Mol Biol Evol 1992 Bowman 285 96

8/3/2019 Mol Biol Evol 1992 Bowman 285 96

1/12

M olecular Evolution of the Fungi: Relationship of theBasidiomycetes, Ascom ycetes, and Chytridiomycetes

Barbara H. Bowman,* John W. Taylor,? Alan G. Brownlee,$Jean Lee,g Shi-Da Lu, * and Thomas J. White*

*Roche M olecular Systems; TD epartment of Plant Biology, University of California,Berkeley; $Division of Animal Production, CSIR O, and $Cetus Corporation

Establishing the phylogeny of fungi and protists o ften has proved difficult owingto the simple m orpholog ies and convergent characters in these organisms. We usedDN A sequences of nuclear small-subunit ribosomal RN A genes to determine phy-

logenetic relationships among three major classes of organisms considered to befungi-Basidiomycetes, Ascomycetes and Chytridiomyce tes-and to assess thetaxonomic position ofNeocallimastix, an economically important anaerobic rumenmicroorganism wh ose classification is controversial. The Basidiom ycetes and As-comycetes, two classes of nonflagellated fungi, are the most closely related taxa .Chytridiom ycetes, though bearing flagella, group with these higher fungi ratherthan with the protists.Neocallimastix, a eukaryo te lacking mitochondria and var-iously classified as a protist or as a fungus, show s closest molecular affinities withthe Chytridiom ycete fungi in the order Spizellomycetales.

Introduction

The fungi, historically considered to be plants because of their form and apparentlack of locomotion, have been recognized as a kingdom in their own right, a classi-fication championed by Whittaker ( 1969). Two of the classes of fungi that lack flgella-Basidiomycetes (mushrooms,shelf fungi, and their allies) and Ascomycetes(mostly molds and yeasts)-are universally allied by morphological characters (Tehl

1988 ) . Also traditionally classified as fungi are three c lasses of flagellated organisms-Chytridiomycetes ( chytrids ), Oom ycetes (water molds), and HyphochytriomycetesOf these, only the chytrids share with the nonflagellated classes their cell-wall polyme( Bartnicki-Garcia 1970) and lysine synthetic pathway (Vogel 1964).

Phylogenetic analysis based on the- 120-base 5s ribosomal RN A (rRNA ) sug-gested that the nonflagellated fungi do not form a natural (monophyletic) group (Hoand Osaw a 1987); indeed, in that analysis, green plants branched between the twfungal classes. Other researchers (Van de Peer et al. 1990) have found these nonfl

gellated classes to form a monophyletic group, also based on 5s rRN A. Howevebecause of the small amount of information in the 5s molecule, neither of these resuis likely to be strongly supported (Halanych 199 1; Steele et al. 199 1). Hendriks et ( 199 1)) using 18s rDNA , found that the nonflagellated classes were not monophyletiwhen a limited number of sequence positions were used in the analysis but that theclasses became so when a less conservative subset was used. To determine wheth

1. Key words:molecular evolution, 18s ribosomal R NA , rumen fungi, chytrid, basidiomyce te.

Address for correspondenceand reprints: Barbara H. Bow man, Roche Molecular Systems, 1145 AtlantAvenue, Suite 100, Alameda, California 945 0 1.

Mol. Biol. Evol. 9(2):285-296. 1992.0 1992 by The University of Chicago. All rights reserve d.

b y g u e s t on N

o v e m b e r 3 0 ,2 0 1 1

h t t p : / / m b e . oxf or d j o ur n a l s . or g /

D o wnl o a d e d f r om
http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/


2/12

286 Bowman et al.

these conflicting observations can be resolved with any confidence, we have used osequence of nuclear small-subunit ribosomal DN A ( 18s rDN A) from the basidiomycete shelf fungusSpongipel is unicolor, comparing it to sequences already known fromAscomycetes and other organisms.

Phylogenetic placement of the class Chytridiomycetes is a matter of current debat

These organisms are claimed for the protists on the basis of possession of flagel(Margulis and Schwartz 1988, pp. 77, 140-141, and 153) and are claimed for thfungi on the basis of characters that they share with the nonflagellated classes. Forstet al. ( 1990), using 18s rDN A sequences, placed the chytridBlas toc lad ie l la emersoni ias a sister group to the ascomycetes, but the degree of support for this conclusion wnot addressed. We present two new sequences for 18s rDN A from members of thChytridiomycetes and determine with a high degree of confidence their relationshito the nonflagellated fungi.

