Ascobotryozyma cognata sp. nov., a new ascomycetous yeast

associated with nematodes from wood-boring beetle galleries

Julia KERRIGAN1*, Maudy T. SMITH2, Jack D. ROGERS1, Ge A. POOT2 and Greg W. DOUHAN3

1Department of Plant Pathology, Washington State University, Pullman, WA 99164, USA.2Centraalbureau voor Schimmelcultures, Yeast Identification Service, Uppsalalaan 8, 3584 CT Utrecht,The Netherlands.3Department of Plant Pathology, University of California at Davis, Davis, CA 95616, USA.E-mail : Ascobotryozyma@hotmail.com

Received 20 March 2003; accepted 28 June 2003.

A new species of Ascobotryozyma, A. cognata sp. nov. (anamorph Botryozyma cognata), was isolated from beetle galleriesin Idaho, USA. A. cognata was found on the surface of free-living nematodes, Panagrellus dubius, collected from galleries

created by the long-horned beetle Saperda calcarata in Populus (aspen), and the weevil Cryptorhynchus lapathi in Salix(willow). A. cognata isolates were collected from similar habits and in relatively close proximity to those of A. americana,the only species described from North America. The recognition of A. cognata as a distinct species was supported by

morphological and molecular data. Thallus cells of A. cognata were significantly shorter than those of A. americana. LowDNA reassociation values, notably different randomly amplified polymorphic DNA (RAPD), inter-sequence simple repeat(ISSR), and amplified fragment-length polymorphic (AFLP) fingerprints, and sequence divergence in both the D1/D2domain of the nuc-LSU rDNA and an additional unidentified region were all consistent with the recognition of a new

species.

INTRODUCTION

Botryozyma is an anamorphic ascomycetous yeastgenus characterized by its unique thallus morphology,which features determinate growth and a branchedbasal cell that attaches to nematodes (Smith et al.1992). The genus contains two species, B. nematodo-phila and B. americana, both found attached to free-living nematodes of the genus Panagrellus. The typespecies, B. nematodophila, was discovered in Verona,Italy, from the disease complex sour-rot of grapes,vectored by Panagrellus redivivoides nematodes associ-ated with Drosophila spp. (Shann 1987). No teleo-morph has been reported for B. nematodophila.Ascobotryozyma americana, the teleomorph of B. ame-ricana, was discovered on P. dubius nematodes frombeetle galleries in Populus tremuloides Michx. (trem-bling aspen) in eastern Washington (Kerrigan et al.2001). The sexual state is characterized by globose ascibearing four lunate ascospores; asci develop directlyfrom thallus cells after two or more thalli have fused

(Kerrigan et al. 2001). A. americana differs from B. ne-matodophila in its distribution, habit, nematode host,cell sizes, DNA reassociation values, and inability togrow at 37 xC (Kerrigan et al. 2001).

The fungus has a commensal relationship withthe nematodes, apparently taking nutrition from thenematodes’ environment rather than the nematodeitself. Thalli are attached to the nematode cuticle bybranched cells that do not penetrate the cuticle (Kerri-gan et al. 2001). Thallus cells of the anamorph aloneare more commonly encountered and can be grown inculture. Asci, when present, are only produced shortlyafter isolation and cannot be maintained in pure cul-ture; ascospore germination has not been observed(Kerrigan et al. 2001).

Additional Ascobotryozyma isolates were obtainedfrom beetle activity in Populus and Salix species inwestern Idaho, relatively close to where A. americanawas found. The objectives of this research were to in-vestigate the genetic relationships among A. americanaisolates from Washington and Ascobotryozyma isolatesfrom Idaho. The new species Ascobotryozyma cognatais described based on morphological, molecular, andphysiological data.

* Corresponding author. Present address: Department of Botanyand Plant Pathology, Oregon State University, Corvallis, OR 97331,USA.

Mycol. Res. 107 (9): 1110–1120 (September 2003). f The British Mycological Society 1110

DOI: 10.1017/S0953756203008323 Printed in the United Kingdom.

MATERIALS AND METHODS

Fungal isolation and culture conditions

Ascobotryozyma isolates were collected in easternWashington and western Idaho (Fig. 1) from Populusand Salix species showing signs of beetle infestation.Infestations were identified by the presence of ex-pelled frass and/or oozing slime near beetle entrancesites. Wood plugs were extracted from within andadjacent to entrance sites with the aid of an in-crement hammer; associated frass also was collected.To sample from galleries, trees with heavy infes-tations were felled and split, and wood and frass werecollected. Samples were plated onto 2% corn mealagar (CMA; Difco, Sparks, MD) acidified with lacticacid and maintained at ambient room temperature(22–25 x). Plates were examined daily under a com-pound microscope and serial transfers of nematodesbearing Ascobotryozyma thalli were made until cul-tures were free of contaminants except for organismsnecessary for the nematodes’ nutrition, mainly otheryeasts. Pure cultures were obtained by transferringthalli to plates of potato dextrose agar (Difco) with0.05% (w/v) yeast extract (PDYA). Single thallus

isolates were generated by streaking out thalli andtransferring a single thallus to PDYA. Specimennumbers, isolation information, and accession num-bers are listed in Table 1.

Morphological examinations

Thallus dimensions from ten isolates from NorthAmerica, including at least one representative isolatefrom each collection site (CBS 8461, CBS 8560, CBS8706, CBS 8707, CBS 8708, CBS 8709, CBS 8750, CBS8752, CBS 8753, and CBS 8754), and both isolates fromItaly (CBS 7426, and CBS 7442) were measured. Someof the isolates produced the teleomorph after beingplated, but only the anamorph was present when cellswere measured. Isolates were grown on PDYA, andfrom each isolate five individual subcultures were gen-erated by collecting cells from five different locationsin the culture. For examinations, cells were collectedwith a loop from two different locations in each sub-culture, and each mounted separately in a drop ofwater on a glass microscope slide. Five mature thalliwere randomly selected from each drop with yeast ; athallus was arbitrarily determined to be mature if atleast five cells were present. From each thallus thelength of the branched cell and the next two apical cells,called the first and second cells, were measured. Toanalyze statistical differences among isolates, Fisher’sanalysis of variance and a two-sample t-test were per-formed using a 95% confidence level.

