Liberomycespistaciae sp. nov., the causal agent of pistachio cankers and decline in Italy

Abstract A new canker and decline disease of pistachio (Pistaciavera) is described from Sicily (Italy). Observations of the disease and sampling of the causal agent started in spring 2010, in the area where this crop is typically cultivated, Bronte and Adrano (Catania province) and later extended to the Agrigento and Caltanissetta provinces. Isolations from the margins of twig, branch and stem cankers of declining plants resulted in fungal colonies with the same morphology. Pathogenicity tests on 5-year-old potted plants of Pistaciavera grafted on P.terebinthus reproduced similar symptoms to those observed in nature and the pathogen was confirmed to be a coloniser of woody plant tissue. Comparison of our isolates with the type of the apparently similar Asteromellapistaciarum showed that our isolates are morphologically and ecologically different from A.pistaciarum, the latter being a typical member of Mycosphaerellaceae. Asteromellapistaciarum is lectotypified, described and illustrated and it is considered to represent a spermatial morph of Septoriapistaciarum. Multi-locus phylogenies based on two (ITS and LSU rDNA) and three (ITS, rpb2 and tub2) genomic loci revealed isolates of the canker pathogen to represent a new species of Liberomyces within the Delonicicolaceae (Xylariales), which is here described as Liberomycespistaciaesp. nov. (Delonicicolaceae, Xylariales). The presence of this fungus in asymptomatic plants with apparently healthy woody tissues indicates that it also has a latent growth phase. This study improves the understanding of pistachio decline, but further studies are needed for planning effective disease management strategies and ensuring that the pathogen is not introduced into new areas with apparently healthy, but infected plants.


Introduction
Cases of pistachio tree decline with gummosis, leaf canopy thinning and fruit losses have been observed for several years in the area of Bronte (Catania province, Sicily, Italy), which is considered the most typical area where high-quality pistachios are produced in Italy (http://www.dibartolosrl.it/bronte-pistachios/). Although pistachio is characterised by good rusticity, it is subject to several fungal diseases known to afflict pistachio trees in the Mediterranean area. Of these, the most commonly reported are phylloptosis, leaf spots mainly caused by Septoria pistaciae, S. pistaciarum and Pseudocercospora pistacina, gum cankers by Cytospora terebenthi and branch and twig cankers by Botryosphaeria dothidea (Chitzanidis 1995, Teviotdale et al. 2002, Vitale et al. 2007, Crous et al. 2013). The latter is widespread and already present as a latent pathogen in numerous plant communities in various parts of the world (Marsberg et al. 2017). Amongst soil-borne pathogens, Verticillium dahliae and Phytophthora spp. are reported to be particularly damaging in California (Holtz 2008). Moreover, recently a new blight was reported on pistachio fruit caused by Arthrinium xenocordella in the Agrigento province, southern Italy (Aiello et al. 2018).
From spring 2010 onwards, surveys have been carried out in 15 pistachio orchards of Catania, Agrigento and Caltanissetta provinces, Sicily, where declining trees were present. Declining plants showed twig, branch and stem cankers associated with vascular necrosis and tree decline. Abundant gummosis often occurred in association with cankered lesions. The cankered area resulted in localised, sunken lesions with several central cracks. After removing the bark, discolouration and necrotic tissue were evident and lesions deepened into the woody tissue. A coelomycetous fungus with pycnidial conidiomata was consistently isolated from these lesions.
The aims of this study were thus to investigate the aetiology of the decline syndrome observed in Bronte and to provide morphological, taxonomic, phylogenetic and pathogenic evidence of the causal organism which proved to be an undescribed species of Liberomyces, which was initially misidentified as Asteromella pistaciarum.

