Sie sind auf Seite 1von 8

Evolutionary mechanisms for

establishing eukaryotic cellular


complexity
Fred D. Mast
1,2,3
, Lael D. Barlow
1
, Richard A. Rachubinski
1
, and Joel B. Dacks
1
1
Department of Cell Biology, University of Alberta, Edmonton, Alberta T6G 2H7, Canada
2
Seattle Biomedical Research Institute, 307 Westlake Avenue North, Seattle, WA 98109-5240, USA
3
Institute for Systems Biology, 401 Terry Avenue North, Seattle, WA 98109-5219, USA
Through a comparative approach, evolutionary cell biol-
ogy makes use of genomics, bioinformatics, and cell
biology of non-model eukaryotes to provide new ave-
nues for understanding basic cellular processes. This
approach has led to proposed mechanisms underpin-
ning the evolution of eukaryotic cellular organization
including endosymbiotic and autogenous processes
and neutral and adaptive processes. Together these
mechanisms have contributed to the genesis and com-
plexity of organelles, molecular machines, and genome
architecture. We review these mechanisms and suggest
that a greater appreciation of the diversity in eukaryotic
form has led to a more complete understanding of the
evolutionary connections between organelles and the
unexpected routes by which this diversity has been
reached.
Bringing together cell biology and evolutionary biology
The emergence of the eukaryotic state nearly 2 billion
years ago transformed life on Earth. Efforts to unravel
the evolutionary mechanisms that have shaped, and con-
tinue to shape, eukaryotic cells are beginning to address
this monumental evolutionary shift. Understanding these
mechanisms will help us to make conceptual connections
between the cell biology of taxonomically diverse modern
eukaryotes, porting knowledge derived in model systems to
less studied organisms of agricultural (e.g., crops, plant
pathogens), environmental (e.g., aquatic primary produ-
cers like haptophytes and diatoms), or medical (e.g., para-
sites such as Plasmodium falciparum, the causative agent
of malaria) relevance. This broad comparative approach
known as evolutionary cell biology (see Glossary) facili-
tates the generation of hypotheses that attempt to explain
the cell biological functions shared among the full range of
eukaryotes.
This approach has been applied successfully to many
aspects of the eukaryotic cell (e.g., [1]). The combination of
ultrastructure and molecular cell biology with genomic
data from a sampling of organisms spanning the taxonomic
breadth of eukaryotes [2,3] (Figure 1) has provided a
wealth of knowledge regarding the evolution of eukaryotic
cell biology and its diversity. From the perspective of a cell
biologist, this wealth of data allows the integration of
established evolutionary theory with the study of cellular
mechanisms.
Review
Glossary
Complexity: a measure of the number of components and interactions of one
system relative to another equivalent system.
Endosymbiosis (primary): the process whereby a prokaryotic cell (endosym-
biont) is incorporated into the cytoplasm of a eukaryotic cell (host), with a
relationship being established via metabolic integration and EGT such that
neither partner can survive on its own.
Endosymbiosis (secondary): the same process as primary endosymbiosis
except that the endosymbiont is a eukaryotic cell possessing a primary plastid.
The process can be extended to tertiary endosymbiosis (the endosymbiont is a
cell possessing a secondary plastid) and serial secondary endosymbiosis (a
lineage possessing one type of secondary plastid replaces its secondary
plasmid with a secondary plastid of a different lineage).
Endosymbiotic gene transfer (EGT): a special case of horizontal gene transfer
(see below), whereby the gene in question is acquired by the host lineage from
the genome of the endosymbiont.
Evolutionary cell biology: an emerging discipline that incorporates compara-
tive perspectives and techniques from cell biology, protistology, molecular
evolution, and mathematical evolutionary theory to address questions of the
origins and diversity of cells.
First eukaryotic common ancestor (FECA): the cell (or population of cells)
belonging to the lineage that gave rise to the modern line of eukaryotes at the
earliest point at which it possessed cell biological features distinct from those
in prokaryote-like cells. Although this organism is deduced to have existed, a
useful way to treat the FECA is as a theoretical reconstruction with the traits
defining it as an exciting open research question.
Horizontal gene transfer: the acquisition of a gene by a genome froma source
other than the immediate parental lineage.
Last eukaryotic common ancestor (LECA): the cell (or population of cells)
belonging to the lineage that gave rise to the modern line of eukaryotes at the
latest point at which the various descendent lineages diverged to leave the
extant eukaryotic lineages. Again, this concept is most useful as a theoretical
reconstruction or reference point to assess the antiquity of various cell
biological features.
Monophyletic: a group is considered monophyletic when it encompasses all
descendants of a single ancestor.
Paraphyletic: a group is considered paraphyletic when it encompasses some,
but not all, descendants of a single ancestor.
Paralog: genes that are the result of a gene duplication process.
Selection: the process by which a factor (including the presence of another
organism) presents a circumstance that results in the preferential death of
some organisms in the environment over others.
0962-8924/$ see front matter
2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tcb.2014.02.003
Corresponding author: Dacks, J.B. (dacks@ualberta.ca).
Keywords: constructive neutral evolution; endosymbiosis; evolutionary cell biology;
organelle paralogy hypothesis; protocoatomer; transfer-window hypothesis.
