The cup fungus Pestalopezia brunneopruinosa is Pestalotiopsis gibbosa and belongs to Sordariomycetes

Pestalopezia brunneopruinosa, the type species of Pestalopezia in Leotiomycetes, produces typical cup-shaped ascomata. Because its asexual morph has conidia comprised of five cells including apical and basal appendages and three pigmented median cells, it was first described as Pestalotia gibbosa, which belongs to Sordariomycetes. This contradiction has not been resolved due to the difficulty in isolating this fungus in culture. In this study, we isolated separate strains from the sexual morph and the asexual morph for molecular analysis. Phylogenetic trees of Sporocadaceae based on internal transcribed spacer, partial β-tubulin, and partial translation elongation factor 1-alpha sequence datasets revealed that both strains fall into the same taxon, in a clade in Pestalotiopsis sensu stricto alongside P. gaultheriae and P. spathulata. We provide the first evidence that fungi producing cup-shaped ascomata in Pestalotiopsis belong to Sordariomycetes, and we have proposed the transfer of Pestalopezia brunneopruinosa to Pestalotiopsis gibbosa.


Introduction
Pestalopezia brunneopruinosa (Zeller) Seaver is a leaf spot pathogen on salal (Gaultheria shallon Pursh) that produces asci on an apothecium as a sexual morph [1]. The asexual morph of Pestalopezia brunneopruinosa resembles that of Pestalotiopsis sensu lato (s. lat.) and was first described independently by Harkness [2] as Pestalotia gibbosa. Thus, it has been suspected that Pestalopezia brunneopruinosa and Pestalotia gibbosa are the same fungus, because the two fungi have been found in close proximity on the same leaves. Bonar [3] demonstrated that cultures from germinated ascospores of Pestalopezia brunneopruinosa produced conidia that were the same as that of Pestalotia gibbosa. Seaver [4] likewise concluded that Pestalopezia brunneopruinosa was the sexual morph of Pestalotia gibbosa. However, phylogenetic analyses of both fungi to clarify their relationship has not been previously conducted.
The genus Pestalotia was established by De Notaris [5]. Subsequently, Steyaert [6] split the genus Pestalotia into Pestalotia sensu stricto (s. str.) (conidia composed of 6 cells), Pestalotiopsis (5 cells) and Truncatella (4 cells), although many species were still retained in Pestalotia s. PLOS  lat. without reconsideration. Recently, Pestalotiopsis s. lat. was further split into three genera, Pestalotiopsis s. st., Neopestalotiopsis, and Pseudopestalotiopsis, based on morphology and molecular phylogeny [7]. These fungi belong to Sporocadaceae within Sordariomycetes [8]. The Harkness description of Pestalotia gibbosa conidia (three pigmented median cells in fivecelled versicoloured conidia, with septa darker than the rest of the cell), is similar to that of Neopestalotiopsis. However, the current taxonomic position of Pestalotia gibbosa is unclear, especially since the disposition of this fungus in Sordariomycetes was made without molecular data support. The sexual morph of Pestalotiopsis s. lat. was determined by Barr [9] to be Pestalosphaeria which produces three celled-ascospores and perithecial ascocarps. Réblová et al. [10] proposed using the name Pestalotiopsis rather than Pestalosphaeria as the currently accepted name, following recent botanical code changes, but there was no mention of the name Pestalopezia in this argument. Pestalopezia, Pestalotiopsis, and Pestalosphaeria, are, however, included in a "without-prejudice list of generic names of fungi for protection under the International Code" [11]. Pestalopezia brunneopruinosa, the sexual morph, was classified as a member of the Leotiomycetes [12] because it produces cup-shaped ascomata. Thus, the genus names of the sexual and asexual morphs are currently forced into different taxonomic classes. Beimforde et al. [13] conducted a phylogenetic analysis by combining fossil data and molecular data (18S rDNA, 28S rDNA, RPB1, and RPB2) and showed estimated lineages of both families diverged during the Permian or Carboniferous periods and Leotiomycetes and Sordariomycetes are sister clades. Their results indicate that these families, both of which produce inoperculate asci, are closely related in the molecular phylogenetic tree. However, there is no report that fungi belonging to Sordariomycetes can produce cup-shaped ascomata. The aim of this study was to clarify the taxonomic position of Pestalopezia brunneopruinosa with respect to Pestalotia gibbosa, and to determine the name for this fungus based on the concept of one fungus, one name [14,15].

