ABSTRACT Since most of the examples of “exon shuffling” are between vertebrate genes, the view is often expressed that exon shuffling is limited to the evolutionarily recent lineage of vertebrates. Although exon shuffling in plants has been inferred from the analysis of intron phases of plant genes [Long, M., Rosenberg, C. & Gilbert, W. (1995) Proc. Nati. Acad. Sci. USA 92, 12495-12499] and from the comparison of two functionally unknown sunflower genes [Domon, C. & Steinmetz, A. (1994) Mol. Gen. Genet. 244, 312-317], clear cases of exon shuffling in plant genes remain to be uncovered. Here, we report an example of exon shuffling in two important nucleus-encoded plant genes: cytosolic glyceraldehyde-3-phosphate dehydrogenase (cytosolic GAPDH or GapC) and cytochrome cl precursor. The intron-exon structures of the shuffled region indicate that the shuffling event took place at the DNA sequence level. In this case, we can establish a donor-recipient relationship for the exon shuffling. Three amino terminal exons of GapC have been donated to cytochrome ci, where, in a new protein environment, they serve as a source of the mitochondrial targeting function. This finding throws light upon an old important but unsolved question in gene evolution: the origin of presequences or transit peptides that generally exist in nucleus-encoded organelle genes.

This finding also shows a clear role for exon shuffling in the origin of presequences or transit peptides in the nuclearly encoded organellar proteins, as speculated previously (13, 14). Cytochrome cl is a nuclearly encoded enzyme in eukaryotes. Although the presequences of human and plant cytochrome cl are unrelated, the intron patterns suggest a common ancestral nuclear gene for the mature protein. Of five introns in this region of the two genes, one is identical in position (position 1), a second pair lies within two amino acids (position 2), and a third pair lies within five amino acids (position 3) (Fig. 2b). The intron patterns suggest the possibilities of a common ancestral nuclear gene or a common ancestral transfer from the mitochondrion to the nucleus. The ancestral cytochrome cl gene in plants must have been targeted to the mitochondrion; thus this targeting sequence was replaced in the line leading to the potato by the GapC gene. This replacement may have been selected by some advantage in using the GapC promoter.

Although the origin of the targeting sequence is clear, the initial function is not, since it does not seem to have a targeting function in the donor:

Does the donor shuffled sequence from GapC have an organelle targeting function? Although the shuffled sequence exists in all GAPDH genes, including GapA and GapB that are chloroplast-specific, both GapA and GapB have additional N-terminal elements, commonly believed to be responsible for the transit of the proteins to the organelle. Furthermore, GapC does not require organelle targeting because it functions in the cytosol. If some GapC were to occur in the mitochondrion, the “presequence” in the donor gene would have served as an organelle targeter. However, the existence of GapC proteins in an organelle has not been reported to date although GapC genes are probably of mitochondrial origin (25, 30), and the general belief is that all GapC is cytoplasmic. Thus we suggest that the donor shuffled sequence is important but not sufficient for organelle targeting activity.

For me, the basic results are 1) MRP and CRP exhibit extensive homology, 2) they show conserved intron positions in line with a common descent, 3) MRPs and CRP between themselves contain considerable number of conserved introns.

From the abstract:  In this paper, we tried to gain insights into intron evolution from a novel perspective by comparing the gene structures of cytoplasmic ribosomal proteins (CRPs) and mitochondrial ribosomal proteins (MRPs), which are held to be of archaeal and bacterial origin, respectively. We analyzed 25 homologous pairs of CRP and MRP genes that together had a total of 527 intron positions. We found that all 12 of the intron positions shared by CRP and MRP genes resulted from parallel intron gains and none could be considered to be “conserved,” i.e., descendants of the same ancestor. This was supported further by the high frequency of proto-splice sites at these shared positions; proto-splice sites are proposed to be sites for intron insertion. Although we could not definitively disprove that spliceosomal introns were already present in the last universal common ancestor, our results lend more support to the idea that introns were gained late. At least, our results show that MRP genes were intronless at the time of endosymbiosis. The parallel intron gains between CRP and MRP genes accounted for 2.3% of total intron positions, which should provide a reliable estimate for future inferences of intron evolution.

