Multilocus Genotyping of 'Candidatus Phytoplasma Solani' Associated with Grapevine Bois Noir in Iran.

Grapevine Bois noir (BN) is associated with 'Candidatus Phytoplasma solani'. It has been recorded in vineyards throughout Europe as well as in different countries in Asia, where it now constitutes a threat to Iranian viticulture. BN is strictly dependent on 'Ca. P. solani' strains, wild host plants, and insect vectors. The molecular typing of 'Ca. P. solani', based on the nonribosomal gene tuf and the two hypervariable markers vmp1 and stamp, is valuable for the reconstruction and clarification of the pathways of BN spread. In this study, an RFLP analysis was performed on the vmp1 gene, and a single-nucleotide polymorphism analysis confirmed new vmp types in 'Ca. P. solani'. A stamp gene phylogenetic analysis allowed us to distinguish between the new genotype infections in the grapevines and the 'weeds' Convolvulus arvensis and Erigeron bonariensis in Iranian vineyards, highlighting the close genetic relatedness of the strains of 'Ca. P. solani' found in Iran and Azerbaijan. The most common genotype in the grapevines was tuf b/V24/stamp III, which was associated with C. arvensis. This information contributes toward the identification of further routes of introduction of 'Ca. P. solani' in Iran to sustain the control measures for the management of BN.

1. Introduction

Grapevine Bois noir (BN) is a ‘Candidatus Phytoplasma solani’ (‘Ca. P. solani’)-associated disease that causes significant crop losses [1,2,3,4]. BN has been recorded in vineyards throughout Europe as well as in different countries in Asia (e.g., Israel, Jordan, Lebanon, Syria, Iran, Azerbaijan, Turkey, Georgia, Russia, and China) [5,6,7,8,9,10,11,12,13,14]. The epidemiology of ‘Ca. P. solani’ is highly correlated with the host plant reservoirs, from which the insect vectors (mainly belonging to the Cixiidae family such as Hyalesthes obsoletus [15] and Reptalus panzeri [16]) acquire a phytoplasma inoculum from feeding.

Over the past 20 years or so, several studies have been published regarding potential vectors [17,18,19,20,21]. Recently, Quaglino et al. [22] demonstrated that nine new insect vectors could transmit ‘Ca. P. solani’, which has raised several questions regarding the biological cycle of ‘Ca. P. solani’.

Multilocus sequence typing (MLST) analyses of nonribosomal genes have been widely applied to molecular epidemiology [23,24,25,26]. The molecular typing of ‘Ca. P. solani’ isolates associated with BN is important to clarify the pathways of BN spread [27,28,29,30]. The study of the molecular epidemiology of ‘Ca. P. solani’ isolates considers the genetic diversity of the different molecular typing markers. Among these, the most widely used are the tuf gene, which encodes the translation elongation factor TU [27]; the stamp gene, which encodes the antigenic membrane protein AMP [30]; and the vmp1 gene, which encodes a variable membrane protein [31,32].

According to the tuf gene characterization, ‘Ca. P. solani’ isolates are grouped into two main genotypes, tuf-a and tuf-b, which were shown to be related to the different natural epidemiological cycles of ‘Ca. P. solani’ in western Germany [24]. Isolates of the tuf-a type were spread from the stinging nettle (Urtica dioica), but isolates of the tuf-b type were disseminated from bindweed (Convolvulus arvensis). A third minor genotype, tuf-c, was shown to propagate from the bindweed Calystegia sepium [26,27]. Recently, different tuf-b sub-genotypes (molecular variants) have been proposed to be involved in an epidemiological cycle of BN [29,33,34]. Sequence analyses of vmp1 and stamp, which are variable genes, have provided evidence of a high variability among the BN strains in terms of the tuf types [35,36]. Based on RsaI RFLP and sequencing analyses of partial vmp1 genes, 25 vmp molecular types of ‘Ca. P. solani’ were described in vineyard ecosystems around the world [35]. Based on the phylogenetic analyses of the concatenated nucleotide sequences of the genes vmp1 and stamp for 76 ‘Ca. P. solani’ strains, 49 vmp1/stamp sequence variants were then grouped into 5 vmp1/stamp clusters [37]. The vmp1/stamp IV cluster included ‘Ca. P. solani’ (tuf type a) and was associated with the nettle-related biological cycle, whereas the other four clusters, that included ‘Ca. P. solani’ (tuf type b), were associated with the bindweed-related biological cycle. Recent studies and continuous surveys from vineyards in Iran have highlighted a significant increase in BN and, consequently, considerable yield losses based on production quantity [38]. To select effective control strategies, comprehensive investigations into BN to clarify the ambiguous aspects of the disease and the epidemiological cycle are critical. Due to the complexity of BN, the numerous weeds associated with vineyards, and the lack of detailed investigations into the weeds in vineyards, we performed the molecular typing of ‘Ca. P. solani’ isolates associated with grapevines and the most prevalent weeds in vineyards in this study.

