Horizontal gene transfer in trypanosomatids

Opinion

TRENDS in Parasitology

Vol.23 No.10

Horizontal gene transfer in trypanosomatids Fred R. Opperdoes and Paul A.M. Michels Research Unit for Tropical Diseases, Christian de Duve Institute of Cellular Pathology and Laboratory of Biochemistry, Universite´ catholique de Louvain, Avenue Hippocrate 74-75, B-1200 Brussels, Belgium

Trypanosomes harbour a large number of structural and biochemical peculiarities. Kinetoplast DNA, mitochondrial RNA editing, the sequestration of glycolysis inside glycosomes and unique oxidative-stress protection mechanisms (to name but a few) are found only in the members of the order Kinetoplastida. Thus, it is not surprising that they have provoked much speculation about why and how such oddities have evolved in trypanosomes. However, the true reasons for their existence within the eukaryotic world are still far from clear. Here, Fred Opperdoes and Paul Michels argue that the trypanosome-specific evolution of novel processes and organization could only have been made possible by the acquisition of a large number of foreign genes, which entered a trypanosomatid ancestor through lateral gene transfer. Many different organisms must have served as donors. Some of them were viruses, and others were bacteria, such as cyanobacterial endosymbionts and non-phototrophic bacteria. Peculiarities of the Trypanosomatidae Trypanosomatidae, a subdivision of the Kinetoplastida, have attracted considerable attention because they harbour a large number of peculiarities not found in other organisms. The kinetoplast, a specialized region of the mitochondrion, contains a large mass of mitochondrial DNA (mtDNA) consisting of several thousand catenated minicircles and several dozen maxicircles. It is the only mtDNA that is visible under the light microscope [1,2]. Moreover, these organisms have peroxisome-like organelles that contain (in addition to the enzymes of b-oxidation of fatty acids and ether-lipid biosynthesis, typical of peroxisomes) several enzymes of the glycolytic pathway, the hexose-monophosphate pathway and the pyrimidine biosynthetic pathway [3,4]. Recently, several cyanobacterial or plant-like genes were found in trypanosomes [5,6]. This observation was surprising because these organisms were not considered to be related to plants and other chloroplast-containing organisms and remnants of plastids have never been observed in the Kinetoplastida. Certain Euglenida, which belong (together with the Kinetoplastida) to the Euglenozoa, have secondary plastids. Therefore, Hannaert et al. [5] proposed that a common ancestor of both the Euglenida

Corresponding author: Opperdoes, F.R. ([email protected]). Available online 7 September 2007. www.sciencedirect.com

and Kinetoplastida would have acquired a secondary plastid through endosymbiosis with a photosynthetic alga. However, recent phylogenetic and ultrastructural analyses of various euglenoid taxa suggest that the euglenoids’ acquisition of the secondary chloroplast must have occurred after, and not before, the separation of Kinetoplastida and Euglenida [7]. In addition, the recent completion of the TriTryp genome-sequencing projects (i.e. projects aimed at sequencing the genomes of Leishmania major, Trypanosoma brucei and Trypanosoma cruzi) has not revealed convincing evidence for the presence of remnants of an alga in the trypanosomatid genome [8]. Nevertheless, in the TriTryp genomes, we find an intriguingly large number of genes that seem to be the result of lateral transfer from a cyanobacterium or chloroplast to a predecessor of trypanosomatids (Table 1). Lateral transfer of plant-like genes A large number of plant-like genes have been identified in the genomes of the three trypanosomatids (see Table 1 for an outline of those found in L. major). (i) ATP-dependent phosphofructokinase (PFK) is related to the PPi-dependent PFKs of some anaerobic bacteria, anaerobic protists and plants [9]. (ii) The glycosomal glyceraldehyde-phosphate dehydrogenase (GAPDH) is related to the cyanobacterial GAPDH1 [10]. (iii) Not only is fructose-bisphosphate aldolase phylogenetically related to the aldolase of chloroplasts, but it also shares a unique active-site residue with all the enzymes that operate in the Calvin cycle in cyanobacteria and chloroplasts [5]. (iv) Another Calvin-cycle enzyme, sedoheptulose-1,7-bisphosphatase (SBPase), was also identified in the genome of T. brucei and T. cruzi, and it has been suggested that this enzyme functions in a modified pentose-phosphate pathway [5]. However, after this was reported, SBPase genes were detected in ciliates as well (F.R.O., unpublished), and it was recently proposed that ciliates also once harboured an algal endosymbiont, which was lost again [11–13]. (v) Like plants, trypanosomatids contain a cofactor-independent phosphoglycerate mutase, and the trypanosomatid and plant enzymes are phylogenetically related [14]. (vi) In all other organisms where this enzyme has been found, it is known to be magnesium dependent; only the trypanosomatid and plant enzymes depend on the presence of cobalt ions for their activity [15]. (vii) All trypanosomatids have one or several biopterin or folate transporters, which so far have only been found in cyanobacteria and in the chloroplasts of plants [16]. (viii) There is also gene-structural evidence for an event of lateral

