Mycologia, 96(6), 2004, pp. 1339–1351. q 2004 by The Mycological Society of America, Lawrence, KS 66044-8897
Phylogenetic relationships among Rhizophydium isolates from North America and Australia Peter M. Letcher 1 Martha J. Powell James G. Chambers Wallace E. Holznagel
zoospore ultrastructure among isolates in the six subclades and an unresolved polytomy group within the Rhizophydium clade, thus evaluating the application of zoospore ultrastructure for lower level taxonomic decisions. All isolates were examined by transmission electron microscopy, and four types of zoospores were found. Thus, within the well-supported Rhizophydium clade, zoospore ultrastructure appeared divergent. Because similar zoospore types also were found in two distinct subclades, zoospore structure might be interpreted superficially as convergent. However, unresolved polytomys indicated molecular divergence among these taxa and the need for a more diverse taxa and gene sampling to resolve relationships. One of the zoospore types characterized represents the most simplified form of zoospore described so far in the Chytridiales. The range in molecular secondary structure composition and in zoospore morphology suggested that isolates we provisionally placed in Rhizophydium actually represent multiple genera. Key words: Chytridiales, Chytridiomycota, large subunit, ultrastructure, zoospore
Department of Biological Sciences, University of Alabama, Tuscaloosa, Alabama 35487
Abstract: The order Chytridiales is the largest and most diverse of five orders in phylum Chytridiomycota. Rhizophydium is one of two genera in the Chytridiales with more than 220 described species. Because thallus characters used in classical descriptions of Rhizophydium species often intergrade into other species, as well as other genera, species distinctions frequently are unclear. Species often are delimited solely on substrate or host; many described species consequently may be synonymous. On the other hand, because the thallus is relatively simple morphologically similar forms actually may be genetically distinct. As a beginning for the revision of the genus Rhizophydium, this study used molecular and ultrastructural analyses to characterize cultures identified as Rhizophydium species. A broad geographic sampling of Rhizophydium-like organisms from North American and Australian soils was studied as a foundation for enhanced identification of soil chytrids. The first objective was to ascertain the genetic variability of Rhizophydium isolates with spherical to angular sporangia and multiple discharge pores, using nuclear large subunit rRNA gene sequence analysis. Sequences of 45 isolates of Chytridiales, including 29 isolates in the Rhizophydium clade were analyzed. Alignment based on LSU rRNA secondary structure revealed a similar reduced stem and loop structure in the C1p3 helix region that distinguished morphologically similar Rhizophydium clade members from other members of the Chytridiales. In our parsimony analysis, the Chytriomyces clade was sister of the Nowakowskiella, Lacustromyces and Rhizophydium clades. Six subclades within the Rhizophydium clade were resolved. Several closely related isolates appeared geographically widespread because North American and Australian isolates were found together in three of the six subclades. The second objective was to sample
INTRODUCTION
Chytrid fungi (phylum Chytridiomycota) are abundant in soils where they decay refractory materials such as cellulose, chitin and keratin, and parasitize plants and other fungi (Sparrow 1960, Powell 1993). Our understanding of soil chytrids within the structure of the total microbial soil community, however, is limited (Letcher and Powell 2001). In a study of the association of chytrids with mosses in forest soils, differences in distribution at the microhabitat level were detected (Letcher and Powell 2002). One limitation in understanding the role of chytrids in soil communities, however, is the difficulty in distinguishing genera and species of many common soil chytrids (Letcher and Powell 2001). Thallus structural features used to discriminate genera such as Rhizophydium, Rhizidium, Phlyctochytrium and Rhizophlyctis, including presence or absence of an apophysis and other rhizoidal and sporangial features, often intergrade (Karling 1932; Miller 1968, 1976; Barr 1969; Willoughby 1971; Powell and Koch 1977; Sparrow and Lange 1977). Moreover, speciation often is based
Accepted for publication June 14, 2004. 1 Corresponding author. E-mail:
[email protected]
1339
1340
MYCOLOGIA
solely on specific substrate or host, without examination of nutritional ranges; consequently, many described species may be synonymous (Sparrow 1960). On the other hand, because the thallus is relatively simple and potential responses to selective pressure are limited, thallus morphology may be convergent (Powell and Koch 1977). Thus, isolates from similar habitats with similar morphologies actually may be genetically distinct. This study is the beginning of a long-term study to revise the genus Rhizophydium, to determine its boundaries and to determine genetic variability of morphologically similar isolates. Schenk (1858) proposed the genus, separating inoperculate members from the genus Chytridium, and Rabenhorst (1868) described the genus. Clements and Shear (1931) designated the algal parasite Rhizophydium globosum as the type. Although the thallus of Rhizophydium is relatively simple, consisting of a sporangium and a single tapering rhizoid that often branches, prolific descriptions of new species have expanded the morphological limits of the genus resulting in intergrades into other genera (i.e., Phlyctochytrium). The genus Rhizophydium now includes five sections and more than 225 species, but its generic boundaries are far from clear (Sparrow 1960, Karling 1977, Longcore 1996). As Sparrow (1960) noted, ‘‘This is the largest and most complex genus of the chytrids. . . . It is well realized that the present treatment of this most difficult genus is far from adequate.’’ Recent comparisons of molecular sequences of genes have helped resolve relationships among fungi and the phylum Chytridiomycota, with chytrids consistently basal within the fungi (Bowman et al 1992, Bruns et al 1992, Li and Heath 1992, Wainwright et al 1993, Paquin et al 1997, Jensen et al 1998, Tehler et al 2000, Tehler et al 2003). Only two broad spectrum molecular studies of chytrids have been conducted, one using the nuclear small-subunit (SSU) ribosomal RNA gene ( James et al 2000) and another using combined SSU and nuclear large-subunit (LSU) ribosomal RNA gene data aligned based on rRNA secondary structure (Chambers 2003). Both studies identified four monophyletic clades (Chytriomyces [5 Chytridium], Rhizophydium, Nowakowskiella and Lacustromyces) within the order Chytridiales. The clade that James et al (2000) designated as the Chytridium clade is now the Chytriomyces clade because the clade contains no authentic species of Chytridium. Chytridium confervae included in the clade ( James et al 2000) was transferred to Chytriomyces (Batko 1975). The first objective of our study was to use molecular analysis to resolve relationships among a broad geographical sampling of organisms difficult to dis-
