Cytospora paraplurivora sp. nov. isolated from orchards with fruit tree decline syndrome in Ontario, Canada

A new species of Cytospora was isolated from cankered wood of Prunus spp. during a survey of orchards exhibiting symptoms of fruit tree decline syndrome in southern Ontario, Canada. We found isolates that are morphologically similar to species in the Cytosporaceae family, which is characterized by single or labyrinthine locules, filamentous conidiophores or clavate to elongate obovoid asci and allantoid, hyaline conidia. Multi-gene phylogenetic analysis of ITS, LSU, act and tef1- α showed that the isolates form a distinct clade, sister to Cytospora plurivora. Morphologically, our isolates showed differences in the length of conidia and culture characteristics compared to C. plurivora, suggesting the establishment of a new species. The species is described as Cytospora paraplurivora sp. nov. and placed in the family Cytosporaceae of Diaporthales. Additionally, we sequenced, assembled and characterized the genome of the representative isolate for this new species. The phylogenomic analysis confirms the species order and family level classification. C. paraplurivora sp. nov. has the potential to severely affect stone fruits production, causing cankers and dieback in stressed trees, and eventually leads to tree decline. Pathogenicity tests show that the species is pathogenic to Prunus persica var. persica.


Introduction
Stone fruit trees are economically important crops cultivated in Ontario, Canada. The Niagara Peninsula produces more than 90% of its peaches, nectarines and apricots [1]. Peach occupy the largest production areas, followed by nectarines and apricots. Fruit tree decline syndrome (FTDS) has been recently discovered in several orchards in this area of the province. The common symptoms include stem canker and dieback, wilting, increased suckering, leaf discolouration and eventual total collapse of the tree. Disease incidence ranged from 49 to 72% for apricots and nectarines (Fig 1). Similar disease symptoms were observed in stone fruits and apple trees in other regions of North America [2,3]. A number of studies reported that Cytospora spp. was among the causative agents of canker and dieback diseases in fruit trees including Prunus spp. [4,5]. Generally, members of the Cytospora spp. infect trees that are stressed by extreme weather conditions (drought or freezing), and by invading wounds in the bark caused by insect damage and improper pruning. These fungi commonly overwinter in a form of pycnidia embedded in the bark of cankered branches [6]. The genus Cytospora (Cytosporacea, Diaporthales) was established by Ehrenberg in 1818. There are currently 676 records of Cytospora spp. registered in the IndexFungorum (www. indexfungorum.org; accessed Nov 2021). However, 163 species of the genus with available nucleotide sequences are listed in the NCBI taxonomy database (www.ncbi.nlm.nih.gov/ taxonomy; accessed May 2022). Approximately, 150 species in the genus have been reported to cause diseases in more than 120 woody plants, which can result in significant commercial losses for growers [7][8][9]. The taxonomic species identification of Cytospora spp. is mainly based on morphological characteristics and molecular phylogenetic analysis. Recent studies described a number of new species of Cytospora on different hosts using multi-gene (ITS, LSU, act, tef1-α, tub2, and rpb2) phylogenetic analysis [5,[10][11][12]. This approach allows for the identification of cryptic and novel species within the genus. Obtaining high-quality, cost-effective genomic data is now possible through rapidly developing sequencing technologies. Wholegenome sequence data used in phylogenomics can contribute to resolving taxonomic uncertainties or support species reclassifications [13]. One such study found that a polyphagous plant pathogen, Corynespora olivacea was previously misclassified on family level [14].
The objective of this paper was to identify and characterize a novel Cytopora species associated with the decline of Prunus armeniaca, P. persica var. persica, P. persica var. nucipersica in Ontario, Canada. The characterization of the new species was performed following recently published guidelines [15,16].