1471-4922/$ – see front matter ! 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.pt.2007.08.002

Opinion


Vol.23 No.10

471

Table 1a. Leishmania major genes of possible plant or cyanobacterial origin Enzyme name 2,3-bisphosphoglycerate-independent phosphoglycerate mutase 6-phosphogluconate dehydrogenase adenylate kinase (glycosomal) adenylate kinase, putative delta 6 fatty acid desaturase-like fructose-1,6-bisphosphate aldolase glutathione (trypanothione) peroxidase glyceraldehyde 3-phosphate dehydrogenase, glycosomal hypothetical protein, unknown function orotidine-50 -phosphate decarboxylase-orotatephosphoribosyltransferase, probable phosphofructokinase (ATP-dependent) putative proton-translocating inorganic pyrophosphatase sugar transporter-like protein trypanothione reductase

Accession LmjF36.6650 LmjF35.3340 LmjF36.1360 LmjF21.1250 LmjF14.1340 LmjF36.1260 LmjF26.0800 LmjF30.2980 LmjF30.2080 LmjF16.0550 LmjF29.2510 LmjF31.1220 LmjF18.0040 LmjF05.0350

Support S W S S S W W I/S S S W I I S

Table 1b. Leishmania major genes of possible bacterial origin Enzyme name 2,4-dienoyl-coa reductase-like 3-ketoacyl-coa thiolase, possible 5-methyltetrahydropteroyltriglutamate-homocysteine methyltransferase argininosuccinate synthase ATP-NAD kinase biotin/lipoate protein ligase carbonic anhydrase family protein, putative cysteine synthase coproporphyrinogen iii oxidase, putative deoxyribose-phosphate aldolase dephospho-CoA kinase deoxyuridine triphosphatase dUTPase dihydroorotate dehydrogenase ferrochelatase-like protein b-fructosidase fumarate hydratase fumarate hydratase class I, anaerobic gamma-glutamyl-phosphate reductase galactokinase glutamate dehydrogenase (NADP), putative glycerol-3-phosphate dehydrogenase (NAD) homocysteine s-methyltransferase inosine-50 -monophosphate dehydrogenase IMPdh iron superoxide dismutase (SODB1 glycosomal) isocitrate dehydrogenase isopentenyl-diphosphate delta-isomerase l-ribulokinase n-acyl-l-amino acid amidohydrolase phosphoglycerate kinase phosphoglycerate mutase cofactor dependent-like protoporphyrinogen oxidase-like protein pteridine reductase PTR1 putative gamma-glutamyl kinase, probable putative pyruvate/indole-pyruvate decarboxylase ribose 5-phosphate isomerase serine acetyltransferase sucrose hydrolase-like protein tagatose-6-phosphate kinase tagatose-6-phosphate kinase thermostable carboxypeptidase 1 glycosyl hydrolase udp-glc 40 -epimerase xylulokinase, putative

Accession LmjF06.0930 LmjF23.0690 LmjF31.0010 LmjF23.0260 LmjF06.0460 LmjF31.1070 LmjF06.0610 LmjF36.3590 LmjF06.1270 LmjF06.1070 LmjF22.1530 LmjF06.0560 LmjF16.0530 LmjF17.1460 LmjF23.0870 LmjF24.0320 LmjF29.1960 LmjF02.0650 LmjF35.2740 LmjF28.2910 LmjF10.0510 LmjF36.6310 LmjF17.0725 LmjF32.1820 LmjF33.2550 LmjF35.5330 LmjF36.0060 LmjF20.1560 LmjF30.3380 LmjF08.0060 LmjF06.1280 LmjF23.0270 LmjF26.2710 LmjF34.3250 LmjF28.1970 LmjF34.2850 LmjF27.2340 LmjF25.2440 LmjF02.0030 LmjF33.2540 LmjF36.5240 LmjF33.2300 LmjF36.0260

L + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +

Tb + + ! ! + + ! ! ! + + + + ! ! + + ! ! ! + ! + + ! + ! + + + ! + ! ! + ! ! ! ! + + + !

