Revision of the genus Aciculosporium (Clavicipitaceae) with a description of a new species on wavyleaf basketgrass, and proline-containing cyclic dipeptide production by A. take

Abstract

The genus Aciculosporium (Clavicipitaceae, Hypocreales, Ascomycota)was established in 1908 for A. take, which is the causal fungus of witches’ broom of bamboo. Although the original description was valid at that time, a type specimen for A. take has not been designated. To standardize the use of this genus and species name, a neotypification and reference specimen of A. take are proposed. Multilocus phylogenetic analyses based on DNA sequences from 28S rDNA, TEF, Tub2, Mcm7, and RPB2 revealed that A.sasicola is from a different lineage to A. take, and other specimens from wavyleaf basket grass (Oplismenus undulatifolius) represent a distinct species proposed here as Aciculosporium oplismeni sp. nov. Chemical analysis using mass spectrometry and nuclear magnetic resonance spectroscopy showed that A. take produces four proline-containing cyclic dipeptides, which are moieties of ergot alkaloids. However, ergot alkaloids, lolines, peramine, indole-diterpenes, and lolitrem were not detected in the culture solvent. This study offers clarification of the lineage and morphology of this genus.

1. Introduction

The genus Aciculosporium is one of the phytopathogenic clavicipitaceous genera (Clavicipitaceae, Hypocreales, Ascomycota). The type species Aciculosporium take I. Miyake is the causal agent of witches’ broom in bamboo (Fig. 1). Unfortunately, no type material was designated in the original description to fix the precise application of the name (Miyake, 1908). Aciculosporium take lives endophytically in the vegetative shoots of bamboo (Poaceae) and forms conidiostromata and sessile ascostromata on the apices of infected twigs (Tanaka, 2009). In culture, the species produces two-celled holoblastic macroconidia with two dichotomously branched apical appendages that separate by breaking at the middle septum and germinate from the basal end (Tsuda et al., 1997). The appendaged conidia of Aciculosporium can be distinguished from those of other Clavicipitaceae genera. Molecular phylogenetic studies of the Clavicipitaceae have revealed that members of the genus Aciculosporium are closely related to Claviceps species, but many characteristics of A. take are quite different from those of Claviceps. For example,Clavicepsspecies are non-systemic pathogens, which generally produce enteroblastic Sphacelia-type conidia, which are associated with the honeydew production. They often possess holoblastic secondary conidia (Pažoutová et al., 2004) and infect the florets of grasses, sedges, and rushes.

Fig. 1 - Aciculosporium take. A: Witches' broom symptom on Phyllostachys bambusoides. B: Witches' broom symptom (arrows) on Phyllostachys pubescens (TNS-F-60468). C: Ascostroma on shoot apex of P. bambusoides. D–F: Asci. D: Intact ascus. E: Eight filiform ascospores emerged from an ascus. F: Apical hemispherical thickened tip of asci. G: Colony (MAFF 241224) on PDA at 25˚C for 1 mo. H: Apical appendagedconidia. D–F, H: These spores were stained with lactophenol cotton blue. Bars: C, G 1 cm; D, E, H, 50 µm; F10 µm.

Three species with apical branched appendaged conidia have joined the genus Aciculosporium during the 21^st century, although A. take had been the lone species in this genus for over 90 years. One of these newly described species is Aciculosporium sasicola Oguchi, a pathogen causing witches’ broom in small bamboo Sasa sect. Sasa (Arundinarieae, Bambusoideae) species in Japan(Oguchi, 2001). Even though A. take also parasitizes many bamboo species (Arundinarieae, Bambusoideae ), A. sasicola is considered to be a separate species, as the morphological characteristics of A. sasicola differ slightly from those of A. take. However, whether they are conspecific or not has not been discussed based on molecular phylogenetic analysis. Later, the multilocus phylogenetic study has shown that A. take, Cepsiclava phalaridis (J. Walker) J. Walker (formerly known asClaviceps phalaridis J. Walker), and Neoclaviceps monostipa J. White, G. Bills, S. Alderman & J. Spatafora form a monophyletic clade (Píchová et al., 2018) and transferred to Aciculosporium. Aciculosporium monostipum(J. White, G. Bills, S. Alderman & J. Spatafora) M. Kolařík & Píchová infects individual florets of the panicoid grass in Costa Rica; it replaces the host ovules, produces stipitate ascostromata directly (i.e. without formation of any stroma or sclerotium) from the parasitized ovaries, and has one type of anamorph, represented by conidia with apical appendages (Sullivan et al., 2001). Aciculosporium phalaridis (J. Walker) M. Kolařík & Píchováis a systemic endophyte of pooid grasses in southeastern Australia; it forms sclerotia in florets, produces stipitate ascostromata from the sclerotia, and has two anamorphs, one of which produces holoblastic conidia with apical branched appendages (Walker, 2004).

Many phytoparasitic clavicipitaceous fungi produce alkaloids, which probably play a role in the surrounding ecology. For example, some Claviceps species produce ergot alkaloids and indole-diterpenes that may protect ergots. However, in the phylogenetically related A. phalaridis, no ergot alkaloids were found in stromata or cultures (Píchová et al., 2018). The production of indole-3-acetic acid and related compounds were confirmed in A. take cultures (Tanaka et al., 2003). Genome analysis has indicated that A. take does not contain a gene cluster encoding the biosynthesis of ergot alkaloids, lolines, or peramine but contains a cluster of biosynthesis genes for indole-diterpenes or lolitrem (Schardl et al., 2013). However, this gene cluster is predicted to be nonfunctional, because the geranylgeranyl diphosphate synthase gene, encoding the first step in the biosynthetic pathway, is a pseudogene (Schardl et al., 2013). As A. take cannot produce ergot alkaloids, lolines, indole-diterpenes or peramine, this species may produce different secondary metabolites with ecological roles. Therefore, the alkaloids or related compounds produced by this fungus should be analyzed.

In the present study, we summarized recent knowledge about Aciculosporium, revised its taxonomy using newly obtained collections and analyzed secondary metabolites of A. take. To do this, we conducted a multilocus phylogenetic analysis and a nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry (MS)-based chemical analysis of a culture solution containing A. take.

