| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Systematics and Phytogeography |
Department of Biology, Clark University, 950 Main Street, Worcester, Massachusetts 01610 USA
Received for publication December 14, 2004. Accepted for publication June 2, 2005.
ABSTRACT
Mitrula species represent a group of aquatic discomycetes with uncertain position in the Helotiales and an unknown life history. Mitrula species were studied using a combination of cultural, morphological, and molecular techniques. Pure colonies were isolated from Mitrula elegans, and conidia were induced in vitro. Herbarium materials from Europe, Asia, and North America were studied. Sequences of rDNA, including partial small subunit rDNA, large subunit DNA and ITS, were used to infer phylogenetic relationships both within Mitrula and between Mitrula and other inoperculate discomycetes, with special attention to fungi that resemble Mitrula in morphology or ecology. Equally weighted parsimony analyses, likelihood analyses, constrained parsimony analyses, and Bayesian analyses were performed. Results suggest that (1) the anamorph of M. elegans produces brown bicellular conidia, (2) a new subalpine species M. brevispora is distinct, (3) more than six lineages and clades can be recognized in Mitrula, (4) the morphological species M. elegans is not monophyletic, (5) a close relationship between Mitrula and either Geoglossaceae or Sclerotiniaceae is not supported, (6) the Helotiaceae is paraphyletic, and (7) Mitrula belongs to a clade within the Helotiales that also includes other aero-aquatic genera, Cudoniella, Hydrocina, Vibrissea, Ombrophila, and Hymenoscyphus.
Key Words: aquatic fungi • decomposition • ecology • mitosporic fungi • vernal pools
Mitrula Fr. species, the so-called swamp beacons, represent a group of unique nonlichenized discomycetes in the Helotiales. Relatively common in northern temperate forests, they are distributed from Europe to Asia and North America (Redhead, 1977
). Mitrula species produce yellow to pinkish-orange, club-shaped spore-producing bodies (apothecia), which mostly occur on submerged, decaying vegetation in standing or slowly moving, shallow water. The apothecia emerge above the water level and are termed "aero-aquatic." In North America, Mitrula species have been collected from April to October (Seaver, 1951
; Redhead, 1977
).
Ascomycetes generally have a stable haploid anamorph stage, with the ability to produce asexual spores (conidia). The mycelium produced by a single conidium can recycle the anamorph, or it can fuse with another haploid individual to produce the dikaryotic teleomorph stage. A huge number of conidia can be produced in asexual life cycles, and thus anamorphs play an important role in the distribution, dikaryotization, and fertilization of ascomycetes. Anamorphs associated with aquatic members in the Helotiales have been treated as an ecological group, the aquatic hyphomycetes, which also are known as amphibious or Ingoldian fungi, because they were first described by Ingold (1942)
.
Fungi play a major role in decomposition of wood and leaf litter in forest ecosystems. Many wood- and litter-decaying fungi are members of the Basidiomycota (Alexopoulos et al., 1995). However, the extracellular wood-decaying enzymes produced by Basidiomycota require aerobic conditions, and substrates submerged under water do not favor the growth of these fungi (Kirk and Farrell, 1987
; Pointing et al., 2004
). Ingoldian fungi degrade large amounts of leaf litter and woody debris in woodland streams and rivers using a form of decomposition that has been classified as "soft rot," and which has been reported in Ascomycota and anamorphic fungi (Blanchette, 1994
). In one study, 75% of the fungal biomass on underwater substrates was made up of ascomycetes (Nikolcheva et al., 2003
). Thus, Mitrula and other Ingoldian fungi play an important role in nutrient cycling in certain freshwater ecosystems. Because of their important ecological roles, Ingoldian fungi have been studied extensively, and more than 300 species have been described since 1942 (Harrington, 1997
).
Despite the attention they have received, the teleomorphs of most aquatic anamorphs remain poorly known. In the families of the Helotiales, there is little correlation between the classifications of teleomorphs and anamorphs (http:// www.fm5web.life.uiuc.edu:23523/mitosporic/; Sutton and Hennebert, 1994
). Many anamorphic fungi have never been observed to have a sexual life cycle in the field, and it is possible that they may have completely lost the sexual state through evolution.
Molecular methods have been used to investigate the linkages between anamorphs and teleomorphs (Pfister, 1997
; Goodwin, 2002
; Harrington and McNew, 2003
). In the Helotiales, aero-aquatic species of Cudoniella, Hymenoscyphus, Hydrocina, Loramyces, Ombrophila, and Vibrissae have been reported to have aquatic anamorphs in the form-genera Anavirga, Anguillospora, Articulospora, Helicodendron, and Tricladium (Abdullah et al., 1981
; Fisher and Webster, 1983
; Descals et al., 1984
; Digby and Goos, 1987
; Fisher and Spooner, 1987
; Hamad and Webster, 1988
; Webster et al., 1995
; Marvanová and Descals, 1996
). Thousands of ascocarps of Mitrula have been collected, but an anamorphic state has not been reported in the life history.
