Contrasting Mitochondrial Genome Organizations and Sequence Affiliations among Green Algae: Potential Factors, Mechanisms, and Evolutionary Scenarios1

The three green algal mitochondrial genomes completely sequenced to date—those of Chlamydomonas reinhardtii Dangeard, Chlamydomonas eugametos Gerloff, and Prototheca wickerhamii Soneda & Tubaki—revealed very different mitochondrial genome organizations and sequence affiliations. The Chlamydomonas genomes resemble the ciliate/fungal/animal counterparts, and the Prototheca genome resembles land plant homologues. This review points out that all the green algal mitochondrial genomes examined to date resemble either the Chlamydomonas or the Prototheca mitochondrial genome; the Chlamydomonas-like mitochondrial genomes are small and have a reduced gene content (no ribosomal protein or 5S rRNA genes and only a few protein-coding and tRNA genes) and fragmented and scrambled rRNA coding regions, whereas the Prototheca-like mitochondrial genomes are larger and have a larger set of protein-coding genes (including ribosomal protein genes), more tRNA genes, and 5S rRNA and conventional continuous smallsubunit (SSU) and large-subunit (LSU) rRNA coding regions. It appears, therefore, that the differences previously observed between the mitochondrial genomes of C. reinhardtii and P. wickerhamii extend to the two green algal mitochondrial lineages to which they belong and are significant enough to raise questions about the causes and mechanisms responsible for such contrasting evolutionary strategies among green algae. This review suggests an integrative approach in explaining the occurrence of distinct evolutionary strategies and apparent phylogenetic affiliations among the known green algal mitochondrial lineages. The observed differences could be the result of distinct genetic potentials differentiated during the previous evolutionary history of the flagellate ancestors and/or of subsequent changes in habitat and life history of the more advanced green algal lineages. Key index words: Chlamydomonas; Prototheca; green algae; mitochondrial genome origin and evolution The eukaryotic cell is an associative system comprising at least two or three main subsystems with different evolutionary histories (Margulis 1981, Gray 1992). Well accepted now is the eubacterial (alphaproteobacterial and cyanobacterial, respectively) endosymbiotic origin of two of the eukaryotic cell’s or1 Received 20 May 1997. Accepted 30 October 1997. 2 Author for reprint request; e-mail [email protected]. ganelles, namely, the mitochondria and the plastids, although their monoor polyphyletic origin is still debated (Dayhoff and Schwartz 1981, Stewart and Mattox 1984, Gray et al. 1989, Lockhart et al. 1992, Morden et al. 1992). Single endosymbiotic events accounting for the origin of mitochondria and plastids, respectively, would imply that some common ancestral characters should be present in all the extant lineages and that distinct derived traits should be developed within and shared among related lineages. In addition, monophyletic origins for the mitochondria and plastids would require that phylogenies based on organellar traits be consistent with those based on nuclear or nucleus-encoded features; in other words, all the compartments within an eukaryotic cell should resemble their corresponding counterparts in the same compared lineage. However, there are reports of lineages in which the organelles and the nucleo-cytosolic compartment do not display the same phylogenetic affiliations, specifically, the plastids of the cryptomonads (Douglas 1992, 1994, McFadden and Gilson 1995) and the protist Euglena gracilis Klebs (Gibbs 1978, 1981, Morden et al. 1992) and the mitochondria of land plants (Gray 1989), the protozoan Acanthamoeba castellanii (Lonergan and Gray 1994), and the green alga Chlamydomonas reinhardtii Dangeard (Gray 1992). The question to be addressed in such cases is: Are these examples of incongruence between the nuclear and organelle phylogenies the result of very divergent evolutionary strategies among closely related lineages or, rather, of different evolutionary origins? An interesting case with which to address these types of evolutionary questions is represented by the green algae. The only three green algal mitochondrial genomes completely sequenced to date— namely, those of C. reinhardtii (Boer and Gray 1988a, b, c, Gray and Boer 1988, Michaelis et al. 1990), Chlamydomonas eugametos Gerloff (Denovan-Wright et al. 1998), and Prototheca wickerhamii Soneda & Tubaki (Wolff et al. 1994) revealed very different mitochondrial genome organizations and sequence affiliations. The mitochondrial genomes of the two Chlamydomonas taxa on the one hand and that of Prototheca on the other resemble more the ciliate/ fungal/animal and land plant mitochondrial types, 17 GREEN ALGAL MITOCHONDRIAL GENOME EVOLUTION respectively, than one another. In other words, there is an unexpected incongruence between the phylogenetic relationships suggested by the nucleocytosolic compartments of these two green algal lineages and those suggested by their mitochondria. The phylogeny of the green algal group is progressively being deciphered, and the new information gathered through molecular approaches will probably trigger the reconsideration of the traditional green algal