Trends in Cell Biology, July 2014, Vol. 24, No. 7 435
Signicant progress has been made in revealing the
mechanisms underpinning the emergence of complexity
in the eukaryotic cell. The number and sophistication of
internal compartments in eukaryotes distinguish them
from prokaryotes and it is this sophistication that trans-
lates into vastly greater complexity in eukaryotic cellular
conguration, which is measured by the number of com-
ponents in a system and the number of interactions be-
tween them [4,5] compared with another, equivalent
system (e.g., cell versus cell). There is general agreement
from both molecular phylogenetic analyses [6,7] and evi-
dence from the fossil record ([8,9], but see [10]) that pro-
karyotes predated eukaryotes. This evidence suggests that
eukaryotes must have arisen from a state resembling that
of a prokaryote; that is, that the acquisition of organelles
and complex cellular machines in eukaryotes must be
explained from a cellular state lacking the extensive pres-
ence of these features. Evolution from less complexity
(prokaryote state) to increased complexity of cellular ar-
chitecture (eukaryote state) can be divided into at least
three successive evolutionary stages (Figure 1). First, a
transition from an organism lacking some or all internal
membranes to an organism that possesses some cellular
features that dene it as eukaryotic [i.e., a rst eukaryotic
common ancestor (FECA)] must occur. Next, the organism
must transition from the FECA to a last eukaryotic com-
mon ancestor (LECA). Finally, evolution and divergence
after the LECA must occur to form the major lineages of
extant eukaryotes [2,11].
Research on the prokaryote-to-eukaryote transition
(e.g., [1]) has reconstructed a surprisingly sophisticated
LECA possessing a well established actin/tubulin cytoskel-
eton, an elaborate endomembrane system, a nucleus,
mitochondria, and machinery for processes such as intron
splicing and meiosis (Figure 1). Furthermore, at least one
additional major cellular innovation has inuenced post-
LECA increases in complexity: the acquisition of plastids.
Therefore, an important question for evolutionary cell
biology is what mechanisms drive increased cellular com-
plexity at the level of molecular machines and the forma-
tion of organelles. Here we review the progress that has
been made in addressing this question.
Mechanisms for organelle acquisition
Before 1974, the null hypothesis for the evolution of inter-
nal membrane-bound compartments (organelles) within
the eukaryotic cell was an autogenous process; that is,
eukaryotic cells are built in a stepwise manner from indi-
vidual building blocks present in the pre-eukaryotic ances-
tor [12,13]. However, evidence demonstrating that both
mitochondria and chloroplasts are of bacterial origin (en-
dosymbiosis) [14,15] shifted the theoretical basis of the
eukaryotic evolutionary eld such that both endosymbiotic
and autogenous explanations are now viable alternatives
to entertain when addressing the origins of a eukaryotic
organelle. Our scientic understanding of both mecha-
nisms is maturing, with endosymbiosis being far better
understood. Below, we discuss the current standing on the
emergence of complexity of eukaryotic cells through endo-
symbiosis and autogenous processes.
Endosymbiosis
Endosymbiosis is the incorporation and residence of one
organism, the endosymbiont, inside another, the host.
Beyond the familiar examples of endosymbiosis (i.e., the
chloroplast of plants and the nearly ubiquitous mitochon-
drion), a diverse array of organisms with organelles de-
rived from additional endosymbiotic events also exists
Prokaryote
First eukaryoc common ancestor
(FECA)
Last eukaryoc common ancestor
(LECA)
Opisthokonta
Amoebozoa
Excavata
Archaeplasda
SAR
CCTH
Extant eukaryoc lineages
(A) (B)
TRENDS in Cell Biology
Figure 1. Retracing the emergence of modern eukaryotes. (A) The emergence of eukaryotes can be divided into at least three successive evolutionary stages: (i) transition
from a prokaryote-like starting point to the first eukaryotic common ancestor (FECA); (ii) transition from the FECA to the last eukaryotic common ancestor (LECA); and (iii)
evolution and divergence after the LECA to form the major lineages of eukaryotes as we know them today [2,11]. (B) After the LECA, six extant eukaryotic lineages (Box 1)
have been characterized [11]. These are the Opisthokonta, Amoebozoa, Excavata, and Archaeplastida, the stramenopiles, alveolates and rhizarians (SAR), and
the contentious grouping of cryptomonads, centrohelids, telonemids, and haptophytes (CCTH). Cartoon images of representative organisms fromeach supergroup are not
to scale.
Review
Trends in Cell Biology July 2014, Vol. 24, No. 7
436
(Figure 2A). Recent cell biological and genomic studies of
these organisms have revealed much about the mechanism
of endosymbiosis. Indeed, one reason why endosymbiosis is
better understood than autogenous mechanisms of organ-
elle acquisition is the wealth of endosymbiotic intermedi-
ates available for study (e.g., [16]). Recent and independent
occurrences of endosymbiosis have revealed the earliest
stages of the process, including several examples of prima-
ry (e.g., Paulinella) and secondary (e.g., Hatena) plastid-
derived organelles as well as transiently acquired plastids
termed kleptoplasts [16]. Other examples of where the host
and symbiont are at the beginning of their integration
include dinotoms, algae wherein the host lineage is a
dinoagellate that possesses a minimally reduced diatom
endosymbiont. Recent work has begun to uncover the
extent and nature of their organellar and metabolic inte-
gration [17]. Genome sequencing of organisms such as the
cryptophyte and chlorarachniophyte algae, whose photo-
synthetic organelle contains both a secondary plastid ge-
nome and the remnant of the red or green algal nuclear
genome (nucleomorph) and cytoplasm [18], have also
allowed examination of genome reduction and cellular
integration in endosymbiosis (Box 2). Because photosyn-
thesis has been gained, stolen and co-opted throughout the
history of eukaryotes, the acquisition of plastids has been a
particularly useful model for understanding the early
stages of endosymbiosis. However, a rare example of a
potential secondary mitochondrial endosymbiont has re-
cently been described by genomic methods [19]. The sh
pathogen Neoparamoeba contains what appears to be an
intracellular symbiont related to Ichthyobodo necator, a
kinetoplastid. Although the atypical mitochondrion of this
symbiont occupies nearly half of its cytoplasmic volume,
the extent to which the endosymbiont has progressed to
become an organelle is unclear. The nuclear genome of the
symbiont does not appear to have undergone extensive
reduction compared with that of other kinetoplastids. All of
these examples help focus the question of how reduced an
endosymbiont has to be for it to be considered an organelle
and no longer an organism.