The systematic position of Mitrula in the Helotiales also is uncertain. Because of the club-shaped apothecia, Mitrula species have been accepted as members of the family Geoglossaceae, which contains fungi commonly referred to as "earth tongues" (e.g., Geoglossum glutinosum Pers.: Fr.) (Imai, 1941
; Dennis, 1968
; Korf, 1973
; Benkert, 1983
). The concept of the Geoglossaceae has been modified extensively by recent studies and classifications (Hawksworth et al., 1995
; Kirk et al., 2001
), and it seems likely that the family will be restricted and excluded from a monophyletic Helotiales in future classifications (Platt, 2000
; Pfister and Kimbrough, 2001
; Z. Wang et al., unpublished manuscript). Based on ultrastructure of the ascus (a saclike cell containing ascospores) from selected taxa in the Helotiales, Verkley (1994)
concluded that Mitrula paludosa Fr. is closely related to members of the Sclerotiniaceae, and this placement was accepted by some recent classifications (Hawksworth et al., 1995
; Eriksson et al., 2004
). However, based on internal transcribed spacer (ITS) rDNA sequences, Mitrula elegans (Berk.) Fr. was placed as the sister group of Chloroscypha species (Helotiaceae) instead of having a close relationship with Sclerotinia sclerotiorum (Lib.) de Bary (Sclerotiniaceae) (Gernandt et al., 1997
). The family Helotiaceae was resolved as paraphyletic in a study based on 18s rDNA phylogeny (Gernandt et al., 2001
), and the position of Mitrula in the family was accepted with a question mark in The Dictionary of the Fungi, vol. 9 (Kirk et al., 2001
).
Fifty-five species names have been recorded under the genus Mitrula (http://www.indexfungorum.org). Some species described in Mitrula have been transferred to other genera, such as Bryoglossum, Scleromitrula (= Verpatinia), and Heyderia (Imai, 1941
; Maas Geesteranus, 1964
; Dennis, 1968
; Korf, 1973
; Redhead, 1977
; Schumacher and Holst-Jensen, 1997
). Morphological characters used for defining Mitrula species are limited and include characters of asci, ascospores, and apothecia. Noting that the measurements of asci and spores by different authors were not consistent, Imai (1941)
, Maas Geesteranus (1964)
, and Korf (1973)
suggested that Mitrula is a monotypic genus, with M. paludosa as the sole species.
The present concept of Mitrula (Redhead, 1977
) includes four species that are united by possession of orange, clavate, fleshy apothecia, a white stipe, inflated stipe hyphae, hyaline ascospores, an amyloid apical ascal tip, and an aquatic habitat. In Redhead (1977)
, characters of spores and apothecia, and geographic distributions were used to distinguish species. Four species were recognized, M. elegans, M. borealis S.A. Redhead, M. lunulatospora S.A. Redhead, and one European species M. paludosa, which may be referred to Asian material as well (Redhead, 1977
). Mitrula borealis has a boreal and subalpine distribution in North America and Europe, while the other three species are believed to be restricted to shallow water at a low altitude. In addition to M. borealis and M. paludosa, two Mitrula species with dark brown or pink hymenophores, M. multiforme (E. Henning) Massee and M. omphalostoma E.-H. Benedix, were accepted in Europe by Benkert (1983)
. The genus Bryoglossum was erected based on the bryophilous (moss-inhabiting) species Mitrula gracilis Karsten. Bryoglossum gracilis (Karsten) S.A. Redhead may be closely related to Mitrula species, but shows some significant differences from the four Mitrula species in producing minute cauloscales (scale-like stipe hairs), having a gelatinized free margin, a yellowish stipe with narrow stipe hyphae, and releasing a yellow to reddish brown pigment in 10% KOH solution (Kankainen, 1969
; Redhead, 1977
).
The goals of this study were to (1) isolate Mitrula strains from field collections and induce its anamorphic stage in vitro, (2) evaluate the number of species within Mitrula, and (3) position Mitrula in the Helotiales using data from three rDNA regions, small subunit rDNA (ssu-rDNA), large subunit rDNA (lsu-rDNA), and 5.8S rDNA, from diverse taxa representing the major groups and many aero-aquatic genera in the Helotiales.
MATERIALS AND METHODS
Morphological and cultural studies
Morphological descriptions are based on observations of fresh, dried, or rehydrated specimens. Microscopic studies use squashed tissues. Measurements were made in Melzer's reagent (Korf, 1973
) using bright field microscopy (Olympus CH-2, Olympus, Tokyo, Japan). Methods for collecting, isolating, and culturing Mitrula species basically followed Goos et al. (1986)
. Ascospores were collected in petri dishes on 2% malt extract agar sealed with parafilm, and resulting colonies were kept at room temperature (16–20°C) under ambient light. ITS sequences of the colonies were checked to detect contamination. Colonies were transferred into flasks containing sterilized leaves of Fagus americana Sweet. submerged in water, and were incubated at room temperature to induce sporulation. Mycelia were removed from the flasks onto moist filter paper and incubated under ambient light at room temperature for 2 weeks.
The nuclear condition of the mycelia was examined using the fluorescent stain 4',6'-diamidino-2-phenylindole hydrochloride (DAPI, Sigma, St. Louis, Missouri, USA). Sterile cover slides were placed close to selected colonies for a week, and the fungal cells were allowed to grow onto the slides. Attached fungal cells were fixed with formaldehyde solution (10% formaldehyde in 0.1% Triton X-100) overnight and washed twice with phosphate-buffered saline (PBS; pH 7.4, 138 mM NaCl, 2.6 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4). The cells were then resuspended in DAPI solution (1 : 10 000 DAPI in PBS) for 10 min in the dark followed by two washes with PBS. Slides with attached fungal cells were mounted with 50% glycerol solution and sealed with nail polish. Images were captured on a Nikon E600 microscope (Nikon, Augusta, GA, USA) with bright field and epifluorescence optics and a SPOT RT Slider cooled colored digital (CCD) camera using SPOT software (Diagnostic Instruments, Sterling Heights, Michigan, USA).
