See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/230869915

Insect symbionts as hidden players in insect-plant interactions

Article in Trends in Ecology & Evolution · September 2012

DOI: 10.1016/j.tree.2012.08.013 · Source: PubMed

CITATIONS READS

309 1,850

3 authors:

Enric Frago Marcel Dicke

Cirad - La recherche agronomique pour le développement Wageningen University & Research
50 PUBLICATIONS 2,086 CITATIONS 927 PUBLICATIONS 56,252 CITATIONS

SEE PROFILE SEE PROFILE

Charles Godfray
University of Oxford
521 PUBLICATIONS 62,887 CITATIONS

SEE PROFILE

All content following this page was uploaded by Enric Frago on 05 November 2019.

The user has requested enhancement of the downloaded file.

 Review

Insect symbionts as hidden players in

insect–plant interactions
Enric Frago1, Marcel Dicke2, and H. Charles J. Godfray1
1
Department of Zoology, University of Oxford, South Parks Road, Oxford OX1 3PS, UK
2
Laboratory of Entomology, Wageningen University, P.O. Box 8031, 6700EH Wageningen, The Netherlands

There is growing evidence of the importance of microbi- and become a pest. Experimental exchange of the gut sym-
al mutualistic symbioses in insect–plant interactions. biont ‘Candidatus Ishikawaella capsulata’ between the two
Mutualists may affect host plant range and enable species showed that the ability to grow succesfully and
insects to manipulate plant physiology for their own develop on the crop is purely determined by the bacterium
benefit. The plant can also be a route for the horizontal [7]. The pea aphid (Acyrthosiphon pisum) is known to be
transfer of mutualistic microorganisms among their host associated with at least seven species of facultative symbi-
insects. Where this occurs, selection for improved trans- ont and the chestnut weevil (Curculio sikkimensis) with six;
mission might cause the insect mutualist to damage the their frequency of occurrence varies across the multiple host
plant and become a plant pathogen. Insect microbial plants on which the insects feed [6,8]. There is some evidence
associates can influence ecological communities by suggesting that these symbionts may affect host plant use,
changing the way the plant interacts with their hosts’ although a recent systematic study of A. pisum failed to find
competitors and natural enemies. We review recent any consistent effects [9]. However, Tsuchida and coworkers
research in this field and suggest that insect mutualists [10] showed recently that injection of a symbiont from a
may be more important ‘hidden players’ in insect–plant clover-adapted pea aphid allowed another aphid species
interactions than is currently realised. that normally could not feed on clover to use this host plant.
The extent to which symbionts directly influence host range
Insect symbionts and host plant use thus remains unclear and requires more experimental stud-
Plants are unpromising food for insects, because they large- ies. A particular issue is that associations may reflect not
ly comprise indigestible structural compounds such as cel- causation but the influence of correlated third factors, or
lulose and lignin marinated in a diversity of deterrent or might be a function of the history of symbiont colonisation. It
toxic chemicals [1]. Insects have evolved many different will be interesting to see whether symbionts are important
strategies to feed on plants including associations with in allowing their hosts to exploit novel ecological niches such
mutualistic symbionts (see Glossary), which can be impor- as non-native crops, as was found in the stinkbug.
tant mediators of direct and indirect interactions between The effects of microbial symbionts on their insect hosts
herbivorous insects and their host plants [2] (Figure 1). For may be considerably subtler than allowing them to utilise
example, almost all aphids carry an obligate bacterial sym- different plant species and here we argue that their role as
biont (Buchnera) that synthesizes certain essential amino hidden players in insect–plant interactions may be far
acids and hence allows the insect to feed on a purely phloem greater than currently appreciated. We begin by showing
diet that otherwise would not sustain development. In the that symbionts may actively manipulate food plant physi-
past decade, it has become apparent that these symbiotic ology and antiherbivore defences to their hosts’ advantage.
interactions can be much more complex than hitherto appre-
ciated and that they can have ramifications that affect whole
ecosystems [3–5]. Although obligatory symbionts like Buch-
nera are required for survival in every individual of a Glossary
species, and in many ways are more like organelles than
Commensalistic symbiosis: a symbiotic relationship in which one partner
independent organisms, many facultative or secondary sym- benefits with no associated costs to the other.
bionts have been discovered. These are not essential for host Facultative (or secondary) symbiont: a symbiont that is not essential for the
survival and are often found in only a fraction of the popu- survival of its host, although it is often beneficial.
Horizontal transmission: transmission of symbionts (including mutualists and
lation, in which they can have striking effects on host pathogens) other than from parent to offspring (may be interspecific or
phenotype [5,6], including host plant utilisation. intraspecific).
Indirect plant-mediated interaction: the effect of one species of herbivore on
The stinkbugs Megacopta punctatissima and M. cri-
another through changes in the quality of a shared host plant.
braria provide a clear example of insect symbiont involve- Insect vector: an insect that carries and transmits a pathogen.
ment in host plant use. Both species feed on native plants Mutualistic symbiosis: a reciprocally beneficial symbiosis.
Obligate (or primary) symbiont: a symbiont that is required for the survival of
in Japan, but only M. punctatissima can colonize soy bean its host.
Pathogen: an antagonistic microorganism that causes disease.
Corresponding author: Frago, E. (enric.frago@zoo.ox.ac.uk). Symbiosis: a lasting interaction between two species in an intimate associa-
Keywords: bacterial symbionts; fungal symbionts; community ecology; facultative tion. The term is sometimes restricted to cases in which both partners benefit.
symbiosis; horizontal transmission; host plant use; indirect plant-mediated interac- Vertical transmission: transmission of symbionts from parent to offspring.
tions; mutualism; induced plant defences.

