The Neomuran Revolution and Phagotrophic Origin of ...system, cytoskeleton, and associated motors....

The Neomuran Revolution and PhagotrophicOrigin of Eukaryotes and Cilia in the Lightof Intracellular Coevolution and a RevisedTree of Life

Thomas Cavalier-Smith

Department of Zoology, University of Oxford, Oxford OX1 3PS, United Kingdom

Correspondence: [email protected]

Three kinds of cells exist with increasingly complex membrane-protein targeting: Uni-bacteria (Archaebacteria, Posibacteria) with one cytoplasmic membrane (CM); Negibacte-ria with a two-membrane envelope (inner CM; outer membrane [OM]); eukaryotes with aplasma membrane and topologically distinct endomembranes and peroxisomes. I combineevidence from multigene trees, palaeontology, and cell biology to show that eukaryotes andarchaebacteria are sisters, forming the clade neomura that evolved �1.2 Gy ago from aposibacterium, whose DNA segregation and cell division were destabilized by mureinwall loss and rescued by the evolving novel neomuran endoskeleton, histones, cytokinesis,and glycoproteins. Phagotrophy then induced coevolving serial major changes making eu-karyote cells, culminating in two dissimilar cilia via a novel gliding–fishing–swimmingscenario. I transfer Chloroflexi to Posibacteria, root the universal tree between them andHeliobacteria, and argue that Negibacteria are a clade whose OM, evolving in a greenposibacterium, was never lost.

THE FIVE KINDS OF CELLS

The eukaryotic cell originated by the mostcomplex set of evolutionary changes since

life began: eukaryogenesis. Their complexityand mechanistic difficulty explain why eukary-otes evolved 2 billion years or more after pro-karyotes (Cavalier-Smith 2006a). To under-stand these changes, we must consider the cellbiology of all five major kinds of cells (Fig. 1);determine their correct phylogenetic relation-ships; and explain the causes, steps, and detailed

mechanisms of the radical transitions betweenthem. Figure 1 highlights three fundamentallydifferent kinds of prokaryote differing great-ly in membrane topology and membrane andwall chemistry. In all cells, the major membranelipids are glycerophospholipids having two hy-drophobic hydrocarbon tails attached to a hy-drophilic phosphorylated glycerol head, butglycerol-phosphate stereochemistry differs inarchaebacteria (sn-glycerol-1-phosphate) fromthat in all other cells (sn-glycerol-3-phosphate).Negibacteria and posibacteria (collectively called

Editors: Patrick J. Keeling and Eugene V. Koonin

Additional Perspectives on The Origin and Evolution of Eukaryotes available at www.cshperspectives.org.

Copyright # 2014 Cold Spring Harbor Laboratory Press; all rights reserved; doi: 10.1101/cshperspect.a016006

Cite this article as Cold Spring Harb Perspect Biol 2014;6:a016006

1

on August 31, 2020 - Published by Cold Spring Harbor Laboratory Press http://cshperspectives.cshlp.org/Downloaded from

http://cshperspectives.cshlp.org/

neomura

BACTERIA

Negibacteria

acyl ester phospholipids; SRP protein-targeting to CM ; cell division by FtsZ and divisomeDnaA replication initiator; Xer DNA replication terminator; anoxygenic green bacterial photosynthesis;

Smc condensins; parABS origin segregation; DNA translocases; conjugation via type IV secretion;

(=prokaryotes)

microtubules

flagellum

OM

CM

Archaebacteria

hyperthermophily:isoprenoid ether phospholipds;reverse gyrase

Euglenozoaneokaryotes

EUKARYOTA

murein peptidoglycan

CM periplasmic space

eubacteria

murein peptidoglycan

TFs IIA, F; Smc 5/6, CENPA,

reticulinmRNA SL trans-splicing; loss of

transcriptional control

ventral feeding groove, split right

root

neomuran revolution N-linked glycoprotein

histones, Cdc6, MCM, PCNA; ESCRT III GTPase; SRP arrest; losses of Xer termination, DnaA, G, DNA gyrase, SecA

mureinloss

Posibacteria

1.2 Gy ago

3.5 Gy ago

phagotrophy

eukaryogenesis: endomembranes, peroxisomes, endoskeleton, nucleus, centrioles, cilia, cyclins, mitosis, sex

nucleus

2 cilia, centrioles

OM Omp85

mitochondrion

mureinloss

1-haem cytochrome c

type IV secretion lost

loss of PI, ACP, CL, sterols, and many genes

porins

cytopharynx

OM

CM

Figure 1. Relationships between the five major cell types, showing key evolutionary innovations in the transitionsmaking them. Rigid murein cell walls originated before the cenancestor of all life using both D- and L-aminoacids in the first cell, a posibacterium with acyl ester glycerophospholipids that divided using FtsZ, possibly aphotoheterotroph similar to Heliobacterium. Negibacteria evolved by acquiring an outer membrane (OM) withcomplex targeting of porins and other b-barrel proteins inserted by Omp85-dependent machinery never lost inthe history of life, being retained when eukaryotes enslaved phagocytosed negibacteria to make mitochondriaand subsequently chloroplasts (even kept in secondarily anaerobic DNA-free hydrogenosomes and mitosomesthat evolved by drastically modifying aerobic mitochondria). The neomuran revolution was arguably a stabi-lizing response to traumatic loss of murein. Histones H3/4 ensured passive negative DNA supercoiling (makingnucleosomes) to replace eubacterial ATP-driven supercoiling by DNA gyrase; this stabler DNA coiling forceddrastic coevolutionary changes in RNA polymerase and especially DNA replication machinery: repair polymer-ase d replaced DNA polymerase III, the b-clamp became PCNA, the replication fork helicase Mcm replacedDnaB, the unrelated Pol primase replaced DnaG primase, and Cdc6 replaced the replication initiator DnaA;Cdc6 possibly evolved from a gene duplicate of the eubacterial clamp loader DnaX, itself undergoing minormodification to neomuran RFC. (Legend continues on following page.)

T. Cavalier-Smith

2 Cite this article as Cold Spring Harb Perspect Biol 2014;6:a016006



“eubacteria”) are mutually closer in cell enve-lope chemistry and informational machinerythan either is to archaebacteria, whose basicinformational machinery is eukaryote-like de-spite their cells being fully prokaryotic in struc-ture and DNA segregation machinery. Eubac-teria, unlike eukaryotes and archaebacteria,generally have cell walls of the peptidoglycanmurein that forms a covalently cross-linkedbag (sacculus) completely surrounding the cy-toplasmic membrane (CM). Murein hydrolaseenzymes must repeatedly cleave, and other en-zymes reseal, murein covalent bonds so that eu-bacteria can grow without bursting under highinternal osmotic pressure (Egan and Vollmer2013).

Some derived methanogenic archaebacte-ria have covalently cross-linked walls of pseudo-murein, a different peptidoglycan with similarcleavage-resealing growth. However, other arch-aebacteria and all eukaryotes have cell surfaces of

non-cross-linked globular glycoproteins con-taining hydrophilic oligosaccharides covalentlylinked to asparagine (N) residues. N-linked gly-coproteins are made cotranslationally, oligosac-charides being attached during trans-membraneprotein secretion by membrane-associated ri-bosomes. I argued that this shared characterevolved in the last common ancestor (cenances-tor) of eukaryotes and archaebacteria, whichjointly constitute the putative clade neomura.The neomuran theory of eukaryote origins (Ca-valier-Smith 1987c; revised and updated: Ca-valier-Smith 2002c, 2009, 2010c) has a phylo-genetic part and a causal mechanistic part, asshould any scientific explanation of megaevo-lutionary events. Many inadequate “theories” ofeukaryote origin focus exclusively on phylogenyand have no explanatory part or only a cursory,unconvincing one. Phylogenetically, the neo-muran theory asserts that (1) eubacteria aresubstantially older than and ancestors of neo-

Figure 1. (Continued) Novel TATA-box-binding transcription factors (TBP and others) (Ouhammouch et al.2009) replaced the eubacterial transcription regulator CrtA. Murein loss freed MreB filaments that maintaineubacterial rod shape (or related ParM filaments that segregate some plasmids) to become the actin endoskel-eton, conferring osmotic stability; new ESCRT-III filaments helped membrane division, allowing loss of FtsZ ineukaryotes and some archaebacteria. Novel cotranslationally made N-linked glycoprotein enabled archaebac-teria to make rigid S-layer-like walls and eukaryotes a flexible cell surface coat, allowing phagotrophy andingestion of prey cells to evolve, triggering a cascade of eukaryogenic changes associated with coated vesicleorigins. These mediated endomembrane differentiation, internal digestion, targeted vesicle fusion, and nuclearenvelope evolution to protect chromatin internalized by phagocytosis (see Fig. 2); a-tubulin, b-tubulin, and g-tubulin evolved from posibacterial plasmid-segregating TubZ GTPase, enabling DNA segregation by mitosis,drastically changing chromosome organization; cohesins enabling mitosis and eukaryotic cell-cycle controlsevolved from duplicated Smc condensins. Archaebacteria replaced acyl ester lipids by heat-stable isoprenoidtetraethers to become the first extremophiles, but lost so many lipids and proteins that they could never haveevolved directly into eukaryotes, as did the transient neomuran ancestor, retaining far more eubacterial char-acters. Archaebacteria kept fatty acid (FA) synthesis (Lombard et al. 2012a) but lost acyl-carrier protein (ACP),which enables rapid bulk FA synthesis in eubacteria and eukaryotes, no longer needed for the trace FA amountsthat sufficed after archaebacteria lost acyl esters, including phosphatidylinositol (PI) and cardiolipin (CL).Neokaryotes retained bacterial transcription regulation but evolved new transcription factors (TFs). Soon aftermitochondrially donated group II self-splicing introns became spliceosomal introns in the cenancestral eukary-ote (Cavalier-Smith 1991c), Euglenozoa evolved trans-splicing of spliced-leader (SL) miniexons for all mRNAs(Cavalier-Smith 1993) and lost transcriptional control of gene expression. Neokaryotes alone replaced centro-meric histone H3 by CENP-A and evolved Smc5/6 for DNA repair. Archaebacterial flagella are not homologousto, and evolved independently of, eubacterial flagella, which must have evolved in Negibacteria, probably in earlyGracilicutes (Cavalier-Smith 2006c); so if the tree is correctly rooted within Posibacteria, they were presumablyacquired by Posibacteria by lateral gene transfer (LGT) subsequently, but before actinobacteria diverged fromTeichobacteria (see Fig. 3). Ancestral green bacteria lacked flagella but could probably glide and thus makestromatolites, yielding the oldest fossil evidence for eubacteria. Absence of photosynthetic carbon fixation inarchaebacteria means that, unlike the much older eubacteria, they could never have fueled an extensive globalecosystem alone.

The Neomuran Revolution and Origin of Eukaryotes

Cite this article as Cold Spring Harb Perspect Biol 2014;6:a016006 3



mura; and (2) archaebacteria are sisters of eu-karyotes, not their ancestors. This explainswhy eukaryotes are a blend of eubacteria-like(e.g., membrane chemistry) and archaebacte-ria-like characters (e.g., N-linked glycoproteins,information-processing machinery) plus en-tirely novel features (cell structure, mitosis).Eubacteria-like characters, ancestral for all life,were inherited vertically by eukaryotes, whilearchaebacteria secondarily evolved unique lip-ids; shared archaebacteria-like characters origi-nated only in the neomuran cenancestor (lastcommon ancestor); and uniquely eukaryoticcharacters evolved immediately after eukaryotesdiverged from archaebacteria. No other theoryexplains that as simply. Assuming that archae-bacteria were directly ancestral to eukaryotes(Van Valen and Maiorana 1980) requires non-parsimonious assumptions that eubacteria-likecharacters were regained during eukaryogenesis.

PHAGOTROPHY, CYTOSKELETON,MOTORS, AND THE COEVEOLUTIONARYTHEORY OF EUKARYOGENESIS

Three key innovations made eukaryotes: (1) aninternal cytoskeleton of formin-associated actinfilaments cross-linked by actin-related proteins(Arp 2/3) plus microtubules (a,b-tubulin) nu-cleated by g-tubulin; (2) cytoskeleton-associat-ed molecular motor ATPases: myosin for actin;kinesin and subsequently dynein for micro-tubules; (3) the eukaryote-specific endomem-brane system (endoplasmic reticulum [ER],Golgi complex, and lysosomes—the bags of di-gestive enzymes discovered by De Duve thatmediate intracellular digestion in eukaryotesbut no bacteria). Only subsequently could thenucleus, centrioles, mitosis, and eukaryotic ge-netic system evolve, using and stimulated bythis novel machinery. The decisive evolution-ary mediator of eukaryogenesis was the originof phagocytosis and intracellular prey digestionfor the first time in history, which also enableduptake and transformation of an a-proteobac-terium into mitochondria, greatly improvingaerobic utilization of intracellular digestion pro-ducts simultaneously with the origin of the nu-cleus and cilia (Cavalier-Smith 2002c).

These ideas arose before archaebacteria wererecognized (Woese and Fox 1977). De Duveand Wattiaux (1966) suggested that rough ERevolved by budding from the CM of a wall-lessbacterium when evolving an ability to ingestother cells by a primitive version of phagocy-tosis, with subsequent differentiation of inter-nalized vesicles producing the Golgi apparatusand lysosomes for more efficient internal diges-tion. Stanier (1970) noted that such phagotro-phy, absent from all prokaryotes, must have im-posed novel selective forces favoring larger cellsize and increased internal complexity, provid-ing sufficient explanation of the greater inter-nal complexity of eukaryotic cells, especially theorigin of the internal cytoskeleton. I explainedhow phagotrophy, by internalizing CM withattached prokaryotic chromosomes, must havedisrupted bacterial cell cycles and DNA segre-gation even more dramatically than wall-loss,so phagotrophy imposed novel selective forcescausing evolution of mitosis, meiosis, and novelgenetics and chromosome organization of eu-karyotes (Cavalier-Smith 1975) and much larg-er genomes (Cavalier-Smith 1978b). John andWhatley (1975, 1977) argued that mitochondriaoriginated from endosymbiotic purple–non-sulfur bacteria (now called a-proteobacteria).

I argued that these novel eukaryotic geneticfeatures all evolved as coevolutionary responsesto disruption of the prokaryotic cell-surface-lo-cated genetic system by the new endomembranesystem, cytoskeleton, and associated motors.Moreover, the nucleus evolved by incompleteER fusion around chromatin to protect it fromshearing damage by cytoplasmic motors, andperoxisomes stemmed from subdivision andspecialization of endomembranes. As murein-wall-loss necessitated osmotically and cell-cy-cle-stabilizing innovations, I proposed that mi-crotubules evolved to stabilize DNA segregationby premitotic mechanisms, and actin was re-quired for phagotrophy (not then known) andcytokinesis. I argued that actin was the prima-ry and microtubules the next major molecu-lar innovation enabling eukaryotes to evolve(Cavalier-Smith 1975). We now know that ac-tin originated slightly before eukaryotes duringthe neomuran revolution, from MreB filaments

T. Cavalier-Smith




that help maintain shape for rod-like eubacteria(Lowe and Amos 2009; Wickstead and Gull2011). However, changing MreB to actin wastrivial, MreB sometimes being called “bacterialactin.” More radically, gene triplication madea,b,g-tubulins subsequently, in the eukaryotecenancestor, probably from posibacterial plas-mid-segregating GTPase TubZ, rather than relat-ed FtsZ, to make spindle microtubules for mi-totic DNA segregation (Cavalier-Smith 2010c).

I now suggest that the most fundamentaleukaryogenic molecular innovation was theorigin of formin rings (type A2) (Chalkia etal. 2008) to promote rapid actin polymeriza-tion (by encircling the filaments’ fast-growingbarbed ends, drawing in monomers from pro-filin complexes) to extend pseudopodia aroundprey cells during phagotrophy, plus profilinto bind G-actin as a soluble store allowing sud-den F-actin extension through formin-binding.Secondarily, actin duplicated, yielding Arp2/3for branching by linking pointed ends to otheractin filaments; with actin-capping proteins forbarbed ends, branching made an osmoticallystabilizing three-dimensional (3D)–gel mesh-work. Figure 2 updates the intracellular coevo-lutionary consequences of phagotrophy. Likesingle-headed myosin I, actin proved essentialfor phagocytosis, but its involvement in cytoki-nesis in the contractile ring of podiate eukary-otes probably resulted from much later evolu-tion of two-headed myosin II in early podiates(Fig. 3). Originally, eukaryote cytokinesis usedthe membrane-bending dynamin GTPase thatevolved in the ancestral eubacterium (lost bycrenarchaeotes) plus ESCRT-III GTPase fila-ments for membrane scission, which arguablyevolved in the ancestral neomuran (Cavalier-Smith 2010c); crenarchaeotes lost ESCRT (andmost lost FtsZ).

This coevolutionary phagotrophy theory ofeukaryogenesis, enunciated before evidence forsymbiogenetic origins of chloroplasts and mi-tochondria became compelling (Gray and Doo-little 1982; Gray 1992; Cavalier-Smith 2013b),initially wrongly assumed cyanobacterial ances-try of the whole eukaryotic cell (Cavalier-Smith1975). I did not then appreciate the doublenessof the negibacterial envelope and likely mecha-

nistic impossibility of evolving unimembra-nous eukaryotes by wall-loss from any negibac-terium. Although that defective phylogeny isconsigned to history, the central logic of phago-trophy being the driving force behind the originof endomembranes and cytoskeleton and of theradically transformative consequences of bothnovelties for eukaryotic cell cycles and geneticswas almost certainly correct, providing the onlylogically coherent explanation for eukaryogen-esis that is both mechanistically and selectivelyconvincing.

Initially I argued that centrioles and ciliaevolved from microtubules substantially afterthe first eukaryotes (Cavalier-Smith 1975,1978a, 1981, 1982), but that was wrong. Almostcertainly the first ciliated eukaryote had onecilium only (Cavalier-Smith 1975, 1978a), butthe assumption that some primitively uniciliateeukaryotes still survive (Cavalier-Smith 1987c,2002c) was wrong. Probably all uniciliates aresecondarily simplified, and the eukaryote cen-ancestor had two centrioles bearing dissimi-lar cilia, which take two cell cycles to developto maturity, the anterior cilium being youngerand forming in the first cell cycle, the older pos-terior one being modified in structure, position,and roots in the second cell cycle (ciliary trans-formation: Brugerolle 1992; Cavalier-Smith andKarpov 2012; Cavalier-Smith 2013a). Beforeeach cell division, two new centrioles assem-ble beside old ones at the beginning of S phasewhen DNA is replicated, this being controlledby cyclin proteins unique to eukaryotes. Cyclinsshare a helical domain with the neomurantranscription factor TFIIb and perhaps evolvedfrom it.

Proteolysis of proteins connecting the twoolder cilium-bearing centrioles allows their sep-aration to opposite spindle poles during mitoticprophase. Possibly parent centrioles are held to-gether by the same loop-like cohesin proteinsas sister chromatids (Nasmyth 2011; Eichingeret al. 2013), both assembling at S-phase onset,and both cleaved by the enzyme separase, em-phasizing the deep coevolution of chromosomeand centriole cycles. However, in Drosophila atleast, cohesin cleavage is inessential for centrioledisengagement, which requires a drop in cyclin-





Figure 2. Intracellular coevolution during phagotrophy-driven eukaryogenesis. (A) Eubacteria segregate DNAby actively moving replicon origins (O) by ParABS machinery, DNA condensation by Smc condensin rings, andmoving termini (T) by the DNA translocase FtsK anchored at the mid-cell nascent division site, marked by theGTPase FtsZ ring for membrane scission by the divisome after XerCD recombinase resolves daughter DNAsinto covalently separate molecules. Neomuran loss of murein disrupted orderly linear arrangement of chromo-some origins and termini on a rigid wall, causing FtsK loss and allowing replicon numbers per chromosome toincrease, and ESCRT-III GTPase filaments replaced the eubacterial divisome; simultaneously, the posibacterialparacrystalline S layer became novel N-linked glycoproteins. (B) In eukaryotes only, these glycoproteins (yellow)became flexible and specialized for binding prey, initially digested by enzymes secreted externally by membrane-attached ribosomes. MreB (or its plasmid segregation ParM relative) (Yutin et al. 2009) evolved into actin,yielding linear filaments stimulated by formins for extending cytoplasm partly around prey, thereby increasingdigestion product absorption, and duplications yielded Arp2/3, generating an osmotically stabilizing branchingendoskeleton (blue). (Legend continues on following page.)