Here you can see the several assumptions that are made. Interestingly, they say they only need one clear example of conserved intron positition. They find >10, but then do a parsimony analyses that syas that they are not real conserved intron position.

Mitochondrial ribosomes are considered to be of bacterial origin (that is, they are a product of endosymbiosis), as evidenced by the considerable homology that exists between mitochondrial ribosomal proteins (MRPs) and bacterial RPs [21,22]. Cytoplasmic ribosomal proteins (CRPs), on the other hand, are thought to have evolved independently from archaea, although there is sufficient homology between MRPs and CRPs to allow a comparison of their gene structures. Like most mitochondrial genes, MRP genes were transferred to the nuclear genome after endosymbiosis [23] and, like their cellular counterparts, contain spliceosomal introns. Thus, by comparing the intron/exon structures of MRP and CRP genes, it may be possible to determine whether spliceosomal introns existed in their last common ancestor. If at least one clear case of intron position conservation is found (i.e., introns at this position are descendants of the same ancestral intron), then it can be concluded that spliceosomal introns existed in the last common ancestor of CRP and MRP genes (introns-early).

A total of 79 MRP genes were found in the human genome [21,22]. Of these, 43 were homologous to bacterial genes, and among these 43, 25 were homologous to eukaryotic CRP genes. We compared the gene structures of these 25 homologous pairs. The multiple sequence alignments showed that, out of the total of 527 intron positions, only 12 (2.3%) were shared by CRP and MRP genes; i.e., an intron was present at each of these positions in at least one CRP gene and in at least one MRP gene (Figure 1B, Figure S1, and Dataset S1).

Here they basically state that there sample is too small. Also, I cannot relate to the parsimony approach since I don’t think evolution is parsimonious and it is a highly anthropomorf and teleological term.

The maximum parsimony method was used to infer the most parsimonious scenarios of these nine shared positions. We recently proposed a maximum likelihood approach for inferring the evolution of introns; however, we believe that maximum parsimony is the best choice for the current dataset because maximum likelihood uses only patterns of intron position in the conserved regions of the multiple sequence alignment, and our dataset is not large enough to make valid statistical inferences using this method.

Figure 4 has made it to my gallery of Trees of Life, and a nice illustration of how people deal with evolution.

From the abstract: Unlike most organisms, the mitochondrial DNA (mtDNA) of Chlamydomonas reinhardtii, a green alga, does not encode subunit 6 of F0F1-ATP synthase. We hypothesized that C. reinhardtii ATPase 6 is nucleus encoded and identified cDNAs and a single-copy nuclear gene specifying this subunit (CrATP6, with eight exons, four of which encode a mitochondrial targeting signal). Although the algal and human ATP6 genes are in different subcellular compartments and the encoded polypeptides are highly diverged, their secondary structures are remarkably similar. When CrATP6 was expressed in human cells, a significant amount of the precursor polypeptide was targeted to mitochondria, the mitochondrial targeting signal was cleaved within the organelle, and the mature polypeptide was assembled into human ATP synthase. In spite of the evolutionary distance between algae and mammals, C. reinhardtii ATPase 6 functioned in human cells, because deficiencies in both cell viability and ATP synthesis in transmitochondrial cell lines harboring a pathogenic mutation in the human mtDNA-encoded ATP6 gene were overcome by expression of CrATP6.

It looks like no mitochondrial gene is essential for mitochondrion function and they can all be nuclear-encoded. The hydrophobicity problem could have been the main driver for transfer to the mitochondria from the nucleus.

