The Incidence and Genetic Analysis of Two Betaflexiviruses Capillovirus alphavii and Tepovirus tafpruni in Iran
Article information
Abstract
Viral diseases have emerged as a serious threat to cherry trees production in Iran. To determine which virus(es) are present, three leaves from a sweet cherry tree (Qz5) with diffuse white blotch spots and deformation were subjected to high-throughput sequencing. After de novo assembly, blast analysis revealed that 12 contigs ranging from 360 to 7,433 nucleotides (nts) shared 78–96% nt identities with Capillovirus alphavii (cherry virus A, CVA) and seven contigs, ranging from 350 to 6,844 nts, shared 79–88% nt identities with Tepovirus tafpruni (prunus virus T, PrVT). During a survey, CVA, PrVT, and CVA + PrVT infections were detected in 12.6%, 5.1%, and 7.9% of 724 sour and sweet cherry samples. Phylogenetic analysis revealed that Iranian CVA was grouped into GIIIB, whereas PrVT fell into a distinct branch, which was confirmed by diversity plots. The within-population diversity was lower than the between-population diversity suggesting the contribution of a founder effect on diversification of CVA isolates. Host-specific codon adaptation analysis revealed the highest adaptation of CVA to sour cherry. This could suggest that sour cherry may be one of the closest Prunus species to wild progenitors. It raises the possibility that viruses such as CVA may have exerted evolutionary pressures influencing domestication processes. Additionally, the similarity index indicated that the common plum (Prunus domestica) may have exerted significant evolutionary pressure on CVA and PrVT. The association of CVA and PrVT was reported for the first time in the mid-Eurasian region, specifically in Iran, which represents an issue in phytosanitary certification of cherry plants.
Stone fruits (Prunus spp. in the family Rosaceae) are soaring in economic emphasis (Rubio et al., 2017). Viruses affect fruit quality and yield and decrease the economic life of infected trees (Umer et al., 2019). These crops accumulate many viruses, mainly because they are vegetatively propagated (Maliogka et al., 2018). For these reasons, viruses are easily distributed in new areas. The accessibility of complete genome sequences has accelerated the study of population structure, genetic diversity, and evolutionary forces, such as recombination, mutation, and selection pressure. More than 50 viruses and virus-like agents that infect stone fruit trees have been identified, and this list is constantly being updated (Çağlayan et al., 2019; Kinoti et al., 2020; Maliogka et al., 2018; Rubio et al., 2017). The Betaflexiviridae family (in the order Tymovirales) is one of the most diverse positive-strand RNA virus ensembles.
The host range of betaflexiviruses is very large, and most of them infect fruit trees, such as apples, stone fruits, and grapevines (Da Silva et al., 2019; Liu et al., 2019; Marais et al., 2015a, 2015b; Rwahnih et al., 2019). Capillovirus alphavii (cherry virus A, CVA), is one of the most widespread betaflexiviruses affecting sweet cherry (P. avium) and sour cherry (P. ceracus) worldwide (Marais et al., 2011; Sabanadzovic et al., 2005). This virus has also been reported to a lesser extent from noncherry hosts, including apricot (P. armeniaca), cherry plum (P. cerasifera), flowering cherry (P. serrulata), Japanese apricot (P. mume), Japanese plum (P. salicina), and peach (P. persica) (Ben Mansour et al., 2023; James and Jelkmann, 1998; Marais et al., 2008a). Despite the high prevalence of CVA infection, no natural vector has been reported, however, it can be readily transmitted from plant to plant through vegetative propagation and grafting (Ben Mansour et al., 2023; Marais et al., 2008b).
To date, there is no clear evidence of specific symptoms associated with CVA, but studies have shown that in mixed infections with other viruses, the severity of symptoms is greater, and rootstock/scion compatibility is affected (Costa et al., 2022; Jelkmann, 1995; Marais et al., 2008a, 2012; Noorani et al., 2022). The genome of CVA comprises a positive single-stranded RNA of 7,383 nucleotides (nt), except for the 3′-terminal poly (A) tail (Jelkmann, 1995). The genomic structure of CVA is similar to that of Apple stem grooving virus, the type species of the Capillovirus genus, and includes two overlapping open reading frames (ORFs). A large polyprotein (266 kDa), including the RNA-dependent RNA polymerase (RdRp) fused in frame to the coat protein (CP), and a 52 kDa putative movement protein (MP) are encoded by ORF1 and ORF2, respectively (James and Jelkmann, 1998; Jelkmann, 1995). High genomic diversity has been reported in the CVA population, and the nucleotide identity of the entire genome varies between 79% and 99% (Kesanakurti et al., 2017; Kinoti et al., 2020; Noorani et al., 2022). CVA isolates have been classified into five molecular groups based on their RdRp (Marais et al., 2012); however, six different molecular groups have been reported according to their full genome sequences (Kesanakurti et al., 2017). At present, CVA is grouped into seven phylogenetic clusters (Gao et al., 2017).
