molecular phylogeny of laetiporus and other brown rot

Molecular phylogeny of Laetiporus and other brown rot polypore genera inNorth America

Daniel L. Lindner1

Mark T. BanikU.S.D.A. Forest Service, Madison Field Office of theNorthern Research Station, Center for Forest MycologyResearch, One Gifford Pinchot Drive, Madison,Wisconsin 53726

Abstract: Phylogenetic relationships were investigat-ed among North American species of Laetiporus,Leptoporus, Phaeolus, Pycnoporellus and Wolfiporiausing ITS, nuclear large subunit and mitochondrialsmall subunit rDNA sequences. Members of thesegenera have poroid hymenophores, simple septatehyphae and cause brown rots in a variety of substrates.Analyses indicate that Laetiporus and Wolfiporia arenot monophyletic. All North American Laetiporusspecies formed a well supported monophyletic group(the ‘‘core Laetiporus clade’’ or Laetiporus s.s.) withthe exception of L. persicinus, which showed littleaffinity for any genus for which sequence data areavailable. Based on data from GenBank, the southernhemisphere species L. portentosus also fell welloutside the core Laetiporus clade. Wolfiporia dilatohy-pha was found to represent a sister group to the coreLaetiporus clade. Isolates of Phaeolus, Pycnoporellusand members of the core Laetiporus clade all fellwithin the Antrodia clade of polypores, whileLeptoporus mollis and Laetiporus portentosus fell withinthe phlebioid clade of polypores. Wolfiporia cocosisolates also fell in the Antrodia clade, in contrast toprevious studies that placed W. cocos in the corepolyporoid clade. ITS analyses resolved eight cladeswithin Laetiporus s.s., three of which might representundescribed species. A combined analysis using thethree DNA regions resolved five major clades withinLaetiporus s.s.: a clade containing conifer-inhabitingspecies (‘‘Conifericola clade’’), a clade containing L.cincinnatus (‘‘Cincinnatus clade’’), a clade contain-ing L. sulphureus s.s. isolates with yellow pores(‘‘Sulphureus clade I’’), a clade containing L.sulphureus s.s. isolates with white pores (‘‘Sulphureusclade II’’) and a clade containing L. gilbertsonii andunidentified isolates from the Caribbean (‘‘Gilbertso-nii clade’’). Although there is strong support forgroups within the core Laetiporus clade, relationshipsamong these groups remain poorly resolved.

Key words: evolution, Fungi, Macrohyporia,Polyporaceae, Poria, root rot, sulfur shelf, Wolfiporiaextensa

INTRODUCTION

The genera Laetiporus Murrill, Leptoporus Quel.,Phaeolus (Pat.) Pat., Pycnoporellus Murrill and Wolfi-poria Ryvarden & Gilb. contain species that possesssimple septate hyphae, cause brown rots and produceannual, polyporoid fruiting bodies with hyalinespores. These shared morphological and physiologi-cal characters have been considered important intraditional polypore taxonomy (e.g. Gilbertson andRyvarden 1986, Gilbertson and Ryvarden 1987,Ryvarden 1991). However recent molecular workindicates that Laetiporus, Phaeolus and Pycnoporellusfall within the ‘‘Antrodia clade’’ of true polyporesidentified by Hibbett and Donoghue (2001) whileLeptoporus and Wolfiporia fall respectively within the‘‘phlebioid’’ and ‘‘core polyporoid’’ clades of truepolypores (Binder et al 2005).

Recent molecular and mating studies also have ledto a revision of genus Laetiporus in North America(Banik et al 1998, Banik and Burdsall 1999, Banik andBurdsall 2000), which formerly contained only L.sulphureus (Bull.) Murrill and L. persicinus (Berk &M.A. Curtis) Gilbertson. Burdsall and Banik (2001)concluded that genus Laetiporus contains at least sixmorphologically and ecologically distinct species.Despite this recent work relationships remain unre-solved among North American Laetiporus species andamong other brown rot polyporoid species lackingclamps. This study had two goals: (i) to determinerelationships among North American species ofLaetiporus and (ii) to determine whether any NorthAmerica polyporoid species that cause brown rots andlack clamps are closely related to Laetiporus. Toaccomplish these goals species of Laetiporus, Lepto-porus, Phaeolus, Pycnoporellus and Wolfiporia weresequenced in three regions of rDNA: the intergenictranscribed spacer (ITS), the nuclear large subunit(nLSU) and the mitochondrial small subunit(mtSSU).

Of the genera included in this study the mostwidely known is Laetiporus. As it is currently definedLaetiporus includes annual polypore species thatproduce a brown rot, have dimitic binding hyphaeand lack cystidia and clamp connections (Gilbertson

Accepted for publication 26 February 2008.1 Corresponding author. E-mail: [email protected]

Mycologia, 100(3), 2008, pp. 417–430. DOI: 10.3852/07-124R2# 2008 by The Mycological Society of America, Lawrence, KS 66044-8897

417

and Ryvarden 1986, Ryvarden 1991). With theexception of L. persicinus, all North AmericanLaetiporus species produce brightly colored, conspic-uous fruiting bodies that have been regarded tradi-tionally as part of the L. sulphureus s.l. speciescomplex. Species in the L. sulphureus s.l. complexare popular edibles that frequently are collectedunder the common name ‘‘Sulfur Shelf’’ or ‘‘Chickenof the Woods’’ (Arora 1986). Laetiporus sulphureus s.l.also has been investigated for novel antimicrobial andmedicinal compounds (Turkoglu et al 2007) and itsbright pigments have been examined for potential asfood colorants (Davoli et al 2005). Various authorsfrom the late 19th and early 20th centuries describedadditional varieties and species within the L. sulphur-eus s.l. complex to account for the wide range ofmorphological and ecological variation exhibited bythis species, although few of these names were usedwidely or consistently. Burdsall and Banik (2001)recognize five species and an additional variety withinthe L. sulphureus s.l. complex: L. cincinnatus (Mor-gan) Burds., Banik & T.J. Volk, L. conifericola Burds.& Banik, L. gilbertsonii Burds., L. gilbertsonii var.pallidus Burds., L. huroniensis Burds. & Banik and L.sulphureus s.s.

With the exception of Laetiporus cincinnatus, allspecies in the L. sulphureus s.l. complex typicallyproduce sessile to laterally substipitate pilei withbright orange upper surfaces and yellow or whitepore layers (Burdsall and Banik 2001). Laetiporuscincinnatus produces centrally stipitate, rosette-shaped fruiting bodies with a white pore layer; thesefruiting bodies arise near the base of large diameterhardwood trees, usually Quercus species, in theeastern and midwestern United States. In NorthAmerica Laetiporus sulphureus s.s. (as defined byBurdsall and Banik 2001 and the present study) isfound in the eastern and midwestern United Statesand often occurs on Quercus species althoughspecimens occasionally are found on other hard-woods. Preliminary evidence based on a limitednumber of ITS sequences suggests that NorthAmerican and European populations of L. sulphureuss.s. are conspecific (unpubl data), however furthercomparative work is needed. If species barriers existbetween North American and European populationsof L. sulphureus s.s., nomenclatural revision of theNorth American species would be required given thatL. sulphureus s.s. is typified based on Europeanmaterial.

The recently described species L. conifericola and L.huroniensis (Burdsall and Banik 2001) are restrictedto conifers, with L. huroniensis being found in theupper midwestern United States and L. conifericolabeing restricted to western North America. Laetiporus

gilbertsonii occurs in the southern and western regionsof North America and occurs on a wide range ofhardwoods including Eucalyptus. The upper pileussurface of L. gilbertsonii is generally more salmon topink than other Laetiporus species while the porelayer is either yellow or white, which differentiates thetwo color forms, L. gilbertsonii with a yellow pore layerand L. gilbertsonii var. pallidus with a white pore layer.