For scanning electron microscopy (SEM), nema-todes bearing thalli were placed in a fixative of 2.5%glutaraldehyde and 0.1 M KPO4 buffer (pH 7.2) and,after approximately 30 min, the nematodes were cutwith a razor blade to reduce shrinkage. The samplesand fixative were poured into a Prep-Eze (Pelco,Redding, CA) mesh tissue basket, rinsed three timesin a similar phosphate buffer, post-fixed in 1% osmiumtetroxide and 0.05 M KPO4 buffer (pH 7.2) for 2 h atroom temperature, rinsedwith distilled water, and dehy-drated in a graded ethanol series. Specimens werecritical-point-dried, carefully affixed to mounts withconductive adhesive tabs, coated with 150–200 A ofgold, and observed with a Hitachi S-570 scanningelectron microscope operated at 20 kV acceleratingpotential.

Molecular analyses

DNA extractions

Single thallus cultures were grown for 10–14 d onPDYA at room temperature. Specimen numbers forcultures included in the analyses are listed in Table 1and in the appropriate figures. To determine DNA basecompositions and DNA–DNA reassociation values,DNA was isolated and purified according the protocoldescribed by Smith et al. (1995). For RAPD- andISSR-PCR analysis, DNA was isolated according themethod of Moller et al. (1992) and the concentration

8706

87058750

87548709

8708

8461

8462

8463

8560

8751

8752

9135

= Populus spp. - Saperda calcarata

= Salix spp. - Cryptorhynchus lapathi

WASHINGTON IDAHO

84608753

8707

25 km

Fig. 1. Ascobotryozyma collection sites in North America.

Numbers are CBS accession numbers. The lower image in-dicates the orientation of the enlarged area.

J. Kerrigan and others 1111

was evaluated by gel electrophoresis. For AFLP andsequence analyses, DNA was extracted following theprotocol of Lee & Taylor (1990) with the eliminationof 2-mercaptoethanol in the lysis buffer. DNA wasquantified with a Fluorolite 1000 (Dynatech Labora-tories, Chantilly, VA) fluoremeter using a FluorescentDNA Quantitation Kit (Bio-Rad, Hercules, CA).

DNA base composition and rehybridization

The mol% G+C of isolates was measured followingthe procedure of Smith et al. (1995). The levels ofDNA–DNA homology were determined spectro-photometrically using the method by Seidler & Mandel(1971) as modified by Kurtzman et al. (1980). Theoptimal reassociation temperature (Tm x25 x), 50 x,was determined according to the method of Kurtzmanet al. (1980).

RAPD and ISSR

The core 10-mer primer OPA-13 (Operon Technol-ogies, Alameda, CA) was used for RAPD-PCR andthree microsatellite primers (ATG)5 (Garbelotto et al.1993), (GTG)5 (Welsh & McClelland 1990), and M13(Vassart et al. 1987) were used for ISSR-PCR analysis.Amplification reactions were performed in a final vol-ume of 50 ml containing 10 mM Tris-HCl, 50 mM KCl,1.5 mM MgCl2, 0.001% gelatin, 0.2 mM of each dNTP(Boehringer Mannheim, Mannheim), 10 pM of each

primer, 1 U of Super-therm DNA polymerase (ITKDiagnostics, Withorn) and 20–30 ng of genomic DNA.The thermocycler (BioMed Thermocycler 60; BioMed,Thewes) was programmed for 35 cycles for primersM13 and (GTG)5 : denaturation of 20 s at 94 x, an-nealing of 1 min at 55 x, and extension of 2 min at 72 x.Forty amplification cycles were performed with primers(ATG)5 and OPA-13: 1 min at 94 x, 1 min at 48 x forprimer (ATG)5 or 1 min at 34 x for primer OPA-13,followed by 2 min at 72 x. The PCR products wereanalyzed by electrophoresis in 1.7% agarose gelschilled at 14 x. To avoid ambiguous results, the ampli-fication reactions of all strains were processed simul-taneously from stock solution of premixed reagentsin a single PCR assay, as suggested by Messner et al.(1994).

RAPD- and ISSR-PCR bands were scored using abinary code (1, 0) corresponding to positive and nullalleles, respectively, and patterns of each isolate werecombined in a composite fingerprint with the computerprogram GelCompar 4.0 (Applied Maths). Geneticsimilarities were calculated using the Pearson product-moment correlation and cluster analysis was performedusing the unweighted pair-group method with arith-metic average (UPGMA) clustering algorithm.

AFLP

The AFLP procedure outlined by Vos et al. (1995) wasfollowed, with the exception the restriction enzyme

Table 1. Isolates of Ascobotryozyma and Botryozyma used for morphological, molecular, and/or physiological examinations.

Species CBS no.a

Origin

mol%

G+Ce

GenBank accession no.

