See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/325858928

Universal Tree of Life

Chapter · June 2018

DOI: 10.1002/9780470015902.a0001525.pub3

CITATIONS READS

0 1,619

3 authors:

Patrick Forterre Violette Da Cunha

Institut Pasteur International Network Institut de Biologie Intégrative de la Cellule
423 PUBLICATIONS   14,748 CITATIONS    92 PUBLICATIONS   937 CITATIONS   

SEE PROFILE SEE PROFILE

Morgan Gaia
Genoscope - Centre National de Séquençage
79 PUBLICATIONS   514 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Structure, function and evolution of the Sua5/TsaC universal enzyme family View project

The first identified Integrative and Conjugative Element using a DDE transposase View project

All content following this page was uploaded by Morgan Gaia on 31 January 2019.

The user has requested enhancement of the downloaded file.

Universal Tree of Life Introductory article

Patrick Forterre, Unité de Biologie Moléculaire du Gène chez les Extrêmophiles Article Contents
• Introduction
(BMGE), Département de Microbiologie, Institut Pasteur, Paris, France
• The Root of the Tree and the Nature of the Last
Violette Da Cunha, Unité de Biologie Moléculaire du Gène chez les Extrê- Universal Common Ancestor (LUCA)

mophiles (BMGE), Département de Microbiologie, Institut Pasteur, Paris, France • Specific Molecular Features of Modern
Organisms Originated in the Various Branches of
Morgan Gaïa, Unité de Biologie Moléculaire du Gène chez les Extrêmophiles the Universal Tree
• The Universal Tree and the Origin of Eukaryotes
(BMGE), Département de Microbiologie, Institut Pasteur, Paris, France
• Where Are Viruses in the Universal Tree of Life?

Online posting date: 14th June 2018

The first tree of life based on the comparison of 1990). Archaea and Bacteria correspond to prokaryotic cells in
ribosomal RNA (ribonucleic acid) allowed classi- which transcription and translation both take place in the cyto-
fying all ribosome-encoding organisms into three plasm, whereas Eukarya are composed of eukaryotic cells in
domains: Archaea, Bacteria and Eukarya. Phyloge- which transcription takes place in a nucleus while translation
nomics analyses later on show that these domains takes place in the cytoplasm. Importantly, despite being prokary-
otes, Archaea are much more similar to Eukarya than to Bacteria
probably emerged from a last universal common
at the molecular level, as nicely illustrated by the many ribo-
ancestor (LUCA) simpler than modern organisms.
somal proteins (r-proteins) specifically shared by Archaea and
Different results are however still obtained regard- Eukarya, when none are specifically shared between Bacteria and
ing the relationships between Archaea and Eukarya another domain (Lecompte et al., 2002; see as follows). Dur-
when relying on concatenations of universal pro- ing the last two decades, the advances in DNA (deoxyribonu-
teins. Recent analyses have revealed that the topol- cleic acid) sequencing techniques allowed recovering more and
ogy obtained in these studies is strongly depen- more protein sequence data, first from fully sequenced genomes
dent on the universal proteins and species data and more recently from genomes reconstructed from metage-
sets. On the basis of this observation, our recent nomics data. This favoured the construction of universal trees
phylogenetic analyses presently support the sis- based on concatenations of universal protein sequences, and led
terhood of Archaea and Eukarya. Viruses, defined to new views and controversies about the topology of the univer-
sal tree, especially concerning the relationship between Archaea
as virion-encoding organisms, cannot be formally
and Eukaryotes (Gribaldo et al., 2010).
located in the tree of life but populate it entirely
from its trunk to the leaves.

