Complete Genome Sequence Analysis of Guangdong Isolates of Cymbidium Mosaic Virus

Yonglu WEI, Rui REN, Jie GAO, Weiping LIU, Qi XIE, Genfa ZHU, Fengxi YANG*

1. Environmental Horticulture Research Institute, Guangdong Academy of Agricultural Sciences/Guangdong Key Laboratory of Ornamental Plant Germplasm Innovation and Utilization, Guangzhou 510640, China; 2. Guangdong Wengshan Orchid Research Co., Ltd, Wengyuan 512600, China

Abstract [Objectives] The paper was to explore the genetic information and evolution of CymMV, and to provide an important scientific basis for monitoring and early warning of orchid virus disease and anti-virus genetic engineering of orchid in Guangdong Province. [Methods] RT-PCR and DASELISA were used to detect and identify CymMV from leaves with suspected virus disease of Cymbidium sinense collected from Guangzhou area. The genome sequence assembly, annotation, phylogeny and selection pressure analysis of CymMV isolates were performed with related molecular biology software. [Results] Two CymMV isolates (GZV013 and ZC29) were found in Guangdong Province for the first time in this study. The genome of both GZV013 and ZC29 were 6 227nt in length, encoding 5 functional proteins. The similarity analysis of the full sequence showed that the nucleotide sequence identity of GZV013 and Taiwan isolate M2 was 97.03% and that of ZC29 and Nanjing isolate NJ-1 was 97.11%. The complete genome sequence identity among CymMV isolates ranged from 86.85% to 98.31%, and the differentiation of diverse populations was closely related to host species and geographical isolation. Each region of CymMV genome was affected by negative selection and conformed to the neutral evolution model. The genes encoding RdRp, TGB1 and TGB2 had the highest mutation rates in the genome. [Conclusions] GZV013 was most closely related to Taiwan isolate M2, and ZC29 was most closely related to Nanjing isolate NJ-1, belonging to the same branch of a family.

Key words Orchid; Virus detection; Cymbidium mosaic virus (CymMV); Sequence analysis; Selection pressure

1 Introduction

Cymbidium mosaic virus (CymMV), belonging toPotaxvirus, is one of the most widespread and harmful viral pathogens of orchid. Orchid virus diseases caused by CymMV infection are widely prevalent in the Netherlands, the United States, South Korea, Japan, Southeast Asia, as well as Zhejiang, Guangdong, Hainan and other major orchid producing areas in China[1-11]. CymMV has a wide range of hosts and can infect many orchids inCymbidium,Cattleya,Dendrobium,PhalaenopsisandVanda[12-13]. CymMV infection of orchids often causes symptoms such as leaf chlorosis, dry spot, necrosis, dwarfing and deformity of plants, which seriously affects their growth and ornamental value and causes great economic losses[14]. At present, there is no effective drug for preventing and controlling plant virus diseases in the world, but frequent international and domestic trade, exchange of germplasm resources and propagation of plant strains have aggravated the widespread of CymMV and other orchid virus diseases, posing a serious threat to the industrial development of orchids. Therefore, it is of great significance for virus detection, monitoring and early warning and development of virus-free technique by studying the complete genome of CymMV.

Studies on the prevalence, detection and molecular biology of orchid virus diseases have been gradually carried out at home and abroad. It was found that the CymMV genome contains a single-stranded sense RNA (about 6 200 nt), which encodes multiple functional proteins, including RNA-dependent RNA polymerase (RdRp), movement protein (MP) and coat protein (CP), among which MP encoding gene contains 3 open reading frames (TGB1, TGB2 and TGB3)[15]. RdRp has the function of mediating virus replication, while MP and CP proteins can mediate intercellular and long distance movement of virus, respectively. Recent studies have shown that when CymMV and odontoglossum ringspot virus (ORSV) co-infectPhalaenopsis, its small RNA plays an important role in host infection and immune escape[16]. Analysis ofCymbidiumorchids infected by CymMV showed that CymMV infection can indeed promote the expression of one nonexpressor of pathogenesis-related genes 1 (NPR1) and two pathogenesis related genes 1 (PR1)[17], and it is possible to develop a rapid detection sensor based on the interaction between virus and host[18-19]. The research on the complete genome sequence of virus and its functional protein coding genes will lay a foundation for the excavation of key pathogenic factors of viruses, analysis of virus-orchid molecular interaction mechanism and antivirus genetic engineering[20-24].