Locationb Substratec Vectord nuc-LSU Unknown seq

A. americana 8461 Cheney, WA, USA Pt Sc 29.7

A. americana 8462 Cheney, WA, USA Pt Sc 28.3 AF538878 AF538890

A. americana 8463 Cheney, WA, USA Pt Sc 29.6

A. americana 8560 Cheney, WA, USA Pt Sc 29.8 AF538879 AF538891

A. americana 8751 Cheney, WA, USA Pt Sc

A. americana 8752 Cheney, WA, USA Pt Sc 30.9

A. americana 9135 Cheney, WA, USA Pt Sc AF538880 AF538892

A. cognata 8460 St. Maries, ID, USA Ss Cl 28.7 AF538881 AF538893

A. cognata 8705 Priest Lake, ID, USA Pt Sc 27.6 AF538882

A. cognata 8706 Priest Lake, ID, USA Pb Sc AF538883 AF538894

A. cognata 8707 St. Maries, ID, USA S Cl 28.8 AF538884 AF538895

A. cognata 8708 Grangemont, ID, USA S Cl AF538885 AF538896

A. cognata 8709 Grangemont, ID, USA S Cl AF538886

A. cognata 8750 Priest Lake, ID, USA Pt Sc

A. cognata 8753 St. Maries, ID, USA Ss Cl 27.1 AF538887 AF538897

A. cognata 8754 Grangemont, ID, USA S Cl

B. nematodophila 7426 Pescantina, Verona, Italy Vv D 31.4 AF538876 AF538888

B. nematodophila 7442 Pedemonte, Verona, Italy Vv D 29.4 AF538877 AF538889

Outgroups

Pichia stipitis U45741

Trigonopsis variabilis U45827

a CBS, Centraalbureau voor Schimelcultures, Utrecht.b WA, Washington (state); ID, Idaho.c Pt, Populus tremuloides ; Ss, Salix scouleriana ; Pb, Populus balsamifera ; S, Salix sp. ; Vv, Vitis vinifera.d Sc, Saperda calcarata ; Cl, Cryptorhynchus lapathi ; D, Drosophila.e ¡S.D.<0.5%, calculated from two determinations.

Ascobotryozyma cognata sp. nov. 1112

Tru 91 (Boehringer Mannheim) was used instead ofMse I (Tru 91 is an isoschizomer ofMse I). Two primercombinations were used with the addition of one ortwo selective nucleotides at the 3k end: EcoR 1+CT/Tru 91+C and EcoR 1+AG/Tru 91+C. Each 20 mlreaction contained 1rPCR buffer (Life Technologies,Grand Island, NY), 2.5 mM MgCl2, 2.5 mM each dNTP(Life Technologies), 3.75 mM of each primer, 0.5 U ofTaq polymerase (Life Technologies), and 10 ml of tem-plate DNA. Thermocycling conditions consisted of aninitial hold at 72 x and 94 x for 1 and 4 min, respect-ively, followed by 44 cycles with an annealing tem-perature of 65 x for 30 s, the temperature reduced by 1 x

for the first 9 cycles, and an extension temperatureof 72 x for 1 min. All amplifications were performed ina PE-9700 thermocycler (Perkin Elmer, Norwalk, CT).Following amplification, 5 ml of formamide loading dyewas added to each reaction and denatured for 4 min at94 x and cooled on ice. Three to four ml of each samplewere loaded and separated on 4% acrylamide gelsbuffered in TBE (1.35 M Tris pH 8.0, 0.45 M boric acid,25 mM EDTA) for approximately 2 h at 1800–1900volts using a Bio Rad Sequi-Gen GT SequencingCell (Bio-Rad Laboratories). The gels were silver-stained by the method of Bassam, Caetano-Anolles &Gresshoff (1991) and dried overnight before scoringbands visually.

AFLP bands were scored using a binary code (1, 0)corresponding to positive and null alleles, respectively.The data were analyzed with NTSYS-pc 2.02 (ExeterSoftware, Setavket, NY). Genetic distance was calcu-lated with the similarity for qualitative data function(SIMQUAL) using the Dice (1945) coefficient. Clusteranalyses were performed by means of the UPGMAclustering algorithm.

DNA sequencing

A portion of the divergent D1/D2 domain (Guadetet al. 1989) at the 5k end of the nuc-LSU rDNA genewas amplified and sequenced by means of the nuclearlarge primers NL-1 and NL-4 (O’Donnell 1993).Amplification of the 5k end of the mt-LSU rDNA genewas attempted with the mitochondrial large univer-sal primers ML1 and ML6 (White et al. 1990). Afterpreliminary analysis showed informative sequence di-vergence among isolates, a reverse primer was designedto generate a sequence of approximately 500–600 bpfor analysis in a single run. The reverse primer ML1R(5k-AGCGCAATAAGTCCGATTCC) was designedwith the Primer3 website (Rozen & Skaletsky 1998).PCR amplification was performed in 40 ml reactionscontaining 4 ml of a 25-fold dilution of genomic DNA.PCR and thermocycling conditions are as describedabove with the exception that a constant annealingtemperature of 52 x for 30 cycles was used. PCR pro-ducts were electrophoresed in 1% agarose gel contain-ing ethidium bromide, illuminated using UV light, andclean reactions were purified using a QIAquick PCR

Purification Kit (Quiagen, Valencia, CA). Sequencingreactions with both forward and reverse primersfor both gene regions were performed following themanufacturers instructions using the BigDye Termin-ator Cycle Sequencing Ready Reaction kit (AppliedBiosystems, Foster City, CA). PCR products werepurified with Centriflex Gel Filtration Cartridges (EdgeBioSystems, Gaitherburg, MD). DNA sequencing re-actions were analyzed on an Applied Biosystems model377 DNA Sequencer (PE Biosystems) by the Labora-tory for Biotechnology and Bioanalysis, School ofMolecular Biosciences, Washington State University.

