Endosymbiotic Theory

A Bodył and P Mackiewicz, University of Wrocław, Wrocław, Poland

© 2013 Elsevier Inc. All rights reserved.

Glossary plastids contain chlorophyll c and are surrounded by four

Alveolata A monophyletic taxon composed primarily of membranes, the outermost bearing ribosomes.
dinoflagellates (half of them containing the peridinin Endosymbionts Organisms residing in the cells or tissues of
plastid), parasitic apicomplexans (with a other organisms, which can come from either eukaryotic or
nonphotosynthetic plastid known as the apicoplast), and prokaryotic lineages; in contrast to cell organelles, their
ciliates (completely aplastidal); a common feature of genomes encode all their required proteins and, in the case
alveolates is the presence of alveoli, flattened of obligatory endosymbionts, they are integrated with their
membranous sacs that subtend their plasma membrane host cells only at the metabolic level.
and fulfill cytoskeletal functions. Endosymbiotic gene replacement There are two kinds of
Archaea One of three domains of life, parallel to Bacteria such replacement: one is when a host gene present in the
and Eucarya (Eukaryota); it contains unicellular nucleus, takes over the function of an endosymbiont gene,
microorganisms devoid of a nucleus. These organisms resulting in loss of the latter. The second is when a host
formerly were classified under the name Archaebacteria nuclear gene is replaced by an endosymbiont gene
along with Bacteria (previously Eubacteria) as prokaryotes; transferred to the host nucleus. In the first case, this
however, archaeons differ from bacteria in the biochemistry process generally is accompanied by duplication of a host
of their cell walls and membranes, and many of their nuclear gene, followed by one copy is acquiring a proper
informational genes (i.e., those encoding DNA replication, endosymbiont/organelle-targeting signal.
transcription, and translation components) are more Endosymbiotic gene transfer Gene transfer occurring
closely related to eukaryotes than to bacteria. during endosymbiosis in which a gene from an
Archaeplastida A eukaryotic kingdom (also known as endosymbiont or organelle genome is transferred and
Plantae) comprising three groups of eukaryotes harboring integrated into the host’s nuclear genome.
primary plastids, that is, glaucophytes, red algae, and green ERAD system The endoplasmic reticulum-associated
plants (including green algae and higher plants). Their degradation (ERAD) transport system characteristic of all
plastids are surrounded by two membranes and contain eukaryotes; it participates in the export of misfolded
phycobiliproteins (glaucophytes, red algae) or chlorophyll b proteins from the lumen of the endoplasmic reticulum
(green plants) in addition to chlorophyll a; moreover, into the cytosol, where they are degraded by the
glaucophyte plastids still retain the bacterial cell wall (of proteasome. ERAD-like translocons also operate in
peptidoglycan) of their cyanobacterial ancestor, a conserved peroxisomes and secondary plastids.
ancient trait in the evolution of primary plastids. Hydrogenosomes Anaerobic forms of mitochondria that
Cell organelles Intracellular structures surrounded by a perform H2-producing fermentation reactions to
membrane or membranes and fulfilling specific metabolic synthesize ATP via substrate-level phosphorylation; they
functions. Some organelles, such as mitochondria and occur in some anaerobic protists.
plastids, have an endosymbiotic origin; their genomes Mitosomes Highly reduced mitochondria of several
encode only a very restricted pool of proteins, with the anaerobic and parasitic protozoan lineages; they have lost
remaining proteins encoded in the host nuclear genome all bioenergetic functions characteristic of mitochondria
(some originally from the endosymbiont and others from and hydrogenosomes. The only functions known for
the host) and imported into the organelles through mitosomes are sulfate metabolism and Fe–S cluster
translocons in their envelope membranes. biogenesis.
Chromista A historic term for a eukaryotic kingdom Primary plastids Plastids derived directly from
containing algae with red alga-derived plastids, such as cyanobacteria; they are found in all representatives of the
cryptophytes, heterokonts, and haptophytes. Their Archaeplastida.