2. Materials and methods

2-1.Specimen sampling, isolation, and morphological studies

Scientific names of the bamboo species followed the recent taxonomic study of Kobayashi (2017). Most specimens and cultures of A. take were previously used in our studies of Tsuda et al. (1997),Tanaka et al. (2002) and Tanaka (2009)Fresh specimens of A. sasicola (Fig. 2) were collected in Hokkaido (type locality), Gifu and Kyoto, Japan. Although the A. sasicola culture obtained from Hokkaido did not survive, the culture obtained from Kyoto was deposited in the NARO Genebank (National Agriculture and Food Research Organization, Japan) as MAFF 246967. A fresh specimen of Aciculosporium sp. ex Oplismenus undulatifolius (Fig. 3) was obtained from the Institute for Nature Study at Tokyo, Japan, and deposited at the mycological herbarium of the National Museum of Nature and Science (Tsukuba, Japan; TNS-F-87159; Table 1). The ascospores of the Aciculosporium sp. ex O. undulatifolius were spread on potato dextrose agar in Petri dishes and incubated at 18 °C in the dark. Three months after inoculation, a colony made of conidia was isolated. The culture was deposited in the NARO Genebank as MAFF 246966. Sections of ascostroma were prepared as described previously (Tanaka et al., 2020). We could not analyze A. monostipum, because ATCC culture of A. monostipum (ATCC MYA-621) is no longer available. Bright-field images were obtained using a differential interference contrast microscope (E-800 or ECLIPSE Ni, Nikon, Tokyo, Japan).

Fig. 2 - Aciculosporium sasicola. A: Witches’ broom symptom on Sasa palmata (TNS-F-91306). B: Conidiostroma on shoot apex of S. palmata (TNS-F-91306). C: Colony (MAFF 246967) on PDA at 25˚C for 1 mo. D: Apical appendagedconidia stained with lactophenol cotton blue. Bars: B,C 1 cm; D 50 µm.

Fig. 3 - Aciculosporium oplismeni. A, B: Ascostroma on shoot apex of Oplismenus undulatifolius (TNS-F-87159). C: Cross section of ascostroma stained by periodic acid-Schiff reaction. D–F: Asci. D: Unmatured ascus. E: Apical hemispherical thickened tip of ascus. F: Matured ascus with germinated ascospores. G: Colony (MAFF 246966) on PDA at 25˚C for 1 mo. H: Apical appendaged conidia stained with lactophenol cotton blue. C, D, F: Multiple images were combined into a single image using GraphicConverter (Remke Software, Peine, Germany). Bars: A, G 1 cm; B 5 mm; C 100 µm; D, F, H 50 µm; E 10 µm.

Table 1 Aciculosporium species and its allies used in this study. Data obtained for this study are in bold font.

				GenBank accession numbers
Species	Specimen / strainb	Host	Location	ITS	28S rDNA	MCM7	TEF	TUB2	RPB2
A. take	MAFF 241224	Phyllostchys pubescens	Japan, Ishikawa, Kanazawa	LC571753	LC571753	LC572027	LC572034	LC572041	LC572048
A. take	TNS-F-60465	Phyllostchys pubescens	Japan, Kagoshima, Kajiki	LC571755	LC571756	LC572028	LC572035	LC572042	LC572049
A. take	MAFF 241223	Phyllostachys bambusoides	Japan, Kyoto, Sakyo	AB065423	LC571754	LC572029	LC572036	LC572043	LC572050
A. take	TNS-F-60469	Phyllostachys bambusoides	Japan, Kagoshima, Amami	LC571756	LC571756
A. take	Ma-Uji	Phyllostachys bambusoides	Japan, Kyoto, Uji	AB065423
A. take	Ma-Okayama	Phyllostachys bambusoides	Japan, Okayama, Takahashi	AB065423
A. take	Ma-Chiba	Phyllostachys bambusoides	Japan, Chiba, Sakura	AB065423
A. take	Ginmeichiku	Phyllostachys bambusoides var. castillonis-inversa	Japan, Tokyo, Chiyoda	AB065424
A. take	Inyouchiku	Hibanobambusa tranquillans	Japan, Kyoto, Sakyo	AB065423
A. take	Okamezasa	Shibataea kumasasa	Japan, Kyoto, Nishigyo	AB086846
A. take	Nezasa	Pleioblastus gramineus	Japan, Kyoto, Uji	AB065422
A. take	Nambusuzu	Neosasamorpha shimidzuana	Japan, Gifu, Yoro	AB066292
A. take	ZJHZ	Arundinaria fargesii	China, Zhejiang	MK874908
A. take	B1	Pyllostachys pubescens	China, Guangxi	EF363682
A. take	B2	Pyllostachys pubescens	China, Guangxi	EF363683
A. monostipum	INBio 6-141	unknown panicoid grass	Costa Rica		AF245293		AY986983
A. oplismeni	MAFF 246966a	Oplismenus undulatifolius	Japan, Tokyo, Minato	LC571760	LC571760	LC572033	LC572040	LC572047	LC572054
A. phalalidis	CCC 293	Phalaris tuberosa	Australia, New South Wales	AJ133399	AJ133399	LT216474	LT216524	FJ711476	LT216598
A. sasicola	MAFF 246967	Sasa palmata	Japan, Kyoto, Nantan	AB066293	LC571757	LC572030	LC572037	LC572044	LC572051
A. sasicola	TNS-F-60466	Sasa senanensis	Japan, Hokkaido, Sapporo	LC571758	LC571758	LC572031	LC572038	LC572045	LC572052
A. sasicola	MAFF 247297	Sasa palmata	Japan, Gifu, Hida	LC571759	LC571759	LC572032	LC572039	LC572046	LC572053
A. sasicola	MAFF 247298	Sasa palmata	Japan, Ishikawa, Hakusan	LC589716
Corallocytostroma ornithocopreoides	WAC 8706	Astrebla pectinata	Australia, Western Australia	AJ557075	AJ557075	LT216496	LT216546	FJ711475	LT216620
Claviceps africana	CCC 489	Sorghum bicolor	Mexico, Guanajuato	LT216447	LT216447	LT216466	LT216515	FJ711458	LT216562
Claviceps arundinis	CCC 956	Phragmites australis	Lithuania, Aukshaitia	JX083477	JX083477	JX083546	JX083684	JX083408	JX083615
Claviceps fusiformis	CCC 525	Pennisetum glaucum	Zimbabwe, Shamva	AJ626727	AJ626727	LT216456	LT216506	EF473867	LT216576
Claviceps humidiphila	CCC 434 a	Dactylis sp.	Germany, Phillipsreuth	JX083497	JX083497	JX083566	JX083704	JX083428	JX083635
Claviceps maximensis	CCC 816	Panicum coloratum	USA, Texas	FJ686000	FJ686000	LT216471	LT216521	FJ711506	LT216590
Claviceps purpurea	CCC 771 a	Secale cereale	Czech Republic, Bezdědice	JX083524	JX083524	JX083593	JX083731	JX083455	JX083662
Claviceps pussila	CCC 602	Bothriochloa insculpta	Zimbabwe, matopos	FJ685996	FJ685996	LT216476	LT216526	FJ711490	LT216599

