• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Diverse Eukaryotes Have Retained Mitochondrial Homologues of the Bacterial Division Protein FtsZ

Posted on: Saturday, 12 June 2004, 06:00 CDT

Mitochondrial fission requires the division of both the inner and outer mitochondrial membranes. Dynamin-related proteins operate in division of the outer membrane of probably all mitochondria, and also that of chloroplasts - organelles that have a bacterial origin like mitochondria. How the inner mitochondrial membrane divides is less well established. Homologues of the major bacterial division protein, FtsZ, are known to reside inside mitochondria of the chromophyte alga Mallomonas, a red alga, and the slime mould Dictyostelium discoideum, where these proteins are likely to act in division of the organelle. Mitochondrial FtsZ is, however, absent from the genomes of higher eukaryotes (animals, fungi, and plants), even though FtsZs are known to be essential for the division of probably all chloroplasts. To begin to understand why higher eukaryotes have lost mitochondrial FtsZ, we have sampled various diverse protists to determine which groups have retained the gene. Database searches and degenerate PCR uncovered genes for likely mitochondrial FtsZs from the glaucocystophyte Cyanophora paradoxe, the oomycete Phytophthora infestans, two haptophyte algae, and two diatoms - one being Thalassiosira pseudonana, the draft genome of which is now available. From Thalassiosira we also identified two chloroplast FtsZs, one of which appears to be undergoing a C- terminal shortening that may be common to many organellar FtsZs. Our data indicate that many protists still employ the FtsZ-based ancestral mitochondrial division mechanism, and that mitochondrial FtsZ has been lost numerous times in the evolution of eukaryotes.

Abbreviations of FtsZ proteins used in text.

CfFtsZ-mt: Cylindrotheca fusiformis mitochondrial FtsZ; CpFtsZ- mt: Cyanophora paradoxa mitochondrial FtsZ; GoFtsZ-mt: Gephyrocapsa oceanica mitochondrial FtsZ; PcFtsZ-mt: Pleurochrysis carterae mitochondrial FtsZ; PiFtsZ-mt: Phytophthora infestans mitochondrial FtsZ; TpFtsZ-cp 1 & 2: Thalassiosira pseudonana plastid FtsZ; & plastid FtsZ2; TpFtsZ-mt: Thalassiosira pseudonana mitochondrial FtsZ.

Introduction

Mitochondria are the descendants of an a-proteobacterium that was engulfed by another cell about two billion years ago; plastids have a similar origin, but one that was more recent and where the endosymbiont was a cyanobacterium (Martin et al. 2001). These essential organelles thus cannot be made de novo and must divide to maintain their numbers in daughter cells. In recent years it has become apparent that some of the proteins that act in the division of mitochondria and plastids are derived from their respective bacterial ancestors. Bacterial cytokinesis relies on over a dozen proteins: the earliest acting and most phylogenetically widespread of these is FtsZ - a GTPase that assembles into polymeric filaments in vitro (Addinall and Holland 2002). Very early in the division of Escherichia coli cells, FtsZ forms a cytoplasmic ring at the plane of division and, as division progresses, the Z-ring contracts, presumably drawing the inner membrane to fission (Bi and Lutkenhaus 1991).

The first identification of a eukaryotic FtsZ came with discovery of a plastid-targeted version of the protein in Arabidopsis thaliana (Osteryoung and Vierling 1995). The Arabidopsis nuclear genome is now known to encode three stromal-targeted copies of FtsZ, two of which at least are required for normal plastid division and form medial division rings similar to bacterial FtsZ rings (Osteryoung et al. 1998). Analyses of plant and green algal genomes have revealed that the possession of two or more plastidtargeted FtsZ genes is a general feature of these organisms, and that the predicted proteins of these genes form two distinct families, FtsZ1 and FtsZ2 (Araki et al. 2003; Osteryoung and McAndrew 2001; Stokes and Osteryoung 2003; Wang et al. 2003). Since no cyanobacterium is known to encode two ftsZs, the separate families are likely to be derived from an ancient gene duplication that may have occurred soon after the appearance of the green algae (Gilson and Beech 2001; Osteryoung and McAndrew 2001). Multiple plastid FtsZs are also found in red algae (Miyagishima et al. 2003), and here we identify two plastid FtsZs in the genome of the diatom Thallasiosira pseudonana, indicating that all plastids (with the exception of apicoplasts, which have no FtsZ; Gardner et al. 2002) may have a requirement for more than one FtsZ. Though FtsZs from separate families co-localize in Arabidopsis (AtFtsZ1-1 and AtFtsZ2-1; Vitha et al. 2001), the specific functional requirements for multiple FtsZs in plastid division are unknown. It is interesting to note, however, that in most instances plastid FtsZ1 proteins have lost the conserved C-terminal core domain of FtsZ, which is believed to be involved in protein-protein interactions (Osteryoung and McAndrew 2001 ; Stokes and Osteryoung 2003).

