585

Mycoscience 41: 585-593, 2000

Molecular phylogeny of in the genera and

Masanobu Tabata 1), Thomas C. Harrington 21, Wei Chen 21 and Yasuhisa Abe 3)

1) Shikoku Research Center, Forestry and Forest Products Research Institute, 2-91 5 Asakura-nishi, Kochi 780-8077, Japan 2) Department of Plant Pathology, Iowa State University, Ames, Iowa 50011, U.S.A. 3) Forestry and Forest Products Research Institute, P. O. Box 16, Tsukuba, Ibaraki 305-8687, Japan

Accepted for publication 18 September 2000

Analyses of DNA sequences from the internal transcribed spacer (ITS) region of the nuclear rDNA and from a portion of a manganese-dependent peroxidase gene were used to assess the species in Amylostereum, including isolates from the mycangia of , decay, and basidiomes. Four species are recognized: A. areolatum, A. chailletii, A. laevigatum, and A. ferreum. An unidentified Amylostereum isolate from the of Xoanon matsumurae had an ITS se- quence identical to that of A. areolatum. Another unidentified Amylostereum isolate from the mycangium of areolatus was near A. laevigatum, which appears to be the mycangial symbiont for those horntails attacking cedar-like trees. The other isolates, primarily from , proved to be either A. areolatum or A. chailletii. The DNA sequences of , E. tsugicola and E. japonicum were similar to those of the Amylostereum spe- cies, and Amylostereum species are now recognized as members of the family rather than the family . Echinodontium taxodii was found to be distinct from the Echinodontiaceae and , and E. taxodii is recognized as a species.

Key Words Amy/ostereum; Echinodontium; Echinodontiaceae; internal transcribed spacer; manganese-dependent peroxidase.

Excluding Dextrinocystidium sacratum (G. H. Cunnin- (1998) placed Amylostereum in the monotypic family gham) S. H. Wu (1995), there are four recognized spe- Amylostereaceae. Analyses of sequences of mitochon- cies of Amylostereum (Stereaceae): A. areolatum (Fr.: drial small subunit rDNA (Hsiau, 1996) suggested an Fr.) Boidin, A. chailletii (Pers.: Fr.) Boidin, A. ferreum affinity between A. chailletii and a group of Aphyl- (Berk. & Curt.) Boidin & Lanquetin, and A. laevigatum Iophorales informally recognized as "group 2" by Hibbett (Fr.: Fr.) Boidin (Boidin and Lanquetin, 1984). All of the and Donoghue (1995). In comparing the sequences of Amylostereum species occur on coniferous trees. the internal transcribed spacer (ITS) region of the nuclear , A. chailletii, and A. laevigatum rDNA, the sequence of A. chailletii was found to be very are associated with decay of Pinaceae and other similar to the sequence of Echinodontium tinctorium (Ell. in the Northern Hemisphere (Breitenbach and & Ev.) Ell. & Ev. (Harrington, unpublished). Kranzlin, 1988; Chamuris, 1988; Eriksson and Ryvarden, Gross (1964) monographed the monotypic family 1973; Eriksson et al., 1978; Ginns and Lefebvre, 1 993), Echinodontiaceae and recognized six Echinodontium spe- and A. ferreum decays wood of spp. in Latin cies with smooth to spinose hymenophores. Echinodon- America (Boidin and Lanquetin, 1984). Some tium tinctorium and E. tsugicola (P. Henn. & Shirai) Imaz. Amylostereum species are also known as symbionts of cause white heartrot of living Pinaceae in USA and mycophagus horntails [Sirex and species, Japan, respectively (Gilbertson and Ryvarden, 1986). (Hymenoptera: Siricinae)], which carry their fungal sym- Echinodontium ballouii (Banker) Gross was described biont in mycangia and inoculate the wood of the plant from (L.) B. S. P. in the eastern host with hyphal fragments or arthrospores as they USA (Gross, 1964). Echinodontium japonicum Imaz. is oviposit (Gaut, 1969, 1 970; Sano et al., 1995; Tabata known to occur on Quercus spp. in Japan (Imazeki, and Abe, 1997; Tabata and Abe, 1999; Terashita, 1935). Two Echinodontium species, E. taxodii (Lentz & 1970). Mckay) Gross and E. sulcatum (Burt) Gross, are often The Amylostereum has been traditionally placed in the genus Laurilia, which is in the Stereaceae placed in the Stereaceae (Donk, 1964), and species in (Parmasto, 1968; Pouzar, 1959) or in the Echinodontia- Amylostereum superficially resemble Stereum. How- ceae (J•lich, 1981). ever, molecular systematics has called for re-evaluation We used phylogenetic analyses to re-evaluate the of many of the families of Aphyllophorales. Boidin et al. genera Amylostereum and Echinodontium. We se- 586 M. Tabata et al.