Tc + + + ! + + ! + ! + + + + ! ! + + + + + + ! + + ! + + + + + ! + ! ! ! + ! ! ! + + + !

Support S S S I S S S S S S I S S S S S S S S S S S I I I S I S I S S S S I S S S S S S S S S

Annotations and accession codes are available in GeneDB. All genes listed in Table 1a have been identified in both L. major, T. brucei and T. cruzi. In Table 1b, we have indicated in which of the three trypanosomatids the gene is present by using ‘+’ or ‘!’. Homologous genes were identified by BLAST searches of the trypanosomatid databases (http://www.genedb.org). Possible events of horizontal gene transfer were identified as follows: all protein sequences of the L. major protein databank as available in GeneDB were used for searching the SwissProt database for homologues with the BlastP program and a cut-off value of 1 " e!10. When the Blast output reported either a bacterium or a plant homologue among the best ten hits, the Leishmania sequence was retained and used for searching the Uniref90 and the Uniref50 databases (release 10.4, 1 May, 2007), which are subsets of the Uniprot database, of which each random sequence pair has less than 90% and 50% identity, respectively. The first 30 or 50 sequences reported in the BLAST output with these databases were aligned together with the query sequence using the program ClustalW [32]. Neighbour-joining trees were constructed for each multiple alignment after the removal of all positions with gaps, and correction for multiple substitutions was performed with the tree option of ClustalW. All neighbour-joining trees generated are presented with the program NJPlot. Trees are available at http://www.icp.ucl.ac.be/#opperd/lgt/. Support scores are as follows: S, strong support by the phylogenetic tree (no conflicting taxa identified); I, intermediate support (conflicting taxa at a distance of several branch points); W, weak support (alternative explanations possible). www.sciencedirect.com

472

Opinion

TRENDS in Parasitology Vol.23 No.10

gene transfer from a cyanobacterium to the common ancestor of both bodonids and trypanosomatids. In T. cruzi, five genes (Pyr1, Pyr2, Pyr3, Pyr4, and Pyr6–5 ), encoding all six enzymes involved in de novo pyrimidine biosynthesis, are tandemly linked. This is similar to the gene organization of the Pyr operon of many bacteria (Box 1, Figure I). Moreover, in a phylogenetic analysis each half of the trypanosomatid bifunctional protein orotate decarboxylase/orotidine phosphoribosyltransferase, encoded by Pyr6–5 , robustly clusters with its cyanobacterial homologues (not shown) and the cyanobacterial proteins are, respectively, 46%– 49% identical to their trypanosomatid counterparts,

whereas the eukaryotic OPRTases share only 26%–28% identity. These observations suggest that possibly the entire bacterial Pyr operon entered an ancestral kinetoplastid via a mechanism of horizontal transfer (Box 1). (ix) Finally, the strongest evidence for an event of lateral transfer from a cyanobacterium or chloroplast to a trypanosomatid is the presence of a gene encoding the L. major hypothetical protein LmjF30.2080 (Figure 1). This gene only has homologues in cyanobacteria and chloroplast-bearing eukaryotes, such as algae, diatoms and plants. In algae and diatoms, this gene (YCF45 ) is positioned on the chloroplast genome itself, whereas in plants the gene has moved to the nucleus and