^a Ex-type

^b Abbreviations for collections: CCC, Culture Collection of Clavicipitaceae, Institute of Microbiology, Academy of Sciences of the Czech Republic, Prague; INBio, Instituto Nacional de Biodiversidad, Costa Rica; MAFF, Ministry of Agriculture, Forestry and Fisheries of Japan, Tokyo, Japan; TNS-F, the mycological herbarium of the National Museum of Nature and Science, Tsukuba, Ibaraki, Japan; WAC, Western Australian Plant Pathology Reference Culture Collection, Australia, Perth

2-2. DNA extraction, amplification, and sequencing

DNA was extracted from culture according to the protocol described byIzumitsu et al. (2012). PCR was conducted using KOD FxNeo or KOD One DNA polymerase (Toyobo, Osaka, Japan). The PCR products were purified and then sequenced using an ABI PRISM 3130 genetic analyzer (Applied Biosystems, Foster City, CA). Sequencing reactions were conducted using the ABI PRISM BigDye Terminator version 3.1 (Applied Biosystems). The nucleotide sequences of the primers were listed in Supplementary Table S1. Five genes were used for phylogenetic analyses, as in a previous study of Claviceps and Aciculosporium (Píchová et al., 2018). The internal transcribed spacer (ITS) and the part of the 28S rDNA regions were amplified using the primers ITS1F/ITS5 and LR5+ and sequenced using primers ITS1F, ITS3, ITS4, NL1, NL4, LR2SM, LR2RSM, and LR5+. A fragment of the Mcm7 gene (Mcm7) was amplified and sequenced with Aci-Mcm7F and Aci-Mcm7R primers. A region of the elongation factor 1α gene (TEF) was amplified using primers TEF-983F and TEF-2218R and sequenced using primers TEF-983F, TEF2218R, TEF-1577F-Cla, and TEF-1567R-Cla. Part of β-tubulin (Tub2) was amplified with Aci-T1 and Aci-T22 primers and sequenced with Aci-T1, T2, T12, and Aci-T22 primers. Part of the second largest subunit of the RNA polymerase II (RPB2) region was amplified using primers Aci-fRPB2-5F and fRPB2-7cR-Cla and sequenced using primers Aci-fRPB2-5F, RPB2-6F2, RPB2-6R2, and fRPB2-7cR-Cla. The sequences were deposited in the DDBJ database (http://www.ddbj.nig.ac.jp/; accession numbers) (Table 1).

2-3. Phylogenetic analyses

The nucleotide sequences of Aciculosporium species were compared with the known sequences of species from this genus and related Corallocytostroma and Claviceps fungi obtained from GenBank (https://www.ncbi.nlm.nih.gov/genbank/; Table 1). The nucleotide sequences of the five genes were individually aligned using CLC Main Workbench 8 (CLC bio, Aarhus, Denmark), and sites that did not align and introns were excluded manually. The alignments were concatenated for phylogenetic analyses. The alignments were deposited in the TreeBASE repository (https://treebase.org/; Submission ID 26685). Maximum likelihood (ML) analysis was conducted in CLC Main Workbench 8 (CLC bio), using the GTR+G+T model that was recommended by the Model Test (AIC model). Support for the tree nodes was determined by analyzing 1000 bootstrap replicates. Maximum parsimony (MP) analyses were conducted by PAUP 4.0a164 (Swofford, 2003) using 1000 heuristic search replicates. The resultant phylogenetic tree (Fig. 3) was rooted with Claviceps species.

2-4. Culturing of A. take for chemical analysis

Aciculosporium take (MAFF 241224) was precultured on peptone–yeast–glucose (PYG) agar, which was formulated using (per L of water) 1.0 g polypeptone (Nihon Pharmaceutical Co., Ltd, Tokyo, Japan), 1.0 g yeast extract (Oriental Yeast Co., Ltd., Tokyo, Japan), 10.0 g d-glucose (Wako Pure Chemical Industries, Ltd.), and 15.0 g agar. The precultured fungus was placed in a culture flask containing 500 mL of PYG broth together with the agar medium and cultured at 25 °C for 7 d while shaking at 150 rpm. The culture broth was dispensed into 156 culture flasks containing 500 mL of PYG medium, and cultured at 25°C for 7 d while shaking at 150 rpm.

2-5. Chemical analysis

Optical rotations were measured on a JASCO DIP-1000 (JASCO Co., Ltd., Tokyo, Japan). Electrospray ionization-mass spectrometry spectra were obtained using a JEOL JMS-600W spectrometer (JEOL Ltd., Tokyo, Japan) (Supplementary Fig S3). NMR spectra were obtained using a Bruker AVANCE-400 (400.13 MHz for ¹H) spectrometer (Bruker Bio Spin K. K., Kanagawa, Japan). The multiplicity of the signals was abbreviated as follows: s = singlet, d = doublet, dd = doublet of doublets, dt = doublet of triplets, t = triplet, q = quartet, m = multiplet, and br = broad. Column chromatography was conducted using sephadex LH-20 resin (GE healthcare Japan Co., Ltd., Tokyo, Japan). Thin-layer chromatography (silica gel, Merck Ltd., Japan) was conducted using van Urk reagent (Ehmann, 1977). Preparative high-performance liquid chromatography (HPLC) was conducted using an LC-20AT prominence pump (flow rate 2 mL min^-1, Shimadzu Corp., Kyoto, Japan) and an Inertsil ODS-P column (10 × 250 mm, GL science Inc., Tokyo, Japan) in a CO-965 column oven (30 °C, JASCO Co., Ltd., Tokyo, Japan,) and monitored with an SPD-20A UV detector (Shimadzu Corp., Kyoto, Japan). A BioShaker PERSONAL-11 (TAITEC Co., Ltd., Saitama, Japan) was used for the incubation of A. take.

