mycological research 113 (2009) 73–81

journal homepage: www.elsevier.com/locate/mycres

Characterisation and phylogenetic relationships of virgultorum and A. anomala in the ()

Heike DE SILVAa, Lisa A. CASTLEBURYb,*, Sarah GREENa, Jeffrey K. STONEc aForest Research, Northern Research Station, Roslin, Midlothian, EH25 9SY, Scotland, United Kingdom bSystematic Mycology and Microbiology Laboratory, USDA ARS, Beltsville, MD 20705, USA cDepartment of Botany and Plant Pathology, Oregon State University, Corvallis OR 97331-2902, USA article info abstract

Article history: The two diaporthalean fungi Anisogramma virgultorum and A. anomala are biotrophic Received 28 November 2007 parasites. A. virgultorum causes stromatal cankers on young shoots of birch whereas Received in revised form A. anomala infects young branches of . Although previous classifications 1 July 2008 based on morphological characteristics placed both species in the , Diapor- Accepted 12 August 2008 thales, their taxonomic position within the order and their relationship to each other re- Corresponding Editor: quired further clarification. We determined the nucleotide sequences of the ITS and Brenda Wingfield partial LSU nu-rDNA regions of both species. A putative second teleomorph form of A. vir- gultorum, described in the literature as the ‘single perithecial form’, was also included in Keywords: the analysis. Based on phylogenetic analyses of LSU sequences, the stromatal forms of Betula A. virgultorum and A. anomala were part of a well-supported monophyletic sister clade to Corylus the Gnomoniaceae. The single perithecial form was placed within a clade containing repre- Eastern filbert blight sentative members of the Gnomoniaceae, separate from species of Anisogramma. These re- Gnomoniaceae sults indicate that the single perithecial form of A. virgultorum actually represents an unrelated and as yet unidentified species of Gnomoniaceae. A morphological description of asci and ascospores of the three species is given. A Wilcoxon two sample test revealed that asci of the stromatal form of A. virgultorum were significantly shorter than those of the single perithecial species. Ascospores of the stromatal form of A. virgultorum were signifi- cantly shorter and wider than those of the single perithecial species. Published by Elsevier Ltd on behalf of The British Mycological Society.

Introduction eastern filbert blight on commercially grown European - nut (Corylus avellana) in the eastern United States and more The genus Anisogramma comprises two well-known species: recently in Oregon and Washington, and British Columbia A. virgultorum and A. anomala. A. virgultorum produces dark (Johnson et al. 1996). A third species, originally described as brown to black erumpent stromatal cankers on young shoots Diaporthe apiospora has been placed in Anisogramma as A. apio- of Betula pendula and B. pubescens and has been observed to spora, but Barr (1978) did not include it in Anisogramma and cause shoot dieback on young trees in Scotland (H.D.S. & later placed it in the genus Apioporthella (Barr 1991). A. apio- S.G. unpubl.). A. anomala causes the devastating disease spora has a valsoid arrangement of perithecia that does not

* Corresponding author. E-mail address: [email protected] 0953-7562/$ – see front matter Published by Elsevier Ltd on behalf of The British Mycological Society. doi:10.1016/j.mycres.2008.08.008 74 H. De Silva et al.

