Epidemiological Role of Dictyophara europaea (Hemiptera: Dictyopharidae) in the Transmission of ‘ Candidatus Phytoplasma solani’

: Bois noir, an economically important disease of grapevine yellows that causes signiﬁcant economic losses in wine production, is associated with ‘ Candidatus Phytoplasma solani’ and transmitted to grapevines by cixiids Hyalesthes obsoletus and Reptalus panzeri . Polyphagous planthopper Dictyophara europaea , commonly found in natural habitats, harbors phytoplasmas from distinct groups and is an alternative vector in the open epidemiological cycles of the Flavescence dor é e phytoplasma in grapevine in European vineyards. This study addresses the role of D. europaea in the transmission cycle(s) of ‘ Ca . P. solani’ among wild habitats, natural reservoir plants, and the vineyard agroecosystem using MLSA and transmission trials with naturally infected adults to grapevine and Catharanthus roseus . The infection rates of D. europaea ranged from 7% to 13% in diverse locations, while reservoir herbaceous plants were infected in the amount of 29%. A total of 13 CaPsol MLSA genotypes were detected in D. europaea (7) and plants (8). Nine of them corresponded to previously identiﬁed genotypes. Two new genotypes were found in D. europaea (tuf-b1/S1/V14/Rqg50-sv1 and tuf-b1/S18/V14/Rqg50-sv1) and one in Convolvulus arvensis (tuf-b1/S1/V2-TA/Rqg31-sv1), whereas one was shared by two hosts, Crepis foetida and Daucus carota (tuf-b1/S1/V2-TA/STOL-sv1). Naturally infected D. europaea successfully transmitted the tuf-b1/S1/V2-TA/STOL type to ﬁve grapevines and six periwinkles, tuf-b1/S1/V2-TA/Rqg31 to one grapevine, and tuf-b1/S1/V2-TA/Rqg50 to one periwinkle, indicating that D. europaea is an intermediate vector in CaPsol epidemiological cycles.


Introduction
'Candidatus Phytoplasma solani' ('Ca. Phytoplasma solani'; CaPsol) [1], formerly known as stolbur phytoplasma, belongs to the 16S rRNA XII-A subgroup. It is endemic to Europe and the Mediterranean region, inhabiting a diverse host range of herbaceous and woody plants, and is the causative agent of diseases on a variety of cultivated plants, including solanaceous crops, grapevine, maize, and lavender [2][3][4][5][6][7]. Bois noir (BN), a disease belonging to the grapevine yellows (GY) complex, induced by 'Ca. Phytoplasma solani' is widespread in central Europe, the Mediterranean region, and the Middle East. It is known to cause severe damages in vineyards, with infection rates reaching 50-80 percent in some areas and significant economic losses in wine production [8,9]. The molecular typing of CaPsol strains associated with various hosts (plant or insect) or of different geographic origins is largely used in research studies of BN epidemiology, including disease spread, outbreaks, and increases in the incidence [6,[10][11][12][13]. The multilocus sequence analysis (MLSA) approach, based on (i) two housekeeping genes, tuf and secY, and (ii) two variable membrane proteins, which are under selection pressure and are presumably involved in host recognition, vmp1 and stamp [2,6,14,15], is used for the characterization of CaPsol in epidemiological studies [7,[10][11][12]16,17].
Dictyophara europaea (Linnaeus, 1767) (Hemiptera: Dictyopharidae), also known as the European lantern fly, is a polyphagous planthopper species found throughout the Western Palaearctic and is commonly associated with a variety of plant species and ecosystems [21]. It is frequently found, particularly in the late summer dry season, in association with drought-tolerant deciduous plant species such as Clematis vitalba L., which commonly occupy vineyard ecosystems [22]. This planthopper is known to harbor a number of different phytoplasmas from phylogenetically distant 16S rRNA groups, including 16SrII, 16SrV, and 16SrXII [23][24][25] and is a proven vector of Flavescence dorée (FD; 16SrV-C subgroup) transmission from natural reservoir C. vitalba to grapevine in southeastern Europe [23,26]. Moreover, a recent study in northern Italy that included the experimental testing of eight alternative CaPsol vectors in grapevine suggested that D. europaea could transmit CaPsol to grapevine [27]; however, its epidemiological role remains unknown. Considering all the above-mentioned findings, the primary goal of our study was to determine the role of this dictyopharid planthopper in the 'Ca. P. solani' transmission cycle(s) among wild habitats, natural reservoir plants, and vineyard agroecosystem and to provide insight into the influence of these associations in BN dynamics within the vineyard.
This was investigated through a survey on the occurrence and infection incidence of D. europaea and tentative reservoir plants in natural habitats and vineyard surroundings, as well as by combining molecular epidemiology data based on (MLSA) multilocus sequence analysis on four genes (tuf, secY, vmp1, and stamp) of 'Ca. P. solani' isolates and the experimental transmission of the pathogen by D. europaea.