In 2015, Marais et al. (2015a) proposed using Tepovirus tafpruni (prunus virus T, PrVT) species in the Tepovirus genus. However, biological information concerning PrVT is still scarce, but molecular screening of prunus samples has allowed the identification of three hosts (P. domestica, P. avium, and P. cerasifera) in Italy and Azerbaijan, with an overall prevalence of 1%, and from Pyrus communis in Kyrgyzstan (Costa et al., 2022). Since all reported host plants of PrVT were simultaneously infected with apricot pseudochlorotic leaf spot virus - APCLSV (Trichovirus), apple chlorotic leaf spot virus - ACLSV (Trichovirus), CVA (Capillovirus), prune dwarf virus - PDV (Ilarvirus), or little cherry virus - LChV1 (Closteroviridae) which are well-known prunus-infecting viruses, symptoms associated with PrVT infection have not yet been described, and it is difficult to associate specific symptoms with PrVT infection (Marais et al., 2015a, 2015b). The genome organization is similar to that of some members of the family Betaflexiviridae, with three overlapping ORFs (RNA polymerase, MP, and capsid protein).
Iran is one of the main areas of commercial sweet cherry production in West Asia. Sweet cherry orchards are widely spread across nearly all geographical areas of Iran, covering 14,589 hectares, with an annual production of 105,390 tonnes and a yield of 7.2 tonnes/ha (Food and Agriculture Organization of the United Nations, 2022). During the past decade, the average yield of cherry trees in Iran has decreased from 8.7 tonnes per hectare in 2013 to 7.2 tonnes per hectare in 2022 (Food and Agriculture Organization of the United Nations, 2022). Recently, the prevalence of virus-like symptoms, including leaf blotch, leaf deformation, and vein yellowing (Fig. 1), was reported in sweet cherry orchards and nurseries in some provinces of Iran. Despite the economically important role of viruses infecting sweet cherry, molecular analysis of these pathogens is less studied in Iran, possibly due to a lack of genetic sequence data. This study was carried out to assess the presence of viruses infecting sweet cherries via high-throughput sequencing (HTS). Knowledge of the identification of viruses in the two major growing sweet cherry areas is important for expanding effective control plans for these crops. Furthermore, in this study, host-specific codon adaptation analysis was conducted using two CVA and PrVT detected viruses in comparison to all previously reported sequences of CVA and PrVT that provides more information about host-pathogen coevolution, and phylogenetic relationships among specified species (Biswas et al., 2019; He et al., 2019; Xu et al., 2008; Yang et al., 2022).
Virus-like symptoms found in Iranian cherry plants.
Materials and Methods
Plant materials and viral source
In the early summer of 2023, three leaves with virus-like symptoms were collected from sweet cherry orchards and nurseries in Qazvin Province, Central Iran, among the main sweet cherry production districts in Iran. The cherry leaves showed virus-like symptoms, including diffuse white blotch spots, yellows, and deformation (Fig. 1). Three symptomatic leaves of an individual plant (Qz5) were submitted to HTS analysis. In addition, infection of CVA and PrVT were checked in sour (n = 26) and sweet (n = 32) cherry gardens in seven provinces of Iran (Alborz, Azarbaijan-sharghei, Kurdistan, Qazvin, Semnan, Tehran, and Zanjan). Totally, 724 leaf samples were collected from sour (n = 353) and sweet (n = 371) cherry trees (Table 2).
