1

Probing early eukaryoticevolution using phylogenetic

methods

The Universal SSU rRNA TreeWheelis et al. 1992 PNAS 89: 2930

Archezoa

The SSU Ribosomal RNA Tree for Eukaryotes

Mitochondria?

Prokaryoticoutgroup

AnimalsFungi

Ciliates + ApicomplexaStramenopiles

Euglenozoa

GiardiaTrichomonas

Plants / green algae

Red algae

Entamoebae

Choanozoa

Dictyostelium

Physarum

Microsporidia

Percolozoa

The Archezoa Hypothesis - primitivelyamitochondriate eukaryotes

(Tom Cavalier-Smith 1983)

Trichomonas Microsporidia

Giardia‘Tree’ courtesy of W. Ford Doolittle Entamoeba

The Archezoa HypothesisT. Cavalier-Smith (1983)

• The Archezoa hypothesis would fall if:

– Find mitochondrial genes on archezoangenomes

– Find that archezoans branch amongaerobic species with mitochondria

– Find mitochondrion-derived organellesin archezoans

2

Mitochondrial HSP 70 is encoded by thehost nucleus but is of endosymbiotic origin

N C

ATPase DOMAIN PEPTIDE BINDINGDOMAIN

Targeting sequences Retention

sequences

mitochondrial HSP70

alpha-proteobacteria

chloroplasts

cyanobacteria

Cytosolic HSP70

RER HSP70

other proteobacteria

Gram +ve & Archaea

Endosymbioticorigin

Mitochondrial genes in Archezoa

Giardia /Spironucleus

Trichomonas

Microsporidia

Entamoeba

Heat shock 70, Chaperonin 60

Heat shock 70, Chaperonin 60

Heat shock 70

Heat shock 70, chaperonin 60

*defined as forming a monophyletic group with mitochondrialhomologues in a non-controversial species phylogeny

Proteins of mitochondrial origin*Archezoa

Chaperonin 60 Protein Maximum Likelihood Tree(PROTML, Roger et al. 1998, PNAS 95: 229)

A case ofEukaryote Eukaryote

HGT?

Note 100% Bootstrap support

Long branches may cause problems forphylogenetic analysis

• Felsenstein (1978) made a simple model phylogeny including fourtaxa and a mixture of short and long branches

• Methods which assume all sites change at the same rate(e.g. PROTML) may be particularly sensitive to thisproblem

A

B

C

DTRUE TREE WRONG TREE

A B

C D

ppq

qq p > q

Chaperonin 60 Protein Maximum Likelihood Tree(PROTML, Roger et al. 1998, PNAS 95: 229)

Longestbranches

3

• Does the Cpn60 tree topology change:

– If we remove long-branch outgroups

– If we remove sites where every specieshas the same amino acid

A simple experiment:Cpn-60 Protein ML tree (PROTML) from variable

sites with outgroups removedGiardia

Entamoeba

Dictyostelium

30

31

Plants

Apicomplexa

Euglena & Trypanosoma

Trichomonas

Animals & Fungi

The Archezoa HypothesisT. Cavalier-Smith (1983)

• The Archezoa hypothesis would fall if:

– Find mitochondrial genes on archezoangenomes

– Find that archezoans branch amongaerobic species with mitochondria

– Find mitochondrion-derived organellesin archezoans

a-tubulin trees place Microsporidia with FungiEdlind TD, et al. (1996). Mol Phylog Evol 5 :359-67Keeling PJ, Doolittle WF. (1996). Mol Biol Evol 13 :1297-305.

Microsporidia

Fungi

Other eukaryotes

65

Microsporidia are primitive and earlybranching eukaryotes or relatives of fungi?

• Microsporidia branch deep in some trees but not in others– Deep: SSU rRNA, EF-2– With fungi: tubulin and HSP70

• Microsporidia contain a mitochondrial HSP70– They once contained the mitochondrial endosymbiont and thus are

not Archezoa sensu Cavalier-Smith– Hirt et al., 1997

– Or they obtained this gene (and the one for tubulin) from fungi viaHGT

– Sogin, 1997 Curr. Op. Genet. Dev. 7: 792-799

4

75

83

88

The tree shownis the maximumlikelihood treeobtained using aprogram calledPROTML

Maximum likelihood and phylogeneticinference

• The maximum likelihood tree is the one withthe highest probability of giving rise to thedata under a particular model

• It is not the probability that it is the “true”tree

Assumptions underpinning thePROTML model

• Assumes all sites in the molecule can change• Assumes that all sequences are evolving in

the same way• EF-2 violates both of these assumptions:

