Prokaryotic microorganisms are widespread in all the
environments on Earth. Given their ecological ubiquity,
it is not surprising to find many prokaryotic species in
close relationships with members of many eukaryotic
taxa, often establishing a persistent association, which
is known as symbiosis (BOX 1). According to the fitness
effects on the members of the symbiotic relationship,
associations can be referred to as parasitism, mutualism
or commensalism and, depending on the location of the
symbiont with respect to host cells, as ectosymbiosis or
endosymbiosis. Among intracellular symbioses, there are
differences regarding the extent of dependence between
the animal host and the symbiont and the age of the asso-
ciation, leading to the dichotomic classification between
obligate primary endosymbionts (P-endosymbionts)
and facultative secondary symbionts (S-symbionts).
P-endosymbionts generally have long evolutionary histo-
ries with their hosts, whereas S-symbionts seem to have
established more recent associations, retaining the ability
to return to a free-living condition1–4. In some cases, an
S-symbiont can evolve to become an obligate partner
and, if a P-endosymbiont is already present, microbial
consortia can be established.
For several decades, the idea that microbial associa-
tions might be central to eukaryote evolution remained
controversial. However, the work of Lynn Margulis since
the late 1960s not only contributed to the establishment
of a symbiotic theory for cell evolution5, but also revived
earlier proposals that were forgotten by biologists6.
Today, there is a wide consensus on the essential role of
symbioses during the origin and evolution of eukaryotic
cells, although controversies about the details persist7.
In addition, on the basis of its wide distribution across
major phylogenetic taxa (BOX 1), symbiosis could have
an important role in the evolution of species that are
involved in such partnerships. In all the major biological
phenomena that are classified as symbiotic, new cellular
structures and/or metabolic capabilities emerge as a
result of evolutionary forces, favouring the maintenance
of the association8. The interplay of both partners or, in
some cases, between a single host and more than one
symbiont, forms an evolving biological community that
changes throughout time.
In most cases of symbioses between prokaryotes and
eukaryotes, the relationship between host and symbiont
is so close that the microorganisms cannot be cultured,
making them difficult to study. However, in the past
few years, genome sequencing, transcriptomics and
the recently emerged field of metagenomics — which
avoid having to culture microbial cells — have offered
new opportunities for symbiosis research. Whole
genomes of several intracellular bacterial symbionts
have been sequenced, allowing comparisons among
the evolutionary innovations of these bacteria, from
being free living to various stages of integration with
their respective hosts9–23 (TABLE 1). Such genomics
approaches to the study of symbiosis will allow several
questions to be tackled, contributing to an understand-
ing of the evolutionary relevance of this phenomenon.
In particular, these questions relate to the nature of the
*Institut Cavanilles de
Biodiversitat i Biologia
Evolutiva, Universitat de
València, Apartado de correos
22085. 46071 València and
CIBER de Epidemiología y
Salud Pública, Spain.
‡Departament de Genètica and
§Departament de Bioquímica
i Biologia Molecular,
Universitat de València,
Dr. Moliner, 50. 46100
Burjassot (València), Spain.
Correspondence to A.M.
12 February 2008
From the Greek, sym ‘with’ and
biosis ‘living’. A long-term
association between two or
more organisms of different
species that is integrated at the
behavioural, metabolic or
genetic level. According to
the level of dependence on the
host, symbiosis can be obligate
or facultative. The term was
introduced by Anton de Bary
and Albert Bernard Frank when
discussing lichens and
mycorrhizae, respectively, at
the end of the 1870s.
Learning how to live together:
genomic insights into
Andrés Moya*‡, Juli Peretó*§, Rosario Gil*‡ and Amparo Latorre*‡
Abstract | Our understanding of prokaryote–eukaryote symbioses as a source of
evolutionary innovation has been rapidly increased by the advent of genomics, which
has made possible the biological study of uncultivable endosymbionts. Genomics is
allowing the dissection of the evolutionary process that starts with host invasion then
progresses from facultative to obligate symbiosis and ends with replacement by, or
coexistence with, new symbionts. Moreover, genomics has provided important clues on
the mechanisms driving the genome-reduction process, the functions that are retained
by the endosymbionts, the role of the host, and the factors that might determine
whether the association will become parasitic or mutualistic.
Nature Reviews Genetics | AOP, published online 12 February 2008; doi:10.1038/nrg2319
218 | AdvAncE OnLInE PubLIcATIOn
© 2008 Nature Publishing Group
Box 1 | Spread of symbiosis in the biosphere
Stable symbioses have independently evolved many times in diverse
groups of eukaryotes. Although the scope of this review is restricted to
prokaryotic partners that are associated with animal hosts, there are many
other remarkable examples of symbiosis between eukaryotic organisms,
as exemplified by fungal associations with other fungi, protists, plants
(that is, mycorrhizae), algae (that is, lichens) and animals (for an overview,
see rEF. 90). Most symbioses have a proven biochemical foundation. In
some cases, one of the partners benefits from compounds that are
produced by the other; in other cases, the waste products of one partner
(especially nitrogen compounds) are recycled by the other.
The eukaryotic nucleocytoplasm has limited metabolic capabilities. The
mitochondrial respiratory chain and oxygenic photosynthesis in
the chloroplast are two examples of metabolic functions that have been
acquired through symbiotic associations with prokaryotic partners
during early eukaryotic evolution8. Countless symbiotic relationships
have emerged more recently, or are in the process of developing, and
they complement the limited metabolic networks of most eukaryotes
with several prokaryotic metabolic capabilities, such as dinitrogen
fixation91, methanogenesis92, chemolithoautotrophy93, nitrogen
assimilation94 and essential-nutrient anabolism95,96. In particular, animal
heterotrophic metabolism is relatively narrow and some amino acids,
vitamins and fatty acids must be obtained from an external source. Thus,
symbiotic associations with microorganisms have allowed animals to
adapt to specialized feeding behaviours by providing the nutrients that
are deficient in their restricted diets. In fact, all intracellular mutualistic
symbioses between bacteria and animals that have been analysed at the
genomic level9–23 (TABLE 1) are related to nutrient provision and waste
recycling. In the best-studied cases — those of symbioses involving
insects97 — the presence of such associations throughout most of insect
evolutionary history suggests that symbiosis has been a driving force in
the diversification of this group.
The figure shows the phylogenetic distribution of symbioses, indicating
the bacterial and archaeal classes within which there are associations
with eukaryotic hosts. Data were collected from the literature and are
the result of a long tradition of studies that have used ecological,
developmental, morphological, biochemical and genomic approaches to
investigate symbiosis (a list of examples is given in the Supplementary
information S2 (table)). The uneven distribution of symbiosis highlights
the diverse interests and motivations of the scientific community.
Advances in genomics have had a large impact on the research into
microbial symbioses, which has implications for biotechnology (for
example, sponge-associated prokaryotes98), agriculture (for example,
nitrogen fixation in plant-associated bacteria91) and biomedicine (for
example, the human intestine microbiome99). Metagenomic methods
have notably increased our knowledge of the biodiversity of non-
cultivable bacteria in symbiotic consortia, including the description of
completely new candidate phyla — Endomicrobia100 and Poribacteria101.
Postulated symbiotic events leading to the evolutionary origin of
organelles (mitochondria and chloroplasts) are indicated. The branching
order in the eukaryotic phylogenetic tree has been adapted from rEFs
102,103. The phylogenetic tree of animals is an adaptation from rEF. 104,
modified by recent molecular phylogenies105,106. The lengths of branches
are not to scale. An asterisk indicates that complete genomes are
available (see TABLE 1).
Nature Reviews | Genetics
A, B*, Fu, S, α, β, γ*, δ, ε, M
Cy, α, β
A, Cy, α, β, γ
Cy, α, β, γ, δ, M
A, B, Ch, Fi, α, β, γ, ε
Cy, β, M
B, F, Fi, γ, M
B, α, γ, δ, ε, S
Ac, A, Cl, Cy, Fi, G, N, α, β, γ, δ, S, V, P, T
Fi, S, M
nATurE rEvIEWS | genetics
AdvAncE OnLInE PubLIcATIOn | 219
© 2008 Nature Publishing Group
symbioses in which one
species is increasing its fitness
while the fitness of the other
species is adversely affected.
symbioses in which both
species increase their fitness.
symbioses in which one
partner is increasing its fitness
without affecting the other
genome-reduction process, the type of genes that are
retained by the endosymbiont, the molecular pathways
that are used by the host to control the endosymbiont
population and the complex set of factors that deter-
mine whether the final outcome of the association is
parasitic or mutualistic.
Here we focus on the study of mutualistic endosym-
bioses of animals for which genomic studies have been
carried out, providing new biological and evolutionary
insights. Although only one host genome has been
sequenced, complete genomic sequences for several
bacterial symbionts are available. comparisons with
free-living relatives have revealed dramatic changes in
genome size and content, and have begun to reveal the
mechanisms by which these changes take place and their
functional consequences. Furthermore, the mechanisms
that are involved in the establishment, maintenance and
evolution of mutualistic associations are being scruti-
nized; this has led to the emergence of new insights into
aspects such as metabolic interdependencies between
partners and the survival of the prokaryote in the face
of the host immune response.