suggest a relationship with Anisogramma, and therefore, is not branches in severely infected trees leading to branch dieback considered in this paper. within a few years (Gottwald & Cameron 1979; Gottwald & A. virgultorum has been reported from birch in Germany, Cameron 1980a; Johnson et al. 1994). Disease incidence may Italy, Finland, Switzerland, UK, Sweden (Massee 1914; Thei- be as high as 100 % at some sites and in some cases entire or- ßen & Sydow 1916; Vleugel 1917; Froidevaux & Mu¨ ller 1972; chards have been killed by the disease (Gottwald & Cameron Witzell & Karlsson 2002) and New Hampshire (Barr 1978). Al- 1980b; Johnson et al. 1996). though widely distributed, it has been considered a minor Historically, morphological characteristics of the stromata, pathogen of birch in the UK because of its rare occurrence perithecia, and ascospores have been used to distinguish gen- (Dennis 1968; Ellis & Ellis 1985). However, in a recent survey, era within the Diaporthales (Barr 1978; Monod 1983). However, this species was found abundantly at nine planted birch emphasis on different morphological characters by different stands across Scotland and was associated with crown die- authors and the great variation in phenotypic features has back of affected trees (De Silva et al. 2008). Ascospores formed lead to much confusion in the determination of families and within flask-shaped perithecia are released in spring from genera. For example, A. virgultorum was listed in the family stromatal cankers on infected birch shoots of the previous Valsaceae by Eriksson (1992) while Dennis (1968) and Froide- season’s growth. Spore release studied over a two-year period vaux & Mu¨ ller (1972) included this species in the Diaporthaceae. coincided with shoot elongation of birch seedlings, a period More recently, A. virgultorum was placed in the Gnomoniaceae when the host appears to be most susceptible to infection (Barr 1978) while in The Dictionary of the Fungi, the genus Ani- (H.D.S. & S.G. unpubl.). Ascospores are released in spring to in- sogramma is listed in the Valsaceae (Hawksworth et al. 1995). fect young expanding shoots, and the first symptoms, such as The aim of this present study is to determine whether the dark brown staining of the phloem and splitting of the epider- two forms of A. virgultorum are actually conspecific, to charac- mis, are observed on infected current season shoots during terise both forms in comparison with A. anomala, and to infer mid to late July, approximately eight to ten weeks after infec- the phylogenetic relationships of these species to other taxa in tion. Ascostromata then develop on infected birch shoots by the Diaporthales using nucleotide sequences of the complete mid-August (H.D.S. & S.G. unpubl.). ITS region and LSU nu-rDNA gene. A species with similar ascospores that has been identified as a second teleomorph form of A. virgultorum was found in Switzerland and described by Froidevaux & Mu¨ ller (1972) as Materials and methods having small individual fruiting bodies, each containing a sin- gle perithecium, that are scattered around the infected birch Fungal material shoot. In the more common stromatal form numerous peri- thecia occur densely packed in rows within each stroma. Eighteen isolates of Anisogramma virgultorum, one isolate of Whereas the stromatal form was commonly found on birch the single perithecial species and one isolate of A. anomala trees infected with A. virgultorum across Scotland (De Silva were included in this study. Isolates of both birch pathogens et al. 2008), the rare single perithecial species has, to date, were collected from planted and site-natural birch stands only been observed on two B. pubescens trees at one site in across Scotland and A. anomala on infected shoots of Corylus avellana were obtained from one location in Oregon (Table 1). the north of Scotland. Froidevaux & Mu¨ ller (1972) do not men- tion how frequently the single perithecial form was encoun- All infected birch shoots were stored at 20 C until use. Cul- tered, but note that it was not as common as the stromatal turing of A. virgultorum and the single perithecial species was form. attempted by streaking ascospores onto malt agar (2 % malt In North America, A. anomala is indigenous on native hazel agar, 2 % sucrose, 2 % bacteriological agar and 0.2 % yeast ex- (C. americana) and causes a disease of the commercial crop tract) amended with 0.5 % activated charcoal following auto- species, C. avellana, known as eastern filbert blight that pre- claving. While ascospores of all isolates of A. virgultorum vents establishment of this crop in the northeastern USA germinated, their germ tubes lysed after a few days on this (Barss 1930). In 1970 this disease was first discovered on com- medium. Monoascospore cultures could only be isolated for mercially grown European hazelnut trees in the state of the single perithecial species. These were maintained in an Washington (Davison & Davidson 1973) and has since spread incubator at 20/15 C day/night temperatures with a 12 h to Oregon. In Oregon and Washington, ascospores mature photoperiod. Light was supplied as cool white fluorescent and are released from cankers during periods of rain from and near-UV light. Colonies were subcultured to obtain actively early winter to late spring. However, new infections by asco- growing mycelium for DNA extraction. As A. virgultorum could spores occur only on young, developing shoots over a period not be cultured, molecular analyses were conducted by DNA of several weeks in spring. Infection occurs on succulent, extraction using perithecial contents taken directly from newly emerged leaves and shoots soon after breaking of the a freshly cut stroma or by direct PCR methods. Similarly, vegetative buds. The disease cycle requires one or more years, DNA of A. anomala was extracted from perithecial contents including a 12–24 month latent period from the time of infec- taken from infected C. avellana shoots. tion to expression of first symptoms on affected branches (Stone et al. 1992; Johnson et al. 1994). The disease spreads by Voucher material new infections from ascospores and by perennial expansion of existing cankers along and around infected branches. Pe- Herbarium specimens of Anisogramma virgultorum and rennial cankers can measure from a few centimetres to over A. anomala are stored at the USDA ARS Systematic Mycology 2 m in length. Larger cankers often girdle the scaffolding Laboratory, in Beltsville MD, USA. Shoot material infected Phylogenetics of Anisogramma 75

Table 1 – Fungal isolates used for ITS and LSU analyses Isolate Fungal species Host species Origin Collection date Source

AV01 Anisogramma virgultorum Betula pubescens Carroch, Scotland 06.09.2006 H. De Silva AV02 A. virgultorum B. pubescens and B. pendula Edderton, Scotland 14.08.2006 H. De Silva AV03 Single perithecial species B. pubescens Edderton, Scotland 16.08.2004 H. De Silva AV04 A. virgultorum B. pubescens Glen Rinnes, Scotland 17.08.2006 H. De Silva AV05 A. virgultorum B. pendula Grantown, Scotland 15.08.2006 H. De Silva AV06 A. virgultorum B. pendula Strahanna, Scotland 05.09.2006 H. De Silva AV07 A. virgultorum B. pubescens Townhead, Scotland 08.09.2006 H. De Silva AV08 A. virgultorum B. pubescens Ulzieside, Scotland 04.09.2006 H. De Silva AV09 A. virgultorum B. pendula Dunkeld, Scotland 12.09.2006 H. De Silva AV10 A. virgultorum B. pubescens and B. pendula Dornoch, Scotland 12.09.2006 H. De Silva AV11 A. virgultorum B. pubescens Glentress, Scotland 23.05.2006 S. Green AV12 A. virgultorum B. pubescens Glen Loy, Scotland 30.08.2006 H. De Silva AV13 A. virgultorum B. pubescens Cowie Water, Scotland 11.07.2006 H. De Silva AV14 A. virgultorum B. pubescens Leanachan, Scotland 30.08.2006 H. De Silva AV15 A. virgultorum B. pubescens Drumsyniebeg, Scotland 27.06.2006 H. De Silva AV16 A. virgultorum B. pubescens Buchromb, Scotland 22.09.2005 H. De Silva AV17 A. virgultorum B. pubescens Dundeugh, Scotland 06.09.2006 H. De Silva AV18 A. virgultorum B. pubescens Townhead Wood, Scotland 06.09.2006 H. De Silva AV19 A. virgultorum B. pubescens and B. pendula Glen Artney, Scotland 21.08.2006 H. De Silva AA01 A. anomala Corylus avellana Oregon, USA Nov. 2006 J. Stone