Introduction I. E. Wallin) toward the end of the nineteenth and the beginning
of the twentieth century. Although initially abandoned, the idea
The endosymbiotic theory posits that at least some organelles in was revived after the discovery of DNA molecules in mitochondria
eukaryotic cells, mitochondria and plastids, in particular, evolved and plastids in the 1960s. In 1967, Lynn Margulis suggested not
from free-living organisms that were enslaved by other cells only the endosymbiotic origin of mitochondria and plastids, but
functioning as their hosts (Figure 1). The endosymbiotic (or also an endosymbiotic origin of the eukaryotic cell itself. In her
exogenous) origin of mitochondria and plastids was proposed model, eukaryotic cells originated in a series of endosymbioses
by several scientists (C. Mereschkovsky, A. F. W. Schimper, involving distinct bacterial cells. A hallmark of this scenario was a

484 Brenner’s Encyclopedia of Genetics, 2nd edition, Volume 2 doi:10.1016/B978-0-12-374984-0.01678-8

 Endosymbiotic Theory 485

(a) (b)

Peptidoglycan

α-Proteo­
bacterium

Gram-positive Peptidoglycan-less
bacterium anaerobic bacterium

Spirochete

Phagotrophic

eukaryote

α-Proteo­
bacterium

Phagotrophic
eukaryote

Cyano­
bacterium

Nucleus
Mitochondrion

Plastid
Cilium

Golgi Endoplasmic
apparatus reticulum

Cytoskeleton
Figure 1 Two main models explaining the origin of the eukaryotic cell. (a) The non-symbiotic model posits that the characteristic organelles and
structures of the eukaryotic cell are not mitochondria and plastids, but the endomembrane system (including peroxisomes), the nucleus, and the cilium
along with cytoskeleton. It is hypothesized that all these structures evolved autogenously (= endogenously) from a Gram-positive bacterial ancestor
(termed the neomuran) after loss of its peptidoglycan wall, and were driven by endo- and exocytosis. In this model, only an already phagocytotic
eukaryotic cell could engulf the bacterial progenitors of mitochondria and plastids. Consequently, endosymbioses only resulted in the diversification of the
eukaryotic cell, but not in its origin. (b) By contrast, the endosymbiotic model postulates that the eukaryotic cell originated by a series of endosymbioses
occurring between distinct bacterial cells. Its original version, formulated by L. Margulis, stated that the first endosymbiosis involved an anaerobic
bacterial host devoid of cell wall and an aerobic bacterium as an endosymbiont. The nucleus evolved autogenously after this endosymbiosis and such a
proto-eukaryotic cell could then acquire a cilium and cytoskeleton from a spirochete (a kind of ectosymbiosis). According to this model, the last
endosymbiosis involved a cyanobacterium as the source of primary plastids. Some versions of the endosymbiotic scenario suggest an endosymbiotic
origin for almost all organelles (including even the endoplasmic reticulum) within the eukaryotic cell.

spirochete origin of the eukaryotic cytoskeleton and cilium. Thus, within an already established eukaryotic host, whereas the second
there are two main versions of the endosymbiotic theory: one suggests the eukaryotic cell initially assembled through a process
proposes that some cell organelles evolved by endosymbiosis of endosymbiosis (Figure 1).
486 Endosymbiotic Theory

Bacterial Origins for Mitochondria and Primary Plastids these controversies is the chimerism of bacterial genomes; it
probably results from horizontal gene transfers between distinct
Mitochondria and primary plastids almost certainly evolved
bacterial lineages on the one hand, and lineage-specific gene
from α-proteobacteria and cyanobacteria, respectively
duplications and losses on the other hand.
(Figure 2), but a more exact determination of their bacterial
α-Proteobacteria and cyanobacteria are Gram-negative bac­
sources remains elusive. In the case of mitochondria, it has
teria, which means the ancestors of mitochondria and primary
been suggested they evolved from Rickettsia-, Pelagibacter-, or
plastids were very likely surrounded by two membranes, a
Rhodospirillum-like species. A similar uncertainty concerns pri­
plasma membrane and an outer membrane, with a peptidogly­
mary plastids, with proposal of an Anabena-like species versus
can wall between them.
an unknown ancient cyanobacterial lineage. A major to resolving

Phagosomal membrane
Outer membrane Ribosome
Peptidoglycan Mitochondrion
Plasma membrane
Phycobilisome
Thylakoid