Field survey and isolation
Surveys of 15 pistachio orchards were conducted from 2010 to 2017 in Bronte and Adrano (Catania province, eastern Sicily) and Agrigento and Caltanissetta provinces (western Sicily). Approximately 10 samples per orchard showing cankered twigs and branches from declining pistachio plants were randomly collected for analysis (Fig. 1). Sub-cortical and wood fragments (about 5 × 5 mm) were cut from the lesion margins between affected and healthy tissues. In addition, from one orchard in Bronte, twigs were also sampled from asymptomatic pistachio plants. Subsequently, tissue pieces were disinfected by soaking in 70% ethanol for 5 s, 4% sodium hypochlorite for 90 s, rinsed in sterile water for 60 s and dried on sterile filter paper in a laminar flow cabinet. The fragments were placed on to 1.5% (w/v) malt extract agar (MEA, Oxoid, Basingstoke, UK) and 2% potato dextrose agar (PDA, Oxoid), incubated at room temperature (25 ± 5 °C) and examined for fungal growth. Numerous slow-growing cultures were obtained and single-conidial isolations were performed with conidia collected from pycnidia produced on those cultures within one month of incubation at room temperature under natural light conditions. More than 80 single-spore isolates were obtained from symptomatic and asymptomatic tissue isolations. Amongst these, 71 isolates were characterised by molecular and phylogenetic analysis (Table 1) and the four isolates ISPaVe1958, ISPaVe2105, ISPaVe2106 and ISPaVe2148 were considered for morphological, taxonomic and pathogenic studies. For a summary of sampling information of these isolates, see Suppl. Material 1.

Morphological characterisation
For morphological investigations, cultures were grown on MEA, PDA and 2% corn meal agar (CMA, Sigma-Aldrich) supplemented with 2% w/v dextrose (CMD). Moreover, pycnidial formation was assessed on artificially inoculated sterilised pistachio twigs incubated in a moist chamber. The isolates used in this study are maintained in the culture collections of the Dipartimento di Agricoltura, Alimentazione e Ambiente, University of Catania (PV) and of the CREA-DC (ex CREA-PAV), the extype isolate ISPaVe1958 of the new pistachio pathogen was deposited at the Westerdijk Fungal Biodiversity Institute (CBS), Utrecht, The Netherlands and the holotype specimen in the Fungarium of the Department of Botany and Biodiversity Research, University of Vienna (WU).
For investigations of temperature-growth relationships of the new pistachio pathogen, the holotype isolate ISPaVe1958 and the more recent isolate ISPaVe2148 were used. Agar plugs (5 mm diam.) were taken from the edge of actively growing cultures on MEA and transferred on to the centre of 9 cm diam. Petri dishes containing 1.5% MEA. Three replicate plates were incubated at 10, 15, 20, 25, 30 and 35 °C in the dark and measurements were taken after 21 d at right angles along two lines intersecting the centre of the inoculum and the mean growth rates plus and minus the standard deviation were calculated.
The holotype isolate ISPaVe1958 (CBS 128196) of the new pistachio pathogen and the type specimens of Asteromella pistaciarum deposited in the Natural History Museum of Vienna (W) were morphologically examined. For light microscopy, squash mounts and hand sections of pycnidia were made using a razor blade and observed in tap water or in 3% KOH. Methods of microscopy included stereomicroscopy using a Nikon SMZ 1500 equipped with a Nikon DS-U2 digital camera and light microscopy with Nomarski differential interference contrast (DIC) using the compound microscope Zeiss Axio Imager.A1 equipped with a Zeiss Axiocam 506 colour digital camera. Images were captured and measured with NIS-Elements D v. 3.0 or with the Zeiss ZEN Blue Edition software. For certain images of pycnidia, the stacking software Zerene Stacker v. 1.04 (Zerene Systems LLC, Richland, WA, USA) was used. Measurements are reported as maximum and minimum in parentheses and the range representing the mean plus and minus the standard deviation of a number of measurements given in parentheses.

