Characterization and a RT-RPA assay for rapid detection of Chilli Veinal mottle virus (ChiVMV) in tobacco

For the isolation of potential viral pathogens infecting tobacco plants, the symptomatic leaf samples were collected from Luzhou, Sichuan province, China in 2018. Total RNA was extracted using RNAiso Plus reagent (TaKaRa, Dalian, China) from tobacco leaves following the manufacturer’s guidelines. Viruses were detected by RT-PCR with four pairs of degenerate primers [27], including CMVCPf/CMVCPr (for cucumber mosaic virus), Tob-Uni1/Tob-Uni2 (for tobamoviruses), PotyF/PotyR (for potyviruses) and TSWVf/TSWVr (for tomato spotted wilt virus) (Additional file 1: Table S1). The amplified PCR products were purified with EasyPure Quick Gel Extraction kit (Transgen Biotech, Beijing, China), cloned into pEASY-T1 Cloning Vector (Transgen Biotech) and directly sequenced (Sangon Biotech, Shanghai, China), respectively.

The crude extract was obtained by homogenizing ChiVMV-infected tobacco leaf sample in phosphate buffer (0.01 M, pH 7.2) at 1:5 (w/v) ratio. The crude extract was inoculated onto five healthy plants of tobacco, pepper, tomato (Solanum lycopersicum L.) and alkekengi (Physalis alkekengi L.) by rubbing carborundum-dusted leaves according to a previous report, respectively [28]. Inoculated plants were grown under greenhouse conditions at 24~26 °C, and observed daily for symptom development. To verify the presence of ChiVMV-LZ in inoculated plants, the systemically infected leaves were harvested for RNA isolation and RT-PCR was performed as described [29].

The RT-PCR was performed using six degenerate primers for ChiVMV genome amplification (Additional file 1: Table S1). The 5′- and 3′-terminal sequences of ChiVMV were obtained through rapid amplification of cDNA ends (RACE) assays using SMARTer™ RACE cDNA Amplification kit (Clontech, Mountain View, USA) according to the manufacturer’s protocol. The obtained PCR products were cloned into pEASY-T1 Cloning Vector (Transgen Biotech) and sequenced (Sangon Biotech). The DNAMAN package (Version 8.0.8, Lynnon Biosoft, USA) was used for the full-length genome alignment. Amino acid sequence of ChiVMV-LZ was compared with those of other eleven ChiVMV isolates (GenBank Accession No. AGN92430.1, AGN92431.1, AGE47830.1, NP982308.1, AMV75279.1, ACX53640.1, ALH24936.1, CAB43195.2, CAP39938.1, ADP30806.1, ADP30807.1) obtained from GenBank. Four published pepper-infecting potyviruses (BAJ10980.1, ACE80681.1, CAJ57401.1 and ABG56784.1) were used as outgroup members. A total of sixteen sequences were aligned using the ClustalW algorithm of MEGA7 [30]. The phylogenetic tree of ChiVMV was constructed using 1000 bootstrap repeat sequences as matrices in the phylogenetic process of the adjacent connections [31].

