Deep Sequencing of Banana Bract Mosaic Virus from ﬂowering Ginger (Alpinia Purpurata) and Development of an Immunocapture RT-LAMP Detection Assay

Arch Virol DOI 10.1007/s00705-016-2830-y

ORIGINAL ARTICLE

Deep sequencing of banana bract mosaic virus from ﬂowering ginger (Alpinia purpurata) and development of an immunocapture RT-LAMP detection assay

1,2 1 2 1 Jingxin Zhang • Wayne B. Borth • Birun Lin • Kishore K. Dey • 1 2 2 2 Michael J. Melzer • Huifang Shen • Xiaoming Pu • Dayuan Sun • John S. Hu1

Received: 9 October 2015 / Accepted: 14 March 2016 Ó Springer-Verlag Wien 2016

Abstract Banana bract mosaic virus (BBrMV) has never hypothesis that the A. purpurata isolate arrived in Hawaii been reported in banana plants in Hawaii. In 2010, how- from Southeast Asia. ever, it was detected in a new host, flowering ginger (Alpinia purpurata). In this study, we characterize the A. purpurata isolate and study its spread in flowering ginger Introduction in Hawaii. A laboratory study demonstrated that BBrMV could be transmitted from flowering ginger to its natural Banana bract mosaic virus (BBrMV) was first isolated from host, banana, therefore raising a serious concern about the bananas (Musa spp. L., Musaceae, Zingiberales) in the potential risk to the rapidly growing banana industry of Philippines in 1979 [1]. It has subsequently been detected Hawaii. To quickly monitor this virus in the field, we in other countries including India, Samoa, Sri Lanka, developed a robust immunocapture reverse transcription Thailand, and Vietnam [2]. Recently, the virus was dis- loop-mediated isothermal amplification (IC-RT-LAMP) covered in Colombia and Ecuador [3], indicating that this assay. Deep sequencing of the BBrMV isolate from A. pathogen can spread quickly and cause significant eco- purpurata revealed a single-stranded RNA virus with a nomic losses worldwide. In June 2009, BBrMV was also genome of 9,713 nt potentially encoding a polyprotein of detected in red- and pink-flowering ginger (Alpinia pur- 3,124 aa, and another predicted protein, PIPO, in the ?2 purata (Vieill.) K. Schum.) on the island of Oahu, Hawaii, reading-frame shift. Most of the functional motifs in the USA. The virus induced symptoms of mosaic, streaking, Hawaiian isolate were conserved among the genomes of and a severe cupping of leaves, with a browning of flowers isolates from one found in the Philippines and India. and reduction in their size and shelf life [4]. However, the A. purpurata isolate had an amino acid BBrMV belongs to the genus Potyvirus, family Po- deletion in the Pl protein that was most similar to the tyviridae. It has a poly-(A) tract at the 3’ terminus of the Philippine isolate. Phylogenetic analysis of an eastern monopartite, linear, single-stranded RNA (ssRNA) positive Pacific subpopulation that included A. purpurata was sense (?) genome [5, 6]. The virions lack an envelope and closest in genetic distance to a Southeast Asian subpopu- are flexuous, filamentous rods 720 to 850 nm long and lation, suggesting frequent gene flow and supporting the 12-15 nm in diameter (Viralzone, http://viralzone.expasy. org/viralzone/all_by_species/50.html). Only two com- pletely sequenced BBrMV genomes, one from the Philip- & John S. Hu pines [7] and one from India [8], have been published. In [email protected] both instances, the virus was isolated from banana. The

1 genome isolated in the Philippines (BBrMV-PHI) was Department of Plant and Environmental Protection Sciences, obtained using degenerate primers for members of the College of Tropical Agriculture and Human Resources, University of Hawaii, Honolulu, HI, USA genus Potyvirus [7], and the second, from India (BBrMV- TRY), was obtained using primers based on the nucleotide 2 Key Laboratory of New Technique for Plant Protection in Guangdong, Institute of Plant Protection, Guangdong sequence of the BBrMV-PHI isolate [8]. In our previous Academy of Agricultural Sciences, Guangzhou, China study [4] cloning of partial sequences of the BBrMV 123 J. Zhang et al. isolate from A. purpurata revealed 95-98 % nucleotide and elected to develop a loop-mediated isothermal ampli- sequence identity to the available sequences of BBrMV in fication (LAMP) assay for BBrMV. LAMP is a newly NCBI, indicating that the isolate might differ from isolates emerging molecular tool that uses Bst DNA polymerase found in other plant hosts or geographical locations. More with DNA strand-displacement activity to generate ampli- importantly, BBrMV has not been reported infecting fication products within 60 minutes [25]. It is simple, rapid, bananas in Hawaii, nor has A. purpurata been previously specific and cost-effective when compared to PCR [26]. reported as a host of this virus. We have fully characterized RT-LAMP (reverse transcription LAMP) synthesizes the genome of this A. purpurata isolate from Hawaii to cDNA from a viral RNA template in a one-step reaction determine its relatedness to other known strains. We used a [27]. It has a high-throughput diagnostic capacity that different strategy from previous BBrMV studies, applying meets the needs of the agricultural industry and quarantine deep sequencing and beginning with the isolation of dou- inspection protocols. ble-stranded RNA (dsRNA) from infected A. purpurata. Deep sequencing has been used widely to identify and characterize novel plant viruses. Published techniques have Materials and methods used DNA [9, 10], RNA [11, 12], dsRNA [13, 14], or short- interfering RNA [15, 16] as templates. We have previously Isolation of double-stranded RNA used dsRNA from infected plants to deep-sequence the genomes of four closely related closteroviruses infecting a Leaves of A. purpurata with symptoms of BBrMV were single common green ti plant, Cordyline fruticosa L. [17]. collected from the University of Hawaii at Manoa campus These previous studies indicated the great potential for in Honolulu. We isolated dsRNA from symptomatic leaves diagnostics and discovery of viruses by deep sequencing using the procedures described by Morris and Dodds [28] [18]. We used this procedure not only to characterize the as modified by Hu et al. [29]. Five microliters of dsRNA genome of BBrMV from A. purpurata but also to help was analyzed using 1 % (w/v) agarose gel electrophoresis, understand the differences between this virus and similar and the remainder was stored at -80 °C. viruses infecting banana. A further goal was to detect any undiscovered RNA viruses that also might be associated Library generation and deep sequencing with the disease symptoms described above. Another important question was the prevalence of this One microliter of dsRNA was heat-denatured at 95 °C for disease on A. purpurata. A reverse transcription poly- 10 minutes with 10 pmol of primer universal-dN6 [30] and merase chain reaction (RT-PCR) assay revealed that then immediately chilled on ice. First-strand cDNA syn- BBrMV is widespread on A. purpurata in Hawaii [4]. Red- thesis was performed using AMV reverse transcriptase and pink-flowering ginger (A. purpurata) is an ornamental (RNaseH?) (Promega, USA) according to the manufac- plant in the diverse family Zingiberaceae [19]. It is com- turer’s instructions. cDNA was then concentrated on a YM- monly grown for cut flowers in the home and also is in 50 column (Millipore, USA) together with 200 llof commercial production in Hawaii, where sales were valued nanopure H2O. The eluate was used as a primer/template in at $1.61 million in 2005-2006 [20]. BBrMV is transmitted a 20-ll hot-start overlap-extension PCR reaction using the by several species of aphids in a non-persistent manner [1, Easy-AÒ high-fidelity PCR cloning enzyme (Stratagene, 21]. It is also spread by vegetative propagation of the USA), with the following program: 95 °C for 7 minutes; rhizome and its offshoots and thus is easily spread by 10 cycles at 95 °C for 60 seconds, 55 °C for 60 seconds, humans. and 72 °C for 60 seconds; followed by 72 °C for 7 min- Bananas are an essential source of food in many tropical utes. Using the Easy-AÒ high-fidelity PCR cloning and subtropical countries [22]. The Hawaiian Islands are by enzyme, a 200-ll PCR reaction with 20 pmol of universal far the largest banana producer in the United States, fol- rPCR primer as the single primer [30] and 10 llofthe lowed by Florida. Banana production in Hawaii reached overlap-extension PCR reaction as the template was run 8,090 million metric tonnes in 2010 [23]. In this study, we according to the following cycling program: 95 °C for also tested whether BBrMV could be transmitted from A. 7 minutes; 35 cycles at 95 °C for 60 seconds, 58 °C for 60 purpurata to bananas by aphids, to determine if this virus seconds, and 72 °C for 60 seconds; followed by 72 °C for isolate could be a serious threat to banana production in 7 minutes. PCR products were fractionated by size (400- Hawaii. 700 bp) by agarose gel electrophoresis, ligated into To monitor spread of the disease, a robust detection pGEMÒ-T Easy Vector (Promega), and sequenced to assay for this virus is necessary. We considered the existing determine if BBrMV sequences were present. Following range of diagnostic assays for use in our study, including treatment with ExoSAP-ITÒ (USB/Affymetrix), the puri- ELISA [6], RT-PCR [7] and immunocapture RT-PCR [24] fied PCR products were sequenced directly on an Ion PGM 123 Banana bract mosaic virus from flowering ginger