Study Sites
During a four-year period, from 2016 to 2019, a field survey was carried out in a total of eight different locations in central, eastern, and south-eastern Serbia ( Table 1). Three of the selected locations were in the surroundings of vineyard agroecosystems previously found to be affected by grapevine yellows disease [22,24]. The remaining five locations had a variety of habitat types, including stony and semi-mesic meadows, abandoned pastures, and xerothermic ruderal slopes ( Table 1). The composition of the vegetation was distinct to each location, with a wide variety of plant covers and an intricate spatial layout.  1 The number preceding the location name corresponds to the number denoting the location in Figure 1. 2 Genotype designations are in accordance with previously published nomenclature for the tuf gene [17,28], the secY gene [7,28], and vmp1 RFLP profiles [6,11], while stamp genotypes are indicated by their corresponding reference strains [11,15] or as SNP variants (sv) when they include a single-nucleotide mutation. The number of isolates at a location with the same MLSA genotype, if more than one, is indicated in brackets next to the stamp genotype.

'Ca. Phytoplasma solani' Genotypes Associated with Dictyophara europaea and Reservoir Plants
All 'Ca. P. solani' isolates identified in D. europaea, reservoir plants, and experimental test plants were characterized through the multilocus typing of four genes: tuf, secY, vmp1, and stamp. The sequence analyses of the tuf gene region of CaPsol strains identified in D. europaea from eight locations revealed that all were infected with the tuf-b1 genotype ( Table 1). Among plant isolates, tuf-b1 was also the most prevalent gen-

Survey of D. europaea and Reservoir Plants
Insect collections were carried out from mid-July to mid-September from 2016 to 2018, in accordance with previously observed data on adult flight periods [22]. Due to the adults' polyphagous feeding behavior, specimens were collected using entomological sweep nets and mouth aspirators from dominant broadleaf weedy plants in each locality. Collected specimens were placed in 2 mL plastic vials (Sarstedt) filled with 96% ethanol and transported to a laboratory in mobile refrigerators at 10 • C, where they were identified at the species level using taxonomic keys [21,29]. The insects were stored in ethanol at −20 • C after identification until DNA extraction.
In all studied areas, no plants expressed symptoms of phytoplasma infection. The most abundant plant species from the four selected locations were sampled in early September to determine their status as 'Ca. P. solani' reservoir plants and inoculum sources ( Table 2). Each plant's leaf samples were stored in a separate plastic bag and transported in the field refrigerator at 10 • C. Each sample was processed for DNA extraction upon arrival at the laboratory and then stored in a freezer at −20 • C for further analysis.

DNA Extraction from Insects, Herbaceous Hosts, and Experimental Plants
Total DNA was extracted from 357 individual insect specimens using a previously reported nondestructive extraction method [13]. Under a stereomicroscope Leica MZ7.5 (Leica Microsystems, Wetzlar, Germany), each D. europaea specimen was perforated between the hind legs and mesothorax and put into 2 mL tubes (Sarstedt, Nümbrecht, Germany) containing 800 µL of extraction buffer (20 mM TRIS, 10 mM EDTA, 0.5% SDS) and 8 µL of proteinase K. After overnight incubation at 56 • C, the specimens were removed and stored in 96% ethanol, and DNA was extracted from the solution using a chloroformisopropanol procedure. In the final extraction stage, isolated DNA was dissolved in 150 µL of TE buffer (10 mM TRIS, 1 mM EDTA; pH 7.6).
Using a previously reported CTAB procedure [30], total DNA was extracted from the leaf samples of field-collected plants and of experimental plants (grapevines and periwinkles). The DNA pellets were resuspended in 100 µL of TE buffer. The DNA extracted from insects and plants was stored at −20 • C until further analysis.