0169-5347/$ – see front matter ß 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tree.2012.08.013 Trends in Ecology and Evolution, December 2012, Vol. 27, No. 12 705
 Review Trends in Ecology and Evolution December 2012, Vol. 27, No. 12

(a) (b) Insect symbionts may manipulate plant physiology

A4 Insect herbivores induce various species-specific defences
A1
in plants that are expressed both locally and systemically.
They affect the herbivore directly, but also indirectly by
A5 promoting the effectiveness of its natural enemies
[1,12,13]. The expression of this plastic phenotype, which
B1
takes the form of physiological changes in the plant, is
B2
mediated by a complex signal-transduction network in
A2 A3
which one antagonist can modulate the changes caused
by another [11,13,15]. There is mounting evidence that
insect-associated microorganisms might benefit their hosts
(c) (d) by interfering with the plant’s response to insect attack.
Perhaps the best known examples are the mutualistic
fungi of bark and ambrosia beetles, which attack wood
and make it digestible for their hosts’ larvae and which
may also assist the beetles in overcoming tree resistance
mechanisms [16]. The community composition of the sym-
biotic yeasts associated with Dendroctonus bark beetles is
strongly influenced by the genus of conifer they attack [17]
and it would be fascinating to explore whether the associ-
ation is adaptive by evaluating fitness consequences after
TRENDS in Ecology & Evolution
reciprocal exchange experiments. Endosymbiotic bacteria
Figure 1. Insect symbionts (represented by an insect carrying a bacterium) might have a similar though less well understood role.
influence insect–plant interactions at different levels through direct interactions Western corn rootworm, Diabrotica virgifera, is a major
(solid lines) as well as through indirect plant-mediated interactions (dashed lines).
Yellow lines represent symbiont-mediated interactions, deep green lines represent pest of maize that typically carries bacterial endosym-
insect–plant interactions, and pale green lines represent changes in plant state or bionts of the genus Wolbachia [18]. A microarray experi-
physiology. (a) Insect symbionts can directly influence host plant use in ment demonstrated that the activation of defence-related
herbivorous insects (A1), but also indirectly through changes to plant state or
physiology (A2). Such changes can affect other insects sharing the same host plant genes in the plant was higher when fed on by Wolbachia-
(A3). Insect symbionts can directly affect the host’s interactions with natural cured beetles [19]. Similarly, in the tomato psyllid, Bacter-
enemies (A4), but also indirectly through changes in plant physiology and the
icerca cockerelli, high concentrations of the bacterial endo-
emission of herbivore-induced plant volatiles (A5). (b) Insect symbionts can
colonize plants, which is a likely route for horizontal transmission (B1). Similarly, symbiont ‘Candidatus Liberibacter psyllaurous’ correlates
plant pathogens can be vectored by insects and this may evolve into mutualism if with reduced expression of plant defensive pathways in
the insect benefits from a diseased host plant (B2). (c) Different insect symbionts
can differentially affect insect host plant use and ultimately modulate interactions
tomato [20]. Although in these studies no effect on insect
between insects. (d) Communities of insect symbionts, including bacteria, fungi, fitness was observed, it would be interesting to explore
and viruses, are found in both insects and plants, where they can engage in these phenomena in the wild relatives of maize and toma-
complex interactions.
to; especially under stressful field conditions in which the
costs of plant-induced defences might be higher.
Insect herbivores have been reported to manipulate
Insights into how microorganisms can tilt the balance in directly host plant physiology for their own benefit [13].
favour of the insect are provided by recent advances in our However, it is important to check whether hidden players
mechanistic understanding of the signalling pathways and might be involved. A symbiotic partner could introduce
biochemical responses involved in plant defence and insect new genetic resources to give its host an advantage, in
counterattack [11,12], and how these responses can be particular the ability to synthesize bioactive molecules
modulated by microbial plant pathogens [13] or non-path- that had evolved in other interactions. In chewing insects,
ogenic plant microorganisms [14]. We go on to discuss for example, the most well-studied elicitors of plant
whether symbionts might more often than currently appre- defences are found in insect saliva and regurgitant, an
ciated use a food plant as a conduit from one insect host to important class being N-acylamino acids [21]. Although
another. The biology of these symbionts might come to any benefits for the insect host have yet to be identified, the
resemble the many plant pathogens that use insects as symbiotic gut flora has the potential to modulate the plant
vectors for transfer among host plants, and in the next response, because 50% of 23 strains of symbiotic bacteria
section we explore the fluid boundaries between these two isolated from the guts of noctuid caterpillars were able, at
life histories. In the last section, we consider unexpected least in vitro, to synthesize N-acylamino acids [22]. Anoth-
community interactions involving insect symbionts. In er possible example involves leaf-mining moths whose
addition to direct effects on host fitness, changes to plant larvae feed late in the autumn. They prevent the local
physiology brought about by microbial symbionts may area of the leaves in which they feed from senescing,
affect their hosts’ risk of attack by natural enemies such leading to conspicuous ‘green islands’ in autumnal-
as parasites, parasitoids, and pathogens. These changes coloured leaves. It has been known for some time that
can affect the whole community of insects feeding on green islands are caused by the leaf-miner secreting cyto-
particular host plants. We also explore complex community kinins, plant hormones that inhibit senescence, maintain
interactions among symbionts inside the same insect host chlorophyll, and control nutrient mobilization [23]. Recent
and their consequences for the plants on which they feed. studies have revealed that, in the apple leaf-mining moth
706
Review Trends in Ecology and Evolution December 2012, Vol. 27, No. 12

