Research Papers – 12th Special issue on Grapevine Trunk Diseases

Summary. Grapevine Trunk Diseases (GTDs) are major threats in Mediterranean countries, causing economic losses due to reduced grape yields and long-term vine productivity, as well as death of grapevines. A survey was conducted in Piedmont (Northern Italy) during 2021-2022 to investigate the species diversity and distribution of GTD pathogens in this important Italian wine region. Morphological and multi-locus phylogenetic analyses (based on ITS, tef1 , tub2 , act and rpb2 ) identified species of Botryosphaeriaceae at high frequency, including Botryosphaeria dothidea , Diplodia mutila, Diplodia seriata and Neofusicoccum parvum . Other pathogens commonly associated with GTDs, including Eutypa lata , Fomitiporia mediterranea and Phaeomoniella chlamydospora , were also isolated. Less commonly isolated species included Neocu-curbitaria juglandicola, Paraconiothyrium brasiliense , Seimatosporium vitis-viniferae and Truncatella angustata . Pathogenicity tests with two representative isolates of each species were carried out using one-year-old potted grapevine cuttings (‘Barbera’). All isolates (except N. juglandicola ) caused brown wood necrotic vascular discolourations on inoculated plants and were successfully re-isolated. Effects of temperature on colony growth were also assessed. For all tested isolates there was no growth at 5°C, only four isolates ( Botryosphaeriaceae ) grew at 35°C, and optimum growth temperatures were between 20 and 25°C. This is the first record of Paraconiothyrium brasiliense and Neo-cucurbitaria


INTRODUCTION
Grapevine (Vitis vinifera L.) is an important cultivated crop, with a worldwide vineyard area of 6.73 million ha (FAOSTAT, 2021), mainly grown for wine and table grape production.Reports have increased of diseases caused by grapevine trunk diseases (GTDs) associated fungi causing severe economic and yield losses as a result of reduced grape quality and early plant death.Grapevine trunk diseases are severely destructive in Europe and Mediterranean countries (Guerin-Dubrana et al., 2019), representing major threats to vineyard productivity.
Grapevine trunk diseases are associated with different vascular xylem-colonizing pathogenic fungi.Phaeomoniella chlamydospora and several Phaeoacremonium spp.are responsible for Petri disease and Esca complex, the major GTDs reported in all European and Mediterranean countries.Fomitiporia mediterranea, in the same areas, is the most common lignin-degrading Basidiomycete fungus responsible for white rot (Surico et al., 2004).Different dieback diseases such as Botryosphaeria dieback, Diaporthe dieback and Eutypa dieback are associated with different species of related fungal families (Claverie et al., 2020).
Among Fungi commonly associated with GTDs, several recent studies have highlighted the association of other wood-degrading fungi with symptomatic plants (Raimondo et al., 2019;Bekris et al., 2021).Numerous fungi belonging to Neopestalotipsis, Truncatella, Seimatosporium and Sporocadus have recently been reported as part of the grapevine microbiome, while their roles as causes of symptoms need to be clarified (Maharachchikumbura et al., 2017;Geiger et al., 2022;Vanga et al., 2022).
With more than 718,000 ha of wine and table grapevines (O.I.V, data 2022, https://www.oiv.int/it/whatwe-do/country-report?oiv), Italy is leading grapevine production worldwide and represent the fourth largest vineyard acreage after Spain, France and China.Among all Italian regions, Piedmont (Northern Italy, with the Langhe area included in UNESCO's World Heritage list), is a renowned wine production region.In the last 30 years, incidence of GTDs has increased in all Italian regions, and several fungi have been reported associated with grapevines showing various symptoms (Surico et al., 2000;Guerin-Dubrana et al., 2019).Since the 1990s, studies have reported high disease incidence and mortality of plants in the first year of planting.The Esca complex, including apoplexy, is frequent and widespread in all grape growing Italian regions (Guerin-Dubrana et al., 2019), and reaches high incidence in climatical-ly favourable seasons, up to 80% in mature vineyards (Romanazzi et al., 2009).
Botryosphaeria dieback associated with different pathogens, including Diplodia seriata, Neofusicoccum parvum and Lasiodiplodia theobromae, was reported in Apulia, Marche, Molise, Tuscany, Sardinia and Sicily (Burruano et al., 2008;Romanazzi et al., 2009;Spagnolo et al., 2014;Carlucci et al., 2015;Linaldeddu et al., 2015;Mondello et al., 2020), but no investigations have been carried out in Piedmont.Due to limited information on distribution of GTD related pathogens in Piedmont, the present research aims were: 1) to investigate the species diversity and distribution of GTD pathogens in Piedmont, focusing on canker agents and wood-degrading fungi associated with dead cordons, independently from detected foliar symptoms; 2) to characterize obtained isolates; and 3) to test representative isolates for pathogenicity to healthy grapevine plants.