In contrast to the almost ubiquitous use of FtsZ for plastid division, FtsZ has been identified in the mitochondria of relatively few organisms. Complete genomes of fungi and animals contain no FtsZ, and higher plants have only those that are clearly related to cyanobacterial ftsZ which encode plastid-targeted proteins. In all of these organisms, and indeed perhaps all eukaryotes (Miyagishima et al. 2003), a dynamin-related protein (DRP) that is likely to have arisen from the host cell acts at the mitochondrial outer membrane at a late stage of division. Dynamins are GTPases that may have a general role in membrane modeling, and mitochondrial DRPs form a ring at the mitochondrial constriction site where they interact with other proteins to probably bring about the final severance of the organelle (Griparic and van der Bliek 2001; Yoon and McNiven 2001; Arimura and Tsutsumi 2002; Shaw and Nunnari 2002; Legesse-Miller et al. 2003; Nishida et al. 2003). DRPs are likely to have been recruited early in the evolution of mitochondria, perhaps even at the time of endosymbiosis (van der Bliek 2000) - a view that is supported by the widespread phylogenetic distribution of these proteins (Miyagishima et al. 2003).

To date, the organisms known to contain mitochondrial FtsZs are certain groups of protists: the chromophyte (stramenopile) alga Mallomonas splendens (Beech et al. 2000), the red alga Cyanidioschyzon merolae (Takahara et al. 2000), and the slime mould Dictyostelium discoideum from which we have recently described two mitochondrial FtsZs (Gilson et al. 2003). These proteins are most closely related to FtsZs of the ancestors of mitochondria, the ?- proteobacteria; and the FtsZs of M. splendens, C. merolae and at least one FtsZ in D. discoideum localize early within dividing mitochondria as patches and/or rings around the site of division (Beech et al. 2000; Gilson et al. 2003; Takahara et al. 2000). Genetic evidence for a role of these proteins in mitochondrial division comes from D. discoideum amoebae, where disruption of mitochondrial FszA and/or FszB lead to a greatly increased proportion of elongated mitochondria (Gilson et al. 2003).

To begin to understand why mitochondrial FfsZs have been lost from the higher eukaryotes such as animals, plants and fungi, and from some protists, we have begun a survey of eukaryotes can to see determine where which groups still harbor mitochondrial FfsZ genes. Here we describe six novel mitochondrial FtsZ sequences, and discuss if any there are any mitochondrial characteristics that differ between groups that contain mitochondrial FtsZ, and those that do not. In addition, we describe two plastid FfsZ sequences from the Thallasiosira pseudonana genome, one of which appears to be in the process of losing its C-terminal core domain. ;

Results

Genome Databases Reveal Two Novel Mitochondrial FtsZ Genes

Database searches were used to uncover FfsZs in the centric diatom Thalassiosira pseudonana and the oomycete Phytophthora infestans. The M. splendens MsFtsZ-mt was used to perform tBLASTn (Altschul et al. 1997) searches of the T. pseudonana draft genome (http://genome.jgi-psf.org/thapsO/ thaps0.home.html). Thalassiosira pseudonana was found to have a full-length FtsZ gene of 1497 bp, which we have called Thalassiosira pseudonana mitochondrial FtsZ (TpFtsZ-mt; Figure 1 sequence available on request). The predicted amino acid sequence of TpFtsZ-mt is predicted by iPSORT (Bannai et al. 2001, 2002) and TargetP (Emanuelsson et al. 2000) to be targeted to mitochondria. Also found were two genes with high homology to plastid FtsZs, that we call Thalassiosira pseudonana plastid FfsZ1 and FtsZ2 (TpFtsZ-cp1 and TpFtsZ-cp2; sequence available on request), of 1233 and 1368 bp respectively. A tBLASTn search of the P. infestans EST database (http://www.vbi.vt.edu/~estap/), using MsFtsZ-mt, located a 627 bp cDNA with high homology to mitochondrial FfsZs; we have called this partial cDNA Phytophthora infestans mitochondrial FfsZ (P/FfsZ-mf; sequence available on request).

Degenerate PCR Amplified Four Possible Mitochondrial RsZ Genes

Primers for degenerate PCR were designed to specifically amplify genes encoding the conserved central region of mitochondrial FtsZs (- 200 aa). This region contains, among other conserved motifs, the GTPbinding region typical of all FtsZs (GGGTGT/SG; Fig. 1). The degenerate PCR primers amplified -550 bp fragments of possible mitochondrial FfsZ genes from DNA o\f the glaucocystophyte Cyanophora paradoxa, the pennate diatom Cylindrotheca fusiformis, and the haptophytes P/eurochrysis carterae and Gephyrocapsa oceanica. The Cyanophora paradoxa PCR returned a 579 bp product: we called this partial gene Cyanophora paradoxa mitochondrial FfsZ (CpFfsZ-mf; GenBank accession: AY487241). Single-primer PCR (Keeling and Doolittle 1997) was then used to extend the genomic sequence of CpFtsZ-mt by 502 bp in the 5' direction. We predicted that if CpFtsZ- mt was a nuclear-encoded, mitochondrial-targeted protein, then it should ? encode a targeting sequence at its N-terminus. When the 502 bp region was translated, however, no putative mitochondrial targeting sequences that formed a continuous open reading frame with the rest of CpFtsZ-mt could be identified (data not shown). Translation of another short open reading frame within the 5' region revealed it had properties of a mitochondrial targeting peptide, and that the removal of a putative (239 bp) spliceosomal intron, bordered by GT and AT, reconstituted an open reading frame with the rest of the CpFtsZ-mt gene. The programs TargetP and iPSORT predicted that this theoretical FtsZ was targeted to the mitochondrion. Only the conserved middle region of CpFtsZ-mt was used for phylogenetic reconstruction (Figs 1, 2).