Laetiporus persicinus is the one North AmericanLaetiporus species not considered part of the L.sulphureus s.l. complex. This is due to strikingmacroscopic differences, including a brown to pink-ish-brown pileus surface, a pinkish-cream pore layerand flesh and tubes that bruise blackish-brown.Laetiporus persicinus originally was described byBerkeley and Curtis in 1872 as Polyporus persicinusbut was transferred to Laetiporus in 1981 by R.Gilbertson (Gilbertson 1981). Laetiporus persicinus isfound in the southeastern United States occurringprimarily on Quercus although it is found occasionallyon conifers. Burdsall and Banik (2001) hypothesizedthat L. persicinus might not be closely related tospecies in the L. sulphureus s.l. complex based on thedark pigmentations in the fruiting bodies, theappearance of the binding hyphae and preliminaryRFLP data.

Work by Hibbett and Donoghue (1995) withmtSSU rDNA sequences indicated that genus Phaeo-lus is closely related to Laetiporus sulphureus s.l. Thisresult has been observed in many analyses (Hibbettand Donoghue 2001, Hibbett and Binder 2002,Binder et al 2005), which confirmed that Phaeolus isnot aligned with the Hymenochaetales despite super-ficial similarities (Wagner and Fischer 2001). Thesingle North American species of Phaeolus treated byGilbertson and Ryvarden (1987), P. schweinitzii (Fr.)Pat., is an important root rot pathogen in coniferecosystems. Like L. cincinnatus and L. persicinus, P.schweinitzii has a centrally stipitate fruiting bodyfound at the base of trees. Unlike Laetiporus species,it possesses a monomitic hyphal system and produceshymenial cystidia. Young Phaeolus fruiting bodies areyellow to orange but become predominantly brownwith age (Gilbertson and Ryvarden 1987).

Genus Pycnoporellus also contains species withmonomitic, yellow to orange fruiting bodies, leadingGilbertson and Ryvarden (1987) to suggest a closerelationship between Phaeolus and Pycnoporellus.Gilbertson (1981) included both North Americanspecies of Pycnoporellus, P. alboluteus (Ellis & Everh.)Kotl. & Pouzar and P. fulgens (Fr.) Donk, in genusPhaeolus (Gilbertson 1981). However Gilbertson andRyvarden (1987) placed these species in Pycnoporellus.Like Phaeolus both Pycnoporellus species grow primar-ily on coniferous wood although neither forms the

418 MYCOLOGIA

centrally stipitate fruiting bodies typical of Phaeolus.Pycnoporellus alboluteus is resupinate and occursprimarily in western North America while P. fulgensis a pileate species found predominantly in easternNorth America. A distinguishing characteristic ofPycnoporellus species is a bright red reaction pro-duced by 2% KOH on fruiting bodies; Phaeolusschweinitzii in contrast turns dark brown to black in2% KOH (Gilbertson and Ryvarden 1987).

Species in genus Wolfiporia show microscopicsimilarities to species of Laetiporus, Phaeolus andPycnoporellus despite producing macroscopic fruitingbodies that are thin, resupinate and lacking in brightpigmentations. The two species of Wolfiporia reportedfrom North America, W. cocos (F.A. Wolf) Ryvarden &Gilb. and W. dilatohypha Ryvarden & Gilb., are dimiticand possess greatly inflated skeletal hyphae, adefining characteristic of this genus (Ryvarden1991). Although Wolfiporia cocos also has been knownas W. extensa (Peck) Ginns, Redhead and Ginns(2006) proposed to conserve the basionym Poria cocosagainst Daedalea extensa due to a lack of acceptance ofthe nomenclaturally correct name W. extensa. Theconservation of the epithet cocos should stabilize thenomenclature associated with this species and endyears of nomenclatural ambiguity (see Ginns andLowe 1983, Ginns 1984). Wolfiporia cocos is mostwidely known for producing large, edible sclerotiathat have been referred to as ‘‘tuckahoes’’ or ‘‘Indianbread’’ in North America (Weber 1929). The sclerotiaof W. cocos have been reported from China, usuallyunder the name Poria cocos, where they have beencollected and used medicinally (Zhang et al 1997,Wang et al 2004, Wu et al 2004, etc.).

Leptoporus is a monotypic genus in North Americacontaining species L. mollis (Pers.) Quelet. Fruitingbodies of L. mollis are sessile to effused-reflexed anddisplay reddish-purple to purple-brown coloration(Gilbertson and Ryvarden 1986, Ryvarden 1991).Leptoporus is similar to Ceriporia microscopically,having a monomitic hyphal system and lacking bothclamp connections and cystidia (Gilbertson andRyvarden 1986). Ceriporia has been delimited fromLeptoporus based solely on the type of decay, withCeriporia species producing a white rot (Gilbertsonand Ryvarden 1986). Leptoporus mollis is widespreadin North America but seems to be collected rarelybased on the number of collections in herbaria.

Although the genera examined in this study areknown to be a polyphyletic assemblage, they werechosen as a starting point in the search for speciesthat might be closely related to Laetiporus. The goalof this study was not to complete an exhaustiveevolutionary investigation of all of these genera but todetermine whether a realignment of species within

these genera might be warranted in light of recentwork on Laetiporus. Many studies have demonstratedthat nonpolyporoid genera, including the spathulategenus Sparassis and numerous corticioid genera, areclosely related to the genera in this study (Hibbettand Donoghue 2001, Binder et al 2005). Thisemphasizes the need for future work that includes awide range of species with diverse morphologies aswell as species that produce clamp connections. Thiswork is a first step toward increasing taxon samplingof species that might fall within the Antrodia clade oftrue polypores while also determining relationshipsamong species in genus Laetiporus.

MATERIALS AND METHODS

DNA isolation, PCR and sequencing.—Cultures to besequenced were obtained from the culture collection ofthe Center for Forest Mycology Research, U.S.D.A. ForestService, Madison Field Office of the Northern ResearchStation, which is housed at the Forest Products Laboratoryin Madison, Wisconsin. Each culture was grown on potato-dextrose agar and a small amount of the resulting myceliumwithout associated agar was prepared for use as templateDNA in PCR following the protocol of Volk et al (1996).

Two regions of nuclear rDNA and one of mitochondrialrDNA were amplified in PCR. Except where otherwise notedprimer designations are those of White et al (1990). PCRwas performed with 53 Green GoTaq reaction buffer andGoTaq DNA polymerase (Promega, Madison, Wisconsin).GoTaq reaction buffer was diluted to a 13 workingconcentration and 0.025 units of GoTaq DNA polymerasewere added per microliter of reaction volume. Each primerhad a final concentration of 0.2 mM and each dNTP(Promega, Madison, Wisconsin) had a final concentrationof 200 mM. Thermocycler conditions were: initial denatur-ing at 94 C for 2 min; 30 cycles of denaturing at 94 C for40 s, annealing at 53 C for 40 s and extension at 72 C for120 s; and a final extension step of 72 C for 5 min.

A section of nLSU rDNA was amplified with primersLR16/LROR. For most of the isolates a section of ITS wasamplified with ITS4/ITS5. Partial ITS sequences fromisolates PR-2 and PR-6326 were amplified with primersITS3/ITS4, while ITS2/ITS5 and ITS3/ITS4 were used forWolfiporia cocos isolates. A section of mtSSU rDNA wasamplified with ms-1/ms-2. Attempts at amplification of eachof these three regions were made from cultures of thesetaxa: Laetiporus cincinnatus, L. conifericola, L. gilbertsonii, L.gilbertsonii var. pallidus, L. huroniensis, L. persicinus, L.sulphureus s.s., Leptoporus mollis, Phaeolus schweinitzii,Pycnoporellus alboluteus, P. fulgens, Wolfiporia cocos, W.dilatophya and the unidentified isolates PR-2, PR-6326, PR-914, GDL-1, EUC-1 and KOA-1. Preliminary morphologicaldata placed all unidentified isolates in genus Laetiporus.Collection information for all isolates is provided (TABLE I).