with the single perithecial species will be collected this August For the stromatal form of Anisogramma virgultorum, geno- in Scotland and send to USDA ARS Systematic Mycology mic DNA was extracted directly from perithecial contents Laboratory, in Beltsville MD, USA. for using the MoBio Ultra Clean Plant DNA Isolation kit (MoBio, Carlsbad, CA). After thawing of the infected birch shoots, Morphological examination of Anisogramma virgultorum a stroma was sliced open with a clean scalpel under a dissect- types and statistical analysis ing microscope. The contents taken from eight perithecia were placed into the lid of a MoBio bead solution tube and Birch shoots that had developed cankers caused by the stro- extracted according to the manufacturer’s instructions. Geno- matal form of Anisogramma virgultorum (Fig 1A) were collected mic DNA of A. anomala was extracted as described above for in May 2007, when ascospore release was occurring naturally. the stromatal form of A. virgultorum, except the extractions Birch shoots infected with the single perithecial species (Fig 1B) were performed under quarantine conditions. were collected in August 2007. For the morphological examina- Genomic DNA and PCR amplicons were visualised by UV tion of asci and ascospores of the two birch pathogens, fluorescence following 1 % TAE (Tris–acetate buffer) agarose a stroma was sliced open with a sterile scalpel under a dissect- gel electrophoresis containing ethidium bromide. Genomic ing microscope and perithecial contents of one perithecium DNA quantified with Lambda DNA (Fermentas, Lithuania) or Ò were transferred to a microscope slide and covered with a Spectrophotometer ND1000 (Nanodrop , Wilmington, DE). a drop of distilled water. The lengths and widths of 50 asci Relative sizes of PCR amplicons were estimated by compari- and 100 ascospores were measured for each of the two birch son with a GeneRuler 100 bp DNA ladder (Fermentas, pathogens using a graticule in the eyepiece of a compound mi- Lithuania). croscope at 400 magnification. Differences between the total length and width of asci and ascospores of the two birch path- PCR amplification and sequencing ogens were determined in a Wilcoxon two sample test using PROC NPAR1WAY in SAS version 9.1 (SAS Institute Inc. 2000). For all fungal isolates the ITS regions 1 and 2 of the nrDNA in- The morphology of A. anomala on Corylus avellana shoots (Fig cluding the 5.8S gene regions were amplified using the primer 1C) has been described in detail by Gottwald & Cameron (1979). pair ITS5 (50-GGAAGTAAAAGTCGTAACAAGG-30) and ITS4 (50-TCCTCCGCTTATTGATATGC-30)(White et al. 1990) and the DNA extraction LSU regions of the nu-rDNA were amplified using the primer pair LR0R (50-ACCCGCTGAACTTAAGC-30) and LR3 Genomic DNA of the single perithecial species was extracted (50-CCGTGTTTCAAGACGGG-30)(Rehner & Samuels 1994). from approximately 30 mg mycelium harvested from colonies For the direct PCR method, frozen birch shoots infected on malt agar. Fungal cells were mechanically ruptured using with the stromatal form of Anisogramma virgultorum were the Fast Prep system (model FP120, BIO 101 Systems, Vista, thawed and a stroma sliced open with a clean scalpel. For CA) in 600 ml PuregeneÒ Cell Lysis Solution (Gentra Systems, each isolate, the contents of one perithecium were placed Minneapolis, MN) according to the manufacturer’s instruc- into the lid of a 0.2 ml Eppendorf tube. The perithecial con- tions with the exception of rehydrating the DNA pellet in tents were covered with approximately 0.2 ml ddH2O and 30 ml ddH2O (double-distilled water) instead of the supplied stored at 20 C overnight. For the PCR reaction, 10 ml Master- buffer. Mix 2.5x (Eppendorf, Hamburg), 1.25 ml (10 mM) each of forward 76 H. De Silva et al.

Amplified fragments were purified for sequencing by adding 2 ml ExoSAP-IT (USB, Cleveland, OH) to 6 mlofeachampliconand incubated according to the manufacturer’s instructions. Se- quencing reactions for the PCR amplicons from A. virgultorum and the single perithecial species were prepared using the Big- Dye Terminator kit version 3.1 (Applied BioSystems, Foster City, CA). Each reaction mix contained 1 ml BigDye, 0.75 mlof 5 sequencing buffer, 0.75 mlof2.5mM magnesium chloride,

2 mlof10mM primer, 1 ml DNA extract, and 4.5 mlddH2O. Se- quencing primers used were the forward and reverse PCR primers (ITS5, ITS4, LR0R, and LR3). Sequencing reactions were performed according to the following protocol: 96 Cfor 1min,40cyclesat96C for 10 s, 50 Cfor5s,60Cfor4min,fol- lowed by cooling at 4 C. Amplicons for A. anomala were sent to the School of Biological Sciences Sequencing Service at the Uni- versity of Edinburgh for purification and sequencing. Manual editing of sequences and verification of all informative sites were performed using the software Sequencher (version 4.2 or 4.7; Gene Codes Corp., Ann Arbor, MI).