At present, the complete genome sequence of 14 CymMV isolates have been recorded in GenBank database, including 2 Japanese isolates Cat (LC125633.1) and Japan (AB197937.1), 1 Korean isolate Korean_type_2 (AF016914.1), 1 Indian isolate plm1 (AM055720.1), 1 Singapore isolate Singapore (U62963.1), 3 American isolates Clone 18-29 (EF125180.1), Clone 18-1 (EF125178.1) and Clone 18-30 (EF125179.1), as well as 6 Chinese isolates NJ-1(JQ860108.1), China(KR185347.1), SMi2(AM055640.2), HNXL(HQ681906.1), Taiwan(AY571289.1) and M2(EU314803.1). The above 6 Chinese isolates are from Nanjing, Haikou, Yunnan and Taiwan, and their natural hosts includeCephalotaxushainanensisandVanillaplanifolia. So far, the complete genome sequence of CymMV isolates from Guangdong Province has not been reported yet. In order to explore the genetic information and evolution ofCymbidiumCymMV, the study first carried out virus complete genome analysis, sequence homology analysis, phylogenetic analysis and selection pressure analysis of CymMV isolate from Guangdong Province based on the investigation and detection of viruses in the main production areas of orchid flowers in Guangdong Province, in order to further clarify the virus types and classification status of CymMV in Guangdong. The study will provide important scientific basis and experimental materials for monitoring and early warning of orchid virus diseases and antiviral genetic engineering of orchid in Guangdong Province.

2 Materials and methods

2.1 MaterialsCymMV isolates GZV13 and ZC29 were obtained from Tianhe District and Zengcheng District, Guangzhou City, Guangdong Province, respectively. Both isolates were isolated and purified from the leaves ofCymbidiumsinensesuspected of virus disease. Reverse transcription polymerase chain reaction (RT-PCR) and double-antibody sandwich enzyme-linked immunosorbent assay (DA-SELISA) showed that they were all CymMV-positive[25-30]. Total RNA extraction kit andEscherichiacoliDH5α competent cells were purchased from Tiangen Biotech (Beijing) Co., Ltd. PrimeScript®1stStrand cDNA Synthesis Kit, PrimeSTAR®Max DNA Polymerase, TA clone vector PMD19-T Vector, 5’ RACE and 3’ RACE kits were purchased from TaKaRa Bio, Japan. DNA purification kit and plasmid small volume extraction kit were purchased from Omega Bio-Tek, USA. Other chemicals were analytical pure made in China.Nicotianabenthamianaseeds were preserved by Guangdong Key Laboratory of Ornamental Plant Germplasm Innovation and Utilization.

2.2 Methods

2.2.1Primer synthesis. The published complete genome sequences of 14 CymMV were obtained from GenBank database (Table 1), and the sequences were analyzed online on NCBI website. The relatively conserved regions were selected and specific primers were designed using Primer Premier 5.0 software[31]for complete genome sequencing of 2 CymMV isolates (Table 2). There was at least 800 bp overlapping region between two adjacent amplified fragments. The primers were synthesized by Invitrogen, USA.

Table 1 The complete genomes of 16 CymMV that have been sequenced

Table 2 Information of primer sequence for CymMV sequencing

2.2.2Total RNA extraction and first strand synthesis of cDNA. The CymMV isolates GZV13 and ZC29 were inoculated toN.benthamianafor 20-30 d. The diseased fresh upper leaves were taken to extract the total RNA of the plant by Total RNA Kit, and cDNA was synthesized by reverse transcription using Oligo (dT) Primers and PrimeScript®1stStrand cDNA Synthesis Kit.

2.2.3RT-PCR amplification and gel electrophoresis. The cDNA obtained by reverse transcription was used as the template, and the primer combinations CymMV-N-F/CymMV-N-R, CymMV-M-F/CymMV-M-R and CymMV-C-F/CymMV-C-R-3 were used for RT-PCR assay. PCR assay was performed in a 50 μL system containing 5 μL of 10× PCR buffer, 2 μL of dNTPs (2.5 mmol/L), 1 μL of positive and reverse primers (10 μmol/L), 2 μL of reverse transcription template cDNA, 0.5 μL ofTaqDNA polymerase (2.5 U/μL), and 38.5 μL of ddH2O. The amplification conditions were as follows: pre-denaturating at 94 ℃ for 4 min; denaturating at 94 ℃ for 45 s, annealing at 54-68 ℃ for 45 s, extension at 72 ℃ for 1 min, 30 cycles; extension at 72 ℃ for 10 min. 5 μL of PCR product was added to 2% agarose gel and detected by electrophoresis.