PCR products were purified with QIAquick PCR Purifica-tion Kit (QIAGEN) according to the manufacturer’sprotocol. Sequencing reactions were performed following

LINDNER AND BANIK: LAETIPORUS AND OTHER GENERA 419

TA

BL

EI.

Iso

late

sse

qu

ence

dfo

rth

isst

ud

y

Spec

ies

Co

llec

tio

nN

um

ber

Lo

cali

ty

Gen

Ban

kA

cces

sio

nN

um

ber

s

ITS

nL

SUm

tSSU

Laet

ipor

us

cin

cin

natu

sD

A-3

7D

ane

Co

.,W

isco

nsi

n,

USA

EU

4025

57E

U40

2521

EU

4024

85L

.ci

nci

nn

atu

sM

AS-

1Su

ffo

lkC

o.,

Mas

sach

use

tts,

USA

EU

4025

58L

.ci

nci

nn

atu

sW

S-1

Was

hin

gto

nC

o.,

Wis

con

sin

,U

SAE

U40

2559

L.

cin

cin

natu

s46

-110

4O

rlea

ns

Par

ish

,L

ou

sian

a,U

SAE

U40

2560

EU

4025

22E

U40

2484

L.

con

ifer

icol

aA

K-1

Mu

nic

ipal

ity

of

An

cho

rage

,A

lask

a,U

SAE

U40

2574

L.

con

ifer

icol

aC

A-8

Men

do

cin

oC

o.,

Cal

ifo

rnia

,U

SAE

U40

2575

EU

4025

23E

U40

2487

L.

con

ifer

icol

aJA

M-1

Ket

chik

anG

atew

ayB

oro

ugh

,A

lask

a,U

SAE

U40

2577

EU

4025

24E

U40

2486

L.

con

ifer

icol

aN

V-2

Was

ho

eC

o.,

Nev

ada,

USA

EU

4025

78E

U40

2488

L.

con

ifer

icol

aT

RE

E-5

Ken

aiP

enin

sula

Bo

rou

gh,

Ala

ska,

USA

EU

4025

76L

.gi

lber

tson

iiC

A-1

3N

apa

Co

.,C

alif

orn

ia,

USA

EU

4025

49E

U40

2527

EU

4024

96L

.gi

lber

tson

iiO

R-2

Mar

ion

Co

.,O

rego

n,

USA

EU

4025

50E

U40

2495

L.

gilb

erts

onii

var.

pall

idu

sT

JV20

00-1

01W

aku

lla

Co

.,F

lord

ia,

USA

EU

4025

53E

U40

2528

EU

4024

93L

.gi

lber

tson

iiva

r.pall

idu

sL

-129

54Sa

nJo

seP

rov.

,C

ost

aR

ica

EU

4025

51L

.gi

lber

tson

iiva

r.pall

idu

sF

P15

0268

St.

An

dre

ws

Par

ish

,Ja

mai

caE

U40

2552

EU

4025

29E

U40

2494

L.

hu

ron

ien

sis

HM

C-1

Mar

qu

ette

Co

.,M

ich

igan

,U

SAE

U40

2569

L.

hu

ron

ien

sis

HM

C-2

Mar

qu

ette

Co

.,M

ich

igan

,U

SAE

U40

2570

EU

4024

90L

.hu

ron

ien

sis

HM

C-3

Mar

qu

ette

Co

.,M

ich

igan

,U

SAE

U40

2571

EU

4025

40L

.hu

ron

ien

sis

MI-

7M

arq

uet

teC

o.,

Mic

hig

an,

USA

EU

4025

72L

.hu

ron

ien

sis

MI-

14G

oge

bic

Co

.,M

ich

igan

,U

SAE

U40

2573

EU

4025

39E

U40

2489

L.

per

sici

nu

sH

HB

9564

Leo

nC

o.,

Flo

rdia

,U

SAE

U40

2579

EU

4025

13L

.per

sici

nu

sH

HB

9668

Leo

nC

o.,

Flo

rdia

,U

SAE

U40

2580

EU

4025

03L

.per

sici

nu

sR

LG

1472

5E

ast

Bat

on

Ro

uge

Par

ish

,L

ou

sian

a,U

SAE

U40

2581

EU

4025

12E

U40

2502

L.

per

sici

nu

sR

LG

1473

9E

ast

Bat

on

Ro

uge

Par

ish

,L

ou

sian

a,U

SA(?

)E

U40

2582

L.

sulp

hu

reu

sC

T-1

Fai

rfie

ldC

o.,

Co

nn

ecti

cut,

USA

EU

4025

65E

U40

2532

EU

4024

79L

.su

lphu

reu

sD

A-4

1D

ane

Co

.,W

isco

nsi

n,

USA

EU

4025

66E

U40

2533

EU

4024

81L

.su

lphu

reu

sE

RT

-713

Mis

siss

ipp

i,U

SAE

U40

2564

L.

sulp

hu

reu

sG

R-1

2G

ran

tC

o.,

Wis

con

sin

,U

SAE

U40

2561

EU

4025

34E

U40

2480

L.

sulp

hu

reu

sN

J-2

Mo

rsic

Co

.,N

ewJe

rsey

,U

SAE

U40

2562

L.

sulp

hu

reu

sT

JV95

-84

Dan

eC

o.,

Wis

con

sin

,U

SAE

U40

2563

L.

sulp

hu

reu

sM

AS-

2M

idd

lese

xC

o.,

Mas

sach

use

tts,

USA

EU

4025

68E

U40

2531

EU

4024

91L

.su

lphu

reu

sT

JV99

-150

La

Cro

sse

Co

.,W

isco

nsi

n,

USA

EU

4025

67E

U40

2530

EU

4024

92L

.sp

.G

DL

-1G

uad

elo

up

e,L

esse

rA

nti

lles

,F

ran

ce(C

arib

bea

n)

EU

4025

47E

U40

2525

EU

4024

83L

.sp

.P

R-9

14L

uq

uil

loM

ts,

Pu

erto

Ric

o,

USA

EU

4025

48E

U40

2526

EU

4024

82L

.sp

.E

UC

-1M

aui

Co

.,H

awai

’i,

USA

EU

4025

45E

U40

2541

L.

sp.

KO

A-1

Haw

ai’i

Co

.,H

awai

’i,

USA

EU

4025

46E

U40

2542

Lep

topor

us

mol

lis

RL

G71

63P

ima

Co

.,A

rizo

na,

USA

EU

4025

83E

U40

2511

EU

4025

06L

.m

olli

sT

JV93

-174

Jeff

erso

nC

o.,

Was

hin

gto

n,

USA

EU

4025

84E

U40

2510

EU

4025

07P

haeo

lus

schw

ein

itzi

iD

A-3

8D

ane

Co

.,W

isco

nsi

n,

USA

EU

4025

85E

U40

2514

EU

4025

08P

.sc

hw

ein

itzi

iH

HB

1892

4A

lask

a,U

SAE

U40

2586

EU

4025

15E

U40

2509

420 MYCOLOGIA

the BigDye terminator protocol (ABI Prism). The purifiedPCR products for each isolate were sequenced from boththe 39 and 59 direction with the same primers used in thePCR amplification. Sequencing products were analyzed atthe University of Wisconsin Biotechnology Center (Madi-son, Wisconsin) with an ABI Prism 377 sequencer.Complementary sequences for each isolate were comparedand discrepancies resolved with the sequence electrophe-rograms. For analysis purposes all sequences were termi-nated at the base immediately adjacent to the primerattachment point. In addition to the primers used inamplification two primers were synthesized: FPLms-4(TGTACTTTAGGTCCTAAA) and FPLms-7 (ACAGGATGTGCGACTTGC). These primers were used to se-quence long mtSSU amplification products that occurred inL. persicinus, P. fulgens and two of three isolates of W.dilatohypha.

