Identification and Characterization of New ‘Candidatus Phytoplasma solani’ Strains Associated with Bois Noir Disease in Vitis vinifera L. Cultivars Showing a Range of Symptom Severity in Georgia, the Caucasus Region
- Fabio Quaglino , Department of Agricultural and Environmental Sciences–Production, Landscape, Agroenergy (DISAA), University of Milan, Milano, Italy
- David Maghradze , Institute of Horticulture, Viticulture and Oenology, Agricultural University of Georgia, Tbilisi, Georgia
- Paola Casati , DISAA, University of Milan
- Nona Chkhaidze , Laboratory of Plant Anatomy and Physiology, Agricultural University of Georgia, Tbilisi
- Mzagho Lobjanidze , Institute of Entomology, Agricultural University of Georgia, Tbilisi
- Adriano Ravasio
- Alessandro Passera
- Giovanni Venturini
- Osvaldo Failla
- Piero Attilio Bianco , DISAA, University of Milan
Abstract
Evidence from a preliminary survey highlighted that ‘Candidatus Phytoplasma solani’, the etiological agent of bois noir (BN) disease of grapevine, infects grapevine varieties in Georgia, a country of the South Caucasus. In this study, field surveys were carried out to investigate the BN symptom severity in international and Georgian native varieties. ‘Ca. P. solani’ was detected and identified by polymerase chain reaction-based amplification and restriction fragment length polymorphism analysis of 16S ribosomal DNA, and further characterized by multiple gene typing analysis (vmp1 and stamp genes). Obtained data highlighted that the majority of Georgian grapevine varieties showed moderate and mild symptoms, whereas international cultivars exhibited severe symptoms. Molecular characterization of ‘Ca. P. solani’ from grapevine revealed the presence of 11 distinct phytoplasma types. Only one type (VmGe12/StGe7) was identical to a strain previously reported in periwinkle from Lebanon; the other ‘Ca. P. solani’ types are described here for the first time. Phylogenetic analyses of vmp1 and stamp gene concatenated nucleotide sequences showed that ‘Ca. P. solani’ strains in Georgia are associated mainly with the bindweed-related BN host system. Moreover, the fact that ‘Ca. P. solani’ strains are distributed in grapevine cultivars showing a range of symptom intensity suggests a different susceptibility of such local cultivars to BN.
Grapevine yellows (GY) are a phytoplasma-associated disease complex that induces severe crop losses in almost all varieties used for wine production. GY have been described in Europe, Australia, and several American countries, and more than seven genetically different phytoplasmas have been associated with such disease complex (Laimer et al. 2009). Typical GY symptoms include berry shrivel, desiccation of inflorescences, color alterations and curling of the leaves, reduction of growth, and irregular ripening of wood (Belli et al. 2010).
Among GY, bois noir (BN) is responsible for serious crop losses in the Euro-Mediterranean area and on other continents (Belli et al. 2010; Foissac et al. 2013). Phytoplasmas are cell-wall-less obligate intracellular parasites of the Mollicutes class, classified through the nucleotide sequence analysis of their housekeeping genes (Davis et al. 2013; IRPCM Phytoplasma/Spiroplasma Working Team–Phytoplasma Taxonomy Group 2004). On the basis of unique biological properties and exclusive molecular markers within multiple genes, the etiological agent of BN disease has been attributed to phytoplasma strains (BNp) belonging to the species ‘Candidatus Phytoplasma solani’, subgroup 16SrXII-A (Quaglino et al. 2013). In Euro-Mediterranean regions, BNp strains are transmitted to grapevine by Hyalesthes obsoletus Signoret (Homoptera: Cixiide), a polyphagous vector living preferentially on nettle (Urtica dioica L.), bindweed (Convolvulus arvensis L.), and chaste tree (Vitex agnus-castus L.) inside or around vineyards (Kosovac et al. in press; Langer and Maixner 2004; Sharon et al. 2005). Moreover, Reptalus panzeri has been reported as a natural vector of BNp in Serbian vineyards (Cvrković et al. 2014). Additionally, Anaceratagallia ribauti and R. quinquecostatus were experimentally confirmed as vectors of ‘Ca. P. solani’ but not to grapevine (Pinzauti et al. 2008; Riedle-Bauer et al. 2008); therefore, currently, such insects are not considered to be involved in BNp transmission to grapevine. Other studies showed that several weeds host BNp and can play a role in BN diffusion (Mori et al. 2015). The biological complexity of BN disease has stimulated research on molecular markers of BNp genetic diversity. Multilocus sequence typing, based on molecular characterization of tuf (Langer and Maixner 2004), secY, vmp1 (Fialová et al. 2009), and stamp (Fabre et al. 2011) genes, highlighted the presence of numerous genetically distinct strains characterized by different distribution and prevalence in the Euro-Mediterranean basin (Atanasova et al. 2015; Foissac et al. 2013; Kosovac et al. in press; Kostadinovska et al. 2014; Murolo and Romanazzi 2015).
Due to this complexity, it is difficult to design efficient control strategies against BN. Because insecticides applied to the grapevine canopy influence neither the disease nor the presence of H. obsoletus, the management of H. obsoletus host plants in the vineyards and surrounding areas is considered crucial for BN control (Mori et al. 2012). Thus, preventive measures such as checking the sanitary status of propagation materials and treating diseased mother plants through thermotherapy are applied to limit long-distance dissemination and in-field spread of the disease. Other strategies for reducing BN spread or incidence are based on (i) preventive removal of the grape suckers on which H. obsoletus could feed after grass mowing, (ii) trunk cutting above the engagement point of the symptomatic grapevines, and (iii) treatments by resistance inducers (Belli et al. 2010).
An ambitious strategy for phytoplasma disease control is based on the selection of resistant, tolerant, or not susceptible plant varieties, cultivated or not-cultivated, as source of resistance genes for plant breeding programs. Previous studies identified the presence of plant species or varieties showing low susceptibility to phytoplasma infection (Bianco et al. 2011). Unfortunately, up to now, none of the examined Vitis spp. and Vitis vinifera L. varieties have been found immune or resistant to the phytoplasma associated with GY (Laimer et al. 2009).