'Ca. P. solani' Identification
The initial identification of 'Ca. P. solani' in field-collected D. europaea adults was conducted using the 'Ca. P. solani'-specific Stol11 nested PCR protocol with the Stol11F2/R1 and Stol11F3/R2 primer pair [31,32], following a previously published protocol [33]. The same 'Ca. P. solani' detection protocol was utilized for all field samples and experimental plants from the transmission trials. DNA extracts of 'Ca. P. solani'-infected grapevine and insects from Serbia [11,12] were used as positive controls in all amplification reactions.

Experimental Transmission of 'Ca. P. solani' with Field-Collected Dictyophara europaea
To determine whether D. europaea naturally harboring CaPsol could act as a natural vector within an ecosystem and given that its vector role for FD phytoplasma has previously been established [23,26], we undertook transmission trials with CaPsol-infected specimens. Experiments were carried out with adults from two locations where D. europaea populations were dense and naturally carried phytoplasma at a rate of around 10%: (i) Topola, in the grape-growing region of central Serbia, with a rate of CaPsol-infected specimens of 9%, and (ii) Rogljevo in eastern Serbia, with a rate of CaPsol-infected specimens of 11% (Table 1). In the middle of August 2018, a total of 576 D. europaea adults were collected using entomological nets and mouth aspirators in Topola and Rogljevo.
The experimental plants were kept in the growth chamber under the same conditions as during the IAP and monitored for symptom development. Ten months later, the leaves of all test and control plants were sampled for the identification of 'Ca. P. solani' using the Stol11 protocol.

Multilocus Sequence Analysis (MLSA) of 'Ca. P. solani' Isolates
Field-collected weeds and insects, and experimentally infected grapevines and periwinkles were subjected to molecular typing to assess the epidemiological relevance of 'Ca. P. solani' isolates associated with D. europaea. Isolates were characterized by sequence and/or RFLP analysis of the following four phytoplasma genomic loci: (i) the two housekeeping genes-tuf, which encodes translation elongation factor Tu, and secY, encoding the preprotein translocase subunit SecY-and (ii) two membrane protein genes-antigenic membrane protein gene stamp and variable membrane protein vmp1 gene of 'Ca. P. solani'.
The tuf gene of 'Ca. P. solani' was amplified using nested PCR with primer Tuf1f/r, followed by TufAYf/r, as described previously [2].
The secY gene was amplified using nested PCR with primers PosecF1 and PosecR1, followed by PosecF3 and PosecR3 [34], while the stamp gene was amplified using the StampF/R0 and StampF1/R1 primer pairs [15]. The nested PCR amplicons were sequenced in both directions with the primers used for tuf gene amplification, reverse primer PosecR3 for secY, and forward primer StampF1 for the stamp gene. The sequencing was performed by Macrogen Inc., and the sequences were deposited in the GenBank database under accession numbers ON804249-ON804271, ON804278-ON804304, and ON804313-ON804341 for the stamp, tuf, and secY genes, respectively (Table S1). The sequences were compared to those previously published [11,17,20,35,36] using the BLASTn algorithm (https://blast.ncbi.nlm.nih.gov, accessed on 26 February 2022).
Using NETWORK software (Network 10.2.0.0, Forster P.&M., Colchester, England, www.fluxus-engineering.com, accessed on 18 May 2022), a phylogenetic network was constructed to assess the evolutionary relatedness and resolve the genealogy of the stamp genotypes of all 'Ca. P. solani' types involved in the epidemiological cycles in the studied locations. The network was constructed using the strains found in this study and the Balkan reference strains [11,16,20,35,37,38]. Gene genealogies were inferred by performing median joining (MJ) calculations [39] with parameter e = 0 and maximum parsimony (MP) post-processing to obtain a network containing all the shortest trees.
A fragment of the vmp1 gene encoding a putative 'Ca. P. solani' membrane protein was amplified using the StolH10F1/R1 primer pair [14], followed by the TYPH10F/R inner primer pair [34]. The nested products of all characterized isolates were digested with the RsaI restriction enzyme, and isolates expressing the V2 profile were also digested with TaqI and AluI [11,20]. The digested products were separated using automated capillary electrophoresis QIAxcel Advanced System (Qiagen) according to conditions described previously [13], and restriction profiles were compared with the reference controls of the V2-TA, V3, V4, V7-A, V14 and V17 vmp1 profiles [11,20].
Primer pairs that were used for amplification and sequencing of each analyzed gene region are listed in Supplementary Table S2.