RT-PCR results on leaf samples collected from sour and sweet cherry gardens in seven provinces
RNA library preparation
Total RNA from a pool of three symptomatic sweet cherry leaf samples (Qz5) (Fig. 1) was selected for RNA-Seq. A Ribo-zero rRNA Removal Kit (Epicenter, Madison, WI, USA) was used to delete rRNA from the total RNA. RNA-Seq libraries were constructed using TruSeq Stranded Total RNA for Illumina according to the manufacturer’s instructions. Paired-end reads of 150 bp were generated by sequencing on the Illumina HiSeq 2000 platform (Novogene, Beijing, China). CLC Genomics Workbench version 20 (QIAGEN, Venlo, The Netherlands) was used for data analysis of FASTQ files. A de novo assembly of the reads into contigs using CLC Genomics Workbench version 9 (QIAGEN) and a BLASTN search of GenBank (https://www.ncbi.nlm.nih.gov/genome/viruses) were used for the viral genome detection. A nucleotide BLAST of the obtained contigs was performed using Genious version 22 (Biomatters, Auckland, New Zealand). The selected contig from the assembly mapped to viral sequences in the GenBank database was analyzed for the presence of putative viruses using the NCBI database. The results were validated using the Virus Detect tool (http://virusdetect.feilab.net/cgi-bin/virusdetect/vdo_home.cgi) and we also validated the result by mapping the clean reads to the viral genome that was detected in the blast result. As a confirmation, the 5′- and 3′-termini of the CVA and PrVT genomes were obtained using 5′- and 3′-RACE kits (Roche Diagnostics, Mannheim, Germany) according to the manufacturer’s instructions. The complete genomes of the viral ORFs and untranslated regions (UTRs) at the 5′- and 3′-ends of each gene were obtained (Table 1). The virus-related sequences of CVA (n = 126) and PrVT (n = 5), were aligned using the ClustalX program for further analysis (Larkin et al., 2007).
Genome features and ORFs of Iranian CVA-Qz5, and PrVT-Qz5 isolates
Total nucleic acid extraction and reverse transcription polymerase chain reaction
To reconfirm the existence of the viruses, specific primers were designed using the sequence assemblies and available genome sequences to detect putative viruses by reverse transcription polymerase chain reaction (RT-PCR). Total RNA extraction, was conducted using TRIzol (Thermo Fisher, Invitrogen, Waltham, MA, USA). First-strand cDNA synthesis was conducted using Easy cDNA Synthesis Kit (Parstous Co., Tehran, Iran) according to the manufacturer’s protocol. The cDNA (3 μl) was then amplified using 0.5 μM (20 pmol/μl) each of the CVA-MPF (5′-TGTCGATCATACCAGTYAAGAAG-3′), and CVA-MPR (5′-GTCAATGAGTGCCACAGTT-3′) primers for CVA (designed in this study); and the Tep-1R (5′-ACTTTATTAAGTTAAAGCTAAGCCGCC-3′) and Tep-1F (5′-CCIGGIYTIGGIGGIMGICAYGGICARATG-3′) primers for PrVT (Marais et al., 2015a) in a 50-μl reaction volume. After the first step of denaturation (5 min at 95°C), 40 cycles of 30 s at 95°C, 30 s at 50°C, and 45 s at 72°C were applied, followed by a final step of 10 min at 72°C. The amplified fragments were observed by agarose gel electrophoresis, and their nucleotide sequences were determined.
Phylogenetic and recombination analysis
Phylogenetic reconstructions of the full genome sequences of CVA isolates (n = 114) were performed via the neighbor-net method in SplitsTree4 v.4.12.6 (Huson and Bryant, 2006), with 1,000 replicates used to evaluate the statistical significance of the branches. Cherry necrotic rusty mottle virus, the type species of the genus Robigovirus, was used as an outgroup. Based on the CP gene, the phylogenetic grouping of PrVT isolates was determined by the neighbor-joining (NJ) method implemented in MEGA X (Kumar et al., 2018). Pairwise nucleotide diversities were estimated using the Kimura two parameters implemented in the phylip3.67 program and are shown using color plots. The number of synonymous differences per synonymous site from averaging over all sequence pairs between phylogroups was conducted using the Nei-Gojobori (assumed transition/transversion bias = 2) model (Zhang et al., 1998). Furthermore, the Nei-Gojobori model was used for mean diversity calculations within and between subpopulations. These evolutionary analyses were conducted in MEGA X (Kumar et al., 2018).