2. Materials and Methods

2.1. Sample Collection and DNA Extraction

From August to October 2016, 2017, and 2018, 225 BN-symptomatic grapevine leaf samples and 100 symptomatic and asymptomatic weeds (Convolvulus arvensis and Erigeron bonariensis) were collected (the detailed database is listed in Supplementary Material Table S1) and the molecular detection and further multilocus sequence typing of ‘Ca. P. solani’ isolates were undertaken (Table 1). The grapevine and herbaceous samples were collected from the seven main grapevine-growing provinces in Iran—Azarbaijan Gharbi, Azarbaijan Sharghi, Zanjan, Qazvin, Fars, Lorestan, and Khorasan Razavi—from the same areas reported by Jamshidi et al. [39]. Asymptomatic plant materials were also collected as negative controls. The total DNA was extracted from the petioles of the grapevine leaf samples and the roots of the weeds using the CTAB DNA extraction procedure [39].

Table 1

‘Candidatus Phytoplasma solani’ nucleotide sequences used in the phylogenetic analysis of the vmp1 and stamp genes from the investigated vineyards.

Isolate	Host	Origin	Accession Number
			vmp1	stamp
a8	Vitis vinifera	Azarbaijan Gharbi	OM920494
AG5	Vitis vinifera	Azarbaijan Gharbi	OM920495
AG15	Vitis vinifera	Azarbaijan Gharbi	OM920496	OM791207
DG12	Vitis vinifera	Qazvin	MK547295	OM791210
DG23	Vitis vinifera	Zanjan	OM920497	OM791211
C4	Vitis vinifera	Zanjan	MK547294	OM791208
C6	Vitis vinifera	Qazvin	OM920498	OM791209
L3	Vitis vinifera	Lorestan	MK547300	OM791218
KG11	Vitis vinifera	Khorasan Razavi	MK547299	OM791217
FZ10	Vitis vinifera	Fars	MK547297	OM791215
FZ11	Vitis vinifera	Fars	MK547298	OM791216
DW3	Convolvulus arvensis	Qazvin	MK547296	OM791212
DW5	Convolvulus arvensis	Qazvin	OM920499	OM791213
DW12	Erigeron bonariensis	Zanjan	OM920500	OM791214

2.2. Molecular Typing and Phylogeny of ‘Candidatus Phytoplasma solani’ Strains

A preliminary analysis was performed on the 225 collected samples of grapevine and the 100 collected samples of weeds using the universal P1/P7 primer pair followed by a nested PCR with rStol/fStol-specific primers (Table S2) [40]. The positive samples were subjected to further molecular typing based on tuf, vmp1, and stamp hypervariable genes. The tuf gene was amplified using the STOLTUF-F0/STOLTUF-R0 primer pair for a direct PCR [33] and the TufAYf/r primers for a nested PCR following Langer and Maixner [27]. For vmp1, amplification was performed with the primer pair StolH10F2/R2 [16], followed by the primer pair TYPH10F/R [41]. A PCR-RFLP analysis was performed with RsaI. The digested fragments were visualized on 2.5% agarose gels. The stamp gene was amplified by a nested PCR with the primer pair StampF/R0 followed by StampF1/R1, as per Fabre et al. [42]. The amplified tuf, vmp1, and stamp genes were sequenced in both directions by a genomics sequencing service (Genewiz UK, Takeley, UK; https://www.genewiz.com/ accessed on 22 May 2022). These Iranian vmp1 and stamp sequences were compared with the sequences available in the database. For the vmp1 sequences, a double-check was performed using a virtual RFLP analysis by applying the pDRAW32 software (http://www.acaclone.com/ accessed on 22 May 2022), which performs a restriction analysis based on the nucleotide sequences. For both vmp1 and stamp, the nucleotide sequences were used for the phylogenetic analysis. Multiple sequence alignments were identified using CLUSTAL-W software. Using MEGA7 [43], we calculated the phylogenetic relationships. The maximum parsimony trees were obtained with a Subtree–Pruning–Regrafting (SPR) algorithm [44]; the trees were drawn to scale with branch lengths calculated using the average pathway method.