'Ca. Phytoplasma solani' Identification in Dictyophara europaea and Natural Reservoir Plants
The survey of D. europaea adults and reservoir plants naturally harboring 'Ca. P. solani' was undertaken in eight sites, three of which were located in the grape-growing regions of central and eastern Serbia. Over the course of four years, specimens of D. europaea and weeds were collected from mid-July to mid-September (from 2016 to 2019). There were 357 specimens of D. europaea collected from the various regions that were studied, and 35 (10%) of them tested positive for CaPsol. To be more specific, infection rates ranged from 7% to 13% in seven diverse localities, while none of the individuals collected in Dobanovci (in the Srem region) tested positive for CaPsol (Table 1).
In vineyard region of Topola in central Serbia (Table 1 and Figure 1, location 2), Salvia pratensis L. (Lamiaceae) was the most abundant plant, evenly covering the surface of the meadow, intermixed with field bindweed (Convolvulus arvensis L. (Convolvulacea)) and climbing shrub (Clematis vitalba L. (Ranunculaceae)) at the edges of the meadow. In this location, the majority of D. europaea specimens were obtained from S. pratensis, while the incidence of insect infection was 9% (Table 1). In total, 2 of 20 S. pratensis and 8 of 20 C. arvensis carried CaPsol. In contrast, none of the six examined C. vitalba was infected with CaPsol ( Table 2).
The vegetation in the grape-growing region of Šumadija, at the location of Krnjevo (Table 1 and Figure 1, location 3), consisted of various plant species, with nearly equal proportions of grasses and deciduous plants. Only ten D. europaea were collected as a result of survey at this location, one of which tested positive for CaPsol infection (Table 1).
In the abandoned pasture on Danube River at Donji Milanovac (Table 1 and Figure 1, location 4), C. arvensis intermixed with various grass plants covered most of the surface [40]. CaPsol infection was detected in four of twelve asymptomatic bindweeds, two of six Sorghum halepense and Cichorium intybus, and one of six Daucus carota and Elymus repens ( Table 2). The infection rate of D. europaea in this location was 12% (Table 1).
In a moderate hillside meadow at location Brza Palanka (Table 1 and Figure 1, location 5), near Danube River in eastern Serbia, we collected 12 D. europaea, one of which was infected with CaPsol ( Table 1).
The ruderal hillside at the vineyard region of Rogljevo in eastern Serbia (Table 1 and Figure 1, location 6) had plant coverage that consisted of C. foetida and P. hieraciodes as dominant plants, followed by various Asteraceae, Poaceae, and Apiaceae plants. Climbing shrub C. vitalba was found along the meadow's edges. The D. europaea population was dense, with 105 specimens collected in 2016-2018, of which 12 (11%) were positive for CaPsol infection (Table 1). CaPsol was detected in 12 of 20 C. foetida, 7 of 16 D. carota, and 6 of 16 P. hieraciodes. CaPsol was also found in two of six S. halepense, and C. intybus, and one of six E. repens. None of the analyzed C. vitalba, Achillea millefolium, and Falcaria vulgaris tested positive for CaPsol (Table 2).
In the Jasenovik village, near the town of Niš in southern Serbia, specimens of D. europaea were collected from the stony meadow (Table 1 and Figure 1, location 7), where Crepis foetida L. and Picris hieracioides L. (Asteraceae) covered approximately 30% of the surface. Other members of the Asteraceae family were also abundant in this location, as well as a variety of legume plant species. CaPsol was found in 10 of the 18 C. foetida and 8 of the 20 P. hieraciodes plants sampled. Among the other reservoir plants collected, Cynodon dactylon, Daucus carota, and Erigeron annuus tested positive for CaPsol at rates of 13-17%, while Carlina vulgaris and Senecio erucifolius were not infected ( Table 2). We collected 48 D. europaea at this location, 5 of which (10%) carried CaPsol (Table 1).
At a xerothermic ruderal slope near the town of Vlasotince, also in eastern Serbia (Table 1 and Figure 1, location 8), we collected 25 D. europaea, 2 of which tested positive for CaPsol infection (Table 1).
Finally, none of the 15 D. europaea specimens collected in the semi-mesic meadow at the location of Dobanovci in Vojvodina (northern Serbia, location 1), tested positive for CaPsol infection (Table 1).

