﻿A new Arthrinium-like genus of Amphisphaeriales in China

﻿Abstract Species of Arthriniums. l. are usually known as endophytes, pathogens or saprobes occurring on various hosts and substrates and are characterised by globose to subglobose, sometimes irregular, dark brown and smooth-walled or finely verruculose conidia, always with a truncate basal scar. Currently, Arthriniums. l. contains two phylogenetically distinct clades, namely, Apiospora and Arthriniums. s. However, Arthriniumtrachycarpi and Ar.urticae have still not been properly classified. With new isolates from diseased leaves of Lithocarpusglaber collected in China, we propose the new Arthrinium-like genus Neoarthrinium in Amphisphaeriales. Based on the morphology and phylogeny of multiple loci, the new genus is established with the type species, N.lithocarpicola and three new combinations, N.moseri (syn. Wardomycesmoseri), N.trachycarpi (syn. Ar.trachycarpi) and N.urticae (syn. Ar.urticae) are added to this genus.


Introduction
Apiosporaceae, including Arthrinium-like taxa, was proposed to accommodate genera with apiosporous hyaline ascospores and a basauxic, Arthrinium-like conidiogenesis (Hyde et al. 1998). In a recent outline of Sordariomycetes, Hyde et al. (2020) a combined matrix of ITS, LSU, tef1 and tub2 sequences. The aim of this study was to determine the phylogenetic placement of Ar. trachycarpi, Ar. urticae and our new isolates within Amphisphaeriales, which resulted in the identification of a new phylogenetic lineage with isolates belonging to neither Arthrinium nor Apiospora. As a result, a new genus is established for these isolates.

Isolation and morphology
Diseased leaves of Lithocarpus glaber were observed and collected in Guangdong Province of China (39 m elevation; 23°8'52"N, 113°27'18"E), packed in paper bags and transferred to the laboratory for pure culture isolation. The samples were first surfacesterilised for 1 min in 75% ethanol, 3 min in 1.25% sodium hypochlorite and 1 min in 75% ethanol, rinsed for 2 min in distilled water and blotted on dry sterile filter paper. Then, the diseased areas of the leaves were cut into 0.5 × 0.5 cm pieces using an aseptic razor blade, transferred on to the surface of potato dextrose agar plates (PDA; 200 g potatoes, 20 g dextrose, 20 g agar per litre) and incubated at 25 °C to obtain pure cultures. The cultures were deposited in the China Forestry Culture Collection Center (CFCC; http://cfcc.caf.ac.cn/) and the specimen was deposited in the Herbarium of the Chinese Academy of Forestry (CAF; http://museum.caf.ac.cn/).
The morphology of the isolates was studied, based on sporulating axenic cultures grown on PDA in the dark at 25 °C. The conidiomata were observed and photographed under a dissecting microscope (M205 C, Leica, Wetzlar, Germany). The conidiogenous cells and conidia were immersed in tap water and then the microscopic photographs were captured with an Axio Imager 2 microscope (Zeiss, Oberkochen, Germany), equipped with an Axiocam 506 colour camera using differential interference contrast (DIC) illumination. For measurements, 50 conidiogenous cells and conidia were randomly selected. Culture characteristics were recorded from PDA after 10 d of incubation at 25 °C in the dark.