Molecular techniques
DNA was isolated from dried fruiting bodies and cultures. Approximately 20–30 mg of fungal tissue (herbarium materials) or a small amount of mycelia from cultures was ground in liquid nitrogen and extracted in 600 µL of extraction buffer (1% SDS, 0.15 M NaCl, 50 mM EDTA) at 75°C for 1 h, purified with phenol-chloroform-isoamyl alcohol (25 : 24 : 1), and precipitated with 95% ethanol and 3 M NaCl overnight. Crude DNA extracts were purified with GeneClean (Bio 101, La Jolla, California, USA). Cleaned DNA samples were diluted with distilled water up to 500 fold for use as PCR templates.
Sequence data were generated from three regions: (1) partial nuclear small subunit (nuc-ssu) rDNA bounded by primers PNS1 and NS41 (White et al., 1990
; Hibbett, 1996
), from 36 isolates of 31 genera; (2) partial nuclear large subunit (nuc-lsu) rDNA bounded by primers JS-1 and LR5 (Vilgalys and Hester, 1990
; Landvik, 1996
), from 47 isolates of 31 genera; (3) complete internal transcribed spacers 1 and 2 and the 5.8S rDNA (nuc-ITS rDNA) bounded by primers ITS1F and ITS4 (White et al., 1990
), from 47 isolates of 31 genera. Sequences generated in this study were submitted to GenBank (accession numbers AY789276-AY789434). Twenty-nine additional sequences were downloaded from GenBank (Appendix 1).
Thirteen isolates of Mitrula were included in molecular studies, representing M. lunulatospora, M. borealis, and M. elegans from North America, M. paludosa from Europe, and a subalpine collection from China that is described here as M. brevispora sp. nov. Of the 15 families accepted in the recent Helotiales classification (Kirk et al., 2001
), 11 families, including Bulgariaceae, Leotiaceae, Cudoniaceae, Geoglossaceae, Hemiphacidiaceae, Rutstroemiaceae, Sclerotiniaceae, Helotiaceae, Hyaloscyphaceae, Vibrisseaceae, and Dermateaceae were sampled. Species of aquatic genera such as Vibrissea, Hydrocina, Cudoniella, and Ombrophila also were included, as were species with morphological similarity to Mitrula, such as Bryoglossum gracile, Heyderia abietis (Fr.) Link, and Scleromitrula shiraiana (Henn.) S. Imai. Species from Rhytismatales, Dothideales, Mycocaliciales, and Pezizales were sampled as well.
PCR reaction mixes (Promega Corp., Madison, Wisconsin, USA) contained 2.5 µL 10x PCR buffer, 5 µM dNTP, 12.5 pM of each PCR primer, and 5 µL DNA in 25 µL. The amplification program included 40 cycles of 94°C for 30 s, 50°C for 30 s, and 72°C for 1 min. PCR products were purified using Pellet Paint (Novagen, Madison, Wisconsin, USA) and sequenced using the ABI Prism BigDye-terminator cycle sequencing kit (Applied Biosystems, Foster City, California, USA) according to the manufacturer's protocols. Primers used for sequencing were PNS1, NS19bc, NS19b, NS41, JS-1, LR3, LR3R, LR5, ITS1F, and ITS4 (Vilgalys and Hester, 1990
; White et al., 1990
; Landvik, 1996
). Sequencing reactions were purified using Pellet Paint and were run on an Applied Biosystems 377XL automated DNA sequencer. Sequences were edited with Sequencher version 3.1 (GeneCodes Corp., Ann Arbor, Michigan).
Phylogenetic analyses
Two data sets were prepared, one for higher-level analyses (HLA) and one for lower-level analyses (LLA). The data set for HLA included sequences of 36 isolates from three genes, nuc-ssu rDNA, nuc-lsu rDNA and 5.8S rDNA. The HLAs were intended to resolve the placement of Mitrula in the Helotiales. The HLA data set was rooted using Neolecta irregularis (Perk) Korf & J.K. Rogers (Landvik, 1996
). The nuc-ssu rDNA sequences of Ciboria sp. G36, Bisporella sp. G39, and Scleromitrula shiraiana G123 were about 360–560 base pairs (bp) shorter than sequences of the other taxa. The nuc-lsu rDNA sequence of Rutstroemia bolaris (Batsch) Rehm was about 527 bp shorter than sequences of other taxa.
The data set for the LLA included sequences of nuc-lsu rDNA and ITS from 24 isolates. LLA were intended to resolve the relationships among Mitrula species. Mitrula lsu-rDNA and ITS sequences were used as BLAST queries of the GenBank database, but no sequences with high similarity (>95%) were found. The LLA data set was rooted using Heyderia abietis based on the results of the HLA.