Nucleus

Cyanobacterium Endoplasmic reticulum

Transport vesicle
Golgi apparatus
Phagotrophic ancestor
of Archaeplastida

TP
Phagosomal + outer membrane

Plasma membrane

Toc

Tic

EGT

Tic
Primary plastid

SP

Cell wall

Higher plant cell

Figure 2 Evolution of primary plastids. Primary plastids evolved from cyanobacteria that were engulfed by phagotrophic protozoans. Driving forces
behind the establishment of primary plastids could have been photosynthesis, as well as nonphotosythetic functions such as nitrogen fixation or fatty acid
synthesis. Numerous genes of the cyanobacterial endosymbiont have moved to the host nucleus via the endosymbiotic gene transfer (EGT). Almost all
their protein products are equipped with transit peptides (TPs) and imported into the plastid with the help of Toc and Tic translocons; however, some
plastid pre-proteins carry signal peptides (SPs)(e.g., α-carbonic anhydrase) that deliver them to primary plastids via the endoplasmic reticulum and/or the
Golgi apparatus. The inner membrane of primary plastids is certainly derived from the endosymbiont plasma membrane, but the origin of their outer
membrane is still controversial. This membrane could have come from the host phagosomal membrane (as demonstrated by the endomembrane
system-mediated targeting of some proteins) or the outer membrane of the cyanobacterial endosymbiont (the membrane contains porin-like proteins such
as the Toc75 channel). It is possible, however, that the outer membrane of primary plastids has a chimeric eukaryotic–bacterial origin, because the
engulfed cyanobacteria (with their outer and plasma membranes) must initially have been surrounded by a phagosomal membrane, and, after its
disruption, their outer membrane could have acquired features of the host phagosomal membrane.
 Endosymbiotic Theory 487