System (Ion 314 chips) at the University of Hawaii’s using PhusionÒ High-Fidelity DNA Polymerase (NEB) in a Advanced Studies of Genomics, Proteomics, and Bioin- 20-ll reaction containing 2 ll of dA-tailed cDNA, 10 ll formatics Laboratory, Short-length (\65 nt) and low- PhusionÒ mix, 10 pmol oligo-dT primer [31], with incu- quality reads and primer sequences were trimmed, as were bation at 98 °C for 3 minutes, 45 °C for 10 minutes, and low-quality basecalls at the ends of reads. The remaining 68 °C for 5 minutes. Double-stranded cDNA was used for reads were assembled de novo using GeneiousÒPro 7.1.5 PCR with PhusionÒ High-Fidelity DNA Polymerase in a (Biomatters Ltd., Auckland, New Zealand), and the tar- 20-ll reaction mixture containing 2 ll of double-stranded geted contigs were aligned with BBrMV reference cDNA, 10 ll PhusionÒ mix, 10 pmol of primer 667 [31], sequences. Low-coverage regions, or regions where unex- and 10 pmol of the virus-specific primers BBrMV-304Rev pected stop codons or frameshifts occurred were confirmed (for BBrMV-384Rev synthesized cDNA) or BBrMV- by Sanger sequencing by amplifying the flanking regions in 384Rev (for BBrMV-1076Rev synthesized cDNA). This question. mixture was then incubated at 98 °C for 3 minutes, fol- To obtain 3’-terminal sequences, an oligo-dT primer [31] lowed by 35 cycles of amplification at 98 °C for 30 sec- was used to initiate cDNA synthesis. This primer was also onds, 55 °C for 30 seconds, 68 °C for 40 seconds, and a used in subsequent PCR reactions along with the virus- final extension at 68 °C for 10 minutes. PolyA tails were specific primer BBrMV-9473, which was designed to added by PCR using Taq DNA polymerase and dATP, and anneal near the 3’ terminus of the available BBrMV amplicons were ligated into pGEMÒ-T Easy Vector and sequence. To obtain the complete 5’ terminus of the gen- then sequenced by the Sanger method. The details of the ome, fresh RNAs were used to synthesize cDNAs with primers used are shown in Table 1. AMV reverse transcriptase (RNaseH?, Promega) and The complete genome sequence of BBrMV obtained specific primers (BBrMV-384Rev or BBrMV-1076Rev), from A. purpurata and the other two complete genome which were designed to anneal near the 5’ termini of known sequences of BBrMV from banana were used to predict BBrMV ORFs. cDNA synthesis and purification were per- random repeats, using Tandem Repeats Finder [32]. Their formed as described above. The purified cDNA was dA- predicted polyprotein sequences were then used to analyze tailed using terminal deoxynucleotidyl transferase (Pro- transmembrane domains with the TMpred (Prediction of mega, USA) according to the provided protocol. The dA- Transmembrane Regions and Orientation) [33] program. In tailed cDNA was used for second-strand synthesis of cDNA another case, 5’- and 3’-UTR sequences of these three

Table 1 Primers used for genome sequencing and virus detection Primer Sequence Use References

Universal-dN6 5’-GCCGGAGCTCTGCAGAATTCNNNNNN-3’ rPCR [30] Universal 5’-GCCGGAGCTCTGCAGAATTC-3’ rPCR [30] Oligo d(T) 5’-CACTCCCTATTATCCAGG(T)16-3’ First strain cDNA synthesis for [31] 3-temini; Second strain cDNA synthesis for 5-termini BBrMV-9473For 5’-ATATGCACTCTCTGCTTGGGG-3’ 3-temini specific forward primer This study BBrMV-384Rev 5’-TCCTCTCATCCGCAAACCAC-3’ 5’Race-specific primer This study BBrMV-1076Rev 5’-TCTCTTGCGTCAACAAGCCT-3’ 5’Race-specific primer This study Primer 667 5’-CACTCCCTATTATCCAGG-3’ 5’Race general primer [31] BBrMV-304Rev 5’-CATCATCTGGTGTCGCCACT-3’ 5’Race-specific primer This study BBrMV-CPF 5’-TCTGGAACGGAGTCAACC-3’ Reverse-transcription PCR This study BBrMV-CPR 5’-CCGTGACATTACGCATTG-3’ Reverse-transcription PCR This study BBrMV-CP-F3 5’-CTGGGAAAATGCGTCTCC-3 IC-RT-LAMPa This study BBrMV-CP-B3 5’-CCACCTAACAAGGTTGTGAA-3 IC-RT-LAMP This study BBrMV-CP-FIP 5’-TGACAAATCAAACTGATCAGGCTTA- IC-RT-LAMP This study AAGGTATCGAGGAAAAACTGC-3 BBrMV-CP-BIP 5’-AATGCTATCGCAACTAGGGAGC- IC-RT-LAMP This study CTGTTCTTCGTCCTCTATGG-3 BBrMV-CP-LF 5’-ATTGAAGCAAGAACTCGACGTT-3 IC-RT-LAMP This study BBrMV-CP-LB 5’-TGATGCATGGTGTGATGCTGTA-3 IC-RT-LAMP This study a IC-RT-LAMP is immunocapture reverse-transcription loop-mediated isothermal amplification