T. Cavalier-Smith




dependent kinase activity, and another centro-somal separase target may be the crucial linkerof new and old centrioles (Oliveira and Na-smyth 2013); in Caenorhabditis, cohesin seemsinvolved only in certain developmental stages(Cabral et al. 2013). In mammals, the giantcoiled-coil kendrin is a separase target mediat-ing centriole disengagement (Matsuo et al.2012), but its sequence is highly conserved onlyin vertebrates, which have a clearly related family

of A-kinase anchor proteins important for manyaspects of cell structure; but as neither is reliablytraceable beyond vertebrates, it is possible thatwhen early animals expanded centrosome sizecleavage of more than one, possibly novel cen-trosome protein became necessary, even if cohe-sin may have been the ancestral target (still aconjecture). Cohesin rings comprise Y-shapedSmcs and kleisin cross-linkers that evolved (afterpre-eukaryotes diverged from archaebacteria)

Figure 2. (Continued) (C) Evolution of a surface membrane protein channel (Derlin) enabled partially digestedproteins to be pulled across the membrane and fully digested on its cytosolic face by cylindrical proteasomes.Digestion products of prey completely internalized into a phagosome (center) was most efficiently absorbed;phagosome-associated V-SNAREs and CM-associated T-SNAREs evolved to refuse phagosome membraneswith the surface. (D) Accidental phagocytic internalization of membrane-attached DNA was made permanentby evolving COP-coated vesicles that returned membrane only to the cell surface; after exclusion of ribosomesand DNA from COP vesicles, continued phagocytosis removed all from the plasma membrane, and the inter-nalized ribosome/DNA-associated membrane became protoER. Membrane fragmentations generated separatecompartments specializing in b-oxidation of fatty acids (peroxisomes) and cytochrome P450 oxidation ofaromatics and protein secretion (ER). Bacterial Sec61/SRP for extruding unfolded proteins was retained byER and TAT machinery for unfolded proteins modified for peroxisome biogenesis (for more details, seeCavalier-Smith 2009). FtsZ-related posibacterial plasmid-segregated TubZ GTPase evolved by gene duplicationinto eukaryotic mitotic segregator (a-,b-tubulin microtubules, nucleated at minus ends by g-tubulin-contain-ing centrosomes), microtubule rigidity mechanistically replacing peptidoglycan rigidity. Initially, microtubulepolymerization forced sister centrosomes and associated DNA apart. COPs were also used for pinocytosis, andpreexisting dynamin and ESCRT-III coopted for membrane scission generating protoendosomes (pE). Single-head myosin I and kinesin diverged from a common posibacterial ATPase ancestor to form motors for inter-nalizing phagosomes along actin filaments or minus-to-plus movement of vesicles on microtubules for exocy-tosis, respectively. (E) Kinesin was coopted to push apart antiparallel microtubules from sister centrosomes,improving segregation, cytokinesis being by preexisting neomuran ESCRT-III GTPase ring orthogonal to thespindle. (F) Simultaneously, coordinate gene duplications of COP proteins and SNAREs multiplied the numberof topologically and chemically distinct compartments developmentally interlinked by vesicle transport: copIIfor ER to Golgi, CopI for recycling membrane from Golgi, and clathrin for making protoendosomes (pE) andlysosomes (L) (Faini et al. 2013). (E and F) Aspects of the same stage. (G) A protonuclear envelope formed bypartial ER cisternal fusion onto the surface of chromatin, centrosomes duplicating into centrin-connecteddistinct microtubule-nucleating centers (MNC) for cell-surface cortical microtubules and nuclear-envelope-associated spindle poles. COPII coats were retained by the protonuclear envelope, evolving into nuclear porecomplexes (NPC: their origin and that of importin- and RanGTP-gradient-based nucleocytoplasmic proteinimport using nuclear localization signals [NLS] were fully explained in Cavalier-Smith 2010c). Novel rapid DNAsegregation by mitosis in anaphase replaced two-stage rigid-wall-associated prokaryotic segregation via newspindle kinesins, causing chromosome linearization and telomeres. Minus-end-directed dynein ATPase motorsevolved to move vesicles along microtubules to centrosomes, fusing to form centrosome-attached, stacked Golgicisternae (G) specializing in subsequent glycosylation stages. (H ) Transition fibers attached a ring of microtu-bules to the cell surface forming a protocilium, a novel heterotrimeric kinesin-2 evolving to move them relativeto protociliary membrane glycoproteins adhering to the substratum, initiating protociliary gliding to carry cellsto fresh prey; recruitment of septins to the protociliary base and evolution of a transition zone plate (TP) andcollar, plus anterograde (IFTB) and retrograde (IFTA) transport particles from COPI coats, and modification ofnuclear protein-targeting machinery for ciliary protein import established a discrete protociliary compartment.Figure 5 shows how this could have evolved into 9 þ 2 cilia in the cenancestral eukaryote. Division of laboramong coevolving peroxisomes (P, ancestrally attached to and segregated with the nuclear envelope in closedmitosis), endomembranes, and mitochondria (M: derived from phagocytosed, undigested a-proteobacteria)optimized aerobic metabolic utilization of phagotrophy digestion products.





Figure 3. Expanded tree of life showing major subdivisions of ancestral eubacteria and derived neomura. Keyinnovations in cell evolution primarily involve membranes and cell skeleton. Unibacteria with single membranesevolved three different CM chemistries: Endobacteria (thick-walled Teichobacteria plus derived wall-free my-coplasmas and spiroplasmas [i.e., Mollicutes]) and Chloroflexi, the two most ancient posibacterial subphyla,retained ancestral hopanoids as membrane rigidifiers. Actinobacteria evolved sterols and phosphaphatidylino-sitol. Archaebacteria, the youngest bacterial phylum, sister to eukaryotes, replaced acyl ester phospholipidbilayers by a stabler isoprenyl ether monolayer to become the first hyperthermophiles. Numerous proteinswere lost during their origin and early diversification. (Legend continues on following page.)

T. Cavalier-Smith




by gene duplications from homologous con-densin rings that universally mediate higher-or-der DNA folding (bacterial nucleoid-folding;eukaryote chromatin-folding). Anaphase chro-mosome separation is initiated later than cen-triolar disengagement by ubiquitin-controlledproteasome-mediated proteolysis of separaseinhibitors and hundreds of other proteins, allmarked by anaphase-promoting complex (APC)

ubiquitin ligase (Oliveira et al. 2010) and timedby changes in cyclin phosphorylation (a unique-ly eukaryotic cell-cycle control principle) (Na-smyth 1995).

These universally shared cohesin-based fea-tures of higher-order DNA folding and segrega-tion strongly argue against past wild speculationthat neomura evolved DNA replication inde-pendently of eubacteria, quite apart from the

Figure 3. (Continued) The neomuran common ancestor probably arose from a stem actinobacterium byreplacing covalently cross-linked cell-wall peptidoglycan by more flexible glycoproteins via an antibiotic-resis-tant but traumatically wall-less, DNA-segregationally defective intermediate that recovered through revolution-ary change in ribosomes/SRPs and evolving histone-stabilized chromatin, causing radical changes to DNA-handling enzymes: the neomuran revolution. Eukaryotes arose by exploiting the new flexible glycoproteinsurface to trap and phagocytose bacteria; phagotrophy internalized their digestive system (endomembranes)and genetic system, stabilized by additional histones, novel endomembrane attachments, and nuclear-porecomplexes and a novel 3D internal cytoskeleton and novel motors, used for mitosis/cell division and vesicleand ciliary motility, and internalized ana-proteobacterium for enslavement as a mitochondrion (synergisticallyimproving food-energy conversion). Eukaryotes diverged early into Euglenozoa (which retained ancestralciliary gliding on surfaces and divergent DNA replication initiation and mitochondrial protein import machin-ery, but evolved specialized feeding apparatus for a surface-associated lifestyle) and Excavata, which lost glidingand evolved planktonic feeding by a posterior ciliary groove. Excavata comprise nonamoeboid Loukozoa, oftenwith posterior cilium vanes, plus vane-free Percolozoa ancestrally with alternating amoeboid and flagellatestages (sometimes differentially lost). From a vaned Malawimonas-like loukozoan that simplified cytochromec biogenesis by evolving unimolecular heme lyase stemmed two derived supergroups of contrasting morphologyand lifestyle: (1) Corticates specializing on photic zone planktonic living by evolving cortical alveoli andenslaving cyanobacteria to form chloroplasts (first Plantae [almost all lost phagotrophy] then a secondaryenslavement of a red alga to generate photophagotrophic Chromista) (Cavalier-Smith 2013b); many corticatesevolved a fourth microtubular ciliary root (R4) absent from podiate and eozoan supergroups. (2) Exclusivelyheterotrophic podiates, by origin of ventral pseudopodia, and dorsal pellicle associated with reevolved posteriorciliary gliding, with subsequent loss of posterior cilium and its roots to create opisthokonts (names in red) withradically simplified cytoskeleton. Vanes were lost by all neozoa but Colponema, which retained the loukozoanfeeding method. Ancestrally, chromists had four kinds of ribosome, four genomes, and novel membranetopology with nuclear-coded proteins imported across the periplastid membrane by novel mechanisms derivedby duplications from the ERAD machinery that evolved to export unfolded proteins for proteasome digestion inthe first eukaryote (Fig. 2C,D,F); many evolved tubular ciliary hairs that modified feeding in heterokonts(Cavalier-Smith and Scoble 2013). All except cryptomonads lost the nucleomorph and periplastid ribosomes.Long-tailed myosin II that forms antiparallel aggregates mediating contraction of podiate pseudopodia andcytokinetic contractile actomyosin rings probably evolved near the ancestral podiate, assuming that the perco-lozoan Naegleria got myosin II by LGT from podiates. Very different reticulose/filose pseudopodia evolved inthe chromist infrakingdom Rhizaria. Amoebozoa and opisthokonts, formerly grouped as “unikonts,” evolvedfrom biciliate Sulcozoa by independently losing gliding (eukaryote cytoskeletal diversification and its coevolu-tion with changing feeding modes are detailed elsewhere: Cavalier-Smith and Chao 2012; Cavalier-Smith andKarpov 2012; Cavalier-Smith 2013a). Ancestrally, photosynthetic Negibacteria retained hopanoids and diver-sified into eight phyla differing in IM photosynthetic machinery, OM chemistry, and flagellar organization.Eurybacteria comprise Negativicutes (Marchandin et al. 2010) (formerly Selenobacteria: Cavalier-Smith 1992,2002b, 2006c), Fusobacteria, and Thermotogales. Filarchaeota comprise crenarchaeotes, thaumarchaeotes, andkorarchaeotes. Although the origin of the first (stem posibacterial) cell was probably as early as 3.5 Gy ago, themajor eubacterial radiation producing their modern (crown) phyla likely occurred subsequently, possibly�2.7–2.5 Gy ago; its essential simultaneity accounts for almost nonexistent resolution at the base of theeubacterial tree (Pace 2009), which coupled with a quantum-evolution-stretched neomuran stem in manysequence trees makes it very hard to place neomura anywhere robustly within the eubacterial tree.





fact that the genetic complexity of any eubac-terial/neomuran intermediate with probablyat least a thousand genes could not have beensupported by a purely RNA-based genome with-out the conservatism and evolutionary stasisallowed by efficient DNA repair. The key im-portance of cell-cycle continuity during eukar-yogenesis has too often been ignored; manyimaginary intermediates could not have repro-duced.

The complex origin of phagotrophy waspossibly simplified before full phagocytosis byan intermediate stage importing extracellularprey proteins through Derlin surface-membranepores, coupled with digestion internally by sur-face-attached proteasomes before lysosomesevolved (Fig. 2C) (Cavalier-Smith 2009). Pro-teasomes, originating before actinobacteriaand neomura diverged (Cavalier-Smith 2006c),were radically complicated by numerous geneduplications in the eukaryote cenancestor whenfirst participating in novel eukaryotic cell-cyclecontrols.

L-FORMS AND THE NEOMURANREVOLUTION AND EUKARYOGENESIS:CELL CYCLE/STRUCTURE COEVOLUTION

The neomuran revolution was initiated by ac-cidental loss of the posibacterial murein wall.Posibacteria can spontaneously lose murein tobecome naked L-forms; similar protoplasts canbe generated in the laboratory by penicillin,preventing enzymes from resealing murein dur-ing growth (Dominguez-Cuevas et al. 2012).Such naked cells survive without bursting in os-motically protected environments and undergomultiform drastic changes, far more extensivethan the effects of most DNA mutations: ex-perimentally tractable analogs for the initiatingphase of the neomuran revolution.

During eubacterial cell cycles, coordinatedDNA segregation, cell growth, and division de-pend on CM attachment of sister chromosomes,cell wall rigidity, and its geometrically controlledgrowth and division (Fig. 2A) (Egan and Voll-mer 2013). The fission site for divisomes ismarked by a GTPase FtsZ filament ring (Adamsand Errington 2009; Buske and Levin 2013);

ParA ATPase filaments move replicon originsvia ParB-binding condensin-rich centromere-like parS DNA segments (Gruber and Erring-ton 2009; Sullivan et al. 2009); after replicationterminates with eubacteria-specific dimer res-olution by Xer-recombinase, surface-associatedDNA translocases separate termini (Kaimeret al. 2009; Sherratt et al. 2010; Grainge et al.2011). Murein loss grossly disrupts that elab-orate cell-cycle mechanism and osmoticallydestabilizes cells. L-forms therefore bleb mem-branes, yielding DNA-less gobbets and poly-ploid cells with several chromosomes. Thesetraumatized physically and genetically unstablecells over weeks or months can spontaneouslybecome stable L-forms without murein, whichcan multiply for years and have even been iso-lated from nature; that indicates extremelystrong selection for cell-cycle stabilization in ac-cidental L-forms. Eubacterial chromosome or-ganization (a single replicon with one originand terminus only) is fundamentally coadaptedto progressive DNA segregation throughout thecell cycle and active movement of origins sepa-rately from termini (Cavalier-Smith 1987a); thatoriginally unfashionable idea of active move-ment and orderly cell-surface association of eu-bacterial DNA is vindicated (Stouf et al. 2013).

Simultaneously, I argued that unique fea-tures of eukaryote genetics and cell cycles werecaused by rescuing segregationally defectiveL-forms by evolving mitosis with one-stepsudden anaphase segregation of all DNA, neces-sitated by phagotrophy-induced internaliza-tion of membrane DNA-attachment sites (Cava-lier-Smith 1987c). Indirectly, novel cell structurechanged genetics by shifting selective forces.As adumbrated for eukaryotes alone (Cavalier-Smith 1975), nucleosomes with histones H3/4evolved immediately following murein wall lossto stabilize neomuran chromosomes. HistonesH2a,b and H1 evolved in eukaryotes only for mi-totic chromosome condensation cycles: prophasecompaction to avoid shearing in sudden ana-phase, plus telophase loosening for interphasetranscription (Cavalier-Smith 2010c). In its neo-muran form, the unified phagotrophy and co-evolutionary theoryofeukaryogenesiscoherentlyexplained 48 eukaryotic innovations (Cavalier-

T. Cavalier-Smith




Smith 1987c), and subsequently still more (Ca-valier-Smith 2002c, 2010c).

THE NEOMURAN REVOLUTION,EUKARYOGENESIS, ANDARCHAEBACTERIAL DIVERGENCE

Mechanistically, the neomuran theory compris-es three distinct, interlinked theories, explaining(1) the origin of eukaryotes; (2) the origin ofarchaebacteria; and (3) that both origins werecaused by a single dramatic initiating event,the “neomuran revolution”: a radical changein eubacterial cell structure generating a tran-sient, unstable L-form-like, evolutionary inter-mediate—the neomuran cenancestor (Cavalier-Smith 1987c). The core ideas of the neomurantheory are as follows: (1) Murein loss and cell-cycle destabilization happened immediatelybefore Eukaryote/Archaebacterial divergence.(2) Stabilization was rapid (over a few years,not thousands or millions) but extended longenough for some changes to occur in commonbefore the phylogenetic split (Table 1) and some(in radically divergent directions) subsequently,separately in eukaryote and archaebacterial sis-ter lineages (Fig. 1). The neomuran revolutionwas a rescue from trauma that changed cellsmore radically than standard evolutionary di-vergence by single-gene mutations or gettingpreexisting genes from foreign bacteria byLGT. (3) Evolution of phagotrophy simultane-ously explains the novel cell structure and ge-netic system of eukaryotes, both mechanisticallyand selectively. Transient instability and poly-ploidy before mitosis was perfected made in-numerable simultaneous gene duplications (al-lowing major genetic innovations) and genelosses. (4) The evolutionary success and cellulardivergence of eukaryotes and archaebacteria

were because the stabilized neomuran’s imme-diate descendants adopted two entirely novel,contrasting life styles, freeing them from com-petition with their long-established eubacterialancestors: phagotrophy and hyperthermophily.

The naked protoeukaryotic lineage inventedphagotrophy by phagocytosis and internal di-gestion of other cells, thereby becoming moreefficient predators than any eubacteria, andsexual cell fusion, thereby creating more sharp-ly defined “biological species” with radicallydifferent population structure from typicallyclonal prokaryotes (Cavalier-Smith 1991a). Ar-chaebacteria invented hyperthermophily vianovel heat-stable lipids: isoprenoid ethers withtwo glycerophospholipids covalently linked bytheir hydrophobic tails, making double-headedtetraether lipids forming a stable membranemonolayer. Life originated as eubacteria undereasy mesophilic conditions, although Thermo-togales and Aquificales became secondary hy-perthermophiles by thermostable-protein LGTfrom archaebacteria. Fatty acid and isoprenoidsynthesis evolved in the cenancestral cell (Lom-bard and Moreira 2011; Lombard et al. 2012a,b;like many others, they incorrectly root thetree between neomura and eubacteria). Youngerarchaebacteria were ancestrally hyperthermo-philes, although some subsequently becamemesophilic by LGT from eubacteria, revertingto lipid bilayers, unavoidably retaining isopre-noid diethers because their hyperthermophiliccenancestor irretrievably lost ancestral acyl esterbiosynthetic machinery. Because evolving arch-aebacterial heat-stable lipids was enzymatical-ly simple, they became the world’s first hyper-thermophiles by gradually modifying numerousproteins enabling spread into ever-hotter hab-itats free of competitors. The neomuran me-valonate pathway for isoprenoid synthesis was

Table 1. Key stabilizing innovations during the neomuran revolution

1. MreB becomes actin; stabilizes against osmotic stress2. ESCRT-III filaments for membrane division during cytokinesis3. DNA gyrase (ATP-using active supercoiler) lost; replaced by histones H3/4 (later lost by crenarchaeotes);

passive wrapping of DNA around nucleosomes protects; Mcm proteins4. SRP adds cotranslation arrest domain5. N-linked glycoproteins





already present in their posibacterial ancestor(archaebacteria replaced some enzymes) as wasa relative of the enzyme making sn-glycerol-1-phosphate to which archaebacteria alone attachisoprenoid tails (Pereto et al. 2004). Probablymevalonate and methylerythritol phosphate(MEP) isoprenoid synthesis pathways both orig-inated in early posibacteria and were differen-tially lost (notably MEP during the neomuranrevolution and mevalonate by manyeubacteria);LGT was rarer than earlier claimed (Lombardand Moreira 2011).

Membrane stability also involved retain-ing contiguous laterally interacting semicrystall-ine surface glycoproteins as an S layer, like thenonglycosylated S layer outside the thick mureinwalls of their posibacterial ancestors. Wall rigid-ity enabled some archaebacteria to retain the an-cestral eubacterial FtsZ cell division machinerybut prevented phagotrophy or sexual cell fusionevolving. Therefore archaebacterial cell struc-ture, growth, division, and genetics remainedfundamentally bacterial or prokaryotic. Earlyclaims that archaebacteria are a “third form oflife” in addition to eukaryotes and prokaryotes/bacteria are thus falsified, despite misleading,confusing, purely propagandistic name changesthat some of us never accepted (deleting “bacte-ria” from archaebacteria and “eu” from eubac-teria); as Archaea Koch and Baerendt, 1854 isa genus of Madagascan spiders first discoveredin Baltic amber (Dippenaar-Schoeman andJocque 1997), it was doubly confusing to usethe same name for a group of bacteria and tomake it ambiguous whether “bacteria” refers toall prokaryotes, as it properly does (Cavalier-Smith 2007a), or just eubacteria (Woese et al.1990). In fact, archaebacteria are a third form ofbacteria (Figs. 1 and 3). Contrary to early mis-conceptions, still sadly widespread, their dis-covery was irrelevant to the origin of life, yetcrucial for understanding eukaryote originsbecause their striking molecular differencesfrom eubacteria and marked partial similaritiesto eukaryotes enabled four strong deductions,universally accepted:

1. In conjunction with strong similarities of mi-tochondria to a-proteobacteria (John andWhatley 1975; Gray 1992), it showed that mi-

tochondria could not have evolved from thesame ancestor as the rest of the eukaryoticcell. One must accept that eukaryotes are evo-lutionary chimeras of a moderately changeda-proteobacterium (negibacterium) and farmore complex, radically different host cell(Cavalier-Smith 2002c).