Neoca l l imas t i x and related anaerobic gut microbes also have controversial classifications. Although these agriculturally important, cellulolytic microbes share chaacters with Chytridiomycetes, many of their characteristics are unique, i.e., multiplflagella (Braune 19 13 ), genomic G+C content ~20% (Brownlee 1989)) lack of mtochondria (Heath et al. 1983 ), strict anaerobiosis (Orpin 1975 ), and hydrogen-generating hydrogenosomes (Yarlett et al. 1986). The zoospores of these gutmicrobesshare ultrastructural features with one order of Chytridiomycetes (Spizellomycetales)(Barr 1980; Heath et al. 1983) and share developmental features with another (Blas-tocladiales) ( Wubah et al. 199 1); but intense interest in these microbes has resultedin many new descriptions, and their classification is anything but settled (Barr et al.1989; Breton et al. 1990; Ho et al. 1990). To determine the phylogenetic placementof N eoca l l imas t i x , w e sequenced its nuclear 18s rDN A, for comparisonwith ourchytrid sequences and other know n sequences.

We therefore address three questions that are well suited to molecular evolutionarinvestigation: ( 1) Are the Basidiomycetes, Ascomycetes, and Chytridiomycetes paof a monophyletic fungal lineage? (2) Are the anaerobic gut microorganisms mem beof the Chytridiomycetes? (3) Are gut chytrids most closely related to the order Spzellomycetales?

Mat erial and M ethods

Neoca l l imas t i x species, culture LM -2, was obtained from the Anaerobic CultureCollection, CSIRO, Prospect, New South Wales.Sp ize l l omyces acumina tu s , strainBarr 62A, was a gift from Donald Barr.Chyt r id ium confervae 8l- 1 was obtained fromthe U.C. Microgarden (University of California, Berkeley). Total DN A isolated froC. conferae (Lee and Taylor 1990) andNeoca l l imas t i x (Brownlee 1988 ) and pSK2plasmid DNA fromSpongipe l i s un ico lor (Kwok et al. 1986) was amplified using con-ditions which minimize the errors introduced by theT aq DN A polymerase ( < 1 error/5 kb after 35 cycles) (Gelfand and W hite 1990) with modified versions of primerNS 1 and NS8 (W hite et al. 1990): SL2 l-CCGAATTCGTAGTCATATGCTTGTCT;SL27-CCAAGCTTAAACCTTGTTACGACTT. Amplified DNA was purified bprecipitation, treated with Klenow fragment, purified by agarose electrophoresis, anblunt-end ligated into the HincII site of the pUCl8 vector. TransformedEscherichiacoli strain DH5 was screened with the probe NS2 (White et al. 1990), and clonewere sequenced from both strands (Sequenase protocol; U .S. Biochemical). Sequencobtained from the cloned polymerase chain reaction (PCR) products ofNeoca l l imas t i x ,C. confervae, and Spongipe l i s un ico lor were verified against partial sequences obtained

b y g u e s t on N o v e m b e r 3 0 ,2 0 1 1


http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/


3/12

Molecular Evolution of Fungi 287

from direct sequencing of 800- 1,100 bases each from single-stranded DN A, asymmetrically amplified (Kwok et al. 1986; Gyllensten and Erlich 1988) either from 10 ng of genomic DN A or from the clonedSpongipe l i s un ico lor rDNA repeat.

PCR reactions onS p i z e l l o m y c e s a c u m i n a t u s total DN A were performed usingprimers NS 1 (White et al. 1990) and NS24( AAACCTTGTTACGACTTTTA ) (A .Gargas, personal communication), one of which was biotinylated in each of two rciprocal reactions. Spize l lomyces acumina tus DN A was diluted 1: 100 with autoclaveddistilled water from the extracted stock DNA . Fifty microliters of diluted D NA wamplified in a loo-u1 reaction containing 1 X GeneAmp Buffer (Perkin Elmer-Cetus); 62.5 pM each dGTP, dATP, dTTP, and dCTP; 5% glycerol (Smith et a1990); 50 pmol of each primer; and 2.5 units AmpliTaq (Perkin Elmer-Cetus). Threaction mixture was heated to 95C for 5 min and then w as subjected to 40 cycleeach of 40 s at 95C 25 s at 50C and 3 min at 72C; these were followed by a singIO-m in extension at 72C . Products were stored at 4C briefly before single-strandedDNA preparation.

S p i z e l l o m y c e s a c u m i n a t u s single-stranded 18s rDNA for sequencing was preparedusing a streptavidin agarose (SA ) technique based on that of Mitchell and Merri( 1989), with the variation that columns were eliminated, and the entire procedur was done in microcentrifuge tubes. All procedures were performed at room tempe ature. After each step, the mixture was spun briefly in a microcentrifuge to pellet teSA beads, and the supematant was removed with an Eppendorf micropipetter. SAbeads (Bethesda Research Laboratories 5942SA ) were rinsed five times with storaebuffer (20 mM Tris, 200 mM NaCl, 1 mM ethylenediaminetetraacetate, pH 7.)before use. The slurry of SA and storage buffer[ - 1:l (v:v)] was stored at 4C. Ap-proximately 90 ~1 of double-stranded, biotinylated PCR product was captured on 20-~1 SA slurry in a 2-ml microcentrifuge tube (Sarstedt 72,689) by agitation onaLabquake rotator (Labindustries, Berkeley, Calif.) for 30-40 min. After being washetwice with 500 ~1 storage buffer, the double-stranded DN A (still attached to SA) wdenatured by rinsing the beads twice, for 6 min each time, with 150 ~1 freshly prepare0.2 M NaO H, and the nonbiotinylated single strand of DN A w as recovered in tsupernatants. The combined supernatant containing the ssDN A was neutralized wi200 pl of 5 M amm onium acetate, pH 6.8, and was desalted and concentrated tovolume of 35-60 ~1 by using three 2-ml washes with autoclaved distilled water oCentricon- (Amicon 42 12). Of the resulting -40 ~1 of single-stranded DN A slution, 7.5 ~1 was sequenced using standard Sequenase 2.0 protocol (U.S. Biochemicalwith the exception that the labeling mix was diluted 1:20 before use. Bases at 1,4positions were sequenced on both strands of the DN A. Single-stranded sequence waccepted when the data were clear, were read identically by two investigators who hno knowledge of the others reading, and were alignable w ith published fungal squences.