Field sampling and isolation of fungi
Surveys were carried out from July 2021 to November 2022, in five vineyards in the Alba and Alessandria areas of Piedmont, Northern Italy.Wood samples were collected from necrotic cordons, and from trunk portions of declining vines of 12 different grapevine culti-vars (Table 1).Sampled vines were aged between 10 and 25 years, and showed dieback symptoms, including foliar discolourations, dieback and internal dark V-shaped wood necroses.The sampling method was destructive; vines were cut and transverse sections of the affected trunk and branches of each plant were examined to check for wood necroses.Each sample was reduced into small fragments, including the necrotic zone, and was sterilized in a sodium hypochlorite solution (1%) for 1 min and then rinsed in sterilized distilled water for 30 sec.Excess water was removed using sterilized filter paper.The wood samples were cut into small pieces from the margins of the necrotic zones.Five wood fragments from each sample were plated onto the surface of Potato Dextrose Agar (PDA, Merck) in a Petri plate, supplemented with streptomycin sulphate (25 ppm L -1 , PDA-S), and incubated at 25 ± 1°C.After 5 d the plates were examined.From the margins of resulting fungal colonies, single hyphal tips were cut and placed on new PDA plates, to obtain pure cultures.

DNA extraction, polymerase chain reaction (PCR) amplification, and sequencing
For recurring fungal colonies, mycelium from each 10-d-old pure culture on PDA was scraped and collected into a 2 mL capacity centrifuge tube.Total DNA was extracted directly from fresh mycelium using the E.Z.N.A.® Fungal DNA Mini Kit (Omega Bio-Tek), following the manufacturer's instructions.DNA amplification and sequencing of different loci were carried out to achieve species identification.Botryosphaeriaceae-like isolates were characterized through DNA amplification and sequencing of the partial translation elongation factor-1α (tef1) gene, using the primers EF1-728F and EF1-986R (Carbone and Kohn, 1999).For the remaining isolates, the nuclear ribosomal internal transcribed spacer (ITS) region was amplified using universal primers ITS1 and ITS4 (White et al., 1990), while the primers T1 and Bt2b (Glass and Donaldson, 1995;O'Donnell and Cigelnik, 1997), fRPB2-5f and fRPB2-7cr (Liu et al., 1999), and ACT512f and ACT783r (Carbone and Kohn, 1999), were used to amplify, respectively, genes for partial beta-tubulin (tub2), the fragment of the RNA polymerase II subunit 2 (rpb2), and γ-actin (act).The PCR reactions and conditions adopted for all the loci were described in the above respective cited studies.Polymerase chain reaction assays were carried out in a final 25 μL volume, using a Taq DNA polymerase kit (Qiagen) and 25 ng of DNA.

Phylogenetic analyses
To give an overview of isolated genera, an initial phylogenetic analysis was conducted with sequences of the partial translation elongation factor-1α (tef1) gene for Botryosphaeriaceae-like isolates, and with the nuclear ribosomal internal transcribed spacer (ITS) gene for other isolates.Subsequently, a subset of representative isolates was then selected based on the previous results, to distinguish the isolates at species level.A multilocus phylogenetic analysis was conducted using the following locus combinations: ITS and tef1 for members of Botryosphaeriaceae (Pintos et al., 2018;Guarnaccia et al., 2020); ITS, tub2 and act were amplified for Paraconiothyrium (Verkley et al., 2004); ITS, tub2 and rpb2 for Neocucurbitaria (Jaklitsch et al., 2018); and ITS, tub2 and tef1 for isolates related to the family Sporocadaceae including the genera Truncatella and Seimatosporium (Raimondo et al., 2019).Isolate sequences, including ref-erences downloaded from GenBank, were aligned with the software MAFFT v. 7 online server (http://mafft.cbrc.jp/alignment/server/index.html)(Katoh and Standley, 2013), and were then manually adjusted in MEGA v.7 (Kumar et al., 2016).Multi-locus analyses, based on Maximum Parsimony (MP) were performed using Phylogenetic Analysis Using Parsimony (PAUP) v. 4.0b10 (Cummings, 2004), while MrModeltest v. 2.3 (Nylander, 2004) and MrBayes v. 3.2.5 (Ronquist and Huelsenbeck, 2003) were used for the Bayesian Inference (BI) analyses.The best nucleotide substitution model for each gene was estimated using MrModeltest.Based on obtained results for optimal setting criteria for each locus, BIs were performed using the Markov Chain Monte Carlo (MCMC) method.Four simultaneous Markov Chains were run for 1,000,000 generations starting from a random tree topology.Trees were saved each 1000 generation, while preburn and heating parameters were set, respectively, to 0.25 and 0.2.Based on burn-in fraction, the remaining trees were used to calculate the majority rule consensus tree and posterior probability (PP).The analyses stopped once the average standard deviation of split frequencies fell below 0.01.For Maximum Parsimony, phylogenetic relationships were estimated using the heuristic search option with 100 random addition sequences.Tree bisection reconnection (TBR) was used with branch swapping option as "best trees"; characters were treated as equally weighted, and gaps as fifth base.Parsimony and the bootstrap analyses were based on 1,000 replications, and tree lengths (TL), consistency indices (CI), retention indices (RI) and rescaled consistence indices (RC) were calculated.Resulting trees were visualized with FigTree version 1.4.4 (Page, 1996).Sequences generated in this study were deposited in GenBank (Table 2).