At the other extreme of endosymbiotic integration exist
organelles apparently reduced from the canonical eukary-
otic state, such as the non-photosynthetic apicoplasts of
apicomplexans [16] and the hydrogenosomes and mito-
somes, some of which no longer possess organellar gen-
omes. Initially these latter organelles were seen as distinct
classes; however, recent studies have clearly established
them as derivatives of mitochondria and found various
intermediates possessing aerobic or anaerobic metabo-
lisms and different genomic organizations (e.g., [20]).
The range of genomic and cytoplasmic minimalization
found for endosymbiotically derived organelles raises the
question of what mechanism determines and limits the
extent of this reductive trend in any given lineage. The
passage of time cannot explain this reduction because a
wide range of reduction is observed in organelles clearly
derived from the same founding event (e.g., the mitochon-
drion). However, it was proposed that because the main
Box 1. Eukaryotic diversity
Eukaryotic diversity (Figure 1B in main text) is currently divided into
six large taxonomic groupings, or supergroups [2]. The Opistho-
konta encompasses the lineages of animals and fungi, as well as
their single-celled relatives. The Amoebozoa houses a diversity of
amoeboid lineages with, and without, flagellated stages. It includes
the pathogens Balamuthia, Acanthamoeba, and, most famously,
Entamoeba histolytica, the causative agent of amoebic dysentery.
The Opisthokonta and Amoebozoa are united in large-scale
molecular phylogenetic analyses and thought to represent a
monophyletic grouping, named the Amorphea [2]. The Archae-
plastida incorporates the lineages of red algae, green algae
(including land plants), and the glaucophytes, which are derived
from a single founding primary-endosymbiotic event. The SAR
clade unites the seemingly disparate lineages of stramenopiles
(diatoms, brown algae, and the causative agent of the Irish Potato
Famine, Phythophthora) and alveolates (ciliates like Paramecium,
the dinoflagellates that cause red tides, and apicomplexans such as
Plasmodium, which causes malaria). The supergroup Excavata
includes important disease-causing agents such as Trypanosoma,
Leishmania, Giardia, and Trichomonas, as well as their free-living,
or nonpathogenic, relatives. Finally, the CCTH supergroup currently
contains the lineages of cryptophytes, centrohelids, telonemids, and
haptophytes; however, the most recent large-scale molecular-
evolutionary analyses have cast doubt on the unity of these in a
single group [3] and the CCTH should be treated as tentative at best.
Box 2. Endosymbiosis
The types of endosymbiosis are classified based on the nature of the
host and of the endosymbiont. The simplest form, or primary
endosymbiosis, involves a eukaryotic host and a bacterial endo-
symbiont (Figure 2A in main text). Two such primary events have
been transformative in the history of eukaryotes and involved the
incorporation of an a-proteobacterium and a cyanobacterium to
give rise to mitochondrion- and plastid-derived organelles, respec-
tively. Both events are known to have occurred early in eukaryotic
history, with the mitochondrial event now convincingly shown to
have predated the LECA [20]. A primary plastid endosymbiosis is
very likely to have occurred at the base of the Archaeplastida
lineage, conferring photosynthetic capacity and giving rise to all red
and green algae and land plants. The photosynthetic ability was
clearly advantageous, as it spawned the subsequent evolution of
complex plastids [16] through secondary and tertiary endosym-
bioses (Figure 2A in main text).
As a mechanism, the process of endosymbiosis can be divided
into initiation and integration. Initiation may stem from various
possible microbial associations, including mutualistic exchange of
metabolites, intracellular invasion of the host by a parasite, or
predatory ingestion of an eventual endosymbiont by a phagotroph.
After initiation, the success of the resulting chimera depends on the
ability to synchronize the cell growth and division cycles of the host
and endosymbiont. In all cases, gradual transfer of genetic material
from the endosymbiont genome to the host genome promotes this
synchronization (Figure 2B in main text). This ratchet-like mechan-
ism of EGT drives the establishment of an obligate relationship
between the endosymbiont and its host. After acquiring an
endosymbiont, the organism has two genomes, one in the nucleus
and one in the endosymbiont (Figure 2B, i in main text). Whether by
lysis or by improper fission and fusion events of the endosymbiont
during replication, endosymbiont DNA released into the cytoplasm
can be integrated into the host genome (Figure 2B, ii in main text).
With an endosymbiont gene now encoded and expressed by the
host, it must be successfully retargeted to the endosymbiont
(Figure 2B, iii in main text). When this occurs, the endosymbiont
copy is redundant and sustains mutational decay, and the
endosymbiont genome is reduced (Figure 2B, iv in main text). The
directionality imposed by this transfer results in an iterative ratchet-
like mechanism. The window of opportunity permitting EGT
remains open until only a single endosymbiont genome remains
(Figure 2B, v and vi in main text) [16,20]. Loss of genes, coincident
with loss of function, in the organism-to-organelle transition is also
a major source of genome reduction [16,20].