Recombination events, putative parental isolates of recombinants, and recombination breakpoints were analyzed via several methods implemented in RDP4 version 4.39 (Martin et al., 2015), with the default configuration and Bonferroni corrected P-value cutoffs of 0.05 and 0.01. Only recombination events supported by at least three methods with an associated P-value of 1.0 × 10−6 were considered.
Estimation of population genetic parameters
DnaSP v.6 was used to estimate population genetic parameters for CVA (Rozas et al., 2017). Nucleotide and haplotype diversities were determined for all populations (n = 114 sequences) and for each population. Furthermore, other parameters including nucleotide diversity (per site) (Pi), the average number of nt differences between the sequences (k), the number of haplotypes (h), and haplotype diversity (Hd) were also estimated for each population. Fu and Li’s D (Fu and Li, 1993) and Tajima’s D values were calculated to find the demographic expansion. The genetic differentiation between populations was estimated using Kst*, Ks*, Z*, Snn, and Fst parameters according to geographical regions where CVA was isolated (Hudson, 2000; Tajima, 1989). There is no genetic differentiation if Kst* is near zero (null hypothesis) (Tsompana et al., 2000). A smaller mean of Z* and Snn values indicates low genetic differentiation and vice versa (Hudson, 2000). The null hypothesis in Ks*, Kst*, and Z*, is rejected by a significant P-value < 0.05 (Hudson, 2000; Tsompana et al., 2000). In addition, genetic differentiation between populations was estimated using Fst values, wherein it considered as very high for Fst > 0.25, high for 0.15 < Fst < 0.25, moderate for 0.05 < Fst < 0.15, and low for Fst < 0.05 (Hudson et al., 1992). Concerning the criterion by Hudson (Hudson, 2000), gene flow is considered low for Nm < 1, high for 1 < Nm < 4, and very high for Nm > 4. Owing to the small number of sequences, this analysis was not performed on the PrVT.
Host-specific codon adaptation analysis
The codon adaptation index (CAI) predicts the relative adaptation of a virus to its potential host, which means that sequences with higher CAIs are considered to have great adaptability to the host (Sharp and Li, 1987). The CAI values ranged from 0 to 1 and were computed using the CAIcal SERVER (Puigbò et al., 2008) (http://genomes.urv.cat/CAIcal/RCDI/). Furthermore, the similarity of codon usage between CVA and a reference genome was calculated using relative codon deoptimization index (RCDI) analysis. The RCDI values of the CVA were obtained using the RCDI/eRCDI server (Puigbò et al., 2010) (http://genomes.urv.cat/CAIcal/RCDI/). An RCDI value of 1 indicates complete adaptation of the virus to its host. Moreover, RCDI values higher than 1 indicate deoptimization codon usage patterns between the virus and its host(s) (Mueller et al., 2006). The reference datasets for the hosts were obtained from the Codon Usage Database (http://www.kazusa.or.jp/codon/). Owing to the small number of PrVT sequences, CAI/RCDI analysis was not performed. The similarity index (SiD) is a recent method for estimating the resemblance of the overall codon usage of the hosts to that of a certain virus. The SiD was calculated to determine the effect of the overall host codon usage pattern on the CVA and PrVT isolates. The value of the SiD analysis ranges between 0 and 1, where a higher value indicates that the host has a dominant effect on the usage of codons and vice versa (Zhou et al., 2013).