2. That host was more closely related to ar-chaebacteria than to eubacteria.

3. Differences between it and eubacteriaevolved in two stages: those shared witharchaebacteria first; organizationally moreradical, uniquely eukaryotic inventions sub-sequently.

4. In conjunction with marked similarities be-tween chloroplasts and cyanobacteria, onemust accept that the host component was aheterotroph, chloroplasts being implantedsubsequently by phagocytosis and radicaltransformation of cyanobacteria (Cavalier-Smith 2000, 2013b).

ACTINOBACTERIA, LIKELY NEOMURANSISTERS

The neomuran theory identified exospore-forming actinobacteria (e.g., Mycobacterium,Streptomyces), all possessing phosphatidylino-sitol (unlike other prokaryotes), crucial foreukaryote cell signaling, and cholesterol, andmany with 40S proteasomes (unlike other eu-bacteria), as more likely posibacterial ancestorsof neomura than Teichobacteria, the posibac-terial class with thick teichoic-acid-containingwalls and endospores (e.g., Bacillus), althoughValas and Bourne (2011) gave reasons favor-ing Teichobacteria instead. Their and my argu-ments are simultaneously satisfied if the an-cestor was a phylogenetic intermediate, sisterto actinobacteria but derived from a Teichobac-teria-like ancestor (Fig. 3); this also reconcilesneomuran theory with sequence trees that typ-ically place neomura in an unresolved positionclose to the base of the eubacterial radiation inwhich branching orders among the three mainposibacterial groups Actinobacteria, Endobac-teria, and Chloroflexi and negibacterial phy-la are ill-resolved (Pace 2009; Woese 2013), aswell as with actinobacteria-specific signatures

T. Cavalier-Smith




(Gao and Gupta 2005, 2012; Gao et al. 2006,2009; Servin et al. 2008). Actinobacteria are im-mediately adjacent to archaebacteria on two ofthe five protein-family splits networks of Daganet al. (2010) based on 191 bacterial genomesand separated from them by Endobacteria ona third, so the majority support a neomuranrelationship with Posibacteria, with actinobac-teria favored as the closest. Cyanobacteria arealso adjacent on three, although their negibac-terial cell structure makes a direct relationshipless likely, whereas Gracilicutes are overall moredistant; thus, of the limited bacterial mega-groups sampled, posibacteria and cyanobacte-ria were the best candidates for the closest rela-tives to neomura. It is a pity that the analysesexcluded Eurybacteria and Hadobacteria (datafor Chloroflexi were then unavailable). Actino-bacteria alone in prokaryotes have a CTP-de-pendent cardiolipin synthetase needed for mi-tochondria (Sandoval-Calderon et al. 2009).

COTRANSLATIONAL SECRETION DURINGTHE NEOMURAN REVOLUTIONAND EUKARYOGENESIS

The neomuran cenancestor evolved shared fea-tures of ribosomes (anisomycin inhibition ofpeptidyl transferase; chloramphenicol resis-tance; new proteins), ribosomal rRNA process-ing (fibrillarin; U3 and C/D- and H/AC-boxsnoRNA guides for pre-rRNA cleavage andbase modification), tRNA protein-spliced in-trons, and signal recognition particle (helix 6;SRP19 protein) that are absent in eubacteria andevolved from the simpler system of Posibacteria(Cavalier-Smith 1987c, 2002b). Neomuran gly-coproteins are made by ribosomes docked ontoa membrane by the ribonucleoprotein signalrecognition particle (SRP) that recognizes theiramino-terminal hydrophobic signal sequenceand membrane-embedded SRP receptors andtransfers the signal hairpin into the membrane.The rest of the unfolded protein then crossesthe membrane via an openable protein channel,folding correctly on the other side. Before signalpeptidase removes its signal peptide, oligosac-charyltransferase attaches a presynthesized hex-ose-rich oligosaccharide by its basal N-acetyl-

glusosamine to one or more asparagines twoamino acids upstream of a serine or threonine.The oligosaccharide is synthesized sequentiallyon the ribosomal face of the membrane whilecovalently attached (in eukaryotes always viaN-acetylglucosamine; in archaebacteria ances-trally thus) to a phosphorylated polyisoprenolcarrier (dolichol phosphate or pyrophosphate).This hydrophilic core oligosaccharide is flippedacross the membrane by flippase protein. Thiscomplex machinery is strongly conserved andhomologous between archaebacteria and eu-karyotes, proving the central thesis of the neo-muran theory and refuting the progenote idea(Woese and Fox 1977). The common ancestorof neomura, undeniably with all of these mech-anisms, was an advanced cell with extremely so-phisticated translation, transcription, DNA rep-lication and segregation, and a lipid membranewith numerous embedded proteins. There cer-tainly was an historical changeover betweenacylester membrane lipids (eukaryotes and eu-bacteria) and isoprenoid ethers of archaebac-teria: one was not converted into the other(chemically impossible). Instead, one replacedthe other, hyperthermophily being the strongselective advantage for replacing eubacterial byarchaebacterial lipids (Cavalier-Smith 1987b,c),there being none for the reverse; the transientintermediate had both types (Cavalier-Smith2002b).

Eubacterial murein peptidoglycan is anal-ogous to neomuran N-linked glycoproteins:both comprise amino acids, sugars, and aminosugars; synthesis involves transfer of hydrophil-ic core units containing N-acetylglusosamineacross the hydrophobic CM, after being synthe-sized on its cytosolic face by sequential covalentattachment to a phosphorylated polyprenol.Murein consists of an alternating copolymer ofamino sugars N-acetylglusosamine and N-ace-tyl muramic acid covalently cross-linked by oli-gopeptides: its precursors flipped across CM onundecaprenol are muramopeptides with twoamino sugars and four amino acids. Commonbiogenetic involvement of N-acetylglucosamineand isoprenoid carriers suggested an evolution-ary relationship (Cavalier-Smith 1987c); I pro-posed that the ancestral neomuran posibacte-





ria-derived L-form lost muramic acid, omittedthe amino acids, and added extra hexoses,and the same oligosaccharyltransferase addedit to preexisting S-layer proteins, yielding thefirst N-linked glycoproteins (Cavalier-Smith1987c). BLAST analysis now strongly supportsthat: as predicted, eukaryotic and archaebac-terial dolichol-dependent oligosaccharyltrans-ferases making N-linked glycoproteins are ho-mologous, strongly conserved, and homologouswith, but somewhat less similar to, the unde-caprenol-associated oligosaccharyltransferasesof teichobacterial and actinobacterial posibac-teria that make murein. Therefore, almost cer-tainly, neomuran N-linked glycoproteins didevolve from preexisting posibacterial oligo-saccharyltransferases that previously made mu-rein. The similarity of neomuran transferasesis much greater to those of posibacteria thanto any Negibacteria except a few d-proteobacte-ria, for example, the myxobacterium Stigma-tella; these few d-proteobacterial oligosacchar-yltransferases perhaps underwent LGT fromneomura to myxobacteria.

The eukaryote rough ER cotranslationalprotein channel comprises Sec61 and ancillarya,g-subunits; these evolved from the SecYEGprokaryote translocon when phagotrophy in-ternalized the CM during eukaryogenesis. Theplasma membrane also kept prokaryotic ABCtransporters for rare posttranslational proteinsecretion, for example yeast mating hormones,but other protein secretion mechanisms werelost or transformed beyond confident recogni-tion during eukaryogenesis. Loss of type IV se-cretion (responsible for bacterial conjugationin archaebacteria and eubacteria) largely ex-plains why LGT is much rarer in eukaryotesthan bacteria. The TatAC bacterial pathway forexporting folded proteins (using the same sig-nal peptidase as Sec for unfolded ones) (Palmerand Berks 2012) is absent in eukaryotes, butthe more complex homologous negibacterialTatABC was retained by chloroplast thylakoids.

Universal Sec arose in the first cell, as didTAT and ABC transporters, and two types ofsignal peptidases: type I for most proteins andtype II specific for lipoproteins. Type III, usedfor secreting prokaryotic flagellar proteins and

pilins and in archaebacteria also for many otherproteins, probably evolved subsequently withflagella. Types II and III signal peptidases werelost during eukaryogenesis. Eubacterial SecAthat unfolds folded proteins for threadingthrough SecY was lost when the neomuran an-cestor made Sec more cotranslational, by thenovel neomuran arrest domain of SRP (Cava-lier-Smith 2002b). Bacterial TAT secretes foldedglobular proteins across CM. Cavalier-Smith(2006a, 2009) suggested that eukaryotes mod-ified TAT for protein import into peroxisomes.Import of soluble peroxisomal enzymes usestwo alternative topogenic sequences: a nona-peptide recognized by type II receptors and thecarboxy-terminal (type I) SKL motif by recep-tors; although TAT recognizes a carboxy-ter-minal RR motif, Pex5, the SKL receptors likeTAT use tricopeptide repeats (Rucktaschel et al.2010) and thus might be related.

COEVOLUTION OF MITOCHONDRIAL,ER, AND PEROXISOMAL RESPIRATIONAND SEGREGATION

These three oxygen-consuming respiratoryorganelles—the mitochondrion, the ER, andthe peroxisome—coevolved in early eukaryotes,partitioning aerobic metabolism among them,making aerobic phagotrophy more efficient.Early mitosis was open or semiopen, with per-oxisomes segregated to daughter cells by at-tachment to the nuclear envelope through thecell cycle. Space constraints prevent discuss-ing the origin of mitochondria, probably froma photosynthetic a-proteobacterium contem-poraneously with the origin of cilia and nu-cleus (Cavalier-Smith 2006b, 2007b), and howits negibacterial reproduction (originally usingFtsZ-associated division) was modified and in-tegrated into the eukaryotic cell cycle, partlydivergently in excavates and Euglenozoa (Kine-toplastea segregate mitochondria by centrioleattachment) (Gluenz et al. 2011); they and dip-lonemids evolved glycosomes from peroxisomes(Cavalier-Smith 1997; Gualdron-Lopez et al.2012). Peroxisomes originated marginally beforemitochondria, not long before, as De Duve oncethought. Although they probably evolved by pri-

T. Cavalier-Smith




mordial divergence from the ER (Cavalier-Smith1975; Gabaldon 2010), not by symbiogenesis ofbacteria as De Duve proposed, they evolved noveldivision machinery, multiplying mainly bygrowth and division (Pieuchot and Jedd 2012),thus becoming effectively distinct genetic mem-branes (Cavalier-Smith 2004). Mitochondria,and subsequently plastids, apparently cooptedthis division machinery (Pan and Hu 2011), justas mitochondria arguably duplicated the per-oxisomal ADP/ATP exchanger when evolvingIM carriers, the key to their enslavement (Cava-lier-Smith 2006b), subsequently donating dupli-cates to chloroplasts (Cavalier-Smith 2000). An-cestrally, peroxisomes were probably segregatedthrough specific nuclear-envelope attachments,retained by diverse protists (Cavalier-Smith andOates 2012).

Separating ATP-generating and degradativerespiratory cytochromes in mitochondria andER following the differentiation of peroxisomesand endomembranes (Fig. 2) made aerobic ex-ploitation of prey more efficient, but, contraryto Lane and Martin (2010), mitochondria didnot cause large eukaryote genomes. Increasedgenome size resulted from the origin of the nov-el phagotrophy-driven cytoskeleton, eukaryoticcell cycle, and nuclear architecture with nuclearpores, which collectively transformed the ma-chinery and selective forces acting on chromo-somes, and by allowing marked (initially phago-trophy-driven) increases in cell size imposedupward coevolutionary pressures on genomesize, which controls nuclear volume, which in-variably coevolves with cell size (Cavalier-Smith2005, 2010c). Symbiogenesis did not drive theseinnovations; for its importance to eukaryote cellevolution, see Cavalier-Smith (2013b).

DIVERSE CELL BIOLOGY OF FILARCHAEOTACLARIFIES NEOMURAN EARLY EVOLUTION

Archaebacterial multigene trees show Korarch-aeota and Thaumarchaeota forming a cladewith Crenarchaeota (Podar et al. 2013), all heregrouped as subphylum Filarchaeota (Table 2) toemphasize that they ancestrally had two eukary-ote-like filamentous cytoskeletal proteins: actin(Yutin et al. 2009) and ESCRT-III filaments for

cytokinesis (Hobel et al. 2008; Samson etal. 2008; Ettema and Bernander 2009; Makarovaet al. 2010). A direct eukaryotic origin from fil-archaeotes is implausible because they lackthe right lipids and scores of proteins (e.g., dy-namin, Hsp90) vital for making eukaryotes.The archaebacterial cenancestor was not a crudegene-poor progenote but had .1000 genes,undergoing extensive genomic reduction af-ter diverging from eukaryotes (Cavalier-Smith2002b, 2007b): differential losses during ar-chaebacterial radiation reduced genes from an-cestrally �2000 (Csuros and Miklos 2009; Wolfet al. 2012), yielding lineages lacking differentsubsets of proteins necessary for making eu-karyotes, all originally present in the more com-plex ancestral neomuran. The mechanisticallyimplausible idea of host archaebacterial lip-id replacement by mitochondria (Martin andMuller 1998) can “explain” hypothetical lossof isoprenoid ether lipids but not acquisitionof phosphatidylinositol, cardiolipin (actinobac-teria alone in prokaryotes have CTP-dependentcardiolipin synthetase), and sterols, all readilysupplied by the neomuran ancestor. It is farsimpler that this ancestor stabilized its genomeby evolving chromatin, its cytoplasm by ac-tin, and division by ESCRT before eukaryotesand archaebacteria diverged, filarchaeotes los-ing dynamin, and euryarchaeotes losing actinand ESCRT. Many trees suggest archaebacterialholophyly (Yutin et al. 2008), but despite incon-trovertible evidence for multiple losses withinFilarchaeota, some investigators ignore the pos-sibility of eurybacterial actin/ESCRT loss andsuppose that eukaryotes evolved from filar-chaeotes. Contradictorily, others assumed thateukaryotes evolved from euryarchaeotes (whereeukaryotic-like glycosyltransferase is more wide-spread than in crenarchaeotes) because of his-tone similarities (Reeve et al. 2004; Cubonovaet al. 2005) or the phylogenetically refuted andmechanistically implausible hypothesis that eu-karyotes evolved from a methanogenic euryar-chaeote (Martin and Muller 1998).

Given huge problems in cell biology thatmassive replacement of host lipids and proteinsentails in a hypothetical archaebacterial origin,one must not uncritically accept certain mul-





tigene trees emphasizing ribosomal proteinsshowing eukaryotes as sisters to Filarchaeotaonly, not to all archaebacteria (Williams et al.2012), as refuting the neomuran theory. As Va-las and Bourne (2011) explain, most of those 31proteins (Williams et al. 2012) underwent rapid(possibly tree-biasing) quantum-evolutionarychanges in ribosomes during the origins ofeukaryotes and archaebacteria, as earlier arguedfor rRNA (Cavalier-Smith 2002b); the very longstem at the base of eukaryotes in these treesis precisely the condition most likely to ensurethat a genuine outgroup (sister) is misplacedwithin the ingroup (here archaebacteria), as de-cisively shown by Shavit et al. (2007). I there-fore consider this probable artifact the simplestway to reconcile contradictory trees showingarchaebacterial holophyly (probably right forreasons previously detailed) (Cavalier-Smith2010c) and paraphyly (probably this long-stem artifact)—see also Gribaldo et al. (2010),who do not yet take seriously the possibility

that neomura arose from eubacteria, despite itsimply resolving the impasse that they discuss(although unlike most overinfluenced by theprogenote myth that ignored cell biology andpalaeontology, they are open-minded enoughto mention it as a potential solution to whichthey may eventually be driven!). Even the Wil-liams et al. tree suggests that eukaryotes andarchaebacteria arose nearly contemporaneously,reconcilable with fossil dates and eukaryotemultigene trees only by accepting that archae-bacteria are at least twice as young as eubacteriaand evolved from them (Cavalier-Smith 2006a,2013a). Neomuran theory holds that eukaryotesand archaebacteria diverged almost immedi-ately after neomura originated, and euryar-chaeotes and filarchaeotes diverged immediate-ly thereafter. It therefore predicts that it shouldbe extremely hard for multigene trees to de-cide whether eukaryotes are sisters to or nesteddeeply within archaebacteria, expecting sometrees to show one and some the other, exactly

Table 2. Classification of new archaebacterial subphylum Filarchaeotaa

Class 1 ‘Korarchaeota’Name suggested by Barns et al. (1996) at the grossly inflated ‘kingdom’ rank; should be a class, which cannot be

made formally until ‘Candidatus Korarcheum’ or another contained genus is legitimately published.

Class 2. ‘Thaumarchaeota’Name suggested by Brochier-Armanet et al. (2008) at the unwisely inflated rank of ‘phylum’; should be a class,

but not formally possible until a genus within it is legitimately published.

Order 1. ‘Cenarchaeales’ Cavalier-Smith, 2002 was illegitimate as the widely used ‘type genus’ ‘Cenarchaeum’ isstill not formally published and thus illegitimate.

Order 2. ‘Caldiarchaeales’ may be a suitable name for the clade containing ‘Candidatus Caldiarchaeum’, whichdoes not deserve the phylum rank suggested by Nunoura et al. (2011) as ‘Aigarchaeota’.

Class 3. Crenarchaeota Cavalier-Smith, 2002 (later synonym: Thermoprotei [Reysenbach 2002])Their stronger hyperthermophily than other archaebacteria probably caused more extensive loss of ancestral

proteins (e.g., histones).aDiagnosis: Archaebacteria containing actin and/or ESCRT-III filaments, unlike Euryarchaeota. Type order

Thermoproteales Zillig and Stetter, 1982. Etymology: Filum L. thread, to show they have one or both of these cytoskeletal

filaments: archae Gk ancient. Comment: Equivalent to the too highly ranked informal ‘TACK superphylum’ (Guy and Ettema

2011), who demonstrate numerous independent losses of neomuran proteins within Filarchaeota and Euryarchaeota,

postulating loss of actin and ESCRT-III by Euryarchaeota, which retained FtsZ and histones, unlike most Filarchaeota, is

much more parsimonious than accepting that eukaryotes are sisters to Filarchaeota, as some trees suggest (Williams et al.

2012). I give a formal definition despite no taxa above classes currently having standing in prokaryotic nomenclature; the code

needs changing to encompass phylum and kingdom names (as for classes, do not apply priority to them). Clades DSAG/AAG/

MHVG of Guy and Ettema (2011) should not be a separate phylum, but may belong in Thaumarchaeota when better

characterized. Filarchaeota contains all archaebacteria except subphylum Euryarchaeota (with five classes: Cavalier-Smith

2002b), in which I now place ‘Nanoarchaeum’, which will not deserve separation from other archaebacteria at higher rank than

an order after being formally named, contrary to those who treat it as a ‘phylum’; its deep divergence from other euryarchaeotes

looks like a typical artifact of rapid evolution after parasitic reduction, as in mycoplasmas, microsporidia, and Mikrocytos.

T. Cavalier-Smith




as is observed. Above all, these trees provideno evidence that archaebacteria are as old aseubacteria. If they were, they would be two tothree times older than eukaryotes, which shouldmake multigene trees nest eukaryotes shallowlywithin archaebacteria, which they do not.