In the last decade, the Georgian grapevine cultivars and minor germplasm have been objects of intense research activities from both a scientific and a vinicultural perspective (Chkhartishvili and Maghradze 2012; Maghradze et al. 2009). The Georgian native germplasm consists of more than 500 cultivars, grown and vinified today only in part (Chkhartishvili and Maghradze 2012), but the best cultivars, such as ‘Saperavi’ (red) and ‘Rkatsiteli’ (white), are well known outside of the country and are cultivated in eastern Europe and central Asia. From the genetic point of view, the Georgian native germplasm demonstrated a very unique genetic pool, distinct from the European (Imazio et al. 2013) and central Asiatic (Bacilieri et al. 2013) ones. This specificity, and some relationships among the Georgian wild (representative in the native flora) and cultivated grapevines, are in agreement with archaeological evidence related to the antiquity and originality of Georgian viniculture and cultivar assortment. These genetic characteristics have a phenotypic correspondence in terms of vine morphology, phenology, and cultural and enological traits, which show a large variability inside the Georgian germplasm associated with a distinctness from neighboring as well as geographically distant grapevine germplasm (Maghradze et al. 2009). Moreover, recent studies reported that grapevine varieties selected in domestication centers of V. vinifera such as Georgia (the country of the South Caucasus) showed possible tolerance or resistance to plant pathogens such as Plasmopara viticola (Berk. & M. A. Curtis) Berl. & De Toni, associated with downy mildew (Bitsadze et al. 2014). For that reason, among others, Georgian native V. vinifera varieties were employed in breeding-based constitution of new varieties (Vakhtangadze et al. 2010). Evidence from a preliminary survey on GY in Georgia highlighted that ‘Ca. P. solani’ infects international and local varieties (Giorgadze 2005; Quaglino et al. 2014).
In the present study, field surveys and multiple gene-typing analyses were carried out to study (i) the GY symptom severity in international and Georgian native varieties and (ii) the genetic diversity among ‘Ca. P. solani’ strain populations in Georgia.
Materials and Methods
Symptom observation and plant sampling.
Surveys on GY symptoms were carried out from 14 to 20 September 2013 in vineyards and V. vinifera germplasm collections in Khaketi and Shida (Inner) Kartli regions in eastern Georgia. Based on the severity of typical GY and GY-like symptoms observed, international and native Georgian grapevine varieties were classified into group I (mild symptoms: color alterations and curling of the leaves; normal production), group II (moderate symptoms: color alterations and curling of the leaves, partial irregular ripening of wood, and mild berry shrivel; partially reduced production), and group III (severe symptoms: berry shrivel, desiccation of inflorescences, color alterations and curling of the leaves, reduction of growth, and irregular ripening of wood; complete loss of production). Leaf samples were collected from 81 symptomatic grapevine plants of four western European or international (‘Chardonnay’, France; ‘Freisa’, Italy; ‘Carignano’ or ‘Carignan’, Spain; and ‘Muscat a Petites Grains Blanc’ or ‘Moscato Bianco’, international) and 37 native Georgian varieties. Moreover, in Mukhrani vineyards (Shida Kartli region), leaf samples were collected from six plants of the wild species C. arvensis, known as one of the main wild plants involved in BNp diffusion (Table 1).
Table 1. Symptom severity and phytoplasmas in Georgian vineyards
Phytoplasma detection based on polymerase chain reaction and restriction fragment length polymorphism analysis of the 16S ribosomal RNA gene.
Total DNA was extracted from examined plants as described by Angelini and colleagues (2001), with some modifications. Briefly, leaf veins and petioles (0.5 g) were separated from the lamina with sterile scalpels, immersed in liquid nitrogen, and ground using sterile pestles and mortars. Prewarmed cetyltrimethylammonium bromide (CTAB)-based buffer (2.5% [wt/vol] CTAB, 100 mM Tris [pH 8.0], 1.4 M NaCl, 50 mM EDTA [pH8], 1% PVP-40, and 0.5% ascorbic acid) was added to the crushed tissues, homogenized by mechanical pestle, and held at 60°C for 20 min. After incubation, DNA was extracted by adding chloroform/isoamyl alcohol (24:1, vol/vol) solution and precipitated by incubation with isopropanol at −20°C for 20 min. A nucleic acid pellet was washed with 70 and 80% ethanol, air dried, suspended in 50 μl of deionized autoclaved water, and maintained at −30°C until use.
Detection of phytoplasmas was carried out by means of amplification of 16S ribosomal DNA (rDNA) in nested polymerase chain reaction (PCR) assays primed by universal primer pairs P1/P7 and R16F2n/R16R2, and subsequent AluI-, BfaI-, BstUI-, and MseI-restriction fragment length polymorphism (RFLP) assays on the obtained amplicons. PCR and RFLP conditions were as previously described (Quaglino et al. 2009). PCR assays were performed by using Taq polymerase (Promega, Milan, Italy) in an automated thermal cycler (MasterCycler Gradient; Eppendorf, Milan, Italy). PCR and enzymatic digestion products were electrophoresed through 1 and 3% agarose gel, respectively, in Tris-borate-EDTA (TBE) buffer, stained with Midori Green Advance (Biosigma, Venice, Italy), and visualized under a UV transilluminator. Total nucleic acids from periwinkle (Catharanthus roseus (L.) G. Don) plants infected by phytoplasma strains elm yellows 1 (EY1, ‘Ca. P. ulmi’, subgroup 16SrV-A), stolbur (STOL, ‘Ca. P. solani’, subgroup 16SrXII-A), and aster yellows 1 (AY1, ‘Ca. P. asteris’, subgroup 16SrI-B) were used as reference controls. Total nucleic acids extracted from healthy periwinkle and PCR mixture devoid of nucleic acids were used as negative controls.
Molecular characterization of ‘Ca. P. solani’ strains through sequence analysis of vmp1 and stamp genes.
BNp strains, identified in the present study by PCR-RFLP-based assays on 16S rDNA, were typed by nested PCR amplification of the gene vmp1 and subsequent molecular characterization by RFLP assays and sequence analyses. Reaction mixtures and PCR conditions were as previously described (Fialová et al. 2009). Phytoplasma reference controls and visualization of PCR products were as described above for 16S rDNA. Amplified PCR products were analyzed by RFLP technique using the enzyme RsaI (Fialová et al. 2009). Enzymatic digestion products were electrophoresed through 3% agarose gel in TBE buffer, stained with Midori Green Advance (Biosigma), and visualized under a UV transilluminator.