One pair of primers (RPA-ChiVMV-F: 5′-AAGATATGGGCTTCAAAGAAACCTTACCG-3′ and RPA-ChiVMV-R: 5′-CCTACCCTACCGTCCAGTCCGAACATCCT-3′) used for RT-RPA was designed based on the conserved coat protein genes for ChiVMV-LZ detection. The general guidelines for designing the RPA primers were followed in RPA manufacturers’ websites (http://​www.​twistdx.​co.​uk/​en/​support/​rpa-assay-design-2), which recommended one pair of primers with 30–35 nucleotides for the optimal formation of recombinase/primer. To optimize the reaction time, RPA reactions were executed for 10, 20 and 30 min respectively using the cDNA from ChiVMV-infected tobacco plants. Conventional RT-RPA assays were performed in a 38 °C water bath for 20 min using a TwistAmp Basic kit (TwistDX, Cambridge, UK). The 50 μL reaction volume included 29.5 μL of rehydration buffer, 2.4 μL of each RPA-ChiVMV-F/R primer (10 μM), 2.5 μL of magnesium acetate (280 mM), 12.2 μL of nuclease-free water, and 1.0 μL of cDNA. RPA amplicons were purified using a SanPrep Column PCR Product Purification kit (Sangon Biotech) and then analyzed by 2% agarose gel electrophoresis. To evaluate the specificity of the established RT-RPA assay, ChiVMV-, TSWV-, potato virus Y (PVY)- or tobacco vein banding mosaic virus (TVBMV)-infected tobacco plants were tested, with the healthy tobacco plants as negative controls. For the sensitivity assessment, total RNA extracted from ChiVMV-infected tobacco plant was diluted to 1 ng, 10 pg, 1 pg, 100 fg, 10 fg and 1 fg, respectively. The diluted templates were then used for the first-strand cDNA synthesis with the M-MLV reverse transcriptase as instructed (Promega, Madison, USA). Each cDNA template was detected using RT-RPA and RT-PCR, respectively. To evaluate the feasibility of RT-RPA for ChiVMV detection in the field, a total of 71 tobacco plants with ChiVMV or viral-like symptoms were detected by RT-RPA and RT-PCR, respectively.

During the investigation of viral diseases in tobacco in Luzhou, Sichuan in Jun 2018, leaf samples with yellowing and shrinking symptoms were collected from field-grown tobacco (Fig. 1 a). Four pairs of universal primers (CMVCPf/CMVCPr, Tob-Uni1/Tob-Uni2, PotyF/PotyR and TSWVf/TSWVr) were used to detect the collected tobacco samples by RT-PCR, and an expected 320-bp fragment was obtained only with primers PotyF/PotyR (Fig. 1 b). After cloning and sequencing, the obtained nucleotide sequences of PCR products were aligned by BLASTn in NCBI database. The results showed that the sequences had over 98% homology with that of ChiVMV. Therefore, we preliminarily determined that the symptomatic tobacco samples were infected by ChiVMV, which was named ChiVMV-LZ.

To explore the pathogenicity of ChiVMV-LZ, the virus was mechanically inoculated onto different host plants. After ten days post inoculation (dpi), ChiVMV infection resulted in various symptoms in all tested solanaceous plants, including tobacco, pepper, tomato and alkekengi. Many round bright spots appeared on the leaves of ChiVMV-infected tobacco plants, with the manifestation of chlorosis and yellowing. Moreover, the leaves shrank with droopy and folded leaf margin, especially for the lower leaves. ChiVMV-infected pepper leaves prominently showed darker vein stripes, shrinkage of leaf margin and deformity of leaf surface, especially in young leaves. For ChiVMV-infected tomato plants, the upper leaves shrank, rolled up and stretched slowly. The alkekengi plants infected by ChiVMV showed crimped, yellowed, and almost completely unextended tender leaves (Fig. 2 a). With RT-PCR detection using the specific primers Chi6-F/R (Additional file 1: Table S1), all the tested host plants showed positive infection of ChiVMV (Fig. 2 b). These results demonstrated the pathogenicity of ChiVMV-LZ on different host plants.