Review Trends in Cell Biology July 2014, Vol. 24, No. 7
437
Independent acvity Binding and presuppression Mutaon and dependence Ratchet-like increase in dependence
Acvity A
Factor A Factor A
Factor B
Acvity A
Factor A
Factor B
Acvity A
Factor A
Factor B
Acvity A
(i)
(i) (ii) (iii)
(ii) (iii) (iv)
x
y
z
x
y
z
x
y
z
x y
z
x y
z
x
y
z
(v)
x
x
y
z
x
y
z
(vi)
x
y
z
y
z
y
z
y
z
x
z
x
y
Endosymbiosis (A)
(B)
(C)
(D)
EGT and transfer window hypothesis
Organelle paralogy hypothesis
Construcve neutral evoluon
Primary Secondary Terary
(i) (ii) (iii)
(i) (ii) (iii) (iv)
TRENDS in Cell Biology
Figure 2. Mechanisms of cellular evolution. (A) The variety of plastids arising from iterative acquisition of photosynthetic endosymbionts. (i) Primary endosymbiosis is
established following engulfment of a cyanobacterium (green) by a eukaryotic host cell. A similar primary endosymbiotic process would also have produced the
mitochondrion from a proteobacterium. (ii) In secondary plastid endosymbiosis, a green or red algal cell is engulfed by a new host cell. (iii) This process is repeated in
tertiary endosymbiosis. Although primary, secondary, and tertiary endosymbioses are conceptually interconnected, they are not consecutive steps of a single colonization.
(B) The steps of endosymbiotic gene transfer (EGT) from a newly acquired endosymbiont: lysis of the endosymbiont and (ii) transfer of the gene to the host nucleus; (iii)
retargeting and (iv) endosymbiont-encoded gene loss; and (v) repetition until (vi) a single endosymbiont remains. (C) The organelle-paralogy hypothesis (OPH). (i) Different
protein families interact cooperatively to specify organelle-defining properties such as tethering, docking, fission, or fusion. (ii) Specificity-encoding protein families evolve
by gene duplication and divergence, as represented by this hypothetical phylogeny. (iii) Increases in the complexity of specificity-encoding protein families are mirrored by
increases in the complexity of the membrane-trafficking system. Paralogs of the specificity-encoding protein family reside in and have their effect on distinct compartments.
Modified from[59]. (D) A generalized outline of constructive neutral evolution (CNE). (i) Protein factor A possesses a given activity. (ii) Through randomsteric collisions, a
stochastic interaction with a separate factor B occurs that has little or no effect on the activity of factor A. (iii) A mutation (represented by a yellow star) occurs in factor A
that reduces its activity, but due to the interaction of factor A with factor B, the mutation is suppressed and the activity of factor A is maintained at near-original levels. This
could be due to stabilization of the structure of factor A, masking of its charge or exposed hydrophobic residues, or altered localization of factor A allowing better access to
its substrate. (iv) Subsequent mutations of the original factor A, and compensatory mutations in the interacting factor B, further integrate factor B in the activity of factor A
via a ratchet-like mechanism that may also lead to the recruitment of additional factors.
Review
Trends in Cell Biology July 2014, Vol. 24, No. 7
438
mechanism of DNA transfer to the host nucleus comes from
lysed organelles, the rate of transfer is proportional to the
copy number of the endosymbiont in the cell (Figure 2B).
This idea became known as the transfer-window hypothe-
sis [21], which implies that transfer cannot continue once
the number of organelles has reached a single copy. Indeed,
experimental [22] and comparative [23] genomic analyses
revealed far fewer transfers from plastids to nuclear gen-
omes in organisms possessing a low plastid copy number.
In addition, the nuclear genomes of a cryptophyte and of a
chlorarachniophyte alga, each possessing a single nucleo-
morph, have been sequenced and were reported in 2012
[24]. These ndings revealed a complete lack of recent DNA
transfer from either the plastid or the nucleomorph ge-
nome despite evidence of transfer from mitochondria,
which is consistent with a reduction to a single organelle
that is responsible for halting endosymbiotic gene transfer
(EGT) and hence organelle reduction.
Endosymbiosis has repeatedly allowed for increased
overall complexity in eukaryotic cells compared with their
pre-merged state. Ironically, because the cell is at its most
complex state immediately after endosymbiosis begins,
with integration progressing principally via EGT or gene
loss, the process of endosymbiosis actually involves
decreases in complexity.
Autogenous (non-endosymbiotic) organelles
Although endosymbiosis has undoubtedly been a powerful
force in building some aspects of eukaryotic cellular com-
plexity, it does not explain them all. A simpler, alternative
explanation for the origin of organelles delimited by a
single lipid bilayer and devoid of genetic material is that
they are autogenous. The organelles most commonly
proposed to have an autogenous origin are those of the
membrane-trafcking system, including the endoplasmic
reticulum (ER), Golgi apparatus, endosomes, and plasma
membrane [25]. Although these endomembrane organelles
are dynamically connected to one another, they are main-
tained as distinct compartments through the action of
membrane trafcking machineries such as Rabs, SNAREs,
coatomer, and adaptin (AP) complexes [26]. These speci-
city-encoding protein families have different members
that perform the same function (e.g., inducing membrane
curvature or facilitating membrane fusion) at distinct
locations within the membrane-trafcking system [26].
