- Email: [email protected]
Molecular characterization of two badnavirus genomes associated with Canna yellow mottle disease
Virus Research 243 (2018) 19–24
Contents lists available at ScienceDirect
Virus Research journal homepage: www.elsevier.com/locate/virusres
Short communication
Molecular characterization of two badnavirus genomes associated with Canna yellow mottle disease
MARK
Dulanjani Wijayasekaraa, Peter Hoytb, Austin Gimondoc, Bruce Dunnc, Aastha Thapaa, ⁎ Hannah Jonesa, Jeanmarie Verchota, a b c
Department of Entomology & Plant Pathology, 127 Noble Research Center Oklahoma State University, Stillwater, OK 74078, United States Department of Biochemistry and Molecular Biology, Oklahoma State University, Stillwater, OK 74078, United States Department of Horticulture & Landscape Architecture, Oklahoma State University, Stillwater, OK 74078, United States
A R T I C L E I N F O
A B S T R A C T
Keywords: Badnavirus Potyvirus Subviral DNAs Satellite DNA Plant virus Cannaceae Musa virology Badnavirus Bacilliform DNA containing plant viruses Ornamental gingers Banana streak virus Canna virus Caulimoviridae Tropical plant virus
Members of the genus Badnavirus have a single non-covalently closed circular double-stranded DNA genome of 7.2–9.2 kb. The genome encodes three open reading frames (ORFs) on the positive DNA strand. Canna yellow mottle virus (CaYMV) is a badnavirus that has been described as the etiological cause of yellow mottle disease in canna, although only a 565 bp fragment of the genome has been previously reported from cannas. In this report, concentrated virions were recovered from infected canna plants and nucleic acids were extracted. Two fulllength sequences represent two badnavirus genomes were recovered and were determined to be 6966 bp and 7385 bp in length. These DNAs represent a virus strain belonging to Canna yellow mottle virus and a novel species tentatively termed Canna yellow mottle associated virus. Phylogenetic analysis indicates that these two viruses are closely related to sugarcane bacilliform GD virus, pineapple bacilliform comosus virus, banana streak MY virus, and cycad leaf necrosis virus. We also showed naturally grown canna plants to be frequently co-infected by these two badnaviruses along with a potyvirus, Canna yellow streak virus.
Canna species are native to Central and South America and have been extensively hybridized for floriculture production (Cooke, 2001). Hybrid cannas are traded globally and are subject to serious virus diseases. Accumulation of viruses in vegetatively propagated rhizomes has hindered international trade of germplasm. The three most common viruses reported in canna plants are Bean yellow mosaic virus (BYMV), Canna yellow streak virus (CaYSV) and Canna yellow mottle virus (CaYMV) (Rajakaruna et al., 2013). While researchers have reported the complete genome sequences for several isolates of BYMV and CaYSV, only a 565 bp portion of the CaYMV genome has been described from infected canna plants. This fragment was used to develop diagnostic PCR primers known as CaYMV-3 and CaYMV-4 (Momol et al., 2004). These primers have since been widely employed for diagnostic detection of the virus associated with canna yellow mottle disease. Based on the 565 bp fragment of CaYMV, researchers reported this virus belongs to the Badnavirus genus. The genome of a typical member of the genus Badnavirus, consists of single circular double stranded DNA molecules ranging from 7.2 to 9.2 kb with three open reading frames
(ORF) (Borah et al., 2013; King et al., 2011). The ORF3 encodes a long polyprotein that contains the aspartic acid protease, reverse transcriptase and RNase H domains. Recently, Zhang et al. (2017) reported the complete genome sequence which they proposed to be a variant of CaYMV identified in Alpinia purpurata (Vieill.) K. Schum. in Hawaii. The complete genome of canna yellow mottle virus in A. purpurata (CaYMVAp) was 7, 120 bp (Zhang et al., 2017). To facilitate virus indexing of vegetatively propagated canna plants, we maintain more than 1000 plants per year, of several varieties, for the past seven years in a greenhouse. We developed a workflow for screening large sample populations to identify and segregate infected and healthy plants. We optimized a method for extracting total nucleic acids from leaves of 8-week old plants and developed a reliable twostep multiplex reverse transcription, polymerase chain reaction (RTPCR) to simultaneously detect BYMV, CaYSV, and CaYMV. This method employs primer concentrations to ensure the reliable and sensitive detection of as little as 0.1 pg of virus nucleic acids (Chauhan et al., 2015). The results are managed in a database that contains information
Abbreviations: CaYMV, Canna yellow mottle virus; CaYSV, Canna yellow streak virus; NGS, next generation sequencing; ORFs, open reading frames ⁎ Corresponding author. Present Permanent address: Texas A & M Agrilife Research, Dallas Center, 17360 Coit Road, Dallas, TX 75252, United States. E-mail address: [email protected] (J. Verchot). http://dx.doi.org/10.1016/j.virusres.2017.10.001 Received 7 August 2017; Received in revised form 29 September 2017; Accepted 2 October 2017 Available online 05 October 2017 0168-1702/ © 2017 Elsevier B.V. All rights reserved.
Virus Research 243 (2018) 19–24
D. Wijayasekara et al.
Badnavirus. There are two small ORFs encoding the P1 and P2 proteins. P1 has an unknown function and P2 has nucleic acid binding properties and is reported to be associated with virions. The larger ORF3 encodes a P3 polyprotein (Fig. 1A) that is suggested to contain the movement protein, coat protein, aspartic acid protease, and RT/RNAase H1 domains (King et al., 2011). The CaYMAV-1 ORF1 (nucleotide positions 564–1085) produces a protein of 173 amino acids. The CaYMAV-1 ORF2 (nucleotide position 1085–1486) encodes a protein of 133 amino acids. The translation stop codon for ORF1 and start codon for ORF2 reside in a common TAATGA element within the CaYMAV-1 genome, and a similar combination of ORF1 stop and ORF2 start codons is seen in the genome of banana streak virus from Musa acuminata Colla (BSAcYNV). The CaYMAV-1 ORF3 (nucleotide position 1491–6545) encodes a polyprotein of 1682 amino acids (Fig. 1B) (Zhuang et al., 2011). For CaYMV-Ap01, ORF1 (nucleotide positions 544–1060) produces a protein of 172 amino acids and ORF2 (nucleotide position 1059–1463) encodes a protein of 134 amino acids. The translation stop codon for ORF1 and start codon for ORF2 resides in a common TGATGA element (Fig. 1B). The CaYMV-Ap01 ORF3 coding region (nucleotide position 1463–6848) encodes a 1795 amino acid polyprotein. The numbering of CaYMAV-1 and CaYMV-Ap01 genomes starts with the putative tRNAmet-binding site located in the intergenic region which is a defining feature of badnavirus genomes (Fig. 1A). The CaYMAV-1 putative tRNAmet-binding site has the highly conserved 18 nucleotide consensus sequence TGGTATCAGAGCGAGGTT. A similar consensus occurs in the CaYMV-Ap01 intergenic region, with three mismatches TGGTATCAGAGCTGAGTT (the mismatches are underlined). The putative poly-adenylation signal (AATAAA) in the CaYMAV1 genome is located near nucleotide 6873, and in the CaYMV-Ap01 genome is near position 7291 (Fig. 1A) (Borah et al., 2013; King et al., 2011). To better confirm the existence of these two virus genomes in infected cannas, RNA was isolated from a subset of seven symptomatic plants. RT-PCR using primers that differentially detect both genomes. We recently reported that PCR detection of viral cDNAs is highly sensitive and can detect as little as 0.1 pg of viral RNA sequences (Chauhan et al., 2015). Since badnaviruses replicate through an RNA intermediate and produce transcripts as templates for translation, there is a greater abundance of viral RNAs than DNAs in infected cells (Periasamy et al., 2006). To test for both CaYMAV-1 and CaYMV-Ap01, we used three sets of PCR primers (Table S1). The primer set named CaYMV-3/4 detects a 565 bp fragment of both CaYMAV-1 and CaYMV-Ap01genomes and all seven samples tested positive for badnavirus infection using these primers (Fig. 1A and C) (Chauhan et al., 2015). The 0036F/R primer pair hybridizes to unique sequences in the CaYMV-Ap01 genome and produce a 1288 bp PCR product. Four samples produced PCR products with the 0036 F/R primer set indicating the presence of CaYMV-Ap01 (Fig. 4A and C and data not shown). The CaYMV-7-F/R primer pair hybridizes to unique sequences in the CaYMAV-1 genome and produces a 978 bp fragment of the CaYMAV-1 genome (Fig. 4A and C and data not shown). Interestingly, two samples tested positive for both CaYMAV-1 and CaYMV-Ap01 (Fig. 1C). These data indicate that both badnaviruses can simultaneously infect canna plants. Controls for PCR cycling included cDNA synthesized from healthy control host plants or plasmids containing 978-bp CaYMAV-1 or 1288-bp CaYMV-Ap01 genome fragments (Fig. 1C). A maximum likelihood (ML) tree was constructed using the fulllength genomes of 39 badnavirus species and included the genomes of CaYMAV-1 and CaYMV-Ap01 to learn about their relatedness to other badnavirus species (Fig. 2 and Table S2) (Hung and Weng, 2016). The GenBank accession numbers for the sequences are in Table S2. The ML tree contains three major clades. Clade 1 contains 19 species including CaYMV-Ap, CaYMV-Ap01 and CaYMAV-1. The genome sequences of CaYMV-Ap and CaYMV-Ap01 share 94% identity which is well above the 80% threshold for demarcation of species, established by the International Committee on Virus Taxonomy (ICTV) (King et al., 2011).