Sample collection and isolation
Diseased leaves of salal (Fig 1) were collected from Sandcut Beach trail near Shirley, Vancouver Island BC, Canada in 2013. Several isolates that originated from single conidia in acervuli were cultured from diseased leaves. Isolates were also initiated from ascospores in an ascus, but ejected ascospores failed to individually germinate. Subsequent transfers from the ascus isolate were made from single conidia. Isolates obtained from the asexual morph: NOF 3175/ TAP13K_P3, and from the sexual morph: NOF 3176/TAP13K_ca_as2 were maintained on PDA (potato dextrose agar, Eiken, Tokyo, Japan) at 15˚C, examined to assess taxonomic position, and deposited in The Fungus Culture Collection of the Northern Forestry Centre, Edmonton, Alberta, Canada and Tamagawa University, Machida, Tokyo, Japan. A voucher specimen containing both apothecia and acervuli was deposited in the Pacific Forestry Centre Forest Pathology Herbarium (DAVFP 29689). Information of new combination in Nomenclature was deposited in the Mycobank (http://www.mycobank.org/defaultinfo.aspx?Page= Home: MB#824630).
To confirm the culture was isolated from the sexual morph, DNA was extracted from a single apothecium from DAVFP 29689 by CTAB [20], and ITS was amplified using our designed primer PES3 (5'-GGCCTACCCTGTAGCGCCTT-3') and ITS4.
Polymerase chain reaction (PCR) products were purified using ExoSAP-IT (GE Healthcare Japan, Tokyo, Japan) and sequenced using the ABI 310 DNA sequencer (ABI, Tokyo, Japan). These sequences have been deposited in the DNA Data Bank of Japan (https://www.ddbj.nig. ac.jp/index-e.html: accession numbers are shown in Table 1).
The results of the preliminary sequence homology search using BLAST were that the two Vancouver Island salal isolates, NOF 3175/TAP13K_P3 and NOF 3176/TAP13K_ca_as2, fell into Pestalotiopsis s. str. Additional sequence data for phylogenetic analysis were obtained from 7 other previously unpublished strains (listed in bold in Table 1), and 43 other strains published in previous studies [7,21]. To generate phylogenies based on ITS, β-tubulin, and tef1 sequences, Seiridium spp., members of Amphisphaeriaceae (outgroup) and Phlogicylindriaceae, were chosen because they are phylogenetically close to Sporocadaceae. The dataset of each genomic region (ITS, β-tubulin, and tef1) was aligned using MAFFT [22]. All positions containing gaps and missing data were deleted from the analysis. The strength of internal branches from the resulting tree was tested using the bootstrap analysis [23] with 1,000 replications.
Sequence data comprising the aligned dataset were subjected to maximum-likelihood (ML), neighbor-joining (NJ) and maximum-parsimony (MP) phylogenetic analyses using MEGA software Version 7 [24]. Molecular analyses using the ML method were performed using HKY+G+I nucleotide substitution model for ITS, β-tubulin, and tef1. Initial trees for the heuristic searches were automatically generated by applying the NJ and BioNJ algorithms to a matrix of pairwise distances estimated using the maximum composite likelihood approach and then selecting the topology with a higher log-likelihood value. Evolutionary history was inferred using the NJ method [25]. The tree was drawn to scale with branch-length units equivalent to those of the evolutionary distances used to infer phylogeny. Evolutionary distances were computed using the Kimura 2-parameter method [26] as the number of base substitutions per site. MP trees were generated using the tree-bisection-regrafting (TBR) algorithm and search level 3, which generates initial trees by randomly adding sequences (10  replicates). Consistency, retention, homoplasy, and composition indices were calculated for parsimony-informative sites. The resulting trees were printed using TreeView v. 1.6.6 [27] and, together with the alignments, deposited as S21431 in TreeBASE (https://www.treebase. org/treebase-web/home.html).

Morphological observations
Morphological observations were made from symptomatic salal leaves collected in 2013 (DAVFP 29689) and from a single dried herbarium specimen DAVFP 11308. The latter was collected in 1959, also from Vancouver Island, and determined as Pestalopezia brunneopruinosa by W. Ziller (S1 Fig). The asexual and sexual morphs were observed and measured in water using light microscopy (BX 51, Olympus Tokyo, Japan).

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 MycoBank from where they will be made available to the Global Names Index. The unique MycoBank number can be resolved and the associated information viewed through any standard web browser by appending the MycoBank number [urn:lsid:mycobank.org:Mycobank:824630] contained in this publication to the prefix http://www.mycobank.org/MB/. The online version of this work is archived and available from the following digital repositories:PubMed Central, LOCKSS.

Phylogenetic analysis
In addition to the Vancouver Island collections preliminarily identified as Pestalotia gibbosa (NOF 3175/TAP13K_P3, Culture from conidia) and Pestalopezia brunneopruinosa (NOF 3176/TAP13K_ca_as2, Culture from ascospores), a total of 52 strains, including Pestalotiopsis (30 strains with two obtained from sexual morphs), Neopestalotiopsis (11 strains), and Pseudopestalotiopsis (4 strains including one obtained from the sexual morph), were examined (accession numbers shown in Table 1). The sequence matrix used for phylogenetic analyses contained at least 1258 nucleotide positions for final data set from sequences 550 bp of ITS, 560 bp of β-tubulin, and 530 bp of tef1. In ML method, the highest log-likelihood was -6657.97. The optimal tree generated using the NJ method had a branch-length of 0.665. An MP tree had a length of 909, consistency index of 0.547, retention of 0.87 and composite index of 0.509. Only the ML tree (Fig 2) is shown here, because the ML, NJ, and MP methods generated similar topologies. Pestalotia gibbosa and Pestalopezia brunneopruinosa were placed in the same clade with Pestalotiopsis gaultheriae (ML/NJ/MP: 100/100/100). Pestalotiopsis spathulata was also closely placed to Pestalotia gibbosa and Pestalopezia brunneopruinosa with highly supported bootstrap values (ML/NJ/MP: 100/100/99). Furthermore, the ITS sequence obtained from Pestalotia gibbosa (NOF 3175/TAP13K_P3) and Pestalopezia brunneopruinosa (NOF 3176/TAP13K_-ca_as2) were the same as the ITS sequence obtained from DNA extracted directly from an apothecium of DAVFP 29689 (epitype specimen) (S2 Fig).