123 J. Zhang et al. complete genome sequences were used to find RNA fold- edible ginger (Zingiber officinale) plants for a 48-hour ing structures using Mfold [34]. inoculation access period (IAP), after which the aphids were killed with an insecticidal spray. The inoculated Phylogenetic analysis plants were periodically examined for symptoms for four months and were tested by RT-PCR for the presence of the The coat protein (CP) of BBrMV is involved in aphid virus three weeks and 16 weeks after IAP. transmission, cell-to-cell and systemic movement, encap- sidation of viral RNA, and viral RNA amplification [35]. Immunocapture RT-LAMP (IC-RT-LAMP) This protein interacts with its plant hosts and vectors and diagnostic assay may be under selection pressure with shifts in geographic distribution and plant hosts [36, 37]. Therefore, 63 CP gene Leaves of A. purpurata with typical symptoms of BBrMV sequences were retrieved from NCBI for analysis and infection (designated BBrMV) and other A. purpurata aligned using ClustalW [38]. These sequences were trun- plants with the different virus-like symptoms (designated cated into 406-nt pieces, since some of the CPs from SV1 and SV2, Fig. 5) were collected from the grounds of Southeast Asia and Ecuador were only partial sequences. the University of Hawaii at Manoa. One-hundred micro- The CP sequences were analyzed for evidence of recom- liters of BBrMV polyclonal antibodies (AGDIA, USA) was bination using RDP4 Beta 4.46 software [39]. Phylogenetic pipetted into each well of a PCR plate and incubated analysis was performed using MEGA 5.0 software [40], overnight in a humid container at 4 °C. The wells were using the maximum-likelihood algorithm with the units of then emptied and washed three times with 19 PBST. the number of nucleotide substitutions per site in the Leaves of A. purpurata (50 mg) were ground in 1 ml of Tamura-Nei model and bootstrapped with 1,000 replica- extraction buffer, and 100 ll of the liquid was added to tions. We used DnaSPÒ 5.10 [41] to calculate the number each pre-coated well and incubated for 2 hours at room of polymorphic sites, the amount of nucleotide sequence temperature. The incubated wells were then washed seven diversity (Pi), and Ka and Ks values, which were used for times with 19 PBST followed by two washes with DEPC- analyzing non-synonymous and synonymous mutations, treated sterile water. One-step RT-LAMP was performed in respectively, at the nucleotide level,, using the CP the wells as described above. RT-LAMP reactions (25 ll) sequence of the A. purpurata isolate as a reference. The contained 2.5 llof109 ThermoPolÒ Reaction Buffer 2? sequence-based statistical tests Ks*, Z, and Snn were used (New England Biolabs, USA, Mg free), 10 mM MgSO4, for genetic differentiation among different virus popula- 1.2 mM dNTP mix, 0.2 lM primer BBrMV-CP-F3/B3, tions [42, 43]. Fst (the interpopulational component of 1.6 lM primer BBrMV-CP-FIP/BIP, 0.8 lM primer genetic variation or the standardized variance in allele BBrMV-CP-LF/LB, 1 M betaine (Affymetrix, USA), 8 U frequencies across populations) measured for the level of Bacillus stearothermophilus DNA polymerase (New Eng- gene flow between populations was also calculated in land Biolabs), 450 nM EvaGreenÒ (Biotium, USA), 40 U Ò DnaSP 5.10 [41]. An absolute value of Fst \ 0.33 sug- M-MLV reverse transcriptase (Promega) and ddH2Otoa gests frequent gene flow [44]. final volume of 25 ll. RT-LAMP was run using an iQ5Ò Multicolor Real-Time PCR Detection System (Bio-Rad, Transmission of BBrMV USA). The reaction program was as follows: 65 °C for 60 minutes, with fluorescence readings taken each minute, Mechanical inoculations: Tissue extracts from BBrMV- and then inactivation for 2 minutes at 85 °C. A DAS- infected A. purpurata were used to mechanically inoculate ELISA kit (AGDIA) specific for BBrMV was used leaves of 15 herbaceous plants, including Cassia occiden- according to the manufacturer’s instructions to confirm the tals, Chenopodium amaranticolor, Brassica campestris, IC-RT-LAMP assay. Capsella rubella, Solanum lycopersicum, Cucumis sativus, Total RNA isolation (RNeasyÒ Plant Mini Kit, QIA- Vigna unguiculata, Phaseolus vulgaris, A. purpurata and GEN, USA) and RT-PCR (M-MLV First-Strand cDNA others. The inoculated plants were grown for 30 days in an Synthesis Kit, Promega) were performed according to the insect-free greenhouse at 20-25 °C with a 12-16 h pho- manufacturer’s instructions. toperiod and evaluated for local and systemic symptoms. Vector transmission: Transmission experiments were Detecting BBrMV in A. purpurata and Musa spp. conducted by placing non-viruliferous black banana in the field aphids, Pentalonia nigronervosa, on detached leaves of A. purpurata infected with BBrMV for a 48-hour acquisition Leaves of A. purpurata, including red- and pink-flowering access period (AAP). The leaves, each with 50 or more varieties, were sampled from different locations on the aphids, were then placed in cages with young banana or island of Oahu. The samples included those with symptoms 123 Banana bract mosaic virus from flowering ginger