about plant varieties, individual plant symptoms monitored over the life time of the plant, and the diagnostic results for every individual. This workflow enabled us to determine which plants and varieties were best to use for this study. This research was undertaken to recover complete badnavirus genome sequences from greenhouse grown ornamental canna plants. We collected 80–100 g of infected leaves (cultivars ‘Red Futurity’ and ‘Striped Beauty’) that tested positive by RT-PCR (using the CaYMV-3 and -4 primer) for CaYMV for virus purification. We employed two published methods involving extraction and differential centrifugation for the recovery of bacilliform virions that was developed for isolation of members of the genus Caulimovirus (Covey et al., 1998; Zhuang et al., 2011). DNA was extracted from virions and resuspended in deionized water using conventional procedures. The pooled DNA preparations were sequenced using a Roche 454Junior™ Genomic Sequencer. There were 375,342 raw reads with median read lengths of 441 bp were cleaned using the FastQC software (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) on the iPlant/CyVerse website (https://user.iplantcollaborative.org/) to eliminate primer and adapter sequences, and to identify and restrict sequencing reads to those with phred scores ≥ 30. These were automatically assembled by the Roche Genome Assembly software (v. 2.8) and then reassembled using Newbler v2.7 resulting in 13,253 cleaned contigs. These sequences were used for similarity BLASTn searches in the NCBI GenBank databases using viruses (Taxid: 10239) as a limiting organismal name. While three contigs were identified with high similarity (e < 1 × 10−30) to CaYMV isolate V17 ORF3 gene (EF189148.1), and Sugarcane bacilliform virus isolate BataviaD, (FJ439817.1), only one of these, contig0028 (6966 bp) represented a full-length virus genome (Genbank Accession: NC030462.1). We ascribed a provisional name to this genome of canna yellow mottle associated virus 1 (CaYMAV-1). As a second approach to recovering badnavirus nucleic acids, RNA was extracted from infected (variety “Red Futurity”) leaves that tested positive by RT-PCR for CaYMV using the CaYMV-3/4 primer set. Total RNA was isolated using the Roche® SeqCap RNA system. Oligo(dT)12-18 primers were used for reverse transcription to prepare cDNA and then random hexamers were used to synthesize the second DNA strand. The dsDNA was ligated to sequencing adapters. Using the Roche 454Junior™ Genomic Sequencer, we obtained 163,870 reads with an average read length of 359 bp. Sequence reads were assembled into 10,745 contigs with an average length of 687 bp. These contigs were submitted to BLASTn suite and the resulting output was imported to MEGAN6 software (Huson et al., 2016) which automatically assigns output reads using a Lowest Common Ancestor algorithm to their respective taxons. Of the 10,745 contigs, 6249 were assigned to taxons and we identified a subset ascribed to virus taxons. Within this subset of data, we identified and assembled four contigs to obtain a second fulllength circular viral genome that was 7385 bp in length. The phylogenetic data presented in the following paragraphs suggest that this virus is a variant of the recently reported CaYMV-Ap. In this report, this genome was provisionally identified as variant 1 of the canna yellow mottle virus in Alpinia purpurata (CaYMV-Ap01) (Genbank Accession: MF074075) (Fig. 1A). To clone the full-length genomes designated as CaYMAV-1 and CaYMV-Ap01, overlapping PCR products were generated using DNA extracted from infected leaves and primers listed in Table S1. Individual PCR products were introduced into pGEM-T Easy vectors (Promega, USA). Several colonies of each transformant were sequenced (Fig. 1B) and the resulting transformant sequences were aligned using SDSC Biology Work Bench and MEGA 6.0 software (Hall, 2013; Sohpal et al., 2010; Subramaniam, 1998). The full-length of the genomes recovered by direct cloning where then aligned with the CaYMAV-1 and CaYMVAp01 genomes obtained by de novo sequencing (Edgar, 2004). The genomes of CaYMAV-1 and CaYMV-Ap01 each show an organization that is consistent with species belonging to the genus 20
Virus Research 243 (2018) 19–24
D. Wijayasekara et al.
Fig. 1. Properties of CaYMAV-1 and CaYMV-Ap01 genomes. (A) Linear representations of the CaYMAV1 and CaYMV-Ap01 genomes. The open boxes represent ORFs1, 2 and 3. Nucleotide position 1 is identified as the start of the tRNA-met binding site, as for other badnaviruses. The nucleotide positions for start and stop of each ORF is identified above the boxes in the linear diagram. The regulatory sequences representing the ORF1/2 Stop/Start site and the poly A tail coding sequence are presented below the boxes. The ORF3 encodes a polyprotein. The positions for the zinc finger binding domain (ZnF), the aspartic acid protease (Pro) domain, the reverse transcriptase (RT) and RNAse H domains are identified. (B) Representative ethidium bromide stained 1% gel showing PCR products amplifying sequential and overlapping segments of an entire CaYMAV-1 genome. These products were cloned into pGEM-T and then sequenced. (C) RNA was extracted from naturally infected canna leaves (lane 1) and healthy (lane 2) canna leaves. Representative 2% agarose gel electrophoresis presenting RT-PCR products of naturally infected canna plants grown from rhizomes in the greenhouse. Nucleic acids were extracted approximately 8 weeks after planting. Primers used in each test are identified on the right of each gel. Lane number identify leaf samples taken from different plants, but tested in each gel.