Sequence data were aligned by means of the multiplealignment program ClustalX 1.8 (Thompson et al.1997). Unaligned regions were checked for accuracymanually by comparing the computerized sequenceread data with the chromatographs. Phylogeneticrelationships were calculated with PHYLIP version3.5c (Felsenstein 1993). All data inputs were ran-domized, and bootstrap tests were performed in 1000replicates using SEQBOOT. Analyses were performedwith maximum parsimony (DNAPARS) (1000 datasets analyzed), maximum likelihood (DNAML) (100data sets analyzed), and distance matrix (DNADIST)with neighbor-joining (NEIGHBOR) (1000 data setsanalyzed, each) as the criteria. A majority-rule con-sensus tree was generated for each analysis usingCONSENSE. Pichia stipitis (Kurtzman & Robnett1997) and Trigonopsis variabilis (Kurtzman & Robnett1995) were used as outgroups for the nuc-LSU. Noorganism aligned with the region amplified with theML1 andML1R primer pair with enough similarity forinclusion in the analyses. Phenograms were drawn withTreeView 1.6.5 (Page 1996).

Physiological characterization

The physiological characteristics of isolates weredetermined following the methods of Yarrow (1998).Growth tests were performed at least three times foreach isolate.

RESULTS

Isolates

Ascobotryozyma isolates were collected from beetleentrance sites and galleries in Populus and Salix speciesfrom seven locations in western Idaho (Fig. 1). Isolatesfrom Populus trees were found in two locations innorthern Idaho and were collected from galleries thatappeared characteristic of those created by the poplarborer, Saperda calcarata (Coleoptera : Cerambycidae), along-horned, wood-boring beetle whose larvae carveout galleries for two or more years (Metcalf & Flint1939). Isolates from Salix were found in five locations,spanning a distance of about 125 km. Willows wereinfested by the poplar and willow borer, Crypto-rhynchus lapathi (Coleoptera : Curculionidae), an exotic,

J. Kerrigan and others 1113

ubiquitous beetle commonly found in shrubby willows(Smith & Stott 1964). All isolates were found associatedwith the free-living nematode Panagrellus dubius.

Morphology

The sizes of thallus cells from the Idaho isolates (fromSalix and Populus) (Figs 2–3) were significantly shorter(P<0.001) than those from the Washington isolates(from Populus) (Table 2). The first and second cellsapical to the basal cell measured, respectively, 10.6–13.0 mm and 12.0–14.8 mm for the Idaho isolates and12.0–14.5 mm and 13.9–16.6 mm for the Washingtonisolates. Basal cell sizes were not significantly different.The two isolates from Populus spp. in Idaho wereintermediate in size between those from Populus inWashington and Salix in Idaho. These incremental celllengths were generated through calculation of meansand standard deviations ; sizes could not be measuredat a precision greater than 0.5 mm.

Isolates from Idaho (CBS 8460, CBS 8705, CBS8707, CBS 8709, and CBS 8753) formed mature ascithat appeared similar to those of Ascobotryozymaamericana (from Washington), with globose asci andlunate ascospores (Figs 4–7). In isolates CBS 8460and CBS 8753, however, the teleomorph differedmorphologically, bearing multiple asci from one cell(Figs 8–10) rather than only one ascus per cell as inA. americana (Kerrigan et al. 2001). Ascus-bearingcells were sometimes inflated and irregular in shape(Fig. 10). No regular order of ascus production wasnoted, as young and mature asci occurred in differentorientations. Ascospore germination has not beenobserved, and it is not known whether this yeast isheterothallic or homothallic. After the teleomorphstopped forming, thalli often exhibited irregular

morphologies as previously observed in A. americana(Kerrigan et al. 2001).

Molecular Data

DNA rehybridization

DNA reassociation values of 99–100% were observedamong representative isolates from Idaho, which werecollected from both Cryptorhynchus lapathi damagein Salix and Saperda calcarata damage in Populus.Isolates from Idaho had low reassociations values withthose from Washington (Ascobotryozyma americana)(47–63%) and Italy (Botryozyma nematodophila)(48–62%) (Fig. 11). Values ranging from 39–54% werereported between A. americana and B. nematodophila(Kerrigan et al. 2001).

RAPD, ISSR, and AFLP

Three distinct clades were found based on clusteringanalysis using RAPD and ISSR (Fig. 12) and AFLP(Fig. 13) markers. All isolates from Idaho clusteredtogether, as did the Ascobotryozyma americana iso-lates from Washington and the two Botryozyma nema-todophila isolates from Italy.

DNA sequence data

Sequencing the D1/D2 domain of the nuc-LSU rDNAyielded a 555 bp region for alignment. Sequences ofBotryozyma nematodophila (from Italy) from thisstudy were identical to those deposited in GenBank(accession no. U40105) from the same isolate (Kurtz-man & Robnett 1995). Sequences from this domainwere highly conserved and sequences were identicalamong isolates within each region. B. nematodophiladiffered from Ascobotryozyma americana (from

Figs 2–3. Botryozyma cognata. Fig. 2. A nematode covered with thalli. Bar=25 mm. Fig. 3. One thallus. Bar=5 mm.

Ascobotryozyma cognata sp. nov. 1114

Washington) by 10 nucleotides (1.80%) and from theIdaho isolates by six nucleotides (1.08%). Washingtonisolates had four nucleotide substitutions (0.72%)when compared with the Idaho isolates. Bootstrapvalues from all phylogenetic analyses, including maxi-mum parsimony, maximum likelihood, and neighborjoining, strongly supported the division of isolates intothree clades representing distinct species (Fig. 14).

The region amplified with the ML1 and ML1Rprimer pair yielded a 569 bp region. ABLAST similaritysearch of the sequences did not reveal significant simi-larity to other fungal sequences; therefore, the identityof these sequences is not known. This region was foundto be informative for elucidating phylogenetic iso-late relationships among the Ascobotryzyma isolatesand was congruent with the nuc-LSU rDNA region.

Table 2. Summary of Ascobotryozyma and Botryozyma species.