SUCCESSIVE GLIDING, FISHING, ANDSWIMMING STEPS IN THE PHAGOTROPHY-RELATED ORIGIN OF CILIA

Ciliary biogenesis and locomotory undulationare extremely complex, requiring more than 660genes (Carvalho-Santos et al. 2010, 2011), andevolved autogenously by modifying the micro-tubular cytoskeleton and associated motors.Early theories inadequately explained how thatcomplexity evolved (Cavalier-Smith 1978a,1982). Centrioles are so intricately constructed(Li et al. 2012) that their ontogeny proba-bly closely recapitulates phylogeny: amorphousgerminative discs containing g-tubulin initiateassembly of ninefold cartwheels, then singlets,then doublets/triplets (Cavalier-Smith 1974;Jerka-Dziadosz et al. 2010; Gogendeau et al.2011). Intraflagellar transport (IFT) machineryfor intraciliary protein movement evolved fromCOPI-coated vesicles (Jekely and Arendt 2006;van Dam et al. 2013), probably contemporane-ously with nuclear pore complexes from COPIIvesicle coats (Cavalier-Smith 2010c); ciliarygliding, using the same machinery, probablyarose before swimming (Cavalier-Smith 2009).In remarkable harmony with the simultaneouscoevolutionary origin of cilia and nuclei thesis(Cavalier-Smith 1987c), cilia are a distinct cellcompartment into which ciliary proteins aretargeted across a basal “ciliary pore complex”by machinery containing importin and nucleo-porins used for nuclear import, similarly gov-erned by a RanGTP/GDP gradient, and recog-nizing ciliary localization signals (CLS) relatedto nuclear localization signals (NLS) (Fan et al.2011; Kee et al. 2012). Nuclear import machin-ery was either partially coopted and modifiedfor cilia or nuclear and ciliary import divergedsimultaneously from vesicle coats.

Like bird feathers, which originated to in-sulate small running dinosaurs (easier), subse-

quently modified for flight (harder), Figure 4outlines an autogenous origin of centrioles andcilia with two functional shifts and three succes-sive selective forces: (1) kinesin-driven ciliarygliding and compartmentation for surface mo-tility; (2) protociliary duplication and transfor-mation making a differentiated prey trappinganterior ciliary fishing rod (as Phalansteriumcilium does today: Smirnov et al. 2011) anda posterior gliding cilium for enhancing pha-gotrophy; and (3) ciliary bending, initially toimprove fishing and subsequently allow swim-ming. Thus, improved phagotrophy probablypreceded swimming as a selective advantage,and final stages in centriole and ciliary evolu-tion followed ciliary duplication, unlike previ-ous theories. The resulting cenancestral eukary-ote immediately diverged into (a) swimmingexcavates by losing gliding, evolving a ventralgroove with a split right centriolar root and aposterior ciliary undulation drawing in food;and (b) posteriorly gliding Euglenozoa by evolv-ing a ventral root-supported cytostome for sur-face feeding, consistently with rooting the treebetween excavates and Euglenozoa (Figs. 1 and3), with contrasting centriole-anchoring cyto-skeletons (Cavalier-Smith 2013a). Because glid-ing and ciliary surface feeding are mechanis-tically simpler, and gliding could be initiatedby preexisting cytoskeletal properties (Fig. 4A),their historically preceding ciliary beating al-lowed the seemingly irreducible complexity ofciliary beating to evolve in simple stages; eachhad clear selective advantages, improvable grad-ually by blind mutations, needing no symbio-genesis (Margulis 1970), no viral input (Satiret al. 2008), and no intelligent design.

Centriolar duplication and mitosis co-evolved with new cell-cycle controls (Fig. 5).Although archaebacteria share eukaryote-likeDNA replication machinery, cycle controls re-mained prokaryotic because they retain cell-surface-attached DNA, ParA P-loop ATPasefilaments, and typically ParB centromere-likeproteins for moving chromosomes (Bernanderet al. 2010), although the ancestral parS DNAsites that they bind are rarely recognizable(Livny et al. 2007), probably through histone-stimulated divergence. Eukaryotic cycle control





Figure 4. Autogenous origin of cilia with successive origins of different ciliary components under three con-trasting selective forces. (A) During or just after the origin of nuclear pore complexes (npc Fig. 2), singletmicrotubules from the g-tubulin protocentrosome push out the plasma membrane as a protocilium by whichplus-end-directed kinesin-2 motors (Verhey et al. 2011) attached by their tails to glycoprotein surface adhesinssticking to the substratum propel its microtubules forward (arrow) in primitive gliding motility, enabling cells tofind fresh prey on the substratum as phagotrophy locally depletes it. (B) Posterior ciliary gliding was improved byattaching microtubules firmly to the cell surface by protocentriolar transitional fibers (proximally) and Y-shaped membrane connectors (slightly distally) plus ciliary compartmentation dependent on novel diffusionbarriers (septin filaments in ciliary membrane base: Carvalho-Santos et al. 2011), central distal transition plate(TP), and a peripheral dense collar at the distal end of the Y-connector region, associated with npc proteins withnovel ciliary import machinery using npcs and importin-b2 and NLS-related CLS targeting sequences (locatedin TP and/or collar). (Legend continues on following page.)




(including cyclins, origin of ubiquitin fromancestral neomuran ubiquitin-like proteins[Pearce et al. 2008; Maupin-Furlow 2013], andproteasome complexification), were thereforeconsequences not of novel neomuran chro-matin and DNA-replication machinery but

of radical eukaryote-specific cell structural in-novations, especially phagotrophy-driven DNAinternalization to endomembranes and ori-gins of TubZ-GTPase-related microtubules, g-tubulin centrosomes, centrioles/cilia, and mi-tosis.

Figure 4. (Continued) Dynein 1b, the first dynein, evolved (by gene-duplicating its common ancestor with therelated AAAþATPase midasin/REA1 that mechanically strips biogenesis factors from 60S preribosomal subunitsjust before they exit the nucleus: Garbarino and Gibbons 2002; Kressler et al. 2012) to recycle distal adhesins left atthe protociliary tip by axoneme gliding. Anterograde movement of dynein improved by carriage on kinesin-2-driven IFTB particles that evolved from CopI coat proteins, and special SNAREs evolved for basal delivery ofciliary membrane precursor vesicles with distinctive proteins and lipids. Protocentrioles were rigidified by a hub-spoke core, microtubules fixed at nine by ninefold SAS-6 hub assembly (Guichard et al. 2012). Cross-linksrigidified the axoneme, the two on its substratum face having numerous extra-rigid linkers linking them likesled runners (subsequently becoming doublets 1 and 2 of Chlamydomonas: Lin et al. 2012). Novel proteins (e.g.,Rib43a: Norrander et al. 2000) stabilized A-tubules and the bases of centriolar root microtubules (dorsal andventral fans) compared with transient spindle microtubules. (C) Ciliary duplication produced a younger “fish-ing” cilium projecting into the medium for trapping swimming bacteria, pulling them baseward by minus-directed dynein 1b, for phagocytosis. Centrin plus novel proteins orthogonally rigidly connected the two pro-tocentrioles now with doublets (architecture [Nicastro et al. 2011] perhaps determined by novel microtubuleinner proteins, for example PACRG (Ikeda et al. 2007; Ikeda 2008), stabilized by a scaffold containing 1-tubulin[blue]). PACRG-interacting Rib72 (Ikeda et al. 2003) differentiated ciliary from centriolar doublets. Centriolartransformation temporally and physiologically differentiated the two cilia and made separate left and rightventral microtubular roots. (D) Successive dynein duplications generated inner arms (top: sufficient for bending[Heiss et al. 2013] as doublet 1–2 linker excluded arms from doublet 1, destroying ninefold symmetry; then thenexin–dynein regulatory complex [Heuser et al. 2009, 2012; Lin et al. 2011; Bower et al. 2013] for calciumregulation of beat; then outer dynein arms for greater power, and the center pair [nucleated by g-tubulin on TP,and fixed so did not rotate] with new arms [Carbajal-Gonzalez et al. 2013] and kinesin-9 [Wickstead et al. 2010]and spokes [Barber et al. 2012] to modulate beat mechanics [not inherently needed for planar beat: Idei et al.2013]) to draw in more prey by water currents. For pictorial simplicity, the cell body is proportionally too smalland ahead of the cilia, but probably at stage b/c the cytoskeleton geometrically rearranged to lift the cell bodyfrom the substratum and put the ciliated cell apex at the front, which is mechanically stabler (found in all extantposterior ciliary gliding eukaryotes), entailing a basal stable bend ([E] mechanism unknown) for the posteriorgliding cilium. Many complexities of present cilia (Mizuno et al. 2012) probably evolved subsequently to improveefficiency but would not have been essential for their origin, for example, association of IFTA/B into one complexand of these into distinct anterograde and retrograde trains (Pigino et al. 2009; Buisson et al. 2013), and additionof BBsomes, likely adaptors for improving retrograde transport of some proteins (Lechtreck et al. 2013) andsensory functions (sensation [Jekely and Arendt 2006] is less plausible than gliding for the original function), andbeat pattern modulators. (E) The cenancestral eukaryote diverged to form swimming excavates that abandonedgliding and undulate the posterior cilium to draw prey into the ventral groove supported by a split right ventralcentriolar root R2 (blue), and Euglenozoa that ancestrally retained gliding, added a cytopharnyx supported byancestrally unsplit R2 and dissimilar paraxonemal rods (probably attached to the specially linked doublet 1–2homologs) to broaden and further rigidify the posterior cilium for stabler gliding, and parallelized their centri-oles within a ciliary pocket. After losing phagotrophy, saprophytic, parasitic, and photosynthetic Euglenozoa lostgliding and developed swimming by the anterior (Euglenophyceae) or posterior (trypanosomatids) cilium;some bacterivores (petalomonads) lost the posterior cilium, presumably recruiting dynein 1b for anterior ciliarygliding. All other eukaryotes evolved from excavates (Fig. 3); Apusozoa, Cercozoa, and the heterokont Caecitellusreevolved posterior ciliary gliding, presumably using kinesin-2. The V-fiber, with associated acorn-base attacheddistally to centriole 1–2 triplets (at least in neokaryotes: Geimer and Melkonian 2005), demarcates the centriolefrom the transition zone and perhaps evolved in the cenancestral eukaryote; its rotational asymmetry and that ofcentriolar root attachments to specific triplets probably reflect an asymmetric doublet “numbering machinery”that probably evolved in the earliest gliding protocilium (B).





Figure 5. Eukaryote cell-cycle logic and evolution. Complexes of cyclin-dependent kinase (CDK) and cyclinscontrol the eukaryotic cell cycle by phosphorylating numerous proteins, timed by growth-dependent increasesin cyclins and their sudden proteolysis (red curves) (Nasmyth 1995). Cyclins share a domain with neomuran-specific transcription factor TFII, and CDKs evolved from posibacterial serine-threonine (S/T) kinases. Orig-inally one cyclin could have controlled S-phase initiation and anaphase onset using a lower threshold for theformer switch (Novak et al. 1998; Tyson and Novak 2008; Harashima et al. 2013); gene duplication enabledbetter control by cyclinE/cdk2 (for DNA replication and centriole duplication) and cyclin B/cdk1 to activate theanaphase promoting complex (APC), the ubiquitin ligase that initiates anaphase resetting of cell-cycle controlsvia proteasome degradation of cyclins and numerous key cycle proteins. Phosphorylation-cum-proteolytic cell-cycle controls originated in posibacteria, but a novel Cdc-6-mediated control over replication initiation evolvedin ancestral neomura after histones H3/4 and MCM DNA helicase, replacing eubacterial DnaA/CrtA control;proteolysis by proteasomes that originated in the neomuran/actinobacterial cenancestor replaced eubacterialClpXP proteolysis. Ancestral eukaryotes evolved origin recognition complexes (ORCs) more complex than thesingle protein Cdc-6 of archaebacteria, probably because the suddenness of mitotic anaphase (rapidly segre-gating all parts of chromosomes at once, unlike the temporally separate segregation of origins and the generallysingle terminus in bacteria) required concerted replication initiation at hundreds of origins, ensuring replicationcompletion and tighter chromosome folding (using extra histones and novel heterochromatin machinery:Cavalier-Smith 2010c) well before mitosis; uniquely in eukaryotes, mitosis demands a temporally discrete Sphase. Formerly, only neokaryotes were thought to have ORCs (Cavalier-Smith 2010b), but extremely divergentversions of most constituents are now known in trypanosomes, whose cell-cycle controls are the most divergentwithin eukaryotes (Li 2012), consistent with eukaryotic rooting between Euglenozoa and neokaryotes (Figs. 1and 3). Successively more complex controls and checkpoints evolved with novel polo-like and aurora S/Tkinases playing multifarious roles in mitosis and cytokinesis and a multiplicity of kinesins evolving to improvespindle assembly and function. Probably all proteins shared by trypanosomes (Li 2012) and opisthokontsevolved before the eukaryote cenancestor. Mitosis (upper panel) was converted to meiosis by two innovations(lower panel): homologous chromosome pairing by the synaptonemal complex and blocking centromericcohesin digestion at meiosis 1 anaphase, which automatically bypassed cell-cycle resetting caused by anaphasecentromere splitting so that the next interphase had no S phase as previously explained (Cavalier-Smith 1981),thereby halving ploidy—the original meiotic function (Cavalier-Smith 2002a, 2010c). (Legend continues onfollowing page.)

T. Cavalier-Smith




MISUNDERSTANDINGS OF THENEOMURAN THEORY OF CELLEVOLUTION

Acceptance that archaebacteria are not a pri-mary domain but evolved from posibacteriais growing. Indel analysis (Skophammer et al.2006, 2007; Lake et al. 2009; Valas and Bourne2009, 2011) supports that and that negibacteriaare a derived clade. Palaeontology and multi-gene trees when critically combined prove thateubacteria historically preceded neomura (Cav-alier-Smith 2002b), and eubacteria are aboutthree times as old as neomura (Cavalier-Smith2006a, 2013a). Thus, there is only one primary“domain” of life: eubacteria (Cavalier-Smith1987b,c, 2006a,c). Archaebacteria are certainlymore recent. The neomura theory is probablycladistically historically correct and coherentlylogically explains the origins of both eukaryotesand archaebacteria. The transformations it en-tails are cell-biologically reasonable, selectivelycomprehensible, and unexplained by alternativeideas. A stimulus through antibiotic warfareamong bacteria is likely because penicillin cancause wall loss and many antibiotics target ri-bosomes and lead to resistance through changesin their structure (Cavalier-Smith 1992). De-spite its unequalled explanatory power, theneomuran theory was repeatedly ignored, mis-understood, or misrepresented, but remainsunrefuted. Most discoveries since 1987 greatlystrengthen its core ideas.

Yet, the lingering dogma/prejudice thatarchaebacteria are a “primary domain,” neverwith any sound evidence from paleontologyor phylogeny, still makes some reluctant to ac-cept that archaebacteria are younger than andevolved from eubacteria. Some still think ar-chaebacteria are ancestors, not sisters of eukary-otes, an idea subconsciously favored by wrong-ly calling shared features “archaebacterial” not

“archaebacteria-like.” Some skeptics confusefundamental tenets of the neomuran /phago-trophy/coevolutionary theory of eukaryoteorigins with the logically independent, now-dis-proved idea that some eukaryotes were primi-tively without mitochondria, proposed earlierwhen classifying amitochondrial eukaryotes asArchezoa and arguing for the first time (correctlyit turned out) that the mitochondrial OMevolved from that of a-proteobacteria (Cava-lier-Smith 1983a,b). Discoveries of mitochon-drial relics in all “Archezoa” showed that noprimitively amitochondrial eukaryotes persist,so I abandoned that idea long ago (Cavalier-Smith 1998, 2002c). That disproved “archezoanhypothesis” concerned the root of the eukaryotetree, not the causes of eukaryogenesis; causalexplanation of eukaryogenesis by the probablycorrect phagotrophy/coevolutionary theory islogically independent, as is the now-associatedneomuran interpretation of the phylogenetic re-lationship between eukaryotes, archaebacteria,and posibacteria, which also I argue is correct.Whatdiffers in my present interpretation(Fig. 1)is the root position for (a) eukaryotes, and(b) the whole tree of life. The position of theeukaryotic root between Euglenozoa and neo-karyotes (excavates and their descendants) is ar-guably more secure than previously, being basedon a dozen independent reasons, detailed else-where (Cavalier-Smith 2010a,c, 2013a), notjust one as in previous guesses, and also stronglysupported by all prokaryote-rooted ribosomalmultiprotein trees using 13, 432 amino-acid po-sitions (Lasek-Nesselquist and Gogarten 2013).An alternative, slightly different position of theeukaryote root within loukozoan excavates, be-tween jakobids and Malawimonas, based solelyon mitochondrial protein 42-gene trees (Zhaoet al. 2013), risks being somewhat inaccuratebecause of long-branch artifacts (Shavit et al.2007).

Figure 5. (Continued) It is unclear whether centromeres evolved from posibacterial plasmid partition proteinsassociated with GTPase TubZ (Aylett et al. 2010, 2013) or de novo; CENP-A was probably added only inneokaryotes after they diverged from Euglenozoa (Fig. 3) (Cavalier-Smith 2010c). As the text notes, recentevidence suggests that cohesin digestion is not invariably necessary for centriole separation in animals, raisingdoubt as to the ancestral eukaryotic process.





Differences in ribosomal RNA sequencesthat separated the so-called ‘three domains’ aresuperficial compared with deep similarities incell secretory machinery. Shared glycoproteinsynthesis involving dolichol phosphate car-riers was only one of 16 supporters of the neo-muran clade (Cavalier-Smith 1987c); sincethen at least as many more have been discovered(Cavalier-Smith 2010c); even those few stillimagining that archaebacteria and eubacteriaevolved separately from a progenote (Martinand Russell 2003) accept neomura as a cladeand that archaebacteria and eukaryotes can-not both be “primary domains.” Cavalier-Smith(1987c) listed 24 cellular and molecular char-acters shared by eubacteria and archaebacteriathat decisively refuted the progenote hypo-thesis, proving a direct transition between eu-bacteria and archaebacteria via a highly devel-oped cellular intermediate with lipid membraneand cytochrome-based membrane-dependentbioenergetics. Such complexity required reli-able translation and DNA genomes with effi-cient replication, repair, and transcription. Sub-sequently, many others also proved the lastcommon ancestor of life must have been a com-plex bacterium with well-developed membranebioenergetics, growth, and division machinery(e.g., Jekely 2006; Lombard et al. 2012b), re-futing the entirely untenable idea of separateprecellular origins of neomura and eubacteria(Martin and Russell 2003).

Thiergart et al. (2012) tendentiously falselyattributed the branching of only 15 eukaryot-ic proteins with actinobacteria as the “level ofsupport” for the neomuran theory, seriouslymisrepresenting its predictions. In fact, their145 genes branching closest to archaebacteriastrongly support it; given equal evolutionaryrates and gene losses within euryachaeotes andfilarchaeotes, it predicts that an equal numbershould appear as sister to each subphylum, as-suming that they and eukaryotes diverged at vir-tually the same time; purely by chance, somewill group with one and some the other (Cava-lier-Smith 2007b). Because crenarchaeote pro-teins mostly evolve faster and are more oftenlost, one expects some bias toward euryarch-aeotes, exactly as found: 68 trees put eukaryotes

closer to filarachaeotes and 77 closer to eu-ryarchaeotes. If eukaryotes evolved substantiallymore recently than archaebacteria from anancestor in either archaebacterial subphylum,all 145 trees should put them within that onegroup. Unwittingly, Thiergart et al. (2012) sup-plied the strongest evidence yet that archaebac-teria cannot be substantially older than eukary-otes, contrary to Martin’s assumptions (Martinand Muller 1998; Martin and Russell 2003).Thiergart et al. (2012) misunderstood and mis-represented the neomuran theory by assert-ing that it originally postulated that eukaryotesevolved from cyanobacteria, citing Cavalier-Smith (1975). That paper, predating archaebac-terial recognition (Woese and Fox 1977) did notpropose the neomuran idea.

Very detailed explanations of stepwise ori-gins of endomembranes, nucleus, mitosis, cellcycle, and sex (both mechanisms and selectiveadvantages) in earlier publications (Cavalier-Smith 2009, 2010c), including detailed criti-cisms of less plausible alternatives, cannot beadequately summarized here. The rest of thisarticle outlines new insights into bacterial cellevolution that sharpen distinctions between Ne-gibacteria and Posibacteria, making me con-clude that Negibacteria and neomura bothevolved from Posibacteria, the first cells. Thisis necessary to explain how bacterial phylogenymaps onto the fossil record, the only directevidence for evolutionary timing, and therebyperhaps more convincingly than before defendthe well-substantiated recency of archaebacteriacompared with eubacteria, and simultaneousorigin of archaebacteria and eukaryotes froma radically transformed actinobacteria-relatedposibacterium.