2-6. Isolation of chemical compounds

After cultivation, the culture broth was separated into mycelium and filtrate by paper filtration. The filtrate was extracted twice with EtOAc, and then, the EtOAc solution was concentrated in vacuo. The EtOAc extract (2.1 g) was suspended in MeCN, and the filtered MeCN solution was defatted with hexane and then concentrated in vacuo. The MeCN extract (1.2 g) was chromatographed on 100 g of sephadex LH-20 resin using a stepwise method with mobile phases of hexane–CHCl₃ (1:4), CHCl₃–acetone (3:2), CHCl₃–acetone (1:4), acetone, and MeOH. The hexane–CHCl₃(1:4) fraction (0.2 g) was suspended in a hexane–acetone (1:1) mixture to give 1 (26.1 mg) by filtration. The CHCl₃–acetone (3:2) fraction (50 mg) was purified by HPLC on a silica gel column using CH₃Cl–MeOH (15:1) and (25:1), followed by hexane–acetone (1:1); 2 (1.1 mg), 3 (0.5 mg), and 4 (1.4 mg) were obtained from this purification.

3. Results

3-1. Phylogenetic analysis

The multilocus phylogenetic tree (28S rDNA, Mcm7, TEF, Tub2, and RPB2) demonstrated that Aciculosporium species form a monophyletic group, and showed that Aciculosporium sp. ex O. undulatifolius has a distinct lineage, described below as A. oplismeni (Fig. 4). Phylogenetic trees based on individual loci showed a similar topology (Supplementary Fig S1). The overall ITS1-5.8S rDNA-ITS2 sequence similarity between A. take (LC571753) and A. oplismeni (LC571760) was observed to be 97.1% (442/455), and two inclusion or deletion were found in the ITS1 region. This phylogenetic analysis also supported a close, but distinct, relationship between A. take and A. sasicola. In comparison, the sequence of A. take had two transitions in the ITS1 region (2/140) and one transition in the ITS2 region (1/159). The overall ITS1-5.8S rDNA-ITS2 sequence similarity between A. take and A. sasicola (LC571758) was observed to be 99.3% (452/455). Additionally, DNA analysis showed that there were few intraspecific variations in ITS region among samples of A. take (Supplementary Fig S2).

Fig. 4 - A maximum-likelihood phylogenetic tree of a concatenated data set of five loci (28S rDNA, Mcm7, TEF, Tub2, and RPB2). Sequence data obtained in this study are shown in bold face. Bootstrap support values (ML/MP) are given at nodes and based on 1000 replicates. The scale represents the number of nucleotide substitutions per site. ^T ex-type isolates.

3-2. Chemical analysis

We did not detect any indole alkaloids, lolines, peramine, indole-diterpene and lolitrem in the solvent of the A. take culture. We did, however, identify 4 proline-containing cyclic dipeptides in the culture of A. take, as follows. Compounds 1-4 were identified as [1:cyclo-(l-Pro-l-Val)], [2:cyclo-(l-Pro-l-Leu)], [3:cyclo-(l-Pro-l-Phe)], and [4:cyclo-(l-Pro-l-Ile)] by detailed comparison of the analytical data (NMR, MS, optical rotation, etc.) and literature values (Kwon et al., 2001; Fdhila et al., 2003; Yan et al., 2004; Takaya et al., 2007).

1:cyclo-(l-Pro-l-Val): Colorless amorphous: [α]²⁰_D -134.3°(c 1.0, MeOH); EIMS m/z: 196 (M⁺). ¹H-NMR (CDCl₃) d3.61(1H, m, H-3), 3.51(1H, m, H-3), 2.01(1H, m, H-4), 1.88 (1H, m, H-4), 2.34 (1H, m, H-5), 2.01(1H, m, H-5), 4.05(1H, brt, J=8.0 Hz, H-6), 6.61 (1H, s, NH), 3.91 (1H, brs, H-9), 2.60 (1H, m, H-10), 0.88 (3H, d, J=7.0 Hz, CH₃), 1.06 (3H, d, J=7.0 Hz, CH₃) (Kwon et al., 2001).

2:cyclo-(l-Pro-l-Leu): Colorless amorphous: [α]²⁰_D -127.9°(c 1.0, MeOH); EIMS m/z: 210 (M⁺). ¹H-NMR (CDCl₃) d3.57(2H, m, H-3), 1.91(1H, m, H-4), 2.01 (1H, m, H-4), 2.12 (1H, m, H-5), 2.34(1H, m, H-5), 4.12(1H, t, J=8.0 Hz, H-6), 6.92 (1H, brs, NH), 4.01 (1H, brd, J=9.5 Hz, H-9), 1.54 (1H, m, H-10), 2.01 (1H, m, H-10), 1.74 (1H, m, H-11), 1.00 (3H, d, J=6.5 Hz, CH₃), 0.96 (3H, d, J=6.5 Hz, CH₃) (Yan et al., 2004).

3:cyclo-(l-Pro-l-Phe): Colorless amorphous: [α]²⁰_D -89.5°(c 1.0, MeOH); EIMS m/z: 244 (M⁺). ¹H-NMR (CDCl₃) d3.64(1H, m, H-3), 3.57(1H, m, H-3), 1.88 (2H, m, H-4), 2.30 (1H, m, H-5), 1.99 (1H, m, H-5), 4.06 (1H, m, H-6), 5.82 (1H, brs, NH), 4.27 (1H, brd, J=8.0 Hz, H-9), 3.54 (1H, m, H-10), 2.87 (1H, m, H-10), 7.26 (5H, m, Ar) (Fdhila et al., 2003).

4:cyclo-(l-Pro-l-Ile): Colorless amorphous: [α]²⁰_D -116.9°(c 1.0, MeOH); EIMS m/z: 210 (M⁺). ¹H-NMR (CDCl₃) d3.62(1H, m, H-3), 3.53(1H, ddd, m, H-3), 2.01(1H, m, H-4), 1.90 (1H, m, H-4), 2.35 (1H, m, H-5), 2.01(1H, m, H-5), 4.05(1H, m, H-6), 6.50 (1H, brs, NH), 3.95 (1H, brs, H-9), 2.29 (1H, m, H-10), 1.40 (1H, m, H-11), 1.19 (1H, m, H-11), 0.90 (3H, t, J=6.5 Hz, CH₃), 1.05 (3H, d, J=6. 5 Hz, CH₃) (Takaya et al., 2007).

3-3. Taxonomy

AciculosporiumI. Miyake, Botanical Magazine Tokyo 22: 307 (1908). [MB 41].

= Albomyces I. Miyake, Gifuken Nokai-Zasshi 20: 14 (1908) (nom. inval., Art. 36.1).