Box 1. A highly unusual pathway of pyrimidine biosynthesis The pyrimidine biosynthetic pathway comprises six enzymes: glutamine-dependent carbamoyl-phosphate synthetase (CPSase), aspartate carbamoyltransferase (ACT), dihydroorotase (DHOase), dihydroorotate dehydrogenase (DHODase), orotate phosphoribosyltransferase (OPTase) and orotidine-50 -phosphate decarboxylase (ODCase) (Figure Ia). The corresponding genes are labelled Pyr1–6 in eukaryotes and PyrA–F in bacteria. In mammals, five of the six enzymes are associated into two different multifunctional proteins, CAD and UMP synthase. A single gene, CAD (CPSase-DHOaseACTase), encodes the equivalents of Pyr1–3-2, and another gene, Pyr5–6, encodes UMP synthase (ORPTase-ODCase). Pyr4 occurs separately and encodes a monofunctional protein with DHODase activity; the protein is present in the mitochondrion, where it delivers its electrons to the respiratory chain. In some bacteria the Pyr genes are scattered over the chromosome, whereas in others they are clustered and organized as an operon. In T. cruzi, the genes Pyr1, Pyr3, Pyr6–5, Pyr2 and Pyr4 (encoding the enzymes of the pyrimidine biosynthetic pathway) are all found on a 25 kb segment of genomic DNA [34]. A similar operon-like organization of the Pyr genes exists in L. major and in T. brucei, but in T. brucei the dihydroorotase is located elsewhere in the genome. Interestingly, in a BlastP search of the UniProt database (http://www.expasy.org), the only other organisms with a fused Pyr6–5 (PyFE) gene in the reversed order are the cyanobacteria: Anabaena sp., Synechocystis sp. and Gloeobacter violaceus. The trypanosomes form a robust clade with these cyanobacteria in phylogenetic-tree reconstruction. A similar analysis, using the ODCase half of the trypanosomatid bifunctional protein, also revealed that this one is most closely related to the cyanobacterial fused Pyr6–5 (PyFE) gene (39% identity). The trypanosomatid Pyr3 and Pyr4 genes also seem to be of bacterial origin [34], but here no specific affiliation with cyanobacteria could be established. By contrast, trypanosomatid Pyr1 and Pyr2 seem to be of a eukaryotic origin [35]. In trypanosomatids, as in mammals, the first three activities of the pathway are cytosolic; however, in trypanosomes the fourth enzyme, DHODase, is present in the cytosol [36] (rather than in the mitochondrion as in other eukaryotes) and utilizes fumarate as an electron acceptor [37]. The ODCase/OPRTase genes of all three trypanosomatids predict a peroxisome-targeting signal and, indeed, the pathway’s final two steps (OPRTase and ODCase) occur inside glycosomes [36]. These observations have been interpreted by Annoura et al. [37] with the following evolutionary scenario (Box 1, Figure Ib). A common ancestor of euglenids and kinetoplastids had a mitochondrial DHODase whose descendant still exists in Euglena gracilis. In the kinetoplastid line of descent, an ancestor of the bodonids and trypanosomatids subsequently acquired a cytosolic DHODase from a bacterium by horizontal gene transfer. Interestingly, in Bodo caudatus the gene has fused with the gene coding for aspartate carbamoyltransferase, whereas this is not the case in the trypanosomatids. It is not clear whether this event of horizontal transfer is the same as the one that led to the acquisition of the other Pyr genes by the Kinetoplastida or whether this occurred in one or more separate events. The acquisition of the DHODase gene then might have contributed to adaptation to anaerobiosis in the kinetoplastid lineage [37] and has allowed such organisms as www.sciencedirect.com

T. brucei and Phytomonas characias to survive and thrive without needing a functional mitochondrial respiratory chain. It is interesting to note that yeast, the only other eukaryote known to have a soluble DHODase, has also adapted to a lifestyle without functional mitochondria when it is involved in alcoholic fermentation.

Figure I. (a) Schematic representation of the genomic organization of the genes involved in the pyrimidine biosynthetic pathway in various organisms. In both cyanobacteria and Trypanosomatidae, the Pyr5 (red) and Pyr6 (blue) genes have been fused in reverse order, whereas in mammals fusion has occurred in forward order. (b) Evolutionary scenario for the acquisition by Kinetoplastida from (cyano)bacteria of genes for the pyrimidine biosynthetic pathway. ACT, aspartate carbamoyltransferase; DHODase, dihydroxyorotate dehydrogenase; mt, mitochondrial; ODCase/OPRTase, orotate decarboxylase/orotidinemonophosphate ribosyltransferase.