Transformation of Endosymbionts into Organelles membrane insertion, they acquired N-terminal targeting signals
permitting them to use insertion machinery. In a last step, these
The first step in the evolution of mitochondria and primary
pathways, along with an amino acid pre-existing channel in the
plastids was establishment of permanent endosymbioses between
inner endosymbiont membrane, could have given rise to the
their bacterial ancestors and eukaryotic (or prokaryotic) host cells.
import of proteins with N-terminal targeting signals into
Under the traditional model, these bacteria were initially engulfed
the endosymbiont matrix (or stroma).
as food by eukaryotic cells and digested in phagosomes.
Under the classical view, the outer membrane of mitochon­
Therefore, establishment of stable α-proteobacterial and cyano-
dria and primary plastids corresponds strictly to the outer
bacterial endosymbioses required disturbing the digestion process membrane of Gram-negative bacteria. In higher plants, however,
by disrupting the phagosomal membrane or modifying the endo­ a small number of nuclear-encoded plastid proteins carry signal
cytotic pathway. Bacteria are known (mainly parasitic in these peptides and are imported via the endoplasmic reticulum (ER)
cases, but similar to bacteria suggested as the mitochondrial and/or Golgi apparatus. This suggests that the outer plastid
ancestor) that invade eukaryotic cells without formation of membrane could have a chimeric eukaryotic–bacterial origin.
phagosomes and/or secrete proteins that modify endocytotic
pathways. In contrast, the mechanisms by which bacteria integrate
inside their prokaryotic host cells are unknown.
Complex Evolutionary Pathways of Plastids: Eukaryotic
After the establishment of permanent endosymbioses, the
Alga-Derived Plastids and Disappearing Nuclei
acquired bacteria began to undergo transformation into true cell
organelles. Such transformations involved two processes: gene The evolution of plastids is much more complicated than that
transfer from the endosymbiont to the host nuclear genome, of mitochondria. Primary plastid endosymbiosis resulted in the
and the origin of translocons in the endosymbiont envelope plastids of glaucophytes, red algae, and green plants (Figure 2),
that could import proteins encoded by these transferred genes. which constitute the kingdom Archaeplastida; however, this
Known genomes of mitochondria and plastids contain from 3 to was only the starting point for the further reticulate plastid
273 protein-encoding genes, whereas those of modern evolution. In a second layer of plastid endosymbioses, both
α-proteobacteria and cyanobacteria encode, for example, 3788 red and green algal plastids were acquired independently by
(Rhodospirillum rubrum) and 5043 (Anabaena variabilis), respec­ different heterotrophic lineages, creating secondary plastids
tively. Numerous endosymbiont genes required by free-living surrounded by three or four envelope membranes (Figure 3).
bacteria were lost completely, while most others were transferred Green alga-derived secondary plastids are present in euglenids,
(or ‘escaped’) to the host nuclear genome via endosymbiotic gene some dinoflagellates (e.g., Lepidodinium viride), and chlorarach­
transfer. Various hypotheses have been advanced to explain this niophytes, whereas those descended from red algae occur
process, for example, that such transfers protected endosymbiont in cryptophytes, heterokonts, and haptophytes (the former
genes from mutations induced by free radicals generated during kingdom Chromista), as well as in most plastid-bearing dino­
redox reactions occurring in mitochondria and plastids. flagellates and apicomplexans (in the supergroup Alveolata).
After transfer to the nucleus, each prokaryotic In a third layer of plastid endosymbioses, secondary
endosymbiont-derived gene needed to acquire nuclear promoter plastid-containing algae (e.g., cryptophytes and haptophytes)
and polyadenylation signals. Acquistions of these signals by the were enslaved by additional eukaryotic hosts belonging to the
transferred genes enabled their expression in the host cytosol dinoflagellates. Good examples of tertiary endosymbioses are
and sometimes resulted in the replacement of a host gene by a the fucoxanthin plastids of Karenia brevis and Karlodinium
bacterial homolog, a process known as endosymbiotic gene micrum, which evolved from a haptophyte alga.
replacement. In a next evolutionary stage, transferred genes A ‘missing-link’ stage in the evolution of eukaryotic
acquired sequences encoding targeting signals (e.g., via exon shuf­ alga-derived plastids is exemplified by chlorarachniophytes
fling) that allowed their protein products to be imported into the and cryptophytes. These algae preserved remnants of the cyto­
mitochondrion or the primary plastid. Most proteins targeted plasm of the eukaryotic alga, including its vestigial nucleus
to mitochondria and primary plastids carry N-terminal transit known as the nucleomorph (Figure 3). Although nucleomorph
peptides that are later removed in the organelle matrix. genomes have undergone drastic reduction, they preserve
Mitochondria and primary plastids are surrounded by two several typical eukaryotic features, including linear chromo­
membranes. Consequently, their import machineries are com­ somes with telomeres and spliceosomal introns. Moreover,
posed of two translocons: one for the outer membrane and the molecular phylogenetic analyses of nucleomorph genes clearly
other for the inner membrane; these are Tom and Tim (mito­ demonstrate that they are derived from green algal (chlorar­
chondria) and Toc and Tic (primary plastids), respectively achniophytes) and red algal (cryptophytes) nuclei.
(Figure 2). Each of these translocons is built from multiple
proteins fulfilling channel, receptor, and regulatory functions.
Modifications, Reductions, and Losses of
Some of these components can be traced to the endosymbiont,
Endosymbiont-Derived Organelles
whereas others are evolutionary inventions of the host lineage,
suggesting that the import machineries of mitochondria and Although most eukaryotes contain mitochondria, there are
primary plastids evolved via molecular tinkering. Proteins initi­ several lineages (e.g., diplomonads, some amoebozoans,
ally imported into mitochondria and primary plastids probably microsporidians) that are devoid of typical mitochondria. In
were host-derived metabolic carriers devoid of N-terminal tar­ the 1980s, Cavalier-Smith erected the superkingdom Archezoa,
geting signals. These could cross the outer membrane through whose members were proposed never to have contained mito­
pre-existing porin-like channels and then insert spontaneously chondria; in other words, they represented a premitochondrial
into the inner membrane. Perhaps to improve efficiency of their stage in the evolution of eukaryotic cell. A characteristic
488 Endosymbiotic Theory