1- F, CO 7~ 1.1_ < ~'~, ~" ~ LO 0"~ O0 r~ ~1 O~ ~4 0 0 0 0 0 0 T- v-

g~ oo oo oo oo ~o oO ~0 o0o0 OO en . e- ffl N N N ~ N N N NN 7~ oO < < < < < < < << t- O e- ~o 0 (3 o e- ~ 0 ~ ~ ~ ~ ~ ~0 n ~ ~ ~ ~ ~ ~0 ~ 000 e-I-- en . 0 0 0 < <<< < < < ~ <<~ ~ ~ E

<~ ~ < m~ m m m m ~<<~ oo ._ ~.~ ~ ~< ._

.s

o 0 ..J

E t~ EEE -~ E ~ .~.~.~ E E E ~E e~.~= ~===~,&~-N-N-N ...... c'~ = = o o m ~ ~ E ~

t~ ~~ ~,~,~,~

~'Z~ ~ ~

O~

:~:~~ ~~ o Q~ ~'~ ~ ~ ~ ~ ~ ~ o ~ ~ ~ ~ ~ ~

r- .~.~ ~ .~ ~ ..... ~,~,~ ~ ~ .-~ ~, ~~.0.,~~~~-~ ....~~ ~,,~ ~E ~=~ .~ .s

~~ ~.~ .~,~~.~.~ .~ .~~~~ ~ ,_ 0 0 "1-

Q~ ~ ~ ~ ~0~ o ~0 e- 0 0~ 0~ ~ ~ o~8oo~8~8~ ~o~ ~gooo~~ooo o~~o ~- 6~ ~ ~ Q~ ~8 OZ ooo ~~mmaaaaaaa ~ 0 ~0~0~~~~ ~ Q~

~~~~0~~ 0 0 ~z

J~ E r- Q~ o 0 ~0 Molecular phylogeny of Amylosterem and Echinodontium 587 quenced the ITS regions and a portion of a peroxidase as outlined for the ITS sequences, except that 1000 gene from isolates obtained from the Institute for Fer- bootstrap replications were made. mentation, the Centraalbureau voor Schimmelcultures, the Canadian Collection of Fungal Cultures, and from iso- Results lates we obtained from horntails and woody substrata in Japan. We also examined the morphology of ITS sequences The ITS sequences of each of the basidiomes of A. laevigatum collected from Japan and Amylostereum species aligned well with each other. In Sweden. order to find potential outgroup taxa and identify genera that may be closely related to Amylostereum, we com- Materials and Methods pared these ITS sequences with those generated in our studies (Harrington et al., 1998) of Heterobasidion. Het- Isolates The source and substrate of isolates are given erobasidion and other genera were placed in "group 2" in Table 1. Extracted DNA of Stereum annosum Berk. & based on mitochondrial rDNA sequences (Hibbett and Br., S. hirsutum (Willd.: Fr.) S. F. Gray, and Bondarzewia Donoghue, 1995; Hsiau, 1996). Extracted DNA from montana (Fr.) Sing. was kindly provided by David Hib- representatives of group 2 genera was provided by Dr. D. bert. Hibbett, and we generated ITS sequences for representa- ITS sequences Mycelia were grown at room tempera- tives of Bondarzewia, Lentinellus, Auriscalpium, Herici- ture (20-25~ for 10-12 days in 15-20 ml of MY liquid urn, Echinodontium, and Russula. Of these genera, the medium (2~ malt extract, l~ yeast extract). A fresh ITS sequence of E. tinctorium matched closely to ITS se- mycelial mat was collected by vacuum filtration, ground quences of Amylostereum. We subsequently generated to a fine powder in liquid nitrogen with a mortar and pes- ITS sequences for isolates of E. tinctorium, E. tsugicola, tle, and DNA was extracted following the protocol of and E. japonicum, and these sequences were easily DeScenzo and Harrington (1994). Template DNA of aligned with those of Amylostereum species. However, most of the isolates was extracted from liquid cultures, the ITS sequence of E. taxodii did not match closely to but mycelia of three Amylostereum isolate (B29, B1358, those of the other Echinodontium species. Because B1360) were scraped lightly with a pipette tip for tem- Amylostereum has been placed in the Stereaceae, we plate DNA (Harrington and Wingfield, 1995). also included ITS sequences of two Stereum species in A fragment of nuclear rDNA about 700 bp long from the analyses, and the ITS sequence of Perenniporia the 3' end of the 18S (small subunit gene) to the 5' end of subacida (Pk.) Donk was also compared as a potential the 28S (large subunit gene) was amplified and se- outgroup taxon. quenced using the primers ITS1-F (Gardes and Bruns, In the complete data set of ITS sequences of 1993) and ITS4 (White et al., 1990). The polymerase Amylostereum, Echinodontium, Stereum, and P. subaci- chain reaction (PCR) products were purified using QIA- da, there were 614 aligned characters with the insertions quick PCR Purification Kit (QIAGEN Inc., USA) and se- of gaps. However, 261 of these characters were ex- quenced with the ABI PRISM 377 DNA sequencer (Per- cluded from the initial analysis because of ambiguous kin-Elmer) in the DNA Sequencing Facility at Iowa State alignment. Of the included characters, 259 were con- University. Both the coding and template strands were stant, 39 were parsimony uninformative, and 55 