≡ Mitosporium (I. Miyake) Clem. & Shear, The genera of fungi, Edn 2 (Minneapolis): 82 (1931). [MB 3219].

Note: See notes below for explanation of Albomyces nomenclature.

Aciculosporium take I. Miyake, Botanical Magazine Tokyo 22: 307 (1908). [MB 191932]. Fig. 1.

≡ Balansia take (I. Miyake) K. Hara, Shizuokaken Nogyo-Zassan, p. 220 (1919).

≡ Mitosporium take (I. Miyake) Clem. & Shear, The genera of fungi, p. 285 (1931). [MB 431842]

=Albomyces take I. Miyake, Gifuken Nokai-Zasshi 20: 14 (1908) (nom. inval., Art.36.1). [MB 548207].

≡ Albomyces take (I. Miyake) I. Hino, Transactions of the Mycological Society of Japan 3: 113 (1962). (nom. inval., Art. 38.5).

Neotype [MBT 394743]: On Phyllostachys bambusicola, JAPAN. Tokyo, Hachioji, Takao (present address), 4 Jul 1936, T. Ogawa (TNS-F-195715).

Host: various taxa of Arundinarieae in Bambusoideae (Hibanobambusa, Neosasamorpha Phyllostachys, Pleioblastus, Pseudosasa, Semiarundinaria, Shibataea, ).

Distribution: China, Japan, Korea, Taiwan

Specimen examined: On Phyllostachys bambusoides, JAPAN. Tokyo, Hachioji, Takao (present address, 35°38'N 139°16'E), 4 Jul 1936, T. Ogawa, (neotypedesignated here [MBT 394743], TNS-F-195715). On P. pubescens var. pubescens, JAPAN. Ishikawa, Kanazawa, Nukadani (36°30'24.75"N, 136°37'58"E), 16 Jul 2007, E. Tanaka (reference specimen designated here [MBT 393135], TNS-F-60468, culture MAFF 241224); On Phyllostachys bambusoides, JAPAN. Ishikawa, Kanazawa, Nukadani (36°30'30"N 136°37'58"E), 16 Jul 2007, E. Tanaka (TNS-F-60467); On P. bambusoides, JAPAN. Kyoto, Kyoto, Sakyo, Ginkakuji (35°01'34"N 135°47'51"E), Oct 2004, E. Tanaka (culture MAFF 241223); On P. pubescens var. pubescens, JAPAN. Kagoshima, Aira, Kajiki (31°44'59.6"N, 130°40'10.5"E), 1 Jun 2004, E. Tanaka (TNS-F-60465); On P. bambusoides, JAPAN. Kagoshima, Amami, Naze, Okuma (28°24'35.2"N, 129°31'21.2"E), 27 Nov 2019, E. Tanaka (TNS-F-60469); On Pseudosasa japonica (new host), JAPAN, Ishikawa, Kanazawa, UchiyamaToge (36°37'04"N, 136°46'04"E), 13, Dec 2008, E. Tanaka (TNS-F-91308) ; On P. pubescens var. pubescens, JAPAN, Tokyo, Oi-machi, 11, Aug 1898, (TNS-F-25693).

Description: Ascostromata reddish-brown, sessile, pulvinate, fleshy pseudoparenchymatous, 2–3 mm diam and 3–6 mm long, formed on conidiostromata. Perithecia immersed at marginal part of stroma, oval or pyriform, 375–525 × 100–125 µm, tips opened toward outside. Asci narrowly cylindrical, 270–330 × 5.0–6.0 µm, with a hemispherical tip, containing eight ascospores. Ascospores filiform, 230–300 × 1.5–2.0 µm, hyaline, multiseptated. Conidiostromata whitish, pseudoparenchymatous, fusiform, forming irregular cavities, enclosed by leaf sheath, formed on shoot apices of the host. Conidiophores arising from the innermost cells of stomatal wall, hyaline, simple or branched, one-celled, 11.5–24 × 1–1.8 µm. Conidia on conidiostromata filiform, hyaline, 2–3 septate, slightly constricted at the septa, slightly obtuse at both ends, 35–62 × 1.0–1.8 µm. Culture on PDA slow growing, yeast-like. Mycelial growth occurred. Conidia in culture holoblastic, hyaline, narrowly cylindrical to filiform, 35–61 × 1.6–2 µm, with apical branching appendages.

Notes: The anamorph of Aciculosporium take has been called as Albomyces take I. Miyake. This anamorphic name was originally given as an assumed name of the causal fungi in-perfecti of witches` broom of bamboo (Hara, 1908), and was not accepted by Prof. I. Miyake in the original publication. Consequently, the name Albomyces take I. Miyake is invalid (Art. 36.1). Miyake (1908) determined the teleomorph and validly described the holomorphic name, but did not mention Albomyces. Hino (1962)tried to re-describe Albomyces take as the anamorph name of A. take; however, he did not designate the type and the species name Albomyces take (Art. 40.1), and subsequently, the name of the genus Albomyces has been invalid (Art. 38.5). We were unable to successfully extract and sequence DNA from the neotype specimen.

Aciculosporium sasicola Oguchi, Mycoscience 42: 219 (2001). [MB 476668]. Fig. 2.

   = Albomyces sasicola Oguchi, Mycoscience 42: 220 (2001). [MB 474669]. (nom. inval., Art. 35.1).

Holotype: On Sasa senanensis, JAPAN, Hokkaido, Sapporo, Mt. Moiwa, 3 Aug 1999, T. Oguchi (NTS-392 in SAPA).

Host. Sasa sect. Sasa spp.

Distribution: Japan

Description: Ascostromata pale brown, sessile, pulvinate, fleshy pseudoparenchymatous, 5 mm diam and 8–15 mm long, formed on conidiostromata. Perithecia immersed at marginal part of stroma, oval or pyriform, 350–520 × 90–180 µm, tips opened toward outside. Asci narrowly cylindrical, 112–275 × 5.0–7.5 µm, with a hemispherical tip, containing eight ascospores. Ascospores filiform, 75–175 × 1.0–2.0 µm, hyaline, multiseptated. Conidiostromata whitish or pale brown, pseudoparenchymatous, fusiform, forming irregular cavities, enclosed by leaf sheath, formed on shoot apices of the host. Conidiophores arising from the innermost cells of stomatal wall, hyaline, simple, one-celled, 10–25 × 1–2 µm. Conidia on conidiostromata filiform, hyaline, 2-septate, slightly constricted at the septa, slightly obtuse at both ends, 40–50 × 1–2 µm. Culture on PDA slow growing, yeast-like. Conidia in culture holoblastic, hyaline, narrowly cylindrical to filiform, 17–61 × 1.5–3 µm, with or without apical branching appendages.

