Figure 3.13 Unrooted phylogenetic tree, constructed using the neighbour-joining principle, for ATP-dependent Lon protease, a gene suspected of lateral transfer. The tree shows two clusters of archaeal (top right) and bacterial (bottom left) genes; however, the gene from the archaeon M. mazei clusters with the bacteria, rather than with the Archaea. This greatly supports the hypothesis that ATP Lon protease was acquired by M. mazei from a bacterial donor. Scale bar indicates the proportion of amino acid difference. After Deppenheimer et al. (2002), by permission of Horizon Press.
DNA of 'foreign' origin dating from a transfer event millions of years ago is still visible to the present-day genome investigator. At the same time, lateral gene transfer complicates the construction of phylogenetic trees, because the phylogenetic reconstruction of one gene may be different from the reconstruction of another gene if one of them underwent lateral gene transfer. Even the classical gene used for prokaryote phylogeny, the 16 S rRNA gene, which is assumed to be less prone to lateral gene transfer than operational genes, has caused some problems. Phylogenies based on 16 S rRNA are not always consistent with those derived from another essential gene, RNA polymerase. Consequently, the phylogenetic history of a gene is not always a correct indicator of the phylogenetic history of the organisms themselves. In a much-discussed paper, Doolittle (1999) proposed a radical way out of this dilemma, which is to abolish the whole idea of a universal phylogenetic tree of life. Instead, Doolittle (1999) argued that the accepted taxonomic categories, for example Bacteria and Archaea, may be used as convenient descriptors of shared genes, but not as diagnostic indicators of common ancestry. Doolittle (1999) presented a sketch of early evolution in which the base of life is seen as a highly reticulate structure, a network of promiscous gene exchange, from which the three main domains of life finally rise (Fig. 3.14).
Despite the fact that the importance of lateral gene transfer is now well established, the argument that it is an impediment to classifying prokaryotes is not necessarily true. Snel et al. (1999) developed a genome-based phylogenetic approach that goes one step further than just comparing the sequences of genes. These authors considered the fraction of genes shared between genomes as a measure of distance between two species, as follows:
Ma where dAB is the distance between genomes A and B, nAB is the number of genes shared (using an arbitrary threshold level for orthology), and nA is the number of genes in the smallest genome of the two. So in this method the phylogenetic distance between two species is characterized by a single parameter, not by as many parameters as there are shared genes. The phylogeny based on a matrix of pairwise distances calculated in this way was
called a genome phylogeny. Snel et al. (1999) applied this approach to 12 fully sequenced prokaryotes and showed that the resulting tree was actually quite similar to the tree generated by the 16 S rRNA gene. The authors concluded that, despite lateral gene transfer, there is still a very strong phylogenetic signature in the gene content of prokaryotic genomes. However, we should realize that a genome phylogeny captures the central trend of evolutionary history, but does not provide the complete picture (Wolf et al. 2002).
The haploid, circular structure of prokaryotic genomes extends to the genomes of mitochondria and chloroplasts, which have a similar arrangement but are smaller due to loss of many genes. Gray et al. (1999) indicated that the most probable ancestor of the mitochondrion is to be found in the order Rickettsiales of the Alphaproteobacteria. Members of this group include various obligate intracellular parasites such as Rickettsia and Wolbachia. The species Rickettsia prowazekii, the causative agent of a form of typhus transmitted by lice, was long considered to have the most mitochondrion-like genome. However, when more members of the group were sequenced, such as Wolbachia pipientis, an obligate intracellular parasite of Diptera (Wu et al. 2004), doubt was cast on the grouping of mitochondria within the Rickettsiales. Still, evolutionary analysis supports the hypothesis that mitochondria share a common ancestor with the Alphaproteobacteria. Also, the common view remains that the collective mito-chondrial genomes are monophyletic; that is, they all originate from the same ancestor.