Phyllonorycter blancardella, the bacterial endosymbiont example of a bacterial symbiont using this route was found.
Wolbachia is involved in the production of cytokinins. A Rickettsia, one of numerous endosymbionts carried by
When the symbiont is removed, cytokinin concentrations the whitefly Bemisia tabaci, can colonize phloem cells,
are reduced in the mine and the green islands disappear, spread throughout the plant, and infect a symbiont-free
leading to increased moth mortality [24]. It would be insect [31]. B. tabaci is a highly polyphagous species, but
interesting to study whether the same strategy is used plant-mediated horizontal transmission could be demon-
by other leaf miners. strated only in cotton. If horizontal transmission is influ-
The molecular techniques needed to study insect–mi- enced by plant species, and if symbionts influence insect
crobe–plant interactions have only become available for fitness, this three-way interaction may affect selection for
non-model organisms in the past decade and we suspect host plant specialisation and population differentiation on
that many more examples of microbial manipulation of the different food plants.
host plant will supplement the relatively few examples we With few exceptions, natural selection acts on host and
review here. In particular, modern high-throughput mo- vertically transmitted symbionts in exactly the same way:
lecular methods that allow the complete metabolome or what is good for one is good for the other and their
transcriptome of the insect or plant to be profiled in the evolutionary interests are identical. However, horizontal
presence or absence of a symbiont is a highly efficient transmission through the plant implies that their evolu-
technique to discover potential new ways that microbial tionary interests may differ, with important ramifications
symbionts use to manipulate eukaryotes. for the biology of the symbiosis. In particular, the selection
on the symbiont to become an insect or plant mutualist or a
Plant-mediated transmission of insect symbionts parasite may be stronger. Symbionts found in the gut flora
Although insect symbionts are predominantly transmitted of insects are often closely related to microorganisms found
vertically, phylogenetic studies show that most facultative in the environment and in many cases the gut may be one
and many obligate symbionts can also be transmitted of a continuum of habitats they exploit and in which they
horizontally [25]. However, in many cases exactly how replicate. Some insect bacterial pathogens differ little from
horizontal transmission occurs is not known, although relatives in the soil except that they possess cassettes of
there are several examples where the host plant is impli- genes or ‘pathogenicity islands’ that enable them to attach
cated. The simplest way the plant may be involved is as a to and invade the host [32]. Were an insect symbiont to be
passive surface. For example, strict vertical transmission able to use a host plant not just as a passive route for
is not possible for some gut symbionts and in these species transmission but also for replication, it might greatly
successful inoculation of the next generation requires increase its rate of horizontal transmission. However,
juveniles to feed on material inoculated or contaminated when replication in the plant harms the plant, we are
with the adult gut flora. In the stink bug Riptortus pedes- more likely to think of the microorganism as a plant
tris, obligate gut mutualists are deposited in capsules pathogen than as an insect mutualist. By classifying micro-
beside the eggs on the leaf surface, where they are eaten organisms as one or the other, we risk missing an impor-
by the nymphs just after they hatch [26]. From phyloge- tant part of their biology, especially if they are studied by
netic data, the facultative secondary symbionts of aphids different research communities.
are known to be regularly transferred among host lineages
[27]. Although the mechanism involved in horizontal trans- When is a plant pathogen also an insect mutualist?
mission is not fully understood, bacteria can be found in In the same way that there are insect symbionts and
honeydew and siphuncular fluids, and can survive on plant pathogens that use the plant to infect new hosts, there
surfaces, where they might be picked up by other aphids are many examples of plant pathogens that use insects as
[28]. Transmission through surface contamination of the vectors to move among their hosts. Vectoring a plant
host plant is a relatively unspecialised means of passing on pathogen can directly or indirectly affect the vector’s fit-
symbionts and may facilitate the sharing of symbionts ness. The direct effects involve the costs or conceivably the
among different insect species feeding on the same plant benefits of pathogen carriage; the indirect effects include
species. the consequences for the insect of feeding on a diseased
There has been much discussion about whether insect plant. There are some examples of highly virulent patho-
symbionts can move between herbivore hosts systemically gens compromising transmission by causing plant mortal-
via the interior of the plant, especially the mutualists of ity [33,34], but others in which the diseased plant is less
species such as homopterans that feed with their mouth- able to defend itself against insect attack and becomes a
parts embedded in the plant’s vascular system. Systemic better food source for the herbivorous insect [35]. The plant
colonization of plant tissues might not only provide sym- pathogen is now also an insect mutualist. Such a mutual-
bionts with the opportunity to reach new insect hosts but ism may occur through crosstalk between signalling path-
could also provide a refuge for the microorganism at times ways involved in plant responses to plant pathogens and
when the insect is rare or absent. Symbionts might be herbivorous insects, in particular the salicylic and the
selected to reside in long-lived plants in a manner analo- jasmonic acid pathways [11,13,15,36]. These two signal-
gous to that of delayed germination in seeds. Numerous transduction pathways interact and upregulation of one
mutualistic insect viruses are known to be transmitted via can result in the attenuation of defence responses mediat-
plants [29] and entomopathogenic fungi may do likewise, ed by the other, with important implications for insect
because they can be found inside plant tissues where they fitness [36]. Numerous insects, including aphids, thrips,
seem to have no detrimental effects [30]. Recently, the first whiteflies, and leafhoppers, that vector plant viruses or
707
Review Trends in Ecology and Evolution December 2012, Vol. 27, No. 12

