again the great genetic diversity of prokaryotes even at the
species level. It is important to remind that two
representatives of the same species with similar lifestyle
and occupying the same ecological niche may present
important genomic variations. This is for instance the case
of Bradyrhizobium. These symbionts of legumes have very
similar lifestyles, but they use a different set of genes to
establish their symbiotic interaction with their host.
12.4.2 HGT in Eukaryotes
Our knowledge about the impact of HGT on the genome
content and the evolution of eukaryotes remains very limited
in comparison with that of prokaryotes. In fact, despite the
relatively large number of complete eukaryotic genome
sequences available since several years ago, very few studies
have addressed this question in a systematic way. Thus, the
data about HGT in eukaryotes are often anecdotal and most
likely do not reflect the actual importance of this phenomenon in eukaryotes. Several reasons can explain this delay in
the investigation of HGT in eukaryotes. The first one is
historical since, for long time, the scientific community
was dominated by the idea that eukaryotes are naturally
reluctant to HGT due to the presence of the nuclear membrane. This membrane, an essential character of all eukaryotic cells, represents an additional barrier that exogenous
DNA has to overcome to have the opportunity to integrate
in the cell genome. The absence of such a membrane in
prokaryotes was considered as a reason to explain the facility with which these microorganisms are able to acquire
exogenous DNA. The second reason is linked to the classical
view of the major evolutionary mechanisms acting on
eukaryotes. For a long time, eukaryotic evolution has been
studied using multicellular species with sexual reproduction
(most often animals). In these species, an HGT can be
successful only if it occurs in the germ cells, which is a
very improbable event. Thus, even the most fervent
advocates of a major role of HGT in evolution conclude
that this phenomenon is certainly marginal in eukaryotes
(Bapteste et al. 2009). Hence, the evolution of eukaryotes
and prokaryotes would follow different ways because of
their different cellular structures. However, this vision does
not take into account that all eukaryotes are subjected (at
present and all along their history) to a type of event
extremely rare in prokaryotes but which represents a very
rich source of HGT: symbiosis and, more particularly,
endosymbiosis.
12.4.2.1 Symbioses and Acquisition
of Bacterial Genes
Symbiotic associations are very frequent in eukaryotes and
have been described in all eukaryotic lineages. The
eukaryotic cell itself is the product of a very ancient
endosymbiosis. In fact, it is now widely accepted that all
contemporary eukaryotes have, or have had in their past,
mitochondria. These organelles derive from the endosymbiosis of an alphaproteobacterium within a proto-eukaryotic
host (cf. Chap. 4, Sect. 4.3.1) (Roger 1999; Philippe et al.
2000). That endosymbiosis had very important consequences
for eukaryotes, in particular by allowing them to respire
oxygen and, thus, to conquer numerous ecological niches.
Without aerobic respiration, it seems likely that multicellular
eukaryotes had never evolved. A second endosymbiosis has
also played a very important role in the evolutionary history
of eukaryotes. It concerns the acquisition by the ancestor of
the supergroup Plantae (red algae, green algae and plants,
and glaucophytes; cf. Chap. 7) of a photosynthetic plastid by
the endosymbiosis of a cyanobacterium. Plastids derived
from this event are called primary plastids. It is thanks to
them that eukaryotes have acquired the capacity to carry out
oxygenic photosynthesis. Other eukaryotes also have this
capacity (e.g., Euglenozoa, Cryptophyta, Haptophyta, etc.).
These are eukaryotic lineages that have acquired plastids
through secondary endosymbioses involving green or red
algae (these plastids are called secondary plastids).
A characteristic common to both the mitochondrial and
plastidial endosymbionts is that their genomes are extremely
reduced in comparison to those of their bacterial
counterparts (Alphaproteobacteria and Cyanobacteria). In
general, mitochondrial and plastidial genomes only have a
few dozens of genes, whereas alphaproteobacterial and
cyanobacterial species have several thousands. The first
eukaryotic genome sequences have shown that this genome
reduction had two origins: the clearest one is that many
genes originally present in the ancestral mitochondrial and
plastidial genomes were lost because they become nonessential in the new environment provided by the eukaryotic host
cytoplasm. In addition, many other genes were not lost but
transferred into the host genome. These genes are expressed
by the host and their products are carried to the plastids and
mitochondria allowing the functioning and maintenance of
these organelles. This implies that, during eukaryotic
evolution, complex targeting and transport systems have
emerged (Cavalier-Smith 1999). HGT events linked to
these endosymbioses are called endosymbiotic gene
transfers (EGTs). These EGTs provide selective advantages
to the host because it takes the control of the endosymbiont
(of its cell cycle, apoptosis, metabolic activity, etc.).
Although this is an important question, few studies have
tried to quantify in an accurate way the number of genes in the
eukaryotic genomes acquired by EGT from mitochondria and
plastids. This seems simple because it just requires identifying
the evolutionary origin of each gene in a eukaryotic genome,
but very often it is very difficult or even impossible to identify
precisely the bacterial donors. That is, it can be very difficult
476
C. Brochier-Armanet and D. Moreira
species level. It is important to remind that two
representatives of the same species with similar lifestyle
and occupying the same ecological niche may present
important genomic variations. This is for instance the case
of Bradyrhizobium. These symbionts of legumes have very
similar lifestyles, but they use a different set of genes to
establish their symbiotic interaction with their host.