Results and Discussion

We obtained sequences for 18s rRN A genes, complete except for an estimate38 bases at the 5 end and 49 bases at the 3 end, forN e o c a l l i m a s t i x species, the twochytrids Chyt r id ium confe rvae and Spize l lomyces acumina tus , and the basidiomycete

shelf fungus Spo ngipel is un icolor ( f ig. 1) . These could be unambiguously aligned withpublished sequences from the chytridBlas t oc lad iel la em ersoni i (Forster et al. 1990),the ascomycete bread moldNeurospora c rassa , the soybean G l y c i n e m a x , and the

b y g u e s t on N

o v e m b e r 3 0 ,2 0 1 1


http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/


4/12

FIG. 1 -Aligned small subunit rRNA sequences from four species. Sequence data were obtained fthe anaerobic rumen fungusNeocallimustix ( I,7 17 bases, GenBank accession number MS9761 ), the chytridsChytridium confervae ( 1,707 bases; M59 758),Spizellomyces acuminatus ( 1,718 bases; M59759), and thebasidiomycete shelf fungusSpongipelis unicolor ( 1,718 bases; M5 9760 ), between PC R primers N Sl andNS24. These were aligned with published sequences of the ascomycete bread moldNeurospora crussu(GenBank N EURRN AS), the chytridBlusrocladiellu emersonii (Forster et al. 1990), the ciliate protistStylonychiu pustulata ( GenBank SLURGSS ), and the soybean Glycine mux ( GenBank SOYRGE) .(Thesefour sequences are not shown, bu t the complete alignment will be supplied on request.) Sequences waligned manually, using the Eyeball Sequence Editor ESE E version 1.09 (Cabot and Beckenbach 1989 )

b y g u e s t on N o v e m

b e r 3 0 ,2 0 1 1


D o wnl o a

d e d f r om
http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/


5/12


S.un1cok.r CAAAcTTGGTCATTT*GAGGliRG 1718C.co erf. .......................Neoc.llIma .......................s.acum1nat .......................

the IBM PC. A dot indicates a base both homologous to and identical to the corresponding base in t

Spongipelis unicolor (reference) sequence. A capital letter indicates a base that differs from the base in threference sequence. A lowercase letter indicates a base at a position whose homology to positions in reference sequence is undetermined . These include areas in which the sequences are sufficiently divergthat positional homology cannot be inferred directly from sequence similarity and in which, in additiosmall or large length variations are evident. When short (< 1 2 bases) a reas of substantial sequence divergenwere bounded by regions of certain homology and included no length variation, these were consideralignable (i.e., positional homology was inferred-e.g., see positions 743-747 in fig. 1). Unalignab le araswere omitted from all phylogenetic analyses. In some regions, a subset of the sequences was alignable, adfor those sequences these positions were included in parsimony analyses. Positions used in distance analyes(neighbor joining) are underlined and include only those positions alignable in all eight species (includigNeurospora crassa, B. emersonii, Sty lonychia pustulat a, and G. max). The boundary of a sequence that wasalignable despite high sequence variation was differentiated from an adjacent unalignab le area if ( 1) theof the four bases at the margin of the aligned region were identical to those at the sam e positions in at lestone other sequence of certain alignment, (2) these three matching bases included the position bounding ealigned area, and (3) no length variation was involved (e.g., see C.confervae at positions 605-608, whichis identical both to the reference sequence at positions 606 and 608 and to the well-alignedNeocakmastix sequence a t position 605 ).

pairwise substitutions in the aligned regions was 169 ( 12.4%) between the bread moand the ciliate protist.

We used three different m ethods of phylogenetic analysis to determine evolutionary relationships and to evaluate the strength of the branching order: the neighbojoining method (Saitou and Nei 1987), based on genetic distances; a bootstrap teof the neighbor-joining method (Whittam 199 1); and the parsimony-based WinningSites Test (Prager and Wilson 1988) .