The vmp1 gene fragments amplified from 15 BNp strains, representative of the obtained RsaI-RFLP profiles, were selected for nucleotide sequence analysis. The vmp1 gene amplicons were sequenced in both senses by a commercial service (Primm, Milan, Italy) to achieve at least 5× coverage per base position. Nucleotide sequence data were assembled by employing the Contig Assembling program of the software BioEdit version 7.0.5 (http://www.mbio.ncsu.edu/bioedit/bioedit.html) and deposited in the GenBank database (http://www.ncbi.nlm.nih.gov/) under accession numbers shown in Table 2. The vmp1 nucleotide sequences of ‘Ca. P. solani’ strains previously published were retrieved from GenBank (Supplementary Table S1) and used for comparison with the sequences obtained in this study. In detail, nucleotide sequences were compiled in FASTA format and trimmed to the fragment delimited by annealing sites of the primers TYPH10F and TYPH10R (Fialová et al. 2009). To comprehensively compare RsaI-RFLP profiles of BNp strains from Georgia with profiles previously described (Aryan et al. 2014; Cimerman et al. 2009; Cvrković et al. 2014; Kostadinovska et al. 2014; Murolo et al. 2013), trimmed nucleotide sequences were searched for single-nucleotide polymorphisms in recognition sites for the enzyme RsaI by virtual RFLP analyses using the software pDRAW32 (http://www.acaclone.com/). The association between the vmp1-RFLP profiles and BN symptom severity was evaluated by χ2 test using IBM SPSS Statistics V. 22.0 for Windows (IBM Corporation, Armonk, NY).
Table 2. Bois noir phytoplasma types identified in Georgia by nucleotide sequence analyses of the membrane protein coding genes vmp1 and stamp
The stamp gene was amplified from the same 15 BNp strains selected for vmp1 gene sequence characterization. Reaction mixtures and PCR conditions were as previously described (Fabre et al. 2011). Phytoplasma reference controls and visualization of PCR products were as described above for 16S rDNA. The stamp gene amplicons were sequenced, assembled, and deposited in the National Center for Biotechnology Information (NCBI) GenBank database under accession numbers shown in Table 2. Nucleotide sequences of the ‘Ca. P. solani’ stamp gene previously published were retrieved from GenBank and utilized for comparison with the sequences obtained in this study. Nucleotide sequences were compiled in FASTA format and trimmed to the fragment delimited by annealing sites of the primers StampF1 and StampR1 (Fabre et al. 2011).
The vmp1 and stamp gene nucleotide sequences were aligned using the ClustalW Multiple Alignment application and analyzed for sequence identity determination using the Sequence Identity Matrix application of the software BioEdit, version 7.0.5. Based on sequence identities, ‘Ca. P. solani’ strains were grouped into vmp1 and stamp genetic variants, and into collective vmp/stamp types for which nomenclature was designated in the present study (Table 3; Supplementary Tables S2 and S3). Strains of each variant or type shared 100% sequence identity.
Table 3. ‘Candidatus Phytoplasma solani’ vmp/stamp types identified by collective nucleotide sequence analyses of the membrane protein coding genes vmp1 and stamp
Phylogenetic analysis and selective pressure of ‘Ca. P. solani’ from Georgia and other geographical regions.
Nucleotide sequences of vmp1 and stamp genes of BNp strains identified in the present study (Table 2) and of one strain representative of each vmp1 and stamp genetic variant identified among ‘Ca. P. solani’ strains previously described in GenBank were employed for phylogenetic analyses. GenBank accession numbers are shown in Supplementary Table S1 and on the phylogenetic trees. Moreover, vmp1 and stamp gene sequences of ‘Ca. P. solani’ from this and previous studies (Tables 2 and 3) were concatenated with the BioEdit software and used for phylogenetic analyses. In detail, minimum evolution analysis was carried out using the neighbor-joining method and bootstrap was replicated 1,000 times with the software MEGA6 (Tamura et al. 2013).
The ratio between the proportion of nonsynonymous and synonymous substitutions (dN/dS ratio) for each gene was determined for the nucleotide sequences of (i) the overall ‘Ca. P. solani’ strain populations (both strains identified in Georgia in this study and previously identified strains from Euro-Mediterranean areas), (ii) the ‘Ca. P. solani’ strain populations identified in Georgia, and (iii) the ‘Ca. P. solani’ strain populations from Euro-Mediterranean areas. The dN/dS ratios and the null hypothesis of no selection (H0: dN = dS) versus the positive selection hypothesis (H1: dN > dS) were calculated using the Nei-Gojobori method in a codon-based Z-selection test (Nei and Gojobori 1986). The analysis was carried out in MEGA6, and the variance of the differences was computed using the bootstrap method (1,000 replicates; Tamura et al. 2013). The synonymous and nonsynonymous nucleotide substitution rates for the nucleotide sequences of the genes vmp1 and stamp were calculated. Positive selection happens when dN/dS ratio >1.0 and P value for the Z test < 0.05; on the other hand, a ratio < 1.0 suggests a purifying selection process (Nei and Kumar 2000).
Results
Symptoms observed on grapevine in Georgia.
Severe symptoms were observed in 3 international varieties (Chardonnay, Carignano, and Freisa) and 1 local Georgian variety (‘Kisi’) (Fig. 1A and B); moderate symptoms were observed in 4 local Georgian varieties (‘Buera’, ‘Goruli Mtsvane’, Saperavi, and ‘Saperavi Pachkha’) (Fig. 1C and D); and mild symptoms were observed in 22 local Georgian varieties and 1 international variety (Moscato Bianco) (Fig. 1E and F; Table 1). Leaf samples were collected from 25, 19, and 37 plants of varieties showing severe, moderate, and mild symptoms, respectively (Table 1). Leaf samples were also collected from six bindweed plants showing yellowing, reddening, dwarfism, and leaf malformation (Fig. 2).
Fig. 1. Intensity of grapevine yellows symptoms observed in Georgian vineyards. Severe symptoms on cultivars A, Chardonnay and B, Kisi; moderate symptoms on C, Saperavi and D, Goruli Mtsvane; and mild symptoms on E, Rkatsiteli and F, Tsitska.
Fig. 2. Symptoms observed on bindweed (Convolvulus arvensis L.) plants inside the Georgian vineyard located in Mukhrani. Main symptoms are dwarfism, color alteration, and malformation of the leaves.
GY phytoplasma identification.