3. Results

Molecular Typing of ‘Candidatus Phytoplasma solani’ Strains Based on the Tuf, vmp1, and stamp Genes

From the preliminary analysis performed on 16Sr DNA using the fStol/rStol primer pair in the nested PCR, 218 grapevine and weed samples were defined as infected by ‘Ca. P. solani’. According to the tuf gene analysis, most of the grapevines and weeds were tuf type b1. Strains a8 (tuf b5) and DG23 (tuf b6) [34] were included in the current study for the molecular typing analysis. The positive samples then underwent a PCR with STOLH10F2/R2 followed by TYPH10F/R primers. Polymorphic amplicons that ranged from 1100 bp to 1250 bp were obtained from approximately 142 grapevine and 27 weed samples, which corresponded with 65% of the samples analyzed. A restriction digestion with the RsaI enzyme allowed the identification of eight vmp1 profiles of ‘Ca. P. solani’ infecting the grapevines, C. arvensis, and E. bonariensis. Based on the vmp1 typing, the already known types V1, V3, V4, V10, V15, and V20 were found in the Iranian vineyards sampled. Two new additional vmp restriction patterns, named V24 and V27, were proposed according to the SEE-ERANET database [35]. These profiles were determined from the actual RFLP analysis on the amplified vmp1 gene and confirmed by a virtual RFLP analysis on the vmp1 nucleotide sequences (Figure S1). A vmp1 gene single-nucleotide polymorphism analysis confirmed the new V24 types for strains a8, AG5, and DG23, which only had two repeated domains. The vmp1 sequence of strain AG15 had an additional RsaI site at position 621 and three repeated domains and constituted the new V27 RFLP type. The phylogenetic relationship clustered the strains in this study with the ‘Ca. P. solani’ reference strains. Strains C4, DW3, DW12, and KG11 were defined as vmp1 type V1; C6 and DW5 were defined as vmp1 type V3. Sequences with RFLP type V24—namely, the a8, AG5, and DG23 strains—were grouped in the phylogenetic tree. The vmp1 sequence of strain AG15 did not match any reference strain (Figure 1).

Figure 1

Phylogenetic tree inferred from the ‘Candidatus Phytoplasma solani’ strain nucleotide sequences of the vmp1 gene. The evolutionary history was inferred using the maximum parsimony method. The MP tree was obtained using the Subtree–Pruning–Regrafting (SPR) algorithm. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (500 replicates) are shown next to the branches. The names of phytoplasma strains included in the phylogenetic analysis are given in the tree image. The GenBank accession numbers of the sequences are given; the gene sequences obtained in our investigation are indicated by filled circles.

A stamp gene phylogenetic analysis distinguished six ‘stamp clusters’. Four of these have already been reported and belong to clusters I, II, III, and IV [30,42], but two additional molecular clusters were exclusively represented by these Iranian and Azerbaijan ‘Ca. P. solani’ strains in different clusters. This was shown by a maximum parsimony phylogenetic analysis. The six stamp genotypes detected in the grapevines (KG11, DW3, C4, AG15, L3, and DG23) belonged to stamp cluster III. DG12 was unique, with a CAAAAAGAAGCT deletion at position 373 and a CAAAAAGAAGCT deletion at position 361. The DW5, DW12, and C6 genotypes also had a CAAAAAGAAGCT deletion at position 373 and an ACC insertion at position 256 (Figure 2).

Figure 2

Phylogenetic analysis of the stamp sequences using maximum parsimony. Iranian ‘Candidatus Phytoplasma solani’ isolates are indicated in red. The most parsimonious tree is shown with the bootstrap values indicated along the branches. Insertions and deletions (see Key, bottom right) for the isolates or phylogenetic branches are indicated at the top of the isolate names or along the branches. The stamp genetic clusters are those described in Fabre et al. [42].