Opinion


Vol.23 No.10

473

Figure 1. Phylogenetic tree of YCF45-related proteins, which are found only in cyanobacteria, algae, diatoms, plants and trypanosomatids. The L. major query sequence was used for searching the UniProt database for homologues via the BlastP algorithm. The query sequence together with all homologous sequences in the database were aligned in ClustalW, and after the removal of sites with gaps, a phylogenetic tree was created with the program MrBayes v3.1.2 [33]. Numbers at the nodes represent the probability of the partition. The horizontal bar represents ten accepted mutations per 100 amino acids.

acquired coding information for a chloroplast import signal, indicating that the protein has retained its chloroplast location. Also in trypanosomatids, an N-terminal extension is included, but it is not clear whether this serves in any routing toward an intracellular compartment. The YCF45 homologue contains an AAA-type ATPase and a cytochrome P450 domain, but currently, its function is unknown. Together, the results from phylogenetic tree construction, the presence of a unique Pyr6–Pyr5 fusion common to cyanobacteria and Kinetoplastida, and the presence in trypanosomatids of the chloroplast YCF45 gene homologue provide compelling evidence for the hypothesis that the Kinetoplastida must have once acquired genes from either a cyanobacterium or a plastid. Lateral transfer of bacterial genes The TriTryp genome-sequencing projects have identified several bacterial sugar-kinase genes [17]. Both T. cruzi and L. major possess a ribulokinase and a galactokinase, whereas a xylulokinase is present only in L. major. Leishmania also has a sucrase and a b-fructosidase, which are involved in the cleavage of disaccharides. T. brucei lacks them all. Ribulokinase, xylulokinase, sucrase and fructosidase all share a relatively high percentage of identical residues (50%–60%) with their bacterial counterparts, whereas eukaryotic homologues were not found in the combined SwissProt/Uniprot database (Table 1). This renders it likely that all these sugar-metabolism genes might have originated from bacteria. L. major, but not www.sciencedirect.com

the two trypanosomes, contains two isoenzymes of tagatose-6-phosphate kinase belonging to the fructose-6phosphate kinase B family, which has, so far, only been reported in bacteria. All these sugar-kinase genes predict a glycosome-targeting signal [4]. Inside glycosomes, the presence of so many sugar kinases specific for sugars other than glucose probably reflects trypanosomatids’ early adaptation to the plant-derived, sugar-rich diets of their insect hosts. This adaptation must have been facilitated by multiple events of horizontal gene transfer between bacterial (endo)symbionts present in either the insect midgut or the parasite itself. The subsequent acquisition of peroxisome-targeting signals allowed these enzymes to be redirected to the glycosome, which became a specialized organelle of carbohydrate metabolism. Subsequently, the African trypanosome further adapted to the midgut of the tsetse fly, which feeds solely on glucose-containing blood, and to the glucose-rich tissue fluids of the mammalian host. As a result, the need for the metabolism of other sugars disappeared, and the genes for the enzymes involved in their metabolism were lost again [17]. Trypanosomatids parasitize a great variety of insects with different feeding habits; such insects include bloodfeeding triatomines and mosquitoes; beetles feeding on plant juices only; and sandflies feeding on blood, nectar and aphid honeydew. Moreover, Crithidias, leptomonads and herpetomonads not only inhabit the midgut of a wide variety of insects but also are capable of establishing a transient infection in plants [18]. Phytomonads are

474

Opinion


digenetic parasites that alternate between using insects as hosts and using plants as hosts. The acquisition of several bacterial sugar-kinase genes, and possibly other genes for enzymes with a function in carbohydrate metabolism (Table 1), might explain the wide host range that characterizes the trypanosomatids as a family. Lateral transfer of viral genes For the replication and maintenance of its kinetoplast DNA, the trypanosomatid mitochondrion requires at least six DNA polymerases, five helicases and three DNA ligases

[1,2,19,20], whereas mitochondria of other eukaryotes have only one copy of each. This multiplicity of enzymes is unprecedented and might reflect the challenges encountered by the protein machinery in guaranteeing a faithful replication of the kDNA structure. Moreover, the original mitochondrial polymerase g found in the mitochondria of all other eukaryotes has been lost from the trypanosomatid mitochondrion and replaced by six other DNA polymerases. So where did these enzymes come from? Phylogenetic tree construction provides a partial answer to this question. Figure 2 shows that both the trypanosomatid