The Root of the Tree and the

Introduction Nature of the Last Universal
Common Ancestor (LUCA)
More than 160 years ago, in 1857, Darwin had already anticipated
‘the time [ … ] when we shall have very fairly true genealog- The root of the universal tree corresponds to the position of
ical trees of each great kingdom of nature’. This nonethe- the last universal common ancestor (LUCA). This root was first
less remained an unfulfilled fantasy until the second part of positioned using phylogenetic analyses with paralogous proteins
the twentieth century, when the molecular biology revolution that diverged by duplication before the emergence of LUCA
offered the possibility to compare sequences of informational (Iwabe et al., 1989; Gogarten et al., 1989; Brown and Doolittle,
macromolecules, proteins and/or nucleic acids. The first trees, 1997). From these studies, it was proposed to position LUCA
which were based on ribosomal RNA (ribonucleic acid) sequence between a lineage leading to Bacteria and a lineage leading
comparison, allowed to classify all ribosome-encoding organisms to both Archaea and Eukarya. With this root, Archaea and
(sensu Raoult and Forterre, 2008) into three domains: Archaea, Eukarya form a monophyletic clade that Forterre has recently
Bacteria and Eukarya (Woese and Fox, 1977; Woese et al., suggested calling Arkarya (Forterre, 2015). The phylogenetic
analyses that originally supported the rooting between Bacte-
eLS subject area: Evolution & Diversity of Life ria and Arkarya were nonetheless plagued by potential artificial
How to cite: attractions between the long branches leading to Bacteria and the
Forterre, Patrick; Da Cunha, Violette; and Gaïa, Morgan (June long branches leading to the paralogous proteins used to root the
2018) Universal Tree of Life. In: eLS. John Wiley & Sons, Ltd: trees (Forterre and Philippe, 1999). More recently, the rooting of
Chichester. the universal tree was tentatively deduced from the distribution
DOI: 10.1002/9780470015902.a0001525.pub3 of r-proteins in modern organisms (Forterre, 2015). Besides the

eLS © 2018, John Wiley & Sons, Ltd. www.els.net 1

 Universal Tree of Life

2015). Further insights about the nature of LUCA can be obtained

Eukaryotes by comparative molecular biology. Importantly, the three major
DNA replication proteins (replicase, primases and helicases) are
11 not homologous between Bacteria and Arkarya, suggesting that
LUCA had still an RNA genome if the tree is rooted between
these two clades (Forterre, 2006). If correct, this implies that the
33 0
ancestor of the homologous DNA-dependent RNA polymerases
34 that transcribe DNA into RNA in the three domains worked in
1 23 LUCA as an RNA replicase. In agreement with this hypothesis,
0
some modern RNA viruses use eukaryotic RNA polymerase II
Archaea Bacteria to replicate their genomes (Chang et al., 2008). The basal tran-
scription factors are also nonhomologous between Bacteria and
(a)
Arkarya, suggesting that RNA polymerase performed nonspe-
cific initiation of transcription in LUCA (Werner and Grohmann,
Archaea Eukaryotes 2011). In agreement with an RNA-based LUCA, it is worth noting
Bacteria
that the only DNA topoisomerases family that can be traced back
to LUCA, that is the type IA family (Forterre and Gadelle, 2009),
+1 is also the only family of DNA topoisomerases with members
having RNA topoisomerase activity (Garnier et al., 2018).
+11
The metabolism of LUCA remains controversial and elu-
+23 Arkarya sive. A recent comparative analysis of carbon monoxide
34 +33
LUCA
dehydrogenase/acetyl-CoA synthase complex in Bacteria and
Archaea indicates that at least four of the five proteins in this
(b) complex were present in LUCA, and the authors concluded
that their results are equally compatible with an autotrophic,
Figure 1 (a) Venn diagram representing the number of shared or specific mixotrophic or heterotrophic LUCA (Adam et al., 2018). In any
ribosomal subunits in the ribosomes of the three domains. (b) Hypothesis case, it is probable that LUCA had no membrane-bound ATP
of distribution of the r-proteins along the evolution of the ribosomes, from (adenosine triphosphate) synthase and used a simpler system
LUCA to the three cellular domains.
(fermentation) to produce ATP (Mulkidjanian et al., 2007).
Indeed, the proteins involved in the rotor-based mechanism of
ATP synthesis in the bacterial F0 /F1 ATP synthase complex
34 universal r-proteins, bacterial ribosomes contain 23 specific and in the arkaryal ATP synthase (in Archaea) or V-ATPase (in
r-proteins, whereas the archaeal and eukaryotic ribosomes share Eukarya) are not homologous.
33 r-proteins that are absent from bacterial ribosomes (Lecompte Notably, if LUCA was much simpler than Archaea and Bacte-
et al., 2002) (Figure 1a). This makes the rooting between Bac- ria in terms of cellular organisation and fundamental molecular
teria and Arkarya more parsimonious, since the present pattern features, some traits common to these two domains, and often
of r-proteins distribution can be simply explained by the addition associated to the ‘prokaryotic’ phenotypes (a small circular DNA
of the 23 bacterial-specific r-proteins in the branch of the tree genome organised into operons, the coupling of transcription and
leading to Bacteria, and of the 33 arkaryal-specific r-proteins in translation, small size and high macromolecular turnover, etc.),
the branch leading to Eukaryotes and Archaea (Figure 1b). If the could have originated from convergent evolution.
root is located in the specific eukaryotic or archaeal branch, one
should imagine that these 33 r-proteins were already present in
LUCA and later on replaced by 23 nonhomologous r-proteins in
Bacteria. There is no obvious selection pressure to explain such Specific Molecular Features of
dramatic nonhomologous replacements, especially considering Modern Organisms Originated in
that these proteins are often located at similar positions in the
ribosome. The same reasoning can be made for the origin and
the Various Branches of the
evolution of other molecular features that are not homologous Universal Tree
between Bacteria and Arkarya (see as follows).
Rooting the tree of life between Bacteria and Arkarya suggests A common misconception in discussing early-life evolution is to
that LUCA was an organism simpler than modern cells, with confuse Bacteria with primitive organisms (first forms of life on
ribosomes containing at least the ancestors of the 34 r-proteins our planet). Indeed, Bacteria are much more complex than the
present in the ribosomes of all organisms living today. However, simple LUCA that we just described. The branch of the tree lead-
LUCA was probably already equipped to synthesise faithfully ing from LUCA to the last bacterial common ancestor (LBCA)
sophisticated proteins since the set of universal proteins (pro- marked the development of complex bacterial traits, such as
teins hence most likely present in LUCA) also contains some the peptidoglycan, a formidable type of complex cell wall, an
tRNA (transfer ribonucleic acid) modification enzymes known extremely sophisticated flagellum, a very fast replication machin-
to improve translation fidelity (Grosjean et al., 2007; Forterre, ery (with Okazaki fragments 10 times longer than in Archaea