charac- sequenced, with complete overlap of the complementary ters were informative. Ten most parsimonious trees of sequences. Sequences were visually aligned and ana- 150 steps were found, with a consistency index (CI) of lyzed using heuristic searches in PAUP 4.0, with step- 0.8200, a retention index (RI) of 0.8767, and a rescaled wise additions (simple) and tree-bisection-reconnection consistency index (RC) of 0.7189 (Fig. 1). Perenniporia (Swofford, 1998). Uninformative characters were ig- subacida, the only polypored species studied, was select- nored. Gaps were coded as a newstate (fifth character) ed as an outgroup taxon, rooting the tree at an internal because the gaps were consistently found within species node with basal polytomy. and, except for outgroup taxa, most gaps were only one The ITS sequence of E. taxodii was quite distinct or two bases in length. Bootstrapping (100 bootstrap from those of the other ingroup taxa. Strong bootstrap replicates) was used to determine confidence in the support was found for the inferred clade that included branches. two Stereum species, the Amylostereum species, and Manganese-dependent peroxidase In an earlier study the Echinodontium species, exclusive of E. taxodii (Fig. (Maijala et al., 1998), a portion of a manganese 1 ). The inferred clade containing the Amylostereum spe- (Mn)--dependent peroxidase gene was identified in an cies, E. tinctorium, E. tsugicola, and E. japonicum was isolate of E, tinctorium, and a homologous gene was well supported, as was the clade of E. tinctorium and E. found in each of the Echinodontium and Amylostereum tsugicola. The two isolates of E, japonicum formed a species studied except E. taxodii. A PCR fragment of well-supported clade, as did the nine tested isolates of A. about 570 bp was amplified using degenerative primers areolatum. The inferred Amylostereum clade was only and the reaction conditions outlined by Maijala et al. weakly supported, but there were few phylogenetically (1998). Amplified fragments were cloned into the informative characters among the Amylostereum species pGEM-easy vector (Promega, Madison, Wisconsin, USA) after deletion of the ambiguously aligned characters. and sequenced at the DNA Sequencing Facility with the Each of the branches with bootstrap support (Fig. 1) was vector primers T7 and SP6. Phylogenetic analyses were found in each of the 10 most parsimonious trees. Fur- 588 M. Tabata et al.

B1350 Japan B1351 Japan B1395 Japan 99 B1373 Japan B1385 Germany Amylostereum B1352 Germany B1353 Denmark areolatum B1386 Australia B1394 New Zealand B1359 Brazil B1360 Brazil Amylostereum ferreum i B1358 UK B1354 UK I B1356 Sweden B1387 Germany B1355 Canada Amylostereum B1388 Canada B1389 Canada chailletii 1 change B1392 Canada B1390 Canada B1391 Canada B29 USA B105 USA B1393 USA Amylostereum sp. B1366 Japan B1363 Japan B1361 Japan B1362 Japan 64 Japanese B1364 Japan Amylostereum B1368 Japan B1369 Japan laevigatum B1370 Japan B1365 Japan B1367 Japan B1397 France I French B1371 France I Amylostereum 81372 France laevigatum 100 100 Bl122B13 USA USA .I E. ttnctorum" i B1377 Japan 93 58~B1378 Japan I E. tsuglco/a 100 B1375 Japan i~--,,,~,,-l~c"i~o~n§ B1374 Japan japonicum 78 Stereum hirsutum Stereum annosum B~376Echinodontium taxodii B37 Perenniporia subacida Fig. 1. One of 10 most parsimonious trees based on 355 alignable characters of the ITS-l, 5.8S, and ITS-2 regions of the nuclear rDNA of Amylostereum, Echinodontiurn, and Stereum species. The tree is rooted to Perenniporia subacida. Bootstrap values (100 replicates) greater than 50~ are indicated above the branches. Isolate numbers in bold are isolates from horntails.