The most bacterium-like mitochondrial genome is found in the flagellate Reclinomonas americana (Sarcomastigophora, Histionida). When the mito-chondrial DNA (mtDNA) of this species was sequenced (Lang et al. 1997) it was considered by some as a missing link between bacteria and mitochondria, because of its unusually large number of genes (97, including all the proteins found in other sequenced mtDNAs). The organelle genome database (GOBASE), coordinated by the
University of Montreal in Canada (http: / / megasun.bch.umontreal.ca/gobase), has now collected information on some 50000 mitochondrial sequences, including 429 complete organelle genomes. The mitochondrial genomes of protists probably hold the key to the evolution of the group, because they comprise most of the phylogenetic diversity within the eukaryotes (Gray et al. 1999).
Mitochondria obey the rule that endosymbiosis is accompanied by genome miniaturization (see Section 3.1). In the course of evolution, most of the genetic information for mitochondrial biogenesis and function has moved to the nuclear genome; the proteins needed by the mitochondrion are synthesized in the cytoplasm and then transported across the mitochondrial membrane. Still, the mitochondrial genome encodes several RNAs and proteins essential for mitochondrial function, mostly respiratory complexes of the electron transport chain such as NADH ubiquinone oxi-doreductase, succinate ubiquinone oxidoreductase, ubiquinol cytochrome c oxidoreductase, and cytochrome c oxidase.
Comparisons among mitochondrial genomes are troubled by the fact that loss of genes has occurred many times independently. For example, genome miniaturization in the bacterial rickettsias has taken place independently of miniaturization in the mitochondrial lineages of protists. To complicate things further, some plant mitochondria contain genes that originate from chloroplasts; this holds for two tRNA genes in the mtDNA of Arabidopsis. The result is that the size and the gene content of mitochondria are remarkably divergent between the eukaryote lineages.
Gray et al. (1999) pointed out that mtDNAs come in two basic types, designated as ancestral and derived. Ancestral mitochondrial genomes (for example, the one from Reclinomonas americana) have retained clear vestiges of their eubacterial ancestry, with many non-animal genes, tightly packed in a genome with no or few introns. Derived mitochon-drial genomes are characterized by substantial reduction in genome size, marked divergence of rRNA genes and adoption of biased codon-usage patterns in protein genes. All metazoan and most fungal mtDNAs fall into this category.
The size of mitochondrial genomes varies between less than 6 kbp in Plasmodium falciparum (the human malaria parasite, belonging to the group Apicomplexa) to more than 200 kbp in land plants. Gene content is similarly variable across species. In angiosperms, the mitochondrial genome has evolved to become recombinationally active, which has led to extensive rearrangements of genes, breaking up bacterial gene clusters, and loss of tRNA genes. The mitochondrial genome of Arabidopsis is among the largest of the eukaryotes, but it does not encode many more genes than some of the protist mtDNAs. An overview of the diversity of mitochondrial genome size and content is given in Fig. 3.15.
Interestingly, mitochondrial genomes may fragment into different molecules. This is thought to be due to recombination between repeat segments in different parts of the genome. An example is found in the potato cyst nematode, Globodera pallida (Tylenchida, Heteroderidae), which has a mitochondrial genome consisting of six different circular small chromosomes each 6.3-9.5 kbp. The 12 mitochondrial genes are scattered over the six units, but the ribosomal genes are all concentrated on one of them (Armstrong et al. 2000). Even more surprising is that the frequencies of these mito-chondrial components differ between populations of the nematode. Such small mitochondrial genomes are called subgenomic-sized mtDNAs; they are also found in some green algae and higher plants. Another peculiar situation is due to the presence of plasmids inside mitochondria. Mitochondrial plasmids are especially ubiquitous among filamentous fungi and some of them cause a syndrome of growth loss and early senescence when inserted into ribosomal genes of the mitochondrial genome (Maas et al. 2005).