tinguish because of thallus morphological plasticity and intergrades. Rhizophydium isolates from soils from two distant regions of the world, North America and Australia, are analyzed to explore phylogenetic implications of speciation and provenance in disparate environments. Such analyses can lead to better recognition of species and genera of soil chytrids. In this study we generated a phylogenetic hypothesis for the genus Rhizophydium based on analysis of LSU rRNA secondary structure alignment. Regions of the LSU evolve more rapidly than the SSU and consequently provide an opportunity to assess phylogenetic relationships among more recently evolved taxa (Hillis and Dixon 1991). LSU rRNA has been shown to have adequate variation among related chytrids to resolve close relationships (Chambers 2003). To evaluate the generic boundaries of Rhizophydium and to provide structure to the cladograms, among 45 isolates included in the analysis were 29 isolates in the Rhizophydium clade, 11 isolates in the Chytriomyces clade, two isolates in the Nowakowskiella clade, two isolates in the Lacustromyces clade and Phlyctochytrium planicorne. Phlyctochytrium is included in a molecular phylogenetic analysis for the first time (Letcher 2003, Letcher and Powell 2004b). Rhizophlyctis harderi was included in the sampling because previous ultrastructural studies (Powell and Roychoudhury 1992, Roychoudhury and Powell 1992) indicated that its placement in Rhizophlyctis is questionable and its various thallus forms include a form morphologically similar to Rhizophydium. However, several identified isolates included in this analysis, as well as Rhizophlyctis harderi and Rhizophydium brooksianum, have thallus morphology distinctive from the type of Rhizophydium. This is the first molecular analysis of a large sampling of chytrids within a single genus. The second objective is to characterize zoospore ultrastructure for subclades among the isolates in the Rhizophydium clade. Earlier studies (Barr and Hartmann 1976, Barr and Hadland-Hartmann 1978b, Barr 1980) attempted to use zoospore ultrastructure to produce a unified view of the genus Rhizophydium (Group III type zoospore, Barr 1980), but isolates that were classified as Rhizophydium based on thallus structure exhibited a range of zoospore types (Barr and Hadland-Hartmann 1978b). Because features of zoospore ultrastructure appear to be stable, these results suggest that the genus is actually an assemblage of several genera. Thus, in this study zoospore ultrastructure was examined for all isolates among the subclades within the Rhizophydium clade to test both the consistency of the Group III type zoospore (Barr 1980) and the reliability of zoospore ultrastructure for lower level taxonomic decisions.
LETCHER
ET AL:
PHYLOGENY
MATERIALS AND METHODS
Taxonomic sampling.—Partial LSU rRNA gene sequences from 45 taxa in the order Chytridiales (TABLE I) were elucidated. The outgroup taxa were Monosiga brevicollis (Choanoflagellida) and Ichthyophonus hoferi (Mesomycetozoa) (Chambers 2003). DNA was extracted from pure cultures obtained from chytrid collections maintained at the American Type Culture Collection, Canadian Collection of Fungal Cultures, The University of Alabama, and University of Maine. LSU rRNA gene sequences for the outgroup taxa were obtained from GenBank (http://www.ncbi.nlm.nih. gov/). Sample preparation.—Cultures were grown on these media: PmTG, mPmTG, ½ CM1, ¼ YpSs and ½ YpSs (Barr 1987, Longcore 1995), and DNA isolation was accomplished using an established protocol (Letcher and Powell 2004a). Molecular data analysis.—Sequencher 3.0 (GeneCodes) was used to assemble contiguous sequences. LSU rRNA gene sequences obtained were aligned automatically using Clustal X 1.81 (Thompson et al 1997) and manually using BioEdit (Hall 1998) based on the secondary structure model of Saccharomyces cerevisiae (Gutell et al 1994, van de Peer et al 1997), GenBank U53879, available online at http:// www.rna.icmb.utexas.edu. The alignment has been submitted to TreeBASE (http://www.treebase.org/treebase/). Two phylogenetic analyses were conducted. Phylogenetic trees were constructed using PAUP* 4.0b10 (Swofford 2001) and PAUPRat (Sikes and Lewis 2001) to accommodate computer memory restraints and time constraints. Search strategy involved five ratchet runs of 200 iterations each (200 iterative trees plus one starting tree each run, for 1005 collectively) to increase the probability of hitting the shortest tree. Maximum parsimony (MP) was the criterion, and initial trees were constructed with TBR branch swapping. Uninformative characters were deleted, and a bootstrap analysis of informative characters used 1000 replicates with random taxon addition sequence and TBR branch swapping. In addition, a neighbor joining (NJ) tree was constructed using the HKY85 model. Fixation for electron microscopy.—Of the 45 isolates in the Chytridiales for which LSU rRNA sequences were elucidated, zoospores from 27 isolates in the Rhizophydium clade isolates were examined by transmission electron microscopy. Fixation of zoospores followed an established protocol (Letcher and Powell 2004a). Ultrastructural analysis.—The correspondence between molecular subclades of taxa in an inferred phylogeny and zoospore ultrastructural features was assessed. Among 29 isolates in the Rhizophydium clade, character states of the MLC cisterna (simple or fenestrated), a kinetosome associated structure (KAS: an electron-opaque spur or shield), position of lipid globules and number of microtubular roots were superimposed on terminal taxa. RESULTS
Sequence analysis.—From 45 chytrid taxa (TABLE I), LSU rRNA gene sequences determined in this study
OF
RHIZOPHYDIUM
1341
included 1441 characters, of which 459 were parsimony informative. Of 1005 total trees, 913 MP trees were 1712 steps in length, with a consistency index (CI) of 0.596 and a retention index (RI) of 0.668. The inferred parsimony phylogeny (FIG. 1) shows the relationship among the Chytriomyces clade, Nowakowskiella clade, Lacustromyces clade, Rhizophydium clade and Phlyctochytrium planicorne. The Chytriomyces clade was sister of all other clades and taxa included in the study. The four monophyletic clades were well supported (Chytriomyces 5 92%, Lacustromyces 5 93%, Nowakowskiella 5 94%, Rhizophydium 5 99%) as were deeper nodes. In the strict consensus MP tree, three Kappamyces isolates (PL 74, PL 75 and PL 98; see Letcher and Powell 2004a) were basal to the other 26 isolates in the Rhizophydium clade (arrowed A, FIG. 1). A NJ tree (FIG. 2) inferring amount of evolutionary change among isolates in the Rhizophydium clade indicated that the Kappamyces isolates were sister of, but highly divergent from, 24 isolates that formed the core of the Rhizophydium clade (arrowed B, FIG. 1). Alignment of sequences based on LSU rRNA secondary structure revealed an informative stem and loop structure in the helix C1p3 region of the C domain (Ben Ali et al 1999), located downstream from the 59 end at bases 808–953 in the aligned data. The structure was extensive in Rhizophydium brooksianum, the Lacustromyces clade, Phlyctochytrium planicorne, all isolates of the Chytriomyces clade and outgroup taxa. In comparison, it was reduced in all other provisional Rhizophydium species, Rhizophlyctis harderi, and the Nowakowskiella clade. Within the Rhizophydium clade, six subclades (I– VI) were delineated (FIG. 3). Subclade I consisted of three isolates of Kappamyces, had 100% support and was basal to subclade II. Subclade II consisted of Rhizophydium brooksianum and Rhizophlyctis harderi. These two isolates grouped with 97% support in the MP analysis (FIG. 1) and were sister of the grouping of 24 isolates that comprised the core of the Rhizophydium clade. In the NJ tree (FIG. 2), these two isolates also grouped, but the group was sister of the remaining 27 isolates. In the core group of the Rhizophydium clade (composed of 24 isolates), four subclades had support values of 59–90% (bootstrap support: III 5 90%, IV 5 59%, V 5 70%, VI 5 72%), and five isolates were in an unresolved polytomy. Morphology.—From the NJ analysis (FIG. 2), arrowed B indicates a core of 24 Rhizophydium isolates, most of which are on short branches. These are isolates that are characterized by sporangia that are variable in size, range in shape from spherical to angular, release zoospores from multiple discharge pores, have