Figure 2. Phylogenetic-tree reconstruction of DNA polymerases (DNA polymerase 1 [a]) and DNA ligases (DNA ligase K [b]). A trypanosomatid query sequence was used for searching the UniProt database for homologues with the BlastP algorithm. The query sequence together with the sequences of the 50 best hits were aligned in ClustalW. After the removal of incomplete sequences or fragments, a phylogenetic tree was created with the program MrBayes v3.1.2 [33] after removal of sites with gaps. Numbers at the nodes represent the probability of the partition. www.sciencedirect.com

Opinion


DNA polymerases 1B, 1C and 1D genes and the DNA ligase k-a and k-b genes are most closely related to the corresponding genes of bacteriophages and African swine fever (ASF) virus, with whom they form a single and robust clade. ASF virus is transmitted by arthropods, and its DNA ligase is apparently closely related to that of the Caudovirales viruses, which seem to be the donors of the trypanosomatid DNA polymerase. The Caudovirales comprise viruses of many bacteria (including plant pathogens, such as Xanthomonas spp) and are all transmitted by arthropods, as are the trypanosomatids themselves. Euglenozoa consist of two major divisions: the Euglenida and the Kinetoplastida. Kinetoplastida comprise the polyphyletic group of Bodonina and the monophyletic subdivision Trypanosomatina [21]. Although the members of the Bodonina suborder also have huge kinetoplasts, their DNA consists of loose and large conglomerates of monomeric circles filling most of the mitochondrial lumen [22]. Kinetoplast-like mtDNA packaging by catenation of mini- and maxi-circles seems to be an exclusive characteristic of Trypanosomatina [23]. Therefore, a possible time point for the acquisition of both DNA ligase and DNA polymerase by trypanosomatids should be placed after the separation of the free-living Bodonina and the parasitic Trypanosomatina. Thus, the horizontal transfer of genes from one or more of these viruses to a predecessor of the present-day members of the trypanosomatid family might have occurred inside the midgut of an arthropod host that was feeding on plants. Indeed, another event of horizontal transfer from a plant pathogen, most likely an ancestor of Ralstonia eutropha, has been reported recently in the case of the plant trypanosomatid Phytomonas [24]. After the acquisition of the DNA-polymerase and DNAligase genes by lateral transfer, these genes underwent duplications, leading to the formation of separate Pol1B, Pol1C and Pol1D and ligase k-a and k-b genes (Figure 2), although the possibility of multiple gene acquisitions cannot be excluded. It is only after these events had taken place that trypanosomatids diverged from each other. It is not clear how these events could have triggered kDNA evolution. Possibly, with each gene duplication the complexity of the kDNA network was allowed to increase, which resulted in the complex and catenated kDNA structure that we find in the present-day trypanosomatids. Alternatively, the acquired viral gene products might have induced a catenation of the mtDNA circles, and gene duplications might have been subsequently necessary for dealing with the increased complexity. The transfer of viral genes to the nuclear genome, followed by the acquisition of a transit sequence allowing the redirection of their gene products to the mitochondria, is by no means unique. Also in Trypanosomatidae, the mitochondrial ligase 1 involved in RNA editing seems to have been derived from a T4 phage protein [25,26]. However, this must have been a much earlier event because RNA editing occurs in bodonids as well [27,28]. In addition, very early in the evolution of the eukaryotic cell, the original bacterial-type RNA polymerase that was inherited from the a-proteobacterial ancestor of mitochondria was replaced by an RNA polymerase gene of a T3 or T7-type phage [29]. Apparently, many events of horizontal gene www.sciencedirect.com

