Evolution of Cooperation in
Ambrosia Beetles
Inauguraldissertation
der Philosophisch-naturwissenschaftlichen Fakultät
der Universität Bern
vorgelegt von
Peter Hans Wilhelm Biedermann
von Trofaiach / Österreich
Leiter der Arbeit:
Prof. Dr. Michael Taborsky
Institut für Ökologie und Evolution
Abteilung Verhaltensökologie
Universität Bern
Evolution of Cooperation in
Ambrosia Beetles
Inauguraldissertation
der Philosophisch-naturwissenschaftlichen Fakultät
der Universität Bern
vorgelegt von
Peter Hans Wilhelm Biedermann
von Trofaiach / Österreich
Leiter der Arbeit:
Prof. Dr. Michael Taborsky
Institut für Ökologie und Evolution
Abteilung Verhaltensökologie
Universität Bern
Von der Philosophisch-naturwissenschaftlichen Fakultät angenommen.
Bern, 20. März 2012
Der Dekan:
Prof. Dr. Silvio Decurtins
Supervised by:
Prof. Dr. Michael Taborsky
Department of Behavioural Ecology
Institute of Ecology and Evolution
University of Bern
Wohlenstrasse 50a
CH-3032 Hinterkappelen
Switzerland
Reviewed by:
Prof. Dr. Jacobus J. Boomsma
Section for Ecology and Evolution
Institute of Biology
University of Copenhagen
Universitetsparken 15
2100 Copenhagen
Denmark
Examined by:
Prof. Dr. Heinz Richner, University of Bern (Chair)
Prof. Dr. Michael Taborsky, University of Bern
Prof. Dr. Jacobus J. Boosma, University of Copenhagen
Copyright
Chapter 1 © PNAS 2011 by the National Academy of Sciences of the United States of America,
Washington, USA
Chapter 2 © Mitt. Dtsch. Ges. allg. angew. Ent. 2011 by the DGaaE, Müncheberg, Gernany
Chapter 4 © Zookeys 2010 by Pensoft Publishers, Sofia, Bulgaria
Chapter 5 © Behav. Ecol. & Sociobiol. by Springer-Verlag GmbH, Heidelberg, Germany
Chapter 9 © J. Bacteriol. by the American Society for Microbiology, Washington, USA
General Introduction, Chapter 3, 6, 7, 8, Appendix 1,2, and Summary & Conclusion © Peter H.W.
Biedermann
Cover drawing © by Barrett Anthony Klein, Entomoartist, Department of Biology, University of
Konstanz, Germany. http://www.pupating.org
Layout by Peter H. W. Biedermann
Printed in Bern, Switzerland by Kopierzentrale der Universität Bern
Für meine Eltern
die meine Begeisterung für die Natur erkannt haben
Für Tabea
die mich bedingungslos unterstützt
Und Helene Francke-Grosmann, Karl E. Schedl, Dale M. Norris und Richard A. Roeper
für ihre Pionierarbeiten auf dem Gebiet der Ambrosiakäfer-Forschung
“There is grandeur in this view of life, with its several powers, having been originally breathed into a
few forms or into one; and that, whilst this planet has gone cycling on according to the fixed law of
gravity, from so simple a beginning endless forms most beautiful and most wonderful have been, and
are being, evolved.”
C. Darwin, Origin of Species 1859
“Ich habe ihm nun geraten, künftig in der Natur nie einen einzelnen Gegenstand alleine
herauszuzeichnen, nie einen einzelnen Baum, einen einzelnen Steinhaufen, eine einzelne Hütte,
sondern immer zugleich einigen Hintergrund und einige Umgebung mit. Und zwar aus folgenden
Ursachen: Wir sehen in der Natur nie etwas als Einzelheit, sondern wir sehen alles in Verbindung mit
etwas anderem, das vor ihm, neben ihm, hinter ihm, unter ihm und über ihm sich befindet. Auch fällt
uns wohl ein einzelner Gegenstand als besonders schön und malerisch auf; es ist aber nicht der
Gegenstand allein, der diese Wirkung hervorbringt, sondern es ist die Verbindung, in der wir ihn
sehen.“
Johann Wolfgang von Goethe, zu Eckermann, 5. Juni 1826
Contents
3
General Introduction
11
Chapter 1
Biedermann PHW and M Taborsky (2011): Larval helpers and age polyethism in ambrosia
beetles. Proceedings of the National Academy of Science, USA 108(41): 17064-17069.
27
Chapter 2
Biedermann PHW, K Peer and M Taborsky (2011)
Female dispersal and reproduction in the ambrosia beetle Xyleborinus saxesenii Ratzeburg
(Coleoptera; Scolytinae). Mitteilungen der deutschen Gesellschaft für allgemeine und
angewandte Entomologie 18: in press.
33
Chapter 3
Biedermann PHW and M Taborsky (manuscript in work)
Responses to artificial selection on dispersal in a primitively eusocial beetle.
53
Chapter 4
Biedermann PHW (2010)
Observations on sex ratio and behavior of males in Xyleborinus saxesenii Ratzeburg
(Scolytinae, Coleoptera). In: Cognato AI, Knížek M (Eds) Sixty years of discovering scolytine
and platypodine diversity: A tribute to Stephen L. Wood. Zookeys 56: 253-267.
69
Chapter 5
Biedermann PHW, KD Klepzig and M Taborsky (2011)
Costs of delayed dispersal and alloparental care in the fungus-cultivating ambrosia beetle
Xyleborus affinis Eichhoff (Scolytinae: Curculionidae). Behavioral Ecology and Sociobiology
65:1753–1761.
79
Chapter 6
Biedermann PHW and M Taborsky (manuscript in work)
Social fungus farming varies among ambrosia beetles.
101
Chapter 7
Biedermann PHW, KD Klepzig, M Taborsky and DL Six (manuscript in work)
Dynamics of filamentous fungi in the complex ambrosia gardens of the primitively eusocial
beetle Xyleborinus saxesenii Ratzeburg (Scolytinae; Curculionidae).
123
Chapter 8
De Fine Licht HH and PHW Biedermann (in review)
Patterns of functional enzyme activity show that larvae are the key to successful fungus
farming by ambrosia beetles. Frontiers in Zoology, submitted
143
Chapter 9
Grubbs KJ, Biedermann PHW, Suen G, Adams SM, Moeller JA, Klassen JL, Goodwinm LA,
Woyke T, Munk AC, Bruce D, Detter C, Tapia R, Han CS & CR Currie (2011)
The complete genome sequence of Streptomyces cf. griseus (XyelbKG-1 1), an ambrosia
beetle-associated actinomycete. Journal of Bacteriology 193(11): 2890-91.
145
Appendix 1
Biedermann PHW, M Taborsky M and DL Six (manuscript in work)
Fungal associates and their effect on the behaviours and success of the ambrosia beetle
Xyleborus affinis Eichhoff (Scolytinae: Curculionidae).
167
Appendix 2
Biedermann PHW, M Taborsky and CR Currie (manuscript in work)
Mechanisms of fungus gardening in ambrosia beetles.
175
Summary & Conclusion
183
Acknowledgements
185
Contributions & Funding
189
Curriculum Vitae
© 2000 G. Larson, The Far Side
Curriculum Vitae
General Introduction
“Again as in the case of corporeal structure, and conformably with my theory, the instinct for each
species is good for itself, but has never, as far as we can judge, been produced for the exclusive good
of others” C. Darwin, Origin of Species 1859.
Since I got familiar with Darwin’s concept of the survival of the fittest, I am very much fascinated by
the question how a behaviour can persist in nature that obviously reduces the direct success (fitness)
of an actor, but instead greatly benefits others. We are surrounded by such examples of the success
of cooperation: packs of cooperatively hunting carnivores, helpers at bird nests that support the
breeding pair in protection and food provisioning, the great success of social insects, where some
individuals even refrain from reproduction, and the incredible diverse examples of symbioses, from
gut microbes that help humans to digest their food, to pollination of flowers by bees and the
fascinating fungus garden that provide nutrients for their insects farmers for getting tended, weeded
and provisioned with new substrate. Perhaps less obvious, cooperation can not only be found among
selfish biological entities, but is also the foundation of the major evolutionary transitions to stages of
higher complexity. The evolution of chromosomes from independent genes, of multicellular
organisms from individual cells and of societies from individuals involves a transition such that
“entities that were capable of independent replication before the transition can replicate only as part
of a larger whole after it” (Maynard-Smith and Szathmáry 1995).
The transition from egoistic individuals to symbioses and to animal societies was already noticed by
Darwin as a challenge for his theory of natural selection (Darwin 1859), and has puzzled generations
of evolutionary biologists ever since. After the concept of cooperation among animals as “good for
the species” has been rejected, W.D. Hamilton’s inclusive fitness theory (also known as kin selection
theory; Hamilton 1964) has prevailed as the currently best and most widely accepted theory for
understanding the evolution of cooperation. Even though it is perhaps difficult to accept on first
sight, there is tremendous evidence that selection acts on the gene level of individuals, meaning that
in evolutionary times organisms persist that do not behave for the benefit of themselves, but instead
for the highest possible success of the genes they carry. Thus, individuals are expected to maximize
the success of their own genes (direct fitness) plus the success of the same genes in other (related)
organisms (indirect fitness).
This thesis illustrates the power and validity of Hamilton’s inclusive fitness theory by critically testing
its predictions with the social life of ambrosia beetles and their symbioses with fungi. My thesis is a
first step for explaining the evolution of sociality and fungus farming in a lineage of beetles, which
ancestors are solitary living and feeding on plants. On the next few pages, I will briefly introduce the
inclusive fitness concept in more detail and present ecological and genetic factors that promote the
evolution of cooperation in nature.
1
General Introduction
Hamilton’s inclusive fitness theory
„Die Theorie bestimmt was wir beobachten können.“ Albert Einstein.
Modern evolutionary theory on cooperation1 originated with Hamilton’s inclusive fitness theory and
his ground-breaking article on the genetic evolution of social behaviour (Hamilton 1963; 1964). This
led biologists to realize that altruistic behaviour is based on fundamentally selfish interests, often
through helping relatives that possess the same genes, and are never done for the good of the
species. (Darwin 1859). This can be also formulated by Hamilton’s rule (Hamilton 1964):
C<r×B
Where C is the fitness cost to the actor, r is the genetic relatedness between the actor and the
recipient, and B is the fitness benefit to the recipient.
This formula shows that reproductive altruism can evolve even if it is associated with sterility of the
actor, given costs of own reproduction are below the indirect fitness gains via helping a close
relative. Hence, if individuals are clones (r = 1), reproductive altruism can evolve simply if it is more
efficient than personal reproduction. Hence, theoretically, totipotent clonal individuals are
indifferent about who reproduces (Frank 1995). That means, on the other hand, that the higher the
relatedness, the lower the threat of cheating (i.e. one partner is not reciprocating the benefit it gets
in the relationship) to evolve (e.g., Hamilton 1978; Bourke 2011). The general assumption that
relatedness facilitates altruism and mitigates selfishness is only challenged under limited dispersal,
when local competition between relatives may fully outweigh the benefits of altruism (Hamilton
1975; West et al. 2002).
Hamilton’s famous theory on kin selection can only be partly applied to the evolution of interspecific
mutualism (symbiosis; i.e. when individuals of different species cooperate with each other), however
(Foster and Wenseleers 2006). In groups of unrelated individuals, mutually beneficial acts, but not
altruism (i.e. a behaviour with net costs for an actor), can evolve under the assumptions of inclusive
fitness, because relatedness is needed (r > 0) for altruism to be selected (Hamilton 1964; 1972;
Bourke 2011). Therefore, reproductive division of labour (= altruism) can only evolve within societies
of related individuals and not in interspecific mutualism, whereas mutually beneficial acts are
ubiquitously found in both intra- and interspecific cooperation. Hence, it is important to note that
interspecific relationships will only evolve to be mutualistic if the interests of both partners match
somehow (e.g. Herre et al. 1999; Foster and Wenseleers 2006). This might be due to a feedbackbenefit between the partner’s successes: e.g. if a fungus farmed by insects will produce more fruiting
bodies to feed more insects (increasing the fitness of the insects), it will be later dispersed by more
insects (increasing its own fitness). Additional to relatedness and aligned fitness interests, there are
1
Here I define a behaviour as cooperative if social partners potentially benefit from its performance,
independently of whether this behaviour entails net costs to the actor (Brosnan and de Waal, 2002). This
definition includes (i) mutualistic behaviours that are regarded as selfish acts generating benefits to other
individuals (common goods) as a by-product , and (ii) altruistic behaviours that bring about net costs to the
actor, which are compensated through indirect fitness benefits via kin-selection; they will thus only evolve in
groups of relatives (Hamilton, 1964). In the course of social evolution and task specialization shaped by kin
selection, mutualistic behaviours may lose their original function and change into truly altruistic behaviors (Lin
and Michener, 1972).
2
General Introduction
also ecological and other genetic factors that facilitate inclusive fitness gains for an organism to be
higher when joining a group than breeding independently, which will be outlined in the following.
Ecological factors in the evolution of cooperation
Ecological factors influence the survival and fecundity of solitary individuals and ones living in
differently-sized groups, the costs of dispersal, and the costs and benefits of division of labour (Emlen
1982; Vehrencamp 1983a,b). Certain ecological factors have been repeatedly found to facilitate the
evolution of sociality (Korb and Heinze 2008). The first set of factors predisposes for social life
because of constraints on solitary breeding (ecological constraint hypothesis; Emlen 1982)): these are
(i) limitation of high quality nesting sites, (ii) high mortality during dispersal and difficult nest
foundation (e.g. harsh environments), and (iii) demographic factors like minimal group size or
population density. The other set of factors facilitates social evolution due to inclusive fitness
benefits of philopatry (benefits of philopatry hypothesis (Stacey and Ligon 1991): these are (iv)
alloparental brood care, (v) a better common defense under high parasite or predation pressure, (vi)
direct fitness through nest inheritance, and (vii) direct and indirect fitness through monopolization of
a long-lasting, non-diminishable (bohnanza-like) food source. Ecological factors that have been
shown to constrain social evolution (despite other traits that would favour it) are (a) a short breeding
season that may allow only one cycle of solitary breeding but no overlapping generations (e.g.
facultative eusociality in halictid bees depending on altitude and latitude; Eickwort et al. 1996; Field
et al. 2010), and (b) long life spans of individuals relative to duration of the nest, which selects for
maintenance of behavioural flexibility (phenotypic plasticity) because inclusive fitness benefits from
alternative social strategies should change substantially over time – thus totipotency of helpers
should be maintained (e.g. cooperative breeding birds, mammals, lower termites and some ambrosia
beetles (Alexander et al. 1991; Choe and Crespi 1997; Korb and Hartfelder 2008, this thesis).
In contrast to the evolution of sociality, there has not been much effort to establish ecological factors
that facilitate the evolution of interspecific mutualism (Bourke 2011). It has been noted, however,
that (i) mutualisms more commonly evolve in stable than fluctuating environments (e.g. tropical vs.
temperate environments), probably because perturbations can hinder the establishment and
maintenance of a stable relationship (May 1976), and (b) life-history associated with a sessile habit
apparently predisposes species to evolve interspecific mutualisms (Osman and Haugsness 1981;
Janzen 1985; Bourke 2011). Three underlying reasons have been proposed: First, an organism with
limited dispersal abilities can greatly profit from an association with a partner that exchanges
mobility (e.g., evolution of pollen and seed dispersal by animals). Second, sessile organisms often
profit from mutualisms that help them acquiring nutrients since they cannot forage actively (e.g.,
Sachs et al. 2004). Third, being sessile assures partner fidelity and facilitates evolution of mutualism
(Nowak and May 1992; Bourke 2011).
Genetic factors in the evolution of cooperation
Sub-sociality, monogamy, haplodiploidy and inbreeding have been proposed to predispose for
evolution of cooperation – why will be outlined in the following. High relatedness is crucial for the
evolution of altruism, but it is not essential for, although facilitates, the evolution of cooperation (see
above). Therefore, it must have played a particularly important role in the transition from
3
General Introduction
cooperative breeding (primitive / facultative eusociality; Michener 1974; Crespi and Yanega 1995)) to
obligate eusociality. Cooperative breeders and eusocial societies are both characterized by
overlapping parental and offspring generations and alloparental brood care, but they differ in
whether the reproductive altruism of some individuals is reversible or permanent (Batra 1966;
Wilson 1971). Obligatory eusociality has arisen independently in at least 24, mostly invertebrate,
lineages, whereas cooperative breeding evolved many more times in vertebrates (Ligon and Burt
2004; Crozier 2008; Bourke 2011). Current data suggests that permanently sterile eusocial castes
only evolved via (i) sub-social colony foundation and (ii) life-time monogamy (Hughes et al. 2008;
Boomsma 2009). First, sub-social group formation by a mother and her offspring facilitates the
evolution of altruism, because of high within-group relatedness compared to the alternative semisocial group formation of non-siblings, where relatedness is usually low or insecure. Second, if nest
foundation is by a monogamous mother, relatedness between daughters and daughter’s own
offspring is equal (average r = 0.5; also in haplodiploid systems with unbiased sex-ratios), which
implies that any constraint on individual reproduction can favor the evolution of staying, helping and
finally sterility (Boomsma 2009). Despite that, multiple queen-mating can be found in many
obligatory eusocial clades, but this has been shown to be a derived trait, which evolved after the
transition to obligate worker sterility (Hughes et al. 2008; Boomsma 2009).
In haplodiploid species fertilized, diploid eggs develop into females, whereas unfertilized, haploid
eggs develop into males. This leads to relatedness asymmetries within subsocial groups of
monogamous mothers, which has been proposed to promote social evolution by kin-selection
(haplodiploidy hypothesis; (Hamilton 1964): Males pass all their genes to their daughters (r = 1). Full
sisters share more genes with each other (r = 0.75), then they would share with their own offspring (r
= 0.50). Therefore, if there are opportunities to help relatives producing more offspring, females
should prefer to raise sisters compared to selfish reproduction (Hamilton 1964). However, this is only
the case under female biased sex-ratios, because under non-biased sex-ratios higher relatedness
with sisters (r = 0.75) would be offset by lower relatedness with brothers (r = 0.25) (Trivers and Hare
1976). Female-biased colony sex-ratios compared to population sex-ratio may arise through partial
bivoltinism (two overlapping generations per year) and split sex ratios (females contribute different
sex-ratios to the same offspring generation) (Grafen 1986; Pamilo 1991; Bourke and Franks 1995).
Inbreeding in haplodiploids would not change relatedness asymmetries because relatedness
between sisters and to offspring increases to the same degree (Craig 1982). Hence, although
inbreeding favours sociality by kin selection as long as local competition between relatives does not
outweigh the benefits of altruism (Hamilton 1975; West et al. 2002), inbreeding does not change the
proposed role of haplodiploidy (i.e. under female biased colony sex-ratios) for social evolution.
Finally, asymmetries in relatedness caused by haplodiploidy that normally facilitate the evolution of
female biased helping behaviour (Hamilton 1964; Alexander 1974) are reduced by inbreeding, which
increases the likelihood of male helpers to evolve (e.g. thrips; Hamilton 1972; Chapman et al. 2000).
In contrast to intraspecific cooperation the absence of indirect fitness gains (kin selection) in
interspecific interactions led to the currently accepted view that mutualisms are reciprocal
exploitations that nonetheless provide net benefits to each partner (e.g. Maynard-Smith and
Szathmáry 1995; Herre et al. 1999). The interests of interacting partners are hardly ever the same,
creating potential for conflicts of interest that shape or destabilize associations. Foster and
Wenseleers ( 2006) identified three key factors for the evolution of mutualism in their theoretical
model: (1) high benefit to cost ratio, (2) high within-species relatedness (within a species feedbacks
benefit those that caused them or close relatives) and (3) high between species fidelity (species stay
4
General Introduction
together long enough for feedbacks to benefit those that caused them). Positive relatedness is
essential for all feedback benefits between the partners. Increasing species number on one side (or
guild) has the same effect as reducing within species relatedness. An individual will only provide aid
to another species, if the individual itself or other positively related individuals receive the feedback
aid (Foster and Wenseleers 2006).
A brief introduction to insect sociality and fungus agriculture
“A well-flavoured vegetable is cooked, and the individual is destroyed; but the horticulturist sows
seeds of the same stock, and confidently expects to get nearly the same variety…. Thus I believe it has
been with social insects: a slight modification of structure, or instinct, correlated with the sterile
condition of certain members of the community: consequently the fertile males and females of the
same community flourished, and transmitted to their fertile offspring a tendency to produce sterile
members having the same modifications.” C. Darwin, Origin of Species 1859.
Over the last decades hymenoptera emerged as the main model system for studying social evolution,
because the order consists of species in varying states of sociality, from solitary breeding to
obligatory eusociality. It is now clear that genetic (e.g. haplodiploidy, inbreeding), ecological
constraints on independent reproduction and benefits of philopatry can predispose individuals to be
selected for higher sociality. However, despite a decade of research, there is still discordance of the
general importance of genetic factors, like haplodiploidy, relative to ecological factors (e.g. Korb and
Heinz 2008). Ecological factors were often underestimated or not studied in insects, while genetic
relatedness was sometimes overestimated (e.g. Korb and Lenz 2004).
Insect agriculture is one of the best examples for the success of cooperation in evolution (Farrell et
al. 2001; Mueller and Gerardo 2002; Mueller et al. 2005). In total, there are about 3500 species of
beetles (the weevil subfamilies Scolytinae and Platypodinae; Farrell et al. 2001), ants (220 described
species in the subfamily Myrmicinae; e.g. Weber 1972) and termites (about 330 species; e.g. Aanen
et al. 2002) known to grow, tend and harvest fungi, which apparently span the whole range of interand intraspecific cooperative relationships (e.g. Wilson 1971; Kent and Simpson 1992; Kirkendall et
al. 1997; Hölldobler and Wilson 2010). All three insect farming lineages have evolved the highest
levels of intra- and interspecific cooperation, namely eusociality and mutualism with fungi. Advanced
fungiculture appears not possible without intraspecific cooperation (Mueller et al. 2005). Obligatory
eusociality is the rule in the attine ants and the macrotermitine termites, and has been found in one
species of ambrosia beetles up to now (Austroplatypus incompertus; Kent and Simpson 1992).
Division of labor sets the stage for the origin of fungiculture in fungus-growing ants and termites
(Mueller et al. 2005). Ambrosia beetles originate from solitary or colonial ancestors, and their fungus
agriculture may have coevolved with sociality (Kirkendall et al. 1997; Mueller et al. 2005; Peer and
Taborsky 2007). They live inside trees, a habitat which extraordinarily favors the evolution of
inbreeding, haplodiploidy and sociality of arthropods (Hamilton 1967; 1978). The reason is that a tree
often offers a huge and safe food-resource for the tiny exploiters that they can settle and defend for
multiple generations. Furthermore, wood is typically low on nutrients (Hunt and Nalepa 1994) and
high on poisonous secondary metabolites, including alkaloids, terpenes, and phenols (Highley and
5
General Introduction
Kirk 1979), which has repeatedly selected for symbioses with decomposing and detoxifying
microorganisms (e.g. ambrosia beetles, termites; Hunt and Nalepa 1994; Choe and Crespi 1997;
Grunwald et al. 2010). Living within wood has apparently fostered at least eight independent origins
of fungiculture in ambrosia beetles (Farrell et al. 2001; Hulcr et al. 2007). Hence, they represent a
unique model system to study the evolution of sociality in relation to fungiculture. Interestingly,
ambrosia beetles vary in their mating system (inbreeding vs. outbreeding species) and ploidy level
(haplodiploid vs. diploid species), which are factors that have been assumed to contribute to social
evolution, although their respective roles in social evolution are controversial (see above).
Three types of methods have been suggested for the analysis of ultimate evolutionary questions for
the study of adaptation (modified from Crespi and Choe (1997)): (i) functional design analyses
encompass descriptive, correlative data collection and manipulative experiments (Williams 1966), (ii)
measurement of selection involves quantification of phenotypes and fitness components (Lande and
Arnold 1983) and (iii) comparative analysis and phylogenetics. Phylogenetically based tests allow
inference of evolutionary trajectories, and thus may help disentangling causes from effect. Among
ambrosia beetle species or clades, effects of inbreeding, haplodiploidy or other ecological variables
may be analyzed, for example, through comparisons of sister taxa that differ in their social system.
Before such comparative studies can be conducted, however, one needs to establish some basic
knowledge on the beetle-fungus interactions, as well as the social system. Four years ago, little was
known about the cooperative behaviours within ambrosia beetle nests and whether there is any
fungus gardening behaviour of the beetles. Associated fungi had been identified in only a few species
of ambrosia beetles. Studies about the role fungal associates play in the nutrition of ambrosia beetles
had been missing, and it had not been investigated whether ambrosia beetles can selectively favour
or inhibit the growth of specific fungi. However, this knowledge is crucial for the understanding of
cooperation inside the beetle galleries. Observations or indications rather than rigorous or scientific
analysis had dominated many articles on the social and farming behaviour of ambrosia beetles
published in the last 150 years – this is not surprising because it is almost impossible to study these
beetles within their natural habitat. Laboratory observation techniques have allowed me to observe
the beetles in detail, and study functions of social behaviours, as well as roles of fungi by performing
correlational studies and manipulative experiments. Finally, two-year consecutive laboratory rearing
enabled artificial selection on dispersal (a correlate of sociality) and its associated traits.
Ambrosia beetles are not only a great model system for evolutionary questions (Crespi and Choe
1997), but invasive fungus-associated beetles pose also a huge threat on naive forests that are
already stressed by climate change (Hulcr and Dunn 2011; Six et al. 2011; Choi 2011).
Literature
Aanen DK, Eggleton P, Rouland-Lefévre C, Guldberg-Frøslev T, Boomsma JJ & Rosendahl S (2002) The evolution
of fungus-growing termites and their mutualistic fungal symbionts. PNAS 99: 14887-14892.
Alexander RD (1974) The Evolution of Social Behavior. Annual Review of Ecology and Systematics 5: 325-383.
Alexander RD, Noonan KM & Crespi BJ (1991) The evolution of eusociality. The Biology of the Naked Mole Rat
(Sherman PW, Jarvis JUM & Alexander RD, eds), pp. 1-44. Princeton University Press, Princeton.
Batra SWT (1966) Nests and social behaviour of halictine bees of India. Indian J Enthomology 28: 375-393.
6
General Introduction
Boomsma JJ (2009) Lifetime monogamy and the evolution of eusociality. Philosophical Transactions of the
Royal Society B-Biological Sciences 364: 3191-3207.
Bourke AFG (2011) Principles of Social Evolution. Oxford University Press, Oxford.
Bourke AFG & Franks NR (1995) Social Evolution in Ants. Princeton University Press, Princeton.
Brosnan SF & de Waal FBM (2002) A proximate perspective on reciprocal altruism. Human Nature-An
Interdisciplinary Biosocial Perspective 13: 129-152.
Chapman TW, Crespi BJ, Kranz BD & Schwarz MP (2000) High relatedness and inbreeding at the origin of
eusociality in gall-inducing thrips. PNAS 97: 1648-1650.
Choe JC & Crespi BJ (1997) The Evolution of Social Behaviour in Insects and Arachnids. Cambridge University
Press, Cambridge UK.
Craig R (1982) Evolution of Eusociality by Kin Selection - the Effect of Inbreeding Between Siblings. J. Theor. Biol.
94: 119-128.
Crespi BJ & Choe JC (1997) Explanation and evolution of social systems. The Evolution of Social Behavior in
Insects and Arachnids (Choe JC & Crespi BJ, eds), Cambridge University Press, Cambridge.
Crespi BJ & Yanega D (1995) The definition of eusociality. Behav. Ecol. 6: 109-115.
Crozier RH (2008) Advanced eusociality, kin selection and male haploidy. Australian Journal of Entomology 47:
2-8.
Darwin C (1859) The origin of species. Gramercy Books, New York.
Eickwort GC, Eickwort JM, Gordon J & Eickwort MA (1996) Solitary behavior in a high altitude population of the
social sweat bee Halictus rubicundus (Hymenoptera: Halictidae). Behav. Ecol. Sociobiol. 38: 227-233.
Emlen ST (1982) The evolution of helping. I. An ecological constraints model. Am. Nat. 119: 29-39.
Farrell BD, Sequeira AS, O'Meara BC, Normark BB, Chung JH & Jordal BH (2001) The evolution of agriculture in
beetles (Curculionidae : Scolytinae and Platypodinae). Evolution 55: 2011-2027.
Field J, Paxton RJ, Soro A & Bridge C (2010) Cryptic Plasticity Underlies a Major Evolutionary Transition. Current
Biology 20: 2028-2031.
Foster KR & Wenseleers T (2006) A general model for the evolution of mutualisms. J. Evol. Biol. 19: 1283-1293.
Frank SA (1995) Mutual policing and repression of competition in the evolution of cooperative groups. Nature
377: 520-522.
Grafen A (1986) Split sex ratios and the evolutionary origins of eusociality. J. Theor. Biol. 122: 95-121.
Grunwald S, Pilhofer M & Holl W (2010) Microbial associations in gut systems of wood- and bark-inhabiting
longhorned beetles [Coleoptera: Cerambycidae]. Systematic and Applied Microbiology 33: 25-34.
Hamilton WD (1963) Evolution of Altruistic Behavior. Am. Nat. 97: 354-&.
Hamilton WD (1964) The genetical evolution of social behaviour. I+II. J. Theor. Biol. 7: 1-52.
Hamilton WD (1967) Extraordinary sex ratios. Science 156: 477-488.
Hamilton WD (1972) Altruism and related phenomena, mainly in social insects. Ann. Rev. Ecol. Syst. 3: 193-232.
7
General Introduction
Hamilton WD (1975) Innate social aptitudes of man: an approach from evolutionary genetics. Biosocial
Anthropology (Fox R, ed), pp. 133-155. John Wiley and Sons.
Hamilton WD (1978) Evolution and diversity under bark. Diversity of insect faunas (Mound LA & Waloff N, eds),
pp. 154-175. Blackwell, Oxford.
Herre EA, Knowlton N, Mueller UG & Rehner SA (1999) The evolution of mutualisms: exploring the paths
between conflict and cooperation. Trends Ecol. Evol. 14: 49-53.
Highley TL & Kirk TK (1979) Mechanisms of Wood Decay and the Unique Features of Heart-Rots.
Phytopathology 69: 1151-1157.
Hölldobler B & Wilson EO (2010) Der Superorganismus. Springer-Verlag, Berlin, Heidelberg.
Hughes WOH, Oldroyd BP, Beekman M & Ratnieks FLW (2008) Ancestral monogamy shows kin selection is key
to the evolution of eusociality. Science 320: 1213-1216.
Hulcr J, Kolarik M & Kirkendall LR (2007) A new record of fungus-beetle symbiosis in Scolytodes bark beetles
(Scolytinae, Curculionidae, Coleoptera). Symbiosis 43: 151-159.
Hunt JH & Nalepa CA (1994) Nourishment and Evolution in Insect societies. Westview Press, Boulder.
Janzen DH (1985) The natural history of mutualisms. The biology of mutualism: ecology and evolution. (Boucher
DH, ed), pp. 40-99. Oxford University Press, New York.
Kent DS & Simpson JA (1992) Eusociality in the beetle Austroplatypus incompertus (Coleoptera: Platypodidae).
Naturwissenschaften 79: 86-87.
Kirkendall LR, Kent DS & Raffa KF (1997) Interactions among males, females and offspring in bark and ambrosia
beetles: the significance of living in tunnels for the evolution of social behavior. The Evolution of Social Behavior
in Insects and Arachnids (Choe JC & Crespi BJ, eds), pp. 181-215. Cambridge University Press.
Korb J & Hartfelder K (2008) Life history and development - a framework for understanding developmental
plasticity in lower termites. Biological Reviews 83: 295-313.
Korb J & Heinze J (2008) The ecology of social life: a synthesis. Ecology of social evolution. (Korb J & Heinze J,
eds), pp. 245-259. Springer, Berlin Heidelberg.
Korb J & Lenz M (2004) Reproductive decision-making in the termite Cryptotermes secundus (Kalotermitidae)
under variable food conditions. Behav. Ecol. 15: 390-395.
Lande R & Arnold SJ (1983) The Measurement of Selection on Correlated Characters. Evolution 37: 1210-1226.
Ligon DJ & Burt DB (2004) Evolutionary origins. Ecology and Evolution of Cooperative Breeding in Birds (Koenig
WD, ed), pp. 5-34. Cambridge University Press, Cambridge.
Lin N & Michener CD (1972) Evolution of sociality in insects. Q. Rev. Biol. 47: 131-159.
May RM (1976) Models for two interacting populations. Theoretical ecology: principles and applications. (May
RM, ed), pp. 49-70. Blackwell Scientific Publications, Oxford.
Maynard-Smith J & Szathmáry E (1995) The major transitions in evolution. Oxford University Press, Oxford.
Michener CD (1974) The Social Behavior of Bees. Harvard University Press, Cambridge.
Mueller UG & Gerardo N (2002) Fungus-farming insects: Multiple origins and diverse evolutionary histories.
Proceedings of the National Academy of Sciences of the United States of America 99: 15247-15249.
8
General Introduction
Mueller UG, Gerardo NM, Aanen DK, Six DL & Schultz TR (2005) The evolution of agriculture in insects. Annual
Review of Ecology Evolution and Systematics 36: 563-595.
Nowak MA & May RM (1992) Evolutionary Games and Spatial Chaos. Nature 359: 826-829.
Osman RW & Haugsness JA (1981) Mutualism Among Sessile Invertebrates - A Mediator of Competition and
Predation. Science 211: 846-848.
Pamilo P (1991) Evolution of the Sterile Caste. J. Theor. Biol. 149: 75-95.
Peer K & Taborsky M (2007) Delayed dispersal as a potential route to cooperative breeding in ambrosia beetles.
Behav. Ecol. Sociobiol. 61: 729-739.
Sachs JL, Mueller UG, Wilcox TP & Bull JJ (2004) The evolution of cooperation. Q. Rev. Biol. 79: 135-160.
Stacey PB & Ligon JD (1991) The benefits-of-philopatry hypothesis for the evolutioni of cooperative breeding:
variation in territory quality and group size effects. Am. Nat. 117: 831-846.
Trivers RL & Hare H (1976) Haplodiploidy and the evolution of the social insects. Science 191: 263.
Vehrencamp SL (1983a) A model for the evolution of despotic versus egalitarian societies. Anim. Behav. 31:
667-682.
Vehrencamp SL (1983b) Optimal degree of skew in cooperative societies. Am. Zool. 23: 327-335.
Weber NA (1972) Gardening ants: The attines. American Philosophical Society, Philadelphia.
West SA, Pen I & Griffin AS (2002) Conflict and cooperation - Cooperation and competition between relatives.
Science 296: 72-75.
Williams GC (1966) Adaptation and natural selection. Princeton University Press, Princeton, New Jersey.
Wilson EO (1971) The insect societies. Belknap Press of Harvard University Press, Cambridge.
9
General Introduction
10
Chapter 1
Larval helpers and age polyethism in ambrosia beetles
Peter H. W. Biedermann1 and Michael Taborsky
Department of Behavioral Ecology, Institute of Ecology and Evolution, University of Bern, CH-3012 Bern, Switzerland
Edited by Bert Hölldobler, Arizona State University, Tempe, AZ, and approved September 1, 2011 (received for review May 14, 2011)
which is a habitat extraordinarily favoring social evolution (12),
apparently having fostered at least seven independent origins of
fungiculture in beetles (13). Hence, they represent a unique model
system to study the evolution of sociality in relation to
fungiculture. Interestingly, ambrosia beetles vary in their mating
system (inbreeding vs. outbreeding species) and ploidy level
(haplodiploid vs. diploid species), which are factors that have been
assumed to contribute to social evolution, although their
respective roles in social evolution are controversial (1). The
ambrosia beetle subtribe Xyleborini is characterized by regular
inbreeding, haplodiploidy, and fungiculture (8, 14). High
relatedness and haplodiploidy in combination with an extremely
female-biased sex ratio are factors predisposing them to advanced
sociality (1). Additionally, cooperation in fungi-culture is likely
because a single individual can hardly maintain a fungus garden on
its own (10). Indeed, it was shown that adult female offspring
delay dispersal from their natal gallery, which results in an overlap
of eggs, larvae, pupae, and at least two generations of adult
offspring within a colony (8). A helper effect of philopatric females
has been indicated by the fact that the number of staying females
that do not reproduce correlates positively with gallery
productivity in Xyleborinus saxesenii Ratzeburg (8). Behavioral
observations of ambrosia beetles within their galleries have been
missing so far, however, because it is virtually impossible to
observe them in nature inside the wood. The only report on
eusociality in ambrosia beetles is not based on behavioral data,
but reproductive roles have been inferred by destructive sampling
of active nests of Austroplatypus incompertus (9). To facilitate
observations of beetle behaviors inside galleries, we developed
artificial observation tubes to contain entire colonies of
reproducing beetles (15, 16). Here we use this breeding technique
of X. saxesenii to ask (i) whether offspring produced in a gallery
engage in alloparental brood care and fungus maintenance, (ii)
whether different types of individuals specialize in divergent tasks,
and (iii) whether decisions to help and to disperse relate to the
number of potential beneficiaries and the number of potential
workers present in the colony. Furthermore, we evaluate
experimentally (iv) whether female dispersal depends on the
presence of an egg-laying foundress, because her removal should
affect the need for alloparental care. We compare our results with
the patterns of sociality known from other major insect taxa.
Division of labor among the workers of insect societies is a
conspicuous feature of their biology. Social tasks are commonly
shared among age groups but not between larvae and adults with
completely different morphologies, as in bees, wasps, ants, and
beetles (i.e., Holometabola). A unique yet hardly studied holometabolous group of insects is the ambrosia beetles. Along with one
tribe of ants and one subfamily of termites, wood-dwelling ambrosia
beetles are the only insect lineage culturing fungi, a trait predicted to
favor cooperation and division of labor. Their sociality has not been
fully demonstrated, because behavioral observations have been
missing. Here we present behavioral data and experiments from
within nests of an ambrosia beetle, Xyleborinus saxesenii. Larval and
adult offspring of a single foundress cooperate in brood care, gallery
maintenance, and fungus gardening, showing a clear division of labor
between larval and adult colony members. Larvae enlarge the gallery
and participate in brood care and gallery hygiene. The cooperative
effort of adult females in the colony and the timing of their dispersal
depend on the number of sibling recipients (larvae and pupae), on
the presence of the mother, and on the number of adult workers.
This suggests that altruistic help is triggered by demands of brood
dependent on care. Thus, ambrosia beetles are not only highly social
but also show a special form of division of labor that is unique among
holometabolous insects.
altruism | cooperative fungiculture | insect agriculture | larval workers | mutualism
D
ivision of labor enhances work efficiency and is fundamental
to the biological evolution of social complexity (1). This
conspicuous feature of social insects largely explains their ecological success (2). Task specialization in insects usually occurs
between different age groups. Individuals either pass through
consecutive molts during development (Hemimetabola; e.g.,
termites, aphids), with immatures resembling small adults and
division of labor occurring typically between such larval nymphs
that may show morphological specializations [e.g., first-instar
soldier aphids (3)]; or individuals metamorphose by dramatically
reorganizing their morphology during the pupal stage (Holometabola; e.g., bees, wasps, ants, beetles), which predestines
larvae and adults to specialize in different tasks because of their
morphological and physiological differentiation. Indeed, larvae of
the eusocial Hymenoptera, for example, may produce nestbuilding silk [weaver ants (4)] and may supply adults with extra
enzymes and nutrients [trophallaxis in several wasps and ants (2,
5, 6)]. However, all known cooperative actions of holometabolous
larvae are apparently completely under adult control (2), and
despite the potential for the evolution of highly specialized
immature helper morphs, larvae of these taxa are largely immobile and helpless and in need of being moved, fed, and cared
for by adults (7). In ambrosia beetles, however, in which cooperative breeding (8) and eusociality (9) also have been assumed,
larvae can move and forage independently in the nest, which
provides great potential for division of labor between larvae and
adults.
Division of labor sets the stage for the origin of fungiculture in
fungus-growing ants and termites (10). Ambrosia beetles originate
from solitary or colonial ancestors, and their fungus agriculture
may have coevolved with sociality (8, 10, 11). However, the role of
division of labor is unknown. Ambrosia beetles live inside trees,
Results
Age- and Sex-SpecificBehavior.
Gallery maintenance and brood care were allocated differently
between different age classes within a nest(Fig.1andTable S1). The
gallery was extended mainly by larvae (digging), which also
reduces the spread of mold (SI Text and Fig. S1), whereas fungus
care (cropping) and brood protection (blocking) were exclusively
conducted
by
adults.
Blocking
was
only
Author contributions: P.H.W.B. and M.T. designed research; P.H.W.B. performed
research; P.H.W.B. analyzed data; and P.H.W.B. and M.T. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
1
11
To whom correspondence should be addressed. E-mail: [email protected].
Chapter 1
Fig. 1. Division of labor and age polyethism between age and sex classes in X. saxesenii. Bars show the mean (±SE) proportions of time larvae, teneral females, mature females,
and males performed different cooperative tasks. Statistically significant differences between the classes are denoted by different letters (P < 0.05; GEE, details in Table S1).
Note scale differences between A-C and D-F.
performed by foundresses and mature females, and in 33 of 35 cases only
one individual blocked at a time. In the 25 cases (in 19 of 93 galleries) when
we did not observe a blocking female, the following behavioral scan
revealed that larvae had crawled out of the gallery. At least 71 larvae (x =
3.7 larvae per gallery) died in this manner, which suggests that an
important function of blocking is to prevent larvae from getting
accidentally lost.
Larvae and adults took different shares in hygienic behaviors. Only
larvae compressed dispersed waste into compact balls (balling; Movie S1).
Frass-balls, pieces of wood, or parts of dead siblings were moved within
the gallery and pushed out of the entrance (shuffling; Movie S2) mainly by
mature females but to some extent also by larvae and teneral females (Fig.
1 and Table S1). The ultimate waste disposer was always the mature
female that was blocking the entrance. Cannibalism was directed toward
adult beetles (1 case), pupae (3 cases), and larvae (37 cases) that were
already dead (in most cases) or that did not respond to being groomed. In
such cases the groomer (larva or adult) would use its mandibles to open
the body of the groomed sibling within seconds. Allogrooming was shown
by all stages and both sexes, and it seemed to be crucial for individual
survival: in an experiment with 10 pupae that were singly kept either with
one or six larvae, the pupae survived in five of five cases with six larvae, but
they survived in only one of five cases with one larva (Fisher exact test: P =
0.048, n = 10; details in SI Text and Fig. S2). In the other four cases with one
larva, the pupae were overgrown and killed by fungi. In addition, the body
surface of single foundresses that had not yet successfully established a
brood was overgrown by a fungal layer (mainly Paecilomyces variotii and
Fusarium merismoides) within a few weeks, which caused the death of at
least 7 of 29 solitary females. Whenever individuals of different stages
encountered each other, they removed the visible fungal layer on each
other's
bodies by allogrooming. In males, allogrooming was frequently followed
by a mating attempt; it may thus serve also to obtain information about
female mating status.
Adult Female Behaviors Depending on Gallery Composition.
The proportion of time adult females exhibited cropping and allogrooming
and the occurrence of blocking were higher during gallery stages when
brood dependent on care was present in the colony (preadult and larvaladult gallery stages) than during other times (postlarval gallery stage; Table
S2). Shuffling was shown equally often before, during, and after larval
presence within a gallery, whereas adult female digging tended to increase
per capita after pupation of the last larva. Cannibalism was not shown
before the hatching of adult daughters, and it occurred more often when
dependent offspring were still present. Dispersal of adult females seemed
to be contingent on brood care demands: it significantly increased in the 10
d after a major proportion of larvae had completed their development
relative to the 10 d before (Wilcoxon: z = -3, P < 0.001, n = 23; Fig. S3).
