Proc. Natl. Acad. Sci. USAVol. 96, pp. 580–585, January 1999EvolutionMicrosporidia are related to Fungi: Evidence from the largestsubunit of RNA polymerase II and other proteinsROBERT P. HIRT*†,JOHN M. LOGSDON,JR.†‡,BRYAN HEALY*, MICHAEL W. DOREY‡,W.FORD DOOLITTLE‡,AND T. MARTIN EMBLEY*§*Department of Zoology, The Natural History Museum, London SW7 5BD, United Kingdom; and‡Canadian Institute for Advanced Research, Program inEvolutionary Biology and Department of Biochemistry, Halifax, NS, B3H 4H7, CanadaEdited by Estella B. Leopold, University of Washington, Seattle, WA, and approved December 1, 1998 (received for review August 12, 1998)ABSTRACT We have determined complete gene se-quences encoding the largest subunit of the RNA polymeraseII (RBP1) from two Microsporidia, Vairimorpha necatrix andNosema locustae. Phylogenetic analyses of these and otherRPB1 sequences strongly support the notion that Microspo-ridia are not early-diverging eukaryotes but instead arespecifically related to Fungi. Our reexamination of elongationfactors EF-1aand EF-2 sequence data that had previouslybeen taken as support for an early (Archezoan) divergence ofthese amitochondriate protists show such support to be weakand likely caused by artifacts in phylogenetic analyses. TheseEF data sets are, in fact, not inconsistent with a Microsporidia1 Fungi relationship. In addition, we show that none of theseproteins strongly support a deep divergence of Parabasaliaand Metamonada, the other amitochondriate protist groupscurrently thought to compose early branches. Thus, the phy-logenetic placement among eukaryotes for these protist taxais in need of further critical examination.Microsporidia are highly specialized eukaryotic unicells, livingonly as obligate intracellular parasites of other eukaryotes (1).They lack the mitochondria and peroxisomes typical of mosteukaryotes. Thus, in 1983 Cavalier-Smith (2) included theMicrosporidia with other amitochondriate protists, Parabasa-lia (e.g., Trichomonas), Metamonada (e.g., Giardia), and Ar-chamoebae (e.g., Entamoeba) in the kingdom Archezoa. Theseprotists were presumed to have diverged from other eu-karyotes before the acquisition of mitochondria and weresuggested as the earliest eukaryotic lineages.Phylogenetic trees based initially on small-subunit ribosomalRNA (SSUrRNA) (3) and then on protein translation elon-gation factor (EF-1aand EF-2) (4, 5) sequences showed thatMicrosporidia indeed diverged early, along with the Parabasa-lia and Metamonada. Thus, these data apparently confirmedthe archezoal hypothesis in general and what we call the‘‘Microsporidia-early’’ hypothesis in particular—Archamoe-bae were eliminated from the Archezoa when their SSUrRNAsneither placed them together nor as early branches (6).However, more recent trees constructed from tubulin (7, 8)and Hsp70 data (9, 10) placed Microsporidia in the eukaryotic‘‘crown,’’ favoring a position within, or as the sister group to,Fungi. Although the support for the ‘‘Microsporidia 1 Fungi’’hypothesis (or M 1 F) from the Hsp70 data is not compelling[when judged by maximum-likelihood (ML) difference tests orother support values] (9, 10), support froma-tubulin is firm(7). If Microsporidia are truly related to Fungi, then theirsimplified cell structures and small genomes are degeneratefeatures permitted, or perhaps encouraged, by their parasiticlifestyles.Here we report sequences for the largest subunit of RNApolymerase II (RPB1) from two Microsporidia, Vairimorphanecatrix and Nosema locustae. Several phylogenetic methodsapplied to these and other RPB1 sequences strongly support aclose relationship of Microsporidia and Fungi. A reanalysis ofthe apparently conflicting EF data show that the support thatthese sequences lend to a deeply diverging Microsporidia isweak and attributable to artifacts. Furthermore, the EF-1agene from the microsporidian Glugea plecoglossi (5) carries aninsertion encoding 11 amino acids that is otherwise only foundin, and is diagnostic for, the EF-1agenes of Fungi and Metazoa(animals) (11), which form a clade based on other criteria (6, 11).An alternative hypothesis to reconcile the apparently con-flicting gene trees and to allow Microsporidia to retain theirearly status is that they might have borrowed genes from hostgenomes (6). We conclude that such chimeric theories are notnecessary because there is no significant conflict betweenthese proteins concerning the position of Microsporidia.Where a convincing phylogenetic signal is present, it relatesthem to Fungi.Numerous recent reports indicate that the Microsporidia,Metamonada, and Parabasalia possess nuclear genes ofa-pro-teobacterial provenance whose products normally (in mito-chondriate eukaryotes) serve mitochondrial functions; thepresence of such genes indicates mitochondrial loss from theseprotists (9, 10, 12–16). Although the ancestral presence ofmitochondria does not itself preclude a deep divergence of anyof these taxa, our analyses of RPB1 and reanalyses of EF data(in addition to challenging the deep placement of Microspo-ridia) show that the support for an early divergence forTrichomonas and Giardia is also weaker than generally sup-posed. The inferred early branching positions of Metamonadaand Parabasalia largely depend on a single data set, i.e.,SSUrRNA. The Microsporidia thus may be the only memberof the (now former) Archezoa (sensu Cavalier-Smith, ref. 17)about whose phylogenetic position we now have confidencebased on multiple molecular data sets.MATERIALS AND METHODSIsolation of Genomic Clones. Two sets of oligonucleotideprimers—RPB1-F1 (cgGACTTYGAYGGNGAYGARATG-The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked ‘‘advertisement’’ inaccordance with 18 U.S.C. §1734 solely to indicate this fact.PNAS is available online at www.pnas.org.This paper was submitted directly (Track II) to the Proceedings office.Abbreviations: BP, bootstrap proportions; CSR, constant-site re-moval; ISR, invariant-site removal; CTD, C-terminal domain; EF,elongation factor (protein translation); LBP, local bootstrap propor-tions; ML, maximum likelihood; MP, maximum parsimony; RPB1,RNA polymerase II largest subunit; SSUrRNA, small subunit ribo-somal RNA; M 1 F, Microsporidia 1 Fungi; FSR, fast-site removal.Data deposition: The sequences reported in this paper have beendeposited in the GenBank database (accession nos.