Sequence analyses

Pairwise sequence comparisons of LSU sequences were con- ducted in GenBank with BLASTn (Altschul et al. 1990) using de- fault settings and BLAST2 (Tatusova & Madden 1999) with the following settings differing from the defaults to ensure the en- tire sequences were aligned: match ¼ 1, mismatch ¼1, open gap ¼ 3, and extension gap ¼ 2. LSU nu-rDNA gene trees were inferred by NJ using the Kimura two-parameter distance as implemented in MEGA version 3.1 (Kumar et al. 2004) and by MP using Close-Neighbour-Interchange option with 100 ran- dom addition sequences (Nei & Kumar 2000). Relative support for branches was estimated with 1 K BS replications (Felsen- stein 1985) for both NJ and MP BS analysis. Sequences gener- ated in this study were deposited in GenBank as Anisogramma virgultorum LSU ¼ EU683065, ITS ¼ EU683062; A. anomala LSU ¼ EU683066, ITS ¼ EU683064 and single peri- thecial species LSU ¼ EU683067, ITS ¼ EU683063. Sequences obtained from GenBank are listed by taxon name and Gen- Bank accession number in the tree.

Fig 1 – (A) Stromatal canker of Anisogramma virgultorum on Results a young living shoot of Betula pubescens and (B) the single perithecial species with individual perithecia scattered Morphological examination around a young living shoot of B. pubescens. (C) Stromata of A. anomala on living branches of Corylus avellana. Colonies of the single perithecial species grew very slowly, reaching a diameter of only a few millimetres after three months of incubation. The mycelium is medium to dark and reverse primer and 12.3 ml ddH2O were added to each brown in colour and dense with an almost felt-like appear- sample to obtain a final volume of 25 ml. For genomic DNA ex- ance (Fig 2). Sporulation has not been observed in culture. tractions, the ITS and LSU regions were amplified in 50 ml reac- Asci of Anisogramma spp. contain eight ascospores that are hy- tions containing 20 ml MasterMix 2.5x (Eppendorf), 2.5 ml aline and unequally two-celled with the septum near the nar-

(10 mM) each of forward and reverse primer, 24 ml ddH2O and rower end of the spore. Asci of the stromatal form of 1–3 ml DNA extract. The samples were centrifuged for 1 min A. virgultorum are broadly clavate and ranged between 58– at 700 g and placed into a thermal cycler (BIO-RAD, Hercules, 80 mm (mean 67 mm) in length and between 12–20 mm (mean CA), preheated to 94 C for 4 min. The thermal cycler pro- 15 mm) in width (n ¼ 50). In ten of the 50 asci assessed, gramme was set as follows: 94 C for 30 s, 35 cycles of 55 C a thin, petiole-shaped stipe was observed at the ascus base for 30 s, 72 C for 1 min, with a final extension period at that measured between 15–23 mm in length and between 1.5– 72 C for 30 min, followed by cooling at 4 C. 2 mm in breadth. Asci of the single perithecial species are Phylogenetics of Anisogramma 77

with a long, thread-like stipe and measure 45–65 mm long and 10–15 mm broad (Gottwald & Cameron 1979). Mature ascospores of A. virgultorum (Fig 3A) ranged between 12–15 mm in length with a mean value of 13 mm and between 6–9 mm in breadth with a mean of 7.6 mm(n ¼ 100). The larger cell is ovoid, whereas the smaller cell is more rounded. In comparison, ascospores of the single perithecial species (Fig 3B) measured between 15–18 mm in length with a mean value of 17 mm and between 5–8 mm in breadth with a mean value of 6.9 mm(n ¼ 100). Whereas the smaller cell is rounded like the one of the stromatal form of A. virgultorum, the larger cell is obovoid with an acute apex. A Wilcoxon two sample test showed that ascospores of the stromatal form of A. virgultorum were significantly shorter (P < 0.0001) and broader (P < 0.0001) than those of the single perithecial Fig 2 – Mycelium of the single perithecial species after nine species. Mature ascospores of A. anomala (Fig 3C) have an ellip- months on 2 % malt agar amended with 0.5 % activated tic larger cell, measuring between 8–12 mm long and 4–5 mm charcoal. wide, whereas the smaller cell resembles a hemispherical cap measuring only 1–1.5 mm(Gottwald & Cameron 1979). clavate with a thin stipe at the ascus base and ranged between DNA analysis 70–90 mm in length with a mean value of 81 mm and between 12–16 mm in breadth with a mean value of 14 mm(n ¼ 50). A The final edited sequences for the stromatal form of Anisog- Wilcoxon two sample test revealed that asci of the stromatal ramma virgultorum contained 555 bp for the ITS and 588 bp form of A. virgultorum were significantly shorter (P < 0.0001) for the LSU gene region. For the single perithecial form of than those of the single perithecial species, whereas there A. virgultorum, the ITS gene region contained 604 bp and was no difference in ascal breadth (P ¼ 0.1467) between the the LSU region 522 bp. The final edited sequence for A. two species. Mature asci of A. anomala are broadly clavate anomala contained 539 bp for the ITS and 537 bp for the

Fig 3 – Micrographs of hyaline, unequally two-celled ascospores of (A) the stromatal form of Anisogramma virgultorum; (B) the single perithecial species; and (C) A. anomala. 1 10 20 30 40 50 60 | | | | | | | ------GGAT-ATTGCTG Anisogramma virgultorum TTGGAAGTAAAAAGTCGTAACAAGGTCTCCGTTGGTGAACCAGCGGAGGGATCATTGCTG Single perithecial species ------CGTAACAAGGTCTCCGTTGGTGAACCAGCGGAGGGATCATTGTTG

61 70 80 90 100 110 120 | | | | | | | Anisogramma anomala GAGCAAACGCCCCCCCCTTCCCCGGGGGGGGCGCTACCCAGA-AACCCTTTGTGAATCTT Anisogramma virgultorum GAGCAAACGCCCTCGC------GGGCGCTACCCAGAAAACCCTTTGTGAATTTT Single perithecial species GAACAAACGCCCTC------GGGCGCTATCCAGA-AACCCTTTGTGAATTAT