Because brood numbers likely correlate with fungus productivity, dispersal
could be triggered by alternative factors, however, like the quality of the
wood/medium or fungus. Nevertheless, in summary there are hints that
some behavioral tasks of adult females functionally relate to the demands
of care-dependent brood.
In a second analysis we tested for the relationship between adult female
behaviors and the numbers of adult females and brood (pupae and larvae)
present in the galleries during the larval-adult gallery stage only (Table S2).
Analyzed per capita, adult female digging, shuffling, cannibalism, and
blocking were all independent of the numbers of adult females and
younger nestmates. However, per capita allogrooming and fungus cropping
activities significantly increased with increasing numbers of pu-
12
Chapter 1
brood produced by the latter. Dispersal propensity of adult females (i)
correlates positively with low numbers of dependent young and high
numbers of adult workers, and (ii) is increased by experimental removal of
the foundress. This suggests that philopatry may be related to indirect
fitness benefits, because the group members sharing a gallery are all very
highly related (for another xyleborine beetle see ref. 14). There is no
morphological differentiation among adult females, and they are all fully
capable of breeding and establishing their own gallery. However,
dissections of all females in a number of field galleries showed that
daughters cobreed in their natal gallery in only a quarter of colonies (17). In
summary, the social and reproductive patterns of X. saxesenii conform to
primitive eusociality, defined by overlapping generations, cooperative
brood care, and some reproductive division of labor, despite totipotency in
reproduction of all individuals.
pae and larvae, whereas with increasing adult female numbers
allogrooming significantly decreased. Adult female dispersal correlated
negatively with the number of dependent offspring [generalized estimation
equation (GEE): P = 0.003; Fig. 2A and Table S2] and positively with adult
female numbers (GEE: P < 0.001; Table S2).
In 34 of 43 galleries with mature offspring, females delayed their
dispersal from the natal nest after maturation. This philopatric period (i.e.,
the latency from the first appearance of a mature female within a gallery
until the first dispersal event) correlated positively with the average
number of dependent brood (eggs, larvae, pupae) per adult female present
during this period (Spearman rank correlation: R S = 0.379, P = 0.027, n =
34).
Effect of Foundress Removal.
Division of Labor Between Larvae and Adults.
Eggs are produced primarily by the foundress, and in 4 of 16 galleries
dissected in the field eggs were produced also by at least one daughter (on
average reproduced 23.9% of all females in these 4 galleries: range, 2-4
egg-layers on 4-22 females in total; details in ref. 17). If fewer eggs are produced, the need for alloparental care declines. Therefore, we predicted
that the number of dispersing daughters will increase if the foundress is
experimentally removed from the gallery. Our experimental interference
raised the dispersal activity of daughters (relative to the same galleries
before the manipulation; Fig. 2B) in the treatment (Wilcoxon: z = -2.637, P
= 0.008, n = 10) and in the control groups (Wilcoxon: z = -1.82, P = 0.034, n
= 10). Removal of the foundress, however, raised the dispersal of
daughters much more strongly than the control situation (i.e., removal of
the medium without the foundress; U test: z = -3.708, P < 0.001, n = 10 +
10).
Task sharing in X. saxesenii is unequal between the sexes and age classes
for most of the social behaviors described (Fig. 1 and SI Text). Regarding
gallery hygiene, for example, larvae form balls from dispersed frass,
whereas the transport and removal of these balls is mainly performed by
adult females. Gallery extension is almost exclusively accomplished by
larvae. A social role of larvae has never been reported in ambrosia beetles,
and gallery maintenance and fungiculture have been attributed solely to
the foundress (11, 15, 18). Larvae might serve a cooperative function not
only in Xyleborini but also in other beetles, as anecdotal observations and
speculations suggest from the Scolytinae [e.g., Dendroctonus sp.(19); the
Xyloterini (20)], Platypodinae [e.g., Trachyostus ghanaensis (21);
Doliopygus conradti (22)], and Passalidae [Passalus cornatus (23)]. The
ultimate cause for the larval specialization in digging may relate to their
repeated molting: as the mandibles gradually wear during excavation (for
wood-dwelling termites see ref. 24), adult females showing extensive
digging behavior would suffer from substantial long-term costs. In contrast,
larval mandibles regenerate at each molt.
Teneral and mature females take over fungus care. Blocking of the
gallery entrance is done exclusively by mature females and almost only by
the foundress (see ref. 14 for similar observations in Xylosandrus
germanus). Direct brood care by allogrooming is performed by all age
classes. Males groom females at high rates, which may primarily serve
courtship because mating attempts are always initiated by allogrooming(cf.
20). Males do not take part in other social behaviors except for low levels
of digging and cropping. Asymmetries in relatedness caused by
haplodiploidy should favor females to become helpers (25, 26). Inbreeding
can nevertheless reduce relatedness asymmetries and thus favor also
males to help (27). Male soldiers and brood-caring males have been
documented in haplodiploid and sib-mating thrips and Cardiocondyla ants,
respectively (27). In Xyleborini, very few males are produced
[approximately 5-12% of offspring (28)], and males do not seem to
contribute significantly to gallery function, hygiene, and fungus growth.
Instead, they seem to specialize in their reproductive role by continually
attempting to locate and fertilize their sisters. Nevertheless, both male
larvae and adults are fully capable of performing all of the behaviors
exhibited by females, as demonstrated in galleries that contain only males
[approximately 2% of X. saxesenii galleries are founded by unfertilized
females that produce solely male broods (16, 28)].
The life-history trajectory of X. saxesenii is most similar to that of social
aphids and some termite families, where individuals serve as helpers (e.g.,
workers and soldiers) in their natal colonies before maturation (Table 1).
Either sex may help in these taxa, and their flexible developmental period
allows individuals to adjust the length of their immature stage to maximize
their
Discussion
Here we report on division of labor and age polyethism in Coleoptera,
which apparently relate to the fungiculture in ambrosia beetles. It confirms
the predicted high degree of sociality of ambrosia beetles (9, 11). In X.
saxesenii, all group members contribute to the divergent tasks of gallery
maintenance and brood care, and there is correlative evidence that female
offspring delay their dispersal depending on the number of potential
beneficiaries present in their natal gallery. Tasks are typically shared
differentially between larval and adult colony members, which resembles
the division of labor reported from termites and other highly social insect
taxa (Table 1). Among holometabolous insects, however, X. saxesenii is the
only species to date known to exhibit an active behavioral task
specialization between larvae and adults. Furthermore, adult daughters in
this species specialize in different tasks than the colony foundress, and
they seem to adjust their work effort flexibly to the varying size of the
Fig. 2. (A) Adult female dispersal correlates negatively with the number of cared
brood in the gallery (larvae and pupae; GEE: P = 0.003; statistical details in Table S2).
(B) Effects of removal of the blocking foundress. The numbers of dispersing X.
saxesenii adult females (medians and quartiles) are shown 4 d before and 4 d after
the experimental treatments. Wilcoxon tests: *P < 0.05, **P < 0.01; Mann-Whitney U
test: ***P < 0.001; sample sizes were 10 control galleries and 10 galleries from which
the foundress was removed.
13
Chapter 1
Table 1. Forms of division of labor in exemplary species of the most prominent social insect taxa
In Xyleborini, ideal conditions for the evolution of both mutualism and
altruism seem to prevail: (i) they are well pre-adapted to brood care
because they originate from a beetle lineage (Scolytinae) with parental
care (38); (ii) they breed in isolated galleries that are founded by one
reproductive that dominates offspring production [i.e., the foundress (17)];
(iii) they are haplodiploid and mate predominantly among full siblings [e.g.,
inbreeding coefficients of approximately 0.9 in Xylosandrus germanus
Reiter (8)], which increases relatedness within natal colonies; and (iv) they
disperse solitarily after maturation to reproduce elsewhere. In addition, (v)
colony members benefit greatly from cooperation due to their dependence
on fungi-culture; and (vi) the wood used as a resource for shelter and
substratum for fungi is virtually non-depreciable, which renders resource
competition negligible. Larval digging, for example, which might be
regarded as a by-product of selfish larval feeding, reduces competition
because it generates a common good (i.e., space and substratum for
fungiculture). Similarly, other group members benefit from gallery hygiene
resulting as a by-product from the seemingly selfish adult feeding activities
cropping and cannibalism. By contrast, balling, shuffling, allogrooming,
blocking, and adult digging may rather be altruistic. These behaviors apparently benefit other group members at the expense of time and energy
costs to the actors, without immediate benefits to the latter. Particularly
dangerous are allogrooming and blocking, because they expose the
performers to pathogens, parasites, and predators.
Adult X saxesenii females showed a strong incentive to stay and
cooperate in a productive natal nest. Partly this may be
inclusive fitness. They may either remain in an immature helper phase at
the nest or develop into adults that can disperse or reproduce in their natal
colony (3, 29-32). Likewise, the developmental period of second- and thirdinstar larvae of X. saxesenii can vary between 4 and 17 d (16). In addition,
there is a striking similarity of these taxa in ecology and mating patterns.
They all inhabit defensible nests, often within wood, which favors local
mating and inbreeding – conditions claimed as strongly favoring social
evolution (12). In many of these cases helping tasks do not seem to curtail
a helper's future reproduction (i.e., because helping is risk free and does
not reduce a helper's energy stores), which may weaken the tradeoff
between helping and future reproduction (33, 34).
Role of Individual Selection and Kin Selection.
We defined a behavior as cooperative if social partners potentially benefit
from its performance, independently of whether this behavior entails net
costs to the actor (35). This definition includes (i) mutualistic behaviors that
are regarded as selfish acts generating benefits to other individuals
(common goods) as a by-product (36), and (ii) altruistic behaviors that bring
about net costs to the actor, which are compensated through indirect
fitness benefits via kin-selection; they will thus only evolve in groups of
relatives (25). In the course of social evolution and task specialization
shaped by kin selection, mutualistic behaviors may lose their original
function and change into truly altruistic behaviors (36). High relatedness
and spatial separation between groups are very favorable to the evolution
of cooperative behaviors (12), as long as local competition between
relatives does not oppose this force (37).
14
Chapter 1
Adult female behaviors depending on gallery composition. To check for potential
effects of gallery age on behavior, we discriminated between the following
successive stages of gallery development: (i) preadult gallery stage: founder
female and dependent offspring (larvae and pupae) with or without eggs present
(n = 2 galleries); (ii) larval-adult gallery stage: all age classes (mature/ teneral
beetles, larvae, and pupae) with or without eggs present (n = 3 1 galleries); and
(iii) postlarval gallery stage: only adults without eggs, larvae, or pupae present (n
= 16 galleries). Variation in sample size of galleries was caused by the fact that not
all age classes were visible in every gallery. We used a first series of GEEs to
identify the effect of the particular stage of gallery development on the
proportion of time adult females (= teneral and mature females) spent with a
certain behavior. In a second GEE series performed only with data of the larvaladult gallery stage, we analyzed whether and how task performance of adult
females related to the number of adult females present and to the number of
pupae and larvae present in the gallery (Table S2).
Female dispersal. For each gallery we measured the retention period between
the first appearance of a mature daughter (fully sclerotized, ready to disperse)
within the gallery and the first female dispersal event (i.e., the female had left
through the gallery entrance and sat on top of the medium, where it tried to
escape through the cap). Using a Spearman rank correlation analysis we tested
whether the dispersal delay interval related to the mean number of offspring
attended, divided by the mean number of adult females present at that time.
Influence of foundress on offspring dispersal. In the entrance tunnel foundresses
either block (sit still and fully close the tunnel) or move back and forth when
shuffling material to the dumps. To determine the influence of the foundress'
presence on the behavior and dispersal propensity of mature females, we
experimentally removed the first centimeter of the entrance tunnel when a
female was present in there (n = 11 galleries), at a stage when eggs, larvae, and
adult daughters were present together with the foundress. We determined the
reproductive status of the removed female by dissection to check whether we
had successfully removed the foundress. We excluded one gallery from the
treatment group, where we found an immature female blocking instead of the
foundress. In the control group (n = 10 galleries) we removed the first centimeter
of medium when no female was present in this part of the entrance tunnel.
Experimental galleries were randomly assigned to treatment and control groups.
Dispersal of the daughters was measured in both groups for 4 d before and 4 d
after the intervention by collecting the females on top of the medium that had
left the gallery through the entrance and tried to disperse through the cap of the
tube (Fig. 2B).
selfish, because up to one quarter of females may get the chance to
reproduce (17), and females might build up reserves during their
philopatric period for subsequent dispersal and nest foundation. An
experimental study in the ambrosia beetle Xyleborus affinis suggests,
however, that females do not build up reserves during their philopatric
period but rather suffer direct fitness costs (39). Alternatively, indirect
fitness benefits may be involved in helping to raise siblings (8). Three
results suggest that adult female cooperation is triggered by the
demands of brood dependent on care: (i) helping effort of adult
females rose with a greater number of brood dependent on care, (ii)
adult females dispersed at a higher rate with an increasing number of
workers in the colony, and (iii) dispersal rate of females increased in
response to experimental removal of the foundress, which indicates
that delayed dispersal of females is not primarily motivated by the
potential to breed in the natal gallery (see also ref. 8).
In conclusion, the high degree of sociality in ambrosia beetles seems
to result from a combination of four major factors: (i) parental care as a
preadaptation for the evolution of sociality in the ancestors of modern
ambrosia beetles (38); (ii) very high relatedness within families due to
haplodiploidy and inbreeding; (iii) a proliferating, monopolizable
resource providing ample food for many individuals, which needs to be
tended and protected; and (iv) high costs of dispersal (for another
scolytine beetle see ref. 40) due to the difficulties of finding a suitable
host tree, of nest foundation, and a successful start of fungiculture (11,
16), which render predispersal cooperation particularly worthwhile. X.
saxesenii larvae are predisposed to assume certain tasks like balling of
frass and gallery enlargement (digging) because of their body
morphology and the frequent renewal of mandibles by molting. Thus,
behavioral tasks are shared between larval and adult stages. This has
not been shown for beetles, and the described division of labor
between immature and adult stages seems to be unique among
holometabolous social insects at large.
Materials and Methods
Study System. X. saxesenii galleries are founded by individual females that
transmit spores of the species-specific ambrosia fungus Ambrosiella sulfurea Batra
in their gut from the natal to the new gallery (41, 42). This fungus forms a yellow
layer of fruiting cells on the surface of gallery walls (Fig. S4A). After landing on a
tree trunk, the foundress excavates a straight tunnel into the xylem with a small
egg niche at its end. As soon as fungus beds emerge, she feeds on them and starts
egg-laying (16). During offspring development the egg niche is enlarged to a single
flat brood chamber of up to a few square centimeters in size and ss1-mm height.
There, most of the fungus garden grows, and individuals of all age classes live in
close contact with each other (Fig. S4C). In this study we bred X. saxesenii in glass
tubes filled with artificial medium that mainly contained sawdust (for details on
this method see SI Text and ref. 16).
Statistics. We used a series of GEEs [lmer in R (43)], which are an extension of
generalized linear models with an exchangeable correlation structure of the
response variable within a cluster (= gallery identity), to analyze effects of
dependent variables on correlated binary response variables (proportional data
were transformed to binary data) and to identify factors affecting the relative
behavioral frequencies per class (larvae, adult females, and males) (44). First, we
tested whether the larvae, teneral females, mature females, and males show
different tendencies to express the cooperative behaviors (Fig. 1 and Table S1).
Second, we compared these frequencies between foundresses and their mature
daughters (SI Text and Table S4). In a third series of GEEs we determined the
influence of a particular developmental stage of the gallery (preadult, larval-adult,
and postlarval gallery stages) on the relative behavioral frequency per class (Table
S2). Finally, we modeled whether larvae and adult female numbers affected the
relative frequencies of cooperative behaviors in adult females (Table S2). For the
removal experiment we compared behavioral frequencies and dispersal activity
between the groups using Mann-Whitney U tests, and within groups using
Wilcoxon matched-pairs signed-ranks tests (Fig. 2B). All statistical analyses were
performed with SPSS version 15.0 and R version 2.8.1 (43).
Behavioral Recordings and Analyses. In total, 93 of roughly 500 galleries were
founded successfu lly in the laboratory (i.e., eggs were laid), and in 43 of these
galleries individuals reached adulthood. These galleries were used for behavioral
analyses. We distinguished 11 behaviors (Table S3), which comprised seven
cooperative behaviors that apparently raise the fitness of colony members. Every
second to third day we performed scan observations of all individuals visible
within a gallery (details in SI Text).
Age- and sex-specific behavior. We used GEEs (details below) with an exchangeable correlation structure of the response variable within a cluster (=
gallery identity) to identify effects of the four classes of individuals (larvae, teneral
females, mature females, males) on the proportion of time spent with a certain
behavior, using binomial error distributions (Fig. 1 and Table S1).
ACKNOWLEDGMENTS. The previous work of K. Peer provided the basis of this
study. We thank C. Arnold and S. Knecht for help in a pilot study, and U.G.
Mueller, T. Turrini, R. Bergmueller, K. Peer, and two anonymous reviewers for
helpful comments on the manuscript. P.H.W.B. was partially supported by the
Roche Research Foundation and a DOC grant from the Austrian Academy of
Science.
15
Chapter 1
1. Bourke AFG (2011) Principles of Social Evolution (Oxford Univ Press, Oxford).
2. Holldobler B, Wilson EO (1990) The Ants (Belknap Press of Harvard Univ Press, Cambridge,
MA).
3. Stern DL (1997) Foster WA. Social Behaviour in Insects and Arachnids, eds Choe JC, Crespi BJ
(Cambridge Univ Press, Cambridge, U.K.), pp 150-165.
4. Wilson EO, Holldobler B (1980) Sex differences in cooperative silk-spinning by weaver ant
larvae. Proc Natl Acad Sci USA 77:2343-2347.
5. Ishay J, Ikan R (1968) Food exchange between adults and larvae in Vespa orientalis F. Anim
Behav 16:298-303.
6. Hunt JH, Baker I, Baker HG (1982) Similarity of amino-acids in nectar and larval saliva - the
nutritional basis for trophallaxis in social wasps. Evolution 36:1318-1322.
7. Wilson EO (1971) The Insect Societies (Belknap Press of Harvard Univ Press, Cambridge, MA).
8. Peer K, Taborsky M (2007) Delayed dispersal as a potential route to cooperative breeding in
ambrosia beetles. Behav Ecol Sociobiol 61:729-739.
9. Kent DS, Simpson JA (1992) Eusociality in the beetle Austroplatypus incompertus
(Coleoptera: Platypodidae). Naturwissenschaften 79:86-87.
10. Mueller UG, Gerardo NM, Aanen DK, Six DL, Schultz TR (2005) The evolution of agriculture
in insects. Annu Rev Ecol Evol Syst 36:563-595.
11. Kirkendall LR, Kent DS, Raffa KF(1997) The Evolution of Social Behavior in Insects and
Arachnids, eds Choe JC, Crespi BJ (Cambridge Univ Press, Cambridge, U.K.), pp
28. Biedermann PHW (2010) Observations on sex ratio and behavior of males in Xyleborinus
saxesenii Ratzeburg (Scolytinae, Coleoptera). Zookeys 56:253-267.
29. Thorne BL (1997) Evolution of eusociality in termites. Annu Rev Ecol Syst 28:27-54.
30. Crespi B J (1997) Mound LA. The Evolution of Social Behaviour in Insects and Arachnids, eds
Choe JC, Crespi B J (Cambridge Univ Press, Cambridge, U.K.), pp 166-180.
31. Chapman TW, Crespi BJ, Perry SP (2008) Ecology of Social Evolution, eds Korb J, Heinze J
(Springer, Berlin), pp 57-83.
32. Korb J (2008) Ecology of Social Evolution, eds Korb J, Heinze J (Springer, Berlin), pp 151174.
33. Queller DC, Strassmann JE (1998) Kin selection and social insects. Bioscience 48: 165-175.
34. Korb J (2008) Ecology of Social Evolution, eds Korb J, Heinze J (Springer, Berlin), pp
245-259.
35. Brosnan S, de Waal F (2002) A proximate perspective on reciprocal altruism. Hum Nat
13:129-152.
36. Lin N, Michener CD (1972) Evolution of sociality in insects. Q Rev Biol 47:131 -159.
37. Hamilton WD (1975) Biosocial Anthropology, ed Fox R (John Wiley & Sons, New York),
pp 133-155.
38. Jordal BH, Sequeira AS, Cognato AI (2011) The age and phylogeny of wood boring weevils
and the origin of subsociality. Mol Phylogenet Evol 59:708-724.
39. Biedermann PHW, Klepzig KD, Taborsky M (2011) Costs of delayed dispersal and alloparental care in the fungus-cultivating ambrosia beetle Xyleborus affinis Eichhoff
(Scolytinae: Curculionidae). Behav Ecol Sociobiol 65:1753-1761.
40. Garraway E, Freeman BE (1981) Population-dynamics of the juniper bark beetle
Phloeosinus neotropicus in Jamaica. Oikos 37:363-368.
41. Batra LR (1966) Ambrosia fungi: Extent of specificity to ambrosia beetles. Science 153: 193195.
42. Francke-Grosmann H (1975) Zur epizoischen und endozoischen Übertragung der
symbiotischen Pilze des Ambrosiakäfers Xyleborus saxeseni (Coleoptera: Scolitidae).
Entomologica Germanica 1:279-292.
43. R Development Core Team (2008) R: A Language and Environment for Statistical
Computing (R Foundation for Statistical Computing, Vienna, Austria).
44. Zeger SL, Liang KY, Albert PS (1988) Models for longitudinal data: A generalized estimating
equation approach. Biometrics 44:1049-1060.
45. Crosland MWJ, Traniello JFA, Scheffrahn RH (2004) Social organization in the dry-wood
termite, Cryptotermes cavifrons: Is there polyethism among instars? Ethol Ecol Evol 16:117132.
46. Machida M, Miura T, Kitade O, Matsumoto T (2001) Sexual polyethism of founding
reproductives in incipient colonies of the Japanese damp-wood termite Hodotermopsisjaponica (Isoptera: Termopsidae). Sociobiology38:501-512.
47. Crosland MWJ, Ren SX, Traniello JFA (1998) Division of labour among workers in the
termite, Reticulitermes fukienensis (Isoptera: Rhinotermitidae). Ethology104:57-67.
48. Badertscher S, Gerber C, Leuthold RH (1983)Polyethism in food-supply and processing in
termite colonies of Macrotermes subhyalinus (Isoptera). Behav Ecol Sociobiol 12: 115-119.
49. Benton TG, Foster WA (1992) Altruistic housekeeping in a social aphid. Proc R Soc
Lond B Biol Sci 247:199-202.
50. Crespi B J (1992) Behavioral ecology of Australian gall thrips (Insecta, Thysanoptera).
J Nat Hist 26:769-809.
51. Choe JC, Crespi B J (1997) The Evolution of Social Behaviour in Insects and Arachnids
(Cambridge Univ Press, Cambridge, U.K.).
52. Lindauer M (1953) Division of labour in the honey bee colony. Bee World 34:63-73.
53. Korb J, Heinze J (2008) Ecology of Social Evolution (Springer, Berlin).
181 -215.
12. Hamilton WD (1978) Diversityof Insect Faunas, eds Mound LA, Waloff N (Blackwell,
Oxford), pp 154-175.
13. Farrell BD, et al. (2001) The evolution of agriculture in beetles (Curculionidae: Scolytinae
and Platypodinae). Evolution 55:2011 -2027.
14. Peer K, Taborsky M (2005) Outbreeding depression, but no inbreeding depression in
haplodiploid Ambrosia beetles with regular sibling mating. Evolution 59:317-323.
15. Saunders JL, Knoke JK (1967) Diets for rearing the ambrosia beetle Xyleborus ferrugineus
(Fabricius) in vitro. Science 15:463.
16. Biedermann PHW, Klepzig KD, Taborsky M (2009) Fungus cultivation by ambrosia beetles:
Behavior and laboratory breeding success in three xyleborine species. Environ Entomol
38:1096-1105.
17. Biedermann PHW, Peer K, Taborsky M (2011) Female dispersal and reproduction in the
ambrosia beetle Xyleborinus saxesenii Ratzeburg (Coleoptera; Scolytinae). Mitt Dtsch Ges
Allg Angew Entomol.
18. Francke-Grosmann H (1967) Symbiosis, ed Henry SM (Academic Press, New York), pp
141 -205.
19. Deneubourg JL, Gregoire JC, LeFort E (1990) Kinetics of larval gregarious behaviour in the
bark beele Dendroctonus micans (Coleoptera: Scolytidae). J Insect Behav 3:169-182.
20. Wichmann HE (1967) Die Wirkungsbreite des Ausstossreflexes bei Borkenkäfern. Anzeiger
für Schädlingskunde. J Pest Sci 40:184-187.
21. Roberts H (1968) Notes on biology of ambrosia beetles of genus Trachyostus Schedl
(Coleoptera - Platypodidae) in West Africa. Bull Entomol Res 58:325-352.
22. Browne FG (1963) Notes on the habits and distribution of some Ghanaian bark beetles and
ambrosia beetles (Coleoptera: Scolytidae and Platypodidae). BullEntomol
Res 54:229-266.
23. Miller WC (1932) The pupa-case building activities of Passalus cornutus Fab. (Lamellicornia). Ann Entomol Soc Am 25:709-712.
24. Roisin Y (2000) Termites: Evolution, Sociality, Symbioses, Ecology, eds Abe T, Bignell DE,
Higashi M (Kluwer Academic, Dordrecht, The Netherlands), pp 95-119.
25. Hamilton WD (1964) The genetical evolution of social behaviour. I. J Theor Biol 7:
1 -16.
26. Alexander RD (1974) The evolution of social behavior. Annu Rev Ecol Syst 5:325-383.
27. Hamilton WD (1972) Altruism and related phenomena, mainly in social insects. Annu
Rev Ecol Syst 3:193-232.
16
Chapter 1
survival of pupae, and that (ii) six digging larvae should be more
successful than one digging larva to hinder the spread of mold on the
chamber walls.
Survival of pupae was significantly affected by the number of
coinhabiting larvae (Fisher exact test: P = 0.048, n = 10). All five pupae
survived in the chambers with six larvae, whereas only one of the five
pupae survived in the chamber with one larva (Fig. S2). Pupae died
because they were overgrown by fungi.
The percentage of chamber wall area covered with mold after 14 d
was significantly lower when the chamber was inhabited by six larvae
(median 24.1%) than by one larva [median 90.5%; generalized
estimation equation (GEE): coefficient ± SE = -3.072 ± 0.163, z = 18.86, P
< 0.001, n = 26; Fig. S1]. Mold fungi covered 93% of the chamber area in
two chambers that were left empty for 14 d.
In summary, this experiment revealed the importance of (i) larval
allogrooming for the survival of pupae and of (ii) larval digging to hinder
the spread of mold fungi within the gallery.
Supporting Information
Biedermann and Taborsky 10.1073/pnas.1107758108
SI Text
Activity of Individuals Dependent on Light and Gravity. All behavioral
observations of Xyleborinus saxesenii in this study were done under a
microscope with a 6-W artificial light source. For storage, however,
tubes were wrapped in paper to keep the gallery dark as if in wood.
Therefore, we tested whether the changing of light conditions and of
the axis of gravity would affect beetle activity.
Five galleries during the larval-adult gallery stage were used to obtain
five activity measures after three different treatments: the numbers of
active and inactive individuals (larvae and adults combined) in each
gallery were counted (i) right after uncovering the tubes, (ii) after 30
min of exposure to a 6-W light source, and (iii) after 60 min of light
exposure, right after changing the axis of gravity by 90°.
Separate pairwise comparisons (Wilcoxon matched-pairs signed-ranks
test) of the five observations within each gallery were combined in a
metaanalysis according to Sokal and Rohlf (1): χ2(2 x N) = -2 x ∑ ln(π).
This revealed no significant influence of the light [χ2(10) < 18.31; P >
0.05] and gravity treatments (P > 0.05) on the activity of both larvae and
adults.
Behaviors of the bark beetle Ips pini also do not differ between
individuals kept in the dark or exposed to light, nor between individuals
reared in chambers in a vertical position or in a horizontal position (2).
Behaviors of Foundresses vs. Mature Daughters. Usually the foundress
cannot be distinguished from her mature daughters once the latter are
fully sclerotized. In six galleries, however, the foundress could be
distinguished from her mature daughters because they remained lighter
than their mothers for an extended period. In these galleries we made
10 focal observations of foundresses and 14 of mature daughters (10
min each). GEE models showed that foundresses and their mature
daughters differed significantly in the frequency of cooperative
behaviors, except for allogrooming (Table S4) and cannibalism, which
were never observed by foundresses. The individual blocking the gallery
entrance was always the foundress during the focal animal
observations. In the removal experiment, however, 1 of 11 dissections
of the blocking female revealed immature ovaries, indicating that
daughters may also block the gallery entrance occasionally. Per capita,
foundresses showed more fungus cropping and shuffling behavior but
less gallery extension behavior (digging) than their mature daughters. In
summary, foundresses spent more time with cooperative behaviors
than there mature daughters.
Tendency of X. Saxesenii Larvae to Aggregate. For this experiment we
removed all individuals and tunnels from seven galleries during the
larval-adult gallery stage, leaving only some centimeters of fungusinfested medium within the tubes. In these seven tubes we created
three artificial chambers of «20 mm2 (and «1-mm height) within the
medium, extending next to the glass from the top of the medium
alongside the tube wall. Six to twelve second/third-instar larvae were
placed on top of the medium in each of the tubes and allowed to move
freely. After 24 h we recorded the location of all larvae.
In six replicates, all larvae had moved inside one and the same of the
three chambers. In only one replicate had larvae split up and were
distributed over two chambers. This suggests that X. saxesenii larvae
show a strong tendency to aggregate.
Details on Function of Different Cooperative Behaviors.
Gallery extension – digging. An important common good created by
larval digging is the enlargement of the gallery. Nevertheless, it is
probably a mutually beneficial behavior with selfish benefits for the
larvae because it is also part of their feeding on fungal hyphae
penetrating the wood. The consumed wood passes the gut without any
sign of digestion (3, 4), however, because the enzymes required to
digest wood are likely missing (5, 6). Because hyphens of several fungi
co-occur together with the ambrosia fungus in the medium/wood
(especially when the gallery gets older), larval digging apparently also
hindered the spread of mold (see above and Fig. S1). Mold fungi did not
completely overgrow chambers in the presence of larvae. They probably
serve the larvae as additional food source, despite the fact that these
fungi are usually toxic to arthropods. Previous studies have shown that
toxic secondary metabolites produced by mold fungi as a response to
feeding by arthropods can be overcome if the arthropods feed
gregariously on these fungi (7-9). X. saxesenii larvae showed a strong
tendency to aggregate within the gallery (see above).
The digging of adults is clearly different from feeding and solely
serves gallery extension. Gallery enlargement by digging is a mutual
benefit to colony members, because it increases the surface where the
ambrosia
Effect of Larval Numbers on the Survival of Pupae and on Fungus
Growth. For this experiment we removed all individuals from 13
galleries during the larval-adult gallery stage. In each of these galleries
we created two flat chambers with a height of « 1 m m and an area
between 22.86 mm2 and 91.89 mm2 next to one of the existing tunnels
and next to the tube glass (to allow observations). Thereafter, one
chamber was filled with one larva and the other one with six larvae
(chambers were randomly assigned). In 5 of the 13 galleries we also
placed one pupa each in both chambers, which contained one or six
larvae, respectively. For the next 14 d we tracked the survival of the
pupae (in 5 galleries) and the appearance of mold on the walls of the
chambers (in all 13 galleries). Several fungi coexist in the medium at this
gallery stage, and mold fungi are expected to overtake if they are not
controlled by larvae. Therefore, we took pictures of the chambers on
day 14 after the treatment and analyzed the chamber area covered and
uncovered by mold according to larval numbers, by using the program
ImageJ (version 1.44p). We predicted that (i) fungal layers growing on
the body surface of pupae should be removed more frequently by six
allogrooming larvae than by one allogrooming larva, which may affect
17
Chapter 1
fungi can grow, in this way lowering within-family competition for food
and space (10). Gallery surface area positively affects fungus and thus
colony productivity (11-13). The contribution of adults to gallery
extension is negligible compared with larvae.
because females dispersed at higher rates after the removal of the
blocking foundress (Fig. 2B).
Brood care and body hygiene - allogrooming. Allogrooming was frequently observed between individuals of all stages, and its importance
was shown by (i) pupae being overgrown by fungi and dying more
frequently in the presence of one allogrooming larva compared with six
allogrooming larvae (see above), and (ii)the repeated observation of
single foundresses (before a brood was successfully produced) dying
due to a lack of getting groomed. Four solitary foundresses died because
they got adhered to the gallery wall by a fungal layer on their elytra, or
fungi grew underneath the elytra and caused them to swell so that
moving was prevented. Additionally, it has been reported from
ambrosia beetles that eggs do not hatch (41) and larvae die (21) in the
absence of the grooming foundress. The gregarious life of all age classes
that constantly groom each other might also explain the puzzling low
frequency of parasitoids and mites found in nests of ambrosia beetles
compared with those of their close relatives, the non-gregarious,
phloem-feeding bark beetles (35, 42, 43). Recent studies of other taxa
have shown that grooming can greatly decrease the success of
parasitoids [e.g., by grooming their eggs (44)], parasites [e.g., phoretic
mites (45, 46) and entomopathogenic fungi (47, 48)]. Costs of grooming
include the time and energy spent and the risk of parasite transmission
to the groomer (30, 49, 50).
Fungus care – cropping. Adult females are constantly walking and
screening the fungal layers with their large, disk-shaped antennae (Fig.
S4D). They stop frequently to move their hairy, comb-like mouthparts
through the fungus (Fig. S4E), probably brushing off sprouting ambrosia
structures (Fig. S4B). This cropping behavior of the adults apparently
serves both nutrient intake and fungus care. It induces the characteristic
ambrosial growth of the fungus, that is, the formation of copious
ambrosia cells (fruiting bodies) and sporodochia [clusters of fungus
spores (14-18)] (Fig. S4 A and B). The presence of beetles is thus crucial
for the growth of consumable ambrosia fungus structures (15, 19). Additionally, cropping is likely to prevent invasions of the ambrosia garden
by foreign fungi and microbes (14, 20, 21). Oral secretions of X. saxesenii
contain various bacteria (e.g., refs. 22 and 23) that possibly support
fungus growth either by providing nutrients or by controlling the spread
of pathogens, as shown for bacterial mutualists of fungus-culturing ants
(24-26).
Gallery hygiene - balling, shuffling, and cannibalism. Gallery hygiene is
important for a flourishing fungus garden as well as a healthy colony.
The larval balling behavior has not been described before. Bending their
bodies ventrally, larvae form cordlike frass into balls that can be shifted
within the gallery or dumped out of the entrance more easily (Movie
S1). Frass balls are shifted by shuffling (Movie S2) them toward certain
areas, where they are used either to close parts of tunnels, possibly to
regulate humidity (14); to isolate diseased areas [so-called "death chambers" (27, 28)]; or to be recycled by the fungus. We observed that the
wood chippings in the larval frass are often integrated into the fungal
beds (see also ref. 27), which suggests that the mastication of the wood
may facilitate resource utilization by the ambrosia fungi. This would
explain why nitrogen from beetle excretions had been detected in
growing fungi (29). Perhaps most importantly, frass and debris are
disposed, because space is required for fungi to grow and for beetles to
move (3, 12). Occasionally we observed sibling cannibalism, which may
serve both nutrient recycling and the removal of dead and diseased
specimens. The latter is probably essential for hindering the spread of
diseases and parasites [including mold (30, 31)], which is particularly
threatening for highly inbred communities (32, 33).
Detailed Materials and Methods.
Study system and laboratory breeding. Of 350 galleries of X. saxesenii
studied in the field, the majority produced approximately 10-25
dispersing individuals, 3.6% produced more than 100 individuals, and
one gallery exceeded 300 individuals (51). The beetles show a strong
sexual dimorphism: the rare males [the average sex ratio is
approximately 1:8 to 1:20 male/female (52)] do not fully sclerotize, stay
small, and are unable to fly. Most of them die in their natal gallery after
the females have dispersed (52).
We bred X. saxesenii in glass tubes filled with artificial medium that
mainly contained sawdust ("test medium" described in ref. 53). We used
one founder female per tube that had either been collected in the field
or originated from a brood raised in the laboratory. Females excavated
galleries and brood chambers largely along the transparent tube wall,
which enabled observations of activity and development (Fig. S4A).
When a gallery had been successfully established, the tube was
wrapped in paper to keep it dark as if in wood, but light could enter
through the entrance at the top of the tube. The tube was kept in a
constant light/dark cycle (11 h light/13 h dark) at 28 °C/22 °C and 70%
humidity, and the paper wrap was only removed during observations
under a microscope (x6.4 to x16 magnification) with an artificial light
source (maximum 6 W). Behavior of scolytine beetles is affected neither
by light (kept in the dark vs. exposed to light) nor when the gallery is
turned by 90°, changing the axis of gravity (see above and ref. 2).
Behavioral recordings and analyses. Every second to third day we
performed scan observations of all individuals visible within a gallery:
we noted the gallery identity, the number of visible individuals, and the
respective behaviors they showed at that moment, the number of
dispersing individuals found on the surface of the medium (where they
tried to escape through the cap), and we counted the number of eggs
and pupae. In each scan all gallery parts were browsed one time for
visible individuals as described. We did not discriminate between the
three larval instars. Pupae and adult beetles were sexed on the basis of
morphology and size. We discerned teneral adult females that had
recently hatched and showed weak sclerotization and brownish elytra,
and mature adult females that were fully sclerotized with dark brown to
black elytra.
From each scan observation we noted the proportion of individuals
per class (larvae, teneral females, mature females, males) performing a
respective behavior. In total we conducted 500 scan observations of 43
galleries per age class (x = 11.9 scans per gallery, range 2-40).
Gallery protection - blocking. Blocking of the entrance by a gallery
member is ubiquitous in ambrosia beetles, although it exposes the
blocking individual to predation, for instance by birds (34, 35) or
predatory beetles (36). Parasitoids also have been reported to lay eggs
on blocking ambrosia beetles (37). Blocking provides a variety of
essential services to the gallery members (see ref. 38 for review), like
regulating the microclimate, preventing larvae from falling out of the
gallery, and excluding parasites, parasitoids, predators, and foreign fungi
from the gallery, which are common threats in bark and ambrosia
beetles. Additionally, it may hinder other ambrosia beetles from
entering a proliferating gallery and foreign males from mating with
females in the gallery, which can detrimentally affect their future
reproductive success due to an outbreeding depression (39). Our study
cannot distinguish between these potential, not mutually exclusive
functions, but the data suggest that blocking by adult females is
essential for the safety of larvae. Larvae are very mobile, and in addition
to our own results, the only previous removal experiment of a blocking
female in scolytine beetles we know of also resulted in the loss of larvae
[and eggs; species: Coccotrypes dactyliperda Fabricius (40)]. In line with
this, blocking is most common during the presence of larvae and tends
to decrease in frequency after the first female offspring has matured
(Table S2). In addition, we found that blocking might also retain workers
in
the
nest,
18
Chapter 1
1. Sokal RR, Rohlf F J (1995) Biometry: The Principles and Practice of Statistics in Biological
29. Kok LT, Norris DM (1972) Symbiotic interrelationships between microbes and ambrosia
beetles 6. Aminoacid composition of ectosymbiotic fungi of Xyleborus ferrugineus
(Coleoptera, Scolytidae). Ann Entomol Soc Am 65:598-602.
30. Cremer S, Armitage SAO, Schmid-Hempel P (2007) Social immunity. Curr Biol 17: R693R702.
31. Wilson-Rich N, Spivak M, Fefferman NH, Starks PT (2009) Genetic, individual, and group
facilitation of disease resistance in insect societies. Annu Rev Entomol 54: 405-423.
32. Hamilton WD (1972) Altruism and related phenomena, mainly in social insects. Annu
Rev Ecol Syst 3:193-232.
33. Hamilton WD, Axelrod R, Tanese R (1990) Sexual reproduction as an adaptation to resist
parasites (a review). Proc Natl Acad Sci USA 87:3566-3573.
34. Moore GE (1972) Southern pine beetle mortality in North Carolina caused by parasites and
predators. Environ Entomol 1:58-65.
35. Kenis M, Wermelinger B, Gregoire JC (2004) Bark and Wood Boring Insects in Living Trees in
Europe, A Synthesis, ed Lieutier F (Kluwer Academic Publishers, Dordrecht, The
Netherlands), pp 237-290.
36. Wichmann HE (1967) Die Wirkungsbreite des Ausstossreflexes bei Borkenkäfern. Anzeiger
für Schädlingskunde. J Pest Sci 40:184-187.
37. Beaver RA(1986) The taxonomy, mycangia and biology of Hypothenemus curtipennis
(Schedl), the first known cryphaline ambrosia beetle (Coleoptera: Scolytidae). Ent Scand
17:131-135.
38. Kirkendall LR, Kent DS, Raffa KF (1997) The Evolution of Social Behavior in Insects and
Arachnids, eds Choe JC, Crespi BJ (Cambridge Univ Press, Cambridge, U.K), pp 181-215.
39. Peer K, Taborsky M (2005) Outbreeding depression, but no inbreeding depression in
haplodiploid Ambrosia beetles with regular sibling mating. Evolution 59:317-323.
40. Herfs A (1950) Studien an dem Steinnussborkenkäfer, Coccotrypes tanganus Eggers.
Höfchen Briefe 3 and 1:1-57.
41. Roeper R, Treeful LM, O'Brien KM, Foote RA, Bunce MA (1980) Life history of the ambrosia
beetle Xyleborus affinis (Coleoptera: Scolytidae) from in vitro culture. Great Lakes Entomol
13:141-144.
42. Schedl W (1964) Biologie des gehöckerten Eichenholzbohrers, Xyleborus monographus
(Scolytidae, Coleoptera). Z Angew Entomol 53:411-428.
43. Eichhorn O, Graf P (1974) Über einige Nutzholzborkenkäfer und ihre Feinde. Anzeiger für
Schädlingskunde-Journal of Pest Science 47:129-135.
44. Vincent CM, Bertram SM (2010) Crickets groom to avoid lethal parasitoids. Anim
Behav 79:51 -56.
45. Buchler R, Drescher W, Tornier I (1992) Grooming behavior of Apis cerana, Apis mellifera
and Apis dorsata and its effect on the parasitic mites Varroa jacobsoni and Tropilaelaps
clareae. Exp Appl Acarol 16:313-319.
46. Delfinado-Baker M, Rath W, Boecking O (1992) Phoretic bee mites and honeybee grooming
behavior. Int J Acarol 18:315-322.
47. Rosengaus RB, Maxmen AB, Coates LE, Traniello JFA (1998) Disease resistance: A benefit of
sociality in the dampwood termite Zootermopsis angusticollis (Isoptera: Termopsidae).
Behav Ecol Sociobiol 44:125-134.
48. Fernandez-Marin H, Zimmerman JK, Rehner SA, Wcislo WT (2006) Active use of the
metapleural glands by ants in controlling fungal infection. Proc Biol Sci 273: 1689-1695.