Sample collection and isolation
Wood samples were collected in 2018-2021 from 30 apricot, 6 peach and 6 nectarine trees exhibiting extensive tree fruit decline symptoms from nine commercial orchards located in southern Ontario. Roots were symptom-free. The cankered and diseased wood sections from trunk and rootstock were cut into 1 cm pieces, and surface disinfested with 70% ethanol for 30 sec, followed by 1% NaClO for 20 min and three rinses in sterile distilled water. The samples were air-dried and placed on a 2% potato dextrose agar (PDA, Difco, USA) supplemented with kanamycin (50 mg L−1). The PDA plates were incubated at 22˚C for 5 days in the dark. All fungal colony-forming units were hyphal-tip transferred to individual PDA plates and incubated at 22˚C for 7 days in the dark. Purified mycelial isolates were classified into morphotypes prior to molecular identification. C. paraplurivora was characterized by fast-growing, white to cream with uneven lobate growth margin colonies.
Tree branches with fruiting structures (conidiomata) were checked separately. Single conidia isolations were performed using the protocol described by Chomnunti et al. [17]. For long-term preservation, fungal cultures were stored at −80˚C in 30% glycerol. The holotype specimen was deposited in the Canadian National Mycological Herbarium (DAOM) and the living culture collection maintained by the Canadian Collection of Fungal Cultures (DAOMC).

Morphological examination
Conidiomata formed on the tree branches were described prior to sectioning with a sterile surgical scalpel. The macro-morphological structures were measured using a dissecting microscope (OMAX 2000X Infinity Compound Siedentopf Microscope with a Built-in Camera). In order to calculate the mean size of the structures, 10 conidiomata, 20 conidiogenous cells and 50 conidia were measured. The obtained measurements were recorded with minimum and maximum values and means were calculated. Radial growth of fungal colonies was estimated with two perpendicular measurements after 5 days of incubation. Morphology description [18] and color characterisation [19] were performed after 7 days.

DNA extraction, PCR amplification and Sanger sequencing
Genomic DNA was extracted from mycelium using the Plant/Fungi DNA Isolation Kit (Norgen Biotech Corp., Thorold, Canada) with the following modifications: fungal tissue was vortexed for 15 minutes with 1 mm glass beads and 500 μL lysis buffer and 1 μL RNase A prior to incubation at 65˚C. After incubation on ice, the fungal mixture was centrifuged at 10,000 rpm to separate the lysate from the beads and biomass. During the column wash, the resin was dried by spinning for 10 minutes at 14,000 rpm. DNA was eluted at 10,000 rpm for 2 minutes. PCR amplifications were executed using C1000 Touch PCR thermal cycler (Bio-Rad, Hercules, USA) under conditions described in the references for each region. The internal transcribed spacer (ITS) region was amplified with the primer pair ITS1/ITS4 [20]. The primer pair LR0R/ LR5 [21] was used to amplify the large subunit rRNA gene (LSU). The partial actin (ACT) region was amplified with the primer pair ACT512F/ACT783R [22], and the primer pair EF1-728F/ EF1-986R [22] was used to amplify partial translation elongation factor 1-alpha (tef1-α) gene sequences. The quality of the PCR products was examined using electrophoresis in 1% agarose gel. Sanger sequencing was carried out at Genome Quebec's Sequencing Facility (Montreal, Canada).

Sequence alignment and phylogenetic analysis
The consensus sequences were built from Sanger chromatograms using BioEdit [23]. The ITS, LSU, act, tef1-α sequences from our study were uploaded on GenBank (http://www.ncbi.nlm. nih.gov/) on the Basic Local Alignment Search Tool (BLAST) [24] to find other similar sequences. The sequences of Cytospora spp. from the top BLAST hits were then extracted and added to recently published sequence datasets [5,9]. Diaporthe vaccinii (CBS 160.32) of Diaporthaceae was used as the outgroup. All the sequences were initially aligned using CLUS-

DNA isolation, genome sequencing and characterization
Highly purified genomic DNA from the isolate FDS-564 was extracted from fungal mycelium grown in a liquid Potato Dextrose Broth for 5 days at room temperature, filtered and freezedried before following the modified DNA extraction protocol as previously described (Norgen Biotech Comp., Thorold, Canada). The library was prepared using a SMRT bell Express Template Prep Kit (PacBio, Menlo Park, USA). The library pool was sequenced on one SMRT cell using the PacBio Sequel II platform in the SickKids sequencing facility (Toronto, ON, Canada). The reads were assembled into contigs with the Canu assembler v.