Phagosomal membrane
TP

Plasma membrane

Plastid envelope
Mitochondrion
Phycobilisome

Thylakoid Toc
Tic

Primary plastid Nucleus

Rough endoplasmic reticulum

Red alga

Cryptophyte ancestor

Plastid endoplasmic reticulum TP

Perplastid membrane

Perplastid compartment

Plastid envelope

Toc
Phycobilin proteins
Tic
Nucleomorph
Thylakoid
EGT

Secondary plastid EGT

Tic
Toc
Der
Sec

TP Cryptophyte

SP TP
Figure 3 Evolution of secondary plastids in cryptophyte algae. Cryptophytes are a peculiar group of algae containing secondary plastids of red algal
origin. Their plastids are surrounded by four membranes: the two innermost membranes (or plastid envelope) originated from the envelope membranes of
the endosymbiont primary plastid, the third (or periplastid) membrane from the endosymbiont plasma membrane, and the outermost membrane from the
phagosomal membrane of the host. Since the outermost membrane is covered with ribosomes, it is suggested that the phagosomal membrane underwent
fusion with the ER, resulting in the plastid ER membrane. Cryptophyte plastids still retain the vestigial cytoplasm of the red algal endosymbiont (the
periplastid compartment) along with its reduced nucleus known as the nucleomorph. The nucleomorph genome encodes some plastid proteins with
transit peptides that are imported into the stroma with the help of Toc and Tic translocons. During secondary endosymbiosis, most endosymbiont genes
were transferred out of the endosymbiont’s nucleus (subsequently reduced to the nucleomorph) via endosymbiotic gene transfer (EGT). Host
nucleus-encoded, plastid-targeted proteins of cryptophytes carry bipartite presequences composed of a signal peptide followed by a transit peptide. The
signal peptide is responsible for protein translocation across the outermost membrane containing the Sec translocon, whereas the transit peptide enables
passage across the three remaining membranes that are equipped with the Der translocon (in the periplastid membrane), and the Toc and Tic translocons
(in the plastid envelope). The Der translocon is derived from the ERAD system and was probably relocated from the endosymbiont’s ER.

representative of this clade was the diplomonad Giardia lam- common ancestor. Hydrogenosomes and mitosomes are spe­
blia; however, further studies identified mitochondrion-related cialized mitochondria that arose as adaptations to anaerobic
organelles, such as hydrogenosomes and mitosomes, in each conditions. They are usually devoid of a genome, although the
eukaryotic supergroup (including all archezoan lineages), tra- hydrogenosome of the ciliate Nyctotherus ovalis still retains its
cing the origin of mitochondria back to the eukaryotic DNA. Formation of Fe–S clusters is a common and vital feature
 Endosymbiotic Theory 489

(a) Organics

Glucose

Glucose ATP
Glycolysis Anabolic
ATP ATP carbohydrate
H2 pathway
Pyruvate
Anaerobic CO2
respiration

ATP Methanogenesis
Oxidative
Facultatively respiration ATP

anaerobic

O2 H2O
α-proteobcterium CO
Only in aerobic
conditions
2
Methanogenic
CH4 archaeon

Organics

(b)
1 Organics

Glucose

Glucose

ATP 2 Glucose ATP

Pyruvate
Glycolysis Anabolic
ATP ATP carbohydrate
H2 pathway
Pyruvate
Anaerobic CO2
respiration

ATP Methanogenesis
Oxidative
respiration ATP

Methanogenic
archaeon H 2O
Only in aerobic
CH4
O2 conditions
with symbiotic CO2
α-proteobacterium

(c) Organics

Glucose
Glycolysis
ATP
Pyruvate 3
ATP Loss of anabolic
carbohydrate H2
Pyruvate
Anaerobic and CO2
respiration methanogenesis
pathways
ATP
Oxidative
respiration

Protoeukaryote
with ancestral Only in aerobic
O2 H 2O conditions
mitochondrion CO2