Fig. 1 Organization of the complete BBrMV genome from A. indicate cleavage by NIa-Pro. The residues below indicate polymor- purpurata and a comparison of polymorphic sites with two other phic sites occurring in the low-compositional-complexity region complete BBrMV genomes. Pretty Interesting Potyviridae ORF within P1 or the predicted motifs within VPg. ‘‘BBrMV-Ginger’’ (PIPO) was expressed by a ?2 ribosomal frameshift from the P3 indicates the genome of BBrMV from A. purpurata. ‘‘BBrMV-PHI’’ ORF [51]. The number below the box representing each protein and ‘‘-TRY’’ indicate the genomes of BBrMV from bananas in the corresponds to the start of each protein and the end of the CP. Blue Philippines and India, respectively (color figure online) triangles indicate the dipeptides of cleavage sites and black triangles typical of BBrMV, leaves with other virus-like symptoms, a 132-nt 5’-UTR and a 209-nt 3’-UTR, excluding the and asymptomatic leaves. Any banana plants growing near polyA tail, that consisted of a single large open reading the flowering ginger were also sampled. The banana vari- frame (ORF) of 9,375 nt (Fig. 1). The 5’-UTR was A/T- eties sampled included ‘Dwarf Brazilian’ (AAB, Pome rich (59.85 %) and shared 96 % nucleotide sequence subgroup), ‘Mysore’ (AAB, Mysore), ‘Giant Plantain’ identity with the complete genome of BBrMV from the (AAB, Plantain), ‘Foconah’ (AAB, Pome), ‘Pome’ (AAB, Philippines (BBrMV-PHI) [7] and 97 % sequence identity Pome), ‘Kifutu’ (AAB, Pome) and ‘Goldfinger’ (AAAB, with the complete genome of BBrMV from India (BBrMV- hybrid). Samples were also collected from plants with TRY) [8]. BBrMV-Ginger shared a higher nucleotide typical symptoms of banana bunchy top virus (BBTV) or sequence similarity with BBrMV-PHI (99 %) identity than banana streak virus (BSV). Virus detection was done by the with BBrMV-TRY (93 %) identity in the 3‘-UTR. The 5’- IC-RT-LAMP method as described above. UTR of the BBrMV-Ginger isolate also contained two potybox-like blocks, ATCTCAaCAAGACATTCA and ACCTTACGCAACT (the ‘‘potybox-a and potybox-b’’-like Results sequences are underlined, respectively) [8]. A single G residue in BBrMV-PHI and BBrMV-TRY is replaced by an Genome organization of BBrMV from A. purpurata A residue (lowercase) in the sequence of BBrMV-Ginger. The ORF encodes a putative polyprotein of 3,124 amino A total of 289,674 high-quality reads were generated from acids with a molecular mass of 354.34 kDa and an iso- the cDNA library, with a median length of 217 bp. Of these electric point of 8.09. The amino acid sequence of BBrMV- reads, 44,586 were assembled de novo into a contig of Ginger was highly similar to those of BBrMV-PHI (98 %) 9,797 bp (average coverage of 1,114.89) and the other identity and BBrMV-TRY (97 %) identity. An initiation 19,528 reads assembled into a contig of 6,586 bp (average codon was located in the first in-frame CAAATGG, which coverage of 728.99). Among these reads, more were found is the same as in BBrMV-PHI and BBrMV-TRY. How- in the internal regions of the viral genome, including the ever, the termination codon AATAAATG of BBrMV- positions of the 6K1 protein, CI protein, 6K2 protein, VPg Ginger and BBrMV-PHI was modified to AATAGATA in protein and NIb protein. These two contigs were aligned to BBrMV-TRY. a BBrMV reference genome sequence (NC_009745.1) and The cleavage sites of the putative polyprotein, which low-coverage regions or areas where unexpected stop divide it into 10 proteins, were the dipeptides Y/S, G/G, codons or frameshifts occurred were confirmed by PCR, Q/S, Q/N, Q/N, E/N, E/G, Q/H and Q/S at the C-terminus and 5’-terminal sequences were obtained by RLM-RACE of each predicted protein (Fig. 1), the same as in BBrMV- (RNA ligase-mediated rapid amplification of cDNA ends). PHI and BBrMV-TRY. BBrMV-Ginger had more A complete 9,713-nucleotide (nt) genome sequence of sequence similarity to BBrMV-PHI than to BBrMV-TRY BBrMV from A. purpurata (designated BBrMV-Ginger (Table 2); only one amino acid deletion occurred in the under GenBank accession KT456531) was generated with central region of the P1 protein when compared to the other

123 J. Zhang et al.

Table 2 Amino acid sequence identities for each individual protein of three complete BBrMV genomes P1 HC-Pro P3 6K1 CI 6K2 VPg NIa NIb CP

BBrMV-Ginger - BBrMV-PHI 90 % 98 % 97 % 98 % 99 % 100 % 99 % 99 % 98 % 99 % BBrMV-Ginger - BBrMV-TRY 88 % 98 % 95 % 100 % 99 % 96 % 96 % 99 % 97 % 98 % BBrMV-PHI - BBrMV-TRY 88 % 98 % 95 % 98 % 98 % 96 % 97 % 98 % 97 % 98 %

two genomes. An overlapping coding sequence for pipo tests, Ks*, Z, and Snn, indicated a higher divergence (Pretty Interesting Potyviridae ORF) that was predicted between the Indian isolates and the other subpopulations to be initiated at a G1A7A motif was also found. This from the eastern Pacific or Southeast Asia (Table 3). FST ORF, situated within the P3 cistron (polyprotein posi- values among the eastern Pacific and Southeast Asia tion: 2,491-3,528), was predicted to extend from position isolates were less than 0.33, indicating frequent gene 2,949 to 3,194 and was translated as a PIPO of 82 aa in flow. However, gene flow among the Indian subpopula- the ?2 reading-frame shift relative to the polyprotein tion and the other two subpopulations was infrequent (Fig. 1). ([0.33). Many conserved characterized functional motifs were predicted to occur within the BBrMV-Ginger genome; Aphid transmission of BBrMV most of these motifs were highly conserved in BBrMV-PHI and BBrMV-TRY. The conserved motifs included H-X8-E- BBrMV could not be transmitted to any of the tested X30-GWSG in the P1 protein, GYCY-X71-H, RISC, PSA, herbaceous plants by mechanical inoculation. In aphid IGR, and CCC in the HC-Pro protein, conserved motifs I, transmission tests, BBrMV acquired by aphids from

II, V, and VI in the CI protein, H-X34-D-X67-GDCG-X14-H infected A. purpurata could be transmitted to A. purpurata in NIa, GDD in NIb, and DAG and FRQ-X41-FDF in the and banana by P. nigronervosa (Fig. 3). BBrMV infection CP protein [7, 35, 45]. An exception was that residue N was confirmed by RT-PCR in some of the inoculated from the motif NMYG of VPg [7, 46] in both BBrMV- bananas16 weeks after IAP, but not in any of the inoculated Ginger and BBrMV-PHI was replaced with an S residue in bananas three weeks after IAP (Fig. 3). BBrMV could not BBrMV-TRY. be transmitted to edible ginger, either mechanically or by aphids. Phylogenetic analysis Development of an IC-RT-LAMP diagnostic assay We used various recombination detection algorithms to assess 406-nt CP sequences from the 63 BBrMV isolates. We attempted to use an RT-LAMP assay to detect banana Recombination events were only detected by the Gene- bract mosaic virus sequences from total RNAs isolated conv and Siscan method in virus isolate TN21 and TN18, from A. purpurata. We were unable to detect the virus, respectively. Therefore, all 63 isolates were used for however, using primers designed for use with cardamom phylogenetic analysis with the A. purpurata isolate. The [47] or with the newly designed LAMP primers (Fig. 4a). India isolates from either banana or cardamom showed Therefore, an immunocapture procedure was integrated great divergence, while all of the isolates from Southeast into the RT-LAMP assay to enrich BBrMV virions. Using Asia, Ecuador, Samoa, and Hawaii were grouped closely this IC-RT-LAMP approach, four A. purpurata samples into the same close clade (Fig. 2). Compared to the with typical BBrMV symptoms, BBrMV-1 through sequence from A. purpurata, the isolate from Ecuador had BBrMV-4, produced exponential PCR amplifications, eight polymorphic sites and a Pi value of 0.020, whereas indicating positive reactions (Fig. 4b). The other suspected the isolate from Samoa had six polymorphic sites and a Pi virus-infected samples, the healthy control, and the ddH2O value of 0.015. Both of their Ks values were zero against blanks all produced negative reactions (RFU = 0). This the A. purpurata isolate. We therefore defined them as IC-RT-LAMP amplification process required only about an eastern Pacific subpopulations according to their geo- hour to complete. The results of the IC-RT-LAMP assay graphical distribution. The isolates from Southeast Asia, were confirmed by DAS-ELISA. Four BBrMV-infected including the Philippines, Thailand, and Vietnam, had samples were also positive (Fig. 5) in this assay, while the 5-11 polymorphic sites and Pi values of 0.012-0.027, healthy controls and the samples from plants with other compared to the A. purpurata isolate. Three statistical virus-like symptoms were all negative. These results

123 Banana bract mosaic virus from ﬂowering ginger

b Fig. 2 Phylogenetic analysis of the coat protein (CP) nucleotide sequences of BBrMV isolates from different locations in the world. The analysis used the maximum-likelihood algorithm with units of the number of nucleotide substitutions per site in the Tamura-Nei model, and bootstrapped with 1,000 replications. Isolates from eastern Paciﬁc are shown in red; isolates from Southeast Asia, blue; isolates from India, black (color ﬁgure online)

indicated that the IC-RT-LAMP diagnostic assay could reliably detect BBrMV from infected A. purpurata plants with no cross-reactivity to other viruses.