pathogenic bacteria and fungi, perform better on, and are mounting evidence that microorganisms associated with
more attracted to, plants infected with the pathogen insects can affect community structure and processes in a
[29,35,37–39]. A particularly clear example is provided similar manner. When symbionts manipulate plant phys-
by the success of the invasive B biotype of the whitefly iology in ways that benefit their insect hosts, other herbi-
B. tabaci, which is spreading rapidly around the world vores sharing the same plant might also be affected.
partially due to mutualism with a virus. The B biotype has Successful attack of jack pines, Pinus banksiana, by the
increased fitness on diseased plants, whereas the fitness of bark beetle, Dendroctonus ponderosae, is thought to be
the native biotype is harmed [40,41]. Virus suppression of facilitated by its fungal associate Grosmannia clavigera.
the jasmonic acid-mediated defences in the plant by the B However, inoculation of the fungus by the beetle leads to
biotype may provide it a competitive advantage and ac- increased monoterpene concentrations in needles, which in
count for its invasion and spread [42]. turn increases the rate of feeding by the jack pine bud-
The establishment of new symbioses can also lead to worm, Choristoneura pinus pinus, presumably to compen-
novel plant pathogens and hence affect the importance of sate for decreased food quality [53]. This example suggests
an insect as a pest. Particularly dramatic examples occur that D. ponderosae might benefit from its symbiotic part-
in invasive bark and ambrosia beetles, where novel beetle– ner not only through increased success when infesting the
fungus associations can markedly increase the damage the tree, but also by reduced food quality for interspecific
beetle causes and in some cases allows beetles that nor- competitors.
mally attack dead trees to colonise live ones [43]. The In a similar manner, insect symbionts may benefit their
beetle Dendroctonus valens has recently colonized China, hosts by influencing the degree to which they can vector
where it has acquired novel indigenous isolates of its plant pathogens. Given that virus infection of plants has an
fungal associate Leptographium procerum. The new mutu- enormous influence on host plant quality, the symbiont
alisms make the beetle a much more serious pest than it is status of these insects will have major consequences for the
in its original home, partly because the fungus alters the trophic web of insects based on that host plant. In aphids
tree’s profile of volatile chemicals in a way that recruits and whiteflies, GroEL chaperone proteins are produced by
more beetles to the tree [44,45]. symbiotic bacteria and are known to bind plant viruses
Phylogenetic studies provide further evidence of the [54,55]. The whitefly B. tabaci harbours a range of different
porous boundary between plant pathogens and insect facultative symbionts known to produce GroEL proteins,
mutualists. Most bacteria in the genus Spiroplasma are but the tomato yellow leaf curl virus binds only to those
insect mutualists or pathogens, although some species are synthesised by the bacterium Hamiltonella, so that virus
known to be plant pathogens vectored by leafhoppers or transmission is most efficient in B. tabaci biotypes carrying
psyllids [46]. The clade Arsenophonus is one of the most this species [56].
widely distributed bacterial symbionts colonising insects
and other arthropods, but has two known species that Interactions with natural enemies
infest plants [47]. One of the latter, ‘Candidatus A. phy- Insect symbionts are known to affect interactions among
topathogenicus’ is vectored among plants by the planthop- herbivores and their natural enemies. Some parasitoid
per Pentastiridius leporinus, in which it can also be species are known to use volatiles produced directly or
transmitted vertically. This intimate association with indirectly by their host’s symbiont at the host location [57–
the insect may reflect its evolutionary origin as an arthro- 59]. This was first demonstrated in the laboratory with
pod symbiont [48]. parasitoids of the wood wasp Sirex noctilio, which home in
Thus, evidence is accumulating that at least some mu- on volatiles produced by the mutualistic wood-rotting
tualistic insect symbionts might have more intimate con- fungus Amylostereum [60]. Recent studies of bark beetles
nections with their host’s food plant than previously have shown that the composition of the parasitoid com-
appreciated and that some insect-transmitted plant patho- munity attracted to logs treated with different fungal
gens may benefit their vectors. Due to the extent that these mutualists is influenced by the particular fungal species
different interactions are studied by different research present [61,62].
communities, we may falsely be seeing a dichotomy in Insect symbionts may also affect interactions with nat-
what is actually a continuum of interactions. ural enemies through their influence on their host’s sus-
ceptibility and resistance. This has been most intensively
Symbiont interactions at the community level studied in the pea aphid A. pisum, in which one facultative
Plant-mediated indirect effects symbiont, Hamiltonella defensa, confers resistance to
Ecologists have for many years built food webs that de- parasitoids [63], whereas another, Regiella insecticola,
scribe the trophic interactions between herbivores and provides protection from a fungal pathogen [64]. An endo-
their natural enemies. However, recent studies have symbiont in the genera Rickettsiella is also likely to affect
shown that there are many further unsuspected non-tro- aphid interactions with higher trophic levels, because it
phic indirect interactions in such webs, often mediated by affects aphid colour [65], a phenotype known to influence
changes in plant quality [49,50]. Symbionts can further the rate at which insects are predated or parasitized.
modulate these interactions; for instance, the structure of Symbionts that reduce risk of natural enemy mortality
the food web based on a particular plant species (a source are particularly common in aphids on certain host plant
web) can be influenced by whether the plant is infected by species, although the hypothesis that varying natural
an endophytic fungus or a soil-borne microorganism enemy pressure on different plant species selects for dif-
[51,52]. Although currently less well understood, there is ferent symbiont communities [66,67] has yet to be tested.
708
Review Trends in Ecology and Evolution December 2012, Vol. 27, No. 12