Figure 4 (Continued)
490 Endosymbiotic Theory

of typical mitochondria, hydrogenosomes, and mitosomes, Number of Endosymbiotic Origins of Mitochondria and Plastids
and apparently the main reason they are maintained by all
Available data clearly demonstrate that all mitochondria, along
eukaryotic cells.
with their hydrogenosomal and mitosomal derivatives, evolved
Plastids are known mainly as photosynthetic organelles,
from a single α-proteobacterial endosymbiosis. A single cyano-
but they also fulfill numerous vital nonphotosynthetic func­
bacterial origin seems to have given rise to the primary plastids
tions, such as the biosynthesis of fatty acids, heme groups, and
of glaucophytes, red algae, and green plants, but independent
amino acids. Consequently, they are retained by their hosts
primary endosymbioses and even secondary origin of some of
even when photosynthesis is lost. In addition to higher plants,
these plastids are still considered possible.
there are several well-characterized examples of nonphotosyn­
Available evidence indicates that the green algal plastids of
thetic plastids in euglenids (e.g., Euglena (Astasia) longa),
euglenids, some dinoflagellates (e.g., L. viride), and chlorarachnio­
cryptophytes (e.g., Cryptomonas paramecium), and heterokonts
phytes resulted from separate secondary endosymbioses. In
(e.g., Pteridomonas danica). Interestingly, such plastids are
addition, the tertiary plastids found in several dinoflagellate
found even in parasitic organisms, such as the green alga
lineages certainly evolved via independent endosymbioses. More
Helicosporidium sp., most apicomplexans (e.g., Plasmodium fal­
controversy surrounds the origin of chromist and alveolate plas­
ciparum and Toxoplasma gondii) and the apicomplexa-related
tids, termed collectively chromalveolate plastids. One model
perkinsid Perkinsus marinus. They are considered good targets
posits that all these plastids are derived from a single red algal
for the development of efficient drugs against diseases caused
secondary endosymbiosis; however, an increasing number of data
by these parasites.
favor an alternative scenario involving a secondary origin specifi­
As with mitochondria, there are no known examples of
complete plastid loss from algae with primary plastids. The cally in cryptophytes, with the remaining chromalveolate plastids
possibility of such losses is still debated in the cases of several evolving by multiple tertiary (or serial) endosymbioses.
protist lineages (e.g., ciliates, goniomonads) that are closely
related to algae containing secondary plastids. Plastid loss is
highly unlikely in such free-living organisms, however, because Endosymbiotic Origin of the Eukaryotic Cell: Nucleus or
of numerous plastid nonphotosynthetic functions. It appears Mitochondrion First?
more likely to have occurred in parasitic forms (e.g., in the There are two kinds of models for the endosymbiotic origin of
apicomplexan Cryptosporidium parvum) that receive abundant the eukaryotic cell, those suggesting that the nucleus was the
different compounds from their host cells. first endosymbiont-derived organelle to evolve and those
assuming that the first endosymbiotic organelle was the
mitochondrion.
Cyanobacterial Endosymbionts of Paulinella chromatophora:
Under the ‘nucleus first’ models (as in non-symbiotic models
An Interesting Example of Cell Organelles in Progress
for the origin of eukaryotic cells), the mitochondrion could only
The thecate amoeba P. chromatophora harbors two be acquired by a complex (eukaryotic) cell with a developed
cyanobacterium-derived endosymbionts that are deeply inte­ nucleus, an endomembrane system, and cytoskeleton along
grated with the host cell; for example, they divide with phagocytotic abilities they conferred. The evolution of all
synchronously with the host cell. Moreover, the genomes of these features must have been extremely expensive energetically
these endosymbionts have lost �75% of their initial coding (it required the origin and expression of �3000 new protein
capacity. Interestingly, this process was accompanied by the families), however, and thus could not have happened without
efficient transfer of at least 32 endosymbiont genes to the an energy source to satisfy these needs. Thus, it now appears that
host nucleus, including some that encode N-terminal targeting only the acquisition of a mitochondrion by a prokaryotic host,
signals. For example, an N-terminal extension of nucleus- with subsequent expansion of its inner membrane and massive
encoded PsaE from Paulinella strongly resembles signal transfer and/or loss of its non-bioenergetic genes could provide
peptides (and it is absent from homologous cyanobacterial sufficient energy sources to support, in energetic terms, the origin
proteins), suggesting protein import into these endosymbionts of typical eukaryotic features. Consequently, the alternative
via the endomembrane system. All these data indicate that ‘mitochondrion first’ models are much more probable based
Paulinella endosymbionts are undergoing transformation into on the idea of syntrophy between an archaeal (= archaebacterial)
true cell organelles. host and an α-proteobacterial symbiont (Figure 4).

Figure 4 The hydrogen hypothesis for the origin of the first eukaryotic cell by a syntrophic symbiosis between an α-proteobacterium (the symbiont) and
a methanogenic archaeon (the host). (a) The methanogen was an obligate autotroph dependent on H2 (used in the synthesis of ATP via methanogenic
respiration) and CO2 (incorporated into carbohydrates in the anabolic carbohydrate pathway). The α-proteobacterium was a facultative anaerobe capable
of carrying out both oxidative and anaerobic respiration. Because the methanogen was an obligate anaerobe, these bacteria could only meet in an
anoxygenic environment, where the symbiont produced H2 and CO2 and the host utilized them. (b) After the disappearance of the geological source of H2,
the methanogen became strictly dependent on the α-proteobacterium and the selection favored its enclosure by the host to increase the diffusion of H2.
These changes decreased the influx of organic compounds into the α-proteobacterium, and later could have triggered the relocation of organic importers
to the host plasma membrane (1). At this stage, the methanogen imported organic compounds and transported them into the α-proteobacterium, but
most were certainly incorporated by its anabolic carbohydrate pathway. A subsequent evolutionary step would have been relocation of glycolysis to the
host cytosol (2). (c) The presence of two carbohydrate pathways (anabolic and catabolic) in the methanogen cytosol operating in opposite directions
resulted in a futile carbon cycle, and selection eliminated the anabolic pathway (3). The resulting proto-eukaryotic cell was a heterotroph, with
proto-mitochondrion having features of both mitochondria (oxidative respiration) and hydrogenosomes (anaerobic respiration).
 Endosymbiotic Theory 491