The neighbor-joining method (Saitou and Nei 1987) was used to construct, frothe unambiguously aligned positions, the branching order shown in figure 2. This trunites the two classes of nonflagellated fungi as closest relatives, in accordance wimorphological and physiological criteria. The anaerobic rumen fungusNeocullimastixfalls within a monophyletic group formed by the flagellated class ChytridiomycetesThis group is united with the nonflagellated fungal classes, to the exclusion of tprotist.

Any phylogenetic tree must be considered a hypothesis; it is necessary to examinrigorously any specific inferences drawn from this tree, to see whether the data asufficiently robust to support definitive conclusions. The neighbor-joining methousually returns the minimum-length tree (Saitou and Nei 1987; Saitou and Imanis1989), but this is not necessarily the correct tree. (Neighbor-joining is guaranteed return the correct tree only if distance data satisfy the condition of being additive )

b y g u e s t on N

o v e m b e r 3 0 ,2 0 1 1


http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/


6/12

290 Bowman et al.

Shlonvchia vustulata _ -- ..-(chate prbtist)

\

80

Glycine max(soybean)

1Spongipelis unicolor 9.Neocallimastix

Spizellomyces acuminatus 17

61

\

\

%

&Chytridium confervae 2

6.

\ Blastocludiellia emersonii J

FIG. 2.-Neighbor-joining tree of eight species. The computer program NJTREE (Jin an d Ferguson

1990) was used to estimate the branching order and branch lengths for the eight species. O f the 1,368alignable sequence positions, a maximum of 169 ( 12.4 % ) differed in any pairwise comparison. Uncorrected

differences were used to construct the tree. Branch lengths have been rounded to the nearest wh ole numberof substitution s, and branch es are drawn proportional to their lengths . A scale is provided to show the

inferred nu mber of substitutions per sequence position. In this tree, nontlagellated fungi form a m on ophyleticgroup. The anaerobic rumen flagellate Neocuiiimastix falls within the monophyletic group for m ed by theChytridiomycetes (flagellated fungi). Chytrids and the nonflagellated fungi form a monoph yleti c group to

the exclusion of the protist and green plant. The two significant bootstrap results (Whittam 199 1) , returnedafter 1,000 trials, are shown in brackets. Bootstrap support for other branches was not significant at the 95%level (Felsenstein 1985): there was 93.1 % support for the branch u niting Neocallimast ix w ith Spizellomy ces

acumi nat us, 8 1.3% support for the branch uniting these two with Chytridium confervae. and 93.4% support

for the monophyly of the chytrids (flagellated fungi).

joining tree, we used both a bootstrapping test based on the neighbor-joining algorithm(Whittam 199 1) and the parsimony-based Winning Sites Test (Prager and Wson 1988).

The Winning Sites Test determines how strongly the branching order is supporteby the data, by contrasting the number of sequence positions that support alternativunrooted tree topologies. For the simplest test, four taxa are arranged into the thrpossible topologically distinct, bifurcating networks (e.g., see fig. 3 ). Phylogeneticallinformative sequence positions for this test are those at which two species share obase while the two others share a different base. When a common base is shared the two species on the same side of the network, the data can be explained by a sinsubstitution event, occurring on the central branch between the nodes connecting tpairs of sequences. The other two topologies would require at least two events. Tnetwork requiring a single event is counted as winning at that sequence positioThe total numbers of sequence positions favoring each of the three networks acompared, to see whether support for one topology significantly exceeds support the alternatives ( Li and Gouy 199 1). Because the Winning Sites Test can be led astray

by base-compositional bias among the sequences compared (Irwin and Wilson, acepted) the base compositions of the sequences analyzed were evaluated.Neocullimastix

enomic DNA is known to have an unusuall low ercenta e of G+C ( 18%; Brownl



http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/


7/12

Rooted trees

aStyionychiu puetuiatu

Spongipelis unicolor

Glycinc nu.r

I Neurospom craw

bNeurospom cmssu

Spongipelis unicolor

Glycine mar

sty10?lychiu uetulutu

Molecular Evolution of Fungi 29 1

Unmated trees:number of supporting positions

Stylonychiu pustulutu Giycinenrax

11

Spongipelis unicolor X Neumspom cmesuNeurospom mssa

(bread mold)

XGlycinc tax(soybean)

30

Spongipelie unicolor Stylonychiu push&u(shelf fungus) (ciliate protist)

E

Neurospora crasso x Glycine mox9Styionychiu puetuluta Spongipclis u&color

Ftc. 3.-Eva luation of relationship of Ascomycetes and Rasidiomycetes. The Winning Sites Test (Pragr

and Wilson 1988 ) was used to compare support for the branching order suggested by 5s rRNA sequenecomparisons and for the morphologically based branching order, which is also that returned by the neighbo-joining method using 185 rDN A sequences. a, SS-based rooted tree (left) and corresponding unroote network. b, One possible rooted tree that unites nonflagellated fungi in monophyletic group. The unroote network (right) represents any rooted tree uniting these two taxa. The third possible network is also show.Fifty sequence positions are phylogenetically informative for these four taxa, with two taxa sharing one beand the other two sharing a different base. Of these, 1I support (require only one substitution in) networka, 30 support network b, and 9 support network c. When the four-taxon method of Li and Gouy ( 199 1used, this result is significant at the 0.15% level.

