Abstract

Materials and methods

Results

Molecular phylogeny

Of the 5188 nucleotide positions included in the four-loci sequence matrix, 1838 were parsimony informative (590 from ITS-LSU, 566 from rpb2, 682 from tef1). The best ML tree (lnL = –48924.555) revealed by the RAxML analysis is shown as phylogram in Fig. 1. Maximum parsimony analyses revealed 12 MP trees 9903 steps long (not shown). In both ML and MP analyses, Caudospora is placed within the Sydowiellaceae with high support, but its closest relatives remain uncertain due to lack of bootstrap support of most nodes within Sydowiellaceae.

Open in a separate window

Fig. 1

Phylogram of the best maximum likelihood tree (lnL = -48924.555) revealed by RAxML from an analysis of the combined ITS-LSU-rpb2-tef1 matrix of 75 selected members of Diaporthales, with two members of Magnaporthaceae (Gaeumannomyces graminis, Kohlmeyeriopsis medullaris) selected as outgroup. Familial classification follows Senanayake et al. (2017a). Caudospora (bold) is revealed as a member of Sydowiellaceae with high support. ML and MP bootstrap support above 50 % are given above or below the branches.

Of the 8683 nucleotide positions included in the nine-loci sequence matrix, 72 were parsimony informative (6 from SSU-ITS-LSU, 5 from cal, 15 from ms204, 9 from rpb1, 14 from rpb2, 15 from tef1, 8 from tub2). The MP analysis revealed a single tree 1559 steps long (Fig. 2); it is identical in topology to the best ML tree (lnL = –18010.352) revealed by RAxML. Sister group relationship of Caudospora iranica to C. taleola received maximum and high (93 %) bootstrap support in MP and ML analyses (Fig. 2), respectively, confirming their status as distinct species.

Open in a separate window

Fig. 2

Phylogram of the single MP tree 1559 steps long, obtained by PAUP from an analysis of the combined nine-loci matrix (SSU-ITS-LSU, cal, ms204, rpb1, rpb2, tef1, tub2) of Caudospora, with Hapalocystis berkeleyi selected as outgroup according to Fig. 1. MP and ML bootstrap support above 50 % are given at the first and second position, respectively, above or below the branches.

Taxonomy

Caudospora iranica Mehrabi & Voglmayr, sp. nov. – Figs. 3, 5a, b, e.

MycoBank no.: MB 823458

Type. – IRAN. East Azerbaijan Province, Arasbaran, on dead branches of Quercus sp., 11 July 2015, leg. M. Mehrabi (IRAN 16716 F holotype; WU 39950 isotype; ex-holotype culture IRAN 2552 C, ex-isotype culture CBS 143507).

Description. – Pseudostromata 1–3.5 mm diam., immersed in the bark of dead branches (ca. 1.5 cm thick), erumpent, circular to irregular, separate, scattered, sometimes confluent, delimited from surrounding bark by a black line, the latter visible on the bark surface. Ectostromatic discs white, convex, circular to oval, tissue between ostiolar necks distinctly gray to dark grey, powdery. Central column whitish to grey. – Ostioles dark, at the same level as the disc surface, rarely projecting, opening separately. – Perithecia 2–7 per stroma, 300–700 μm diam., subglobose to ovoid, black, monostichous, arranged circinately in brown entostromata, with long ostiolar necks converging towards the ectostromatic disc. – Paraphyses elongate, hyaline, filiform, septate, evanescent. – Asci (170)181–211(230) × 15–18 μm ( = 196 × 16.7 μm; n=20), cylindrical, with short or obsolete stalks and a conspicuous apical ring, containing eight uniseriate ascospores. – Ascospores (18)21–27(32) × (8)10.5–13(15) μm ( = 23.7 × 11.7 μm; n=42), l/w = (1.6)1.8–2.3(2.5), broadly ellipsoid, two-celled, constricted at the septum, hyaline or pale yellowish, coarsely verrucose, in SEM with isolated warts 0.4–1 μm diam., with a tubular gelatinous appendage 10–17 × 1–1.3 μm long at each end and 2–3 (rarely 4) similar median tubular gelatinous appendages arising from near the septum.