2 eLS © 2018, John Wiley & Sons, Ltd. www.els.net

 Universal Tree of Life

Bacteria Archaea Eukaryotes

Specific lipids
ATP synthase
Reverse gyrase…
LBCA LECA
LACA Nucleus
Peptidoglycan es
ar yot Nucleolus
Complex flagellum uk
to-e Spliceosome
Complex cell division Pro V-ATP synthase
F0/F1 ATP synthase
Mitosis
Gyrase…
Meiosis…

Figure 2 Some specific molecular features that originated in the different branches of the universal tree of life.

and Eukarya), an elaborated mechanism of cell division and seg- Bacteria (Brochier-Armanet and Forterre, 2007). The emergence
regation based on the FtsK motor protein, and a sophisticated of this enzyme has probably been a major step in the evolution
mechanism of ATP production, the F0 /F1 ATP synthase. Another of life, allowing the colonisation of high temperature biotopes.
critical event that took place in the lineage leading to Bacteria is Eukarya emerged late in the evolution of life, since the last
the emergence of DNA gyrase, a type II DNA topoisomerase that eukaryotic common ancestor (LECA) already contained mito-
introduces negative supercoiling in DNA and directly couples the chondria (derived from an internalised alpha-Proteobacteria) and
energetic state of the cell to its transcription pattern via its control all complex features that are universal hallmarks of eukaryotes,
of intracellular DNA topology. such as the nucleus and nucleolus, spliceosomal introns, mito-
Several critical innovations also emerged in the lineage leading sis, sex and meiosis (Koonin, 2010). Notably, recent discoveries
to Archaea, such as another unique mechanism of ATP production of animal-like macrofossils in oxygenated environments from 2.1
(the archaeal ATP synthase) and new types of lipids that maintain Gya ago and of fungus-like mycelial fossils in 2.4-Gya-old vesic-
the impermeability of membranes at very high temperatures. ular basalt (El Albani et al., 2010; Bengtson et al., 2017) revealed
This observation suggests that the last archaeal common ancestor that LECA might have originated much earlier than previously
(LACA) was a hyperthermophilic organism, in agreement with assumed.
the presence of hyperthermophiles in most archaeal phyla. The It is presently unclear if proto-eukaryotes emerged early or
presumed hyperthermophilic phenotype of LACA could explain late compared to Archaea and Bacteria and the timing of the
why no RNA viruses are presently known to infect Archaea, as emergence of the major eukaryotic hallmarks is still controversial
RNA is much more sensitive than DNA to high temperatures. (Pittis and Gabaldón, 2016) (Figure 2). However, one can assume
The LACA and the LBCA were possibly both thermophiles that this emergence was a slow process that took time, suggesting
and/or hyperthermophiles, as suggested by phylogenetic that proto-eukaryotes (those that evolved before LECA) evolved
reconstructions of their putative rRNA (ribosomal ribonu- upon a rather long period of time and might have emerged more
cleic acid) and protein sequences (Boussau et al., 2008; Groussin than 3 Gya ago.
and Gouy, 2011). In contrast, the same strategy suggested that
LUCA was a mesophile or a moderate thermophile. This result
would imply that the ability to adapt to high temperature envi-
ronments had played a major role in the selection of the two
The Universal Tree and the Origin
prokaryotic domains, possibly favouring thermoreduction mech- of Eukaryotes
anism (Forterre, 1995, 2013). All hyperthermophilic archaea
and bacteria contain reverse gyrase, an ATP-dependent type I The stem branches leading from the trifurcation point connecting
DNA topoisomerase that introduces positive supercoiling into the three domains to the last common ancestors of each domain
DNA in vitro and is essential for life at very high temperatures are strikingly different in universal trees based on proteins. The
(Brochier-Armanet and Forterre, 2007). Phylogenetic analysis bacterial branch is by far the longest one, whereas the specific
revealed that this enzyme was probable not present in LUCA, archaeal branch is very short. For some authors, this suggests
but probably emerged in Archaea and was later on transferred to that the short archaeal branch is due to a long-branch attraction