In addition to mitochondria, the other main eukaryotic organelle of bacterial origin is the chloroplast. Comparative studies on bacterial and chloroplast genomes have now demonstrated convincingly that chloroplasts are derived from a cyanobacterium related closely to the present species Nostoc punctiforme. It is also evident that chloroplast genomes jointly are one monophyletic group; that is, they all descend from the same
Porphyra Acanthamoeba Schizosaccharomyces Plasmodium
Figure 3.15 Genome size and gene content of mitochondrial DNAs across a wide range of species. (a) Circles and lines represent circular and linear chromosomes, with the ORFs of known function shown as dark lines. The major groups to which the species mentioned belong are as follows: Rickettsia (Alphaproteobacteria), Arabidopsis (angiosperm plant), Marchantia (liverwort), Jakoba, Reclinomonas (flagellates), Allomyces (fungus), Prototheca (green alga), Tetrahymena (ciliate), Acanthamoeba (amoeba), Ochromonas (golden alga), Phytophthora (oomycote), Chondrus (red alga), Chlamydomonas eugamatos, Chlamydomonas reinhardtii (green algae), Schizosaccharomyces pombe (yeast), Homo (human) and Plasmodium (malaria parasite). (b) Gene complement of mitochondrial genomes, showing the overlap between species. Each ellipse corresponds to one organism and includes all the mitochondrial genes of that organism. For example, the tiny mitochondrial genome of Plasmodium has four genes, which are also found in all other mitochondria, while all known mitochondrial genes are found in the mitochondrial genome of Reclinomonas. The genes are designated by code names. Reprinted with permission from Gray et al. (1999). Copyright 1999 AAAS.
ancestor. This is not to imply that the symbiotic event that gave rise to the chloroplast occurred only once. In fact, it is assumed that the initial inclusion of a cyanobacterium (which led to red algae, green algae, and higher plants) was followed by a second endosymbiotic event, probably involving a red alga, which produced cells in which the chloroplast was surrounded by four, rather than two, membranes. From this type of cell three different evolutionary lineages are assumed to have originated (Tudge 2000; Raven and Allen 2003; Falkowksi et al. 2004):
• Brown algae (Phaeophyta), diatoms (Bacillario-phyta), golden algae (Xanthophyta), and water molds (Oomycota), with the chloroplast secondarily lost in the Oomycota.
• Dinoflagellates (Dinoflagellata), Apicomplexa (Plasmodium and other parasites), and ciliates
(Ciliata), with a strong reduction of the chloroplast in apicomplexans and a complete loss of the chloroplast in ciliates.
• Kinetoplastids (parasites like Trypanosoma) and Euglenozoa, with secondary loss of the chloroplast in kinetoplastids and some euglenoids.
So, the present-day scattered distribution of photosynthetic capacity across the eukaryotes is explained not only by gains but also by losses of chloroplasts. Losses are especially prominent in the lines that obtained the chloroplast through double symbiosis. Different degrees of chloroplast reduction can be observed in the phylum Apicomplexa. These unicellular organisms are characterized by an organelle called an apicoplast, an assumed relict of a chloroplast, with a greatly reduced genome; however, not all species have this organelle. A recent genomic survey of the human parasite, Cryptosporidium parvum, which lacks an apicoplast, demonstrated the presence of 31 genes of likely cyanobacterial and other prokaryote origin in the genome, confirming the theory that apicomplexans evolved from a plastid-containing lineage (Huang et al. 2004).
The genome of a chloroplast typically encodes 60-200 proteins, which is more than an order of magnitude less than the genome of a cyano-bacterium, which encodes at least 1500 proteins. Genes in the preplastid were either lost or transferred to the nucleus. Studies on Arabidopsis nuclear and chloroplast genomes (Martin et al. 2002) have shown that thousands of genes have been transferred. Some 4500 genes in the nuclear genome of Arabidopsis, or 18% of the genes, appear to have a bacterial (chloroplast) origin. This is not to say that the products of these genes are all functional in the chloroplast; actually, more than half of the originally chloroplastic genes are now targeted to other cell compartments. To complicate things further, the protein products of many nuclear genes that were not acquired from the plastid ancestor are now targeted to the plastid! In this complicated interplay between genomes, many issues remain unresolved, including the fundamental question of why chloroplasts have a genome at all, if genes can be transferred so easily to the nucleus (Raven and Allen 2003). There must be some crucial selective advantage in retaining some genes in chloroplasts, but not others.
Was this article helpful?