Vol.23 No.10

475

transfer involving viral genes have been instrumental in shaping the present trypanosomatid mitochondrion. Concluding remarks The previously reported plant-like traits found in the trypanosomatid genome are most likely the remnants of a very early event of primary endosymbiosis with a photosynthetic bacterium. This event probably took place before the separation of trypanosomatids and bodonids, and it must have led to the formation of glycosomes. In the ancestor of the trypanosomatids, the primary photosynthetic endosymbiont was eventually lost but some of its DNA ended up in the host nucleus. Much later, by an event of secondary symbiosis, the related euglenoids acquired their chloroplasts, which persist today. Trypanosomatids have undergone many more events of lateral gene transfer. They have succeeded in acquiring additional metabolic complexity through multiple acquisitions of genes that are involved in carbohydrate metabolism and that probably originated from many different bacterial donors. Many of the gene products were routed to the glycosomes, which became the specialized organelles unique to these organisms. These events have allowed the trypanosomatids, as a family, to adapt to a vast range of different hosts. The donor bacteria for these enzymes could have been one or more of the endosymbionts that still survive in some of the present-day trypanosomatids, such as Blastocrithidia and Crithidia spp. [30,31], or could have been encountered in the midgut of one or more ancestral insect hosts. Additional events of lateral transfer (involving one or more viruses that were probably picked up in the arthropod midgut and were followed by a series of gene duplications) might have led to the complex architecture of the kDNA complex, which we find in the present-day trypanosomes and which requires multiple polymerases and ligases to replicate. Acknowledgements The authors thank Paul Englund and Ken Stuart for providing information on annotations of mitochondrial proteins, stimulating discussions and the critical reading of the manuscript. This work was supported by an Inter-University Attraction Poles grant from the Belgian government to F.R.O.

References 1 Liu, B. et al. (2005) Fellowship of the rings: the replication of kinetoplast DNA. Trends Parasitol. 21, 363–369 2 Lukes, J. et al. (2005) Unexplained complexity of the mitochondrial genome and transcriptome in kinetoplastid flagellates. Curr. Genet. 8, 1–23 3 Michels, P.A. et al. (2006) Metabolic functions of glycosomes in trypanosomatids. Biochim Biophys Acta 1763, 1463–1477 4 Opperdoes, F.R. and Szikora, J.P. (2006) In silico prediction of the glycosomal enzymes of Leishmania major and trypanosomes. Mol. Biochem. Parasitol. 147, 193–206 5 Hannaert, V. et al. (2003) Plant-like traits associated with metabolism of Trypanosoma parasites. Proc. Natl. Acad. Sci. U.S.A. 100, 1067–1071 6 Leander, B.S. (2004) Did trypanosomatid parasites have photosynthetic ancestors? Trends Microbiol. 12, 251–258 7 Waller, R.F. et al. (2004) More plastids in human parasites? Trends Parasitol. 20, 54–57 8 El-Sayed, N.M. et al. (2005) Comparative genomics of trypanosomatid parasitic protozoa. Science 309, 404–409 9 Michels, P.A. et al. (1997) The glycosomal ATP-dependent phosphofructokinase of Trypanosoma brucei must have evolved from

Opinion

476

10

11 12 13 14

15

16

17 18

19 20 21 22 23 24


an ancestral pyrophosphate-dependent enzyme. Eur J Biochem. 250, 698–704 Fast, N.M. et al. (2001) Nuclear-encoded, plastid-targeted genes suggest a single common origin for apicomplexan and dinoflagellate plastids. Mol. Biol. Evol. 18, 418–426 Bhattacharya, D. et al. (2004) Photosynthetic eukaryotes unite: endosymbiosis connects the dots. Bioessays 26, 50–60 Aury, J.M. et al. (2006) Global trends of whole-genome duplications revealed by the ciliate Paramecium tetraurelia. Nature 444, 171–178 Eisen, J.A. et al. (2006) Macronuclear genome sequence of the ciliate Tetrahymena thermophila, a model eukaryote. PLoS Biol. 4, e286 Chevalier, N. et al. (2000) Trypanosoma brucei contains a 2,3 bisphosphoglycerate independent phosphoglycerate mutase. Eur. J. Biochem. 267, 1464–1472 Guerra, D.G. et al. (2004) Characterization of the cofactor-independent phosphoglycerate mutase from Leishmania mexicana mexicana. Histidines that coordinate the two metal ions in the active site show different susceptibilities to irreversible chemical modification. Eur. J. Biochem. 271, 1798–1810 Klaus, S.M. et al. (2005) Higher plant plastids and cyanobacteria have folate carriers related to those of trypanosomatids. J. Biol. Chem. 280, 38457–38463 Berriman, M. et al. (2005) The genome of the African trypanosome Trypanosoma brucei. Science 309, 416–422 Catarino, L.M. et al. (2001) Classification of trypanosomatids from fruits and seeds using morphological, biochemical and molecular markers revealed several genera among fruit isolates. FEMS Microbiol. Lett. 201, 65–72 Downey, N. et al. (2005) Mitochondrial DNA ligases of Trypanosoma brucei. Euk. Cell 4, 765–774 Klingbeil, M.M. et al. (2002) Multiple mitochondrial DNA polymerases in Trypanosoma brucei. Mol Cell. 10, 175–186 Simpson, A.G. et al. (2006) The evolution and diversity of kinetoplastid flagellates. Trends Parasitol. 22, 168–174 Lukes, J. et al. (2002) Kinetoplast DNA network: evolution of an improbable structure. Eukaryot Cell 1, 495–502 Roy, J. et al. (2007) Unusual mitochondrial genome structures throughout the euglenozoa. Protist 158, 385–396 Molinas, S.M. et al. (2003) The multifunctional isopropyl alcohol dehydrogenase of Phytomonas sp. could be the result of a horizontal