1989); nevertheless, the percentage of G+C inN e o c u l l i m a s t i x 18s rDNA is 41.9%and does not differ markedly from those of the other fungi evaluated, whose percentagof G+C are 44.3%-48.2%.

Using the W inning Sites Test, we first evaluated whether the nonflagellated fungclasses are closest relatives. In addition to these fungi, green plants and ciliate protiswere represented both in the present study and in the 5s rRN A study by Hori anOsaw a ( 1987) that found the nonflagellated fungi not to be monophyletic. The 5rRN A tree grouped the Ascomycetes (represented here by the bread moldNeurosporacrassa) with green plants (the soybean G. max) and grouped the Basidiomycetes(represented here by the shelf fungusSpongipe l i s un ico lor ) with ciliate protists (Sty-l onych ia pustuluta) . (This branching order, obtained using a simplified unweightedpair-group method, was not strongly supported, having substan tial error bars on thclosely spaced branch points.) W e compared the network representing the 5s tree (f3a) both with that representing the 18s neighbor-joining tree (fig. 3b) and with ththird alternative. Of 50 sequence positions phylogenetically informative for these fotaxa, 11 supported the 5s branching order, 30 supported the i8S branching ordeand 9 supported the third a lternative. This support for the unity of the nonflagellatefungal classes is significant at the 0.15% level, by the four-taxon method of Li an

b y g u e s t on N

o v e m b e r 3 0 ,2 0 1 1




8/12

292 Bowman et al.

Because the protists are a paraphyletic group, their relationship to the fungclasses migh t not be accurately represented by a single phylum. We repeated this teusing the chrysophyte algaOchromonas dan i ca (GenBank OCHAB), whose divergencefrom other lineages was close in time to the divergence of plant and fungal lineag(Gunderson et al. 1987). ThusO c h r o m o n a s stands the best chance of disrupting the

monophyletic grouping of the fungi. Of 43 phylogenetically informative positions, grouped the two nonflagellated fungal lineages, while only 6 and 11 positions, resptively, supported the two alternatives (support again was significant at the 0.15% lev(Li and Gouy 199 1) . Thus the Winning Sites Test strongly favors the branching ordsuggested by morphological and physiological criteria.

In addition, a bootstrapping test using the neighbor-joining algorithm (Whittam199 1) supported the association of the nonflagellated fungal classes, in 976 out 1,000 replicates (fig. 2). The greater quantity of information in the 18s molecu

(relative to 5s) can indeed be used to resolve with a high degree of confidence tbranching order of these four taxa, supporting the morphological classification unitithe two classes of nonflagellated fungi.

A second inference from the neighbor-joining tree is that the Chytridiomyceteare the closest relatives of the nonflagellated fungi. Chytrids traditionally have beclassified as flagellated fungi, along with the Oomycetes (water molds) and Hyphochy-triomycetes. Margulis and Schwartz ( 1988, p. 76) show a morphologicallybased treein which chytrids are not closely related to the nonflagellated fungi and areclassed asprotists. Recent molecular studies( Kwok et al. 1986; Fiirster et al. 1990) havesuggestedinstead both that chytrids are part of a monophyletic lineage that includesthe non-flagellated fungal classes and that Oomycete water molds have their closest relativesamong the protists.

We used the Winning Sites Test to determine whether the proposition that chytridsshared their most recent common ancestry with nonflagellated fungi, rather than wprotists, could be supported with a high degree of confidence. For this analysis, compared the soybean 18s rDNA sequence both with those of the ciliate protistS t y -l o n y c h i a p u s & d a t a , a species from the Chytridiomycete type genusC h y t r i d i u m , andwith the nonflagellated shelf fungusSpongipe l i s un ico lor. Twenty-five sequence po-sitions supported grouping C .c o n & w e with the shelf fungus, while only 12 united iwith the protist, and 7 favored the third alternative, supporting, at the 0.9% level, association of the chytrids with the nonflagellated fungi (Li and Gouy 199 1). Subtution of the bread mold for the shelf fungus gave similar results, as did using chrysophyte 0. danicu in place of the ciliate. In addition, a bootstrapped neighborjoining test (Whittam 199 1) supported (in 99 1 of 1,000 replicates) the branch unifythe flagellated (Chytridiomycete) and nonflagellated fungal lineages, to the exclusiof the protist and green plant (fig. 2).

A third inference tested us ing the Winning Sites Test is that the anaerobic rumfungus Neoca l l imas t i x is part of a monophyletic fungal lineage, rather than being protist. Because of the paraphyly of the protists, testing the simple association Neocu l l imas t i x with another fungus, to the exclusion of a representative protist, insufficient. It is necessary to demonstrate thatNeoca l l imas t i x branches within thefungi and is not an exterior branch.