Phylogenetic analyses.—Sequences were aligned manuallywith Sequence Alignment Editor (Se-Al) v2.0a9. Sequenceswere deposited in GenBank (see TABLE I) and sequencealignments for nLSU, mtSSU, ITS and a combined datasetof all three regions were deposited in TreeBase (accession #S1994). Maximum parsimony was implemented in PAUP4.0b10 (Swofford 2002) and Bayesian analysis in MrBayes 3.1.2(Huelsenbeck and Ronquist 2001, Ronquist and Huelsenbeck2003).

The nLSU sequences were aligned with 15 additionalsequences from GenBank: Antrodia carbonica AF287844,Ceriporia purpurea AF287852, Dacryobolus sudans AY293176,Fomes fomentarius AF287857, Grifola frondosa AY629318, G.frondosa AY826982, Laetiporus portentosus AY293191, Par-mastomyces transmutans AF518635, Pleurotus ostreatusU04140, Polyporus squamosus AF393069, Sarcodon imbrica-tus AY586711, Sparassis crispa AY218386, S. spathulataAF287889, S. spathulata AY218396 and Wolfiporia cocosAF393081. The sequences from GenBank represent aselection of species from the three major polyporoid clades(Antrodia, phlebioid and core polyporoid) defined byBinder et al (2005), as well as two sequences from outsidethe polyporoid clade, which were included as outgroups.Sequences of nLSU from two isolates originally identified asLaetiporus persicinus (PR-2 and PR-6326, ‘‘Polyporus talpae’’in TABLE I) were excluded from analyses because they couldnot be aligned with other species; BLAST analyses of thesesequences produced no significant matches.

For nLSU data, heuristic searches were conducted inPAUP with characters unordered and of equal weight andgaps treated as missing data. Default settings were used withthese exceptions: stepwise-addition option was set torandom with 100 replicates, steepest descent was used withthe TBR branch swapping option and the number of treessaved was set to automatically increase. Bootstrap supportfor clades (Felsenstein 1985) was estimated from 1000heuristic searches with the same settings described above,with the exception that the stepwise-addition option was setto random with 50 replicates. Bayesian inference wasimplemented in MrBayes using default settings with theGTR model. Two million generations were performed withsamples taken in increments of 100. The first 5000 trees

Spec

ies

Co

llec

tio

nN

um

ber

Lo

cali

ty

Gen

Ban

kA

cces

sio

nN

um

ber

s

ITS

nL

SUm

tSSU

P.

schw

ein

itzi

iM

I-5

Go

geb

icC

o.,

Mic

hig

an,

USA

EU

4025

87‘‘

Pol

ypor

us

talp

ae’

’P

R-2

Pu

erto

Ric

o,

USA

EU

4025

43‘‘

P.

talp

ae’

’P

R-6

326

Lu

qu

illo

Mts

,P

uer

toR

ico

,U

SAE

U40

2544

Pyc

nop

orel

lus

alb

olu

teu

sH

HB

1281

6M

un

icip

alit

yo

fA

nch

ora

ge,

Ala

ska,

USA

EU

4025

88E

U40

2538

P.

alb

olu

teu

sH

HB

1759

8K

enai

Pen

insu

laB

oro

ugh

,A

lask

a,U

SAE

U40

2589

EU

4025

37P

.alb

olu

teu

sO

KM

6406

Fla

thea

dC

o.,

Mo

nta

na,

USA

EU

4025

90P

.fu

lgen

sC

A-2

0M

arip

ose

Co

.,C

alif

orn

ia,

USA

EU

4025

91P

.fu

lgen

sF

P10

1689

Vil

asC

o.,

Wis

con

sin

,U

SAE

U40

2592

EU

4025

36E

U40

2504

P.

fulg

ens

HH

B17

342

Ken

aiP

enin

sula

Bo

rou

gh,

Ala

ska,

USA

EU

4025

93E

U40

2535

EU

4025

05W

olfi

por

iaco

cos

MD

-106

Ben

ton

Co

.,O

rego

n,

USA

EU

4025

94E

U40

2519

EU

4025

01W

.co

cos

MD

-275

Ben

ton

Co

.,O

rego

n,

USA

EU

4025

95E

U40

2520

EU

4025

00W

.dil

ato

hyp

ha

CS-

63C

lark

Co

.,K

entu

cky,

USA

EU

4025

55E

U40

2516

EU

4024

97W

.dil

ato

hyp

ha

FP

7216

2T

enn

esse

e,U

SAE

U40

2556

EU

4025

17E

U40

2498

W.

dil

ato

hyp

ha

FP

9408

9G

reen

bri

erC

o.,

Wes

tV

irgi

nia

,U

SAE

U40

2554

EU

4025

18E

U40

2499

TA

BL

EI.

Co

nti

nu

ed


(25%) were considered the burn-in and were excluded fromconstruction of the consensus tree. After running theanalysis the standard deviation of the split frequencies wasexamined to confirm it was below 0.01, the potential scalereduction factor was examined to confirm all parameterswere close to 1.0, and the plot of the generations versus thelog probability of the data (the log likelihood values) wasexamined to confirm it had reached the stationary phase(was neither increasing nor decreasing). Three indepen-dent Bayesian runs were performed and posterior proba-bilities were averaged across runs.

For the ITS region, sequences from species in the coreLaetiporus clade (see FIG. 1) were found to align easily withWolfiporia dilatohypha sequences, while sequences from allother species were found to be more than 25% dissimilarbased on sequence identity, making them difficult orimpossible to align. Analyses for the ITS region thereforewere performed with species in the core Laetiporus cladewith W. dilatohypha as outgroup. Ten Laetiporus sulphureusITS sequences from GenBank (AF229196, AM269785,AM269786, AY089742, AY218417, AY835667, AY835668,DQ221108, DQ450876 and EF088657) were found to alignwith our dataset and were included in analyses. Maximumparsimony and Bayesian analyses were run on the ITSsequences using the methods described for nLSU data withthese exceptions: 1 400 000 generations were run in theBayesian analysis and the burn-in was set to 3500 (25%).

Maximum parsimony and Bayesian analyses were run onthe mtSSU sequences with the methods described for ITSdata. All species (TABLE I) were included with the exceptionof species for which sequence data were not successfullyobtained (see RESULTS). Leptoporus mollis was used asoutgroup based on the results of the nLSU analysis.

A combined analysis was conducted with all isolates forwhich complete nLSU, mtSSU and ITS data were available.Species were included in this analysis if sequence data couldbe aligned across all three regions. Potential conflictsamong DNA regions were identified by first running anindependent parsimony analysis on each of the threeregions. The parsimony analysis included running aheuristic search in PAUP using default settings with theseexceptions: stepwise-addition option was set to random with100 replicates, steepest descent was used with the TBR

branch swapping option and the number of trees savedwas set to automatically increase. A strict consensus tree wasconstructed for each region, and the trees were examinedfor potential conflict. Bootstrap analyses were run for eachdataset with the methods described for the nLSU analysis.No conflicts were observed among the datasets (i.e. noconflicting branches were observed among the consensustrees and no conflicting branches were supported with abootstrap value .70), so the three regions were combinedinto one dataset. Maximum parsimony and Bayesiananalyses were run on the combined dataset with themethods described for the ITS region.