'Ca. Phytoplasma solani' Genotypes Associated with Dictyophara europaea and Reservoir Plants
All 'Ca. P. solani' isolates identified in D. europaea, reservoir plants, and experimental test plants were characterized through the multilocus typing of four genes: tuf, secY, vmp1, and stamp. The sequence analyses of the tuf gene region of CaPsol strains identified in D. europaea from eight locations revealed that all were infected with the tuf-b1 genotype ( Table 1). Among plant isolates, tuf-b1 was also the most prevalent genotype, detected in 93% of infected plants. Beside tuf-b1, at the location of Rogljevo, ruderal hillside in a traditional vine growing district of east Serbia, the tuf-d type was detected in 25% of infected C. foetida and 28% of infected D. carota (Table 2). This type was recently identified in sugar beet, parsley, and parsnip from northern Serbia [38].
The secY genotyping revealed the presence of three genotypes in D. europaea and reservoir plants (Tables 1 and 2). The most prevalent was the S1 genotype (STOL; AM992086) [15,28], which is common in and endemic to Europe and is associated with various crops and reservoir plants [1,7,11,15,17]. This genotype was detected in all D. europaea isolates from six locations (except Vlasotince) and was prevalent (96%) in reservoir plants from all natural habitats. Two CaPsol-positive D. europaea collected at the xerothermic ruderal slope near Vlasotince carried the secY S18 genotype corresponding to isolate LOZA2 (FN813282) [7,15], which was previously identified in grapevine sampled in eastern Serbia in 2006. In 25% of C. arvensis CaPsol-infected plants from vineyards region in the locality of Topola and in 25% of CaPsol-infected C. arvensis from the abandoned pasture of Donji Milanovac, the S4 genotype was detected (GGY; AM992093) [15,28]. This genotype is also endemic to Europe and is associated with various crops and reservoir plants.
The vmp1 RFLP typing using RsaI, TaqI, and AluI endonucleases resulted in the identification of two different profiles, namely, V2-TA, and V14, both previously identified and designated as such [6,11] (Tables 1 and 2). Most of the CaPsol isolates of D. europaea (86%) and reservoir plants (89%) belonged to the V2-TA genotype. This profile was detected in plants and insects from all surveyed locations. Beside V2-TA, profile V14 was detected in one of eleven D. europaea from Topola, two of three D. europaea from Donji Milanovac, one of two D. europaea from Vlasotince, and in one specimen collected in Brza Palanka, as well as in five of eight and three of twelve infected bindweeds from Topola and Donji Milanovac, respectively.
The sequence analysis of the stamp gene marker resulted in the identification of seven different CaPsol genotypes among all D. europaea and reservoir plant isolates (Tables 1 and 2; Figure 2). Four genotypes (STOL, Rqg50, Rqg31, and Rpm35) were identical to previously recorded reference strains [11,15]. The STOL genotype was the most abundant and present in D. europaea and the weeds from four locations. In Krnjevo, Rogljevo, and Jasenovik, this was the only genotype present in D. europaea, while in the vineyard region of Topola, this genotype was detected in 72% of infected insects. The STOL genotype was found in all reservoir plants with the exception of Convolvulus arvensis, as expected, given that this plant harbors specific 'Ca. P. solani' genotypes due to its association with Hyalesthes obsoletus as a primary vector, driving the pathogen transmission and host-plant-associated epidemiology [12,20,36,40]. Genotype Rqg50 was found in CaPsol-infected C. arvensis and D. europaea isolates from Topola and Donji Milanovac. All isolates associated with C. arvensis and D. europaea from Donji Milanovac carried this genotype, as well as 50% of C. arvensis and 18% of D. europaea isolates from Topola. This genotype was also identified in one of two CaPsol-positive D. europaea from Vlasotince. Genotype Rqg31 was sporadically found in isolates of D. europaea and bindweed from Topola, while the Rpm35 type was found only in one isolate of C. arvensis from the same location. The new stamp genotype (Rqg50-sv1), an SNP variant of Rqg50, was detected in a single specimen of D. europaea from Vlasotince and Brza Palanka. One of eight CaPsol-positive C. arvensis isolates from Topola harbored new stamp genotype Rqg31-sv1, which is an SNP variant of Rqg31. One of twelve CaPsol-positive C. foetida and one of seven CaPsol-positive D. carota from Rogljevo had a unique stamp genotype (STOL-sv1) representing an SNP variant of the STOL genotype.  [11,16,20,35,37,38]. Genotypes PO and StolC originating from France [15] were used for stabilization of the network. Genotypes are represented by circles and paired genotypes are connected by a line. Circle sizes are proportional to the strain frequency found in this study and colors correspond to the associated host as shown in the figure legend. Each line connecting the circles is one mutational difference. For more than one mutational difference between strains, numbers of SNP mutations are presented in brackets; black dot vertices represent median vectors (mv). Gray circles represent reference strains not detected in this study. Strain and vector abbreviations are indicated. Newly detected strains designated in this study are in bold.
All isolates carrying the tuf-d genotype were affiliated to the stamp gene STOL genotype, while in other 'Ca. Pytoplasma solani' isolates of the same host plant the tuf-b1 type of the tuf gene occurred syntopically, i.e., coexisting on location.