LxxANLEYLDLA(G/S)Q, which is located in the alignment between amino acid positions 20 and 33 (Fig. 3A). Second is a PKxIE sequence which we found lies between amino acid positions 139 and 143 in the alignment (Xu et al., 2011). Amino acid position 141, represented by ‘x’ has Iso in CaYMAV-1 and Phe in CaYMV-Ap and CaYMV-Ap01 ORF1 proteins. Overall, the combination of changes in ORF1 supports the hypothesis that CaYMAV-1 is a unique species, while CaYMV-Ap and CaYMV-Ap01 are more closely related variants. The ORF2 sequences are approximately 134 amino acids and begins with a conventional ATG start codon and contain characteristic Leu-rich sequences (Fig. 3B). CaYMV-Ap and -Ap01 share 94% similar or identical amino acids, whereas CaYMAV-1 ORF2 shows 75% similarity with CaYMV-Ap01 ORF2. The first sequence is LNSxLFLLVRDL, which is located between positions 45 and 55 in the alignment. The second sequence is LxDKLLRL, which is located between positions 61and 69. Two addition conserved motifs among Badnavirus ORF2 proteins are a KQLNxxI near amino acid position 40 and a C-terminal KDPY sequence. Among the three ORF2 proteins in the alignment, there is RQLNxxL motif which contains a K to R substitution (Fig. 3B). The CaYMV-Ap and −Ap01 ORF2 proteins have RNPY and the CaYMAV-01 ORF2 protein has KSPY near position 125 in the alignment (Fig. 2B) (Borah et al., 2009; Xu et al., 2011). The large ORF3 polyproteins of CaYMV-AP, CaYMV-AP01, and CaYMAV-1 are between 1693 and 1795 amino acids in length (Supplementary Fig. 1). The ORF3 polyprotein sequences of CaYMV-Ap and -Ap01 share 93% similar amino acid and 64% similar amino acids with CaYMAV-1. There are six motifs that define most badnavirus ORF3 polyproteins. First is the QIRDYR sequence which lies in the predicted movement protein region. For all three genomes in the alignment, this motif lies between amino acid position 27–32. There is a canonical zinc
Thus the CaYMV-Ap01 genome, which was isolated from cannas, appears to be a variant of the CaYMV-Ap genome which was isolated from A. purpurata (flowering ginger) (Zhang et al., 2017). On the other hand, the CaYMAV-1 and CaYMV-Ap01 genomes share approximately 65% identity suggesting that these are different but related badnavirus species (Fig. 2). The ML tree also identifies three related subgroups in Clade 1 which primarily consist of viruses that infect tropical plant hosts of the order Zingiberales [canna and banana (families Cannaceae and Musaceae, respectively)] or Poales [pineapple and sugarcane (families Bromeliaceae and Poaceae)] (Iskra-Caruana et al., 2014; Laney et al., 2012). The closest relative for the three canna virus isolates: Sugarcane bacilliform GD virus (SCBGDV), Pineapple bacilliform comosus virus (PBCoV), Banana streak MY virus (BSMYV), and Cycad leaf necrosis virus (CyLNV). To better understand the relatedness of these virus sequences, the translation products of CaYMV-Ap01 and CaYMAV were analyzed using ExPASy Translation tool (Artimo et al., 2012) and then aligned with CaYMV-Ap using CLUSTALW in SDSC Biology Workbench (Fig. 3 and Supplementary Fig. 1). Within the ORF1 alignment, CaYMV-Ap and CaYMV-Ap01 share 95% similar or identical residues. There were 9 amino acids (5%) that were not identical among CaYMV-Ap and CaYMV-Ap01, although three of these were similar charged residues (Fig. 1). CaYMAV and CaYMV-Ap01 shared 70% similar residues. There were 53 amino acid positions (30%) that were unique residue in comparison to CaYMV-Ap and CaYMV-Ap01. Of these 53 residues, thirteen were similar in charge to residues in the same position in either CaYMVAp or CaYMV-Ap01. The aligned ORF1 proteins contained two highly conserved motifs that are recognized across members of the genus Badnavirus (Borah et al., 2009) First is a Leu-rich region, 21
Virus Research 243 (2018) 19–24
D. Wijayasekara et al.
Fig. 2. ML trees including 38 badnavirus species. ML tree comparing complete nucleotide sequences of 30 badnavirus species, including the canna infecting viruses, listed in Table S2. The ML tree was generated using MEGA 6.0 software, selecting the Kimura model with 1000 bootstraps.
finger-like RNA binding domain is CxCx2Cx4Hx4C and, a second Cysrich sequence which lies further downstream. The alignment of CaYMVAp, -Ap01, and CaYMAV-1 polyprotein sequences shows a zinc fingerlike domain located between amino acid positions 768 and 783: CKCY (V/A)CGEEGHFAxDC. Interestingly, within this sequence, Ala772 and Arg781 are maintained in CaYMV-Ap and -Ap01 but are replaced by Val and Trp in CaYMAV-1. The second Cys-rich sequence lies between positions 873 and 917: Cx13CRACxLxARxGxRMxCx2Cx4CC LCSxYC (Supplementary Fig. 1) (Borah et al., 2009; Xu et al., 2011). The remaining motifs provide enzymatically activities for the aspartic protease, reverse transcriptase, and RNAase H domains. The aspartic protease has a canonical DTG catalytic motif located near amino acid
position 1097 and an I(I/L)G motif near position 1169 within the alignment (Supplementary Fig. 1). The YIDD catalytic site of the reverse transcriptase is located near position 1440 within the alignment, and is found within a highly conserved F(I/V)AVYIDDI(L/S)(V/I)FS sequence. The canonical RNAse H sequence is defined as Dx42Ex25Dx48D. Among badnaviruses the first D is located at position 1589, within a well conserved (V/I/M)xLETDGCMEGWGG(V/I)CKW sequence. The next E is located at position 1632 in the alignment, within a conserved KSTI(D/N)AE sequence. The D at position 1656 lies within a RTDCQA(I/L) sequence, which is also highly conserved among badnaviruses. Then the final D lies at position 1684 within a QNKPSRVRWLTFSD sequence that is also highly conserved among 22
Virus Research 243 (2018) 19–24
D. Wijayasekara et al.
Fig. 3. Amino acid sequence alignments of predicted (A) ORF1 and (B) ORF2 proteins encoded by CaYMVAp01, CaYMV-Ap and CaYMAV-1. The black shading points to identical amino acids whereas grey shading identifies the most common amino acids at the same position. The consensus sequence identifies in capital letters all residues that are conserved among all three proteins, and in lower case the residues that are conserved among three of the three proteins in the alignment.
healthy C. indica (a natural species) were included. Total nucleic acids were extracted from leaves after 4–8 weeks of growth. RT-PCR was carried out using CaYSV detecting primers, CaYMV-7F/R and 0036F/R primers (Table 1) (Chauhan et al., 2015). The data in Table 1 shows that the CaYMAV-1 and CaYMV-Ap01 most often occurred in variegated plants (‘Pretoria’ and ‘Striped Beauty’) alongside CaYSV. Considering only the two variegated plant varieties, 96% of the ‘Striped Beauty’ plants tested positive for CaYSV, CaYMAV-1, and CaYMV-Ap01, while 93% of the ‘Pretoria’ plants tested positive for CaYMAV-1 and CaYMV-Ap01. These plants show symptoms include yellowing and necrosis (Fig. 4). CaYMV-Ap01 transcripts were detected as sole infecting virus in four cultivars: ‘Australia’, ‘President’,
some badnaviruses including BSMYV, CyLNV and CoYMV (Supplementary Fig. 1) (Xu et al., 2011). We previously reported that many cannas were co-infected with CaYSV and CaYMV (Chauhan et al., 2015). Now that we have obtained evidence that the CaYMV-3 and -4 primers likely detect both CaYMVAp01 and CaYMAV-1, we undertook a new study to determine if CaYSV is seen alongside these two viruses in infected canna plants. The analysis reported in Table 1 used primers that separately detect CaYMAV-1 and CaYMV-Ap01, alongside CaYSV, to learn if these three genomes occur in hybrid varieties grown from rhizomes. Two hundred and twenty-seven hybrid cannas, belonging to eight cultivars, were grown from rhizomes in the greenhouse. As a negative control, five plants of
Table 1 Number (%) of naturally infected canna plants identified by RT-PCR diagnostic detection of plants grown from rhizomes or seed. RT-PCR Results
Australia
Burning Ember
President
Striped Beauty
Eureka
Picaso
Pretoria
Richard Wallace
Tangelo
C. indica
CaYMAV-1 CaYMV-Ap01 CaYSV only CaYMAV-1 + CaYMV-Ap01 CaYMAV - 1 + CaYSV CaYMV-Ap01 + CaYSV CaYMAV-1 + CaYMV-Ap01 + CaYSV Clean/Negative Total Samples