RESULTS

PCR and sequencing.—Complete ITS sequences were

obtained from 51 isolates (TABLE I) yielding 30sequence variants. All ITS sequences were 555–720 bp long except for Wolfiporia cocos, which hadan ITS length of 1863 bp. The extreme length of W.cocos ITS was due to an approximately 1000 bpinsertion in ITS 1 and an approximately 270 bpinsertion in ITS 2 when compared to Phaeolusschweinitzii.

Sequences of nLSU were obtained from 35 isolates(TABLE I) yielding 20 sequence variants. Sequences ofnLSU were 662–734 bp. Sequences of mtSSU wereobtained from 31 isolates; Pycnoporellus alboluteus andisolates PR-2 and PR-6326 did not yield mtSSUsequences. Most mtSSU sequences were 491–610 bp.However L. persicinus, P. fulgens and two of threeisolates of W. dilatohypha had large insertions thatresulted in sequence lengths in excess of 1700 bp;these insertions occurred at the same location withineach sequence. Insertions were removed to alignsequences. (GenBank accession numbers can befound in TABLE I.)

Phylogenetic analyses.—The heuristic search usingnLSU sequences produced 44 equally parsimonioustrees (one of which is shown in FIG. 1) (length 5 946,CI 5 0.517, RI 5 0.757). Five of the seven Laetiporusspecies included in the analysis (L. cincinnatus, L.conifericola, L. gilbertsonii, L. huroniensis and L.sulphureus) fell in the group referred to as the coreLaetiporus clade or Laetiporus s.s., while two Laeti-porus species (L. persicinus and L. portentosus) felloutside the core Laetiporus clade. The unidentifiedLaetiporus isolates from Hawai’i and the Caribbeanfell within the core Laetiporus clade. Wolfiporiadilatohypha isolates formed a sister group to the coreLaetiporus clade.

Although a relatively small number of species wereincluded in the nLSU analysis, the tree reflects theoverall structure of the polyporoid clade as defined byBinder et al (2005), including Antrodia, phlebioid andcore polyporoid clades. Laetiporus persicinus, Phaeolusschweinitzii, Pycnoporellus alboluteus, P. fulgens and theisolates of Wolfiporia cocos sequenced for this study allfell within the Antrodia clade. The one W. cocossequence taken from GenBank (AF393081) fell withinthe core polyporoid clade (FIG. 1). Phaeolus schweinit-zii formed a clade with the W. cocos isolates sequencedfor this study, although this clade has weak statisticalsupport (bootstrap $70 or posterior probability$0.95). Pycnoporellus alboluteus and P. fulgens formeda clade with strong statistical support (bootstrap $70and posterior probability $0.95). Leptoporus mollis,Ceriporia purpurea and Laetiporus portentosus formed astrongly supported clade representing the phlebioidclade of polypores.

422 MYCOLOGIA

The heuristic search with ITS data produced 98equally parsimonious trees (one of which is shown inFIG. 2) (length 5 191, CI 5 0.843, RI 5 0.941). Eightclades were found, seven of which have strongstatistical support. These eight clades correspond to

five described species (Laetiporus cincinnatus, L.conifericola, L. gilbertsonii, L. huroniensis and L.sulphureus s.s.) as well as three apparently unde-scribed species of Laetiporus. Relationships amongLaetiporus species were poorly supported, although

FIG. 1. Analysis based on nLSU sequences. The tree shown is one of 44 equally parsimonious trees (length 5 946, CI 5

0.517, RI 5 0.757). Data were taken from GenBank for isolates not included in TABLE I; these isolates are labeled withGenBank accession numbers. All other isolates were sequenced for this study and are labeled with collection numbers.Quotation marks around a species indicate a suspect identification.


there is strong support for a clade containing L.gilbertsonii and the apparently undescribed Caribbeanisolates. There is moderate support for a cladecontaining L. conifericola and L. huroniensis. Twosequences (AM269785 and AM269786) listed inGenBank as Laetiporus sulphureus clustered stronglywith other Laetiporus spp. (L. gilbertsonii and L.cincinnatus respectively). The GenBank sequenceAY835668, which is derived from a L. sulphureus

isolate collected from Quercus in northern Germany,clustered with North American isolates of L. sulphur-eus s.s.

Results from the heuristic search with mtSSUsequences produced few statistically supported clades.The core Laetiporus clade was weakly supported, whilerelationships within the clade were unresolved.Alignment of mtSSU sequences from species outsidethe core Laetiporus clade was difficult due to large

FIG. 2. Analysis based on ITS sequences. The tree shown is one of 98 equally parsimonious trees (length 5 191, CI 5 0.843,RI 5 0.941). Data were taken from GenBank for isolates not included in TABLE I; these isolates are labeled with GenBankaccession numbers. All other isolates were sequenced for this study and are labeled with collection numbers. Quotation marksaround a species indicate a suspect identification.

424 MYCOLOGIA

variable regions; alignment within the core Laetiporusclade showed little variation among isolates, with theexception of one variable region approximately 30base pairs long. Within this variable region a 16 bpindel (TTTAATTTTAAATTCA) was found to occur infour isolates: two L. conifericola isolates (JAM-1 andCA-8) and two L. sulphureus s.s. isolates with whitepores (MAS-2 and TJV99-150); this indel did notoccur in any of the other isolates sampled.

The heuristic search with the combined datasetproduced 31 equally parsimonious trees (one ofwhich is shown in FIG. 3) (length 5 244, CI 5

0.812, RI 5 0.892). Five clades were resolved withweak to strong statistical support. Laetiporus gilbertso-nii isolates and the Laetiporus isolates from theCaribbean formed a weakly supported clade (Gilbert-sonii). Laetiporus cincinnatus isolates formed astrongly supported clade (Cincinnatus) that did notgroup consistently with any other species (the branchcontaining Gilbertsonii and Cincinnatus clades col-lapses in a strict consensus tree). Laetiporus sulphur-eus s.s. isolates split into two strongly supportedclades, one with yellow pores (Sulphureus clade I)and one with white pores (Sulphureus clade II). Thetwo conifer-inhabiting species, L. conifericola and L.huroniensis, formed a weakly supported clade (Con-ifericola). The three branches indicating relation-ships among major clades collapse in a strictconsensus tree.

DISCUSSION

Sequence analysis confirms that Laetiporus sulphureuss.l. represents a large species complex in NorthAmerica, which supports work by Burdsall and Banik(2001) using mating compatibility, RFLP, morpholo-gy and ecology. Although five major clades encom-passing eight species groups were identified withinLaetiporus s.s., further work is needed to resolverelationships among species and to determine thenumber of Laetiporus species in North America.Laetiporus sulphureus s.l. is known from manycontinents, including Africa, Asia and Europe (Ry-varden and Johansen 1980, Ryvarden and Gilbertson1993, Nunez and Ryvarden 2001) so additionalspecies undoubtedly remain uncharacterized world-wide.

Of all species in the core Laetiporus clade,Laetiporus sulphureus s.s. displayed the largest num-ber of sequence variants and also showed variation inpore color, either white or yellow. Burdsall and Banik(2001) describe Laetiporus sulphureus s.s. as possess-ing lemon yellow pores but mention rare specimenswith white pores that are indistinguishable from L.sulphureus s.s in morphology, habitat and RFLP

patterns. The two isolates of this type included inthis study (MAS-2 and TJV99-150) cluster with L.sulphureus s.s. in the ITS analysis (FIG. 2) and areclearly differentiated from both L. cincinnatus and L.gilbertsonii var. pallidus, the two species that areknown to possess white pore layers. ITS data indicatethat these white-pored isolates fall within L. sulphur-eus s.s., yet both the nLSU (FIG. 1) and combineddatasets (FIG. 3) support separation of white poredforms of L. sulphureus s.s. from isolates with yellowpores. Additional isolates are needed to determinewhether this differentiation warrants the descriptionof a new species or variety. Unfortunately singlespores from white-pored L. sulphureus s.s. fruitingbodies have low germination rates across a wide pHrange of agar media (unpubl data), which hashampered mating studies. Although names have beenintroduced to describe white-pored forms of L.sulphureus s.l., such as Peck’s L. sulphureus var.semialbinus (Peck 1906), it is likely these namesdescribe the more common and morphologicallydistinct species L. cincinnatus (Burdsall and Banik2001).