ther, neighbor-joining analysis of the same data resulted found, with C1=0.8279, R1=0.9161, and RC=0.7585. in a tree with the same topology as that in the most par- The inferred clade containing the Amylostereum species simonious tree (Fig. 1), except for minor changes in the was well supported, as was the clade containing E. tin- relationships among A. chailletii, A. laevigatum and A. ctorium and E. tsugicola (Fig. 2). With rooting the tree ferreum. at an internal node with basal polytomy, E. japonicum did A second ITS data set of fewer taxa was used to look not clearly group with the other Echinodont/um species more closely at the relationships among the Echinodonti- or with Arnylostereum. urn and Amylostereurn species. The sequences of S. hir- The ITS sequence of the A. ferreum isolates was sutum, E. taxodii, and P. subacida were eliminated. Of near that of isolates of A. chailletii and A. laevigatum. the aligned 564 characters, including gaps, 364 charac- Three of the isolates of A. chailletii from Europe had ters were constant, and 77 variable characters were par- slightly different ITS sequences from the North American simony-uninformative, leaving 123 parsimony-informa- and a single German isolate of A. chailletii, but there was tive characters. Using S. annosurn as an outgroup tax- no bootstrap support (< 500/oo) for the branch connecting on, 152 most parsimonious trees of 308 steps were these two A. chailletii groups. Similarly, A. laevigatum Molecular phylogeny of Amylosterem and Echinodontium 589

B1350 Japan B1351 Japan B1395 Japan 99 B1373 Japan B1385 Germany Amylostereum B1352 Germany areolatum B1353 Denmark B1386 Australia B1394 New Zealand 5 changes ~_~ B1359 Brazil IAmylostereum B1360 Brazil ferreum B1358 UK B1354 UK B1356 Sweden

100 B1387 Germany B1355 Canada Amylostereum 5; B1388 Canada B1389 Canada chailletii B1392 Canada B1390 Canada B1391 Canada B29 USA B105 USA -- B1393USA Amy/ostereum sp. B1397 France French 35 B1371 France 50 Amylostereum B1372 France laevigatum B1366Japan B1363 Japan B1361 Japan 5~ B1362 Japan B1364 Japan Japanese B1368 Japan Amylostereum 89 B1369Japan laevigatum B1370Japan _ B1385 Japan B1367 Japan

B1122 USA I Echinodontium tinctorium 100 / ] B13 USA 11001- B1377 Japan I Echinodontium tsueicola L B1378 Japan I 100 i B1375 Japan B1374 Japan iEchinodontium japonicum Stereum annosum Fig. 2. One of 152 most parsimonious trees based on 564 alignable characters of the ITS-l, 5.8S, and ITS-2 regions of the nuclear rDNA of Amylostereum and Echinodontium species. The tree is rooted to Stereum annosum. Bootstrap values (100 replicates) greater than 50~ are indicated above the branches. Isolate numbers in bold are isolates from horntails. isolates from Japan and France were found in two well- tiffed Amylostereum isolate (B1373) from Xo. rnatsu- supported branches. An unidentified Amylostereum iso- murae Rohwer was identical to that of A. areolaturn (Fig. late (B1393) from S. areelatus Cr, in California, USA had 2). a distinct ITS sequence. The ITS sequence of an uniden- The strict consensus tree included branches for A. 590 M. Tabata et al. areolatum, A. ferreum, A. laevigatum from Europe, A. A. chailletii, and these branches also had no bootstrap laevigatum from Japan, A. chailletii, E. tinctorium, E. support in parsimony analysis (Fig. 2). tsugicola, and E. japonicum. The branch grouping E. tin- Mn-dependent peroxidase Portions of four apparently ctorium and E. tsugicola was found in each of the most nonorthologous genes of Mn-dependent peroxidase, des- parsimonious trees, and the branch grouping all of the ignated A, C, D and E, were identified among amplifica- Amylostereum species was also found in the strict con- tion products of Echinodontium and Amylostereum spe- sensus tree. However, the relationships among cies using the degenerative primers (Maijala et al., 1998). Amylostereum, E. tinctorium/E, tsugico/a, and E. japoni- These nonorthologous genes were distinguished by cure were not resolved in the consensus tree. The differences in their putative amino acid sequences and neighbor-joining analysis gave a tree with the same topol- unique sequences of three included introns (Maijala et al., ogy as the most parsimonious tree shown in figure 2, ex- 1998). The peroxidase A gene was the most commonly cept in the branches relating A. areolaturn, A. ferreum, A. encountered among the hundreds of cloned fragments laevigatum from Europe, A. laevigatum from Japan, and that were sequenced, and we were able to generate per-

B1350 Japan 100 Amylostereum B1385 Germany areolatum B1352 Germany

B1359 Brazil Amylostereum 100 ferreum 5 changes 100 L B1354 UK IAmylostereum B1391 Canada chailletii

100 - B1393 USA Amylostereum sp.

1361Japan I Japanese 88 B1365 Japan i Amylostereu m 100 ~i1369 Japan I laevig atum B1371 France I French 8.r I A mylostereum B1372 France ~ laevigatum

100 B13 USA I Bl122 USA I Echinodontium tinctorium