Pathogenicity
Pathogenicity tests with four fungal strains of the undescribed pistachio pathogen were performed to satisfy Koch's postulates. Trials were carried out outdoors and in a growth chamber at 25 ± 1 °C. Potted 5-yr-old plants of Pistacia vera grafted on to P. terebinthus were used for artificial inoculations. Three plants for each isolate and six inoculation sites for each plant were considered.
Inoculations were made on stems and twigs after removing a 5 mm diam. bark disc with a cork borer, replacing it with a 5 mm plug from a 14-d-old PDA culture and covering it with sterile wet cotton, wrapped with parafilm (Pechney Plastic Packaging Inc., Chicago, USA) and aluminium foil to prevent contamination and desiccation. An equivalent number of plants and inoculation sites were inoculated with sterile PDA plugs as controls. The inoculated plants were observed every week. Symptom typology and the length of lesions were assessed after 12 months. To fulfil Koch's postulates, re-isolation was conducted following the same procedure as described above for isolations. Tissue fragments were plated on MEA or PDA and morphological and molecular identifications by sequencing the ITS rDNA were performed.

DNA extraction, PCR amplification and sequencing
The extraction of genomic DNA from pure cultures was performed as reported in previous studies (Voglmayr and Jaklitsch 2011) by using the DNeasy Plant Mini Kit (QIAgen GmbH, Hilden, Germany) or the Wizard Genomic DNA Purification Kit (Promega Corporation, WI, USA). For the ex-type strain of the new species, the complete internal transcribed spacer region (ITS1-5.8S-ITS2) and a ca. 0.9 kb fragment of the large subunit nuclear ribosomal DNA (nLSU rDNA) were amplified and sequenced as a single fragment with primers V9G (de Hoog and Gerrits van den Ende 1998) and LR5 (Vilgalys and Hester 1990); the complete ITS region of the other strains was amplified with primers ITS5 and ITS4 (White et al. 1990); the RNA polymerase II subunit 2 (rpb2) gene was amplified with primers fRPB2-5F2 and fRPB2-7cR (Liu et al. 1999, Sung et al. 2007 or dRPB2-5f and dRPB2-7r (Voglmayr et al. 2016); and the beta-tubulin (tub2) gene with primer pairs T1 andT22 or Tub2Fd andBt-2b (O'Donnell andCigelnik 1997, Aveskamp et al. 2009). The PCR product was purified using an enzymatic PCR cleanup (Werle et al. 1994) as described in Voglmayr and Jaklitsch (2008). DNA was cycle-sequenced using the ABI PRISM Big Dye Terminator Cycle Sequencing Ready Reaction Kit v. 3.1 (Applied Biosystems, Warrington, UK) with the same primers as in PCR; in addition, primers ITS4, LR2R-A ) and LR3 (Vilgalys and Hester 1990) were used for the ITS-LSU fragment. Sequencing was performed on an automated DNA sequencer (3730xl Genetic Analyser, Applied Biosystems).

Phylogenetic analyses
NCBI BLASTn searches of the ITS and LSU of the undescribed pistachio pathogen revealed members of Xylariales as closest matches. For phylogenetic analyses, two combined matrices were produced; GenBank accession numbers of the sequences used in the phylogenetic analyses are given in Table 1. A combined ITS-LSU matrix was generated to reveal the phylogenetic position of the undescribed pistachio pathogen within Xylariales. For this, representative GenBank sequences of Xylariales were selected from Jaklitsch et al. (2016) and supplemented with some additional GenBank sequences; six taxa of Sordariomycetes were added as outgroup. The second combined matrix contained three loci (ITS, rpb2, tub2) sequenced for 68 isolates of the undescribed pistachio pathogen; in addition, GenBank sequences of four accessions of Delonicicolaceae and of six additional members of Xylariales were added and two species of Diaporthales were used as outgroup Crous 2017, Voglmayr et al. 2017).
Maximum Likelihood (ML) analyses were performed with RAxML (Stamatakis 2006) as implemented in raxmlGUI v. 1.3 (Silvestro and Michalak 2012), using the ML + rapid bootstrap setting and the GTRGAMMAI substitution model with 1000 bootstrap replicates.
Maximum Parsimony (MP) analyses were performed with PAUP v. 4.0a161 (Swofford 2002), using 1000 replicates of heuristic search with random addition of sequences and subsequent TBR branch swapping (MULTREES option in effect, steepest descent option not in effect, COLLAPSE command set to MINBRLEN). Molecular characters were unordered and given equal weight; gaps were treated as missing data. Bootstrap analyses with 1000 replicates were performed in the same way, with 5 rounds of replicates of heuristic search with random addition of sequences and subsequent TBR branch swapping during each bootstrap replicate, with each replicate limited to 1 million rearrangements in the analysis of the three-loci matrix.