eLS © 2018, John Wiley & Sons, Ltd. www.els.net 3

 Universal Tree of Life

the long eukaryotic and bacterial branches. The rRNA molecules,

Bacteria Eukarya with only four nucleotide letters, contain very limited information
‘Eocytes’ to clearly discriminate such events, and indeed different authors
have obtained either the Woese or eocyte trees using rRNA.
Universal proteins are theoretically better suited to identify the
correct topology using the 20 amino acid letters alphabet. How-
ever, single-protein trees of universal proteins provide trees that
are often not robust and contradict each other (Gribaldo et al.,
Archaea
2010). Most authors have thus used concatenations of different
sets of universal proteins to try recovering the correct universal
LUCA tree topology, by accreting the signal embedded in each protein
sequence. In the last 10 years, several groups have published
(a)
trees of life supporting the eocyte topology based on this strategy
(Cox et al., 2008; Guy and Ettema, 2011; Raymann et al., 2015).
Bacteria Archaea Eukarya More recently, the group of Thijs Ettema in Upsalla has pub-
lished universal trees in which Archaea emerged from a novel
archaeal superphylum called Asgard (the Olympus of Nordic
Gods) (Spang et al., 2015; Zaremba-Niedzwiedzka et al., 2017).
The genomes of Asgard archaea have been reconstructed in
silico from environmental DNA. Methods used for these recon-
structions assemble together pieces of DNA based on sequence
Arkarya signatures supposed to be constant all along the same genome
and markers abundance assumed to be similar for a given genome.
LUCA However, it is difficult to completely prevent the introduction in
(b) such genomes of DNA pieces from the genome of a different
organism with similar abundance and genomic signature present
in the same environment. Da Cunha and colleagues detected sev-
Figure 3 (a) Schematic representation of an eocyte tree of life. (b)
Schematic representation of a Woese tree of life.
eral insertions of putative eukaryotic origin in the elongation
factor EF2 from the three Asgard archaea analysed by Spang
and colleagues, possibly due to contamination (Da Cunha et al.,
(LBA) artefact produced by the attraction of the long eukaryotic 2017). They noticed that removing EF2 from the 36 universal
and bacterial branches. This was first suggested by Jim Lake who proteins data set used by these authors was sufficient to break
proposed that Eukarya emerged from a specific archaeal group, the affiliation between Eukarya and Asgards in the phylogenetic
which he called ‘eocytes’ (cells of dawn) because it was originally trees. This highlights the potentially dramatic, yet often underes-
only composed of thermophilic organisms, supposed by some timated impact that a single protein can have on a phylogenetic
authors to be at the ‘dawn of life’ (Lake, 1988). In the eocyte tree, analysis based on the concatenation of multiple proteins (Shen
Archaea are paraphyletic (not a valid taxonomic group) unless et al., 2017).
Eukarya themselves are considered to be Archaea (Figure 3a). In recent years, a myriad of new sequences from nanosized
The eocyte version of the universal tree of life has often been bacteria and archaea, whose genomes have been reconstructed
labelled the 2-D (two primary domains) scenario, as opposed to from environmental DNA, has also been added to some trees
the Woese version of the tree, frequently dubbed the 3-D scenario of life (Hug et al., 2016; Williams et al., 2017). It has been
(Figure 3b). However, this nomenclature is somehow confusing argued that these nanospecies represent most of the life diversity
since the Woese tree can also be considered as a 2-D tree with because they branch at the base of their respective domains and
Arkarya forming a primary domain besides Bacteria. can be divided in a very high number of phyla based on rRNA
Several theoretical arguments have been put forward to support sequence similarity (Hug et al., 2016). However, adding these
either the Woese or the eocyte tree of life (Kurland et al., 2006; nanospecies in the tree of life is challenging since they are most
Lane and Martin, 2010; Forterre, 2013). A priori, the distinction likely fast-evolving parasitic species. Indeed, besides their small
between the Woese and eocyte topologies can be made using size, they have reduced genomes lacking many critical metabolic
phylogenetic analyses based on either the rRNA or the universal pathways (Baker et al., 2010). It is well known that the presence
proteins conserved in the three domains of life. However, this of fast-evolving species in data sets can produce LBA artefact.
task is made difficult by the differences in length between the Notably, it has been shown that even the most recent sophisticated
various branches of the universal trees obtained. Indeed, the models of phylogenetic reconstruction cannot cope with LBA
choice between the two models depends on the existence or not when the branch of the out-group is too long (Gouy et al.,
of the short specific archaeal branch. If this branch indeed exists, 2015). This is precisely the case of the universal trees in which
it can be only detected from a priori rare mutational events, since the very long bacterial branch tends to attract long branches of
these events should have occurred in this short branch at positions fast-evolving archaea (Da Cunha et al., 2017). It is therefore hard
that otherwise remained identical all along the evolution in both to find a balance between representing life diversity in a tree while