Although each protein family could play an individual role,
part of the information encoding specicity in membrane
trafcking appears to result from combinatorial protein
protein interactions between members of the different
families [27]. Comparative genomic and phylogenetic anal-
yses of these various protein families have revealed details
of their primary diversication by gene duplication (e.g.,
[1]). Surprisingly, the duplications giving rise to paralogs of
the various specicity-encoding proteins associated with
each cellular location occurred before the LECA. However,
examination of the endocytic paralogs of the SNARE, Rab,
and AP families revealed a pattern whereby some organ-
elle-specic paralogs had not duplicated before the LECA,
with parallel duplications occurring instead in lineages
after the LECA [28]. These patterns provide an under-
standing of the timing of these events and suggest a
possible mechanism underpinning them, which is formal-
ized in the organelle-paralogy hypothesis (OPH) [28,29].
The OPH (Figure 2C) proposes that a set of specicity-
encoding proteins with complementary functions that de-
ne organelle properties produce sets of interacting para-
logs by undergoing duplications. Through coevolution,
these sets of specicity-encoding proteins accumulate
mutations that x their specic functional binding, thus
dening separate organelles [30]. Iterations of this process
could therefore account for the array of organelles in the
endomembrane systems of extant eukaryotes that arose
via differentiation from an original prototypical internal
compartment in the FECA.
Recently, the OPH has been tested by computer simu-
lation. Mathematical modeling of specicity-encoding
genes in populations of vesicles showed that gene duplica-
tion and differential interactions between paralogs pro-
duced novel vesicular compartments [31]. The OPH further
predicts that the order of evolutionary emergence for each
member of a specicity-encoding protein family should
correspond to the order of emergence of the different
organelles they dene and on which they have effect.
Two recent studies have reported phylogenetic resolution
for important specicity-encoding protein families, thereby
allowing hypotheses to be proposed based on empirical
evidence regarding an order of evolutionary emergence
beyond the establishment of extensive complexity in mem-
brane trafcking in the LECA. AP complexes aid in sorting
the vesicular trafc between organelles found between and
including the plasma membrane and the trans-Golgi net-
work [32,33]. Comparative genomic and phylogenetic anal-
ysis resolved the order of emergence of the members of the
AP complex family, with AP3 and AP5 rst diverging from
the remaining AP complexes, followed by AP4 and AP1/2
[32]. Based on their known locations of action, this order
suggests that adaptins rst acted at an organellar inter-
face between the secretory system and the phagocytic
system, before the establishment of the trans-Golgi net-
work. In addition, recent evidence provides clues to the
conservation among the Rab family of GTPases, which are
molecular switches involved in specifying organelle iden-
tity in the membrane-trafcking system [34]. Although it is
well established that Rab GTPases are ancient and that
the LECA possessed a large complement of such proteins
[35], the extent to which Rab families are conserved
remained unknown. Rigorous homology searching resulted
in the expansion of the Rab complement in LECA to 15
subfamilies [36]. However, robust phylogenetic resolution
between the paralogs of the Rab gene families increased
the estimated number of Rab subfamilies in the LECA to
between 19 and 23 [37]. Surprisingly, this analysis also
revealed two ancient sets of Rabs, one inferred to be
involved in exocytosis and one predominantly in endocyto-
sis, potentially reecting the earliest establishment of
these pathways. As improved comparative and phyloge-
netic methods are applied to other trafcking families, it
will be important to compare the evolutionary patterns
that emerge and to delve further into events pre-LECA.
Although the OPH is a mechanism for evolving in-
creased compartment number and specialization within
an organellar system, it is currently limited to the
Review Trends in Cell Biology July 2014, Vol. 24, No. 7
439
membrane-trafcking system. However, an idea that com-
plements the OPHis the protocoatomer hypothesis, which
proposes, based on protein-structural evidence, that ho-
mology exists between the membrane deformation com-
ponents of vesicular trafcking and the nuclear pore [38].
Specically, proteins integrated into the COP I, COP II,
clathrin, and nuclear pore complexes share a structure of
b-propellers followed by a-solenoid domains. These pro-
teins are suggested to be homologous and therefore de-
rived from a single ancestral protocoatomer protein [39].
Recent analyses have also rmly established relation-
ships between protocoatomer-derived proteins of the
intraagellar complex [40]. These proteins, which are
dispersed throughout the cell and essential for organ-
elle-specic functions, appear to have expanded along
with their organelles via the process described in the
OPH. Therefore, the overlap between the two hypotheses
extends the mechanism of autogenous organelle evolution
to potentially all organelles for whicha non-endosymbiotic
origin appears likely.
Examples exist of organelles whose origins blur the
divisions of autogenous and endosymbiotic organellar evo-
lution. The origin of the peroxisome has been contentiously
explained by both mechanisms. Although the evidence,
both functionally [41] and evolutionarily [42], strongly
favors an autogenous origin for peroxisomes, there have
undoubtedly been, and continue to be, molecular and
functional interactions between peroxisomes and orga-
nelles of endosymbiotic origin, notably the mitochondrion
[43]. Many proteins that localize to the peroxisome are
encoded by genes of bacterial origin and function in meta-
bolic processes shared with mitochondria (e.g., fatty-acid
oxidation). Determining how endosymbiotic organelles
have become integrated within the cell and interact with
non-endosymbiotically derived systems is an emerging
area of investigation for cell biology and evolutionary cell
biology. Work in the past few years has uncovered several
protein complexes mediating protein, lipid, and ion trans-
port between the ER and mitochondria [44] and it was
recently shown that protein complexes bridging the ER
and mitochondria in fungi are more widely present in
eukaryotes than previously suspected [45,46].
Constructive neutral evolution (CNE)
Evolutionary processes are not limited to the organellar
level. Individual cellular machines in the eukaryote (e.g.,
ribosomes, proteasomes) also show increased complexity
over their prokaryotic counterparts. In some cases, this
increased complexity could result in new functions, pro-
viding a selective advantage to the eukaryotic cell. How-
ever, the role of selection as the only driver in the evolution
of complexity is increasingly being questioned.
