Molecular and biological characterization of Chilli leaf curl virus and associated Tomato leaf curl betasatellite infecting tobacco in Oman

Total nucleic acid was extracted from leaf samples using a cetyltrimethylammonium bromide-based method [12] and kept at − 20 °C. Extracted DNA was used as a template in polymerase chain reaction (PCR) with primer pairs for the detection of begomoviruses (TYLCD-356 (5′-ATCATTTCCACKCCCGYCTCGA-3′/TYLCD-1044 5′-GCRTGMGTACABGCCATATACA-3′), amplifying an ~ 800 nt product, betasatellites (Sat101/Sat102), amplifying an ~ 1350 nt product [13], and alphasatellites (DNA101/DNA102), amplifying an ~ 1380 nt product [14]. Additionally the primer pair βC1F (5′-AGACCCGGGATGACGATCAGATATAATAACA-3′)/βC1R (5′-ACGTCGACTCACACACACACTTTCGTACA-3′), amplifying a ~ 350 nt product, was used in PCR for the detection of betasatellites.

Circular DNA molecules in nucleic acid samples were enriched using rolling circle amplification (RCA) as described earlier [15]. Restriction of the high molecular weight concatameric RCA products with BamHI resulted in ~ 2.7 kb fragments, which were eluted from agarose gels using a GeneJet Gel Extraction Kit (Thermo Fisher Scientific) and cloned in BamHI restricted pGEM-3zf (+) (Promega Madison, USA). Potentially full-length clones resulting from RCA (for begomovirus) and PCR (for betasatellite) were sequenced commercially using the primer-walking approach (Macrogen Inc., South Korea).

Sequences were assembled using SeqMan, part of the Lasergene package of sequence analysis software (DNA Star Inc., Madison, WI, USA). Related sequences available in the GenBank database were identified using the Basic Local Alignment Search Tool nucleotide (BLASTn) [16] run on-line (http://​blast.​ncbi.​nlm.​nih.​gov/​Blast.​cgi). Open reading frames (ORFs) in sequences were identified using the ORF Finder program run on-line (https://​www.​ncbi.​nlm.​nih.​gov/​orffinder/​). The Species Demarcation Tool (SDT), with the MUSCLE option, was used to calculate sequence identity values [17]. Pairwise multiple sequence alignments were produced using the MUSCLE algorithm implemented in MEGA6 [18]. The evolutionary relationships between sequences were determined by constructing phylogenetic trees using Clustal X (neighbor-joining method) and displayed using Treeview [19].

Clone Tob11was digested with HindIII and SalI to release a fragment of ~ 1000 bp which was ligated into the binary vector pGreen0029 [20]. Then the full-length insert of Tob11, released using HindIII, was ligated into the pGreen0029 clone containing the fragment, linearized using HindIII, to yield a partial direct repeat of the virus genome. A similar strategy was used to produce a construct for clone Tob44 in pGreen0029 using a ~ 425 bp KpnI-XbaI fragment and the full-length insert released using KpnI.

Binary vector constructs were electroporated into Agrobacterium tumefaciens strain LBA4404. Agrobacterium inocula were prepared and inoculated to Nicotiana benthamiana, N. glutinosa and N. tabacum seedlings by agroinfiltration as described previously [21]. Infiltrated plants were maintained in an insect-free growth room at 28 °C with a photoperiod of 16 h and monitored daily for the appearance of symptoms.

Total genomic DNA (~ 10 μg) extracted from plant tissue was resolved on 1.5% agarose gels, transferred overnight by capillary action onto Hybond N⁺ membrane (Amersham) and UV cross-linked. Probes were produced by PCR using a DIG-high prime kit (Roche Applied Science) and primers βC1F (5′-AGACCCGGGATGACGATCAGATATAATAACA-3′)/ βC1R (ACGTCGACTCACACACACACTTTCGTACA-3′) for betasatellite and (qPCRChV F (5′-AGTTCATATAGGGAAGGTTATGTGTA-3′) / qPCRChV R (5′- GGACAAGGAAGAACATCACACTATTC-3′) for virus. Membranes were hybridized and washed as described previously [22, 23] and hybridization was detected by colorimetric detection with nitro blue tetrazolium chloride and 5-bromo-4-chloro-3-indolyl-phosphate, toluidine-salt (Roche Applied Science).

During a survey for virus-infected plants in December 2015 on an organic farm in Sohar (Al Batinah region, Sultanate of Oman), fields of tobacco (Nicotiana tabacum) were observed with plants showing severe downward leaf curling, leaf thickening, vein swelling, mild yellowing and stunting (Fig. 1). In the fields between 60 and 70% of plants were exhibiting symptoms.