Phylogenomic analysis and genome alignment
The proteomes of 17 species of ascomycetes, including five species in the Cytosporaceae, were retrieved from the Mycocosm portal [42] (accessed May 2022) and the NCBI Genome database (accessed Dec 2021). Orthofinder v. 2.5.4 [43] was used to identify the single-copy orthogroups  [46] and displayed as a dot-plot graph using D-Genies [47].

Excised branch pathogenicity trials
Excised branches collected from non-symptomatic apricot, nectarine and peach trees were used to test the pathogenicity of C. paraplurivora isolates following the protocol by Arzanlou and Narmani [62]. Thirteen green lateral branches, with a mean diameter of 1 cm and 20 cm in length, were excised from each tree species. The leaves were removed, and the branches were surface disinfested with 70% ethanol for 10 min, rinsed 3 times with sterile water and airdried. Ten branches were wounded and inoculated with 4 mm mycelium agar plug from a 5-day old cultures of each isolate and wrapped with Parafilm (Parafilm 1 "M", MilliporeSigma,

PLOS ONE
Canada). Three controls were inoculated with sterile PDA plugs. Excised branches were placed inside a clear plastic container with moist paper towels and incubated at room temperature in the dark. The branches were checked 12 days post-inoculation to measure the necrotic lesions.
Longitudinal sections were made from the inoculation point for further lesion examination.

Living plants pathogenicity trials
The peach seedlings (cv. Loring) were grown from seeds and placed in a biosafety plant growth chamber under a temperature of 23˚C and a photoperiod of 16 h. A completely randomized design was used for the experiment. Inoculations were performed on four-month-old seedlings. A total of five seedlings were wounded and inoculated with 4 mm mycelium agar plug from a 5-day old cultures of each isolate and wrapped with Parafilm (Fig 7). Five seedlings were inoculated with sterile PDA plugs as negative control. The developing lesions were measured 7, 10, 13 and 27 days post-inoculation (dpi).

Re-Isolation of the pathogen
C. paraplurivora was re-isolated from the excised branches and seedlings stem fragments and from fruiting structures (conidiomata). Three tissue fragments (0.5 cm long) were taken from the advanced necrosis and from the lesion edge of each branch and seedling. The tissue fragments were surface disinfested with 70% ethanol for 30 sec, followed by 1% NaClO for 20 min and three rinses in sterile distilled water. The samples were air-dried and placed on a 2% PDA supplemented with kanamycin (50 mg L −1 ). The PDA plates were incubated at 22˚C for 5 days in the dark. All fungal colony-forming units were hyphal-tip transferred to individual PDA plates and incubated at 22˚C for 7 days in the dark. One conidiomata was taken from each infected seedling and the pathogen was re-isolated using the protocol described by Chomnunti et al. [17]. The identification of C. paraplurivora re-isolated from lesions and fruiting structures was made on the basis of morphological features using reference cultures and PCR with the primer pair EF1-728F/ EF1-986R [22], as described above, to fulfil Koch's postulates.

Nomenclature
The electronic version of this article in Portable Document Format (PDF) in a work with an ISSN or ISBN will represent a published work according to the International Code of Nomenclature for algae, fungi and plants, and hence the new names contained in the electronic publication of a PLOS ONE article are effectively published under that Code from the electronic edition alone, so there is no longer any need to provide printed copies. In addition, new names contained in this work have been submitted to Index Fungorum and MycoBank from where they will be made available to the Global Names Index. The unique Index Fungorum number can be resolved and the associated information viewed through any standard web browser by appending the Index Fungorum number contained in this publication to the prefix www.indexfungorum.org/. The online version of this work is archived and available from the following digital repositories: PubMed Central, LOCKSS.

Results
The decline disease has affected up to 72% of the trees in the surveyed stone fruits orchards in southern Ontario. C. paraplurivora was the only Cytospora spp. isolated from symptomatic trees in all surveyed orchards. The incidence of C. paraplurivora on symptomatic trees was 50% in peaches, 57% in apricots and 100% in nectarines.