A 579 bp product was amplified from the diatom C.fusiformis: we call this partial gene, Cylindrotheca fusiformis mitochondrial FtsZ (CfFtsZ-mt; GenBank accession: AY487244). From G. oceanica a 570 bp partial FtsZ was produced, called Gephyrocapsa oceanica mitochondrial FtsZ (GoFtsZ-mt; Fig. 1 ; GenBank accession: AY487237), and from P. carterae a 573 bp partial FtsZ was amplified that we have called Pleurochrysis carterae mitochondrial FtsZ (PcFtsZ-mt; Fig. 1 ; GenBank accession: AY487242). Degenerate PCR was also performed on DNA from the dinoflagellate Alexandrium catenella (Whedon and Kofoid) Balech, the euglenophyte Euglena gracilis Klebs, and the primitive green algae Nephroseimis minuta (Carter) Butcher and Mesostigma viride Lauterborn: however, no FfsZ- like products were produced.

We are confident that the partial FtsZs presented here represent bona fide mitochondrial FfsZs for a number of reasons. In the case of C. paradoxa, the partial FtsZ appears to have an intron - a feature exclusive to eukaryotic genes. The predicted amino acid sequence of Cylindrotheca fusiformis mitochondrial FtsZ is very similar to the mitochondrial FtsZs of other stramenopiles, in particular that of another diatom, T. pseudonana, and the chromophyte M. splendens (Fig. 2). The haptophyte sequences, GoFtsZ- mt and PcFtsZ-mt, both group phylogenetically with ?- proteobacterial FtsZs, and in some analyses form a clade with the mitochondrial FtsZ of Cyanidioschyzon merolae (data not shown).

Phylogenetic Analyses of Mitochondrial FtsZs

The predicted amino acid sequences of the putative mitochondrial and plastid FtsZ genes were aligned with known FtsZs from plastids, eubacteria and archaea using ClustalX (Thompson et al. 1997). Translations of the conserved central domain of RsZs amplified by our degenerate PCRs were used to generate phylogenetic trees, because the level of amino acid conservation within this region provides the most informative data. Phylogenetic trees were produced using the predicted amino acid sequences of the FfsZs found here and those available from GenBank, using an alignment of approximately 300 conserved amino acids (underlined in Fig. 1). Phylogenetic analyses using the neighbour-joining, maximum parsimony, and quartet puzzling algorithms were performed on PAUP*v10b (Swofford 1999). Bayesian Inference was performed using MRBAYES (Huelsenbeck and Ronquist 2001) using the Markov Chain Monte Carlo algorithm. A consensus tree built from these methods is shown in Figure 2. For each branch, the respective neighbour-joining and maximum parsimony bootstrap values, quartet puzzling confidence values and Bayesian Inference frequencies are shown ('-' denotes an absent node or a bootstrap value of less than 50%; Fig. 2). all of the mitochondrial FtsZ sequences identified in this study form a clade with the ?- proteobacteria and mitochondrial FtsZs; this clade is well supported by Bayesian Inference (67/52/41/100) and was evident in all other phylogenies produced (data not shown). The branchings of the stramenopile FtsZs by Bayesian Inference were also robustly maintained (52/-/72/100) and the four were consistently grouped (despite differing branch order) in all analyses performed (data not shown). The mitochondrial FtsZs of P. carterae, G. oceanica, and C. merolae all branched from an aproteobacterial group containing FtsZs from Sinorhizobium and Agrobacterium species. The mitochondrial FtsZs of Dictyostelium were the only mitochondrial sequences that did not cluster with other mitochondrial FtsZs. all the methods robustly grouped TpFtsZ-cp1 and TpFtsZ-cp2 together (100/97/99/ 100), in the same clade as the M. splendens and red algal plastid FtsZs.

Figure 1. Alignment of the predicted amino acid sequences of mitochondrial, [alpha]-proteobacterial, C. merolae plastid and E. coli ftsZs. Similarity is shown by grey fill, and identity by black fill. The region showing the highest degree of similarity is foubnd in the central 200 amino acides of the proteins. The aligned region used for phylogenetic analysis is underlined, and the amino acids used to design degenerate PCR primers (1R and 4L) are boxed. The sequence obtained using these primers corresponds to the conserved region of FtsZ. GenBank accession numbers of all FtsZ sequences used are available upon request.

Figure 2. Phylogenetic tree of FtsZ protein sequences. The values at each node represent bootstrap/confidence values for neighbour- joining/parsimony/quartet-puzzling/Markov Chain Monte Carlo analyses; a bootstrap value less than 50 is represented as '-'. In all analyses, the cyanobacteria group with the plastid FtsZs, and the a-proteobacteria group with the mitochondrial FtsZs. Within the a-proteobacterial/mitochondrial group, the stramenopile FtsZs form a clade. The FtsZs of plastids form three clades: green-plastid FtsZ1, green-plastid FtsZ2, and the red and chromophyte algal group. The green-plastid FtsZ2 and red/chromophyte (Cavalier-Smith 1998) plastid FtsZ clades are the most similar to cyanobacterial FtsZs (also shown in Wang et al. 2003), while the greenplastid FtsZ1 group is the most distantly branching. Note that while the green-plastid FtsZs form two distinct groups, the multiple red/chromophyte plastid FtsZs are monophyletic.