Sequences were aligned with ClustalX (Thompson et al., 1997
) and adjusted by eye in the data editor of PAUP* 4.0b (Swofford, 2002
). Alignments are available at TreeBASE (accession nos. M2285 and M2286). Both data sets were analyzed in PAUP* 4.0b (Swofford, 2002
) and MrBayes 3.0 (Huelsenbeck and Ronquist, 2001
), with gaps treated as missing data. Ambiguously aligned positions were excluded from the data sets before performing the analyses.
Parsimony analyses were performed using equal weighting of characters and transformations. Heuristic searches were performed with 1000 replicate searches, each with a random taxon addition sequence, MAXTREES set to autoincrease, and tree-bisection-reconnection (TBR) branch swapping. A bootstrap analysis was performed with 1000 replicates, each with 10 random taxon addition sequences, MAXTREES set to autoincrease, and TBR branch swapping.
Likelihood analyses were performed using a model and model parameters estimated with Modeltest version 3.5 (Posada and Crandall, 1998
) and MrModeltest 2.0 (Nylander, 2004
). A likelihood command block was copied from MrModeltest 2.0, which configures PAUP* to search for a maximum likelihood (ML) tree under the GTR + 
+ I model with 100 heuristic search replicates, each with a random taxon addition sequence, MAXTREES set to autoincrease, and TBR branch swapping. The rate matrix (A-C, 0.722900; A-G, 2.077300; A-T, 0.939200; C-G, 0.801700; C-T, 5.015800; G-T, 1.000000), assumed nucleotide frequencies (A = 0.26370, C = 0.22000, G = 0.27150, T = 0.24480), assumed proportion of invariable sites (=0.424) and gamma distribution (shape parameter alpha = 0.5056) were estimated by hLRT in MrModeltest 2.0.
Constrained analyses were performed using trees constructed with constraint command in PAUP* to test two alternative hypotheses: (1) Mitrula species are members of Sclerotiniaceae. In this case, a group including two Mitrula species, Ciboria, Sclerotinia, Monilinia, Scleromitrula, and Rutstroemia was forced to be monophyletic. (2) Mitrula belongs to the family Geoglossaceae. A group including Mitrula, Geoglossum and Trichoglossum was forced to be monophyletic. Tree searching in the constrained analyses used the same parsimony analyses settings described previously. Equally parsimonious trees from unconstrained and constrained analyses were compared using the SH test in PAUP* (Shimodaira and Hasegawa, 1999
; Goldman et al., 2000
). The SH test was performed for 1000 replicate bootstrap data sets by using full likelihood optimizations. Based on the overall likelihood value, trees significantly different (P < 0.05) from the best tree were rejected (Table 1).
|
+ I model in MrBayes 3.0 by running four chains with 500 000 generations using the program default priors on model parameters. Trees were sampled every 100 generations. Likelihoods converged to a stable value after 10 000 generations, and all trees prior to this convergence were discarded as the "burn-in" phase before computing a majority rule consensus tree in PAUP*. The same analytical settings of parsimony analyses as those for the HLA were applied to the LLA for combined nuc-lsu rDNA and ITS data.
RESULTS
Mitrula brevispora Zheng Wang sp. nov. (Fig. 1)
Ascoma solitaria, 14–32 mm altum. Ascogenea forma cylindracea, subalbea vel lutea, 4–12 x 2–4.5 mm. Stipes 0.8–1 mm crassus. Hymenium 97–102 µm crassus. Asci cylindraceo-clavati, apice attenuati, jodo abturaculo minutissimo caerulescentes, 4.9– 6.7 µm crassus. Paraphyses filiformes, 2.4–2.6 µm crassus. Sporae 8: ellipticae vel subfusoideae, 2.4–3.1 5.5–10.1 µm.
|
Ascocarps solitary, flesh, 14–32 mm high. Ascogenous part cylindrical to slightly clavate, smooth, beige to yellow, 4–12 x 2–4.5 mm. Stipe unbranched, 0.8–1 mm wide, hyaline to white. Hymenium about 97–102 µm thick. Asci eight-spored, elongate-clavate, 4.9–6.7 µm wide, croziers present, apical pores amyloid. Paraphyses filiform, slightly enlarged above, 2.4–2.6 µm wide. Ascospores elliptical-fusoid to broadly cylindrical, one-celled, bi-guttulate to multi-guttulate, rarely with a gelatinous sheath, 2.4–3.1 x 5.5–10.1 µm. Subhymenium not distinct. Clavula medulla hyphae 4.8–17 µm wide. Stipe medulla hyphae 14–24 µm wide, stipe inner cortical hyphae 3.5–7.5 µm wide. On mossy soil and dead foliage in coniferous forests in alpine or subalpine areas.
Additional specimens examined: On duff in conifer forests. Satan Forest farm (3300 m a.s.l.), Bailongjiang River, Gansu, China, 10 July 2002; leg. Zheng Wang No. ZW02-020 (Clark University).