The only other Laetiporus species found to havevariation in the color of the pore layer was L.gilbertsonii. Both color forms of L. gilbertsonii arealmost indistinguishable in these analyses, with onlytwo base pairs of variations in both the ITS and nLSUregions. This suggests that a change in pore colora-tion can occur with little additional genetic differen-tiation in this group. Laetiporus gilbertsonii also showsvariation in host preference, being found on bothQuercus spp. and Eucalyptus spp. (Burdsall and Banik2001). The only other isolate in this study collectedon Eucalyptus, EUC-1, was collected in Hawai’i andclustered with an addition Hawai’ian isolate from anative Acacia koa (koa tree). These Hawai’ian isolates,along with the Caribbean isolates in the Gilbertsoniiclade and the white-pored isolates of L. sulphureuss.s., represent three likely candidates for undescribedspecies in the core Laetiporus clade.

A fourth undescribed Laetiporus sp. was identifiedin the ITS analysis (FIG. 2, L. ‘‘sulphureus’’ AY218417,AY835667, DQ450876, EF088657), although all rep-resentatives in this group are GenBank sequences forwhich little geographic or host data exist. Onesequence of this type (AY835667) was obtained froma Laetiporus sulphureus s.l. isolate collected on Piceaabies in southern Germany (Davoli et al 2005), raisingthe possibility that these isolates are conspecific withthe European conifer-inhabiting Laetiporus sulphur-eus s.l. isolates investigated by Rogers et al (1999).Burdsall and Banik (2001) report that L. monticolaCerny. was described in 1989 to accommodateEuropean Laetiporus specimens found on conifers,


although apparently this species was never validlypublished due to a lack of a Latin diagnosis. Furthertaxonomic work is needed to determine whetherconifer-inhabiting Laetiporus isolates from Europe

constitute an unrecognized species. GenBank se-quence AY835668, which is derived from a L.sulphureus isolate collected on Quercus in northernGermany (Davoli et al 2005), clustered with North

FIG. 3. Combined analysis based on nLSU, mtSSU and ITS sequences. The tree shown is one of 31 equally parsimonioustrees (length 5 244, CI 5 0.812, RI 5 0.892). Isolates were included in this analysis only if sequence data were available for allthree regions.

426 MYCOLOGIA

American L. sulphureus s.s. isolates (FIG. 2), support-ing the hypothesis that L. sulphureus s.s. is restrictedto angiosperms and is found in both North Americaand Europe.

GenBank ITS data also produced two sequences ofLaetiporus ‘‘sulphureus’’ (AM269785 and AM269786)that clustered strongly with known Laetiporus isolatesidentified independently with mating compatibilityand fruiting body morphology (Banik and Burdsall2000, Burdsall and Banik 2001). Laetiporus ‘‘sulphureus’’sequence AM269785 is from a Californian isolateassociated with an unknown host (Guglielmo et al2007) and clustered with L. gilbertsonii sequences; L.‘‘sulphureus’’ sequence AM269786 is from an isolatecollected on Quercus in Wisconsin (Guglielmo et al2007) and clustered with L. cincinnatus sequences(FIG. 2). The phylogenetic placement of AM269785and AM269786, taken together with the host andgeographic data associated with their parent isolates,indicate the species names associated with theseGenBank sequences are in need of revision.

The discovery of Wolfiporia dilatohypha as a sistergroup to the core Laetiporus clade was fortuitous andmight help to define the generic limits of Laetiporusas sequence data accumulate from other regions ofthe world. Two species of Laetiporus, L. portentosusand L. persicinus, fell well outside the core Laetiporusclade, indicating that Laetiporus as it is currentlydefined represents a polyphyletic assemblage. Theother North American polypore species sequencedfor this study, including Leptoporus mollis, Phaeolusschweinitzii, Pycnoporellus alboluteus and P. fulgens,also do not appear to be closely allied with the coreLaetiporus clade; however, with the exception of L.mollis, all these species fall in the Antrodia clade ofpolypores. The Wolfiporia cocos isolates sequenced forthis study also fell within the Antrodia clade, althoughthis is in contrast to earlier studies that placed W.cocos in the core polyporoid clade (this point isaddressed in detail later in the discussion). Furthertaxon sampling and additional sequences analyses areneeded to confirm the placement of these brown rotpolypore species in broader basidiomycete phyloge-nies.

A lack of monophly in Laetiporus first was demon-strated by Hibbett and Donoghue (2001) withsequence analysis of two rDNA regions from thesouthern hemisphere species L. portentosus. Using asingle isolate of L. sulphureus s.l., Hibbett andDonoghue (2001) found that L. portentosus is notclosely related to L. sulphureus s.l. Although both L.portentosus and L. persicinus show macroscopic,microscopic and physiological similarities to otherLaetiporus species, sequence data suggest these areconvergent characters and that neither species is

closely related to any of the species in the coreLaetiporus clade. Although L. portentosus was placedin Piptoporus, Rajchenberg (1995) transferred it toLaetiporus based on morphological and culturalcharacteristics. Unfortunately these characters appearto be phylogenetically uninformative at the genericlevel for these species groups. Hibbett and Dono-ghues’ (2001) analysis indicated that L. portentosusfalls within the ‘‘Antrodia clade’’ of true polypores,although the nLSU analysis in this study placed it inthe phlebioid clade. More gene regions from addi-tional L. portentosus isolates need to be sequenced toresolve this issue.

Like Laetiporus portentosus, L. persicinus fallsoutside the core Laetiporus clade in the nLSUanalysis. This result is consistent with L. persicinus’divergent morphological traits, including a lack ofbright orange coloration in the pileus surface andflesh and tubes that stain blackish-brown. BLASTanalyses of the GeneBank (NCBI) databases usingITS, nLSU and mtSSU regions from L. persicinusrevealed no significant similarity to any known genus,suggesting that placement in a new genus may bewarranted. However further work is needed toconfirm that an appropriate genus does not alreadyexist. Since being described as Polyporus persicinus byBerkeley and Curtis in 1872, this species has beenplaced in numerous genera, including Meripilus(Ryvarden 1972) and Buglossoporus (Corner 1984),before being transferred to Laetiporus by Gilbertson(1981).

Wolfiporia dilatohypha, which formed a sister groupto the core Laetiporus clade, was the only speciesoutside genus Laetiporus with an ITS sequence thataligned easily with core Laetiporus species. Wolfiporiadilatohypha isolates were so similar to species in thecore Laetiporus clade that future work must addresswhether W. dilatohypha is congeneric with species inthe core Laetiporus clade. The Wolfiporia cocos isolatessequenced for this study were found to be distantlyrelated to both W. dilatohypha and all Laetiporusspecies examined. This indicates that Wolfiporia, likeLaetiporus, is polyphyletic. Although five species arepresently found in genus Wolfiporia, only two wereincluded in this study. Further taxon sampling isneeded to determine whether any of the threeremaining Wolfiporia species, W. cartilaginea Ryvar-den, W. curvispora Y.C. Dai, and W. sulphurea (Burt)Ginns, are closely allied with other Wolfiporia speciesor genus Laetiporus.