Transmission Trials
To determine whether D. europaea could be a natural vector of 'Ca. P. solani' and to elucidate its epidemiological role, transmission trials using naturally infected adults were performed with specimens collected in Topola and Rogljevo, where CaPsol was

Rqg50-sv1
Ho-Vac1  Figure 2. Median-joining network inferred from stamp genotypes of 'Candidatus Phytoplasma solani' strains detected in naturally infected Dictyophara europaea and reservoir plants in this study and those previously identified in the Balkans [11,16,20,35,37,38]. Genotypes PO and StolC originating from France [15] were used for stabilization of the network. Genotypes are represented by circles and paired genotypes are connected by a line. Circle sizes are proportional to the strain frequency found in this study and colors correspond to the associated host as shown in the figure legend. Each line connecting the circles is one mutational difference. For more than one mutational difference between strains, numbers of SNP mutations are presented in brackets; black dot vertices represent median vectors (mv). Gray circles represent reference strains not detected in this study. Strain and vector abbreviations are indicated. Newly detected strains designated in this study are in bold.
All isolates carrying the tuf-d genotype were affiliated to the stamp gene STOL genotype, while in other 'Ca. Pytoplasma solani' isolates of the same host plant the tuf-b1 type of the tuf gene occurred syntopically, i.e., coexisting on location.