Cultural characteristics. – Colonies felty, with irregular margins, first white, later primrose (23,,b) at the surface, often deeply immersed into the agar, reverse primrose, becoming olivaceous black (27,,,,m) in the dark at 24 °C, reaching the margin of a 30 mm Petri-dish after 30 days at 24 °C. No asexual morph seen.

Etymology. – Referring to its origin from Iran.

Habitat. – On dead corticated twigs of Quercus sp.

Distribution. – Only known from the type collection in Iran.

Notes. – Caudospora iranica resembles C. taleola in stroma and ascoma morphology and ascospore appendages. However, there are pronounced differences between these species in the ascospore ornamentation. In C. iranica, the surface of the ascospore wall is covered with prominent isolated warts up to 1 μm diam. that are easily visible by light microscopy, whereas in the latter a finely verruculose ornamentation is scarcely visible in light microscopy without staining; the much smaller (up to 0.4 μm diam.) and densely disposed warts are clearly revealed only in SEM. In addition, asci (196 × 16.7 vs. 162 × 12.1 μm on average) and ascospores (23.7 × 11.7 vs. 23.9 × 9 μm on average) of C. iranica are wider than those of C. taleola.

Caudospora taleola (Fr.) Bih. K. Svenska Vetensk Akad. Handl., Afd. 15(2): 11 (1889). – Figs. 4, 5c, d, f–m.

Basionym. – Sphaeria taleola Fr., Syst. mycol. (Lundae) 2(2): 391 (1823).

Synonyms. – Aglaospora taleola (Fr.) Tul. & C. Tul., Select. fung. carpol. (Paris) 2: 168 (1863).

Chorostate taleola (Fr.) Traverso, Fl. ital. crypt. 1(2): 212 (1906).

Diaporthe taleola (Fr.) Sacc., Atti Soc. Veneto-Trent. Sci. Nat., Padova, Sér. 4 4: 112 (1875).

Hercospora kornhuberi Bäumler, Annln K. K. naturh. Hofmus. Wien 13: 440 (1898).

Hercospora taleola (Fr.) E. Müll., in Müller & von Arx, Beitr. Kryptfl. Schweiz 11(no. 2): 728 (1962).

Melanconis taleola (Fr.) Speg., Atti Soc. Crittogam. Ital., Série 2 3(1): 54 [no. 77] (1881).

Valsa taleola (Fr.) Fr., Summa veg. Scand., Sectio Post. (Stockholm): 391 (1849).

Type. – AUSTRIA. Oberösterreich, Raab, Wetzlbach, on dead branches of Quercus robur, 24 December 2015, leg. H.Voglmayr (WU 39951, neotype here designated, MBT381444; ex-neotype culture D185 = CBS 143508)

Description. – Pseudostromata 1–4 mm diam., immersed in the bark of dead branches (1–3 cm diam.), flat or erumpent, circular to irregular, separate, scattered, sometimes confluent, delimited from surrounding bark by a black line. Ectostromatic discs white, flat, circular to ovoid, tissue between ostiolar necks distinctly gray to dark grey, powdery. Central column whitish to grey. – Ostioles dark, at the same level as the disc surface, rarely projecting, opening separately, sometimes necks of perithecia united below the disc and discharging through a single ostiole. – Perithecia 2–16 per stroma, 300–700 μm diam., subglobose to ovoid, black, monostichous, arranged circinately in brown entostromata, with long ostiolar necks converging towards the ectostromatic disc. – Paraphyses elongate, hyaline, filiform, septate, evanescent. – Asci (140)145–180(207) × 10–13 μm ( = 163 × 12.1 μm; n = 20), cylindrical, with short or obsolete stalks and a conspicuous apical ring, containing eight uniseriate ascospores. – Ascospores (17)20–28(31) × (6.5)8–10(11) μm ( = 23.9 × 9 μm; n = 30), l/w = (2.1)2.3–3.0(3.4), ellipsoid, two-celled, constricted at the septum, hyaline, appearing smooth in light microscopy but distinctly verrucose in SEM with densely disposed warts 0.1–0.4 μm diam., with a hyaline tubular gelatinous appendage 6–17 × 1–1.5 μm long at each end and 2–3 (rarely 4) similar median tubular gelatinous appendages arising from near the septum.