DNA extraction, PCR amplification and phylogenetic analyses
Genomic DNA was extracted from colonies grown on cellophane-covered PDA using a cetyltrimethylammonium bromide (CTAB) method (Doyle and Doyle 1990). DNA was checked by electrophoresis in a 1% agarose gel and the quality and quantity were measured using a NanoDrop 2000 (Thermo Scientific, Waltham, MA, USA). The following primer pairs were used for amplification of the gene regions sequenced in the present study: ITS1/ITS4 for the ITS1-5.8S-ITS2 nrDNA region (ITS) (White et al. 1990); LR0R/LR5 for the 28S nrDNA region (LSU) (Vilgalys and Hester 1990); EF1-728F/EF2 for the translation elongation factor 1-α (tef1) gene (O'Donnell and Cigelnik 1997;Carbone and Kohn 1999); Bt2a/Bt2b for the beta-tubulin (tub2) gene (Glass and Donaldson 1995). The PCR conditions were set as follows: an initial denaturation step of 5 min at 94 °C, followed by 35 cycles of 30 s at 94 °C, 50 s at 52 °C (ITS and LSU) or 54 °C (tef1 and tub2) and 1 min at 72 °C and a final elongation step of 7 min at 72 °C. The PCR products were assayed via electrophoresis in 2% agarose gels. DNA sequencing was performed using an ABI PRISM 3730XL DNA Analyser with a BigDye Terminator Kit v.3.1 (Invitrogen, USA) at the Shanghai Invitrogen Biological Technology Company Limited (Beijing, China).
The phylogenetic analyses of the combined loci were performed using Maximum Likelihood (ML) and Bayesian Inference (BI) methods. The ML was implemented on the CIPRES Science Gateway portal (https://www.phylo.org) using RAxML-HPC BlackBox 8.2.10 (Stamatakis 2014), employing a GTRGAMMA substitution model with 1000 bootstrap replicates. The Bayesian posterior probabilities (BPP) were determined by Markov Chain Monte Carlo (MCMC) sampling in MrBayes v. 3.2.6 (Ronquist et al. 2012). The six simultaneous Markov chains were run for 1 M generations, starting from random trees and sampling trees every 100 th generation and 25% of aging samples were discarded, running until the average standard deviation of the split frequencies dropped below 0.01. The phylogram was visualised in FigTree v.1.3.1 (http://tree.bio.ed.ac.uk/software) and edited in Adobe Illustrator CS5 (Adobe Systems Inc., USA). The newly-generated nucleotide sequences were deposited in GenBank (Table 1).

Phylogenetic analyses
The combined sequence dataset (ITS, LSU, tef1 and tub2) was analysed to infer the phylogenetic placement of our new isolates within Amphisphaeriales. The dataset consisted of 136 sequences, including two outgroup taxa, Clypeosphaeria mamillana (CBS 140735) and Pseudosporidesmium knawiae (CBS 123529). A total of 3526 characters, including gaps (793 for ITS, 859 for LSU, 762 for tef1 and 1112 for tub2), were included in the phylogenetic analysis. Of these characters, 1543 were constant, 284 were variable, but parsimony-uninformative and 1699 were parsimony-informative. The best ML tree (lnL = -72640.48) revealed by RAxML is shown in Fig. 1. The topologies resulting from ML and BI analyses of the concatenated dataset were congruent (Fig. 1). Isolates CFCC 54456 and CFCC 55883 from the present study, together with CFCC
Culture characters. Colonies on PDA flat, spreading, with flocculent aerial mycelium forming concentric rings, edge entire, mouse grey to greyish-green, reaching 60 mm diam. after 10 d at 25 °C, forming abundant conidiomata.

MycoBank No: 844772
Basionym. Wardomyces moseri W. Gams, Beih. Sydowia 10: 67 (1995) Notes. Based on a placement within Xylariales in phylogenetic analyses, Sandoval-Denis et al. (2016) excluded this species from the genus (Microascales); however, they did not suggest an alternative generic classification. The blastic hyaline, smooth, lageniform conidiogenous cells aggregated in clusters and the subglobose to ellipsoid dark brown conidia with a longitudinal germ slit (Gams 1995) fully matched the genus Neoarthrinum. The ITS, LSU and tub2 sequences of the ex-holotype strain of N. moseri (CBS 164.80) are almost identical to those of N. trachycarpi, indicating that they may be synonymous. Both species were isolated from petioles of palms: N. moseri from Mauritia minor Burret in Colombia and N. trachycarpi from Trachycarpus fortunei (Hook.) H.Wendl. in China. However, the two species were reported to differ in conidial size (10-14 × 3-4.5 μm in N. moseri vs. 6.1-8.5 × 4.2-5.8 μm in N. trachycarpi;Gams 1995;Yan et al. 2019) and for the time being, we therefore kept them separate.  Ellis, Mycol. Pap. 103: 16 (1965) Notes. The possibility that Apiosporella urticae (Rehm) Höhn. is the sexual morph of Arthrinium urticae is raised by the fact that both share the same host (Urtica) and are classified as members of the Apiosporaceae (Index Fungorum, accessed 4 July 2022). This evidence would have far reaching nomenclatural consequences not only for species, but also for generic classification, as Apiosporella (Höhnel 1909) may then qualify for an older genus name to be used for Neoarthrinium. However, according to L. Holm, the holotype specimen of its basionym, Apiospora urticae (S-F12119), represents a very different fungus, Didymella eupyrena (Didymellaceae, Pleosporales, Dothideomycetes; https://herbarium.nrm.se/specimens/F12119, accessed 4 July 2022). The status of the genus Apiosporella is still unclear because Höhnel (1909) did not choose a type from the six different species included in the genus. However, none of the original species is a close relative of Apiosporaceae or Neoarthrinium; therefore, Apiosporella should be excluded from Apiosporaceae.
No sequence data are available for isolates from the type host Urtica dioica L. (Urticaceae). The single culture sequenced (IMI 326344) was isolated from unidentified leaf litter collected in India. Additional molecular studies on verified isolates from Urtica collected in Europe are necessary to reveal whether IMI 326344 represents true N. urticae. However, N. urticae appears to be very rare and we are unaware of any additional collections with the exception of the type.