With but few exceptions, a “canonical” set of mtDNA-encoded proteins, typically six subunits of NADH dehydrogenase ubiquinone oxidoreductase, the cytochrome b subunit of ubiquinone-cytochrome c oxidoreductase, three subunits of cytochrome c oxidase, and two subunits of ATP synthase, is remarkably conserved among all species examined, ranging from the protist Reclinomonus americana to mammals. Although the reason for this conservation has been the subject of speculation, the most widely held view is that these proteins are so hydrophobic that they are unable to be imported from the cytoplasm, and therefore this set was constrained by evolutionary pressure to remain in the mitochondrial genome (Claros et al., 1995 , 1996 ; Perez-Martinez et al., 2000 , 2001 ).

However, the rules determining which hydrophobic protein genes are retained in the mtDNA are not hard and fast. For example, the mtDNA of the yeasts S. cerevisiae and Schizosaccharomyces pombe contain the ATP9 gene, which encodes subunit c of ATP synthase, whereas subunit c is nucleus encoded in mammals. Even more striking has been the finding that some organisms among the algae, ciliates, apicomplexans, and flowering plants lack mtDNA-encoded COX II, COX III, and/or ATP6, all highly hydrophobic proteins. Because these three subunits are necessary for the functioning of COX and ATP synthase, respectively, it is almost certain that these genes have been transferred to the nuclear DNA in these organisms. In fact, the nucleus-encoded genes specifying COX II and COX III (Perez-Martinez et al., 2000 , 2001 ; Watanabe and Ohama, 2001 ), and, as reported herein and elsewhere (Funes et al., 2002 ), ATPase 6, have been identified in algal species, including C. reinhardtii.

From the Abstract. In the unicellular green alga, Chlamydomonas reinhardtii, cytochrome oxidase subunit 2 (cox2) and 3 (cox3) genes are missing from the mitochondrial genome. We isolated and sequenced a BAC clone that carries the whole cox3 gene and its corresponding cDNA. Almost the entire cox2 gene and its cDNA were also determined. Comparison of the genomic and the corresponding cDNA sequences revealed that the cox3 gene contains as many as nine spliceosomal introns and that cox2 bears six introns. Putative mitochondria targeting signals were predicted at each N terminal of the cox genes. These spliceosomal introns were typical GT–AG-type introns, which are very common not only in Chlamydomonas nuclear genes but also in diverse eukaryotic taxa. We found no particular distinguishing features in the cox introns. Comparative analysis of these genes with the various mitochondrial genes showed that 8 of the 15 introns were interrupting the conserved mature protein coding segments, while the other 7 introns were located in the N-terminal target peptide regions. Phylogenetic analysis of the evolutionary position of C. reinhardtii in Chlorophyta was carried out and the existence of the cox2 and cox3 genes in the mitochondrial genome was superimposed in the tree. This analysis clearly shows that these cox genes were relocated during the evolution of Chlorophyceae. It is apparent that long before the estimated period of relocation of these mitochondrial genes, the cytosol had lost the splicing ability for group II introns. Therefore, at least eight introns located in the mature protein coding region cannot be the direct descendant of group II introns.

Both genetically and biochemically, group I introns are rather special. At the RNA level, they are inherently autocatalytic, mediating their own removal from transcripts containing them and effecting the ligation of flanking exons (self-splicing) (1). Many group I introns are mobile elements, able to spread in genetic crosses to alleles that do not contain them via a process known as intron homing (2). Homing is initiated by a site-specific endonuclease encoded by the intron. Additionally, several group I introns specify protein cofactors (maturases) that function in the splicing of the intron RNA that encodes them (1, 2). Although a given group I intronic reading frame almost always specifies either endonuclease or maturase activity, there are a few cases known in which the encoded protein can perform both functions (3, 4).