As discussed above, some insect vectors of plant viruses Box 1. Research challenges
benefit from the disease the pathogen causes. In the case of Ecological understanding of insect–plant interactions has made
the western flower thrips, Frankliniella occidentalis, the tremendous progress in recent decades. A major development has
insect exhibits increased growth rates when feeding on been to investigate insect–plant interactions in a multitrophic
infected plants, although its size at maturity is not affect- perspective comprising dynamic communities of plant-associated
ed. It appears that the chief advantage of growing faster is species [75–78]. In recent years, appreciation of the complexity of
such systems has increased profoundly with the recognition that
that larvae are vulnerable to the predatory mites Neoseiu- insect–plant interactions may be decisively influenced by microbial
lus cucumeris and Iphiseius degenerans for a shorter period symbionts that in themselves may comprises complex commu-
of time, leading to reduced predation on virus-infested nities, as reviewed here. Integrating research on such hidden
plants [68]. players is likely to provide new breakthroughs in our understanding
of the ecology of insect–plant interactions. Some of the research
challenges ahead are:
Complex interaction among symbiont communities  obtaining a more integrated ecological and mechanistic under-
Multiple infections of symbionts in the same insect host are standing of the role of facultative symbionts of herbivorous
common and may lead to complex community interactions insects in manipulating host plant physiology, in particular
with important consequences for plants. The plant-associ- through effects on the plant’s induced response to herbivory;
 discovering the routes by which facultative symbionts are
ated bacteria Xylella fastidiosa and Alcaligenes xylosoxi-
transmitted horizontally in the plant, in particular the frequency
dans subsp. denitrificans both colonise the cibarial region with which plants act as conduits for symbiont movement
of the alimentary track of their vector, the glassy-winged between insects;
sharpshooter, Homalodisca vitripennis, where they com-  understanding what determines when a symbiont transmitted via
pete for adhesion sites [69]. Their relative success may a plant becomes pathogenic or mutualistic to both the insect and
the plant;
have important consequences for the plant, because X.
 unravelling the effects of insect symbionts on their hosts
fastidiosa causes Pierce’s disease in grapes whereas A. competitors through changes in plant quality;
xylosoxidans colonizes plants as a commensal [69,70].  understanding how the viruses (including phages) and other
Complex interactions among insect symbionts can also mobile genetic elements found in symbionts may mediate the
occur between microorganisms as distant as fungi and fitness effects they have on their host;
 exploring how the domestication of plants has affected symbiont–
bacteria. The southern pine beetle, Dendroctonus frontalis, insect–plant interactions (to the benefit or detriment of agricultur-
has two fungal associates: one (Entomocorticium sp.) that al productivity); and
digests wood for the developing larvae and another  understanding why the creation of novel symbioses can affect the
(Ophiostoma minus) that is thought to be involved in importance of the insect host as a pest.
combating tree defences during initial colonization. Once
the beetle has become established, O. minus ceases to be
beneficial to the insect because it competes with Entomo- the intricate responses of plants to herbivore and pathogen
corticium [71]. It has been found that yet another beetle attack, and how plant antagonists manipulate signalling
symbiont, a bacterium (Streptomyces sp.), produces an pathways for their own benefit. We are only just beginning
antibiotic that inhibits O. minus growth [72,73]. This to understand how insect mutualists may modulate these
illustrates the complexity of the effects of symbiont com- interactions and the next decade is likely to see major
munities on the phenotype of their host. progress in unravelling the mechanistic basis of these
interactions. In Box 1, we list what we believe are some
Concluding remarks and future perspectives of the most interesting challenges faced by researchers in
Aphids are among the best-studied of all insects because of the immediate future.
their great importance as pests in temperate agriculture. Finally, how insect mutualists interact with plants is
Yet in the last two decades, our understanding of their potentially of major applied significance. The invasion of
biology has been overturned: their capacity to withstand new pests has often been facilitated by their mutualists
environmental insults and to protect themselves from and some novel interactions have resulted in new and more
parasitoids and pathogenic fungi is greatly influenced by virulent insect pests [41,43,45,74]. Manipulating sym-
the facultative symbionts they carry. Although we knew bionts may be exploited to improve pest control. Finding
before that their primary symbiont Buchnera affects their out more about insects, plants, and their microbial associ-
nutritional biology, we now know that their facultative ates will be both fascinating and useful.
symbionts also play a role and that they have other com-
plex effects on the aphid’s interaction with the host plant. Acknowledgements
M.D. was supported by the ESF EUROCORES programme EUROVOL
As with aphids, so with whiteflies, leaf miners, and bark and an NWO TOP-grant (854.10.010). E.F. was funded by the ESF
beetles; increasingly, we are finding unexpected influences Frontiers In Speciation Research (Short Visit Travel Grants) programme
of symbiotic microorganisms – hidden players in the inter- (FroSpects4090) and by the Conselleria d’Educació de la Generalitat
actions between herbivorous insects, their food plants and Valenciana (Spain) Programa VALi+d per a Investigadors en Fase
Postdoctoral programme (APOSTD/2010/062). The authors are grateful
their natural enemies. Although it would be wrong to
to Paul Craze and four anonymous reviewers for their insightful and
expect symbionts to be involved in all aspects of their hosts’ helpful comments on an earlier version of this manuscript, and to Sara
biology, the possibility of a hidden player should always be Mitri for help in designing the figure.
kept in mind.
The revolution in our understanding of the role of References
symbionts has been made possible by the many advances 1 Schoonhoven, L.M. et al. (2005) Insect-Plant Biology, Oxford University
in molecular biology. These same advances have revealed Press