4 eLS © 2018, John Wiley & Sons, Ltd. www.els.net

 Universal Tree of Life

getting heavily exposed to LBA artefacts or building robust trees organism cell or REO) becomes a virocell (Forterre, 2012). How-
yet lacking information regarding new phyla. ever, living or not, viruses cannot be practically placed in the
Notably, Forterre and colleagues switched from an eocyte TOL because there is no universal protein in the viral world (a
to a Woese tree when they removed archaeal species known protein that would be shared by all viruses and could be used
to be fast-evolving from their 35-proteins tree (without EF2) as a phylogenetic marker for their evolution). Recently, viruses
(Da Cunha et al., 2017). This suggests that the eocyte tree is have been tentatively classified into major lineages, based on
favoured by LBA artefacts induced by these species. Analysis of their capsid proteins (Bamford, 2003; Abrescia et al., 2012;
single-protein trees from this data set revealed two subsets of pro- Forterre et al., 2014). Indeed, capsid (or nucleocapsid) proteins
teins favouring either the Woese (11 proteins) or the eocyte tree are the major virion components, and production of virions can
(24 proteins). The number of ‘Woese proteins’ was lower, but they be considered to be the hallmark of viruses (Raoult and Forterre,
were on the average larger than the ‘eocyte’ proteins and provided 2008; Krupovic and Bamford, 2010). As a consequence, the
better resolution of the archaeal phylogeny (hence containing a origin of viruses can be assimilated to the origin of virions,
priori more signal). Notably, some ‘Woese proteins’ are absent as a specific strategy for genome dissemination. A dozen of
from universal proteins data sets used by other authors who recov- nonhomologous capsid proteins have been already identified in
ered the eocyte tree in their analyses (Raymann et al., 2015; Hug the viral world, defining as many viral lineages and indicating
et al., 2016) suggesting that further work is needed to determine that viruses are polyphyletic (e.g. different viral lineages origi-
the actual extent of conflicting signals among concatenations and nated independently) (Krupovic and Koonin, 2017). In opposition
figure out their weights. to ribosome-encoding organisms, it is therefore not possible to
Another problem with most recently published universal tree of define a last common ancestor for all viruses. See also: Virocell
life is that the number of species used is very different from one Concept, The
domain to another, with usually a much higher number of archaeal Members of the three domains of life are infected by a
species. Such imbalance in the taxon sampling can influence the plethora of viruses corresponding to three ensembles of viruses:
outcome of the analysis by impacting downstream experiments archaeoviruses, bacterioviruses (bacteriophages) and eukary-
(Nasir et al., 2016). To avoid this pitfall, universal trees should be oviruses (Raoult and Forterre, 2008). Notably, several viral
built using more balanced data sets between the three domains. lineages include viruses from these three ensembles, suggesting
This strategy has been used by Forterre and colleagues on a that their ancestor was already thriving at the time of LUCA
concatenation of the two RNA polymerase large subunits (using (Bamford, 2003; Abrescia et al., 2012). A convenient metaphor
RNA polymerase II as marker for Eukarya). The authors obtained to place viruses in the TOL is thus to consider that they are here,
a robust Woese tree with all methods of tree reconstruction used there and everywhere, as sung by the Beatles, from the trunk of
(Da Cunha et al., 2017). the TOL to the leaves.
At the moment, the topologies of the universal tree of life
are very contrasted and continuously moving, and the current
dominant view, the eocyte tree, has been challenged for the Glossary
several biases it may contain (Spang et al., 2018; Da Cunha
et al., 2018). However, it should be reminded that none of the Clade A group of organisms that originated from the same
studies published until now used all possible universal proteins ancestor and encompasses all descendants of this ancestor.
but different partly overlapping protein data sets. One should also Mono-/paraphyletic A monophyletic taxon corresponds to a set
keep in mind that the discovery of new deep branching lineages of sequences grouping their shared common ancestor and all
of Archaea, Bacteria or Eukarya could still change the tree of life the descending sequences. A paraphyletic taxon also
(TOL) topology in the future. encompasses a shared common ancestor yet lacks some
sequences or groups of sequences.
Ortho-/paralogous proteins Homologous proteins that
Where Are Viruses in the Universal diverged from an ancestral protein by speciation (orthologous)
or after a duplication event within a genome (paralogous).
Tree of Life? Distinguishing orthologous and paralogous proteins is hence
It has been disputed if viruses should be, or not, included in the critical for inferring their evolutionary history.
tree of life. For some authors, this should not be done because Phylogenomics Studies combining phylogenetic and
viruses are not alive (Moreira and López-García, 2009). For oth- comparative genomic analyses.
ers, viruses are alive and those encoding their own RNA poly- Thermoreduction The thermoreduction hypothesis posits that
merase can be included in the tree of life using this protein the prokaryotic phenotype may result from reductive evolution
as a universal marker (Boyer et al., 2010). Defining the bor- of both genomes and cells (notably coupling of transcription
ders of what is a cellular organism is a difficult problem as and translation, and faster macromolecular turnover) from
shown, for instance, by the case of mitochondria (Forterre, 2016). mesophilic ancestors in adaptation to high temperatures.
Indeed, one cannot determine at what moment a living intracel- Tree topology Pattern of connections between the different
lular bacterium became an organelle that is not alive. Viruses can clades of a phylogenetic tree. Trees can be rooted by adding
be viewed as living organisms if one focus on the intracellular an out-group of sequences, known or supposed to be distantly
stage of the viral infection, when the ribocell (ribosome-encoding related to the ones of interest.