1342 TABLE I.
MYCOLOGIA Taxon sampling for LSU rRNA phylogenetic analysis of 45 isolates from four clades in the Order Chytridiales Taxon
Clade
Culture number
Outgroup: Ichthyophonus hoferii Monosiga brevicollis Ingroup: Allochytridium luteum Chytriomyces sp. Chytriomyces clade Chytriomyces confervae Chytriomyces appendiculatus Chytriomyces hyalinus Chytriomyces hyalinus Chytriomyces sp. Chytriomyces sp. Chytriomyces spinosus Kappamyces laurelensis b Kappamyces sp. Kappamyces sp. Karlingiomyces sp. Nowakowskiella elegans Obelidium mucronatum Phlyctochytrium planicorne c Polychytrium aggregatum Rhizophlyctis harderi Rhizophydium brooksianum Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium elyensis Rhizophydium sp. Rhizophydium sp. Rhizophydium macroporosum Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Rhizophydium sp. Siphonaria petersenii a
New South Wales, Australia. See Letcher and Powell 2004a. c See Letcher and Powell 2004b. d Tasmania, Australia. b
Accession #
Origin
AY026370 AY026374 Nowakowskiella Chytriomyces Chytriomyces Chytriomyces Chytriomyces Chytriomyces Chytriomyces Chytriomyces Chytriomyces Chytriomyces Rhizophydium Rhizophydium Rhizophydium Lacustromyces Nowakowskiella Chytriomyces Lacustromyces Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Rhizophydium Chytriomyces
ATCC 60989 JEL 176 JEL 186 ATCC 24931 JEL 165 PL 13 PL AUS 14 JEL 91 PL 06 JEL 59 PL 98 PL 74 PL 75 JEL 93 M 29 JEL 57 JEL 47 JEL 109 ATCC 24053 JEL 136 JEL 281 PL AUS 2 PL AUS 3 PL AUS 6 PL AUS 7 PL AUS 8 PL AUS 9 PL AUS 12 PL AUS 18 PL AUS 21 PL AUS 22 PL AUS 24 PL 01 PL 03 PL 04 PL 05 PL 08 PL 10 PL 11 PL 42 PL 72 PL 73 PL 88 PL 102 JEL 102
AY439066 AY439078 AY439070 AY439074 AY439076 AY439075 AY442956 AY439077 AY439055 AY439073 AY439034 AY439033 AY439053 AY439069 AY439067 AY439071 AY439028 AY439068 AY349087 AY349086 AY439032 AY439044 AY439045 AY439046 AY439047 AY439048 AY439049 AY439035 AY439051 AY439040 AY439030 AY439052 AY439042 AY439041 AY439057 AY439058 AY439059 AY439037 AY439038 AY439056 AY439031 AY439039 AY439036 AY439043 AY439072
Virginia, USA Maine, USA Michigan, USA Ontario, Canada Maine, USA Alabama, USA NSW, AUSa Maine, USA Tennessee, USA Maine, USA Georgia, USA Virginia, USA Virginia, USA Maine, USA Maine, USA Maine, USA Maine, USA Michigan, USA Canada Maine, USA Maine, USA NSW, AUS NSW, AUS NSW, AUS NSW, AUS NSW, AUS NSW, AUS NSW, AUS NSW, AUS NSW, AUS TAS, AUSd TAS, AUS Vermont, USA Virginia, USA Virginia, USA Tennessee, USA Alabama, USA Alabama, USA Florida, USA North Carolina, USA Virginia, USA Alabama, USA Alabama, USA Georgia, USA Maine, USA
LETCHER
ET AL:
PHYLOGENY
OF
RHIZOPHYDIUM
1343
FIG. 1. Partial LSU rRNA gene parsimony analysis of 45 taxa in the order Chytridiales. Strict consensus of 913 trees (L 5 1712 steps, CI 5 0.596, RI 5 0.668). Values are bootstrap values (1000 replicates). ‘‘A’’ at fifth node on backbone indicates point of origin for FIG. 3; ‘‘B’’ indicates point of origin of core Rhizophydium isolates.
a single rhizoidal axis early in development and have a soil habitat. Examples of these morphological forms have been illustrated previously (FIGS. 12–14, Letcher and Powell 2001; FIGS. 38, 40, 41, 45, 46, Letcher et al 2004). Ultrastructural characters.—Zoospore ultrastructure of 27 of the 29 isolates in the Rhizophydium clade was examined. Ultrastructure of Rhizophlyctis harderi (Powell and Roychoudhury 1992, Roychoudhury and Powell 1992) and Rhizophydium brooksianum (Longcore 2004) were available through previous publications. TEM analysis revealed four distinct types of zoospores (FIG. 3, FIG. 4a–d). Zoospores of 12 isolates contained both a fenestrated microbody-lipid globule complex (MLC) cisterna and an electron-opaque kinetosome-associated structure (KAS), a spur (FIG.