Discussion
Arthrinium and related genera are important fungal taxa whose concepts and classification have undergone many changes and additions (e.g. Cooke 1954;Samuels et al. 1981;Larrondo and Calvo 1990;Hyde et al. 1998;Jaklitsch and Voglmayr 2012;Crous and Groenewald 2013;Singh et al. 2013;Sharma et al. 2014;Dai et al. 2016Dai et al. , 2017Hyde et al. 2016;Jiang et al. 2018Jiang et al. , 2020Wang et al. 2018;Pintos et al. 2019;Pintos and Alvarado 2021). In recent years, substantial changes in classification were implemented in the course of unitary nomenclature. A large number of newly-discovered species have been described as a result of extensive sampling of new isolates, based on multigene phylogenies (e.g. Crous and Groenewald 2013;Wang et al. 2018;Pintos and Alvarado 2021). Currently, Arthrinium-like asexual morphs are shared by three distinct lineages within Amphisphaeriales, viz. Apiospora, Arthrinium s. s. and Neoarthrinium as shown in Fig. 1. Arthrinium s. s. is the sister genus to Nigrospora, which morphologically differs from Apiospora, Arthrinium and Neoarthrinium in conidial ontogeny (Wang et al. 2017). The phylogram shown in Fig. 1 is consistent with that shown in Tian et al. (2021) in placing Apiospora, Arthrinium and Nigrospora within a clade that is distinct from the new genus Neoarthrinium, although Apiospora and Arthrinium share conidial morphology similar to that of Neoarthrinium.
Morphologically, Apiospora, Arthrinium and Neoarthrinium are similar in having basauxic conidiogenesis. Conidia of Apiospora and Neoarthrinium are generally more or less rounded in face view and lenticular in side view, while those of Arthrinium are variously shaped, viz. globose, angular, polygonal, curved, fusiform or navicular (Yan et al. 2019;Pintos and Alvarado 2021). However, the conidiophores of several Arthrinium and Neoarthrinium species have thick blackish septa, which are rarely observed in Apiospora (Ellis 1965;Wang et al. 2018;Pintos and Alvarado 2021). Hence, these three genera are difficult to distinguish by only asexual morphology.
Regarding their hosts, there are some tendencies in host preferences, while Arthrinium species are predominantly found in Cyperaceae and Juncaceae (Pintos and Alvarado 2021) and species of Apiospora primarily occur on Poaceae (but also on many other hosts; Wang et al. 2018). Four Neoarthrinium species were discovered on four hosts from three distantly-related host families (i.e. N. lithocarpicola from Lithocarpus glaber (Thunb.) Nakai, Fagaceae; N. moseri from Mauritia minor Burret, Arecaceae; N. trachycarpi from Trachycarpus fortune (Hook.) H.Wendl., Arecaceae; and N. urticae from Urtica dioica L., Urticaceae; Ellis 1965;Yan et al. 2019). Hence, host association is not a fully reliable feature to distinguish Apiospora, Arthrinium and Neoarthrinium.
Compared to species, generic delimitation is much more subjective. However, there is a broad agreement that genera, along with all taxonomic classification units at all ranks, should be monophyletic. As morphology is frequently insufficient for phylogenetic classification, molecular evidence is regarded as significant data or even an essential characteristic in the classification and identification of fungal taxa. In the present study, Neoarthrinium is proposed as a new genus for a group of species phylogenetically distinct from Apiospora, Arthrinium and Nigrospora to maintain monophyletic Arthrinium-like genera. Using morphological and phylogenetic data, however, we need more samples to improve our understanding of Arthrinium-like taxa and genera in the Amphisphaeriales.