EVOLUTIONARY SIGNIFICANCE OF THENEGIBACTERIAL CELL ENVELOPE

The names “Negibacteria” and “Posibacteria,”coined to distinguish eubacteria with double orsingle bounding membranes (Cavalier-Smith1986, 1987b,c), are phylogenetically more pre-cise than the older “Gram-negative” and “Gram-positive” that were evolutionarily confusing be-cause some Posibacteria stain Gram-negatively

T. Cavalier-Smith




(e.g., mycoplasmas) and some negibacteria (e.g.,Deinococcus) stain Gram-positively. I arguedthat membrane number and biogenesis werefundamentally more important than staining,that OM evolved only once, and was lost atmost once since life began. I argue here it wasnever lost. My key distinction was ignored fordecades, partly because Woese’s repeated histor-ically inaccurate, naıve tirades against morphol-ogy (Woese 1994) led many biochemists to ig-nore bacterial envelope biology. Confusionscaused by such myopia are clearing; renaissanceof molecular cell ultrastructural anatomy andmembrane biogenesis confirms the fundamen-tal negibacterial/posibacterial dichotomy (ex-cellent reviews: Desvaux et al. 2009; Sutcliffe2010). Gupta (1998a,b,c) made the same dis-tinction much later, inventing informal terms“diderm” and “monoderm,” but astoundinglywrongly asserted that their fundamental phy-logenetic distinction was not recognized before1998 (Gupta 2011). OMs are simple phospho-lipid bilayers in Hadobacteria (Deinococcus,Thermus) and Thermotogales, which I thereforenow place in that phylum, but in all other Ne-gibacteria (as here revised by excluding Chloro-flexi), only its inner leaflet is phospholipid (itsouter is lipopolysaccharide).

The inner membrane (CM) of negibacteriais homologous with CM of unibacteria in bear-ing SecYEG channels, SRP receptors, TAT, ABCtransporters, type IV secretion, and membrane-bound respiratory and/or photosynthetic elec-tron transport chains, supporting proton gra-dients and proton-driven phosphorylation bypartially membrane-embedded ATP synthetas-es. Contrastingly, negibacterial OMs lack allthose and are penetrated by large cylindricalproteinaceous pores (porins), allowing freetraverse of small molecules, and have b-barrelproteins, not a-helical globular proteins of theCM and eukaryotic plasma membrane, endo-membranes, and peroxisomes. OM b-barrelproteins move across the CM via Sec or ABCcarriers, then insert into OM by using the b-barrel OM protein Omp85 (attached to the ABCtransporter by a “membrane fusion” proteinconnecting CM and OM). The complexity andconservation of this b-barrel insertion machin-

ery mean that negibacterial OMs had a singleorigin and were inherited by membrane hered-ity (Cavalier-Smith 2004) for .2.5 billion years,being kept throughout mitochondrial and chlo-roplast evolution (Cavalier-Smith 2013b). Ne-gibacteria evolved four other secretion mecha-nisms absent in Unibacteria: types II, V, VI, andVII using different OM porins.

The negibacterial cell plan is outstanding-ly stable. If the universal tree is rooted in Uni-bacteria (Fig. 1), the first cell had one mem-brane, and OMs were never lost in the historyof life. If the first cell were negibacterial (Blo-bel 1980), the OM arose with the first cell (forpossible mechanisms, see Cavalier-Smith 2001)and must have been lost by Posibacteria, forwhich the only plausible mechanism proposedis hypertrophy of murein, thickening the wallenough to break mechanically all contacts be-tween the CM and OM, preventing OM bio-genesis (Cavalier-Smith 1980). No theory ofnegibacterial eukaryote origins can be taken se-riously without a physically plausible mecha-nism of OM loss, sufficient reason for firmlyrejecting evolution of a phagotrophic wall-freeeukaryote from a cyanobacterium (Cavalier-Smith 1975), d-proteobacterium (Lopez-Garcıaand Moreira 1999), or planctobacterium (sim-ilarities between eukaryotes and planctobacte-ria are superficial misinterpretations: Cavalier-Smith 2010c; Santarella-Mellwig et al. 2013):Planctomycetes independently lost murein andFtsZ and some acquired tubulins and a kinesinby LGT. Elsewhere, I explained other fatal flawsof the d-proteobacterium fusion theory (Cava-lier-Smith 2010c).

The unrelated extra membrane in the cren-archaeote Ignicoccus without Omp85 ma-chinery (Nather and Rachel 2004) is a trivialconvergence irrelevant to the monophyly anduniqueness of negibacteria, contrary to super-ficial assertions that it invalidates the funda-mental importance, evolutionary stability, anduniqueness of negibacterial architecture, andcomplex biogenesis. More important excep-tions to the usual posibacterial body plan areMycobacteria, actinobacteria that probably con-vergently with Negibacteria evolved an OM.Mycobacteria have a very thick murein wall, fur-





nished with posibacterial sortases for lipopro-tein secretion like related teichobacteria (Cava-lier-Smith 2006a); their OM is a lipid bilayer,not phospholipid, but outer leaflet glycolipidand inner mycolic acid linked by arabinoga-lactan to underlying peptidoglycan (Silhavyet al. 2010). Presumably, it originated by secre-tion across the wall like the waterproofing lipidcuticles of land plants and insects. The key bio-physical problem in its origin was permeabi-lizing it to hydrophilic molecules: by b-barrelporins (Song et al. 2008)—but did these origi-nate independently of negibacterial porins orvia LGT from them? I predict that Omp85 in-sertion machinery is absent and consider themPosibacteria.

Theories of negibacterial origin generatetheir OM contrastingly: from endospore-form-ing posibacteria (Endobacteria) by prespore en-gulfment by its sister cell (Dawes 1981); by fold-ing up an inside-outcell with external genes andribosomes to form the first cell (Blobel 1980).They give radically different answers to thefundamental question: Did posibacteria pre-cede negibacteria (Fig. 1), or vice versa (Blobel1980)? Neomura remain a clade in either view.Because Blobel’s idea offered a way of evolvingcells unhampered by phospholipid bilayer im-permeability to nucleotides and amino acids,I initially preferred it (Cavalier-Smith 1987b,2001). However, that problem is now circum-vented because FAs probably originally wereshorter than today (Budin et al. 2009, 2012),making a single membrane more plausible forthe first cell because it avoids Blobel’s sealing-off gastruloid phase, which the OM biogenesiscomplexity now revealed makes unlikely. Recentendobacterial discoveries also support Dawes’sidea, although some (Errington 2013) overlookhis priority. Overall, a unibacterial root for pro-karyotes now seems mechanistically simpler forenvelope evolution.

GREEN BACTERIA AND THE ORIGINOF LIFE

Chloroflexi (chlorobacteria: non-sulfur greenbacteria plus non-photosynthetic relatives) arenot Negibacteria, despite their Gram-negative

staining and some electron micrographs sug-gesting an OM, but divergent Posibacteria (Sut-cliffe 2011). A long-running mistake consideredthem negibacteria (Cavalier-Smith 1987b,c,1991a,b, 1992, 2002b, 2006a,c, 2010a). Theirearly 16S rDNA grouping with Hadobacteria(oddly still believed by Woese 2013) seems tohave been a misleading long-branch problem.A 31-protein tree (Ward et al. 2009) groupsthem with Teichobacteria and Actinobacteria(Fig. 3). Transfer of Chloroflexi to Posibacteriamakes Negibacteria coterminous with Omp85OM biogenesis machinery, more sharply defin-ing the subkingdom. Accepting Chloroflexi asPosibacteria removes the polarization wronglyplacing them at the base of Negibacteria (Cava-lier-Smith 2006c) but does not alter earlier ar-guments for their being the most primitive of allbacteria.

Indeed, a novel aerobic green photohetero-trophic bacterium Candidatus Chloracidobac-terium (Bryant et al. 2007), a Negibacteriumin deep-branching class Acidobacteria of Pro-teobacteria sensu (Cavalier-Smith 2002b), re-inforces the view that green bacteria were at theroot of the universal tree (Cavalier-Smith 1985,1991a,b, 1992). Because Chloracidobacteriumhas chlorosomes and a type I photosyntheticreaction center, like Chloroflexi and Chloro-bi (green sulfur bacteria: negibacteria) (GarciaCostas et al. 2011; Tsukatani et al. 2012), andAcidobacteria are sisters to purple proteobac-teria and their colorless relatives (Ward et al.2009), the lineage connecting negibacteria andposibacteria clearly had chlorosomes and type Ireaction centers, as also did that joining Chlor-obi to the base of Proteobacteria, assuming ver-tical inheritance of photosynthesis. Congruenceof type I reaction-center (Bryant et al. 2007) and31-protein (Ward et al. 2009) trees contradictsinvoking photosynthetic machinery LGT to ne-gate that.

A universal root in Negibacteria, which aflagellar polarization argument favored (Ca-valier-Smith 2006c), would most plausibly bewithin Gracilicutes between Chlorobi andChloracidobacterium. Were it in Posibacteriaas I now prefer (Fig. 3) and some indels suggest(Skophammer et al. 2007), it is probably be-

T. Cavalier-Smith




tween Chloroflexi and Teichobacteria (whichinclude the green chlorosome-lacking photo-heterotrophic Heliobacteria as well as hetero-trophic Bacillales and Clostridiales), and LGTfrom gracilicute Negibacteria presumably gavePosibacteria flagella. Thus, ancestral cells wereprobably photosynthetic green bacteria, wheth-er the universal root is in Posibacteria or Negi-bacteria. Unless this root is within Gracilicutes,the unique purple bacterial photosynthetic ma-chinery must have green bacterial ancestry, asprobably do cyanobacterial photosystems. The31-protein tree and posibacterial root togetherfit conclusions from photosystem evolutionthat heliobacterial photosynthesis is the mostprimitive, followed by Chloroflexi, then Cyano-bacteria, gracilicute photosynthesis being morederived (Gupta 2012). If cell lineages divergedas in Figure 3, gene duplications probably cre-ated photosystem II (Sousa et al. 2013) in thecommon ancestor of Cyanobacteria and Graci-licutes, system II being lost by Chlorobi andChloracidobacterium (and type I reaction cen-ters by Proteobacteria); galactolipid synthesisfor photosynthetic membranes probably origi-nated in green posibacteria (monogalactosyl-diacylglycerol synthase homologs are conservedthroughout posibacteria) (Yuzawa et al. 2012).Multiple losses during photosynthetic diver-sification were probably more and LGT less fre-quent than many assume. Whether the first cellwas a Heliobacterium-like posibacterium (mostlikely) or Chlorobium-like Negibacterium, itlacked flagella and RuBisCo.

INTEGRATING MULTIGENE SEQUENCETREES AND FOSSIL DATA FOR REAL TIMINGOF CELL MEGAEVOLUTION

The eubacterial radiation is extremely bush-likewith very little basal resolution even on multi-gene trees (Battistuzzi et al. 2004; Ciccarelliet al. 2006). The 25-protein tree of Battistuzziet al. (2004) showed clear bipartition betweenGracilicutes (phyla Spirochaetae, Proteobacte-ria, Sphingobacteria, Planctobacteria, includingthe monophyly of each except Planctobacteria)and all other bacteria, but incorrectly placedcyanobacteria and Deinococci (with Thermus

constituting the phylum Hadobacteria) with-in Posibacteria, suggesting that they might bemore closely related to Posibacteria than areGracilicutes. The 31-protein tree of Ward et al.(2009) groups Hadobacteria (bootstrap sup-port 100% for their monophyly) and Thermo-toga/Aquifex as sisters with 69% support, butwrongly places Cyanobacteria and Fusobac-teria deeply within Posibacteria (insignificant12% support); that tree showed monophyly ofGracilicutes (81%, support), 97% support formonophyly of Planctobacteria, 87% for Proteo-bacteria sensu Cavalier-Smith (2002b) (includ-ing Acidobacteria), and 100% for Spirochaetaeand for Sphingobacteria (Chlorobi þ Bacterio-detes and relatives), and for the latter two beingsisters. From cell evolution perspectives, thatmultigene tree is the best I know for Eubacteria.

To eliminate potential LGT problems, Bap-teste et al. (2008) selected 22 compatible mark-ers from 31, unfortunately with narrower taxo-nomic sampling and all branching with ,50%support unwisely collapsed; they found justfour eubacterial clades: Gracilicutes (72%), En-dobacteria (lowGC Posibacteria: 58%), bothwith major anaerobic and aerobic lineages,and two aerobic clades, each with 100% support(Cyanobacteria, Actinobacteria). Higher sup-port for aerobic clades is because they branchless deeply than mixed aerobic/anaerobicclades, with a longish bare stem. That contrasthas a simple evolutionary explanation, their re-spective radiation timings: crown cyanobacteriaarguably radiated near the time of the great oxy-genation event they caused (2.4 Gy ago), where-as actinobacteria (with proportionately longerbare stem) radiated somewhat later, after oxy-genation enabled their aerobic heterotrophy. Incontrast, Gracilicutes and Endobacteria, withanoxygenic photosynthesizers and numerousanaerobic heterotrophs or chemotrophs, radiat-ed earlier, probably .2.7 Gy ago, but ,3.5 Gyago. Stem lineages of both aerobic groups wereprobably anaerobes; I suggest anaerobic greenbacteria for cyanobacteria and anaerobic het-erotrophic endobacteria for actinobacteria. My-coplasmas are sisters to Bacillales within Teicho-bacteria, confirming their murein loss; theyhave longer branches than any other Posibacte-





ria, showing that wall loss and/or genome min-iaturization somewhat released their proteinsfrom stabilizing selection, analogously to theelevated evolutionary rates of neomura. Evolu-tionary rates in Bapteste et al. (2008) are similarwithin other major clades, slightly greater inGracilicutes than Cyanobacteria or Posibacteria(Posibacteria apparently a weakly supportedclade collapsed into two during figure prepara-tion). Applying a crude evolutionary clock toeubacteria indicates that cyanobacteria initiallyradiated almost contemporaneously with pur-ple bacteria. Although backward-extrapolatingclock estimates are problematic because theydo not allow for quantum evolution that caninflate ages (Cavalier-Smith 2002b) or for sys-tematic differences between rates across do-mains and phyla, Feng et al. (1997) estimated2.1–2.5 Gy ago for the date of the Posibacteria/Cyanobacteria/other Negibacteria radiation,in surprisingly close harmony with the robustpalaeontological date of �2.4 Gy ago for thegreat atmospheric oxidation event (Schirrmeis-ter et al. 2013).

Eukaryotes cannot have originated signi-ficantly before a-proteobacteria, which mustbe younger than purple bacteria generally. Thea-proteobacterial radiation on the 22-gene treeis about half as old as for purple bacteria col-lectively. That implied age of �1.25 Gy agomust be somewhat too low because of limitedtaxon sampling. From the more taxon-rich 191-genome tree of Ciccarelli et al. (2006), the sameargument gives a maximal eukaryote age of�1.8 Gy ago. Given the resolution within jointmitochondrial/a-proteobacterial gene trees(Derelle and Lang 2012; Zhao et al. 2013), mi-tochondria probably did not arise from the firsta-proteobacteria, but are probably ,80% asold as a-proteobacteria, perhaps younger still.That gives �1.44 Gy ago as a reasonable upperlimit for eukaryote age, close to the oldest fos-sils (1.45 Gy ago) that several palaeontologistsregard as eukaryotic (Javaux et al. 2001), al-though I consider them more likely all misinter-preted bacteria (Cavalier-Smith 2006a). Inde-pendent estimates of minimum eukaryote agefrom the morphological fossil record and eukar-yote multigene trees (Cavalier-Smith 2013a) of

�1.2 + 0.25 Gy are congruent and surprisinglyclose. These estimates bracket the origin ofeukaryotes between 0.95 and 1.45 Gy ago, mak-ing them 2.4–3.7� younger than stem eubac-teria (�3.5 Gy), although possibly no more thantwice as young as the probably explosive crowneubacterial radiation (here estimated as �2.7 Gyago). The archaebacteria branch is far longerthan all eubacterial clades: more than threetimes as long as that of most eubacteria andabout five times longer than cyanobacteria. Be-cause cyanobacteria are .2.4 Gy old (Schirr-meister et al. 2013), more likely �2.7 Gy (Bosaket al. 2009, 2012; Petroff et al. 2010; Bosak et al.2012), that date objectively calibrates the meanlength of crown cyanobacterial branches; be-cause the stem connecting the base of crown eu-bacteria to the base of crown archaebacteria is justover five times longer, it would represent 12 Gyof evolution (conservatively assuming 2.4 Gy:2.6� the 4.6 Gy age of the Earth and 3.4� theage of all life: 3.5 Gy) if these 22 proteins werethen evolving at the same rate as in cyanobacte-ria for the past 2.4 Gy. Likewise, crown Archae-bacterial branches are substantially longer thancrown cyanobacterial ones and would represent�6.4 Gy of evolution; stem plus crown wouldrepresent 18 Gy, greater than the 13 Gy age ofthe universe. From these necessarily underesti-mates, the logically inescapable conclusion isthat these proteins evolved substantially fasterin crown eukaryotes and crown archaebacteriathan in eubacteria, and that neomuran stemlengths on most sequence trees are hugely in-flated by quantum evolution, causing false con-clusions about timing if that is overlooked (e.g.,Doolittle 1995).

That gross rate inequality invalidates mid-point rooting of prokaryotic distance networks(using subsets of gene families from 191 ge-nomes) by Dagan et al. (2010). Their favoritenetwork (Fig. 5) shows the archaebacteria stemmuch longer than the whole cyanobacterialbranch and stem plus crown �3� longer, sotheir statistical argument for uniform evolu-tionary rates is fallacious and their rooting nobetter than “reading chicken entrails,” to use theapposite phrase introduced by the senior author(Graur and Martin 2004). Dagan et al. falsely

T. Cavalier-Smith




claim that I “dismiss molecular data” “whereconvenient” by invoking quantum evolution.On the contrary, I extensively used moleculardata combined with objective fossil evidence ontiming, which proved the reality of quantumevolution in many molecules and disprovedtheir gradualist assumptions. They selectivelyinterpret fossil evidence that provides no hardevidence that archaebacteria are older than1.5 Gy, notwithstanding earlier carbon isotoperatios overconfidently interpreted as evidencefor archaebacterial methanogenesis (Cavalier-Smith 2006a). Their Figure 4C net for thegene family tranche conserved 35%–40% clear-ly bipartitions Negibacteria from Unibacteria.

Compared with the sophisticated multigenetrees discussed above, the three-domain single-gene rDNA trees shown by Pace (2009), Woese(2013), and Fournier and Gogarten (2010) arearchaic, of only historical interest representinga pioneering but primitive three-decades-oldphase of phylogenetics, suitable only for Pace’sT-shirts. In all, and even the technically bet-ter but taxonomically undersampled concaten-ated rDNA trees of Williams et al. (2012), thebranching within eukaryotes is certainly sub-stantially wrong ( judged by numerous multi-gene trees based on many scores of proteins),that within eubacteria completely unresolvedbush, and assumptions of rooting probably er-roneous, naıvely relying on probably artifactualparalog duplicate trees. Referring to them as theuniversal tree of life (Pace 2009) is more con-fusing than enlightening.

The trees based on 34 universal ribosomalproteins in Lasek and Gogarten (2013) are muchbetter in emphasizing a two-domain structure(eubacteria versus neomura; neomura share 30new proteins entirely absent in the ribosomallysimpler and older eubacteria) and having analmost correct branching order within eukary-otes (their root constantly between Euglenozoaand neokaryotes). They have no basal resolutionwithin eubacteria, as expected if the universalroot is among them followed by extremely rapidradiation and or saturation effects. Williams etal. (2012, 2013, 2014), though accepting that thethree primary domains idea is wrong (they cor-rectly treat eukaryotes as derived, not primary)

and that neomura are a clade (oddly withoutusing that name), still argue for two primarydomains, mistakenly viewing eubacteria as aclade and regarding eubacteria and archaebac-teria both as primary domains. I suspect thatwidespread failure to accept eubacteria aloneas the single primary domain of life, as I haveconsistently argued since 1987, owes more to themistaken dogmas briefly discussed below thanto the evidence, which is admittedly hard formany specialists to evaluate as it depends oncritical integration of palaeontology and neon-tology, both phylogenetics and molecular andespecially cell biology of both bacteria and eu-karyotes, as well as critical judgments aboutthe mechanistic and selective plausibility of thevarious transitions theoretically possible amongthe five major cell types shown in Figure 1.

METHODOLOGICAL BIASES ANDMOLECULAR CHRONOMETRIC MYTHS

It is important to note that even if there is noresolution whatever in input data for sequencetrees, as would be the case if (as I argue) eukary-otes, filarchaeotes, and euryarchaeotes divergedessentially simultaneously, phylogenetic algo-rithms can be biased to introduce apparent-ly strong support for meaningless conclusions(Yang 2007). Too often sequence trees are treat-ed as factual data, rather than indirect inferencesbased on oversimplified evolutionary assump-tions, because they always inevitably are, andgiven excessive emphasis compared with otherevolutionary evidence, morphological, cell bio-logical and palaeontological; all kinds of evi-dence must be treated critically and integratedif we are to solve such difficult historical prob-lems.