0 42 (78) 0 0 0 3 (5) 4 (7) 5 (9) 54
0 0 29 (94) 0 0 0 0 2 (6) 31
0 21 (72) 0 0 0 0 0 8 (28) 29
0 1 (4) 0 0 0 0 27 (96) 0 28
0 1 (5) 14 (64) 0 0 7 (32) 0 0 22
0 0 12 (86) 0 0 1 (7) 0 1 (7) 14
0 0 0 14 (93) 0 0 1 (7) 0 15
0 0 6 (35) 0 1 (6) 10 (59) 0 0 17
0 0 16 (94) 0 0 0 0 1 (6) 17
0 0 0 0 0 0 0 5 5
23
Virus Research 243 (2018) 19–24
D. Wijayasekara et al.
Fig. 4. Images of CaYMV-Ap01, CaYMAV-1, and CaYSV infected cannas. Examples of symptoms in two canna cultivars.
‘Striped Beauty’, and ‘Eureka’, (Table 1). In the burgundy leaved ‘Australia’, the presence of CaYMV-Ap01 transcripts is not associated any visible symptoms (Fig. 4). CaYSV was detected as the sole infecting virus in five cutlivars: ‘Burning Ember’, ‘Eureka’, ‘Picasso’, ‘Richard Wallace’, and ‘Tangelo’ (Table 1). CaYSV infected plants show mosaic disease. Prior to this study researchers predicted that there was a single badnavirus known as CaYMV, which infected canna plants. In this study, we investigate the etiology of canna yellow mottle disease by next generation sequencing using nucleic acids isolated from affected canna plants. Analysis of the generated data revealed the presence of sequences corresponding to two species of badnaviruses. These data were used to generate PCR primers for amplification of two full length genome sequences. Based on whole genome organization, and sequence similarity analysis, these two new genomes appear to belong to the genus Badnavirus and appear to be most closely related to CaYMV-Ap. The final outcomes reported in Table 1 indicate that both badnaviruses have a reasonable prevalence in commercial canna plants. The diagnostic analysis of more than 230 plants presented in Table 1, also indicated that several viruses may occur in mixed infections. Further studies are needed to understand the etiology of disease caused by single or mixed infection in canna plants.
Xenarios, I., Stockinger, H., 2012. ExPASy: SIB bioinformatics resource portal. Nucleic Acids Res. 40, W597–W603 Web Server issue. Borah, B.K., Johnson, A.M., Sai Gopal, D.V., Dasgupta, I., 2009. Sequencing and computational analysis of complete genome sequences of citrus yellow mosaic badna virus from acid lime and pummelo. Virus Genes 39 (1), 137–140. Borah, B.K., Sharma, S., Kant, R., Johnson, A.M., Saigopal, D.V., Dasgupta, I., 2013. Bacilliform DNA-containing plant viruses in the tropics: commonalities within a genetically diverse group. Mol. Plant Pathol. 14 (8), 759–771. Chauhan, R.P., Wijayasekara, D., Webb, M.A., Verchot, J., 2015. A reliable and rapid multiplex RT PCR assay for detection of two potyviruses and a pararetrovirus infecting canna plants. Plant Dis. 99, 1–9. Cooke, I., 2001. The Gardener's Guide to Growing Cannas. David & Charles; Timber Press, Devon, England] Portland, Or 160 p. Covey, S.N., Noad, R.J., al-Kaff, N.S., Turner, D.S., 1998. Caulimovirus isolation and DNA extraction. Methods Mol. Biol. 81, 53–63. Edgar, R.C., 2004. Muscle: a multiple sequence alignment method with reduced time and space complexity. BMC Bioinform. 5, 113. Hall, B.G., 2013. Building phylogenetic trees from molecular data with MEGA. Mol. Biol. Evol. 30 (5), 1229–1235. Hung, J.H., Weng, Z., 2016. Sequence alignment and homology search with BLAST and ClustalW. Cold Spring Harb. Protoc. http://dx.doi.org/10.1101/pdb.prot093088. Huson, D.H., Beier, S., Flade, I., Górska, A., El-Hadidi, M., Mitra, S., Ruscheweyh, H.J., Tappu, R., 2016. MEGAN community edition – interactive exploration and analysis of large-scale microbiome sequencing data. PLoS Comput. Biol. 12 (6), e1004957. Iskra-Caruana, M.L., Chabannes, M., Duroy, P.O., Muller, E., 2014. A possible scenario for the evolution of banana streak virus in banana. Virus Res. 186, 155–162. King, A., Lefkowitz, E., Adams, M., Carstens, E. (Eds.), 2011. Virus Taxonomy: Ninth Report of the International Committee on Taxonomy of Viruses, 9th ed. Elsevier Academic Press, London, UK. Laney, A.G., Hassan, M., Tzanetakis, I.E., 2012. An integrated badnavirus is prevalent in fig germplasm. Phytopathology 102 (12), 1182–1189. Momol, M.T., Lockhart, B.E.L., Dankers, H., Adkins, S., 2004. Canna yellow mottle virus detected in Canna in Florida. Plant Health Prog. 2–4 Online. Periasamy, M., Niazi, F.R., Malathi, V.G., 2006. Multiplex RT-PCR, a novel technique for the simultaneous detection of the DNA and RNA viruses causing rice tungro disease. J. Virol. Methods 134 (1–2), 230–236. Rajakaruna, P., Shafiekhani, M., Kim, T., Payton, M., Chauhan, R., Verchot, J., 2013. Production of discernible disease phenotypes in Canna by five plant viruses belonging to the genera Potyvirus, Cucumovirus and Badnavirus. Plant Pathol. 63 (4), 821–830. Sohpal, V.K., Dey, A., Singh, A., 2010. MEGA biocentric software for sequence and phylogenetic analysis: a review. Int. J. Bioinform. Res. Appl. 6 (3), 230–240. Subramaniam, S., 1998. The Biology Workbench–a seamless database and analysis environment for the biologist. Proteins 32 (1), 1–2. Xu, D., Mock, R., Kinard, G., Li, R., 2011. Molecular analysis of the complete genomic sequences of four isolates of gooseberry vein banding associated virus. Virus Genes 43 (1), 130–137. Zhang, J., Dey, K.K., Lin, B., Borth, W.B., Melzer, M.J., Sether, D., Wang, Y., Wang, I.C., Shen, H., Pu, X., Sun, D., Hu, J.S., 2017. Characterization of Canna yellow mottle virus in a New Host, Alpinia purpurata, in Hawaii. Phytopathology 107 (6), 791–799. Zhuang, J., Wang, J.H., Zhang, X., Liu, Z.X., 2011. Molecular characterization of banana streak virus isolate from Musa Acuminata in China. Virol. Sin. 26 (6), 393–402.