Morphological analyses
Two isolates of each species identified using molecular analyses were selected for the morphological observation.Mycelium plugs (each 6 mm diam.) were taken from each 10-d-old fungal colony growing on PDA and were transferred to Petri dishes containing different media.To enhance sexual sporulation or conidium production, for Botryosphaeriaceae-like isolates, 2% water agar supplemented with sterile pine needles (Pine Needle Agar or PNA, Smith et al., 1996) was used, with incubation at 25°C under near-UV light (Crous et al., 2006).For Neocucurbitaria, malt extract agar (MEA) and PDA were used, with incubation at 20°C and alternating lightdark periods (Jaklitsch et al., 2018).Corn meal agar (CMA), oat agar (OA) and MEA were used for Paraconiothyrium, incubated at 25°C under UV light.PDA, MEA   and OA, with incubation at 25°C in dark, were used for Seimatosporium isolates (Kanetis et al., 2022).For Truncatella isolates, CMA and MEA were used, and colonies were incubated at 21°C with alternating light-dark periods (Liu et al., 2019).

Effects of temperature on fungal colony radial growth
To investigate the effect of temperature on the colony radial growth of selected fungal isolates (Table 2), each isolate was grown on PDA amended with streptomycin sulphate (25 ppm L -1 ) for 7 d in the dark at 25°C.Mycelium plugs were taken from the margins of 10-d-old colonies using a cork borer (0.6 cm diam.) and were placed upside down at the centers of 9 cm diam.Petri dishes, each containing 10 mL of PDA-S medium.All plates were then incubated for 7 d at 5, 10, 15, 20, 25, 30 or 35°C, and each isolate was tested using seven replicate plates per temperature.Following incubation, the Petri plates were examined without being opened, and the mean colony diameters (minus the diameter of the initial inoculation plugs) of each growing mycelial colony were measured in two perpendicular directions at the end of the 4th and 7th day.Radial growth rates (mm d -1 ) were calculated for each temperature.The variations in mycelium growth rates at different temperatures were analyzed using the generalized Analytics Beta model (López-Moral et al., 2017).Based on this analysis, optimum growth temperature and the corresponding maximum growth rate were calculated for each isolate.Box-Cox transformation was applied to optimum growth temperature data.To satisfy ANOVA assumptions, normality and homogeneity of variance were evaluated with, respectively, Shapiro-Wilk and Levene's tests.One-way ANOVA was carried out, followed by Tukey's test for evaluation of statistically significance differences between means (at P < 0.05), as both ANOVA assumptions were satisfied for the growth rate data.Welch's ANOVA was performed on optimum growth temperature data because only the normality assumption was satisfied.Statistical differences (P < 0.05) were analyzed with the Games-Howell post hoc test.All statistical analyses were carried out using R (https:// www.R-project.org/).

Pathogenicity tests
Fifteen representative isolates from seven identified fungal species were used to inoculate one-year-old potted 'Barbera' grapevine cuttings grafted on K5BB rootstock (Table 2).Ten plants were inoculated with each isolate.The inoculations were carried out in May 2022.Cuttings were inoculated above the grafting point by forming a slit (1.0-1.5 cm long) using a sterile scalpel as described by Carlucci et al. (2015) and Bezerra et al. (2021).Agar plugs (6 mm diam.) were taken from 10-d-old fungal cultures grown on PDA and plugs were placed with mycelium in contact with plant tissues, under the stem bark.Each inoculated wound was wrapped with wet sterile cotton wool soaked in sterilized distilled water and was then sealed firmly with Parafilm® (American National Can) to maintain high humidity at the inoculation point.Control plants were inoculated with sterile agar plugs.Inoculated plants were the placed in a greenhouse at 25 ± 3°C, from May to November 2022.After 180 d. from inoculation, the plants were examined after bark removal and lengths of any visible necrotic wood lesions were measured from the inoculation points.Small tissue pieces (0.5 cm) from the necrotic area were placed on PDA supplemented with streptomycin sulphate (25 ppm L -1 ), and incubated at 25 ± 1°C.To fulfil Koch's postulates, resulting colonies were identified based on their morphological characteristics.Data of necrotic lesion lengths were subjected to statistical analysis.Shapiro-Wilk (W) tests were used to determine if the data followed normal distributions.Levene's tests were carried out to assess the homogeneity of the variances of the dataset.A Welch's ANOVA was performed because the dataset was normally distributed, but data were not homoscedastic.The Games-Howell post hoc test was used to evaluate statistically significant differences among mean lesion lengths caused by the different fungal isolates (at P ≤ 0.05).All statistical analyses were carried out using R (https://www.R-project.org/).