4a). Zoospores of 12 isolates contained both a simple MLC cisterna and a KAS, a spur (FIG. 4b). Zoospores of five isolates had two lipid globules, one larger globule located anteriorly or laterally with a simple MLC cisterna, and a second, smaller globule located posteriorly with a fenestrated MLC cisterna (FIG. 4c); those isolates also had a KAS, a spur. Zoospores of three isolates had a simple MLC cisterna and no KAS (FIG. 4d). Ultrastructural features mapped on the phylogeny.—The inferred LSU rRNA gene phylogeny (FIG. 3) shows the relationship among six subclades in the Rhizophydium clade. Character states of these zoospore ultrastructural characters are (i) MLC cisterna, simple or fenestrated; (ii) KAS, spur, shield or none; (iii) number of microtubular roots; and (iv) number and
1344
MYCOLOGIA
FIG. 2. Rooted, NJ consensus phylogeny estimated among 45 taxa in the order Chytridiales. Branch lengths are proportional to the inferred amount of evolutionary change. Scale bar represents 0.05 nucleotide substitutions per site. ‘‘A’’ at fifth node on backbone indicates point of origin for FIG. 3; ‘‘B’’ indicates point of origin of core Rhizophydium isolates.
FIG. 3. Portion of MP LSU rRNA gene phylogeny of 29 isolates in the Rhizophydium clade, with ultrastructure characters and character states (a cisterna, either fenestrated or simple; and a KAS—kinetosome associated structure, either a spur, a shield or absent). Tree excerpted from FIG. 1 at point ‘‘A’’.
LETCHER
ET AL:
PHYLOGENY
relative size of lipid globules were mapped against the subclades to infer derived character states. A simple MLC cisterna was present and a KAS was absent in all taxa in subclade I (Kappamyces). A simple MLC cisterna and a spur were present in all taxa in subclades III and V, with the exception of isolate PL AUS 12, which had a fenestrated MLC cisterna. A fenestrated MLC cisterna and spur were present in all taxa examined in subclades IV and VI. All taxa had at minimum one large lipid globule, which in all isolates of subclade I was located in the center of the zoospore and in most other isolates was located laterally. In addition, five isolates (Rhizophlyctis harderi [subclade II], PL AUS 2, PL AUS 6 and PL AUS 9 [in the unresolved polytomy] and JEL 281 [subclade IV]) had a second, smaller lipid globule, which in all cases was partially covered by a fenestrated MLC cisterna and a microtubular root extended from the MLC cisterna to the kinetosome. The proximity of the globules varied within and among the isolates. In isolates PL AUS 2, PL AUS 6 and PL AUS 9, the smaller lipid was proximal to and closely associated with the larger lipid. In Rhizophlyctis harderi and isolate JEL 281, the two globules were variously located, with the larger globule generally located in the anterior portion of the zoospore and the smaller lipid in a lateral-posterior portion. With the exception of three isolates (PL AUS 9 [subclade II], and PL 88 and PL AUS 7, [subclade III]), all isolates having a lipid globule with a fenestrated MLC cisterna also had at least one microtubular root. Rhizophlyctis harderi had possibly three roots, Rhizophydium brooksianum had possibly two roots and PL AUS 2 had possibly three roots. DISCUSSION
Ultrastructural data from previous studies (Powell and Roychoudhury 1992, Roychoudhury and Powell 1992, Longcore 2004) and molecular data from this study suggest that the taxonomic affinities of the two isolates in subclade II, Rhizophlyctis harderi and Rhizophydium brooksianum, are unresolved. Rhizophlyctis harderi has a fenestrated MLC cisterna and a KAS (a shield) over the kinetosome that differs from the KAS (a spur) present in most isolates of the Rhizophydium clade. Rhizophydium brooksianum has a simple cisterna and a spur similar to the KAS, which is present in most of the Rhizophydium clade isolates. The disparity is due possibly to the large distance inferred for Rhizophydium brooksianum. Further analysis of these two organisms is beyond the scope of this project. The objective of this investigation was to explore the boundaries of the complex and difficult genus
OF
RHIZOPHYDIUM
1345
Rhizophydium. Our results revealed subclades within the Rhizophydium clade that can focus more specific ultrastructural analysis and more wide-ranging molecular sequencing. The type of Rhizophydium (R. globosum) was based on an algal parasite in a true aquatic habitat (Rabenhorst 1868). Rhizophydium globosum is a typical ‘‘little round chytrid’’ (Longcore 2004) having a spherical sporangium, multiple discharge papillae and a fairly extensive, branched rhizoidal system arising from a single axis (Sparrow 1960, Karling 1977). Many of the isolates that formed the core Rhizophydium group in the molecular data analyses (arrowed B, FIGS. 1–3) shared basic morphological thallus features of the type, although angular-shaped sporangia and a greater range in number of discharge pores were found among the isolates. Until the type of the genus R. globosum is found and isolated and zoospore ultrastructure and molecular signatures analyzed, the boundaries of the genus Rhizophydium cannot be confidently established. Without characterization of the type beyond morphology, placement of isolates into the genus Rhizophydium has to be considered provisional. Ribosomal RNA LSU gene analysis.—The complex secondary structure folding of rRNA sequences is well documented (Hillis et al 1991, Gutell et al 1994). Because the secondary structure of rRNA determines its functionality and must be conserved to maintain its functionality, it is this conservation of structure that can be used to generate hypotheses about the evolutionary history of these organisms. LSU analysis clearly resolved the Rhizophydium clade from the Chytriomyces clade. In the MP analysis, within the Rhizophydium clade, subclade I, with bootstrap support of 100%, was basal to a well supported core group of 24 isolates. The three isolates that compose subclade I have been delineated as Kappamyces gen. nov. (Letcher and Powell 2004a) because those isolates are clearly outside the boundary of the genus. In our alignment, the relative reduced size of the stem and loop structure in the LSU C1p3 region structure served as a comparative molecular marker in the Chytridiales for members of the Rhizophydium clade. Although the stem and loop structure also was reduced in Rhizophlyctis harderi and the Nowakowskiella clade, other regions of the sequence distinguished those taxa from provisional Rhizophydium isolates. The variation in statistical support for isolates contributing to a polytomy in the core group of isolates suggested that the divergence of these organisms was such that the LSU rRNA data do not resolve lower level taxonomic relationships. It might be possible to further resolve relationships with nuc-ITS15.8S-ITS2 rRNA (Adams et al 2002) and/or EF1alpha