Acknowledgements Research was funded by Oklahoma Center for Advancement of Science and Technology Applied Research Program Phase II AR 132053-2; and by the Oklahoma Department of Agriculture Specialty Crops Research Grant Program. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.virusres.2017.10.001. References Artimo, P., Jonnalagedda, M., Arnold, K., Baratin, D., Csardi, G., de Castro, E., Duvaud, S., Flegel, V., Fortier, A., Gasteiger, E., Grosdidier, A., Hernandez, C., Ioannidis, V., Kuznetsov, D., Liechti, R., Moretti, S., Mostaguir, K., Redaschi, N., Rossier, G.,
24
Contents lists available at ScienceDirect
Virus Research journal homepage: www.elsevier.com/locate/virusres
Short communication
Molecular characterization of two badnavirus genomes associated with Canna yellow mottle disease
MARK
Dulanjani Wijayasekaraa, Peter Hoytb, Austin Gimondoc, Bruce Dunnc, Aastha Thapaa, ⁎ Hannah Jonesa, Jeanmarie Verchota, a b c
Department of Entomology & Plant Pathology, 127 Noble Research Center Oklahoma State University, Stillwater, OK 74078, United States Department of Biochemistry and Molecular Biology, Oklahoma State University, Stillwater, OK 74078, United States Department of Horticulture & Landscape Architecture, Oklahoma State University, Stillwater, OK 74078, United States
A R T I C L E I N F O
A B S T R A C T
Keywords: Badnavirus Potyvirus Subviral DNAs Satellite DNA Plant virus Cannaceae Musa virology Badnavirus Bacilliform DNA containing plant viruses Ornamental gingers Banana streak virus Canna virus Caulimoviridae Tropical plant virus
Members of the genus Badnavirus have a single non-covalently closed circular double-stranded DNA genome of 7.2–9.2 kb. The genome encodes three open reading frames (ORFs) on the positive DNA strand. Canna yellow mottle virus (CaYMV) is a badnavirus that has been described as the etiological cause of yellow mottle disease in canna, although only a 565 bp fragment of the genome has been previously reported from cannas. In this report, concentrated virions were recovered from infected canna plants and nucleic acids were extracted. Two fulllength sequences represent two badnavirus genomes were recovered and were determined to be 6966 bp and 7385 bp in length. These DNAs represent a virus strain belonging to Canna yellow mottle virus and a novel species tentatively termed Canna yellow mottle associated virus. Phylogenetic analysis indicates that these two viruses are closely related to sugarcane bacilliform GD virus, pineapple bacilliform comosus virus, banana streak MY virus, and cycad leaf necrosis virus. We also showed naturally grown canna plants to be frequently co-infected by these two badnaviruses along with a potyvirus, Canna yellow streak virus.
Canna species are native to Central and South America and have been extensively hybridized for floriculture production (Cooke, 2001). Hybrid cannas are traded globally and are subject to serious virus diseases. Accumulation of viruses in vegetatively propagated rhizomes has hindered international trade of germplasm. The three most common viruses reported in canna plants are Bean yellow mosaic virus (BYMV), Canna yellow streak virus (CaYSV) and Canna yellow mottle virus (CaYMV) (Rajakaruna et al., 2013). While researchers have reported the complete genome sequences for several isolates of BYMV and CaYSV, only a 565 bp portion of the CaYMV genome has been described from infected canna plants. This fragment was used to develop diagnostic PCR primers known as CaYMV-3 and CaYMV-4 (Momol et al., 2004). These primers have since been widely employed for diagnostic detection of the virus associated with canna yellow mottle disease. Based on the 565 bp fragment of CaYMV, researchers reported this virus belongs to the Badnavirus genus. The genome of a typical member of the genus Badnavirus, consists of single circular double stranded DNA molecules ranging from 7.2 to 9.2 kb with three open reading frames
(ORF) (Borah et al., 2013; King et al., 2011). The ORF3 encodes a long polyprotein that contains the aspartic acid protease, reverse transcriptase and RNase H domains. Recently, Zhang et al. (2017) reported the complete genome sequence which they proposed to be a variant of CaYMV identified in Alpinia purpurata (Vieill.) K. Schum. in Hawaii. The complete genome of canna yellow mottle virus in A. purpurata (CaYMVAp) was 7, 120 bp (Zhang et al., 2017). To facilitate virus indexing of vegetatively propagated canna plants, we maintain more than 1000 plants per year, of several varieties, for the past seven years in a greenhouse. We developed a workflow for screening large sample populations to identify and segregate infected and healthy plants. We optimized a method for extracting total nucleic acids from leaves of 8-week old plants and developed a reliable twostep multiplex reverse transcription, polymerase chain reaction (RTPCR) to simultaneously detect BYMV, CaYSV, and CaYMV. This method employs primer concentrations to ensure the reliable and sensitive detection of as little as 0.1 pg of virus nucleic acids (Chauhan et al., 2015). The results are managed in a database that contains information
Abbreviations: CaYMV, Canna yellow mottle virus; CaYSV, Canna yellow streak virus; NGS, next generation sequencing; ORFs, open reading frames ⁎ Corresponding author. Present Permanent address: Texas A & M Agrilife Research, Dallas Center, 17360 Coit Road, Dallas, TX 75252, United States. E-mail address: [email protected] (J. Verchot). http://dx.doi.org/10.1016/j.virusres.2017.10.001 Received 7 August 2017; Received in revised form 29 September 2017; Accepted 2 October 2017 Available online 05 October 2017 0168-1702/ © 2017 Elsevier B.V. All rights reserved.
Virus Research 243 (2018) 19–24
D. Wijayasekara et al.
Badnavirus. There are two small ORFs encoding the P1 and P2 proteins. P1 has an unknown function and P2 has nucleic acid binding properties and is reported to be associated with virions. The larger ORF3 encodes a P3 polyprotein (Fig. 1A) that is suggested to contain the movement protein, coat protein, aspartic acid protease, and RT/RNAase H1 domains (King et al., 2011). The CaYMAV-1 ORF1 (nucleotide positions 564–1085) produces a protein of 173 amino acids. The CaYMAV-1 ORF2 (nucleotide position 1085–1486) encodes a protein of 133 amino acids. The translation stop codon for ORF1 and start codon for ORF2 reside in a common TAATGA element within the CaYMAV-1 genome, and a similar combination of ORF1 stop and ORF2 start codons is seen in the genome of banana streak virus from Musa acuminata Colla (BSAcYNV). The CaYMAV-1 ORF3 (nucleotide position 1491–6545) encodes a polyprotein of 1682 amino acids (Fig. 1B) (Zhuang et al., 2011). For CaYMV-Ap01, ORF1 (nucleotide positions 544–1060) produces a protein of 172 amino acids and ORF2 (nucleotide position 1059–1463) encodes a protein of 134 amino acids. The translation stop codon for ORF1 and start codon for ORF2 resides in a common TGATGA element (Fig. 1B). The CaYMV-Ap01 ORF3 coding region (nucleotide position 1463–6848) encodes a 1795 amino acid polyprotein. The numbering of CaYMAV-1 and CaYMV-Ap01 genomes starts with the putative tRNAmet-binding site located in the intergenic region which is a defining feature of badnavirus genomes (Fig. 1A). The CaYMAV-1 putative tRNAmet-binding site has the highly conserved 18 nucleotide consensus sequence TGGTATCAGAGCGAGGTT. A similar consensus occurs in the CaYMV-Ap01 intergenic region, with three mismatches TGGTATCAGAGCTGAGTT (the mismatches are underlined). The putative poly-adenylation signal (AATAAA) in the CaYMAV1 genome is located near nucleotide 6873, and in the CaYMV-Ap01 genome is near position 7291 (Fig. 1A) (Borah et al., 2013; King et al., 2011). To better confirm the existence of these two virus genomes in infected cannas, RNA was isolated from a subset of seven symptomatic plants. RT-PCR using primers that differentially detect both genomes. We recently reported that PCR detection of viral cDNAs is highly sensitive and can detect as little as 0.1 pg of viral RNA sequences (Chauhan et al., 2015). Since badnaviruses replicate through an RNA intermediate and produce transcripts as templates for translation, there is a greater abundance of viral RNAs than DNAs in infected cells (Periasamy et al., 2006). To test for both CaYMAV-1 and CaYMV-Ap01, we used three sets of PCR primers (Table S1). The primer set named CaYMV-3/4 detects a 565 bp fragment of both CaYMAV-1 and CaYMV-Ap01genomes and all seven samples tested positive for badnavirus infection using these primers (Fig. 1A and C) (Chauhan et al., 2015). The 0036F/R primer pair hybridizes to unique sequences in the CaYMV-Ap01 genome and produce a 1288 bp PCR product. Four samples produced PCR products with the 0036 F/R primer set indicating the presence of CaYMV-Ap01 (Fig. 4A and C and data not shown). The CaYMV-7-F/R primer pair hybridizes to unique sequences in the CaYMAV-1 genome and produces a 978 bp fragment of the CaYMAV-1 genome (Fig. 4A and C and data not shown). Interestingly, two samples tested positive for both CaYMAV-1 and CaYMV-Ap01 (Fig. 1C). These data indicate that both badnaviruses can simultaneously infect canna plants. Controls for PCR cycling included cDNA synthesized from healthy control host plants or plasmids containing 978-bp CaYMAV-1 or 1288-bp CaYMV-Ap01 genome fragments (Fig. 1C). A maximum likelihood (ML) tree was constructed using the fulllength genomes of 39 badnavirus species and included the genomes of CaYMAV-1 and CaYMV-Ap01 to learn about their relatedness to other badnavirus species (Fig. 2 and Table S2) (Hung and Weng, 2016). The GenBank accession numbers for the sequences are in Table S2. The ML tree contains three major clades. Clade 1 contains 19 species including CaYMV-Ap, CaYMV-Ap01 and CaYMAV-1. The genome sequences of CaYMV-Ap and CaYMV-Ap01 share 94% identity which is well above the 80% threshold for demarcation of species, established by the International Committee on Virus Taxonomy (ICTV) (King et al., 2011).