Genomics coverage of prokaryotic endosymbionts
Obligate endosymbionts of animals. In recent years,
the genomes of several bacterial P-endosymbionts of
animals have been completely sequenced (TABLE 1).
P-endosymbionts are inherited by strict vertical trans-
mission and, in most cases, reside in specialized host
cells called bacteriocytes. Functional analyses have cor-
roborated the nutritional role of the associations; each
bacterium provides the nutrients that are deficient in
the diet of its animal-host, whereas the bacterium gains
a permanent supply of a wide range of metabolites that
are provided by the host.
Most of the endosymbionts that have been studied
are γ-proteobacteria that live in obligate association
with insects, and they provide various functions to
Table 1 | Genomic data for mutualistic symbionts of animals
Baizongia pistaciae (aphid)Heterotroph61825.35073329
Cinara cedri (aphid)Heterotroph42220.23623313
Heterotroph 70627.45833 374
Heterotroph 792 29.66103394CP000016
Pachypsylla venusta (psyllid)
Brugia malayi (nematode)#
Heterotroph 68633.2595639 9
The data shown are for those symbionts that have been sequenced as of January 2008. Data were retrieved from the National Center for Biotechnology Information
(NCBI). *γ-Proteobacteria. ‡Genome sequence in progress. §These bacteria are called Candidatus. ||Bacteroidetes. ¶α-Proteobacteria. #Complete genome available,
see rEF. 78. **ε-Proteobacteria; although it is unclear whether these two isolates are epibiotic symbionts or another variation of symbionts, many genome features
strongly support their symbiotic lifestyle. ‡‡Estimated from rEF. 20. CDS, coding sequences; rRNA, ribosomal RNA.
220 | AdvAncE OnLInE PubLIcATIOn
© 2008 Nature Publishing Group
A symbiosis in which the
symbiont lives on the body
surface of the host, including
internal surfaces such as the
lining of the digestive tube and
the ducts of glands.
symbioses in which a
prokaryote symbiont lives
inside a eukaryotic cell.
bacterial endosymbionts that
live inside specialized animal
host cells called bacteriocytes.
The association is obligate for
that coexists with a
located in syncitial cells near
the bacteriocyte and in various
other insect tissue types.
secondary symbionts are not
essential for host survival and
are transferred horizontally
among individuals of both the
host species and other species.
The application of genomic
analyses to uncultured
microorganisms. Also referred
to as environmental genomics.
The endosymbionts are
maternally transferred, that
is, directly from a host to
specialized cells of the host
species in which symbiotic
The metabolic mode in which
the carbon source is organic
matter. By extension, this is a
metabolic mode in which
organic matter is the source of
carbon, electrons and energy
The metabolic mode in
which CO2 is the carbon
source and an inorganic
chemical reaction is both the
electron and energy source.
their host, including: the provision of essential amino
acids9,10,12–14,16,21,23, vitamins and cofactors9–12,15 that are
absent in the host diet; nitrogen recycling and storage13,14;
and the provision of metabolic factors that are required
for survival and fertility19. Whereas most of these species
are heterotrophic endosymbionts, two recently sequenced
endosymbionts that are harboured by deep-sea clams of
the genus Calyptogena17,18 are chemolithoautotrophic, and
provide almost all nutrients to the host by fixing cO2
and using H2S as a source of energy and reducing power
(FIG. 1 and Supplementary information S1 (figure)).
Facultative symbionts. Occasionally, hosts tolerate
S-symbionts that coexist with a P-endosymbiont and
that can be either deleterious or beneficial to the host24.
In aphids, a number of facultative symbionts that reside
in multiple host tissues, cells surrounding primary
bacteriocytes or in their own bacteriocytes have been
described25. Although they are normally vertically trans-
mitted, their distribution patterns suggest that sporadic
horizontal transmission among host individuals and host
species must have occurred26. Among the possible
beneficial roles that have been identified, some of these
S-symbionts can rescue the host from heat damage27–29,
provide defence against natural enemies (parasitoids and
pathogens)30–32 and participate in host specialization33–36.
In other cases, facultative symbionts can spread among
lineages without conferring a benefit, by manipulating
host reproduction (reviewed in rEF. 37). For example,
in Wolbachia infections of arthropods, the symbiont
undergoes transfer among host lineages. remarkably,
Wolbachia can also be a P-endosymbiont in filarial
nematodes19, indicating that the same prokaryotic line-
age can take part in more than one type of symbiotic
At present, the only completely sequenced genome
of an S-symbiont is that of Sodalis glossinidius (TABLE 1).
This bacterium coexists in the gut lumen of tsetse flies
with the P-endosymbiont Wigglesworthia glossinidia,
although they occupy different areas of the gut and can
be found both intra- and extracellularly22. It has been
suggested that S. glossinidius has an important role in
the acquisition of trypanosome infections38.
Endosymbiotic consortia. Although the importance of
syntrophy between unrelated organisms has been recog-
nized in several cases39, the new field of metagenomics
provides the ability to study bacterial endosymbionts
that coexist in the same host, enabling the discovery
of complex and stable associations, and analysis of the
contribution of each partner to the relationship.
One endosymbiotic consortium that has recently
been reported involves Buchnera aphidicola bcc
and Serratia symbiotica Scc, which coexist in the
bacteriome of the aphid Cinara cedri40. S. symbiotica is
a facultative symbiont in other aphid species, but has
been found in all the cedar-aphid populations that
have been analysed, casting doubts over its faculta-
tive status in this species. Functional and evolutionary
comparative analyses of the sequenced B. aphidicola
genomes, as well as microscopic analysis of the two
cedar-aphid endosymbionts (FIG. 2), led Pérez-brocal
and co-workers21 to conclude that S. symbiotica Scc
might have the potential to replace B. aphidicola bcc.
In a second example, genomic sequencing has been
carried out for Baumannia cicadellinicola and the
bacteroidetes species Sulcia muelleri, the co-resident
P-endosymbionts of the xylem-feeding sharpshooter
Homalodisca coagulata15,23. Genomic analysis has
revealed that the two symbionts have complementary
biosynthetic capabilities, which are needed to provide
their host with nutrients that are lacking in xylem sap.
Phylogenetic analysis revealed that Sulcia was ancestrally
present in a host lineage that acquired Baumannia at the
same approximate time as the switch to xylem feeding,
consistent with the view that its nutrient-provisioning
capabilities were a requirement for the host to evolve
towards this lifestyle41. As in the case of Buchnera–
Serratia, the two symbionts live in close proximity within
the host bacteriome15.
Metagenomic approaches have also been used to
analyse more complex consortia, including the four co-
occurring symbionts from the segmented worm Olavius
algarvensis42. In this instance, the host belongs to a group
of oligochaetes that lack a mouth, gut and anus, and
that are unique among annelids in having reduced their
nephridial excretory system. O. algarvensis harbours
at least four symbionts, two γ-proteobacteria (sulphur
oxidizing) and two δ-proteobacteria (sulphate reducing),
which fix cO2 and provide the host with nutrients. Almost
all amino acids and several vitamins can be synthesized by
the symbionts and then provided to their host, probably
through controlled digestion of the bacteria as suggested
by the vacuole localization of the host43. The symbionts
are also likely to be involved in host-waste recycling.
Genomic changes during endosymbiont evolution
Genomic changes experienced by endosymbionts. A gen-
eral feature of intracellular pathogenic and mutualistic
bacteria is that they have smaller genomes with a higher
AT content than their free-living relatives44. The causes
of reductive-genome evolution are related to both the
genetic information that is needed in the new environ-
ment and to population dynamics, although the relative
importance of these factors has not been determined45,46.
The change from a free-living environment to one that is
intracellular and protected implies that many genes are
rendered unnecessary, whereas others become redun-
dant because their functions can be supplied by the host.
Therefore, some genetic material can be lost without a
detrimental effect, and can accumulate mutations owing
to the lack of effective natural selection to purge them. In
addition, owing to the strict vertical transmission of obli-
gate endosymbionts, only a few bacteria are transmitted,
which generates continuous bottlenecks that favour the
action of random genetic drift. The combination of these
factors allows the accumulation of mutations in genes
that can be beneficial but that are not essential — such
as genes involved in dnA repair and recombination
(see below) — further increasing the repertoire of genes
that can be lost, and reducing the possibility of genetic
exchange by homologous recombination. Furthermore,
nATurE rEvIEWS | genetics
AdvAncE OnLInE PubLIcATIOn | 221
© 2008 Nature Publishing Group
some endosymbionts retain a
generalized ability to colonize
and persist in multiple hosts,
that is, their transmission is
between individuals of the
same or different host species,
rather than from parent to
Emergence of new metabolic
capabilities as a result of
symbiosis, it is often
essential for the survival
of the consortium.