121 130 140 150 160 170 180 | | | | | | | Anisogramma anomala --CT------CCGTTGCCTCGGC-CTGGTTGCCCCCGAACGGGGGCCCCTCCTCCTCCG Anisogramma virgultorum -CCT-----ACCCGTTGCCTCGGTA-TGGTTGCCCCCGAACGGGGGCCCCTCCTGC—-CC Single perithecial species ACCTTAAAAAAACGTTGCCTCGGCACTGGTTGGCTTCCT-CGGAAGCCCC-CCTACTTTG

181 190 200 210 220 230 240 | | | | | | | Anisogramma anomala GGAG-GGAGCTGACCGGCCGGCGGCCCCATAAACACTGCTCCT---GT-TACGA-TATCT Anisogramma virgultorum GGAG-GGAGCCGACCGGCCGGTGGCCCGATAAACTCTTGTTGTTTGGTGTACCACTATCT Single perithecial species GTAGCGGAGCAGGCCTGTCGCTGGCCCTATAAATCCTTGTTTTT---TGTAATATCATCT

241 250 260 270 280 290 300 | | | | | | | Anisogramma anomala GAGCCTTTGACAG---AAATGAATCAAAACTTTCAACAACGGATCTCTTGGTTCTGGCAT Anisogramma virgultorum GAGCC-AAAACAAAATAAATCAATCAAAACTTTCAACAACGGATCTCTTGGTTCTGGCAT Single perithecial species GAG---TAAACAAGCCAAATAAATCAAAACTTTCAACAACGGATCTCTTGGTTCTGGCAT

301 310 320 330 340 350 360 | | | | | | | Anisogramma anomala CGATGAAGAACGCAGCGAAATGCGATAAGTAATGCGAATTGCAGAATTCAGTGAATCATC Anisogramma virgultorum CGATGAAGAACGCAGCGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATC Single perithecial species CGATGAAGAACGCAGCGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATC

361 370 380 390 400 410 420 | | | | | | | Anisogramma anomala GAATCTTTGAACGCACATTGCGCCCGGCGGCATTCCGCCGGGCATGCCTGTTCGAGCGTC Anisogramma virgultorum GAATCTTTGAACGCACATTGCGCCCGGCGGTATTCCGCCGGGCATGCCTGTTCGAGCGTC Single perithecial species GAATCTTTGAACGCACATTGCGCCCGGTGGTATTCCACCGGGCATGCCTGTTCGAGCGTC

421 430 440 450 460 470 480 | | | | | | | Anisogramma anomala ATGTCACCCCTCGGAGCCTCGGCTTTGGTGTTGGAGGAACAGCCT--CC------G Anisogramma virgultorum ATTTCACCCCTCAAAGC-TTTGCTTTGGTGTTGGAGGAGCAGCCTGTCCCCTTGGGGACG Single perithecial species ATTTCAACCCTCAAAGCCTCGGCTTTGGTGTTGAAGGAATACCCTGTCAAACGG-----G

481 490 500 510 520 530 540 | | | | | | | Anisogramma anomala GGCTGCCCTCTGAAATTCAGTGGCGGGCTCGCTGGAATTTTGAGCGTAGTAATCTT--TG Anisogramma virgultorum GGCTGCCCTCTGAAATCCAGTGGCGGGCTCGCTAGAATTTTGAGTGTAGTAATCT---TG Single perithecial species GGGTACCCTTCTAAATTAATCGGCGGGCTCGCTAGCATTTTGAGCGCAGTAATTTACATA

541 550 560 570 580 590 600 | | | | | | | Anisogramma anomala CCTCGCTTTGAAGGAC-TGGCGGGCAC----AGCCGTTTAAAAAAAAAAAAAACACCCCA Anisogramma virgultorum CCTCGCTTCTAAAGAC-TGGTGGG-AC--CTGGCCGT------AAAACACCCCCCCA Single perithecial species CCTCGTTTATAAAGACTTAGCGGG-ACTTCTTGCCGT------AAAACCCCCCTCTA

601 610 620 630 640 650 660 | | | | | | | Anisogramma anomala C-GTCTGACAAGATGACCTC-GATCCGG------Anisogramma virgultorum CTCTCTGAAAA--TGACCTC-GATCAGGTAGGAATACCCGCTGAACTTAAGCATATCAAT Single perithecial species TTTTCTGAAAAT-TGACCTCGGATCAGGTAGGAATACCCGCTGAACTTAAGCATATCA--

Fig 4 – Comparisons of nucleotide sequences of the 5.8S gene region and its flanking ITS1 and ITS2 regions of Anisogramma virgultorum, the single perithecial species and A. anomala. The 5.8S gene region lies between the red arrows. Shading rep- resents nucleotide conspecificity, black being identical for all three species and grey being identical for only two species.