BBrMV in A. purpurata and Musa Spp. ﬁeld samples

We were only able to detect BBrMV from samples of A. purpurata with symptoms of bract mosaic. These samples all produced clear exponential increases in their RFU within 60 cycles, while samples from asymptomatic plants or other samples from A. purpurata with different virus- like symptoms were negative (Fig. 6). BBrMV could not be detected from any banana samples, including asymptomatic bananas and bananas with symptoms of either BBTV or BSV.

Discussion

In this study, we isolated dsRNA from A. purpurata with typical symptoms of BBrMV infection and used a deep- sequencing strategy to characterize the complete genome of the virus. A double-stranded DNA virus of the genus Badnavirus was also present in A. purpurata plants that were both symptomatic and asymptomatic for BBrMV, suggesting that the DNA virus was not responsible for the symptoms caused by BBrMV. A draft genome sequence of the A. purpurata isolate (BBrMV-Ginger) was obtained using deep sequencing, but the 5’ and 3’ termini remained to be sequenced. To conﬁrm the 5’-terminal sequence, we used RLM-RACE with dsRNA [48]. However, the longest sequences obtained using RLM-RACE were 19 nucleotides shorter than the sequences of BBrMV-PHI and BBrMV-TRY. We then used RLM-RACE with total RNA [49] to obtain a more complete 5’-terminal sequence. The longest sequences in this cDNA library had three more adenine residues at the 5’ terminus when compared to the reference genomes, indicating that the partial 5’ terminus of the viral RNA might have been truncated during replication into dsRNA. We therefore concluded that the BBrMV-Ginger sequence is 9,713-nt in length.

123 J. Zhang et al.

Table 3 Genetic differentiation and gene flow measurement for geographical BBrMV populations Eastern Pacifica -Southeast Asia Eastern Pacific - India Southeast Asia - India

Ks* 2.07889 2.41454*b 2.36033** Z 51.47658 612.03980** 699.62072** Snn 0.73333 0.91987 0.88798**

FST 0.04305 0.39798 0.38375 Nucleotide sequence diversity 0.01909 0.03484 0.03621 a The eastern Paciﬁc subpopulation includes the isolates from Hawaii, Ecuador, and Samoa b *0.05[P [ 0.01, ** P \ 0.01

Fig. 3 Aphid transmission of BBrMV from infected A. purpurata to B3 (3 weeks after IAP); lane 9, inoculated banana B3 (16 weeks after other hosts in insect cages. a, transmission of BBrMV from A. IAP); lane 10, inoculated banana B4 (sucker, 16 weeks after IAP); purpurata to banana. b, transmission of BBrMV from A. purpurata to lane 11, inoculated banana B5 (16 weeks after IAP); lane 12, water edible ginger. c, RT-PCR for BBrMV in inoculated banana and A. template; lane 13, A. purpurata (infected control); lanes 14 and 15, purpurata. Lanes 1 and 2, healthy banana; lane 3, inoculated banana healthy A. purpurata; lane 16, A. purpurata from the greenhouse; lane B1 (3 weeks after IAP); lane4, inoculated banana B1 upper leaf (16 17, inoculated A. purpurata (16 weeks after IAP); lane 18, ‘Polyne- weeks after IAP); lane 5, inoculated banana B1 lower leaf (16 weeks sian Princess’; M, marker Ladder II. The size of the positive amplicon after IAP); lane 6, inoculated banana B2 (3 weeks after IAP); lane 7, was 631 bp inoculated banana B2 (16 weeks after IAP); lane 8, inoculated banana

Within their predicted polyprotein sequences, the P1 occurring at the NMYG motif, whereas an N residue was proteins of these three virus isolates were the least con- replaced with an S residue in BBrMV-TRY. This change, served, and, like other potyviruses, most of the polymor- however, did not occur at the important tyrosine phic sites occurred at the N-terminus [8, 32]. The other (Y) residue, which has been experimentally conﬁrmed to proteins, however, were highly conserved, including the be the functional linking residue for the covalent attach- characterized functional motifs in the polyprotein [7, 35, ment of VPg to the viral RNA [7, 46]. It has been suggested 50]. All motifs were highly conserved between BBrMV- that this genetic change in the VPg domain might reﬂect a Ginger and BBrMV-PHI, with only a minor change host adaption or geographic isolation of these three

123 Banana bract mosaic virus from ﬂowering ginger

Fig. 4 Development of RT-LAMP and IC-RT-LAMP assays for indicating they were positive, while the other suspected virus-infected detection of BBrMV in A. purpurata. a, evaluation of RT-LAMP. The samples (SV1, SV2), healthy control and ddH2O template all showed tested samples did not show typical exponential PCR standard curves flat lines (RFU = 0), indicating negative amplification. Symptomatic of RFU; but after 60 min, an exponential increase in RFU was BBrMV samples are shown in red; suspected virus-infected samples detected by RT-LAMP for the healthy control (H); b, evaluation of SV1 and SV2, yellow and blue respectively, healthy control, green; IC-RT-LAMP. Samples BBrMV-1–BBrMV-4 with typical BBrMV and ddH2O template, black (color figure online) symptoms produced clear exponential PCR standard curves of RFU,

Fig. 5 DAS-ELISA detection results for samples from A. purpurata with different symptoms. a, leaves; b, results of DAS-ELISA for samples with typical BBrMV-symptoms and the healthy control (H); c, negative results of DAS-ELISA for SV1 and SV2, and a non- template control (NTC)

isolates, since the VPg domain is essential for viral repli- production, or even the virulence determinant [51, 52], is cation and host genotype speciﬁcity [7, 35, 50]. Another unclear. protein PIPO was predicted to occur within the P3 cistron We found only a few changes in these conserved motifs of both BBrMV-Ginger and BBrMV-PHI in the ?2 read- and therefore searched for short tandem repeats, trans- ing-frame shift; but the highly conserved G1-2A6-7 motif at membrane domains, and secondary structure elements in the 5’ end of pipo was replaced with AA7 in BBrMV-TRY. the UTRs to more fully characterize the different virus Whether this replacement affects the level of P3N?PIPO isolates. A comparison of short tandem repeat regions from