An obstacle faced by all endosymbiotic models for the providing additional support for their autogenous (or endo­
origin of the eukaryotic cell is the inability of known bacterial membrane) origin.
or archaeal cells to employ phagotrophic feeding and, there­
fore, to be able to incorporate potential endosymbionts. There
Perspectives
do exist predatory bacteria that are able to invade other bacter­
ial cells (e.g., Bdellovibrio), however, and examples are known of The endosymbiotic theory is usually used to explain the origin
bacterial endosymbiosis (e.g., the γ-proteobacterium Moranella of eukaryotic cells, but it can also be applied to bacterial cells.
inside the β-proteobacterium Temblaya). Thus, it is important For example, Gram-negative bacteria could have evolved via an
to note that α-proteobacteria, the ancestors of mitochondria, endosymbiosis between a clostridium and an actinobacterium,
include Bdellovibrio-like species such as Midichloria mitochondrii. implying that their inner membrane is derived from the plasma
membrane of the endosymbiotic bacterium, whereas the outer
membrane originated from the plasma membrane of the bac­
Cilium, Nucleus, and Peroxisomes: Symbiotic versus terial host.
Autogenous Origins Darwin did not consider the significance of symbiotic
associations in his theory of evolution. Moreover,
The motility symbiosis between the parabasalid Mixotricha
endosymbiosis-mediated fusion of evolutionarily distinct
paradoxa and spirochetes attached to its plasma membrane is
lineages (netlike or reticulate evolution) contrasts with his
sometimes regarded as a model for symbiotic origin of the idea of bifurcating divergence from common ancestors (treelike
eukaryotic cilium and cytoskeleton. The movement of spiro­ evolution). Thus, endosymbiotic associations are sometimes
chetes is driven by typical bacterial flagella, however, which are treated as examples of non-Darwinian evolution via the inheri­
absent from eukaryotic cilia. Although these bacteria have tance of acquired characteristics (e.g., acquisition of new genes
microtubule-like structures in their cytosol, they are neither and membranes) or macrogenesis involving ‘hopeful monsters’
composed of tubulins nor involved in cell movement. (e.g., protozoans containing red or green algal endosym­
Moreover, spirochetes are surrounded by two membranes bionts). Nevertheless, each endosymbiotic association comes
(with flagella located between them) and it is difficult to ima­ under natural selection, resulting in the survival (and reproduc­
gine how they could be transformed into cilia that are not even tion) of only those best adapted to their environments.
completely bound by a single membrane. Finally, tubulins Regardless of these general evolutionary considerations,
probably evolved from the bacterial FtsZ proteins, which are available data clearly indicate that endosymbioses have had
involved in bacterial cell division. an enormous impact on the evolution of biosphere of our
Eukaryotic informational genes encoding replication, tran­ planet. It seems that a special role in this process was played
scription, and translation components predominantly have an by the mitochondrial endosymbiosis, which not only enabled
archaeal origin. Based on this finding, it was proposed that the the origin of the first eukaryotic cell but also facilitated a dra­
nucleus evolved from an archaeal endosymbiont; however, the matic increase in the complexity of the eukaryotic world
nuclear envelope contains highly permeable pore complexes through the evolution of multicellularity.
and disintegrates during open mitosis, unique features absent
from all prokaryotic cells and endosymbiont-derived orga­
nelles. Moreover, it originates from the endomembrane See also: Archaea; Archaeal Genetics; Bacteria; Chloroplasts,
system during the eukaryotic cell cycle. Available data clearly Genetics of; Cytoskeleton; Eukaryotes; Evolution: Eukaryotes;
indicate that the host component of eukaryotic cells evolved Flagella; Gene Duplication; Mitochondria, Genetics of;
from an archaeon, whereas the transformation of its genome to Glycolysis; Horizontal Gene Transfer; Introns and Exons;
the eukaryotic nucleus could have been driven by the endo­ Mitochondria; Mitochondrial Genetics/Evolution; Mitochondrial
symbiosis with an α-proteobacterium (later mitochondrion). DNA; Mitochondrial Genome; Nuclear Envelope, Transport;
Transfer of group II self-splicing introns from the Nuclear Pores; Nucleomorph Genomes; Nucleus; Organelles;
α-proteobacterium to the archaeal host genome, and their sub­ Progenote; Prokaryotes; Signal Sequence; Splicing; Symbionts,
sequent evolution into spliceosomal introns, would have Genetics of; Symbiosome.
resulted in the direct translation of these foreign sequences to
nonfunctional chimeric proteins. Only spatial separation of
translation from intron excision by the formation of a nuclear
Further Reading
envelope could solve these problems.
Considering their ability to divide and import matrix pro­
Archibald JM (2009) The puzzle of plastid evolution. Current Biology 19: R81–R88.
teins posttranslationally, it was suggested that peroxisomes, Archibald JM (2011) Origin of eukaryotic cells: 40 Years on. Symbiosis 54: 69–86.
which are surrounded by a single membrane, also have an Bodył A, Mackiewicz P, and Milanowski R (2010) Did trypanosomatid parasites contain a
endosymbiotic origin. Most proteins responsible for their bio­ eukaryotic alga-derived plastid in their evolutionary past? Journal of Parasitology 96:
465–475.
genesis, however, are imported into them cotranslationally via
Bodył A, Mackiewicz P, and Stiller JW (2009) Early steps in plastid evolution: Current
the ER. Moreover, the peroxisome membrane possesses an ideas and controversies. BioEssays 31: 1219–1232.
ER-derived translocon (or ERAD system) involved in protein Cavalier-Smith T (2009) Predation and eukaryote cell origins: A coevolutionary
import, but is devoid of bacterial porin-like proteins character­ perspective. International Journal of Biochemistry & Cell Biology 41: 307–322.
Embley TM and Martin W (2006) Eukaryotic evolution, changes and challenges. Nature
istic of the outer membranes of mitochondria and plastids. It
440: 623–630.
was also demonstrated that, unlike plastids and mitochondria, Gabaldón T, Snel B, van Zimmeren F, et al. (2006) Origin and evolution of the
peroxisomes can develop de novo from the ER after they are lost, peroxisomal proteome. Biology Direct 1: 8.
492 Endosymbiotic Theory