2.2.4Complete genomic sequencing of CymMV and genomic structure prediction. The viral genome fragments cymMV-N, CymMV-M and CymMV-C of the expected size were cut and recycled, then connected to the PMD19-T vector and transformed intoE.coliDH5α competent cells. Positive clones were selected for shaking, and the correct monoclonal plasmid detected by bacterial liquid PCR was extracted and sequenced. The sequencing was completed by Shenzhen BGI Co., Ltd. After Blast alignment, the sequence was confirmed as CymMV genome fragments. Afterwards, the genomic sequences of CymMV isolates GZV13 and ZC29 were assembled using Contig Express software, and the Open Reading Frame (ORF) and genome structure of each functional protein encoding gene of CymMV complete genome sequence were predicted by ORFfinder program of NCBI website, which was then submitted to GenBank database.

2.2.5Complete genome sequence analysis, phylogenetic analysis and selection pressure analysis of CymMV. Sequence alignment and homology analysis of complete genome sequence and each function protein encoding genes of CymMV were conducted through online software Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/). Phylogenetic analysis was performed by MEGA 5.10 software, and phylogenetic tree was established using neighbor-joining method; the bootstrap value was 1 000, and the values with confidence level less than 70% were hidden. Genomic nucleotide polymorphism analysis and selection pressure neutral test were conducted using DNASP 5.0 software, andPivalue was calculated to estimate nucleotide diversity.

3 Results and analysis

3.1 Genome fragment amplification and structure prediction of CymMV isolatesUsing cDNA obtained by reverse transcription as the template, 3 pairs of primers were used for PCR amplification, and finally the viral genome amplified fragments CymMV-N, CymMV-M and CymMV-C with the lengths of 2 800, 2 825 and 3 102 bp were obtained (Fig.1). Sequencing and Blast alignment showed that these fragments were all CymMV genome sequences. The CymMV genome sequences were assembled by Contig Express software.

Except for the 3’-terminal poly (A), the complete genome length of the CymMV isolates GZV13 and ZC-29 were 6 227 nt, with 5’- and 3’-untranslated regions (UTR) of 73 and 76 nt, respectively. The contents of A, G, C and T bases in the complete genome nucleotide sequence of GZV13 were 26.53%, 19.59%, 29.05% and 24.83%, respectively. The contents of A, G, C and T bases in the complete genome nucleotide sequence of ZC-29 were 26.4%, 19.83%, 29.03% and 24.73%, respectively. The complete genome encoded 5 functional proteins, including RdRp (4 254 nt), TGB 1 (701 nt), TGB 2 (339 nt), TGB 3 (276 nt) and CP (276 nt). The above results indicate that GZV13 and ZC-29 have the same genomic structure as most CymMV isolates that have been reported.

3.2 Sequence identity analysisIn order to explore the differences between CymMV isolates (GZV013 and ZC-29) and other reported isolates, the genome nucleotide sequences and amino acid sequences of each functional protein were analyzed for identity.

Note: A. Structure prediction; B. Amplified PCR products.

Complete genome nucleotide sequence analysis showed that Clone 18-1 and Clone 18-30 had the highest sequence identity (98.31%) among the 16 CymMV isolates; HNXL and Clone 18-30 had the lowest sequence identity (86.85%); the identity between GZV013 and ZC-29 was 97.01%. Korean_type_2 and NJ-1 had the highest identity with GZV013 and ZC-29, which were 97.36% and 97.69%, respectively. Clone 18-30 had the lowest identity with GZV013 and ZC-29 (87.40%).