Pure cultures and conidia of Mitrula elegans
Ascospores from two isolates, G45 from Massachusetts and G146 from West Virginia, germinated on 2% malt extract agar (MEA) plates in 24 h with germ tubes extending from both ends of the ascospores. All single spore isolates grew slowly, and colonies from single spores grew poorly even after transfer onto a new MEA plate. Multispore cultures reached 10 mm in diameter in 3 to 5 days at 16–20°C and became orange to brownish-orange because of brown incrustations produced by some cells. The function of brown incrustations and brown-walled cells is unknown. Brown pigments were detected in the medium as well. Hyphae became dark colored when the media became dry. Some cells became swollen to form chains of chlamydospore-like structures, and hyphae were found to grow out from those chlamydospore-like cells with or without rehydrating the media (Fig. 2a). A single nucleus was revealed in each cell by DAPI (Fig. 3).
|
|
Anamorph of Mitrula elegans: Aquatic. Conidiophores not distinguishable from the vegetative hyphae. Conidiogenous cells terminal, hyaline, integrated. Conidia in early stage solitary, dark brown, filiform, septate, sometimes dichotomously branched, often of highly variable length, 10–20 µm wide. Mature conidia may be released by disarticulation of a septum or may simply be cleaved from the conidiogenous cell, brown to dark brown, ellipsoid to ovoid, bicellular, septa becoming distinct during development, 12–17 x 7.0–7.5 µm. No germination of the conidia was observed.
Phylogenetic inference from the HLA data set
The placement of Mitrula was estimated using three rDNA regions (lsu + ssu + 5.8S; Fig. 4). The combined genes had an aligned length of 2024 base pairs (22 positions were excluded from the analyses) with 220 uninformative variable positions and 395 parsimony-informative positions.
|
Maximum likelihood analysis based on the HLA data set under the GTR + + I model generated two trees of equal likelihood value (–lnL = 11383.91954). The majority rule consensus tree of 4900 Bayesian trees under the same model supported the same tree topology as the maximum likelihood analysis, which differed from the MP tree mainly by the placement of Mitrula and the Rhytismatales (Fig. 4b). However, the conflict between MP and ML trees concerning the placement of Mitrula was not supported by high bootstrap or posterior probability (PP) values. In the ML tree (Fig. 4b), the genus Mitrula was supported as monophyletic (PP = 1.0) and was weakly supported (PP = 0.68) as the sister group of a clade including Ascocoryne and Chloroscypha. Members of the family Helotiaceae were present in four clades in the ML tree. The families Sclerotiniaceae, Rutstroemiaceae, and Hemiphacidiaceae and Heyderia abietis formed a well supported clade (PP = 1.0). In agreement with the result of MP analyses, the Geoglossaceae formed a basal branch of the inoperculate ascomycetes in the ML tree.
Constrained analyses under the assumption that Mitrula is a member of the Sclerotiniaceae generated a single most parsimonious tree. Constrained analyses under the assumption that Mitrula is a member of the Geoglossaceae generated three equally parsimonious trees. Compared with the maximum parsimony tree generated from the unconstrained analyses, trees under both assumptions were rejected by the S-H test (Table 1).
Phylogenetic inference from the LLA data set
The relationships among Mitrula species were inferred using two rDNA regions (lsu + ITS). The combined data had an aligned length of 1390 base pairs with 143 uninformative variable positions and 259 parsimony-informative positions. Equally weighted parsimony analysis based on this data set generated 26 equally parsimonious trees of 1046 steps and consistency index (CI) = 0.578 (Fig. 5). All Mitrula isolates formed a clade (BP = 100%), and the aquatic species Hydrocina chaetocladia was the sister group to Mitrula species, but with weak support (BP < 50%). More than six lineages were detected among Mitrula species. Isolates of M. borealis from North America differed by 4 bp in the ITS region with M. brevispora from China, and these three subalpine species were well supported as a clade (BP = 100%). Mitrula lunulatospora and M. elegans G47 from eastern USA formed two independent lineages. Eight isolates formed a monophyletic group (BP = 94%), including six isolates of M. elegans from North America and two isolates of M. paludosa from Europe. Mitrula paludosa is monophyletic with moderate support (BP = 76%). Except for M. elegans G4 collected from Canada, isolates in the M. elegans-M. paludosa clade differed by less than 2 bp. Two isolates of M. elegans with different hymenophore colors, G45 (yellow) and G46 (pink), were from the same locality and only 1 bp differed.
|
Life history of Mitrula
The asexual state of Mitrula elegans has been successfully maintained in vitro, and an asexual sporulating aquatic stage has been observed to produce bicellular, brown conidia. Plasmogamy and dikaryotic hyphae have not been observed, and no apothecia have been produced from the Mitrula cultures. Hyphae from the multispore colonies of Mitrula are uninucleate and are probably haploid. However, the possibility that they are diploid cannot be ruled out, although this would be very unusual for ascomycetes. The bicellular, brown conidia that are released from club-shaped conidia chains in Mitrula elegans are unknown in other aquatic hyphomycetes in the Helotiales. Dark aerial and septate conidia (Anavirga dendromorpha Descals & Sutton) were reported for Vibrissea flavovirens (Hamad and Webster, 1988
). Conidia from aquatic species Hymenoscyphus malawiensis P.J. Fisher & Spooner are somewhat similar to that of M. elegans, but conidia of H. malawiensis are hyaline and multicellular (Fisher and Spooner, 1987
).
Spores of some aquatic hyphomycetes are usually branched or sigmoid, which is presumably an adaptation to aquatic habitats in flowing water. The conidia of M. elegans do not appear to be able to float in the water and may be an adaptation to aquatic habitats with standing or slow-moving water, such as vernal pools. Vernal pools provide unique systems for the wetland species, particularly aquatic plants (Keeley and Zedler, 1998
; Weyembergh et al., 2004
). Because the major carbon source for streams and forested pools is not in-pool photosynthesis but leaf litter from the surrounding forest (Allan, 1995
). Thus, aero-aquatic fungi such as Mitrula species are important in carbon flow in vernal pools and similar habitats.