Discussion

The six mitochondrial FtsZs described here now increase to ten the number of mitochondrial FtsZs so far discovered in eukaryotes. Three of these genes come from the stramenopiles Cylindrotheca fusiformis, Thalassiosira pseudonana and Phytophthora infestans] as expected, the predicted proteins from these three sequences branched together within the mitochondrial/a-proteobacterial group. The data also support earlier predictions, arising from the discovery of mitochondrial FtsZ in Mallomonas splendens, that most, if not all, members of this diverse eukaryotic group would use FtsZ for mitochondrial division (Gilson and Beech 2001).

Our identification of possible mitochondrial FfsZs in Gephyrocapsa oceanica, Pleurochrysis carterae and Cyanophora paradoxa are the first identifications of such genes from two major groups of eukaryotes, the glaucocystophytes and the haptophytes. The predicted mitochondrial FtsZ proteins of G. oceanica and P. carterae cluster with that of Cyanidioschyzon merolae, and the C. paradoxa mitochondrial FtsZ groups with the those of the stramenopiles. Dictyostelium FszA and FszB are the next closest relatives to the C. paradoxa/stramenopile group at the base of the mitochondrial/a- proteobacterial FtsZ tree. Given the evidence in support of mitochondrial monophyly (Gray et al. 1999), it is curious that mitochondrial FtsZs are distributed among the [alpha]- proteobacteria. A possible explanation is that the rates of evolution among the genes are variable, and the two billion years of evolution separating some of the sequences has resulted in anomalous trees due to long-branch attraction effects (Moreira et al. 1999). In this respect, it is interesting to note that, even though plastids owe their origins to a more recent endosymbiosis than mitochondria, a phylogenetic anomaly is also apparent in the plastid portion of FtsZ trees: the green plastid FtsZ1 group appears more ancient than the cyanobacterial FtsZs from which it must be derived (Figure 2; Osteryoung et al. 1998; Stokes and Osteryoung 2003; Wang et al. 2003). This, too, may be an example of unequal evolution that has allowed the plastid FtsZIs to become the 'most different' member of the plastid and cyanobacterial FtsZ sister groups.

Figure 3. Multiple sequence alignment of various FtsZs. The balckened areas correspond to the conserved C-terminal amino acid sequence (D/E)(I/V)PX(F/Y)L, or part thereof. this conserved C- terminal core domain of FtsZs is found in the plastid group, but is absent from the plastid has a C-terminal core domain-like sequence (shaded).

An alternative scenario to long-branch attraction effects is that the mitochondrial FtsZs may be paralogous, and that they should in fact form two or more groups derived from ancient gene duplications. Support for this comes from Dictyostelium, which possesses two mitochondrial FtsZs: FszA and FszB. However, if the last common ancestor of all mitochondria contained two FtsZs, and each Dictyostelium FtsZ is a descendent of these, then it would be expected that one of the Dictyostelium FtsZs should have clustered with the C. paracfoxa/stramenopiles group and the other with the C. merolae/P, carterae/G. oceanica group. This was not observed in our trees and, instead, FszA and FszB grouped together. It is therefore more likely that FszA and FszB have arisen from a \more recent gene duplication, even though the Fsz proteins are not particularly similar (47.9% over the region used for phylogenetic analysis). This lack of similarity could be a classic example of gene duplication, where one member of a protein pair, no longer functionally constrained, begins to diverge quickly (Lynch and Conery 2003).

The proposal that protein paralogy causes the mitochondrial FtsZs to not all group together is further weakened by data from the genome sequence of T. pseudonana. This diatom appears to encode only one mitochondrial FtsZ, implying that mitochondrial FfsZ duplications might be uncommon and confined to only some eukaryotic lineages, such as the amoebozoa to which Dictyostelium belongs. Ultimately, more mitochondrial FtsZ sequences will be required to improve their derived phylogeny.

Duplication and Functional Diversification is Characteristic of Plastid FtsZs

Two distinct plastid FtsZ families, FtsZ1 and FtsZ2, exist in the green algae and plants (Osteryoung et al. 1998; Stokes and Osteryoung 2003; Wang et al. 2003). The green plastid FtsZ2 group shares with many bacterial species a conserved C-terminal core domain, thought to be involved in protein-protein interactions (Ma and Margolin 1999; Osteryoung and McAndrew 2001). The FtsZ C- terminus consists of a variable spacer domain, which is hydrophobic, and the C-terminal core domain, which has at least part of the sequence (D/E)(I/V)PX(F/Y)L (Din et al. 1998; Ma and Margolin 1999; Osteryoung and McAndrew 2001). The green plastid FtsZ1 family, on the other hand, appears to lack this C-terminal domain (Osteryoung and McAndrew 2001 ; Stokes and Osteryoung 2003). There is also a third phylogenetically distinct plastid FtsZ group, one that is comprised of plastid FtsZs from the red and chromophyte algae (Miyagishima et al. 2003; Stokes and Osteryoung 2003).

The plastid FtsZs of some red algae, like those of their green counterparts, have multiple versions, some of which have lost their C-terminal core domain (e.g., Cyanidium caldarium plastid FtsZ2; Fig. 3). Here we have shown that the T. pseudonana genome also has two plastid FtsZs, TpFtsZcpl and TpFtsZcp2. While both of these proteins have the Cterminal consensus sequence, TpFtsZcpl does not have the variable hydrophilic spacer (Fig. 3). The diatom plastid FtsZs group separately from the red plastid FtsZs within the red/ chromophyte plastid FtsZ group, implying that the dual stramenopile plastid FtsZs are the product of their own independent gene duplication. Given that plastid FtsZ duplications in other organisms have apparently led to the loss of the C-terminal core domain in one of the duplicates, it is possible that TpFtsZcpl is in the act of losing the C-terminal core domain altogether. The possession of multiple plastid FtsZs implies that there is an evolutionary advantage to having two plastid FtsZs, one of which has lost the conserved C-terminal core domain. It is interesting to note that among full-length mitochondrial FtsZs, only Dictyostelium FszA possesses a pronounced C-terminal extension and part of the C- terminal core domain (Fig. 3). As the only mitochondrial FtsZ present in the genome of T. pseudonana has no C-terminal core domain, it would seem that this domain is not necessary for diatom (and possibly all stramenopilar) mitochondrial FtsZ function.