49. Rosengaus RB, Traniello JFA (1997) Pathobiology and disease transmission in dampwood
termites Zootermopsis angusticollis (Isoptera: Termopsidae) infected with the fungus
Metarhizium anisopliae (Deuteromycotina: Hypomycetes). Sociobiology 30:185-195.
50. Schmid-Hempel P (1998) Parasites in Social Insects (Princeton Univ Press, Princeton).
51. Hosking GB (1972) Xyleborus saxeseni, its life-history and flight behaviour in New
Zealand. N Z J For Sci 3:37-53.
52. Biedermann PHW (2010) Observations on sex ratio and behavior of males in Xyleborinus
saxesenii Ratzeburg (Scolytinae, Coleoptera). Zookeys 56:253-267.
Research (W. H. Freeman, New York).
2. Schmitz RF (1972) Behavior of Ips pini during mating, oviposition, and larval development
(Coleoptera, Scolytidae). Can Entomol 104:1723-1728.
3. Francke-Grosmann H (1956) Hautdrüsen als Träger der Pilzsymbiose bei Ambrosiakäfern. Z
Morphol Oekol Tiere 45:275-308.
4. Batra LR (1963) Ecology of ambrosia fungi and their dissemination by beetles. Trans
Kans Acad Sci 66:213-236.
5. Martin MM, Kukor JJ, Martin JS, O'TooleTE, Johnson MW (1981) Digestive enzymes of fungusfeeding beetles. Physiol Zool 54:137-145.
6. Martin MM (1983) Cellulose digestion in insects. Comp Biochem Physiol A 75:313-324.
7. Rohlfs M, Churchill ACL (2011) Fungal secondary metabolites as modulators of interactions
with insects and other arthropods. Fungal Genet Biol 48:23-34.
8. Rohlfs M, Obmann B, Petersen R (2005) Competition with filamentous fungi and its
implication for a gregarious lifestyle in insects living on ephemeral resources. Ecol
Entomol 30:556-563.
9. Rohlfs M, Hoffmeister TS (2004) Spatial aggregation across ephemeral resource patches in
insect communities: An adaptive response to natural enemies? Oecologia 140:654-661.
10. Bright DE (1973) The Bark and Ambrosia Beetles of California, Coleoptera: Scolytidae and
Platypodidae (Univ California Press, Berkeley, CA).
11. Kajimura H, Hijii N (1994) Reproduction and resource utilization of the ambrosia beetle,
Xylosandrus mutilatus,in-field and experiment populations. Entomol Exp Appl 71:121-132.
12. Kajimura H, Hijii N (1992) Dymamics of the fungal symbionts in the gallery system and the
mycangia of the ambrosia beetle, Xylosandrus mutilatus (Blandford) (Coleoptera,
Scolytidae). Ecol Res 7:107-117.
13. Tarno H, et al. (2010) Types of frass produced by the ambrosia beetle Platypus quercivorus
during gallery construction, and host suitability of five tree species for the beetle. J For Res
16:68-75.
14. Schneider-Orelli O (1913) Untersuchungen über den pilzzüchtenden Obstbaumborkenkäfer
Xyleborus (Anisandrus) dispar und seinen Nährpilz. Zentralblatt für Bakteriologie,
Parasitenkunde. Infektionskrankheiten und Hygiene II 38:25-110.
15. Batra LR, Michie MD (1963) Pleomorphism in some ambrosia and related fungi. Trans Kans
Acad Sci 66:470-481.
16. Batra LR (1966) Ambrosia fungi: Extent of specificity to ambrosia beetles. Science 153: 193195.
17. French JRJ, Roeper RA (1972) Interactions of ambrosia beetle, Xyleborus dispar (Coleoptera,
Scolytidae), with its symbiotic fungus Ambrosiella hartigii (Fungi imperfecti). Can Entomol
104:1635-1641.
18. Norris DM (1979) Nutrition, Mutualism, and Commensalism, ed Batra LR (Allanheld, Osmun
& Company, Montclair, NJ), pp 53-63.
19. Francke-Grosmann H (1967) Symbiosis, ed Henry SM (Academic Press, New York), pp 141205.
20. Lengerken H (1939) Die Brutfürsorge- und Brutpflegeinstinkte der Käfer (Akademische
Verlagsgesellschaft, Leipzig).
21. Norris DM (1993) Xyleborus ambrosia beetles—a symbiotic ideal extreme biofacies with
evolved polyphagous privileges at monophagous prices. Symbiosis 14:229-236.
22. Haanstad JO, Norris DM (1985) Microbial symbiotes of the ambrosia beetle Xyletorinus
politus. Microb Ecol 11:267-276.
23. Grubbs KJ, et al. (2011) Genome sequence of Streptomyces griseus strain XylebKG-1, an
ambrosia beetle-associated actinomycete. J Bacteriol 193:2890-2891.
24. Currie CR, Stuart AE (2001) Weeding and grooming of pathogens in agriculture by
ants. Proc Biol Sci 268:1033-1039.
25. Mueller UG, Gerardo N (2002) Fungus-farming insects: Multiple origins and diverse
evolutionary histories. Proc Natl Acad Sci USA 99:15247-15249.
26. Mueller UG, Gerardo NM, Aanen DK, Six DL, Schultz TR (2005) The evolution of agriculture
in insects. Annu Rev Ecol Evol Syst 36:563-595.
27. Hubbard HG (1897) in Some Miscellaneous Results of the Work of the Division of
Entomology(US Department of Agriculture Bureau of EntomologyBulletin No. 7), ed Howard
LO (US Department of Agriculture, Washington, DC), pp 9-13.
28. Peer K, Taborsky M (2007) Delayed dispersal as a potential route to cooperative breeding in
ambrosia beetles. Behav Ecol Sociobiol 61:729-739.
53. Biedermann PHW, Klepzig KD, Taborsky M (2009) Fungus cultivation by ambrosia beetles:
Behavior and laboratory breeding success in three xyleborine species. Environ Entomol
38:1096-1105.
19
Chapter 1
Fig. S1. Effect of X. saxesenii larvae on the growth of mold fungi. The number of larvae present within a chamber negatively affected the percentage of wall area covered with
mold fungi of artificially created chambers in fungus infested medium. Generalized estimation equation: coeff. ± SE = -3.072 ± 0.163, z = 18.86, P < 0.001, n = 26 chambers in 13
galleries (paired design). Box-whisker plots with medians (bold), 10th, 25th, 75th and 90th percentiles are shown .
Fig. S2. Survival of experimentally isolated pupae of X. saxesenii with one (Left) or six (Right) tending larvae present. Single pupae kept with only one larva were significantly more
often overgrown by fungi and killed within 14 d compared with pupae kept with six larvae (Fisher exact test: P = 0.048, n = 10 chambers).
20
Chapter 1
Fig. S3. Relationship between larval numbers present in a gallery and the dispersal activity of adult females. In 23 galleries with a single offspring generation larval numbers
dropped below three at 21-77 d (mean 46 d) after gallery foundation. Numbers of dispersing females were counted 10 d before and 10 d after this date. Significantly more
females dispersed after than before this date, that is, when all but the last two larvae had pupated (Wilcoxon matched-pairs signed-ranks tests: z = -3, P < 0.001, n = 23 galleries).
This suggests that female dispersal might be triggered by demands of larvae dependent on female care. Productivity of the fungus, which is probably decreasing at the same time
(because no new eggs are laid), is a confounding factor of this analysis, however, that we could not control for. Box-whisker plots with medians, 10th, 25th, 75th, and 90th
percentiles are shown.
Fig. S4. Morphology of X. saxesenii galleries and their inhabitants (beetles and fungi). (A) Morphology of a brood chamber in artificial medium, with different larval instars and a
teneral female. The orange layer on the gallery walls is formed by fruiting structures of the ambrosia fungus Ambrosiella sulfurea Batra. (B) Morphology of the fungal layer
depicted by scanning electron microscopy with x300 magnification. The round "balls" are fruiting structures of A. sulfurea.(C) Morphology of a brood chamber in the field, with
several third-instar larvae, teneral (light brown), and mature (black) females. A thin, orange fungus layer is lining the gallery walls. (D and E) Head and mouthparts of a X. saxesenii
female depicted by scanning electron microscopy with x200 and x500 magnification, respectively. ant., antenna; c.e., compound eye; la., labium; la.p., labial palp; ma., mandible;
max., maxilla; max.p., maxillary palp.
21
Chapter 1
Table S1. Separate GEE models to examine differences (P < 0.05) between the proportion of time the
different age and sex classes spent with the observed cooperative behaviors (see Fig. 1 in main text)
22
Chapter 1
Table S2. GEE models for examining differences (P < 0.05) between the proportion of time adult females spent with each cooperative behavior, according
to stage (see footnote)
23
Chapter 1
Table S3. Ethogram of the observed behaviors of larvae (L), females (F), and males (M)
Table S4. Separate GEE models for examining differences (P < 0.05) between the proportion of time
mature daughters and foundresses spent with cooperative behaviors
Movie S1. Balling behavior by a larval worker within the brood chamber. This sequence shows an example of balling by a third-instar larva that is surrounded by other larvae and a
pupa. With this behavior larvae collect waste material (sawdust and frass) to form balls that can be shifted in the gallery or dumped out of its entrance.
24
Chapter 1
Movie S2. Shuffling behavior by an adult female in the entrance tunnel of a gallery. This sequence shows the typical shuffling behavior by an adult female, which serves to shift balls and
pieces of waste material (sawdust and frass) out of the entrance tunnel for final disposal. Note that during this video seque nce a third-instar larva (first scene: 0-12 s) and a teneral female
(last scene: 55-59 s) are providing the shuffling female with new material for disposal.
25
Chapter 1
26
Chapter 2
Female dispersal and reproduction in the ambrosia beetle
Xyleborinus saxesenii RATZEBURG (Coleoptera; Scolytinae)
Peter H.W. Biedermann, Katharina Peer & Michael Taborsky
Abteilung Verhaltensökologie, Institut fur Ökologie & Evolution, Universität Bern
Zusammenfassung: Untersuchungen zur kooperativen Brutpflege des Ambrosiakäfers
Xyleborinus saxesenii RATZEBURG (Coleoptera; Scolytinae)
Kooperative Brutpflege und überlappende Generationen bei Arthropoden sind
vorwiegend in den Gruppen der Hautflügler und Termiten zu finden. Weniger bekannt ist,
dass auch Ambrosiakäfer (Coleoptera; Scolytinae und Platypodinae) in sozialen Gruppen
leben. Adulter Nachwuchs verbleibt vor dem Ausfliegen für einige Zeit im Nest, um sich
der Brutpflege, der Nestverteidigung, sowie der Zucht von Pilzgärten zu widmen, die
Ambrosiakäfer zu Nahrungszwecken an den Wänden ihrer Gangsysteme im Holz anlegen.
Mit Hilfe einer optimierten Bruttechnik in Glasröhrchen dokumentieren wir
überlappende Generationen von Weibchen in Galerien des Ambrosiakäfers Xyleborinus
saxesenii Ratzeburg (Coleoptera; Scolytinae). Nach Erlangung der Geschlechtsreife
verzögerten adulte Tochter den Ausflug für durchschnittlich 23 Tage; nach dem Tod des
Gründerweibchens übernahmen einige Töchter die Galerie. Da bei X. saxesenii die
Weibchen im mütterlichen Nest durch ihre Brüder besamt werden, können sie ihre Eier
auch dort ablegen. In unserer Studie haben sich jedoch nur 25% der Töchter im
mütterlichen Nest fortgepflanzt, unabhängig von der Brutgrösse, was darauf hinweist,
dass der verzögerte Ausflug vorwiegend indirekte Fitnessvorteile bietet. Diese Annahme
wird durch frühere Studien unterstützt, die zeigen, dass der Verbleib der Tochter den
Bruterfolg der Mutter verbessert. Dies ist vermutlich darauf zurückzuführen, dass
Pilzzucht von sozialen Gruppen effektiver bewerkstelligt werden kann. Es ist daher
anzunehmen, dass Sozialität und Pilzzucht bei Ambrosiakäfern in engem Zusammenhang
entstanden sind.
Key-words: eusociality, insect agriculture, delayed dispersal, cooperative fungiculture,
alloparental care, helping, haplodiploidy
Peter H.W. Biedermann, Institut fur Ökologie & Evolution, Abteilung Verhaltensökologie,
Baltzerstrasse 6, CH-3012 Bern, Schweiz, E-Mail: [email protected]
Introduction
Eusociality stands for the highest level of social organization, which is characterized by
cooperative brood care, overlap of parental and offspring generations, and the presence
of non-reproductive castes (BATRA 1966). Among insects it is found primarily in
Hymenoptera and Isoptera (WILSON 1971). In Coleoptera, the most species-rich order of
insects, sociality is surprisingly rare and eusociality has only been described for one
species of Platypodid ambrosia beetle (KENT & SIMPSON 1992). Ambrosia beetles live in the
heartwood of trees and cultivate fungi for food. Due to their cryptic life inside wood
tunnels they are difficult to study, but a recently optimized laboratory breeding technique
allows to observe their behaviour (SAUNDERS & KNOKE 1967, BIEDERMANN & al. 2009).
Delayed dispersal of adult daughters, and thus overlapping generations, are typical for
the ambrosia beetles Xyleborinus saxesenii and Xyleborus affinis, although daughters are
fully capable of founding their own nests (PEER & TABORSKY 2007, BIEDERMANN & al. 2011).
Both species are members of the Scolytinae subtribe Xyleborini that is predisposed for
kin-selected sociality because of high relatedness of families within one nest, due to sibmating and haplodiploidy. Indeed, delayed dispersing daughters in several Xyleborini have
been found to engage in a multitude of cooperative behaviours (e.g. maintenance and
protection of the gallery, care of fungi and brood) during their philopatric period (BISCHOFF
2004,
27
Chapter 2
BIEDERMANN 2007). Apparently, this imposes costs on their future reproduction. Mature X.
affinis-daughters, for instance, produce more eggs when induced to disperse early
compared to voluntarily dispersing females that have undergone a philopatric period
(BIEDERMANN & al. 2011). Apart from investing in siblings, however, daughters could
potentially also invest in own reproduction before dispersing from the natal nest.
Here we describe the typical phenology of Xyleborinus saxesenii in laboratory galleries,
which illustrates that parental and offspring generations overlap. Furthermore we
dissected the ovaries of all daughters present in field galleries to determine if they
reproduced in the natal nest.
Material and Methods
Collection and artificial rearing
We collected adult females of Xyleborinus saxesenii from field galleries by dissecting
stumps of beech (Fagus sylvatica) with active galleries in the Spilwald forest near Bern,
Switzerland. Individuals used for artificial rearing were immediately transferred to the
laboratory, surface-sterilized (by submerging them first in ethanol (95%) and then in
distilled water for a few seconds), and afterwards individually put in glass tubes filled with
an artificial rearing medium based on sawdust and agar (modified medium; for details see
BIEDERMANN & al. 2009).
Typically, adult females put onto the medium immediately start to excavate a tunnel
system (gallery), the walls of which they inseminate with spores of their mutualistic
ambrosia fungus Ambrosiella sulfurea BATRA and several other auxiliary fungi (FRANCKEGROSMANN 1975). After establishment of the fungus layers, which is a delicate process that
is successful only in about 20% of cases, successful foundresses start to lay eggs. Offspring
development and brood care can be observed when tunnels and brood chambers are
constructed next to the glass of the tubes.
Gallery phenology
For our observations we selected six laboratory galleries that provided almost full insight.
We observed these galleries every second day for a period of four months and recorded
the numbers of eggs, larvae, pupae, immature females, mature females, and males.
Females have a brownish coloration as long as they are immature and turn black when
fully sclerotized / mature. Sexes are easily discernable by morphology and body size (see
figures in FISCHER 1954). The dates of the first visible egg, the first mature female
offspring, the first dispersal of female offspring and the death of the foundress were
determined. Dispersal was defined as emergence from the gallery, i.e., when individuals
were found on the surface of the medium under the cap of the tube (BIEDERMANN 2007). A
single dead female appeared in one third of all galleries within the first 100 days after
gallery foundation. Most likely this was the foundress, because (i) they were the oldest
members of the gallery, (ii) it was always only one individual and (iii) dead adult beetles
are only very rarely found inside galleries in general (pers. obs.). Therefore, the moment
of 'death of the foundress' was defined as the day when a single dead female was found
on the surface (dead individuals, sawdust and faeces are thrown out of the nest by adult
group members). To increase the sample size for a reliable estimate of this date we used
our whole sample of laboratory galleries, which comprised 66 observations in total.
Reproduction of daughters
For 16 dissected field galleries we recorded the number of eggs, larvae, pupae, immature
and mature females and males. Dispersal had been monitored with dispersal traps (Peer
& Taborsky 2007) in 13 of these galleries from the time the gallery was founded until it
was dissected ( mean = 72 days, range = 19 -309 d). Dispersal had started in nine of these
galleries in the meantime and all dispersing females were preserved. After gallery
collection, females were stored in 95% ethanol. Dissection was accomplished from the
dorsal surface with high precision tweezers under a binocular (6.4 x - 40 x magnification).
We discriminated between immature ovaries (no oocytes visible) and egg-carrying ovaries
(the four ovarioles contain one or more oocytes; see figures in FISCHER 1954).
In order to determine the number of breeders in the galleries, we dissected all females
collected from galleries containing eggs. For galleries without eggs, a minimum of 15
females each were dissected. Females that were accidentally damaged when the gallery
was opened could not be dissected (proportion
28
Chapter 2
dissected females: Mean = 93.5%, Range = 19 -100%, N = 16 galleries). Thus, we consider
the number of egg-layers found in a gallery as a minimum estimate. However, we found
that the galleries that could not be fully dissected all had more than one egg-layer; i.e. the
number of galleries with only a single breeder was correctly determined. We also
dissected 45 dispersing females from four different galleries to check their reproductive
status.
Results
Gallery phenology
After the successful establishment of a fungal layer on the gallery walls, females started
to lay eggs between days 11 and 26 (x ± se: 18 ± 2.2; N = 6 galleries) after gallery
foundation. First mature daughters appeared between 16 and 34 days (x ± se: 26.2 ± 2.8;
N = 6) after the first egg-laying. These daughters delayed their dispersal from the natal
nest for 17 to 38 days ( x ± se: 22.8 ± 3.2; N = 6). Foundresses were found dead on the
gallery surface between 34 and 97 days (x ± se: 72.2 ± 1.8; N = 66) after gallery
foundation.
Two to three peaks of egg-laying, followed by two clear peaks of larval numbers
(Fig.1), were found by averaging the offspring numbers of the six selected galleries every
day, over the whole observation period (day 18 as the mean date of the first visible egg
was set as reference for all six galleries). Egg numbers were smaller than larval numbers
because (i) developmental periods are different (eggs : 5 days; 1st to 3rd larval instars: 8 17 days) and (ii) eggs are less visible especially at the beginning of gallery development.
The first offspring peak was obviously produced by the foundress, but the others were
probably the result of egg-laying daughters, as the foundress likely had died before. Note
that the second phase of egg-laying started at about the same time when the first mature
daughters started to disperse (Fig.1). Offspring production strongly decreased about 90 to
100 days after gallery foundation, when the medium dried out and fungal growth ceased.
Fig.1. Average phenology of six laboratory galleries of Xyleborinus saxesenii. We
synchronised the phenologies of the six galleries by matching the appearance of the first
eggs (see text).
Reproduction of daughters
A direct benefit of delayed dispersal might be reproduction inside the natal gallery. Four
of 16 field galleries contained at least two egg-laying females, (with ovarioles containing
oocytes). By contrast, all dispersing females (N = 45 females from 4 galleries) did not lay
eggs, i.e. their ovarioles contained no oocytes. More than one egg-layer was only found in
galleries where female dispersal had already
29
Chapter 2
started (N = 9). Before the onset of dispersal, there was always a single reproductive
female (i.e., the foundress; N = 4). In all galleries, numbers of sub-adult offspring (only
larvae and pupae) and eggs present at the time of dissection were positively associated
with the number of egg layers found. This was not a by-effect of gallery size, as there was
no correlation between the total number of mature females in the gallery and the
number of egg-layers (Table 1).
Table 1. Correlations (Kendall’s Tau) between total number of eggs, sub-adults (larvae
and pupae), mature females (fully sclerotized) and the number of egg laying females in
field galleries. Significant p-values are printed in bold; α = 0.0083 (Bonferroni corrected α
= 0.05/6; N = 14 -16 galleries as noted).
Eggs total
Sub-adults total
Mature ♀♀ total
T
p (2-tailed)
N
T
p (2-tailed)
N
T
p (2-tailed)
N
Sub-adults
total
0.536
0.007
16
Mature
females total
0.000
1.000
16
0.199
0.295
16
Egg-layers
total
0.681
0.003
14
0.638
0.004
14
0.131
0.556
14
Discussion
The first mature female offspring in X. saxesenii galleries stayed and helped in the care of
fungi and brood for at least 17 days, which is longer than reported for Xyleborus affinis (7
days; ROEPER & al. 1980). Even after dispersal had started, females accumulated in the
gallery, which supports the hypothesis that females usually delay their dispersal. This was
already suggested by an earlier study, where we showed that delayed dispersal does not
serve the accumulation of reserves, but, on the contrary, reduces the fertility of X. affinis
females (BIEDERMANN & al. 2011). As these females are mature and potentially fully
capable of breeding independently, delayed dispersal must entail other fitness gains if it
was not maladaptive. Here we showed that, besides indirect fitness benefits to females
through cooperative behaviours raising the production of close kin (BIEDERMANN 2007),
some females also benefit directly by producing own offspring. The numbers of females
and egg layers did not correlate with each other, however, and on average about three
quarters of females were not laying eggs. Interestingly, both in the laboratory and field
daughters apparently did not reproduce before the onset of dispersal in the respective
galleries. Nevertheless, aggressive interactions, which would suggest competition over
reproduction, have never been observed.
Future studies need to explore the factors affecting dispersal, helping and breeding
decisions of female Xyleborini. Sociality in ambrosia beetles probably evolved in close
association with fungus agriculture in wood. Indeed, advanced sociality within Coleoptera
is only known from the Scolytinae, Platypodinae, and Passalidae, which are all wood-living
insects associated with microbial symbionts. Interestingly, it has been hypothesized that
sociality of ants and termites also evolved within the same habitat (HAMILTON 1978).
Acknowledgements
This manuscript benefitted greatly from comments of Tabea Turrini. PHWB is a recipient
of a DOC fellowship of the Austrian Academy of Sciences at the Department of
Behavioural Ecology, University of Bern, and was partly funded by a fellowship of the
Roche Research Foundation during this research.
30
Chapter 2
References
BATRA, S.W.T. (1966): Nests and social behaviour of halictine bees of India. - Indian Journal
of Entomology 28: 375-393.
BIEDERMANN, P.H.W. (2007): Social behaviour in sib mating fungus farmers.- Master thesis,
Zoological Institute, University Bern.
BIEDERMANN, P.H.W., KLEPZIG, K.D. & TABORSKY, M. (2009): Fungus cultivation by ambrosia
beetles: Behavior and laboratory breeding success in three xyleborine species. Environmental Entomology 38: 1096-1105.
BIEDERMANN, P.H.W., KLEPZIG, K.D. & TABORSKY, M. (2011): Costs of delayed dispersal and
alloparental care in the fungus-cultivating ambrosia beetle Xyleborus affinis
EICHHOFF (Scolytinae: Curculionidae). Behavioral Ecology & Sociobiology, in press.
BISCHOFF, L.L. (2004): The social structure of the haplodiploid bark beetle, Xylosandrus
germanus. - Diploma thesis, Zoological Institute, University Bern.
FISCHER, M. (1954): Untersuchungen über den kleinen Holzbohrer (Xyleborus saxeseni). Pflanzenschutzberichte 12: 137-180.
FRANCKE-GROSMANN, H. (1975): Zur epizoischen und endozoischen Übertragung der
symbiotischen Pilze des Ambrosiakäfers Xyleborus saxeseni (Coleoptera:
Scolitidae). - Entomologica Germanica 1: 279-292.
HAMILTON, W.D. (1978): Evolution and diversity under bark, pp. 154-175 In L.A. MOUND AND
N. WALOFF (eds.), Diversity of insect faunas. - Blackwell, Oxford.
KENT, D.S. & SIMPSON, J.A. (1992): Eusociality in the beetle Austroplatypus incompertus
(Coleoptera: Platypodidae). - Naturwissenschaften 79: 86-87.
PEER, K. & TABORSKY, M. (2007): Delayed dispersal as a potential route to cooperative
breeding in ambrosia beetles. - Behavioral Ecology & Sociobiology 61: 729-739.
ROEPER, R., TREEFUL, L.M., O'BRIEN, K.M., FOOTE, R.A. & BUNCE, M.A. (1980): Life history of
the ambrosia beetle Xyleborus affinis (Coleoptera: Scolytidae) from in vitro
culture. - Great Lakes Entomologist 13: 141-144.
SAUNDERS, J.L. & KNOKE, J.K. (1967): Diets for rearing the ambrosia beetle Xyleborus
ferrugineus (FABRICIUS ) in vitro. - Science 15: 463.
WILSON, E.O. (1971): The insect societies. - Belknap Press of Harvard University Press,
Cambridge.
31
Chapter 2
32
Chapter 3
Responses to experimental selection on dispersal in a primitively
eusocial beetle
Peter H.W. Biedermann1 and M. Taborsky1
1
Department of Behavioural Ecology, Institute of Ecology & Evolution, University of Berne,
Baltzerstrasse 6, CH-3012 Bern, Switzerland
Correspondance: [email protected]
Abstract
A fundamental goal of evolutionary biology is to explain the existence of non-reproductive helpers or
workers, i.e., individuals that invest resources in helping others to reproduce.
Benefits of philopatry and constraints on independent breeding are important forces for delayed
dispersal of offspring in many primitively eusocial groups. From an individual viewpoint, delayed
dispersal should be favored by natural selection, if inclusive fitness gains by remaining in the natal
group excel gains of independent breeding. This is likely under stable food conditions at home, high
costs of independent breeding and the possibility of indirect fitness gains through helping kin.
We used the primitively eusocial ambrosia beetle Xyleborinus saxesenii as a model system in an
artificial selection experiment on dispersal. In this species adult daughters show a high plasticity in
philopatry at the nest, but whether timing of dispersal is heritable and associated with certain other
traits is unknown. We therefore selected one line for early dispersal of daughters (dispersers) and the
other line for late dispersal and staying with the mother (cooperators). Despite that laboratory
breeding significantly delayed the timing of dispersal in both strains within six laboratory
generations, artificial selection on dispersal was proven successful because cooperators emerged
from the natal nest significantly later than dispersers. Most remarkably, foundresses in the
cooperative strain also produced higher numbers of immature and adult brood, and adult females
showed a higher overall activity and engagement in cooperative nest protection. In comparison to
the disperser strain, these investments at home, however, traded-off with a lower feeding-rate and a
reduced success when founding a nest independently. Hence, we hypothesize early and late dispersal
reflect two alternative social strategies, optimized either for selfish or cooperative breeding. In
ambrosia beetles both strategies are probably maintained because their individual inclusive fitness
advantage depends greatly on the stability of the nest. Ambrosia beetles settle dead wood, which
may be suitable for the production of multiple broods (favoring cooperators) or not even a single one
(favoring dispersers).
Introduction
A fundamental question in evolutionary biology is how natural selection has facilitated the evolution
of individuals that refrain from dispersal and own reproduction and instead help others to reproduce
(Darwin 1859; Wilson 1971; Wilson 1975). Current knowledge suggests that ecological constraints on
dispersal (Emlen 1982), benefits of philopatry (Stacey and Ligon 1991) and indirect fitness payoffs
(i.e. kin selection; Hamilton 1964) can facilitate the evolution of delayed dispersal, helping, and
reproductive altruism (Bourke 2011). The relative importance of these factors depends on the
biology and ecology of the species, as suggested by interspecific comparisons. Comparing
hymenopteran species at different sociality levels, for example, has shown that the transition to
eusociality, i.e. obligatorily sterile worker castes, has occurred solely in groups founded by lifetime
monogamous parents and their offspring (Hughes et al. 2008; Boomsma 2009). Ecological factors and
relatedness, however, cannot completely account for the variation in sociality among species, i.e., in
33
Chapter 3
the timing of dispersal, the frequency of helping behaviour and reproductive skew (Komdeur 2006;
Charmantier et al. 2007).
Among group plasticity in timing of dispersal and frequency to help are indications for underlying
genetic variance and/or phenotypic plasticity (e.g. Via and Lande 1985; Price et al. 2003). The latter is
expected to predominate under predictable environmental fluctuations, especially, if they occur
within a life-time of an organism (Bradshaw 1965; Levins 1968; but not only, see Scheiner 1993).
Conditional dispersal and cooperative brood care, for example, should provide a strong advantage in
variable environments, as long as habitat deterioration can be accurately predicted from
environmental cues (e.g. Ronce 2007). Nevertheless, for evolution of dispersal to occur genetic
differentiation is essential and it is expected to be relatively more important (i) in constant
environments if maintenance of plasticity is costly (Dewitt et al. 1998) and (ii) if the plastic response
for extreme phenotypes is incomplete (Price et al. 2003).
In most social species periods of social living are frequently offset by dispersal events followed by
solitary phases. The relative duration of these periods depends on the timing of an individuals’
dispersal from the natal group after reaching adulthood. Delayed dispersal and staying with the
mother will only be selected, if inclusive fitness benefits for an individual to remain in the natal group
outweigh inclusive fitness benefits of independent breeding. This is likely when constraints on
independent breeding or dispersal risks are high and individuals can gain direct or indirect benefits
by group living (Stacey and Ligon 1991; Heg et al. 2004). In many species where environmental
constraints on independent breeding are relatively stable or cannot be estimated prior to dispersal,
individuals may rely on available cues of local resource availability or presence of brood dependent
on care (Biedermann et al. 2011).
Dispersal is not only important for evading adverse environmental conditions, but it is also essential
for long-term survival of populations (Bullock et al. 2002). Dispersing individuals avoid costs of kin
competition, inbreeding, and habitat instability (Belichon et al. 1996; Gandon 1999; West et al.
2001). Thus, dispersal allows tracking patches of favorable habitat and escape from deteriorating
local conditions (Ronce 2007). On the other hand, dispersal is a costly behaviour: First, mortality
during dispersal is usually increased due to adverse environmental conditions and predation (e.g.
Heg et al. 2004; Ronce 2007). Second, energy allocated to dispersal capacity cannot be allocated to
other functions and therefore it commonly trades off with other life-history traits like delayed
maturation, reduced fecundity and shorter lifespan (e.g. migratory syndrome, see Dingle 1996; Roff
and Fairbairn 2001; 2007; Dingle 2006). Dispersal by flight in insects, for example, is energetically
expensive and a substantial amount of resources is allocated to the flight muscles (e.g. Kammer and
Heinrich 1978; Dudley 2000). Hence, many insects, histolyse their flight muscles during reproduction
(Johnson 1969; Roff 1989; Zera and Denno 1997; for Scolytinae, see Robertson 1998; Lopez-Guillen
et al. 2011). A suite of life-history traits co-evolve with dispersal (Dingle 1996) and are thus expected
to be genetically correlated, either by pleiotropy or by linkage disequilibrium (e.g. Fuller et al. 2005).
Nevertheless, while dispersing and philopatric individuals exhibit different sets of life-history traits,
they may still achieve the same lifetime reproductive success (see review in Belichon et al. 1996).
In cooperatively breeding animals, dispersal is associated with a solitary settlement of new habitat,
whereas non-dispersal (staying) is associated with a social life-style and cooperative brood care.
Although not necessarily linked, the decision to stay is associated with the decision to help in many
social species (e.g. Choe and Crespi 1997; Cockburn 1998). Several studies have shown that many
fitness-related traits, like fecundity, are genetically correlated with dispersal (reviewed in Roff and
Fairbairn 2001). Therefore, apart from morphological changes that go together with dispersal and
non-dispersal (and a social versus a solitary life-style), other life history, behavioural and
physiological traits should be correlated (Roff and Fairbairn 2001; Dingle 2006). (i) Dispersers are
expected to have better abilities to colonize new habitat than staying individuals (when latter are
forced to disperse). (ii) Fecundity is expected to be reduced in dispersers relative to stayers and the
difference will depend on the energetic costs of adaptations for dispersal. Despite a short-term
fitness disadvantage, dispersal is still preserved in a population because of its long-term fitness
benefit by the colonization of new habitats. (iii) In cooperatively breeding groups, individuals with a
higher dispersal propensity than average should be less inclined to engage in cooperative brood care
and sustainable resource use. Among group variability for the propensity to leave and help is found
in many social species. Underlying genetic differences are expected because such crucial life history
34
Chapter 3
traits should be subject to selection. Parental effort and helping behaviour, for instance, have been
found to be heritable in cooperatively breeding birds (Maccoll and Hatchwell 2003; Charmantier et
al. 2007). In social insects, genetic differences for helping behaviour are indicated by recent studies
that have identified genes that influence social behaviour (reviewed in Robinson et al. 2008; Keller
2009). However, hitherto we lack information about the evolvability of dispersal and its correlated
traits in social taxa, although this is essential for understanding group formation as the first step in
social evolution.
Scolytine beetles are an ideal model system for the study of the evolution of dispersal in combination
with sociality. Many species may switch between a social and a solitary life within a life-time, which
seems to depend largely on ecological conditions (Kirkendall et al. 1997). Sociality is particularly
facilitated in the fungus farming tribe Xyleborini. They live in haplodiploid, inbred kin-groups in
isolated galleries, which increases relatedness within natal colonies. Furthermore, colony members
benefit greatly from cooperation due to their dependence on fungiculture in dead wood, used as a
resource for shelter and substrate for fungi that is virtually non-depreciable in the short-term
(Biedermann and Taborsky 2011). In the long-run, however, individuals need to maintain their
dispersal abilities if degradation of the dwelling forces them to leave and start a solitary life. In the
best studied species Xyleborinus saxesenii, adult offspring of a foundress usually delay dispersal from
the natal nest to benefit from helping close kin and partly from own reproduction (Biedermann and
Taborsky 2011; Biedermann et al. 2012). Constraints on independent breeding in this primitively
eusocial species appear to be very high, as a result of high costs of dispersal (for another scolytine
beetle see Garraway and Freeman 1981) due to the difficulties of (i) finding a suitable host tree, (ii)
nest foundation, and (iii) a successful start of fungiculture (Kirkendall et al. 1997; Biedermann et al.
2009). Nevertheless, duration of philopatric periods of adult daughters within a nest appear highly
variable (also under optimal conditions), ranging from at least twelve days after maturation to a
whole life (> 60 days). During these times daughters seem to invariably engage in helping behaviour
at the natal nest (Biedermann and Taborsky 2011), and up to a quarter of them also lay eggs
(Biedermann et al. 2012). Despite apparently high levels of inbreeding and genetic homogeneity
within galleries, this might suggest genetic variance within these family groups underlying the timing
of dispersal and the tendency to engage in cooperative behaviours.
Here we test the evolvability of dispersal and its associated traits and trade-offs in an artificial
selection experiment with a laboratory population of X. saxesenii. If dispersal is heritable, selecting
one line of daughters for early dispersal (dispersers) and one line for late dispersal (cooperators),
which is a correlate of cooperation, is expected to reveal significant differences in the mean timing of
offspring dispersal between both groups after some generations. Furthermore, early dispersal,
relative to late dispersal, may trade-off with fecundity and be associated with traits like reduced
engagement in helping behaviours at the natal nest, for example. On the other hand, dispersers may
be better adapted to transfer their fungi to new galleries and found nests independently.
Material & Methods
Study system and general laboratory breeding
The fruit-tree pinhole borer (X. saxesenii; Scolytinae) is a temperate species with a world-wide
distribution. It breeds in tunnel systems (galleries), excavated in recently dead wood of a wide variety
of host trees. New galleries are always founded by individual females transmitting spores of a
complex of fungi, dominated by the species-specific ambrosia fungus Raffaelea sulphurea (formerly
Ambrosiella sulfurea; Ascomycota), in specialised organs (mycetangia or gut) from the natal to the
new gallery (Biedermann et al. 2012). Broods are produced after a thin fungal layer of fruiting cells
has established on the surface of gallery walls. Successful planting of fungus often fails, however,
either because no spores of the ambrosia fungus have been transmitted or the substrate is not
suitable for the spores to germinate (Biedermann et al. 2009). If planting fails the female sometimes
disperses again, but mostly dies. Successful females may, however, produce large broods and lay
eggs for several weeks in common chambers excavated by the larval feeding activity (on fungus
infested wood; (Biedermann and Taborsky 2011). In the laboratory (25° C) first daughters will reach
adulthood about 30 to 40 days after gallery foundation, will be inseminated by a brother and will
remain with their sibs in the following for a period of at least two, sometimes many more weeks. If
35
Chapter 3
the woody substrate is durable, a few adult daughters seem to overtake the mother’s gallery and
proceed breeding after her death (in the laboratory on average 70 to 80 days after gallery
foundation; (Biedermann et al. 2012). If successful, field and laboratory galleries are comparable in
size producing approximately 10-50 dispersing female offspring (Biedermann et al. 2009); galleries
with several hundred dispersers are possible in the field, but exceptional (Hosking 1972). The small,
wingless males are in the minority (the average sex ratio is approximately 1:8 to 1:20; (Biedermann
et al. 2009). They disperse if no more offspring is produced, but can only wander the natal log and
may introduce in neighbouring galleries there (Biedermann 2010).
We bred consecutive laboratory generations of X. saxesenii in transparent plastic tubes filled up to
two-thirds with artificial medium that mainly contained agar and sawdust (“test medium” described
in Biedermann et al. 2009). One founder female was used per tube and is sat onto the medium after
3 sec rinses in bleach, 90% ethanol and distilled water, to sterilize their body surface from weed
fungi. Afterwards tubes are closed by plastic caps and wrapped in paper to keep them dark as if in
wood, but light could enter through the tops of the tubes. Tubes are kept in a constant light/dark
cycle (13 h light/11 h dark) at 28 °C/22 °C and the paper wrap is only removed during observations
under a microscope (×6.4 to ×16 magnification) with an artificial light source (maximum 6 W). Beetles
excavate their tunnels largely along the transparent tube wall, which enables observations of their
behaviours. Experiments showed that behaviour of scolytine beetles within their galleries is affected
neither by light nor by changing the axis of gravity while doing observation (Schmitz 1972;
Biedermann and Taborsky 2011). Dispersal is through the entrance tunnel and can be recorded by
collection of these individuals from the top of the medium.
Pilot studies suggested that periods of diapause are essential for successful long-term breeding of X.
saxesenii within the laboratory (see also data on an obligate diapause in Xylosandrus germanus;
Weber and McPherson 1983). Therefore and because of logistical reasons (i.e. periods when we were
not able to record dispersal every day) we reduced breeding temperature during the F1 to F5
generation once for 8 weeks to 8°C and several times at irregular intervals for 2-3 weeks to 16°C. Egglaying probably ceased during these periods, because overwintering field galleries of X. saxesenii
contain only larvae, pupae and adults (Fischer 1954; unpubl. data). Beetles do not disperse below
18°C (Faccoli and Rukalski 2004); unpubl. data).
Recording of dispersal and artificial selection
Our laboratory population stemmed from females collected in November 2009 from overwintering
galleries in beech stumps (Fagus sylvatica) in the Spil-forest (560 m asl, 46°95’, 7°31’) close to Berne,
Switzerland. We started our selection experiment with its second laboratory generation, making up
30 successfully established galleries (control group, starting generation F0). Every 1-2 days we
collected all dispersing adult females that had emerged onto the medium with insect handling
tweezers, transferred them to sterilized micro tubes with a moistened piece of filter paper and
afterwards stored them in the refrigerator until they were either disposed or used for further
breeding. In each generation we aimed to select for (i) early dispersal (= disperser strain) and (ii)
remaining within the nest (= cooperator strain). In the disperser strain we used about the first third of
all emerging daughters per gallery (starting generation: x = 10.6 females, range 0-35) as founder
females for the next generations. The cooperator strain was made up of the last daughters remaining
in the natal nest (starting generation: x = 5.0 females per gallery, range 0-15). Remaining daughters
were collected by gallery dissections close to the end of gallery life. In cases when galleries were
already empty upon dissection, we continued breeding with the last dispersing females from the
respective gallery that we had stored. After we had recorded the dispersal of the F6 generation, we
used a Cox regression model to detect potential effects of long-term laboratory breeding (comparing
F0 with F6) and artificial selection (cooperators vs. dispersers in F6) on the dispersal rate of
daughters, by controlling for gallery identity and weather parameters for Berne (Source:
MeteoSchweiz).
A way to increase the incentive of adult daughters to stay within a natal nest and to continue brood
care and breeding in the cooperator strain was to extend gallery durability with fresh substrate for
the fungus: First, tubes were cut apart 2-3mm below the chamber. Subsequently, a second tube filled
36
Chapter 3
with freshly prepared, sterile and not yet solidified medium was attached to the first tube and tightly
fixed (Fig. 1). We extended tubes about 70 days after gallery foundation and in a few cases a second
time about 50 days later. Only about one third of tubes had visible chambers and could be extended
this way, however.
Fig. 1. Illustration of the method to increase the longevity of galleries in the cooperators strain.
Behavioural recordings
Behaviours were recorded by scan observations during which all gallery parts were browsed one time
for inhabitants: we noted the gallery identity and counted the number of eggs and pupae, the
number of visible larvae and adults and the respective behaviours they showed at that moment. In
our analysis we did not discriminate between the three larval instars. Pupae and adult beetles were
sexed on the basis of morphology and size. We discerned teneral adult females that had recently
hatched and showed weak sclerotization and brownish elytra, and mature adult females that were
fully sclerotized with dark brown to black elytra. From each scan observation we noted the
proportion of individuals per class (larvae, teneral females, mature females, males) performing a
respective behaviour. We decided between (i) the larval behaviours: digging (= feeding), shuffling,
balling, allogrooming, cannibalism, inactivity and locomotion, (ii) the adult female behaviours:
blocking, shuffling, allogrooming, cannibalism, digging, cropping fungus (= feeding), inactivity and
locomotion, and (ii) the adult male behaviours: allogrooming, cannibalism, cropping fungus, mating
attempt, copula, inactivity and locomotion (for detail descriptions of these behaviours see
(Biedermann and Taborsky 2011)). In the starting generation (F0) we conducted nine scan
observations of all 30 galleries on random days 35-80 days after gallery foundation. The selection
lines in the F6 generation we scanned on 35 random days 12-80 days after gallery foundation. We
intermixed strains and allocated random numbers to the 33 galleries of the cooperators strain and to
the 67 galleries of the disperser strain to be blind on their identity before conducting behavioural
observation. We used a series of GEE models to detect potential effects of long-term laboratory
breeding (comparing F0 with F6) and artificial selection (cooperators vs. dispersers in F6), by
controlling for gallery identity, gallery age and numbers of other immatures and adults present at the
respective scan.
Influence of artificial selection on other parameters
Apart from dispersal and behaviours we also collected data on gallery composition, enzyme profiles
of the fungus gardens, sex ratios and gallery founding success. We dissected galleries in the F6
generation on day 45, 62 and 87 after gallery foundation. All eggs, larvae, pupae, males, teneral and
adult females were counted and then stored in 70% ethanol. Numbers of dispersed daughters until
this date were also recorded. One piece of fungus garden material from the brood chamber was
removed and used for enzyme profiling (see also (De Fine Licht and Biedermann 2012)). We used a
series of GLM models to detect potential effects of artificial selection (cooperators vs. dispersers) on
gallery composition and the ratio between dispersed and remaining daughters.