25

26 27

28 29 30

31 32

33

34

35 36

37

gene transfer from a bacterium to the trypanosomatid lineage. J. Biol. Chem. 278, 36169–36175 Panigrahi, A.K. et al. (2001) Association of two novel proteins, TbMP52 and TbMP48, with the Trypanosoma brucei RNA editing complex. Mol. Cell Biol. 21, 380–389 Ho, C.K. et al. (2004) Structure and mechanism of RNA ligase. Structure 12, 327–339 Lukes, J. et al. (1994) Novel pattern of editing regions in mitochondrial transcripts of the cryptobiid Trypanoplasma borreli. EMBO J. 13, 5086–5098 Blom, D. et al. (1998) RNA editing in the free-living bodonid Bodo saltans. Nucleic Acids Res. 26, 1205–1213 Hedtke, B. et al. (1997) Mitochondrial and chloroplast phage-type RNA polymerases in Arabidopsis. Science 277, 809–811 Du, Y., Maslow, D.A. and Chang, K.P. (1994) Monophyletic origin of beta-division of proteobacterial endosymbionts and their coevolution with insect trypanosomatid protozoa Blastocrithidia culicis and Crithidia spp. Proc. Natl. Acad. Sci. U. S. A. 91, 8438–8441 de Souza, W. and Motta, M.C. (1999) Endosymbiosis in protozoa of the Trypanosomatidae family. FEMS Microbiol Lett. 173, 1–8 Thompson, J.D. et al. (1994) Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positionspecific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673–4680 Ronquist, F. and Huelsenbeck, J.P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 12, 1572–1574 Gao, G. et al. (1999) Novel organization and sequences of five genes encoding all six enzymes for de novo pyrimidine biosynthesis in Trypanosoma cruzi. J. Mol. Biol. 285, 149–161 Nara, T. et al. (2000) Evolutionary implications of the mosaic pyrimidine-biosynthetic pathway in eukaryotes. Gene 257, 209–222 Hammond, D.J. et al. (1981) A novel location for two enzymes of de novo pyrimidine biosynthesis in trypanosomes and Leishmania. FEBS Lett. 128, 27–29 Annoura, T. et al. (2005) The origin of dihydroorotate dehydrogenase genes of kinetoplastids, with special reference to their biological significance and adaptation to anaerobic, parasitic conditions. J. Mol. Evol. 60, 113–127

AGORA initiative provides free agriculture journals to developing countries The Health Internetwork Access to Research Initiative (HINARI) of the WHO has launched a new community scheme with the UN Food and Agriculture Organization. As part of this enterprise, Elsevier has given hundreds of journals to Access to Global Online Research in Agriculture (AGORA). More than 100 institutions are now registered for the scheme, which aims to provide developing countries with free access to vital research that will ultimately help increase crop yields and encourage agricultural self-sufficiency. According to the Africa University in Zimbabwe, AGORA has been welcomed by both students and staff. "It has brought a wealth of information to our fingertips", says Vimbai Hungwe. ‘‘The information made available goes a long way in helping the learning, teaching and research activities within the University. Given the economic hardships we are going through, it couldn’t have come at a better time.’’

For more information, visit www.aginternetwork.org www.sciencedirect.com