Specimen examined: On S. senanensis , JAPAN, Hokkaido, Sapporo, Mt. Moiwa (type locality, 43˚01’33”N 141˚18’57”E, Alt. 350 m), 4 Jun 2006, E. Tanaka (TNS-F-60466); On Sasa palmata, JAPAN, Kyoto, Nantan, Miyama, Ashiu (35˚20’48”N 135˚45’16”E, Alt. 680 m), Jul 1996, E. Tanaka (culture MAFF 246967); On S. palmata, JAPAN, Gifu, Hida, NaraToge (36˚20’51”N 137˚04’01”E, Alt. 1230 m), 22 Jun 2020, E. Tanaka (TNS-F-91306; culture MAFF 247298); On S. palmata, JAPAN, Ishikawa, Hakusan, Shiramine, Mt. Hakusan (36˚08’05”N 136˚44’39”E, Alt. 1809 m), 19 Aug 2020, E. Tanaka (TNS-F-91307; culture MAFF 247297).

Notes: Species of Sasa sect. Sasa are usually distributed in snowy regions of Hokkaido and the Honshu islands in Japan. Aciculosporium sasicola differs morphologically from A. take by having a smaller ascospore size. The name Albomyces sasicola is invalid for the reason mentioned above (Art. 36.1).

Aciculosporium oplismeni E. Tanaka, sp. nov. Fig. 3.

MycoBank no.: MB 836229.

Holotype: On Oplismenus undulatifolius, JAPAN, Tokyo, Minato, Shiroganedai, Institute for Nature Study, 6 Jul 2018, E. Tanaka (TNS-F-87159, culture ex-holotype MAFF 246966).

Host: Oplismenus undulatifolius

Etymology: Named after the host genus Oplismenus.

Description: Ascostoromata pale brown, sessile, pulvinate, fleshy pseudoparenchymatous, 2–3 mm diam and 3–5 mm long, formed on shoot apices of the host. Perithecia immersed at marginal part of stroma, oval or pyriform, 300–380 × 80–180 µm, tips opened toward outside. Asci narrowly cylindrical, 165–305 × 4.5–7.0 µm, with a hemispherical tip, containing eight ascospores. Ascospores filiform, 150–300 × 1.5–2.0 µm, hyaline, filamentous, multiseptated. Conidiostromata have never been seen. Cultures on PDA slow growing, yeast-like. Mycelial growth occurred. Conidia in culture, holoblastic, hyaline, narrowly cylindrical to filiform, 18–94 × 1.9–2.5 µm, with apical branching appendages.

Specimen examined: On O. undulatifolius, JAPAN, Tokyo, Minato, Shiroganedai (35°38'23"N 139°43'11"E), 6 Jul 2018, E. Tanaka (holotype designated here TNS-F-87159, culture ex-holotype MAFF 246966) ; On O.undulatifolius, JAPAN, Tokyo, Minato, Shiroganedai (35°38'23"N 139°43'11"E), 1 Jul 2017, Y. Dagawa and K. Okunishi (TNS-F-91311).

Distribution: Japan

Notes: Aciculosporium oplismeni differs from A. take or A. sasicola in its host plant. The host plant O. undulatifolius is known as wavyleaf basketgrass in English and as Chijimi-zasa in Japanese. This grass is perennial and belongs to the tribe Paniceae (Poaceae). Host plants infected with A. oplismeni did not show any symptoms, although host plants infected with A. take or A. sasicola showed symptoms of witches’ broom, with dwarf leaves and twigs. The conidia with dichotomously branched appendages are formed by microcyclic conidiation in that they germinate from the basal end, in the same manner with A. take (Tsuda et al., 1997) and A. phalaridis (Walker, 2004). This species has only been found in one location so far.

Aciculosporium monostipum (J. White et al.) M. Kolařík & Píchová, Molecular Phylogenetics and Evolution 123: 81 (2018). [MB 823457].

Basionym: Neoclaviceps monostipa J. White et al., Mycologia 93: 92 (2001). [MB 467728]. Detailed description in Sullivan et al. (2001).

Aciculosporium phalaridis (J. Walker) M. Kolařík & Píchová, Molecular Phylogenetics and Evolution 123: 81 (2018). [MB 823479].

Basionym: Claviceps phalaridis J. Walker, Proceedings of the Linnean Society of New South Wales 82: 326 (1957). [MB 294988]. Detailed description in Walker (2004).

≡ Cepsiclava phalaridis (J. Walker) J. Walker, Australasian Plant Pathology 33: 228 (2004). [MB 371342].

4. Discussion

4-1. Neotype and reference specimen forAciculosporium take

The genus Aciculosporium was introduced to accommodate A. take, the type species of the genus (Miyake, 1908The original description of A. take did not reference any collections that keep a type specimen nor showed any illustrations to enable the lectotype for A. take to be designated. According to Hara (1918), its original description was based on a sample on P.bambusoides collected in Tamagawa, Kanagawa, Japan. The specimen that may be designated as a lectotype is also missing. As A. take is the type species of the genus, type material for A. take should be designated to standardize taxonomic studies of the genus. We thus selected a sample on P. bambusoides collected in Takao, Tokyo, Japan as the neotype (TNS-F-195715) of A. take. The selected neotype possesses a sexual morph, and its morphological characteristics are almost consistent with those of the original description. The distance between the locality of the neotype and the locality mentioned by Hara (1918)is close enough, about 23 km. Besides, we selected a sample on P. pubescens collected in Kanazawa, Ishikawa, Japan as a reference specimen (TNS-F-60468). This specimen also possesses a sexual morph, and its morphological characteristics are almost consistent with those of the original description (Miyake, 1908). The characteristics of this sample have been studied in detail (Tanaka, 2009, 2010). The culture isolate (MAFF 241224) from the sample has been whole-genome sequenced and used in some studies (Schardl et al., 2013; Schardl et al., 2014). Additionally, ITS sequence analysis showed that there was almost no intraspecific variation among A. take in Japan (Supplementary Fig S2). Therefore, this specimen is considered to have sufficient qualifications to be a reference specimen for both the species standard and practical use.