An organ-like structure formed
Inner and outer
(except pantothenate and UQ)
to the host
Vacuole of the
Vacuole of the host cell
the host cell
the host cell
Reduced carbon, non-essential amino acids
to the host
(except Thr and Ile)
Inner and outer
Nature Reviews | Genetics
Figure 1 | comparative biochemistry of symbiont–host interdependence. As a study case, we use information from
complete genome sequences to infer and compare the metabolisms that are encoded by the reduced genomes of an
autotrophic and a heterotrophic endosymbiont. Blue indicates the presence of a metabolic process, orange indicates its
absence. a | Candidatus Ruthia magnifica17 uses electrons from H2S to drive an autotrophic metabolism based on the
reductive pentose phosphate pathway (Calvin–Benson cycle), supplies essential nutrients (including amino acids and
cofactors) to its host, and recycles nitrogen. As is the case in the Calyptogena okutanii endosymbiont18, it has the potential
for anabolism of all essential biomolecules except threonine (Thr), isoleucine (Ile) and ubiquinone (UQ). The missing steps
in those pathways could be explained by the function of an unidentified enzyme, or by complementation with the host
metabolism. b | Buchnera aphidicola BCc uses reduced-carbon molecules and non-essential amino acids to fuel its
heterotrophic metabolism and synthesize essential nutrients, with the exception of tryptophan (Trp) and riboflavin21.
These molecules could be provided by a second co-existing endosymbiont, Serratia symbiotica. The self-sufficiency of
R. magnifica sharply contrasts with the incomplete metabolic abilities of B. aphidicola BCc. Among the few similarities
between both metabolisms, note the absence of nutrient-transport systems to the host. A more detailed version can be
found in the Supplementary information S1 (figure). PTS, phosphotransferase system; Sec system, secretory system.
222 | AdvAncE OnLInE PubLIcATIOn
© 2008 Nature Publishing Group
the intracellular environment isolates the symbiont from
other bacterial species and thus reduces, and can even
eliminate, the possibility of gaining new genetic mate-
rial by horizontal gene transfer. Thus, gene losses can
In addition to having reduced genomes, an increased
AT content is another general feature of the endosymbi-
ont genomes that have been sequenced so far. The bias
towards AT is probably due to the loss of genes for dnA
repair, leading to a mutational Gc to AT pressure47 (for
a metabolic hypothesis see rEF. 48). Such a bias has a
notable effect on many genes by altering the structure
and function of the corresponding proteins. In fact, it
has been proposed that GroEL, a chaperonin that is
constitutively overexpressed in B. aphidicola, helps to
post-translationally correct the altered structure of many
proteins that is caused by this bias49. A by-product of the
base-composition bias of these genomes is the loss of
the codon-usage bias that is typical of free-living bacte-
ria: it is almost absent from B. aphidicola, and is highly
reduced in P-endosymbionts with larger genomes and
The degree of both genome-size reduction and
increase in AT content vary among endosymbionts,
and correlate with the age of the association.
P-endosymbionts that are partners in old associations
have generally smaller genomes and an AT content higher
than 70%, whereas S-symbionts and P-endosymbionts
that are part of younger associations have genome sizes
and AT contents intermediate with respect to older
P-endosymbionts and free-living relatives2,25,50.
The genome-reduction process examined. To understand
the different stages of the genome-reduction process it is
necessary to analyse a range of genome sequences, from
endosymbionts that still have genomes similar in size
to those of their free-living relatives, to those that have
undergone the most dramatic reductions. Such compara-
tive analyses suggest that the process of genome shrink-
age might have taken place in two separate stages.
A massive gene loss must have occurred soon after
the establishment of the obligate symbiosis, probably by
means of large deletions that would cause the elimination
of a series of contiguous genes52. The accumulation of
mobile elements, representing a source for chromosomal
rearrangements and gene inactivation, seems to have an
important role at this first stage. This is suggested by
the fact that mobile elements are relatively abundant in
bacteria that have recently acquired an obligate, intra-
cellular way of life53,54. Mobile elements, such as bacte-
rial insertion sequences (IS), are fairly abundant in the
S-symbiont S. glossinidius22, but even more so (estimated
at more than 25% of genome content) in its close rela-
tives SOPE (Sitophilus oryzae primary endosymbiont)
and SZPE (Sitophilus zeamais primary endosymbiont).
These species are P-endosymbionts of the rice and maize
weevils, respectively, and both have recently established
obligate associations with their hosts55,56 (sequencing of
the SOPE genome is in progress).
These data indicate that a common symbiotic ancestor
of these two lineages must already have possessed these
elements. The increase in frequency of these elements in
the newly established P-endosymbiont must be due to
an increase in the replicative transposition of elements
that were resident at the onset of symbiosis24. All analysed
endosymbiotic bacteria that are involved in older associa-
tions possess genomes that are free of mobile elements.
Presumably, therefore, the expansion of mobile
elements must at some point have become deleterious
and they must have been removed as part of the process of
genome degradation. The difference in the abundance
of transposable elements in two strains of Wolbachia
with different lifestyles and genome sizes supports this
statement. In Wolbachia pipientis wMel57, a reproductive
parasite of Drosophila melanogaster, 14% of the 1.27-Mb
symbiont genome is occupied by repetitive dnA and
mobile elements, whereas in W. pipientis wbm19, the obli-
gate endosymbiont of the nematode Brugia malayi, these
elements account for just 5.4% of the 1.08-Mb genome.
during the second stage of the genome-reduction
process, genome shrinkage seems to have mostly
occurred through a process of gradual gene loss, scat-
tered along the genome. Such loss seems to follow
a pattern that starts with the inactivation of a gene
(pseudogenization) by single-nucleotide mutations,
Nature Reviews | Genetics
Figure 2 | Bacteriocytes of the aphid Cinara cedri.
a | A 1.5 mm semi-thin section that shows both primary
and secondary bacteriocytes, easily identifiable
by their different tones under toluidine blue staining.
b | Electron micrographs of the same individual that
show the round shape of Buchnera aphidicola and
Serratia symbiotica, respectively, in their bacteriocytes.
P, primary symbiont (B. aphidicola); S, secondary symbiont
(S. symbiotica); n, nuclei of both bacteriocytes;
m, mitochondria. Electron micrographs courtesy of
A. Lamelas, Universitat de València, Spain.
nATurE rEvIEWS | genetics
AdvAncE OnLInE PubLIcATIOn | 223
© 2008 Nature Publishing Group
and continues with a rapid reduction in length until the
original gene is completely eroded58,59. Even in advanced
stages of the reductive process, genome-length reduc-
tion is mainly due to the loss of protein-coding genes,
not to their shortening21. regarding intergenic regions,
there is a slight size reduction in the four sequenced
B. aphidicola strains when compared with their close
relative Escherichia coli60. The shortening of intergenic
regions is more evident in extremely reduced genomes,
as observed in Carsonella ruddii16, in which this has
been so extreme as to have led to overlapping genes.
In C. ruddii, the annotated OrFs are also consider-
ably shorter than their orthologues in other bacteria,
although the functionality of these shortened OrFs is
Functional changes in hosts and symbionts
The first step towards the establishment of an obligate
endosymbiosis takes place when a free-living bacterium
infects the host. Then, both organisms co-evolve to adapt
to the new situation. On the bacterial side, genomics
studies have revealed that the endosymbiont genome
gets smaller during this adaptive process, owing to the
loss of genes that are rendered unnecessary in the new
environment; however, a symbiotic bacterium must still
retain a number of functions to allow its continued sur-
vival in the host. by contrast, our understanding of how
the animal host is affected at the genetic level, and how it
has solved the problem of having a bacterium (or more
than one) inside its cells, is poorly understood. However,
host specialization implies that it is now dependent on
the bacteria to survive and reproduce. Adaptive changes
in the host include the development of specialized cells
where bacteria reside, adequate systems to control bacte-
rial populations, plus the modification of its immune
response against the intruder cells.
General changes in endosymbiont gene content. The
catalogue of individual genes that are shared by all
analysed endosymbionts suggests that the molecular
mechanisms that are necessary for survival in an intra-
cellular environment share broad commonalities across
all endosymbiotic associations13. Only about one-third
of the coding capacity of each endosymbiont genome
that has been sequenced so far is devoted to the require-
ments of the specific symbiosis, with these genes mainly
reflecting differences in host lifestyle, nutritional needs
and location within host cells.
Across the sequenced endosymbiont genomes, genes
that are involved in essential functions — such as those
involved in dnA replication, transcription and transla-
tion — are more likely to be retained than are genes
of other functional classes, and sometimes account for
more than one-third of the genome21. chaperone sys-
tems and all essential components of the protein trans-
location machinery are also retained in these genomes,
ensuring the correct folding and localization of protein
Although gene losses affect all functional categories,
they are nonrandom. The most dramatic losses affect
genes that are involved in metabolism but are not
required for host survival, as we discuss later. Another
general feature of all sequenced endosymbiont genomes
is the loss of most dnA repair and transcriptional regula-
tory mechanisms. Furthermore, losses affect most genes
of the latter category, indicating that there is no need
for transcriptional regulation in a stable environment.