1009 85l B1377 Japan I Echinodontium tsugicola B1378 Japan I

1O0 f B1375 Japan I B1374 Japan Echinodontium japonicum

Bondarzewia montana- Peroxidase 2

Bondarzewia montana - Peroxidase B Fig. 3. One of two most parsimonious trees based on 410 characters of exons of a partial sequence of manganese-dependent peroxi- dase A of Amylostereum and Echinodontium species. The tree is rooted to Bondarzewia montana peroxidase 2 and peroxidase B sequences. Bootstrap values (1000 replicates) greater than 50~ are indicated above the branches. Isolate numbers in bold are isolates from horntails. Molecular phylogeny of Amylosterem and Echinodontium 591 oxidase A sequences for each of the Amylostereum and lyses confirm that A. areolatum is the associate of S. Echinodontium species, except for E. taxodii. In com- nitobei, S. juvencus and S. noctilio from Japan and Eu- paring the putative amino acid sequence of peroxidase A rope. Sirex noctilio and its symbiont A. areolatum have with that of other Mn-dependent peroxidase genes, the also been introduced to South America, South Africa, peroxidase B and peroxidase 2 genes of B. montana Australia, and New Zealand (Gilbert and Miller, 1952; Gil- (Maijala et al., 1998) were most similar, and these DNA moure, 1965; Slippers et al., 1998; Vibrans, 1991). sequences were used as outgroups in the parsimony An unidentified Amylostereum isolate (B1373) from analysis of the putative exons of peroxidase A (Fig. 3). the mycangium of Xo. matsumurae in Japan had an ITS The peroxidase A introns of the Echinodontium and sequence identical to that of other A. areolatum isolates. Amylostereum species could not be aligned without am- Isolate B1373 from Xo. matsumurae and two isolates biguity. (FD-241 and FD-242) from S. nitobei were grown on Of the included 410 characters from the four exons, potato-dextrose agar in Petri plates at 25~ and the cul- 213 were constant and 57 were parsimony-uninforma- tural characteristics were noted. These three isolates tive, leaving 140 parsimony-informative characters. formed mycelia that were white at first, later becoming Two most parsimonious trees of 348 steps and identical brown, cottony, and with a sweet odor. They had pale topology were found, with C1=0.7356, R1=0.8221, and brown, thick-walled, clavate cystidia with an acute to RC=0.6047. The inferred relationships among the in- somewhat rounded edge, encrusted with granular materi- group taxa (Fig. 3) were very similar to those inferred al. They produced arthrospores that were 5-21 x 1- from the ITS analysis. Echinodontium japonicum was 4 f~m. From the ITS sequences and morphological com- basal to the other ingroup taxa, E. tinctorium and E. parisons, we conclude that the fungal symbiont of Xo. tsugicola formed a strongly supported group, and the ge- matsumurae is A. areolatum. This is the first report of a nus Amylostereum was well supported by a 100~ boot- mycangial from any Xoanon species, but only one strap value. As in the ITS analysis, A. areolatum was isolation was made from a single horntail. distinct from the other Amylostereum species (Fig. 3). Amylostereum chailletiiis a common decay fungus in The Japanese and French isolates of A. laevigatum both North America and Europe, and it also has been re- formed two distinct, sister clades, and these two clades corded as a symbiont of S. areolatus, S. cyaneus Fab., grouped strongly with the unidentified Amylostereum iso- Urocerus augur augur Klug, U. augur sah Mocs~ry, U. late (B1393) from California. Neighbor-joining of the californicus Nort., and U. gigas L. (Gaut, 1970). Isolates exon DNA sequences gave a tree identical to that from from U. gigas in Europe had ITS and peroxidase A se- parsimony analysis (Fig. 3), except that E. japonicum quences that were similar to those from decayed wood or grouped with the other two Echinodontium species in the basidiomes. neighbor-joining analysis. The putative amino acid se- Gaut (1970) found by anastomosis, dikaryotization, quences for peroxidase A were included in a much larger and interfertility tests that the fungus associated with S. data set for another study (Maijala, Harrington, and areolatus was A. chailletii. An Amylostereum isolate Raudaskoski, unpublished), and parsimony analysis of (B1393=CCFC010375) from