To date, most studies of group I intron mobility have dealt with intraorganismal transfer occurring between intron+ and intron alleles of particular genes during genetic crosses. Little is known about the frequency and extent of horizontal transfer of group I introns between organisms that do not mate. Most such identified cases involve transfer into the same genome (e.g., mitochondrial) in taxa that at least belong to the same phylum (see ref. 5). There is, however, one reported instance in which interphylum (and interorganellar) transfer of group I introns appears to have occurred (6). Even so, nothing remotely approaching the extraordinary intron radiation reported by Cho et al. in this issue of the Proceedings (5) has been documented previously.

This is the commentary on another article that showa that a common intron position between a fungal nuclear (?) donor and a plant mitochondrial gene. This is interpreted as wide-spread intron insertion but can be easily explained by a loss in most but a few, and in more agreement with an original gene that possessed introns.

What these workers have uncovered is an explosive invasion of plant mitochondrial DNA (mtDNA) by a particular group I intron. The authors were led to the present study by a previous finding (7) of this curious intron in the gene encoding subunit 1 of cytochrome oxidase (cox1) in the mtDNA of an angiosperm (flowering plant), Peperomia polybotrya. Not only is this the sole group I intron so far reported in the mtDNA of vascular plants (in contrast to the frequent presence of group II introns in plant mtDNA), it clearly is of a different evolutionary origin than the gene in which it resides. In fact, phylogenetic evidence suggests that this intron arose recently by horizontal transfer from a fungal donor species (7). In the initial study (7), the intron was not found in cox1 from 19 other diverse plant species, and a follow-up investigation (8) indicated that it was restricted to the single genus Peperomia within the order Piperales. […] From this survey, the authors infer 32 separate cases of intron acquisition among the 278 genera and 281 species of angiosperm examined in the botanical equivalent of a “zoo blot.” Extrapolating to angiosperms as a whole, Cho et al. (5) come to the startling conclusion that this intron has invaded the cox1 gene >1,000 times among the >13,500 genera and >300,000 species of extant flowering plants.

Here a nice example of circular reasoning: the article proposes that extensive intron homing is necessary, but for intron homing, the genes need to be in the same physical location, but they are not. So instead of concluding that intron homing is not possible, he concludes that the genes had to be in physical proximity since intron homing was observed and simply calls this promiscuity. And he continues with ‘that being the case …’.

(ii) How does horizontal transfer actually take place at the cellular level? For intron homing to occur, both the intron-containing donor DNA and the intronless recipient DNA must be in the same physical location; the intron+ DNA must be satisfactorily transcribed; and the encoded endonuclease must be correctly translated. Because cox1 is encoded in the mitochondrial genome, this implies that the cox1 intron homing described by Cho et al. (5) takes place within mitochondria. That being the case, the mitochondrial transcription and translation systems…

The mitochondrial origin of mitosome is supported by (i) presence of two membranes surrounding these organelles, (ii) localization of proteins of iron sulfur cluster assembly machinery within these organelles (IscU, IscS, ferredoxin), (iii) targeting of proteins into the mitosome by means of N-terminal leader sequences with similar properties of mitochondrial leader sequences, (iv) FeS cluster assembly activity identified in mitosome-rich cell fraction.

From the abstract: Phylogenetic analysis showed a close relationship among all eukaryotic IscS genes including those of amitochondriates. IscS of proteobacteria formed a sister group to the eukaryotic clade, suggesting that isc-related genes were present in the proteobacterial endosymbiotic ancestor of mitochondria and hydrogenosomes. NifS genes of nitrogen-fixing bacteria, which are IscS homologs required for specific formation of FeS clusters in nitrogenase, formed a more distant group. The phylogeny indicates the presence of a common mechanism for FeS cluster formation in mitochondriates as well as in amitochondriate eukaryotes. Furthermore, the analyses support a common origin of Trichomonas hydrogenosomes and mitochondria, as well as secondary loss of mitochondrion/hydrogenosome-like organelles in Giardia.