Teleomorphs in the Helotiales, such as Vibrissea flavovirens (Pers.) Korf & J.R. Dixon, Mollisia gigantea P.J. Fisher & J. Webster, and Hydrocina chaetocladia Scheuer, were found in association with phialophora-like anamorphs (microconidial synanamorph), which may have a spermatial function and can fertilize monokaryotic or dikaryotic hyphae produced by other conidia or sclerotia (Fisher and Webster, 1983
; Hamad and Webster, 1988
; Webster et al., 1991
; Harrington and McNew, 2003
). No microconidial synanamorph was observed from the cultures of M. elegans.
Position of Mitrula in the Helotiales
Phylogenetic analyses indicate that Mitrula is a member of the Helotiales-Rhytismatales clade and may be closely related to the small, brightly colored, saprotrophic, and aquatic or hygrophilous members in the Helotiaceae. The precise placement of Mitrula requires a reclassification of the Helotiaceae and the Helotiales, which needs to include many more taxa.
Both parsimony and likelihood analyses suggested that the Helotiales-Rhytismatales is monophyletic and that there were multiple origins of aero-aquatic habitat in the Helotiales (Fig. 4). The BSA clade, which is recognized without strong support, includes aero-aquatic species of Vibrissea, Hydrocina, Cudoniella, and Ombrophila and, in addition, Ascocoryne, Bisporella, Hyaloscypha, and Hymenoscyphus species, which are associated with highly humid habitats. Various anamorphs are known in the BSA clade, but the anamorphs are not consistent with the phylogeny of related teleomorphs (Marvanová, 1997
). More taxa and molecular data from additional loci are necessary to fully understand the evolution of these fungi.
In the HLA parsimony tree, Mitrula species share a clade with other two aquatic genera, Vibrissea and Hydrocina. Vibrissea species are found on substrate submerged in quickly moving streams, and produce filiform (thread-like) ascospores up to 470 µm in length (Sánchez and Korf, 1966
; Korf, 1990
; Iturriaga, 1997
). Hydrocina chaetocladia produces minute, cup-shaped fruitbodies on submerged twigs and produces conidia of Tricladium chaetocladium and a microconidial synanamorph (Webster et al., 1991
).
The Helotiaceae is not a monophyletic group based on our results. Presently, the Helotiaceae is a large family with about 100 genera, more taxa should be included for clarifying the phylogenetic structure in this family, which should be restricted to a much narrower sense as suggested by Kirk et al. (2001)
. Previous hypotheses of relationships between Mitrula species and either members of the Geoglossaceae or species of the Sclerotiniaceae (Korf, 1973
; Verkley, 1994
) were rejected. The Geoglossaceae occurs as a basal branch of inoperculate ascomycetes, and it should be excluded from the Helotiales. The Sclerotiniaceae, Rutstroemiaceae, Hemiphacidiaceae, and Heyderia are closely related, and there are some Mitrula-like fungi in this clade. Bryophilous Bryoglossum is closely related to Lachnum rather than Mitrula. No data of the aquatic family Loramycetaceae in the Helotiales were available.
Species of Mitrula
This study partially supports the morphological species delimitation of Mitrula in North America by Redhead (1977)
. The isolates representing M. borealis, M. lunulatospora, and M. elegans, M. brevispora from Asia, and M. paludosa from Europe, represented more than six lineages, whereas Redhead (1977)
divided Mitrula into four species.
Two isolates of Mitrula borealis from North America (including the isotype) and the new species M. brevispora from China, formed a basal clade among Mitrula species. Mitrula borealis has also been reported from Europe (Redhead, 1977
), but no recent collections are available. The Chinese collection is the first record of a subalpine species of Mitrula in Asia. Despite the small genetic divergence (4 bp) between M. borealis and M. brevispora based on partial lsu-rDNA and ITS sequences, the latter is significantly different morphologically from M. borealis. Mitrula brevispora occurs on mossy substrates, which are not submerged in water, and produces ascocarps that are more cylindrical rather than clavate, with much smaller ascospores than those of M. borealis (5.5–10.1 x 2.4–3.1 µm vs. 10.5–18 x 2.5–5 µm). Thus, the recognition of M. brevispora as a distinct species is warranted, although it is a close relative of M. borealis, based on rDNA sequences.
Mitrula elegans represented by seven isolates from eastern and western North America is not monophyletic based on our analyses. Isolate G47 collected from eastern North America fits the description of M. elegans morphology well, but it is not closely related to the other isolates in the clade that includes M. elegans isolates. Given the conservation of lsu-rDNA and ITS regions observed in the M. borealis–M. brevispora clade, the variation among M. elegans isolates is fairly high. Two European collections of M. paludosa showed a single base-pair difference to North American M. elegans in the ITS region. Mitrula paludosa has been reported from Asia as well (Imai, 1941
; Teng, 1996
), but no recent Asian collections are available to verify this occurrence with molecular methods. The close relationship between M. elegans in North America and M. paludosa in Europe suggested by previous morphological studies is supported by the molecular data.