Along with Neoca l l imas t i x , the ciliate protist S t y l o n y c h i a p u s t u l a t a , the bread

mold Neurospora c rassa , and the chytrid Sp ize l l omyces acumina tu s were used to testthis inference (fig. 4). If the tree unitingNeoca l l imas t i x with the protist is favored(fig. 4a), then Neoca l l imas t i x might not be a fungus. If it unites with either of th





9/12

Rooted trees

aNeurospora crass(~

Spizellomyces euminatus

Neocailimastix

Sty lonchia ushdata

b Neocaltimasti

Spizdlomyces acuminatus

Neurospora crassa

C

Neocallimastix

Neurospora cmssa

Spizllomyces acuminatus

Molecular E volution of Fungi 293

Unrooted trees:number of supporting positions

Neurospora crassa

Spizellomyces cuminutus XNeocallimaslix

4

Srylonychia ust ulatu

Neocallimastix(rumen fungus)

Spizellomyces cuminatus n Sty lonychia ustuluta(cbytrid fungus) (ciliate protist)

Neocallimaslix

Neurospora crassaX

Spizellomyces euminatus

4

Siylonychia ushda t a

FIG. 4.-Evaluation of placement of anaerob ic rumen flagellateNeocallimastix in Fungi kingdom. TheWinning Sites Test (Prager an d Wilson 19 88 ) was used to evalua te the strength of the inference that Ne-cullimastix belongs in the Fung i kingdom and is not a protist. Sup port of network a (topologically equivalentto the rooted tree to its left) wou ld allow the conclusion thatNeocullimastix is a protist; support for eithernetwork b or network c would placeNeocullimustix within the Fungi kingdom. Of the positions phyloge-netically informative for these taxa, 1 9.favor placing Neocullimustix within the fungi and associated withthe chytrid Spizellomyces acuminutus, and 4 positions favor each of networks a and c. When the four-taxonmethod of Li and Gouy ( 19 9 1) is used, this result is significant at the 0.1 5% level.

other species, it can with confidence be called a fungus (fig. 4b and c). O f the 2

positions phylogenetically informative for these four taxa, 19 uniteNeoca l l imas t i xwith the Chytridiomycete fungus, while only 4 place it with the protist, and 4 associatit with the bread mold (this distribution is significant at the 0.15% level) (Li and Gou199 1). This result supports the placement ofNeocu l l imas t i x within the fungi. Specif-ically, N e oca l l i ma s t i x is allied with the chytrid branch, to the exclusion of the nonfla-gellated fungal lineage. The more conservative bootstrap test of neighbor-joining(Whittam 199 1)) however, provided nonsignificant support (e.g., uniting the fouflagellated fungi as a monophyletic group in only 93 1 of 1,000 trials.)

To evaluate with which orderNeoca l l imas t i x is allied w ithin the c lass Chytridi-omycetes, we compared its sequence with those of three o ther fungi: ( 1)B. emersoni i ,representing the order Blastocladiales, which shares some developmental features wiNe oca l l i m a s t i x ( Wubah et al. 199 1); (2) Sp ize l l omyces acumina tu s , representing theorder Spizellomycetales, into which Heath et al. ( 1983) proposed placingNeoca l l i -m a s t i x ; on the bas is of zoospore ultrastructure; and (3) the shelf fungus as outgroupThe network unitingNeocu l l im as t i x with the order Spizellomycetales was overwhelm-ingly supported (22 sequence positions favoringSp ize l l om y ces acum ina tu s v s . 4and

3 favoringB. em erson i i

and the outgroup, respectively), verifying this aspect of theneighbor-joining tree (with support at the 0.15% level) (Li and Gouy 199 1) and suporting Heath et al.s ( 1983) classification based on zoospore ultrastructure. Whe



D

o wnl o a d e d f r om


10/12

294 Bowm an et al.

we substituted C.confervae (Order Chytridiales) forB. emersoni i in the Winning SitesTest, the test could not exclude the possibility thatNeoca l l imas t i x united with theorder Chytridiales instead of with the Spizellomycetales. Bootstrap analysis (Whitta199 1) returned nonsignificant results, unitingNeoca l l imas t i x with Spizellomyces in934 of 1,000 trials. Nevertheless, the phylogenetic position ofNeoca l l imas t i x is as a

late branch within the fungi, suggesting that its lack of mitochondria is due to seconary loss.In summ ary, we have presented four new essentially full-length fungal 18s rRN

sequences from one basidiomycete and three chytrid fungi. Within the organismrepresented by these sequences, our data support the association of the nonflagellatfungal classes Ascomycetes and Basidiomycetes, in keeping with morphological clsification. We have further shown, w ith a high degree of confidence, that the flagellaclass Chytridiomycetes and the nonflagellated fungal classes together form a monphyletic unit. A lthough the Chytridiomycetes were the first class to diverge at the bof the fungal branch, there is no strong reason to exclude them from the Fungi kingdo

We demonstrate with the present study an instance in which molecular data habeen used successfully to discern the phylogenetic status of an unusual organism whounique biochemical properties made its classification by morphological and physlogical criteria a matter of contention. The rumen flagellateNeoca l l imas t i x is certainlya Chytridiomycete fungus. It has been possible, with a high probability, to assign thisgenus to the order Spizellomycetales, to the certain exclusion of the orderBlastocla-diales.