Despite the apparent differences in sequence between the green- plastid FtsZIs and green-plastid FtsZ2s, both protein families appear to localise to the dividing region of plastids and form co- aligned or heteropolymeric rings (Vitha et al. 2001). It is unknown if this also applies to the plastid FtsZs of red algae and stramenopiles, but the FtsZ homologues in the mitochondria of Dictyostelium do not appear to co-localize. Dictyostelium FszA, which contains a semblance of the conserved C-terminal core domain, appears to be a canonical Z-ring-forming FtsZ, whereas FszB-GFP, which lacks this domain, is localized to an electron-dense sub- mitochondrial body of unknown function (Gilson et al. 2003). Thus, there may be a propensity for duplication of FtsZ, twice in the plastids (green and non-green) and at least once in mitochondria, and for the loss of the core C-terminal domain in one of the duplicates.

Multiple Losses of Mitochondrial FtsZ in Eukaryotes

While some eukaryotes have increased their number of FfsZ genes, others have dispensed with them altogether. Mitochondrial FtsZ genes have been lost at least three separate times in the evolution of eukaryotes (Arimura and Tsutsumi 2002; McFadden and Ralph 2003; Miyagishima et al. 2003). The first loss occurred at some point during the evolution of green algae and plants. Land plants, the descendents of green algae, do not contain mitochondrial FtsZ genes, and no green alga for which significant gene data is available (e.g., Chlamydomonas sp. and Chlorella sp.) encodes a mitochondrial FtsZ. Phylogenetic analyses indicate that red, green and glaucocystophyte algae are sister groups that have radiated from the organism that first acquired a primary plastid (Douglas 1998). From the presence of a mitochondrial FtsZ in a primitive red alga (C. merolae), and our discovery that the glaucocystophyte C. paradoxa still retains a mitochondrial FtsZ, it seems likely the green- plastid lineage lost the protein after it diverged from the glaucocystophytes and red algae (Arimura and Tsutsumi 2002; McFadden and Ralph 2003).

To determine how early mitochondrial FtsZ was lost in the evolution of the green-plastid lineage, we searched for potential mitochondrial FtsZs in the primitive green algae, Nephroselmis minuta and Mesostigma viride, using PCR conditions that were successful for other organisms. We failed to amplify any candidate genes. Although these negative results are not proof that the primitive greens we sampled lack mitochondrial FtsZ, we cautiously predict that the loss of mitochondrial FtsZ occurred very early during the evolution of the green-plastid lineage.

The second loss of mitochondrial FtsZ from a eukaryofic lineage appears to have happened before the diversification of the opisthokonts, of which the animals and fungi are members. This loss, however, cannot have greatly predated the diversification of the opisthokonts, since their next closest relatives, the Amoebozoa (Baldauf et al. 2000), include the mitochondrial FtsZ-containing Dictyostelium.

The third group known to have abandoned the use of FtsZ for dividing its mitochondria is the Apicomplexa. The complete genome of Plasmodium falciparum does not contain a mitochondrial FtsZ (Gardner et al. 2002) and the gene does not appear in the significant amounts of gene data available for other apicomplexans, such as Toxoplasma gondii. We have used degenerate PCR to attempt to determine if a mitochondrial FtsZ was present in an alveolan relative of the apicomplexa, the dinoflagellate Alexandrium catenella; however, no candidate genes were amplified.

Why have Certain Eukaryotes Lost Mitochondrial FtsZ?

A comparison of the mitochondrial morphologies of organisms that have lost mitochondrial FtsZ with those of organisms that have retained the protein may provide clues as to why the protein was lost in certain lineages. While there are no consistent differences between the morphologies of mitochondria using FtsZ, and those that do not (in plants, opisthokonts or apicomplexans), it is important to keep in mind that the loss of mitochondrial FtsZ may have occurred for different reasons in different lineages, each of which was evolving independently.

Despite two billion years of evolution, mitochondria are essentially enslaved bacteria and, in bacteria, FtsZ divides cells at the mid-point (Bi and Lutkenhaus 1991). This is most evident in the red algal unicell, C. merolae, which uses FtsZ to divide it's single mitochondrion at the mid-point (Nishida et al. 2003), and is still apparent in both M. splendens and D. discoideum (Beech et al. 2000; Gilson et al. 2003). It is curious to note that in the genomes of D. discoideum and T. pseudonana, no homologues of the Min proteins, required for placement of the FtsZ ring in plastids and bacteria (Harry 2001; Miyagishima et al. 2003), are apparent, and there have been no molecules identified in mitochondria that may act in this capacity. The reticulate mitochondria of opisthokonts are decidedly non-bacterial-like, and in these cells mitochondrial division and fusion appear to occur at seemingly random positions (Nunnari et al. 1997). It is also of interest that the tachyzoites of the apicomplexan, Toxoplasma gond/7, have a reticulated mitochondrial network that extends throughout daughter cells (MeIo et al. 2000). It may be that mitochondrial FtsZ is particularly well suited to dividing mitochondria at their mid-points, and has been discarded by organisms not requiring such precision. On the other hand, the haptophyte P. carterae, has a reticulated mitochondrial network (Beech and Wetherbee 1984), and thus no easily recognisable mid-point. Further analyses are underway to determine if the mitochondrion of P. carterae is as plastic as those of yeast and animals. In any case, given that P. carterae may represent a species that has managed to maintain FtsZ while evolving a reticulated mitochondrion, it will be important to determine exactly where in the mitochondrion FtsZ localizes in this organism, to better understand the role of mitochondrial FtsZs in mitochondrial division.