Conclusions
In vitro cultural studies reveal that the life cycle of the aero-aquatic Mitrula elegans involves production of brown bicellular conidia, which may be an adaptation to habitats with standing or slow-moving water and vernal pools. Phylogenetic analyses based on three rDNA regions (lsu, ssu, and 5.8S) reject previously proposed relationships between Mitrula and Sclerotiniaceae or Geoglossaceae and imply multiple origins of the aquatic habitat in the Helotiales. More than six lineages can be recognized in Mitrula based on molecular data and morphological characters. Low variation is evident at both the ITS and lsu rDNA loci among species with significant morphological differences, but comparatively high molecular divergences are obvious at the same loci within the morphological species M. elegans. Further morphological, ecological and molecular studies are required to clarify the phylogenetic relationships both within Mitrula and within the Helotiales and to reconstruct the evolution of Mitrula in the aero-aquatic habit.
APPENDIX 1
Taxa, vouchers, and GenBank accession numbers for all sequences analyzed with a dash (—) for missing information or unpublished data. Sequences of isolates (GenBank) were downloaded from GenBank. Sequences of isolates (AFTOL: Assembling Fungal Tree Of Life) were downloaded from the AFTOL database (http://aftol.biology.duke.edu/pub/ dataIndex). After each Mitrula species is the tracking number (starting with the letter G) that is used in the lower-level analyses (Fig. 5).
FOOTNOTES
1 This study was supported by National Science Foundation grants DEB-0228657 to D. S. H. and DEB-0128925 to D. S. H. and M. B., and a National Geographic Society grant 7192-02 to Z. W. and D. S. H. The authors thank two anonymous reviewers for their critical review of the manuscript. We thank S. Redhead, P. Johnston, J. Spatafora, D. Mitchel, D. Mitchell, D. Pfister, H. Knudsen, A. Holst-Jensen, Ch. Scheuer, P. B. Matheny, T. Schumacher, D. Hewitt, A. Wilson, J. Slot, M. Takahasi, S. O. Khattab, and the curators of DAOM, PDD, UWH, CUP, FH, HMAS, HKAS, OSC, NIFG, MBH, and WTU for providing collections. 
2 Author for correspondence (e-mail: zwang{at}clarku.edu
) additional address: Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang, 110016, China
LITERATURE CITED
Allan J. D. 1995 Stream ecology, structure and function of running water. Chapman & Hall, London, UK
Abdullah S. K. E. Descals J. Webster 1981 Teleomorphs of three aquatic hyphomycetes. Transactions of the British Mycological Society 77: 475-483[ISI]
Blanchette R. A. 1994 Degradation of the lignocellulose complex in wood. Canadian Journal of Botany 73: (Supplement 1) S999-S1010
Benkert D. 1983 Bemerkenswerte Ascomyceten der DDR VI. Die weißsporigen Geoglossaceen. Gleditschia 10: 141-171
Dennis R. W. G. 1968 British ascomycetes. Cramer, Lehre, Germany
Descals E. P. J. Fisher J. Webster 1984 The Hymenoscyphus teleomorph of Geniculospora grandis. Transactions of the British Mycological Society 83: 541-546[ISI]
Digby S. R. D. Goos 1987 Morphology, development and taxonomy of Loramyces. Mycologia 79: 821-831[CrossRef][ISI]
Eriksson O. E. H.-O. Baral R. S. Currah K. Hansen C. P. Kurtzman G. Rambold T. Laessøe [eds.] 2004 Outline of Ascomycota: 2004. Myconet 10: 1-99
Fisher P. J. B. Spooner 1987 Two new ascomycetes from Malawi. Transactions of the British Mycological Society 88: 47-54[ISI]
Fisher P. J. J. Webster 1983 The teleomorphs of Helicodendron giganteum and H. paradoxum. Transactions of the British Mycological Society 81: 656-659[ISI]
Gernandt D. S. F. J. Camacho J. K. Stone 1997 Meria laricis, an anamoph of Rhabdocline. Mycologia 89: 735-744[CrossRef][ISI]
Gernandt D. S. J. L. Platt J. K. Stone J. W. Spatafora A. Holst-Jensen R. C. Hamelin L. M. Kohn 2001 Phylogenetics of Helotiales and Rhytismatales based on partial small subunit nuclear ribosomal DNA sequences. Mycologia 93: 915-933[CrossRef][ISI]
Goldman N. J. P. Anderson A. G. Rodrigo 2000 Likelihood-based test of topologies in phylogenetics. Systematic Biology 49: 652-670[CrossRef][ISI][Medline]
Goodwin S. B. 2002 The barley scald pathogen Rhynchosporium secalis is closely related to the discomycetes Tapesia and Pyrenopeziza. Mycological Research 106: 645-654[CrossRef][ISI]
Goos R. D. S. K. Abdullah P. J. Fisher J. Webster 1986 The anamorph genus Helicoon. Transactions of the British Mycological Society 87: 115-122[ISI]
Hamad S. R. J. Webster 1988 Anavirga dendromorpha, anamorph of Apostemidium torrenticola. Sydowia 40: 60-64