 Amitochondriate eukaryotes can be divided into two metabolic types (Martin and Müller 1998 ; Müller 1998 ). Type I organisms such as Giardia and Entamoeba lack organelles involved in core energy metabolism, while type II organisms (trichomonads, some ciliates, and chytrid fungi) harbor a double-membrane limited organelle, the hydrogenosome (Müller 1993 ; Hackstein et al. 1999 ; Kulda 1999 ). The hydrogenosome is the site of the FeS protein-mediated metabolism of pyruvate and the formation of molecular hydrogen, which is accompanied by substrate-level phosphorylation ATP synthesis. In type I amitochondriates, the FeS protein-dependent pyruvate metabolism takes place in the cytosol (Reeves 1984 ; Ellis et al. 1993 ).

It is also interesting if the type I (mitosomes) and type II (hydrogenosomes) and mitochondriate organism (mitochondria) are all unique in the sense that the organelles do not mix. Would be in line with the gradual evolution of mitosome>hydrogenosome and/or mitochondrion.

A common origin of the two organelles [mitochondria and hydrogenosomes] is supported by a number of similarities in their structure, function, and biogenesis (Johnson, Lahti, and Bradley 1993 ; Benchimol, Johnson, and De Souza 1996 ; Bui, Bradley, and Johnson 1996 ; Bradley et al. 1997 ; Dyall et al. 2000 ), as well as by phylogenetic analysis of several hydrogenosomal metabolic enzymes (Länge, Rozario, and Müller 1994 ; Hrd and Müller 1995a, 1995b ) and heat shock proteins (Müller 1997 ; Embley and Hirt 1998 ). Although neither mitochondria nor hydrogenosomes have been found in type I organisms, genes of probable mitochondrial origin have been identified in Giardia (Roger et al. 1998 ) and Entamoeba (Clark and Roger 1995 ). Moreover, a putative mitochondrial “remnant,” the mitosome (Tovar, Fischer, and Clark 1999 ) or crypton (Mai et al. 1999 ), has recently been detected in Entamoeba (Müller 2000 ).

The phylogenetic relationship shows that protists are generally grouped together and they are close to plants and metazoa and fungi.

In all global phylogenetic reconstructions, IscS/NifS-like homologs formed two distinct groups that were previously designated groups I and II (Mihara et al. 1997 ). The IscS sequences of Trichomonas and Giardia and those of the mitochondrial homologs in other eukaryotes formed a single clade (group I) with a high bootstrap value (99%) using the local rearrangement option of the PROTML program (fig. 3 ). Within this clade, Trichomonas and Giardia formed a subgroup together with Plasmodium falciparum and Arabidopsis thaliana. The second eukaryotic subgroup consisted of metazoan IscS, and the third group comprised homologs in fungal mitochondria. The -proteobacterium Rickettsia prowazeki, often considered a close relative to the mitochondrial ancestor, clustered together with metazoan mitochondrial IscS (Andersson and Kurland 1999 ).

Abstract: […] by the discovery of mitochondria-like double membrane-bound organelles called mitosomes. Here, we report that proteins targeted into mitosomes of Giardia intestinalis have targeting signals necessary and sufficient to be recognized by the mitosomal protein import machinery. Expression of these mitosomal proteins in Trichomonas vaginalis results in targeting to hydrogenosomes, a hydrogen-producing form of mitochondria. We identify, in Giardia and Trichomonas, proteins related to the component of the translocase in the inner membrane from mitochondria and the processing peptidase. A shared mode of protein targeting supports the hypothesis that mitosomes, hydrogenosomes, and mitochondria represent different forms of the same fundamental organelle having evolved under distinct selection pressures.