LSU gene region. A GenBank BLASTn search using LSU gene close relationships with the Gnomoniaceae. Three fungal spe- sequences from isolates of A. virgultorum and the single cies were returned as the best-scoring matches for A. anom- perithecial species found Sirococcus clavigignenti-juglandacea- ala: S. clavigignenti-juglandacearum, Gnomonia leptostyla, and rum as the best-scoring match for both taxa, suggesting G. rostellata. Phylogenetics of Anisogramma 79

Sequence alignment and analysis Table 2 – Percent identities between the LSU regions of Anisogramma virgultorum, the single perithecial species, Although DNA of 19 isolates of the stromatal form of Anisog- A. anomala and other representative taxa in the ramma virgultorum was extracted, only six LSU and nine ITS se- Gnomoniaceae quences could be determined. All six LSU and all nine ITS Pairwise comparisons Identities Gapped Percent sequences obtained for A. virgultorum were identical. ITS re- positions identity gions 1 and 2 were sequenced for the stromatal and single Anisogramma virgultorum versus 487/519 0/519 93 perithecial forms of A. virgultorum and A. anomala and aligned single perithecial species (Fig 4). The sequences of the flanking ITS1 and ITS2 regions A. virgultorum versus A. anomala 520/546 0/546 95 surrounding the 5.8S rRNA gene show marked sequence vari- Single perithecial species 475/519 0/519 90 ation among the three species. Sequence identity in this gene versus A. anomala region between A. virgultorum and A. anomala and between A. virgultorum versus 542/568 0/568 95 A. virgultorum and the single perithecial species was 95 %, Gnomonia gnomon A. virgultorum versus 539/568 0/568 94 compared with 90 % between A. anomala and the single peri- Melanconis alni thecial species (Table 2). Single perithecial species 487/519 0/519 93 The dataset for the LSU analysis consisted of 57 taxa, eight versus G. gnomon of which were representative members of the Gnomoniaceae. Single perithecial species 478/519 0/519 92 The outgroup taxa used were Gaeumannomyces graminis versus M. alni (AF362557) and Magnaporthe grisea (AB026819) (Magnaportha- A. anomala versus G. gnomon 740/790 0/790 93 ceae). In the phylogenetic analysis based on the LSU nu- A. anomala versus M. alni 737/790 0/790 93 rDNA, the single perithecial species was placed within the Gnomoniaceae, although it did not form a well-supported group between the two species is still not clear as only the BS value with any of the representative taxa and could not be assigned from the MP analysis was statistically significant. This may, in to a genus based on this analysis. In contrast, A. virgultorum part, be explained by the fact that hosts and host ranges for and A. anomala formed a well-supported (82 %) monophyletic the two species are quite different. Furthermore, both patho- clade outside of the Gnomoniaceae, but within the Diaporthales gens form stromatal cankers on living shoots of their host (Fig 5). trees, but there are some significant differences in the biology of the two fungi. Anisogramma virgultorum has a shorter latent period of only two to three months before symptoms occur Discussion (H.D.S. & S.G. unpubl.) whereas the latent period of A. anomala is between 12 and 24 months long (Stone et al. 1992). Stroma- Analysis of the partial LSU gene revealed that Anisogramma tal cankers of A. virgultorum have not been observed to virgultorum and A. anomala formed a well-supported mono- expand in length, whereas those produced by A. anomala are phyletic clade outside of the Gnomoniaceae representing a pre- perennial and individual cankers can increase in length viously unknown lineage within the Diaporthales. Only the from a few centimetres up to 1 m year1 depending on their single perithecial species was grouped with other taxa in the location within a tree (Gottwald & Cameron 1980b; Johnson Gnomoniaceae, such as Gnomonia gnomon and Apiognomonia et al. 1994). errabunda. Based on the results from this study, Anisogramma The variation in the length of mature asci (58–79 mm) and virgultorum produces well-developed stroma, like A. anomala, ascospores (12–15 mm) of A. virgultorum assessed in this study and is a true representative of the genus Anisogramma. How- was slightly greater than length measurements given in the ever, the single perithecial species does not belong to the ge- literature. Theißen & Sydow (1916) gave measurements of nus Anisogramma, but cannot be placed in a genus with any 45–55 12 mm in size for asci, whereas Vleugel (1917) and Den- confidence at this time. Ascospore morphology is similar to nis (1968) stated measurements of up to 75 12 mm. Only Apiognomonia; however, comparison with ITS sequences of Vleugel (1917) described the petiole-shaped, sterile stipe at known species of Apiognomonia, as well as other gnomonia- the base of the ascus measuring 20–40 mm observed on some ceous genera available in GenBank, does not suggest it is con- asci of A. virgultorum in this study. It appears to be very fragile generic with Apiognomonia or any other genus. and is likely to break easily as it was only observed in ten out Fungi belonging to the Diaporthales have brown to black of 50 asci assessed in this study. Similarly, the variation in perithecial ascomata immersed in the substrate or stroma length of mature ascospores of A. virgultorum was slightly and lack true paraphyses at maturity. The asci are unitunicate greater in this study than length measurements given by Thei- and have a refractive ring at the apex (Barr 1978). Only recently ßen & Sydow (1916) (10–12 mm) and Petrak (1934) (9–13 mm). the following nine families have been outlined in the Diapor- The variation in breadth of the ascospores assessed in this thales based on molecular characteristics: Gnomoniaceae, Mel- study (6–9 mm) was also greater than those given by Theißen anconidaceae, Schizoparmeaceae, Cryphonectriaceae, Valsaceae, & Sydow (1916) and Petrak (1934) of 4–5.5 mm. The slight differ- Diaporthaceae, Pseudovalsaceae, , and Togniniaceae ences in ascospore size may be due to their maturation, as it is (Rossman et al. 2007). The molecular data demonstrated that not known during which time of year infected shoot material the genus Anisogramma is more closely related to members was collected and examined by the authors. During this study of the Gnomoniaceae than any other of the nine families. ascospores were assessed at the end of May, when ascospores Although Anisogramma virgultorum and A. anomala formed are naturally released from the stromatal cankers to ensure a monophyletic clade within the Diaporthales, the relationship the ascospores were mature. Melanconis stilbostoma AF362567 99 Melanconidaceae Although A. virgultorum and the single perithecial species 100 Melanconis alni AF362566 produce morphologically similar ascospores and are biotrophic Melanconis marginalis AR3442 parasites of birch, the symptoms they cause on affected trees 72 Gnomonia gnomon CBS199.53 78 Gnomonia setacea AF362563 vary. For the single perthecial species, the clustering of single Cryptodiaporthe aesculi AR3580 perithecia was observed at the basal ends of young shoots, as Single Perithecial Species EU683067 well as mid-way up shoots with black, slightly sunken lesions Gnomoniaceae Apiognomonia errabunda AR2813 surrounding the location of the perithecia. Dieback occurred Cryptosporella hypodermia AR3552 97 during the growing season of the host as dead leaves and wilt- 98 Plagiostoma conradii AR3488 ing tips were observed on affected current season shoots. In 99 Ditopella ditopa AR3423 A. virgultorum 100 Phragmoporthe conformis AR3632 contrast, infection by is less conspicuous. Stro- matal cankers occur rather linearly along the shoot and can 82 Anisogramma virgultorum EU683065 Anisogramma anomala EU683066 vary in length from a few millimetres up to 11 cm (H.D.S. &