Nucleotide sequence analysis of each fragment showed that the identity rates of 5’UTR, 3’UTR, RdRp, TGB1, TGB2, TGB3 and CP genes were 86.11%-100%, 82.86%-100%, 87.02%-96.67%, 82.88%-98.58%, 87.09%-98.52%, 89.49%-99.28% and 89.43%-99.11%, respectively. As shown in Table 3, Korean_type_2 had the highest identity with 5’UTR of GZV013 (97.36%), while ZC29 had the lowest identity (91.67%). The 3’UTR was exactly identical with ZC29, China, HNXL and Korean_type_2 (100%), but had the largest difference with Taiwan (82.86%). Taiwan (97.27%), Japan (97.86%), Taiwan (98.23%) and Singapore (99.64%) had the highest gene sequence identity with RdRp, TGB1, TGB2, TGB3 and CP of GZV013, while Clone 18-30 had the lowest gene sequence identity, which were 87.09%, 84.45%, 89.09%, 90.94% and 97.92%, respectively.

Amino acid sequence analysis of each region showed that the concordance rates of RdRp, TGB1, TGB2, TGB3 and CP of each coding protein were 94.71%-97.81%, 84.48%-99.57%, 94.64%-100.00%, 87.91%-100.00%, and 97.31%-99.10%, respectively. As shown in Table 3, Taiwan (97.67%), ZC29 (99.57%) and Taiwan (100%) had the highest concordance rates with RdRp, TGB1 and TGB2 of GZV013, respectively, and its TGB3 sequence was completely consistent with that of NJ-1 and China (100%), while Clone 18-30 had the lowest concordance rates with RdRp, TGB1, TGB2 and TGB3, which were 94.71%, 84.48%, 94.64% and 87.91%, respectively. ZC29, NJ-1, Cat and Japan had the highest amino acid sequence identity with CP gene of GZV013, which was 99.1%; Indian isolate plm1 and American isolate Clone 18-29 had the lowest amino acid sequence identity, which was 97.31%.

Table 3 Nucleotide and amino acid sequence similarity between GZV013 and other CymMV isolates %

Based on the above results, the complete genome sequence identity of the 16 CymMV isolates ranged from 86.85% to 98.31%. The nucleotide and amino acid sequences of GZV013 and ZC-29 were highly consistent with those of China (NJ-1, China, HNXL,etc.) and Korea (Korean_type_2) and other regional isolates adjacent to Guangdong Province, and had the lowest identity with Clone 18-1 and Clone 18-30 from USA.

3.3 Phylogenetic analysisThe phylogenetic analysis of 16 CymMV isolates based on the complete genome sequence was performed using MEGA5.0 software. The results (Fig.2) showed that the 16 isolates could be co-clustered into groups A and B. Among them, 8 Chinese isolates (ZC29, GZV013, China, NJ-1, SMi2, HNXL, Taiwan and M2), 2 Japanese isolates (Cat and Japan), 1 Korean isolate (Korean_type_2), 1 Singapore isolate (Singapore) and 1 Indian isolate (plm1) and 1 American isolate (Clone18-29) were clustered into group A; 2 American isolates (Clone 18-1 and Clone 18-30) were clustered into group B.

Fig.2 Phylogenetic analysis of complete genome nucleotide sequences of CymMV isolates

Group A can be divided into subgroups I and II. Two Chinese isolates (HNXL and China), 2 Japanese isolates (Cat and Japan), 1 Korean isolate (Korean_type_2), 1 Indian isolate (plm1) and 1 American isolate (Clone 18-29) were clustered into subgroup I. The natural hosts of these isolates includeC.hainanensis,V.planifolia,Dendrobium,Cattleya,etc.ZC29 was most closely related to NJ-1 from Nanjing, GZV013 was most closely related to M2 from Taiwan, and they clustered into subgroup II with SMi2 from Yunnan, China. Most of the hosts of these 5 isolates belonged toCymbidiumandPhalaenopsis. The results showed that the differentiation of diverse populations of CymMV was closely related to host species and geographical isolation.

3.4 Genomic polymorphism and selection pressure analysis

In order to study the evolution and genetic variation of CymMV genome under natural selection pressure, the complete genome sequences of 16 CymMV isolates were performed nucleotide diversity analysis and neutral test.

According to the polymorphism maps based onPivalues of CymMV in different regions (Fig.3), the C-terminus of RdRp, TGB1 and TGB2 had the highestPivalues in the complete genome, that is, the variability was the highest. ThePivalues of 5’UTR, 3’UTR, N-terminus of RdRp, TGB2 and CP were relatively low, that is, the polymorphism was lower.