709
Review Trends in Ecology and Evolution December 2012, Vol. 27, No. 12

2 Barbosa, P. et al. (1991) Microbial Mediation of Plant-Herbivore 29 Roossinck, M.J. (2011) The good viruses: viral mutualistic symbioses.
Interactions, Wiley-Interscience Nat. Rev. Microbiol. 9, 99–108
3 Janson, E.M. et al. (2008) Phytophagous insect-microbe mutualisms 30 Vega, F.E. (2008) Insect pathology and fungal endophytes. J. Invertebr.
and adaptive evolutionary diversification. Evolution 62, 997–1012 Pathol. 98, 277–279
4 Gibson, C.M. and Hunter, M.S. (2010) Extraordinarily widespread and 31 Caspi-Fluger, A. et al. (2012) Horizontal transmission of the insect
fantastically complex: comparative biology of endosymbiotic bacterial symbiont Rickettsia is plant-mediated. Proc. R. Soc. Lond. B: Biol. Sci.
and fungal mutualists of insects. Ecol. Lett. 13, 223–234 279, 1791–1799
5 Feldhaar, H. (2011) Bacterial symbionts as mediators of ecologically 32 Grkovic, S. et al. (1995) Genes essential for amber disease in grass
important traits of insect hosts. Ecol. Entomol. 36, 533–543 grubs are located on the large plasmid found in Serratia entomophila
6 Oliver, K.M. et al. (2010) Facultative symbionts in aphids and the and Serratia proteamaculans. Appl. Environ. Microbiol. 61, 2218–2223
horizontal transfer of ecologically important traits. Annu. Rev. 33 Sacristan, S. and Garcia-Arenal, F. (2008) The evolution of virulence
Entomol. 55, 247–266 and pathogenicity in plant pathogen populations. Mol. Plant Pathol. 9,
7 Hosokawa, T. et al. (2007) Obligate symbiont involved in pest status of 369–384
host insect. Proc. R. Soc. Lond. B: Biol. Sci. 274, 1979–1984 34 Froissart, R. et al. (2010) The virulence-transmission trade-off in
8 Toju, H. and Fukatsu, T. (2011) Diversity and infection prevalence of vector-borne plant viruses: a review of (non-)existing studies. Philos.
endosymbionts in natural populations of the chestnut weevil: relevance Trans. R. Soc. B 365, 1907–1918
of local climate and host plants. Mol. Ecol. 20, 853–868 35 Stout, M.J. et al. (2006) Plant-mediated interactions between
9 McLean, A.H.C. et al. (2011) Effects of bacterial secondary symbionts pathogenic microorganisms and herbivorous arthropods. Annu. Rev.
on host plant use in pea aphids. Proc. R. Soc. Lond. B: Biol. Sci. 278, Entomol. 51, 663–689
760–766 36 Pieterse, C.M.J. et al. (2009) Networking by small-molecule hormones
10 Tsuchida, T. et al. (2011) Interspecific symbiont transfection confers a in plant immunity. Nat. Chem. Biol. 5, 308–316
novel ecological trait to the recipient insect. Biol. Lett. 7, 245–248 37 Ebbert, M.A. and Nault, L.R. (2001) Survival in Dalbulus leafhopper
11 Kessler, A. and Baldwin, I.T. (2002) Plant responses to insect vectors improves after exposure to maize stunting pathogens. Entomol.
herbivory: the emerging molecular analysis. Annu. Rev. Plant Biol. Exp. Appl. 100, 311–324
53, 299–328 38 Kluth, S. et al. (2002) Insects as vectors of plant pathogens: mutualistic
12 Dicke, M. and Baldwin, I.T. (2010) The evolutionary context for and antagonistic interactions. Oecologia 133, 193–199
herbivore-induced plant volatiles: beyond the ‘cry for help’. Trends 39 Fereres, A. and Moreno, A. (2009) Behavioural aspects influencing
Plant Sci. 15, 167–175 plant virus transmission by homopteran insects. Virus Res. 141, 158–
13 Pieterse, C.M.J. and Dicke, M. (2007) Plant interactions with microbes 168
and insects: from molecular mechanisms to ecology. Trends Plant Sci. 40 Jiu, M. et al. (2007) Vector-virus mutualism accelerates population
12, 564–569 increase of an invasive whitefly. PLoS ONE 2, e182
14 Pineda, A. et al. (2012) Rhizobacteria modify plant–aphid interactions: 41 Liu, J. et al. (2009) Differential indirect effects of two plant viruses on
a case of induced systemic susceptibility. Plant Biol. 14, 83–90 an invasive and an indigenous whitefly vector: implications for
15 Dicke, M. et al. (2009) Chemical complexity of volatiles from plants competitive displacement. Ann. Appl. Biol. 155, 439–448
induced by multiple attack. Nat. Chem. Biol. 5, 317–324 42 Zhang, T. et al. (2012) Begomovirus-whitefly mutualism is achieved
16 Paine, T.D. et al. (1997) Interactions among scolytid bark beetles, their through repression of plant defences by a virus pathogenicity factor.