The very long eukaryotic stems in the 29-protein trees of Williams et al. (2012), stronglydependent on ribosome-associated proteins,are attributable substantially to exceptionallyrapid quantum evolution consequential on theorigin of the nucleus, as explained previously(Cavalier-Smith 2002b); as also explained there,similar quantum evolution during the neo-muran revolution in SRP cotranslational se-cretion may partially explain the comparably





long stem separating eubacteria and Neomura,whose length (whatever its causes) arguably vir-tually ensures a Shavit et al. (2007) type of arti-fact, failing to group neomura with the single(!) included posibacterium. Neither of theselong stems accurately reflects long evolutionarytimes, but rapid quantum evolutionary stretch-ing; proteins chosen for many three-domaintrees are the least clock-like of all and are biasedtoward the minority that clearly differentiatearchaebacteria and eubacteria (Cavalier-Smith2002b). The two posibacterial lineages (Actino-bacteria, Chloroflexi) that are closest to Neo-mura on classical rRNA trees (Woese 2013)and closer than any sampled were not even in-cluded by Williams et al. (2012), and the latter(the closest) was absent from Fournier and Go-garten (2010), making both poor tests of neo-muran relationships.

One need only look at the wildly high evo-lutionary rates of 16S rDNA in microsporidiathat led to their being misconstrued as the ear-liest eukaryotes (Vossbrinck et al. 1987), whenthey really are advanced derivatives of Choano-zoa of comparable recency to fungi (Cavalier-Smith 2013a; James et al. 2013), or of chloro-plast 16S rDNA in dinoflagellates (Zhang et al.2000), to appreciate the absurdity of the claimfor small-subunit ribosomal rRNA as a mo-lecular chronometer, or the falseness of the as-sumption that big changes in its structure mustnecessarily be ancient (Woese and Fox 1977).The whole Woesian “three-domain” perspec-tive was biased and fundamentally misled byhis profoundly mistaken assertion that rRNAis a “molecular chronometer”; the precise op-posite is true, rRNA being possibly the leastclock-like of all molecules in evolutionary rates.The extraordinarily elongated branch lengthsfor Microsporidia (James et al. 2013) and foran independently ultrarapidly evolving haplo-sporidian, the rhizarian parasite Mikrocytosmackini (Burki et al. 2013), on multigene treesemphasize that no proteins universally evolve ina uniform clock-like manner. One must use fos-sils and genuine atomic decay clocks in igneousrocks for objective timing, as Woese never did.

The foundation of the three-domain dogmais excessively narrow, depending mainly (not

quite exclusively) on ribosome-related charac-ters subject to temporally misleading quantum-evolutionary biases during the neomuran re-volution and eukaryogenesis (Cavalier-Smith1981, 1987c, 2002b). In a survey of 80 univer-sally conserved proteins, Harris et al. (2003)found that of the 50 whose trees agreed withthat of rDNA in showing archaebacteria andeukaryotes as sisters, all except two transcrip-tional and four DNA replication/repair geneswere ribosome-associated; none of the 30 pro-teins that disagreed with their rRNA tree wereribosome-associated. They unwisely dismissedall 28 of those showing phylogenetically in-termixed eubacteria and neomura from seri-ous consideration as likely to have undergoneLGT at some stage, downplaying and almostignoring the alternative reasonable explanationthat metabolic genes generally did not undergoquantum evolution during the neomuran revo-lution, and therefore lack the stretched stem thatthrough grossly violating molecular-clock as-sumptions greatly enhances ribosome-relatedsequence tree evidence for the monophyly ofneomura (see Cavalier-Smith 2002b). In fact,both their and my explanations of this impor-tant, sub-carpet-swept, incongruence betweenmetabolic and ribosomal genes are probablycorrect for different branches; I suspect thatinherently poor resolution caused by lack ofresolution-enhancing quantum evolution (com-bined with archaebacteria actually being youn-ger than their eubacterial ancestors) is the majorreason and LGT the minor one.

Two more open-minded devotees of theWoesian perspective (Doolittle and Zhaxy-bayeva 2013) emphasize that “a much largernumber [than the few informational genesthat differentiate archaebacteria and eubacte-ria], many hundreds, of ‘operational’ genesmake up a shared resource that is common toBacteria and Archaea” and that all prokaryoteshave “a vast shared pool of genes encoding di-verse metabolic functions seldom if ever usedby eukaryotes.” They candidly admit that for“metabolism, regulation, population genetics,and ecology. . .without knowing in advance itwould, in many instances, be hard to say wheth-er a particular published paper on these topics

T. Cavalier-Smith




had a bacterium or an archaeon as its studyobject.” Differences between archaebacteriaand eubacteria have been grossly exaggerated.Doolittle and Zhaxybayeva (2013) cogently ar-gue that the notion of a fundamental dichot-omy between ancestral prokaryotes and derivedeukaryotes in cell organization remains perfect-ly valid and evolutionarily sound, despite over-confident and superficial propaganda against itfrom archaeaologists, mainly Pace. They alsohighlight the tacit refusal by Woese and Paceto recognize neomura as a clade, a failure thatDoolittle and Zhaxybayeva (2013) emphasize canbe understood only in sociological, rhetorical,or propagandistic, not scientific terms, becauseit would entail explicitly admitting the un-doubted fact that the idea of three primary do-mains was simply wrong. (So is the idea of twoprimary domains, which Doolittle and Zhaxy-bayeva still espouse.) They equally overlookedthe key concept of Negibacteria; neither Woese(2013) nor Pace (2009) (nor even Doolittle andZhaxybayeva) mentions the two-membranecharacter of their envelope, vital for understand-ing bacterial evolution and eukaryogenesis.

As far as I know, Woese and Pace never citedany of my papers or attempted to refute theirarguments; under a tree in Cold Spring Harborin 1987, George Fox (coinventor of archae-bacteria) told me not to send my first neo-muran paper to Woese, despite complimentingits “many good arguments,” because “he wouldnot bother to read it.” Even his swansong(Woese 2013) restricted discussion of the ar-chaebacteria–eukaryote relationship to one ex-tremely superficial paragraph that merely(rightly) rejected the ideas of Hyman Hartmanand Martin and Russell (2003) (sadly, Woese’sown attempts at “explaining” cell evolution—latterly by vague invocation of rampant LGT—were devoid of significant cell biological contentor logic). His assertion that “Evolution does notproceed by suddenly and drastically altering agiven cell design (at least a fairly advanced one)”could hardly be more wrong for eukaryogenesis;but that prejudice simply explains why he wasunprepared even to contemplate the possibilityof a derived nature for archaebacteria, as someother influential and otherwise good scientists

sadly still are. Nonetheless, Woese (2013) greatlypleased me by advocating “bacteriology” not“microbiology” for the whole science of pro-karyotes (a name he also used without objec-tion, unlike Pace), and using “bacteria” in histext exactly as I long have to refer to all prokary-otes, including “archaea,” and reinstating both“eubacteria” and “eukaryotes” (but his figurecontradictorily used “bacteria” and pointless“eukarya”)—welcome partial recantation ofearlier nomenclatural unwisdom (Woese et al.1990). Likewise, his fundamental misinterpre-tations of the tree of life and the evolutionarysignificance of archaebacteria that so severelymisled a generation of researchers will notmaintain their distorting stranglehold over phy-logenetic thinking, if the rising generation ofvigorous, less-prejudiced younger researchersdemands proper emphasis on cell-biological,palaeontological, and 3D ultrastructural andcrystallographical evidence of molecular mor-phology (Jekely 2006, 2008; Valas and Bourne2009; Keeling 2013), in addition to the limitedpresently dominant one-dimensionality of se-quences, when reconstructing the history oflife. Morphology at all levels underlies evo-lution: electrical attractions and repulsions be-tween different shapes vivify cells, not mathe-matical abstractions like sequences.

Doolittle (2000), especially, has much exag-gerated the degree to which LGT confuses thetree of life, best seen as an organismal tree basedon predominantly vertically inherited cell line-ages (Fig. 1), not a multigene tree. Even in pro-karyotes, cell lineages devoid of anastomoses bysex or symbiogenesis were simply tree-like for 2billion years before eukaryotes.

Vertical inheritance of distinctive prokary-otic cell structures and integrated cell cycles wasthe phylogenetic framework within which LGTfiddled around with relatively restricted meta-bolic details, creating a genic but not organis-mal web: both by near-neutral replacements ofroughly equivalent enzymes catalyzing a stan-dard reaction and by truly novel acquisitionof adaptively valuable, phylogenetically dis-tant catalysts. But these gene-by-gene transfersdo not alter the fact that the basic differencesin cell structure between eubacterial phyla like





Cyanobacteria, Spirochaetae, Posibacteria, andProteobacteria have been stable for billions ofyears despite LGT, as has their distinctivenessfrom Archaebacteria for more than a billion,and can be used in interpreting bacterial or-ganismal megaevolution (Fig. 3) in much thesame way as we can reliably and meaningfullyfor eukaryotes despite the temporarily confus-ing effects of sex and symbiogenesis (althoughLGT, greater substitutional saturation over thelonger timescale, and paucity of ultrastructuralcomplexity do make it harder for the truly an-cient eubacteria than for the younger neomura).On the neomuran interpretation, the primaryphase of “progressive Darwinian evolution” thatimproved the basic machinery of life after cellsfirst evolved, as speculatively discussed by Doo-little and Zhaxybayeva (2013) and hordes ofothers (including me: Cavalier-Smith 1987b,2001) in the past three decades, took place notin the stem connecting neomura and eubacte-ria, as they and so many others wrongly assume,but much earlier in the stem eubacteria neces-sarily missing from sequence trees, during theearly Archaean period 3.5–2.7 Gy ago (whichdoes not mean the period when “Archaea”lived—they are purely Protoerozoic [mid andlate, not even early] and Phanerozoic bacteria).Some of these speculations are sensible if ap-plied to stem bacteria but totally misleading ifapplied to stem neomura, whose evolution wecan reliably infer by critical and multievidencialcomparative biology. Contrary to pervasive, of-ten dogmatic, misconceptions stemming pri-marily from Woese’s chronometric myth andaversion to accepting megaevolutionary changein advanced cells, the last universal commonancestor of life (LUCA) did not lie on the stemconnecting neomura and eubacteria, but withinthe eubacterial radiation.

The derived nature and recency of archae-bacteria have been long concealed by pervasivebiases in citing or interpreting contradictoryparalog duplicate trees, as explained previously(Cavalier-Smith 2002b, 2006c); numerous pa-ralog trees, which place the root within eubac-teria, as I consider correct, and thus contradictthe Woesian paradigm, are either ignored ordismissed by vaguely invoking LGT (Zhaxy-

bayeva et al. 2005), and those that agree, whichare probably dominated by misleading long-branch artifacts (Cavalier-Smith 2006c), are cit-ed as support! Trees that place Archaebacteriacloser to Posibacteria than to Negibacteria, inconcordance with neomuran theory, are toolightly dismissed as implying LGT (e.g., Fenget al. 1997). The seemingly objective but prob-ably artifactual rooting by the first paralog du-plicate trees (Gogarten et al. 1989; Iwabe et al.1989) eclipsed my still-cogent earlier argumentsfor a eubacterial root (Cavalier-Smith 1987c),and subsequent dogma and uncritical inertiaallowed Doolittle and Zhaxybayeva (2013) towrite that “most biologists today” would believeone of their Figures 2.3f–h, which all wrongly(as argued here and since 1987) show archae-bacteria as old as eubacteria. Contrary to thatwidely held assumption there is no historicallyconvincing evidence for archaebacteria being asold as Eubacteria; instead, their Figures 2.3b,d(elaborated in my Figs. 1 and 3) are probablycorrect for reasons summarized above. Anotherattempt at rooting the universal tree using ami-no acid composition biases depends on severalhighly questionable assumptions (Fournier andGogarten 2010); from my perspective, their datasimply indicate that there was a bigger change inribosomal protein composition during the neo-muran revolution than during either eukaryo-genesis or the origin of archaebacteria, and fallshort of proving where the root lies.

Two other biases prejudice bioinformaticsinterpretations: the prevalence of gene loss(underestimating it inflates LGT estimates andgrossly deflates estimates in the number of genespresent in the cenancestor of each domain); andthe prevalence in megaevolution like eukaryo-genesis of radical gene evolution beyond bio-informatics recognition, which led to hugelywrong conclusions about the relative contribu-tions of neomura and a-proteobacteria to theeukaryotic chimera (see Cavalier-Smith 2007b).

I must stress that if more convincing evi-dence than now exists were to appear and estab-lish that eukaryotes are really derived from ar-chaebacteria (Williams et al. 2012, 2013, 2014),contrary to my arguments that they are sisters,that would prove even more strongly that neo-

T. Cavalier-Smith




mura are a clade. But it would not provide anyevidence that archaebacteria are a primary do-main as old as eubacteria. Nor would it alter theessential cell biological or selective bases for themajor innovations in cell biology during eukar-yogenesis discussed here. It would just make amechanistically rather trivial change in the pre-cise phylogenetic origins of some of the precur-sor proteins for these dramatic megaevolution-ary changes, and introduce a slightly greatertime interval between the origin of eukaryotesand archaebacteria than I have assumed. Themost innovative core features of my cell evolu-tionary arguments for the secondary origins ofboth groups would remain valid.

CONCLUDING REMARKS

Understanding eukaryote origins requires bal-anced synthesis of cell biology, palaeontology,and phylogeny into a complete logically consis-tent picture, here outlined. Partial approachesfrom single perspectives spawned many over-simplified or unnecessarily complex ideas, oftenbiasing interpretations or obscuring the basicsimplicity of cell history. Especially difficulthas been determining the root of the tree forall life and eukaryotes (Figs. 1 and 3), essentialfor deducing phenotypes for early eukaryotesand their ancestors and sound reasoning abouteukaryogenesis. Errors ensure false interpreta-tions, hard to correct because tree-root posi-tions are often taken as “obviously true” wheninterpreting other data. We must be ready tochange conclusions radically if new evidenceor insights demand it.

REFERENCES

Adams DW, Errington J. 2009. Bacterial cell division: As-sembly, maintenance and disassembly of the Z ring. NatRev Microbiol 7: 642–653.

Aylett CH, Wang Q, Michie KA, Amos LA, Lowe J. 2010.Filament structure of bacterial tubulin homologue TubZ.Proc Natl Acad Sci 107: 19766–19771.

Aylett CH, Izore T, Amos LA, Lowe J. 2013. Structure of thetubulin/FtsZ-like protein TubZ from Pseudomonas bac-teriophage fKZ. J Mol Biol 425: 2164–2173.

Bapteste E, Susko E, Leigh J, Ruiz-Trillo I, Bucknam J, Doo-little WF. 2008. Alternative methods for concatenation ofcore genes indicate a lack of resolution in deep nodes ofthe prokaryotic phylogeny. Mol Biol Evol 25: 83–91.

Barber CF, Heuser T, Carbajal-Gonzalez BI, Botchkarev VVJr, Nicastro D. 2012. Three-dimensional structure of theradial spokes reveals heterogeneity and interactions withdyneins in Chlamydomonas flagella. Mol Biol Cell 23:111–120.

Barns SM, Delwiche CF, Palmer JD, Pace NR. 1996. Perspec-tives on archaeal diversity, thermophily and monophylyfrom environmental rRNA sequences. Proc Natl Acad Sci93: 9188–9193.

Battistuzzi FU, Feijao A, Hedges SB. 2004. A genomic time-scale of prokaryote evolution: Insights into the origin ofmethanogenesis, phototrophy, and the colonization ofland. BMC Evol Biol 4: 44.

Bernander R, Lundgren M, Ettema TJ. 2010. Comparativeand functional analysis of the archaeal cell cycle. CellCycle 9: 794–806.

Blobel G. 1980. Intracellular protein topogenesis. Proc NatlAcad Sci 77: 1496–1500.

Bosak T, Liang B, Sim MS, Petroff AP. 2009. Morphologicalrecord of oxygenic photosynthesis in conical stromato-lites. Proc Natl Acad Sci 106: 10939–10943.

Bosak T, Liang B, Wu TD, Templer SP, Evans A, Vali H,Guerquin-Kern JL, Klepac-Ceraj V, Sim MS, Mui J.2012. Cyanobacterial diversity and activity in modernconical microbialites. Geobiology 10: 384–401.

Bower R, Tritschler D, Vanderwaal K, Perrone CA, Mueller J,Fox L, Sale WS, Porter ME. 2013. The N-DRC forms aconserved biochemical complex that maintains outerdoublet alignment and limits microtubule sliding in mo-tile axonemes. Mol Biol Cell 24: 1134–1152.

Brochier-Armanet C, Boussau B, Gribaldo S, Forterre P.2008. Mesophilic Crenarchaeota: Proposal for a third ar-chaeal phylum, the Thaumarchaeota. Nat Rev Microbiol6: 245–252.

Brugerolle G. 1992. Flagellar apparatus duplication and par-tition, flagellar transformation during division in Ento-siphon sulcatum. BioSystems 28: 203–209.

Bryant DA, Costas AM, Maresca JA, Chew AG, Klatt CG,Bateson MM, Tallon LJ, Hostetler J, Nelson WC, Heidel-berg JF, et al. 2007. Candidatus Chloracidobacteriumthermophilum: An aerobic phototrophic acidobacte-rium. Science 317: 523–526.

Budin I, Bruckner RJ, Szostak JW. 2009. Formation of pro-tocell-like vesicles in a thermal diffusion column. J AmChem Soc 131: 9628–9629.

Budin I, Debnath A, Szostak JW. 2012. Concentration-driv-en growth of model protocell membranes. J Am Chem Soc134: 20812–20819.

Buisson J, Chenouard N, Lagache T, Blisnick T, Olivo-MarinJC, Bastin P. 2013. Intraflagellar transport proteins cyclebetween the flagellum and its base. J Cell Sci 126: 327–338.

Burki F, Corradi N, Sierra R, Pawlowski J, Meyer GR, AbbottCL, Keeling PJ. 2013. Phylogenomics of the intracellularparasite Mikrocytos mackini reveals evidence for a mito-some in Rhizaria. Curr Biol 23: 1541–1547.

Buske PJ, Levin PA. 2013. A flexible C-terminal linker isrequired for proper FtsZ assembly in vitro and cytokinet-ic ring formation in vivo. Mol Microbiol 89: 249–263.





Cabral G, Sans SS, Cowan CR, Dammermann A. 2013. Mul-tiple mechanisms contribute to centriole separation inC. elegans. Curr Biol 23: 1380–1387.

Carbajal-Gonzalez BI, Heuser T, Fu X, Lin J, Smith BW,Mitchell DR, Nicastro D. 2013. Conserved structural mo-tifs in the central pair complex of eukaryotic flagella.Cytoskeleton (Hoboken) 70: 101–120.

Carvalho-Santos Z, Machado P, Branco P, Tavares-Cadete F,Rodrigues-Martins A, Pereira-Leal JB, Bettencourt-DiasM. 2010. Stepwise evolution of the centriole-assemblypathway. J Cell Sci 123: 1414–1426.

Carvalho-Santos Z, Azimzadeh J, Pereira-Leal JB, Betten-court-Dias M. 2011. Evolution: Tracing the origins ofcentrioles, cilia, and flagella. J Cell Biol 194: 165–175.

Cavalier-Smith T. 1974. Basal body and flagellar develop-ment during the vegetative cell cycle and the sexual cycleof Chlamydomonas reinhardii. J Cell Sci 16: 529–556.

Cavalier-Smith T. 1975. The origin of nuclei and of eukary-otic cells. Nature 256: 463–468.

Cavalier-Smith T. 1978a. The evolutionary origin and phy-logeny of microtubules, mitotic spindles and eukaryoteflagella. BioSystems 10: 93–114.

Cavalier-Smith T. 1978b. Nuclear volume control by nucle-oskeletal DNA, selection for cell volume and cell growthrate, and the solution of the DNA C-value paradox. JCell Sci 34: 247–278.

Cavalier-Smith T. 1980. Cell compartmentation and the or-igin of eukaryote membranous organelles. In Endocytobi-ology: Endosymbiosis and cell biology, a synthesis of recentresearch (ed. Schwemmler W, Schenk HEA), pp. 893–916. de Gruyter, Berlin.