4-2. Phylogenetic analysis ofAciculosporium species

Multilocus phylogenetic analysis revealed the interspecific relationships among the Aciculosporium species. Among them, the analysis supported that Aciculosporium sp. on O. undulatifolius is clearly an independent species. Therefore, we regarded it as a new species and made a taxonomic proposal for this species as A. oplismeni. It was also found that A. take and A. sasicola , taxa with low intraspecies differences, formed two well-supported lineages, which is supported by the differences in ascospore size. Therefore, we decided to leave A. sasicola as a distinct species.

4-3. Locality ofAciculosporium oplismeni

The new species A. oplismeni has not been found outside of its type locality, even though its host O. undulatifolius is common in Eurasia. Aciculosporium oplismeni is characterized by its sessile ascostromata bearing on shoot apices of O. undulatifolius.Píchová et al. (2018)stated that the ancestral state of Aciculosporium and Claviceps was endophytism and tended to evolve toward strict ovarian parasitism. In Aciculosporium, this trend can be explained by the evolution toward the balanced symbiosis, where the host sexual reproduction is not dramatically affected. In common grasses, the stromata formed on a shoot apices disable sexual reproduction, whereas the ovarian parasites (i.e. A. monostipum, A, phalaridis) just reduce the seed number. That disadvantage can be compensated by the asexual reproduction as in the case of the bamboo and its relatively common pathogen A. take. On the other side, infection by A. oplismeni inhibits the host sexual reproduction, which can also explain the restricted geographical distribution of A. oplismeni.

4-4. Chemical analysis ofAciculosporium take

The present study is the first to report the detection of chemical compounds other than indole-3-acetic acid and the related compounds in A. take. As suggested by the genome analysis, indole alkaloids, peramine, and lolitrem were not found in the culture solvent of A. take. Instead, unlike other phytoparasitic clavicipitaceous fungi, A. take produces cyclic dipeptides containing L-proline. Interestingly, these compounds are known as moieties of peptide ergot alkaloids and ergotamine, which are often produced by Claviceps species. The production of cyclic dipeptides indicates that A. take shares the ability to produce these compounds with Claviceps spp. The identified cyclic dipeptides belong to the class of Proline–Xaa cyclodipeptides (Xaa denotes unspecified amino acids). These molecules have been reported from various fungal genera and have phytotoxic (compounds 1, 2, and 3) or antifungal (compound 2and 4) activity (see Wang et al., 2017 for review;Carrieri et al., 2020). Additionally, compounds 2 and 3 have been identified from the culture filtrate of endophytic Epichloë typhina (Song et al., 2015; Zhou et al., 2015) and have been reported to show antifungal activity (Wang et al., 1999; Magnusson et al., 2003). Aciculosporium take also produces indole-3-acetic acid (auxin) (Tanaka et al., 2003), a well-known plant growth regulator involved in both aggressive (i.e., pathogenic) as well as non-aggressive (i.e., endophytic) plant/fungal interactions (Chanclud & Morel, 2016). Apart from indole-3-acetic acid, the other newly identified cyclic dipeptides may be responsible for the host growth defects seen in plants infected with A. take. Fungi can also use cyclic dipeptides to compete with other fungi, by altering antifungal resistance of systemically infected plants.

Acknowledgement

This work was supported by JSPS KAKENHI Grant Number 16K07238. This study was also partially supported by "Studies on Fauna and Flora of the Institute for Nature Study, National Museum of Nature and Science, Tokyo".We are grateful to Dr. Tsuyoshi Hosoya (National Museum of Nature and Science, Japan) for helping us find a specimen that should be a neotype.

