Introduction

Plum pox virus (PPV), a potyvirus in the family Potyviridae, is the etiologic agent of sharka (also known as ‘plum pox’), the most important and detrimental viral disease of stone fruits (genus Prunus in the Rosaceae family). The main hosts of PPV are commercial cultivars and rootstocks of major Prunus species, such as apricot (P. armeniaca), peach (P. persica), plum (P. domestica), sweet (P. avium) and sour cherry (P.cerasus; García et al., 2014). In addition, PPV has been reported in several ornamental and wild Prunus species, such as blackthorn (P. spinosa) and cherry plum (P. cerasifera; James and Thompson, 2006; Sebestyén et al., 2008), as well as in two cases in Tilia sp. and Spirea sp. (Çıtır et al., 2018; Pigliónico et al., 2021). In Prunus species, PPV infection has been often associated with other viruses (e.g., apple chlorotic leaf spot virus, ACLSV, prune dwarf virus, PDV, Prunus necrotic ringspot virus, PNRSV) and viroids (e.g., hop stunt viroid, HSVd, peach latent mosaic viroid, PLMVd; Hadidi et al., 2011).

The PPV spread at local scale is mainly due to aphid species such as Myzus persicae, Aphis spiricola and Hyalopterus pruni, which transmit the virus in a non-persistent manner, whereas the long-distance spread of PPV has taken place through global trade of infected plant material (García et al., 2014). Identified for the first time in Bulgaria in 1918 on plum trees (Atanasoff, 1932), nowadays PPV is widespread in Europe, in the Mediterranean basin, in the Middle East, in North and South America, and in Central, South and East Asia (EPPO, 2022).

Sharka symptoms commonly occur on Prunus leaves and fruits as chlorotic (e.g., blotches, rings, oak-leaves patterns, vein clearing) and morphological alterations (e.g., distortion on leaves, arabesque depressions with necrotic areas on fruits; Revers and García, 2015). Therefore, although sharka poses no danger to consumers, it can make stone fruits unmarketable, also by causing acidity and deformities. Moreover, the only way to manage the disease is to destroy all infected tree. The impact of sharka disease over 30 years was thus estimated to be over 10,000 million euros, taking into account the economic losses in production and the costs of sanitary control and surveillance, and of eradication programmes run in affected countries worldwide (Cambra et al., 2006; Rubio et al., 2017). For these reasons, PPV was ranked as quarantine pest in European Union since 2000, although in 2019 it was regraded to regulated non-quarantine pest (RNQP) due to its widespread endemic presence (European Union, 2019).

With a filamentous flexible-shape capside of approximatively 750 × 15 nm (Sochor et al., 2012), the PPV genome consists of a positive sense single-strand RNA (+ ssRNA) of about 9,700-9,800 nucleotides (Schneider et al., 2011; Glasa et al., 2013). PPV sequencing allowed to detect the presence of one main ORF coding for a 355 kDa polyprotein (García et al., 2014). Polyprotein processing gave ten mature proteins renamed in the same way of the nucleotide sequences: P1, HCPro, P3, 6K1, CI, 6K2, VPg, NIapro, Nib, CP. There was also an additional protein, named P3N-PIPO, deriving from a short ORF embedded within the P3 encoding region (Revers and García, 2015). Among these proteins, P1, HCPro and Nlapro are viral proteases that operate for the polyprotein cleavage releasing proteins for virus replication and spreading (García et al., 2014). According to the well-known high mutation rates and high evolutionary potential of potyvirus (LaTourette et al., 2022), the worldwide PPV genetic characterization identified ten main strains: PPV-Marcus (M), PPV-Dideron (D), PPV-Recombinant (Rec), PPV-El Amar (EA), PPV-Cherry (C), PPV-Winona (W), PPV-Turkey (T), PPV-Cherry Russian (CR), PPV-Ancestor (An) and a recent PPV-Cherry Volga (CV; Chirkov et al., 2018). PPV-M, PPV-D and PPV-Rec have been mostly reported, with PPV-M resulting the most aggressive, commonly associated with severe disease symptoms and efficiently transmitted by aphids, whereas PPV-D being the most widely distributed, and PPV-Rec being the recombinant product of PPV-M with PPV-D. Conversely, the other strains are worldwide distributed but restricted to local areas (James et al., 2013).