PCR-based amplification of the 16S rRNA gene showed that 55 of 81 examined grapevines were infected by phytoplasmas. In detail, phytoplasmas were detected in all grapevine plants showing severe symptoms and in 58% (11 of 19 plants) and 51% (19 of 37 plants) of grapevine plants showing moderate and mild symptoms, respectively (Table 1). Moreover, all six bindweed samples have been found infected by phytoplasmas. DNA amplification was obtained from periwinkles infected by phytoplasma reference strains STOL, AY1, and EY1. Reliability of the results was supported by the absence of DNA amplification in the reactions of healthy periwinkle and the negative control (PCR mixture devoid of DNA). AluI-, BfaI-, BstUI-, and MseI-digestion analysis showed that all the phytoplasma strains identified in grapevine and bindweed belonged to the species ‘Ca. P. solani’, taxonomic subgroup 16SrXII-A, because they had restriction patterns indistinguishable from one another and from the patterns characteristic of the STOL (16SrXII-A) reference strain (data not shown).
‘Ca. P. solani’ strain characterization by vmp1 and stamp gene sequence analysis.
The vmp1 gene fragment was amplified from 43 of 55 infected grapevines, and from all six infected bindweeds. In detail, amplification was obtained in 22, 9, and 12 grapevine plants showing severe, moderate, and mild symptoms, respectively (Table 1). Enzymatic digestions of the 49 vmp1 amplicons showed the presence of six RsaI profiles among BNp strains from Georgia. Such profiles were confirmed by virtual RFLP analysis carried out on vmp1 nucleotide sequences amplified from BNp strains representative of each actual RFLP pattern. Virtual RFLP-based comparison of vmp1 RFLP profiles obtained in this study with profiles reported in literature showed that Georgian BNp strains showed previously described (V1, V14, and V15) and new (undescribed [und]1, und2, and und3) restriction patterns (Fig. 3). Strains showing profiles V1, V14, and und2 were prevalent within BNp strain populations in Georgia and were identified with significantly different distribution in grapevine varieties showing severe, moderate, and mild symptoms in both the Khaketi and Shida Kartli regions (χ2 = 16.671, df = 10, P = 0.029; Fig. 4). On the other hand, strains showing profile und1 were found only in Chardonnay showing severe symptoms and in symptomatic bindweed from the Shida Kartli region, while strains showing profiles V15 and und3 (identified in a single grapevine plant) were found only in varieties showing mild symptoms in the Khaketi region (Table 1; Fig. 4).
Fig. 3. Virtual RsaI restriction fragment length polymorphism (RFLP) profiles of vmp1 gene fragments obtained from bois noir attributed to phytoplasma (BNp) strain populations in Georgia. Attribution of vmp1 restriction patterns of Georgian BNp strains (Carv1, Carv2, Char7, Kisi38, Amla77, and Tsol89) was carried out through comparison with virtual RsaI-RFLP profiles of ‘Candidatus Phytoplasma solani’ reference strains (S13, STOL, Aa25, Mca21, 19-25, PO, LG, Mp49, B2035, Mvercer2, Mp46, P7, T2-92, CrHo12_721, Mag1, Vv12_274, and Vv12_I116). Designation of vmp1 RsaI-RFLP profiles is according to the SEE-ERANET nomenclature (X. Foissac, INRA, Bordeaux, France). Profile V-unr (strain Vv12_I116): V = unreported and M = marker Φx174 digested with the enzyme HaeIII.
Fig. 4. Distribution of Georgian bois noir (BN) attributed to phytoplasma (BNp) types, determined by restriction fragment length polymorphism (RFLP) analysis of vmp1 gene amplicons, in grapevine cultivars exhibiting severe, moderate, and mild symptoms. Asterisk (*) indicates that the counts of plants showing different BN symptoms within the vmp1-RFLP profile are different at a P < 0.1 (χ2 test).
Nucleotide sequence analysis, performed on 15 BNp strains from Georgia selected on the basis of the diversity found through RFLP analysis of vmp1 gene amplicons, revealed the presence of 12 and 7 genetic variants of vmp1 (here designated as VmGe1 to VmGe12) and stamp (here designated as StGe1 to StGe7) genes, respectively (Table 2). Average sequence identity among genetic variants was 92.3 and 97.8% for vmp1 and stamp genes, respectively. Eleven Georgian BNp vmp/stamp types were described as the combination of vmp1 and stamp genetic variants (Table 2). Two of such types (VmGe8/StGe3 and VmGe12/StGe7) were identified more frequently (5 of 15 BNp strains); the other types were found in single plant samples.
Sequence identity calculation among nucleotide sequences of vmp1 and stamp genes of ‘Ca. P. solani’ strains retrieved from the NCBI GenBank highlighted the presence of 63 vmp1 genetic variants (here designated as Vm1 to Vm63) of 105 analyzed sequences and 35 stamp genetic variants (here designated as St1 to St35) of 89 analyzed sequences. Average sequence identity among genetic variants was 90.5 and 95.5% for vmp1 and stamp genes, respectively. Collective analysis of vmp1 and stamp nucleotide sequences, available in the NCBI GenBank for 58 ‘Ca. P. solani’ strains, allowed the identification of 38 vmp/stamp types, here designated as the combination of vmp1 and stamp genetic variants (Table 3). Comparison with representative strains of each variant from GenBank nucleotide sequences showed that BNp strains from Georgia showed 11 vmp1 (VmGe1 to VmGe7 and VmGe9 to VmGe12) and 6 stamp (StGe1 to StGe6) novel genetic variants, previously unreported. Only BNp phytoplasma strains Tsol89 and Kiqu94 (VmGe12/StGe7) shared 100% sequence identity with vmp1 and stamp genes of ‘Ca. P. solani’ strain P7 (Vm53/St15; Table 3), identified in periwinkle in Lebanon (Cimerman et al. 2009). Nucleotide sequences of one BNp strain from Georgia, representative of each vmp1 and stamp genetic variant, were deposited at the NCBI GenBank (accession numbers shown in Table 2).
Phylogenetic analysis and selective pressure of ‘Ca. P. solani’ strains from Georgia and other geographical regions.