about plant varieties, individual plant symptoms monitored over the life time of the plant, and the diagnostic results for every individual. This workflow enabled us to determine which plants and varieties were best to use for this study. This research was undertaken to recover complete badnavirus genome sequences from greenhouse grown ornamental canna plants. We collected 80–100 g of infected leaves (cultivars ‘Red Futurity’ and ‘Striped Beauty’) that tested positive by RT-PCR (using the CaYMV-3 and -4 primer) for CaYMV for virus purification. We employed two published methods involving extraction and differential centrifugation for the recovery of bacilliform virions that was developed for isolation of members of the genus Caulimovirus (Covey et al., 1998; Zhuang et al., 2011). DNA was extracted from virions and resuspended in deionized water using conventional procedures. The pooled DNA preparations were sequenced using a Roche 454Junior™ Genomic Sequencer. There were 375,342 raw reads with median read lengths of 441 bp were cleaned using the FastQC software (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) on the iPlant/CyVerse website (https://user.iplantcollaborative.org/) to eliminate primer and adapter sequences, and to identify and restrict sequencing reads to those with phred scores ≥ 30. These were automatically assembled by the Roche Genome Assembly software (v. 2.8) and then reassembled using Newbler v2.7 resulting in 13,253 cleaned contigs. These sequences were used for similarity BLASTn searches in the NCBI GenBank databases using viruses (Taxid: 10239) as a limiting organismal name. While three contigs were identified with high similarity (e < 1 × 10−30) to CaYMV isolate V17 ORF3 gene (EF189148.1), and Sugarcane bacilliform virus isolate BataviaD, (FJ439817.1), only one of these, contig0028 (6966 bp) represented a full-length virus genome (Genbank Accession: NC030462.1). We ascribed a provisional name to this genome of canna yellow mottle associated virus 1 (CaYMAV-1). As a second approach to recovering badnavirus nucleic acids, RNA was extracted from infected (variety “Red Futurity”) leaves that tested positive by RT-PCR for CaYMV using the CaYMV-3/4 primer set. Total RNA was isolated using the Roche® SeqCap RNA system. Oligo(dT)12-18 primers were used for reverse transcription to prepare cDNA and then random hexamers were used to synthesize the second DNA strand. The dsDNA was ligated to sequencing adapters. Using the Roche 454Junior™ Genomic Sequencer, we obtained 163,870 reads with an average read length of 359 bp. Sequence reads were assembled into 10,745 contigs with an average length of 687 bp. These contigs were submitted to BLASTn suite and the resulting output was imported to MEGAN6 software (Huson et al., 2016) which automatically assigns output reads using a Lowest Common Ancestor algorithm to their respective taxons. Of the 10,745 contigs, 6249 were assigned to taxons and we identified a subset ascribed to virus taxons. Within this subset of data, we identified and assembled four contigs to obtain a second fulllength circular viral genome that was 7385 bp in length. The phylogenetic data presented in the following paragraphs suggest that this virus is a variant of the recently reported CaYMV-Ap. In this report, this genome was provisionally identified as variant 1 of the canna yellow mottle virus in Alpinia purpurata (CaYMV-Ap01) (Genbank Accession: MF074075) (Fig. 1A). To clone the full-length genomes designated as CaYMAV-1 and CaYMV-Ap01, overlapping PCR products were generated using DNA extracted from infected leaves and primers listed in Table S1. Individual PCR products were introduced into pGEM-T Easy vectors (Promega, USA). Several colonies of each transformant were sequenced (Fig. 1B) and the resulting transformant sequences were aligned using SDSC Biology Work Bench and MEGA 6.0 software (Hall, 2013; Sohpal et al., 2010; Subramaniam, 1998). The full-length of the genomes recovered by direct cloning where then aligned with the CaYMAV-1 and CaYMVAp01 genomes obtained by de novo sequencing (Edgar, 2004). The genomes of CaYMAV-1 and CaYMV-Ap01 each show an organization that is consistent with species belonging to the genus 20
Virus Research 243 (2018) 19–24
D. Wijayasekara et al.
Fig. 1. Properties of CaYMAV-1 and CaYMV-Ap01 genomes. (A) Linear representations of the CaYMAV1 and CaYMV-Ap01 genomes. The open boxes represent ORFs1, 2 and 3. Nucleotide position 1 is identified as the start of the tRNA-met binding site, as for other badnaviruses. The nucleotide positions for start and stop of each ORF is identified above the boxes in the linear diagram. The regulatory sequences representing the ORF1/2 Stop/Start site and the poly A tail coding sequence are presented below the boxes. The ORF3 encodes a polyprotein. The positions for the zinc finger binding domain (ZnF), the aspartic acid protease (Pro) domain, the reverse transcriptase (RT) and RNAse H domains are identified. (B) Representative ethidium bromide stained 1% gel showing PCR products amplifying sequential and overlapping segments of an entire CaYMAV-1 genome. These products were cloned into pGEM-T and then sequenced. (C) RNA was extracted from naturally infected canna leaves (lane 1) and healthy (lane 2) canna leaves. Representative 2% agarose gel electrophoresis presenting RT-PCR products of naturally infected canna plants grown from rhizomes in the greenhouse. Nucleic acids were extracted approximately 8 weeks after planting. Primers used in each test are identified on the right of each gel. Lane number identify leaf samples taken from different plants, but tested in each gel.