The presence of a common type of FeS assembly machinery in Giardia mitosomes, trichomonad hydrogenosomes, and mitochondria argues for a common evolutionary history of these organelles (4); however, it does not refute contentions that these organelles each arose independently from related species of bacterial endosymbionts (15). One problem is the absence of knowledge concerning the biogenesis of the mitosomes, the evidence that provided strong arguments for a common progenitor of hydrogenosomes and mitochondria (16, 17). Proteins targeted into the mitochondria are synthesized in cytosol with an N-terminal extension for protein targeting; however, many have internal targeting signals. Both sorts of targeting information are recognized by the outer (TOM) and inner (TIM) membrane translocases (18, 19). The mitochondrial matrix proteins are further translocated through the TIM23 complex, with energy supplied by a PAM complex. The PAM complex includes an integral membrane protein with a J domain referred to either as Pam18 (20) or Tim14 (21). After translocation, N-terminal presequences are then cleaved by a matrix-located processing peptidase (MPP) (22). Proteins targeted to hydrogenosomes have N-terminal extensions that carry targeting information (23).

To provide insight into the biogenesis of Giardia mitosomes, we investigated and compared targeting of GiiscS, GiiscU, and [2Fe2S] ferredoxin to Giardia mitosomes and to hydrogenosomes in Trichomonas vaginalis. We show that mitosomes and hydrogenosomes share a common mode of protein targeting that, like protein import into mitochondria, can make use of N-terminal or internal targeting signals. Initial sequence analysis and cell localization studies suggests that Giardia and Trichomonas have protein import machinery that shares common components with the protein import machinery of mitochondria and mitochondria-like processing peptidases.

Conservation of Protein Targeting in Mitosomes and Hydrogenosomes. To determine whether the mitosomal targeting sequences on GiiscU and Gifdx can function to target proteins to hydrogenosomes, the giardial genes were overexpressed in T. vaginalis. Immunofluorescence labeling of trichomonad cells expressing tagged GiiscU, Gifdx, and GiiscS localized these proteins to discrete structures surrounding trichomonad nuclei and cytoskeletal structures, the cell distribution typical for hydrogenosomes (Fig. 3A). The labeling of tagged proteins also colocalized with malic enzyme, a marker protein for hydrogenosomes. Stronger malic enzyme signal corresponds to its abundance in hydrogenosomes (30). In contrast, the absence of N-terminal leader sequences on GiiscU and Gifdx abrogated the delivery of the proteins into the target organelle with the majority of each protein accumulating in the cytosol (Fig. 3B).
 

From the abstract: Microsporidia were long considered to be amitochondriate, but recently a tiny mitochondrion-derived organelle called the mitosome was detected. The molecular function of this organelle remains poorly understood. The mitosome has no genome, so it must import all its proteins from the cytosol. In other fungi, the mitochondrial protein import machinery consists of a network series of heterooligomeric translocases and peptidases, but in microsporidia, only a few subunits of some of these complexes have been identified to date. Here, we look at targeting sequences of the microsporidian mitosomal import system and show that mitosomes do in some cases still use N-terminal and internal targeting sequences that are recognizable by import systems of mitochondria in yeast. Furthermore, we have examined the function of the inner membrane peptidase processing enzyme and demonstrate that mitosomal substrates of this enzyme are processed to mature proteins in one species with a simplified processing complex, Antonospora locustae. However, in Encephalitozoon cuniculi, the processing complex is lost altogether, and the preprotein substrate functions with the targeting leader still attached.

Also some nice overview of the proteins in it.

The number of proteins estimated to be acting within yeast mitochondria is 800-1,000 (19, 20). In contrast, the number of putative mitochondrial proteins identified within all microsporidian species amounts to 21 thus far (21, 22). Based on the whole genome sequence of Encephalitozoon cuniculi, the only clear function of the organelle from these proteins is iron-sulfur cluster (ISC) assembly, and it seems to have lost the capacity for ATP production through oxidative phosphorylation (21).

The identified proteins can largely be categorized into those that act in protein and metabolite import (TOM70, TIM22, TOM40, Imp2, mitochondrial Hsp70, and ATM1-ABC transporter proteins) and those involved in ISC assembly and export (frataxin, ferredoxin, ISCU, ISCS, ERV1, and ferredoxin NADPH oxido-reductase [FNR]). The two subunits of pyruvate dehydrogenase, PDH and -, and mitochondrial glycerol-3-phosphate dehydrogenase (mtG3PDH) are involved in metabolic processes, and manganese-containing superoxide dismutase (MnSOD) is involved in protection against oxidative stress (Fig. 1).