Leaves from four symptomatic N. tabacum plants, originating from two independent but neighbouring fields (500 m apart), were collected, as well as a non-symptomatic plant from each field. A high molecular weight product was obtained by RCA with DNA extracted from all four symptomatic plants but not from the non-symptomatic plants (results not shown). Three potentially full-length clones (Tob11, Tob12 and Tob13), resulting from digestion of RCA product with BamHI, were obtained, one from each of three N. tabacum plants (Table 1). The sequence of each clone was 2761 nucleotide (nt) in length and are available in the nucleotide sequence databases under the accession numbers given in Table 1. Analysis of sequences showed them to have an organization of genes typical of the genomes of monopartite begomoviruses originating from the OW, with two predicted genes in the virion-sense and four in the complementary-sense (Table 1). SDT analysis showed the three sequences to have greater than 99% nucleotide sequence identity, showing them to be isolates of a single begomovirus species, based on current guidelines for species demarcation [24]. Comparison of the sequences to sequences available in the GenBank database using BLASTn showed them to have high levels of identity to isolates of the monopartite ChiLCV. Pairwise sequence analysis using SDT, and selected sequences from the database, showed the isolates from tobacco have the highest levels of sequence identity (99.9 to 100%) to an isolate of ChiLCV recently identified in watermelon in Oman (KX787939 [25];). This was supported by a phylogenetic analysis based upon an alignment of the full-length sequences of ChiLCV obtained from tobacco and selected sequences from the databases (Fig. 2a). Isolates of ChiLCV originating from Oman form a clade distinct from the ChiLCV isolates originating from the Indian sub-continent, with the sequences from tobacco being most closely related to the ChiLCV isolate from watermelon.

Three clones were obtained by PCR with betasatellite primers β101/102, one each from the three plants from which the virus clones were obtained (Additional file 1: Table S1). The sequences are 1375 (clone Tob45) and 1377 nt (clones Tob44 and Tob46) in length and are available in the nucleotide sequence databases under the accession numbers given in Additional file 1: Table S1. Analysis of the sequences showed them to have an organization and features typical of betasatellites [5], consisting a single conserved (between all betasatellites) ORF in the complementary-sense (known as βC1) having the capacity to encode a product of 118 amino acid, a region of sequence rich in adenine (coordinates 722–968 nt) and a sequence well conserved between all betasatellites (known as the satellite conserved region; coordinates 1252–21 nt) that contains a predicted stem-loop structure with, as part of the loop, the nonanucleotide sequence TAATATTAC with similarity to the origin of replication of geminiviruses.

SDT analysis showed the three betasatellites isolated from tobacco to have between 99.7 and 99.8% nucleotide sequence identity with each other. Comparison of the three sequences from tobacco to sequences available in the GenBank database using BLASTn showed them to have high levels of nucleotide sequence identity to isolates of the betasatellite ToLCB. Pairwise sequence analysis using SDT, and selected ToLCB sequences from the databases, showed the isolates from tobacco have the highest levels of sequence identity (99.9%) to an isolate of ToLCB recently identified in watermelon in Oman (KX787940 [25];). Overall the sequences of ToLCB from tobacco showed higher levels of sequence identity to isolates of ToLCB originating from the Arabian Peninsula than to isolates either from Iran or the sub-continent. This is supported by a phylogenetic analysis (Fig. 2b) which shows ToLCB isolates from the Arabian Peninsula, including those identified here in tobacco, form a clade separate from other ToLCB isolates but to be most closely related to isolates from Iran.

To satisfy Koch’s postulates a partial direct repeat construct of ChiLCV was inoculated to Nicotiana benthamiana, N. glutinosa and N. tabacum plants, either in the presence of absence of a construct for the associated ToLCB (Table 2; Fig. 3). Inoculation of just the ChiLCV construct to N. benthamiana seedlings resulted in severe upward leaf curling and vein swelling on the undersides of young, newly developing leaves initiating at 16 days post-inoculation. Co-inoculation of N. benthamiana seedlings with the constructs for both ChiLCV and ToLCB resulted in severe downward leaf curling and crumpling young, newly developing leaves at 12 dpi.

Inoculation of N. glutinosa with only the ChiLCV construct resulted in leaf crumpling which initiated at 21 dpi. As was the case for N. benthamiana, co-inoculation of N. glutinosa with the construct for ToLCB resulted in symptoms which appeared earlier (18 dpi) and the symptoms showed a pronounced downward leaf curling. Similarly for N. tabacum inclusion of ToLCB in the inoculum led to a decrease in latent period (from 21 to 18 dpi) and an increase in symptom severity, with slightly more severe downward leaf curling, although this was less pronounced than for either N. benthamiana or N. glutinosa. However, for both singly inoculated plants and co-inoculated plants, the infectivity was significantly less to N. tabacum (50 and 60%, respectively) than to either of the other Nicotiana species.

The presence of either ChiLCV, or ChiLCV and ToLCB, was confirmed in all symptomatic inoculated plants by diagnostic PCR, whereas neither component was identified in non-inoculated and non-symptomatic inoculated plants. Southern blot analysis of DNA samples extracted from inoculated plants showed the presence of large amounts of viral DNA in both singly and co-inoculated plants and high amounts of betasatellite DNA for co-inoculated plants for N. benthamiana and N. glutinosa (Fig. 4). However, for N. tabacum, although there was a large amount of viral DNA detected for the plant inoculated with ChiLCV alone, much less was detected in the plant co-inoculated with ChiLCV and ToLCB. Nevertheless, the amount of viral DNA detected was comparable to the amount of viral DNA detected in field infected tobacco plant, although the amount of betasatellite detected was significantly less. For all inoculated plants, as well as the field collected plant, there was a pronounced smear of small molecular weight material below the viral and betasatellite ssDNA forms.