associated fungi, and live host conifers. Annu. Rev. Entomol. 42, 179– Mol. Ecol. 21, 1294–1304
206 43 Hulcr, J. and Dunn, R.R. (2011) The sudden emergence of
17 Rivera, F.N. et al. (2009) Gut-associated yeast in bark beetles of the pathogenicity in insect-fungus symbioses threatens naive forest
genus Dendroctonus Erichson (Coleoptera: Curculionidae: Scolytinae). ecosystems. Proc. R. Soc. Lond. B: Biol. Sci. 278, 2866–2873
Biol. J. Linn. Soc. Lond. 98, 325–342 44 Lu, M. et al. (2010) Complex interactions among host pines and fungi
18 Clark, T.L. et al. (2001) Occurrence of Wolbachia in selected vectored by an invasive bark beetle. New Phytol. 187, 859–866
Diabroticite (Coleoptera: Chrysomelidae) beetles. Ann. Entomol. Soc. 45 Lu, M. et al. (2011) Do novel genotypes drive the success of an invasive
Am. 94, 877–885 bark beetle-fungus complex? Implications for potential reinvasion.
19 Barr, K.L. et al. (2010) Microbial symbionts in insects influence down- Ecology 92, 2013–2019
regulation of defense genes in maize. PLoS ONE 5, e11339 46 Gasparich, G.E. (2010) Spiroplasmas and phytoplasmas: microbes
20 Casteel, C.L. et al. (2012) Manipulation of plant defense responses by associated with plant hosts. Biologicals 38, 193–203
the tomato psyllid (Bactericerca cockerelli) and its associated 47 Novakova, E. et al. (2009) Arsenophonus, an emerging clade of
endosymbiont Candidatus Liberibacter psyllaurous. PLoS ONE 7, intracellular symbionts with a broad host distribution. BMC
e35191 Microbiol. 9, 143
21 Bonaventure, G. et al. (2011) Herbivore-associated elicitors: FAC 48 Bressan, A. et al. (2009) Vector transmission of a plant-pathogenic
signaling and metabolism. Trends Plant Sci. 16, 294–299 bacterium in the Arsenophonus clade sharing ecological traits with
22 Spiteller, D. et al. (2000) Gut bacteria may be involved in interactions facultative insect endosymbionts. Phytopathology 99, 1289–1296
between plants, herbivores and their predators: microbial biosynthesis 49 Ohgushi, T. (2005) Indirect interaction webs: herbivore-induced effects
of N-acylglutamine surfactants as elicitors of plant volatiles. Biol. through trait change in plants. Annu. Rev. Ecol. Evol. Syst. 36, 81–105
Chem. 381, 755–762 50 Utsumi, S. et al. (2010) Linkages among trait-mediated indirect effects: a
23 Giron, D. et al. (2007) Cytokinin-mediated leaf manipulation by a new framework for the indirect interaction web. Popul. Ecol. 52, 485–497
leafminer caterpillar. Biol. Lett. 3, 340–343 51 Omacini, M. et al. (2001) Symbiotic fungal endophytes control insect
24 Kaiser, W. et al. (2010) Plant green-island phenotype induced by leaf- host-parasite interaction webs. Nature 409, 78–81
miners is mediated by bacterial symbionts. Proc. R. Soc. Lond. B: Biol. 52 Pineda, A. et al. (2010) Helping plants to deal with insects: the role of
Sci. 277, 2311–2319 beneficial soil-borne microbes. Trends Plant Sci. 15, 507–514
25 Jaenike, J. (2012) Population genetics of beneficial heritable 53 Colgan, L.J. and Erbilgin, N. (2011) Tree-mediated interactions
symbionts. Trends Ecol. Evol. 27, 226–232 between the jack pine budworm and a mountain pine beetle fungal
26 Kikuchi, Y. et al. (2011) Specific developmental window for associate. Ecol. Entomol. 36, 425–434
establishment of an insect-microbe gut symbiosis. Appl. Environ. 54 Vandenheuvel, J.F.J.M. et al. (1994) Endosymbiotic bacteria
Microbiol. 77, 4075–4081 associated with circulative transmission of potato leafroll virus by
27 Russell, J.A. et al. (2003) Side-stepping secondary symbionts: Myzus persicae. J. Gen. Virol. 75, 2559–2565
widespread horizontal transfer across and beyond the Aphidoidea. 55 Morin, S. et al. (1999) A GroEL homologue from endosymbiotic bacteria
Mol. Ecol. 12, 1061–1075 of the whitefly Bemisia tabaci is implicated in the circulative
28 Darby, A.C. and Douglas, A.E. (2003) Elucidation of the transmission transmission of tomato yellow leaf curl virus. Virology 256, 75–84
patterns of an insect-borne bacterium. Appl. Environ. Microbiol. 69, 56 Gottlieb, Y. et al. (2010) The transmission efficiency of tomato yellow
4403–4407 leaf curl virus by the whitefly Bemisia tabaci is correlated with the