eLS © 2018, John Wiley & Sons, Ltd. www.els.net 5

 Universal Tree of Life

References Forterre P (2012) Virocell concept, The. In: eLS. Chich-

ester: John Wiley & Sons, Ltd. http://www.els.net [DOI:
Abrescia NG, Bamford DH, Grimes JM and Stuart DI (2012) Struc- 10.1002/9780470015902.a0023264].
ture unifies the viral universe. Annual Review of Biochemistry 81: Forterre P (2013) The common ancestor of Archaea and Eukarya
795–822. was not an archaeon. Archaea 2013: Article ID 372396. DOI:
Adam PS, Borrel G and Gribaldo S (2018) Evolutionary history of 10.1155/2013/372396.
carbon monoxide dehydrogenase/acetyl-CoA synthase, one of the Forterre P, Krupovic M and Prangishvili D (2014) Cellular domains
oldest enzymatic complexes. Proceedings of the National Academy and viral lineages. Trends in Microbiology 22 (10): 554–558.
of Sciences of the United States of America 115 (6): E1166–E1173. Forterre P (2015) The universal tree of life: an update. Frontiers in
Baker BJ, Comolli LR, Dick GJ, et al. (2010) Enigmatic, ultrasmall, Microbiology 6: 717. DOI: 10.3389/fmicb.2015.00717.
uncultivated Archaea. Proceedings of the National Academy of Forterre P (2016) To be or not to be alive: how recent discoveries
Sciences of the United States of America 107 (19): 8806–8811. challenge the traditional definitions of viruses and life. Studies in
Bamford DH (2003) Do viruses form lineages across different History and Philosophy of Biological and Biomedical Sciences 59:
domains of life? Research in Microbiology 154 (4): 231–236. 100–108.
Bengtson S, Rasmussen B, Ivarsson M, et al. (2017) Fungus-like Garnier F, Debat H and Nadal M (2018) Type IA DNA
mycelial fossils in 2.4-billion-year-old vesicular basalt. Nature Topoisomerases: a universal core and multiple activi-
Ecology & Evolution 1 (6): 141. DOI: 10.1038/s41559-017-0141. ties. Methods in Molecular Biology 1703: 1–20. DOI:
Boussau B, Blanquart S, Necsulea A, Lartillot N and Gouy M (2008) 10.1007/978-1-4939-7459-7_1.
Parallel adaptations to high temperatures in the Archaean eon. Gogarten JP, Kibak H, Dittrich P, et al. (1989) Evolution of the
Nature 456 (7224): 942–945. vacuolar H+-ATPase: implications for the origin of eukaryotes.
Boyer M, Madoui M-A, Gimenez G, La Scola B and Raoult D Proceedings of the National Academy of Sciences of the United
(2010) Phylogenetic and phyletic studies of informational genes States of America 86 (17): 6661–6665.
in genomes highlight existence of a 4th domain of life including Gouy M, Baurain D and Philippe H (2015) Rooting the tree of life:
giant viruses. PLoS One 5 (12): e15530. the phylogenetic jury is still out. Philosophical Transactions of the
Brochier-Armanet C and Forterre P (2007) Widespread distribution Royal Society of London. Series B, Biological Sciences 370 (1678):
of archaeal reverse gyrase in thermophilic bacteria suggests a 20140329.
complex history of vertical inheritance and lateral gene transfers. Gribaldo S, Poole AM, Daubin V, Forterre P and Brochier-Armanet
Archaea 2 (2): 83–93. C (2010) The origin of eukaryotes and their relationship with
Brown JR and Doolittle WF (1997) Archaea and the the Archaea: are we at a phylogenomic impasse? Nature Reviews
prokaryote-to-eukaryote transition. Microbiology and Molecular Microbiology 8 (10): 743–752.