Cavalier-Smith T. 1981. The origin and early evolution ofthe eukaryotic cell. In Molecular and cellular aspects ofmicrobial evolution (ed. Carlile MJ, et al.), pp. 33–84.Cambridge University Press, Cambridge.

Cavalier-Smith T. 1982. The evolutionary origin and phy-logeny of eukaryote flagella. In prokaryotic and eukaryoticflagella 35th symposium of the society of experimental bi-ology (ed. Amos WB, Duckett JG), pp. 465–493. Cam-bridge University Press, Cambridge.

Cavalier-Smith T. 1983a. A 6-kingdom classification and aunified phylogeny. In Endocytobiology II (ed. Schwemm-ler W, Schenk HEA), pp. l027–l034. de Gruyter, Berlin.

Cavalier-Smith T. 1983b. Endosymbiotic origin of themitochondrial envelope. In Endocytobiology II (ed.Schwemmler W, Schenk HEA), pp. 265–279. de Gruyter,Berlin.

Cavalier-Smith T. 1985. Introduction: The evolutionary sig-nificance of genome size. In The evolution of genome size(ed. Cavalier-Smith T), pp. 1–36. Wiley, Chichester, UK.

Cavalier-Smith T. 1986. The kingdoms of organisms. Nature324: 416–417.

Cavalier-Smith T. 1987a. Bacterial DNA segregation: Its mo-tors and positional control. J Theor Biol 127: 361–372.

Cavalier-Smith T. 1987b. The origin of cells: A symbiosisbetween genes, catalysts, and membranes. Cold SpringHarb Symp Quant Biol 52: 805–824.

Cavalier-Smith T. 1987c. The origin of eukaryotic and ar-chaebacterial cells. Ann NY Acad Sci 503: 17–54.

Cavalier-Smith T. 1991a. The evolution of cells. In Evolutionof life (ed. Osawa S, Honjo T), pp. 271–304. Springer,Berlin.

Cavalier-Smith T. 1991b. The evolution of prokaryotic andeukaryotic cells. In Fundamentals of medical cell biology(ed. Bittar GE), Vol. I, pp. 217–272. JAI, Greenwich, CT.

Cavalier-Smith T. 1991c. Intron phylogeny: A new hypoth-esis. Trends Genet 7: 145–148.

Cavalier-Smith T. 1992. Origins of secondary metabolism.Ciba Found Symp 171: 64–80.

Cavalier-Smith T. 1993. Evolution of the eukaryotic ge-nome. In The eukaryotic genome (ed. Broda P, et al.),pp. 333–385. Cambridge University Press, Cambridge.

Cavalier-Smith T. 1997. Cell and genome coevolution: Fac-ultative anaerobiosis, glycosomes and kinetoplastan RNAediting. Trends Genet 13: 6–9.

Cavalier-Smith T. 1998. A revised six-kingdom system oflife. Biol Rev Camb Philos Soc 73: 203–266.

Cavalier-Smith T. 2000. Membrane heredity and early chlo-roplast evolution. Trends Plant Sci 5: 174–182.

Cavalier-Smith T. 2001. Obcells as proto-organisms: Mem-brane heredity, lithophosphorylation, and the origins ofthe genetic code, the first cells, and photosynthesis. J MolEvol 53: 555–595.

Cavalier-Smith T. 2002a. Meiosis. In The Oxford encyclopae-dia of evolution, pp. 700–708. Oxford University Press,New York.

Cavalier-Smith T. 2002b. The neomuran origin of archae-bacteria, the negibacterial root of the universal tree andbacterial megaclassification. Int J Syst Evol Microbiol 52:7–76.

Cavalier-Smith T. 2002c. The phagotrophic origin of eu-karyotes and phylogenetic classification of Protozoa. IntJ Syst Evol Microbiol 52: 297–354.

Cavalier-Smith T. 2004. The membranome and membraneheredity in development and evolution. In Organelles,genomes and eukaryote phylogeny (ed. Hirt RP, HornerDS), pp. 335–351. CRC, Boca Raton, FL.

Cavalier-Smith T. 2005. Economy, speed and size matter:Evolutionary forces driving nuclear genome miniaturi-zation and expansion. Ann Bot 95: 147–175.

Cavalier-Smith T. 2006a. Cell evolution and Earth history:Stasis and revolution. Philos Trans Roy Soc B 361: 969–1006.

Cavalier-Smith T. 2006b. Origin of mitochondria by intra-cellular enslavement of a photosynthetic purple bacte-rium. Proc Roy Soc B 273: 1943–1952.

Cavalier-Smith T. 2006c. Rooting the tree of life by transi-tion analysis. Biol Direct 1: 19.

Cavalier-Smith T. 2007a. Concept of a bacterium still validin prokaryote debate. Nature 446: 7257.

Cavalier-Smith T. 2007b. The chimaeric origin of mitochon-dria: Photosynthetic cell enslavement, gene-transfer pres-sure, and compartmentation efficiency. In Origin of mi-tochondria and hydrogenosomes (ed. Martin WF, MullerM), pp. 161–199. Springer, Berlin.

Cavalier-Smith T. 2009. Predation and eukaryote cell ori-gins: A coevolutionary perspective. Int J Biochem Cell Biol41: 307–322.

T. Cavalier-Smith




Cavalier-Smith T. 2010a. Deep phylogeny, ancestral groups,and the four ages of life. Philos Trans Roy Soc B 365: 111–132.

Cavalier-Smith T. 2010b. Kingdoms Protozoa and Chro-mista and the eozoan root of eukaryotes. Biol Lett 6:342–345.

Cavalier-Smith T. 2010c. Origin of the cell nucleus, mitosisand sex: Roles of intracellular coevolution. Biol Direct 5: 7.

Cavalier-Smith T. 2013a. Early evolution of eukaryote feed-ing modes, cell structural diversity, and classification ofthe protozoan phyla Loukozoa, Sulcozoa, and Choano-zoa. Eur J Protistol 49: 115–178.

Cavalier-Smith T. 2013b. Symbiogenesis: Mechanisms, evo-lutionary consequences, and systematic implications.Ann Rev Eco Evol Syst 44: 145–172.

Cavalier-Smith T, Chao EE. 2012. Oxnerella micra sp. n. (Ox-nerellidae fam. n.), a tiny naked centrohelid, and the di-versity and evolution of Heliozoa. Protist 163: 574–601.

Cavalier-Smith T, Karpov SA. 2012. Paracercomonas kinetidultrastructure, origins of the body plan of Cercomona-dida, and cytoskeleton evolution in Cercozoa. Protist163: 47–75.

Cavalier-Smith T, Oates B. 2012. Ultrastructure of Allapsavibrans and the body plan of Glissomonadida (Cerco-zoa). Protist 163: 165–187.

Cavalier-Smith T, Scoble JM. 2013. Phylogeny of Hetero-konta: Incisomonas marina, a uniciliate gliding opalo-zoan related to Solenicola (Nanomonadea), and evidencethat Actinophryida evolved from raphidophytes. Eur JProtistol 49: 328–353.

Chalkia D, Nikolaidis N, Makalowski W, Klein J, Nei M.2008. Origins and evolution of the formin multigenefamily that is involved in the formation of actin filaments.Mol Biol Evol 25: 2717–2733.

Ciccarelli FD, Doerks T, von Mering C, Creevey CJ, Snel B,Bork P. 2006. Toward automatic reconstruction of a high-ly resolved tree of life. Science 311: 1283–1287.

Csuros M, Miklos I. 2009. Streamlining and large ancestralgenomes in Archaea inferred with a phylogenetic birth-and-death model. Mol Biol Evol 26: 2087–2095.

Cubonova L, Sandman K, Hallam SJ, Delong EF, Reeve JN.2005. Histones in crenarchaea. J Bacteriol 187: 5482–5485.

Dagan T, Roettger M, Bryant D, Martin W. 2010. Genomenetworks root the tree of life between prokaryotic do-mains. Genome Biol Evol 2: 379–392.

Dawes IW. 1981. Sporulation in evolution. In Cellular andmolecular aspects of microbial evolution (ed. Carlile MJ,et al.), pp. 85–130. Cambridge University Press, Cam-bridge.

De Duve C, Wattiaux R. 1966. Functions of lysosomes. AnnRev Physiol 28: 435–492.

Derelle R, Lang BF. 2012. Rooting the eukaryotic tree withmitochondrial and bacterial proteins. Mol Biol Evol 29:1277–1289.

Desvaux M, Hebraud M, Talon R, Henderson IR. 2009.Secretion and subcellular localizations of bacterial pro-teins: A semantic awareness issue. Trends Microbiol 17:139–145.

Dippenaar-Schoeman AS, Jocque R. 1997. African spiders:An identification manual. Agricultural Research Council,Pretoria, South Africa.

Dominguez-Cuevas P, Mercier R, Leaver M, Kawai Y, Erring-ton J. 2012. The rod to L-form transition of Bacillus sub-tilis is limited by a requirement for the protoplast to es-cape from the cell wall sacculus. Mol Microbiol 83: 52–66.

Doolittle RF. 1995. The origins and evolution of eukaryoticproteins. Philos Trans R Soc Lond B 349: 235–240.

Doolittle WF. 2000. Uprooting the tree of life. Sci Am 282:90–95.

Doolittle WF, Zhaxybayeva O. 2013. What is a prokaryote?In The prokaryotes—Prokaryotic biology and symbiotic as-sociation (ed. Rosenberg E, et al.), pp. 21–37. Springer-Verlag, Berlin.

Egan AJ, Vollmer W. 2013. The physiology of bacterial celldivision. Ann NY Acad Sci 1277: 8–28.

Eichinger CS, Kurze A, Oliveira RA, Nasmyth K. 2013. Dis-engaging the Smc3/kleisin interface releases cohesinfrom Drosophila chromosomes during interphase andmitosis. EMBO J 32: 656–665.

Errington J. 2013. L-form bacteria, cell walls and the originsof life. Open Biol 3: 120143.

Ettema TJ, Bernander R. 2009. Cell division and the ESCRTcomplex: A surprise from the archaea. Commun IntegrBiol 2: 86–88.

Faini M, Beck R, Wieland FT, Briggs JA. 2013. Vesicle coats:Structure, function, and general principles of assembly.Trends Cell Biol 23: 279–288.

Fan S, Whiteman EL, Hurd TW, McIntyre JC, Dishinger JF,Liu CJ, Martens JR, Verhey KJ, Sajjan U, Margolis B. 2011.Induction of Ran GTP drives ciliogenesis. Mol Biol Cell22: 4539–4548.

Feng DF, Cho G, Doolittle RF. 1997. Determining divergencetimes with a protein clock: Update and reevaluation. ProcNatl Acad Sci 94: 13028–13033.

Fournier GP, Gogarten JP. 2010. Rooting the ribosomal treeof life. Mol Biol Evol 27: 1792–1801.

Gabaldon T. 2010. Peroxisome diversity and evolution.Philos Trans R Soc Lond B 365: 765–773.

Gao B, Gupta RS. 2005. Conserved indels in protein se-quences that are characteristic of the phylum Actinobac-teria. Int J Syst Evol Microbiol 55: 2401–2412.

Gao B, Gupta RS. 2012. Phylogenetic framework and mo-lecular signatures for the main clades of the phylum Ac-tinobacteria. Microbiol Mol Biol Rev 76: 66–112.

Gao B, Paramanathan R, Gupta RS. 2006. Signature proteinsthat are distinctive characteristics of Actinobacteria andtheir subgroups. Antonie Van Leeuwenhoek 90: 69–91.

Gao B, Sugiman-Marangos S, Junop MS, Gupta RS. 2009.Structural and phylogenetic analysis of a conservedactinobacteria-specific protein (ASP1; SCO1997) fromStreptomyces coelicolor. BMC Struct Biol 9: 40.

Garbarino JE, Gibbons IR. 2002. Expression and genomicanalysis of midasin, a novel and highly conserved AAAprotein distantly related to dynein. BMC Genomics 3: 18.

Garcia Costas AM, Tsukatani Y, Romberger SP, OostergetelGT, Boekema EJ, Golbeck JH, Bryant DA. 2011. Ultra-structural analysis and identification of envelope proteins





of “Candidatus Chloracidobacterium thermophilum”chlorosomes. J Bacteriol 193: 6701–6711.

Geimer S, Melkonian M. 2005. Centrin scaffold in Chlamy-domonas reinhardtii revealed by immunoelectron mi-croscopy. Eukaryot Cell 4: 1253–1263.

Gluenz E, Povelones ML, Englund PT, Gull K. 2011. Thekinetoplast duplication cycle in Trypanosoma brucei isorchestrated by cytoskeleton-mediated cell morphogen-esis. Mol Cell Biol 31: 1012–1021.

Gogarten JP, Kibak H, Dittrich P, Taiz L, Bowman EJ, Bow-man BJ, Manolson MF, Poole RJ, Date T, Oshima T, et al.1989. Evolution of the vacuolar ATPase: implications forthe origin of eukaryotes. Proc Natl Acad Sci 86: 6661–6665.

Gogendeau D, Hurbain I, Raposo G, Cohen J, Koll F, BastoR. 2011. Sas-4 proteins are required during basal bodyduplication in Paramecium. Mol Biol Cell 22: 1035–1044.

Grainge I, Lesterlin C, Sherratt DJ. 2011. Activation ofXerCD-dif recombination by the FtsK DNA translocase.Nucleic Acids Res 39: 5140–5148.

Graur D, Martin W. 2004. Reading the entrails of chickens:Molecular timescales of evolution and the illusion ofprecision. Trends Genet 20: 80–86.

Gray MW. 1992. The endosymbiont hypothesis revisited.Int Rev Cytol 141: 233–357.

Gray MW, Doolittle WF. 1982. Has the endosymbiont hy-pothesis been proven? Microbiol Rev 46: 1–42.

Gribaldo S, Poole M, Daubin V, Forterre P, Brochier-Arma-net C. 2010. The origin of eukaryotes and their relation-ship with the Archaea: Are we at a phylogenomic im-passe? Nat Rev Microbiol 8: 743–752.

Gruber S, Errington J. 2009. Recruitment of condensin toreplication origin regions by ParB/SpoOJ promotes chro-mosome segregation in B. subtilis. Cell 137: 685–696.

Gualdron-Lopez M, Brennand A, Hannaert V, Quinones W,Caceres AJ, Bringaud F, Concepcion JL, Michels PA. 2012.When, how and why glycolysis became compartmental-ised in the Kinetoplastea. A new look at an ancient or-ganelle. Int J Parasitol 42: 1–20.

Guichard P, Desfosses A, Maheshwari A, Hachet V, DietrichC, Brune A, Ishikawa T, Sachse C, Gonczy P. 2012. Cart-wheel architecture of Trichonympha basal body. Science337: 553.

Gupta RS. 1998a. Life’s third domain (Archaea): An estab-lished fact or an endangered paradigm? A new proposalfor classification of organisms based on protein sequenc-es and cell structure. Theor Popul Biol 54: 91–104.

Gupta RS. 1998b. Protein phylogenies and signature se-quences: A reappraisal of evolutionary relationshipsamong archaebacteria, eubacteria, and eukaryotes. Mi-crobiol Mol Biol Rev 62: 1435–1491.

Gupta RS. 1998c. What are archaebacteria: Life’s third do-main or monoderm prokaryotes related to Gram-posi-tive bacteria? A new proposal for the classification ofprokaryotic organisms. Mol Microbiol 29: 695–708.

Gupta RS. 2011. Origin of diderm (Gram-negative) bacte-ria: Antibiotic selection pressure rather than endosymbi-osis likely led to the evolution of bacterial cells with twomembranes. Antonie Van Leeuwenhoek 100: 171–182.

Gupta RS. 2012. Origin and spread of photosynthesis basedupon conserved sequence features in key bacteriochlor-

ophyll biosynthesis proteins. Mol Biol Evol 29: 3397–3412.

Guy L, Ettema TJ. 2011. The archaeal “TACK” superphylumand the origin of eukaryotes. Trends Microbiol 19: 580–587.

Harashima H, Dissmeyer N, Schnittger A. 2013. Cell cyclecontrol across the eukaryotic kingdom. Trends Cell Biol23: 345–356.

Harris JK, Kelley ST, Spiegelman GB, Pace NR. 2003. Thegenetic core of the universal ancestor. Genome Res 13:407–412.

Heiss AA, Walker G, Simpson AG. 2013. The flagellar appa-ratus of Breviata anathema, a eukaryote without a clearsupergroup affinity. Eur J Protistol 49: 354–372.

Heuser T, Raytchev M, Krell J, Porter ME, Nicastro D. 2009.The dynein regulatory complex is the nexin link and amajor regulatory node in cilia and flagella. J Cell Biol 187:921–933.

Heuser T, Barber CF, Lin J, Krell J, Rebesco M, Porter ME,Nicastro D. 2012. Cryoelectron tomography reveals dou-blet-specific structures and unique interactions in the I1dynein. Proc Natl Acad Sci 109: E2067–E2076.

Hobel CF, Albers SV, Driessen AJ, Lupas AN. 2008. TheSulfolobus solfataricus AAA protein Sso0909, a homo-logue of the eukaryotic ESCRT Vps4 ATPase. BiochemSoc Trans 36: 94–98.

Idei M, Osada K, Sato S, Nakayama T, Nagumo T, Mann DG.2013. Sperm ultrastructure in the diatoms Melosira andThalassiosira and the significance of the 9 þ 0 configu-ration. Protoplasma 250: 833–850.

Ikeda T. 2008. Parkin-co-regulated gene (PACRG) productinteracts with tubulin and microtubules. FEBS Lett 582:1413–1418.

Ikeda K, Brown JA, Yagi T, Norrander JM, Hirono M,Eccleston E, Kamiya R, Linck RW. 2003. Rib72, a con-served protein associated with the ribbon compartmentof flagellar A-microtubules and potentially involvedin the linkage between outer doublet microtubules. JBiol Chem 278: 7725–7734.

Ikeda K, Ikeda T, Morikawa K, Kamiya R. 2007. Axonemallocalization of Chlamydomonas PACRG, a homologue ofthe human Parkin-coregulated gene product. Cell MotilCytoskeleton 64: 814–821.

Iwabe N, Kuma K, Hasegawa M, Osawa S, Miyata T. 1989.Evolutionary relationship of archaebacteria, eubacteria,and eukaryotes inferred from phylogenetic trees of du-plicated genes. Proc Natl Acad Sci 86: 9355–9359.

James TY, Pelin A, Bonen L, Ahrendt S, Sain D, Corradi N,Stajich JE. 2013. Shared signatures of parasitism and phy-logenomics unite cryptomycota and microsporidia. CurrBiol 23: 1548–1553.

Javaux EJ, Knoll AH, Walter MR. 2001. Morphologicaland ecological complexity in early eukaryotic ecosystems.Nature 412: 66–69.

Jekely G. 2006. Did the last common ancestor have a bio-logical membrane? Biol Direct 1: 35.

Jekely G. 2008. Origin of the nucleus and Ran-dependenttransport to safeguard ribosome biogenesis in a chimericcell. Biol Direct 3: 31.

T. Cavalier-Smith




Jekely G, Arendt D. 2006. Evolution of intraflagellar trans-port from coated vesicles and autogenous origin of theeukaryotic cilium. BioEssays 28: 191–198.

Jerka-Dziadosz M, Gogendeau D, Klotz C, Cohen J, BeissonJ, Koll F. 2010. Basal body duplication in Paramecium:The key role of Bld10 in assembly and stability of thecartwheel. Cytoskeleton (Hoboken) 67: 161–171.

John P, Whatley FR. 1975. Paracoccus denitrificans and theevolutionary origin of mitochondria. Nature 254: 495–498.

John P, Whatley FR. 1977. Paracoccus denitrificans Davis(Micrococcus denitrificans Beijerinck) as a mitochondri-on. Adv Bot Res 4: 51–115.

Kaimer C, Gonzalez-Pastor JE, Graumann PL. 2009. SpoIIIEand a novel type of DNA translocase, SftA, couple chro-mosome segregation with cell division in Bacillus subtilis.Mol Microbiol 74: 810–825.

Kee HL, Dishinger JF, Blasius TL, Liu CJ, Margolis B, VerheyKJ. 2012. A size-exclusion permeability barrier and nu-cleoporins characterize a ciliary pore complex that regu-lates transport into cilia. Nat Cell Biol 14: 431–437.

Keeling PJ. 2013. The number, speed, and impact of plastidendosymbioses in eukaryotic evolution. Annu Rev PlantBiol 64: 583–607.