81 Coniella musaiensis AR3534 S.G. unpubl.). The majority of infected young shoots were killed 80 98 Schizoparme straminea AF362569 within a year from initial infection with dieback occurring 99 Pilidiella castaneicola CBS143.97 Schizoparmeaceae 97 mainly during the dormant phase of the host after the leaves 99 Pilidiella granati CBS152.33 have fallen (H.D.S. & S.G. unpubl.). Therefore, the pathogen Coniella fragariae AR3382 may be present at a site for a few years without being noticed. 82 Cryphonectria cubensis CBS101281 88 Cryphonectria havanensis CBS505.63 The stromatal form of A. virgultorum expressed character- istics of an obligate parasite as isolations have consistently 77 73 Endothia eugeniae AF277142 99 73 Endothiella gyrosa AF362555 been unsuccessful. A number of amendments to the culture Cryphonectria- Cryptodiaporthe corni AR2814 ceae medium, such as activated charcoal (at 0.1 % and 0.05 %), Cryphonectria parasitica ATCC38755 bovine serum albumin (at 0.1 % and 0.05 %), cyclodextrin Cryphonectria macrospora AR3444 (0.5 %), and birch extract (obtained from fresh, crushed birch Chromendothia citrina AR3446 shoots soaked in water overnight and filtered), have been Cryphonectria nitschkei AR3433 used in an attempt to induce growth of the stromatal form 73 Cryptodiaporthe vepris AR3559 Leucostoma nivea AR3512 of A. virgultorum in standard malt agar cultures (S.G. 70 Valsa ceratosperma AR3426 unpubl.). However, the germ tubes, which always emerged 81 Valsella salicis AR3514 from the lateral end of the larger cells, lysed after 24 to Valsaceae Leucostoma auerswaldii AR3428 72 h on the growing medium. Culturing of A. anomala was 97 Valsa ambiens AF362564 99 sporadic and chemical adsorbants, such as bovine serum Valsa mali AF362559 albumin and activated charcoal, were necessary to induce Mazzantia napelli AR3498 ascospore germination and stimulate growth of fungal Diaporthe perjuncta AR3461 92 70 hyphae (Stone et al. 1994). Diaporthe pustulata AR3430 Diaporthaceae 77 Diaporthe arctii AF362562 In the literature, the single perithecial species was

Diaporthe decedens AR3459 described as a second teleomorph stage of A. virgultorum Diaporthe eres AF362565 (Froidevaux & Muller 1972). However, results from this study 86 ¨ 97 Diaporthe medusaea AR3422 show that the stromatal form of A. virgultorum and the single Pseudovalsa umbonata AR3897 94 perithecial species are not conspecific or even closely related. 99 Pseudovalsa longipes AR3541 Pseudovalsaceae The single perithecial species represents a distinct species and Pseudovalsa modonia AR3558 should no longer be referred to as A. virgultorum to avoid Chapeckia nigrosa AR3809

74 Hapalocystis berkeleyi AR3851 future confusion. More collections of this are needed 86 Prosthecium innesii AR3639 to determine whether it represents an undescribed species Sydowiellaceae Rossmania ukurunduense AR3484 or genus within the Gnomoniaceae. Multigene phylogenetic Sillia ferruginea AR3440 studies need to be conducted to clarify the relationship of Sydowiella depressula CBS813.79 A. virgultorum and A. anomala to the Gnomoniaceae to determine Sydowiella fenestrans AR3777 whether they truly represent a new family-level lineage Togninia fraxinopennsylvanica AY761083 99 within the Diaporthales. 100 Togninia minima AY761082 Togniniaceae 93 Togninia novae-zealandiae AY761081