Asexual morph on natural substrate acervular, phomopsis-like. – Conidiomata ca. 1 mm diam., visible as darker spots in surface view, showing the same organisation as the sexual morph, i.e. a central whitish stromatic column with a white ectostromatic disc and a brown pseudostroma delimited from surrounding bark by a black line, preceding the sexual morph; conidiogenous region in section pale olivaceous green. – Conidiophores branched, hyaline, septate, formed on the upper margin of the central column. – Conidiogenous cells phialidic, (6.5)11.5–18.5(26) × (1.4)1.7–2.5(3.5) μm ( = 14.8 × 2.1 μm; n = 70), hyaline, smooth. – Conidia (15)17–25 (32) × 0.7–1.1(1.6) μm ( = 21 × 0.9 μm; n=50), l/w = (14.8)18.2–29.7(37.2), filiform with variable shape from more or less straight, sigmoid, falcate, semicircular to circular, unicellular, hyaline, smooth.

Cultural characteristics. – Colonies felty, irregular, often deeply immersed into the agar, white at the surface and primrose (23,,b) to olivaceous black (27,,,,m) at the reverse in the dark at 24 °C, reaching the margin of 30 mm Petri-dish after 30 days at 24 °C.

Habitat. – On dead corticated twigs of Quercus sp.

Distribution. – Widely distributed in Europe, Asia, North America.

Other specimens examined. – AUSTRIA. Oberösterreich, St. Willibald, Gr. Sallet, on dead branches of Quercus robur, 26 December 2015, leg. H. Voglmayr (WU 39952, culture D186); St. Willibald, Geitzedt, on dead branches of Quercus robur, 31 December 2015, leg. H. Voglmayr (WU 39953, culture D187); Niederösterreich, Mannersdorf im Leithagebirge, Schweingraben, on dead branches of Quercus petraea, 12. March 2016, leg. H. Voglmayr (WU 39957, culture D216). FRANCE. Ardèche (07), Saint-Jean-Roure, Mandon, on dead branches of Quercus robur, 27 December 2015, leg. A. Gardiennet A.G. 15171 (WU 39955, culture D197); Saint-Jean-Roure, Grange de Sagne, on dead branches of Quercus robur, 28 December 2015, leg. A. Gardiennet A.G. 15173 (WU 39956, culture D198). IRAN. East Azerbaijan Province, Arasbaran, on dead branches of Quercus sp., 11 July 2015, leg. M. Mehrabi (IRAN 16715 F, WU 39954, culture IRAN 2551 C).

Notes. – No type specimen of Sphaeria taleola is preserved in the Fries collection at UPS (A. Kruys, pers. comm.), which necessitates neotypification. We here select a well-developed, abundant Austrian collection from the most likely original host, Quercus robur, as neotype, for which a culture and sequence data are available.

To our knowledge, C. taleola has not been previously been recorded from Iran; it is not listed in Ershad (2009). For comparison with C. iranica, see Notes there.

Discussion

Although morphologically undoubtedly a member of Diaporthales (Kirk et al. 2008), the phylogenetic affiliation and familial classification of Caudospora remained unclear up to now in the lack of DNA sequence data. Based on similarities in stromata and perithecia, Müller & Arx (1962) classified Caudospora in Hercospora (Lamproconiaceae), which was challenged by Rogers (1984), who recognized it to be substantially different from Hercospora in verrucose ascospores and a phialidic phomopsis-like asexual morph, vs. the holoblastic conidiation of the rabenhorstia-like asexual morph in Hercospora. However, although Rogers (1984) noted some similarities to the quercicolous Diaporthe leiphaemia var. raveneliana in verrucose ascospore ornamentation, he considered taxonomic rearrangements premature due to insufficient data. In our phylogenies, Caudospora is clearly revealed as a member of Sydowiellaceae (Fig. 1). Remarkably, already Munk (1957) noted a high similarity of centrum morphology of Caudospora to that of Sydowiella fenestrans, to which it is closely related.

After transfer of several lineages originally classified in other families of Diaporthales (e.g. Gnomoniaceae, Melanconidaceae), the Sydowiellaceae have in recent years become a rather heterogeneous assemblage which is difficult to define morphologically (Rossman et al. 2007). It currently comprises about 15 genera occurring as saprotrophs, endophytes and parasites of various herbaceous and woody hosts (Senanayake et al. 2017b). Most genera and species are characterized by reduced, poorly developed stromata containing few to a single perithecium, but there are few exceptions like Sillia and Chapeckia with well-developed multiperitheciate stromata. Also Caudospora has well-developed pseudostromata, which are delimited from the surrounding host tissue by a conspicuous black line, which is so far unique in Sydowiellaceae.