Species Location Nematode Plant host Vector

Thallus cell sizes (mm)a

Basal First Second

americana Washington, USA P. dubius Populus Saperda calcarata 6.8–9.5 12.0–14.5 13.9–16.6

cognata Idaho, USA P. dubius Populus, Salix Saperda calcarata,

Cryptorhynchus lapathi

6.5–9.4 10.6–13.0 12.0–14.8

nematodophila Verona, Italy P. redivivoides Vitis vinifera Drosophila 9.0–10.3 12.8–15.9 15.4–18.6

a Incremental cell lengths generated through statistical analysis.

Figs 4–10. Ascobotryozyma cognata. Bar=5 mm. Fig. 4. Fused thalli with developing asci (A) on the cuticle of a nematode (N).Fig. 5. An ascus formed from two fused thalli, regions of fusion indicated with arrows. Fig. 6. Two fused (arrows) thallibearing two asci (A) with portions of ascospores visible inside. Fig. 7. Three ascospores. Fig. 8. A thallus with four asci, two

immature and two deliquesced, formed from one cell. Fig. 9. SEM of thalli with numerous asci (A), intact and deliquesced.Fig. 10. An inflated thallus cell with four asci of different ages.

J. Kerrigan and others 1115

Sequences of the isolates from Italy were identical toeach other. Sequence data from theWashington isolateswere identical except for one isolate (CBS 8462) thatdiffered by two nucleotides. This isolate was collectedin 1992 and has been maintained in pure culture ; it nolonger forms branched cells. The Idaho isolates wereidentical except for one isolate (CBS 8706) that differedby three base pairs ; it is the only isolate from a differentPopulus species and the most northern collection. Thisregion was more divergent than the D1/D2 domain,exhibiting a greater number of nucleotide substitutionsbetween groups. Isolates again broke down into threewell-supported clades (Fig. 15) using all phylogeneticanalyses. Washington and Idaho isolates exhibited the

greatest difference, with 13.34% sequence divergence.Isolates from Washington and Italy differed in 11.42%of their basepairs, and Idaho and Italy differed in9.67%.

Physiology

Physiological capacities, listed in Table 3, are limited,as with the Washington (Ascobotryozyma americana)and Italian (Botryozyma nematodophila) isolates(Kerrigan et al. 2001). The production of carbondioxide from glucose was not observed, and only nineout of 46 tested carbon compounds and two of ninetested nitrogen compounds were utilized (Table 3).Utilization profiles of the isolates from Idaho were thesame as those for A. americana and B. nematodophilaexcept none of the Idaho isolates were able to utilizetrehalose and, like A. americana, the ability to grow at37 x was variable.

Taxonomy

Ascobotryozyma cognata J. Kerrigan, M. T. Sm. &J. D. Rogers, sp. nov. (Figs 4–10).

Etym. : ‘cognata ’, referring to the relatedness of thespecies.

Thalli ad paginam nematodorum ab cellulis basalibus affixi.Cellulae basales obovoideae, corporibus 5–12r2–4 mm,basibus furcatis. Cellulae ceterae subcylindricae vel sub-

ovoideae, 6.5–19r2–4.5 mm, ad apicem pullulantes. Asciglobosi vel late ellipsoidei, tetraspori, 7–9r6–8 mm, uni usquead quattuor e lateribus et apicibus cellularum pullulantes.Ascosporae hyalinae, lunatae, leves, 7–8.5r1.5–2 mm. A

Ascobotryozyma americana differt in amplitudinibus minor-ibus cellularum et in structura DNA.Fermentatio et cultura in carbonibus et in nitrogeniis et

additiciis characteribus in Tabula 3 monstrantur.Typus : USA : Idaho : Benewah Co., ca 47x 21k N

116x 48kW, e cuticula nematodi Panagrellus dubius Sanwal e

foramene coleopteri Cryptorhynchus lapathi in Salix scoule-riana lectus, 20 June 1998, J. Kerrigan [Lamina vitrea per

55 60 65 70 75 80 85 90 95 100

Similarity

CBS 8462

CBS 8463

CBS 8560

CBS 8461

CBS 8752

CBS 7426

CBS 7442

CBS 8705

CBS 8460

CBS 8753

CBS 8707

A. cognata

B. nematodophila

A. americana

Idaho, USA

Verona, Italy

Washington, USA

Isolate Species Location

Fig. 11. UPGMA phenogram of genetic distances among Ascobotryozyma and Botryozyma isolates from DNA

reassociation values.

55 60 65 70 75 80 85 90 95 100

Similarity

CBS 8462

CBS 8463

CBS 8560

CBS 8461

CBS 8752

CBS 7442

CBS 8705

CBS 8460

CBS 8753

CBS 8707

Ascobotryozyma cognata

Botryozyma nematodophila

Ascobotryozyma americanaCBS 8751

CBS 8754

CBS 8707.1

CBS 8706

CBS 8705.1CBS 8708

CBS 8709

CBS 8709.1

CBS 8750

CBS 7898

35 40 45 50

Fig. 12. UPGMA phenogram of genetic distances obtainedwith RAPD and ISSR analysis using the primers (ATG)5,

(GTG)5, M13, and OPA-13. Isolates with a .1 suffix indicatea subculture.

Ascobotryozyma cognata sp. nov. 1116

microscopium status sexualis cum ascis] (WSP 70305 – holo-typus; CBS 8753, WSP 70306, 70307 – isotypi).

Thalli attached to the nematode by basal, or anchor,cells. Basal cells obovoid with the body 5–12r2–4 mm,with bases branched. Other cells subcylindrical tosubovoid, 6.5–19r2–4.5 mm, budding at the apex.Asci globose to broadly ellipsoid, four-spored, 7–9r6–8 mm, typically one to four, sometimes more, bud-ding from cell sides and apices. Ascospores hyaline,lunate, smooth, 7–8.5r1.5–2 mm.