Final sex ratios of all galleries in the F0 and F6 generation were recorded and analysed whether they
are affected by laboratory breeding and artificial selection using a GEE model that controlled for
gallery of origin in the F0 generation. Furthermore, we recorded the founding success, i.e. the ability
of a founder female to successfully produce at least one adult offspring, of the F1 and the F6
generation in our selection lines. Afterwards we used a GEE model to test the influence of long-term
37
Chapter 3
laboratory breeding (comparing F1 with F6) and artificial selection (cooperators vs. dispersers in F6)
on the binomial success rates, by controlling for gallery identity.
Statistics
We used a Cox regression model to detect potential effects of long-term laboratory breeding
(comparing F0 with F6) and artificial selection (cooperators vs. dispersers in F6) on the dispersal rate
of daughters, by controlling for gallery identity and weather parameters. Furthermore, we analysed
several other life-history variables to detect correlative effects and trade-offs in response to our
artificial selection. We used a series of GEEs (lmer in R; (R Development Core Team 2008), which are
an extension of generalized linear models with an exchangeable correlation structure of the response
variable within a cluster (= gallery identity or origin), to analyse effects of dependent variables on
correlated binary response variables (proportional data were transformed to binary data) and to
identify effects of laboratory breeding and artificial selection (1) on each of the relative behavioural
frequencies, (2) on gallery founding success, and (3) on sex ratios. Furthermore, we used a series of
GLM models with quasi-binomial error distribution to detect potential effects of artificial selection
(cooperators vs. dispersers) on gallery composition and the ratio between dispersed and remaining
daughters. All statistical analyses were performed with SPSS version 15.0 and R version 2.8.1 (R
Development Core Team 2008).
Results
Dispersal of adult daughters from the natal nest
Our selective treatment for delayed dispersal of adult daughters (= cooperator strain) increased the
likelihood of female dispersal within six generations by a factor of 1.96 in relation to the females
selected for early dispersal (= disperser strain; Cox model: p = 0.005; Table 1 and Fig. 1). Half of the
adult daughters had been dispersed at day 64 ( x , CI 63 – 64) after gallery foundation in the disperser
strain, which is five days before it occurred in the cooperator strain ( x = 69, CI 68 – 70). Compared to
the dispersal of the starting generation (= control; x = 59, CI 59 – 61), however, laboratory breeding
overall resulted in a later dispersal of adult daughters (control vs. disperser strain: odds-ratio = 0.48,
p = 0.001; control vs. cooperator strain: odds-ratio = 0.24, p < 0.001). In general, dispersal tended to
occur on certain days when barometric pressure had risen during the last 24 hours (p = 0.1).
Table 1. Results of the Cox regression model for the dispersal of adult daughters.
1
2
Cox Model
Control (N = 226) vs. Cooperators (N = 336)
Control (N = 226) vs. Dispersers (N = 804)
Cooperators (N = 336) vs. Dispersers (N = 804)
5
Fall vs. Rise of barometric pressure
b
-1.41
-0.74
0.67
0.32
2
3
S.E.
0.1
0.08
0.09
0.07
z
-5.29
-3.28
2.8
1.63
p
<0.001
0.001
0.005
0.103
Wald test: χ = 28.3, df = 2, p = <0.001
1
Formula for the final Cox Model in R: coxph(Surv(galleryage, dispersal) ~ Treatment + cluster(Gallerynumber), data).
2
b = regression coefficient of dispersal function for variable.
3
Standard error of regression coefficient.
4
Odds ratio (= exp(b)).
5
Variable excluded from the final model.
38
4
Odds
0.24
0.48
1.96
1.38
Chapter 3
Figure 1. Philopatry of adult females within the natal nest under two selection regimes. The black
dotted line refers to the dispersal of the Control galleries before the start of the selection.
Cooperators refer to the 6th generation of the selected galleries that were bred for cooperation and
late dispersal, respectively. Dispersers refer to the 6th generation of galleries that were selected for
early dispersal. Cox regression analysis showed that in both Cooperators (odds ratio = 0.24; z = -5.29,
p = <0.001) and Dispersers (odds ratio = 0.48; z = -3.28, p = 0.001) adult females delayed dispersal
after six generations of laboratory breeding relative to the females in the Control galleries in the first
generation. Our selective treatment increased the likelihood of philopatry in the Cooperators by a
factor of 1.96 relative to the Dispersers (z = 2.8, p = 0.005).
Behavioural changes in response to the selection regime
Selecting for early and late dispersal of daughters had strong effects on their behaviours within the
natal nests. As expected adult females of the cooperator strain were more active (GEE: p = 0.001 –
0.052; Table 2, Table S1) and were more often engaged in cooperative behaviours than the adult
females in the disperser strain (p = 0.005) and the control group(p < 0.001), which was mainly an
effect of an increased blocking frequency (p = 0.003 – 0.004; see Table S1). Cropping of the fungus
was less commonly observed in the cooperator strain in relation to the dispersers strain (p < 0.001)
and to the control (p < 0.001). Consecutive laboratory breeding led to an overall increase of adult
female activity (= decrease of inactivity; p = 0.001 – 0.18), of cannibalism (p = 0.024 – 0.043) and
locomotion (p < 0.001), but reduced their engagement in cropping of the fungus (p < 0.001).
Behaviours of larval stages were not differently affected by the two selection regimes. In relation to
the control group, however, larvae in both lines decreased their activity (GEE: p = 0.002 – 0.011;
Table S2) , the frequency of locomotion (p < 0.001), and tended to engage less often in allogrooming
(p = 0.028 – 0.18). Male behaviours were unaffected by laboratory breeding and the selection
regimes, except for allogrooming that was shown less often in the dispersers than the controls (p =
0.041; Table S3).
39
Chapter 3
Gallery composition in dependence of the selection regime
The first offspring normally emerges from the pupal stage at a gallery-age between 30 and 40 days.
In our study, significant dispersal activity started about 14 days later (see Fig. 1). Eleven cooperator
and disperser galleries were dissected on day 45. Offspring composition (i.e. number of eggs, larvae,
pupae, males, teneral and adult females) did not differ between the selection regimes at this gallery
stage (GLMs: p < 0.05, Table S4). Adult female dispersal had only started in one exceptional gallery of
the cooperator strain. Around the time when half of the daughters had dispersed from their natal
gallery, at a gallery age of 62 days, we found offspring composition to differ markedly between the
selected strains: Cooperator galleries (N = 11 galleries) contained on average a higher number of
immature offspring (p < 0.001) and adult daughters (p < 0.001) than disperser galleries (N = 40).
Latter is partly caused by the smaller number of dispersed daughters until this date in the cooperator
strain (p = 0.039). Close to the end of gallery life when egg-laying had ceased, at a gallery-age of 87
days, however, the small numbers of immatures and adult females left within the galleries did not
differ between cooperator (N = 13) and disperser strain ( N = 5). The small average productivity of the
disperser galleries ( x = 8.8 ind.) may explain the higher absolute and relative numbers of daughters
that had dispersed from the cooperator galleries until this date.
Table 2. Separate GEE models to examine differences (p < 0.05) between the proportion of time adult
females spent with the observed behaviours.
Behaviour
Parameter
Coeff.  SE
z
p
Adult ♀♀ activity
Intercept (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery
2.91  0.44
0.92  0.29
0.33  0.25
-0.58  0.3
-0.02  0.01
6.63
3.13
1.33
-1.94
-3.08
<0.001
0.001
0.18
0.052
0.002
Intercept (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery
Number of larvae
1.7  0.49
1.35  0.34
0.5  0.31
-0.85  0.3
-0.08  0.01
0.06  0.01
3.45
4
1.6
-2.83
-9.41
4.78
<0.001
<0.001
0.11
0.005
<0.001
<0.001
Intercept (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery
Number of larvae
-0.66  0.27
-1.28  0.16
-0.67  0.14
0.61  0.16
0.02  0.01
-0.02  0.01
-2.45
-8.06
-4.64
3.92
3.98
-2.15
0.014
<0.001
<0.001
<0.001
<0.001
0.031
Adult
♀♀
cooperative
behaviours (blocking, shuffling,
allogrooming and cannibalism)
Adult ♀♀ cropping fungus
Direction
+++
(-)
-+++
---+++
----+++
+++
-
We used GEEs with an exchangeable correlation structure of the response variable within a cluster (= gallery) to identify
effects of the selection regime on the total amount of time each behaviour was observed, by controlling for influences of
gallery-age and number of larvae (for the full model see Table S3). The influences of the regimes on the behavioural
frequencies are displayed as contrasts between classes.
40
Chapter 3
Figure 2. Composition of galleries in the two selected strains at an age of 62 days after gallery foundation.
Mean (± standard errors) numbers, for the cooperator and disperser galleries are given. GLM: *** - p < 0.001, *
- p < 0.05 (for exact values see Table S4).
Gallery founding success
At the beginning of our selection experiment we measured the founding success (i.e. the ability to
produce at least one adult daughter) of the first 269 dispersing daughters (1st generation of disperser
strain) and the 115 last dispersing / remaining daughters (1st generation of cooperator strain) from
the same galleries. Founding success did not differ between the strains in the first generation and
also compared to the daughters in the cooperator strain after six generations (GEE: p > 0.05; Fig. 3,
Table S5). Early dispersing daughters in the disperser strain, however, markedly improved their
founding success until the sixth generation (p < 0.001).
41
Chapter 3
Figure 3. Rate of successful brood establishment by daughters, when reared independently after the first and
the sixth laboratory generation and depending on the selection regime. Different letters denote significant
differences in the success rates (GEEs: p < 0.05; for details see Table S4).
Discussion
The evolution and maintenance of sociality can only be understood by deciphering the interaction
between genetics and environment in shaping a complex phenotype (Komdeur 2006; Charmantier et
al. 2007). Our study is the first to show the response of dispersal to selection in a social taxon. It
helps to understand how social phenotypes are selected over non-social phenotypes when sociality is
favoured by specific environmental conditions. Adult female X. saxesenii selected for early dispersal
(and hence reduced sociality) exhibited smaller offspring numbers and a shorter lifespan compared
to those selected for late dispersal. In the latter, however, only 32% of breeding attempts were
successful, compared to 69% in early dispersers. Under natural conditions, however, the higher
success in nest foundation in early dispersers is likely offset by predation and adverse environmental
conditions (for a scolytine beetle see Garraway and Freeman 1981), potentially resulting in similar
reproductive outputs.
After six generations of artificial selection on early and late dispersal, we found daughters of the
early disperser strain to leave on average five days before those of the philopatric strain. Such a shift
was not detected, however, when we compared dispersal between each parental gallery at the
beginning of the experiment with dispersal in its daughter-galleries at the end. Hence, differences
between the selected strains after six generations are not a result of among-gallery selection on early
and late dispersers, but these arise because of between-gallery differences in timing of dispersal.
Although we did not particularly design our experiment for selection on between-gallery differences,
this result is not surprising. Ambrosia beetles in the tribe Xyleborini are highly inbred and thus withinfamily genetic variability is very low and offering no possibility for selection. The only genetic study
on a xyleborine beetle reported an inbreeding coefficient of F = 0.88 in Xylosandrus germanus, which
suggests that only 3% of all matings are with non-sibs, and individuals within a gallery are almost
clones (Keller et al. 2011).
The selectability of the timing of dispersal in X. saxesenii indicates that this behaviour can respond to
environmental factors. A crucial environmental factor for ambrosia beetle sociality is the
unpredictability of the habitat they settle. Constraints on independent breeding are always very high
in these beetles because dying trees are patchily distributed (and thus difficult to locate) and the
plantation of ambrosia gardens often fails. This has apparently led to females in temperate
xyleborine ambrosia beetles to ubiquitously remain and help in the natal nest for a few days after
reaching adulthood. Dead wood offers a habitat that is ephemeral, but for awhile highly productive.
Depending on the part of the tree, which is settled, galleries may be productive for one or several
beetle generations, which selects for simpler or more complex sociality, respectively. Although X.
42
Chapter 3
saxesenii prefers a relatively long-lived and stable woody substrate (hardwood of large diameter
trees in an early stage of the degradation process; Pfeffer 1995), host quality can deteriorate quickly,
e.g. due to desiccation or competition with other xylophagous organisms, forcing beetles to disperse.
Such an environment with unpredictable, but often extensive durability selects for the maintenance
of both early and late dispersers, since the constraints on independent breeding are high. Ambrosia
beetle females dispersing early have a better chance of locating new hosts and may, therefore, be
favoured by selection late dispersers. In a large-diameter tree stump that is able to nourish multiple
generations of beetles, on the other hand, early dispersal would probably have higher fitness costs,
as the likelihood for not finding a better resource is high. Indeed, current knowledge suggests that
primitive eusociality (=cooperative breeding) in ambrosia beetles is only found under exactly this
condition (e.g. X. saxesenii and X. affinis; Biedermann et al. 2011; 2012; Biedermann and Taborsky
2011). Sub-sociality, on the other hand, appears to be present in species predominantly attacking
short-lived resources (e.g. Xylosandrus germanus; Bischoff 2004), like small diameter branches,
whereas true eusociality has evolved only within living trees, which are a long-lasting and stable
habitat (in Austroplatypus incompertus; Kent and Simpson 1992).
The cooperators strain produced females that lived longer and produced more adult offspring
compared to the dispersers strain. This resembles the syndrome of mobile and sedentary
phenotypes in other insects living on ephemeral resources (e.g. fruit-feeding moths; Gu and
Danthanarayana 1992; Gu et al. 2006; Torriani et al. 2010). A major difference to these solitary
species is, however, that X. saxesenii females may additional gain indirect fitness benefits by staying
and helping in the natal nest. Our recent studies on cooperatively breeding xyleborine ambrosia
beetles (X. affinis and X. saxesenii) showed that adult offspring staying in the natal group invariably
engage in the following cooperative behaviours: (i) gallery protection by blocking, (ii) brood care by
allogrooming, (iii) gallery hygiene by shuffling faeces and sawdust (Biedermann and Taborsky 2011),
and (iv) cropping the fungal layers, which may serve the removal of fungal weeds (Norris 1979).
Furthermore, (v) the mere presence of adult females (and also immatures) profits fungus growth,
most likely through their excretions (Anisandrus dispar; Batra and Michie 1963; French and Roeper
1972). In summary, these behaviours all help to raise closely related kin and are expressed depending
on current needs in X. saxesenii; dispersal and help are partly triggered by the amount of brood
depending on care and the number of other helpers (Biedermann and Taborsky 2011).
Female activity tended to be higher (p = 0.052) in the cooperator than in the disperser strain.
Additionally, we found cooperators to engage slightly more frequently in the combined cooperative
behaviours blocking, shuffling, allogrooming, and cannibalism. This resulted solely from a significant
increase in blocking frequency, however. The magnitude of received help is much higher in
cooperators if the total help is examined: Many more females stayed and helped, for several days
more, within the natal nest than in dispersers. This may be one reason why the females in the
cooperator strain produced more offspring than in the disperser strain in the sixth generation. Only
one of the behaviours that are regarded mutually beneficial – fungus cropping – was more commonly
expressed by the females in the disperser strain. There are two possible explanations: Females
following the non-social strategy (i) might have a higher food requirement, because they need to
build up an energy-expensive flight apparatus, and (ii) they should have less of an interest in longterm management of their fungus cultures compared to cooperators, and thus possibly “overgrazed”
their gardens. It will be interesting to compare the growth and quality (i.e. fungal species diversity) of
the fungus gardens between the strains in a future study. Genetic differences have been repeatedly
shown to be responsible for such distinct behavioural strategies (e.g. in birds; Sih et al. 2004) and
may also underlie our findings. Also adult males exhibited significantly more allogrooming in the
cooperative strain, which suggests that their behavioural changes were also triggered by genetic or
non-genetic effects correlating with our selected traits.
Here we show a heritable component of dispersal (and sociality), which means that each individual
can adjust its behaviour only within a certain range. There are only two other studies in cooperatively
breeding birds that indicate genetic differences to partly underlie individual variation in social
behaviours (i.e. parental care; Maccoll and Hatchwell 2003; Charmantier et al. 2007). Besides genetic
inheritance, however, there are other modes of non-genetic inheritance. (a) The phenotype of an
organism can be determined not only by its genotype and the environment the organism
43
Chapter 3
experiences, but also by the environment and phenotype of its mother. This may be due to mothers
supplying eggs with mRNA or proteins. Insect’s phenotypes have been frequently reported to be
affected by such maternal effects (e.g. Mousseau and Dingle 1991; Li and Margolies 1994) and they
may also play a role in X. saxesenii. (b) Alternatively, in insects living in symbioses with
microorganisms like ambrosia beetles, the symbionts might directly shape behaviours and lifehistories of their hosts. If xyleborine ambrosia beetles found new galleries, they always transfer
spores of the fungal symbionts from the natal nest in their mycetangia (vertical transmission;
Francke-Grosmann 1956). Additionally, a few spores of commensal and parasitic fungi get usually
vectored on the body surface of beetles and this fungi increase in abundance during aging of the
gallery (e.g. Francke-Grosmann 1967; Beaver 1989; Kajimura and Hijii 1992). Hence, a community of
microbes is continuously associated and vertically transferred from one generation to the next,
probably without much horizontal (between galleries) exchange. “Inheritance” of a microbial
community could give rise to repeated patterns of behavioural responses (e.g. dispersal, cooperative
behaviours) and life-history traits (e.g. fecundity, life-span) within families (cf. comparable to
territory inheritance in cooperatively breeding birds; Komdeur 1992; Charmantier et al. 2007). Crossfostering experiments (exchanging offspring or fungal communities) should be used in the future to
reveal the potential role of non-genetic modes of inheritance.
Despite a heritable component, there is also a huge within gallery variation in the timing of dispersal.
The most likely mechanism underlying this variation is variability in individual response thresholds to
environmental conditions. In this study, for example, we found evidence for increased dispersal after
a rise in barometric pressure. Barometric pressure is closely associated with weather conditions and
serves as a dispersal clue for many small insects (e.g. Zettel 1984; Li and Margolies 1994). Also, there
are indications that the quality of the fungal resource, i.e., fungal species composition, affects
dispersal (Kajimura and Hijii 1992). Furthermore, in another study we found that dispersal correlated
with number of gallery inhabitants. Larvae, which depend on care, lead to delayed dispersal, whereas
other helpers promoted dispersal (Biedermann and Taborsky 2011). An alternative explanation for
variable dispersal is enforced philopatry through maternal manipulation. Maternal blocking of the
entrance tunnel of the gallery has been shown to mainly serve (a) the protection of the gallery from
intruders (e.g. predators, parasites, foreign ambrosia beetles), (b) the regulation of the microclimate,
(c) preventing larvae from falling out of the gallery, and (d) possibly preventing helping daughters
from dispersal (for details see Kirkendall et al. 1997; Biedermann and Taborsky 2011). Recently we
found adult daughters to disperse at higher rates after the removal of the blocking female
(Biedermann and Taborsky 2011). Blocking females are usually egg-layers and they might profit from
the presence of the helpers (Biedermann and Taborsky 2011). In this study we found adult female
blocking behaviour to significantly respond to our selective treatment. After six generations it was
exhibited more frequently in the cooperators than the dispersers strain. Hence, we might have
indirectly selected for high and low blocking frequencies that consequently have led to females
dispersing late or early, respectively. The possible role of blocking in enforcing females to help needs
to be explored in future studies.
Foundresses produced more offspring in the cooperators strain than in the dispersers strain. First,
this may be a result of higher adult female activity and frequency of helping, which may have pushed
fungus productivity (see above). Adult female helping behaviours are allocated to brood numbers
(Biedermann and Taborsky 2011). Differences in the amount of help between strains, however, are
not an artefact of the higher offspring numbers in the cooperators, because we controlled for brood
numbers in our model when we compared the frequency of behaviours between the two strains.
Second, foundresses may have adopted different energy allocation strategies: They may have
invested either in helping and reproduction, or in dispersal. Dissections of X. saxesenii females
revealed that dispersing females always have non-mature ovaries (Biedermann et al. 2012). For
females in the disperser strain, on the other hand, a more efficient flight-machinery can be assumed.
Consequently, they face a dispersal-reproduction trade-off, which is typically found in insects (Gu and
Danthanarayana 1992; Zera and Denno 1997; Gu et al. 2006). Although scolytine beetles can
apparently fully histolyse their wing muscles (e.g. Robertson 1998), this may consume energy that is
later missing for reproduction. Furthermore, preparation for dispersal may be associated with a
higher metabolic turnover (e.g. Hanski et al. 2006), which is generally known to constrain energy
reserves and survival (e.g. Van Voorhies 2001). Still it cannot be excluded that females preparing for
44
Chapter 3
dispersal may adopt a risk-spreading strategy (Kisdi 2002) and lay some eggs before departure from
the natal habitat (e.g. Hughes and Dorn 2002; Biedermann et al. 2011), and then produce less
offspring after founding a new nest.
Independent of the artificial selection on timing of dispersal we found consecutive laboratory rearing
to generally lead to later dispersal and behavioural changes. This suggests adaptations to laboratory
conditions. Artificial medium is a nutrient-rich and relatively long-lived breeding substrate, and this
apparently shifted dispersal in both strains over time. Furthermore, it led to larvae being less active,
walking less and engaging less in allogrooming. Adult female activity, locomotion and cannibalism
increased, whereas cropping fungus strongly decreased. The latter, together with the reduced larval
allogrooming, indicates that infestation with fungal parasites decreased during consecutive
laboratory rearing. In summary, laboratory conditions seem to favour the cooperative and latedispersing strategy, as both strains were selected for later dispersal, independent of the treatment.
This is probably a result of increased gallery productivity and longevity. High quality food conditions
have been repeatedly shown to strongly select against a dispersers phenotype in arthropods that live
on ephemeral resources (e.g. Li and Margolies 1994; Gu et al. 2006).
Many ambrosia beetles have recently developed into invasive pests of forests and fruit-tree
plantations (Hulcr and Dunn 2011). Invasiveness is often correlated with a high dispersal ability (e.g.
Sakai et al. 2001).Our study suggests that populations might differ in their dispersal capability and
probably several other correlated life-history traits. It would be interesting to compare invasive and
natal populations with regard to their dispersal and life-history traits. Deciphering the underlying
heritable differences may greatly help for understanding the evolution of high dispersal abilities and
invasiveness and the other extreme philopatry and social evolution. Clearly, ecological conditions
play a major role. In unpredictable environments, populations are expected to exhibit
polymorphisms for dispersal traits (Southwood 1977). A genetic basis for this polymorphism is well
documented in many arthropod populations (Harrison 1980; Roff and Fairbairn 2001; 2007). This
suggests that variation between individuals in the propensity to stay, help and breed in the natal nest
does not arise solely from individuals maximizing inclusive fitness, but are partly heritable.
Heritability found in this study is a prerequisite for evolution of these traits.
Literature
Batra LR & Michie MD (1963) Pleomorphism in some ambrosia and related fungi. Transactions of the Kansas Academy of
Science 66: 470-481.
Beaver RA (1989) Insect-fungus relationships in the bark and ambrosia beetles. Insect-fungus interactions (Wilding N, Collins
NM, Hammond PM & Webber JF, eds), pp. 121-143. Academic Press, London.
Belichon S, Clobert J & Massot M (1996) Are there differences in fitness components between philopatric and dispersing
individuals? Acta Oecologica-International Journal of Ecology 17: 503-517.
Biedermann PHW (2010) Observations on sex ratio and behavior of males in Xyleborinus saxesenii Ratzeburg (Scolytinae,
Coleoptera). Zookeys 56: 253-267.
Biedermann PHW, Klepzig KD, Ott E, Taborsky M & Six DL (2012) Dynamics of filamentous fungi in the ambrosia gardens of
the primitively eusocial beetle Xyleborinus saxesenii Ratzeburg (Scolytinae: Curculionidae). in prep.
Biedermann PHW, Klepzig KD & Taborsky M (2009) Fungus cultivation by ambrosia beetles: behavior and laboratory
breeding success in three xyleborine species. Environmental Entomology 38: 1096-1105.
Biedermann PHW, Klepzig KD & Taborsky M (2011) Costs of delayed dispersal and alloparental care in the fungus-cultivating
ambrosia beetle Xyleborus affinis Eichhoff (Scolytinae: Curculionidae). Behav. Ecol. Sociobiol. 65: 1753-1761.
Biedermann PHW, Peer K & Taborsky M (2012) Female dispersal and reproduction in the ambrosia beetle Xyleborinus
saxesenii Ratzeburg (Coleoptera; Scolytinae). Mitteilungen der Deutschen Gesellschaft für allgemeine und angewandte
Entomologie in review.
Biedermann PHW & Taborsky M (2011) Larval helpers and age polyethism in ambrosia beetles. Proc. Natl Acad. Sci. USA
108: 17064-17069.
45
Chapter 3
Bischoff, L L (2004) The social structure of the haplodiploid bark beetle, Xylosandrus germanus. Diploma thesis, Zoological
Institute, Univ. Bern.
Boomsma JJ (2009) Lifetime monogamy and the evolution of eusociality. Philosophical Transactions of the Royal Society BBiological Sciences 364: 3191-3207.
Bourke AFG (2011) Principles of Social Evolution. Oxford University Press, Oxford.
Bradshaw AD (1965) Evolutionary significance of phenotypic plasticity in plants. Advances in Genetics (Caspari EW, ed), pp.
115-157. Academic Press, New York.
Bullock JM, Kenward RE & Hails RS (2002) Dispersal ecology. Blackwell, Oxford.
Charmantier A, Keyser AJ & Promislow DE (2007) First evidence for heritable variation in cooperative breeding behaviour.
Proceedings of the Royal Society B-Biological Sciences 274: 1757-1761.
Choe JC & Crespi BJ (1997) The Evolution of Social Behaviour in Insects and Arachnids. Cambridge University Press,
Cambridge UK.
Cockburn A (1998) Evolution of Helping Behavior in Cooperatively Breeding Birds. Annual Review of Ecology and Systematics
29: 141-177.
Darwin C (1859) The origin of species. Gramercy Books, New York.
De Fine Licht HH & Biedermann PHW (2012) Patterns of functional enzyme activity suggest that larvae are the key to
successful fungus farming by ambrosia beetles. Frontiers in Zoology in review.
Dewitt TJ, Sih A & Wilson DS (1998) Costs and limits of phenotypic plasticity. Trends Ecol. Evol. 13: 77-81.
Dingle H (1996) Migration: The biology of life on the move. Oxford University Press, New York.
Dingle H (2006) Animal migration: is there a common migratory syndrome? Journal of Ornithology 147: 212-220.
Dudley R (2000) The biomechanics of insect flight. Princeton University Press, Princeton NJ.
Emlen ST (1982) The evolution of helping. I. An ecological constraints model. Am. Nat. 119: 29-39.
Faccoli M & Rukalski JP (2004) Attractiveness of artificially killed red oaks (Quercus rubra) to ambrosia beetles (Coleoptera,
Scolytidae). Invertibrati di una foresta della Pianura Padana, Bosco della Fontana. (Ceretti P, Hardersen S, Mason F, Nardi G,
Tisato M & Zapparoli M, eds), pp. 171-179. 3. Cierre Grafica Editore, Verona.
Fischer M (1954) Untersuchungen über den kleinen Holzbohrer (Xyleborus saxeseni). Pflanzenschutzberichte 12: 137-180.
Francke-Grosmann H (1956) Zur Übertragung der Nährpilze bei Ambrosiakäfern. Naturwissenschaften 43: 286-287.
Francke-Grosmann H (1967) Ectosymbiosis in wood-inhabiting beetles. Symbiosis (Henry SM, ed), pp. 141-205. Academic
Press, New York.
French JRJ & Roeper RA (1972) Interactions of ambrosia beetle, Xyleborus dispar (Coleoptera, Scolytidae), with its symbiotic
fungus Ambrosiella hartigii (Fungi imperfecti). Canadian Entomologist 104: 1635-1641.
Fuller RC, Baer CF & Travis J (2005) How and When Selection Experiments Might Actually be Useful. Integrative and
Comparative Biology 45: 391-404.
Gandon S (1999) Kin competition, the cost of inbreeding and the evolution of dispersal. J. Theor. Biol. 200: 345-364.
Garraway E & Freeman BE (1981) Population-dynamics of the juniper bark beetle Phloeosinus neotropicus in Jamaica. Oikos
37: 363-368.
Gu HN & Danthanarayana W (1992) Quantitative Genetic-Analysis of Dispersal in Epiphyas-Postvittana .2. Genetic
Covariations Between Flight Capacity and Life-History Traits. Heredity 68: 61-69.
Gu HN, Hughes J & Dorn S (2006) Trade-off between mobility and fitness in Cydia pomonella L. (Lepidoptera : Tortricidae).
Ecological Entomology 31: 68-74.
Hamilton WD (1964) The genetical evolution of social behaviour. I+II. J. Theor. Biol. 7: 1-52.
46
Chapter 3
Hanski I, Saastamoinen M & Ovaskainen O (2006) Dispersal-related life-history trade-offs in a butterfly metapopulation.
Journal of Animal Ecology 75: 91-100.
Harrison RG (1980) Dispersal polymorphisms in insects. Ann. Rev. Ecol. Syst. 11: 95-118.
Heg D, Bachar Z, Brouwer L & Taborsky M (2004) Predation risk is an ecological constraint for helper dispersal in a
cooperatively breeding cichlid. Proceedings of the Royal Society of London Series B-Biological Sciences 271: 2367-2374.
Hosking GB (1972) Xyleborus saxeseni, its life-history and flight behaviour in New Zealand. N. Z. J. Forest Science 3: 37-53.
Hughes J & Dorn S (2002) Sexual differences in the flight performance of the oriental fruit moth, Cydia molesta.
Entomologia Experimentalis et Applicata 103: 171-182.
Hughes WOH, Oldroyd BP, Beekman M & Ratnieks FLW (2008) Ancestral monogamy shows kin selection is key to the
evolution of eusociality. Science 320: 1213-1216.
Hulcr J & Dunn RR (2011) The sudden emergence of pathogenicity in insect-fungus symbioses threatens naive forest
ecosystems. Proceedings of the Royal Society B: Biological Sciences.
Johnson CG (1969) Migration and dispersal of insects by flight. Methuen, London.
Kajimura H & Hijii N (1992) Dymamics of the fungal symbionts in the gallery system and the mycangia of the ambrosia
beetle, Xylosandrus mutilatus (Blandford) (Coleoptera, Scolytidae). Ecological Research 7: 107-117.
Kammer AE & Heinrich B (1978) Insect flight metabolism. Advances in insect physiology (Treherne JE, Berridge MJ &
Wigglesworth VB, eds), pp. 133-228. Academic Press, London.
Keller L (2009) Adaptation and the genetics of social behaviour. Philosophical Transactions of the Royal Society B-Biological
Sciences 364: 3209-3216.
Keller L, Peer K, Bernasconi C, Taborsky M & Shuker DM (2011) Inbreeding and selection on sex ratio in the bark beetle
Xylosandrus germanus. submitted.
Kent DS & Simpson JA (1992) Eusociality in the beetle Austroplatypus incompertus (Coleoptera: Platypodidae).
Naturwissenschaften 79: 86-87.
Kirkendall LR, Kent DS & Raffa KF (1997) Interactions among males, females and offspring in bark and ambrosia beetles: the
significance of living in tunnels for the evolution of social behavior. The Evolution of Social Behavior in Insects and Arachnids
(Choe JC & Crespi BJ, eds), pp. 181-215. Cambridge University Press.
Kisdi E (2002) Dispersal: Risk spreading versus local adaptation. Am. Nat. 159: 579-596.
Komdeur J (1992) Importance of Habitat Saturation and Territory Quality for Evolution of Cooperative Breeding in the
Seychelles Warbler. Nature 358: 493-495.
Komdeur J (2006) Variation in individual investment strategies among social animals. Ethology 112: 729-747.
Levins R (1968) Evolution in changing environments. Princeton University Press, Princeton NJ.
Li JB & Margolies DC (1994) Barometric-Pressure Influences Initiation of Aerial Dispersal in the 2-Spotted Spider-Mite.
Journal of the Kansas Entomological Society 67: 386-393.
Li JB & Margolies DC (1994) Responses to Direct and Indirect Selection on Aerial Dispersal Behavior in Tetranychus-Urticae.
Heredity 72: 10-22.
Lopez-Guillen G, Carrasco JV, Cruz-Lopez L, Barrera J, Malo E & Rojas J (2011) Morphology and Structural Changes in Flight
Muscles of Hypothenemus hampei (Coleoptera: Curculionidae) Females. Environmental Entomology 40: 441-448.
Maccoll ADC & Hatchwell BJ (2003) Heritability of parental effort in a passerine bird. Evolution 57: 2191-2195.
Mousseau TA & Dingle H (1991) Maternal Effects in Insect Life Histories. Annual Review of Entomology 36: 511-534.
Norris DM (1979) The mutualistic fungi of Xyleborini beetles. Nutrition, Mutualism, and Commensalism (Batra LR, ed), pp.
53-63. Allanheld, Osmun & Company, Montclair.
Pfeffer A (1995) Zentral- und westpaläarktische Borken- und Kernkäfer (Coleoptera: Scolytidae, Platypodidae). Pro
Entomologia, c/o Naturhistorisches Museum Basel, Basel.
47
Chapter 3
Price TD, Qvarnstrom A & Irwin DE (2003) The role of phenotypic plasticity in driving genetic evolution. Proceedings of the
Royal Society of London Series B-Biological Sciences 270: 1433-1440.
R Development Core Team (2008) R: A language and environment for statistical computing. Vienna, Austria, R Foundation
for Statistical Computing.
Robertson IC (1998) Flight muscle changes in male pine engraver beetles during reproduction: the effects of body size,
mating status and breeding failure. Physiological Entomology 23: 75-80.
Robinson GE, Fernald RD & Clayton DF (2008) Genes and Social Behavior. Science 322: 896-900.
Roff DA & Fairbairn DJ (2001) The genetic basis of dispersal and migration and its consequences for the evolution of
correlated traits. Dispersal (Clobert J, Danchin E, Dhondt AA & Nichols JD, eds), pp. 191-202. Oxford University Press, New
York.
Roff DA & Fairbairn DJ (2007) The evolution and genetics of migration in insects. Bioscience 57: 155-164.
Roff DA (1989) Exaptation and the evolution of dealation in insects. J. Evol. Biol. 2: 109-123.
Ronce O (2007) How does it feel to be like a rolling stone? Ten questions about dispersal evolution.
Sakai AK, Allendorf FW, Holt JS, Lodge DM, Molofsky J, With KA, Baughman S, Cabin RJ, Cohen JE, Ellstrand NC, McCauley
DE, O'Neil P, Parker IM, THOMPSON JN & Weller SG (2001) The population biology of invasive species. Annual Review of
Ecology and Systematics 32: 305-332.
Scheiner SM (1993) Genetics and Evolution of Phenotypic Plasticity. Annual Review of Ecology and Systematics 24: 35-68.
Schmitz RF (1972) Behavior of Ips pini during mating, oviposition, and larval development (Coleoptera, Scolytidae).
Canadian Entomologist 104: 1723-1728.
Sih A, Bell AM, Johnson JC & Ziemba RE (2004) Behavioral syndromes: An integrative overview. Q. Rev. Biol. 79: 241-277.
Southwood TRE (1977) Habitat, Templet for Ecological Strategies - Presidential-Address to British-Ecological-Society, 5
January 1977. Journal of Animal Ecology 46: 337-365.
Stacey PB & Ligon JD (1991) The benefits-of-philopatry hypothesis for the evolutioni of cooperative breeding: variation in
territory quality and group size effects. Am. Nat. 117: 831-846.
Torriani MVG, Mazzi D, Hein S & Dorn S (2010) Direct and correlated responses to artificial selection on flight activity in the
oriental fruit moth (Lepidoptera: Tortricidae). Biol. J. Linn. Soc. 100: 879-889.
Van Voorhies WA (2001) Metabolism and lifespan. Experimental Gerontology 36: 55-64.
Via S & Lande R (1985) Genotype-Environment Interaction and the Evolution of Phenotypic Plasticity. Evolution 39: 505-522.
Weber BC & McPherson JE (1983) Life history of the ambrosia beetle Xylosandrus germanus (Coleoptera: Scolytidae).
Annals of the Entomological Society of America 76: 455-462.
West SA, Murray MG, Machado CA, Griffin AS & Herre EA (2001) Testing Hamilton's rule with competition between
relatives. Nature 409: 510-512.
Wilson EO (1971) The insect societies. Belknap Press of Harvard University Press, Cambridge.
Wilson EO (1975) Sociobiology. The new synthesis. Belknap Press of Harvard University Press, Cambridge.
Zera AJ & Denno RF (1997) Physiology and ecology of dispersal polymorphism in insects. Annual Review of Entomology 42:
207-230.
Zettel J (1984) The Significance of Temperature and Barometric-Pressure Changes for the Snow Surface-Activity of IsotomaHiemalis (Collembola). Experientia 40: 1369-1372.
48
Chapter 3
Supplementary information
Table S1. Separate GEE models to examine differences (p < 0.05) between the proportion of time
larval stages spent with the observed behaviours.
Behaviour
Parameter
Coeff.  SE
z
p
digging / feeding
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery
Number of larvae
Number of staying adult females
0.89  0.15
-0.01  0.1
0.04  0.1
0.05  0.07
-0.01  0.0
-0.02  0.0
0.02  0.01
5.92
-0.06
0.42
0.69
-3.62
-6.96
1.69
<0.001
0.95
0.67
0.49
<0.001
<0.001
0.09
gallery hygiene
(shuffling and balling)
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery*
Number of larvae
Number of staying adult females*
-3.1  0.16
-0.22  0.18
-0.11  0.16
0.11  0.13
0.01  0.01
0.02  0.01
-0.01  0.02
-19.11
-1.25
-0.68
0.8
1.3
3.53
-0.67
<0.001
0.21
0.49
0.42
0.19
<0.001
0.51
allogrooming
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery*
Number of larvae
Number of staying adult females*
-2.28  0.12
-0.19  0.14
-0.29  0.13
-0.1  0.1
-0.0  0.0
0.03  0.0
-0.02  0.01
-18.3
-1.35
-2.2
-1
-0.64
7.26
-1.55
<0.001
0.18
0.028
0.34
0.53
<0.001
0.12
inactivity
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery*
Number of larvae*
Number of staying adult females*
-1.91  0.1
0.31  0.12
0.36  0.12
0.06  0.08
0.0  0.0
-0.0  0.0
-0.02  0.01
-19.2
2.54
3.16
0.66
1.01
-0.09
-1.46
<0.001
0.011
0.002
0.51
0.31
0.93
0.15
locomotion
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery*
Number of larvae
Number of staying adult females
-2.17  0.14
-0.49  0.15
-0.6  0.14
-0.11  0.11
-0.0  0.0
0.03  0.0
-0.03  0.01
-15.5
-3.3
-4.16
-1.05
-0.58
5.36
-2.18
<0.001
<0.001
<0.001
0.3
0.56
<0.001
0.029
We used GEEs with an exchangeable correlation structure of the response variable within a cluster (= gallery) to identify effects of the
selection regime on the total amount of time each behaviour was observed, by controlling for influences of gallery-age, number of larvae
and number of staying adult females. The influences of the regimes on the behavioural frequencies are displayed as contrasts between
classes.
*variable not in the final model after step-wise model reduction using ANOVA analysis of log-likelihood scores.
49
Chapter 3
Table S2. Separate GEE models to examine differences (p < 0.05) between the proportion of time
adult males spent with the observed behaviours.
Behaviour
Parameter
Coeff.  SE
z
p
cropping fungus
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery*
Number of larvae*
Number of males*
Number of teneral females*
Number of staying adult females
-0.81  0.32
-0.42  0.39
-0.06  0.35
0.36  0.36
-0.01  0.02
0.02  0.02
0.06  0.08
-0.05  0.05
-0.05  0.03
-2.55
-1.07
-0.17
1
-0.48
1.15
0.72
-0.89
-1.71
0.011
0.28
0.86
0.32
0.63
0.25
0.47
0.38
0.088
allogrooming
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery
Number of larvae*
Number of males
Number of teneral females*
Number of staying adult females*
0.25  0.9
0.38  0.42
-0.49  0.35
-0.87  0.43
-0.03  0.02
-0.02  0.02
0.13  0.08
0.06  0.04
-0.02  0.03
0.28
0.92
-1.39
-2.04
-1.53
-0.95
1.57
1.39
-0.6
0.78
0.36
0.16
0.041
0.13
0.34
0.12
0.16
0.55
mating attempt and
copula
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery*
Number of larvae
Number of males*
Number of teneral females
Number of staying adult females*
-1.98  0.39
-0.12  0.48
-0.04  0.43
0.09  0.48
0.02  0.02
0.03  0.02
-0.12  0.14
0.08  0.05
0.02  0.04
-5.05
-0.25
-0.08
0.18
0.89
1.56
-0.9
1.57
0.42
<0.001
0.8
0.93
0.86
0.37
0.12
0.37
0.12
0.67
locomotion
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery*
Number of larvae
Number of males*
Number of teneral females*
Number of staying adult females*
-0.98  0.31
-0.57  0.41
-0.38  0.36
0.19  0.41
0.0  0.02
-0.04  0.02
-0.11  0.11
-0.05  0.06
0.03  0.03
-3.13
-1.38
-1.04
0.47
0.14
-1.52
-1.02
-0.84
0.95
0.002
0.17
0.3
0.64
0.89
0.13
0.31
0.4
0.34
We used GEEs with an exchangeable correlation structure of the response variable within a cluster (= gallery) to identify effects of the
selection regime on the total amount of time each behaviour was observed, by controlling for influences of gallery-age, number of larvae,
number of males, number of teneral females and number of staying adult females. The influences of the regimes on the behavioural
frequencies are displayed as contrasts between classes.
*variable not in the final model after step-wise model reduction using ANOVA analysis of log-likelihood scores.
50
Chapter 3
Table S3. Separate GEE models to examine differences (p < 0.05) between the proportion of time adult females
spent with the observed behaviours.