the mycangium of S. these amino acid sequences gave trees with identical areolatus was isolated by Stillwell and was presumed to topology to the parsimony analysis of the exon se- be A. chailletii. However, this isolate has unique ITS and quences with respect to the Amylostereum and peroxidase A sequences that were similar to those of A. Echinodontium species. laevigatum and distinct from those of A. chailletii. Sirex areolatus commonly attacks cedar-like trees such as Discussion Libocedrus, Cupressus, Juniperus, and Sequoia (Furniss and Carolin, 1977), similar to the hosts of U. japonicus The four recognized species of Amylostereum appear to and U. antennatus, which have A. laevigatum as its sym- form a monophyletic group based on both ITS and peroxi- biont. More isolates need to be examined to clarify the dase A sequence analyses. Further, A. ferreum, A. species of Amylostereum associated with S. areolatus, chailletii, and A. laevigatum form a separate group, sister but it appears that A. laevigatum is the primary symbiont to A. areolatum. Similarly, Slippers et al. (1998) found of Sirex and Urocerus horntails attacking cedar-like tree by the sequences of small-subunit mt-rDNA and the first species, while A. chailletii and A. areolatum are the pri- intergenic spacer (IGS-1) region of the nuclear rDNA that mary symbionts of the horntails attacking the Pinaceae. A. ferreum and A. laevigatum are more closely related to Although the ITS sequences of A. laevigatum iso- A. chailletii than to A. areolatum, and Vasiliauskas et al. lates from Japan and France were distinct, isolates from (1999) also found that A. chailletii and A. /aevigatum these two countries had similar peroxidase A se- from Europe have similar ITS sequences. Boidin and quences. It has been speculated that A. laevigatum in Lanquetin (1984) showed that A. ferreum was partially Europe is two distinct taxa, one occurring on Juniperus interfertile with A. laevigatum but was completely inter- (with basidiospores 7-9/~m long) and another on Taxus sterile with A. areolatum. (basidiospores 9-12/~m long) (Eriksson and Ryvarden, Isolates of A. areolatum, A. chailletii, and A. laeviga- 1973). However, Boidin and Lanquetin (1984) did not turn from horntails had similar or identical ITS and peroxi- find the same distinction between Taxus and Juniperus dase sequences to the respective isolates from isolates from Sweden and France. Specimens of A. basidiomes or decayed wood. Our DNA sequence ana- laevigatum from Japan (Forestry and Forest Products 592 M. Tabata et al.

Research Institute, SFM1, SFM2, and SFM3) and those moved from the Stereaceae and placed in the Echinodon- from Juniperus in Sweden (Botanical Museum of Uppsala tiaceae. University, UPS Nos. 142326, 142340, and 142359) The Echinodontiaceae was proposed by Donk (1961 ) were examined, and each had the shorter basidiospores and emended by Gross (1964), recognizing only the ge- typical of the Juniperus type. Thus, the peroxidase A nus Echinodont/um, which included species with smooth sequences and morphological data support the earlier hymenophores as well as hydnoid hymenophores. identification of the symbiotic fungus associated with U. JLilich (1981) adopted Gross's interpretation and de- antennatus and U. japonicus as A. laevigatum (Tabata scribed the family in detail, but he added another genus, and Abe, 1997; Tabata and Abe, 1999). Further studies Laurilia, including E. taxodii and E. sulcatum. The dispo- of more specimens of A, laevigatum from Europe are sition of Laurilia is unclear, but affinities of L. taxodii with needed, however. Echinodontium and Stereum are questioned by our DNA The results of ITS sequence and peroxidase A ana- sequence analyses. Ryvarden (1991) suggested that lyses show that some of the Echinodontium species are the genus Haploporus be placed in the Echinodontiaceae, quite closely related to Amylostereum. Echinodontium but more data are needed to determine if Haploporus is tsugicola is morphologically very similar to E. tinctorium related to Echinodontium and Amylostereum. (Gilbertson and Ryvarden, 1986; Imazeki, 1935), and Acknowledgements We would like to thank Mr. K. Sasaki, these species are sister to the genus Amylostereum, Hokkaido Research Center of Forestry and Forest