1346
MYCOLOGIA
FIG. 4. Summary diagrams of four zoospores of isolates in the Rhizophydium clade showing variation in ultrastructure morphology. N, nucleus; R, ribosomes; L, lipid globule; sL, small lipid globule; M, mitochondrion; ER, endoplasmic reticulum; FC, fenestrated cisterna; SC, simple cisterna; Mb, microbody; Mt, microtubular root; K, kinetosome; Sp, spiral; F, flagellum; TP, terminal plate; P, prop; FB, fibrillar bridge; NFC, non-flagellated centriole; Vac, vacuole; LVe, large vesicle adjacent to kinetosome; Sve, small vesicle in peripheral cytoplasm. (a) Isolate PL AUS 8, with a fenestrated cisterna, kinetosome-associated
LETCHER
ET AL:
PHYLOGENY
sequence analyses (Kauserud and Schumacher 2003), which have been used successfully to resolve relationships at or below the genus level. However, without having the type species available for analysis, determination of taxonomic boundaries (particularly at the family, genus and species levels) is hypothetical because we do not know ultrastructurally or molecularly what a Rhizophydium is. Range of zoospore ultrastructure among Rhizophydium isolates.—Determining a characteristic zoospore type for the genus Rhizophydium has been difficult because species assigned to the genus exhibit a range of zoospore ultrastructural features. In addition, isolates assigned to other genera sometimes produce zoospores like those of Rhizophydium taxa. This confusion is clearly the result of intergrading thallus morphology that makes generic identification often difficult when based on thallus morphology alone. Barr and Hadland-Hartmann (1978b) examined the ultrastructure of 12 species of Rhizophydium, the result of which was ‘‘the typical Rhizophydium zoospore’’ (Barr and Hadland-Hartmann 1978b), later referred to as the ‘‘Group III-type (Rhizophydium) zoospore’’ (Barr 1980). In a tabulation of ultrastructure features of those species (TABLE 1, Barr and Hadland-Hartmann 1978b), most of the 12 species examined had a fenestrated MLC cisterna, all species with a fenestrated MLC cisterna had laterally orientated kinetosome-associated microtubules, and most of the species lacked a spur as a KAS. All species examined in that study except R. patellarium had kinetosome-associated microtubular roots. Barr and Hadland-Hartmann also estimated the number of mitochondria for each species, which ranged from ‘‘one to 5–10 to many.’’ Amon (1984) described R. littoreum (synonymizing Phlyctochytrium sp. [Kazama 1982], and P. aestuarii Ulken [Lange and Olson 1977]), a chytrid having a fenestrated MLC cisterna, a spur as a KAS and a microtubular root. Among four different zoospore forms found among isolates in the Rhizophlyctis rosea complex, Barr and De´Saulniers (1986) identified one (R. rosea subtype C) that was more characteristic of a Rhizophydium-type of zoospore. Beakes et al (1993) described R. planktonicum Canter emend, but because that species had several ultrastructural features present in members of the Chytriomyces clade, yet absent in the Rhizophydium
OF
RHIZOPHYDIUM
1347
clade, R. planktonicum might not be closely related to other Rhizophydium species whose ultrastructure has been documented. Chen and Chien (1996) described the zoospore of R. macroporosum Karling, which had a fenestrated MLC cisterna, spur as a KAS and microtubular root. Longcore (2004) most recently described R. brooksianum, which had a simple MLCcisterna, a spur and at least one and possibly two microtubular roots. In a morphologically intergrading taxon, McNitt (1974) described the ultrastructure of Phlyctochytrium irregulare Koch, which had a fenestrated MLC cisterna, spur and microtubular root. Our study of Rhizophydium clade zoospores similarly revealed at least four types of zoospores exhibiting different combinations of characters and character states. Because serial section analysis is required to determine number of mitochondria in zoospores (Powell and Roychoudhury 1992, Roychoudhury and Powell 1992), analysis of this character was beyond the scope of our survey. The importance of structures associated with the flagellar apparatus (KAS, microtubular roots, transition region, fibrillar bridge between the kinetosome and nonflagellated centriole) and MLC (cisternae, lipids, microbodies; see Powell and Roychodhury 1992) as sources of characters and character states useful in taxonomic characterization is exemplified in the variation of zoospore types found in Rhizophydium clade isolates with simple thallus morphology. The preponderance of ultrastructural and molecular evidence suggests that the Rhizophydium clade clearly includes multiple genera. Not having the type of the genus in culture, makes the question of the boundary between Rhizophydium and other potential new genera problematic. However, Rhizophydium subclades with thallus structure different from the type and basal or sister of the core group, which exhibit morphological features similar to the type species and establish the potential boundary of the genus. One of the Rhizophydium subclade (I) that in the MP analysis is basal to the core (subclades III, IV, V, VI, polytomy) was distinct in both molecular and zoosporic features, and these differences warranted establishing a new genus, Kappamyces (Letcher and Powell 2004a). Another subclade (II) sister of the core in the MP analysis is difficult to interpret. Because the NJ tree shows that Rhizophydium brooksianum and Rhizophlyc-
← electron-opaque spur and microtubular root; (b) isolate PL 08, with a simple cisterna and a spur; (c) isolate JEL 281, with a large lipid globule associated with a simple cisterna, a small lipid globule with a fenestrated cisterna, a spur and a microtubular root; (d) isolate PL 98, Kappamyces laurelensis, with a simple cisterna, no kinetosome-associated structure and no microtubular root.
