International Journal for Parasitology 34 (2004) 1393–1404 www.parasitology-online.com Trypanosomes are monophyletic: evidence from genes for glyceraldehyde phosphate dehydrogenase and small subunit ribosomal RNA* Patrick B. Hamiltona, Jamie R. Stevensb, Michael W. Gauntc, Jennifer Gidleya, Wendy C. Gibsona,* aSchool of Biological Sciences, University of Bristol, Bristol BS8 1UG, UK bSchool of Biological Sciences, University of Exeter, Exeter EX4 4PS, UK cPathogen Molecular Biology and Biochemistry Unit, Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK Received 5 July 2004; received in revised form 26 August 2004; accepted 26 August 2004 Abstract The genomes of Trypanosoma brucei, Trypanosoma cruzi and Leishmania major have been sequenced, but the phylogenetic relationships of these three protozoa remain uncertain. We have constructed trypanosomatid phylogenies based on genes for glycosomal glyceraldehyde phosphate dehydrogenase (gGAPDH) and small subunit ribosomal RNA (SSU rRNA). Trees based on gGAPDH nucleotide and amino acid sequences (51 taxa) robustly support monophyly of genus Trypanosoma, which is revealed to be a relatively late-evolving lineage of the family Trypanosomatidae. Other trypanosomatids, including genus Leishmania, branch paraphyletically at the base of the trypanosome clade. On the other hand, analysis of the SSU rRNA gene data produced equivocal results, as trees either robustly support or reject monophyly depending on the range of taxa included in the alignment. We conclude that the SSU rRNA gene is not a reliable marker for inferring deep level trypanosome phylogeny. The gGAPDH results support the hypothesis that trypanosomes evolved from an ancestral insect parasite, which adapted to a vertebrate/insect transmission cycle. This implies that the switch from terrestrial insect to aquatic leech vectors for fish and some amphibian trypanosomes was secondary. We conclude that the three sequenced pathogens, T. brucei, T. cruzi and L. major, are only distantly related and have distinct evolutionary histories. q 2004 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Trypanosoma; Trypanosomatidae; Phylogeny; 18S rRNA; Evolution; GAPDH 1. Introduction human African trypanosomiasis or sleeping sickness, while Trypanosoma cruzi causes Chagas disease in South and Trypanosomes (genus Trypanosoma) are widespread Central America. The genus Trypanosoma is in the phylum blood parasites of vertebrates, usually transmitted by Euglenozoa (Eukaryota; Excavata), which comprises three arthropod or leech vectors. Several trypanosome species orders, Diplonemida, Euglenida and Kinetoplastida, and a are agents of disease in humans and/or livestock particularly taxon of uncertain placement, Postgaardi (Cavalier-Smith, in the tropics. For example, Trypanosoma brucei causes 1993; Simpson, 1997). Lifestyles within the Euglenozoa range from autotrophic photosynthesizers, such as Euglena * gracilis, to free-living heterotrophs, such as Bodo saltans, Note: Nucleotide sequence data reported in this paper are available in GenBank, EMBL and DDBJ databases under accession numbers: gGAPDH and facultative or obligate parasites, such as genus gene sequences—AJ620245, AJ620247, AJ620251–AJ620253, AJ620255– Trypanosoma (Sleigh, 1989). AJ620264, AJ620266–AJ620270, AJ620272, AJ620273, AJ620275– Trypanosomes are an easily recognizable group, because AJ620278, AJ620280–AJ620291, AJ620293. SSU rRNA gene sequences: they all share vertebrate parasitism and have a characteristic AJ620547, AJ620548, AJ620555, AJ620557, AJ620564. * Corresponding author. Tel.: C44 117 928 8249; fax: C44 117 925 morphology—the trypomastigote form—in the vertebrate 7374. bloodstream. Two different evolutionary origins for trypa- E-mail address: [email protected] (W.C. Gibson). nosomes have been proposed: vertebrate first, whereby they 0020-7519/$30.00 q 2004 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijpara.2004.08.011 1394 P.B. Hamilton et al. / International Journal for Parasitology 34 (2004) 1393–1404 evolved from gut parasites of vertebrates, or invertebrate and several gene sequences are available. In the three first, whereby they evolved from an invertebrate trypano- trypanosomatids studied [T. brucei (Michels et al., 1986); somatid parasite (Hoare, 1972; Vickerman, 1994). Central T. cruzi (Kendall et al., 1990); Leishmania mexicana to the resolution of these theories is the issue of whether (Hannaert et al., 1992)], there are three GAPDH genes, two trypanosomes had a single evolutionary origin and/or gave of which encode the glycosomal enzyme (gGAPDH), while rise to any other trypanosomatids, i.e. are they monophy- the other encodes the cytosolic enzyme (cGAPDH) (Michels letic? The first molecular phylogenetic studies, based on et al., 1986; Kendall et al., 1990; Hannaert et al., 1992, 1998). comparisons of genes encoding mitochondrial and nuclear Cytosolic GAPDH genes are more closely related to bacterial ribosomal RNAs (rRNA), showed trypanosomes to be GAPDH genes than eukaryotic GAPDH genes and thus form paraphyletic (Gomez et al., 1991; Fernandes et al., 1993; a separate lineage (Wiemer et al., 1995; Hannaert et al., Landweber and Gilbert, 1994; Lukes et al., 1994; Maslov 1998). The two gGAPDH genes from T. brucei, T. cruzi and and Simpson, 1995; Maslov et al., 1996). However, the L. mexicana are in a tandem repeat and are identical in inclusion of more taxa from a broader range of host species sequence (Michels et al., 1986; Kendall et al., 1990; Hannaert in subsequent studies based on rRNA genes provided et al., 1992). Likewise the bodonid Trypanoplasma borreli support for monophyly (Lukes et al., 1997; Haag et al., has two gGAPDH genes in a tandem repeat, but the two 1998; Stevens et al., 1998, 1999, 2001; Wright et al., 1999; predicted proteins differ by 17 amino acids (5%) (Wiemer et Simpson et al., 2002), as did studies based on protein-coding al., 1995). The euglenid, E. gracilis, also possesses two genes (Hannaert et al., 1992; Hashimoto et al., 1995; GAPDH enzymes, GapA, involved in the Calvin-cycle in the Wiemer et al., 1995; Alvarez et al., 1996; Adje´ et al., 1998; chloroplasts and GapC in glycolysis in the cytosol (Hallick et Hannaert et al., 1998; Simpson et al., 2002). al., 1993). The GapA gene is similar to Gap2, a gene found in Doubt has now been cast on this consensus by a re- cyanobacteria (Hallick et al., 1993), and no homologue has analysis of SSU rRNA gene sequences (Hughes and been detected in kinetoplastids (Wiemer et al., 1995). Piontkivska, 2003b). Hughes and Piontkivska contend that Although E. gracilis has no glycosomes, the GapC gene is previous SSU rRNA gene trees do not adequately prove orthologous to trypanosomatid gGAPDH (Henze et al., monophyly of trypanosomes, because they either include an 1995). gGAPDH gene orthologues have not been found in inadequate number and selection of taxa, or are rooted representative diplonemids (two species of Diplonema and inappropriately (Hughes and Piontkivska, 2003b). In recent one species of Rhynchopus)(Qian and Keeling, 2001). SSU rRNA gene trees, trypanosomes and trypanosomatids Thus gGAPDH would appear to be a suitable candidate appear paraphyletic (Hughes and Piontkivska, 2003a,b); in gene for reconstruction of trypanosome phylogeny, as it particular, placement of Trypanosoma vivax is problematic evolves slowly and under a different set of evolutionary and it appears outside the main group of trypanosome constraints to the SSU rRNA gene. In addition, several gene species in some trees. This conflicts with the previous sequences are already available from previous studies. Here consensus on the taxonomic position of T. vivax, in which it we have used gGAPDH data to re-examine the questions of is firmly established as part of the African tsetse-transmitted monophyly of trypanosomes and their evolutionary origins: group characterised by antigenic variation (Hoare, 1972; ‘invertebrate first’ or ‘vertebrate first’. Gardiner, 1989). Although existing phylogenies based on protein-coding genes show trypanosomes to be monophy- letic (Hannaert et al., 1992, 1998; Hashimoto et al., 1995; 2. Materials and methods Wiemer et al., 1995; Alvarez et al., 1996; Adje´ et al., 1998), they include too few taxa to be reliable (Hughes and 2.1. Sequence analysis Piontkivska, 2003a). Most of the previous phylogenies of kinetoplastids have SSU rRNA and gGAPDH genes were amplified from been based on analysis of variation in SSU rRNA genes, trypanosome DNA by PCR. Details of trypanosome strains multi-copy genes that evolve by concerted evolution. As and origins are listed in Table 1. SSU rRNA gene PCR and single-copy, protein-coding genes are under a very different sequencing was as described (Stevens et al., 1999). The set of evolutionary constraints, analysis of such genes is gGAPDH gene was amplified with the primers shown in likely to complement analysis based on the SSU rRNA gene. Table 2. Degenerate primers G3–G7 were designed from an Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) is a alignment of the gGAPDH sequences of Crithidia fascicu- ubiquitous and essential glycolytic enzyme and GAPDH lata AF047493, L. mexicana (X65226) and T. brucei genes have a slow rate of molecular evolution making them (X59955). Expand High Fidelity PCR System (Roche) suitable for studying evolution over large time-scales was used for all