Differs from A. americana in the smaller sizes of cellsand in the structure of its DNA.

Physiological characteristics : Fermentation andgrowth on carbon compounds, nitrogen sources, andadditional properties are summarized in Table 3.

DISCUSSION

Our results clearly support the designation of anew species, Ascobotryozyma cognata, and also resolveand support the taxonomic separation of A. americana

from Botryozyma nematodophila (Kerrigan et al.2001). Molecular and morphological data revealed,surprisingly, that isolates separated into three distinctgroups based on location: Italy (B. nematodophila),Washington (A. americana), and Idaho (A. cognata).The separation of isolates from Idaho is noteworthybecause these isolates were collected from different treegenera, Salix and Populus, and were associated withdifferent beetles, Cryptorhynchus lapathi and Saperdacalcarata, respectively. All isolates were collectedfrom two nematode species in the same genus, whichinhabit similar environments. It was hypothesized ifgenetic separations had evolved, they would have beenalong tree-beetle taxonomic lines. This apparently hasnot occurred or at least has not been detected. Isolatesfrom Idaho spanned a relatively long distance, about250 km on a north-south axis, whereas all isolatesfrom Washington were from one site about 50 km westof the nearest Idaho site (Fig. 1). There are no geo-graphical barriers between the two states, and someIdaho sites were more distant from each other thanfrom the Washington site. It is likely the distributionof the two species overlaps, but further sampling isneeded to determine the ranges of the two species.

A. americana differs from B. nematodophila in itsdistribution (different continent), habit (different planthost and disease/damage), nematode host (differentspecies), cells sizes (shorter), DNA reassociation values

A. cognata

B. nem

atodophila

A. am

ericana

B. nem

atodophila

A. am

ericana

A. cognata

ladder

Fig. 13.AFLP banding patterns of representative Botryozyma

and Ascobotryozyma isolates (B. nematodophila CBS 7426,7442; A. americana CBS 8461, 8462, 8463, 8560, 9135; A.cognata CBS 8460, 8705, 8706, 8708, 8753). The DNA finger-

prints were generated using the primer pairs EcoR 1+AG/Tru 91+C and EcoR 1+CT/Tru 91+C. Molecular weightswere determined by comparison to a 30–300 bp ladder.

64

98

93

100

CBS 8462

CBS 8560

CBS 7426

CBS 7442

CBS 8709

CBS 8460

CBS 8753

CBS 8707

A. cognata

B. nematodophila

A. americana

58

10

CBS 8708

CBS 9135

Trigonopsis variabilis

CBS 8706

CBS 8705

Pichia stipitis

Fig. 14. Phylogenetic tree calculated from neighbor-joininganalysis of the nuc-LSU D1/D2 domain. Branch lengths are

proportional to nucleotide differences, as indicated in thescale bar (%). Bootstrap values are indicated at each node.

J. Kerrigan and others 1117

(39–54%), and inability to grow at 37 x (Kerrigan et al.2001). A. cognata differs from A. americana and B.nematodophila in two of these criteria, cell sizes andreassociation values. Although cell sizes are not a stan-dard character used to differentiate yeast species, thesize differences are visually notable and these dataserve as one criterion in support of the designationof a new species. Reassociation values were 47–63%between Idaho and Washington isolates and 48–63%between Idaho and Italian isolates. Values of similarmagnitude have been used to support the differen-tiation of species. However, in a study on nuclear DNAreassociation studies of various heterothallic asco-mycetous yeasts, Kurtzman (1987) reported isolateswith reassociation values of 40–70% are often con-sidered varieties of the same species. Since the NorthAmerican isolates were collected in relatively closeproximity to each other and all were from wood-boringbeetle galleries and the same nematode species, ad-ditional criteria were examined.

Molecular investigations show strong support forthe division of isolates into three taxonomic groups.The D1/D2 domain of the nuc-LSU rDNA has beensequenced for over 500 species and has been shownto be reliable for discerning ascomycetous yeastspecies within genera (Kurtzman & Robnett 1991,1995, 1997, 1998, Kurtzman 1992, 1995, Valente,Ramos & Leoncini 1999). In this region, A. cognataand A. americana isolates were most similar; however,based on the unidentified region, sequences betweenA. cognata and A. americana were the most different.RAPD, ISSR, and AFLP analyses also show the

division of isolates into three distinct clusters thatcorrelate with A. americana, B. nematodophila, and thenewly recognized A. cognata.

Of special note is molecular evidence suggestingA. cognata and A. americana are independently co-evolving with the nematode from which they were iso-lated. Nematodes from each Ascobotryozyma isolationsite were sent to S. Patricia Stock (University of Cali-fornia at Davis Nematode Collection) for identificationand were also included in a study on the morphologi-cal and molecular characterization of the genusPanagrellus. All isolates were identified as P. dubius.Based on sequence data from approximately 1000 bpof the nuc-LSU (28S) rDNA, the nematode isolatesclustered by state, Washington (where A. americanawas isolated) and Idaho (where A. cognata was iso-lated), and the two branches formed a single clade with100% bootstrap support (S. P. Stock & S. A. Nadler,unpubl.). Sequence divergence among the nematodeisolates was the same as among the Ascobotryozymaisolates, with the separation coinciding with yeast spe-ciation. This provides additional evidence of a geneticseparation of the yeast and its nematode associatebased on location, and bears upon the coevolution ofthese two organisms. Thalli of A. americana were ableto adhere to nematodes from which A. cognata was

CBS 8462CBS 8560

CBS 7426CBS 7442

CBS 8706CBS 8460CBS 8753CBS 8707CBS 8708

A. cognata

A. americana

CBS 9135

1

B. nematodophila

Fig. 15. Phylogenetic tree calculated from neighbor-joininganalysis of an unknown genomic region amplified with pri-mers ML1 and ML1R. Branch lengths are proportional to

nucleotide differences, as indicated in the scale bar (%).