1348
MYCOLOGIA
tis harderi are distinctly related based on branch length, their grouping in MP with strong support might be the result of long-branch attraction, which a more diverse genetic sampling of isolates might resolve. Zoospores of R. brooksianum (Longcore 2004) and R. harderi (Powell and Roychoudhury 1992, Roychoudhury and Powell 1992) are different from each other, including differences in KAS and MLC, but do share some character states. Another ultrastructural zoospore feature that might be valuable in systematic comparisons is the structure of the cytoplasm between the kinetosome and ribosomal aggregation. With the exception of the Kappamyces isolates in subclade I, all Rhizophydium clade isolates examined in this study had an electron-transparent, cup-like, vesiculated area proximal and anterior to the kinetosome, into which, if present, an electron-opaque spur protruded. Barr and Hadland-Hartmann (1978a) also described the vesicle-rich zone between the kinetosome base and in-pocketing of the ribosomal aggregation for some species of Rhizophydium (i.e., R. capillaceum, R. subangulosum). Others more recently have shown this region in zoospores of additional Rhizophydium species (Chen and Chien 1996, R. macroporosum; Longcore 2004, R. brooksianum). It might be a more constant feature of Rhizophydium than was realized. The presence or absence of a vesiculated region proximal to the kinetosome, in conjunction with a suite of other ultrastructural features such as type of MLC cisterna, morphology of the KAS, number and morphology of microtubular roots, morphology of the fibrillar bridge that connects the kinetosome with the nonflagellated centriole and presence and location of Golgi complexes, might be useful in subdividing the genus into more manageable taxonomic units. Ultrastructural patterns within subclades.—Subclades in the Rhizophydium clade and suites of ultrastructural features appear to have correlations. However, the more features analyzed, the more nebulous the correlation, indicating subtle variation among taxa within subclades. In subclade I (Kappamyces), all isolates exhibited zoospores with a large, centrally located lipid globule partially covered by a simple MLC cisterna, no KAS, no observed microtubular root, an electronopaque core in both the kinetosome and nonflagellated centriole, and a ring of up to five distinctive vesicles around the kinetosome and nonflagellated centriole (Letcher and Powell 2004a). In subclade III, all isolates had a single large lipid globule with a simple MLC cisterna, and a spur. All isolates in subclades IV and VI had at minimum a single lipid globule with a fenestrated MLC cisterna and a spur. Isolate JEL 281 (subclade IV) was anomalous, having a
second globule partially covered by a simple MLC cisterna. In subclade V, all isolates except PL AUS 12 had a single lipid globule with a simple MLC cisterna and a spur. Isolate PL AUS 12 was anomalous, having a fenestrated cisterna. These anomalies remain unexplained but might be due to inadequate sampling of genetic diversity and might be resolved further as more diverse taxa are added to the analysis. On the other hand, the presence of fenestrae in the MLC cisterna might be a character state that is repeatedly lost or gained. Habitat considerations.—Habitat is potentially a critical consideration because the type of Rhizophydium was aquatic. Of the approximately 225 described species, about 70% are reported from aquatic habitats (as parasites of algae, plants or other fungi, or as saprotrophs of a variety of refractive substrates). Chytrids that are true parasites are extremely difficult to isolate and maintain, because of the obligate nature of the parasite-host relationship, but may represent genetic diversity not manifested in saprotrophic soil chytrids. All of the isolates in this study were saprotrophs from soil and may represent a more homogeneous collection of Rhizophydium taxa than if isolates from aquatic habitats were included and considered at the molecular and ultrastructural levels. Furthermore, in pursuit of questions regarding populations of specific taxa, our isolation of organisms for this study selected for soil chytrids with spherical to angular sporangial morphology, and with the exception of members of Kappamyces and Rhizophlyctis harderi, sporangia of all isolates had multiple discharge pores. The multipored isolates may represent a limited number of prolific and geographically widespread soil taxa. However, this study has shown remarkable genetic complexity within these seemingly simple organisms. In a comparison of zoospore ultrastructure of species of Phlyctochytrium, Barr (1980) identified two distinct types of zoospores representing two different orders. Because one zoospore form was found only for taxa restricted to soils, he segregated these species into new genera and into a new order. However, an examination of zoospore ultrastructure of Rhizophydium isolates from both aquatic and soil habitats did not similarly reveal a correlation between habitat and major form of zoospores (Barr and Hadland-Hartmann 1978b). Our molecular and ultrastructural study further shows that habitat is a divergent character. Biogeographic considerations.—Fifteen chytrids in the Rhizophydium clade in our study were isolated from eastern North American forests, while 11 taxa were isolated from a variety of diverse Australian vegeta-
LETCHER
ET AL:
PHYLOGENY
tion types (subtropical rainforest, wet and dry sclerophyll forests and heath). Many species of Rhizophydium appear to have global distribution (Sparrow 1960), and examination of isolates from extremely distant geographic regions of the world might provide phylogenetic information regarding speciation in disparate environments. Chytrid fungi have a fossil record extending from approximately 400 MaBP (Taylor et al 1992) and may have existed as much as 1000 MaBP (Simon et al 1993). Thus their origins were perhaps 250 000 000 y before the breakup of the continent Pangaea and pandemic distribution of many of these organisms is expected and might explain why North American and Australian isolates were found among three of the six subclades. However, Australia is also an island continent that has been drifting apart from other continents for approximately 80 000 000 y. The unresolved polytomy of our inferred phylogeny contains a preponderance of Australian isolates, which may exhibit genetic diversity accumulated during continental isolation and habitation of vegetation-diverse environments. More extensive sampling from these and other Australian habitats might contribute to resolution of phylogenetic relationships and provenance among unresolved taxa in this study. Phylogenetic relationships.—Barr (1978) hypothesized a phylogeny of the Chytridiales based on thallus developmental patterns and zoospore ultrastructure. In that hypothesis, Barr attempted to evaluate character polarization as ancestral and derived features, positing chytrid evolution as proceeding from simpler to more complex. Monocentric, Rhizophydium-like organisms were ancestors of Chytridium-like organisms, which in turn evolved to polycentric forms such as Nowakowskiella elegans. The LSU rRNA gene phylogeny in this study presents an alternative hypothesis, in which a grouping of organisms (the Rhizophydium, Nowakowskiella and Lacustromyces clades, and P. planicorne) is sister, rather than ancestor, of Chytriomyces-like organisms of the Chytriomyces clade. Zoospore ultrastructure of all members of the Rhizophydium clade studied so far is distinct from that of the Chytridium Group I type zoospore (Barr 1980). Because of the sister status among the Chytriomyces clade and all other taxa in the inferred phylogeny, polarity of characters is unresolved. In the well supported Rhizophydium clade, thallus morphology was conserved, while zoospore ultrastructure was diverse. In the Rhizophydium clade, zoospore structure appeared divergent because more than one type of zoospore was found, which suggested that multiple general comprise the clade. Because a similar type of zoospore (one having a simple cis-