The theory of CNE [47] posits that many biological
phenomena can arise, or be elaborated on, by neutral
evolutionary processes that promote increased complexity
without additional functionality [48]. CNE is predicated on
an idea of presuppression (Figure 2D); that is, interactions
between factors that are the initial result of random colli-
sions or cytosolic overcrowding and that minimally affect
function [49] may become stabilized due to random muta-
tion in a factors partner or in both factors. On their own,
these mutations may be slightly deleterious for the origi-
nal function, but if binding of the partner restores func-
tionality, the interaction becomes xed. Therefore, the
mutation is not selective in the traditional sense but needs
to be sufciently compensatory to avoid negative selection
and to allow the organism to survive. These mutations
may be extremely rare; nevertheless, once established
they result in a ratchet that promotes tighter binding
and, potentially, recruits other factors. These interactions
could involve protein interactions with nearly any mole-
cule or surface in the cell (e.g., diffusible small molecules,
cellular membranes).
Among these biological phenomena, the origin of the
spliceosome has been proposed to require CNE [47,48].
Comparative genomic studies of spliceosomal components
have demonstrated that the spliceosome is a eukaryotic
innovation that was present in its highly elaborate state
before the LECA [50]. Comprising well over 100 different
protein and RNA components, the spliceosome is a candi-
date for one of the most complex cellular machines in
existence. However, it has long been appreciated that
the underlying essential process could have evolved from
a simple self-splicing group II-class intron. Rather than
being the result of selective forces, the spliceosome is best
explained as a product of CNE whereby mutations in the
self-splicing RNA molecule were suppressed through a pre-
existing interaction with a RNA or RNA/protein complex
[47,48]. As protein and RNA components accumulated over
time, the basic function of splicing remained unaltered.
A recently well elaborated example involving experi-
mental testing of hypothesized CNE processes is the vacu-
olar V
0
-ATPase ring of yeast [49,51]. Although the ancestor
of the yeast V
0
-ATPase ring comprises two subunits, sev-
eral yeasts require three subunits, with the third subunit
resulting from an ancient gene duplication that was fol-
lowed by gene suppression. To verify this sequence of
events, investigators reconstructed the common ancestral
gene of extant two-subunit rings and three-subunit rings
and revealed specic suppressive interactions required to
enforce the adoption of the three-subunit system [51].
When suppression succeeds, the system, because of this
dependency, is more complex; however, the net effect of the
increased complexity remains neutral in that no altera-
tions in the cells ability to produce the phenotype have
occurred. Therefore, CNE allows the accumulation of
greater complexity combined with a dilution of responsi-
bility for maintaining a phenotype among multiple factors.
Ironically, this dilution, via redundant functionality of
components, would reduce the risk of negative selection
on a single mutational target and, as such, the CNE
mechanism itself may be under positive selective pressure
[48].
Concluding remarks
The above overview was organized into processes acting at
the level of the organelle or at the level of the underlying
molecular complex, but such divisions are by no means
absolute. Molecular machineries clearly cooperate to build
and dene organelles. At the same time, the compartmen-
talization of specic molecular machineries within a given
organelle limits the range of proteins with which these
Review
Trends in Cell Biology July 2014, Vol. 24, No. 7
440
molecular machines will frequently interact and thereby
increases the opportunities for distinct environments that
would lead to complexity via CNE mechanisms.
At present, tremendous opportunities exist for the ad-
vancement of evolutionary cell biology as a discipline.
While the eld brings evolutionary biology from the popu-
lation and the large organism down to the scale of the cell,
it also brings a comparative approach over species and
space to cell biologists focused on specic organisms or
organelles. However, there is also a potential for miscon-
ceptions. In many ways, the study of cell biology shares
conceptual commonalities with the discipline of reverse
engineering [52]. Cell biology is typically understood from
a reductionist approach whereby the cell is disassembled,
both conceptually and physically, into its components
(proteins, organelles, and complexes) and then laid out,
manipulated, and understood. Therefore, it is unsurprising
that questions regarding evolutionary mechanisms that
give rise to cells are sometimes misframed as a forward-
engineering problem; that is, How did the cell nd the
most efcient way of performing process x? However,
there is a fundamental difference between evolution and
engineering. Evolution does not always proceed along an
optimized path leading to the observed modern state.
Viewing each trait as the result of an iterative and
mechanistic, rather than teleological, process leading to
these solutions changes the way investigations are under-
taken and data are interpreted. Although it may remain
useful to ask What is the selective advantage of a given
trait?, knowing that the evolutionary path is not always
direct and constant allows the investigator to consider
multiple advantages and possibly entertain alternative
explanations beyond selection. Therefore, it may be more
productive to answer the how behind evolutionary cell
biological questions and to reconstruct the steps and evo-
lutionary details for the emergence of a given trait, thus
deriving process from the patterns observed across multi-
ple examples.
Although signicant progress has been made in devel-
oping model cell-biological systems across eukaryotes (e.g.,
Dictyostelium, Toxoplasma, Trypanosoma, Arabidopsis)
and analyzing molecular evolution to deduce the origins
of protein complexes and their resident organelles, yielding
some of the discoveries described above, many areas re-
main unexplored. For example, the consequences of popu-
lation genetics have not been fully explored in the context
of cellular evolution [53]. Similarly, although there have
been attempts to correlate geology with cellular evolution
[54], particularly regarding the origin of life [55], this
aspect is often overlooked by cell biologists. Furthermore,
the mechanisms of emergence of evolutionary innovations,
such as organelle inheritance, that combine multiple, well
adapted cellular components remains to be better eluci-
dated [56]. Finally, as our understanding of systems biolo-
gy matures and omic data types become increasingly
available, we will be able to integrate information about
the timing and context of genes and proteins into various
models of cellular evolution [57].