99 Gaeumannomyces graminis AF362557 100 Magna- Magnaporthe grisea AB026819 porthaceae Acknowledgements

Fig 5 – A phylogram representing the relationship between This paper is an edited portion of a PhD thesis to be submitted Anisogramma virgultorum, the single perithecial species and A. by H.D.S. at the University of Aberdeen. The PhD is funded by anomala (red font) in relation to other taxa in the Diaporthales. the Scottish Forestry Trust, Edinburgh. We thank Aimee Hyten LSU nu-rDNA gene trees were inferred by NJ using the Kimura and Mikhail Sogonov of USDA ARS for technical assistance two-parameter distance and MP using Close-Neighbour-Inter- during this study and Alexandra Schlenzig of the Scottish change option with 100 random addition trees. BS supports Agricultural Science Agency for arranging the import of 70 % or greater are above (MP) and below (NJ) the branches. The Anisogramma anomala into Scotland and use of quarantine tree was rooted to Gaeumannomyces graminis (AF362557) and facilities. We also thank Joan Cottrell and Amy Rossman for Magnaporthe grisea (AB026819) (Magnaporthaceae). helpful comments on the manuscript. Phylogenetics of Anisogramma 81

references Johnson KB, Mehlenbacher SA, Stone JK, Pscheidt JW, Pinkerton JN, 1996. Eastern filbert blight of European hazelnut: it’s becoming a manageable disease. Plant Disease 80: 1308–1316. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ, 1990. Basic Kumar S, Tamura K, Nei M, 2004. MEGA3: integrated software for local alignment search tool. Journal of Molecular Biology 215: molecular evolutionary genetics analysis and sequence 403–410. alignment. Briefings in Bioinformatics 5: 150–163. Barr ME, 1978. The Diaporthales in North America with emphasis on Massee G, 1914. Black-knot of birch. Kew Bulletin 322–323. Gnomonia and its segregates. Mycologia Memoir No. 7. J. Cramer, Monod M, 1983. Monographie taxonomique des Gnomoniaceae (de Lehre. l’ordre des Diaporthales). Beihefte zur Sydowia 9: 1–315. Barss HP, 1930. Eastern filbert blight. California Agriculture Depart- Nei M, Kumar S, 2000. Evolution and Phylogenetics. Oxford Univer- ment Bulletin 19: 489–490. sity Press, New York. Davison AD, Davidson Jr RM, 1973. Apioporthe and Monochaetia Petrak F, 1934. Mykologische notizen XII. Annales Mycologici 12: cankers reported in Western Washington. Plant Disease Re- 317–447. porter 57: 522–523. Rehner SA, Samuels GJ, 1994. and phylogeny of Glio- Dennis RWG, 1968. British Ascomycetes. Cramer, Lehre. cladium analysed from nuclear subunit ribosomal DNA De Silva H, Green S, Woodward S, 2008. Incidence and severity of sequences. Mycological Research 98: 625–634. dieback in birch plantings associated with Anisogramma vir- Rossman AY, Farr DF, Castlebury LA, 2007. A review of the gultorum and Marssonina betulae in Scotland. Plant Pathology phylogeny and biology of the Diaporthales. Mycoscience 48: 57: 272–279. 135–144. Ellis MB, Ellis JP, 1985. Microfungi on Land Plants. Croom Helm, SAS Institute Inc., 2000. SAS System for Mixed Models, 4th edn. SAS London. Institute, Cary, NC. Eriksson OE, 1992. The Non-lichenized Pyrenomycetes of Sweden. Stone JK, Johnson KB, Pinkerton JN, Pscheidt JW, 1992. Natural Btjtryck, Lund. infection period and susceptibility of vegetative seedlings of Felsenstein J, 1985. Confidence limits on phylogeny: an approach European hazelnut to Anisogramma anomala. Plant Disease 76: using the bootstrap. Evolution 39: 783–791. 348–352. Froidevaux L, Mu¨ ller E, 1972. Anisogramma virgultorum (Fr.) Theiss. Stone JK, Pinkerton JN, Johnson KB, 1994. Axenic culture of Ani- et Syd., un ascomyce`te pathoge`ne du Betula pubescens Ehr- sogramma anomala: evidence for self-inhibition of ascospore hardt. European Journal of Forest Pathology 2: 185–187. germination and colony growth. Mycologia 86: 674–683. Gottwald TR, Cameron HR, 1979. Studies in the morphology and Tatusova TA, Madden TL, 1999. Blast 2 sequences d a new tool for life history of Anisogramma anomala. Mycologia 71: 1107–1126. comparing protein and nucleotide sequences. FEMS Microbiol- Gottwald TR, Cameron HR, 1980a. Infection site, infection period, ogy Letters 174: 247–250. and latent period of canker caused by Anisogramma anomala in Theißen FSJ, Sydow H, 1916. Einige nachtra¨gliche Mitteilungen European filbert. Phytopathology 70: 1083–1087. u¨ ber Dothideen sowie u¨ ber Erikssonia und verwandte Formen. Gottwald TR, Cameron HR, 1980b. Disease increase and the dy- Annales Mycologici 14: 444–453. namics of spread of canker caused by Anisogramma anomala in Vleugel J, 1917. Zur Kenntnis der Pilzflora in der Umgegend von European filbert in the Pacific Northwest. Phytopathology 70: Umea˚ und Lulea˚. Svensk Botanisk Tidskrift 11: 304–324. 1087–1092. White TJ, Bruns T, Lee S, Taylor J, 1990. Amplification and direct Hawksworth DL, Kirk PM, Sutton BC, Pegler DN, 1995. Ainsworth & sequencing of fungal ribosomal RNA sequences for phyloge- Bisby’s The Dictionary of the Fungi, 8th edn. CAB International, netics. In: PCR Protocols: a Guide to Methods and Application. Wallingford, Oxon. Academic Press, Dan Diego, pp. 315–322. Johnson KB, Pinkerton JN, Gaudreault SM, Stone JK, 1994. Infec- Witzell J, Karlsson A, 2002. Anisogramma virgultorum on saplings tion of European hazelnut by Anisogramma anomala: site of of Betula pendula and Betula pubescens in a district of northern infection and effect of host developmental stage. Phytopathol- Sweden. Forest Pathology 32: 207–212. ogy 84: 1465–1470.