123 J. Zhang et al.

changes, even an amino acid deletion in one part of the polyprotein might also affect the functions of other parts of the polyprotein, and may even result in reduced virulence [37]. This could be a functional adaptation acting on the viral isolate when the host changed from banana to A. purpurea, because P1 is the functional protein involved in specific virus–host interactions in potyviruses [55, 56]. Similarly, insertions/deletions of amino acids also occurred within the P1 protein of the isolates from different hosts of watermelon mosaic virus [57] and sunflower chlorotic mottle virus [55, 58]. To determine if other genetic differences among these three viral genomes existed, the UTR sequences were Fig. 6 Indexing of BBrMV in A. purpurata and bananas from compared. Discrimination among the putative folding different locations on Oahu, Hawaii, using IC-RT-LAMP. Field patterns and structural elements in the UTR regions has samples of A. purpurata (red lines) included plants with symptoms of been shown to be informative when differentiating virus BBrMV and other possible symptoms of virus infection plus asymptomatic plants; eleven banana samples (blue lines) collected isolates [53], such as discriminating mosquito-borne fla- from different locations on the island of Oahu, two with typical viviruses from tick-borne flaviviruses using this method BBTV-symptoms, five with symptoms of BSV plus asymptomatic [59]. Two different putative RNA-folding structures within plants; and other banana cultivars and genotypes, including Mysore the 5’-UTR sequences of BBrMV-Ginger and BBrMV- (AAB, Mysore), Giant Plantain (AAB, Plantain), Foconah (AAB, Pome), Pome (AAB, Pome), Kifutu (AAB, Pome) and Goldfinger TRY were predicted to occur by Mfold [34], but only one (AAAB, hybrid) (color figure online) of these structures was predicted to occur in BBrMV-PHI. Conversely, three putative RNA-folding structures were predicted to occur within the 3’-UTR sequences of different viral genomes has been used to identify viral BBrMV-TRY, compared to only one similar structure in lineages from distinct geographic regions [53], but we were the other two isolates. One of these putative folding not able to detect any random repeats in the BBrMV- structures had very different structural elements than the Ginger isolate using Tandem Repeats Finder [32]. others (data not shown). Using the TMpred (Prediction of Transmembrane Phylogenetic analysis of the coat proteins of 64 BBrMV Regions and Orientation) [33] program, genome sequences isolates revealed a distinct relationship between different of the three viruses were predicted to contain nine strong geographic areas and their corresponding nucleotide transmembrane domains, eight of which were common to sequence diversities, but not among the isolates and their all three viral genomes. One transmembrane domain was host plants. Therefore, we suggest that the best way to found in both BBrMV-Ginger (residue position 813-834) characterize the relationship between BBrMV-Ginger and and BBrMV-PHI (residue position 814-835) isolates but the other isolates is geographical distance. The isolates can was not found in the BBrMV-TRY isolate because residue be divided into three geographical subpopulations. The A at position 815 was replaced with residue E in this iso- genetic differentiation among these subpopulations could late. Conversely, a unique transmembrane domain was be estimated using statistical tests such as Snn [43], which found in BBrMV-TRY (residue position 921-938) but not can be calculated for individual samples and their geo- in the other viral genomes, where residue N was replaced graphic location [42]. Using this approach, we found that with residue D. Coincidentally, this polymorphic site in the eastern Pacific subpopulations were more similar to potential transmembrane domain (residue position Southeast Asian subpopulations than they were to Indian 921-938) was located immediately before the potential subpopulations, suggesting frequent gene flow between the initiating motif of PIPO, while PIPO started at a site close eastern Pacific and the Southeast Asian populations. to the residues (position 940) of the polyprotein. This Additionally, a comparison of the complete genome transmembrane domain might be a key virulence deter- sequences suggested that BBrMV-Ginger was more closely minant in these virus isolates [51]. related to BBrMV-PHI than to BBrMV-TRY in phyloge- We also found an amino acid deletion between amino netic distance. acid positions 159 and 160 in the polyprotein of BBrMV- BBrMV isolates from India have the highest genetic Ginger that was not present in the other two genomes. This diversity, which suggests that the isolates from India are deletion is not located in a conserved functional motif, but the oldest of the three isolates characterized [43]. Their rather in a region of lower compositional complexity, as geographic origin can be inferred from the genetic diversity predicted by SMART [54] (see Fig. 1). However, any of this virus and its distribution [60]. A previous study 123 Banana bract mosaic virus from flowering ginger indicated that the Philippine isolate moved from India in a Conflict of interest Author Jingxin Zhang declares that he has no single event [61]. We hypothesize that the virus isolates conflict of interest. Author Wayne Borth declares that he has no conflict of interest. Author Birun Lin declares that he has no conflict of interest. from the eastern Pacific moved from Southeast Asia and Author Kishore Dey declares that he has no conflict of interest. Author that during the movement of this virus, a host shift from Michael Melzer declares that he has no conflict of interest. Author banana to A. purpurea might have occurred. Huifang Shen declares that she has no conflict of interest. Author Why BBrMV has not infected bananas since it moved Xiaoming Pu declares that he has no conflict of interest. Author Dayuan Sun declares that he has no conflict of interest. Author John Hu declares from Southeast Asia to Hawaii is unclear. The three aphid that he has no conflict of interest. vectors known to transmit BBrMV to banana, Pentalonia nigronervosa, Aphis gossypii, and Rhopalosiphum maidis, Ethical approval This article does not contain any studies with are all present in Hawaii. However, BBrMV acquired from human participants or animals performed by any of the authors. infected A. purpurata could be transmitted to banana by P. nigronervosa in our study, suggesting that BBrMV might References become established in Hawaii’s valuable banana farms. To protect the agricultural economy of Hawaii, therefore, 1. Magnaye LV, Espino RRC (1990) NOTE: banana bract mosaic, a robust detection methods are necessary [62]. new disease of banana, I. Symptomatology. Philipp Agric IC-LAMP or IC-RT-LAMP have been used for detecting 73(1):55–59 tomato spotted wilt virus [63], potato leafroll virus [64] and 2. Kumar PL, Selvarajan R, Iskra-Caruana ML, Chabannes M, Hanna R (2015) Biology, etiology, and control of virus diseases beet curly top virus [65], and they were thought to be superior of banana and plantain. Adv Virus Res 91:229–269 to ELISA and IC-RT-PCR when time, safety, cost and sim- 3. Quito-Avila DF, Ibarra MA, Alvarez RA, Ratti MF, Espinoza L, plicity were considered [64]. IC-LAMP and IC-RT-LAMP Cevallos-Cevallos JM, Peralta EL (2013) First report of Banana have the advantage of not requiring nucleic acid extraction or bract mosaic virus in ‘Cavendish’ banana in Ecuador. Plant Dis 97:1003 post-amplification treatment of the amplicons [65]. Like- 4. Wang IC, Sether DM, Melzer MJ, Borth WB, Hu JS (2010) First wise, all existing diagnostic assays for BBrMV are more report of banana bract mosaic virus in flowering ginger in time-consuming than the IC-RT-LAMP diagnostic assay we Hawaii. Plant Dis 94:921 have developed. In combination with the visual dye Eva- 5. Rodoni BC, Ahlawat YS, Varma A, Dale JL, Harding RM (1997) Identification and characterization of banana bract mosaic virus Green and use of a closed tube to prevent cross-contamina- in India. Plant Dis 81:669–672 tion, this assay is simple, rapid, specific and cost-effective. 6. Thomas JE, Geering ADW, Gambley CF, Kessling AF, White M This technique also avoids potential problems due to poor (1997) Purification, properties and diagnosis of banana bract RNA quality and subsequent false-negative results (Fig. 4). mosaic potyvirus and its distinction from Abaca mosaic potyvirus. Phytopathology 87:698–705 Even though the bananas we sampled from Hawaii were 7. Ha C, Coombs S, Revil PA, Harding RM, Vu M, Dale JL (2008) not infected with BBrMV, the possibility of future infec- Design and application of two novel degenerate primer pairs for tions remains. Ginger and banana both belong to the order the detection and complete genomic characterization of poty- Zingiberales, as do other common tropical plants such as viruses. Arch Virol 153:25–36 8. Balasubramanian V, Selvarajan R (2012) Complete genome heliconia, bird-of-paradise, and traveler’s palm. Therefore, sequence of a banana bract mosaic virus isolate infecting the these plants may also be at risk from BBrMV-Ginger French plantain cv. Nendran in India. Arch Virol 157:397–400 infections. The IC-RT-LAMP procedure developed in this 9. Wyant PS, Strohmeier S, Scha¨fer B, Krenz B, Assunçaõ IP, de study will be a useful tool in epidemiological studies for Andrade Lima GS, Jeske H (2012) Circular DNA genomics (circomics) exemplified for geminiviruses in bean crops and detecting the presence of BBrMV in bananas cultivated in weeds of northeastern Brazil. Virology 427:151–157 Hawaii. 10. Idris A, Al-Saleh M, Piatek MJ, Al-Shahwan I, Ali S, Brown JK (2014) Viral metagenomics: analysis of begomoviruses by illu- Acknowledgments This work was supported in part by grants from mina high-throughput sequencing. Viruses 12:1219–1236 the National Natural Science Foundation of China (31300118), the 11. Wylie SJ, Jones MG (2011) The complete genome sequence of a President Foundation of the Guangdong Academy of Agricultural Passion fruit woodiness virus isolate from Australia determined Sciences (201306), the National Institute of Food and Agriculture, US using deep sequencing, and its relationship to other potyviruses. Department of Agriculture, Hatch HAW09025-H (1001478), and the Arch Virol 156:479–482 U.S. Department of Agriculture, Agricultural Research Service (58- 12. Marais A, Faure C, Couture C, Bergey B, Gentit P, Candresse T 5320-4-012). (2014) Characterization by deep sequencing of divergent Plum bark necrosis stem pitting virus isolates and development of a Compliance with ethical standards broad-spectrum PBNSPaV detection assay. Phytopathology 104:660–666 Funding This study was funded by grants from the National Natural 13. Al Rwahnih M, Daubert S, Urbez-Torres JR, Cordero F, Rowhani Science Foundation of China (31300118), the President Foundation of A (2011) Deep sequencing evidence from single grapevine plants the Guangdong Academy of Agricultural Sciences (201306), the reveals a virome dominated by mycoviruses. Arch Virol National Institute of Food and Agriculture, US Department of Agri- 156:397–403 culture, Hatch HAW09025-H (1001478), and the US Department of 14. Elbeaino T, Giampetruzzi A, De Stradis A, Digiaro M (2014) Agriculture, Agricultural Research Service (58-5320-4-012). Deep-sequencing analysis of an apricot tree with vein clearing