LxxANLEYLDLA(G/S)Q, which is located in the alignment between amino acid positions 20 and 33 (Fig. 3A). Second is a PKxIE sequence which we found lies between amino acid positions 139 and 143 in the alignment (Xu et al., 2011). Amino acid position 141, represented by ‘x’ has Iso in CaYMAV-1 and Phe in CaYMV-Ap and CaYMV-Ap01 ORF1 proteins. Overall, the combination of changes in ORF1 supports the hypothesis that CaYMAV-1 is a unique species, while CaYMV-Ap and CaYMV-Ap01 are more closely related variants. The ORF2 sequences are approximately 134 amino acids and begins with a conventional ATG start codon and contain characteristic Leu-rich sequences (Fig. 3B). CaYMV-Ap and -Ap01 share 94% similar or identical amino acids, whereas CaYMAV-1 ORF2 shows 75% similarity with CaYMV-Ap01 ORF2. The first sequence is LNSxLFLLVRDL, which is located between positions 45 and 55 in the alignment. The second sequence is LxDKLLRL, which is located between positions 61and 69. Two addition conserved motifs among Badnavirus ORF2 proteins are a KQLNxxI near amino acid position 40 and a C-terminal KDPY sequence. Among the three ORF2 proteins in the alignment, there is RQLNxxL motif which contains a K to R substitution (Fig. 3B). The CaYMV-Ap and −Ap01 ORF2 proteins have RNPY and the CaYMAV-01 ORF2 protein has KSPY near position 125 in the alignment (Fig. 2B) (Borah et al., 2009; Xu et al., 2011). The large ORF3 polyproteins of CaYMV-AP, CaYMV-AP01, and CaYMAV-1 are between 1693 and 1795 amino acids in length (Supplementary Fig. 1). The ORF3 polyprotein sequences of CaYMV-Ap and -Ap01 share 93% similar amino acid and 64% similar amino acids with CaYMAV-1. There are six motifs that define most badnavirus ORF3 polyproteins. First is the QIRDYR sequence which lies in the predicted movement protein region. For all three genomes in the alignment, this motif lies between amino acid position 27–32. There is a canonical zinc
Thus the CaYMV-Ap01 genome, which was isolated from cannas, appears to be a variant of the CaYMV-Ap genome which was isolated from A. purpurata (flowering ginger) (Zhang et al., 2017). On the other hand, the CaYMAV-1 and CaYMV-Ap01 genomes share approximately 65% identity suggesting that these are different but related badnavirus species (Fig. 2). The ML tree also identifies three related subgroups in Clade 1 which primarily consist of viruses that infect tropical plant hosts of the order Zingiberales [canna and banana (families Cannaceae and Musaceae, respectively)] or Poales [pineapple and sugarcane (families Bromeliaceae and Poaceae)] (Iskra-Caruana et al., 2014; Laney et al., 2012). The closest relative for the three canna virus isolates: Sugarcane bacilliform GD virus (SCBGDV), Pineapple bacilliform comosus virus (PBCoV), Banana streak MY virus (BSMYV), and Cycad leaf necrosis virus (CyLNV). To better understand the relatedness of these virus sequences, the translation products of CaYMV-Ap01 and CaYMAV were analyzed using ExPASy Translation tool (Artimo et al., 2012) and then aligned with CaYMV-Ap using CLUSTALW in SDSC Biology Workbench (Fig. 3 and Supplementary Fig. 1). Within the ORF1 alignment, CaYMV-Ap and CaYMV-Ap01 share 95% similar or identical residues. There were 9 amino acids (5%) that were not identical among CaYMV-Ap and CaYMV-Ap01, although three of these were similar charged residues (Fig. 1). CaYMAV and CaYMV-Ap01 shared 70% similar residues. There were 53 amino acid positions (30%) that were unique residue in comparison to CaYMV-Ap and CaYMV-Ap01. Of these 53 residues, thirteen were similar in charge to residues in the same position in either CaYMVAp or CaYMV-Ap01. The aligned ORF1 proteins contained two highly conserved motifs that are recognized across members of the genus Badnavirus (Borah et al., 2009) First is a Leu-rich region, 21
Virus Research 243 (2018) 19–24
D. Wijayasekara et al.
Fig. 2. ML trees including 38 badnavirus species. ML tree comparing complete nucleotide sequences of 30 badnavirus species, including the canna infecting viruses, listed in Table S2. The ML tree was generated using MEGA 6.0 software, selecting the Kimura model with 1000 bootstraps.
finger-like RNA binding domain is CxCx2Cx4Hx4C and, a second Cysrich sequence which lies further downstream. The alignment of CaYMVAp, -Ap01, and CaYMAV-1 polyprotein sequences shows a zinc fingerlike domain located between amino acid positions 768 and 783: CKCY (V/A)CGEEGHFAxDC. Interestingly, within this sequence, Ala772 and Arg781 are maintained in CaYMV-Ap and -Ap01 but are replaced by Val and Trp in CaYMAV-1. The second Cys-rich sequence lies between positions 873 and 917: Cx13CRACxLxARxGxRMxCx2Cx4CC LCSxYC (Supplementary Fig. 1) (Borah et al., 2009; Xu et al., 2011). The remaining motifs provide enzymatically activities for the aspartic protease, reverse transcriptase, and RNAase H domains. The aspartic protease has a canonical DTG catalytic motif located near amino acid
position 1097 and an I(I/L)G motif near position 1169 within the alignment (Supplementary Fig. 1). The YIDD catalytic site of the reverse transcriptase is located near position 1440 within the alignment, and is found within a highly conserved F(I/V)AVYIDDI(L/S)(V/I)FS sequence. The canonical RNAse H sequence is defined as Dx42Ex25Dx48D. Among badnaviruses the first D is located at position 1589, within a well conserved (V/I/M)xLETDGCMEGWGG(V/I)CKW sequence. The next E is located at position 1632 in the alignment, within a conserved KSTI(D/N)AE sequence. The D at position 1656 lies within a RTDCQA(I/L) sequence, which is also highly conserved among badnaviruses. Then the final D lies at position 1684 within a QNKPSRVRWLTFSD sequence that is also highly conserved among 22
Virus Research 243 (2018) 19–24
D. Wijayasekara et al.
Fig. 3. Amino acid sequence alignments of predicted (A) ORF1 and (B) ORF2 proteins encoded by CaYMVAp01, CaYMV-Ap and CaYMAV-1. The black shading points to identical amino acids whereas grey shading identifies the most common amino acids at the same position. The consensus sequence identifies in capital letters all residues that are conserved among all three proteins, and in lower case the residues that are conserved among three of the three proteins in the alignment.
healthy C. indica (a natural species) were included. Total nucleic acids were extracted from leaves after 4–8 weeks of growth. RT-PCR was carried out using CaYSV detecting primers, CaYMV-7F/R and 0036F/R primers (Table 1) (Chauhan et al., 2015). The data in Table 1 shows that the CaYMAV-1 and CaYMV-Ap01 most often occurred in variegated plants (‘Pretoria’ and ‘Striped Beauty’) alongside CaYSV. Considering only the two variegated plant varieties, 96% of the ‘Striped Beauty’ plants tested positive for CaYSV, CaYMAV-1, and CaYMV-Ap01, while 93% of the ‘Pretoria’ plants tested positive for CaYMAV-1 and CaYMV-Ap01. These plants show symptoms include yellowing and necrosis (Fig. 4). CaYMV-Ap01 transcripts were detected as sole infecting virus in four cultivars: ‘Australia’, ‘President’,
some badnaviruses including BSMYV, CyLNV and CoYMV (Supplementary Fig. 1) (Xu et al., 2011). We previously reported that many cannas were co-infected with CaYSV and CaYMV (Chauhan et al., 2015). Now that we have obtained evidence that the CaYMV-3 and -4 primers likely detect both CaYMVAp01 and CaYMAV-1, we undertook a new study to determine if CaYSV is seen alongside these two viruses in infected canna plants. The analysis reported in Table 1 used primers that separately detect CaYMAV-1 and CaYMV-Ap01, alongside CaYSV, to learn if these three genomes occur in hybrid varieties grown from rhizomes. Two hundred and twenty-seven hybrid cannas, belonging to eight cultivars, were grown from rhizomes in the greenhouse. As a negative control, five plants of
Table 1 Number (%) of naturally infected canna plants identified by RT-PCR diagnostic detection of plants grown from rhizomes or seed. RT-PCR Results
Australia
Burning Ember
President
Striped Beauty
Eureka
Picaso
Pretoria
Richard Wallace
Tangelo
C. indica
CaYMAV-1 CaYMV-Ap01 CaYSV only CaYMAV-1 + CaYMV-Ap01 CaYMAV - 1 + CaYSV CaYMV-Ap01 + CaYSV CaYMAV-1 + CaYMV-Ap01 + CaYSV Clean/Negative Total Samples