The vmp1 gene phylogenetic tree showed that BNp strains identified in the present study grouped within four of the eight main clusters identified. The majority of BNp strains from Georgia (8 of 15), belonging to genetic variants VmGe3, VmGe7, VmGe8, VmGe9, and VmGe12, grouped within the cluster vmp1-5 along with genetic variants Vm53 (identical to VmGe12), Vm56, and Vm57 (Fig. 5). The stamp gene phylogenetic tree showed that BNp strains identified in the present study grouped within three of the four main clusters identified. The majority of BNp strains from Georgia (12 of 14), belonging to genetic variants StGe1, StGe2, StGe3, StGe4, and StGe7, clustered within the cluster stamp-4 along with genetic variants St3, St4, St15 (identical to StGe7), and St28 (Fig. 6).
Fig. 5. Unrooted phylogenetic tree inferred from ‘Candidatus Phytoplasma solani’ strain nucleotide sequences of gene vmp1. Minimum evolution analysis was carried out using the neighbor-joining method and bootstrap replicated 1,000 times. Names of phytoplasma strains included in phylogenetic analysis are written on the tree image. GenBank accession number of each sequence is given in parenthesis; gene sequences obtained in the present study are indicated in bold. Clusters are shown as delimitated by parentheses. Acronyms within clusters indicated phytoplasma hosts and origin. Hosts: Ar, Anaceratagallia ribauti; Can, Capsicum annuum; Car, Convolvulus arvensis; Cr, Catharanthus roseus; Ho, Hyalesthes obsoletus; Lv, Linaria vulgaris; Pa, Prunus avium; Rp, Reptalus panzeri; Rq, R. quinquecostatus; Sl, Solanum lycopersicum; Ud, Urtica dioica; and Vv, Vitis vinifera. Origin: AU, Austria; FR, France; GEO, Georgia; GER, Germany; IT, Italy; LE, Lebanon; MK, Makedonia; SE, Serbia; and SL, Slovenia.
Fig. 6. Unrooted phylogenetic tree inferred from ‘Candidatus Phytoplasma solani’ strain nucleotide sequences of gene stamp. Minimum evolution analysis was carried out using the neighbor-joining method and bootstrap replicated 1,000 times. Names of phytoplasma strains included in phylogenetic analysis are written on the tree image. GenBank accession number of each sequence is given in parenthesis; gene sequences obtained in the present study are indicated in bold. Clusters are shown as delimitated by parentheses. Acronyms within clusters indicated phytoplasma hosts and origin. Hosts: Ar, Anaceratagallia ribauti; Can, Capsicum annuum; Car, Convolvulus arvensis; Cr, Catharanthus roseus; Ho, Hyalesthes obsoletus; La, Lavandula angustifolia; Lv, Linaria vulgaris; Pa, Prunus avium; Rp, Reptalus panzeri; Rq, R. quinquecostatus; Sl, Solanum lycopersicum; Ud, Urtica dioica; Vv, and Vitis vinifera. Origin: AU, Austria; BU, Bulgaria; CR, Croatia; FR, France; GEO, Georgia; GER, Germany; GR, Greece; IT, Italy; LE, Lebanon; MK, Makedonia; SE, Serbia; and SL, Slovenia.
Based on phylogenetic analysis of concatenated nucleotide sequences of the genes vmp1 and stamp, five vmp/stamp clusters were identified. The cluster vmp/stamp-4 included ‘Ca. P. solani’ strains associated with the U. dioica-related biological cycle, while the other four clusters (vmp/stamp-1, -2, -3, and -5) included ‘Ca. P. solani’ strains associated with the Convolvulus arvensis-related biological cycle. The majority of Georgian BNp strains (13 of 14) grouped within Convolvulus arvensis-related clusters vmp/stamp-3 and vmp/stamp-5, whereas only strain Amla77 grouped within the U. dioica-related cluster vmp/stamp-4 (Fig. 7). In detail, Georgian BNp strains identified in grapevine and bindweed, grouped within cluster vmp/stamp-3, were found to be closely related to strain P7, previously identified in a naturally infected periwinkle plant (Catharanthus roseus) in Lebanon (Fig. 7).
Fig. 7. Unrooted phylogenetic tree inferred from ‘Candidatus Phytoplasma solani’ strains based on concatenated nucleotide sequences of the genes vmp1 and stamp. Minimum evolution analysis was carried out using the neighbor-joining method and bootstrap replicated 1,000 times. Names of phytoplasma strains, included in phylogenetic analysis, and their vmp/stamp types are written on the tree image. gene sequences obtained in the present study are indicated in bold characters. Clusters are shown as delimitated by parentheses. Acronyms within clusters indicated phytoplasma hosts and origin. Hosts: Ar, Anaceratagallia ribauti; Can, Capsicum annuum; Car, Convolvulus arvensis; Cr, Catharanthus roseus; Ho, Hyalesthes obsoletus; La, Lavandula angustifolia; Lv, Linaria vulgaris; Pa, Prunus avium; Rp, Reptalus panzeri; Rq, R. quinquecostatus; Sl, Solanum lycopersicum; Ud, Urtica dioica; Vv, and Vitis vinifera. Origin: AU, Austria; BU, Bulgaria; CR, Croatia; FR, France; GEO, Georgia; GER, Germany; GR, Greece; IT, Italy; LE, Lebanon; MK, Makedonia; SE, Serbia; and SL, Slovenia.
The genetic variability and selective pressure in the vmp1 and stamp genes were estimated for the ‘Ca. P. solani’ strains according to the abundance of nonsynonymous mutations. The overall ratio between the nonsynonymous and synonymous mutations (dN/dS) was >1.0 for vmp1 (dN/dS = 4.567, P = 0.000) and stamp (dN/dS = 2.436, P = 0.008) (Table 4). These high values of dN/dS (i.e., >1) indicated detection of a high number of nonsilent (dN) mutations. For the stamp gene, less intensive positive selective pressure (dN/dS) was found within ‘Ca. P. solani’ strains from Euro-Mediterranean populations (dN/dS = 2.711, P = 0.004) in comparison with BNp strains from Georgia (dN/dS = 2.953, P = 0.002). An opposite trend was seen for vmp1 with respect to stamp, where the dN/dS ratio decreased from 4.637 (P = 0.000, Euro-Mediterranean population) to 3.618 (P = 0.000, Georgia) (Table 4).