123 J. Zhang et al.

symptoms reveals the presence of a novel betaflexivirus. Virus 35. Urcuqui-Inchima S, Haenni AL, Bernardi F (2001) Potyvirus Res 181:1–5 proteins: a wealth of functions. Virus Res 74:157–175 15. Pantaleo V, Saldarelli P, Miozzi L, Giampetruzzi A, Gisel A, 36. Moury B, Morel C, Johansen E, Jacquemond M (2002) Evidence Moxon S, Dalmay T, Bisztray G, Burgyan J (2010) Deep for diversifying selection in Potato virus Y and in the coat protein sequencing analysis of viral short RNAs from an infected Pinot of other potyviruses. J Gen Virol 83:2563–2573 Noir grapevine. Virology 408:49–56 37. Melgarejo TA, Alminaite A, Fribourg C, Spetz C, Valkonen JPT 16. Yang X, Wang Y, Guo W, Xie Y, Xie Q, Fan L, Zhou X (2011) (2004) Strains of Peru tomato virus infecting cocona (Solanum Characterization of small interfering RNAs derived from the sessiliflorum), tomato and pepper in Peru with reference to gen- Geminivirus/Betasatellite complex using deep sequencing. PLoS ome evolution in genus Potyvirus. Arch Virol 149:2025–2034 One 6:e16928 38. Higgins D, Thompson J, Gibson T, Thompson JD, Higgins DG, 17. Melzer MJ, Sugano JS, Uchida JY, Borth WB, Kawate MK, Hu Gibson TJ (1994) CLUSTAL W: improving the sensitivity of JS (2013) Molecular characterization of closteroviruses infecting progressive multiple sequence alignment through sequence Cordyline fruticosa L. in Hawaii. Front Microbiol. doi:10.3389/ weighting, position-specific gap penalties and weight matrix fmicb.2013.00039 choice. Nucleic Acids Res 22:4673–4680 18. Marais A, Faure C, Mustafayev E, Barone M, Alioto D, Can- 39. Martin DP, Murrell B, Golden M, Khoosal A, Muhire B (2015) dresse T (2015) Characterization by deep sequencing of Prunus RDP4: Detection and analysis of recombination patterns in virus virus T, a novel Tepovirus infecting Prunus species. Phy- genomes. Virus Evol. doi:10.1093/ve/vev003 topathology 105(1):135–140 40. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S 19. Illg RD, Faria RT (1995) Micropropagation of Alpinia purpurata (2011) MEGA5: molecular evolutionary genetics analysis using from inflorescence buds. Plant Cell Tissue Organ Cult maximum likelihood, evolutionary distance, and maximum par- 40(2):183–185 simony methods. Mol Biol Evol 28:2731–2739 20. Paret ML, de Silva AS, Criley RA, Alvarez AM (2008) Ralstonia 41. Librado P, Rozas J (2009) DnaSP v5: a software for compre- solanacearum race 4: risk assessment for edible ginger and hensive analysis of DNA polymorphism data. Bioinformatics floricultural ginger industries in Hawaii. Hort Technol 25:1451–1452 18(1):90–96 42. Hudson RR (2000) A new statistic for detecting genetic differ- 21. Selvarajan R, Jeyabaskaran KJ (2006) Effect of Banana bract entiation. Genetics 155:2011–2014 mosaic virus (BBrMV) on growth and yield of cultivar Nendran 43. Balasubramanian V, Selvarajan R (2014) Genetic diversity and (Plantain, AAB). Indian Phytopathol 59(4):496–500 recombination analysis in the coat protein gene of Banana bract 22. D’Hont A, Denoeud F, Aury J et al (2012) The banana (Musa mosaic virus. Virus Genes 48:509–517 acuminata) genome and the evolution of monocotyledonous 44. Wei TY, Yang JG, Liao FL, Gao FL, Lu LM, Zhang XT, Li F, plants. Nature 488:213–219 Wu ZJ, Lin QY, Xie LH, Lin HX (2009) Genetic diversity and 23. Evans E, Ballen F (2012) Banana Market. Food and Resource population structure of rice stripe virus in China. J Gen Virol Economics Department, UF/IFAS Extension. http://edis.ifas.ufl. 90:1025–1034 edu/fe901 45. Kadare G, Haenni AL (1997) Virus-encoded RNA helicases. 24. Iskra-Caruana M, Galzi S, Laboureau N (2008) A reliable IC J Virol 71:2583–2590 One-step RT-PCR method for the detection of BBrMV to ensure 46. Murphy JF, Klein PG, Hunt AG, Shaw JG (1996) Replacement of safe exchange of Musa germplasm. J Virol Methods 153:223–231 the tyrosine residue that links a potyviral VPg to the viral RNA is 25. Notomi T, Okayama H, Masubuchi H, Yonekawa T, Watanabe K, lethal. Virology 220:535–538 Amino N, Hase T (2000) Loop-mediated isothermal amplification 47. Siljo A, Bhat AI (2014) Reverse transcription loop-mediated of DNA. Nucleic Acids Res 28:e63 isothermal amplification assay for rapid and sensitive detection of 26. Tomita N, Mori Y, Kanda H, Notomi T (2008) Loop-mediated Banana bract mosaic virus in cardamom (Elettaria cardamo- isothermal amplification (LAMP) of gene sequences and simple mum). Eur J Plant Pathol 138:209–214 visual detection of products. Nat Protoc 3:877–882 48. Coutts RHA, Livieratos IC (2003) A rapid method for sequencing 27. Parida M, Posadas G, Inoue S, Hasebe F, Morita K (2004) Real- the 5’- and 3’-termini of double-stranded RNA viral templates time