Methods

Culture conditions and DNA extraction: Cultures of Pleurochrysis carterae (Braarud and Fagerlund) Christensen, Gephyrocapsa oceanica Kamptnerand Cylindrotheca fusiformis Reimann et Lewin were obtained from the CSIRO Collection of living microalgae, Hobart, Australia (CSIRO strains CS-287, CS335 and CS-13 respectively). Cyanophora paradoxa Korshikov was obtained from Uwe Maier, Philipps Universitat, Germany (University of Texas Collection of Algae strain UTEX LB555). Cylindroth\eca fusiformis, P. carterae, and G. oceanica were grown in f/2 media (Guillard and Ryther 1962; Guillard 1975). Cyanophora paradoxa was cultured in MBL media (Nichols 1973). all cultures were grown at 16 C with a 12 hr light/12 dark light cycle. Approximately 20OmI of a log-phase, axenic culture for each cell type was pelleted and DNA extracted using the Plant DNAzol reagent (Life Technologies, Grand Island N.Y.), following the manufacturer's instructions. To remove co-precipitated contaminants, DNA was further cleaned using the ULTRA clean DNA purification kit (Mo Bio Laboratories, California) following the manufacturer's instructions.

Gene amplification, cloning and sequencing: The predicted FtsZ amino acid sequences from selected prokaryotes, plastids and MsFtsZ- mt, were aligned using ClustalX (Thompson et al. 1997). Two specific regions of homology between the [alpha]-proteobacteria and MsFtsZ- mt (boxed in Figure 1) were used as the target of degenerate primers: 1R (forward) 5'-GGIGGIAAYGCIGTIAAYAAYATGAT-3', and 4L (reverse) 5'-ACRTCNGCRAARTCNARRTTDATR-3'. 20-100 ng of each DNA sample and 10 pmol of each primer, in a total volume of 25 l, were used with the following PCR conditions: 1 cycle of 3 min at 94 C, 35 cycles of 30 s at 94 C, 30 s at 50 C, and 1 min at 72 C, followed by 10 min at 72 C for the final extension. Single-primer PCR was performed as described previously (Keeling and Doolittle 1997) on C. paradoxe DNA, using the gene-specific primer; CpFtsZ 2L 5'- TCGTGCGCTCGTTGGC GACGCGGAACA-3'. The amplified products were purified using an ULTRA clean DNA purification kit, and ligated into the pGEM-T-Easy vector (Promega - Madison, USA) before being cloned into BL21(DE3)pLysS-competent cells (Promega). PCR products were sequenced using the Big Dye Terminator kit (ABI-Prism - Connecticut, USA) and analysed with an Applied Biosystems 373-Stretch DNA sequencer, all according to the manufacturer's instructions.

Identification and Retrieval of FtsZ sequences from online databases: The predicted amino acid sequence of MsFtsZ-mt was used to tBLASTn search (Altschul et al. 1997) the genome database of the centric diatom Thalassiosira pseudonana (Hustedt) Hasle and Heimdal (http://genome.jgi-psf.org/ thapsO/thapsO.home.html) and the EST database of the oomycete Phytophthora infestans (Mont.) de Bary (http://www.vbi.vt.edu/~estap/).

Phylogenetic analyses: The predicted amino acid sequences for each PCR product and downloaded sequence were aligned using ClustalX (Thompson et al. 1997). Manual adjustments to the alignment were performed with SeqPup vO.9 (Gilbert 1999). Phylogenetic analyses of a conserved 303 amino acid stretch were performed using the neighbour-joining method (Saitou and Nei 1987; Studie and Keppler 1988), maximum parsimony method (Fitch 1971), and the quartet puzzling method (Strimmer and von Haeseler 1996) using the preset parameters in the PAUP*v10b software package (Swofford 1999). For the neighbour-joining and parsimony methods, 100 bootstrap (Felsenstein 1985) replicates and 100 quartet puzzling replicates were performed. The Bayesian Inference was performed using MRBAYES (Huelsenbeck and Ronquist 2001) with the Markov Chain Monte Carlo algorithm parameters set at 4 chains. The analysis was run for 1.5 10^sup 6^ generations, which was sampled at every 100th tree, and the first 3000 trees were disregarded. The tree produced in the Bayesian Inference was used as the basis of a consensus tree; the inference frequencies of each node were combined with the corresponding bootstrap values of each node from the neighbour- joining and parsimony methods and the quartet puzzling confidence values were added manually using Adobe Photoshop 7.0.

