2013-12-31

Phylogeny of eukaryotes

There have been some interesting developments regarding eukaryote phylogenetics recently. Particularly, it has become quite obvious that some biokonts are more closely related to unikonts than to the other bikonts, and it now seems likely that the last common ancestor of extant eukaryotes was a bikont with two anterior (part of the cell that faces the direction of movement) cilia arising from two basal bodies, one cilium directed forward, the other one backwards (Paps et al. 2013). A few years ago it was unclear if the last common ancestor of eukaryotes was uni- or bikont. Apusomonadida, Ancyromonadida, Mantamonadida, Breviatea, Diphyllatea (Fig. 1), and some other groups were  on morphological grounds regarded as bikonts (although some species among them are with only one or more than two cilia) and therefore more closely related to other bikonts (Diaphoretickes and Excavata) than to unikonts (Amoebozoa and Opisthokonta), but no phylogenomic data were available from these enigmatic lineages. Some limited molecular data supported that (Stechmann & Cavalier-Smith 2002) and some not (Kim et al. 2006). Now, in short succession, several interesting papers have been published analysing phylogenomic datasets including Apusomonadida, Breviatea, and Diphyllatea (Derelle & Lang 2012; Brown et al. 2013; Zhao et al. 2013). It is quite clear now that bikonts, Apusomonadida and Breviatea, are more closely related to Opisthokonta (one of the unikont lineages with two posterior basal bodies, but only one posterior flagellum) than to any other eukaryote group. Phylogenetic position of Diphyllatea is at the moment less certain, but it appears also to be more closely related to unikonts than to other bikonts (Zhao et al. 2013). There is a nice figure in Paps et al. (2013) summarizing evolution of ciliary apparatus in eukaryotes. In short, last common ancestor of extant eukaryotes may have possessed two anterior cilia arising from two basal bodies. In majority of Amoebozoa the cilia were lost (nevertheless there are amoebozoans with one or two anterior cilia) and in Opisthokonta ciliary apparatus with two basal bodies and one cilium moved to posterior part of the cell. Some Diphyllatea have more than two cilia and some Breviatea only one. Diverse group of Bikonta (Diaphoretickes + Excavata) includes many groups with more than two cilia or with none at all, but ancestral condition of two anterior cilia seems quite likely.


Fig. 1. Phylogeny of Eukaryota based on recent phylogenetic analyses (Hampl et al. 2009Katz et al. 2011Burki et al. 2012Laurin-Lemay et al. 2012Price et al. 2012Timme et al. 2012Zhao etal. 20122013; Brown et al. 2013Paps et al. 2013Seenivasan et al. 2013Yabuki et al. 20102013). Amoeboid intracellular algal parasite group Aphelidea, which were thought to belong to Holozoa (Adl et al. 2012), were actually found to group with Fungi (Karpovet al. 2013). An intracellular parasite of oysters, Mikrocytos, has been confirmed to be a rhizarian (Burki et al. 2013). Consulted classifications: Adl et al. 2012Cavalier-Smith (2013). Cavalier-Smith's classifications are quite idiosyncratic, super complicated, and change rather often (he's a very prolific author). At least this is the impression I'm getting. He also uses many (often novel) paraphyletic taxa.