Transmission Trials
To determine whether D. europaea could be a natural vector of 'Ca. P. solani' and to elucidate its epidemiological role, transmission trials using naturally infected adults were performed with specimens collected in Topola and Rogljevo, where CaPsol was previously detected in 11 out of 116 (9%) and in 12 out of 105 (11%) individuals, respectively ( Table 1).
The survival rate of insects was above 90% after 48 h on both grapevine seedlings and periwinkles. Five weeks after inoculation, four C. roseus test plants exposed to D. europaea from Rogljevo and three C. roseus inoculated by D. europaea from Topola developed symptoms of phytoplasma infection in the form of intense leaf yellowing and virescence.
Approximately four months after the experiments ended, grapevine seedlings exposed to both populations of D. europaea developed the first symptoms of phytoplasma infection.
Symptoms appeared as interveinal discoloration and lamina reddening in three grapevine seedlings inoculated with D. europaea from Rogljevo and three grapevine seedlings inoculated with D. europaea from Topola. The subsequent analysis of plant DNA using the Stol11 protocol confirmed that all grapevine and periwinkle plants with symptoms were infected with CaPsol. The 18 symptomless grapevine plants and 17 periwinkles tested negative for phytoplasma presence (Table 3).
All CaPsol-positive grapevine and C. roseus isolates were subjected to genotyping using MLSA. Overall, three different 'Ca. P. solani' genotypes based on tuf/secY/vmp1/stamp typing were found in the experimental plants. All three grapevines and four periwinkles infected by D. europaea individuals from Rogljevo harbored the tuf-b1/S1/V2-TA/STOL type. The D. europaea population from Topola transmitted the same genotype to two grapevine seedlings and two periwinkles. Moreover, individuals from Topola also transmitted the tuf-b1/S1/V2-TA/Rqg31 genotype to single grapevine and tuf-b1/S1/V2-TA/Rqg50 to one periwinkle (Table 3).