Acknowledgements

BRK is the recipient of a Deakin University postgraduate award, PRG was an Australian Research Council postdoctoral fellow, and the study was funded by an ARC Discovery-Project grant to PLB. We thank Connie Mikulski for laboratory assistance, Uwe Maier and Michael Melkonian for protist strains, and the curators of the Thallasiosira pseudonana genome project (the US Department of Energy Joint Genome Institute http://www.jgi.doe.gov/) and the Phytophthora infestans EST project (http://www.vbi.vt.edu/~estap/), for the provision of sequence data.

Note added in proof: A broader definition of the FtsZ C-terminal core domain has recently been published: Vaughan et al. 2004, J Mol Evol 58: 19-29

References

Addinall SG, Holland B (2002) The tubulin ancestor, FtsZ, draughtsman, designer and driving force for bacterial cytokinesis. J MoI Biol 318: 219-236

Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Research 25: 3389- 3402

Araki Y, Takio S, Ono K, Takano H (2003) Two types of plastid ftsZ genes in the liverwort Marchantia polymorpha. Protoplasma 221: 163-173

Arimura S-i, Tsutsumi N (2002) A dynamin-like protein (ADL2b), rather than FtsZ, is involved in Arabidopsis mitochondrial division. Proc Natl Acad Sei USA 99: 5727-5731

Baldauf SL, Roger AJ, Wenk-Siefert I, Doolittle WF (2000) A kingdom-level phylogeny of eukaryotes based on combined protein data. Science 290: 972-977

Bannai H, Tamada Y, Maruyama O, Nakai K, Miyano S (2001) Views: Fundamental Building Blocks in the Process of Knowledge Discovery. In 14th International FLAIRS Conference. AAAI Press, pp 233-238

Bannai H, Tamada Y, Maruyama O, Nakai K, Miyano S (2002) Extensive feature detection of N-terminal protein sorting signals. Bioinformatics 18: 298-305

Beech PL, Wetherbee R (1984) Serial reconstruction of the mitochondrial reticulum in the coccolithophorid, Pleurochrysis carterae (Prymnesiophyceae). Protoplasma 123: 226-229

Beech PL, Nheu T, Schultz T, Herbert S, Lithgow T, Gilson PR, McFadden Gl (2000) Mitochondrial FtsZ in a chromophyte alga. Science 287: 1276-1279

Bi E, Lutkenhaus J (1991) FtsZ ring structure associated with division in Escherichia coli. Nature 354: 161-164

Cavalier-Smith T (1998) A revised six-kingdom system of life. Biol Rev 73: 203-266

Din N, Quardokus EM, Sackett MJ, Brun YV (1998) Dominant C- terminal deletions of FtsZ that affect its ability to localize in Caulobacter and its interaction with FtsA. MoI Microbiol 27: 1051- 1063

Douglas SE (1998) Plastid evolution: origins, diversity, trends. Curr Opin Genet Dev 8: 655-661

Emanuelsson O, Nielsen H, Brunak S, von Heijne G (2000) Predicting subcellular localization of proteins based on their N- terminal amino acid sequence. J MoI Biol 300: 1005-1016

Felsenstein J (1985) Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783-791

Fitch WM (1971) Toward defining the course of evolution: minimum change for a specified tree topology. SystZool20:406-416

Gardner MJ, Hall N, Fung E, White O, Berriman M, Hyman RW, Carlton JM, Pain A, Nelson KE, Bowman S, Paulsen IT, James K, Eisen JA, Rutherford K, Salzberg SL, Craig A, Kyes S, Chan M-S, Nene V, Shallom SJ, Suh B, Peterson J, Angiuoli S, Pertea M, Allen J, Selengut J, Haft D, Mather MW, Vaidya AB, Martin DMA, Fairlamb AH, Fraunholz MJ, Roos DS, Ralph SA, McFadden Gl, Cummings LM, Subramanian GM, Mungall C, Venter JC, Carucci DJ, Hoffman SL, Newbold C, Davis RW, Fraser CM, Barrell B (2002) Genome sequence of the human malaria parasite Plasmodium falciparum. Nature 419: 498- 511

Gilbert DG (1999) SeqPup, biosequence editor & analysis platform, (http://iubio.bio.indiana.edu/soft/molbio/ seqpup/). Bionet.Software

Gilson PR, Beech PL (2001) Cell division protein FtsZ: running rings around bacteria, chloroplasts and mitochondria. Res Microbiol 152: 3-10

Gilson PR, Yu X-C, Hereld D, Barth C, Savage A, Kiefel BR, Lay S, Fisher PR, Margolin W, Beech PL (2003) Two Dictyostelium orthologs of the prokaryotic cell division protein, FtsZ, localize to mitochondria and are required for the maintenance of normal mitochondrial morphology. Eukaryot Cell 2: 1315-1326

Gray MW, Burger G, Lang BF (1999) Mitochondrial evolution. Science 283: 1476-1481

Griparic L, van der Bliek AM (2001) The many shapes of mitochondrial membranes. Traffic 2: 234-244

Guillard RRL (1975) Culture of Phytoplankton for Feeding Marine Invertebrate. In Smith WL, Chanley MH (eds) Culture of Marine Invertebrate Animals. Plenum Press, New York, USA, pp 26-60

Guillard RRL, Ryther JH (1962) Studies of marine planktonic diatoms. I. Cyclotella nana Hustedt and Detonula confervacea Cieve. Can J Microbiol 8: 229-239

Harry EJ (2001) Bacterial cell division: regulating Z-ring formation. MoI Microbiol 40: 795-803

Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754-755

Keeling PJ, Doolittle WF (1997) Widespread and ancient distribution of a noncanonical genetic code in diplomonads. MoI Biol Evol 14: 895-901

Legesse-Miller A, Massol RH, Kirchhausen T (2003) Constriction and Dnm1p recruitment are distinct processes in mitochondrial fission. MoI Biol Cell 14: 1953-1963

Lynch M, Conery JS (2003) The evolutionary demography of duplicate genes. J Struct Funct Genomics 3: 35-44

Ma X, Margolin W (1999) Genetic and functional analyses of the conserved C-terminal core domain of Escherichia coli FtsZ. J Bacteriol 181: 7531-7544

Martin W, Hoffmeister M, Rotte C, Henze K (2001) An overview of endosymbiotic models for the origins of eukaryotes, their ATP- producing organelles (mitochondria and hydrogenosomes), and their heterotrophic lifestyle. Biol Chem 382: 1521-1539

McFadden Gl, Ralph SA (2003) Dynamin: the endosymbiosis ring of power? Proc Natl Acad Sci USA 100: 3557-3559

MeIo EJL, Attias M, Souza WD (2000) The single mitochondrion of tachyzoites of Toxoplasma gondii. J Struct Biol 130: 27-33

Miyagishima S-Y, Nishida K, Kuroiwa T (2003) An evolutionary puzzle: chloroplast and mitochondrial division rings. Trends Plant Sci 8: 432-438

Moreira D, Le Guyader H, Philippe H (1999) Unusually high evolution\ary rate of the elongation factor 1 alpha genes from the Ciliophora and its impact on the phylogeny of eukaryotes. MoI Biol Evol 16: 234-245

Nichols HW (1973) Growth Media - Freshwater. In Stein JR (ed) Handbook of Phycological Methods. Cambridge University Press, pp 16- 17

Nishida K, Takahara M, Miyagishima S-y, Kuroiwa H, Matsuzaki M, Kuroiwa T (2003) Dynamic recruitment of dynamin for final mitochondrial severance in a primitive red alga. Proc Natl Acad Sei USA 100: 2146-2151

Nunnari J, Marshall WF, Straight A, Murray A, Sedat JW, Walter P (1997) Mitochondrial transmission during mating in Saccharomyces cerevisiae is determined by mitochondrial fusion and fission and the intramitochondrial segregation of mitochondrial DNA. MoI Biol Cell 8: 1233-1242

Osteryoung KW, McAndrew RS (2001) The plastid division machine. Annu Rev Plant Physiol Plant MoI Biol 52: 315-333

Osteryoung KW, Vierling E (1995) Conserved cell and organelle division. Nature 376: 473-474

Osteryoung KW, Stokes KD, Rutherford SM, Percival AL, Lee WY (1998) Chloroplast division in higher plants requires members of two functionally divergent gene families with homology to bacterial ftsZ. Plant CeIMO: 1991-2004

Saitou N, Nei M (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. MoI Biol Evol 4: 406-425

Shaw J, Nunnari J (2002) Mitochondrial dynamics and division in budding yeast. Trends Cell Biol 12: 178-184

Stokes, KD, Osteryoung, KW (2003) Early divergence of the FtsZ1 and FtsZ2 plastid division gene families in photosynthetic eukaryotes. Gene 320: 97-108

Strimmer K, von-Haeseler A (1996) Quartet Puzzling: a quartet maximum-likelihood method for reconstructing tree topologies. MoI Biol Evol 13: 964-969

Studie JA, Keppler KJ (1988) A note on the neighborjoining algorithm of Saitou and Nei. J MoI Evol 5: 729-731

Swofford DL (1999) PAUP*. Phylogenetic Analysis Using Parsimony (*and other methods), version 4.0b10. Sinauer Associates, Sunderland, Massachusetts

Takahara M, Takahashi H, Matsunaga S, Miyagishima S, Takano H, Sakai A, Kawano S, Kuroiwa T (2000) A putative ftsZ gene is encoded in the unicellular primitive red alga Cyanidioschyzon merolae. MoI Gen Genet 264:452-460

Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The CLUSTALX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25:4876-4882

van der Bliek A (2000) A mitochondrial division apparatus takes shape. J Cell Biol 151: F1-F4

Vitha S, McAndrew RS, Osteryoung KW (2001) FtsZ ring formation at the chloroplast division site in plants. J Cell Biol 153: 111-119

Wang D, Kong D, Wang Y, Hu Y, he Y, Sun J (2003) Isolation of two plastid division ftsZ genes from Chlamydomonas reinhardtii and its evolutionary implication for the role of FtsZ in plastid division. J Exp Bot 54:1115-1116

Yoon Y, McNiven M (2001) Mitochondrial division: New partners in membrane pinching. Curr Biol 11: R67-R70

Ben R. Kiefel, Paul R. Gilson1, and Peter L. Beech2

Centre for Cellular and Molecular Biology, School of Biological and Chemical Sciences, Deakin University, 221 Burwood Hwy, Melbourne, 3125, Australia

Submitted September 28, 2003; Accepted November 30, 2003

Monitoring Editor: lain R. Wilson

1 Present address: The Walter and Eliza Hall Institute of Medical Research, Parkville, 3050, Australia

2 Corresponding author;

fax 613 9251 7328

e-mail plbeech@deakin.edu.au

Copyright Urban & Fischer Verlag Mar 2004

More News in this Category