systematics (Chapman and Buchheim 1991, Friedl 1995). Chlamydomonas and P. wickerhamii are traditionally considered members of different orders (the Chlamydomonadales and the Chlorococcales, respectively) that belong to the same class (i.e. the Chlorophyceae; sensu Mattox and Stewart 1984). However, phylogenies based on nuclear rDNA sequences (Wilcox et al. 1992, Steinkötter et al. 1994, Friedl 1995) suggest that although some members of the Chlorococcales, such as Scenedesmus obliquus (Turp.) Kutz, do indeed affiliate with chlorophycean taxa, other members, including P. wickerhamii, form a monophyletic group with advanced lineages of the Pleurastrophyceae class (sensu Mattox and Stewart 1984). This latter group was recently defined by Friedl (1995) as a new class: the Trebouxiophyceae. In this context, it appears that Chlamydomonas and Prototheca, the two green algal lineages whose mitochondrial genomes appear unexpectedly different, may not be as closely related as previously thought; in fact, they are members of two distinct green algal evolutionary lineages, namely, the chlorophycean and the trebouxiophycean (sensu Friedl 1995), whose divergence is probably very old. Although the phylogenetic relationships among green algal lineages are not fully deciphered, both Chlamydomonas and Prototheca are unquestionably green algae (as suggested by ultrastructural, biochemical, and molecular data from both the nucleocytosolic and the chloroplast compartments; Devereux et al. 1990, Chapman and Buchheim 1992, Morden et al. 1992), and green algae are the closest relatives of land plants (McCourt 1995). However, in an archaebacterial-eubacterial-chloroplast-mitochondrial-nuclear phylogenetic tree inferred from small-subunit (SSU) rDNA sequence data, the C. reinhardtii nuclear and mitochondrial sequences occupied different positions relative to land plants (Gray et al. 1989). In the nuclear subtree, C. reinhardtii formed a clade with the plant sequences (as it did also in the chloroplast subtree) and branched off at about the same point as animals and fungi. In contrast, in the mitochondrial subtree, C. reinhardtii branched with the ciliate/fungal/animal sequences, far away from higher plants, which clustered very near the root, close to the alpha-proteobacterial clade. The affiliation of the nuclear SSU rDNA sequences of higher plants and C. reinhardtii was seen as consistent with traditional phylogenies that consider green algae the closest relatives of land plants (Ragan and Chapman 1978, Chapman and Buchheim 1991), whereas the green algal/land plant dichotomy in the mitochondrial tree was interpreted as an anomaly. However, this anomaly in branching topology was attributed to the plant rather than C. reinhardtii mitochondrial sequences and was shown not to be a consequence of differential rate of sequence divergence. Moreover, the authors indicated that plant mitochondrial rRNAs resemble more their eubacterial/chloroplast counterparts than they do their homologues in other mitochondria (Gray et al. 1989). To explain the different branching position of plants within the nuclear and mitochondrial lineages, respectively, and to account for the strong eubacterial features of their mitochondrial rRNAs, Gray et al. (1989) suggested two possibilities: either 1) the mitochondrial rRNA genes of plants have diverged relatively little from the rRNA genes of the ancient eubacterial ancestor of all mitochondria (monophyletic origin) or 2) the higher plant mitochondrial rRNA genes or the mitochondria itself have been acquired more recently than those of other eukaryotic lineages (biphyletic origin). In addition, because of the very different way in which genes are organized and expressed in the C. reinhardtii and plant mitochondrial genomes, the authors concluded that there is no indication that the two shared a common mitochondrial ancestor as recently as they shared a common nuclear (or chloroplast) ancestor. The input of other green algal mitochondrial rDNA sequences, namely, of P. wickerhamii (Wolff and Kück 1990, Wolff et al. 1993) and C. eugametos (Denovan-Wright et al. 1996) did not contribute to the dissolution of the land plant/nonplant split observed in the mitochondrial rRNA trees (Gray et al. 1989), nor did these sequences resolve Chlamydomonas and P. wickerhamii as a green algal clade. However, the fact that the Prototheca sequences are relatively AT rich and rapidly evolving in comparison to their land plant counterparts was considered to possibly account for this green algal taxon branching apart from land plants (Gray 1995). On the other hand, although the number of transitional substitutions is probably saturated in the relatively rapidly evolving mitochondrial rRNA sequences, it was suggested that the apparent affiliation of the Chlamydomonas sequences with ciliate/fungal/yeast counterparts and therefore their separation from the land plant sequences is not due to a ‘‘long-branch length attract’’ artifact (Denovan-Wright et al. 1996). However, phylogenetic trees based on COX1 amino acid sequences indicated that the plant and green algal mitochondrial lineage, including P. wickerhamii (Wolff et al. 1993) and the prasinophyte Platymonas (Tetraselmis) subcordiformis (Wille) Hazen (Kessler and Zetsche 1995), do form a monophyletic group. Therefore, the expected congruency of nu-