Results
Viral sequence mapping and genome assembly
From 41,085,134 raw reads, we obtained 2,823,485 quality-filtered and trimmed reads after removing the reads mapped to the host genome. A de novo assembly of the reads into contigs using CLC Genomics Workbench generated 43,551 contigs (with an average length of 861 bp, and about 30% longer than 1 kb). BLAST analysis showed that 12 contigs ranging from 360 to 7,433 nucleotides (nts), shared 78–96% nt identities with CVA. Furthermore, seven contigs, ranging from 350 to 6,844 nts, shared 79–88% nt identities with PrVT. The largest contig (7,433 nts GenBank accession no. PP736769) matching CVA, covering 87% of CVA type isolate genome (NC_003689) (Jelkmann, 1995), and coinfection with PrVT with contig (6,844 nts GenBank accession no. PP736770), covering 86.09% identical to PrVT isolate from Kyrgyzstan (accession no. MW331540). No additional viruses or viroids were retrieved from the transcriptomic reads. The genomic characteristics of viral ORFs, along with 3′- and 5′-UTRs, are presented in Table 1. The Iranian CVA-Qz5 shared 83.37 to 98.94% identity with other CVA isolates obtained from GenBank. Similar to the genus Capillovirus, the Iranian CVA-Qz5 genome has two ORFs. ORF1 is 7,029 nt long, encoding a deduced RdRp of 2,342 amino acids (aa), and ORF2 is 1,392 nt long, encoding a deduced MP of 463 aa. The 5′- and 3′-UTRs contained 106 and 590 nt, respectively (Table 1). The complete genome of Iranian CVA-Qz5 was submitted to GenBank under accession number PP736769. The Iranian PrVT-Qz5 shared 78.69–86.09% identity with four PrVT isolates, obtained from GenBank. PrVT-Qz5 has a typical genome with three ORFs in the genus Tepovirus. ORF1 consists of 5337 nt encoding a polyprotein of 1,779 aa residues, ORF2 in a different reading frame of 1,152 nt encodes the MP with 384 aa residues, and ORF3 with 666 nt encodes the CP with 222 aa residues (Table 1). The 5′- and 3′-UTRs consisted of 44 and 79 nt, respectively. The complete genome sequence of the Iranian PrVT-Qz5 was submitted to GenBank under accession number PP736770.
Detection of CVA and PrVT
The expected amplified DNA fragments of approximately 688 bp, and 473 bp were successfully amplified from our original Qz5 and other samples (Supplementary Fig. 1). No expected bands were observed with the RNA templates of the healthy sweet cherry plants. In addition, CVA and PrVT were detected in all seven investigated provinces (Table 2). Of 353 sour cherry samples, 33 and 17 were infected with CVA and PrVT, respectively, and 17 were infected with both viruses. Also, out of 371 sweet cherry samples, 58 and 20 were infected with CVA and PrVT, respectively, and 33 were infected with both viruses. CVA and PrVT were detected in both symptomatic and non-symptomatic samples. CVA, PrVT, and CVA + PrVT infections were detected in 91 (12.6%), 37 (5.1%), and 57 (7.9%) of 724 sour and sweet cherry samples. These results indicate a high prevalence of infections with these viruses in the surveyed cherry orchards and nurseries in Iran.
Phylogenetic and recombination analysis
After deleting the ungrouped CVA isolates, CVA isolates (n = 114) clustered into seven phylogroups. The Iranian CVA-Qz5 (accession no. PP620725) fell into phylogroup GIIIB, which is consistent with the results of sequence identity analysis (Fig. 2). Both GII and GIII consisted of the CVA isolates from noncherry and cherry hosts; nevertheless, subcluster GIIIA, consisted of noncherry isolates, including P. armeniaca (KY445749 and LC523018) from South Korea and Australia, respectively, as well as, a P. mume isolate (KY286055). The diversity plot revealed the lowest nucleotide diversity (0.0 to 4.3%) for phylogroup I (GI), and the highest nucleotide diversity (21.5 to 24.9%) was found among the phylogroups (GV, GVI, and GVII). Nucleotide diversity ranging from 8.6 to 17.2% was detected for phylogroups GII, GIII, and GIV (Fig. 3). Specific grouping on the basis of host species origin or geographic isolation was not obtained using neighbor-net phylogenetic analysis (Fig. 2). The number of synonymous differences per synonymous site from mean diversity calculations within subpopulations (0.04 ± 0.001) were lower than evolutionary diversity between subpopulations (0.16 ± 0.001). In addition, estimates of codon-based evolutionary divergence over sequence pairs between seven CVA phylogroups differed between 0.210 to 0.296 (Table 3). However, the highest sequence divergence was found between GV and other phylogroups (Table 3).
Neighbor-net phylogenetic networks were obtained from the whole-genome sequences of cherry virus A. Phylogenetic networks were created using the SplitsTree4 v.4.12.6 software program. The scale bar signifies a genetic distance of 0.01 nucleotide substitutions per site. Cherry necrotic rusty mottle virus (CNRMV) was used as an outgroup.
Pairwise nucleotide diversity plot of cherry virus A isolates using the complete genome. Each phylogroup is indicated by different colors.