Table 4. Codon-based test of positive selection for the analysis of nucleotide sequences of the genes vmp1 and stamp within ‘Candidatus Phytoplasma solani’ strain populations
Discussion
The results obtained in this study confirmed the evidence from previous research indicating the presence of BN disease of grapevine in Georgia (Quaglino et al. 2014). In fact, molecular analyses showed a strong association between specific GY disease symptoms and grapevine plant infection by BNp (‘Ca. P. solani’ strains) within the examined vineyards. On the other hand, the presence of PCR-negative plants of grapevine cultivars showing moderate and mild symptoms could be connected with the low titer or sporadic distribution of phytoplasmas in symptomatic plant tissues (Constable et al. 2003), or with the uncertain attribution of unclear symptoms to GY. Moreover, RFLP analysis of 16S rDNA amplicons excluded the presence of phytoplasmas genetically distinct from ‘Ca. P. solani’ associated with other diseases (i.e., Flavescence dorée) of the GY complex (Belli et al. 2010; Laimer et al. 2009).
Surprisingly, in contrast with what was reported from GY surveys in European countries (Belli et al. 2010; Kosovac et al. in press; Mori et al. 2015), in the present study, BNp-infected bindweed in Georgian vineyards was found showing typical symptoms of phytoplasma infection. Interestingly, asymptomatic bindweed from the same vineyards was negative when tested for PCR-based BNp detection (data not shown). This result could suggest the possibility of monitoring symptoms on bindweed, naturally present in Georgia, as a reporter of the presence of ‘Ca. P. solani’ in vineyards where grapevines are not showing typical GY symptoms (i.e., tolerant varieties or plantlets for propagation materials), improving the possibility of containing the disease by preventing movement of symptomless infected material.
In recent years, multiple gene analysis was proposed and employed to describe phytoplasma species distinguished by evident molecular diversity and representing ecologically separated populations. Moreover, this approach was also applied to the investigation of the genetic diversity among phytoplasmas associated with several diseases in order to identify strain-specific molecular markers useful for improving the understanding of complex phytoplasma ecologies (Davis et al. 2013; Quaglino et al. 2013). In order to gain an insight into the genetic diversity among BNp strains in Georgia, nucleotide sequence analysis was performed on two genes (vmp1 and stamp) coding for membrane proteins putatively involved in the recognition and interaction of BNp with its hosts (Cimerman et al. 2009; Fabre et al. 2011). In the present study, such genes were selected because they have higher sequence variability than the tuf gene, classically used to distinguish two genetically divergent BNp strain types (tuf-a and tuf-b) involved in two diverse epidemiological cycles of BN (Langer and Maixner 2004); thus, they are more useful than the tuf gene to accurately describe the genetic diversity among BNp strain populations (Aryan et al. 2014; Cvrković et al. 2014; Kosovac et al. in press; Kostadinovska et al. 2014).
Based on RsaI-RFLP digestions of vmp1 gene amplicons, the profiles V1, V14, and the previously unreported und2 were prevalent among the analyzed BNp strains (Table 1). These data confirmed the specific association of pattern V14 with eastern Europe (Foissac et al. 2013), and highlighted an unexpected diffusion of type V1, reported as the prevalent type in bindweed- and nettle-related BN host systems in Italy, France, and Germany (Foissac et al. 2013) in the Caucasian geographic regions. This evidence, along with the prevalence of type V1 in the international cultivar Chardonnay, could suggest the nonindigenous origin of this type, possibly introduced in Georgia through import of planting material from central or western Europe.
As largely reported for phytoplasma-associated diseases of stone fruit trees (i.e., apple proliferation and European stone fruit yellows), symptom intensity observed in infected plants can be influenced by both the virulence of the pathogen and the susceptibility level of the plant host (Kison and Seemüller 2001; Seemüller and Schneider 2007; Seemüller et al. 2011). In the case of phytoplasma diseases of grapevine, several studies have investigated the susceptibility of V. vinifera cultivars with different approaches (Margaria et al. 2014; Roggia et al. 2014) but genetic diversity and virulence of phytoplasma strains were not considered accurately. In fact, it is reasonable to hypothesize that genetically distinct BNp strains also could show a variable range of virulence. In Georgia, severe symptoms, typically associated with international cultivars (i.e., Chardonnay) in Europe, have been observed on the same cultivars and in the local variety Kisi. Interestingly, the majority of autochthonous Georgian grapevine cultivars were found to be mildly symptomatic, maintaining complete berry production (Table 1; Fig. 1). Intriguingly, the presence of BNp strains showing RsaI-RFLP profiles V1, V14, and und2 of the gene vmp1 (Table 1; Fig. 4) in grapevine cultivars exhibiting severe, moderate, and mild symptoms suggested a different susceptibility of the cultivars to these BNp strains. On the other hand, the mutually exclusive presence of BNp strains showing RsaI-RFLP profiles V15 and und3 in cultivars showing mild symptoms and RsaI-RFLP profile und1 in cultivars showing severe symptoms suggested the hypo- and hypervirulence, respectively, of such BNp strains. The diverse susceptibility of Georgian grapevine cultivars to BNp strains highlighted in the present work needs to be confirmed in further research.
Molecular characterization carried out by nucleotide sequence analysis of vmp1 and stamp genes showed that BNp populations in Georgia consist mainly of previously unreported strains. In fact, only Georgian BNp strains Tsol89 and Kiqu94 shared 100% sequence identity with the sequences of the ‘Ca. P. solani’ strain P7 (vmp/stamp type Vm53/St15), identified in naturally infected periwinkle in Lebanon in 2001 (Cimerman et al. 2009).
Phylogenetic analysis showed that vmp1 genetic variants identified in this and previous studies within the same RsaI-RFLP-based types (V1) can be grouped in separate clusters distinctly associated with nettle- and bindweed-related BN host systems (Fig. 5), confirming that vmp1 genetic variants of the same RsaI-RFLP-based type can be associated with different BN host systems (Murolo and Romanazzi 2015; Kosovac et al. in press; Kostadinovska et al. 2014). Moreover, clustering of several ‘Ca. P. solani’ strains was different in phylogenetic trees built on the basis of vmp1 and stamp genes, decreasing the chance to obtain a comprehensive data interpretation. To improve the significance of sequence analyses, we performed phylogenetic analysis of the concatenated nucleotide sequences of vmp1 and stamp genes, as reported in other studies focused on phytoplasmas (Durante et al. 2012).