Although PPV was identified in Italy already in the 1970s (Albert et al., 1974) and it has been largely reported in all Italian regions, except only for Umbria, Liguria and Aosta Valley, only three studies have focused on the genetic characterization of PPV Italian isolates (Myrta et al., 2005; Dallot et al., 2011; Rizza et al., 2014), identifying a few nucleotide sequences from Apulia, Sicily and other unknown areas (a few sequences from Emilia Romagna, Tuscany, Veneto and unknown areas exist as direct submissions to GenBank). In Tuscany, where PPV was firstly identified in the early 1990s (Ginanni et al., 1993), and which is one of the major regions of Central Italy for stone fruits production (ISTAT, 2021), a broad investigation and genetic characterization of PPV isolates was lacking. To address this knowledge gap, this study aimed to (i) confirm the presence of PPV in Tuscan stone fruit orchards, also detailing the rates of PPV co-infection with other viruses (i.e., ACLSV, PDV, PNRSV) and viroids (i.e., HSVd, PLMVd), and (ii) characterize at genetic level the obtained isolates. We anticipate that outcomes here reported will be useful for the understanding of genetic diversity and evolutionary relationships of PPV, thus supporting the challenging contrast to the detrimental sharka disease.

Materials and methods

Field surveys and sampling

In 2020 and 2021 (June-September), field surveys were carried out in 80 commercial orchards selected in accordance with the Regional Phytosanitary Service of Tuscany and located in the six Tuscan districts where almost all the production of stone fruits occur, i.e., Arezzo, Florence, Grosseto, Lucca, Pisa and Siena (Fig. 1). A total of 690 leaf samples were collected from different Prunus species, namely apricot (134), peach (336), and plum (220). These samples were randomly collected from different branches of tree crowns (only a few samples showed virus-like symptoms), placed in coolers and same-day carried to the Plant Pathology Lab of the Department of Agriculture, Food and Environment, University of Pisa where they were stored at 4 °C until RNA extraction.

Fig. 1
figure 1

Tuscany stone fruit orchards investigated for plum pox virus presence. Orchards with negative (white), and positive peach (grey), plum (black), and peach and plum (grey/black) samples are reported. Districts: AR, Arezzo; FI, Florence; GR, Grosseto; LI, Leghorn; LU, Lucca; MS, Massa-Carrara; PI, Pisa; PO, Prato; PT, Pistoia; SI, Siena

Full size image

RNA isolation and cDNA synthesis

Cetyltrimethylammonium bromide buffer 2% (CTAB) was used to extract total nucleic acids (TNA) from 0.5 g of leaf tissue (Li et al., 2008) with minor modifications (Pedrelli et al., 2021). This tissue was powdered in liquid nitrogen and added to 5 ml 2% CTAB buffer and incubated at 65 °C for 15 min. TNA was extracted by one volume of chloroform:iso-amylalcohol (24:1) and precipitated with one volume of isopropanol. Pellets were then washed with 70% ethanol, air-dried and dissolved in 80 µl of RNase/DNase free water. RNA purification kit (EURx, Gdańsk, Poland) was employed, and samples were stored at -80 °C. cDNA synthesis was finally performed using M-MMLV reverse transcriptase (GeneSpin s.r.l., Milan, Italy), according to the manufacturers’ instructions.

PPV and other virus and viroid detection

Samples were first tested for the PPV presence by quantitative polymerase chain reaction (qPCR). Samples positive to PPV were then tested for co-infections with ACLSV, PDV, PNRSV, HSVd and PLMVd. Virus detections were achieved using protocols targeting coat protein gene (CP), according to Olmos et al. (2005) for PPV (PPV-D, PPV-M and PPV-Rec strains), Osman et al. (2017) for ACLSV, and Kim et al. (2010) for PDV and PNRSV. Viroid detections were obtained amplifying a small portion of HSVd and PLMVd genomes, according to Luigi and Faggioli (2011) and (2013), respectively. Infected and healthy controls were included as references in each assay. All samples were tested to determine the presence and quality of cDNA using 18 S rRNA as endogenous control (Osman et al., 2008). qPCR assays were conducted in 20 µl reaction volume contained ITaq (Bio-Rad Laboratories, Hercules, CA, USA) using a Rotor-Gene Q Thermocycler (Qiagen, Venlo, The Netherlands).