Discussion
The European lantern fly (D. europaea) is a common polyphagous phloem-feeding planthopper found throughout the Western Palaearctic, with records of its range expanding to Northwestern China [41]. It is a univoltine species frequently found on ruderal, xerothermic, and sunny hillsides in association with dicotyledonous herbs, grasses, and shrubs [21,29]. The long period of adult activity in ecologically diverse ecosystems, which lasts from the end of June to the beginning of October (reviewed in [22]), combined with this planthopper's polyphagous feeding behavior, enables it to come into contact with various phytoplasmas invading diverse plant species.
Our study utilized a combination of molecular and experimental approaches to identify and link relevant constituents in the epidemiological transmission routes of CaPsol, clarifying the epidemiological role of this planthopper in the spread of BN in Serbian vineyards. The multilocus typing of four epidemiologically informative genes, tuf, secY, vmp1, and stamp [2,6,14,15], revealed the complexity of CaPsol epidemiology, as indicated by the number of detected pathogen genotypes and the number of hosts shared by the same genotype.
The most frequent CaPsol comprehensive genotype was tuf-b1/S1/V2-TA/STOL. It was the only genotype found in the isolates of D. europaea as well as potential reservoir plants, including C. foetida, P. hieracioides, S. pratensis, E. annuus, D. carota, C. intybus, S. halepense, and E. repens, in the localities of Rogljevo and Jasenovik. This genotype is regarded as the historical reference strain for CaPsol, and it originates from pepper in Serbia [1]. It is the most common and widespread genotype across the Balkan peninsula, and it has previously been linked to three distinct epidemiological cycles of the tuf-b1 CaPsol strain: (i) the R. panzeri-vectored disease cycle, causing maize redness (MR) disease in maize, BN disease in grapevine, and 'stolbur' disease in potato [4,11,36,42]; (ii) the 'Ca. P. solani' disease cycle driven by H. obsoletus associated with C. foetida in BN-affected vineyards in eastern Serbia [12]; and (iii) the 'Ca. P. solani' disease cycle driven by H. obsoletus from Vitex agnus-castus in the coastal zone of Montenegro [20].
Aside from the dominant STOL genotype, two other stamp genotypes of epidemiological importance in the Balkans, Rqg50 and Rqg31, were detected in D. europaea and asymptomatic C. arvensis in areas where field bindweed was the dominant reservoir plant. Overall, six MLSA genotypes were found in D. europaea and C. arvensis from these locations (Topola and Donji Milanovac), one of which (tuf-b1/S1/V14/Rqg50) was shared by both D. europaea and field bindweeds. The Rqg50 and Rqg31 stamp genotypes have previously been proven to be vectored to grapevine and potato plants by H. obsoletus associated with C. arvensis [12,36] and have repeatedly been found in C. arvensis, its reservoir plant [12,20,40].
Regarding the tuf gene variability, in the location of Rogljevo in eastern Serbia, 3 of 12 C. foetida and 2 of 7 D. carota positive for CaPsol were found to be infected with the tuf-d type. This finding represents the first detection of the tuf-d genotype in plants of the natural environment and, even more, in asymptomatic plants which could serve as a reservoir for this genotype. The tuf-d type has previously been explicitly linked to the epidemic outbreaks of rubbery taproot disease in sugar beet (RTD) in northern Serbia, Croatia, Hungary, and Slovakia [38], but without connections to the source plants of the outbreaks. Since in our study this tuf type was found in tentative reservoir plants alongside tuf-b1 and was always associated with the STOL genotype, its epidemiological implications are uncertain. Although more research is needed, it appears that tuf-d is an SNP variant of tuf-b with shared ecological links with host plants.
Vector feeding habits and mobility, as well as the synchrony of host plant and vector phenology, determine whether the relationship between vector and the plant source of phytoplasma infection is well balanced or whether one plant serves as the predominant source for phytoplasma acquisition [43]. The seasonal occurrence and temporal dynamics of D. europaea in natural habitats in Serbia are well documented [22]. Due to the high mobility of adults, in late summer, when herbaceous vegetation is drying, they aggregate on the plants that are better adapted to drought conditions (e.g., S. pratensis) or migrate to shrubs growing close to the meadows (e.g., C. vitalba) [22]. Based on the MLSA matching of CaPsol in D. europaea and reservoir host plants, it could be assumed that deciduous plants susceptible to dry conditions, such as C. foetida, P. hieracioides, and S. pratensis, together with C. arvensis, the major and most widely distributed natural host of CaPsol, are the main reservoirs for CaPsol acquisition throughout the season.
Furthermore, given that C. vitalba has previously been found infected with CaPsol in Italy [44] and that D. europaea aggregates on C. vitalba in the late summer, acquiring FD phytoplasma and transmitting it to grapevine [23,26], it should be expected that C. vitalba could act as a reservoir plant for CaPsol, also in Serbia. Interestingly, none of the twelve analyzed plants from two localities where D. europaea populations were the most abundant was found to be infected with this phytoplasma. In the vineyards of Tuscany, C. vitalba and other weedy plants were found to be CaPsol-infected with the PO stamp genotype [44] belonging to Cluster I [28,45]. This genotype, as all other genotypes from stamp Cluster I, was never found in the Balkans and is characteristic of Western Europe. We can only speculate that C. vitalba is not susceptible to CaPsol genotypes from stamp Cluster II and III, which are widely distributed in Serbia [11][12][13]36,38] and Southeastern Europe [16,20,35,37], but further investigations considering more samples of climbing shrub from a wider area need to be pursued.
In previous research in China, cultivated plants of the Salvia genus (S. miltiorrhiza) were heavily affected by CaPsol, with infection rates ranging from 50 to 70 percent and severe plant damage [46,47]. Moreover, results from France indicated that cultivated crop Salvia sclarea is a host plant for CaPsol and its vector H. obsoletus and could be responsible for the increased incidence of CaPsol in lavender, causing lavender decline [48]. The presence of Salvia pratensis in natural habitats in Serbia, the aggregation of D. europaea on its lush leaves during the dry season [22], and the presence of the STOL genotype, which is widely distributed and experimentally transmitted to grapevine, indicate that this plant could have a significant role in CaPsol epidemiology and BN outbreaks as its reservoir plant and feeding/aggregation host for polyphagous vectors.
Our results from the field-collected data and from the laboratory-controlled transmission assays provide experimental and molecular evidence that D. europaea is involved in pathogen dispersal within natural habitats and between natural habitats and cultivated crops. Combined with the life-history strategy of this planthopper, which involves longliving adults with a polyphagous feeding behavior, there are clear implications that this planthopper is an intermediate vector in the epidemiological cycles of diverse CaPsolinduced diseases and their natural plant reservoirs.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/horticulturae8070654/s1, Table S1: Primer pairs used for identification and characterization of CaPsol isolates; Table S2: List of CaPsol isolates characterized on tuf, secY, vmp1, and stamp genes and their corresponding profiles or GenBank accession numbers sorted according to the hosts.  Table S2. All other relevant data are within the paper and its Supplementary Materials.