The other interesting development is the rooting of the eukaryote tree, which is a difficult problem to solve because of the lack of a close out-group (all the prokaryotes are too distant) and rapid radiation of lineages at the base of extant eukaryotic tree of life (it is more like a bush than a tree in this case) (Brinkmann & Philippe 2007). Derelle & Lang (2012) have compiled an interesting dataset of 42 mitochondrial proteins (encoded by mitochondrial or nuclear genome) of alpha-proteobacterial origin to root the eukaryote tree. All known eukaryotes have mitochondria or their derivatives (mitosomes or hydrogenosomes; Hjort etal. 2010; Shiflett & Johnson 2010) and thanks to this, genes of mitochondrial origin can be used to reconstruct eukaryote phylogeny (Brinkmann & Philippe 2007). And the other good news is that there is a relatively close outgroup available - Alphaproteobacteria. Previous analyses relying heavily on nuclear informational genes (replication, transcription, translation) of archaeal origin (closest prokaryotic group to eukaryote nuclear lineage) were plagued by long-branch attraction (LBA) artefacts. LBA is an artefact of phylogenetic analyses where long branches (measured in the amount of estimated character change, e.g. nucleotide or amino acid substitutions) tend to group together regardless of evolutionary relationships, if the model or method used is not adequate to correct for multiple substitutions at the same site. For example, the fast evolving lineages (evident as long branches in the phylogenetic trees) of eukaryotes (often parasites) tend to be attracted towards the outgroup represented by long branch leading to Archaea, and in consequence distorting true relationships between eukaryotes. Derelle & Lang (2012) and now also Zhao et al. (2013) found that the root of the eukaryote tree falls between bikonts and unikonts, as suggested about 10 years ago (Stechmann & Cavalier-Smith 2002, 2003; Richards & Cavalier-Smith 2005), with the caveat that some bikonts, like Apusomonadida and Diphyllatea, are 'unikonts'. Standard phylogenomic analyses of eukaryotes (based on usually 100­­–200 highly expressed genes; e.g. many references in the caption of Fig. 1) without prokaryote outgroups are consistent with this kind of rooting (it is possible to root the resulting unrooted trees so that unikonts and bikonts are monophyletic). It is remarkable, apart from the rooting, that most of the well supported clades found in standard phylogenomic analyses are upheld by Derelle & Lang (2012) and Zhao et al. (2013), because most of the 42 proteins used in the mitochondrial dataset are different from the other analyses. For example, only 3 genes are in common with a dataset of 143 genes used by Hampl et al. (2009) (Derelle & Lang 2012). This gives some confidence that we are not completely lost in deciphering these ancient phylogenies.

Another fascinating aspect of eukaryote evolution is how they came to be in the first place. A straw man of a traditional view is that eukaryote gradually evolved from prokaryotic ancestor, acquiring cytoskeleton, phagocytosis, internal membranes (nuclear envelope, endoplasmic reticulum, golgi apparatus) and then engulfing bacteria which later became mitochondria and plastids. But now, more and more viable alternative seems to be that the origin of mitochondria marks the origin of eukaryotes. Martin & Müller (1998) proposed the hydrogen hypothesis for the first eukaryote, where symbiotic association between an Archaea and Alphaproteobacteria finally lead to Alphaproteobacteria living inside the Archaea. Nick Lane popularized this hypotheses in a book "Power, Sex, Suicide: Mitochondria and the Meaning of Life" (2005). Although the title might suggest some kind of a new age nonsense, it is in reality amazingly insightful and wonderfully written book about fundamentals of biology. I wish I read it long time ago (what other similar gems I might be missing?). Lane makes a convincing case why prokaryotes could not simply evolve to eukaryotes and that mitochondria were needed to give a possibility to evolve large and dynamic cells with large genomes, which could then lead to multicellularity. This idea has been elaborated and supported with numbers by Lane & Martin (2010). According to this view, most of the defining features of eukaryotes, like dynamic cytoskeleton, phagocytosis, nucleus, endomembrane trafficking, and sex, all evolved because of mitochondria! The arguments are related to energy generation and requirement for mitochondria to retain control over respiration. Internalization of energy production by mitochondria inside the host cell made it more efficient (by overcoming surface-to-volume constraints) and freed its cell membrane for other tasks (prokaryotes do not have this kind of luxury). Because oxygen is very reactive and easily damages cell constituents, it was necessary to tightly regulate oxygen level inside the cell. The most efficient way seems to be that mitochondria retain part of their genome so that they can quickly respond to varying levels of oxygen by controlling synthesis of respiratory complexes. There is the same core set of about 10–30 genes, which all free living aerobic eukaryotes have retained in their mitochondrial genomes (some have additional genes, which others have lost). Granted with specialized energy producing organelles inside the cell, nuclear genome could freely acquire more and more different protein coding genes and regulatory sequences without high energetic costs to express them. This opened the road to higher complexity and eventually to multicellularity. In line with the view that mitochondria started eukaryogenesis are recent (e.g. Cox et al. 2008; Guy & Ettema 2011; Lasek-Nesselquist & Gogarten 2013; Williams et al. 2013) and not so recent (e.g. Lake et al.1984; Rivera & Lake 1992) findings that nuclear lineage of eukaryotes probably originates within Archaea (it appears that Thaumarchaeota, Aigarchaeota, Crenarchaeota, and Korarchaeota are more closely related to eukaryotes than they are to other Archaea, Euryarchaeota and Nanoarchaeota). Nevertheless, these findings themselves do not exclude the possibility that the eukaryotic lineage within Archaea evolved (some) eukaryotic characters before acquiring mitochondria.