reverse transcription loop-mediated isothermal amplification using RLM-RACE. J Phytopathol 151:525–527 for rapid detection of West Nile virus. J Clin Microbiol 49. Frohman MA, Dush MK, Martin GR (1988) Rapid production of 42:257–263 full-length cDNAs from rare transcripts—amplification using a 28. Morris TJ, Dodds JA (1979) Isolation and analysis of double- single gene-specific oligonucleotide primer. Proc Natl Acad Sci stranded RNA from virus-infected plant and fungal tissue. Phy- USA 85:8998–9002 topathology 69:854–858 50. Adams MJ, Antoniw JF, Beaudoin F (2005) Overview and 29. Hu JS, Gonsalves A, Sether D, Ullman DE (1993) Detection of analysis of the polyprotein cleavage sites in the family Po- pineapple closterovirus, a possible cause of mealybug wilt of tyviridae. Mol Plant Pathol 6:471–487 pineapple. Acta Hortic (ISHS) 334:411–416 51. Chung BY, Miller WA, Atkins JF, Firth AE (2008) An over- 30. Froussard P (1992) A random-PCR method (rPCR) to construct lapping essential gene in the Potyviridae. Proc Natl Acad Sci whole cDNA library from low amounts of RNA. Nucleic Acids USA 105(15):5897–5902 Res 20:2900 52. Glasa M, Svoboda J, Nova´kova´ S (2007) Analysis of the 31. Melzer MJ, Borth WB, Sether DM, Ferreira S, Gonsalves D, Hu molecular and biological variability of zucchini yellow mosaic JS (2010) Genetic diversity and evidence for recent modular virus isolates from Slovakia and Czech Republic. Virus Genes recombination in Hawaiian Citrus tristeza virus. Virus Genes 35:415–421 40:111–118 53. Yan Q (2008) Bioinformatics databases and tools in virology 32. Benson G (1999) Tandem repeats finder: a program to analyze research: an overview. In Silico Biol 8:71–85 DNA sequences. Nucleic Acids Res 27:573–580 54. Letunic I, Doerks T, Bork P (2014) SMART: recent updates, new 33. Hofmann K, Stoffel W (1993) TMbase—a database of membrane developments and status in 2015. Nucleic Acids Res. doi:10. spanning proteins segments. Biol Chem Hoppe Seyler 374:166 1093/nar/gku949 34. Zuker M (2003) Mfold web server for nucleic acid folding and 55. Bejerman N, Giolitti F, de Breuil S, Lenardon S (2010) Molecular hybridization prediction. Nucleic Acids Res 31(13):3406–3415 characterization of Sunflower chlorotic mottle virus: a member of

123 Banana bract mosaic virus from ﬂowering ginger

a distinct species in the genus Potyvirus. Arch Virol mosaic potyvirus, development of diagnostic assays and detection 155:1331–1335 of the virus in banana plants from five countries in southeast Asia. 56. Feng X, Poplawsky AR, Nikolaeva OV, Myers JR, Karasev AV Arch Virol 144:725–1737 (2014) Recombinants of Bean common mosaic virus (BCMV) 62. Hu JS, Li HP, Barry K, Wang M (1995) Comparison of dot blot, and genetic determinants of BCMV involved in overcoming ELISA, and RT-PCR assays for detection of two cucumber resistance in common bean. Phytopathology 104:786–793 mosaic virus isolates infecting banana in Hawaii. Plant Dis 57. Desbiez C, Lecoq H (2008) Evidence for multiple intraspecific 79:902–906 recombinants in natural populations of Watermelon mosaic virus 63. Fukuta S, Ohishi K, Yoshida K, Mizukami Y, Ishida A, Kanbe M (WMV, Potyvirus). Arch Virol 153:1749–17540 (2004) Development of immunocapture reverse transcription 58. Bejerman N, Giolitti F, de Breuil S, Lenardon S (2013) loop-mediated isothermal amplification for the detection of Sequencing of two Sunflower chlorotic mottle virus isolates tomato spotted wilt virus from chrysanthemum. J Virol Methods obtained from different natural hosts shed light on its evolu- 121:49–55 tionary history. Virus Genes 46:105–110 64. Almasi MA, Moradi A, Nasiri J, Karami S, Nasiri M (2012) 59. Leyssen P, Charlier N, Lemey P, Billoir F, Vandamme AM, De Assessment of performance ability of three diagnostic methods Clercq E, de Lamballerie X, Neyts J (2002) Complete genome for detection of Potato Leafroll Virus (PLRV) using different sequence, taxonomic assignment, and comparative analysis of the visualizing systems. Appl Biochem Biotechnol. doi:10.1007/ untranslated regions of the Modoc virus, a flavivirus with no s12010-012-9818-1 known vector. Virology 293:125–140 65. Almasi MA, Hosseyni-Dehabadi SM, Aghapour-ojaghkandi M 60. Fargette D, Konate G, Fauquet C, Muller E, Peterschmitt M, (2014) Comparison and evaluation of three diagnostic methods Thresh JM (2006) Molecular ecology and emergence of tropical for detection of beet curly top virus in sugar beet using different plant viruses. Annu Rev Phytopathol 44:235–260 visualizing systems. Appl Biochem Biotechnol 173:1836–18480 61. Rodoni BC, Dale JL, Harding RM (1999) Characterization and expression of the coat protein-coding region of banana bract

123