The molecular data reveal that Caudospora iranica and C. taleola are closely related, and they share a similar ecology and the same host genus, Quercus. However, they can be easily separated by ascospore morphology. Caudospora iranica has coarsely verrucose ascospores with isolated warts (Figs. 3h–j, 5a, b, e), whereas the ascospores of C. taleola appear smooth in light microscopy (Fig. 4g, h) but are shown to be finely verrucose in SEM (Fig. 5c, d, f). Ascospore ornamentation of C. taleola revealed by SEM in our investigations matches that illustrated in Rogers (1984: fig. 5) from a North American collection.

Open in a separate window

Fig. 3

Caudospora iranica (IRAN 16716 F, holotype). a. stromata on dead branch of Quercus sp.; b. ectostromatic disc; c. transverse sections of a pseudostroma; d. vertical section of a pseudostroma; e, f. asci; g. paraphyses; h–j. ascospores, showing coarsely verrucose ornamentation; k. culture on PDA. Bars: a 3 mm; b 300 μm; c, d 1 mm; e–g 30 μm; h–j 10 μm.

Open in a separate window

Fig. 4

Caudospora taleola (IRAN 16715 F). a. stromata on dead branch of Quercus sp.; b. ectostromatic disc; c. transverse sections of a pseudostroma; d. vertical section of a pseudostroma (showing two united necks under ectostromatic disc); e, f. asci; g, h. ascospores; i. culture on PDA. Bars: a 3 mm; b 300 μm; c, d 1 mm; e, f 30 μm; g, h 10 μm.

Open in a separate window

Fig. 5

a-f. SEM pictures of ascospores of Caudospora iranica (a, b, e) and C. taleola (c, d, f), showing the coarsely and finely verrucose ascospore ornamentation in C. iranica and C. taleola, respectively; the median smooth circular areas in c, d. represent the scars of median gelatinous appendages. g–m. Asexual morph of C. taleola. g, h. conidiomata in transverse section, showing the central stromatic column and olive-green conidial masses; i. conidiomata in vertical section, showing pseudostromata delimited by a black line, the central stromatic column and the olive-green conidial masses produced on the stromatic column; j–l. branched conidiophores with phialides producing filiform conidia (in j. with released elongate and circular conidia); m. conidia. a, b, e. IRAN 16716 F (holotype of C. iranica). c, d, f–m. WU 39951 (neotype of C. taleola). Bars: a–d 5 μm; e, f 1 μm; g 500 μm; h, i 200 μm; j–m 10 μm.

To our knowledge, the asexual morph of C. taleola is here described and illustrated for the first time in detail from natural substrate. We did not observe an asexual morph in our cultures, but Rogers (1984) obtained a phomopsis-like asexual morph on oatmeal agar, with unicellular, sigmoid to falcate hyaline conidia (5)10–17 × 1–1.5 μm in size produced on phialides. He compared it with the descriptions of Myxosporium taleola and Libertella taleola, which were formally described by Saccardo (1884) as putative asexual morphs of C. taleola, based on data reported by Tulasne & Tulasne (1863) and Fuckel (1871) from associations of sexual and asexual morphs on natural substrate. Rogers (1984) found some differences of his strain in conidial size; especially conidial width was substantially narrower in his isolate than that reported for Myxosporium taleola (16 × 8 μm) and Libertella taleola (20–30 × 4 μm). However, he was unable to resolve whether these differences related to strain differences, differences caused by conidiation in pure culture vs. on natural substrate, or erroneous connections of asexual morphs of species other than C. taleola. The asexual morph from natural substrate investigated by us agrees well with that reported by Rogers (1984) in its phialidic conidiogenous cells produced on short, branched conidiophores, conidial shape and conidial width, but our conidia are somewhat longer (15–32 vs. (5)10–17 μm in Rogers 1984), which may be caused by the different substrate. It therefore seems likely that the erroneous conidial width reported by Tulasne & Tulasne (1863) and Fuckel (1871) is due to misidentification.

Acknowledgements

References