Phylogenetic analysis of vmp1 and stamp gene concatenated sequences revealed that the majority of BNp Georgian strains (7 of 14) identified in grapevine cultivars and bindweed grouped along with the Lebanese strain P7 within the cluster vmp/stamp-3 (Fig. 7). Interestingly, this cluster is clearly distinct from other vmp/stamp clusters, including bindweed- and nettle-related BN strains previously identified in central and southern Europe (Aryan et al. 2014; Cvrković et al. 2014; Kosovac et al. in press; Murolo and Romanazzi 2015). Other Georgian BNp strains from grapevine and bindweed were associated with the cluster vmp/stamp-5, including bindweed-related BNp strains previously identified in grapevine in Serbia and Italy (Cvrković et al. 2014; Murolo and Romanazzi 2015). Only one BNp strain (Amla77) from Georgia grouped with nettle-related cluster vmp/stamp-4. Such evidence suggested the coexistence of the BN host systems related to nettle and bindweed in Georgia, with a high prevalence of bindweed as the main host plant of ‘Ca. P. solani’ strains associated with BN.
As shown by Murolo and Romanazzi (2015), the overall ratio between the nonsynonymous to synonymous mutations showed that vmp1 and stamp genes in ‘Ca. P. solani’ strains in Georgia are under a positive selection process but the values of dN/dS ratio indicated a more intensive selection for the gene vmp1 (3.62) compared with stamp (2.95). It should be interesting to investigate more accurately this aspect by comparing the dN/dS ratio values among BNp strain populations identified in different geographic regions and in diverse hosts (grapevine, insect vectors, and weeds) in Georgia.
In conclusion, results from the present study showed that BNp strain populations in Georgia consist mainly of new, unreported ‘Ca. P. solani’ strains associated with both nettle- and bindweed-related BN host systems. Moreover, the distribution of BNp strains among grapevine cultivars showing a variable range of symptom intensity suggests a different susceptibility of such local cultivars to BN disease. Further studies are in progress to evaluate this important topic from the perspective of improving breeding programs for the production of novel grapevine cultivars tolerant of or resistant to phytoplasma diseases.
Acknowledgments
This study was supported by the Cost action FA1003-GRAPENET, East-West Collaboration for Grapevine Diversity Exploration and Mobilization of Adaptive Traits for Breeding.
Literature Cited
- 2001. Flavescence dorée in France and Italy: Occurrence of closely related phytoplasma isolates and their near relationships to palatinate grapevine yellows and an alder yellows phytoplasma. Vitis 40:79-86. ISI, Google Scholar
- 2014. An abundant ‘Candidatus Phytoplasma solani’ tuf b strain is associated with grapevine, stinging nettle and Hyalesthes obsoletus. Eur. J. Plant Pathol. 140:213-227. https://doi.org/10.1007/s10658-014-0455-0 Crossref, ISI, Google Scholar
- 2015. The molecular epidemiology of bois noir grapevine yellows caused by ‘Candidatus Phytoplasma solani’ in the Republic of Macedonia. Eur. J. Plant Pathol. 142:759-770. https://doi.org/10.1007/s10658-015-0649-0 Crossref, ISI, Google Scholar
- 2013. Genetic structure in cultivated grapevines is linked to geography and human selection. BMC Plant Biol. 13:25. https://doi.org/10.1186/1471-2229-13-25 Crossref, ISI, Google Scholar
- 2010. Grapevine yellows: Past, present and future. J. Plant Pathol. 92:303-326. ISI, Google Scholar
- 2011. Conventional and novel strategies for the phytoplasma diseases containment. Phytopathogenic Mollicutes 1:77-82. https://doi.org/10.5958/j.2249-4669.1.2.013 Crossref, Google Scholar
- 2014. Screening of the Georgian grape germplasm to susceptibility of downy mildew: Preliminary results. Acta Hortic. 1032:191-196. https://doi.org/10.17660/ActaHortic.2014.1032.25 Crossref, Google Scholar
- 2012. Viticulture and winemaking in Georgia. Vitis Special Issue:169-176. Google Scholar
- 2009. Striking diversity of vmp1, a variable gene encoding a putative membrane protein of the stolbur phytoplasma. Appl. Environ. Microbiol. 75:2951-2957. https://doi.org/10.1128/AEM.02613-08 Crossref, ISI, Google Scholar
- 2003. Seasonal distribution of phytoplasmas in Australian grapevines. Plant Pathol. 52:267-276. https://doi.org/10.1046/j.1365-3059.2003.00849.x Crossref, ISI, Google Scholar
- 2014. Experimental and molecular evidence of Reptalus panzeri as a natural vector of bois noir. Plant Pathol. 63:42-53. https://doi.org/10.1111/ppa.12080 Crossref, ISI, Google Scholar
- 2013. ‘Candidatus Phytoplasma pruni’, a novel taxon associated with X-disease of stone fruits, Prunus spp.: Multilocus characterization based on 16S rRNA, secY, and ribosomal protein genes. Int. J. Syst. Evol. Microbiol. 63:766-776. https://doi.org/10.1099/ijs.0.041202-0 Crossref, ISI, Google Scholar
- 2012. Sequence analyses of S10-spc operon among 16SrV group phytoplasmas: Phylogenetic relationships and identification of discriminating single nucleotide polymorphisms. Ann. Appl. Biol. 161:234-246. https://doi.org/10.1111/j.1744-7348.2012.00568.x Crossref, ISI, Google Scholar
- 2011. The stolbur phytoplasma antigenic membrane protein gene stamp is submitted to diversifying positive selection. Gene 472:37-41. https://doi.org/10.1016/j.gene.2010.10.012 Crossref, ISI, Google Scholar
- 2009. Genetic variability of stolbur phytoplasma in annual crop and wild plant species in South Moravia (Czech Republic). J. Plant Pathol. 91:411-416. ISI, Google Scholar
- , and
STOLBUR-EUROMED Consortium . 2013. ‘Candidatus Phytoplasma solani’ genome project and genetic diversity in the Euro-Mediterranean basin. Pages 11-13 in: Proc. 3rd Eur. Bois Noir Workshop, Barcelona, Spain. E. Torres, A. Laviña, and A. Batlle, eds. Google Scholar - 2005. New disease of grapevine in Georgia. Zashchita I Karantin Rastenii 4:37. Google Scholar