Human CAGGAATCACCTACGATCCTCGGCGGCGTTACGAACCGAA 53%G+CSaccharomyces CAGGAATCACTTACGATCTTAAGCGGCGTTACGAACAGAA 46%G+CChlorella CAGGAATCACCTACGATCCTCAGCGGCGCTACGAACCGGCDictyostelium CAGGAAACACTTTCGATCTTGTTCGGAGTAATTCTGCGGCTrypanosoma CAGGAATCACTTTCGATCCTGAGCGGCGATATGAACCGGCEntamoeba CAGGAATCACTTACGATCCTAAGCGGAGTAACGAACAGCCCryptosporidium CAGGAATCACTTACGATCCTGCGCGGCGTTACGAACCGGCTritrichomonas CAGGAATCACCTACGATCCTAAGCGGCGTTACGAACCGCC 54%G+CTritrichomonas CAGGAATCACCTACGATCCTAAGCGGCGTTACGAACCGCC 54%G+CGiardia CAGGAATCACCTACGATCCTTCGCGGCGTTACGAACCGCA 53%G+CGlugea CAGGAAAGACCTACGATGCTGATCGGAGTAATGAACCGGA 45%G+CSulfolobus CAGGAAACACACAGGAACCTGTGCGGCTGCTTGATACTTC 43%G+CMethanococcus CAGGAAACACTTTCGAAATTTTGCGGCTGCCTGATTGAGA 44%G+CHalobacterium CAGGAAACACCTACGAAACTACGCGGCTGCATGAACGAGA 58%G+C

LogDet/Paralinear distances for EF-2 DNAvariable sites codon positions 1+2

Animals

Chlorella

Trypanosoma

Trichomonas

Giardia

DictyosteliumEntamoebaSaccharomyces

GlugeaCryptosporidium

SulfolobusMethanococcus

Halobacterium

60

25

76

70

Archaebacteriaoutgroups

Microsporidia+ fungi!

A combination of factors (outgroup GC content andsite rate heterogeneity) influence the EF-2 DNA tree

0 20 40 60 80 100

Methanococcus outgroup(low G+C)

0

20

40

60

80

100

Halobacterium outgroupHigher G+C

0

20

40

60

80

100

0 20 40 60 80 100

(Microsporidia, outgroup)

Fraction of constant sites removed

LogDetBootstrap

values

MLestimate

0 20 40 60 80 1000

20

40

60

80

100

0

20

40

60

80

100

0 20 40 60 80 100

(Microsporidia, outgroup)(Microsporidia, Fungi)

Fraction of constant sites removed

Bootstrapvalues

A combination of factors (outgroup GC content & siterate heterogeneity) influence the EF-2 DNA tree

Methanococcus outgroup(low G+C)

Halobacterium outgroupHigher G+C

5

A combination of factors (outgroup GC content & siterate heterogeneity) influence the EF-2 DNA tree

0 20 40 60 80 1000

20

40

60

80

100

0

20

40

60

80

100

0 20 40 60 80 100

(Giardia, Trichomonas, outgroup)Fraction of constant sites removed

Bootstrapvalues

Methanococcus outgroup(low G+C)

Halobacterium outgroupHigher G+C

Are the former Archezoa reallyancient offshoots?

• Probably not:• Microsporidia are related to fungi (Tubulin, RNA

polymerase,LSU rRNA, HSP70, TATA binding protein,EF-2, EF-1 alpha, SSU rRNA)

• Evidence for Giardia and Trichomonas branching deeperthan other eukaryotes is based on trees made usingunrealistic assumptions (often PROTML)

• There is plenty of room for new hypotheses

If former archezoa containgenes from the mitochondrial

endosymbiont what happened tothe organelle?

Microsporidia are obligateintracellular parasites

Life cycle of microsporidia

Tree of Mitochondrial Hsp70

mtHsp70 has diverse roles in mitochondriaPfanner and Geissler 2001 Nature Reviews 2: 339-344

mtHsp70 also plays a role in assembly of Fe-S clusters - an essential function for yeast mitochondria (Lill & Kispal, 2000)

mtHsp70

6

Microsporidial HSP70 have no obviousN-terminal targeting signal

Localisation of a mtHsp70 in Trachipleistophora

Western blot using an antibody to Th-mtHsp70

Infe

cted

cells

Spor

es

Rabb

it

Confocalmicroscopyusing anantibody toTh-mtHsp70

The Hsp70 protein islocalised to membranebound or electron densestructures.

Traditional fixationmethods show a doublemembrane

Microsporidian

Williams et al. 2002 Nature 418: 865-869

Host cell mitochondrion

Microsporidian‘mitochondrial remnant’

Trees are a good way of exploring thehistory of genes but care is needed!

• Making trees is not easy:– Among-site rate heterogeneity, “fastclock” species, shared nucleotide or aminoacid composition biases

– Different data sets may be affected byindividual phenomena to different degrees

– Biases need not be large if phylogeneticsignal is weak

• Phylogenetic analysis is frequently treated as ablack box into which data are fed (often gatheredat considerable cost) and out of which “The Tree”springs

• (Hillis, Moritz & Mable 1996, Molecular Systematics)

Phylogenetic analysis requirescareful thought