Behaviour
Parameter
Coeff.  SE
z
p
Cooperation (blocking,
shuffling,
allogrooming,
cannibalism)
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery
Number of larvae
Number of staying adult females*
1.7  0.49
1.35  0.34
0.5  0.31
-0.85  0.3
-0.08  0.01
0.06  0.01
0.00  0.02
3.45
4
1.6
-2.83
-9.41
4.78
0.24
<0.001
<0.001
0.11
0.005
<0.001
<0.001
0.81
blocking
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery
Number of larvae
Number of staying adult females
0.27  0.54
1.1  0.39
0.1  0.38
-1.01  0.33
-0.05  0.01
0.06  0.01
-0.25  0.05
0.49
2.85
0.26
-3.02
-4.73
4.45
-5.32
0.62
0.004
0.8
0.003
<0.001
<0.001
<0.001
shuffling
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery
Number of larvae
Number of staying adult females
0.33  0.64
0.03  0.45
-0.58  0.45
-0.61  0.44
-0.07  0.01
0.05  0.02
-0.16  0.05
0.51
0.08
-1.28
-1.38
-4.82
2.67
-2.95
0.61
0.94
0.2
0.17
<0.001
0.008
0.003
allogrooming
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery*
Number of larvae
Number of staying adult females
-3.78  0.34
-0.29  0.4
-0.03  0.39
0.27  0.41
-0.01  0.01
0.03  0.02
0.05  0.02
-11.13
-0.73
-0.07
0.65
-0.84
2.13
2.52
<0.001
0.47
0.95
0.52
0.4
0.033
0.012
cannibalism
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery*
Number of larvae*
Number of staying adult females*
-7.01  1.13
2.43  1.2
2.69  1.19
0.25  0.56
-0.02  0.02
-0.03  0.05
0.06  0.05
-6.22
2.03
2.26
0.46
-0.84
-0.68
1.05
<0.001
0.043
0.024
0.65
0.4
0.49
0.3
cropping fungus
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery
Number of larvae
Number of staying adult females
-0.66  0.27
-1.28  0.16
-0.67  0.14
0.61  0.16
0.02  0.01
-0.02  0.01
0.01  0.01
-2.45
-8.06
-4.64
3.92
3.98
-2.15
1.35
0.014
<0.001
<0.001
<0.001
<0.001
0.031
0.18
inactivity
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery
Number of larvae*
Number of staying adult females*
-2.91  0.44
-0.92  0.29
-0.33  0.25
0.58  0.3
0.02  0.01
-0.02  0.01
0.01  0.01
-6.63
-3.13
-1.33
1.94
3.08
-1.14
0.81
<0.001
0.001
0.18
0.052
0.002
0.25
0.42
locomotion
Mean frequency (Control)
Contrast Control vs. Cooperators
Contrast Control vs. Dispersers
Contrast Cooperators vs. Dispersers
Age of gallery*
Number of larvae*
Number of staying adult females*
-2.73  0.15
0.77  0.2
0.87  0.2
0.11  0.18
0.0  0.01
-0.01  0.01
-0.01  0.02
-18.4
3.94
4.47
0.6
0.56
-0.96
-0.52
<0.001
<0.001
<0.001
0.55
0.58
0.34
0.6
We used GEEs with an exchangeable correlation structure of the response variable within a cluster (= gallery) to identify effects of the
selection regime on the total amount of time each behaviour was observed, by controlling for influences of gallery-age, number of larvae
and number of staying adult females. The influences of the regimes on the behavioural frequencies are displayed as contrasts between
classes.
*variable not in the final model after step-wise model reduction using ANOVA analysis of log-likelihood scores.
51
Chapter 3
Table S4. Separate GLM models to examine differences (p < 0.05) between the composition of the galleries in
the two selection regimes on day 45, day 62 and day 87 after gallery foundation.
Variable
Contrasts
Gallery founding success
rate
Cooperators 1 vs. Dispersers 1
th
th
Cooperators 6 vs. Dispersers 6
st
Cooperators 1 vs. Cooperators
th
6
st
th
Dispersers 1 vs. Dispersers 6
st
st
Coeff. ± SE
z
p
Direction
-0.08 ± 0.26
1.82 ± 0.42
-0.46 ± 0.33
-0.32
4.3
-1.4
0.75
<0.001
0.16
ns
+++
ns
1.45 ± 0.34
4.31
<0.001
+++
Model coefficients are reported as Coeff. ± SE (standard error of the estimate). The influence of the selection regime (cooperators vs.
dispersers) on the offspring numbers and proportion of staying females was modelled for each day separately with quasipoisson or
quasibinomial error distributions. A positive coefficient denotes that the mean of the cooperators is higher than the mean of the dispersers;
a negative contrast denotes the reverse. Significant differences (p < 0.05) are displayed in bold; trends for differences (p < 0.05) are
indicated by brackets. C = cooperators, D = dispersers.
Table S5. GEE model to examine differences (p < 0.05) in the success rates of daughters to produce a brood
st
th
when reared independently between the selection regimes and at the start (1 generation) and at the end (6
generation) of laboratory breeding.
Offspring stage
Age of gallery
Coeff. ± SE
t
p
Direction
Number of immature stages (eggs,
larvae, pupae)
45 days
62 days
87 days
0.61 ± 0.39
-2.69 ± 0.67
0.47 ± 0.87
1.56
-4.03
0.54
0.13
<0.001
0.6
C=D
C>D
C=D
Number of staying adult ♀♀
45 days
62 days
87 days
-0.72 ± 0.43
-1.77 ± 0.09
0.39 ± 0.72
-1.7
-19.2
0.54
0.11
<0.001
0.6
C=D
C>D
C=D
Number of dispersed adult ♀♀
45 days
62 days
87 days
dispersal only in one cooperators gallery
1.1 ± 0.49
2.23
0.03
C<D
-1.15 ± 0.51
-2.24
0.039
C>D
Proportion of staying adult ♀♀
45 days
62 days
87 days
dispersal only in one cooperators gallery
-2.87 ± 0.57
-5.04
<0.001
C>D
1.54 ± 0.8
1.93
0.072
(C < D)
Total adult individuals (♂,♀)
45 days
62 days
87 days
-0.68 ± 0.36
-0.51 ± 0.19
-0.52 ± 0.36
-1.93
-2.67
-1.42
0.068
0.01
0.18
(C > D)
C>D
C=D
We used GEEs with an exchangeable correlation structure of the response variable within a cluster (= gallery) and binomial error distribution (successful brood
production yes or no). The influences of the regimes on the success rates are displayed as contrasts between classes.
52
Chapter 4
Abstract
Strongly female-biased sex ratios are typical for the fungal feeding haplodiploid Xyleborini
(Scolytinae, Coleoptera), and are a result of inbreeding and local mate competition (LMC).
These ambrosia beetles are hardly ever found outside of trees, and thus male frequency and
behavior have not been addressed in any empirical studies to date. In fact, for most species
the males remain undescribed. Data on sex ratios and male behavior could, however,
provide important insights into the Xyleborini's mating system and the evolution of
inbreeding and LMC in general.
In this study, I used in vitro rearing methods to obtain the first observational data on sex
ratio, male production, male and female dispersal, and mating behavior in a xyleborine
ambrosia beetle. Females of Xyleborinus saxesenii Ratzeburg produced between 0 and 3
sons per brood, and the absence of males was relatively independent of the number of
daughters to be fertilized and the maternal brood sex ratio. Both conformed to a strict LMC
strategy with a relatively precise and constant number of males. If males were present they
eclosed just before the first females dispersed, and stayed in the gallery until all female
offspring had matured. They constantly wandered through the gallery system, presumably in
search of unfertilized females, and attempted to mate with larvae, other males, and females
of all ages. Copulations, however, only occurred with immature females. From galleries with
males, nearly all females dispersed fertilized. Only a few left the natal gallery without being
fertilized, and subsequently went on to produce large and solely male broods. If broods were
male-less, dispersing females always failed to found new galleries.
Keywords
inbreeding, haplodiploidy, ambrosia beetles, all male broods, LMC, sub-sociality
53
Chapter 4
Introduction
In panmictic (randomly mating) species, natural selection usually favors balanced sex ratios
(Fisher 1930). In settings where male and female offspring are of unequal value to the
mother, however, optimal sex ratios may be unbalanced. For example, in a non-randomly
mating population or species with strictly local offspring reproduction, a mother will gain
highest fitness by producing as many daughters as possible and just enough sons to ensure
fertilization of all sisters in the brood (Hamilton 1967). This extreme economy in the
production of males is common in small arthropods with regular brother-sister mating and
local mate competition (LMC) between brothers (Hamilton 1967; Norris 1993). Usually it is
associated with arrhenotokous haplodiploidy (Hamilton 1978 referred to arthropods
exhibiting these characters as "the biofacies of extreme sex ratios and arrhenotoky").
Arrhenotoky potentially gives a mother precise control over the sex of each offspring, as she
may choose to produce diploid daughters by fertilizing an egg, or haploid sons by leaving it
unfertilized (Hamilton 1967; Charnov 1982). Hamilton (1978) suggested that the ancestral
habitat for the aforementioned biofacies is situated under the bark of dead trees, even if the
support for this quotation is weak and claims on the original habitat of ancient lineages are
highly speculative (Normark et al. 1999).
A prominent group of beetles living under the bark is the weevil sub-family Scolytinae,
that includes species with varying mating systems (Kirkendall 1983; 1993) and ploidy
(diploid, functional haploid, haplodiploid; Kirkendall 1993; Brun et al. 1995; Normark et al.
1999). Both characters are selected to the extremes in all species belonging to the subtribe
Xyleborini, which are a typical example for Hamilton's biofacies since they are characterized
by strong inbreeding, LMC and haplodiploidy. These so-called ambrosia beetles solely feed
on a microbial complex (fungi, yeasts and bacteria; Haanstad and Norris 1985), which they
grow on the walls of self-constructed tunnel systems ("galleries") in the heart- or sapwood of
trees. Until recently, this habitat has virtually been impossible to access and observe without
destruction, which is why basic data on life histories, mating systems and behavior of most
Xyleborini is still missing. The adaptation of an in vitro rearing technique (Norris and Baker
1967; Biedermann et al. 2009) finally may lead to the establishment of an excellent new
model system for studying the independent evolution of inbreeding and haplodiploidy in a
weevil lineage.
The few observational studies on Xyleborini that exist suggest that these beetles behave
sub-socially (Kirkendall et al. 1997; Mueller et al. 2005). Sub-sociality in Xyleborinus saxesenii
Ratzeburg and other Xyleborini is indicated by the fact that mature female offspring
maintain the natal gallery throughout a few days or even weeks at a time when they could
already found their own nests (Kirkendall et al. 1997; Mueller et al. 2005; Biedermann 2007).
Costs of independent breeding have been found to be high in this species due to risky
dispersal and low founding success, which in former studies did not exceed 20% neither in
the field nor when in vitro culturing was used (Biedermann et al. 2009). Therefore, it might
pay for daughters to stay in productive
54
Chapter 4
galleries, where they either reproduce themselves, or gain indirect fitness benefits by
helping to produce more sisters (e.g. Peer and Taborsky 2007). The help of males (if there is)
is assumed to be of minor importance to the family, given their tiny size (figures in Fischer
1954) and their underrepresentation in relation to females.
If males do not play an important role for gallery maintenance and protection, male
numbers should be minimized in order to lower the cost of LMC. The optimal number should
be affected by the maximum number of females one male is able to fertilize, the timing of
male production, male survivorship and longevity, as well as male dispersal (see Kirkendall
1993 for a detailed review; Borsa and Kjellberg 1996). In Xyleborini males are outnumbered
by females by a ratio of 1:5 to 1:30, depending on the species (e.g. Schedl 1962). Males
supposedly hatch before females (observed in Xyleborus affinis Eichhoff; Roeper et al. 1980),
but their life expectancy is unknown. As they lack functional flight wings, they are assumed
to never leave the nest (Roeper et al. 1980; Kirkendall 1983; 1993), however, recent data
suggests that males of Xylosandrus germanus Reiter sometimes leave their natal gallery to
crawl on the host tree in search for outbreeding opportunities, i.e., disperse (Peer and
Taborsky 2004).
The objective of this study was to observe males of X. saxesenii inside their galleries for
the first time to report their number and hatching time relative to females, as well as to
determine whether or not males disperse and if they share in gallery maintenance and
protection. Furthermore, I aimed at clarifying the existence of all male broods, which has
been remarked upon in several studies on Xyleborini (e.g. Fischer 1954; Norris and Baker
1967; Kirkendall 1993). Such broods are assumed to be produced by unfertilized females
that later mate with one of their sons to subsequently produce "normal" mixed broods
(Herfs 1959). Their existence was tested by following the fate of unfertilized females. All
these studies were conducted by consecutively rearing several generations of X. saxesenii
families in an artificial medium in glass tubes that allow for behavioral observations
(Biedermann et al. 2009).
Methods
Study species
In the study species Xyleborinus saxesenii Ratzeburg, the ambrosial complex is typically
dominated by the anamorphic fungus Ambrosiella sulfurea Batra (Batra 1967), and serves as
the sole food for adults as well as the developing larvae. The latter feed xylomycetophagously (on fungus and wood; Schedl 1958), in this way digging out a single large
brood chamber where individuals of all age classes live in close vicinity to each other and
also to their fungal food source. Galleries are always founded by individual females that
usually have been fertilized by a brother prior to their emergence from the natal nest. The
first offspring mostly stay with their mother after reaching adult-hood while the subsequent
offspring generations develop (e.g. Kalshoven 1962; Schedl
55
Chapter 4
1966; Bischoff 2004; Biedermann et al. 2009). Brood care and fungus tending in X. saxesenii
are hitherto unknown, but expected to occur, because in case the foundress dies before the
first brood has eclosed, the brood dies as well and the fungal garden de-grades (Batra and
Michie 1963; Norris 1979; 1993). Furthermore, gallery protection by blocking the entrance
with the abdomen or a plug of frass and brood care behaviors have been observed in adult
daughters of this as well as other xyleborine species (Kirkendall et al. 1997; Bischoff 2004;
Biedermann 2007; personal observations).
Laboratory breeding and data collection
X. saxesenii beetles were bred in artificial agar-sawdust based "standard medium" in glass
tubes (Biedermann et al. 2009). Single females were surface sterilized by sub-merging them
first in ethanol (95%) and then in distilled water for a few seconds and subsequently put
directly onto the medium. They usually started to excavate a tunnel system within two days
and would not lay eggs until their mutualistic ambrosia fungus had started to grow. About
20% of the females started to lay eggs, which resembles the breeding success found in the
field (Biedermann et al. 2009). The developing brood would subsequently enlarge the gallery
system which was often constructed next to the tube wall, which facilitated the observations
necessary for this study. Parts of the gallery inside the medium could not be accessed with
this method, but as individuals move a lot, this should not have influenced my results.
Observations
I scanned the beetles in 70 observable galleries every second to third day and recorded male
and female numbers, allocating the individuals to different developmental stages (eggs,
larvae, male and female pupae, immature females, mature females, and males). Male and
female pupae and adults were easy to differentiate because of their strong sexual
dimorphism in size (Fig. 1, see figures in Fischer 1954). Immature females were identified by
their light brown coloration that would turn black after maturation. As a result of sex specific
mortality rates and siblicide, the so gained "secondary sex ratio" of the immature and
mature offspring may differ from the mother's optimal "primary sex ratio" of the eggs.
Where I could witness male-female interactions, I recorded the age of the partners and
whether it was a mating attempt or successful mating. Whenever possible, I determined the
dates when the first egg was present (N = 70 galleries), when the first male and female
offspring hatched (N = 29 galleries), when the first and last female maturated (N = 26), and
when the first male dispersed (N = 13). Dispersal was defined as emergence from the gallery,
i.e., when individuals were found on the surface of the medium under the cap of the tube
(Biedermann 2007). I stopped monitoring and dissected the galleries either when eclosion of
new beetles ceased within about 3
56
Chapter 4
months (N = 41 galleries), or when all adult females had dispersed (N = 29 galleries). In the
latter case, galleries were used to monitor male dispersal.
I measured the breeding success (defined as the successful start of egg-laying within 40
days) and brood size of 311 daughters out of 33 families (median = 5 daughters/ family,
range = 1—41) to determine whether they differ between daughters of galleries with males
and male-less ones. Subsequently, I correlated the secondary sex ratio of 62 daughterfamilies that had successfully produced a brood with the secondary sex ratio of their
mothers' families to test whether secondary sex ratios might be heritable.
Additionally, I observed the behavior of eight males from different galleries for 10 min
and calculated the proportion of time spent on different behaviors. I differentiated between
shuffling (moving frass and feces with the legs under the body and with the abdomen),
digging (excavating new tunnels), cannibalism (feeding on a larva, pupa, or adult beetle),
courtship behavior (grooming an egg, larva, pupa, or adult beetle with maxillae and labium),
walking, cropping (feeding on the fungal layer covering gallery walls), and mating attempt
(mounting or copulating with a female) (Biedermann 2007).
Statistical analyses
The association between numbers of male and female offspring was explored using linear
regression. The number of males within a brood of a given size should have
57
Chapter 4
a binomial distribution under random sex determination. To test if this was the case, I
compared the variance of the male numbers observed with that expected if the numbers of
males were binomially distributed using a Chi 2 -test (Green et al. 1982). The same test was
also used to analyze whether the number of males per family resembled a Poisson
distribution. Male mating preference with either immature or adult females was analyzed
using Fisher's exact test. Usually the data were not normally distributed (which was
determined with Kolmogorov-Smirnov tests), consequently non-parametric statistics were
conducted. The Mann-Whitney U-test for comparisons between two independent groups
was applied to determine productivity differences between male and male-less broods, and
to check for differences in the timing of male and female hatching and dispersal. The
inheritance of sex ratios between mother and daughter galleries was analyzed with a
Spearman's rank correlation. Analyses were performed with SPSS (Release 14.0; SPSS,
Chicago, IL) and R (R Development Core Team 2008).
Results
Sex ratio and breeding success
Males were extremely scarce in X. saxesenii galleries (mean abundance per gallery = 0.96, SE
= 0.09, N = 70; Fig. 2). Male production by mothers was not random, as their overall number
did not resemble a Poisson distribution. They produced more one- and two-male broods,
fewer male-less and three-male broods than expected, and no four-male brood, which
suggests the optimal number is one or two males per brood. Nearly one third of the mothers
produced no males (N = 21).
Families with males were significantly larger than families without males (Fig. 3). In
accordance with this finding, males were absent from 50% of the galleries with only ten
daughters or fewer, whereas only 8.3% of the galleries with more than ten daughters were
male-less. However, overall the number of males per gallery was only weakly affected by the
number of adult females (Linear regression: F = 3.824, p = 0.055, DF= 1, N = 70; Fig. 4). The
residual mean squared error of the number of males (MSE = 0.597) was much smaller than
the mean variance expected under binomial distribution (0.9398; Chi 2-test: χ2= 43.2, DF = 68,
p > 0.05), which indicates that the production of males was not random, but relatively
constant and precise (around one or two males per brood).
Breeding success (i.e., the likelihood to lay at least one egg within 40 days) of daughters
per family ranged between 0% and 78% (mean = 15.2%, median = 0, N = 311 daughters from
33 families), and was 0% for all male-less broods. A statistical difference was however not
detectable because of the high variation of breeding success across all families tested (Utest: T = 77,p = 0.067, N = 26 vs. 7).
The secondary sex ratios of the 62 successfully founded daughter-galleries did not
correlate with the secondary sex ratios of their mothers galleries (Spearman's rank
correlation: r = 0.038, p = 0.771, N = 62).
58
Chapter 4
Male presence and behavior
The first male per gallery usually eclosed a few days after his first sister did so (Fig. 5), but
before her maturation (N = 29 galleries). Following eclosion, males spent most of their time
with cropping fungus and walking through the gallery system. When encountering a possible
partner, the male started to courtship her, which resembles grooming the body of the
female, and afterwards attempted to mate with her. Females seemed relatively reluctant,
and thus mating attempts usually lasted several minutes and were rarely successful. Thirty
mating attempts were observed, during which males mounted immature (N = 7 of 30 times)
and mature females (N = 11 of 30 times) as well as pieces of wood, larvae, and other males
( N = 12 of 30 times). Successful copulations, however, were only observed with three
immature females, which hints towards a preference to mate with them (Fisher's exact test:
p = 0.09, N = 21; Fig. 6). Only two females were found to emerge unfertilized from their
galleries (indicated by the production of an exclusively male brood later on); they emerged
from galleries with brood sex ratios of 2:16 and 1:86, respectively, which indicates that one
male is capable of fertilizing up to 60 sisters on its own (no unfertilized females were
detected in a gallery with a sex ratio of 1:60). Other male behaviors which were rarely
exhibited included digging, cannibalism, and the shuffling of saw-dust and feces with the legs
(all with median = 0%).
Although males are wingless, some of them (at least one male in 13 out of 29 galleries)
dispersed, i.e., were found on the surface of the medium, when all offspring had matured
and no new eggs were laid. Males did not disperse randomly together with
59
Chapter 4
60
Chapter 4
61
Chapter 4
their sisters throughout gallery life, but in most cases the first male dispersal
occurred after the last female (in a gallery) had matured (Fig. 5).
Discussion
Under inbreeding conditions, natural selection should favor the production of
that number of males that maximizes the mean number of inseminated females
dispersing from a brood. Thus, in case males are able to inseminate only a
limited number of females, the numbers of males and females should be
correlated (e.g. Green et al. 1982; Borsa and Kjellberg 1996). In contrast to
these predictions, male numbers in X. saxesenii families were relatively
independent of the number of sisters to be fertilized, ranging from 0 to 3 males
per 1 to 86 females, and also independent of maternal brood sex ratios. These
facts conform to a strict LMC strategy (e.g. Hamilton 1967; Bernal et al. 2001)
with a relatively precise and constant number of males per brood. Males were
apparently extremely successful to locate all their unfertilized sister. Only ~3%
(2 of 70 galleries) of the successfully breeding females turned out to be
unfertilized (they produced solely male broods), which is approximately the
same ratio found for Xyleborus compactus Eichhoff in the field (Brader 1964).
Addition-ally, in a family with a sex ratio of 1:60 no unfertilized females were
found during consecutive lab rearing. The success of males is probably
explainable by the gallery morphology. Xyleborini in the genus Xyleborinus
usually construct brood chambers ("caves"; Wood 1982; Roeper 1995) instead
of branching tunnel systems as they are excavated by species of the genus
Xyleborus (J. Hulcr personal communication), where males and females
probably meet more regularly and thus fewer males may be sufficient to
fertilize all sisters. In line with this idea, male and female numbers are
correlated in the branching tunnel systems of Xyleborus affinis Eichhoff
(Xyleborina, Scolytinae), as founder females lay eggs in clusters that always
contain a male egg (Roeper et al. 1980).
At least one male per family should, however, be expected under the
assumption of strict inbreeding. Contrary to that, 30% of all families were maleless; a proportion which is not untypical for Xyleborini (19% of all families in
Xylosandrus germanus Reiter in the field; Peer and Taborsky 2004). As these
measures were taken on adult offspring, the easiest explanation for that finding
is that the secondary sex ratios reported, differ from the primary sex ratios
allocated by the mother to the eggs, i.e., males had somehow been eliminated.
Such secondary male killing could have been accomplished either by selectively
cannibalizing offspring or by selecting mechanisms of the fungal cultivars 2 (for
attine ants; Mueller 2002) or of bacteria in the ovaries (Peleg and Norris 1972;
Kawasaki et al. 2009). Given that both microbes are transmitted only from
mothers to daughters, they would presumably profit from female biased
broods (Dyson and Hurst 2004; Normark 2004). Studies on the primary sex ratio
2
The fungal manipulation hypothesis did not hold up in the only explicit tests so far (see
Dijkstra & Boomsma (2008) in Oikos 117: 1892-1906. [This note was added after the publication
of this article.]
62
Chapter 4
allocated by the mother to the eggs are therefore necessary to clarify the role
of secondary sex ratio distorters as a cause for the high frequency of maleless families in Xyleborini. If, however, it is assumed that there is no
secondary male killing and that the absence of males is intended by the mother
beetle from the very beginning, then either the chances for mating with foreign
partners (and thus, outbreeding) must be higher than previously expected, or
selection against male-less broods must be weak. Meeting and outbreeding
with foreign mates is obviously possible, as males dispersed from nearly half of
the galleries when brood production ceased and females sometimes produced
all male broods, whose males also all emerged. It fits these data that X.
saxesenii males can sometimes be seen outside their nests, occasionally walking
on their natal host trees (personal field data). The idea that males disperse from
their natal nest in search for outbreeding opportunities implies, however, that
they are able to enter foreign galleries, what seems so be complicated by the
fact that females or more rarely males (in the genera Coptodryas and
Cyclorhipidion; J. Hulcr personal communication) block the gallery entrances
(e.g. Kirkendall et al. 1997; personal observations). On the other hand, maleless families were always very small (equal to or less than 10 females; see
similar results in Hypothenemus hampei Ferrari by Borsa and Kjellberg 1996),
and in a previous study daughters of small families (less than 20 females) were
found to contribute much less offspring to the next generation than the
daughters from larger families, presumably because the latter are in better
body condition and have a more productive complex of microbes to nourish
them (Biedermann et al. 2009). Therefore, it is possible that these small maleless galleries would anyway not contribute much fitness to the next generation
and hence selection against the absence of males within small broods is
probably weak.
Relatively independent from the fact if secondary manipulations of the sex
ratio occur, male numbers should be higher under conditions that allow
outbreeding, as has been shown in X. germanus (Peer and Taborsky 2004). In
case X. saxesenii mothers also allocate offspring sex according to outbreeding
opportunities, then less males should have been found in my lab study
compared to the field, where outbreeding opportunities can be expected to
arise regularly. Contrary to these expectations, I found a sex ratio of 1:8 (m:f)
on average, which is well below the average 1:20 sex ratio previously found in
the field both in Australia (Hosking 1972) and Switzerland (personal
observations). Although the methods for measuring sex ratios differed between
lab and field studies, these strongly contradicting results are not explainable by
methodological differences alone and may point to other (environmental)
factors influencing sex allocation in X. saxesenii.
Males hatched significantly later than their first sisters did, which contrasts
with findings in X. affinis, whose males hatch before females (Roeper et al.
1980). I hypothesize that in X. saxesenii fertilization of all females by their
brothers was ensured because females need about ten days to fully develop
(Biedermann et al. 2009), and males hatched on average 4 days after females.
Furthermore, fertilization of all females was made possible by the high sexual
activity of males that did not engage in social behaviors. Instead, they
constantly wandered the gallery in search for virgin females, which they started
to courtship (groom) upon encounter, followed
63
Chapter 4
by mounting and copulation. Although rarely seen, these copulations occurred
exclusively with immature females. Concordantly, males did not disperse as
long as new immatures eclosed and could be fertilized, but tended to leave the
natal nest as soon as eclosion ceased, presumably in search for outbreeding
opportunities. This indicates that they try to fertilize as many sisters as
possible, but search for outbreeding opportunities as soon as direct fitness
maximization within the natal nest is no longer possible.
Unfertilized females produced solely male broods, but did not mate with
their sons to subsequently produce "normal" mixed broods. This expectation
was based on observations of Coccotrypes dactyliperda Fabricius (Scolytinae)
females that do so (Herfs 1959). The behavioral difference between these two
species is, however, plausible when one takes into account the extreme
longevity and fertility of C. dactyliperda females, which raise up to five broods,
in this way producing up to 144 individuals in total. In contrast, X. saxesenii
females usually die soon after their first offspring matures (personal
observations), which in case they mated with one of their sons would mean
shortly after doing so. Even in case she was nevertheless able to lay a few
fertilized eggs before her death, the brood would still be doomed as the
mother's presence and her brood care appears crucial for successful brood
development (e.g. Norris 1993; Kirkendall et al. 1997; Biedermann 2007).
Another fascinating explanation could be that male broods are intended by the
females and represent an alternative reproductive tactic under certain
conditions when fitness gains through outbreeding sons are higher than those
that can be achieved with the production and maintenance of a mixed brood.
Such a condition may, for example, be an environment with a lot of small and
male-less broods (e.g., under high density), a situation in which selection would
favor a few females to just produce sons that visit their male-less neighboring
families.
My data shed light on the cryptic life and mating system of xyleborine
ambrosia beetles within their galleries. In the future, molecular studies will be
crucial for determining whether the observed inter- and intraspecific variance
in sex ratios is an adaption to outbreeding in some Xyleborini or a hint for the
existence of sex ratio distorters, for example, their symbionts. My data provide
a starting point for future studies dealing with factors that influence xyleborine
sex ratios and the frequency of outbreeding events. Ambrosia beetles are one
of the best model systems for studying the evolution of inbreeding,
haplodiploidy, sociality and symbioses, but at the same time one of the least
known. Before comparative studies on different species can be done to address
the issues mentioned, further basic data on the behaviors of these beetles
inside their galleries have to be collected.
Conclusion
My data suggest that males in Xyleborinus saxesenii Ratzeburg are extremely
successful in locating and fertilizing all their sisters in the natal gallery. On
average there is only
64
Chapter 4
about one male per family relatively independent of the number of sisters and
the maternal brood sex ratio. Apparently, this is sufficient to fertilize all
females, probably because the morphology of the gallery system with a central
"brood chamber" makes it easy for males to locate them. Only about 3% of the
females from broods with males disperse unfertilized from the natal nest and
subsequently produce all male broods, but do not mate with one of their sons
to produce a mixed brood afterwards.
Despite the fact that males are not the first to hatch within a brood, they do
so before the first females mature, and only disperse from the natal gallery
once the last female has finished her development. They do not engage in
gallery maintenance, gallery protection and brood care, but constantly wander
the gallery, presumably in search of unfertilized females. Although males
attempt to mate with all individuals independent of age and sex, they were
observed to copulate only with immature females.
Acknowledgements
I want to thank J. Hulcr and A. Cognato for giving me the chance to be part of
this festschrift. I hold S. L. Wood's accomplishments in great esteem, and I
regret that I have never had the chance to meet him in person. This manuscript
emerged at the Institute of Ecology and Evolution, University of Bern,
Switzerland, under the supervision of M. Taborsky. Additionally to discussions
with him, it profited a lot from suggestions made by T. Turrini, R. Roeper, K.
Peer, E. Ott, and two anonymous reviewers. During parts of the project, I was
financially supported by a grant of the Roche Foundation and a DOC grant of
the Austrian Academy of Science.
References
Batra LR (1967) Ambrosia fungi - A taxonomic revision and nutritional studies of some
species. Mycologia 59: 976-1017.
Batra LR, Michie MD (1963) Pleomorphism in Some Ambrosia and Related Fungi.
Transactions of the Kansas Academy of Science 66: 470-481.
Bernal JS, Gillogly PO, Griset J (2001) Family planning in a stemborer parasitoid: sex
ratio, brood size and size-fitness relationships in Parallorhogaspyralophagus
(Hymenoptera : Braconidae), and implications for biological control. Bulletin of
Entomological Research 91: 255-264.
Biedermann PHW (2007) Social behaviour in sib mating fungus farmers. Master thesis,
Berne, Switzerland: University of Berne.
Biedermann PHW, Klepzig KD, Taborsky M (2009) Fungus cultivation by ambrosia
beetles: Behavior and laboratory breeding success in three Xyleborine species.
Environmental Entomology 38 (4): 1096-1105.
Bischoff LL (2004) The social structure of the haplodiploid bark beetle, Xylosandrus
germanus. Diploma thesis, Berne, Switzerland: University of Berne.
65
Chapter 4
Borsa P, Kjellberg F (1996) Secondary sex ratio adjustment in a pseudo-arrhenotokous
insect, Hypothenemus hampei (Coleoptera: Scolytidae). Comptes Rendus del
Academie des Sciences Serie Iii-Sciences de la Vie-Life Sciences 319: 11591166.
Brader L (1964) Etude de la relation entre le scolyte des rameaux du cafeir, Xyleborus
compactus Eichh. (X. morstatti Hag.), et sa plantehote. Mededelingen van de
andbouwhogeschool te Wageningen 64: 1-109.
Brun LO, Borsa P, Gaudichon V, Stuart JJ, Aronstein K, Coustau C, Ffrench-Constant RH
(1995) Functional haplodiploidy. Nature 374: 506.
Charnov EL (1982) The Theory of Sex Allocation. Princeton University Press, Princeton,
355 pp.
Dyson EA, Hurst GDD (2004) Persistence of an extreme sex-ratio bias in a natural
population. Proceedings of the National Academy of Sciences of the United
States of America 101: 6520-6523.
Fischer M (1954) Untersuchungen über den kleinen Holzbohrer (Xyleborus saxeseni).
Pflanzenschutzberichte 12: 137-180.
Fisher RA (1930) The Genetical Theory ofNatural Selection. Oxford University Press,
Oxford, 370 pp.
Green RF, Gordh G, Hawkins BA (1982) Precise Sex-Ratios in Highly Inbred Parasitic
Wasps. American Naturalist 120: 653-665.
Haanstad JO, Norris DM (1985) Microbial symbiotes of the ambrosia beetle Xyletorinus
politus. Microbial Ecology 11: 267-276.
Hamilton WD (1967) Extraordinary sex ratios. Science 156: 477-488.
Hamilton WD (1978) Evolution and diversity under bark, In: Mound LA, Waloff N (Eds)
Diversity of insect faunas. Blackwell, Oxford, 154-175.
Herfs A (1959) Über den Steinnussborkenkäfer Coccotrypes dactyliperda F. Anzeiger für
Schädlingskunde - Journal of Pest Science 32: 1-4.
Hosking GB (1972) Xyleborus saxeseni, its life-history and flight behaviour in New
Zealand. New Zealand Journal for Forest Science 3: 37-53.
Kalshoven LGE (1962) Note on the habits of Xyleborus destruens Bldf., the near-primary
borer of teak trees on Java. Entomologische Berichten 22: 7-18.
Kawasaki YIM, Miura K, Kajimura H (2009) Superinfection of five Wolbachia in the alnus
ambrosia beetle, Xylosandrus germanus (Blandford) (Coleoptera:
Curuculionidae). Bulletin of Entomological Research 100: 231-239.
Kirkendall LR (1983) The evolution of mating systems in bark and ambrosia beetles
(Coleoptera: Scolytidae and Platypodidae). Zoological Journal of the Linnean
Society 77: 293-352.
Kirkendall LR (1993) Ecology and evolution of biased sex ratios in bark and ambrosia
beetles. In: Wrensch DL, Ebbert MA (Eds) Evolution and diversity of sex ratio in
insects and mites. Chapman and Hall, New York, 235-345.
Kirkendall LR, Kent DS, Raffa KF (1997) Interactions among males, females and
offspring in bark and ambrosia beetles: the significance of living in tunnels for
the evolution of social behavior. In: Choe JC, Crespi BJ (Eds) The Evolution of
Social Behavior in Insects and Arachnids. Cambridge University Press,
Cambridge, 181-215.
Mueller UG (2002) Ant versus fungus versus mutualism: Ant-cultivar conflict and the
deconstruction of the attine ant-fungus symbiosis. American Naturalist 160:
S67-S98.
66
Chapter 4
Mueller UG, Gerardo NM, Aanen DK, Six DL, Schultz TR (2005) The evolution of agriculture
in insects. Annual Review of Ecology Evolution and Systematics 36: 563—595.
Normark BB (2004) Haplodiploidy as an outcome of coevolution between male-killing cytoplasmatic elements and their hosts. Evolution 58: 790—798.
Normark BB, Jordal BH, Farrell BD (1999) Origin of a haplodiploid beetle lineage.
Proceedings of the Royal Society London B 266: 2253—2259.
Norris DM (1979) The mutualistic fungi of Xyleborini beetles. In: Batra LR (Ed) Nutrition,
Mutualism, and Commensalism. Allanheld, Osmun and Company, Montclair, 55—63.
Norris DM (1993) Xyleborus ambrosia beetles - a symbiotic ideal extreme biofacies with
evolved polyphagous privileges at monophagous prices. Symbiosis 14: 229—236.
Norris DM, Baker JK (1967) Symbiosis: effects of a mutualistic fungus upon the growth and
reproduction of Xyleborus ferrugineus. Science 156: 1120—1122.
Peer K, Taborsky M (2004) Female ambrosia beetles adjust their offspring sex ratio
according to outbreeding opportunities for their sons. Journal of Evolutionary Biology
17: 257—264.
Peer K, Taborsky M (2007) Delayed dispersal as a potential route to cooperative breeding
in ambrosia beetles. Behavioral Ecology and Sociobiology 61: 729—739.
Peleg B, Norris DM (1972) Bacterial symbiote activation of insect parthenogenetic
reproduction. Nature New Biology 236: 111—112.
R Development Core Team (2008) R: A language and environment for statistical
computing. Vienna, Austria, R Foundation for Statistical Computing.
Roeper RA, Treeful LM, O'Brien KM, Foote RA, Bunce MA (1980) Life history of the
ambrosia beetle Xyleborus affinis (Coleoptera: Scolytidae) from in vitro culture. Great
Lakes Entomologist 13: 141—144.
Roeper RA (1995) Patterns of mycetophagy in Michigan ambrosia beetles. Michigan
Academian 27: 153—161.
Schedl KE (1958) Breeding habits of arboricole insects in Central Africa. In: Becker EC (Ed)
Proceedings of the 10th International Congress of Entomology, Montreal (Canada),
August 1956, 183—197.
Schedl KE (1962) Scolytidae und Platypodidae Afrikas, II. Revista de Entomologia de
Mocamique 5: 1—594.
Schedl W (1966) Zur Verbreitung und Autokologie von Xyleborus eurygraphus Ratzeburg.
Bericht des Naturwissenschaftlichen - Medizinischen Vereins Innsbruck 54: 61—74.
Wood SL (1982) The bark and ambrosia beetles of North and Central America (Coleoptera:
Scolytidae), a taxonomic monograph. Great Basin Naturalist Memoirs 6: 1—1359.
67
Chapter 5
68
Chapter 5
Behav Ecol Sociobiol (2011) 65:1753-1761 DOI 10.1007/s00265-011-1183-5
Received: 30 August 2010/Revised: 14 March 2011 /Accepted: 31 March 2011 /Published online: 15 April 2011 ©
Springer-Verlag 2011
Abstract Body reserves may determine the reproductive output
of animals, depending on their resource allocation strategy. In
insects, an accumulation of reserves for reproduction is often
obtained before dispersal by pre-emergence (or maturation)
feeding. This has been assumed to be an important cause of
delayed dispersal from the natal nest in scolytine beetles. In the
cooperatively breeding ambrosia beetles, this is of special
interest because in this group delayed dispersal could serve two
alternative purposes: "selfish" maturation feeding or "altruistic"
alloparental care. To distinguish between these two possibilities,
we have experimentally studied the effect of delayed dispersal
on future reproductive output in the xyleborine ambrosia beetle
Xyleborus affinis. Females experimentally induced to disperse
and delayed dispersing females did not differ in their body
condition at dispersal and in their founding success afterwards,
which indicates that females disperse independently of
condition, and staying adult females are fully mature and would
be able to breed.
However, induced dispersers produced more offspring than
delayed dispersers within a test period of 40 days. This suggests
that delayed dispersal comes at a cost to females, which may
result primarily from alloparental care and leads to a reduced
reproductive output. Alternatively, females might have
reproduced prior to dispersal. This is unlikely, however, for the
majority of dispersing females because of the small numbers of
offspring present in the gallery when females dispersed,
suggesting that mainly the foundress had reproduced. In
addition, "gallery of origin" was a strong predictor of the
reproductive success of females, which may reflect variation in
the microbial complex transmitted vertically from the natal nest
to the daughter colony, or variation of genetic quality. These
results have important implications for the understanding of
proximate mechanisms selecting for philopatry and alloparental
care in highly social ambrosia beetles and other cooperatively
breeding arthropods.
.
.
.
Keywords Resource allocation Capital breeding Bark beetles
.
.
Sociality Fungus gardening Cooperative breeding
Communicated by N. Wedell
P. H. W. Biedermann (*) • M. Taborsky
Behavioural Ecology, Institute of Ecology and Evolution,
University of Berne,
Baltzerstrasse 6,
3012 Bern, Switzerland
*e-mail: [email protected]
Introduction
Living in groups involves three key decisions of totipotent
individuals (Helms Cahan et al. 2002): whether or not to disperse
(Stacey and Ligon 1991; Kokko and Ekman 2002), whether or not
to breed (Keller and Reeve 1994; Hager and Jones 2009) and
whether or not to help other group members to raise their
offspring (Eden 1987;Stacey and Koenig 1990). Dispersal, as the
first and most basal strategic decision, should depend on relative
fitness effects of staying and leaving, which are a function of
ecological conditions like resource availability, population
density and predation risk (Koenig et al. 1992; Heg et al. 2004;
P. H. W. Biedermann : K. D. Klepzig
USDA Forest Service, Southern Research Station,
2500 Shreveport Hwy,
Pineville, LA 71360, USA
K. D. Klepzig
US Forest Service, Southern Research Station,
200 WT Weaver Blvd, Asheville,
NC 28804, USA
69
Chapter 5
Bruintjes et al. 2010). Prior to dispersal, however, it is often
difficult to estimate potential costs imposed by external factors.
Therefore, only internal cues may be available such as body
condition (i.e. reserves), or local resource availability.
Studies of the fitness consequences of sociality typically
involve highly cooperative vertebrate and insect societies, in
which reproduction is restricted to only a few, specialized
individuals per group. An ideal model system to unravel the
importance of intrinsic factors and external causes selecting for
advanced sociality should, however, preferably comprise
totipotent individuals that flexibly engage in dispersal,
reproduction and helping depending on conditions (e.g. Costa
2006). These requirements are met by the polyphyletic ambrosia
beetles, which cultivate ambrosia fungi as food inside galleries
excavated in the wood of freshly dead trees (e.g. Schedl 1956;
Beaver 1989; Farrell et al. 2001). Dispersal from the natal group
in order to find an own gallery is associated with high fitness
costs, as suitable wood is patchily distributed and the success
rate of establishing a new fungus garden is low (about 20% in
Xyleborinus saxesenii Ratzeburg; Biedermann et al. 2009).
Therefore, if fungus productivity in the natal gallery is good and
thus optimal feeding conditions prevail, philopatry should be
favoured. In fact, such age-dependent, delayed dispersal of
scolytine beetles has been witnessed since a long time (e.g.
Eichhoff 1881; Whitney 1971; Botterweg 1982; Krausseopatz et
al. 1995; McNee et al. 2000). Three non-exclusive hypotheses for
this trait have been proposed:
1.
2.
(Kirkendall et al. 1997; Mueller et al. 2005). Evidence for
such cooperative brood care exists in X. saxesenii, where the
number of larvae and pupae is proportional to the number
of adult female helpers in a gallery (Peer and Taborsky
2007), and behavioural observations revealed cooperative
care (Biedermann 2007; Biedermann et al. 2009). If females
dispose of the potential to help raising broods but suffer
from limited fertility, they may benefit by caring for the
brood of relatives instead of taking the risk to disperse and
breed independently (West-Eberhard 1975; Craig 1983;
Roisin 1994). 3. Direct fitness benefits through reproduction
in the natal gallery
Staying adult daughters might also reproduce in the natal
gallery. This was observed in X. saxesenii, where one quarter
of the females were found to lay eggs in their natal nest
(Biedermann 2007).
In our study species, Xyleborus affinis Eichhoff (Xyleborini,
Scolytinae), daughters flexibly disperse over a period of about 50
days. Prior to dispersal, all females cooperatively care for the
brood and fungal cultures in their natal, commonly defended
gallery for at least 1 week (Roeper et al. 1980). Here we aim to
unravel whether prolonged philopatry of females causes costs or
benefits regarding body condition and future reproductive
success, i.e. to distinguish between hypothesis (1) and the two
alternatives (2 and 3) above. To this end, we experimentally
induced dispersal of adult females and measured their body
weight, size and reproductive success in comparison to a control
group that was allowed to disperse deliberately without
experimental interference. If maturation feeding occurred,
induced dispersers should be less successful in gallery foundation
and offspring production than voluntarily delayed dispersers
(hypothesis 1) because the former had stayed shorter in the
natal gallery than the latter before experimental collection. If in
contrast induced dispersers are more successful, this would
reveal that delayed dispersal is associated with a reduction in
body condition, indicating fitness costs by cooperative brood
care (hypothesis 2) or reproduction prior to dispersal (hypothesis
3). The latter possibility we checked by counting offspring
numbers at the time of collection and by determining the ovarian
status of all collected females.
Direct fitness benefits through maturation feeding
Delayed dispersal has been associated with preemergence or maturation feeding (Eichhoff 1881;
Botterweg 1982; McNee et al. 2000). Evidence for
maturation feeding exists from phloem-feeding mountain
pine beetles, where females were experimentally prevented from feeding after eclosion. They matured normally
but were less likely to breed successfully and laid smaller
eggs (Elkin and Reid 2005). Scolytine ambrosia beetles,
however, have a different feeding habit, using ambrosia
fungi as their sole nutritional source. Female ambrosia
beetles were found to lay eggs only after growing their own
fungus garden on which they fed (French and Roeper 1975;
Kingsolver and Norris 1977; Roeper et al. 1980; Beaver
1986). Hence, it is presently unclear whether reserves
accumulated before emergence will raise the productivity of
those beetles sufficiently to outweigh the fitness costs of
delayed dispersal.