Estimates of codon-based evolutionary divergence over sequence pairs between CVA phylogroups
Iranian PrVT-Qz5 was placed on a distinct branch from previously reported isolates using the NJ method, with the highest identity to the PGQP2 isolate (accession no. MW331540) (Fig. 4A and B). The lowest and highest nucleotide diversity was detected for the PGQP2 and PGQP1 isolates from pear trees (Pyrus communis) in Kyrgyzstan and the Italian C21 isolate (accession no. KF700262), respectively.
Phylogenetic analysis of prunus virus T (PrVT). (A) Neighbor-joining phylogenetic tree was constructed from the whole-genome sequences of PrVT. The scale bar signifies a genetic distance of 0.05 nucleotide substitutions per site. (B) Pairwise nucleotide diversity plot of PrVT isolates using the complete genome sequence.
Different methods were used for recombination breakpoint prediction and to provide evidence for recombination events spread across 14 CVA-analyzed genomes (Supplementary Table 1). Remarkably, following the adopted criteria (detectable by at least three different methods), recombination breakpoints were detected for divergent or ungrouped (UN) CVA isolates that may represent further phylogroups. In addition, recombinant isolates were detected for some isolates belonging to GI (KY510893, and KY510866), GII (LC523017, and MZ291923), and GIV (KY510845), however, all four CVA isolates (KY510859, KY510862, LC523003, and MK847263) clustered in GVI were found recombinant (Supplementary Table 1). Some isolates from South Korea and Australia indicate multiple recombination, whereas for a Korean LC752551 isolate, our Iranian isolate was found as the major parent (Supplementary Table 1).
Estimation of genetic parameters
Different parameters were considered for estimating genetic differentiation according to the whole and each phylogroup and where the CVA was isolated. High haplotype diversity (from 0.9987 to 1.000) along with low nucleotide diversity per site (Pi) were found for CVA isolates, indicating a recent selective sweep or population expansion after a recent bottleneck (Table 4). In addition, Tajima’s D and Fu tests are used to check whether DNA sequences from the population of interest evolve randomly or under a non-random process. A negative value of Tajima’s D (or <0) and Fs also confirmed evidence for an excess number of alleles, as would be expected from a recent population expansion or from genetic hitchhiking (Table 4). A positive Tajima’s D and Fu test signifies low levels of both low and high-frequency polymorphisms, and is evidence for a deficiency of alleles, as would be expected from a population bottleneck. Although not statistically significant P > 0.05.
Demographic test statistics for CVA phylogroups and their isolation regions
Population genetic parameters for CVA were calculated according to geographical isolation (Table 5). Pairwise Fst values showed a weak genetic differentiation between CVA populations from different geographical isolation. The Fst varies from 0.03699 (between Asia and Australia populations) to 0.13961 (Australia and Europe) populations. According to Nm values frequent gene flow between CVA populations differed from very high to high. Genetic differentiation and frequent gene flow were also confirmed with Ks*, Kst*, Z*, and Snn statistics values (Table 5). Negative Fst values between (Australia and Canada; and Asia & Canada) should be effectively seen as zero values. Then the high P-value, which indicates non-statistically significant differences in the frequencies of the marker examined between the targeted populations (Table 5).
Genetic differentiation analysis between CVA populations from different geographical origins
Host-specific codon adaptation patterns
Codon usage adaptation (CAI) analysis was performed to assess codon usage optimization and adaptation of CVA to its hosts. The mean CAI values of CVA sequences were 0.815, 0.801, 0.799, 0.795, and 0.729 for sour cherry, plum blossom, sweet cherry, apricot, and common plum, respectively. Furthermore, we performed an RCDI analysis to compare the similarity in codon usage of the CVA sequences and their hosts. The mean RCDI values were highest for common plum (1.834), followed by apricot (1.582), sweet cherry (1.541), plum blossom (1.556), and sour cherry (1.587) (Fig. 5). These values indicated that CVA host adaptation was greater for sour cheery than for the other hosts.
Codon adaptation index (CAI) analysis and relative codon deoptimization index (RCDI) analysis of cherry virus A sequences from the natural hosts (sour and sweet cherries; plum blossom, apricot, and common plum).
SiD analysis was also performed to understand how the CVA and PrVT codon patterns are affected by these hosts’ codon usage patterns. SiD value of the common plum (P. domestica) was higher than that of the other hosts for CVA (Fig. 6A), and PrVT-CP and PrVT-MP (Fig. 6B), indicating that P. domestica had a great impact on the viruses. Furthermore, the strong selection pressure of the common plum is indicative of a gradual adaptation of CVA and PrVT to this host.