Products though these Echinodontium species have prominently Research Institute and Dr. K. Ito, Kyushu Research Center of spinose hymenophores while those of Amylostereum are Forestry and Forest Products Research Institute for their help in smooth. The hymenophore of E, japonicum, which oc- collecting isolates from Xoanon matsurnurae; Mr. J. Steimel, curs on Quercus rather than conifers, has fine teeth and Iowa State University, for technical assistance; and Dr. A. is related to the other Echinodontium species and Shinohara, National Science Museum, for providing taxonomic Amylostereum based on ITS and peroxidase se- information. We also thank the curators of the Institute for Fer- quences. However, it does not clearly fall into the mentation, Osaka, Japan, the Centraalbureau voor Schimmelcul- Amylostereum or Echinodontium group. The tures (Baarn, Netherlands), and the Canadian Collection of Fun- phylogenetic analyses suggest that Ech/nodont/um is gal Cultures (Ontario, Canada) for providing isolates of paraphyletic (contains another genus, i.e., Amylostere- Amylostereum and Echinodontiurn. Dr. D.S. Hibbett, Clark um) if Echinodontium is comprised of E. tinctoriurn, E. University, Worcester, Massachusetts, kindly provided extract- tsugicola, and E. japonicurn. Unfortunately, cultures of ed DNA of several taxa. This study was supported by Special E. ballouii were not available for study. Coordination Funds of Science and Technology Agency of The hymenophore of E. taxodii is smooth, and it has Government, Japan for Promoting Science and Technology (No. been considered a species of Laurilia (Stereaceae) by Par- 74). masto (1 968). The ITS sequence of our isolate of E. tax- odii was very distinct from the other Echinodontium spe- cies and from Stereum, and we were unable to obtain a Literature cited peroxidase A gene fragment from this isolate. This sup- ports the exclusion of E. taxodii from Echinodontium and Boidin, J. and Lanquetin, P. 1984. Le genre Amylostereum Stereum, and the name L. taxodii (Lentz & McKay) Parm. (Basidiomycetes) intercompatibilit~s partielle entre especes is regarded as appropriate. The related E. sulcatum was allopatriques. Bull. Soc. Mycol. France 100:211-236. not studied here, but its similarity to L, taxodii in biology Boidin, J., Mugnier, J. and Canales R. 1998, Taxonomic and morphology (Davidson et al., 1960) suggests that it, molecular des Aphyllophorales. Mycotaxon 66: 445-491. too, may belong outside of Echinodontium, most likely in Breitenbach, J. and Kranzlin, F. 1988. Fungi of Switzerland, Laurilia (Pouzar, 1959). vol. 2. Non-gilled Fungi. Verlag Mykologia, Lecerne. Amylostereum, Echinodontium, and Laurilia have Chamuris, G. 1988. The non-stipitate stereoid fungi in the northeastern United States and adjacent Canada. Mycol. resupinate to effused-reflexed and amyloid Mem. No. 14. J. Cramer, Stuttgart., pp. 41-45. basidiospores and cause white rots. However, these Davidson, R. W., Lentz, P. L. and Mckay, H. H. 1960. The fun- genera are distinguished by the following features. gus causing pecky cypress. Mycologia 52: 260-279. Amylostereum has an even hymenial surface, monomitic DeScenzo, R. A. and Harrington, T. C. 1994. Use of (CAT)5 as or dimitic hyphal system, smooth basidiospores, and, in a DNA fingerprinting probe for fungi. Phytopathology 84: three of four species, symbiotic relationships with horn- 534-540. tails. Echinodontium has a hydnaceous hymenial sur- Donk, M.A. 1961. Four new families of Hymenomycetes. face, dimitic hyphal system, and echinulate basidio- Persoonia 1" 405-407. . Laurilia has an even to tuberculate hymenial sur- Donk, M.A. 1964. A conspectus of the families of Aphyl- face, dimitic or trimitic hyphal system, and echinulate Iophorales. Persoonia 3: 199-324. Eriksson, J., Hjortstam, K. and Ryvarden, L. 1978. The Cor- basidiospores. Boidin et al. (1998) erected a new mono- ticiaceae of North Europe, vol. 5., pp. 890-893. Fun- typic family for (Amylostereaceae), but Amylostereum giflora, Oslo. the close phylogenetic relationship shown in our study Eriksson, J. and Ryvarden, L. 1973. The Corticiaceae of North supports placement of Arnylostereum in the Echinodon- Europe, vol. 2., pp. 90-95. Fungiflora, Oslo. tiaceae, Based on morphological characteristics and se- Furniss, R. L. and Carolin, V. M. 1977. Western Forest Insects. quence analyses, we propose that Arnylostereum be USDA For. Serv. Misc. Publ. No. 1339: 453-457. Molecular phylogeny of Amylosterem and Echinodontium 593