PPV sequencing and in silico assays

Molecular typing of PPV strains (i.e., PPV-D, PPV-M and PPV-Rec) was achieved in a C-1000 touch Thermocycler (Bio-Rad Laboratories) using primers spanning nuclear inclusion bodies and coat protein (NIb/CP) region (Subr et al., 2004). Infected and healthy controls were included as references. The PCR products directly sequenced by Sanger DNA method (Eurofins genomics, Ebersberg, Germany) were compared using BLASTn program in GenBank with available isolates (www.ncbi.nlm.nih.gov) and sequence alignments were carried out in Bioedit (Hall, 1999). The phylogenetic trees were obtained in Mega X (Kumar et al., 2018), using Maximum Likelihood (ML) method (Jukes-Cantor model) and bootstrap replicated 1,000 times. Available isolates containing partial or complete NIb/CP genomic region previously described in GenBank were employed in phylogenetic analysis and their accession numbers composing ad hoc database are reported in Supplementary Table S1. The PPV strains (C, D, EA, T, W) were used as out-groups.

Results and discussion

PPV is still present in Tuscany orchards, often as co-infection with other viruses and viroids

Despite the regulation that since 2000 obliged farmers to destroy orchards when resulting PPV infected, this study showed that PPV is still present in Tuscany, likewise in other Mediterranean and European areas (Yardimci and Culal-Klllc, 2011; Gospodaryk et al., 2013; Zagrai et al., 2022), supporting the recent downregulation to RNQP (European Union, 2019). Virus-like symptoms, i.e., chlorotic fathering and chlorotic ring on leaves, were observed in six peaches and six plums, respectively. The PCR assays reported PPV presence in 25 and 12 peach and plum samples, respectively (more than 5% of tested samples), whereas no positive samples were reported in apricot (internal control 18s RNA confirmed cDNA good quality with Ct values ranging between 10 and 14, data not shown). Of the symptomatic samples, only seven, i.e., one peach and six plums, were found to be infected. PPV positive samples were reported in Arezzo (n = 24), Florence (n = 12) and Lucca (n = 1) districts, whereas all samples from Grosseto, Pisa and Siena tested negative (Fig. 1).

Eighteen of the 37 PPV positive samples (49%) showed mixed infections. In most of these cases, PNRSV was the main co-infection (27%), followed by PLMVd (24%), whereas ACLSV and PDV were notably less reported (3% each), although some straight differences were observed between plum and peach. Only PNRSV was co-present (3 samples) among symptomatic samples. Moreover, PNRSV was the only virus co-detected in plum (6 samples), although previous studies showed lower PNSRV rates compared to other viruses (Salem et al., 2003; Çelik et al., 2022). The higher PNRSV occurrence observed in the present study could be due to different cultivar susceptibility (Borisova and Sotirov, 2021) and/or to the presence of PNRSV unalike isolates (Kinoti et al., 2017). Differently, PLMVd was the most co-detected in peach (9 samples), followed by PNRSV (4 samples) and then ACLSV and PDV (1 sample each). The higher occurrence of PLMVd in peach is in accordance with previous reports from Italian orchards (Faggioli and Barba, 2006), and this outcome may be due to the fact that PLMVd frequently occurs without leaf symptoms, making more difficult its management (Flores et al., 2006; Serra et al., 2017). Mixed infections of three viruses, namely PPV + ACLSV + PLMVd, PPV + PDV + PNRSV and PPV + PLMVd + PNRSV, were each reported only in one peach sample. Here, HSVd was never detected, despite its wide host range (Luigi and Faggioli, 2013) and its earlier reports in Tuscany on kumquat, mandarin-lime, florentine citron, pear and lemon (Ragozzino et al., 2005; Rizzo et al., 2017).

PPV-M strain was mostly reported in peach, PPV-Rec was only detected in plum

After almost 30 years from it first report (Ginanni et al., 1993), this is the first broad molecular characterization of PPV strains in Tuscany, as well as one of the few at the Italian regional level, after Myrta et al. (2005) in Apulia and Rizza et al. (2014) in Sicily. Molecular typing of PPV infected samples generated NIb/CP amplicon corresponding to PPV-M (459 bp) or PPV-Rec strains (605 bp), whereas PPV-D was never obtained. Actually, PPV-M was detected in all the 25 peach samples, but only in one plum sample, as PPV-Rec was found only in the remaining 11 plum samples. This divergence between species is in accordance with Sihelská et al. (2017) who reported different strains in different hosts (i.e., PPV-M in peach, PPV-Rec in plum, although also PPV-D in plum and apricot). No mixed strain infections were observed. The overall higher occurrence of PPV-M than PPV-Rec was likely because PPV-M is characterized by an efficient transmission in fields by aphids (James et al., 2013). This outcome is in accordance with García et al. (2014) describing PPV isolates pathogenicity as influenced by specific genetic features, as well as their epidemiological behaviour affected by host and/or local agro-ecological systems.