Sampling, isolation and morphological identification of isolates
In sampled vineyards, more than 30% of the plants showed Botryosphaeria dieback related symptoms.Approximately 5-10% of plants showed decline with severe dieback and death.Sampled grapevines showed typical dieback symptoms, primarily associated with Botryosphaeria dieback, such as defoliation and wedgeshaped cankers of internal wood tissues and dark streaking of wood.
A total of 248 fungal isolates were obtained from a total of 32 symptomatic vines of 12 cultivars.The first screen and identification of isolates was based on their morphological and cultural characteristics.A group of isolates identified as Botryosphaeriaceae-like showed high isolation frequency (80 isolates, 37% of total isolates obtained) compared with other common GTDs fungi, such as P. chlamydospora (38 isolates), E. lata (5 isolates) and F. mediterranea (12 isolates) that were occasionally present.

Phylogenetic analyses
The multi-locus analyses conducted on all isolates confirmed the genera obtained with the initial phylogenetic analysis of the tef1 and ITS regions.The combined locus analysis of Botryosphaeriaceae-like isolates consisted of 35 sequences and Lecanosticta acicola, which was chosen as the outgroup.A total of 995 characters (ITS: 1-660 and tef1: 666-995) were included in the phylogenetic analyses of Botryosphaeriaceae-like isolates.A total of 1405 characters (ITS, 1-628;tub2, 633-1123;tef1, 1128-1405) were included in the Paraconiothyrium analyses where, for a total of 15 sequences, Alloconiothyrium aptrootii was chosen as the outgroup.For Neocucurbitaria, Pseudopyrenochaeta lycopersici was chosen as outgroup, and a total of 1810 characters (ITS, 1-507;tub2, 511-896;rpb2, 901-1810) were included in the phylogenetic analyses performed with 16 sequences.The combined phylogenetic session for Seimatosporium and Truncatella had a total of 1823 characters (ITS, 1-579; tef1, 583-1071; tub2, 1076-1823) with 25 sequences.Discosia artocreas was chosen as the outgroup.For each session, a tree was created based on a maximum of 1000 equally most parsimonious trees.Bootstrap support values for all MP trees obtained are shown in Figures 2, 3, 4 and 5.
In the Botryosphaeriaceae-like analyses, two isolates (CVG15777 and CVG1753) clustered with D. seriata reference strains, and isolates CVG1582 and CVG1615 clustered with B. dothidea strains.Isolates CVG1588 and CVG1731 were grouped with Neof.parvum, and isolates CVG1739 and CVG1714 clustered with D. mutila reference strains.For Paraconiothyrium, both strains clustered with reference strains of P. brasiliense.In the phylogenetic tree from the Truncatella and Seimatosporium analysis, two isolates (CVG1601, CVG1631) were identified as T. angustata, and two isolates (CVG1681, CVG1682) were grouped with S. vitis-viniferae reference strains.Isolate CVG1779 of Neocucurbitaria clustered in the Neoc.juglandicola clade.The recommended evolutionary model, unique site patterns, number of generations, and tree produced and sampled for each partition of the Bayesian analyses are reported in Table 3, as well as other parameters produced by MB analyses, including tree lengths, consistency, retention, and rescaled consistency indices.Data obtained from the multi-locus analyses carried out on the 15 selected representative isolates gave four Botryosphaeriaceae species, including B. dothidea, D. mutila, D. seriata and Neof.parvum.Among other less frequently isolated taxa, Neocucurbitaria juglandicola, P. brasiliense, S. vitis-viniferae and T. angustata were identified.