Indirect fitness benefits through cooperative brood care
Females delaying dispersal could help to raise siblings,
especially since the fungus cultivation of ambrosia beetles
may benefit from the cooperation of several individuals
Material and methods
Study species
X. affinis is a tropical and subtropical member of the scolytine
subtribe Xyleborini, which are characterized by
70
Chapter 5
they inoculate the tunnel walls with ambrosia fungus spores
from their mycetangia, which results in a fungus layer lining the
tunnel walls within a week after insertion (Roeper et al. 1980).
At that time, females start to produce eggs, while feeding on
fungus providing the essential nutrients (Kingsolver and Norris
1977). The progeny passes through three larval instars and a
pupal stage. After eclosion, it takes a few days until the beetles
fully sclerotize. The first adult offspring appear in the tunnels
around 38 days after gallery foundation. The first daughters start
to disperse around day 50 after gallery foundation, but most of
them delay dispersal from their natal nest for much longer.
Daughters generally delay their dispersal: (1) The first developing
female offspring stay on average for 12 days after full
sclerotization before they start to disperse (Fig. 1; see also
Roeper et al. 1980); (2) also after dispersal has started, females
accumulate in the gallery as the rate of females eclosing from
the pupal stage is higher than the dispersal rate. About 80-90
days from gallery foundation, the medium deteriorates, which is
when production of new progeny has ceased and all individuals
leave the gallery (Fig. 1).
haplodiploidy, strongly female-biased sex ratios (X affinis: 1:8.5
males/females - Roeper et al. 1980; 1:15.2 males/ females Biedermann, unpublished data) and matings between full
siblings in their natal gallery (Xylosandrus germanus: Peer and
Taborsky 2004, 2005). Only female beetles disperse from their
host trees by flight. They transmit spores of species-specific
ambrosia fungi to the new gallery in a spore carrying organ
("mycetangium"; Francke-Grosmann 1956), or in some cases in
the hindgut (Francke-Grosmann 1975). X. affinis galleries deeply
penetrate the wood of deciduous trees with single tunnels
extending over 6 m that may be inhabited by several generations
for up to 4 years (Schneider 1987). Under laboratory conditions,
two to three generations may develop within a gallery, which
will host up to 100 individuals of all developmental stages
concurrently (Biedermann et al. 2009; Biedermann, personal
observation).
Preparation of artificial medium
We filled sterile glass tubes (18 mm diameter x 150 mm length;
Bellco Glass, Vineland, NJ, USA) with standard medium which
consists of 0.35 g streptomycin, 1 g Wesson's salt mixture, 5 g
yeast, 5 g casein, 5 g starch, 10 g sucrose, 20 g agar and 75 g oak
tree sawdust (Biedermann et al. 2009). All ingredients were
mixed and supplemented by 2.5 ml of wheat germ oil, 5 ml 95%
ethanol and 500 ml of deionised water. The mixture was
autoclaved for 20 min at 124°C and covered immediately with
sterile plastic caps (Bellco Glass kap-uts, Vineland, NJ, USA). After
the medium had cooled down, we scratched its surface with a
sterile scalpel to facilitate the onset of tunnel excavation by a
founder female. Then we closed the tubes again with the plastic
caps and left them to set for 4 to 5 days.
Experimental manipulations
We observed seven galleries of the second laboratory generation
during days 57-63 after gallery foundation and immediately
collected the first females emerging on the surface of the
medium to disperse (henceforth called delayed dispersers; Fig.
1). O n t h e s a m e d a y , w e dissected each of these
galleries and collected the same number of mature females (fully
sclerotized) from the
Laboratory rearing of the beetles
All females used in this study were descendants of the first
laboratory generation of females collected from oak logs in
Pineville, LA, USA (123 ft asl; 31°20', 92°24')in summer 2007.
Dispersing females emerged from their natal gallery on the
surface of the artificial medium and were collected for rearing of
consecutive laboratory generations (Biedermann et al. 2009).
Before starting a new gallery, females were surface-sterilized by
washing them first for a few seconds with 95% ethanol and then
with deionised water. Then each female was placed singly on the
prepared medium in a separate glass tube. These tubes were
reclosed with the caps and stored at room temperature (~23°C)
in darkness (wrapped in paper, but light could shine on the
entrance).
Following insertion on the medium, females immediately
started to bore a tunnel. As they penetrate the medium,
Fig. 1 Typical phenology of a laboratory gallery in X. affinis and the
timing of our manipulations. E1 start of egg laying of the founder female,
F first full sclerotization of a daughter, E2 start of daughter egg laying,
while the foundress usually continues to lay eggs, D first daughters start
to disperse from gallery, Ee end of egg laying because of deterioration of
the medium. We collected dispersing and staying females between days
57 and 63, either to take measurements (dry weight, body length, ovary
development) or to let them breed independently for 40 days until we
dissected the daughter galleries to count their offspring
71
Chapter 5
same gallery as had emerged on the surface (henceforth called
induced dispersers). Thus, we sampled two to 24 females per
gallery, resulting in a total of 38 induced and 37 delayed
dispersing females (one delayed disperser died during handling).
All delayed and induced dispersers were then inserted singly into
new tubes to find their own galleries. Forty days later, we
opened all galleries, checked whether a brood had been
produced and counted offspring numbers. In the analyses of
treatment effects on offspring numbers, we excluded females
that did not produce any offspring (N=11) and one family where
only one induced dispersing female reproduced at all (so no
comparison with a sibling delayed dispersing female was
possible).
females) on (1) the ability to found a gallery successfully by using
binomial error distributions and (2) the offspring numbers
(numbers of eggs, larvae, pupae and adults) by using normal
error distributions. In a GEE model, the correlation structure of
the non-independent measurements (i.e. the influence of gallery
identity) is modelled separately and is not of primary interest.
However, as breeding success and productivity appeared to be
highly variable between progeny of different galleries, we
additionally analysed the effects of gallery of origin. We used
gallery of origin as the sole explanatory variable in a logistic
regression (LR) to determine its influence on breeding success
and in a general linear model (GLM) to determine its influence
on productivity. Another GEE model was performed to analyse
dry weight differences between delayed and induced dispersing
females. Statistical analyses were performed with SPSS (Version
15.0, ©SPSSInc., Chicago, IL,USA,1989-2005) and R (R
Development Core Team 2008). Model coefficients are reported
as B±standard error of the estimate (SE) throughout, with the
induced dispersers as the reference category (coefficient set to
zero). Significance level α=0.05.
Measurements of delayed and induced dispersing females
At the time we started our experimental manipulations, we also
dissected eight galleries and stored all dispersing and staying
females (equivalent to the delayed and induced dispersing
females) in 95% ethanol. At a later date, we dry-weighed all
females from four of these galleries with a high-precision scale
(precision 0.01 mg; Sartorius ME215S-OCE, Göttingen, Germany)
after a 24-h drying process (oven, 80°C). We measured their
body length to the nearest 0.01 mm using a microscope (x6.4x40 magnification) with an ocular micrometer. Using the same
microscope, we also dissected the ovaries of all females from the
remaining four galleries from the dorsal side with high-precision
tweezers. We classified ovaries as either immature (ovaries
rudimentarily developed), mature (fully developed ovaries but
no oocytes) or egg carrying (four ovarioles containing one or
more oocytes; see figures in Fischer 1954).
Results
Experimental manipulations
Of the 75 experimental foundresses, 64 produced a brood
successfully. Treatment did not affect whether a brood was
successfully produced or not (seven of 37 delayed dispersing
females vs. four of 38 induced dispersing females failed to
produce a brood; GEE: B±SE 0.653±
2
0.546, χ = 1.429, df =1, P=0.232), and there was no effect of
2
gallery identity on founding success (LR: χ =0.001, df=1,
P=0.97, N=75).
Of the 64 successful foundresses, induced dispersers
produced more offspring than delayed dispersers (GEE: B± SE
2
9.137±4.426, χ =4.263, df=1, P=0.039), which rejects the
maturation feeding hypothesis. This relationship was observed in
four out of six galleries (Fig. 2a). If the total offspring numbers
were split up between different developmental stages, it became
clear that this result was solely caused by the variance in the
2
numbers of laid eggs (GEE: B± SE 5.175±1.6, χ =10.466, df=1,
P=0.001), but not by the numbers of larvae (GEE: B±SE
2
4.263±3.797, χ =1.26, df=1, P=0.262), pupae (GEE: B±SE
2
2.756±2.589, χ =1.133, df=1, P=0.287) and adults (GEE: B±SE
2
2.703±3.492, χ =0.599, df=1, P=0.439) present 40 days after the
treatment (Fig. 2b). This difference in egg numbers is not
Statistical analyses
Our nested design with variable numbers of repeated measures
from the experimental galleries generated matched, nonindependent measurements. Linear mixed models may be used
when the distribution of the repeated responses for a subject
has a multivariate normal distribu-tion. This is unlikely when the
dependent variable is binary or count data (Norusis 2007).
Therefore, generalized estimating equations (GEE), which are an
extension of generalized linear models, were used to analyse
effects of dependent variables on correlated binary or count
response variables (Liang and Zeger 1986; Zeger and Liang 1986).
We used GEEs with an exchangeable correlation structure of the
response variable within a cluster (= gallery identity) to identify
effects of our treatment (induced vs. delayed dispersing
72
Chapter 5
Fig. 2 Comparison of total numbers of offspring produced by the two
experimental groups of females 40 days after the treatment. a Offspring
numbers in dependence of the gallery of origin and the overall total.
Numbers of daughter galleries included (induced dispersers/delayed
dispersers): gallery1 (2/2), g2 (5/6), g3 (5/5), g4 (5/2), g5 (4/4) and g6
(12/11); g7 was omitted because only one female of this gallery
reproduced successfully. b The overall total is split up
in the four developmental offspring stages. The differences between the
numbers of eggs, larvae and pupae reflect the average duration of these
stages (for X. saxesenii: egg stage - 5 days (range=5), three larval instars 11 days (range=8-21), pupal stage - 7 days (range=6-7); Biedermann et al.
2009). Arithmetic means of daughter gallery offspring numbers are
shown with their standard errors. GEE: *P< 0.05; **P<0.001
dispersing females never had eggs in their ovaries (14 adult
females from four galleries).
an artefact from a few galleries, since we found the mean egg
numbers to be higher in the induced dispersers than in the
delayed dispersers in all six galleries analysed (Fig 3 in the
"Appendix"). The total number of offspring produced by
successful foundresses was strongly affected by the gallery of
origin (GLM: F5,,63 = 5.6, P<0.001).
Discussion
The first developing female offspring in all X. affinis galleries
stayed and helped in brood care for at least
Measurements of delayed and induced dispersing females
Body length and weight did not differ significantly between
induced (x = 2.85mm, SE < 0.01;x = 0.41mg, SE = 0.01, N = 31)
and delayed dispersing females (x = 2.85mm, SE < 0.01;x =
0.4mg, SE = 0.02, N = 13; GEE body length: no variability,
2
P>0.05; GEE body weight: B±SE -0.03±0.52, χ ="0.06, df=1,
P=0.95; see Fig. 4 in the "Appendix").
About 24% of the induced dispersing females had ovaries
containing eggs (14 of 59 adult females from four galleries;
Table 1), which suggests that staying females often produce
eggs in their natal gallery. The number of delayed dispersers
was independent of the number of staying females (= induced
dispersers; Spearman: R=-0.32, P=0.68) and of the number of
females among them laying eggs (Spearman: R=-0.21, P= 0.79,
N=4 galleries). The likelihood to breed differed between
galleries (Fisher's exact test: P=0.03, N=59). Delayed
Ovary
development
A
Gallery
B C
D
Total
(%)
Dispersing
females
Immature
Mature
Egg-carrying
Total
0
2
0
2
1
0
0
1
8
0
0
8
0
3
0
3
64.3
35.7
0
100
Staying females
Immature
Mature
Egg-carrying
Total
2
6
1
9
2
2
5
9
3
1
3
7
13
16
5
34
33.9
42.4
23.7
100
Table 1 Developmental status of the ovary of delayed and induced
dispersing females in four dissected galleries
73
Chapter 5
1week (cf. Roeper et al. 1980) and on average for 12 days after
full sclerotization before they started to disperse (Fig. 1). Also
after dispersal has started, the rate of female eclosion from the
pupal stage is higher than the dispersal rate, which causes
females to accumulate in the gallery. This suggests that females
usually delay dispersal, probably in order to help in care of the
brood produced by the foundress (i.e. their mother) or later on
potentially also by sisters.
Given that females delay their dispersal, our data reject the
predictions of the maturation feeding hypothesis. In contrast,
the predictions of the other two hypotheses were confirmed:
Delayed dispersal seems to entail long-term costs, most likely by
the investment in cooperative care, and some females reproduce
in their natal gallery. The data reveal also that philopatric
females are fully capable to found an own gallery and to start
reproducing at any point of time, disproving the assumption that
they are infertile before dispersal (Graham 1961). Hitherto,
prolonged philopatry, which is a common feature of female life
histories in many phloem (e.g. Kirkendall et al. 1997) and
ambrosia feeding Scolytinae (e.g. Kalshoven 1962; Peer and
Taborsky 2007;Biedermannetal.2009), has been attributed
usually to maturation feeding (Eichhoff 1881; Botterweg 1982;
McNee et al. 2000). In contrast, our findings indicate that in X.
affinis other benefits select for philopatry. Firstly, philopatry can
raise the inclusive fitness of females by enhancing the
production of close relatives (Bischoff 2004; Biedermann 2007;
Peer and Taborsky 2007).
Galleries with more adult females produced more offspring in
the closely related species X. saxesenii (Peer and Taborsky 2007),
also over multiple generations (Biedermann et al. 2009). Direct
observations of X. affinis and other xyleborine species revealed
that staying adult females share in brood care and fungus
maintenance (Roeper et al.
1980; Bischoff 2004; Biedermann 2007; Biedermann et al. 2009).
The members of a gallery are almost clones due to mating
occurring virtually exclusively among full siblings (for X.
germanus: sib-mating estimate = 97% of matings, Keller et al.,
submitted for publication; see also Peer and Taborsky 2005).
Secondly, direct fitness benefits can apply for females that
reproduce at home, which was true for nearly a quarter of
staying females in our sample. This is a similar proportion as
observed in X. saxesenii (Biedermann 2007).
There might be alternative causes to costly helping for the
lowered productivity of delayed dispersers: (1) The reduced
post-dispersal productivity of females may have resulted from
their
own egg laying in the natal gallery. However, it is unlikely that
this would fully explain the productivity differences found
between the two groups of females as the total numbers of eggs,
larvae, pupae and teneral beetles present in the galleries in
relation to the total number of adult females were very small at
the time of our treatment (day 60: mean= 2.57, SE=1.15,
range=0-8, N=7 galleries), suggesting that there was only very
sparse egg laying, if any, by females other than the foundress. (2)
The delayed dispersing females may have developed flight
muscles for dispersal which the induced dispersers might have
lacked; the reuse of energy when transferring nutrients from
flight muscles to ovaries may bring about extra synthesis costs
(Zera and Denno 1997), which could affect productivity (e.g. Elkin
and Reid 2005). This possibility needs to be scrutinised in future
studies focusing on the energetics of dispersal, reproduction and
reorganisation of tissue in these beetles.
Independently of whether females might benefit from staying
by raising their inclusive or direct fitness, dispersal, by contrast, is
costly due to high mortality risk (Milne and Giese 1970; Dahlsten
1982). In addition, establishing a fungus culture after dispersal
fails very often (only 20% of founded fungus gardens are
successful in X. saxesenii; Peer and Taborsky 2007; Biedermann
et al. 2009). Therefore, it is conceivable that females might
compete for staying in a productive nest (Kokko and Ekman
2002). If this was the case, dispersing females could be the less
competitive individuals, which might explain why they were less
successful in their breeding attempts after dispersal when
compared to philopatric females. However, two results indicate
that competition for staying does not occur or has little effect on
dispersal in X. affinis. First, we found no difference in size or body
condition between dispersing and staying females. Second, the
numbers of females present in the gallery (including egg layers
and non-reproductives) did not relate to the numbers of
dispersers in our experiment. A relationship between female
density and dispersal would be expected, however, if
competition triggers dispersal, regard-less whether dispersal
occurs voluntarily or is forced by other colony members.
Delayed dispersing females differed in their ovary development from those collected in the gallery, as egg-carrying
ovaries were only found among the latter. This might suggest
two different reproductive strategies of females: (1) A dispersal
phenotype showing delayed ovary maturation and perhaps also a
strongly developed flight apparatus and (2) a philopatric
phenotype that stays, helps in brood care and eventually breeds
in the natal gallery. Such strategies are common in insects
because there is often a trade-off between
74
Chapter 5
construction of the flight apparatus and ovary development
(Zera and Denno 1997). Nevertheless, the existence of two
distinct phenotypes is yet unknown from scolytine beetles, and
we also do not find evidence for this possibility in our data. First,
we did not find any differences in body size and weight, which
contrasts with theoretical predictions that dispersers should be
the stronger competitors under kin competition (Gyllenberg et
al. 2008; Bonte and De La Pena 2009); this has been confirmed
also in several vertebrate (e.g. Kawata 1987; Cote et al. 2007)
and insect taxa (e.g. Sundstrom 1995; Moore et al. 2006). Body
condition is an important factor determining successful host
finding and gallery foundation in scolytine beetles (Dendroctonus
ponderosae; Latty and Reid 2010). Second, selection for fixed
behavioural strategies, which would be associated with a
dispersal polymorphism, is probably weak for Scolytinae because
of their unpredictable environmental conditions. Indeed, there is
a multitude of studies showing their enormous flexibility to react
adaptively to changing environ-mental conditions, for example,
by transferring nutrients between flight muscles and ovaries
back and forth within a few days (e.g. Reid 1958; McNee et al.
2000). This might explain why both female groups showed the
same founding success and about the same timing of first egg
laying (reflected by the equal number of adult offspring in both
treatments), which would be unlikely assuming two distinct
phenotypes. In summary, although there is variation in
reproductive success and probably flight performance, dis-persal
and philopatry are likely rather plastic, condition-dependent
strategies which are predicted to be superior to fixed strategies
in many cases (e.g. Ims and Hjermann 2001; Bowler and Benton
2005).
"Gallery of origin" significantly affected both the number of
offspring produced after dispersal and the number of egg layers
among staying females. It has been suggested that the
mutualistic microbial complex (certain fungi and bacteria)
maintained in the gallery is the major factor influencing gallery
productivity via the quality of the transmitted mutualistic
microbes (Baker and Norris 1968; Kok et al. 1970; Kingsolver and
Norris 1977; Batra 1979; Kajimura and Hijii 1994). The beetles'
genetic quality might be another factor that has not yet been
explored. Unfortunately, our approach does not allow
distinguishing between these two potential causes of the
observed gallery effects.
Xyleborine ambrosia beetles live under the very conditions
where higher sociality has probably evolved multiple times
(Hamilton 1978), which includes high levels of inbreeding,
protection of a common nest and a virtually unlimited food
source (e.g. some hymenoptera, gall-thrips, aphids and lower
termites: Choe and Crespi 1997; Korb and Heinze 2008).
Therefore, this group can provide insights into intrinsic and
ecological factors inducing individuals to stay at home rather than
to disperse and to help rather than to reproduce independently,
which are basic components of social evolution.
Acknowledgements We are grateful to Stacy Blomqvist and Eric Ott for
collecting X. affinis in the field and starting the first lab galleries. This
manuscript benefitted greatly from comments of Tabea Turrini, Dik Heg
and two anonymous reviewers. The study was supported by a cooperative
agreement between the Department of Behavioral Ecology, University of
Bern and the Southern Research Station, USDA Forest Service. PHWB is a
recipient of a DOC fellowship of the Austrian Academy of Sciences at the
Department of Behavioural Ecology, University of Bern, and was partly
funded by a fellowship of the Roche Research Foundation.
Appendix
Fig. 3 Total numbers of eggs produced by the two experimental groups of
females 40 days after the treatment. Induced dispersers laid more eggs
than delayed dispersers overall (GEE: P<0.001). Numbers of daughter
galleries included (induced dispersers/delayed dispersers): galley (2/2), g2
(5/6), g3 (5/5), g4 (5/2), g5 (4/4) and g6 (12/11). Arithmetic means are
shown with their standard errors
75
Chapter 5
Fig. 4 Comparison of the body weight (mg) of females from the two
experimental groups collected in four galleries. There was no significant
difference between induced and delayed dispersers (GEE: P=0.95).
Numbers of galleries included (induced dispersers/delayed dispersers):
gallerya (9/2), gb (12/7), gc (4/1) and gd (6/3). Arithmetic means of dry
weight are shown with their standard errors
References
Baker JM, Norris DM (1968) A complex of fungi mutualistically involved in
nutrition of ambrosia beetle Xyleborus ferrugineus . J Inv Path
11:246-250
Batra LR (1979) Insect-fungus symbiosis. Nutrition, mutualism, and
commensalism. Wiley, New York
Beaver RA (1986) The taxonomy, mycangia and biology of Hypothenemus
curtipennis (Schedl), the first known cryphaline ambrosia beetle
(Coleoptera: Scolytidae). Ent Scand 17:131-135
Beaver RA (1989) Insect-fungus relationships in the bark and ambrosia
beetles. In: Wilding N, Collins NM, Hammond PM, Webber JF
(eds) Insect-fungus interactions. Academic, London, pp 121 -143
Biedermann PHW (2007) Social behaviour in sib mating fungus farmers.
Master thesis, University of Bern
Biedermann PHW, Klepzig KD, Taborsky M (2009) Fungus cultiva-tion by
ambrosia beetles: behavior and laboratory breeding success in
three xyleborine species. Environ Entomol 38:1096-1105
Bischoff LL (2004) The social structure of the haplodiploid bark beetle,
Xylosandrus germanus. Diploma thesis, University of Bern
Bonte D, De La Pena E (2009) Evolution of body condition-dependent
dispersal in metapopulations. J Evol Biol 22:1242-1251
Botterweg PF (1982) Dispersal and flight behavior of the spruce bark beetle
Ips typographus in relation to sex, size and fat-content. Z Angew
Entomol 94:466-489.
76
Bowler DE, Benton TG (2005) Causes and consequences of animal
dispersal strategies: relating individual behaviour to spatial
dynamics. Biol Rev 80:205-225
Bruintjes R, Hekman R, Taborsky M (2010) Experimental global food
reduction raises resource acquisition costs of brood care
helpers and reduces their helping effort. Funct Ecol 24:10541063.
Helms Cahan S, Blumstein DT, Sundstrom L, Liebig J, Griffin A (2002)
Social trajectories and the evolution of social behaviour. Oikos
96:206-216
Choe JC, Crespi BJ (1997) The evolution of social behaviour in insects
and arachnids. Cambridge University Press, Cambridge.
Costa JT (2006) The other insect societies. Belknap Press of Harvard
University Press, Cambridge.
Cote J, Clobert J, Fitze PS (2007) Mother-offspring competition promotes
colonization success. PNAS 104:9703-9708
Craig R (1983) Subfertility and the evolution of eusociality by kin
selection. J Theor Biol 100:379-397
Dahlsten DL (1982) Relationship between bark beetles and their natural
enemies. In: Mitton JB, Sturgeon KB (eds) Bark beetles in
North American conifers. University of Texas Press, Austin,
pp 140-182
Eden SF (1987) When do helpers help - food availability and helping in
the moorhen, Gallinula-chloropus. Behav Ecol Sociobiol
21:191-195
Eichhoff W (1881) Die Europäischen Borkenkäfer. Springer, Berlin
Elkin CM, Reid ML (2005) Low energy reserves and energy allocation
decisions affect reproduction by mountain pine beetles,
Dendroctonus ponderosae. Funct Ecol 19:102-109
Farrell BD, Sequeira AS, O'Meara BC, Normark BB, Chung JH, Jordal
BH (2001) The evolution of agriculture in beetles
(Curculionidae: Scolytinae and Platypodinae). Evolution
55:2011-2027
Fischer M (1954) Untersuchungen über den kleinen Holzbohrer
(Xyleborus saxeseni). Pflanzenschutzberichte 12:137-180
Francke-Grosmann H (1956) Hautdrüsen als Träger der Pilzsymbiose bei
Ambrosiakäfern. Z Morph Ökol Tiere 45:275-308
Francke-Grosmann H (1975) Zur epizoischen und endozoischen
Übertragung der symbiotischen Pilze des Ambrosiakäfers
Xyleborus saxeseni (Coleoptera: Scolitidae). Entomologica
Germanica 1:279-292
French JRJ, Roeper RA (1975) Studies on the biology of the ambrosia
beetle Xyleborus dispar. J Appl Entomol 78:241-247
Graham K (1961) Air-swallowing—mechanism in photic reversal of the
beetle Trypodendron. Nature 191:519-520
Gyllenberg M, Kisdi E, Utz M (2008) Evolution of condition-dependent
dispersal under kin competition. J Math Biol 57:285-307
Hager R, Jones CB (2009) Reproductive skew in vertebrates: proximate
and ultimate causes. Cambridge University Press, Cambridge
Hamilton WD (1978) Evolution and diversity under bark. In: Mound LA,
Waloff N (eds) Diversity of insect faunas. Blackwell, Oxford,
pp 154-175
Heg D, Bachar Z, Brouwer L, Taborsky M (2004) Predation risk is an
ecological constraint for helper dispersal in a cooperatively
breeding cichlid. Proc R Soc Lond B 271:2367-2374
Ims RA, Hjermann DO (2001) Condition-dependent dispersal. In: Clobert
J, Danchin E, Dhondt AA, Nichols JD (eds) Dispersal. Oxford
University Press, Oxford, pp 203-216
Kajimura H, Hijii N (1994) Reproduction and resource utilization of the
ambrosia beetle, Xylosandrus mutilatus, in-field and
experiment populations. Entomol Exp Appl 71:121-132
Kalshoven LGE (1962) Note on the habits of Xyleborus destruens Bldf.,
the near-primary borer of teak trees on Java. Entomol
Berichten 22:7-18
Chapter 5
Kawata M (1987) The effect of kinship on spacing among female
red-backed
voles,
Clethrionomys
rufocanus
bedfordiae. Oecologia 72:115-122
Keller L, Reeve HK (1994) Partitioning of reproduction in animal
societies. Trends Ecol Evol 9:98-102
Kingsolver JG, Norris DM (1977) The interaction of Xyleborus
ferrugineus Fabr. (Coleoptera: Scolytidae) behavior
and initial reproduction in relation to its symbiotic
fungi. Ann Entomol Soc Am 70:1-4
Kirkendall LR, Kent DS, Raffa KF (1997) Interactions among
males, females and offspring in bark and ambrosia
beetles: the significance of living in tunnels for the
evolution of social behavior. In: Choe JC, Crespi BJ
(eds) The evolution of social behavior in insects and
arachnids. Cambridge University Press, Cambridge,
pp 181 -215
Koenig WD, Pitelka FA, Carmen WJ, Mumme RL (1992) The
evolution of delayed dispersal in cooperative
breeders. Q Rev Biol 67:111-150
Kok LT, Norris DM, Chu HM (1970) Sterol metabolism as a basis
for a mutualistic symbiosis. Nature 225:661 -662
Kokko H, Ekman J (2002) Delayed dispersal as a route to
breeding: territorial inheritance, safe havens, and
ecological constraints. Am Nat 160:468-484
Korb J, Heinze J (2008) The ecology of social life: a synthesis. In:
Korb J, Heinze J (eds) Ecology of social evolution.
Springer, Berlin, pp 245-259
Krausseopatz B, Kohler U, Schopf R (1995) The energetic state
of lps typographus L. (Coleoptera, Scolytidae) during
the life-cycle. Z Angew Entomol 119:185-194
Liang KY, Zeger SL (1986) Longitudinal data-analysis using
generalized linear-models. Biometrika 73:13-22
Latty TM, Reid ML (2010) Who goes first? Condition and danger
dependent pioneering in a group-living bark beetle
(Dendroctonus ponderosae). Behav Ecol Sociobiol
64:639-646
McNee WR, Wood DL, Storer AJ (2000) Pre-emergence feeding
in bark beetles (Coleoptera: Scolytidae). Environ
Entomol 29:495-501
Milne DH, Giese RL (1970) Biology of the Columbian timber
beetle,
Corthylus
columbianus
(Coleoptera:
Scolytidae). 10. Comparison of yearly mortality and
dispersal losses with population densities. Entomol
News 81:12-24
Moore JC, Loggenberg A, Greeff JM (2006) Kin competition
promotes dispersal in a male pollinating fig wasp.
Biol Lett 2:17-19
Mueller UG, Gerardo NM, Aanen DK, Six DL, Schultz TR (2005)
The evolution of agriculture in insects. Ann Rev Ecol
Evol Syst 36:563-595
Norusis M (2007) Generalized estimating equations. In: Norusis
M (ed) SPSS 15.0 advanced statistical procedures
companion. Prentice Hall, Upper Saddle River, pp
279-296
Peer K, Taborsky M (2004) Female ambrosia beetles adjust their
offspring sex ratio according to outbreeding
opportunities for their sons. J Evol Biol 17:257-264
Peer K, Taborsky M (2005) Outbreeding depression, but no
inbreeding depression in haplodiploid ambrosia
beetles with regular sibling mating. Evolution
59:317-323
Peer K, Taborsky M (2007) Delayed dispersal as a potential
route to cooperative breeding in ambrosia beetles.
Behav Ecol Sociobiol 61:729-739
R Development Core Team (2008) R: a language and
environment for statistical computing. R Foundation
for Statistical Computing, Vienna
Reid RW (1958) The behaviour of the mountain pine beetle,
Dendroctonus monticolae Hopk., during mating, egg
laying, and gallery construction. Can Entomol
90:505-509
Roeper R, Treeful LM, O'Brien KM, Foote RA, Bunce MA (1980)
Life history of the ambrosia beetle Xyleborus affinis
(Coleoptera: Scolytidae) from in vitro culture. Great
Lakes Entomol 13:141-144
Roisin Y (1994) Intragroup conflicts and the evolution of sterile
castes in termites. Am Nat 143:751-765
Schedl KE (1956) Breeding habits of arboricole insects in Central
Africa. Verh 10 Int Kongr Entomol (Vienna) 1:183197
Schneider I (1987) Distribution, fungus-transfer and gallery
construction of the ambrosia beetle Xyleborus affinis
in comparison with X. mascarensis (Coleoptera,
Scolytidae). Entomologia Generalis 12:267-275
Stacey PB, Koenig WD (1990) Cooperative breeding in birds:
long-term studies of ecology and behavior.
Cambridge University Press, Cambridge
Stacey PB, Ligon JD (1991) The benefits-of-philopatry
hypothesis for the evolution of cooperative
breeding: variation in territory quality and group size
effects. Am Nat 117:831-846
Sundstrom L (1995) Dispersal polymorphism and physiological
condition of males and females in the ant, Formica
truncorum. Behav Ecol 6:132-139
West-Eberhard MJ (1975) The evolution of social behaviour by
kin selection. Q Rev Biol 50:1-33
Whitney HS (1971) Association of Dendroctonus ponderosae
with blue stain fungi and yeasts during brood
development in lodgepole pine. Can Entomol
103:1495-1503
Zeger SL, Liang KY (1986) Longitudinal data-analysis for discrete
and continuous outcomes. Biometrics 42:121 -130
Zera AJ, Denno RF (1997) Physiology and ecology of dispersal
polymorphism in insects. Ann Rev Entomol 42:207230
77
Chapter 6
78
Chapter 6
Social Fungus Farming Varies among Ambrosia Beetles
Peter H.W. Biedermann1,2 and M. Taborsky1
1
Department of Behavioural Ecology, Institute of Ecology & Evolution, University of Berne,
Baltzerstrasse 6, CH-3012 Bern, Switzerland
2
USDA Forest Service, Southern Research Station, 2500 Shreveport Hwy, Pineville/LA, 71360, USA
Correspondance: [email protected]
Running title: Social fungus farming by ambrosia beetles
Abstract
Advanced fungus agriculture independently evolved once in ants, once in termites, and seven
times in weevils. In the latter, the so-called ambrosia beetles, this evolution was accompanied by an
increase of social complexity leading to division of labour. Ambrosia beetles live in subsocial or
eusocial family groups within self-built galleries in wood, where they farm ambrosia fungi for food.
Gallery morphologies vary considerably between species, which is assumed to affect the frequency of
interactions between group members, i.e. adults and brood, as well as their mutualistic ambrosia
fungi.
Here we illustrate the intra- and interspecific variation in ambrosia beetle gallery morphology
and describe how this affects within-family interactions in two exemplary species. Xyleborinus
saxesenii inhabits cave-like galleries, whereas Xyleborus affinis (like most other ambrosia beetles)
digs out branching tunnel systems. Life histories of both species are very similar, but mature X. affinis
engage less in brood care and fungus cropping relative to mature X. saxesenii. Remarkably, however,
reduced brood care by adults is compensated with increased grooming by larvae in X. affinis, which is
likely an effect of the spatial separation between adults and brood as well as fungus in the tunnel
systems. By contrast, in the caves of X. saxesenii individuals of all age classes are mixed and closely
interact with the fungal layers lining the gallery walls. The cave-like gallery pattern results from
larvae feeding on wood and fungus instead of just grazing off the fungus layer from tunnel walls. Two
fundamentally different ways of fungus agriculture are thereby indicated to exist in the subtribe
Xyleborini (Scolytinae) and much more variation may still be uncovered in other ambrosia beetle
lineages.
Introduction
The cultivation of crops for nourishment is not restricted to humans. Three lineages of
insects – fungus-growing ants (Attini), fungus-growing termites (Macrotermitinae), and ambrosia
beetles (Curculionidae: Scolytinae, Platypodinae) – evolved the skills for fungus agriculture. This
includes habitual planting or insemination, cultivation, protection and harvesting of the cultivar, on
which they inevitably depend for nutrition (Mueller et al. 2005). These complex tasks are shared
between different individuals in a gallery. Sociality and division of labour has therefore been
proposed to have played a crucial role in the evolution of fungus agriculture (Mueller et al. 2005;
Biedermann and Taborsky 2011).
Eusociality and division of labour have apparently facilitated the evolution of agriculture in
ants and termites, because agricultural tasks may be partitioned between group members
(Hölldobler and Wilson 1990; Hart et al. 2002). Different worker castes in ants and termites are
specialized in either foraging, chopping and cleaning the substrate before incorporation into the
79
Chapter 6
garden, planting of mycelium onto new substrate, monitoring and weeding of the garden, and
disposal of diseased or senescent parts of the garden (Traniello and Leuthold 2000; Bot et al. 2001;
Hart et al. 2002; Mueller et al. 2005). Ambrosia beetle agriculture must have originated differently,
because they originate from a non-social lineage with only parental care (Farrell et al. 2001).
Interestingly, the transition to a fungal diet evolved convergently at least eight times in weevils
(Farrell et al. 2001). In all three clades that have hitherto been examined more closely – Xyleborini,
Xyloterini, and Platypodinae – this transition was accompanied by an increase in social complexity
(Kirkendall et al. 1997). Female offspring often delay dispersal from the natal nest, resulting in
overlapping generations (e.g. Merkl and Tusnadi 1992; Peer and Taborsky 2007; Biedermann et al.
2009; 2012), and a recent direct behavioural study revealed cooperative breeding and a complex
system of task sharing between adult and larval stages in Xyleborinus saxesenii (Scolytinae:
Xyleborini; (Biedermann and Taborsky 2011). Eusociality, i.e., permanently sterile and reproductive
castes, was not observed in this species, but was suggested in the ambrosia beetle Austroplatypus
incompertus (Platypodinae) after examination of morphological differences between adult females
(Kent and Simpson 1992).
Social evolution is generally triggered by constraints on independent breeding and / or
benefits of philopatry (e.g. Selander 1964; Stacey and Ligon 1987; Kokko and Ekman 2002; Korb
2008). In ambrosia beetles constraints on independent breeding are probably immense. Freshly dead
wood, where ambrosia beetles culture their fungi, is a patchily distributed resource, and mortality
due to predation or adverse environmental conditions during dispersal is assumed to be very high
(Milne and Giese 1970; Dahlsten 1982). Once a suitable host tree has been detected, a founder
beetle has to excavate a gallery, overcome tree defences like resins, and successfully establish the
ambrosia fungus, which fails in 70-80% of all females (e.g. Fischer 1954; Biedermann 2007; Peer and
Taborsky 2007). By contrast, benefits of philopatry in a productive natal gallery may include fitness
gains by (1) co-breeding and (2) helping close relatives to produce more offspring (Biedermann et al.
2011). All Xyleborini are predisposed for kin-selected benefits as a result of haplodiploidy and
obligate brother-sister matings in the natal gallery (Peer and Taborsky 2005), which lead to high
genetic relatedness (Peer and Taborsky 2007). Indeed, delayed dispersal of daughters appears to be
the rule in Xyleborini, which help their mothers with gallery protection and hygiene, brood and
fungus care (Biedermann et al. 2012). The participation of larvae in these cooperative tasks is unique
among Holometabola (Biedermann and Taborsky 2011).
The beetles’ agriculture differs from that of other insects in one important feature: the
longevity/stability of the substrate on which that fungi grow. Ants and termites collect the substrate
(woody branches, leaves and grass) for their crops while the beetles live within the substrate that os
consumed by the fungus (wood). In the first case, substrate is renewable and fungus gardens are
potentially long-lasting, while in the second case resources are locally exhaustible as nutrients wear
out (Kirkendall et al. 1997). Thus, ambrosia beetles that culture their fungi in tunnel systems
(galleries) in the sap- and heartwood of trees need to constantly enlarge the system to continuously
provide new substrate for their fungi. In addition, most ambrosia beetles settle in freshly dead trees,
where their fungi are faced with an increased competition from other microorganisms over time.
Recycling of beetle excretions by fungi (French and Roeper 1973; Norris 1975), or hygienic
behaviours (Cardoza et al. 2006) and bacteria producing antibiotics (Scott et al. 2008) may postpone
the spread of competing microorganisms, but their conquest cannot be hindered completely. Only
the colonization of gradually dying trees, which sometimes occurs in X. saxesenii and X. affinis, may
enable continuous breeding and several beetle generations to stay and overlap within one gallery;
single X. saxesenii galleries with more than 300 adult inhabitants have been reported (Hosking 1972)
and tunnels of a single X. affinis gallery may be inhabited for more than four years and reach a total
length of 6m (Schneider 1987). Nevertheless, such cases are probably rare and not the rule, and
breeding in dead wood certainly constrains the longevity of fungus gardens and the evolution of
social complexity also in these two species. This may explain why eusociality has yet been suggested
80
Chapter 6
only for A. incompertus, a species breeding in long-lasting galleries (> 36 years) in living trees (Kent
and Simpson 1992).
Another important factor potentially affecting ambrosia beetle sociality is the morphology of
their galleries. A gallery made of a branching tunnel system is typical for the majority of Xyleborini
and Platypodinae. Whilst adult stages are apparently randomly spaced, their brood is only found in
the tips of newly expanded tunnels, where the fungus growth is highest (Fig. 2). Like adults, larvae
are mycetophagous and feed on the palisade-like fungal layer on the tunnel walls. By contrast,
species in the xyleborine genera Xyleborinus and Xylosandrus construct cave-like brood chambers,
where adults and offspring live in close contact to each other and their wall-lining fungus gardens
(Fig.1). These chambers result from the effort of larvae that feed xylo-mycetophagously, not only on
the wall-covering fungal layers but also on fungus-infested wood (Schedl 1956; Roeper 1995). There
is little digestion of the swallowed wood (Francke-Grosmann 1967; Roeper 1995), but it is likely that
decomposition of wood is enhanced due to the spreading of larval faeces on the tunnel walls that
may get reutilized by the growing fungi (Hubbard 1897, Francke-Grosmann 1967). In the other
tunnel-constructing ambrosia beetle lineages, social interactions may be constrained if the tunnels
inhibit the free passing of individuals (Kirkendall et al. 1997).
In this study we compare two cooperatively breeding species of fungus-growing ambrosia
beetles that differ in the type of galleries produced. We describe the behavioural caste that is
responsible for digging the galleries and estimate the extent of social behaviours. As model species
we use X. saxesenii that build cave-like brood chambers and X. affinis that build the more common
branching-tunnel galleries. This comparison will hint on whether the complex system of division of
labour found in X. saxesenii (Biedermann and Taborsky 2011) can be assumed to exist only in species
living in brood chambers, or also in the majority of ambrosia beetles, which live in tunnels. Finally, we
describe the interspecific and intraspecific variation of the structure of their gallery systems.
Material & Methods
Study Species
Xyleborinus saxesenii Ratzeburg and Xyleborus affinis Eichhoff are probably the most
common members of the ambrosia beetle tribe Xyleborini (Curculionidae: Scolytinae) in the Southern
states of the US, and among the most widely distributed ambrosia beetles around the world today
(Wood 1982). X. saxesenii originates from temperate Eurasia, whereas X. affinis is native to the
American tropics and subtropics, and both are highly successful invaders of other continents. Like all
ambrosia beetles, they are solely living on a complex of self-cultured fungi, grown within a tunnel
system bored into the xylem of trees (Beaver 1989). Hosts are recently dead or highly stressed trees
that emit ethanol, which attracts founder females with fungus-filled mycetangia (spore carrying
organs; Ranger et al. 2010). As both species are attacking a wide variety of deciduous and coniferous
tree species, it appears that their main mutualistic fungi are very polyphagous (Wood 1982). These
fungi dominate the microbial community within healthy beetle galleries (Kajimura and Hijii 1992).
An unidentified Raffaelea sp. (Roeper and French, 1981) or Cephalosporium pallidum Verrall (Verrall
1943) serve as primary food for X. affinis; Raffaelea sulphurea (L.R. Batra) T.C. Harr. (syn. Ambrosiella
sulfurea; Harrington et al. 2010) serves as primary food for X. saxesenii (Batra 1966; FranckeGrosmann 1975). Both beetles show a strongly female-biased sex-ratio due to local mate
competition as a result of brother-sister mating (1:8-20; Roeper et al. 1980; Biedermann 2010;
Biedermann et al. 2011; cf. Peer and Taborsky 2004).
Illustrations of Gallery Morphology
81
Chapter 6
For studying the morphology of X. saxesenii galleries we dissected a disc (80 × 15cm diameter × height) cut off from a trunk of beech (Fagus sylvatica) near Bern/CH (560 m asl, 46°95’,
7°31’) on 9th November 2008. The trunk was still rooted in the ground and part of a tree felled about
one year before. Boring holes of beetles were solely found on the moist, north-facing side of the
trunk. The disc was brought to the lab, and beetle tunnels were carefully dissected. Apart from at
least 40 galleries of X. saxesenii, the disc contained also galleries of the ambrosia beetles Xylosandrus
germanus (N = 5 galleries) and Anisandrus dispar (N = 2), as well as the ship-timber beetle Hylecoetus
dermestoides (Lymexylidae; N > 50).
Inhabitants of all X. saxesenii galleries were counted and the shape of the galleries was
traced in original size on paper whenever possible. These drawings were scanned and colour
modified with Adobe Photoshop CS2 (Version 9.0, © Adobe Systems Inc. San Jose/CA, USA, 19992005) to suit graphical illustration. Gallery size was analysed by using GSA Image Analyser
(Demoversion 2011-02-10; © Software development & Analytics Bansemer & Scheel GbR, Rostock,
Germany).