4. Discussion

Bois noir is one of the most important and dangerous grapevine diseases in the growing areas of Europe and Asia [41]. BN represents a threat to Iranian viticulture, with outbreaks now recorded in different regions in Iran [39]. A recent study defined vmp types in grapevines and reservoir plants and identified the tuf-b type in Iranian grapevines [34,39]. The epidemiological cycles have not yet been fully elucidated for Iranian vineyard ecosystems.

Considering the complexity of the disease as well as different insect vectors, numerous different weeds as reservoir sources in vineyards, and the vast viticultural areas in the country, we collected and analyzed different grapevine cultivars that showed BN symptoms for this study, along with the most prevalent weeds in the corresponding vineyards. The complex interactions that ‘Ca. P. solani’ has with grapevine varieties, weeds, and the vectors generate a wide genetic diversity in the population, which can be evaluated by an MLST analysis [45]. An MLST analysis was applied to several ‘Ca. P. solani’ studies; it was found to be useful to study and monitor the epidemiology of ‘Ca. P. solani’ [33,36,46,47,48,49].

Our investigations were undertaken based on vmp1 and stamp multilocus sequence typing. These genes are linked to the biological aspects of ‘Ca. P. solani’ because the VMP and AMP proteins are both involved in the interactions between the phytoplasma and their vectors [50,51,52]. These two genes, therefore, have important roles in understanding the ecology of this pathogen; mutations in their sequences are strongly dependent on geographical distribution and the host range, which might be driven by their interactions with local vector(s) [30]. The diversity of ‘Ca. P. solani’ strains in Iranian vineyards was determined through the study of the variability of three genes linked to the epidemiology: tuf, vmp1, and stamp. Phylogenetic analyses of vpm1 and stamp highlighted that the genotypes within the clusters were related to the locations, providing useful information on the molecular epidemiology and the tracking of the ‘Ca. P. solani’ strain distribution.

According to the vmp1 typing, we distinguished six different profiles already reported in previous studies plus two new types—namely, V24 and V27—which have become prevalent in Iranian vineyards. A stamp gene phylogenetic analysis allowed the placement of most of the ‘Ca. P. solani’ strains in cluster II, which included samples from Central Europe and the Balkans (Germany, Czech Republic, Hungary, Croatia, and Bulgaria) [16,35], and cluster III, linking with Azerbaijan, Georgia, Lebanon, Serbia, Bosnia, and Herzegovina, and Austria strains. This confirmed that the ecology of ‘Ca. P. solani’ strains in Iran was associated with a C. arvensis reservoir. This was the case for the DW3 isolate, which belonged to cluster III. The grouping of the DW5 (C. arvensis), DW12 (E. bonariensis), and C6 (grapevine) strains in a different position related to cluster IV and FZ10, FZ11, and DG12 separately from the defined clusters considering their vmp1 type and tuf type might indicate the presence of interconnections between the cycles and/or that they might be linked to genetic recombination between the stamp clusters in Iranian vineyards.

Useful data were obtained from the genotyping of ‘Ca. P. solani’ in the viticultural regions of Iran, supporting the monitoring of the spread of this pathogen. The most common genotype on the grapevines was tuf-b/V24/stamp III, which is associated with C. arvensis. The important role of C. arvensis in the transfer of the tuf-b/V24/stamp III genotype to grapevines in vineyards was also identified in our investigation. The high genetic diversity among ‘Ca. P. solani’ strains indicated that the spread of BN is not through vegetative propagation, highlighting the important role of vectors in the spread of BN as well as the wild hosts as a source of infection. Considering the placement of DW5 and DW12 as wild hosts different from grapevine-associated ‘Ca. P. solani’ might indicate that further investigations are needed into different reservoir plants in Iranian vineyards. Tracing ‘Ca. P. solani’ in insect vectors and distinguishing possible further routes of its introduction into the grapevine ecosystem is required.

5. Conclusions

‘Ca. P. solani’ typing is crucial in grapevine and reservoir plants to clarify the epidemiological cycle(s) of BN, reflecting the important role of insect vectors in disease dispersal. Our work provides information on the molecular variants of ‘Ca. P. solani’ in Iranian vineyards; six known vmp1 variants were found as well as two novel variants, V24 and V27. This information is useful for future investigations to more accurately understand the epidemiological cycle(s) of BN in Iranian vineyards, contributing to the management of the disease.

6. Patents

No patents are related to this work.