This hypothesis is supported by the results of microarray
experiments in B. aphidicola62,63, which show constitu-
tive gene expression of metabolic genes, regardless of
changes in environmental conditions.
Other features of the functional content of endo-
symbiont genomes are more dependent on their spe-
cific lifestyles. For example, the bacterial cell envelope
seems to be highly simplified, being less structured in
bacteria living inside host-derived vesicles (for example,
B. aphidicola) than in endosymbionts that live free in
the cytosol of bacteriocytes (for example, Blochmannia
floridanus and W. glossinidia)13. The simplification can
be extreme; for example, B. aphidicola bcc has lost all
the genes that are involved in the biosynthesis of the
bacterial cell wall21.
Metabolic changes and nutrient transport. The bigger
endosymbiont genomes, belonging to more recently
established partnerships, have retained many genes that
are involved in several intermediary metabolic path-
ways, but most of these have been lost in the smaller
genomes of well-established endosymbionts (FIG. 1;
TABLE 1). The loss of metabolic genes is also affected to a
certain extent by the presence of other bacterial species
in the intracellular environment of the symbiont, in that
the simultaneous presence of S-symbionts can compen-
sate for the metabolic deficiencies of the P-endosymbiont,
as well as those of the host64.
Even though each endosymbiont (or the combina-
tion of symbionts within a consortium) has specifically
retained the pathways that are necessary to synthesize
the nutrients that respective hosts are unable to provide
for themselves, endosymbionts exhibit limited transport
capabilities65 (FIG. 1). This situation has been observed in
B. aphidicola, the two deep-sea-clam endosymbionts
that were recently analysed17,18 and the endosymbionts of
O. algarvensis42. The subcellular localization of endo-
symbionts inside vacuole-like compartments66,67, and
the presence of lysozyme in the host cells68,69, strongly
suggest that, in these systems, the host nutritional needs
could be satisfied by controlled weakening or killing of
symbiont cells — a case of a necrotrophic host–symbiont
Invasion strategies: have pathogens been domesticated?
recent studies of the mechanisms that are used by endo-
symbionts to establish and maintain their infection of
host tissues indicate that invasion strategies are based on
the same molecular tools that are used by well-studied
pathogenic bacteria (reviewed in rEFs 24,47). These
mechanisms include the use of various secretion systems
for the attachment and invasion of host cells64 and the
utilization of the same quorum-sensing mechanisms
for the regulation of virulence or mutualistic traits,
depending on the type of association17,18.
224 | AdvAncE OnLInE PubLIcATIOn
© 2008 Nature Publishing Group
A part of a genome, for
which there is evidence of
its acquisition by horizontal
transfer, that encodes genes
that contribute to the virulence
of a pathogen.
Sitophilus spp. P-endosymbionts, and their close
relative the S-symbiont S. glossindius, encode presum-
ably functional genes of the type III secretion system
(T3SS)70, an indication that the presence of the secretion
system pre-dates the origin of the symbiotic association.
The T3SS is a crucial virulence factor of many Gram-
negative pathogens of plants and animals, including
humans. However, it also seems to be a beneficial factor,
enabling the establishment of a mutualistic, intracel-
lular insect–bacteria association. In the S. glossinidius
genome there are three regions that contain distinct
T3SS systems. The best characterized are SSr-1 and
SSr-2, which contain genes that are related to genes
found in Salmonella enterica and Chromobacterium
violaceum, respectively. Interestingly, both regions have a
complete set of genes encoding a functional T3SS export
apparatus, but they lack many genes encoding effector
proteins in the orthologous pathogenic islands. Moreover,
the only putative effector proteins that are preserved are
specifically involved in the cytoskeletal rearrangements
necessary for bacterial cell invasion24.
both parasitic and mutualistic Wolbachia lack a T3SS,
but they do possess a complete set of genes for another bac-
terial secretion system, T4SS, which is used by many path-
ogens to secrete proteins in order to invade host cells19,57.
B. aphidicola and W. glossinidia — well-established
P-endosymbionts in an advanced stage of genome reduc-
tion and with highly simplified cell envelopes — also lack
a canonical T3SS, but still retain surface mechanisms for
exporting proteins, such as a simplified flagellar appara-
tus71 (FIG. 3). because these structures are quite abundant
at the B. aphidicola surface — even though these are non-
motile bacteria72 — and only those elements homologous
to the T3SS have been retained in all of these reduced
genomes21 it has been proposed that these structures
must be involved in invading new bacteriocytes, ovaries
and embryos to ensure transmission to host offspring9.
The immune response and control of bacterial inva-
sion. The molecular mechanisms that are involved in
the host immune response to bacterial invasion remain
poorly understood73 but, clearly, hosts have adapted to
maintain rather than eliminate endosymbionts. but how
does the host immune system perceive these bacteria
and control their growth and invasion without complete
Hosts have developed molecular systems to recognize
conserved microbial cell-envelope motifs (for example,
peptidoglycan) through receptors such as those that were
identified following the whole-genome sequencing of
D. melanogaster74. Endosymbionts that have an ancient
relationship with their hosts have highly simplified cell
envelopes and, therefore, these old associations are not
good models to study this type of immune control.
However, analyses of younger endosymbiotic associa-
tions provide insights into the early stages of the interplay
between host immunity and bacterial virulence, for exam-
ple, endosymbionts of weevils from the genus Sitophilus.
Heddi et al. carried out an in vivo broad characterization
of the transcriptional response of the bacteriocyte to
intracellular bacteria, using SZPE as a model organism75.
They found that the bacteriome expresses a peptidog-
lycan recognition protein (PGrP) that is homologous
to the PGrP-Lb protein of D. melanogaster, a catalytic
member of the PGrP family. The amidase activity of
PGrP-Lb reduces the biological activity of peptidoglycan,
and therefore downregulates the host immune response
against Gram-negative bacteria76. The specific role of the
weevil PrPG is still under investigation, but the fact that
it contains the amino-acid residues that are responsible
for amidase activity, and that PrPG transcripts only
accumulate when endosymbiotic bacteria are present
in the bacteriome77, indicate that it might work in the
same way, thus — reducing the host’s defence against
the endosymbiont to allow the long-term interaction.
Nature Reviews | Genetics
FliK FlgD FlgE
FlhA FlhB FliOP FliQ FliR
Rotor (ring C)
FliM FliN FliG
FlhA FlhB FliOP FliQ FliR
FliE FlgB FlgC FlgF FlgG
Figure 3 | the flagellum, an example of genome reduction at the structural level.
a | The Escherichia coli flagellum. Protein and structural components are indicated: cap,
filament (formed by ~20,000 copies of flagellin), hook–filament junction, hook and basal
body. The basal body is formed by a set of rings for anchorage to the inner and outer
membranes and to the cell wall, the rod, the stator, the rotor and the export apparatus
(including one ATPase). b | Vestiges of the flagellar apparatus in Buchnera aphidicola
BCc. Elements that are involved in the formation of the flagellar filament and hook, the
stator/motor and the transcriptional regulators are lost. None of them is necessary for a
non-motile bacterium. Many elements that are involved in the anchorage of the export
apparatus to the cell envelope have also been lost, consistent with the absence of a well-
structured outer membrane and cell wall. However, the MS-ring protein (FliF), two out of
three components of the C-ring (FliN and FliG), the P-ring protein (FlgI) and L-ring
protein (FlgH) are present. Only one protein for the rod formation (FlgF) is preserved.
All the retained proteins are homologous to the components of the type III secretion
system, supporting the hypothesis that this structure is used by the bacterium as a
protein export system to help in the invasion of new host cells71.
nATurE rEvIEWS | genetics
AdvAncE OnLInE PubLIcATIOn | 225
© 2008 Nature Publishing Group
Only one sequenced genome is available for a eukary-
otic host of an endosymbiont that also has a sequenced
genome — the genome of B. malayi78, the nematode host
of Wolbachia wbm. consistent with a mitigated immune
response to control rather than eliminate the endosym-
biont, the immune system of B. malayi has lost the ability
to encode the small antibacterial peptides described in
other sequenced nematodes79. The genome sequence
of the aphid Acyrthosiphon pisum will soon be avail-
able and will provide us with additional clues into how
hosts control bacteria in the later stages of the symbiotic
One additional insight into the control of endosymbi-
onts by their hosts comes from the finding that endosym-
bionts that do not reside in host-derived vesicles, but that
live free inside the cytosol, have lost the gene dnaA11,13.
This gene encodes a protein that is involved in the ini-
tiation of dnA replication, which might indicate that
there is a direct control of bacterial growth by the host.
In the case of B. aphidicola, which does reside in host-
derived vesicles, the high level of lysozyme expression
detected in bacteriocytes68,69 might be an important fac-
tor involved in the host control of bacterial abundance,
as this enzyme catalyses the hydrolysis of peptidoglycan
— a system normally used by animals as a constitutive
defence against bacterial pathogens.