To display the morphology of X. affinis galleries we scanned eight original-sized gallery
drawings from (Schedl 1962) (pp. 362-365) and modified them with Adobe Photoshop CS2 in the
same way as described above.
Laboratory Rearing and Observations.
X. affinis females used in this study were collected from oak logs, whereas X. saxesenii
females were caught with Lindgren funnel traps baited with ethanol, both near Pineville/LA, USA
(123ft asl; 31°20’, 92°24’) in Summer 2007. After bringing all females to the lab we prepared them for
laboratory rearing by surface sterilization, rinsing them for some seconds with 95% ethanol and
afterwards with deionised water. This treatment reduces fungal contamination by elimination of
fungal spores sticking to the body surface of the beetles, but does not harm the fungal spores of the
cultivar within the mycetangium, where they are kept for transmission by the beetles (FranckeGrosmann 1956). Afterwards we placed each female singly on a standard rearing medium of agar and
sawdust that was prepared in glass tubes (for more details on this technique see Biedermann and
others 2009). About 50 tubes per species were established in this way. Tubes were closed with
plastic caps, stored at room temperature (~23°C) and wrapped in paper, allowing light only to shine
onto the surface of the media. This way beetles bore tunnels frequently next to the tube glass, which
allows behavioural observations when the paper is removed (Fig. 2C).
A female immediately starts boring an entrance tunnel when placed onto the medium. While
boring it inseminates the tunnels with ambrosia fungus spores from its mycetangium or the gut
(Francke-Grosmann 1956; Francke-Grosmann 1975). If she is successful this results in fungus growth
on the walls within a few days (Roeper et al. 1980). After feeding on the fungus she starts to lay eggs
(Kingsolver and Norris 1977) and usually proceeds to do so for the next 40 days. During that time her
offspring pass three larval instars and a pupal stage, and finally they pass through a short maturation
period until they are fully sclerotized. First adult progeny appears around day 27 after gallery
initiation (Roeper et al. 1980; Biedermann et al. 2011).
Every third to fourth day we observed the behaviour of all individuals in 19 successfully
established galleries per species, between days 22 and 42 after initiation. As not all tubes had tunnels
next to the glass that could be observed from the beginning, observations ranged from 3 to 6 per
gallery. During each observation we noted the number of eggs and pupae and scanned the behaviour
of all 1st instar larvae, 2nd/3rd instar larvae, teneral females (not fully sclerotized), mature females and
males. Age and sex classes can be easily discerned by their appearance (for details see Fischer 1954
and Biedermann et al. 2009).
82
Chapter 6
Behavioural and Statistical Analyses
All ambrosia beetle behaviours have been described by Biedermann and Taborsky (2011).
Here we focus only on the behaviours of larvae and adults that are of interest in a social context (see
Table 1). Digging behaviour of developing offspring increases the size of the egg-niche to a brood
chamber and thus creates wall surface for the fungal gardens to flourish (Bright 1973). Digging is part
of larval feeding on fungal hyphae that penetrate the wood (wood is not digested; (Batra 1966;
Haack and Slansky 1987) and thus cannot be separated from it. In contrast, the digging of adults is
clearly different from feeding, and solely serves gallery extension. By cropping, adults graze off
ambrosia (Fig. 1), which apparently enhances fungus growth (Biedermann and Currie, unpubl. data).
Consequently, weeds do not invade (see references in Lengerken 1939; Norris 1993) and asexual
fruiting (= ambrosial growth) is induced (Schneider-Orelli 1913; Batra and Michie 1963; French and
Roeper 1972). Gallery hygiene is important for a flourishing fungus garden as well as a healthy colony
and can be attributed to three behaviours: balling, shuffling and cannibalising. By balling, larvae
compress faeces and small pieces of wood into compact balls, which facilitates further transport.
Those frass-balls are moved towards fungus beds by shuffling, where some of their nutrients appear
to get recycled (Xyleborus ferrugineus; Kok and Norris 1972). Frass, pieces of wood, and sclerotized
parts of dead siblings get also shuffled towards unused tunnel-parts or to the entrance, where they
are thrown out of the gallery. The function of cannibalising is not completely clear as on the one
hand it often involves dead or hardly moving siblings that probably need to be removed to maintain
gallery hygiene, but occasionally also individuals that seem to move normally are consumed. The
health status of colony members might be checked during allogrooming, which appears crucial for
individual survival, as microorganisms quickly cover uncleaned beetle bodies and brood (Biedermann
and Taborsky 2011). Finally, blocking of the gallery entrance has been suggested to serve a variety of
functions like protection of the brood and regulation of the microclimate inside the gallery (see
Kirkendall 1993 and Kirkendall et al. 1997 for review).
Generalized estimating equations (GEEs) were used to analyse species differences in the
relative behavioural frequency (binary response variable) per class (1st instar larvae, 2nd/3rd instar
larvae, teneral females, mature females, and males) by controlling for gallery age. The latter had no
significant effect on the behavioural frequencies, so it was excluded from the models. GEEs are an
extension of generalized linear models with an exchangeable correlation structure of the response
variable within a cluster, which allows controlling for the variation between observations from a
single gallery. GEEs were performed using R (lmer in R; Version 2.12.1; R Development Core Team
2008).
Results
Gallery Morphology
Eighteen of >40 dissected field galleries of X. saxesenii were successfully traced and analysed.
Interestingly, although distances between tunnels of different wood boring beetles in the disc were
very close, they never merged. X. saxesenii galleries are illustrated in Fig. 1. All galleries were flat
with a clearance of about 1mm and usually expanded within one plane (except Fig. 1F,G), always
along the direction of the wood fibres, often in a right angle to the annual rings. Entrance tunnels
were directed vertically to the tree surface, had a length of 5 to 50mm and were followed by one or
several distinct brood chambers of different shapes. Galleries with a single, large brood chamber
seemed to harbour the most individuals (Fig. 1O-R; Table S1). About half of the galleries (N = 8) were
inhabited by less than 10 individuals, and about a quarter by more than 50 individuals (N = 4). The
number of individuals correlated strongly with gallery area (R2 = 0.62; linear regression: F = 26.2, p <
0.001, DF= 1, N = 18). The galleries encompassed on average 126.1 mm 2 (± 17 se) and inhabited 23.2
± 6.5 individuals, with an adult sex-ratio of 1.2 : 8.4 (m : f). Hence, there was about one individual per
83
Chapter 6
5 mm2, which makes about 10 mm2 gallery wall space per individual (both sides of the gallery). Brood
and adults were randomly distributed within a chamber (Fig. 1R). In contrast, brood in X. affinis
galleries was only located in the tips of the tunnels (Fig. 2B,C). X. affinis galleries apparently merged
sometimes or alternatively females bored a second exit (Fig. 2I; Schedl 1962). Schedl (1962)
described tunnels to usually expand within one plane, but also above and below if attack densities
get very high. Tunnels of X. affinis can be frequently found also in the phloem, but still with their
normal mycetophagous habit. The eight traced galleries (Fig. 2B-I) that were not fully traced, had a
mean total length of 172.8 mm (Range: 71-294 mm), had between 3 and 13 side-tunnels and split the
first time on average 11.8 mm (Range: 4-26 mm) behind the entrance. Schedl (1962) reported a
mean of 61.1 (Range: 4-223; N = 18 galleries) adult females and 1.5 (Range: 0-4) adult males per
gallery; this gave a strongly female-biased mean sex-ratio of 0.024.
Behavioural differences between X. affinis and X. saxesenii
Larval instars
The way of feeding differed between the species. X. affinis larvae are mycetophagous, solely
cropping the fungal layers from their gallery walls. By contrast, larvae of X. saxesenii expand the
gallery by their xylo-mycetophagous feeding behaviour (= digging). When feeding frequency was
compared between species, it turned out that digging was more common than cropping (GEE: p =
0.007; Fig. 3, Table 3) in 2nd/3rd instars.
There were no behavioural differences among the 1st instar larvae of the two species, except that
allogrooming was significantly more common in X. affinis (GEE: p = 0.028). This difference was also
highly significant in 2nd/3rd instars (GEE: p < 0.001; Fig. 3, Table S2). Balling was only found in 2nd/3rd
instars of X. saxesenii and served to compile faeces and sawdust that accumulated in large amounts
from the larval digging habit.
Teneral and mature females, and males
Behavioural analyses revealed no statistically significant difference between the behaviour of teneral
females. Mature females of both species did not differ in their time spent with blocking and digging
(GEE: p > 0.05; Fig. 4, Table S2). Cropping of the fungus (GEE: p = 0.02) and allogrooming (GEE: p <
0.001) was more common in mature females of X. saxesenii, whereas X. affinis mature females
showed more shuffling (GEE: p < 0.001). Males of both species did not differ in any of the behaviours
observed.
Discussion
The existence of long-lived kin groups is the most likely precondition for the evolution of
agriculture (Brock et al. 2011). Examples include fungus-growing ants and termites (Mueller et al.
2005), bacteria-growing social amoeba (Brock et al. 2011) and ambrosia beetles of the Xyleborini
tribe (Scolytinae; (Schneider-Orelli 1913; Francke-Grosmann 1967; Norris 1993; Biedermann et al.
2009). The latter are characterized by fungus agriculture, inbreeding, haplodiploidy, and varying
social organizations (Peer and Taborsky 2007). Cooperative breeding and larval workers have been
recently described for X. saxesenii (Biedermann and Taborsky 2011) and may exist also in other
Xyleborini.
Here we have aimed to illustrate the complex cave-tunnel and branching-tunnel gallery
systems of Xyleborini, which are two of six types of gallery morphologies known from ambrosia
beetles (for details see Fig.5, Table 2). Apart from the mating system (inbreeding within the nest vs.
84
Chapter 6
outbreeding), the main factor affecting gallery structure is the feeding habit of the larvae, which is
either xylomycetophagous on wood and fungus (digging in X. saxesenii) or mycetophagous on fungus
only (cropping in X. affinis). This has implications interactions among gallery members and between
beetles and their fungi. In the cave-tunnel systems of X. saxesenii, where brood and adults are
intermingled and surrounded by their fungi, adults showed more brood care (allogrooming) and
interactions with the fungus (cropping) than X. affinis adults in their galleries. Wood-containing,
larval faeces were either smeared on the tunnel walls (Hubbard 1897) probably for reutilization by
the fungus, or prepared by the larvae for removal (balling). By contrast, brood was constrained to the
tips of the branching-tunnels in X. affinis. This spacial separation between adults and larvae was
probably the reason for larvae partly undertaking the allogrooming of the mature females. The latter
interacted also less frequently with the fungi (cropping), but instead spent half of their time in
queues with others, shuffling the tunnels free of waste. Previously, individuals of both species have
been observed to adjust their behaviours to the demands in their galleries: (i) X. saxesenii mature
females adjust their per capita allogrooming frequency to the number of brood present (Biedermann
and Taborsky 2011), and (ii) X. affinis larvae and adults reduce cropping- and allogrooming-activities
in the entrance part of the gallery relative to the tips of the tunnels, where most of the brood and
fungus is present (Biedermann, Taborsky and Six, in preparation). These observations suggest that
social task sharing between larvae and adults is not restricted to cave-building species (Biedermann
and Taborsky 2011), but might also exist in the majority of ambrosia beetles inhabiting branchingtunnel systems.
Gallery architecture
All X. saxesenii galleries we dissected had an entrance tunnel and a cave-like brood chamber
expanding along the wood grain. The shape of this chamber was highly variable and in some cases
several chambers were lined up in sequence. This suggests that the mycelium of Raffaelea sulfurea,
the mutualistic fungus, develops along the grain, but penetrates only the wood cells close to the
gallery surface (e.g. Leach et al. 1940; Francke-Grosmann 1963); this growth is probably followed by
the larval gnawing activity and thus the chamber gets a drawn-out shape. The existence of multiple
chambers may indicate that the fungus does not grow equally well in all parts of the gallery.
The high numbers of brood and adults found in field galleries in November suggest that X.
saxesenii galleries persist longer than for one season. It is very likely that the galleries had been
founded about 6 months earlier, because the trees had been felled one year before and the main
founding period of galleries for X. saxesenii is spring (Fischer 1954; Peer and Taborsky 2007). Brood
was abundant and the wood around galleries showed no signs of microbial decay when it was
dissected (although it was densely settled by other wood-boring beetles already). Therefore, it seems
likely that in most galleries the adults, which are descendants of the founder female, would have
continued breeding in the following year. In the laboratory, founder females die about two months
after gallery foundation (Biedermann et al. 2012), and mature daughters may then inherit the gallery
(Biedermann et al. 2009; 2012).
Dissections of all mature females in X. affinis laboratory galleries showed that around the
time when the founder female dies more than half of the daughters reproduce (unpublished data;
see Appendix 1 of the thesis). Field galleries of the sub-tropical X. affinis were not studied here, but
previous studies suggest that they may get remarkably large (up to 6 m in length) and old (up to 4
years; Schedl 1962, Scheider 1987). This information is important because it shows that with a
generation time of about a month (in the laboratory; Biedermann et al. 2009), multiple overlapping
generations within one gallery are possible, which increases the probability that complex social
systems evolve (Kirkendall et al. 1997; Peer and Taborsky 2007).
85
Chapter 6
Behavioural differences
Gallery differences morphology between Xyleborus affinis and Xyleborinus saxesenii were
confirmed also when breeding the beetles in artificial medium in the laboratory, suggesting that
divergent gallery structure is not a result of breeding in a different substrate. We never observed
gallery enlargement by larvae of X. affinis during the trials, although they were reared in an agarsawdust-medium that is much easier to gnaw than solid wood. By contrast, in X. saxesenii almost all
gallery excavation is due to the larval digging activity (Biedermann and Taborsky 2011). Digging by
adults was seen in very low and similar frequencies in both species. These observations are in line
with (Roeper 1995) suggestion that different feeding modes of larval instars and not the digging
activity of adults are responsible for the different gallery structures of ambrosia beetles.
Interestingly, X. saxesenii larval digging was more common than X. affinis larval cropping.
Considering the similar size and developmental period length of both species X. saxesenii larvae may
need to spend more cooperative effort than X. affinis larvae. Digging in wood probably demands
more energy expenditure than cropping off fungus, and in addition there are probably less nutrients
in fungus-infested wood material (wood cannot be digested by the larval digestive system; (FranckeGrosmann 1967) than in pure fungal material. So why do larvae of X. saxesenii engage in digging
instead of cropping, if the cost-benefit ratio is presumably worse? First, it appears that Raffaelea
sulfurea (the primary mutualist of X. saxesenii) does not produce a fungal layer comparable in
thickness to the unidentified primary mutualist of X. affinis (which might be an unidentified Raffaelea
sp. Roeper and French (1981), or Cephalosporium pallidum Verrall (1943)), so the lower productivity
of the fungus on the gallery surface could force larvae to dig, which may raise its productivity by
providing more space for sporulation. In this case, selection of the larval behaviour would be driven
by the fungus. Second, in X. saxesenii galleries digestion of the wood may be achieved by the joint
action of larval offspring and Raffaelea sulfurea. X. saxesenii adults and larvae show different
enzymatic profiles, with xylanases only present in larvae (De Fine Licht and Biedermann, submitted).
Additionally, chewed-up wood swallowed by larvae provides more surface for the fungal enzymes to
interact with (similar to wood-feeding termites; Watanabe and Tokuda 2009), both within intestines
of larvae and in their faeces that are partly smeared on the gallery walls (Hubbard 1897; Biedermann
and Taborsky 2011), probably for further utilization by the fungus. Third, larval digging may be
selected because it creates more space for fungal growth and additional brood.
In the field, X. saxesenii gallery size correlated with the number of colony members.
inhabited more individuals. In contrast to X. affinis , in X. saxesenii other group members benefit
from the larval feeding activity because this increases space and thus reduces within-group
competition for food (Biedermann and Taborsky 2011). Larval digging, however, also produces large
amounts of faecal waste, which is removed from the gallery in two steps: (i) by the balling behaviour
of larvae that apparently serves to pile up the faeces in balls, which are then displaced and shuffled
out of the gallery by adult females. Balling was not observed in X. affinis larvae; much less faeces
were produced in their galleries and the faeces did not contain woody parts. Adult X. affinis seem to
be much more important for gallery hygiene given their shuffling behaviour relative to X. saxesenii
adults, because waste can quickly block a tunnel if it is not immediately removed. In X. affinis,
multiple shuffling adults were often queuing within tunnels, passing along faeces and sawdust for
final disposal outside the entrance.
Brood in X. affinis galleries was confined to the ends of the tunnels were fungi seemed to
grow best, while adults usually stayed in other parts of the tunnels. By contrast, in X. saxesenii brood
and adults were fully mixed in the common brood chamber together with the fungus, where adults
engaged in (a) allogrooming and (b) cropping of the fungus. Allogrooming is necessary, because
individuals that are not cleaned by others are quickly overgrown by mold (Paecilomyces sp. and
Fusicolla sp.; Biedermann et al., submitted), which can quickly kill them (Biedermann and Taborsky
2011). Interestingly, in X. affinis larvae show more allogrooming than adults and also more than X.
saxesenii larvae, probably because X. affinis larvae more frequently meet each other due to the
86
Chapter 6
spatial conditions in the galleries. Cropping of the fungus apparently serves both nutrient intake and
fungus care. It has been shown to induce the ambrosial growth of the fungus, which is necessary for
consumption (French and Roeper 1972) and to prevent the invasion of fungal pathogens (SchneiderOrelli 1913; Norris 1993). Oral secretions might be involved, which in other scolytine beetles were
shown to contain bacteria with antibiotic effects against antagonistic fungi (Cardoza et al. 2006; Scott
et al. 2008). X. saxesenii adults showed more cropping of the fungus than X. affinis adults, but this
difference cannot be easily interpreted as these species use different fungi for agriculture.
In conclusion, it seems that living in tunnels does not restrict the evolution of social
complexity in ambrosia beetles (Fig. 5; Table 2). In contrast to previous suggestions (Kirkendall et al.
1997), individuals in X. affinis tunnel systems can easily pass each other, allowing for similarly
complex social interactions as previously found in X. saxesenii (Biedermann and Taborsky 2011), a
species with a cave-like gallery structure. Cooperative task specialisation and age polyethism is also
shown by X. affinis. Most remarkably, even larvae of X. affinis behave cooperatively, suggesting that
larval workers are probably the rule rather than the exception in gregariously living weevils, the only
holometabolous insect group in which active social behaviours by larvae have been reported
(Biedermann and Taborsky 2011). Hence, we propose that there might be modes of fungus
agriculture differing mainly in the contribution of larvae and adult offspring (Fig. 5; Table 2): (1)
Cooperative xylomycetophagous fungus agriculture (in Xyleborinus sp., Xylosandrus sp. and
Platypodinae), where larvae excavate a brood chamber or tunnels by feeding on fungus infested
wood. Large amounts of wood are swallowed and probably partly digested with the help of enzymes
(De Fine Licht and Biedermann 2012), and faeces may be smeared against the walls for reutilization
by the growing fungus. A common brood chamber facilitates interactions between individuals and
with the fungus. Larvae may engage in balling of faeces and sawdust (yet reported only for
Xyleborinus saxesenii). (2) Cooperative mycetophagous fungus agriculture (in Xyleborus sp.), where
larvae feed on fungal layers covering the gallery walls, and adults enlarge the gallery system by
digging. Faeces do not contain woody parts but are still partly reutilized by the growing fungi (Kok
and Norris 1972; Norris 1975). Larvae harvest the fungus and interact with each other in the tips of
the tunnels, and thus are largely separated from the adults that keep the afferent tunnels free of
waste. (3) Subsocial xylomycetophagous fungus agriculture (in Trypodendron sp.), where larvae
excavate a cradle (pupal niche) for themselves and do not interact with other brood, but there are
observations that they interact with the adults (Hubbard 1897; Borden 1988). Parents bore the
tunnels and keep them free of larval frass and protect the nest (Borden 1988; Kirkendall et al. 1997).
The first two modes of fungus agriculture may coincide with a (A) simple or (B) complex social
type, which is reflected in the complexity of the gallery pattern. Species exhibiting a simple social
type (e.g. Xylosandrus sp., Anisandrus sp., Trypodendron sp.) usually use galleries only for the
development of a single offspring generation (Fig. 5A,C,E), whereas species exhibiting a more
complex social type (e.g. Xyleborinus sp., some Xyleborus sp., Platypus sp.) typically grow several
offspring generations within one gallery (Fig. 5B,D,F). Division of labour between larvae and adults
and delayed dispersal of adult offspring may occur in species of both social types, but these traits are
much more pronounced in species of the complex social type (see Table 2). The development of
several offspring generations in succession requires rather long-lived substrate conditions. Thus,
whereas species of the simple social type can settle in dead branches or tree trunks of small
diameters, species of the complex social type appear to settle typically in large-diameter trunks of
recently dead trees or even in living trees (Schedl 1962; Wood 1982). Nevertheless, in dead wood it is
probably unpredictable whether successive offspring generations will be able to use the same nest,
because of the exploitation of the substrate by a plethora of competing microorganisms. Selection
will therefore favour totipotent individuals that may either disperse and breed independently, or stay
and help, or stay and co-breed with relatives (e.g. X. saxesenii: Peer and Taborsky (2007) and
Biedermann and Taborsky (2011); X. affinis: Biedermann et al. 2011). Eusociality, classified by a
permanently sterile and a reproductive caste, will only evolve under stable resource conditions in
living trees, like found for instance in A. incompertus (Kent and Simpson 1992). Hence, the durability
87
Chapter 6
of the substrate may be the major selective force deciding about the evolution of ambrosia beetle
social complexity.
The ambrosia beetle tribe Xyleborini (Scolytinae) radiated about 20 million years ago and
stands for one of eight independent origins of fungus agriculture in the curculionid beetles (Farrell et
al. 2001; Hulcr et al. 2007), which is probably closely associated with the evolution of division of
labour and advanced sociality (Kirkendall et al. 1997; Biedermann and Taborsky 2011). This contrasts
with the other fungus culturing insect groups, attine ants and macrotermitine termites, that lived in
eusocial groups already when fungiculture evolved (Wilson 1971), which makes ambrosia beetles a
unique model system to study the evolution of sociality in relation to fungiculture (Mueller et al.
2005). The evolutionary history of sociality in ambrosia beetles becomes even more interesting when
considering that they vary in their mating patterns (inbreeding vs. outbreeding species) and ploidy
level (haplodiploid vs. diploid species; see Table 2), which are factors proposed to contribute to social
evolution (Choe and Crespi 1997; Bourke 2011), but their respective roles are controversial (West et
al. 2007; Bourke 2011).
Acknowledgements
We thank Stacy Blomquist and Eric Ott for their help in collecting X. affinis in the field and starting
the first laboratory galleries. This study was performed at the Institute of Ecology and Evolution,
University of Berne, Switzerland, and at the Southern Research Station in Pineville. This work
benefited greatly from help and discussions with Kier Klepzig and commentaries on the manuscript
by Tabea Turrini, Henrik de Fine Licht, Florian Menzel and Markus Zöttl. Studies in Pineville, LA, were
supported by a cooperative agreement with the Southern Research Station, USDA Forest Service.
PHWB received a DOC-fellowship from the Austrian Academy of Sciences and a Roche Research
Fellowship at the Department of Behavioural Ecology, University of Berne, Switzerland.
Literature
Batra LR (1966) Ambrosia fungi: extent of specifity to ambrosia beetles. Science 153: 193-195.
Batra LR & Michie MD (1963) Pleomorphism in some ambrosia and related fungi. Transactions of the Kansas
Academy of Science 66: 470-481.
Beaver RA (1989) Insect-fungus relationships in the bark and ambrosia beetles. Insect-fungus interactions
(Wilding N, Collins NM, Hammond PM & Webber JF, eds), pp. 121-143. Academic Press, London.
Biedermann PHW (2007) Towards eusociality in ambrosia beetles. Masters thesis, University of Berne,
Switzerland.
Biedermann PHW (2010) Observations on sex ratio and behavior of males in Xyleborinus saxesenii Ratzeburg
(Scolytinae, Coleoptera). Zookeys 56: 253-267.
Biedermann PHW, Klepzig KD & Taborsky M (2009) Fungus cultivation by ambrosia beetles: Behavior and
laboratory
breeding success in three xyleborine species. Environmental Entomology 38: 1096-1105.
Biedermann PHW, Klepzig KD & Taborsky M (2009) Fungus cultivation by ambrosia beetles: behavior and
laboratory breeding success in three xyleborine species. Environmental Entomology 38: 1096-1105.
Biedermann PHW, Klepzig KD & Taborsky M (2011) Costs of delayed dispersal and alloparental care in the
fungus-cultivating ambrosia beetle Xyleborus affinis Eichhoff (Scolytinae: Curculionidae). Behav. Ecol. Sociobiol.
65: 1753-1761.
88
Chapter 6
Biedermann PHW, Peer K & Taborsky M (2012) Female dispersal and reproduction in the ambrosia beetle
Xyleborinus saxesenii Ratzeburg (Coleoptera; Scolytinae). Mitteilungen der Deutschen Gesellschaft für
allgemeine und angewandte Entomologie in press.
Biedermann PHW & Taborsky M (2011) Larval helpers and age polyethism in ambrosia beetles. Proc. Natl Acad.
Sci. USA 108: 17064-17069.
Borden JH (1988) The striped ambrosia beetle. Dynamics of Forest Insect Populations. (Berryman AA, ed), pp.
579-596. Plenum, New York.
Bot ANM, Currie CR & Boomsma JJ (2001) Waste management in leaf-cutting ants. Ethology Ecology &
Evolution 13: 225-237.
Bourke AFG (2011) Principles of Social Evolution. Oxford University Press, Oxford.
Bright DE (1973) The bark and ambrosia beetles of California, Coleoptera: Scolytidae and Platypodidae.
University of California Press, Berkeley, CA.
Brock DA, Douglas TE, Queller DC & Strassmann JE (2011) Primitive agriculture in a social amoeba. Nature 469:
393-396.
Cardoza YJ, Klepzig KD & Raffa KF (2006) Bacteria in oral secretions of an endophytic insect inhibit antagonistic
fungi. Ecological Entomology 31: 636-645.
Choe JC & Crespi BJ (1997) The Evolution of Social Behaviour in Insects and Arachnids. Cambridge University
Press, Cambridge UK.
Dahlsten DL (1982) Relationship between bark beetles and their natural enemies. Bark Beetles in North
American Conifers (Mitton & Sturgeon, eds), pp. 140-182. University of Texas Press, Austin.
De Fine Licht HH & Biedermann PHW (2012) Patterns of functional enzyme activity suggest that larvae are the
key to successful fungus farming by ambrosia beetles. Frontiers in Zoology in review.
Farrell BD, Sequeira AS, O'Meara BC, Normark BB, Chung JH & Jordal BH (2001) The evolution of agriculture in
beetles (Curculionidae : Scolytinae and Platypodinae). Evolution 55: 2011-2027.
Fischer M (1954) Untersuchungen über den kleinen Holzbohrer (Xyleborus saxeseni). Pflanzenschutzberichte
12: 137-180.
Francke-Grosmann H (1956) Hautdrüsen als Träger der Pilzsymbiose bei Ambrosiakäfern. Z. Morph. u. Ökol.
Tiere 45: 275-308.
Francke-Grosmann H (1963) Some new aspects in forest entomology. Annual Review of Entomology 8: 415-438.
Francke-Grosmann H (1967) Ectosymbiosis in wood-inhabiting beetles. Symbiosis (Henry SM, ed), pp. 141-205.
Academic Press, New York.
Francke-Grosmann H (1975) Zur epizoischen und endozoischen Übertragung der symbiotischen Pilze des
Ambrosiakäfers Xyleborus saxeseni (Coleoptera: Scolitidae). Entomologica Germanica 1: 279-292.
French JRJ & Roeper RA (1972) Interactions of ambrosia beetle, Xyleborus dispar (Coleoptera, Scolytidae), with
its symbiotic fungus Ambrosiella hartigii (Fungi imperfecti). Canadian Entomologist 104: 1635-1641.
French JRJ & Roeper RA (1973) Patterns of nitrogen utilization between the ambrosia beetle Xyleborus dispar
and its symbiotic fungus. Journal of Insect Physiology 19: 593-605.
Haack RA & Slansky F (1987) Nutritional ecology of wood feeding Coleoptera,Lepidoptera and Hymenoptera.
Nutritional Ecology of Insects, Mites and Spiders (Slansky F & Rodriguez JG, eds), pp. 449-486. Wiley, New York.
89
Chapter 6
Harrington TC, Aghayeva DN & Fraedrich SW (2010) New combinations in Raffaelea, Ambrosiella, and
Hyalorhinocladiella, and four new species from the redbay ambrosia beetle, Xyleborus glabratus. Mycotaxon
111: 337-361.
Hart A, Anderson C & Ratnieks F (2002) Task partitioning in leafcutting ants. Acta ethologica 5: 1-11.
Hölldobler B & Wilson EO (1990) The ants. The Belknap Press of Harvard University Press, Cambridge, MA.
Hosking GB (1972) Xyleborus saxeseni, its life-history and flight behaviour in New Zealand. N. Z. J. Forest Science
3: 37-53.
Hubbard HG (1897) The ambrosia beetles of the United States. Some Miscellaneous Results of the Work of the
Division of Entomology (Howard LO, ed), pp. 9-13. U.S. Dept. of Agriculture Bureau of Entomology Bull. No. 7.
Hulcr J, Kolarik M & Kirkendall LR (2007) A new record of fungus-beetle symbiosis in Scolytodes bark beetles
(Scolytinae, Curculionidae, Coleoptera). Symbiosis 43: 151-159.
Kajimura H & Hijii N (1992) Dymamics of the fungal symbionts in the gallery system and the mycangia of the
ambrosia beetle, Xylosandrus mutilatus (BLANDFORD) (Coleoptera, Scolytidae). Ecological Research 7: 107-117.
Kent DS & Simpson JA (1992) Eusociality in the beetle Austroplatypus incompertus (Coleoptera: Platypodidae).
Naturwissenschaften 79: 86-87.
Kingsolver JG & Norris DM (1977) The interaction of Xyleborus ferrugineus (Fabr.) (Coleoptera: Scolytidae)
behavior and initial reproduction in relation to its symbiotic fungi. Annals of the Entomological Society of
America 70: 1-4.
Kirkendall LR, Kent DS & Raffa KF (1997) Interactions among males, females and offspring in bark and ambrosia
beetles: the significance of living in tunnels for the evolution of social behavior. The Evolution of Social Behavior
in Insects and Arachnids (Choe JC & Crespi BJ, eds), pp. 181-215. Cambridge University Press.
Kok LT & Norris DM (1972) Symbiotic interrelationships between microbes and ambrosia beetles 6. Aminoacid
composition of ectosymbiotic fungi of Xyleborus ferrugineus (Coleoptera, Scolytidae). Annals of the
Entomological Society of America 65: 598-602.
Kokko H & Ekman J (2002) Delayed dispersal as a route to breeding: territorial inheritance, safe havens, and
ecological constraints. Am. Nat. 160: 468-484.
Korb J (2008) The ecology of social evolution in termites. Ecology of Social Evolution (Korb J & Heinze J, eds), pp.
151-174. Springer, Berlin, Heidelberg.
Leach JG, Hodson AC, Chilton SJP & Christensen CM (1940) Observations on two ambrosia beetles and their
associated fungi. Phytopathology 30: 227-236.
Lengerken H (1939) Die Brutfürsorge- und Brutpflegeinstinkte der Käfer. Akademische Verlagsgesellschaft
m.b.H., Leipzig.
Merkl O & Tusnadi CsK (1992) First introduction of Xyleborus affinis (Coleoptera: Scolytidae), a pest of
Dracaena fragrans 'Massangeana', to Hungary. Folia Entom. Hung. 52: 67-72.
Milne DH & Giese RL (1970) Biology of the Columbian Timber Beetle, Corthylus columbianus (Coleoptera:
Scolytidae). 10. Comparison f yearly mortality and dispersal losses with population densities. Entomological
News 81: 12-24.
Mueller UG, Gerardo NM, Aanen DK, Six DL & Schultz TR (2005) The evolution of agriculture in insects. Annual
Review of Ecology Evolution and Systematics 36: 563-595.
90
Chapter 6
Norris DM (1975) Chemical interdependence among Xyleborus spp. ambrosia beetles and their symbiotic
microbes. Material und Organismen 3: 479-788.
Norris DM (1993) Xyleborus ambrosia beetles - a symbiotic ideal extreme biofacies with evolved polyphagous
privileges at monophagous prices. Symbiosis 14: 229-236.
Peer K & Taborsky M (2004) Female ambrosia beetles adjust their offspring sex ratio according to outbreeding
opportunities for their sons. J. Evol. Biol. 17: 257-264.
Peer K & Taborsky M (2005) Outbreeding depression, but no inbreeding depression in haplodiploid ambrosia
beetles with regular sibling mating. Evolution 59: 317-323.
Peer K & Taborsky M (2007) Delayed dispersal as a potential route to cooperative breeding in ambrosia beetles.
Behav. Ecol. Sociobiol. 61: 729-739.
R Development Core Team (2008). R: A language and environment for statistical computing. Vienna, Austria, R
Foundation for Statistical Computing.
Ranger CM, Reding ME, Persad AB & Herms DA (2010) Ability of stress-related volatiles to attract and induce
attacks by Xylosandrus germanus and other ambrosia beetles. Agricultural and Forest Entomology 12: 177-185.
Roeper R, Treeful LM, O'Brien KM, Foote RA & Bunce MA (1980) Life history of the ambrosia beetle Xyleborus
affinis (Coleoptera: Scolytidae) from in vitro culture. Great Lakes Entomologist 13: 141-144.
Roeper RA (1995) Patterns of mycetophagy in Michigan ambrosia beetles. Michigan Academian 27: 153-161.
Roeper RA & French JRJ (1981) Ambrosia fungi of the Western United States and Canada - beetle assocaitions
(Coleoptera: Scolytidae), Tree Hosts, and Distribution. Northwest Science 55: 305-309.
Schedl KE (1956) Breeding habits of arboricole insects in Central Africa. Proceedings 10th International
Congress of Entomology, Montreal, pp. 183-197.
Schedl KE (1962) Scolytidae und Platypodidae Afrikas, II. Revista de Entomologia de Mocamique 5: 1-594.
Schneider I (1987) Distribution, fungus-fransfer and gallery construction of the ambrosia beetle Xyleborus
affinis in comparison with X. mascarensis (Coleoptera, Scolytidae). Entomologia Generalis 12: 267-275.
Schneider-Orelli O (1913) Untersuchungen über den pilzzüchtenden Obstbaumborkenkäfer Xyleborus
(Anisandrus) dispar und seinen Nährpilz. Zentralblatt für Bakteriologie, Parasitenkunde, Infektionskrankheiten
und Hygiene II 38: 25-110.
Scott JJ, Oh DC, Yuceer MC, Klepzig KD, Clardy J & Currie CR (2008) Bacterial protection of beetle-fungus
mutualism. Science 322: 63.
Selander RK (1964) Speciation in Wrens of the Genus Campylorhynchus. University of California Publications in
Zoology 74: 1-305.
Stacey PB & Ligon JD (1987) Territory Quality and Dispersal Options in the Acorn Woodpecker, and A Challenge
to the Habitat-Saturation Model of Cooperative Breeding. Am. Nat. 130: 654-676.
Traniello JFA & Leuthold RH (2000) The behavioral ecology of foraging in termites. Termites: Evolution,
Sociality, Symbiosis, Ecology. (Abe T, Bignell DE & Higashi M, eds), pp. 141-168. Kluwer, Dordrecht.
Verrall AF (1943) Fungi associated with certain ambrosia beetles. Journal of Agricultural Research 66: 135-144.
Watanabe H & Tokuda G (2009) Cellulolytic Systems in Insects. Annu. Rev. Entomol. 55: 609-632.
91
Chapter 6
West SA, Griffin AS & Gardner A (2007) Social semantics: altruism, cooperation, mutualism, strong reciprocity
and group selection. J. Evol. Biol. 20: 415-432.
Wilson EO (1971) The Insect Societies. Belknap Press, Cambridge.
Wood SL (1982) The bark and ambrosia beetles of North and Central America (Coleoptera: Scolytidae), a
taxonomic monograph. Great Basin Naturalist Memoirs 6: 1-1359.
Graphs & Tables
Table 1. Behaviours of larval and adult xyleborine ambrosia beetles likely resulting in mutual benefits for other
group members in the natal gallery (modified and extended from Biedermann & Taborsky, 2011).
Class
Behaviour
Definition
Digging
enlarging the brood chamber by feeding on fungusinfested substrate
Larvae
Cropping
grazing on the fungal layer on the gallery walls
forming balls of frass and sawdust by repeated ventral
Balling
body contractions
grooming an egg, larva, pupa, or adult beetle with the
Allogrooming
mouthparts
Adults
Mutual
benefit
X.
saxesenii
gallery
extension
x
fungus care (?)
X.
affinis
x
hygiene
x
brood care,
hygiene
x
x
Blocking
staying inactive in the entrance tunnel and plugging it
with the body (abdomen directed to the outside)
gallery
protection
x
x
Digging
excavating new tunnels
gallery
extension
x
x
fungus care
x
x
hygiene
brood care,
hygiene
x
x
x
x
grazing on the fungal layer on the gallery walls with the
maxillae and/or mandibles
moving frass and sawdust with the legs and elytra
Shuffling
grooming an egg, larva, pupa, or adult beetle with the
Allogrooming
mouthparts (i.e., maxillae, labium)
Cropping
92
Chapter 6
Table 2. Overview of the gallery architecture and modes of fungiculture and social life of
representative ambrosia beetle genera.
Family
Curculionidae
Tribe
Genus
Xyleborini
Lymexylidae
Xyloterini
Cryphalini
Platypodini
Xylosandrus
Xyleborinus
Anisandrus
Xyleborus
Trypodendron
Gnathotrichus
Platypus
Hylocoetus
Cave-tunnel
Cave-tunnel
Branching
tunnels
Branching
tunnels
Branching
tunnels with
larval cradles
Branching
tunnels with
larval cradles
Branching
tunnels with
larval cradles
Single tunnel
Gallery
architectu
re
Type of
gallery
Structure
Simple
Reference
in Fig. 5
A
1
Complex
2
Simple
1
Complex
2
Simple
1
Complex
2
Complex
2
na
B
C
D
E
F
F
Not shown
Haplodiploid
Haplodiploid
Haplodiploi
d
Haplodiploi
d
Diploid
Diploid
Diploid
Diploid
Inbreeding
Inbreeding
Inbreeding
Inbreeding
Outbreeding
Outbreeding
Outbreeding
Outbreeding
Larvae
Xylomycetopha
gy3 (?)
Xylomycetopha
gy3
Xylomycetopha
gy3
Mycetophagy4
Mycetophagy4
Mycetophagy4
Xylomycetopha
gy3
Xylomycetopha
3
gy (?)
Xylomycetopha
gy3
Mycetophagy4
Mycetopha
gy4
Mycetopha
4
gy
Xylomycetopha
gy3
Adults
Mycetopha
gy4
Mycetopha
4
gy
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Brood chamber
Brood chamber
Tips of
tunnels (?)
Tips of
tunnels
Larval cradles
Larval cradles
Larval cradles,
in tunnels
na
No
No
Partly
Partly
Yes
Yes
Partly
na
1
Several
1
Several
1
Several (?)
Several
na
1
Several
1
Several
1
Several (?)
Several (?)
na
Sub-social
Primitive
eusocial
Sub-social
Sub-social,
primitive
eusocial
Sub-social
Sub-social
Sub-social,
eusocial
Solitary
GE, BC, GH
BC(?), GH(?)
BC, GH
-
?
GE, BC(?), GH
na
GP, BC, GH, FC
GE, GP, BC,
GH, FC
GE, GP, BC,
GH, FC
GE, GP, GH,
FC(?)
GE, GP, GH,
FC(?)
GE, GP, BC, GH,
FC
na
I
I
I
GP, GH
GP, GH
GP, GH
na
Ploidy
Breeding
structure
Mode of
feeding
Social
structure
Communa
l nesting
Location
of brood
Separatio
n of larvae
and adults
Number
of
generatio
ns per
nest
Number
of
breeding
females
Social
organizati
on
No feeding
Division of labour
Larvae
Adult
females
Adult
males
GE(?), BC,
GH(?)
GP, BC, GH,
FC(?)
I
1
Simple gallery structure: tunnels branching usually not branching more than two times; 2 Complex gallery: tunnel branching several times;
Xylomycetophagy: eating xylem and fungal biomass; 4 Mycetophagy: eating fungal fruiting structures. Abbreviations: BC – brood care, FC –
fungus care, GE – gallery extension, GP – gallery protection, GH – gallery hygiene, I – insemination.
3
93
Chapter 6
Fig.1. Illustrations of the morphologies of Xyleborinus saxesenii galleries of three size classes in beech (Fagus sylvatica). Drawings made from original,
overwintering galleries (N = 18) dissected from a single trunk in late fall (9th November 2008, in Bern/CH, 560 m asl, 46°95’, 7°31’). Numbers of inhabitants are
given in the lower left hand corner of each figure (immat – immatures (larvae, pupae), ♀ - adult females, ♂ - adult males; for more details see Table 2 in the
Appendix section). The entrance tunnel is always directed to the side of the trunk, except in figures (O) and (P) where beetles bored into the trunk from the
top. Brood chambers of X. saxesenii are flat with a height of ~1mm and mostly span in one or two planes in the direction of the wood grains; chambers
diverging out of the pictured plane are indicated by white dashed lines in figures (F) and (G). Gallery parts that could not be fully traced are indicated by black
dashed lines. Figure (R) is modified from a photo (Biedermann et al. 2009) to illustrate larval and adult inhabitants (the right side could not be traced). The size
of an adult female is given on top of the scale. Original size.
94
Chapter 6
Fig.2. Illustrations of the
morphology of laboratory and
field Xyleborus affinis galleries.
(A) Photo of a laboratory
gallery in artificial medium. (B-I)
Field galleries adapted from
drawings of Schedl (1962; pp.
362-365;N=8). Tunnels have a
diameter of ~1mm and extend
in
different
directions,
independent of the wood grain,
either in xylem or phloem.
There is no communal brood
chamber, but brood is found in
tips of freshly bored tunnels as
indicated in figures (A) and (C).
It is unclear whether galleries
with two entrances (arrows in I)
result from two merging
galleries or the construction of
a second exit tunnel. The size of
an adult female is given on top
of the scale. Original size.
95
Chapter 6
Fig.3. Species comparison for cooperative (probably mutually beneficial) behaviours of the 2nd and 3rd larval instars. § - Digging and Cropping denote the
different feeding modes of Xyleborus affinis and Xyleborinus saxesenii larvae (see text). The bars show the mean (± se) proportion of time. Statistically
significant differences between the species are given (*** - p < 0.001, ** - p < 0.01; GEE, for details see Table 3 in the Appendix).
96
Chapter 6
Fig.4. Species comparison for cooperative (probably mutually beneficial) behaviours of adult females. The bars show the mean (± se) proportion of time of
Xyleborus affinis and Xyleborinus saxesenii larvae. Statistically significant differences between the species are given (*** - p < 0.001, * - p < 0.05; GEE, for details
see Table 3 in the Appendix).
97
Chapter 6
Fig.5. Types of ambrosia beetle galleries and important representative genera. Black tunnels are excavated by adult beetles, whereas pink areas are excavated by larvae (only
tunnels in F are excavated by both adults and larvae). Cooperative stands for division of labour between adults and larvae. Subsocial stands for parental care without division of
labour between adults and larvae. For details on the biology of the groups see Table 2.