Acknowledgments

We thank A. Wilson, D. Barr, E. Zimmer, and N. Amheim, and thereviewersfor their comments o.n the manuscript; D. Spasic, L. Goda, and C. Levenson foroligonucleotide synthesis; M . A. Brow for advice on sequencing; M. Sogin for providthe Blas toc lad ie l la emersoni i sequence in advance of publication; R. Holmquist, Lake, and W.-H. Li for helpful d iscussion; A. Gargas for the sequence of primer NS

and J. Price for encouragement and support. This research was supported in part NIH grant AI2854502.

LITERATUR E CITED

BARR, . J. S. 1980. An outline for the reclassification of the Chytridiales, and fo r a new order,the Spizellomycetales. Can . J. Bo t. 58:2380-2394.

BARR, . J. S., H. K UD O, K. D. JAKOBER,nd K .-J. CHE NG. 1989.Morphology and developmentof rumen fungi: Neocallimastix sp.,Piromyces communis, and Orpinomyces bovis gen. nov.,

sp. nov. Can. J. Bot. 67:2815-2824.BARTNICKI-GARCIA,. 1970. Cell wall composition and other biochemical markers in fungal

phylogeny. Pp. 81-103 in J. P. HARBORNE, d. Phytochemical phylogeny. A cademic Pr ess,London.

BRAUNE,R. 19 13. Untersuchungen tiber die im Wiederkauermagen vorkomm enden Protozoen.Arch. Protistenkde. 32: 11 l- 170.

BRETON,A., A. BERNALIER, . DUSSER,G. FONTY, B. GAILLAR E-MARTINIC,nd J. GUILLO T.1990. Anaeromyces mucoronatus nov. gen., nov. sp: a new strictly anaerobic rumen funguswith polycentric thallus. F EM S Microbial. L &t. 70:177-l 82.

BR OW NL EE, . G. 1988. A rapid DN A isolation procedure applicable to many refractory fila-mentous fungi. Fungal Gen. Newslett. 35:8-9.

-. 1989. Remarkably AT-rich genomic DN A from the anaerobic fungus Neocallimastix.



http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/


11/12


CABOT , E. L., and A. T. BE CKEN BACH .1989. Simultaneous editing of multiple nucleic acidand protein sequences with ESEE . Com put. Appl. Biosci. 5:233-234.

FEL SEN STE IN, . 1985. Confidence limits on phylogenies: an approach using the bootstrapEvolution 39:783-791.

F~RSTER, H., M. D. CO FFEY, H. ELW OOD ,and M . L. SOGIN. 1990 . Sequence analysis of thesmall subunit ribosomal RN As of three zoosporic tingi and implications for fungal evolution

Mycologia 82306-312.GELF AND , D., and T. J. WH ITE. 1990. Thermostable DN A polymerases. Pp. 129- 14 1in M .

INNIS, D. H. GEL FAN D, J. J. SNINSK Y, and T. J. W HITE, eds. PCR protocols. AcademicPress, San Diego.

GUNDERSON, . H., H. ELWOOD,A. INGOLD, K. KIND LE, and M . L. SO GIN. 1987. Phylogeneticrelationships between chlorophytes, chrysophytes, and oomycetes. Proc. Natl. Acad. ScUSA 845823-5827.

GYLLENSTEN,U. B., and H . A. ERLIC H. 1988 . Generation of single-stranded DN A by thepolymerase chain reaction and its application to direct sequencing of theHLA-DQA locus.Proc. Natl. Acad. Sci. USA 85:7652-7656.

HALANYCH,K. M . 199 1. 5s Ribosomal RN A sequences inappropriate for phylogenetic reconstruction. Mol. Biol. Evol. 8:249-253.

HEATH , I. B., T. BAUC HOP,and R . A. SKIP P. 1983. Assignment of the rumen anaerobe Neo-cullimastix frontalis to the Spizellomycetales (Chytridiomycetes) on the basis of its polyflagellate zoospore ultrastructure. Can. J. Bot. 61:295-307.

HEND RIKS, L., R. DE BAER E, Y. VANDE PEER , J. NEEFS, A. GOR IS, and R. DE W ACHTER. 199 1. The evolutionary position of the rhodophytePorphyra umbilicalis and the basidiomy-

cete Leucosporidium scottii among other eukaryotes as deduced from complete sequencesof small ribosomal subunit RN A. J. Mol. Evol. 32:167-l 77.Ho, Y. W., T. BEAUCHOP,N. ABDULLAH,and S. JALU DIN. 1990. Ruminomyces elegans gen.

et sp. nov., a polycentric anaerobic rumen fungus from cattle. Mycotaxon 38:397-405.HO RI, H., and S.O S A WA 1987 . Origin and evolution of organisms as deduced from 5s ribosomal