The evolution of specialized host cells. using transcrip-
tomic analysis, a pioneering study was carried out into
the development and evolution of aphid cells that are
destined to become bacteriocytes. Several transcription
factors were found to be expressed in these cells early
on in development — before colonization by the symbi-
ont population acquired from the mother — following
a pattern that has not been described in any other insect
cells80. Even when bacteria are experimentally removed
by antibiotic treatment, bacteriocytes are still specified
and maintained. The main conclusion of this work is
that bacteriocytes represent an evolutionarily novel cell
fate, which is developmentally determined independ-
ently of the presence of bacteria. regarding the role
of the bacteriocyte itself in the symbiotic relationship,
the identification of a number of bacteriocyte-specific,
highly expressed genes in the aphid A. pisum by tran-
scriptome analysis69 revealed that most of them were
involved in several functions: non-essential amino-acid
metabolism, consistent with this process being among
the most important in symbiotic systems; transport,
indicating that the bacteriocyte mediates exchange
of metabolites and substrates; and production of lys-
ozyme, possibly involved in nutrient uptake by the
host, as already mentioned. A similar study that was
carried out in S. zeamais revealed that the bacteriocyte
also allows increased sugar uptake and metabolism, and
various anti-stress systems that control the resultant
Evolutionary outcomes of symbioses
Once an association has been established and the
genome-reduction process has started, the loss of some
metabolic abilities, and other functions, irreversibly ties
the intracellular bacteria to their host. This scenario
was initially proposed in the 1930s by Lwoff81 and has
been confirmed by genomic studies (see previous sec-
tion). However, the bacterial genome, no matter how
small, must retain the genes that are essential for the
maintenance of the symbiotic relationship (for example,
essential nutrient provision), and a reduced repertoire of
genes for self-maintenance, self-reproduction and evo-
lution. Otherwise, genome reduction might eventually
lead these bacteria towards extinction.
In time, a second symbiont can join the consortium.
Although initially this new association is likely to be
facultative, a new stable relationship can be established,
in which case the two bacterial species will co-evolve.
As the genome-reductive process continues, and genes
that are shared with the second endosymbiont become
redundant, two possible outcomes can occur. The
actual outcome is a matter of chance, depending on
which genome is affected by the loss of genes that are
needed for the synthesis of molecules that are essential
One possibility is that the presence of a second symbi-
ont might accelerate the degenerative process of the first,
leading to its extinction and replacement by the formerly
facultative symbiont, which will then become obligatory.
This replacement has been reported in symbioses of some
weevils56 and might also be taking place in the case of
the Buchnera–Serratia consortium that has been estab-
lished in the aphid C. cedri21. unlike other sequenced
B. aphidicola strains, bcc has partially lost its symbiotic
role, as it cannot synthesize tryptophan and riboflavin
(see Supplementary information S1(figure)), which need
to come from another source, not only for the survival
of the host, but also for that of B. aphidicola. Genes from
S. symbiotica Scc that are involved in the biosynthesis
of tryptophan have been identified, revealing that this
species might be the source of this essential amino acid.
The complete sequence of S. symbiotica Scc (currently
in progress) will tell us whether this bacterium is able to
perform all the metabolic functions needed to maintain
its host’s fitness, or whether it has also lost some path-
ways. In the first case, extinction of the P-endosymbiont
(Buchnera) and replacement by Serratia would be the
most plausible hypothesis.
Alternatively — highlighting the second possibil-
ity for the evolutionary outcome of a second symbiont
establishing itself in a host — the obligate biochemical
interdependence between two endosymbionts can evo-
lutionarily seal the bacterial metabolic complementa-
tion, and the establishment of a stable consortium would
be the expected evolutionary outcome. This appears
to be the case in the sharpshooter H. coagulata, with
two complementary endosymbionts, B. cicadellinicola
and S. muelleri15,23. Whereas B. cicadellinicola is mainly
devoted to the biosynthesis of vitamins, S. muelleri
encodes the enzymes that are involved in the biosynthe-
sis of most essential amino acids. The complementarity
is striking. For example, S. muelleri has lost the pathway
for the biosynthesis of histidine, which is the only bio-
synthetic pathway for essential amino acids retained in
226 | AdvAncE OnLInE PubLIcATIOn
© 2008 Nature Publishing Group
The smallest set of genes that
is necessary and sufficient to
sustain a living cell in the most
favourable conditions; that is,
in the presence of adequate
nutrients and in the absence
of stress factors.
The genome degeneration process seen in endosymbi-
onts can also be accompanied by massive gene transfer to
the host cell nucleus82. In order for a true cell organelle
to be established, several processes must be acquired: new
gene-regulatory mechanisms; specific target sequences
for proteins that are encoded by the host nucleus but
function in the symbiont; and a protein import appa-
ratus that functions at the symbiont surface. These
processes occurred in the α-proteobacterial ancestor
of mitochondria and in the cyanobacterial ances-
tor that gave rise to chloroplast lineages. Is there any
system that is currently en route from endosymbiont
to organelle? At present, the smallest sequenced endo-
symbiont genome belongs to C. ruddii, which only
codes for approximately 180 proteins. However, a care-
ful analysis revealed that many genes that are involved
in essential functions are absent. It remains to be
elucidated whether C. ruddii is on its way to becoming
an organelle, or if it is nearing extinction and replace-
ment by an unidentified symbiont61. The transfer of
genes from an organelle to the nucleus has been demon-
strated under laboratory conditions (see rEF. 83 and ref-
erences therein). There is also evidence of gene transfer
from Wolbachia to the nuclear genome of several insect
and nematode hosts84. These remarkable cases of lateral
gene transfer should be a caution for those carrying out
eukaryotic genome sequencing, and should add new
insights of relevance to the debate about the differences
between cell organelles and endosymbionts85.
Conclusions and future directions
The advent of genomics has significantly contributed
to reinforcing the view of symbiosis as a widespread
biological phenomenon, especially through the genome
sequencing of uncultivable organisms. Symbiosis is a
dynamic process in which the prokaryotic symbiont,
while providing the host with new metabolic capa-
bilities, experiences many genotypic and phenotypic
changes compared with its free-living relatives in order
to adapt to its new lifestyle, as revealed by the specific
differences observed in the completely sequenced
genomes of several endosymbiotic bacteria.
Many questions remain open for future research. On
the host side, a better understanding of how these organ-
isms control and/or domesticate prokaryotic symbionts
is needed. Genomic and transcriptomic studies of model
eukaryotic hosts that have well known and stable sym-
biotic associations will be valuable. until now, only the
genome of B. malayi, the nematode host of W. pipiensis
wbm, has been fully sequenced78, but the sequence of the
aphid A. pisum is on its way. comparative analysis with
model organisms that have already been sequenced,
such as the nematode Caenorhabditis elegans86 and the
arthropod D. melanogaster87, will give new insights
into the changes that are experienced by the host after
the establishment of the symbiosis. On the bacterial
side, comparative studies of the molecular processes
governing the nature of symbiotic associations, either
mutualistic or parasitic, will provide important clues
to understand why microorganisms evolve toward one
lifestyle or the other.
Symbiosis is not always a matter of two organisms:
we are gaining increasing evidence that well-established
prokaryotic endosymbionts also maintain some meta-
bolic interplay with new facultative symbionts. Such
relationships could finish with the extinction of the
P-endosymbiont (as described in insects from the fam-
ily dryopthoridae, to which Sitophilus spp. belongs56, or
the Buchnera–Serratia consortium in cedar aphids21),
or the symbiotic consortia could continue over time
(as described in the Baumannia–Sulcia consortium in
sharpshooters15). Metagenomics and systems biology
are new approaches that can be used in the analysis
of complex symbiotic consortia. For example, the
metagenomes of sponge microbial communities have
been shown to contain genes and gene clusters for the
biosynthesis of biologically active natural products.
There is evidence to suggest that a mutually beneficial
Box 2 | Minimal cells and synthetic biology
The phenomenon of genome downsizing that has been observed in endosymbionts,
intracellular parasites and organelles has inspired a research programme on minimal
life, based on the hypothesis that genomes must retain essential genes that are
involved in housekeeping functions, and a minimum number of metabolic
transactions for cellular survival and replication. Apart from shedding light on this
fundamental topic, the search for minimal genomes is of much value in the context
of synthetic biology. In fact, one of the aims of this emerging field is the definition
and chemical synthesis of a minimal genome, and its incorporation and expression
in a suitable chassis, either derived from a cell (top-down strategy) or starting from a
simple chemical system, such as a liposome (bottom-up approach). The top-down
approach, also called genome-driven cell engineering107 or the minimal cell
project108, is expected to provide methods for fabricating engineered cells, which
will have a big impact both on biotechnology and on our basic understanding of
The first attempt to define a minimal genome based on comparative genomics used
the genomes of two human parasitic bacteria: Haemophilus influenzae and
Mycoplasma genitalium110. Owing to their parasitic lifestyles, these two bacteria have
reduced genomes when compared with their closest free-living relatives. The
analysis led to the proposal of a minimal gene set composed of just 256 genes, most
of them involved in genetic-information storage and processing, protein
chaperoning and a limited metabolic capability. Later, a combined study of all
published research using computational or experimental methods, including the
comparison of reduced genomes from insect endosymbionts, was used to define
the minimal core of essential genes for a free-living bacterium thriving in a
chemically rich environment111. The main difference between this study and previous
efforts (see rEF. 112 and references therein) was the emphasis on the functional
completeness of the minimal metabolism encoded by the proposed gene repertoire
(involving 62 protein-coding genes out of a minimal set of 208 genes). This aspect has
been explored further, demonstrating the stoichiometric consistency of the minimal
metabolic network, as well as providing insights into some of its architectural
properties, such as its size, clustering and robustness113.
