMV; Z73124), Squash vein yellowing virus (SqVYV; YP_001788991), Cucumber vein yellowing virus (CVYV; AY 578085), Cassava brown streak virus (CBSV; FN433930), Ugandan cassava brown streak virus (UCBSV; YP_004063681), Triticum mosaic virus (TriMV; FJ669487), Sugarcane streak mosaic virus (SCSMV, ADE34528), Wheat streak mosaic virus (WSMV; AF057533), Barley yellow mosaic virus (BaYMV; X697579), Blackberry virus Y (BlYV; AY994084), Ryegrass mosaic virus (RGMV-AV; AF035818), Oat necrotic mottle virus (ONMV; AAQ91884), Brome streak mosaic virus (BrSMV; Z48506) and Potato virus Y (PVY-N; P18247). Each polyprotein-coding sequence was aligned using the muscle program for sequence alignments (Edgar, 2004). This tree was constructed based on a maximum likelihood using the phyml software with 100 bootstrap replicates (Guindon & Gascuel, 2003).

Based on these rules of molecular taxonomy, serological cross-reaction between TomMMoV antiserum and TomMMoV-IL CP, and the similar host range of laboratory test plants, it is concluded that the formerly named EMLMV (Dombrovsky et al., 2012) is a distant strain of TomMMoV (Walkey et al., 1994; Monger et al., 2001; Abraham et al., 2012), designated TomMMoV-IL. The current study reports that TomMMoV-IL is the causal agent of a new devastating aubergine disease which is spread in nature by the whitefly vector B. tabaci.


The authors wish to acknowledge Dr Vered Holdengreber from the Department of Plant Pathology, the Volcani Center, for having conducted the ultrathin sections and the transmission electron microscopy analysis; and Dr Adi Doron-Faigenboim from the Department of Plant Science, the Volcani Center, for the assistance in the phylogenetic tree analysis.