710
Review Trends in Ecology and Evolution December 2012, Vol. 27, No. 12

presence of a specific symbiotic bacterium species. J. Virol. 84, 9310– 67 Ferrari, J. and Vavre, F. (2011) Bacterial symbionts in insects or the
9317 story of communities affecting communities. Philos. Trans. R. Soc. B
57 Dicke, M. and Sabelis, M. (1992) Costs and benefits of 366, 1389–1400
chemical information conveyance: proximate and ultimate 68 Belliure, B. et al. (2008) Herbivore benefits from vectoring plant virus
factors. In Insect Chemical Ecology, an Evolutionary Approach through reduction of period of vulnerability to predation. Oecologia
(Roitberg, B.D. and Isman, M.B., eds), pp. 122–155, Chapman 156, 797–806
& Hall 69 Bextine, B. et al. (2005) Establishment of a genetically marked insect-
58 Godfray, H. (1994) Parasitoids: Behavioral and Evolutionary Ecology, derived symbiont in multiple host plants. Curr. Microbiol. 50, 1–7
Princeton University Press 70 Bextine, B. et al. (2004) Delivery of a genetically marked Alcaligenes sp
59 Dicke, M. (1996) The role of microorganisms in tri-trophic interactions to the glassy-winged sharpshooter for use in a paratransgenic control
in systems consisting of plants, herbivores, and carnivores. In strategy. Curr. Microbiol. 48, 327–331
Microbial Diversity in Time and Space (Colwell, R.R. et al., eds), pp. 71 Klepzig, K.D. and Six, D.L. (2004) Bark beetle-fungal symbiosis:
71–84, Plenum Press context dependency in complex associations. Symbiosis 37, 189–205
60 Madden, J.L. (1968) Behavioural responses of parasites to symbiotic 72 Scott, J.J. et al. (2008) Bacterial protection of beetle-fungus mutualism.
fungus associated with Sirex noctilio F. Nature 218, 189–190 Science 322, 63
61 Adams, A.S. and Six, D.L. (2008) Detection of host habitat by 73 Aanen, D.K. et al. (2009) Biological pest control in beetle agriculture.
parasitoids using cues associated with mycangial fungi of the Trends Microbiol. 17, 179–182
mountain pine beetle, Dendroctonus ponderosae. Can. Entomol. 140, 74 Kikuchi, Y. et al. (2012) Symbiont-mediated insecticide resistance.
124–127 Proc. Natl. Acad. Sci. U.S.A. 109, 8618–8622
62 Boone, C.K. et al. (2008) Parasitoids and dipteran predators exploit 75 Price, P.W. et al. (1980) Interactions among 3 trophic levels - Influence
volatiles from microbial symbionts to locate bark beetles. Environ. of plants on interactions between insect herbivores and natural
Entomol. 37, 150–161 enemies. Annu. Rev. Ecol. Syst. 11, 41–65
63 Oliver, K. et al. (2003) Facultative bacterial symbionts in aphids confer 76 Dicke, M. and van Loon, J.J.A. (2000) Multitrophic effects of herbivore-
resistance to parasitic wasps. Proc. Natl. Acad. Sci. U.S.A. 100, 1803– induced plant volatiles in an evolutionary context. Entomol. Exp. Appl.
1807 97, 237–249
64 Scarborough, C. et al. (2005) Aphid protected from pathogen by 77 Van der Putten, W.H. et al. (2001) Linking above- and belowground
endosymbiont. Science 310, 1781 multitrophic interactions of plants, herbivores, pathogens, and their
65 Tsuchida, T. et al. (2010) Symbiotic bacterium modifies aphid body antagonists. Trends Ecol. Evol. 16, 547–554
color. Science 330, 1102–1104 78 van Veen, F.J.F. et al. (2006) Apparent competition, quantitative food
66 Clark, E.L. et al. (2010) Insect endosymbionts: manipulators of insect webs, and the structure of phytophagous insect communities. Annu.
herbivore trophic interactions? Protoplasma 244, 25–51 Rev. Entomol. 51, 187–208

711
View publication stats