Protein import mechanisms are in a more simplified form in mitosomes, in line with gradual evolution from a simpler to a more complex form: mitosomes –> mitochondria.

Of the identified proteins in microsporidia, few of the mitochondrial import and processing proteins are present (Fig. 1). In addition, very few of the microsporidian mitosomal proteins have N-terminal sequences with characteristics expected of a targeting sequence, so it is unclear how import is achieved. Here we have tested to what extent the common mitochondrial targeting system has been retained in microsporidia and how the complexity of the system is degenerating. We show that many mitosomal proteins are targeted appropriately in yeast, confirming that homologous elements of the ancestral system are still used and that the N terminus is at least partly responsible for encoding targeting information. Using the inner-membrane peptidase (IMP) as a model, we also show the complexity of the system has been progressively reduced during microsporidian evolution. IMP usually consists of two catalytic subunits (Imp1 and Imp2) and a noncatalytic regulator (Som1), but this complex is apparently reduced to a single functional protein in Antonospora locustae and lost altogether in E. cuniculi. The protein substrates of this complex are processed in A. locustae but not in E. cuniculi, where they function as unprocessed preproteins. The reduction of the IMP complex is typical of how complex molecular systems have degenerated in microsporidia.

The data that is presented is in line with a gradual evolution of endomembranous structures that eventually evolved into mitochondria and started with an anaerobic variant that acquired more and more functions, including the use of oxygen as a teminal acceptor in the electron transport chain. So instead of gradually losing functions, they were gradually acquired.

This view has changed recently with the finding of mitochondrial remnant organelles termed mitosomes in Entamoeba (Mai et al., 1999; Tovar et al., 1999): this coupled with the presence of mitochondrial chaperonin genes (Arisue et al., 2002) confirms the secondary loss of mitochondria by this organism and this has been reviewed by Müller (2000). The presence of specialized membranes with electron transport functions has also been detected in Giardia (Lloyd et al., 2002). This organism once considered amitochondriate has mitochondrial-like chaperonin genes (Roger et al., 1998; Arisue et al., 2002), a nuclear coded valyl-tRNA synthetase (Hashimoto et al., 1998) and has recently been demonstrated to contain a fully functional mitochondrial iron–sulphur cluster assembly pathway involving the proteins IscS and IscU which are present in a double membrane-bound mitochondrial remnant organelle (Tovar et al., 2003). However, with the exception of Nyctotherus ovalis (Hackstein et al., 1999), all mitochondrial remnant organelles, including hydrogenosomes, lack an organelle genome, which was the major distinction between mitochondria and hydrogenosomes (Müller, 1993). The recently discovered Entamoeba remnant mitochondrial organelle, the mitosome, also lacks an organelle genome (León-Avila & Tovar, 2004), suggesting that reduction in organelle function was accompanied by loss of the genome. It is clear from several studies that these so-called anaerobes do encounter varying amounts of oxygen and therefore must have the ability to survive the effects of oxidative stress, and this is explored in the report by Lloyd et al. (2004). A mitochondrial relict organelle has also been described in cryptosporidia (Riordan et al., 2003). This observation was recently supported by the finding that Cryptosporidium parvum has genes (IscS and IscU) encoding a mitochondrial-type iron–sulphur cluster biosynthetic pathway and that these proteins target the proposed relict organelle. Thus, C. parvum is the latest to join the growing numbers that support the view that there are no truly amitochondriate extant eukaryotes (Müller, 2000). The report of mitochondrial-type hsp70 genes in two microsporidians strongly suggests that this group of amitochondriates has also undergone secondary mitochondrial loss (Arisue et al., 2002).