References

• Carrieri, R. , Borriello, G. , Piccirillo, G. , Lahoz, E. , Sorrentino, R. , Cermola, M. , Censi, S. , Grauso, L. , Mangoni, A. , & Vinale, F. (2020). Antibiotic Activity of a Paraphaeosphaeria sporulosa-Produced Diketopiperazine against Salmonella enterica. Journal of Fungi, 6, 1- 9. https://doi.org/10.3390/jof6020083.
• Chanclud, E. & Morel, J. (2016). Plant hormones: a fungal point of view. Molecular Plant Pathology, 17, 1289-1297. https://doi.org/10.1111/mpp.12393.
• Ehmann, A. (1977). The van URK-Salkowski reagent — a sensitive and specific chromogenic reagent for silica gel thin-layer chromatographic detection and identification of indole derivatives. Journal of Chromatography A, 132, 267-276. https://doi.org/10.1016/S0021-9673(00)89300-0
• Fdhila, F. , Vázquez, V. , Sánchez, J. L. , Riguera, R. , (2003). DD-Diketopiperazines: Antibiotics active against Vibrio anguillarum isolated from marine bacteria associated with cultures of Pecten maximus. Journal of Natural Products, 66, 1299-1301. https://doi.org/10.1021/np030233e.
• Hara, K. , (1908). On witches' broom disease of bamboo(in Japnanese). Gifuken Nokai Zasshi 20, 13-14.
• Hara, K. (1918), Takeno Tengusubyo(in Japnanese). Dainihon Sanrin Kaiho, 422, 40-43.
• Hino, I. (1962) Pathogenic fungus of the witches' broom disease of bamboos (In Japanese). Transactions of the Mycological Society of Japan, 3, 111-113.
• Izumitsu, K. , Hatoh, K. , Sumita, T. , Kitade, Y. , Morita, A. , Gafur, A. , Ohta, A. , Kawai, M. , Yamanaka, T. , Neda, H. , Ota, Y. ,& Tanaka, C. (2012). Rapid and simple preparation of mushroom DNA directly from colonies and fruiting bodies for PCR. Mycoscience, 53, 396- 401. https://doi.org/10.1007/ S10267-012-0182-3.
• Kobayashi, M. (2017). The illustrated book of plant systematics in color. Bambusoideae in Japan. Hokuryukan, Tokyo, pp 435.
• Kwon, O. S. , Park, S. H. , Yun, B. S. , Pyun, Y. R. , & Kim, C. J. (2001). Cyclo(D-Pro-L-Val ), a specific β-glucosidase inhibitor produced by Aspergillus sp. F70609. The Journal of Antibiotics, 54, 179-181 https://doi.org/10.7164/antibiotics.54.179
• Magnusson, J. , Ström, K. , Roos, S. , Sjögren, J. , Schnürer, J. (2003). Broad and complex antifungal activity among environmental isolates of lactic acid bacteria. FEMS Microbiology Letters, 219, 129-135 https://doi.org/10.1016/S0378-1097(02)01207-7.
• Miyake, I. , (1908). On witches' broom disease of bamboo (In Japanese). Botanical Magazine Tokyo, 22, 305-307. https://doi.org/10.15281/jplantres1887.22.259_305
• Oguchi, T. , (2001). Aciculosporium sasicola sp. nov. on witches' broom of Sasa senanensis. Mycoscience, 42, 217-221. https://doi.org/10.1007/BF02464140
• Pažoutová, S. , Kolařík, M. & Kolínská, R. , (2004). Pleomorphic conidiation in Claviceps. Mycological Research 108, 126-135. https://doi.org/10.1017/S0953756203009067
• Píchová, K. , Pažoutová, S. , Kostovčík, M. , Chudíčková, M. , Stodůlková, E. , Novák, P. , Flieger, M. , van der Linde, E. , & Kolařík, M. , (2018). Evolutionary history of ergot with a new infrageneric classification (Hypocreales: Clavicipitaceae: Claviceps). Molecular Phylogenetics and Evolution, 123, 73-87. https://doi.org/10.1016/j.ympev.2018.02.013
• Schardl, C. L. , Young, C. A. , Hesse, U. , Amyotte, S. G. , Andreeva, K. , Calie, P. J. , Zeng, Z. , (2013). Plant-symbiotic fungi as chemical engineers: multi-genome analysis of the Clavicipitaceae reveals dynamics of alkaloid loci. PLoS Genetics 9, e1003323. https://doi.org/10.1371/journal.pgen.1003323
• Schardl, C. L. , Young, C. A. , Moore, N. , Krom, N. , Dupont, P. Y. , Pan, J. , Florea, S. , Webb, J. S. , Jaromczyk, J. , Jaromczyk, J. W. , Cox, M. P. , & Farman, M. L. , (2014). Genomes of plant-associated Clavicipitaceae. Advances in Botanical Research, 70, 291-327.https://doi.org/10.1016/B978-0-12-397940-7.00010-0 .
• Song, Q.-Y. , Nan, Z.-B. , Gao, K. , Song, H. , Tian, P. , Zhang, X.-X. , Li, C.-J. , Xu, W.-B. , & Li, X.-Z. (2015). Antifungal, phytotoxic, and cytotoxic activities of metabolites from Epichloë bromicola, a fungus obtained from Elymus tangutorum grass. Journal of Agricultural and Food Chemistry, 63, 8787-8792.https://doi.org/10.1021/acs.jafc.5b04260.
• Sullivan, R. , Bergen, M. S. , Patel, R. , Bills, G. F. , Alderman, S. C. , Spatafora, J. W. , & White, J. F. (2001).Features and phylogenetic status of an enigmatic clavicipitalean fungus Neoclaviceps monostipa gen. et sp. nov. Mycologia, 93, 90-99. https://doi.org/10.2307/3761608.
• Swofford, D L. (2003). PAUP 4.0b10: Phylogenetic analysis using parsimony. Sunderland , MA: Sinauer Associates.
• Takaya, Y. , Furukawa, T. , Miura, S. , Akutagawa, T. , Hotta, Y. , Ishikawa, N. , & Niwa, M. (2007).Antioxidant Constituents in Distillation Residue of Awamori Spirits. Agricultural and Food Chemistry, 55, 75-79. https://doi.org/10.1021/jf062029d.
• Tanaka, E. (2009). Specific in situ visualization of the pathogenic endophytic fungus Aciculosporium take, the cause of witches’ broom in bamboo. . Applied and Environmental Microbiology, 75. 4829-34https://doi.org/10.1128/AEM.00635-09.
• Tanaka, E. (2010). Mechanisms of bamboo witches’ broom symptom development caused by endophytic/epiphytic fungi. Plant Signaling & Behavior 5, 415-418.
• Tanaka, E. , Shrestha, B. , Shivas, R. G. (2020). Commelinaceomyces gen. nov. for four clavicipitaceous species misplaced in Ustilago that infect Commelinaceae. Mycologia 112, 3,649-660. https://doi.org/10.1080/00275514.2020.1745524.
• Tanaka, E. , Tanaka, C. , Ishihara, A. , Kuwahara, Y. , & Tsuda, M. , (2003). Indole-3-acetic acid biosynthesis in Aciculosporium take, a causal agent of witches' broom of bamboo. Journal of General Plant Pathology, 69, 1-6. https://doi.org/10.1007/s10327-002-0002-2.
• Tanaka, E. , Tanaka, C. , & Tsuda, M. (2002). Transmission and intraspecific variation of Aciculosporium take, the causal agent of witches' broom of bamboo (In Japanese). Forest Research, Kyoto, 74, 13-20.
• Tsuda, M. , Shimizu, K. , Matsumura, K. , Tanaka, E. , Tanaka, C. , & Doi, Y. (1997). Host range of Aciculosporium take, the causal agent of witches' broom of bamboo plants. Bulletin of the National Science Museum. Series B, Botany, 23, 25-34.
• Walker, J. (2004). Claviceps phalaridis in Australia: Biology, pathology and taxonomy with a description of the new genus Cepsiclava (Hypocreales, Clavicipitaceae). Australasian Plant Pathology, 33, 211-239. https://doi.org/10.1071/AP04001.
• Wang, X. , Li, Y. , Zhang, X. , Lai, D. , Zhou, L. (2017). Structural diversity and biological activities of the cyclodipeptides from fungi Molecules, 22, 2026,https://doi.org/10.3390/molecules22122026.
• Wang, Y. , Mueller, U. G. , & Clardy, J. O. N. (1999). Antifungal diketopiperazines from symbiotic fungus of fungus-growing ant Cyphomyrmex minutus. Journal of Chemical Ecology, 25. 935-941. https://doi.org/10.1023/A:1020861221126.
• Yan, P. , Song, Y. , Sakuno, E. , Nakajima, H. , Nakagawa, H. , & Yabe, K. (2004). Cyclo ( L-Leucyl-L-Prolyl ) produced by Achromobacter xylosoxidans inhibits aflatoxin production by Aspergillus parasiticus. Applied and Environmental Microbiology, 70, 7466-7473. https://doi.org/10.1128/AEM.70.12.7466-7473.2004
• Zhou, L. Y. , Zhang, X. X. , Li, C. J. , Christensen, M. J. , & Nan, Z. B. , (2015). Antifungal activity and phytochemical investigation of the asexual endophyte of Epichloë sp. from Festuca sinensis. Science China Life Sciences, 58, 821-826. https://doi.org/10.1007/s11427-015-4845-0.