12.4.2 HGT in Eukaryotes
Our knowledge about the impact of HGT on the genome
content and the evolution of eukaryotes remains very limited
in comparison with that of prokaryotes. In fact, despite the
relatively large number of complete eukaryotic genome
sequences available since several years ago, very few studies
have addressed this question in a systematic way. Thus, the
data about HGT in eukaryotes are often anecdotal and most
likely do not reflect the actual importance of this phenomenon in eukaryotes. Several reasons can explain this delay in
the investigation of HGT in eukaryotes. The first one is
historical since, for long time, the scientific community
was dominated by the idea that eukaryotes are naturally
reluctant to HGT due to the presence of the nuclear membrane. This membrane, an essential character of all eukaryotic cells, represents an additional barrier that exogenous
DNA has to overcome to have the opportunity to integrate
in the cell genome. The absence of such a membrane in
prokaryotes was considered as a reason to explain the facility with which these microorganisms are able to acquire
exogenous DNA. The second reason is linked to the classical
view of the major evolutionary mechanisms acting on
eukaryotes. For a long time, eukaryotic evolution has been
studied using multicellular species with sexual reproduction
(most often animals). In these species, an HGT can be
successful only if it occurs in the germ cells, which is a
very improbable event. Thus, even the most fervent
advocates of a major role of HGT in evolution conclude
that this phenomenon is certainly marginal in eukaryotes
(Bapteste et al. 2009). Hence, the evolution of eukaryotes
and prokaryotes would follow different ways because of
their different cellular structures. However, this vision does
not take into account that all eukaryotes are subjected (at
present and all along their history) to a type of event
extremely rare in prokaryotes but which represents a very
rich source of HGT: symbiosis and, more particularly,
endosymbiosis.
12.4.2.1 Symbioses and Acquisition
of Bacterial Genes
Symbiotic associations are very frequent in eukaryotes and
have been described in all eukaryotic lineages. The
eukaryotic cell itself is the product of a very ancient
endosymbiosis. In fact, it is now widely accepted that all
contemporary eukaryotes have, or have had in their past,
mitochondria. These organelles derive from the endosymbiosis of an alphaproteobacterium within a proto-eukaryotic
host (cf. Chap. 4, Sect. 4.3.1) (Roger 1999; Philippe et al.
2000). That endosymbiosis had very important consequences
for eukaryotes, in particular by allowing them to respire
oxygen and, thus, to conquer numerous ecological niches.
Without aerobic respiration, it seems likely that multicellular
eukaryotes had never evolved. A second endosymbiosis has
also played a very important role in the evolutionary history
of eukaryotes. It concerns the acquisition by the ancestor of
the supergroup Plantae (red algae, green algae and plants,
and glaucophytes; cf. Chap. 7) of a photosynthetic plastid by
the endosymbiosis of a cyanobacterium. Plastids derived
from this event are called primary plastids. It is thanks to
them that eukaryotes have acquired the capacity to carry out
oxygenic photosynthesis. Other eukaryotes also have this
capacity (e.g., Euglenozoa, Cryptophyta, Haptophyta, etc.).
These are eukaryotic lineages that have acquired plastids
through secondary endosymbioses involving green or red
algae (these plastids are called secondary plastids).
A characteristic common to both the mitochondrial and
plastidial endosymbionts is that their genomes are extremely
reduced in comparison to those of their bacterial
counterparts (Alphaproteobacteria and Cyanobacteria). In
general, mitochondrial and plastidial genomes only have a
few dozens of genes, whereas alphaproteobacterial and
cyanobacterial species have several thousands. The first
eukaryotic genome sequences have shown that this genome
reduction had two origins: the clearest one is that many
genes originally present in the ancestral mitochondrial and
plastidial genomes were lost because they become nonessential in the new environment provided by the eukaryotic host
cytoplasm. In addition, many other genes were not lost but
transferred into the host genome. These genes are expressed
by the host and their products are carried to the plastids and
mitochondria allowing the functioning and maintenance of
these organelles. This implies that, during eukaryotic
evolution, complex targeting and transport systems have
emerged (Cavalier-Smith 1999). HGT events linked to
these endosymbioses are called endosymbiotic gene
transfers (EGTs). These EGTs provide selective advantages
to the host because it takes the control of the endosymbiont
(of its cell cycle, apoptosis, metabolic activity, etc.).
Although this is an important question, few studies have
tried to quantify in an accurate way the number of genes in the
eukaryotic genomes acquired by EGT from mitochondria and
plastids. This seems simple because it just requires identifying
the evolutionary origin of each gene in a eukaryotic genome,
but very often it is very difficult or even impossible to identify
precisely the bacterial donors. That is, it can be very difficult
476
C. Brochier-Armanet and D. Moreira