Keeling PJ (2011) Endosymbiosis: Bacteria sharing the load. Current Biology 21: endosymbionts/plastids via the endomembrane system. Journal of Phycology 46:
R623–R624. 847–859.
Kleine T, Maier UG, and Leister D (2009) DNA transfer from organelles to the nucleus: The Margulis L (1993) Symbiosis in Cell Evolution. New York: W. H. Freeman and Company.
idiosyncratic genetics of endosymbiosis. Annual Review of Plant Biology 60: 115–138. Martin W and Koonin EV (2006) Introns and the origin of nucleus-cytosol
Lake JA (2009) Evidence for an early prokaryotic endosymbiosis. Nature 460: 967–971. compartmentalization. Nature 440: 41–45.
Lane N and Martin W (2010) The energetics of genome complexity. Nature 467: 929–934. Martin WF and Müller M (eds.) (2007) Origin of Mitochondria and Hydrogenosomes.
Li JY and Wu CF (2003) Perspectives on the origin of microfilaments, microtubules, the Berlin; Heidelberg: Springer-Verlag.
relevant chaperonin system and cytoskeletal motors – A commentary on the Richards TA and Archibald JM (2011) Cell evolution: Gene transfer agents and the origin
spirochaete origin of flagella. Cell Research 13: 219–227. of mitochondria. Current Biology 21: R112–R114.
Mackiewicz P and Bodył A (2010) A hypothesis for import of the nuclear-encoded PsaE Stiller JW (2007) Plastid endosymbiosis, genome evolution and the origin of green
protein of Paulinella chromatophora (Cercozoa, Rhizaria) into its cyanobacterial plants. Trends in Plant Science 12: 391–396.