Phylogenetic analysis
The ML and BI analyses based on a concatenated alignment of ITS, LSU, act and tef1-α from 119 strains of Cytosporaceae produced phylogenetic trees with similar topology to those in recent taxonomic studies of Cytospora [5,9,12]. The best scoring RAxML tree is depicted in Fig 2. The strains of C. paraplurivora obtained in this study are clustered together and formed a distinct clade with strong bootstrap support values (100% ML, 1.00 BP). The closest species, C. plurivora is grouped with the new species in the well supported clade (100% ML, 1.00% BP). Broadly, the taxa with leucocytosporoid condiomata (C. amydgali, C. davidiana, C. erumpens, C. gigaspora, C. leucostoma, C. nivea, C. paratranslucens, C. rusanovii, C. sorbicola, C. translucens) also constitute a separate clade (88% ML, 1.00 BP).

PLOS ONE
Culture characteristics: colonies on PDA are fast-growing, reaching 6.2 cm in diameter after 5 days at 25˚C in the dark; initially white turning olivaceous after 5-7 days with thick texture at center, uneven lobate growth margin, lacking aerial mycelium; irregular, abundant pycnidia develop after 14 days. Hyphae hyaline, smooth, branched and septate.
Notes: based on morphological characteristics and phylogenetic analysis, C. paraplurivora is closest to C. plurivora isolated from different hosts in California (USA). This species is found to be associated with a canker disease of fruit (including Prunus spp.) and nut trees. Both species have common morphological characteristics but conidia of C. paraplurivora are notably longer than those of C. plurivora (6.4 × 1.3 μm versus 4.1 × 1.0 μm). Lack of olivaceous color of the C. plurivora cultures grown on PDA and incubated at the same conditions also differentiates these species [5].

Genome assembly
A total of 6 854 783 single-end long reads (NCBI SRA accession no. SRR17267775) were generated from PacBio sequencing of FDS-564 and assembled into 211 contigs (> 1000 bp in size), with an N 50 of 658,741 bp, and a genome size of 39.7 Mb, which is within the size range of other species of Cytosporaceae (35.2 Mb for C. mali var. pyri SXYL134 to 41. 9 Mb for the representative genome of C. mali 03-8). Assessment of the completeness of the genome using BUSCO groups for fungi resulted in values of C:98.9% [D: 1.0%], F: 0.1%, n: 1 706 (C:complete [D:duplicated], F:fragmented, n: number of genes) which makes the genome sequence suitable for further analysis. The repeat content was 2.9% in the assembly of C. paraplurivora FDS-564. A total of 10,249 protein-coding genes were predicted, which is 9.2% fewer genes than predicted for the genome of C. mali 03-8 (11,284).

Phylogenomic analysis
A total of 3,482 orthogroups present in all species included in the analysis with 2,279 SCOs. The phylogenomic tree (Fig 4A) constructed from a concatenated amino acid alignment of SCOs showed that C. paraplurivora was closest to C. leucostoma SXYLt. This analysis placed C. paraplurivora within the Cytosporaceae family of the order Diaporthales with strong bootstrap support (100% ML). However, whole genome sequence alignment between C. paraplurivora and C. leucostoma indicated rather low similarity, where 84.31% of the syntenic blocks were less than 50% similar with a number of small gaps and some repeats (Fig 4B).

Pathogenicity tests
Twelve days post-inoculation, the length of the necrotic lesions caused by C. paraplurivora on apricot, peach and nectarine excised branches were 22, 26 and 29 mm respectively. Lesions were significantly larger in nectarine cuttings compared to those in apricot (Fig 5A). No significant differences were observed between the mean lesion length caused by C. paraplurivora isolates FDS-564 and FDS-623 on peach seedlings 7, 10, 13 and 27 dpi (Fig 6). The canker lesions enlarged on both sides of inoculated incision. The infected peach seedlings start wilting 15 dpi and collapsed 30 dpi. The symptoms observed in the inoculated cuttings and seedlings, included flattening and discolouration of the bark, light brown wedge-shaped cankers, gummosis and abundant fruiting structures (conidiomata) development were similar to those observed on trees in the field (Figs 5B-5D and 7). C. paraplurivora was re-isolated from 90-100% of the inoculated cuttings and seedlings, fulfilling Koch's postulates. Control cuttings and seedlings had no symptoms, and the fungus was not isolated from the wood.