Eight PPV nucleotide sequences were reported, among which seven were new isolates

According to the above outcomes, PPV amplicons sequencing allowed to obtain 37 sequences, among which 26 belonged to PPV-M and 11 to PPV-Rec (Table 1). BLASTn analysis of PPV-M isolates allowed the identification of five new isolates (IT-AR 1, 2, 4, 5, IT-LU 1), as well as one isolate (IT-AR 3) identical (100.00% similarity) to isolate 04 (KF840166) previously reported in Slovakia on plum. Twenty-one sequences were identical to IT-AR 1, whereas only one sequence was associated to each of other isolates. IT-AR 1, 2, 4, and 5 showed high rates of similarity (99.75, 99.51, 99.50 and 99.27%, respectively) with PPV-Vegama (MW251494) previously found in Czech Republic on apricot cv. Vegama, whereas IT-LU 1 displayed 98.53% similarity with isolate Xi1/5 (MG458873) recovered in Bulgaria on plum. Importantly, IT-AR 1 was found in both peach and plum, whereas other PPV-M isolates were obtained only in peach. Differently, BLASTn analysis of PPV-Rec isolates allowed the identification of two isolates (IT-FI 1 and 2) occurring only in plum. Specifically, eight and three sequences were associated to IT-FI 1 and IT-FI 2, respectively, them showing a 99.11 and 98.93% similarity to isolate IT-ERMO-SH1907/08 (GU942476) previously detected in Italy on plum cv. Black Top. The identity levels of nucleotide and deduced amino acid sequences among PPV-M isolates were 97.71–99.75% and 95.59–100.00% (Table 2) and in PPV-Rec isolates were 99.82% and 100.00%, respectively. All identified PPV-M and PPV-Rec isolates were deposited in GenBank (www.ncbi.nlm.ni.gov).

Table 1 Strain/subgroup, host, number of sequences, similarity coefficient and accession number of Plum pox virus isolates identified in Tuscany stone fruit orchards
Full size table
Table 2 Identity matrix between plum pox virus Marcus (PPV-M) nucleotide (lower left) and deduced amino acid (upper right) sequences recovered in Tuscany stone fruit orchards
Full size table

Low genetic variability was reported in the identified PPV-M and PPV-Rec populations

The NIb/CP genetic variabilities (π) reported in PPV-M (0.010 ± 0.003) and in PPV-Rec (0.002 ± 0.002) populations were in accordance with low intra-strain diversity previously reported in PPV strains (García et al., 2014). Actually, π here reported are much lower than those reported by Kamenova and Borisova (2019) in Bulgaria (0.020 ± 0.004 and 0.018 ± 0.003, respectively), and even more than those emerging from the database here ad hoc developed to include sequences deposited in GenBank (Supplementary Tables 1, see Materials and Methods section for further details) referred to isolates from Mediterranean countries, namely Albania, Bosnia and Herzegovina, Croatia, France, Greece, Italy, Montenegro and Turkey (0.033 ± 0.004 and 0.026 ± 0.004, respectively). As virus isolates are commonly created through constant genetic drift and selection (LaTourette et al., 2022), the observed restricted π is likely due to efficacious monitoring and eradication activities (mandatory until 2021) to control PPV diffusion (Gougherty and Nutter, 2015).

PPV-Mb subgroup was firstly identified in Italy, and it was even prevalent than PPV-Ma

The positions of synonymous (8606 nt compared to M92280 isolate) and non-synonymous (8775 nt) nucleotide changes in CP region reported by Dallot et al. (2011) were observed in PPV-M isolates to differentiate between the subgroups Ma and Mb, which provenance is associated to Mediterranean and eastern European countries, respectively. The nucleotide changes showed IT-AR 1, 2, 3, 4, 5 belonging to subgroup Mb whereas IT-LU 1 to subgroup Ma. The phylogenetic analysis confirmed that PPV-M isolates were splitted in the two Ma and Mb clades supported by high bootstrap values (80 and 70%, respectively) among which the new isolates distributed in agreement with the previous analysis (Fig. 2). The present study, not only revealed both Ma and Mb subgroups, but unexpectedly Mb resulted prevalent as it was reported from five isolates (IT-AR 1, 2, 3, 4, 5) detected in Arezzo and Florence districts, whereas Ma was reported only in one isolate (IT-LU 1) detected in Lucca area. Differently, all the Italian isolates L9-1b (KJ994236), 9-335b (KJ994238), IT1 (EF626554), IT2 (EF626555), IT3 (EF626556) and CREA-DC_PPV6 (OL771187) previously reported in GenBank belong only to Ma. Therefore, to the best of our knowledge, this is the first record of Mb isolates in Italy.