Morphology
Morphological observations were performed for all the selected species.Colonies characteristic, including edges shape, colony front and reverse color, mycelia appearance and conidia morphology of B. dothidea, D. mutila, D. seriata and Neof.parvum were congruent with previous descriptions of species belonging to Botryosphaeriaceae family (Phillips et al. 2013).Different conidia were observed (cylindrical to fusiform, hyaline to dark brown) and all isolate showed fast growth mycelia, becoming dark with age starting from the center, spreading to the whole colony.
Colonies of Neoc.juglandicola on PDA and MEA showed slow growth with uneven margins.Colony upper surfaces were brown to dark brown with dense zonate mycelium.Pycnidia appeared as dots, which were numerous and centrally located.Reverse colony sides were dark brown.Conidia, produced on PDA measured 2.0-3.1 × 1.3-1.5 μm, mean (± S.D) = 2.5 ± 0.5 × 1.4 ± 0.1 μm, and were unicellular with smooth surfaces, hya-line, and ellipsoid with rounded apices.Based on morphological features, colonies had similar characteristic to those reported by Jaklitsch et al. (2018).Colonies of P. brasiliense on MEA were white-gray, had regular margins with rapid growth and darker aerial mycelium in the centre.Reverse colony sides were light brown-honey to light yellow amber.On OA the colonies were light gray, each with a darker area.On CMA the colonies were white, with mycelium development in the centres showing concentric and radiating patterns.Conidia were cylindrical to ellipsoid with rounded apices, and measured 2.5-4.4 × 1.3-2.8μm, mean (± S.D) = 3.3 ± 0.9 × 1.9 ± 0.7 μm.They were hyaline and unicellular, with smooth walls, and granular contents.Based on morphological features, colonies had similar characteristic to those reported by Kanetis et al. (2022).
Colonies of S. vitis-viniferae on PDA and MEA had entire edges and were light brown to reddish with wooly aerial mycelium with smooth whitish margins.On OA the colonies were slightly to light brown with off-white wooly margins.Conidia were fusiform, each with three septa, and were constricted at each septum, measuring 15.8-22.7 × 4.2-6.1 μm, mean (± S.D) = 18.9 ± 3.5 × 5.1 ± 1.0 μm.The conidia were pale to dark brown, and each basal cell had an appendage while the apical cell had a rounded apex.
Colonies of T. angustata on PDA had entire edges, had pale gray to white fuzzy mycelium from above and grayish to white on the reverse sides, and were fastgrowing.Black pycnidia were observed at the centre of each colony after 7 d.On MEA the colonies had entire edges, and grew slowly, with cottony white to light brown mycelia.Conidia were fusiform (17.5-19.7 × 6.1-7.3 μm, mean (± S.D) = 18.6 ± 1.1 × 6.7 ± 0.6 μm), mostly with three cells and were transversally septate without septal constrictions, and with truncate bases and several appendices.Based on morphological features of colonies and conidia, isolates studied have similar characteristic to those reported by Raimondo et al. (2019).

Effects of temperature on fungal growth
None of the tested isolates grew at 5°C, growth was slow between 10 to 15°C, and was optimum at 20 to 25°C.Four isolates, CVG1577 (D. seriata), CVG1582 (B.dothidea), CVG1588 (Neof.parvum) and CVG1741 (D. mutila), grew at 35°C.A generalized Analytics Beta model was used to describe the relationship between mycelial growth and selected temperatures (Figure 7) and optimum growth temperature, and the corresponding maximum growth rates were calculated.Coefficients of determination (R 2 ) for the Analytics Beta model ranged between 0.88 and 0.99.Analysis of variance (ANOVA) was carried out on data of mycelial growth rates and optimum growth temperatures.).Mean optimum temperatures for mycelial growth were 24.8°C for S. vitis-viniferae and 24.6°C for Neoc.juglandicola.Both the species showed slow growth at, respectively, 0.83and 0.54-mm d -1 .Maximum growth for T. angustata (3.8 mm d -1 ) was obtained at 20.8°C and for P brasiliense (1.69 mm d -1 ) was at 22.8°C.Based on maximum growth rates, Botryosphaeriaceae isolates had the fastest growth rate (>8 mm d -1 ), followed by T. angustata and P. brasiliense (< 5 mm d -1 ).Seimatosporium vitisviniferae and Neoc.juglandicola had the slowest growth rates (< 1 mm d -1 ).

Pathogenicity tests
The fungal isolates used for pathogenicity tests caused brown necroses and vascular discolourations in the wood of inoculated grapevines, 180 d after inoculation.No lesions were observed on inoculation control plants.Neocucurbitaria juglandicola (CVG 1779) was not re-isolated from necrotic areas in the wood, while all the other respective inoculated fungi were successfully re-isolated from the grapevine plants, fulfilling Koch's postulates.Re-isolated identifications were confirmed through morphological and molecular analyses (partial tub gene sequencing) while frequencies of re-isolations of inoculated species ranged from 80% and 90%.A Shapiro-Wilk (W) test was used for data of necrotic lesion lengths on the inoculated plants to determine if they followed normal distributions with W = 0.9807 (P-value = 0.07556).Levene's test showed that the homogeneity of variance was not significant for the dataset (P = 0.001825).Because data were normally distributed but not homoscedastic, a Welch's ANOVA was performed.This showed that statistically significant differences occurred among the inoculated fungi (P= 3.167e-15).Results of the Games-Howell post hoc test to evalu-   ate differences among mean lesion lengths are shown in Figure 6.The longest necrotic lesions were produced by Neof.parvum (mean length = 178.8mm).Aggressiveness of the other Botryosphaeriaceae was also confirmed: these strains produced variable vessel discolouration, with mean lesion lengths of 147.5 mm from D. seriata, 94.6 mm from B. dothidea, and 86.6 mm from D. mutila.Paraconiothyrium brasiliense and S. vitis-viniferae each showed similar aggressiveness compared to the Botryosphaeriaceae isolates, with, respectively, mean lesion lengths of 134.8 mm and 145.0 mm.Neocucurbitaria juglandicola (mean lesion length = 84.4mm) and T. angustata (mean lesion length = 103.4mm) were the least aggressive among the non-reported GTD pathogens.