RN A sequences. Mol. Biol. Evol. 4445-472 .IRW IN, D. M., and A. C. W ILSO N. Limitations of molecular methods for establishing the ph

logeny of animals, with special reference to the position of elephants. In F. S. SZA LAM . J. NO VA CEK , and M . C. MC KEN NA , eds. Mam mal phylogeny. Springer, New Y ork

(accepted).JIN, L., and J. W. H. FERG USO N. 1990. NJT REE (~2.0) (computer program). Center for Po

ulation Genetics, University of Texas Health Science Center at Houston, Houston.KW OK, S., T. J. WHITE, and J. W. TAYL OR . 1986. Evolutionary relationships between fung

red algae, and other simple eukaryotes inferred from total DNA hybridization to a clonebasidiomycete ribosomal DN A. Exp. Mycol. 10: 196-204.

LE E, S. B., and J. W. TA YL OR . 1990. Isolation of DN A from fungal mycelia and single sporePp. 282-287 in M. A. INNIS, D. H. GELFAN D, J. J. SNINSKY ,and T. J. WH ITE, eds. PCRprotocols. Academ ic Press, San Diego.

LI, W.-H., and M . GO UY . 199 1. Statistical methods for testing molecular phylogenies. Pp. 24277 in M. M. MIYAMOTO,and J. CRACRAFT,eds. Phylogenetic analysis of DN A sequences.Oxford University Press.

MA RGU LIS, L., and K. V. SCHWA RTZ. 1988. Five kingdoms. Freeman, New York.MITC HELL, L. G., and C . R. MER RIL L. 1989. Affinity generation of single-stranded DN A f

dideoxy sequencing following the polymerase chain reaction. Anal. Biochem . 178 :239-242OR PIN , C. G. 1975. Studies on the rumen flagellateNeocallimastixfrontalis. J. Gen. Microbial.

9 1 : 2 4 9 - 2 6 2 .

PRAGER,E. M., and A. C. W ILSO N. 1988. Ancient origin of lactalbumin from lysozyme: analysof DN A and amino acid sequences. J. Mol. Evol. 27:326-335.

SAITO U, N., and T. IMANISH I. 1989. Relative efficiencies of the Fitch-Margoliash, maximumparsimony, maxim um-likelihood, minimum-evolution, and neighbor-joining m ethods

b y g u e s t on N

o v e m b e r 3 0 ,2 0 1 1


http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/


12/12

296 Bowman et al

SAITOU, N., and M.N E I 1987 . The neighbor-joining method: a new method for reconstructingphylogenetic trees. Mol. Biol. Evol. 4~406-425.

SMITH, K. T., C . M . LONG, B. BOWM AN,and M . MANOS. 1990. Amplifications: a forum forPCR users. Issue 5 (September), pp. 16-17.

STEEL E, K. P., K . E. HOL SINGER , R. K. JANS EN, and D. W. TAYL OR. 1991. Assessing threliability of 5s rRN A sequence data for phylogenetic analysis in green plants. Mol. Bio

Evol. 8:240-248.TEH LER , A. 1988. A cladistic outline of the eumycota. Cladistics 4~227-277.VAN DE PEER , Y., R. DE BAE RE, J. CA UW ENB ERGH S, nd R. DE W ACHT ER. 1990. Evolutio

of green plants and their relationship with other photosynthetic eukaryotes as deduced fro5s ribosomal RNA sequences. Plant Syst. Evol. 170:85-96.

VO GE L, H. J. 1964. Distribution of lysine pathways among fungi: evolutionary implicationsAm. Nat. 98:435-446.

WH ITE, T. J., T. BRUN S, S. LEE, and J. TAYLO R. 1990. Amplification and direct sequencing fungal ribosomal RNA genes for phylogenetics. Pp. 3 15-322in M. INNIS, D. H. GELFAND,

J. J. SNINSKY,and T. J. WH ITE, eds. PCR protocols. Academic Press, San Diego.WH ITTAKER, R. H. 1969. New concepts of kingdoms of organisms. Science 163: 150- 160.WH ITTAM , T. S. 199 1. PSFIN D and NJB OO T. (computer programs distributed by the author

Pennsylvania State University, University Park.WU BAH, D. A., M. S. FULLE R,and D. E. AKIN. 199 1. Resistant body formation inNeocdimastix

sp., an anaerobic fungus from the rumen of a cow. Mycologia 83:40-47.YARL ETT, N., C. G. ORPIN , E. A. M UNN , N. C. YARLE TT, and C. A. GR EEN WOOD. 1986.

Hydrogenosomes in the rumen fungusNeocallimastix patriciarum. Biochem. J. 236:729-739.

MASATOSHI N E I , reviewing editor

Received March 5, 199 1; revision received July 23, 199 1

Accepted July 25, 199 1 b y g u e s t on N o v e m b e r 3 0 ,2 0 1 1


http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/http://mbe.oxfordjournals.org/

Documents

Mol Biol Evol 1992 Bowman 285 96