0 42 (78) 0 0 0 3 (5) 4 (7) 5 (9) 54
0 0 29 (94) 0 0 0 0 2 (6) 31
0 21 (72) 0 0 0 0 0 8 (28) 29
0 1 (4) 0 0 0 0 27 (96) 0 28
0 1 (5) 14 (64) 0 0 7 (32) 0 0 22
0 0 12 (86) 0 0 1 (7) 0 1 (7) 14
0 0 0 14 (93) 0 0 1 (7) 0 15
0 0 6 (35) 0 1 (6) 10 (59) 0 0 17
0 0 16 (94) 0 0 0 0 1 (6) 17
0 0 0 0 0 0 0 5 5
23
Virus Research 243 (2018) 19–24
D. Wijayasekara et al.
Fig. 4. Images of CaYMV-Ap01, CaYMAV-1, and CaYSV infected cannas. Examples of symptoms in two canna cultivars.
‘Striped Beauty’, and ‘Eureka’, (Table 1). In the burgundy leaved ‘Australia’, the presence of CaYMV-Ap01 transcripts is not associated any visible symptoms (Fig. 4). CaYSV was detected as the sole infecting virus in five cutlivars: ‘Burning Ember’, ‘Eureka’, ‘Picasso’, ‘Richard Wallace’, and ‘Tangelo’ (Table 1). CaYSV infected plants show mosaic disease. Prior to this study researchers predicted that there was a single badnavirus known as CaYMV, which infected canna plants. In this study, we investigate the etiology of canna yellow mottle disease by next generation sequencing using nucleic acids isolated from affected canna plants. Analysis of the generated data revealed the presence of sequences corresponding to two species of badnaviruses. These data were used to generate PCR primers for amplification of two full length genome sequences. Based on whole genome organization, and sequence similarity analysis, these two new genomes appear to belong to the genus Badnavirus and appear to be most closely related to CaYMV-Ap. The final outcomes reported in Table 1 indicate that both badnaviruses have a reasonable prevalence in commercial canna plants. The diagnostic analysis of more than 230 plants presented in Table 1, also indicated that several viruses may occur in mixed infections. Further studies are needed to understand the etiology of disease caused by single or mixed infection in canna plants.
Xenarios, I., Stockinger, H., 2012. ExPASy: SIB bioinformatics resource portal. Nucleic Acids Res. 40, W597–W603 Web Server issue. Borah, B.K., Johnson, A.M., Sai Gopal, D.V., Dasgupta, I., 2009. Sequencing and computational analysis of complete genome sequences of citrus yellow mosaic badna virus from acid lime and pummelo. Virus Genes 39 (1), 137–140. Borah, B.K., Sharma, S., Kant, R., Johnson, A.M., Saigopal, D.V., Dasgupta, I., 2013. Bacilliform DNA-containing plant viruses in the tropics: commonalities within a genetically diverse group. Mol. Plant Pathol. 14 (8), 759–771. Chauhan, R.P., Wijayasekara, D., Webb, M.A., Verchot, J., 2015. A reliable and rapid multiplex RT PCR assay for detection of two potyviruses and a pararetrovirus infecting canna plants. Plant Dis. 99, 1–9. Cooke, I., 2001. The Gardener's Guide to Growing Cannas. David & Charles; Timber Press, Devon, England] Portland, Or 160 p. Covey, S.N., Noad, R.J., al-Kaff, N.S., Turner, D.S., 1998. Caulimovirus isolation and DNA extraction. Methods Mol. Biol. 81, 53–63. Edgar, R.C., 2004. Muscle: a multiple sequence alignment method with reduced time and space complexity. BMC Bioinform. 5, 113. Hall, B.G., 2013. Building phylogenetic trees from molecular data with MEGA. Mol. Biol. Evol. 30 (5), 1229–1235. Hung, J.H., Weng, Z., 2016. Sequence alignment and homology search with BLAST and ClustalW. Cold Spring Harb. Protoc. http://dx.doi.org/10.1101/pdb.prot093088. Huson, D.H., Beier, S., Flade, I., Górska, A., El-Hadidi, M., Mitra, S., Ruscheweyh, H.J., Tappu, R., 2016. MEGAN community edition – interactive exploration and analysis of large-scale microbiome sequencing data. PLoS Comput. Biol. 12 (6), e1004957. Iskra-Caruana, M.L., Chabannes, M., Duroy, P.O., Muller, E., 2014. A possible scenario for the evolution of banana streak virus in banana. Virus Res. 186, 155–162. King, A., Lefkowitz, E., Adams, M., Carstens, E. (Eds.), 2011. Virus Taxonomy: Ninth Report of the International Committee on Taxonomy of Viruses, 9th ed. Elsevier Academic Press, London, UK. Laney, A.G., Hassan, M., Tzanetakis, I.E., 2012. An integrated badnavirus is prevalent in fig germplasm. Phytopathology 102 (12), 1182–1189. Momol, M.T., Lockhart, B.E.L., Dankers, H., Adkins, S., 2004. Canna yellow mottle virus detected in Canna in Florida. Plant Health Prog. 2–4 Online. Periasamy, M., Niazi, F.R., Malathi, V.G., 2006. Multiplex RT-PCR, a novel technique for the simultaneous detection of the DNA and RNA viruses causing rice tungro disease. J. Virol. Methods 134 (1–2), 230–236. Rajakaruna, P., Shafiekhani, M., Kim, T., Payton, M., Chauhan, R., Verchot, J., 2013. Production of discernible disease phenotypes in Canna by five plant viruses belonging to the genera Potyvirus, Cucumovirus and Badnavirus. Plant Pathol. 63 (4), 821–830. Sohpal, V.K., Dey, A., Singh, A., 2010. MEGA biocentric software for sequence and phylogenetic analysis: a review. Int. J. Bioinform. Res. Appl. 6 (3), 230–240. Subramaniam, S., 1998. The Biology Workbench–a seamless database and analysis environment for the biologist. Proteins 32 (1), 1–2. Xu, D., Mock, R., Kinard, G., Li, R., 2011. Molecular analysis of the complete genomic sequences of four isolates of gooseberry vein banding associated virus. Virus Genes 43 (1), 130–137. Zhang, J., Dey, K.K., Lin, B., Borth, W.B., Melzer, M.J., Sether, D., Wang, Y., Wang, I.C., Shen, H., Pu, X., Sun, D., Hu, J.S., 2017. Characterization of Canna yellow mottle virus in a New Host, Alpinia purpurata, in Hawaii. Phytopathology 107 (6), 791–799. Zhuang, J., Wang, J.H., Zhang, X., Liu, Z.X., 2011. Molecular characterization of banana streak virus isolate from Musa Acuminata in China. Virol. Sin. 26 (6), 393–402.
Acknowledgements Research was funded by Oklahoma Center for Advancement of Science and Technology Applied Research Program Phase II AR 132053-2; and by the Oklahoma Department of Agriculture Specialty Crops Research Grant Program. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.virusres.2017.10.001. References Artimo, P., Jonnalagedda, M., Arnold, K., Baratin, D., Csardi, G., de Castro, E., Duvaud, S., Flegel, V., Fortier, A., Gasteiger, E., Grosdidier, A., Hernandez, C., Ioannidis, V., Kuznetsov, D., Liechti, R., Moretti, S., Mostaguir, K., Redaschi, N., Rossier, G.,
24