Morphological comparisons
Our observations of the apothecia from DAVFP 11308 and 29689 are similar to those of Seaver's description [4] of Pestalopezia brunneopruinosa, with few exceptions. Seaver's ascospore measurements were slightly larger than the Vancouver Island DAVFP (VI) specimens at 7-10 x 14-20 um, plus we observed in the VI collections that mature ascospores eventually darken to brown rather than remaining hyaline (Fig 3, S3 Fig). We also observed a ring-shaped ascus apparatus in DAVFP 29689 which stained blue in Melzer's reagent, but only in scattered mature asci. These morphological variations are relatively minor and likely reflective of the state of maturity of Seaver's material (S1 Table). We also compared our observations and measurements of the conidial states of DAVFP 11308 and 29689 from leaves to published descriptions and specimens of conidia of Pestalopezia brunneopruinosa, Pestalotiopsis gaultheriae, and P. spathulata (Table 2). With the exception that P. spathulata has fewer and longer appendages [7], all are morphologically very similar. Brannigan. Ex-epitype NOF 3176/TAP13K_ca_as2 was isolated from a conidium transferred from a colony originating from a single ascus. Ascocarp: Apothecium developing on the upper surface of pale tan to light brown necrotic areas of attached living leaves, sessile or with short stalk approximately 0.5-2 mm in diameter, cup-shaped, with a wood brown to yellowish brown furfuraceous exterior. Hymenium fuscous when immature, becoming black at maturity because of the dark tips of paraphyses forming the epithelium; asci: 115-150 μm in length (including a short stalk) × 11-15 μm in diameter (n = 20), eight-spored, unitunicate, cylindrical, with slightly pointed apex, apical apparatus ring-shaped and staining blue in Melzer's reagent, but only when fully mature; ascospores:

15-50 knobbed
Pestalotiopsis spathulata [7] 24-32 × 7.  [1].This description did not mention the color of mature ascospores. DAVFP 11308 collected by Ziller (as Pestalopezia) in 1959 contains mature ascomata and is in sufficiently intact state to observe brownish ascospores. However neither sample was suitable for DNA extraction, and hence we established an epitype. Since obtaining cultures that originate from single ascospores is difficult, we initiated our culture (NOF 3176/TAP13K_ca_as2) from a monoconidial isolate that was obtained from hyphae grown from ascospores of a single ascus. We were able to germinate single ascospores ejected from mature ascocarps onto agar, but the resulting germinants failed to grow beyond an initial germ tube. We designated the epitype of Pestalotiopsis gibbosa as DAVFP 29689. We consider P. gaultheriae Y.M. Zhang, Maharchch. & K.D. Hyde [28] to be a synonym of P. gibbosa, but the authors [28] were unable to obtain living cultures from the specimen of P. gaultheriae.

Discussion
Our morphological observations and sequence results confirm that Pestalopezia brunneopruinosa and Pestalotia gibbosa are the same fungus. Conidia of Pestalotia gibbosa are strikingly similar to those of Neopestalotiopsis species because the three median cells of the conidia are versicoloured, and they could be classified into the genus Neopestalotiopsis based on morphology. However, in this study, we demonstrate by genomic analysis that P. gibbosa should be transferred to Pestalotiopsis s. str., even though its sexual morph is an apothecium.
The majority of the more than 200 species associated with the well-known genus Pestalotiopsis s. lat. are typified by the asexual morph, while only a few (14) have known sexual states producing perithecial ascocarps typified by the genus Pestalosphaeria [7,21]. Réblová et al. [10] have recommended use of Pestalotiopsis rather than Pestalosphaeria, but this recommendation did not take into consideration the potential of either Neopestalotiosis or Pseudopestalotiopsis also having teleomorphs genetically related to Pestalosphaeria; and the small (three known species), obscure genus Pestalopezia was not mentioned at all in this recommendation. All species of Pestalosphaeria were considered to be linked to Pestalotiopsis s. str. after the three genera Neopestalotiopsis, Pseudopestalotiopsis, and Pestalotiopsis were separated from Pestalotiopsis s. lat. [7]. Silvério et al [29] in 2016 and Nozawa et al. [17] in 2017, found the sexual morphs of Neopestalotiopsis and Pseudopestalotiopsis, both in agreement with the description of Pestalosphaeria. Hence, they reported that Pestalotiopsis s. str., Neopestalotiopsis, and Pseudopestalotiopsis produce the same sexual morph. However, the relationship of these fungi to Pestalopezia, characterized by the production of apothecia, was not considered in these works. In this study, we obtained strains from conidia of Pestalotia gibbosa and from ascospores of