Field survey and isolation
Cankers and decline symptoms caused by the undescribed pistachio pathogen were detected in 10 orchards amongst the 15 investigated. The disease was primarily observed in the winter period and during late spring.
In the Bronte and Adrano areas (Catania province), symptoms were observed during the dormant season. Symptomatic plants showed gum exudation and often bark scaling on trunk and/or branches. When bark scaling occurred, it appeared as cracking and peeling of the bark. On trunks and large branches, cankers first appeared as visible dead circular areas that developed in the bark, which subsequently became dark and sunken. From that point onwards, infected areas expanded in all directions but much faster along the main axis of the stem, branch or twig. Under some environmental conditions, the host produced callus tissue around dead areas limiting the canker. Under the bark, cankers were characterised by discolouration and necrotic tissues and, in some cases, these extended to the vascular tissues and pith (Figs 1,2).
During the active growing season, the symptomatic plants also showed canopy decline. Inflorescences and shoots, originating from infected branches or twigs, wilted and died. When the trunk was girdled by a canker, a collapse of the entire plant occurred (Fig. 1).
More than 80 single-spore isolates were obtained from symptomatic and a few also from asymptomatic pistachio plants. Amongst these, 71 isolates were characterised by molecular phylogenetic analyses and 68 deposited at the Westerdijk Fungal Biodiversity Institute, Utrecht, Netherlands (Table 1).

Pathogenicity
The initial symptom, observed 3 weeks after artificial fungal inoculation, was gum exudation produced around the point inoculated. After removing the bark, a dark   discolouration and necrotic tissue were visible (Figs 2a, b). After 6 months, external cankers were observed in correspondence with the inoculated sites and small cracks were present in the sunken central area (Fig. 2c). After 12 months, symptoms were very obvious and similar to the cracked cankers observed in nature. Long and deep cracks were evident on the sunken area that defined the cankered lesion. After removing the bark, it was evident that the pathogen was able to colonise the wood and long discolourations were present (Figs 2d, e). After 12 months from inoculation, the length of lesions ranged from 12 to 45 mm. For ISPaVe1958 and ISPaVe2105, length lesions averaged 16.7 ± 3.0 and 31 ± 1.0 mm, respectively, while for IS-PaVe2106 and ISPaVe2148, 18 ± 0.0 mm and 29.7 ± 2.0 mm. Controls measured 4.0 ±1.0 mm in average. Cultures, morphologically identical with the new pistachio pathogen, were re-isolated from these cankers, fulfilling Koch's postulates. Moreover, ITS sequence comparison of these re-isolated cultures confirmed the species identity.

Growth rates
The growth rate experiments revealed 30 °C as optimal temperature for both isolates with an evidently better growth of the holotype ISPaVe1958 at this temperature in comparison to ISPaVe2148 (Fig. 3).

Phylogenetic analyses
Of the 1340 nucleotide characters of the ITS-LSU matrix, 519 are parsimony informative. The best ML tree (-lnL = 19486.775), revealed by RAxML, is shown as a phylogram in Fig. 4. Maximum parsimony analyses revealed 14 MP trees 4008 steps long (not shown). The backbone of the MP trees differs in several deeper unsupported nodes from the ML tree (not shown); notably in the MP tree, the Liberomyces clade was not the most basal node of Xylariales, although without support (not shown).
In the ML and MP analyses of the ITS-LSU matrix (Fig. 4), the Xylariales received maximum support, but backbone support within Xylariales was low to absent. The new species clustered within the Liberomyces clade, which was sister to Delonicicola siamense (Delonicicolaceae, Xylariales). The Delonicicolaceae received high support (100% ML and 99% MP bootstrap support), but their closest relatives remained unclear due to lack of significant backbone support in all deeper nodes (Fig. 4). The sister-group relationship of L. saliciphilus and L. macrosporus received moderate support (81% ML and 89% MP bootstrap support).
Of the 1941 nucleotide characters of the ITS-rpb2-tub2 matrix, 743 are parsimony informative (201 from ITS, 343 from rpb2, 199 from tub2). The best ML tree (-lnL = 12820.324), revealed by RAxML, is shown as a phylogram in Fig. 5. Maximum parsimony analyses revealed 6 MP trees 2669 steps long, with a tree backbone identical to that of the ML tree (not shown).
The analyses of the ITS-rpb2-tub2 matrix (Fig. 5) revealed similar topologies to the analyses of the ITS-LSU matrix. The Xylariales and Delonicicolaceae received high support in ML and MP analyses. The new pistachio pathogen formed a genetically homogeneous clade with high to maximum support, confirming that all isolates sequenced belong to the same species. As in the ITS-LSU analyses, it was placed as sister to the highly supported Liberomyces saliciphilus-L. macrosporus clade with moderate support.