Fig. 2
figure 2

Phylogenetic tree of plum pox virus Marcus (PPV-M) isolates from Italy reconstructed from partial NIb/CP genomic region. The trees were generated by Maximum Likelihood (ML), using the Jukes-Cantor model of evolution for nucleotide. The significance of each branch was evaluated by constructing 1,000 trees in bootstrap analysis. Bootstrap values are shown. The isolates sequenced in this study are in bold. The country of origin is shown in brackets after the name of the isolate. Subdivisions between strain and among PPV-M strain subgroups are reported on the right

Full size image

PPV-Rec isolates were phylogenetically close to Italian and Turkish isolates previously detected

The phylogenetic analysis conducted using all the available PPV-M isolates containing NIb/CP genomic region retrieved in Genbank and included in the ad hoc developed database showed low bootstrap values among M strain and IT-AR 1, 2, 4, 5 resulted closed to the Bosnian isolate BOS150Pl (AJ749995) and to the German PPV-CGG-M7 (AY450597), found both in plum. IT-AR 3 was close to the Slovakian isolates 03 (KF840165) and 04 (KF840166), recovered in peach and plum, respectively, as well as to the North Macedonia isolates MK112 (MK562731) and MK175 (MK562732) found in peach and apricot, respectively. IT-LU 1 was close to the Turkish EdMrPl276 (MW415771) and North Macedonia MK41 (MK562730) isolates detected in plum and myrobalan, respectively (Fig. 3 and Supplementary Figure S1). Conversely, the analogous phylogenetic analysis for PPV-Rec strain showed high bootstrap values, with IT-FI 1 and 2 resulting close to the Italian isolates PPVBR (AJ812242) from Apulia and IT-ERMO-SH1907/08 (GU942476) from Emilia Romagna found in apricot and Japanese plum, respectively, as well as to the Turkish isolate Isparta (EF051630) identified in Japanese plum (Fig. 4 and Supplementary Figure S2). This result suggests a common origin for these isolates, and it may allow to hypothesize a propagation of PPV-Rec strains by movement of infected material along Italy and the Mediterranean basin (Sastry, 2012). Nevertheless, despite these high bootstrap values, no straight country or host-based clustering was observed among PPV-Rec population detected in the present study, as well as among PPV-M ones.

Fig. 3
figure 3

Portion of the phylogenetic tree of plum pox virus Marcus (PPV-M) reconstructed from partial NIb/CP genomic region. The trees were generated by Maximum Likelihood (ML), using the Jukes-Cantor model of evolution for nucleotide. The significance of each branch was evaluated by constructing 1,000 trees in bootstrap analysis. Bootstrap values are shown. The isolates sequenced in this study are in bold. The country of origin is shown in brackets after the name of the isolate. The complete PPV-M phylogenetic tree is available in Supplementary Figure S1

Full size image
Fig. 4
figure 4

Portion of the phylogenetic tree of plum pox virus Recombinant (PPV-Rec) reconstructed from partial NIb/CP genomic region. The trees were generated by Maximum Likelihood (ML), using the Jukes-Cantor model of evolution for nucleotide. The significance of each branch was evaluated by constructing 1,000 trees in bootstrap analysis. Bootstrap values are shown. The isolates sequenced in this study are in bold. The country of origin is shown in brackets after the name of the isolate. The complete PPV-Rec phylogenetic tree is available in Supplementary Figure S2

Full size image

Conclusions

Although in 2019 the European Union regraded PPV to RNQP due to its widespread endemic presence, this virus still represents one of the most devastating pathogen of stone fruits, with huge yield and economic losses worldwide. Thus, maintenance and improvement of diagnostic and molecular characterization activities is still crucial, especially considering that viruses and their vectors seem to be strongly favoured by climate change. This research provided a first broad molecular characterization of PPV strains in stone fruits of Tuscany. First, it confirmed PPV presence in this region, often occurring as co-infection with other viruses and viroids, and with PPV-M strain mostly reported in peach and PPV-Rec only detected in plum. Furthermore, starting from the identification of eight PPV nucleotide sequences (among which seven were new isolates), this study firstly identified the PPV-Mb subgroup in Italy, which was even prevalent than PPV-Ma. Finally, PPV-Rec isolates resulted phylogenetically close to Italian and Turkish isolates previously detected. Overall, the results here presented represent an important step to fill knowledge gaps about PPV in Tuscany, and we believe it may encourage other similar research to achieve more accurate data on PPV populations at both national and international levels.