Near full-length genomic sequence of ChiVMV-LZ was obtained by RT-PCR using primers designed based on the sequences of reported ChiVMV isolates. Six sequenced fragments of 1944 bp, 1581 bp, 1728 bp, 1579 bp, 1871 bp and 1780 bp in length were obtained using primers Chi-F1/R1, Chi-F2/R2, Chi-F3/R3, Chi-F4/R4, Chi-F5/R5 and Chi-F6/R6, respectively (Additional file 3: Fig. S1; Additional file 1: Table S1). The 5′- and 3′-terminal genomic sequences of ChiVMV-LZ were determined by 5′ and 3′ RACE assays, respectively (Additional file 3: Fig. S1). The complete genomic sequence of ChiVMV-LZ (GenBank accession No. MK405594.1) was determined to be 9742 nt in length. The nucleotide homology of ChiVMV-LZ with other eleven ChiVMV isolates deposited in GenBank database was compared (Additional file 2: Table S2). The results showed that the nucleotide homology was 79.9~97.8% between ChiVMV-LZ and other eleven isolates. Notably, the nucleotide sequence identity value between ChiVMV-LZ and Sichuan isolates ChiVMV-Yp8 from pepper plants was the highest (97.8%). The identity values were 80.2~82.3% with the isolates from other provinces in China and 79.9~81.1% with the isolates from Korea and India (Additional file 2: Table S2). In addition, the sequence of ChiVMV-LZ was compared with the published sequences of potyviruses infecting pepper plants, including pepper vein mottle virus (PVMV), pepper mottle virus (PepMoV), pepper yellow mosaic virus (PepYMV) and pepper severe mosaic virus (PepSMV), with 49.0~64.7% nucleotide sequence identity (Additional file 2: Table S2). The amino acid-based neighbor-joining phylogenetic analysis showed that the isolates of ChiVMV from Sichuan province were clustered in one branch (Fig. 3), suggesting that ChiVMV-LZ had the closest relationship with the isolates of ChiVMV-Yp8 and ChiVMV-Pp4. Tobacco-infecting ChiVMV-LZ and pepper-infecting ChiVMV-Yp8 and -Pp4 probably came from the same strain, which were geographically related (Sichuan province). These results indicated that ChiVMV-LZ probably originated from infected pepper plants, and tobacco as a ChiVMV reservoir will probably increase the spread likelihood of ChiVMV to other solanaceous crops.

In order to detect ChiVMV conveniently and quickly, a RT-RPA method was established. Based on the optimization of the reaction conditions of RT-RPA assay, 20 min were selected as the reaction time in this study (Fig. 4 a). To evaluate the specificity of the established RT-RPA assay, ChiVMV-, PVY-, TSWV- or TVBMV-infected tobacco plants were used. The results revealed that only ChiVMV-infected tobacco plants showed clear and positive band in the agarose gel electrophoresis assay (Fig. 4 b), indicating that the designed primers for RT-RPA assay were specific to detect ChiVMV. For sensitivity evaluation, a series of dilutions of total RNA (1.0 × 10³ ng/μL) extracted from ChiVMV-infected tobacco plants, including 1 ng, 10 pg, 1 pg, 100 fg, 10 fg and 1 fg, were prepared and amplified by RT-PCR and RT-RPA, respectively. The RT-RPA method could detect transcripts with RNA concentration of 10 fg, while RT-PCR could produce positive reaction with that of at least 100 fg (Fig. 4 c), suggesting that RT-RPA was approximately 10-fold more sensitive than RT-PCR based on the agarose gel electrophoresis assay. These results showed that the established RT-RPA was a reliable and rapid method for ChiVMV detection with high specificity and sensitivity.

To evaluate the feasibility of RT-RPA method for ChiVMV detection in the field, 71 samples with ChiVMV or viral-like symptoms collected from Luzhou (Sichuan, China), Fuling (Chongqing, China) and Bijie (Guizhou, China) were tested by RT-RPA and conventional RT-PCR assays, respectively. The results showed that 56 samples showed ChiVMV-positive reaction by RT-RPA assay, and only 53 samples were positive by RT-PCR (Table 1). However, these three samples with negative ChiVMV in RT-PCR assay also developed typical symptoms soon, indicating that they were infected by ChiVMV but failed to be detected by RT-PCR assay. In addition, the ChiVMV-positive samples by RT-RPA assay showed clear bands in the agarose gel electrophoresis similar to that of RT-PCR (Fig. 5), suggesting its comparable performance to RT-PCR. These results demonstrated that the established RT-RPA assay was a rapid and sensitive technique for ChiVMV detection, and could be successfully applied in the field-collected samples.