Similarity index (SiD) data of the cherry virus A (A); for the coat protein (CP) and movement protein (MP) sequences of prunus virus T (PrVT) concerning their natural hosts (B). The PrVT-CP and PrVT-MP genes are indicated in blue and orange, respectively.
Discussion
Despite reports of some viruses infecting stone fruit trees, including cherries in Iran (Moeini, 2004; Pourrahim and Farzdafar, 2018), no viral species from Betafexiviridae have been previously reported. In this study, using HTS, CVA and PrVT infection was found to be associated with infected sweet cherries showing viral symptoms (Fig. 1). In addition, RT-PCR analysis using specific primers confirmed the prevalence of mixed infection with CVA and PrVT in sweet and sour cherry plants in orchards and nurseries in the seven provinces of Iran (Alborz, Azarbaijan-sharghei, Kurdistan, Qazvin, Semnan, Tehran, and Zanjan) (Table 2). A mixed infection of CVA and PrVT was also indicated in 57 of 724 samples (7.9%) by RT-PCR. CVA and PrVT were also detected in asymptomatic samples, so the relationship between these viruses and symptoms requires more studies. Perennial plants, which are naturally long-lived, provide an excellent opportunity for infection by one or more viruses over time. Therefore, mixed infections by different plant viruses are common, as is often in fruit trees (Diaz-Lara, 2019). Even though no more viruses have been detected among the leaf samples subjected to HTS, the role of mixed infection with CVA and PrVT in viral symptoms and susceptibility of the sweet cherry cultivars must be further studied in more samples. As it has been shown for other stone fruit tree viruses, symptom development is dependent on various factors including virus strain, host susceptibility, temperature, light, and so on (Pallás et al., 2012). The full-length genomes of two Iranian betaflexiviruses, CVA and PrVT, typical of the Capillovirus and Tepovirus genera, respectively, were detected using HTS (Table 1). Our study represents the first report of members of the genera Capillovirus and Tepovirus infecting sweet and sour cherries in Iran.
Studying the genetic diversity of viruses increases our understanding of geographical origin, variations in virulence, the emergence of new viruses or epidemics, and viral evolution. The positions of the Iranian CVA and PrVT isolates were specified using sequence data and phylogenetic analyses. Phylogenetic analysis of the full genome sequences revealed that the CVA isolates clustered into seven distinct phylogroups, which is in accordance with previous studies (Gao et al., 2017; Kesanakurti et al., 2017; Kinoti et al., 2020; Noorani et al., 2022). The Iranian CVA (accession no. PP620725) belongs to phylogroup GIIIB, which contains cherry and noncherry isolates (Gao et al., 2017). The highest similarities (lowest nucleotide diversity) were found for CVA phylogroup I (GI), which may suggest strong selection pressures (e.g., because of fitness) or evolutionary founder effects/bottlenecks. In addition, a number of statistically significant recombination events were detected in CVA complete genome sequences revealing that a genomic exchange is responsible for the emergence of CVA strains as indicated in previous studies (Gao et al., 2017; Kesanakurti et al., 2017). However, no recombination site was detected in Iranian CVA and PrVT isolates.
The within-population diversity was lower than the between-population diversity suggesting the contribution of a recent expansion after a bottleneck, namely a ‘founder effect on diversification of CVA isolates. Furthermore, the recent distribution of the population can be inferred from high haplotype diversity and low genetic differentiation (Table 3). A combination of high haplotype diversity and low genetic diversity, assessed by mitochondrial DNA markers, is taken as evidence of a recent population expansion after a genetic bottleneck and this also was found for turnip mosaic virus (Tomitaka and Ohshima, 2006), and cauliflower mosaic virus (Farzadfar et al., 2014). A negative value of Tajima’s D (or <0) and Fs indicates populations of interest are likely to experience recent selective sweeps or population expansion after the bottleneck. In other words, a negative Tajima’s D signifies an excess of low-frequency polymorphisms relative to expectation as indicated for CVA populations (Gao et al., 2017; this study). A positive Tajima’s D signifies low levels of both low and high-frequency polymorphisms. A negative value of Fs is evidence for an excess number of alleles, as would be expected from a recent population expansion or genetic hitchhiking. A positive value of Fs is evidence of a deficiency of alleles, as would be expected from a population bottleneck. Statistical significance: Not significant P > 0.05. For no significant result, it could be due to a low sample number. A positive Tajima’s D signifies low levels of both low and high-frequency polymorphisms, indicating a decrease in population size and/or balancing selection. In other words, the values of Tajima’s D are interpreted as follows: Tajima’s D = 0: Population evolving as per mutation-drift equilibrium. Tajima’s D < 0: Recent selective sweep or population expansion after a recent bottleneck. Tajima’s D > 0: Balancing selection or sudden population contraction.