Taxonomy
As a result of the morphological and molecular phylogenetic investigations, the undescribed pistachio pathogen is described as a new species, Liberomyces pistaciae. In addition, for comparison, a morphological re-description and illustrations are also provided for the apparently similar, little known pistachio pathogen, Asteromella pistaciarum, based on type material and it is recognised as a synonym of Septoria pistaciarum.   Figure 4. Phylogram of the best ML tree (-lnL = 19486.775) revealed by RAxML from an analysis of the combined ITS-LSU matrix of selected Xylariales, showing the phylogenetic position of Liberomyces pistaciae (bold) within Delonicicolaceae. ML and MP bootstrap support above 50% are given above or below the branches.
Culture characteristics. Colonies slow-growing (about 4 cm in diam. in 1 month on MEA, 4 cm in 2 weeks on CMD at 22 °C), initially white, turning pale to dark brown with age, with a whitish slightly lobed margin ( Fig. 6a and b), surface mycelium sparse. Red to brown pigments diffusing in growth medium. Densely aggregated pycnidia formed after 7 d on the inoculum plug, successively also on the colony surface.
Notes. The classification and description of the lectotype of Asteromella pistaciarum is here added as it is morphologically similar to Liberomyces pistaciae and the latter had therefore initially been misidentified as the former (see e.g. Pažoutová et al. 2012, who included a sequence of Liberomyces pistaciae as Asteromella pistaciarum in their phylogenies). In addition, Asteromella pistaciarum has not been addressed in previous taxonomic accounts. Two isotype specimens are located in the Natural History Museum of Vienna (W) from which W 1973-15537 is here selected as the lectotype based on preservation and abundance of the specimens. In the original description, Bremer and Petrak (1947) reported a close association of Asteromella pistaciarum with Septoria pistaciarum and an immature mycosphaerella-like sexual morph, which they consider to represent the same species. We agree with this treatment and consider Asteromella pistaciarum to be the spermatial morph of Septoria pistaciarum, the former therefore becoming a synonym of the latter based on priority.