Harrington T. J. 1997 Aquatic hyphomycetes of 21 rivers in southern Ireland. Biology and Enviroment: Proceedings of the Royal Irish Academy 97: 139-148
Harrington T. C. D. L. McNew 2003 Phylogenetic analysis places the Phialophora-like anamorph genus Cadophora in the Helotiales. Mycotaxon 87: 141-152[ISI]
Hawksworth D. L. P. M. Kirk B. C. Sutton D. N. Pegler 1995 Ainsworth and Bisby's dictionary of the Fungi, 8th ed. CAB International, Wallingford, UK
Hibbett D. S. 1996 Phylogenetic evidence for horizontal transmission of group I introns in the nuclear ribosomal DNA of mushroom-forming fungi. Molecular Biology and Evolution 13: 903-917[Abstract]
Huelsenbeck J. P. F. Ronquist 2001 MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754-755
Imai S. 1941 Geoglossaceae Japoniae. Journal of the Faculty of Agriculture, Hokkaido Imperial University 45: 155-264
Ingold T. 1942 Aquatic hyphomycetes of decaying alder leaves. Transactions of the British Mycological Society 25: 339-417
Iturriaga T. 1997 Vibrissea pfisteri, a new species with an unusual ecology. Mycotaxon 51: 215-221
Kankainen E. 1969 On the structure, ecology and distribution of the species of Mitrula s. lat. (Ascomycetes, Geoglossaceae). Karstenia 9: 23-34
Keekey J. E. P. H. Zedler 1998 Characterization and global distribution of vernal pools. In C. W. Witham, E. T. Bauder, D. Belk, W. R. Ferren Jr., and R. Ornduff [eds.], Ecology, conservation, and management of vernal pool ecosystems, proceedings from 1996 conference, 1– 14. California Native Plant Society, Sacramento, California, USA
Kirk P. M. P. F. Cannon J. C. David J. A. Stalpers 2001 Ainsworth and Bisby's dictionary of the Fungi, 9th ed. CAB International, Wallingford, UK
Kirk T. K. R. L. Farrel 1987 Enzymatic "combustion": the microbial degradation of ligin. Annual Review of Microbiology 41: 465-505[ISI][Medline]
Korf R. P. 1973 Discomycetes and Tuberales. In G. C. Ainsworth, F. K. Sparrow, and A. S. Sussman [eds.], The Fungi: an advanced treatise, vol. 4A, 249–319. Academic Press, New York, New York, USA
Korf R. P. 1990 Discomycete systematics today: a look at some unanswered questions in a group of unitunicate ascomycetes. Mycosystema 3: 19-27
Landvik S. 1996 Phylogenetic rDNA studies of Discomycetes (Ascomycota). Ph.D. dissertation, Department of Ecological Botany, Umeå University, Umeå, Sweden
Maas Geesteranus R. A. 1964 On some white-spored Geoglossaceae. Persoonia 3: 81-96
Marvanova L. 1997 Freshwater hyphomycetes: a survey with remarks on tropical taxa. In K. K. Janardhanan, C. Rajendran, K. Natarajan, and D. L. Hawksworth [eds.], Tropical mycology, 169–226. Science Publishers, Enfield, New Hampshire, USA
Marvanova L. E. Descals 1996 Hyphomycetes from streams: new taxa and new combination. Mycotaxon 50: 455-469
Nikolcheva L. G. A. M. Cockshutt F. Bärlocher 2003 Determining diversity of freshwater fungi on decaying leaves: comparison of traditional and molecular approaches. Applied and Environmental Microbiology 69: 2548-2554
Nylander J. A. A. 2004 MrModeltest, version 2.0. Program distributed by the author. Evolutionary Biology Centre, Uppsala University, Uppsala, Sweden, website: http://www.ebc.uu.se/systzoo/staff/nylander.html
Pfister D. H. 1997 Castor, Pollux and life histories of fungi. Mycologia 89: 1-23[CrossRef][ISI]
Pfister D. H. J. W. Kimbrough 2001 Discomycetes. In D. J. McLaughlin, E. G. McLaughlin, and P. A. Lemke [eds.], The Mycota VII, part A, Systematics and evolution, 257–281. Springer-Verlag, Berlin, Germany
Pointing S. B. A. L. Pelling G. J. D. Smith K. D. Hyde C. A. Reddy 2004 Screening of basidiomycetes and xylariaceous fungi for lignin peroxidase and laccase gene-specific sequences. Mycological Research 109: 115-124[CrossRef][ISI]
Platt J. L. 2000 Lichens, earth tongues, and endophytes: evolutionary patterns inferred from phylogenetic analyses of multiple loci. Ph.D. dissertation, Oregon State University, Corvallis, Oregon, USA
Posada D. K. A. Crandall 1998 Modeltest: testing the model of DNA substitution. Bioinformatics 14: 817-818
Redhead S. A. 1977 The genus Mitrula in North America. Canadian Journal of Botany 55: 307-325[ISI]
Sanchez A. R. P. Korf 1966 The genus Vibrissea, and the generic names Leptosporium, Apostremium, Apostemidium, Gorgoniceps and Ophiogloea. Mycologia 58: 722-737[CrossRef][ISI]
Schumacher T. A. Holst-Jensen 1997 A synopsis of the genus Scleromitrula (=Verpatinia) (Asomycotina: Helotiales: Sclerotiniaceae). Mycoscience 38: 55-69[CrossRef]
Seaver F. J. 1951 The North American cup-fungi (In-operculates). Seaver, New York, New York, USA
Shimodaira H. M. Hasegawa 1999 Multiple comparisons of log-likelihoods with applicati