DISCUSSION
This study has characterized the different fungal species associated with dieback symptoms observed in representative vineyards in Piedmont, Italy.Some of these fungi are already known to be associated with diseases such as the Esca complex or Botryosphaeria dieback.Among all isolates collected, most were Botryosphaeriaceae, with B. dothidea, D. seriata, D. mutila and Neof.parvum identified through morphological characterization and confirmed by multi-locus phylogenetic analyses.No Lasiodiplodia isolates were found, which may be result because of these fungi is more prevalent in tropical and sub-tropical climatic regions.In Italy, Lasiodiplodia was only reported in Sicily (Burruano et al., 2008).Other pathogens commonly associated with the Esca complex, were sporadically recovered, including Fomitiporia mediterranea and Phaeomoniella chlamydospora.No isolates of Phaeoacremonium spp.were detected, which is another pathogen commonly associated with the Esca complex (Essakhi et al., 2008).Eutypa lata, the Eutypa dieback pathogen was also occasionally isolated.The pathogens commonly associated with the various symptoms observed on sampled grapevines were successfully isolated.The wedge-shape cankers, typically caused by Botryosphaeriaceae-like fungi or Eutypa infection, were the most common wood symptoms, but Esca-complex associated symptoms were also observed.Some species less frequently isolated from affected grapevine wood were detected.These included Neoc.juglandicola, P. brasiliense, S. vitis-viniferae and T. angustata.For some of these isolates, association with grapevine woody tissues had been reported previously (Elena et al., 2018;Raimondo et al., 2019), and their respective pathogenicity related to GTDs was confirmed.
In different countries, including China (Yan et al., 2013), Iran (Arzanlou et al., 2012), Portugal (Phillips, 2002), Spain (Úrbez-Torres et al., 2006), Turkey (Akgul et al., 2014), France (Larignon et al., 2001), and the United States of America (Úrbez-Torres and Gubler, 2009;Trouillas et al., 2010).Botryosphaeria dothidea has been described as one of the species associated with typical V-shaped necrotic wood lesions and brown discolouration of the xylem vessels.In Italy, B. dothidea was reported on grapevine in the southern and central regions, on which its pathogenicity was confirmed (Carlucci et al., 2009).The two B. dothidea strains used in the present study produced dark streaks on inoculated cuttings, similar to those previously reported.
Diplodia seriata is known to be widespread in Europe, as it was described associated with symptomatic grapevine in Spain (Martin and Cobos, 2007), Portugal (Rego et al., 2009), France (Larignon et al., 2001), Turkey (Akgul et al., 2014) and Croatia (Kaliterna and Miličević, 2014).Reported as either pathogenic or saprophytic in different hosts, pathogenicity trials conducted by Taylor et al. (2005) and Carlucci et al. (2015) confirmed its role in causing necrotic wood lesions on V. vinifera.Based on the present results, isolates of D. seriata produced longer lesions than those caused by D. mutila and B. dothidea, confirming the variability in virulence among different isolates (Elena et al., 2015).
While D. seriata has been reported to be associated with grapevine in different Italian regions, including Apulia (Pollastro et al., 2000) and Tuscany (Spagnolo et al., 2011).However, D. mutila has already been reported associated with Vitis vinifera in Hungary (Lehoczky, 1974;Kovács et al., 2017), Spain, California, and Chile (Morales et al., 2012;Díaz et al., 2013) and with grapevine canker and dieback in Italy (Carlucci et al. 2015).In the present study, D. mutila produced wood discolourations after artificial inoculations, so the association with symptomatic grapevine in Italy was confirmed.
Furthermore, pathogenicity trials carried out in the present study showed Neof.parvum to be the most virulent species producing the longest necrotic lesions.This result is similar to those of Billones-Baaijens et al. (2013) and Úrbez-Torres and Gubler (2009), who, after pathogenicity trials conducted in, respectively, New Zealand and California, reported Neof.parvum as one of the most aggressive species associated with Botryosphaeria dieback.In Italy, Carlucci et al. (2015) came to the same conclusion after testing its pathogenicity on two grapevine cultivars.Neofusicoccum parvum was usually reported as an aggressive wood pathogen, able to infect many hosts.This fungus was also described in association with Botryosphaeria dieback symptoms on grape-vine in France (Larignon et al., 2015), Algeria (Berraf-Tebbal et al., 2014), Spain (Luque et al., 2009), Portugal (Phillips, 2002), and Turkey (Akgul et al., 2014).Likewise, it was isolated from symptomatic grapevines in the Italian regions of Apulia and Tuscany (Carlucci et al., 2009;Spagnolo et al., 2011).
The association of Seimatosporium with grapevine is known, as well as its wide distribution and ability to colonize many hosts (Raimondo et al., 2019).Among all Seimatosporium species, recent studies have reported the association of S. vitis with GTD symptoms, where it was isolated from necrotic tissues and dead cordons in California (Lawrence et al., 2018) and Hungary (Váczy, 2017).In Italy, S. vitis was the first Seimatosporium species to be described in association with dead stem of V. vinifera (Senanayake et al., 2015) and Camele and Mang (2019) described it for the first time causing GTDs.In 2022, Kanetis et al. reported another species, S. vitisviniferae, associated with lesion and wood discolouration on grapevine.In Italy, Raimondo et al. (2019) tested its pathogenicity, confirming its association with GTD symptoms.This agrees with results in the present study, which showed S. vitis-viniferae