OF
RHIZOPHYDIUM
1349
terna and a spur) was found in two distinct subclades (subclades III and V), zoospore structure might be interpreted superficially as convergent. However, the polytomy in the unresolved portion of the inferred phylogeny indicates dissimilarities among isolates and the need for more diverse taxa and gene sampling to resolve relationships. In addition the KAS shows a range in substructure, and thus all structures labeled as a ‘‘spur’’ may not be the same. Other ultrastructural details in zoospores (for example, variation in morphology of the fibrillar bridge that connects the kinetosome to the nonflagellated centriole) might not have been resolved, and additional ultrastructural evaluations of more taxa might improve resolution. The correlation of character states of ultrastructural features (MLC cisterna, KAS, number of microtubular roots, and number, size and position of lipid globules) with the subclades was an indication of the diversity of this large and cumbersome genus. When combined datasets (ultrastructure and molecular) were applied to a complex genus, the resulting analysis provided a clearer picture of diversity and natural relationships among taxa that are morphologically homogeneous. In the Rhizophydium clade, there was an apparent simplification in zoospore ultrastructure from ultrastructural features that characterize the Chytriomyces clade. In members of the Rhizophydium clade, a fenestrated MLC cisterna is present in many taxa while a simple MLC cisterna is present in others. In most zoospores examined, mitochondria are outside the ER surrounding the ribosomal mass, and no electronopaque plates are adjacent to the kinetosome, no electron-opaque plug is in the base of the flagellum and no paracrystalline inclusion is in the peripheral cytoplasm. In Barr’s (1978) proposed chytrid phylogeny, Rhizophydium patellarium Scholz exhibited a zoospore that was simpler and less complex than the Group I (Chytridium-type), Group II (Chytridium lagenaria-type) or Group III (Rhizophydium-type) zoospores (Barr 1980). Barr considered the zoospore of R. patellarium to be ancestor to the zoospores of other more advanced and more complex Rhizophydium and Chytridium species. In our analysis of ultrastructural features, the zoospores of the three isolates in subclade I were similar to the zoospore of R. patellarium. The small zoospores of R. patellarium, like the three Kappamyces isolates, have a simple MLC cisterna, no KAS, no microtubular root, and a kinetosome contacting the single, large mitochondrion within the ribosomal core. One other chytrid, R. capillaceum, has a zoospore similar in simplicity to R. patellarium but is distinguished by having a fenestrated MLC cisterna and a KAS. The LSU rRNA gene phylogeny infers that a divergence of Rhizophydium-
1350
MYCOLOGIA
like organisms led to subclade I (Kappamyces) in one lineage and the core of the Rhizophydium clade (subclades III–VI and an unresolved polytomy) in another lineage. The strong bootstrap support for Kappamyces and the relative simplicity of the zoospore support this hypothesis of divergence. Rather than being the most ancestral form of zoospore in the Chytridiales as hypothesized by Barr (1978), the zoospores of Kappamyces might represent the most simplified form of zoospore in the Chytridiales (Letcher and Powell 2004a). This study clearly shows the ultrastructural and genetic complexity present among a limited sampling of taxa in the Rhizophydium clade. The amount of genetic variation demonstrated requires additional sampling of geographic, habitat and substrate diversity to resolve phylogenetic relationships in this complex group of species. ACKNOWLEDGMENTS
This paper is based on a dissertation submitted by the first author to the Graduate School of The University of Alabama in partial fulfillment of the requirements for a doctorate degree in Biology, 2003. Completion of this study was supported by the National Science Foundation through PEET Grant DEB-9978094; The University of Alabama, Department of Biological Sciences Aquatic Ecology and Systematics Graduate Enhancement Program; and a scholarship from Alabama Power Company. We express our appreciation to Dr Joyce E. Longcore for providing chytrid cultures and helpful suggestions and to Drs Peter A. McGee and Frank H. Gleason, School of Biological Sciences, University of Sydney, and Mr David Tribe and Mrs Pam O’Sullivan of Sydney, New South Wales, Australia. LITERATURE CITED
Adams GC, Surve-Iyer RS, Iezzoni AF. 2002. Ribosomal DNA sequence divergence and group I introns within the Leucostoma species L. cinctum, L. persoonii, and L. parapersoonii sp. nov., ascomycetes that cause Cytospora canker in fruit trees. Mycologia 94:947–967. Amon JP. 1984. Rhizophydium littoreum: a chytrid from siphonaceous marine algae—an ultrastructural examination. Mycologia 76:132–139. Barr DJS. 1969. Studies on Rhizophydium and Phlyctochytrium (Chytridiales). I. Comparative morphology. Can J Bot 47:991–997. . 1978. Taxonomy and phylogeny of chytrids. BioSys 10:153–165. . 1980. An outline for the reclassification of the Chytridiales, and for a new order, the Spizellomycetales. Can J Bot 58:2380–2394. . 1987. Allochytridium expandens. In: Fuller MS, Jaworski A, eds. Zoosporic fungi in teaching and research. Athens, Georgia: Southeastern Publishing Corp. p 16–17.
, De´saulniers NL. 1986. Four zoospore subtypes in the Rhizophlyctis-Karlingia complex (Chytridiomycetes). Can J Bot 64:561–572. , Hadland-Hartmann VE. 1978a. The flagellar apparatus in the Chytridiales. Can J Bot 56:887–900. , . 1978b. Zoospore ultrastructure in the genus Rhizophydium (Chytridiales). Can J Bot 56:2380– 2404. , Hartmann VE. 1976. Zoospore ultrastructure of three Chytridium species and Rhizoclosmatium globosum. Can J Bot 54:2000–2013. Batko A. 1975. Zarys Hydromikologii. Warsaw: Pan´stwowe Wydawnictwo Naukowe. p 210. Beakes GW, Canter HM, Jaworski GHM. 1993. Sporangium differentiation and zoospore fine-structure of the chytrid Rhizophydium planktonicum, a fungal parasite of Asterionella formosa. Mycol Res 97:1059–1074. Ben Ali A, Wuyts J, De Wachter R, Meyer A, van de Peer Y. 1999. Construction of a variability map for eukaryotic large subunit ribosomal RNA. Nucl Acids Res 27:2825– 2831. Bowman BH, Taylor JW, Brownlee AG, Lee J, Lu SD, White TJ. 1992. Molecular evolution of the fungi: relationships of the Basidiomycetes, Ascomycetes, and Chytridiomycetes. Mol Biol Evol 9:285–296. Bruns TD, Vilgalys R, Barns SM, Gonzalez D, Hibbett DS, Lane DJ, Simon L, Stickel S, Szaro TM, Weisburg WG, Sogin ML. 1992. Evolutionary relationships within the fungi: analyses of nuclear small subunit rRNA sequences. Mol Phylogenet Evol 1:231–241. Chambers JG. 2003. Ribosomal DNA, secondary structure, and phylogenetic relationships among the Chytridiomycota [Doctoral dissertation]. Tuscaloosa, Alabama: The University of Alabama. 116 p. Chen S-F, Chien C-Y. 1996. Morphology and zoospore ultrastructure of Rhizophydium macroporosum (Chytridiales). Taiwania 41:105–112. Clements FE, Shear CL. 1931. The genera of fungi. New York: HW Wilson Co. 496 p. Gutell RR, Larsen N, Woese CR. 1994. Lessons from an evolving rRNA: 16S and 23S rRNA structures from a comparative perspective. Microbiol Rev 58:10–26. Hall TA. 1998. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl Acids Symp Ser 41:95–98. Hillis DM, Dixon MT. 1991. Ribosomal DNA: molecular evolution and phylogenetic inference. Quarterly Rev Biol 66:411–453. James TY, Porter D, Leander CA, Vilgalys R, Longcore JE. 2000. Molecular phylogenetics of the Chytridiomycota supports the utility of ultrastructural data in chytrid systematics. Can J Bot 78:336–350. Jensen AB, Gargas A, Eilenberg J, Rosendahl S. 1998. Relationships of the insect-pathogenic Order Entomophthorales (Zygomycota, Fungi) based on phylogenetic analyses of nuclear small subunit ribosomal DNA sequences (SSU rRNA). Fun Gen Biol 24:325–334. Karling JS. 1932. Studies in the Chytridiales. VII. The organization of the chytrid thallus. Am J Bot 19:41–74.