Table 3. Summary of phenotypic characteristics of Ascobotryozyma

cognata strains.

Fermentation:

Gas formation not detectable

Growth on:

D-Glucose v D-Mannitol w/+D-Galactose – Galactitol –

L-Sorbose v myo-Inositol –

D-Glucosamine – D-Glucono-d-lactone –

D-Ribose – 2-Keto-D-gluconate +D-Xylose – D-Gluconate –

L-Arabinose – D-Glucoronate –

D-Arabinose – D-Galacturonic acid –

L-Rhamnose – DL-Lactate w/+Sucrose – Succinate +Maltose – Citrate –

a,a-Trehalose – Methanol –

Me-a-D-glucoside – Ethanol v

Cellobiose – Propane 1,2 diol –

Salicin – Butane 2,3 diol –

Arbutin – Quinic acid –

Melibiose – Saccharate –

Lactose – Galactonic acid –

Raffinose – Nitrate –

Melezitose – Nitrite –

Inuline – Ethylamine +Soluble starch – L-Lysine +Glycerol + Cadaverine –

meso-Erythritol – Creatine –

Ribitol – Creatinine –

Xylitol – Glucosamine –

L-Arabinitol – Imidazol –

D-Glucitol w/+ 0.1% Cycloheximide +Growth at:

37 x v

40 x –

w, weak; v, variable.

Ascobotryozyma cognata sp. nov. 1118

isolated and vice versa; therefore, the fungus is notnematode-specific, at least at the species level. Thebeetles are believed to passively vector nematodes, butnothing is known about the population genetics ofthese beetles. Perhaps speciation is driven by distance,which is governed by the distribution of the beetlevectors.

It appears Ascobotryozyma is widely dispersed, butnot densely, in habitats where conditions are suitablefor both the yeast and nematode. An additional Botryo-zyma isolate was obtained from southern Oregon. Itwas collected from slime flux from hybrid poplars, P.deltoidesrtrichocarpa, infested by the dusky sap beetle,Carpophilus lugubris (Coleoptera : Nitidulidae) (MelodyPutnam, pers. comm.). Based on AFLP banding pat-terns and nuc-LSU sequence data this isolate differsfrom the other three species, but more isolates areneeded to examine its relationship in greater detail.Although isolates have been found from different sub-strates in different regions, all have been associatedwith Panagrellus nematodes and a rich microbial biota.Investigations of similar environments will most likelyyield additional isolates and contribute to a betterunderstanding of the distribution and speciation of thisyeast.

ACKNOWLEDGEMENTS

This is paper PPNS No. 0342, Department of Plant Pathology,

College of Agriculture and Home Economics Research Center,

Project No. 0572 (Washington State University, Pullman). We thank

S. Patricia Stock (University of Arizona, Tucson) for identification

of nematode isolates, and Steven J. Brunsfeld (University of Idaho,

Moscow) for identification of Salix scouleriana. We also thank Tobin

Peever and TimMurray (Washington State University) for assistance

with sequencing and use of AFLP equipment, respectively. Lisette

Smit is thanked for technical assistance with RAPD- and ISSR-PCR

experiments.

REFERENCES

Bassam, B. J., Caetano-Anolles, G. & Gresshoff, P. M. (1991) Fast

and sensitive silver staining of DNA in polyacrylamide gels. Annals

of Biochemistry 196 : 80–83.

Dice, L. R. (1945) Measures of the amount of ecologic association

between species. Ecology 26 : 297–302.

Felsenstein, J. (1993) PHYLIP (Phylogeny Inference Package). Ver-

sion 3.5c. Distributed by the author, Department of Genetics,

University of Washington, Seattle.

Garbelotto, M., Bruns, T. D., Cobb, F. W. & Otrosina, W. J. (1993)

Differentiation in intersterility groups by geographic provenances

among isolates of Heterobasidion annosum detected by random

amplified polymorphic DNA assays. Canadian Journal of Botany

71 : 565–569.

Guadet, J., Julien, J., Lafey, J. & Brygoo, Y. (1989) Phylogeny

of some Fusarium species, as determined by large subunit rRNA

sequence comparison.Molecular Biology and Evolution 6 : 227–242.

Kerrigan, J., Smith, M. T., Rogers, J. D. & Poot, G. A. (2001) Asco-

botryozyma americana gen. nov. et sp. nov. and its anamorph

Botryozyma americana, an unusual yeast from the surface of

nematodes. Antonie van Leeuwenhoek 79 : 7–16.

Kurtzman, C. P. (1987) Prediction of biological relatedness among

yeasts from comparisons of nuclear DNA complementarity.

Studies in Mycology 30 : 459–468.

Kurtzman, C. P. (1992) The systematics of ascomycetous yeasts

defined from ribosomal RNA sequence divergence: theoretical and

practical considerations. InThe Fungal Holomorph: mitotic, meiotic

and pleomorphic speciation in fungal systematics (D. R. Reynolds &

J. W. Taylor, eds) : 271–279. CAB International, Wallingford.

Kurtzman, C. P. (1995) Relationships among the genera Ashbya,

Eremothecium, Holleya and Nematospora determined from rDNA

sequence divergence. Journal of Industrial Microbiology 14 :

523–530.

Kurtzman, C. P. & Robnett, C. J. (1991) Phylogenetic relationships

among species of Saccharomyces, Schizosaccharomyces, Debaryo-

myces and Schwanniomyces determined from partial ribosomal

RNA sequences. Yeast 7 : 61–72.