Biology Reviews 61 (4): 456–502. Grosjean H, Marck C and de Crécy-Lagard V (2007) The various
Chang J, Nie X, Chang HE, Han Z and Taylor J (2008) Transcription strategies of codon decoding in organisms of the three domains of
of hepatitis delta virus RNA by RNA polymerase II. Journal of life: evolutionary implications. Nucleic Acids Symposium Series
Virology 82 (3): 1118–1127. 51: 15–16.
Cox CJ, Foster PG, Hirt RP, Harris SR and Embley TM (2008) The Groussin M and Gouy M (2011) Adaptation to environmental tem-
archaebacterial origin of eukaryotes. Proceedings of the National perature is a major determinant of molecular evolutionary rates in
Academy of Sciences of the United States of America 105 (51): archaea. Molecular Biology and Evolution 28: 2661–2674.
20356–20361. Guy L and Ettema TJG (2011) The archaeal ‘TACK’ superphylum
Da Cunha V, Gaia M, Gadelle D, Nasir A and Forterre P (2017) and the origin of eukaryotes. Trends in Microbiology 19 (12):
Lokiarchaea are close relatives of Euryarchaeota, not bridging the 580–587.
gap between prokaryotes and eukaryotes. PLoS Genetics 13 (6): Hug LA, Baker BJ, Anantharaman K, et al. (2016) A new view of the
e1006810. tree of life. Nature Microbiology 11 (1): 16048.
Da Cunha V, Gaia M, Nasir A and Forterre P (2018) Asgard archaea Iwabe N, Kuma K, Hasegawa M, Osawa S and Miyata T (1989) Evo-
do not close the debate about the universal tree of life topology. lutionary relationship of archaebacteria, eubacteria, and eukaryotes
PLoS Genetics 14 (3): e1007215. inferred from phylogenetic trees of duplicated genes. Proceedings
El Albani A, Bengtson S, Canfield DE, et al. (2010) Large colonial of the National Academy of Sciences of the United States of Amer-
organisms with coordinated growth in oxygenated environments ica 86 (23): 9355–9359.
2.1 Gyr ago. Nature 466 (7302): 100–104. Koonin EV (2010) The incredible expanding ancestor of eukaryotes.
Forterre P (1995) Thermoreduction, a hypothesis for the origin of Cell 140 (5): 606–608.
prokaryotes. Comptes rendus de l’Academie des sciences. Serie III, Krupovic M and Bamford DH (2010) Order to the viral universe.
Sciences de la vie 318 (4): 415–422. Journal of Virology 84 (24): 12476–12479.
Forterre P and Philippe H (1999) Where is the root of the universal Krupovic M and Koonin EV (2017) Multiple origins of viral cap-
tree of life? BioEssays 21 (10): 871–879. sid proteins from cellular ancestors. Proceedings of the National
Forterre P (2006) Three RNA cells for ribosomal lineages and three Academy of Sciences of the United States of America 114 (12):
DNA viruses to replicate their genomes: a hypothesis for the E2401–E2410. DOI: 10.1073/pnas.1621061114.
origin of cellular domain. Proceedings of the National Academy Kurland CG, Collins LJ and Penny D (2006) Genomics and the irre-
of Sciences of the United States of America 103 (10): 3669–3674. ducible nature of eukaryote cells. Science 312 (5776): 1011–1014.
Forterre P and Gadelle D (2009) Phylogenomics of DNA topoiso- Lake JA (1988) Origin of the eukaryotic nucleus determined by
merases: their origin and putative roles in the emergence of modern rate-invariant analysis of rRNA sequences. Nature 331 (6152):
organisms. Nucleic Acids Research 37 (3): 679–692. 184–186.