With tractable progress being made on concrete mecha-
nistic questions, this is truly an exciting time as the biology
of the cell can now be parsed in the light of evolution [58].
Acknowledgments
The authors thank the members of the Dacks laboratory, as well as W. Ford
Doolittle and Holly Goodson, for critical comments on the manuscript and
for discussion. J.B.D. and L.D.B. also thank the staff at the Banff Centre for
the Arts for their generosity and hospitality during the ooding that
occurred in Alberta in June 2013, at which time signicant work on the
writing of this manuscript took place. F.D.M. is the recipient of a Vanier
Canada Graduate Scholarship from the Canadian Institutes of Health
Research (CIHR) and a Full-Time Studentship from Alberta Innovates
Health Solutions. L.D.B. was supported by a National Science and
Engineering Council of Canada (NSERC) Undergraduate Student Re-
search Award. J.B.D. is Canada Research Chair (Tier II) in Evolutionary
Cell Biology. Research in the Rachubinski laboratory is supported by grants
9208, 15131, and 53326 fromthe CIHR. Research in the Dacks laboratory is
supported by a NSERC discovery grant and an Alberta Innovates
Technology Futures New Investigator Award to J.B.D.
References
1 Koumandou, V.L. et al. (2013) Molecular paleontology and complexity
in the last eukaryotic common ancestor. Crit. Rev. Biochem. Mol. Biol.
48, 373396
2 Adl, S.M. et al. (2012) The revised classication of eukaryotes. J.
Eukaryot. Microbiol. 59, 429493
3 Burki, F. et al. (2012) The evolutionary history of haptophytes and
cryptophytes: phylogenomic evidence for separate origins. Proc. Biol.
Sci. 279, 22462254
4 Koonin, E. (2011) The Logic of Chance, FT Press
5 McShea, D.W. (2002) A complexity drain on cells in the evolution of
multicellularity. Evolution 56, 441452
6 Williams, T.A. et al. (2013) An archaeal origin of eukaryotes supports
only two primary domains of life. Nature 504, 231236
7 Kelly, S. et al. (2010) Archaeal phylogenomics provides evidence in
support of a methanogenic origin of the Archaea and a thaumarchaeal
origin for the eukaryotes. Proc. Biol. Sci. 278, 10091018
8 Schopf, J.W. (1999) Deep divisions in the Tree of Life what does the
fossil record reveal? Biol. Bull. 196, 351353 discussion 354355
9 Javaux, E.J. (2007) The early eukaryotic fossil record. Adv. Exp. Med.
Biol. 607, 119
10 Javaux, E.J. et al. (2010) Organic-walled microfossils in 3.2-billion-
year-old shallow-marine siliciclastic deposits. Nature 463, 934938
11 Walker, G. et al. (2011) Eukaryotic systematics: a users guide for cell
biologists and parasitologists. Parasitology 138, 16381663
12 Cavalier-Smith, T. (1975) The origin of nuclei and of eukaryotic cells.
Nature 256, 463468
13 De Duve, C. and Wattiaux, R. (1966) Functions of lysosomes. Annu.
Rev. Physiol. 28, 435492
14 Gray, M.W. and Doolittle, W.F. (1982) Has the endosymbiont
hypothesis been proven? Microbiol. Rev. 46, 142
15 Sagan, L. (1967) On the origin of mitosing cells. J. Theor. Biol. 14,
255274
16 Keeling, P.J. (2010) The endosymbiotic origin, diversication and fate
of plastids. Philos. Trans. R. Soc. Lond. B: Biol. Sci. 365, 729748
17 Imanian, B. et al. (2012) Tertiary endosymbiosis in two dinotoms has
generated little change in the mitochondrial genomes of their
dinoagellate hosts and diatom endosymbionts. PLoS ONE 7, e43763
18 Moore, C.E. and Archibald, J.M. (2009) Nucleomorph genomes. Annu.
Rev. Genet. 43, 251264
19 Tanifuji, G. et al. (2011) Genomic characterization of Neoparamoeba
pemaquidensis (Amoebozoa) and its kinetoplastid endosymbiont.
Eukaryot. Cell 10, 11431146
20 Muller, M. et al. (2012) Biochemistry and evolution of anaerobic energy
metabolism in eukaryotes. Microbiol. Mol. Biol. Rev. 76, 444495
21 Barbrook, A.C. et al. (2006) Why are plastid genomes retained in non-
photosynthetic organisms? Trends Plant Sci. 11, 101108
22 Lister, D.L. et al. (2003) DNA transfer from chloroplast to nucleus is
much rarer in Chlamydomonas than in tobacco. Gene 316, 3338
23 Smith, D.R. et al. (2011) Correlation between nuclear plastid DNA
abundance and plastid number supports the limited transfer window
hypothesis. Genome Biol. Evol. 3, 365371
24 Curtis, B.A. et al. (2012) Algal genomes reveal evolutionary mosaicism
and the fate of nucleomorphs. Nature 492, 5965
25 de Duve, C. (2007) The origin of eukaryotes: a reappraisal. Nat. Rev.