The close relationship between Laetiporus andWolfiporia dilatohypha was surprising given that W.dilatohypha fruiting bodies are macroscopically unlikeLaetiporus fruiting bodies. Wolfiporia dilatohyphaefruiting bodies are effused, thin (3–4 mm) and white


to buff colored with relatively large pores (1–5 permm), which is in contrast to the large, brightlycolored, pileate fruiting bodies of Laetiporus. Threecultural isolates of W. dilatohypha were sequenced forthis study and all had identical LSU sequences andnearly identical ITS regions (two base pairs ofvariation). However no sequences have yet beenobtained from fruiting body material and the mostrecent cultural isolate was collected in 1963. Freshfruiting bodies of W. dilatohypha are needed toconfirm that the sequences in this study are consistentwith sequences derived from fruiting structures.Unfortunately fruiting bodies of this species seem tobe rare or are rarely collected.

Although Wolfiporia dilatohypha was found to beclosely related to the core Laetiporus clade based onITS and nLSU data, it is interesting to note thatisolates of W. cocos, the type species of Wolfiporia, donot appear to be closely related to W. dilatohypha.The two W. cocos isolates sequenced for this study(MD-106 and MD-275) both had identical ITS, nLSUand mtSSU regions. An additional 13 isolates fromthe Center for Forest Mycology Research (CFMR)cultural collection (Madison, Wisconsin) were se-quenced in the nLSU region and all displayedsequences that matched the two original isolatesalmost identically (unpubl data). However nLSU andmtSSU data for W. cocos in this study differ stronglyfrom those reported by Hibbett and Donoghue(2001) and Binder and Hibbett (2002). Hibbett andDonoghue (2001) sequenced the nuclear and mito-chondrial SSU of W. cocos (isolate No. FPL4198),while Binder and Hibbett (2002) sequenced thenuclear and mitochondrial LSU of the same isolate(the GenBank nLSU sequence of this isolate isincluded in FIG. 1 as ‘‘Wolfiporia cocos’’ AF393081).Sequences from isolate FPL4198 subsequently havebeen included in many analyses by various authors(Hibbett and Binder 2002, Hibbett 2004, Binder et al2005). Isolate FPL4198 has shown strong affinities forthe core polypore clade, clustering with white rotgenera such as Polyporus and Trametes (see FIG. 1).

Kim and Jung (2000) however sequenced thenuclear SSU of W. cocos (isolate ATCC13490) andfound that it clustered with other brown rot generasuch as Laetiporus, Phaeolus and Sparassis. As pro-posed by Binder et al (2005), this disparity might bedue to the mislabeling or contamination of isolateFPL4198, which does not appear to represent W.cocos. Unfortunately the original isolate of FPL4198housed at CFMR died in culture and is not availablefor further study. Additional cultural isolates, fruitingbodies and sclerotia of W. cocos are needed to clarifythe position of this species and to determine theextent of variability within this species. Unfortunately

such an analysis might be hampered by the size of theITS region in W. cocos, which was found to be greaterthan 1800 base pairs due to insertion events. Inaddition to making the ITS difficult to amplify, theseinsertions make it difficult to align the ITS of W. cocoswith other species.

Although previous investigations have suggestedthat Phaeolus schweinitzii is closely related to Laeti-porus sulphureus s.l., the ITS of P. schweinitzii isdifficult to align with Laetiporus spp. and the nLSU ofP. schweinitzii shows only 89% sequence identity to L.sulphureus s.s. The closest relative of P. schweinitzii inthe nLSU analysis was Wolfiporia cocos, although thisfinding has weak statistical support. Further samplingis needed to determine whether close relatives of P.schweinitzii exist and whether there is variabilitywithin this widespread species. The proposed linkbetween Phaeolus and Pycnoporellus (Gilbertson andRyvarden 1987) was not borne out by this studybecause Phaeolus schweinitzii and Pycnoporellus iso-lates shared only 85% sequence identity for the nLSUand alignments of ITS regions were impossiblebetween these genera. The two Pycnoporellus specieshowever showed similar nLSU sequences (97%

sequence identity) and the ITS regions were alignedeasily. This evidence, together with the similarmicrostructures and chemical staining reactions inKOH, suggests that Pycnoporellus consists of a naturalgrouping of species.

Unlike the other genera investigated, genus Lepto-porus seems to fall outside the Antrodia clade asdefined by Binder et al (2005). The nLSU sequencesof Leptoporus generated for this study match almostidentically (99% identity over 643 base pairs) those ofBinder et al (2005), who found that Leptoporus fellnear Ceriporia in the phlebiod clade. Genus Lepto-porus and Ceriporia share similar microscopic charac-ters and Gilbertson and Ryvarden (1987) state thatthe production of brown rot is the only differentiatingcharacter for Leptoporus. If this placement is correct itwould indicate that Leptoporus arose within a clade ofwhite rot species. Isolates of Leptoporus were found todisplay two ITS types that differed by 18 base pairs,indicating that further work is needed to determinewhether L. mollis represents a species complex.

In this study two isolates (PR-2 and PR-6326) wereencountered that possessed extremely divergentnLSU sequences. These isolates came from collec-tions that had been classified as Laetiporus persicinus,although preliminary sequence analysis indicatedthese isolates varied strongly from L. persicinus andall other species included in this study. Although ITSdata could not be obtained, the nLSU region fromthese isolates was amplified successfully from culturesand fruiting bodies and had no significant BLAST

428 MYCOLOGIA

matches in GenBank. Preliminary parsimony analysisplaces this species in the core polyporoid clade,although an insertion of approximately 150 base pairsmakes alignment of these sequences difficult (Karl-Henrik Larsson pers comm). This species has darkbrown fruiting bodies that grow as basal rosettes fromtree roots in a fashion similar to L. persicinus andPhaeolus schweinitzii; however, unlike L. persicinus orP. schweinitzii, the fruiting bodies can attain tremen-dous dimensions (0.5–0.8 m diam) and have a thick,black pileus cuticle. Preliminary evidence based onexcavating tree roots close to fruiting bodies suggeststhis species produces a white rot rather than thebrown rot associated with L. persicinus (D. Jean Lodgepers comm). The nLSU data confirms that thisspecies is highly divergent from all other polyporespecies for which sequence data are available despiteits morphological and microscopic similarities to L.persicinus. Gross morphology suggests this taxonmight represent Polyporus talpae Cooke (5 Meripilustalpae [Cooke] D.A. Reid) or possibly Meripilustropicalis Guzman & Perez-Silva (Guzman and Silva1975). This species also might have been described byMurrill from Jamaica in 1910 (Murrill 1910) asAmauroderma brittonii Murrill. Both Polyporus talpaeand Amauroderma brittonii long have been consideredsynonyms of Laetiporus persicinus but further work isneeded to determine the legitimacy of this synonymy.

It is clear that many species within the Antrodiaclade of polypores remain poorly characterized.Additional isolates of Laetiporus, Leptoporus, Phaeolus,Pycnoporellus and Wolfiporia are needed from widergeographic regions to clarify the positions of thesebrown rot polypores in larger basidiomycete phylog-enies. Laetiporus sulphureus s.l. in particular requiresfurther attention on a worldwide basis because thisspecies complex is one of the larger and moreeconomically important groups in the Antrodia clade.The sequences provided by this study hopefully willprovide a framework that aids further phylogeneticand taxonomic investigations.

ACKNOWLEDGMENTS

We thank Don Hemmes of the University of Hawai’i at Hilofor providing Laetiporus isolates from Hawai’i; D. JeanLodge of the USDA-Forest Service, Center for ForestMycology Research, for providing isolates and fruitingbodies of Laetiporus and ‘‘Polyporus talpae’’ from PuertoRico; and Thomas J. Volk of the University of Wisconsin atLa Crosse for collecting many of the Laetiporus specimensused in this study. We also thank David Hibbett and ZhengWang of Clark University for providing sequences ofSparassis as well as access to large sequence databases ofhomobasidiomycetes. We are grateful for the help of DavidHibbett and Karen Nakasone; they provided assistance with

phylogenetic analyses and helpful advice on previousversions of this manuscript.