The variation of CymMV genome in each region was estimated by theD-value of Tajima neutral test, and the smaller theD-value, the greater the influence of negative selection. The results (Fig.3) showed that theD-values of the complete genome were all negative, indicating that the complete genome sequence was subjected to negative selection pressure to varying degrees.

Fig.3 Genome polymorphism and selection pressure analysis of CymMV

To further determine whether CymMV populations satisfied the neutral evolution model, theF*value of each region of CymMV genome was calculated by Fu and Li tests. The smaller theF*value, the greater the impact of negative selection. The results (Fig.3) showed that allF*values were negative, consistent with the aboveD-value, indicating that each region of CymMV genome was affected by negative selection and conformed to the neutral evolution model. RdRp, TGB1 and TGB2 had the highest variability among the complete genome.

4 Discussion

Orchids generally refer to the plants in Orchidaceae, which is one of the largest families of angiosperms. Among them, Chinese cymbidium is one of the top ten traditional Chinese famous flowers and one of the most advantageous and high-output flowers in Guangdong Province, occupying an important position in the flower market at home and abroad[32]. CymMV, as one of the most widely distributed and most harmful viral pathogens of orchid, can infect many orchids, such asCymbidium,Cattleya,Dendrobium,PhalaenopsisandVanda. Therefore, detection and molecular biology of orchid virus disease will provide support for the establishment of orchid virus disease control system and the healthy development of orchid industrialization.

China is the source of most terrestrial orchids[33]. Cymbidium, as an important characteristic flower, is exported to South Korea, Japan and Southeast Asian countries[34], andPhalaenopsisaphroditeand other foreign orchids are increasingly popular. Due to frequent international and domestic trade, exchange of germplasm resources, and limitation of detection and quarantine techniques of orchid virus disease, orchid virus diseases have been widely spread in the main production and consumption areas of orchids. In this study, the CymMV isolates GZV013 and ZC-29 from Guangdong Province had the highest sequence identity and genetic relationship with isolates from other provinces such as Taiwan and neighboring countries and regions such as Korea, suggesting that they may share a common origin.

Virus evolution is stochastic to some extent, but it is also stable due to selection pressure. During the transmission of orchid virus diseases with host materials, CymMV and other viruses have evolved into new strains that could infect more orchid plants. It was found in this study that the hosts of the isolates closest to GZV013 and ZC-29 were allCymbidiumandPhalaenopsis, while the hosts of more distant isolates includedC.hainanensis,V.planifolia,DendrobiumandCattleya, indicating that the differentiation of diverse populations of CymMV was closely related to host species. The emergence of new CymMV isolates that can infect other orchids poses new challenges to the conservation and utilization of orchids germplasm resources. Therefore, the development of simple, sensitive and low-cost detection technology for orchid virus diseases and the establishment of effective prevention and control system for orchid virus diseases are the keys to restraint the widespread spread of orchid virus diseases.

CymMV can undergo genomic mutation and genetic evolution under different selection pressures. The study showed that each region of CymMV genome was affected by negative selection, and RdRp, TGB1 and TGB2 had the highest variability in the complete genome sequence, suggesting that these genes had more beneficial mutations under natural selection pressure (such as new hosts). But there is no evidence that these genes were significantly subject to greater selection pressure. Therefore, the roles of RdRp, TGB1 and TGB2 in host infection by CymMV need to be further studied.

5 Conclusions

In this study, RT-PCR, DA-SELISA and 3 pairs of universal primers were used to identifyC.sinensewith suspected symptoms of virus disease collected from Guangzhou area, and the complete genome assembly, functional annotation and nucleotide sequence analysis of strains GZV13 and ZC29 were completed. Based on phylogenetic analysis of the complete genome of 16 strains at home and abroad, the classification status of CymMV collected from Guangdong was clarified. The sequence similarity also suggested that the CymMV damaging orchids in Guangdong might come from Nanjing and Taiwan, or from Japan, South Korea and other countries with frequent agricultural product trade with Guangdong. It is suggested that relevant quarantine measures should be strengthened in domestic inter-regional trade and import and export trade. Genome polymorphism and selection pressure analysis shows that CymMV is prone to produce a large number of low-frequency mutations, which brings great uncertainty to the spread of the disease among orchids, suggesting that comprehensive pest control, screening and identification of key disease-resistant genes, screening of disease-resistant varieties and research and development of target agents should be done in the early stage of orchid planting in Guangdong.