Kurtzman, C. P. & Robnett, C. J. (1995) Molecular relationships

among hyphal ascomycetous yeasts and yeastlike taxa. Canadian

Journal of Botany 73 : S824–S830.

Kurtzman, C. P. & Robnett, C. J. (1997) Identification of clinically

important ascomycetous yeasts based on nucleotide divergence in

the 5k end of the large-subunit (26S) ribosomal DNA gene. Journal

of Clinical Microbiology 35 : 1216–1223.

Kurtzman, C. P. & Robnett, C. J. (1998) Identification and phy-

logeny of ascomycetous yeasts from analysis of nuclear large

subunit (26S) ribosomal DNA partial sequences. Antonie van

Leeuwenhoek 73 : 331–371.

Kurtzman, C. P., Smiley, M. J., Johnson, C. J. & Wickerham, L. J.

(1980) Two closely related heterothallic species, Pichia amylophila

and Pichia mississipiensis : characterization by hybridization and

deoxyribonucleic acid reassociation. International Journal of Sys-

tematic Bacteriology 30 : 206–216.

Lee, S. & Taylor, J. W. (1990) Isolation of DNA from fungal mycelia

and single spores. In PCR Protocols: a guide to methods and ap-

plications (M. A. Innis, D. H. Gelfand, J. J. Sninsky & T. J. White,

eds): 282–287. Academic Press, San Diego.

Messner, R., Prillinger, H., Altmann, F., Lopandic, K., Wimmer, K.,

Molnar, O. & Weigang, F. (1994) Molecular characterization

and application of random amplified polymorphic DNA analysis

of Mrakia and Sterigmatomyces species. International Journal

Systematic Bacteriology 44 : 694–703.

Metcalf, C. L. & Flint, W. P. (1939) Destructive and Useful Insects.

2nd edn. McGraw-Hill, New York.

Moller, E. M., Bahnweg, G., Sandermann, H. & Geiger, H. H. (1992)

A simple and efficient protocol for isolation of high molecular

weight DNA from filamentous fungi, fruit bodies, and infected

plant tissues. Nucleic Acids Research 20 : 6115–6116.

O’Donnell, K. L. (1993) Fusarium and its near relatives. In The

Fungal Holomorph: mitotic, meiotic, and pleomorphic speciation in

fungal systematics (D. R. Reynolds & J. W. Taylor, eds) : 225–233.

CAB International, Wallingford.

Page, R. (1996) TREEVIEW: an application to display phylogenetic

trees on personal computers. Computer Applications in the Bio-

sciences 12 : 357–358.

Rozen, S. & Skaletsky, H. J. (1998) Primer3. Code available at http://

www-genome.wi.mit.edu/genome_software/other/primer3.html.

Seidler, R. J. & Mandel, M. (1971) Quantitative aspects of deoxy-

ribonucleic acid renaturation: base composition, site of chromo-

some replication, and polynucleotide homologies. Journal of

Bacteriology 106 : 608–614.

Shann, C. (1987) Presenza di nematodi in uve affette da marciume

acido. L’Informatore Agrario 43 : 137–138.

Smith, M. T., de Cock, A. W. A. M., Poot, G. A. & Steensma, H. Y.

(1995) Genome comparison in the yeastlike fungal genus Galacto-

myces Redhead et Malloch. International Journal of Systematic

Bacteriology 45 : 826–831.

Smith, M. T., Shann, C., Batenburg-van der Vegte, W. H., Schmitt,

R., Wehrli, E., Roeijmans, H. J. & van Eijk, G. W. (1992)

Botryozyma nematodophila gen. nov., spec. nov. (Candidaceae).

Antonie van Leeuwenhoek 61 : 277–284.

J. Kerrigan and others 1119

Smith, B. D. & Stott, K. G. (1964) The life history and behaviour

of the willow weevil Cryptorhynchus lapathi L. Annals of Applied

Biology 54 : 141–151.

Thompson, J., Gibson, T., Plewniak, F., Jeanmougin, F. & Higgins,

D. (1997) The ClustalX windows interface: flexible strategies

for multiple sequence alignment aided by quality analysis tools.

Nucleic Acids Research 24 : 4876–4882.

Vassart, G., Georges, M., Monsieur, R., Brocas, H., Lequarre, A. S.

& Christophe, D. (1987) A sequence in M13 phage detects

hypervariable minisatelittes in human and animal DNA. Science

235 : 683–684.

Valente, P., Ramos, J. P. & Leoncini, O. (1999) Sequencing as a tool

in yeast molecular taxonomy. Canadian Journal of Microbiology

45 : 949–958.

Vos, P., Hogers, R., Bleeker, M., Reijans, M., Lee van de, T., Hornes,

M., Frijters, A., Pot, J., Peleman, J., Kuiper, M. & Zabeau, M.

(1995) AFLP: a new technique for DNA fingerprinting. Nucleic

Acids Research 23 : 4407–4414.

Welsh, J. & McClelland, M. (1990) Fingerprinting genomes

using PCR with arbitrary primers. Nucleic Acids Research 18 :

7213–7218.

White, T. J., Bruns, T., Lee, S. & Taylor, J. (1990) Amplification and

direct sequencing of fungal ribosomal RNA genes for phylo-

genetics. In PCR protocols: a guide to methods and applications

(M. A. Innis, D. H. Gelfand, J. J. Sninsky & T. J. White, eds) :

315–322. Academic Press, San Diego.

Yarrow, D. (1998) Methods for isolation, maintenance, classification

and identification of yeasts. In The Yeasts, a taxonomic study

(C. P. Kurtzman & J. Fell, eds) : 77–101. Elsevier Science Publi-

cations, Amsterdam.

Corresponding Editor: J. K. Stone

Ascobotryozyma cognata sp. nov. 1120