LITERATURE CITED

Arora D. 1986. Mushrooms demystified. Berkeley, Califor-nia: Ten Speed Press. 959 p.

Banik MT, Burdsall HH Jr. 1999. Incompatibility betweenLaetiporus cincinnatus and L. sulphureus in culture.Mycotaxon 70:461–469.

———, ———. 2000. Incompatibility groups among NorthAmerican populations of Laetiporus sulphureus sensulato. Mycologia 92:649–655.

———, ———, Volk TJ. 1998. Identification of groupswithin Laetiporus sulphureus in the United States basedon RFLP analysis of the nuclear ribosomal DNA. FoliaCryptog Estonica 33:9–14.

Binder M, Hibbett DS. 2002. Higher-level phylogeneticrelationships of Homobasidiomycetes (mushroom-forming fungi) inferred from four rDNA regions. MolPhylogenet Evol 22:76–90.

———, ———, Larsson KH, Larsson E, Langer E, LangerG. 2005. The phylogenetic distribution of resupinateforms across the major clades of mushroom-formingfungi (Homobasidiomycetes). Syst Biodivers 3:113–157.

Burdsall HH Jr, Banik MT. 2001. The genus Laetiporus inNorth America. Harvard Pap Bot 6:43–55.

Corner EJH. 1984. Ad Polyporaceas III. Piptoporus, Buglos-soporus, Laetiporus, Meripilus, and Bondarzewia. Bei-hefte zur Nov Hedwig 78:137–222.

Davoli P, Mucci A, Schenetti L, Weber R. 2005. Laetiporicacids, a family of non-carotenoid polyene pigmentsfrom fruit-bodies and liquid cultures of Laetiporussulphureus (Polyporales, Fungi). Phytochemistry 66:817–823.

Felsenstein J. 1985. Confidence limits on phylogenies: anapproach using the bootstrap. Evolution 39:783–791.

Gilbertson RL. 1981. North American wood-rotting fungithat cause brown rots. Mycotaxon 12:372–416.

———, Ryvarden L. 1986. North American Polypores.Vol 1. Oslo, Norway: Fungiflora. 433 p.

———, ———. 1987. North American Polypores. Vol 2.Oslo, Norway: Fungiflora. p 434–885.

Ginns J. 1984. New names, new combinations and newsynonymy in the Corticiaceae, Hymenochaetaceae andPolyporaceae. Mycotaxon 21:325–333.

———, Lowe JL. 1983. Macrohyporia extensa and itssynonym Poria cocos. Can J Bot 61:1672–1679.

Guglielmo F, Bergemann SE, Gonthier P, Nicolotti G,Garbelotto M. 2007. A multiplex PCR-based method forthe detection and early identification of wood rottingfungi in standing trees. J Applied Microbiol 103:1490–1507.

Guzman G, Silva EP. 1975. Una nueva especie de Meripilus,hongo gigante destructor de la madera en la selvatropical de Mexico. Bol Soc Mex Mic 9:53–59.

Hibbett DS. 2004. Trends in morphological evolution inHomobasidiomycetes inferred using maximum likeli-hood: a comparison of binary and multistate approach-es. Syst Biol 53:889–903.


———, Binder M. 2002. Evolution of complex fruiting-bodymorphologies in homobasidiomycetes. Proc R SocLond B 269:1963–1969.

———, Donoghue M. 1995. Progress toward a phylogeneticclassification of the Polyporaceae through parsimonyanalysis of mitochondrial ribosomal DNA sequences.Can J Bot 73:S853–S861.

———, ———. 2001. Analysis of character correlationsamong wood decay mechanisms, mating systems, andsubstrate ranges in Homobasidiomycetes. Syst Biol 50:215–242.

Huelsenbeck P, Ronquist F. 2001. MrBayes: Bayesianinference of phylogeny. Bioinformatics 17:754–755.

Kim SY, Jung HS. 2000. Phylogenetic relationships of theAphyllophorales inferred from sequence analysis ofnuclear small subunit ribosomal DNA. J Microbiol 38:122–131.

Murrill WA. 1910. The Polyporaceae of Jamaica. Mycologia2:183–197.

Nunez M, Ryvarden L. 2001. East Asian Polypores. Vol. 2.Polyporaceae s.l. Synopsis Fungorum 14. Oslo, Norway:Fungiflora. p 170–522.

Peck CH. 1906. Report of the state botanist. 1905. New YorkState Museum. Bulletin 105:5–106.

Rajchenberg M. 1995. A taxonomic study of the subantarcticPiptoporus (Polyporaceae, Basidiomycetes) II. Nord JBot 15:105–119.

Redhead SA, Ginns J. 2006. (1738) Proposal to conserve thename Poria cocos against Daedalea extensa (Basidiomy-cota). Taxon 55:1027–1028.

Rogers SO, Holdenrieder O, Sieber TN. 1999. Intraspecificcomparisons of Laetiporus sulphureus isolates frombroadleaf and coniferous trees in Europe. Mycol Res103:1245–1251.

Ronquist F, Huelsenbeck P. 2003. MrBayes 3: Bayesianphylogenetic inference under mixed models. Bioinfor-matics 19:1572–1574.

Ryvarden L. 1972. A critical checklist of the Polyporaceae intropical east Africa. Nor J Bot 19:229–238.

———. 1991. Genera of Polypores: nomenclature and

taxonomy. Synopsis Fungorum 5. Oslo, Norway: Fungi-flora. 363 p.

———, Gilbertson RL. 1993. European Polypores. Vol. 1.Oslo, Norway: Fungiflora. 387 p.

———, Johansen I. 1980. A preliminary Polypore flora ofeast Africa. Oslo, Norway: Fungiflora. 636 p.

Swofford DL. 2002. PAUP*: phylogenetic analysis usingparsimony (*and other methods). Version 4. Sunder-land, Massachusetts: Sinauer Associates.

Turkoglu A, Duru ME, Mercan N, Kivrak I, Gezer K. 2007.Antioxidant and antimicrobial activities of Laetiporussulphureus (Bull.) Murrill. Food Chem 101:267–273.

Volk TJ, Burdsall HH Jr, Banik MT. 1996. Armillarianabsnona, a new species from western North America.Mycologia 88:484–491.

Wagner T, Fischer M. 2001. Natural groups and a revisedsystem for the European poroid Hymenochaetales(Basidiomycota) supported by nLSU rDNA sequencedata. Mycol Res 105:773–782.

Wang Y, Zhang M, Ruan D, Shashkov AS, Kilcoyne M,Savage AV, Zhang L. 2004. Chemical components andmolecular mass of six polysaccharides isolated from thesclerotium of Poria cocos. Carb Res 339:327–334.

Weber GF. 1929. The occurrence of tuckahoes and Poriacocos in Florida. Mycologia 21:113–130.

White TJ, Bruns T, Lee S, Taylor J. 1990. Amplification anddirect sequencing of fungal ribosomal RNA genes forphylogenetics. In: Innis M, Gelfand DH, Sninsky JJ,White TJ, eds. PCR protocols: a guide to methods andapplications. San Diego, California: Academic Press.p 315–322.

Wu S, Ng L, Lin C. 2004. Antioxidant activities of somecommon ingredients of traditional Chinese medicine,Angelica sinensis, Lycium barbarum and Poria cocos.Phytotherapy Res 18:1008–1012.

Zhang L, Ding Q, Zhang P, Zhu R, Zhou Y. 1997. Molecularweight and aggregation behaviour in solution of b-D-glucan from Poria cocos sclerotium. Carbo Res 303:193–197.

430 MYCOLOGIA

molecular phylogeny of laetiporus and other brown rot

Documents