Gardes, M. and Bruns, T. D. 1993. ITS primers with enhanced pp. 62-70. specificity for basidiomycetes--application to the identifica- Parmasto, E. 1968. Conspectus systematis corticiacearum. tion of mycorrhizae and rusts. Mol. Ecol. 2:113-118. Inst. Zool. Bot., Acad. Sci. R. P. S. S. Estonicae. Tartu. Gaut, I. P. C. 1969. Identity of the fungal symbiont of Sirex Pouzar, Z. 1959. New gerena of higher fungi II1. Ceska noctilio. Aust. J. Biol. Sci. 22: 905-914. Mycol. 13: 10-19. Gaut, I. P. C. 1970. Studies of siricids and their fungal sym- Ryvarden, L. 1991. Genera of Polypores. Nomenclature and bionts. PhD thesis, Univ. Adelaide, Australia. , pp. 68-69. Fungiflora, Oslo. Gilbert, J. M. and Miller, L. W. 1952. An outbreak of Sirex noc- Sano, A., Mihara, Y. and Ito, S. 1995. The fungus isolated tilio F. in Tasmania. Aust. For. 18: 63-69. from the mycangia of Urocerus japonicus and U. antenna- Gilbertson, R. L. and Ryvarden, L. 1986. North American Poly- tus. Trans. 43th Mtg. Chubu Br. Jpn. For. Soc., pp. 125- pores, vol. 1., pp, 251-255. Fungiflora, Oslo. 126. (In Japanese.) Gilmoure, J. W. 1965. The life cycle of the fungal symbiont of Slippers, B., Wingfield, M.J., Coutinho, T.A. and Wingfield, Sirexnoctilio. N.Z.J. For. 10: 80-89. B.D. 1998. The identify and possible origin of the Ginns, J. and Lefebvre, M. N. L. 1993. Lignicolous corticioid Amylostereurn symbiont of Sirex noctilio in South Africa. fungi () of North America, systematics, distri- 7th Intern. Congr. of Plant Pathology (ICPP), Edinburgh, bution, and ecology. Mycol. Mem., No. 19., p. 21. APS Scotland Aug. 9-16, 3: 55. Press,St. Paul, Minnesota. Swofford, D.L. 1998. PAUP*: Phylogenetic Analysis Using Gross, H.L. 1964. The Echinodontiaceae. Mycopathol. Parsimony (~ and other methods), ver. 4. Sinauer Associ- Mycol. Appl. 24: 1-26. ates, Sunderland, Massachusetts, U.S.A. Harrington, T.C. and Wingfield, B.D. 1995. A PCR-based Tabata, M. and Abe, Y. 1997. asso- identification method for species of Arrnillaria. Mycologia ciated with the Japanese horntail, Urocerus japonicus. 87: 280-288. Mycoscience 38: 421-427. Harrington, T. C., Stenlid, J. and Korhonen, K. 1997. Evolution Tabata, M. and Abe, Y. 1999. Amylostereurn laevigatum asso- in the genus Heterobasidion. 9th Intern. Conf. of Root and ciated with a horntail, Urocerus antennatus. Mycoscience Butt Rots of Forest Trees, Carcans-Maubuisson, France 40: 535-539. Sept. 1-7, pp. 63-74. Terashita, T. 1970. A Basidiomycete symbiotic to a siricid in Hibbett, D. S. and Donoghue, M.J. 1995. Progress toward a Japan. J. Jpn. For. Soc. 52: 313-316. (In Japanese.) phylogenetic classification of the Polyporaceae through par- Vasiliauskas, R., Johannesson, H. and Stenlid, J. 1999. simony analysis of mitochondrial ribosomal DNA se- Molecular relationships within the genus Amylostereum as quences. Can. J. Bot. 73: S853-$861. determined by internal transcribed spacer sequences of the Hsiau, P. T. 1996. The taxonomy and phylogeny of the mycan- ribosomaIDNA. Mycotaxon71: 155-161. gial fungi from Dendroctonus brevicomis and D. frontalis Vibrans, A. 1991. Biological control of the woodwasp (Sirex (Coleoptera: Scolytidae). PhD thesis, Iowa State Universi- noctilio) in Brazil. Forstarchiv 62: 97-99. ty, USA. White, T. J., Bruns, T., Lee, S. and Taylor, J. 1990. Amplifica- Imazeki, R. 1935. Studies on Echinodontium Ellis et Everhart. tion and direct sequencing of fungal ribosomal RNA genes J. Jpn. Bot. 11: 514-521. for phylogenetics. In: PCR protocols: a guide to methods JSlich, W. 1981. Higher taxa of Basidiomycetes. Biblio. and application, (ed. by Innis, M.A., Gefand, D. H., Snin- Mycol. 85. J. Cramer, Stuttgart. sky, J. J. and White, T. J.), pp. 315-322. Academic Press, Maijala, P., Harrington, T. C. and Raudaskoski, M. 1998. Per- California. oxidase gene structure and gene trees in Heterobasidion. Wu, S. H. 1995. Two new genera of corticioid basidiomycetes Fourth Meeting on the Genetics and Cellular Biology of with gloeocystidia and amyloid basidiospores. Mycologia Basidiomycetes, Nijmegen, The Netherlands March 27-30, 87: 886-890.