Genet. 8, 395403
Review Trends in Cell Biology July 2014, Vol. 24, No. 7
441
26 Bonifacino, J.S. and Glick, B.S. (2004) The mechanisms of vesicle
budding and fusion. Cell 116, 153166
27 Cai, H. et al. (2007) Coats, tethers, Rabs, and SNAREs work together to
mediate the intracellular destination of a transport vesicle. Dev. Cell
12, 671682
28 Dacks, J.B. et al. (2008) Phylogeny of endocytic components yields
insight into the process of nonendosymbiotic organelle evolution. Proc.
Natl. Acad. Sci. U.S.A. 105, 588593
29 Dacks, J.B. and Field, M.C. (2007) Evolution of the eukaryotic
membrane-trafcking system: origin, tempo and mode. J. Cell Sci.
120, 29772985
30 Dacks, J.B. et al. (2009) Evolution of specicity in the eukaryotic
endomembrane system. Int. J. Biochem. Cell Biol. 41, 330340
31 Ramadas, R. and Thattai, M. (2013) New organelles by gene
duplication in a biophysical model of eukaryote endomembrane
evolution. Biophys. J. 104, 25532563
32 Hirst, J. et al. (2011) The fth adaptor protein complex. PLoS Biol. 9,
e1001170
33 Hirst, J. et al. (2012) Adaptor protein complexes AP-4 and AP-5: new
players in endosomal trafcking and progressive spastic paraplegia.
Trafc 14, 153164
34 Grosshans, B.L. et al. (2006) Rabs and their effectors: achieving
specicity in membrane trafc. Proc. Natl. Acad. Sci. U.S.A. 103,
1182111827
35 Pereira-Leal, J.B. and Seabra, M.C. (2001) Evolution of the Rab family
of small GTP-binding proteins. J. Mol. Biol. 313, 889901
36 Diekmann, Y. et al. (2011) Thousands of Rab GTPases for the cell
biologist. PLoS Comput. Biol. 7, e1002217
37 Elias, M. et al. (2012) Sculpting the endomembrane system in deep
time: high resolution phylogenetics of Rab GTPases. J. Cell Sci. 125,
25002508
38 Field, M.C. et al. (2011) Evolution: on a bender BARs, ESCRTs, COPs,
and nally getting your coat. J. Cell Biol. 193, 963972
39 Devos, D. et al. (2004) Components of coated vesicles and nuclear pore
complexes share a common molecular architecture. PLoS Biol. 2, e380
40 van Dam, T.J. et al. (2013) Evolution of modular intraagellar
transport from a coatomer-like progenitor. Proc. Natl. Acad. Sci.
U.S.A. 110, 69436948
41 Titorenko, V.I. and Rachubinski, R.A. (2009) Spatiotemporal dynamics
of the ER-derived peroxisomal endomembrane system. Int. Rev. Cell
Mol. Biol. 272, 191244
42 Schluter, A. et al. (2006) The evolutionary origin of peroxisomes: an
ERperoxisome connection. Mol. Biol. Evol. 23, 838845
43 Mohanty, A. and McBride, H.M. (2013) Emerging roles of mitochondria
in the evolution, biogenesis, and function of peroxisomes. Front.
Physiol. 4, 268
44 Michel, A.H. and Kornmann, B. (2012) The ERMES complex and ER
mitochondria connections. Biochem. Soc. Trans. 40, 445450
45 Flinner, N. et al. (2013) Mdm10 is an ancient eukaryotic porin co-
occurring with the ERMES complex. Biochim. Biophys. Acta 1833,
33143325
46 Wideman, J.G. et al. (2013) The ancient and widespread nature of
the ERmitochondria encounter structure. Mol. Biol. Evol. 30, 2044
2049
47 Stoltzfus, A. (1999) On the possibility of constructive neutral evolution.
J. Mol. Evol. 49, 169181
48 Lukes, J. et al. (2011) How a neutral evolutionary ratchet can build
cellular complexity. IUBMB Life 63, 528537
49 Doolittle, W.F. (2012) Evolutionary biology: a ratchet for protein
complexity. Nature 481, 270271
50 Koonin, E.V. (2009) Intron-dominated genomes of early ancestors of
eukaryotes. J. Hered. 100, 618623
51 Finnigan, G.C. et al. (2012) Evolution of increased complexity in a
molecular machine. Nature 481, 360364
52 Hartwell, L.H. et al. (1999) From molecular to modular cell biology.
Nature 402, C47C52
53 Lynch, M. (2012) The evolution of multimeric protein assemblages.
Mol. Biol. Evol. 29, 13531366
54 Cavalier-Smith, T. (2006) Cell evolution and Earth history: stasis
and revolution. Philos. Trans. R. Soc. Lond. B: Biol. Sci. 361, 969
1006
55 Lane, N. and Martin, W.F. (2012) The origin of membrane
bioenergetics. Cell 151, 14061416
56 Mast, F.D. et al. (2012) Emergent complexity in myosin V-based
organelle inheritance. Mol. Biol. Evol. 29, 975984
57 Ryan, C.J. et al. (2012) Hierarchical modularity and the evolution of
genetic interactomes across species. Mol. Cell 46, 691704
58 Dobzhansky, T. (1973) Nothing in biology makes sense except in the
light of evolution. Am. Biol. Teach. 35, 125129
59 Schlacht, A. et al. (2014) Missing pieces of an ancient puzzle: evolution
of the eukaryotic membrane-trafcking system. Cold Spring Harb.
Perspect. Biol. http://dx.doi.org/10.1101/CSHPERSPECT.a01604
Review
Trends in Cell Biology July 2014, Vol. 24, No. 7
442

Das könnte Ihnen auch gefallen