It must be stressed that these latter studies present just one possible form of a
minimal metabolism. Metabolic complexity is ecologically dependent; it is a
function of the chemical richness of the environment and the primary energy
source(s) available to the living system (FIG. 1). Different versions of minimal gene
sets exist, that is, different combinations of metabolic pathways that can perform
the essential biological functions of self-maintenance, self-reproduction and
evolution114 under the same external conditions. However, it is possible to define
which functions should be performed in any living cell in a specified niche, and list
the genes that would be necessary to maintain such functions. In this context,
comparative and evolutionary genomics of endosymbionts provide insights into the
different routes that endosymbionts have taken to solve the challenges of their own
maintenance under the diverse selective pressures and environmental constraints
that are the result of different host lifestyles.
nATurE rEvIEWS | genetics
AdvAncE OnLInE PubLIcATIOn | 227
© 2008 Nature Publishing Group
The design and fabrication of
artificial biological systems,
with the aim of either
optimizing their performance in
the context of their
technological utility or of
deepening our understanding
of the naturally occurring
Hypsa, V. & Dale, C. In vitro culture and phylogenetic
analysis of ‘Candidatus Arsenophonus triatominarum’,
an intracellular bacterium from the triatomine bug,
Triatoma infestans. Int. J. Syst. Bacteriol. 47,
Dale, C. & Maudlin, I. Sodalis gen. nov. and
Sodalis glossinidius sp. nov., a microaerophilic
secondary endosymbiont of the tsetse fly Glossina
morsitans morsitans. Int. J. Syst. Bacteriol. 49,
Dale, C., Beeton, M., Harbison, C., Jones, T. &
Pontes, M. Isolation, pure culture, and
characterization of ‘Candidatus Arsenophonus
arthropodicus’, an intracellular secondary
endosymbiont from the hippoboscid louse fly
Pseudolynchia canariensis. Appl. Environ.
Microbiol. 72, 2997–3004 (2006).
Darby, A. C., Chandler, S. M., Welburn, S. C. &
Douglas, A. E. Aphid-symbiotic bacteria cultured
in insect cell lines. Appl. Environ. Microbiol. 71,
Sagan, L. On the origin of mitosing cells. J. Theor. Biol.
14, 255–274 (1967).
Sapp, J. Evolution by Association. A History of
Symbiosis (Oxford University Press, New York, 1994).
This book is the first historical report on the
symbiosis concept and its central role in shaping
contemporary evolutionary ideas about the origin
and diversification of life.
de Duve, C. The origin of eukaryotes: a reappraisal.
Nature Rev. Genet. 8, 395–403 (2007).
Margulis, L. Symbiosis in Cell Evolution. Microbial
Communities in the Archaean and Proterozoic Eons
(W. H. Freeman and Co., New York, 1993).
Shigenobu, S., Watanabe, H., Hattori, M., Sakaki, Y. &
Ishikawa, H. Genome sequence of the endocellular
bacterial symbiont of aphids Buchnera sp. APS.
Nature 407, 81–86 (2000).
10. Tamas, I. et al. 50 million years of genomic stasis in
endosymbiotic bacteria. Science 296, 2376–2379
11. Akman, L. et al. Genome sequence of the endocellular
obligate symbiont of tsetse flies, Wigglesworthia
glossinidia. Nature Genet. 32, 402–407 (2002).
12. van Ham, R. C. et al. Reductive genome evolution in
Buchnera aphidicola. Proc. Natl Acad. Sci. USA 100,
13. Gil, R. et al. The genome sequence of Blochmannia
floridanus: Comparative analysis of reduced genomes.
Proc. Natl Acad. Sci. USA 100, 9388–9393 (2003).
14. Degnan, P. H., Lazarus, A. B. & Wernegreen, J. J.
Genome sequence of Blochmannia pennsylvanicus
indicates parallel evolutionary trends among bacterial
mutualists of insects. Genome Res. 15, 1023–1033
15. Wu, D. et al. Metabolic complementarity and
genomics of the dual bacterial symbiosis of
sharpshooters. PLoS Biol. 4, e188 (2006).
16. Nakabachi, A. et al. The 160-kilobase genome of the
bacterial endosymbiont Carsonella. Science 314, 267
17. Newton, I. L. et al. The Calyptogena magnifica
chemoautotrophic symbiont genome. Science 315,
18. Kuwahara, H. et al. Reduced genome of the
thioautotrophic intracellular symbiont in a deep-sea
clam, Calyptogena okutanii. Curr. Biol. 17, 881–886
19. Foster, J. et al. The Wolbachia genome of Brugia
malayi: endosymbiont evolution within a human
pathogenic nematode. PLoS Biol. 3, e121 (2005).
20. Nakagawa, S. et al. Deep-sea vent ε-proteobacterial
genomes provide insights into emergence of
pathogens. Proc. Natl Acad. Sci. USA 104,
21. Pérez-Brocal, V. et al. A small microbial genome:
the end of a long symbiotic relationship? Science 314,
This paper, together with reference 16, shows that
the genomes of B. aphidicola BCc and C. ruddii
represent the lower limits of genome size in insect
endosymbionts. Further research in these systems
might shed light on the evolutionary fate of
22. Toh, H. et al. Massive genome erosion and functional
adaptations provide insights into the symbiotic
lifestyle of Sodalis glossinidius in the tsetse host.
Genome Res. 16, 149–156 (2006).
23. McCutcheon, J. P. & Moran, N. A. Parallel genomic
evolution and metabolic interdependence in an
ancient symbiosis. Proc. Natl Acad. Sci. USA 104,
24. Dale, C. & Moran, N. A. Molecular interactions
between bacterial symbionts and their hosts.
Cell 126, 453–465 (2006).
25. Baumann, P. Biology of bacteriocyte-associated
endosymbionts of plant sap-sucking insects.
Annu. Rev. Microbiol. 59, 155–189 (2005).
26. Russell, J. A., Latorre, A., Sabater-Munoz, B.,
Moya, A. & Moran, N. A. Side-stepping secondary
symbionts: widespread horizontal transfer across
and beyond the Aphidoidea. Mol. Ecol. 12,
27. Chen, D.-Q., Montllor, C. B. & Purcell, A. H. Fitness
effects of two facultative endosymbiotic bacteria on
the pea aphid, Acyrthosiphon pisum, and the blue
alfalfa aphid, A. kondoi. Entomol. Exp. Appl. 95,
28. Montllor, C. B., Maxmen, A. & Purcell, A. H.
Facultative bacterial endosymbionts benefit pea
aphids Acyrthosiphon pisum under heat stress.
Ecol. Entomol. 27, 189–195 (2002).
29. Russell, J. A. & Moran, N. A. Costs and benefits
of symbiont infection in aphids: variation among
symbionts and across temperatures. Proc. Biol. Sci.
273, 603–610 (2006).
30. Oliver, K. M., Moran, N. A. & Hunter, M. S. Variation
in resistance to parasitism in aphids is due to
symbionts not host genotype. Proc. Natl Acad. Sci.
USA 102, 12795–12800 (2005).
31. Oliver, K. M., Russell, J. A., Moran, N. A. &
Hunter, M. S. Facultative bacterial symbionts in aphids
confer resistance to parasitic wasps. Proc. Natl Acad.
Sci. USA 100, 1803–1807 (2003).
32. Scarborough, C. L., Ferrari, J. & Godfray, H. C.
Aphid protected from pathogen by endosymbiont.
Science 310, 1781 (2005).
33. Tsuchida, T., Koga, R. & Fukatsu, T. Host plant
specialization governed by facultative symbiont.
Science 303, 1989 (2004).
34. Simon, J. C. et al. Host-based divergence in
populations of the pea aphid: insights from
nuclear markers and the prevalence of facultative
symbionts. Proc. Biol. Sci. 270, 1703–1712
35. Leonardo, T. E. & Muiru, G. T. Facultative symbionts
are associated with host plant specialization in pea
aphid populations. Proc. Biol. Sci. 270 (Suppl 2),
36. Leonardo, T. E. & Mondor, E. B. Symbiont modifies
host life-history traits that affect gene flow. Proc. Biol.
Sci. 273, 1079–1084 (2006).
37. McGraw, E. A. & O’Neill, S. L. Wolbachia pipientis:
intracellular infection and pathogenesis in Drosophila.
Curr. Opin. Microbiol. 7, 67–70 (2004).
38. Welburn, S. C. & Maudlin, I. Tsetse–trypanosome
interactions: rites of passage. Parasitol. Today 15,
39. Moran, N. A. Symbiosis as an adaptive process and
source of phenotypic complexity. Proc. Natl Acad. Sci.