Discussion
Cytospora spp. are important plant pathogens with worldwide distribution causing canker and dieback diseases in woody plants. The genus has a broad host range that includes Castanea, Corylus. Eucaliptus, Juglans, Malus, Olea, Pistacia, Platanus, Prunus, and Salix. [5,10,12,63,64]. Pathogenic species of Cytospora were mainly isolated from wood cankers. The present study identifies C. paraplurivora sp. nov. isolated from both single conidia and cankered wood sampled from symptomatic apricot, nectarine, and peach trees in southern Ontario. Abundant conidiomata of C. paraplurivora sp. nov. were observed on trunk and branches of declining trees. C. paraplurivora sp. nov. spores dissemination mechanism might be similar to that of the sister species C. plurivora [65], other Cytospora spp. and fungal species causing Botryosphaeriaceae canker [66,67]. The presence of conidiomata suggest that this species might spread across fruit tree orchards through spores dissemination by rain splash, wind and insect [65]. The symptoms caused by the new species resemble the leucostoma canker disease (perennial canker) of stone fruits, a widespread plant disease affecting crop yield in Canada and some regions of USA. C. leucostoma and C. cincta were found to be associated with this disease [68] and the species were identified solely by morphological characteristics. The studies on Cytospora spp. with the ITS sequence data refined the species identification of the leucocytosporoid

PLOS ONE
species [69]. However, the later studies revealed that the protein-coding genes included in phylogenetic analysis allow for discrimination of closely related species of Cytospora [70]. In this study, using the sequence data of act and tef1-α genes suggested by Urbez-Torres et al. [39] improved the phylogenetic resolution of this species complex. This approach has also been extensively used to resolve phylogenies of other plant-associated fungal pathogens such as Diaporthe [71], Valsaria [72] and Diplodia [73].
Phylogenomic analysis using the genome sequence data of C. paraplurivora FDS-564 confirmed order and family level classification of the species. It is interesting to note, that the genome size of the new species is slightly larger than the average for Ascomycota (36.9Mb) [41] but significantly smaller compared to plant pathogens of Diaporthales such as Daiporthe citri 63.6Mb [74]. The high-quality genome assembly of C. paraplurivora will provide a valuable resource for the further study of host-pathogen interactions, pathogenicity-related genes, fungal biology, as well as comparative and population genomics analyses of Cytospora spp. and other fungal taxa.
The pathogenicity tests of C. paraplurivora on Prunus spp. performed in this study can be comparable to the trials of other pathogenic species of Cytospora on Malus spp. conducted in vitro under the same conditions with different timelines [75]. The phylogenetically close species C. paratranslucens caused necrotic lesions up to 10 cm on detached apple branches after 21 days [75]. The lesions up to 3.9 cm developed on twig segments of Malus sieversii inoculated with mycelium of C. parasitica after 8 days. This study showed that, highly virulent model species C. mali was able to cause the lesions up to 8.1 cm for the same time period [76]. Considering the time of post-inoculation with C. paraplurivora, it is assumed that the species can quickly colonize the plant tissue and cause the decline of infected trees. The pathogenicity tests on living peach seedlings confirmed the rapid spread of the canker lesions on the trunk of the inoculated plants, leading to the collapse of peach seedlings 30 dpi.
This paper is the first formal report for a new Cytospora species causing the canker of Prunus spp. associated with FTDS in Ontario. The findings suggest that C. paraplurivora has the potential to severely affect stone fruits production in Ontario. Accurate identification of pathogen(s) associated with stone fruit trees decline will support management of the disease. Further research needs to be conducted on emerging fungal pathogens and their associated infection mechanisms, to develop effective disease management strategies applicable to fruit tree orchards.