LETCHER
ET AL:
PHYLOGENY
. 1977. Chytridiomycetarum Iconographia. Monticello, New York: Lubrecht and Cramer. 414 p. Kauserud H, Schumacher T. 2003. Regional and local population structure of the pioneer wood-decay fungus Trichaptum abietinum. Mycologia 95:416–425. Kazama FY. 1982. Ultrastructure and phototaxis of the zoospores of Phlyctochytrium sp. an estuarine chytrid. J Gen Microbiol 71:555–566. Lange L, Olson LW. 1977. The zoospore of Phlyctochytrium aestuarii. Protoplasma 93:27–43. Letcher PM. 2003. Systematic analysis of molecular and ultrastructural characters among two clades of zoosporic fungi [Doctoral dissertation]. Tuscaloosa, Alabama: The University of Alabama. 149 p. , McGee PA, Powell MJ. 2004. Zoosporic fungi from soils of New South Wales. Australasian Mycologist 22: 99–115. , Powell MJ. 2001. Distribution of zoosporic fungi in forest soils of the Blue Ridge and Appalachian Mountains of Virginia. Mycologia 93:1029–1041. , . 2002. Frequency and distribution patterns of zoosporic fungi from moss-covered and exposed forest soils. Mycologia 94:761–771. , . 2004a. Kappamyces, a new genus in the Chytridiales (Chytridiomycota). Nova Hedwigia (in press). , . 2004b. Phylogenetic position of Phlyctochytrium planicorne (Chytridiales, Chytridiomycota) based on zoospore ultrastructure and partial nuclear LSU rRNA gene sequence analysis. Nova Hedwigia (in press). Li J, Heath IB. 1992. The phylogenetic relationships of the anaerobic chytridiomycetous gut fungi (Neocallimasticaceae) and the Chytridiomycota. I. Cladistic analysis of rRNA sequences. Can J Bot 70:1738–1745. Longcore JE. 1995. Morphology and zoospore ultrastructure of Entophlyctis luteolus sp nov (Chytridiales): implications for chytrid taxonomy. Mycologia 87:25–33. . 1996. Chytridiomycete taxonomy since 1960. Mycotaxon 60:149–174. . 2004. Rhizophydium brooksianum sp nov, a multipored chytrid from soil. Mycologia 96:162–171. McNitt R. 1974. Ultrastructure of Phlyctochytrium irregulare zoospores. Cytobiologie 9:297–320. Miller CE. 1968. Observations concerning taxonomic characteristics in chytridiaceous fungi. J Elisha Mitchell Sci Soc 84:100–107. . 1976. Substrate influenced morphological variations and taxonomic problems in freshwater, posteriorly uniflagellate phycomycetes. In: Jones EBG, ed. Recent advances in aquatic mycology. London: Elek Science. p 459–487. Paquin B, Laforest MJ, Forget L, Roewer I, Wang Z, Longcore JE, Lang BF. 1997. The fungal mitochondrial genome project: evolution of fungal mitochondrial genomes and their gene expression. Curr Genet 31:380– 395. Powell MJ. 1993. Looking at mycology with a Janus face: a
OF
RHIZOPHYDIUM
1351
glimpse of Chytridiomycetes in the environment. Mycologia 85:1–20. , Koch WJ. 1977. Morphological variations in a new species of Entophlyctis. I. The species concept. Can J Bot 55:1668–1685. , Roychoudhury S. 1992. Ultrastructural organization of Rhizophlyctis harderi zoospores and redefinition of the type I microbody-lipid globule complex. Can J Bot 70:750–761. Rabenhorst L. 1868. Flora Europaea algarum aquae dulcis et submarinae. Vol. 3. Leipzig. 1864–1868. 451 p. Roychoudhury S, Powell MJ. 1992. Precise configuration of the Rhizophlyctis harderi zoospore. Can J Bot 70:762– 771. Schenk A. 1858. Algologische Mittheilungen. Verhandl. Phys.-Med. Gesell. Wu¨rzburg, A.F., 8:235–259. Sikes DS, Lewis PO. 2001. Beta software, version 1. PAUPRat: PAUP* implementation of the parsimony rachet. Distributed by the authors. Department of Ecology and Evolutionary Biology, University of Connecticut, Storrs, CT. Simon L, Bousquet J, Le´vesque RC, Lalonde M. 1993. Origin and diversification of the endomycorrhizal fungi and coincidence with vascular land plants. Nature 363: 67–69. Sparrow FK. 1960. Aquatic phycomycetes. 2nd revised ed. Ann Arbor, Michigan: University of Michigan Press. 1187 p. , Lange L. 1977. Some bog chytrids. Can J Bot 55: 1879–1890. Swofford DL. 2001. PAUP* Phylogenetic Analysis Using Parsimony (and other methods). Version 4.0. Sunderland, MA: Sinauer Associates. Taylor TN, Remy W, Hass H. 1992. Fungi from the lower Devonian Rhynie chert: Chytridiomycetes. Am J Bot 79: 1233–1241. Tehler A, Farris JS, Lipscomb DL, Kallersjo M. 2000. Phylogenetic analyses of the fungi based on large rRNA datasets. Mycologia 92:459–474. , Little DP, Farris JS. 2003. The full-length phylogenetic tree from 1551 ribosomal sequences of chitinous fungi, Fungi. Mycol Res 107:897–900. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG. 1997. The Clustal X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 24:4876–4882. van de Peer Y, Jansen J, de Rijk P, de Wachter R. 1997. Database on the structure of small ribosomal subunit RNA. Nucleic Acids Res 25:111–116. Vilgalys R, Sun BL. 1994. Ancient and recent patterns of geographic speciation in the oyster mushroom Pleurotus revealed by phylogenetic analysis of ribosomal DNA sequences. Proc Nat Acad Sci USA 91:4599–4503. Wainwright PO, Hinkle G, Sogin ML, Stickel SK. 1993. Monophyletic origins of the Metazoa: an evolutionary link with the fungi. Science 260:340–342. Willoughby LG. 1971. Aquatic fungi from an Antarctic island and a tropical lake. Nova Hedwigia 22:459–487.