USA 104 (Suppl 1), 8627–8633 (2007).
40. Gomez-Valero, L. et al. Coexistence of Wolbachia with
Buchnera aphidicola and a secondary symbiont in the
aphid Cinara cedri. J. Bacteriol. 186, 6626–6633
41. Takiya, D. M., Tran, P. L., Dietrich, C. H. & Moran, N. A.
Co-cladogenesis spanning three phyla: leafhoppers
(Insecta: Hemiptera: Cicadellidae) and their dual
bacterial symbionts. Mol. Ecol. 15, 4175–4191 (2006).
42. Woyke, T. et al. Symbiosis insights through
metagenomic analysis of a microbial consortium.
Nature 443, 950–955 (2006).
This paper and reference 15 are two pioneering
works on the application of genomic analysis to
uncultivable microbial consortia, which opened up
a new field of research on symbiosis.
43. Giere, O. & Erséus, C. Taxonomy and new bacterial
symbioses of gutless marine Tubificidae (Annelida,
Oligochaeta) from the island of Elba (Italy).
Org. Divers. Evol. 2, 289–297 (2002).
44. Wernegreen, J. J. Genome evolution in bacterial
endosymbionts of insects. Nature Rev. Genet. 3,
45. Moran, N. A. Accelerated evolution and Muller’s
rachet in endosymbiotic bacteria. Proc. Natl Acad. Sci.
USA 93, 2873–2878 (1996).
46. Itoh, T., Martin, W. & Nei, M. Acceleration of genomic
evolution caused by enhanced mutation rate in
endocellular symbionts. Proc. Natl Acad. Sci. USA 99,
47. Wernegreen, J. J. For better or worse: genomic
consequences of intracellular mutualism and parasitism.
Curr. Opin. Genet. Dev. 15, 572–583 (2005).
48. Rocha, E. P. & Danchin, A. Base composition bias
might result from competition for metabolic resources.
Trends Genet. 18, 291–294 (2002).
49. Fares, M. A., Ruiz-Gonzalez, M. X., Moya, A., Elena, S. F.
& Barrio, E. Endosymbiotic bacteria: GroEL buffers
against deleterious mutations. Nature 417, 398 (2002).
50. Moya, A., Latorre, A., Sabater-Munoz, B. & Silva, F. J.
Comparative molecular evolution of primary
(Buchnera) and secondary symbionts of aphids
based on two protein-coding genes. J. Mol. Evol. 55,
51. Rispe, C., Delmotte, F., van Ham, R. C. & Moya, A.
Mutational and selective pressures on codon and
amino acid usage in Buchnera, endosymbiotic bacteria
of aphids. Genome Res. 14, 44–53 (2004).
52. Moran, N. A. & Mira, A. The process of genome
shrinkage in the obligate symbiont Buchnera
aphidicola. Genome Biol. 2, RESEARCH0054 (2001).
53. Moran, N. A. & Plague, G. R. Genomic changes
following host restriction in bacteria. Curr. Opin.
Genet. Dev. 14, 627–633 (2004).
54. Plague, G. R., Dunbar, H. E., Tran, P. L. & Moran, N. A.
Extensive proliferation of transposable elements in
heritable bacterial symbionts. J. Bacteriol. 190,
55. Heddi, A., Charles, H., Khatchadourian, C., Bonnot, G.
& Nardon, P. Molecular characterization of the
principal symbiotic bacteria of the weevil Sitophilus
oryzae: a peculiar G+C content of an endocytobiotic
DNA. J. Mol. Evol. 47, 52–61 (1998).
relationship exists, at least between some of the bacte-
ria and the sponges themselves88. Another interesting
example is the symbiotic gut microbiome of mammals,
in which we are starting to have a clearer picture of the
role of the symbiotic microbial consortia in the meta-
bolic phenotype of the mammalian host89. In this case,
an extensive microbial–mammalian co-metabolism
exists. As new complete genomes of eukaryote hosts
become available, we will be able to make new meta-
bolic inferences to understand the interplay between
the host and its associated microbiota. Genomics and
metagenomics are becoming essential tools to provide
new insights into symbiotic associations, including
those that are highly relevant to biotechnology and
Finally, synthetic biology will benefit from the new
insights gained by comparative genomic analyses
of extremely reduced genomes (BOX 2). This type of
research makes important contributions to the concep-
tually intriguing and biologically fundamental question
of the minimal genomic basis for sustaining cellular life.
Many efforts in this area have attracted a great deal of
public attention regarding both biosecurity risks and
228 | AdvAncE OnLInE PubLIcATIOn
© 2008 Nature Publishing Group
56. Lefevre, C. et al. Endosymbiont phylogenesis in the
Dryophthoridae weevils: evidence for bacterial
replacement. Mol. Biol. Evol. 21, 965–973 (2004).
57. Wu, M. et al. Phylogenomics of the reproductive
parasite Wolbachia pipientis wMel: a streamlined
genome overrun by mobile genetic elements. PLoS
Biol. 2, e69 (2004).
58. Silva, F. J., Latorre, A. & Moya, A. Genome size
reduction through multiple events of gene disintegration
in Buchnera APS. Trends Genet. 17, 615–618 (2001).
59. Gomez-Valero, L., Silva, F. J., Simon, J. C. & Latorre, A.
Genome reduction of the aphid endosymbiont
Buchnera aphidicola in a recent evolutionary time
scale. Gene 389, 87–95 (2007).
60. Gomez-Valero, L., Latorre, A. & Silva, F. J. The
evolutionary fate of nonfunctional DNA in the bacterial
endosymbiont Buchnera aphidicola. Mol. Biol. Evol.
21, 2172–2181 (2004).
61. Tamames, J. et al. The frontier between cell and
organelle: genome analysis of Candidatus Carsonella
ruddii. BMC Evol. Biol. 7, 181 (2007).
62. Wilson, A. C. et al. A dual-genome microarray for the
pea aphid, Acyrthosiphon pisum, and its obligate
bacterial symbiont, Buchnera aphidicola. BMC
Genomics 7, 50 (2006).
63. Wilcox, J. L., Dunbar, H. E., Wolfinger, R. D. &
Moran, N. A. Consequences of reductive evolution
for gene expression in an obligate endosymbiont.
Mol. Microbiol. 48, 1491–1500 (2003).
64. Koga, R., Tsuchida, T. & Fukatsu, T. Changing
partners in an obligate symbiosis: a facultative
endosymbiont can compensate for loss of the essential
endosymbiont Buchnera in an aphid. Proc. Biol. Sci.
270, 2543–2550 (2003).
65. Ren, Q. & Paulsen, I. T. Comparative analyses of
fundamental differences in membrane transport
capabilities in prokaryotes and eukaryotes.
PLoS Comput. Biol. 1, e27 (2005).
66. Fiala-Médioni, A. & Métivier, C. Ultrastructure of the
gill of the hydrothermal vent bivalve Calyptogena
magnifica, with a discussion of its nutrition. Mar. Biol.
90, 215–222 (1986).
67. Baumann, P., Moran, N. A. & Baumann, L. in The
Prokaryotes (ed. Dworkin, M.) 1–67 (Springer, New
68. Fiala-Médioni, A., Michalski, J. C., Jollès, J., Alonso, C.
& Montreuil, J. Lysosomic and lysozyme activities in
gills of bivalves from deep hydrothermal vents.
C. R. Acad. Sci. Paris 317, 239–244 (1994).
69. Nakabachi, A. et al. Transcriptome analysis of the
aphid bacteriocyte, the symbiotic host cell that harbors
an endocellular mutualistic bacterium, Buchnera. Proc.
Natl Acad. Sci. USA 102, 5477–5482 (2005).
70. Dale, C., Young, S. A., Haydon, D. T. & Welburn, S. C.
The insect endosymbiont Sodalis glossinidius utilizes a
type III secretion system for cell invasion. Proc. Natl
Acad. Sci. USA 98, 1883–1888 (2001).
71. Young, G. M., Schmiel, D. H. & Miller, V. L. A new
pathway for the secretion of virulence factors by
bacteria: The flagellar export apparatus functions as a
protein-secretion system. Proc. Natl Acad. Sci. USA
96, 6456–6461 (1999).
72. Maezawa, K. et al. Hundreds of flagellar basal bodies
cover the cell surface of the endosymbiotic bacterium
Buchnera aphidicola sp. strain APS. J. Bacteriol. 188,
73. McFall-Ngai, M. Adaptive immunity: care for the
community. Nature 445, 153 (2007).
74. Hoffmann, J. A. The immune response of Drosophila.
Nature 426, 33–38 (2003).
75. Heddi, A. et al. Molecular and cellular profiles of
insect bacteriocytes: mutualism and harm at the initial
evolutionary step of symbiogenesis. Cell. Microbiol. 7,
76. Zaidman-Remy, A. et al. The Drosophila amidase
PGRP-LB modulates the immune response to bacterial
infection. Immunity 24, 463–473 (2006).