Discussion
This study represents the first work determining the causal agent of cankers and decline of pistachio trees in Sicily, the major production area of Italy. In the field, severe symptoms of canker were observed on branches, shoots and trunks. In some cases, decline and death of host plants also occurred. The fungus almost exclusively isolated from these symptoms was Liberomyces pistaciae and the decline syndrome was strictly reproduced by artificial inoculation experiments. Seventy-one isolates recovered from different orchards over a 7-yr period were identified by molecular analysis. The molecular phylogenetic analyses (Figs 4,5) clearly demonstrate that the genus Liberomyces is affiliated with the Xylariales, which confirms the results of previous analyses , Perera et al. 2017. In both of our analyses, the genus Liberomyces is a sister group to Delonicicola siamense with moderate to high support, for which Perera et al. (2017) established a new family and order, Delonicicolaceae and Delonicicolales. However, if the order Delonicicolales is accepted, the Xylariales are unsupported in Perera et al. (2017) as well as in our phylogenetic analyses of the ITS-LSU matrix (Fig. 4). In the order Xylariales, insufficient backbone resolution and support of phylogenies based on ITS-LSU rDNA has been commonly observed (e.g. Voglmayr 2012, Jaklitsch et al. 2016), which often significantly increases if protein-coding genes like rpb2 and tub2 are considered (e.g. Voglmayr et al. 2018, Wendt et al. 2018). However, for most lineages of Xylariales, only ITS-LSU rDNA data are currently available. Remarkably, also in the phylogenetic analyses of Perera et al. (2017), which were inferred from a combined SSU, ITS, LSU and rpb2 matrix, internal support of Xylariales is absent if Delonicicolales are classified as a separate order. This fact may be due to lack of rpb2 sequence data for many lineages within Xylariales. Therefore, we consider the establishment of a separate order Delonicicolales premature and presently we propose the classification of Delonicicolaceae within Xylariales in which this family also fits morphologically, given its conidiomatal morphology and conidiogenesis.
Due to the pycnidial conidiomata and conidia of similar sizes, the current pistachio pathogen, here described as Liberomyces pistaciae, was initially identified as Asteromella pistaciarum, the true identity of which was unclear at that time. No sequence data are available for authentic material of the latter. However, a re-investigation of the type specimen of A. pistaciarum revealed substantial differences between both species, providing a clear distinction between the two organisms. While the type of A. pistaciarum has short reduced conidiophores with intercalary and terminal ampulliform phialides ( Fig. 7h-l), L. pistaciae has densely fasciculate conidiophores with verticillately arranged holoblastic, lageniform to cylindrical conidiogenous cells with sympodial conidial proliferation (Fig. 6j-q). In addition, the type of A. pistaciarum has distinctly more elongate conidia with a l/w of (2.8-)3.5-4.8(-6.1), compared to (2.0-) 2.7-3.5(-4.7) in L. pistaciae. Moreover, the disease symptoms are markedly different. The type collection of A. pistaciarum represents a foliar pathogen causing clearly delimited polyangular leaf lesions with gregarious subepidermal pycnidia on both sides of the leaf (Fig. 7a), whereas L. pistaciae causes a canker disease of stems and branches. Although no recent collections, sequence data or cultures are available for A. pistaciarum, its close association with Septoria pistaciarum and an immature mycosphaerella-like sexual morph on the holotype specimen, which was already noted in the original description (Bremer and Petrak 1947), provides strong evidence that A. pistaciarum represents the spermatial morph of Septoria pistaciarum and it is therefore here considered to be a synonym of the latter.
There are many fungal genera which can act as plant pathogens, but may behave also as latent pathogens, while closely related species are symptomless endophytes (Carroll 1988). This is apparently also the case in the pathogen Liberomyces pistaciae, which might have a latent phase within the host tissues since it was also isolated from asymptomatic pistachio plants. A latent phase represents a specific condition where the fungus can either develop symptoms or induce changes in the physiology of the host plant (Romero et al. 2001, Crous et al. 2015. Furthermore, Millar (1980) and Andrews et al. (1985) observed that certain latent pathogens become pathogenic when the host is stressed and this may be the case in L. pistaciae on pistachio trees in Bronte. In this regard, the ecology of its closest relatives, L. macrosporus and L. saliciphilus, is of interest, as they were isolated as bark and wood endophytes from several woody hosts , indicating that the primary ecology of the genus Liberomyces may be endophytic, from which the pathogenic L. pistaciae may have evolved. However, detailed studies are necessary to evaluate the influence of stress on pathogenicity of L. pistaciae.
On the basis of the high disease incidence and the frequency of this species observed in several orchards in the last years, we believe that L. pistaciae represents a menace to pistachio production in Sicily. As no epidemiological data are yet available, it is not possible to suggest any control strategies to avoid L. pistaciae infections. Nevertheless, the use and distribution of infected propagation material taken from nurseries and mechanical injuries or pruning wounds could play an important role in promoting the infections. The recent increase in importance of this and other diseases of pistachio in Sicily has stimulated further research and studies are in progress to extend the survey to other areas and to obtain important information to formulate effective disease management strategies.