Kressler D, Hurt E, Bergler H, Bassler J. 2012. The power ofAAA-ATPases on the road of pre-60S ribosome matura-tion—Molecular machines that strip pre-ribosomal par-ticles. Biochim Biophys Acta 1823: 92–100.

Lake JA, Skophammer RG, Herbold CW, Servin JA. 2009.Genome beginnings: Rooting the tree of life. Philos TransR Soc Lond B 364: 2177–2185.

Lane N, Martin W. 2010. The energetics of genome com-plexity. Nature 467: 929–934.

Lasek-Nesselquist E, Gogarten JP. 2013. The effects of modelchoice and mitigating bias on the ribosomal tree of life.Mol Phylogenet Evol 69: 17–38.

Lechtreck KF, Brown JM, Sampaio JL, Craft JM, ShevchenkoA, Evans JE, Witman GB. 2013. Cycling of the signal-ing protein phospholipase D through cilia requires theBBSome only for the export phase. J Cell Biol 201: 249–261.

Li Z. 2012. Regulation of the cell division cycle in Trypano-soma brucei. Eukaryot Cell 11: 1180–1190.

Li S, Fernandez JJ, Marshall WF, Agard DA. 2012. Three-dimensional structure of basal body triplet revealed byelectron cryo-tomography. EMBO J 31: 552–562.

Lin J, Tritschler D, Song K, Barber CF, Cobb JS, Porter ME,Nicastro D. 2011. Building blocks of the nexin–dyneinregulatory complex in Chlamydomonas flagella. J BiolChem 286: 29175–29191.

Lin J, Heuser T, Song K, Fu X, Nicastro D. 2012. One of thenine doublet microtubules of eukaryotic flagella exhibitsunique and partially conserved structures. PLoS ONE7: e46494.

Livny J, Yamaichi Y, Waldor MK. 2007. Distribution of cen-tromere-like parS sites in bacteria: Insights from compar-ative genomics. J Bacteriol 189: 8693–8703.

Lombard J, Moreira D. 2011. Origins and early evolution ofthe mevalonate pathway of isoprenoid biosynthesis in thethree domains of life. Mol Biol Evol 28: 87–99.

Lombard J, Lopez-Garcıa P, Moreira D. 2012a. An ACP-in-dependent fatty acid synthesis pathway in archaea: Im-plications for the origin of phospholipids. Mol Biol Evol29: 3261–3265.

Lombard J, Lopez-Garcıa P, Moreira D. 2012b. The earlyevolution of lipid membranes and the three domains oflife. Nat Rev Microbiol 10: 507–515.

Lopez-Garcıa P, Moreira D. 1999. Metabolic symbiosis at theorigin of eukaryotes. Trends Biochem Sci 24: 88–93.

Lowe J, Amos LA. 2009. Evolution of cytomotive filaments:The cytoskeleton from prokaryotes to eukaryotes. Int JBiochem Cell Biol 41: 323–329.

Makarova KS, Yutin N, Bell SD, Koonin EV. 2010. Evolutionof diverse cell division and vesicle formation systems inArchaea. Nat Rev Microbiol 8: 731–741.

Marchandin H, Teyssier C, Campos J, Jean-Pierre H, RogerF, Gay B, Carlier JP, Jumas-Bilak E. 2010. Negativicoccussuccinicivorans gen. nov., sp. nov., isolated from humanclinical samples, emended description of the family Veil-lonellaceae and description of Negativicutes classis nov.,Selenomonadales ord. nov. and Acidaminococcaceae fam.nov. in the bacterial phylum Firmicutes. Int J Syst EvolMicrobiol 60: 1271–1279.

Margulis L. 1970. Origin of eukaryotic cells. Yale UniversityPress, New Haven.

Martin W, Muller M. 1998. The hydrogen hypothesis for thefirst eukaryote. Nature 392: 37–44.

Martin W, Russell MJ. 2003. On the origins of cells: A hy-pothesis for the evolutionary transitions from abioticgeochemistry to chemoautotrophic prokaryotes, andfrom prokaryotes to nucleated cells. Philos Trans R SocLond B 358: 59–85.

Matsuo K, Ohsumi K, Iwabuchi M, Kawamata T, Ono Y,Takahashi M. 2012. Kendrin is a novel substrate for sep-arase involved in the licensing of centriole duplication.Curr Biol 22: 915–921.

Maupin-Furlow JA. 2013. Ubiquitin-like proteins and theirroles in archaea. Trends Microbiol 21: 31–38.

Mizuno N, Taschner M, Engel BD, Lorentzen E. 2012. Struc-tural studies of ciliary components. J Mol Biol 422: 163–180.

Nasmyth K. 1995. Evolution of the cell cycle. Philos Trans RSoc Lond B 349: 271–281.

Nasmyth K. 2011. Cohesin: A catenase with separate entryand exit gates? Nat Cell Biol 13: 1170–1177.

Nather DJ, Rachel R. 2004. The outer membrane of thehyperthermophilic archaeon Ignicoccus: Dynamics, ul-trastructure and composition. Biochem Soc Trans 32:199–203.

Nicastro D, Fu X, Heuser T, Tso A, Porter ME, Linck RW.2011. Cryo-electron tomography reveals conserved fea-tures of doublet microtubules in flagella. Proc Natl AcadSci 108: E845–E853.

Norrander JM, deCathelineau AM, Brown JA, Porter ME,Linck RW. 2000. The Rib43a protein is associated withforming the specialized protofilament ribbons of flagellarmicrotubules in Chlamydomonas. Mol Biol Cell 11: 201–215.

Novak B, Csikasz-Nagy A, Gyorffy B, Nasmyth K, Tyson JJ.1998. Model scenarios for evolution of the eukaryotic cellcycle. Philos Trans R Soc Lond B 353: 2063–2076.





Nunoura T, Takaki Y, Kakuta J, Nishi S, Sugahara J, KazamaH, Chee GJ, Hattori M, Kanai A, Atomi H, et al. 2011.Insights into the evolution of Archaea and eukaryoticprotein modifier systems revealed by the genome of anovel archaeal group. Nucleic Acids Res 39: 3204–3223.

Oliveira RA, Nasmyth K. 2013. Cohesin cleavage is insuffi-cient for centriole disengagement in Drosophila. Curr Biol23: R601–R603.

Oliveira RA, Hamilton RS, Pauli A, Davis I, Nasmyth K.2010. Cohesin cleavage and Cdk inhibition trigger for-mation of daughter nuclei. Nat Cell Biol 12: 185–192.

Ouhammouch M, Hausner W, Geiduschek EP. 2009. TBPdomain symmetry in basal and activated archaeal tran-scription. Mol Microbiol 71: 123–131.

Pace NR. 2009. Mapping the tree of life: Progress and pros-pects. Microbiol Mol Biol Rev 73: 565–576.

Palmer T, Berks BC. 2012. The twin-arginine translocation(Tat) protein export pathway. Nat Rev Microbiol 10: 483–496.

Pan R, Hu J. 2011. The conserved fission complex on perox-isomes and mitochondria. Plant Signal Behav 6: 870–872.

Pearce MJ, Mintseris J, Ferreyra J, Gygi SP, Darwin KH. 2008.Ubiquitin-like protein involved in the proteasome path-way of Mycobacterium tuberculosis. Science 322: 1104–1107.

Pereto J, Lopez-Garcıa P, Moreira D. 2004. Ancestral lipidbiosynthesis and early membrane evolution. Trends Bio-chem Sci 29: 469–477.

Petroff AP, Sim MS, Maslov A, Krupenin M, Rothman DH,Bosak T. 2010. Biophysical basis for the geometry of con-ical stromatolites. Proc Natl Acad Sci 107: 9956–9961.

Pieuchot L, Jedd G. 2012. Peroxisome assembly and func-tional diversity in eukaryotic microorganisms. Annu RevMicrobiol 66: 237–263.

Pigino G, Geimer S, Lanzavecchia S, Paccagnini E, CanteleF, Diener DR, Rosenbaum JL, Lupetti P. 2009. Electron-tomographic analysis of intraflagellar transport particletrains in situ. J Cell Biol 187: 135–148.

Podar M, Makarova KS, Graham DE, Wolf YI, Koonin EV,Reysenbach AL. 2013. Insights into archaeal evolutionand symbiosis from the genomes of a nanoarchaeonand its inferred crenarchaeal host from Obsidian Pool,Yellowstone National Park. Biol Direct 8: 9.

Reeve JN, Bailey KA, Li WT, Marc F, Sandman K, Soares DJ.2004. Archaeal histones: structures, stability and DNAbinding. Biochem Soc Trans 32: 227–230.

Rucktaschel R, Halbach A, Girzalsky W, Rottensteiner H,Erdmann R. 2010. De novo synthesis of peroxisomesupon mitochondrial targeting of Pex3p. Eur J Cell Biol89: 947–954.

Samson RY, Obita T, Freund SM, Williams RL, Bell SD. 2008.A role for the ESCRT system in cell division in archaea.Science 322: 1710–1713.

Sandoval-Calderon M, Geiger O, Guan Z, Barona-Gomez F,Sohlenkamp C. 2009. A eukaryote-like cardiolipin syn-thase is present in Streptomyces coelicolor and in mostactinobacteria. J Biol Chem 284: 17383–17390.

Santarella-Mellwig R, Pruggnaller S, Roos N, Mattaj IW,Devos DP. 2013. Three-dimensional reconstruction ofbacteria with a complex endomembrane system. PLoSBiol 11: e1001565.

Satir P, Mitchell DR, Jekely G. 2008. How did the ciliumevolve? Curr Top Dev Biol 85: 63–82.

Schirrmeister BE, de Vos JM, Antonelli A, Bagheri HC. 2013.Evolution of multicellularity coincided with increaseddiversification of cyanobacteria and the Great OxidationEvent. Proc Natl Acad Sci 110: 1791–1796.

Servin JA, Herbold CW, Skophammer RG, Lake JA. 2008.Evidence excluding the root of the tree of life from theactinobacteria. Mol Biol Evol 25: 1–4.

Sherratt DJ, Arciszewska LK, Crozat E, Graham JE, GraingeI. 2010. The Escherichia coli DNA translocase FtsK. Bio-chem Soc Trans 38: 395–398.

Shavit L, Penny D, Hendy MD, Holland BR. 2007. Theproblem of rooting rapid radiations. Mol Biol Evol 24:2400–2411.

Silhavy TJ, Kahne D, Walker S. 2010. The bacterial cell en-velope. Cold Spring Harb Perspect Biol 2: a000414.

Skophammer RG, Herbold CW, Rivera MC, Servin JA, LakeJA. 2006. Evidence that the root of the tree of life is notwithin the Archaea. Mol Biol Evol 23: 1648–1651.

Skophammer RG, Servin JA, Herbold CW, Lake JA. 2007.Evidence for a Gram-positive, eubacterial root of the treeof life. Mol Biol Evol 24: 1761–1768.

Smirnov A, Chao EE, Nassonova E, Cavalier-Smith T. 2011.A revised classification of naked lobose amoebae (Amoe-bozoa: Lobosa). Protist 162: 545–570.

Song H, Sandie R, Wang Y, Andrade-Navarro MA, Nieder-weis M. 2008. Identification of outer membrane proteinsof Mycobacterium tuberculosis. Tuberculosis (Edinb) 88:526–544.

Sousa FL, Shavit-Grievink L, Allen JF, Martin WF. 2013.Chlorophyll biosynthesis gene evolution indicates pho-tosystem gene duplication, not photosystem merger, atthe origin of oxygenic photosynthesis. Genome Biol Evol5: 200–216.

Stanier RY. 1970. Some aspects of the biologyof cells and theirpossible evolutionary significance. In Organization andcontrol in prokaryotic and eukaryotic cells (Society for Ge-neral Microbiology Symposium 20) (ed. Charles HP, KnightJG), pp. 1–38. Cambridge University Press, Cambridge.

Stouf M, Meile JC, Cornet F. 2013. FtsK actively segregatessister chromosomes in Escherichia coli. Proc Natl Acad Sci110: 11157–11162.

Sullivan NL, Marquis KA, Rudner DZ. 2009. Recruitment ofSMC by ParB–parS organizes the origin region and pro-motes efficient chromosome segregation. Cell 137: 697–707.

Sutcliffe IC. 2010. A phylum level perspective on bacterialcell envelope architecture. Trends Microbiol 18: 464–470.

Sutcliffe IC. 2011. Cell envelope architecture in the Chloro-flexi: A shifting frontline in a phylogenetic turf war. En-viron Microbiol 13: 279–282.

Thiergart T, Landan G, Schenk M, Dagan T, Martin WF. 2012.An evolutionary networkof genes present in the eukaryotecommon ancestor polls genomes on eukaryotic and mi-tochondrial origin. Genome Biol Evol 4: 466–485.

Tsukatani Y, Romberger SP, Golbeck JH, Bryant DA. 2012.Isolation and characterization of homodimeric type-Ireaction center complex from Candidatus Chloracido-bacterium thermophilum, an aerobic chlorophototroph.J Biol Chem 287: 5720–5732.

T. Cavalier-Smith




Tyson JJ, Novak B. 2008. Temporal organization of the cellcycle. Curr Biol 18: R759–R768.

Valas RE, Bourne PE. 2009. Structural analysis of polarizingindels: An emerging consensus on the root of the tree oflife. Biol Direct 4: 30.

Valas RE, Bourne PE. 2011. The origin of a derived super-kingdom: How a Gram-positive bacterium crossed thedesert to become an archaeon. Biol Direct 6: 16.

van Dam TJ, Townsend MJ, Turk M, Schlessinger A, Sali A,Field MC, Huynen MA. 2013. Evolution of modular in-traflagellar transport from a coatomer-like progenitor.Proc Natl Acad Sci 110: 6943–6948.

Van Valen LM, Maiorana VC. 1980. The archaebacteria andeukaryotic origins. Nature 287: 248–250.

Verhey KJ, Dishinger J, Kee HL. 2011. Kinesin motors andprimary cilia. Biochem Soc Trans 39: 1120–1125.

Vossbrinck CR, Maddox JV, Friedman S, Debrunner-Voss-brinck BA, Woese CR. 1987. Ribosomal RNA sequencesuggests microsporidia are extremely ancient eukaryotes.Nature 326: 411–414.

Ward NL, Challacombe JF, Janssen PH, Henrissat B, Cou-tinho PM, Wu M, Xie G, Haft DH, Sait M, Badger J, et al.2009. Three genomes from the phylum Acidobacteriaprovide insight into the lifestyles of these microorgan-isms in soils. Appl Environ Microbiol 75: 2046–2056.

Wickstead B, Gull K. 2011. The evolution of the cytoskele-ton. J Cell Biol 194: 513–525.

Wickstead B, Gull K, Richards TA. 2010. Patterns of kinesinevolution reveal a complex ancestral eukaryote with amultifunctional cytoskeleton. BMC Evol Biol 10: 110.

Williams TA, Embley TM. 2014. Archaeal “dark matter” andthe origin of eukaryotes. Genome Biol Evol 6: 474–481.

Williams TA, Foster PG, Nye TM, Cox CJ, Embley TM. 2012.A congruent phylogenomic signal places eukaryoteswithin the Archaea. Proc Biol Sci 279: 4870–4879.

Williams TA, Foster PG, Cox CJ, Embley TM. 2013. Anarchaeal origin of eukaryotes supports only two primarydomains of life. Nature 504: 231–236.

Woese CR. 1994. There must be a prokaryote somewhere:microbiology’s search for itself. Microbiol Rev 58: 1–9.

Woese CR. 2013. How we do, don’t, and should look atbacteria and bacteriology. In The prokaryotes: Prokaryoticbiology and symbiotic associations (ed. Rosenberg E,et al.), pp. 3–20. Springer, New York.

Woese CR, Fox GE. 1977. Phylogenetic structure of the pro-karyotic domain: The primary kingdoms. Proc Natl AcadSci 74: 5088–5090.

Woese CR, Kandler O, Wheelis ML. 1990. Towards a naturalsystem of organisms: Proposal for the domains Archaea,Bacteria, and Eucarya. Proc Natl Acad Sci 87: 4576–4579.

Wolf YI, Makarova KS, Yutin N, Koonin EV. 2012. Updatedclusters of orthologous genes for Archaea: A complexancestor of the Archaea and the byways of horizontalgene transfer. Biol Direct 7: 46.

Yang Z. 2007. Fair-balance paradox, star-tree paradox, andBayesian phylogenetics. Mol Biol Evol 24: 1639–1655.

Yutin N, Makarova KS, Mekhedov SL, Wolf YI, Koonin EV.2008. The deep archaeal roots of eukaryotes. Mol Biol Evol25: 1619–1630.

Yutin N, Wolf MY, Wolf YI, Koonin EV. 2009. The origins ofphagocytosis and eukaryogenesis. Biol Direct 4: 9.

Yuzawa Y, Nishihara H, Haraguchi T, Masuda S, ShimojimaM, Shimoyama A, Yuasa H, Okada N, Ohta H. 2012.Phylogeny of galactolipid synthase homologs togetherwith their enzymatic analyses revealed a possible originand divergence time for photosynthetic membrane bio-genesis. DNA Res 19: 91–102.

Zhang Z, Green BR, Cavalier-Smith T. 2000. Phylogeny ofultra-rapidly evolving dinoflagellate chloroplast genes: Apossible common origin for sporozoan and dinoflagel-late plastids. J Mol Evol 51: 26–40.

Zhao S, Shalchian-Tabrizi K, Klaveness D. 2013. Sulcozoarevealed as a paraphyletic group in mitochondrial phy-logenomics. Mol Phylogenet Evol 69: 462–468.

Zhaxybayeva O, Lapierre P, Gogarten JP. 2005. Ancient geneduplications and the root(s) of the tree of life. Proto-plasma 227: 53–64.





2014; doi: 10.1101/cshperspect.a016006Cold Spring Harb Perspect Biol Thomas Cavalier-Smith Cilia in the Light of Intracellular Coevolution and a Revised Tree of LifeThe Neomuran Revolution and Phagotrophic Origin of Eukaryotes and

Subject Collection The Origin and Evolution of Eukaryotes

FunctionEukaryotic Gen(om)e Architecture and Cellular The Persistent Contributions of RNA to

Jürgen Brosius

Mitochondrion Acquired?Eukaryotic Origins: How and When Was the

Anthony M. Poole and Simonetta Gribaldo

the Plant KingdomGreen Algae and the Origins of Multicellularity in

James G. Umen

Bacterial Influences on Animal OriginsRosanna A. Alegado and Nicole King

Phylogenomic PerspectiveThe Archaeal Legacy of Eukaryotes: A

Lionel Guy, Jimmy H. Saw and Thijs J.G. Ettemathe Eukaryotic Membrane-Trafficking SystemMissing Pieces of an Ancient Puzzle: Evolution of

Klute, et al.Alexander Schlacht, Emily K. Herman, Mary J.

OrganellesCytoskeleton in the Network of Eukaryotic Origin and Evolution of the Self-Organizing

Gáspár Jékely LifeIntracellular Coevolution and a Revised Tree ofOrigin of Eukaryotes and Cilia in the Light of The Neomuran Revolution and Phagotrophic

Thomas Cavalier-Smith

from Fossils and Molecular ClocksOn the Age of Eukaryotes: Evaluating Evidence

et al.Laura Eme, Susan C. Sharpe, Matthew W. Brown,

Consequence of Increased Cellular ComplexityProtein Targeting and Transport as a Necessary

Maik S. Sommer and Enrico Schleiff

SplicingOrigin of Spliceosomal Introns and Alternative

Manuel Irimia and Scott William Roy

How Natural a Kind Is ''Eukaryote?''W. Ford Doolittle

Eukaryotic Epigeneticsand Geochemistry on the Provenance ofImprints of Bacterial Biochemical Diversification Protein and DNA Modifications: Evolutionary

et al.L. Aravind, A. Maxwell Burroughs, Dapeng Zhang,

Insights from Photosynthetic EukaryotesEndosymbionts to the Eukaryotic Nucleus? What Was the Real Contribution of

David Moreira and Philippe Deschamps

Phylogenomic PerspectiveThe Eukaryotic Tree of Life from a Global

Fabien BurkiComplex LifeBioenergetic Constraints on the Evolution of

Nick Lane

http://cshperspectives.cshlp.org/cgi/collection/ For additional articles in this collection, see

Copyright © 2014 Cold Spring Harbor Laboratory Press; all rights reserved


http://cshperspectives.cshlp.org/cgi/collection/

http://cshperspectives.cshlp.org/cgi/collection/


Date post:	16-Jul-2020
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

The Neomuran Revolution and Phagotrophic Origin of ...system, cytoskeleton, and associated motors....

Documents