77. Anselme, C., Vallier, A., Balmand, S., Fauvarque, M.-O.
& Heddi, A. Host PGRP gene expression and bacterial
release in endosymbiosis of the weevil Sitophilus
zeamais. Appl. Environ. Microbiol. 72, 6766–6772
78. Ghedin, E. et al. Draft genome of the filarial nematode
parasite Brugia malayi. Science 317, 1756–1760
The complete genome sequence of the nematode
B. malayi, together with the previously reported
genome sequence of its Wolbachia endosymbiont,
will enable a systems approach for the study of
79. Kato, Y. & Komatsu, S. ASABF, a novel cysteine-rich
antibacterial peptide isolated from the nematode
Ascaris suum. Purification, primary structure, and
molecular cloning of cDNA. J. Biol. Chem. 271,
80. Braendle, C. et al. Developmental origin and evolution
of bacteriocytes in the aphid–Buchnera symbiosis.
PLoS Biol. 1, e21 (2003).
This paper was the first study of the development
and evolution of a bacteriocyte containing
81. Lwoff, A. L’évolution physiologique. Étude des pertes
des fonctions chez les microorganismes (Hermann,
82. Timmis, J. N., Ayliffe, M. A., Huang, C. Y. & Martin, W.
Endosymbiotic gene transfer: organelle genomes
forge eukaryotic chromosomes. Nature Rev. Genet. 5,
83. Martin, W. Gene transfer from organelles to the
nucleus: frequent and in big chunks. Proc. Natl Acad.
Sci. USA 100, 8612–8614 (2003).
84. Hotopp, J. C. et al. Widespread lateral gene transfer
from intracellular bacteria to multicellular eukaryotes.
Science 317, 1753–1756 (2007).
85. Bhattacharya, D., Archibald, J. M., Weber, A. P. M. &
Reyes-Prieto, A. How do endosymbionts become
organelles? Understanding early events in plastid
evolution. Bioessays 29, 1239–1246 (2007).
86. C. elegans Sequencing Consortium. Genome sequence
of the nematode Caenorhabditis elegans: a platform
for investigating biology. Science 282, 2012–2018
87. Myers, E. W. et al. A whole-genome assembly of
Drosophila. Science 287, 2196–2204 (2000).
88. Kennedy, J., Marchesi, J. R. & Dobson, A. D.
Metagenomic approaches to exploit the
biotechnological potential of the microbial consortia
of marine sponges. Appl. Microbiol. Biotechnol. 75,
89. Martin, F. P. et al. A top-down systems biology view
of microbiome–mammalian metabolic interactions
in a mouse model. Mol. Syst. Biol. 3, 112 (2007).
90. Paracer, S. & Ahmadjian, V. Symbiosis.
An Introduction to Biological Associations
(Oxford University Press, New York, 2000).
91. Kneip, C., Lockhart, P., Voss, C. & Maier, U. G.
Nitrogen fixation in eukaryotes — new models for
symbiosis. BMC Evol. Biol. 7, 55 (2007).
92. Schink, B. Energetics of syntrophic cooperation in
methanogenic degradation. Microbiol. Mol. Biol. Rev.
61, 262–280 (1997).
93. Stewart, F. J., Newton, I. L. & Cavanaugh, C. M.
Chemosynthetic endosymbioses: adaptations to
oxic–anoxic interfaces. Trends Microbiol. 13,
94. Minic, Z. & Herve, G. Biochemical and enzymological
aspects of the symbiosis between the deep-sea
tubeworm Riftia pachyptila and its bacterial
endosymbiont. Eur. J. Biochem. 271, 3093–3102
95. Zientz, E., Dandekar, T. & Gross, R. Metabolic
interdependence of obligate intracellular bacteria and
their insect hosts. Microbiol. Mol. Biol. Rev. 68,
96. Croft, M. T., Lawrence, A. D., Raux-Deery, E.,
Warren, M. J. & Smith, A. G. Algae acquire vitamin
B12 through a symbiotic relationship with bacteria.
Nature 438, 90–93 (2005).
97. Buchner, P. Endosymbiosis of Animals with Plant
Microorganisms (Interscience, New York, 1965).
98. Taylor, M. W., Radax, R., Steger, D. & Wagner, M.
Sponge-associated microorganisms: evolution,
ecology, and biotechnological potential. Microbiol.
Mol. Biol. Rev. 71, 295–347 (2007).
99. Backhed, F., Ley, R. E., Sonnenburg, J. L., Peterson, D. A.
& Gordon, J. I. Host–bacterial mutualism in the
human intestine. Science 307, 1915–1920 (2005).
100. Stingl, U., Radek, R., Yang, H. & Brune, A.
‘Endomicrobia’: cytoplasmic symbionts of termite gut
protozoa form a separate phylum of prokaryotes.
Appl. Environ. Microbiol. 71, 1473–1479 (2005).
101. Fieseler, L., Quaiser, A., Schleper, C. & Hentschel, U.
Analysis of the first genome fragment from the marine
sponge-associated, novel candidate phylum
Poribacteria by environmental genomics.
Environ. Microbiol. 8, 612–624 (2006).
102. Moreira, D. et al. Global eukaryote phylogeny:
Combined small- and large-subunit ribosomal DNA
trees support monophyly of Rhizaria, Retaria and
Excavata. Mol. Phylogenet. Evol. 44, 255–266 (2007).
103. Rodriguez-Ezpeleta, N. et al. Toward resolving
the eukaryotic tree: the phylogenetic positions
of jakobids and cercozoans. Curr. Biol. 17,
104. Lecointre, G. & Le Guyader, H. The Tree of Life.
A Phylogenetic Classification (Harvard University
Press Reference Library) (Belknap Press,
105. Delsuc, F., Brinkmann, H., Chourrout, D. & Philippe, H.
Tunicates and not cephalochordates are the closest
living relatives of vertebrates. Nature 439, 965–968
106. Bourlat, S. J. et al. Deuterostome phylogeny
reveals monophyletic chordates and the new
phylum Xenoturbellida. Nature 444, 85–88
107. O’Malley, M., Powell, A., Davies, J. & Calvert, J.
Knowledge-making distinctions in synthetic biology.
Bioessays 30, 57–65 (2007).
108. Luisi, P. L. Chemical aspects of synthetic biology.
Chem. Biodivers. 4, 603–621 (2007).
109. Peretó, J. & Català, J. The renaissance of synthetic
biology. Biol. Theor. 2, 128–130 (2007).
110. Mushegian, A. R. & Koonin, E. V. A minimal gene set
for cellular life derived by comparison of complete
bacterial genomes. Proc. Natl Acad. Sci. USA 93,
111. Gil, R., Silva, F. J., Pereto, J. & Moya, A.
Determination of the core of a minimal
bacterial gene set. Microbiol. Mol. Biol. Rev. 68,
This Review proposes a minimal genetic
repertoire for a hypothetical heterotrophic
bacterium, based on comparative genomics
of endosymbionts, experimentally determined
gene essentiality and the functions of a coherent
112. Klasson, L. & Andersson, S. G. Evolution of
minimal-gene-sets in host-dependent bacteria.
Trends Microbiol. 12, 37–43 (2004).
113. Gabaldon, T. et al. Structural analyses of
a hypothetical minimal metabolism. Philos. Trans.
R. Soc. Lond., B, Biol. Sci. 362, 1751–1762
114. Ruiz-Mirazo, K., Pereto, J. & Moreno, A.
A universal definition of life: autonomy and
open-ended evolution. Orig. Life Evol. Biosph. 34,
Financial support was provided by grants GV/2007/050 from
Generalitat Valenciana, Spain, to R.G and BFU2006-06003
from Ministerio de Educación y Ciencia (MEC), Spain, to A.L.
R.G is a recipient of a contract in the ‘Ramón y Cajal’ pro-
gramme from the MEC, Spain. Our thanks to H. Escrivá
(CNRS, Banyuls sur Mer), P. López (CNRS, Orsay) and D.
Moreira (CNRS, Orsay) for their advice in the design of the
eukaryotic phylogenetic tree in BOX 1.
Entrez Genome Project:
Acyrthosiphon pisum | Baumannia cicadellinicola |
Blochmannia floridanus | Brugia malayi | Buchnera aphidicola |
Caenorhabditis elegans | Carsonella ruddii | Chromobacterium
violaceum | Drosophila melanogaster | Sodalis glossinidius |
Sulcia muelleri | Wigglesworthia glossinidia |
Andrés Moya’s homepage:
Juli Peretó’s homepage:
Rosario Gil’s homepage:
Amparo Latorre’s homepage: http://www.uv.es/cavanilles/
Cavanilles Institute web page: www.uv.es/cavanilles
Genomes OnLine Database: http://www.genomesonline.org
Microbial Genome Database for Comparative Analysis:
The Sanger Institute Glossina Genome Project:
See online article: S1 (figure) | S2 (table)
All links ARe Active in tHe Online PDf
nATurE rEvIEWS | genetics
AdvAncE OnLInE PubLIcATIOn | 229
© 2008 Nature Publishing Group