- 2013. From the cradle of grapevine domestication: Molecular overview and description of Georgian grapevine (Vitis vinifera L.) germplasm. Tree Genet. Genomes 9:641-658. https://doi.org/10.1007/s11295-013-0597-9 Crossref, ISI, Google Scholar
IRPCM Phytoplasma/Spiroplasma Working Team–Phytoplasma Taxonomy Group . 2004. ‘Candidatus Phytoplasma’, a taxon for the wall-less, non-helical prokaryotes that colonize plant phloem and insects. Int. J. Syst. Evol. Microbiol. 54:1243-1255. https://doi.org/10.1099/ijs.0.02854-0 Crossref, ISI, Google Scholar- 2001. Differences in strain virulence of the European stone fruit yellows phytoplasma and susceptibility of stone fruit trees on various rootstocks to this pathogen. J. Phytopathol. 149:533-541. https://doi.org/10.1046/j.1439-0434.2001.00671.x Crossref, ISI, Google Scholar
- Molecular tracing of the transmission routes of bois noir in Mediterranean vineyards of Montenegro and experimental evidence for the epidemiological role of Vitex agnus-castus (Lamiaceae) and associated Hyalesthes obsoletus (Cixiidae). Plant Pathol. In press. doi:10.1111/ppa.12409 Google Scholar
- 2014. Multiple gene analyses identify distinct “bois noir” phytoplasma genotypes in the Republic of Macedonia. Phytopathol. Mediterr. 53:491-501. ISI, Google Scholar
- 2009. Resistance to viruses, phytoplasmas and their vectors in the grapevine in Europe: A review. J. Plant Pathol. 91:7-23. ISI, Google Scholar
- 2004. Molecular characterisation of grapevine yellows associated phytoplasmas of the stolbur-group based on RFLP-analysis of non-ribosomal DNA. Vitis 43:191-200. ISI, Google Scholar
- 2009. Genetic and phenetic exploration of Georgian grapevine germplasm. Acta Hortic. 827:107-114. https://doi.org/10.17660/ActaHortic.2009.827.14 Crossref, Google Scholar
- 2014. Metabolic and transcript analysis of the flavonoid pathway in diseased and recovered Nebbiolo and Barbera grapevines (Vitis vinifera L.) following infection by Flavescence doree phytoplasma. Plant Cell Environ. 37:2183-2200. https://doi.org/10.1111/pce.12332 Crossref, ISI, Google Scholar
- 2012. Correlation of bois noir disease with nettle and vector abundance in northern Italy vineyards. J. Pest Sci. 85:23-28. https://doi.org/10.1007/s10340-011-0375-0 Crossref, ISI, Google Scholar
- 2015. Investigation on ‘bois noir’ epidemiology in north-eastern Italian vineyards through a multidisciplinary approach. Ann. Appl. Biol. 166:75-89. https://doi.org/10.1111/aab.12165 Crossref, ISI, Google Scholar
- 2013. Genetic variability of the stolbur phytoplasma vmp1 gene in grapevines, bindweeds and vegetables (corrigendum to vol. 109, pg 2049, 2010). J. Appl. Microbiol. 115:631-633. Crossref, ISI, Google Scholar
- 2015. In-vineyard population structure of ‘Candidatus Phytoplasma solani’ using multilocus sequence typing analysis. Infect. Genet. Evol. 31:221-230. https://doi.org/10.1016/j.meegid.2015.01.028 Crossref, ISI, Google Scholar
- 1986. Simple methods for estimating the numbers of synonymous and nonsynonymous nucleotide substitutions. Mol. Biol. Evol. 3:418-426. ISI, Google Scholar
- 2000. Molecular Evolution and Phylogenetics. Oxford University Press, New York. Google Scholar
- 2008. Ability of Reptalus quinquecostatus (Hemiptera: Cixiidae) to inoculate stolbur phytoplasma to artificial feeding medium. Ann. Appl. Biol. 153:299-305. https://doi.org/10.1111/j.1744-7348.2008.00261.x Crossref, Google Scholar
- 2014. First report of ‘Candidatus Phytoplasma solani’ and ‘Candidatus Phytoplasma convolvuli’ associated with grapevine bois noir and bindweed yellows, respectively, in Georgia. Plant Dis. 98:1151. https://doi.org/10.1094/PDIS-01-14-0026-PDN Link, ISI, Google Scholar
- 2009. New 16Sr subgroups and distinct single nucleotide polymorphism lineages among grapevine Bois noir phytoplasma populations. Ann. Appl. Biol. 154:279-289. https://doi.org/10.1111/j.1744-7348.2008.00294.x Crossref, ISI, Google Scholar
- 2013. ‘Candidatus Phytoplasma solani’, a novel taxon associated with stolbur and bois noir related diseases of plants. Int. J. Syst. Evol. Microbiol. 63:2879-2894. https://doi.org/10.1099/ijs.0.044750-0 Crossref, ISI, Google Scholar
- 2008. Transmission of a stolbur phytoplasma by the Agalliinae leafhopper Anaceratagallia ribauti (Hemiptera, Auchenorrhyncha, Cicadellidae). J. Phytopathol. 156:687-690. https://doi.org/10.1111/j.1439-0434.2008.01416.x Crossref, ISI, Google Scholar
- 2014. Flavescence dorée phytoplasma titre in field-infected Barbera and Nebbiolo grapevines. Plant Pathol. 63:31-41. https://doi.org/10.1111/ppa.12068 Crossref, ISI, Google Scholar
- 2011. HfIB gene-based phytopathogenic classification of ‘Candidatus Phytoplasma mali’ strains and evidence that strain composition determines virulence in multiply infected apple trees. Mol. Plant-Microbe Interact. 24:1258-1266. https://doi.org/10.1094/MPMI-05-11-0126 Link, ISI, Google Scholar
- 2007. Differences in virulence and genomic features of strains of ‘Candidatus Phytoplasma mali’, the apple proliferation agent. Phytopathology 97:964-970. https://doi.org/10.1094/PHYTO-97-8-0964 Link, ISI, Google Scholar
- 2005. Vitex agnus-castus is a preferred host plant for Hyalesthes obsoletus. J. Chem. Ecol. 31:1051-1063. https://doi.org/10.1007/s10886-005-4247-z Crossref, ISI, Google Scholar
- 2013. MEGA6: Molecular Evolutionary Genetics Analysis Version 6.0. Mol. Biol. Evol. 30:2725-2729. https://doi.org/10.1093/molbev/mst197 Crossref, ISI, Google Scholar
- 2010. Participation of Georgian native varieties of grapevine in breeding. B. Acad. Agric. Sci. Ga. 27:186-192. Google Scholar
GenBank accession numbers of DNA sequences: KT184867 to KT184885.