as causing wood necroses after artificial inoculation, with similar severity to necroses caused by D. seriata.Truncatella genera, which is phylogenetically close to Seimatosporium, has been revised by Liu et al. (2019) and includes only one species, T. angustata, while other Truncatella species have been reallocated to other genera.Truncatella angustata has been reported in association with grapevine, isolated as an endophyte in Spain (González and Tello, 2011) and Switzerland (Casieri et al., 2009).As a pathogen, the involvement of T. angustata with GTDs has been demonstrated, isolation from symptomatic grapevines in France (Pintos et al., 2018) and Iran (Arzanlou et al., 2013), and in pathogenicity tests (Úrbez-Torres et al., 2009).This fungus is considered an opportunistic pathogen on grapevine which is not primarily involved in GTDs.In Italy, T. augustata was first reported by Raimondo et al. (2019), who after isolation from symptomatic grapevines, confirmed its pathogenicity and involvement in GTDs.In accordance with the above studies, the present study confirmed its weak pathogenicity on grapevine, causing necrotic discolourations (mean length = 103.4mm) after artificial inoculations.
Paraconiothyrium brasiliense was also less frequently isolated from woody tissues, and the role of Paraconiothyrium spp. on grapevine requires clarification.Pathogenicity on fruit trees and other woody hosts has been demonstrated for different species (Damm et al., 2008), while, P. brasiliense has been isolated from symptomatic and non-symptomatic grapevine tissues from Spain (Elena et al., 2018) and the United States of America (DeKrey et al., 2022).In Italy, P. brasiliense was also recently reported associated with dieback of apple trees (Martino et al., 2023).Pathogenicity trials conducted in the present study showed that P. brasiliense produced wood streaking with similar lesion length (mean = 134.8mm) to lesions produced by S. vitis-viniferae (mean = 145.0mm), confirming pathogenicity of P. brasiliense on grapevine.This study has demonstrated the role of P. brasiliense as a weak woody pathogen, and this is the first report of this fungus as a grapevine pathogen in Italy.Neocucurbitaria juglandicola has also been identified in this study.Neocucurbitaria quercina was reported from grapevine in the United States of America (DeKrey et al., 2022), while Neoc.juglandicola has been reported in association with Juglans regia and Quercus rubra (Jaklitsch et al., 2018).In the present study, after artificial inoculations, necrotic discolouration was visible, but it was not possible to re-isolate the fungus from necrotic areas.The presence of Neoc.juglandicola demonstrates its association with grapevine in Italy, while the pathogenicity tests did not prove its virulence or its association with GTDs.This is the first report of Neoc.juglandicola associated with grapevine, however.
Optimum growth temperatures of tested isolates ranged between a maximum of 27°C (for Neof.parvum) to a minimum of 21°C (for T. angustata).The respective virulences, assessed as lesion lengths, had no relationships with optimum growth temperatures in cultures.Several abiotic factors, including plant drought stress or water excess after climate events, or increases in average temperatures, can play roles in disease development, and may influence pathogen wood colonization and virulence.It is well known that global warming and climate change can increase plant stress and generate favourable conditions for the development of many diseases, including grapevine trunk disease (Guarnaccia et al., 2023).
Several fungi, especially Botryosphaeriaceae, are known to be able to switch from endophytic to pathogenic behaviors as a result of triggers connected with environmental stresses, such as drought, extreme temperatures and nutrient deficits (Slippers and Wingfield, 2007).These fungi may therefore benefit from the ongoing global warming scenario.The high percentage of isolation of these pathogens from vineyards located in Piedmont suggests a shift of these fungi may also be occurring in northern regions, as has occurred in Mediterranean areas.This expansion may be related to climatic changes.Factors such as prolonged drought, high summer temperatures, and changes in agronomic practices could favor development, spread, and pathogenicity of these fungi.
Results from the present study have demonstrated the presence of well-known GTD pathogens in Piedmont, one of the most important wine-production regions in Italy.The diversity and virulence of these pathogens in Piedmont was previously unexplored.Association of P. brasiliense and N. juglandicola with grapevine wood in Italy has been reported for the first time.
This first survey in Piedmont aimed to determine the presence and distribution of Botryosphaeria dieback pathogens, and to investigate the occurrence of other fungi associated with symptomatic grapevines.Further research is required to better clarify the distribution of grapevine pathogens in Northern Italy, especially species of Botryosphaeriaceae, and to determine which are the GTDs pathogens present in this region and monitor their possible shifts following climate changes.
Table2.(Continued).Grapevine dieback in Piedmont: fungal agents characterization and pathogenicity transcribed spacers 1 and 2 together with 5.8S nrDNA; act: actin; tef1: translation elongation factor 1-α gene; rpb2: RNA polymerase second largest subunit; tub2: betatubulin.Sequences generated in this study indicated in italics.Ex-type and ex-epitype isolates are indicated in bold font; Genbank accession numbers generated in this study are indicated in italic font.CVG strains marked with * were selected to investigate effects of temperature on fungal radial growth.