6 eLS © 2018, John Wiley & Sons, Ltd. www.els.net

 Universal Tree of Life

Lane N and Martin W (2010) The energetics of genome complexity. Werner F and Grohmann D (2011) Evolution of multisubunit RNA
Nature 467 (7318): 929–934. polymerases in the three domains of life. Nature Reviews Microbi-
Lecompte O, Ripp R, Thierry JC, Moras D and Poch O (2002) ology 9 (2): 85–98.
Comparative analysis of ribosomal proteins in complete genomes: Williams TA, Szöllösi GJ, Spang A, et al. (2017) Integrative
an example of reductive evolution at the domain scale. Nucleic modeling of gene and genome evolution roots the archaeal tree
Acids Research 30 (24): 5382–5390. of life. Proceedings of the National Academy of Sciences of
Moreira D and López-García P (2009) Ten reasons to exclude viruses the United States of America 114 (23): E4602–E4611. DOI:
from the tree of life. Nature Reviews Microbiology 7 (4): 306–311. 10.1073/pnas.1618463114.
Mulkidjanian AY, Makarova KS, Galperin MY and Koonin EV Woese CR and Fox GE (1977) Phylogenetic structure of the prokary-
(2007) Inventing the dynamo machine: the evolution of the otic domain: the primary kingdoms. Proceedings of the National
F-type and V-type ATPases. Nature Reviews Microbiology 5 (11): Academy of Sciences of the United States of America 74 (11):
892–899. 5088–5090.
Nasir A, Kim KM, Da Cunha V and Caetano-Anollés G (2016) Argu- Woese CR, Kandler O and Wheelis ML (1990) Towards a natural
ments reinforcing the three-domain view of diversified cellular life. system of organisms: proposal for the domains Archaea, Bacteria,
Archaea 2016: Article ID 1851865. DOI: 10.1155/2016/1851865. and Eucarya. Proceedings of the National Academy of Sciences of
Pittis AA and Gabaldón T (2016) Late acquisition of mitochondria the United States of America 87 (12): 4576–4579.
by a host with chimaeric prokaryotic ancestry. Nature 531 (7592): Zaremba-Niedzwiedzka K, Caceres EF, Saw JH, et al. (2017) Asgard
101–104. archaea illuminate the origin of eukaryotic cellular complexity.
Raoult D and Forterre P (2008) Redefining viruses: lessons from Nature 541 (7637): 353–358.
Mimivirus. Nature Reviews Microbiology 6 (4): 315–319.
Raymann K, Brochier-Armanet C and Gribaldo S (2015) The
two-domain tree of life is linked to a new root for the Archaea. Further Reading
Proceedings of the National Academy of Sciences of the United
States of America 112 (21): 6670–6675. Forterre P (2016) Microbes from Hell. Chicago, IL: University of
Shen X, Hittinger CT and Rokas A (2017) Studies can be driven by Chicago Press.
a handful of genes. Nature Ecology & Evolution 1 (5): 126. DOI: Sapp J and Fox GE (2013) The singular quest for a universal tree
10.1038/s41559-017-0126. of life. Microbiology and Molecular Biology Reviews 77 (4):
Spang A, Saw JH, Jørgensen S, et al. (2015) Complex archaea that 541–550.
bridge the gap between prokaryotes and eukaryotes. Nature 521 The Evomobil Project (2017) Archaea of the Putative Asgard
(7551): 173–179. Superphylum do not Close the Debate about the Universal Tree
Spang A, Eme L, Saw J, et al. (2018) Asgard archaea are the of Life Topology. https://www.the-evomobil-project.com/single-
closest prokaryotic relatives of eukaryotes. PLoS Genetics 14 (3): post/2017/08/22/Archaea-of-the-putative-Asgard-superphylum-
e1007080. do-not-close-the-debate-about-the-universal-tree-of-life-topology

eLS © 2018, John Wiley & Sons, Ltd. www.els.net 7

View publication stats