Population genetic parameters according to the geographical isolation of CVA showed a weak genetic differentiation or high frequent gene flow between CVA populations from different geographical regions (Table 5). This goes back to the accumulation of many viruses in fruit trees (Maliogka et al., 2018), and readily transmitted viruses from plant to plant via vegetative propagation and grafting (Ben Mansour et al., 2023). The highest gene flow was indicated between Canada, Asia, and Australia because of Negative Fst values which should be effectively seen as zero values. A zero value for Fst means no genetic subdivision between the populations considered.
Full genome analysis using the NJ method revealed that Iranian PrVT-Qz5 was placed in a distinct branch from previously reported isolates, with the highest identity to the noncherry isolate PGQP2 (accession no. MW331540) from P. communis from Kyrgyzstan; however, the lowest identity was found with the Italian PrVT isolate C21 (accession no. KF700262) from P. avium. Considering the long history of wild Prunus species in Iran and the high population variety of prunus rootstocks used for cherry grafting, the genetic diversity of CVA and PrVT isolates in Eurasian Iran needs further study.
Although sweet cherry has always been suggested to be the primary CVA host, a strong link between CVA and sour cherry was observed in this study using CAI and RCDI analyses which revealed that the greatest adaptation of CVA was to sour cherry (P. cerasus) (Fig. 5). This may indicate that CVA entered prunus hosts relatively earlier than PrVT, with more opportunity to accumulate genetic diversity through interactions with various Prunus species. The results demonstrating CVA’s high adaptation to sour cherry also hint at interesting implications for domestication and the evolutionary origins of prunus hosts. This adaptation could suggest that sour cherry may be one of the closest Prunus species to wild progenitors raising the possibility that viruses such as CVA may have exerted evolutionary pressures influencing domestication processes. Although this idea could extend beyond traditional plant pathology, it presents a compelling direction for future studies on the evolutionary interplay between viruses and host domestication. However, the CAI/RCDI analysis of PrVT could not be calculated in this study due to insufficient sequence information. To make a correct judgment about the adaptation of PrVT, obtaining more sequences of this virus from different prunus hosts is necessary. Additionally, SiD analysis revealed that the common plum (P. domestica) exerted greater evolutionary pressure on CVA and PrVT than the other hosts did. This finding may suggest that CVA and PrVT may evolved in P. domestica or another Prunus progenitor and were subsequently vertically or horizontally transmitted to descendant Prunus species, as also suggested for Citrus tristeza virus-CTV (Biswas et al., 2019).
Untargeted descriptions of all viruses by HTS have made considerable progress in discovering and characterizing viruses, which is necessary for developing effective control strategies, early diagnosis, and precise identification of viral agents. The association of CVA and PrVT with virus-like symptoms of sweet and sour cherry leaves was reported for the first time in mid-Eurasia, Iran, and was confirmed by RT-PCR and sequencing. Developing disease-resistant plants is challenging, as mixed infections are common in naturally occurring fruit tree plants. Hence, better knowledge of viral disease could help develop effective control strategies (e.g., cross-protecting mild strains), and the precise identification of viral agents to prevent the spread of infection, especially in nurseries.
Notes
Conflicts of Interest
No potential conflict of interest relevant to this article was reported.
Acknowledgments
This work was carried out in the Plant Virus Research Department, Iranian Research Institute of Plant Protection (IRIPP), Agricultural Research, Education and Extension Organization (AREEO). Financial support from the IRIPP project (Projects No. 010550, 961463, and 83063) is gratefully acknowledged.
Electronic Supplementary Material
Supplementary materials are available at The Plant Pathology Journal website (http://www.ppjonline.org/).