Figure 1 .
Figure 1.Grapevine trunk disease symptoms observed in Piedmont.A. Symptoms attributed to Botryosphaeria dieback on a grapevine shoots, with complete branches dissection, drying, and fall of affected leaves.B. Cross section of a cordon with internal necrotic wood cankers (wedge-shaped) characteristic of Botryosphaeria dieback.

Figure 2 .
Figure 2. Phylogenetic tree for Botryosphaeriaceae, resulting from a Bayesian analysis of the combined ITS, tef1 and tub2 sequence alignment.Bayesian posterior probabilities (PP) and Maximum likelihood bootstrap support values (ML-BS) are indicated at the nodes (PP/ ML-BS).Ex-type strains are indicated in bold font, and species are delimited with coloured blocks.Isolates collected in the present study are indicated in red font.The tree was rooted to Lecanosticta acicola (LNPV252).

Figure 3 .
Figure 3. Phylogenetic tree of Neocucurbitaria sp., resulting from a Bayesian analysis of the combined ITS, tub2 and rpb2 sequence alignment.Bayesian posterior probabilities (PP) and Maximum likelihood bootstrap support values (ML-BS) are indicated at the nodes (PP/ML-BS).Ex-type strains are indicated in bold font, and species are delimited with coloured blocks.The isolate collected in the present study is indicated in red font.The tree was rooted to Pseudopyrenochaeta lycopersici (CBS 306.65).

Figure 4 .
Figure 4. Phylogenetic tree of Paraconiothyrium sp.resulting from a Bayesian analysis of the combined ITS, tef1 and act sequence alignment.Bayesian posterior probabilities (PP) and Maximum likelihood bootstrap support values (ML-BS) are indicated at the nodes (PP/ML-BS).Ex-type strains are indicated in bold font and species are delimited with coloured blocks.Strains collected in this study are indicated in red.The tree was rooted to Alloconiothyrium aptrootii (CBS 980.95).

Figure 5 .
Figure 5. Phylogenetic tree of Seimatosporium sp. and Truncatella sp.resulting from a Bayesian analysis of the combined ITS, tef1 and tub2 sequence alignment.Bayesian posterior probabilities (PP) and Maximum likelihood bootstrap support values (ML-BS) are indicated at the nodes (PP/ML-BS).Ex-type isolates are indicated in bold font and species are delimited with coloured blocks.Isolates collected in the present study are indicated in red font.The tree was rooted to Discosia artocreas (CBS 124848).

Table 3 .
Parsimony and Bayesian parameters for each phylogenetic analysis to identify different fungi.(P < 0.05) obtained with Games-Howell post hoc tests for optimum growth temperatures and Tukey's test for mycelial growth rates are shown in Figure 8.Among the Botryosphaeriaceae species, optimum mean growth temperatures ranged from 23.2°C for D. seriata and 23.9°C for D. mutila, to 25.5°C for Neof.parvum and 27.3°C for B. dothidea.Neocucurbitaria juglandicola did not grow 35°C.At the respective optimum temperatures, D. mutila grew the most rapidly at 10.9 mm d -1 , followed by D. seriata (10.5 mm d -1 ), B. dothidea (10.0 mm d -1 ) and Neof.parvum (8.7 mm d -1

Figure 6 .
Figure 6.Mean necrosis lengths (mm) in grapevine stems resulting from inoculations with different fungi.A Games-Howell post hoc test was performed and means accompanied by different letters are significantly different (P < 0.05).

Figure 8 .
Figure 8. Mean maximum growth rates (A) at respective optimum growth temperature (B) for eight fungi, Vertical bars indicate standard errors.Means in each histogram accompanied by different letters are significantly different (P = 0.05).

Table 1 .
Information on the vineyards surveyed and sampled in Piedmont, with respective fungi from different grapevine cultivars.

Table 2 .
List of isolates used for phylogenetic analyses, and their GenBank accession numbers.