98
Chapter 6
Supplementary material
Table S1. Xyleborinus saxesenii gallery sizes and compositions in winter. For schemes of
gallery morphology see Fig.1. Sizes in brackets denote galleries that could not be fully
measured.
Gallery
nd
st
1 instars
rd
2 &3
instars
Pupae
Teneral
♀♀
Adult
♀♀
♂♂
Size [mm ]
(86.2)
(75.2)
93
A
B
2
1
2
0
5
C
D
1
1
3
2
4
4
8
7
E
F
G
H
3
1
6
2
I
J
1
K
L
1
2
1
1
3
4
4
M
N
O
3
2
8
P
Q
2
5
R
Mean  se
6
11
5
8
2
1
7
12
3
10
11
1
14
22
32
1
8
14
21
10
2
1
7
21
22
66
24
32
1
7
7
16
29
6
1
55
75
12
69
1
2.1  0.8
9.9  4.3
0.3  0.3
5
1.3  0.7
99
8.4  1.8
1.2  0.5
2
Σ
126
26.6
(232)
(77.2)
50.8
47.7
107.2
(89.7)
137.3
(114)
107.4
232.7
172
87
(286.4)
(209)
23.2  6.5
126.13  17
Chapter 6
Table S2. Behavioural differences (p < 0.05) between the proportions of time the two study species
spent with the observed cooperative behaviours (for graphical illustration see Fig.3. and Fig.4.). GEEs
with an exchangeable correlation structure of the response variable within a cluster (= galleryidentity) were used to identify effects of the species on the total amount of time each behaviour was
observed. Model coefficients are reported as coeff.  se (standard error of the estimate), with
Xyleborus affinis in the first row of the model as the reference category (coefficient set to zero). A
positive contrast denotes that the mean of Xyleborinus saxesenii is higher than the mean of X. affinis;
a negative contrast denotes the reverse.
Class
nd
rd
2 &3
instar larvae
adult ♀♀
Behaviour
Model
coeff.  se
z
p
digging
Only in X. saxesenii
cropping
Only in X. affinis
cropping vs.
digging
Intercept (X. affinis)
Contrast X. affinis vs. X. saxesenii
-0.39  0.2
0.84  0.31
-1.9
2.7
0.06
0.007
balling
Only in X. saxesenii
allogrooming
Intercept (X. affinis)
Contrast X. affinis vs. X. saxesenii
-2.12  0.21
-1.67  0.45
-10.3
-3.74
<0.001
<0.001
blocking
Intercept (X. affinis)
Contrast X. affinis vs. X. saxesenii
-2.91  0.21
-0.17  0.63
-13.9
-0.27
<0.001
0.79
digging
Intercept (X. affinis)
Contrast X. affinis vs. X. saxesenii
-5.28  0.68
0.68  1.45
-7.82
0.47
<0.001
0.64
cropping
Intercept (X. affinis)
Contrast X. affinis vs. X. saxesenii
-2.36  0.23
1.04  0.44
-10.5
2.37
<0.001
0.018
shuffling
Intercept (X. affinis)
Contrast X. affinis vs. X. saxesenii
-0.07  0.09
-2.26  0.44
-0.79
-5.17
0.43
<0.001
allogrooming
Intercept (X. affinis)
Contrast X. affinis vs. X. saxesenii
-4.04  0.36
2.03  0.52
-11.3
3.91
<0.001
<0.001
100
Chapter 7
1
2
3
Dynamics of filamentous fungi in the complex ambrosia gardens of the
primitively eusocial beetle Xyleborinus saxesenii Ratzeburg
(Scolytinae: Curculionidae)
Peter H.W. Biedermann1,2, Kier D. Klepzig2,3, Michael Taborsky1 and Diana L. Six4
1
Department of Behavioural Ecology, Institute of Ecology and Evolution, University of Bern,
Baltzerstrasse 6, CH-3012 Bern, Switzerland
2
USDA Forest Service, Southern Research Station, 2500 Shreveport Hwy, Pineville, LA 71360, USA
3
US Forest Service, Southern Research Station, 200 WT Weaver Blvd, Asheville, NC 28804, USA
4
Department of Ecosystem and Conservation Sciences, College of Forestry and Conservation, The
University of Montana, Missoula, MT 59812, USA
Corresponding author: Peter H.W. Biedermann
Baltzerstrasse 6, CH-3012 Bern, Switzerland
phone: 0041 31 631 3015
e-mail: [email protected]
Species identity for Raffaelea-like sp. B and Artrinium sp. are not yet confirmed by sequencing! –
These species have been marked in grey throughout the text.
Abstract
Insect fungus gardens consist usually of a community of interacting microbes with both
beneficial and detrimental effects to the farmers. In contrast to fungus-farming ants and termites the
fungal communities of ambrosia beetles and the effects of particular fungal species on the farmers
are largely unknown. Here we used a laboratory breeding technique for studying the filamentous
fungal garden community of the xyleborine ambrosia beetle, Xyleborinus saxesenii Ratzeburg
(Scolytinae: Curculionidae), which cultures fungi in tunnel systems excavated within recently dead
trees. Raffaelea sulphurea and a Fusicolla acetilerea (Tubaki, C. Booth & T. Harada) (formerly
Fusarium merismoides var. acetilereum) -like fungus were transmitted in spore-carrying organs
(mycetangia) by gallery founding females and established first in new gardens. Abundance of R.
sulphurea had positive effects on the egg-laying of the female and on larval numbers. Over time, four
other fungal species emerged in the gardens, of which two, Paecilomyces variotii and Penicillium
decaturense, became increasingly abundant. P. variotii had a negative impact on larval numbers and
led to the death of parental females by forming biofilms on their bodies. It also comprised the main
portion of garden material removed from galleries by adults. Our data suggests that two mutualistic,
several commensalistic and one to two pathogenic filamentous fungi are associated with X. saxesenii.
Fungal diversity in gardens of ambrosia beetles appears at least ten times lower than in gardens of
fungus culturing ants.
Keywords: insect-fungus agriculture, cooperative breeding, mycetangium, mycangium, ambrosia
fungus gardens, mutualism, social behaviour, commensalism, mycophagy, eusociality, subsocial
Introduction
Mycophagy by insects has evolved in several lineages including springtails, flies, moths, wood
wasps, termites, ants and beetles (Wheeler and Blackwell 1984; Martin 1987; Wilding et al. 1989).
Among these, only attine ants, macrotermitine termites and curculionid ambrosia beetles evolved
advanced fungus agriculture (Mueller et al. 2005). This involves (1) obligate nutritional dependence
101
Chapter 7
on fungal food for adults and their brood, (2) translocation of their fungal crops by spore- or
propagule-carrying organs within nests and when founding new nests, and (3) cultivation and
management of the fungal crops (i.e., continuous monitoring, sustainable management and
protection, weeding and biological control of alien microbes). As the latter can be easier managed by
a group of individuals partitioning the labour, advanced fungus agriculture is often associated either
with a subsocial (most ambrosia beetles) or eusocial life strategy (all farming ants and termites, one
ambrosia beetle; Mueller et al., 2005).
Fungus agriculture has been well studied in the fungus-farming ants and termites, but is yet
little understood in ambrosia beetles. These beetles dwell in the wood of (usually) recently dead or
weakened trees, where they construct tunnel systems (galleries) upon the walls of which they
nurture ambrosia gardens. Ambrosia gardens consist of fungi the beetles carry into trees in their
intestines or, more commonly, in specialized structures called mycetangia or mycangia (FranckeGrosmann 1956; 1975). This vertical transmission of ambrosia fungi from the beetle`s natal galleries
to newly founded nests is associated with the co-evolution between fungi and beetles and often a
species-specific association between partners (Six 2003). For many species, healthy fungus gardens
are dominated by mutualistic ambrosia fungi of the genera Raffaelea and Ambrosiella (Ascomycota;
Harrington et al. 2010). These species usually show an ambrosial- or kohlrabi-like growth form within
the gardens, forming nutrient-rich fruiting structures (e.g. conidiospores) that are grazed by adult
beetles and their offspring. Gardens also contain a complex of other filamentous fungi, yeasts and
bacteria (e.g.Haanstad and Norris 1985), which are often transmitted by spores sticking to the body
of founding females (Francke-Grosmann 1967). The effects of these microbes are largely unknown.
The associated fungi of only three of about 3500 ambrosia beetles worldwide have yet been
investigated (Kajimura and Hijii 1992; Harrington and Fraedrich 2010; Endoh et al. 2011), and the
fungal dynamics in mycangia and galleries have been studied only in one ambrosia beetle,
Xylosandrus mutilatus (Kajimura and Hijii 1992). The interaction between ambrosia beetles and their
associates has not yet been studied, although both adults and larvae can probably alter the fungal
community of their ambrosia gardens, as has been observed in fungus farming ants (Mueller et al.
2001). Adult ambrosia beetles are particularly attracted to their primary ambrosia fungus and
repelled by fungal pathogens (Hulcr et al. 2011). An observational study of ambrosia beetle
behaviours within their nests revealed that adults and their larvae closely interact with their gardens
(Biedermann and Taborsky 2011). Larvae were observed to cooperate in various duties, which is
exceptional for holometabolous insects,; they enlarge the gallery by digging and thereby create space
for the fungi to spread, they fertilize fungi with their excretions, clean colony members and gallery
walls which prevents the spreading of mold, and participate in the removal of waste from the tunnel
system (Biedermann and Taborsky 2011). Adults were observed to block tunnels, thus potentially
regulating the microclimate of the gardens (Kirkendall et al. 1997), and to graze their gardens, which
affects fungal growth (French and Roeper 1972) and potentially also species composition
(Biedermann and Taborsky 2011). Ambrosia fungi have been shown to dominate gardens in recently
founded galleries and newly built tunnel systems, and decrease in abundance relative to invading
weed fungi at the end of gallery life, when beetles leave the nest, and in old parts of the tunnel
system (e.g. Kajimura and Hijii 1992). The abundance of certain fungi may positively or negatively
affect the brood. For example, fungal associates of Xyleborus ferrugineus (Fabricius) vary
considerably in sterol, lipid and amino acid content, and thus in their nutritional quality for the
developing brood (Kok and Norris 1972a; Kok and Norris 1972b; Kok and Norris 1973).
Here we report a comprehensive survey of the filamentous fungi closely associated with the
ambrosia beetle Xyleborinus saxesenii Ratzeburg, and their dynamics in relation to the life history of
this species. In addition we report which fungal associates of X. saxesenii are carried in the
mycetangia and guts of females during their dispersal flight. Based on previous studies we expected
to find Raffaelea sulphurea (L.R. Batra) T.C. Harr. (syn. Ambrosiella sulfurea; (Harrington et al. 2010)
as the primary symbiont (Francke-Grosmann 1956; 1975; Batra 1967; Roeper et al. 1980; Roeper and
French 1981), but hitherto the identity of other fungal associates has been unknown. To follow the
dynamics of all fungi within the beetle gardens and their effect on the brood, we sampled garden
102
Chapter 7
material over the entire developmental period of a brood and recorded brood numbers and offspring
development. This was possible with help of a laboratory rearing technique allowing to observe the
beetles within their galleries (Biedermann et al. 2009). Furthermore, we identified (i) fungi removed
from the galleries by adults (to the dumps; i.e. the material disposed out of the entrance tunnel), and
(ii) detrimental fungi growing on the bodies of beetles. Finally, we discuss our results in comparison
to fungal communities found within gardens of fungus-growing ants.
Materials & Methods
Study species
Xyleborinus saxesenii is one of the most common ambrosia beetles in temperate zones
worldwide. Originally native to Eurasia, over the last 200 years it has been introduced into parts of
Africa, Oceania, as well as South and North America (for the actual distribution see
http://xyleborini.tamu.edu/public/site/scolytinae/home). The species is still spreading, facilitated by
the shipment of woody products around the world, and by characteristics of its own biology. X.
saxesenii shows little host tree preference and probably lacks on inbreeding depression because of a
mating system in which sib-mating is the rule between haploid brothers and diploid sisters in their
natal nest (as in Xylosandrus germanus; Peer and Taborsky 2005). Therefore, the translocation of a
single female may be sufficient for the successful establishment of a new population without
negative effects of a genetic bottle-neck.
Galleries of X. saxesenii are founded by single females which dig a vertical entrance tunnel of a few
centimetres into a tree trunk. They inoculate gallery walls with fungi, lay eggs when fungal gardens
have established, and later care for the developing brood. The larvae feed on fungus-infested wood
and in this way gradually enlarge the tunnel to a flat brood chamber (Roeper 1995). This
xylomycetophagous feeding is typical for larvae in the genus Xyleborinus and likely serves to reduce
kin competition, as it increases the space for ambrosia gardens to grow and improves the breakdown
of wood by enzymes (De Fine Licht and Biedermann 2012). Wood passes the guts of larvae without
being digested, but finely grounded and readily utilized by the fungi. Such woody frass is partly
spread on the ambrosia garden microbes, which probably recycle and fully breakdown this material
(Biedermann and Taborsky 2011).
Overlapping generations are typical in X. saxesenii nests, because adult females delay
dispersal after maturation and fertilization by a brother (Peer and Taborsky 2007; Biedermann et al.
2011). During this time they engage in brood and fungus care, thereby increasing gallery productivity
(Biedermann and Taborsky 2011). Additionally, fungal gardens benefit from the recycling of their
excretions (Abrahamson and Norris 1970). About 20 % of daughters also reproduce in their natal nest
(Biedermann 2007; Biedermann et al. 2011).
Beetle collection and laboratory breeding
About 100 X. saxesenii females were caught alive with Lindgren funnel traps baited with
ethanol in Pineville, LA, USA (38m asl; 31°20’, 92°24’) in Summer 2007. Collection cups were filled
with damp sterile filter paper and emptied twice daily to avoid microbial contaminations of the
beetles (Benjamin et al. 2004). In the lab we surface-sterilized the beetles by rinsing them twice for a
few seconds, first with 70% ethanol and afterwards with deionised water. This treatment does not
harm the fungal spores of the cultivar within the mycetangium, but should reduce fungal
contamination by eliminating some of the spore-load sticking to the body surface of the beetles (e.g.
molds). It is necessary for laboratory breeding of ambrosia beetles, because these “contaminants”
establish relatively easily in our homogenic artificial media than under natural conditions
(Biedermann et al. 2009), where beetles largely surface-sterilize themselves boring through bark rich
in fungitoxins and other antibiotics substances (Berryman 1989). Surface-sterilization – both in the
laboratory and in the field – is likely incomplete, however, because specific contaminants take-over
old galleries (Kajimura and Hijii 1992), which must have been transmitted initially as sticky spores in
pits of the exoskeleton (Bridges et al. 1984). Thirteen females were used for fungal isolations (see
103
Chapter 7
below). The remaining females were placed singly on an agar-sawdust-based rearing medium in glass
tubes (for details on this technique and ingredients of the modified medium we used see
(Biedermann et al. 2009). Tubes were closed with plastic caps, stored vertically, and wrapped in
paper in a way that allowed light to penetrate the tube only from the top. This way beetles
frequently bored tunnels next to walls of the glass tube, allowing observations of brood development
when the paper was removed (Biedermann 2007). Tubes were kept at 23°C.
Fungus isolations from adult females captured during dispersal flight
After surface-sterilization, we aseptically dissected mycetangia from 13 adult females using
fine tweezers under a microscope (6.4× – 40× magnification). The mycetangium in X. saxesenii is a
paired cavity at the basis of the females’ elytra; for isolating the spores present in the mycetangia,
we removed the two elytra and placed their bases on malt agar (MA: 25 g malt extract, 20 g agar, 1 l
deionized H2O) plates. Elytral mycetangia were too small to be dissected successfully, so we cannot
exclude that our isolations also contain fungi sticking to the upper and the bottom side of the elytra.
Guts of the beetles were dissected and squashed in a sterile Petri dish. Gut material was then spread
across the surface of MA plates using a sterile metal loop. All cultures were then incubated at 25°C in
darkness for about two weeks, and purified by subculturing.
Fungal isolations from laboratory galleries
After introduction into tubes, females usually bored an entrance tunnel and inoculated the
medium with fungi. In previous lab studies, it was observed that ambrosia beetle females first feed
on the fungus and then lay eggs (e.g. Kingsolver and Norris 1977) for about 40 days. During that time,
four periods of gallery development can be discerned: (1) three to five days with only the foundress
and eggs present, (2) at least ten days with foundress, eggs and immatures (larvae and pupae)
present, (3) about 40 days with foundress, eggs, immatures and adult offspring present, and (4) the
nest-leaving phase when the foundress has died, offspring have matured and gradually disperse
(Biedermann et al. 2011). We timed our sampling of brood and fungi in accordance with this gallery
development pattern: we dissected eight galleries in period 1, ten galleries in period 2, and nine
galleries in period 4. We did no dissections during period 3, because we expected only gradual
changes in between fungal frequencies found in period 2 and period 4, and thus decided to not
reduce the sample sizes in the other groups any further. Additionally, we dissected eight galleries
that did not produce brood within one month post-introduction of the female, and nine galleries
where all larvae died. From each of the galleries, we took eight samples from the gallery wall of the
entrance tunnel (= oldest part of gallery) and eight samples from the gallery wall of the brood
chamber (where most inhabitants were present) using a sterile needle. Four of these samples we
placed on MA and four on cycloheximide-streptomycin malt agar (CSMA: 10 g malt extract, 15 g agar,
20 ml filter sterilized CSMA stock solution containing 2 mg of Cycloheximide and 1 mg Streptomycin,
1 L deionized H2O) plates. Samples were placed in the middle of the plates.
Additional isolations
Single, live and healthy larvae from five different galleries were squashed in sterile Petri
dishes and then plated on MA and CSMA agar using sterile metal loops. Four live adult females were
found to be overgrown by a biofilm of fungi, which we isolated by scrapping parts off with a sterile
needle and plating four samples from each insect on MA. The fungal composition of eight gallery
dumps (= frass and sawdust shuffled out of the nest onto the surface of the medium) was analysed
by plating four samples of each on MA and CSMA.
Identification of filamentous fungus isolates
Isolated fungi were initially placed into groups based on cultural colony characteristics (i.e.
morphology and color of mycelium and fruiting structures). Representative samples were used for
DNA sequencing. To extract DNA, a small amount of mycelium and conidia was scraped from the
surface of young, relatively unmelanized colonies growing on MA, or hyphae were taken from
cultures grown in 2% malt extract broth (MEB). The mycelium was macerated in 200 µl PrepMan
104
Chapter 7
Ultra (Applied Biosystems, USA), incubated at 95°C for 10 min, and then centrifuged. The supernatant
containing DNA was then used for PCR amplification of a portion of the ribosomal RNA encoding
region and partial b-tubulin gene with the primer pairs ITS3 (White et al. 1990) and LR3 (Vilgalys and
Hester 1990), and Bt2b (Glass and Donaldson 1995) and T10 (O'Donnell and Cigelnik 1997). PCR
conditions used have been described by (Six et al. 2009). Amplicons were purified using a High Pure
PCR Product Purification Kit (Roche, Germany) and sequencing was performed on an ABI 3130
automated sequencer (Perkin–Elmer Inc, USA) at the Murdock Sequencing Facility (University of
Montana, Missoula, MT USA). DNA sequences of representative isolates were deposited in GenBank
(Table 1). Contigs of forward and reverse sequences obtained with each primer pair were aligned in
MEGA4 (Tamura et al. 2007). BLAST searches were done with sequences of each isolate in the NCBI
GenBank database (http://www.ncbi.nlm.nih.gov). Cultures of isolates from this study were
deposited in the culture collection of Diana Six (DLS) at the University of Montana, Missoula, MT,
USA, and the Centraalbureau voor Schimmelcultures (CBS), Utrecht, The Netherlands (Table 1).
Statistical analyses
Using all 16 samples taken from each laboratory gallery we estimated the frequency of each
fungal species. Additionally, we determined whether a fungus was present or not in the eight
samples taken from the main tunnel and the brood chamber of each gallery. For each fungal species
we analysed how its frequency and presence (dependent variables) were affected by the culture
medium (MA, CSMA), the location within the gallery (brood chamber, main tunnel) and the period of
offspring development (foundress with eggs, foundress with larvae, only adult progeny; fixed factors)
by controlling for gallery of origin (random factor). Using data from the first two periods only
(foundress with eggs, foundress with larvae) we also tested if numbers of eggs and larvae correlated
with fungal frequencies. All analyses were done using generalized estimating equations (GEEs) in R
(lmer; Version 2.12.1; (R Development Core Team 2008). GEEs are an extension of generalized linear
models with an exchangeable correlation structure of the response variable within a cluster, which
allows for controlling the variation between observations from a single gallery. This was necessary
because of variation in sample sizes between galleries, as plates had to be excluded from the
analyses if they did not yield any microbial growth.
Results
Our morphological data in combination with DNA sequencing revealed that six species of
filamentous fungi were associated with X. saxesenii (Table 1). As expected, Raffaelea sulphurea
(formerly Ambrosiella sulfurea) was regularly isolated from X. saxesenii bodies and galleries. Both
-tubulin) generated for this fungus matched those in GenBank (accessions of
subject sequences) for this species at 100%. The isolates also exhibited the distinct morphology of
this species which is….. ITS sequences for the other fungus isolated regularly most closely matched
sequences for Fusicolla acetilerea (formerly Fusarium merismoides var. acetilereum) in GenBank
(accessions)
-tubulin sequence generated for this
fungus (closest match, 90% to Fusarium ciliatum). The dark morphospecies isolated from gallery
dumps and biofilms found on dead insects matched morphological descriptions for Paecilomyces
variotii. Sequences for this fungus also matched those for this species in GenBank (accessions ITS,
-tubulin, 99% 1 base pair difference). Less commonly isolated fungi included Penicillium
decaturense (accessions ITS 99% match in GenBank, morphological and cultural characters also well
matched to species description), a Raffaelea-like sp. B fungus (no close match (<90%) in GenBank to
either ITS or b-tubulin sequences; id not finished yet), and an Artrinium sp. (sequencing not finished).
Overall, R. sulphurea (GEE: p < 0.001), the Raffaelea-like sp. B (p = 0.003) and Artrinium sp.
(only on CSMA) were more commonly detected on CSMA than on MA, whereas the opposite was
105
Chapter 7
true for F. acetilerea-like sp. (p < 0.001; Table 2). The other species were isolated equally often from
CSMA and MA.
Mycetangially-transmitted fungi
F. acetilerea-like sp. dominated in mycetangia of all thirteen dissected females and was
found in six of their guts. R. sulphurea was present in mycetangia of only one of these females, but
was isolated from 9 of 13 female guts (Fig. 1).
Fungus dynamics in relation to offspring development
R. sulphurea and F. acetilerea-like dominated the gardens of freshly founded galleries after
the foundresses had started to lay eggs (period 1, Fig. 1). Egg numbers tended to increase with
increasing abundance of R. sulphurea (GEE: pfrequency = 0.09; Table S2). Fungus composition (frequency
and presence of single species) of unsuccessful galleries without any eggs did not differ from galleries
with eggs (p < 0.05).
All six species of fungi were isolated from samples taken during the period after eggs had
hatched (period 2). F. acetilerea-like sp. increased in abundance (period 1 vs. 2: ppresence = 0.02; Table
S1). This had no effect on larval numbers, however (p = 0.92; Table S2). Larval numbers were
positively correlated with the frequency of R. sulphurea (p = 0.035), and tended to correlate
negatively with the frequency of P. variotii (p = 0.063; Fig.2, Table S2). Fungus composition
(frequency and presence of single species) of galleries in which all larvae died during development
did not differ from galleries with successfully developing larvae (p > 0.05).
Abundance of R. sulphurea (period 1 vs. 4: pfrequency = 0.028) and P. variotii (period 2 vs. 4;
pfrequency = 0.001) were significantly lower after maturation of all offspring (Table S1). Only Pe.
decaturense increased towards this period (period 1+2 vs. 4: ppresence = 0.02). Data for F. acetilerealike sp. is contradictory; the number of galleries where it was present decreased (period 2 vs. 4:
ppresence = 0.03), whereas the relative frequency within galleries increased (period 2 vs. 4: pfrequency =
0.002).
Fungal composition in relation to location
R. sulphurea (pfrequency < 0.001, ppresence = 0.12) and F. acetilerea-like sp. (pfrequency = 0.002,
ppresence = 0.08) were more common in the brood chamber than in the main tunnel of the galleries,
while an opposite trend P. variotii (pfrequency < 0.001, ppresence = 0.01; Table S1). Data on frequency and
presence for Raffaelea-like sp. was contradictory (increase pfrequency = 0.09, decrease ppresence = 0.07).
P. variotii was the dominant species growing in the dumps of the beetles (present in 7 of 8 galleries),
followed by F. acetilerea-like sp. (in 2 of 8 galleries), Raffaelea-like sp. (in 1 of 8 galleries) and Pe.
decaturense (in 1 of 8 galleries; Fig.1).
Fungi on the body of adults
We frequently saw foundresses that were overgrown with a thin layer of fungi. If they were
not able to successfully produce offspring (who would have groomed off this layer) this likely led to
the death of these females, because at some point this layer became so thick that it constricted
movements of beetles through the tunnels (Biedermann and Taborsky 2011). P. variotii (present on
4 of 4 beetles) was the main component forming this layer, but F. acetilerea-like sp. (in 2 of 4 beetles;
Fig. 1) was also found.
Discussion
Fungus isolations from laboratory-reared galleries revealed that two filamentous fungi, R.
sulphurea and F. acetilerea-like sp. are regularly associated with X. saxesenii. Nest-founding females
transferred both species from the natal nest in their spore-carrying organs (mycetangia and guts),
which led to the initial prevalence of these species in the fungal gardens and during the period of
106
Chapter 7
larval development. The primary food fungus, R. sulphurea, formed thin ambrosia layers that were
fed upon by the adults (Fig. 3; Biedermann and Taborsky 2011). F. acetilerea-like sp. probably served
as a secondary and/or larval food source. However, over time five other filamentous fungi,
Arthrinium sp., Raffaelea-like sp. B, Pe. decaturense and P. variotii, appeared within the fungal
gardens, of which the first two seem to be commensal. Pe. decaturense is probably weakly parasitic.
P. variotii appears strongly pathogenic to the beetles, because it invaded the fungal gardens over
time and negatively affected larvae and adults.
The mutualistic associates of X. saxesenii
R. sulphurea and F. acetilerea-like were the only fungi isolated from the spore-carrying
organs of dispersing females ready to found new fungus gardens. Their elytral pouches (mycetangia)
contained mostly spores of F. acetilerea-like sp., whereas R. sulphurea dominated in gut samples.
While fungus farming ants and termites actively pick symbiont propagules to found new fungus
gardens (Mueller et al. 2005), ambrosia beetles passively take up spores into external mycetangia
from the surrounding environment when moving within their natal nest before dispersal (Beaver
1989). At that time (which is the period when only adult progeny is present) galleries were already
heavily contaminated by four other non-mutualistic fungi. Selective substances produced in the gut
lumen and by numerous glands lining the beetles` mycetangia (Schneider and Rudinsky 1969;
Schneider 1991) likely assure the exclusive transmission of R. sulphurea and F. acetilerea-like sp.;
both symbionts were isolated from both organs, although at different frequencies. In summary, this
confirmed the existence of two spore carrying modes in X. saxesenii (Francke-Grosmann 1975) and
suggests important roles for both fungi in the life-cycle of this beetle.
Several observations suggest a strong mutualistic role of R. sulphurea in this system. First, it
prevailed in galleries shortly after their foundation and throughout brood development, where it
formed characteristic ambrosial layers of densely packed conidia with large nutritional conidiospores
on the gallery walls (Fig. 3). Ambrosia layers are predominately formed within the brood chambers,
where adult daughters and larval offspring aggregate and crop off the nutritional conidia
(Biedermann and Taborsky 2011). Second, the number of eggs produced by the foundress and larval
numbers tended to correlate positively with the frequency of R. sulphurea. Third, the abundance of
R. sulphurea was lowest in galleries with only adult progeny, suggesting that egg production ceased
when productivity of this fungus dropped below a certain threshold. Fourth, R. sulphurea has been
isolated from mycetangia and galleries of X. saxesenii originating researchers from all over the US
and Europe (Francke-Grosmann 1956; 1975; Batra 1967; Roeper et al. 1980; Roeper and French
1981). Raffaelea (Ascomycota: Ophiostomatales) and Ambrosiella (Microascales) are also the primary
mutualists of many other temperate ambrosia beetles (Roeper and French 1981; Farrell et al. 2001;
Harrington et al. 2010).
The role of F. acetilerea-like sp. for the beetles is less clear. It was part of the fungal layer
that sometimes formed on adults. Its ability to grow on body surfaces might explain why it was so
frequently isolated from the elytral mycetangia of beetles. On the other hand, observations suggest
that Fusarium sensu lato may also grow in a yeast phase within galleries, similar to the ambrosial
growth of Raffaelea species (Abrahamson 1969; R.A. Roeper, personal communication), which might
have been present also in our galleries. It also was common within brood chambers during
development of larvae indicating that it may serve as an additional food source for them. Larvae feed
differently than adults that feed on ambrosial layers, however, with most feeding occurring on
fungus-infested wood and faeces of other larvae (Biedermann and Taborsky 2011). Enzymatic activity
in guts of larvae and adults also appear to differ (De Fine Licht and Biedermann 2012), which
indicates that they may feed on different fungi. Furthermore, Fusarium merismoides has recently
been found to be of some nutritional value for larvae of the ambrosia beetle Platypus quercivorus
(Platypodinae; Qi et al. 2011). While Raffaelea and Ambrosiella are often cited as the main food fungi
associated with ambrosia beetles, Fusarium sensu lato appear to be extremely common, but its role
for the beetles is obscure and might depend on the host. Fusarium solani, for example, is weakly
pathogenic for the bark beetle Dendroctonus frontalis (Moore 1973). In galleries of the ambrosia
107
Chapter 7
beetles Anisandrus dispar and Xyleborus ferrugineus it is very common, however, and in the latter it
can even serve as the primary food fungus (Zimmermann 1973; Norris 1979). Species of the genus
Fusarium sensu lato are frequently isolated from soil and plants, and also are commonly associated
with ambrosia beetles (Norris 1979). Unpublished studies by R.A. Roeper suggest that Fusarium
associated with ambrosia beetles may have higher temperature growth optima than Raffaelea
species, which might explain why Fusarium species are more commonly found as associates of
ambrosia beetles in the tropics and in the southern US than in the northern US (our study was done
in Louisiana; R.A. Roeper, personal communication). Fusarium has long been acknowledged as a
complex, unresolved polyphyletic group of fungi. Recently, the taxonomy of Fusarium has been
revisited with the result that some species have been moved to different genera, including Fusicolla
(Grafenhan et al. 2011). It is likely that many ambrosia beetle associates formerly identified as
Fusarium will be reclassified in new genera. In summary, these observations suggest a secondary
mutualistic role of F. acetilerea-like sp. for X. saxesenii, but experimental studies are needed to
determine whether this is really the case.
The cultivation of two or more mutualists is apparently common in ambrosia beetles (Norris
1979; Haanstad and Norris 1985; Harrington and Fraedrich 2010; Endoh et al. 2011), against
predictions of most hypotheses regarding the formation and maintenance of mutualism. Symbiont
competition can generate selection for symbiont traits that enhance their competitive ability but
harm the host (Frank 1996; Mueller 2002). Additionally, there should be strong selection for a ‘best
symbiont’, over time leading to its fixation with a host. However, while some symbioses, including
fungus-farming ants and termites, involve only one mutualistic partner (e.g. Mueller et al. 2005),
many others involve multiple symbionts indicating mechanisms that allow coexistence. In the case of
symbionts in ambrosia beetle gardens, niche differences of the various fungi may reduce
competition. If the fungi exploit different resources in the tree this may alleviate selection against
any one partner and help to maintain a community of symbionts rather than a single fungal partner.
In X. saxesenii, one fungus species might serve as food for adults and the other one as food for larvae
(De Fine Licht and Biedermann 2012). Cooperation between symbionts is also possible. Laboratory
studies showed that R. sulfurea has a need for B-vitamins to grow, which might be provided by other
microbes (filamentous fungi, yeasts or bacteria; R.A. Roeper, personal communication). In
Dendroctonus bark beetles (Scolytinae), the possession of several apparently redundant fungal
symbionts with differing environmental tolerances may reduce the risk of the host being left
aposymbiotic when environmental conditions shift over a season and from year to year (Six and
Bentz 2007). Experimental studies are needed to clarify the roles and interactions of the various
symbionts associated with bark and ambrosia beetles.
Other fungal associates
Species of the anamorphic genera Arthrinium, Paecilomyces, and Penicillium were also
associated with X. saxesenii, without being transmitted by founder females in their spore-carrying
organs. Instead, spores of such fungi have been found to be vectored in small quantities on females`
body surfaces (Francke-Grosmann 1967). Arthrinium sp. was isolated at low frequencies from all
gallery-classes and also from larval bodies. The presence of Arthrinium sp. did not affect adult beetles
or larvae in this study
Pe. decaturense and P. variotii predominated in old galleries, at the entrance tunnel and in
gallery dumps. We regard both as parasites, although we found no negative effects of Pe.
decaturense in this study. However, its known to produce antiinsectan compounds (Zhang et al.
2003) and previous to this study Pe. decaturense has been only isolated living (and probably feeding)
on a wood decay fungus (Peterson et al. 2004). Penicillium sp. often compete with insects for
ephemeral resources and thus regularly produce compounds against insect feeding (Peterson et al.
2004; Rohlfs and Churchill 2011).Unidentified Penicillium sp. have been frequently reported from old
galleries of X. saxesenii and have been regarded weak antagonists (Fischer 1953; Francke-Grosmann
1975). For P. variotii there is clear evidence for its antagonistic effect on X. saxesenii: Its abundance
negatively affected larval numbers and it formed a deadly fungal layer on the surface of adult
108
Chapter 7
beetles. In a previous study we found this layer to have caused the death of at least 7 out of 29
isolated females (Biedermann and Taborsky 2011). The genus Paecilomyces includes many
entomopathogenic species and also plant saprobes belonging to the earliest colonizers of recently
dead plants (e.g. Kim et al. 2001; Tang et al. 2005).
Fungus dynamics in a laboratory setting vs. field galleries
Ambrosia beetles live in the wood of trees, where they can only be studied by destructive
gallery dissection. Thus a laboratory setting was required to study the fungus dynamics in relation to
the dynamics of the beetles’ life history within galleries. It is intrinsic to all laboratory studies,
however, that results might be influenced by the difference between laboratory and field conditions.
Our artificial breeding medium, for example, is richer in nutrients and moisture than natural wood
(Saunders and Knoke 1967). Therefore it is important to determine whether these differences could
have influenced the conclusions of our study: First, total offspring numbers of X. saxesenii in field and
laboratory galleries are almost identical (Biedermann et al. 2009). Second, different substrate
conditions might influence the abundance of particular fungi relative to others, but should not affect
within species dynamics (e.g. time course of abundance in dependence of gallery stage and
composition). Thus, we believe that our conclusions are generally valid but the results on relative
fungal abundance should not be overinterpreted.
Can beetles influence the community of their gardens?
Larvae and adult X. saxesenii constantly remove the growth of F. acetilerea-like sp. and P.
variotii from their body surface by grooming each other. They also constantly crop their gardens and
hinder the spread of pathogens by dumping old sawdust, faeces, fungal material and dead individuals
out of the gallery entrance (Biedermann and Taborsky 2011). Symbionts may profit from direct
fertilization by the faeces of larvae and adults; nitrogen is known to be recycled by the primary fungi
of X. ferrugineus (Norris 1975). If larvae and adults are removed from a gallery, its fungus gardens are
overrun by saprobic fungi (normally coexisting at low levels) within 1-2 days (Leach et al. 1940; Batra
1979; Norris 1979; Biedermann and Taborsky 2011), which demonstrates that the beetles play an
active role in maintaining the composition of their gardens. As mechanical removal is likely to be only
partially effective against weed-fungi, beetle-associated antibiotics producing bacteria may play a
role in controlling weeds and pathogens as in other fungus-culturing insects (Mueller et al. 2005). In
Dendroctonus bark beetles several bacterial groups are known to reduce the growth of antagonistic
fungi (Scott et al. 2008) and some of them are actively applied within oral secretions during
specialised cleaning behaviours by the adults (Cardoza et al. 2006). Streptomyces griseus, which is
known to produce antibiotics, has been recently isolated from X. saxesenii galleries (Grubbs et al.
2011), but whether bacteria influence the composition of ambrosia gardens should be investigated in
future studies.
This is the first study reporting correlative evidence for fitness effects of particular fungi on
an ambrosia beetle. A relatively high number of filamentous fungi are commonly associated with the
ambrosia beetle X. saxesenii, which suggests a predictable rather than an accidental presence in the
fungus gardens. Interestingly, most of the secondary fungal flora that we reported has been also
isolated from nests of fungus-growing ants (Rodrigues et al. 2008; 2011): Fusarium sp., Paecilomyces
sp. and Penicillium sp. are regularly isolated from different attine ant species. The reason might be
that these genera are often associated with plants, either as endophytes or early saprobes. Thus,
these fungi are likely present in the plant material the ants collect to provision their gardens
(Rodrigues et al. 2011) as well as in the wood surrounding ambrosia beetle galleries; although in the
present study fungi must have been vectored by the dispersing females. The absence of a strong
association between these secondary associates and the farming insects in both systems suggests
that most of these filamentous fungi are transient components of the gardens. This does not mean,
however, that secondary microbes do not influence insect-fungus symbioses. Whereas research on
ambrosia beetle microbial symbioses is still at the beginning, many important mutualists and
109
Chapter 7
antagonists have been reported from the fungus –farming ants (Silva et al. 2006; Pinto-Tomas et al.
2009).
Our study revealed six fungal species within ambrosia beetle gardens, which is at least ten
times less then recently isolated from fungus-growing ant gardens in the field (between 66 and 106
fungal species depending on the ant species; (Rodrigues et al. 2011). Comparing species
accumulation curves of both studies (Fig. S1 vs. Fig. 2 in Rodrigues et al. 2011) suggests that in this
study we have sampled all fungi present, whereas many more might be detected by increasing
sampling effort in the ants. Of course these differences in absolute species numbers between beetles
and ants is in part an effect of laboratory breeding, but this must not hide the fact that the beetle
galleries are a more closed system than the ant nests (U.G. Mueller, personal communication). First,
the ants build their nests in soil, which is heavily contaminated by microbes. Second, they have to
leave the nest to forage and thus are exposed to all kinds of contaminants. Finally, they have to use
substrate that contains all kinds of endophytic and epiphytic microbes. The beetles, in contrast, bury
into dying or recently dead wood, a substrate, which is much less contaminated by microbes than
soil. Thus, it seems that ambrosia beetles have an easier time to sequester their gardens from foreign
fungi than the fungus-farming ants, despite the fact that beetles cannot protect their gardens from
fungi that grow in from the surrounding wood (the ants sequester their gardens from the
surrounding soil; Hölldobler and Wilson 1990) that serves as the food-substrate for their fungi. In
summary, sophisticated techniques for weeding and disinfection like in the ant gardens (e.g. Currie et
al. 1999; Currie and Stuart 2001; Boomsma and Aanen 2009) might not be needed in beetle gardens.
The biggest threat to ambrosia gardens are probably diminishing nutrients, so recycling of excretions
may be important. Indeed, some evidence of nutrient cycling between beetles and fungi exist (Kok
and Norris 1972a; 1972c; Biedermann and Taborsky 2011).
Literature
Abrahamson LP (1969). Physiological interrelationships between ambrosia beetles and their symbiotic fungi.
Ph.D.-thesis, University of Wisconsin, Madison.
Abrahamson LP & Norris DM (1970) Symbiotic interrelationships between microbes and ambrosia beetles.
5.Amino acids as a source of nitrogen to fungi in beetle. Annals of the Entomological Society of America 63:
177-180.
Batra LR (1967) Ambrosia fungi - A taxonomic revision and nutritional studies of some species. Mycologia 59:
976-1017.
Batra LR (1979) Symbiosis, commensalism and aposymbiosis - conclusions. Insect-Fungus Symbiosis. Nutrition,
Mutualism, and Commensalism (Batra LR, ed), pp. 259-265. John Wiley & Sons, New York.
Beaver RA (1989) Insect-fungus relationships in the bark and ambrosia beetles. Insect-fungus interactions
(Wilding N, Collins NM, Hammond PM & Webber JF, eds), pp. 121-143. Academic Press, London.
Benjamin RK, Blackwell M, Chapela IH, Humber RA, Kevin G, Klepzig KD, Lichtwardt RW, Malloch D, Noda H,
Roeper RA, Spatafora JW & Weir A (2004) Insect-and Other Arthropod-Associated Fungi. Biodiversity of
Fungipp. 395-433. Academic Press, Burlington.
Berryman AA (1989) Adaptive pathways in scolytid-fungus associations. Insect-Fungus Interactions (Wilding N,
Collins NM, Hammond PM & Webber JF, eds), pp. 145-159. Academic Press, London.
Biedermann, PHW (2007) Social behaviour in sib mating fungus farmers. Master thesis, University of Berne,
Switzerland.
Biedermann PHW, Klepzig KD & Taborsky M (2009) Fungus cultivation by ambrosia beetles: Behavior and
laboratory
110
Chapter 7
breeding success in three xyleborine species. Environmental Entomology 38: 1096-1105.
Biedermann PHW, Klepzig KD & Taborsky M (2009) Fungus cultivation by ambrosia beetles: behavior and
laboratory breeding success in three xyleborine species. Environmental Entomology 38: 1096-1105.
Biedermann PHW, Peer K & Taborsky M (2011) Female dispersal and reproduction in the ambrosia beetle
Xyleborinus saxesenii Ratzeburg (Coleoptera; Scolytinae). Mitteilungen der Deutschen Gesellschaft für
allgemeine und angewandte Entomologie in review.
Biedermann PHW & Taborsky M (2011) Larval helpers and age polyethism in ambrosia beetles. Proc. Natl Acad.
Sci. USA 108: 17064-17069.
Boomsma JJ & Aanen DK (2009) Rethinking crop-disease management in fungus-growing ants. Proceedings of
the National Academy of Sciences of the United States of America 106: 17611-17612.
Bridges JR, Marler JE & McSparrin BH (1984) A quantitative study of the yeasts and bacteria associated with
laboratory-reared Dendroctonus frontalis Zimm. (Coleopt., Scolytidae)1. Zeitschrift für Angewandte
Entomologie 97: 261-267.
Cardoza YJ, Klepzig KD & Raffa KF (2006) Bacteria in oral secretions of an endophytic insect inhibit antagonistic
fungi. Ecological Entomology 31: 636-645.
Currie CR, Scott JA, Summerbell RC & Malloch D (1999) Fungus-growing ants use antibiotic-producing bacteria
to control garden parasites. Nature 398: 701-704.
Currie CR & Stuart AE (2001) Weeding and grooming of pathogens in agriculture by ants. Proceedings of the
Royal Society of London Series B-Biological Sciences 268: 1033-1039.
De Fine Licht HH & Biedermann PHW (2012) Patterns of functional enzyme activity suggest that larvae are the
key to successful fungus farming by ambrosia beetles. Front.Zool., in review.
Endoh R, Suzuki M, Okada G, Takeuchi Y & Futai K (2011) Fungus Symbionts Colonizing the Galleries of the
Ambrosia Beetle Platypus quercivorus. Microbial Ecology 62: 106-120.
Farrell BD, Sequeira AS, O'Meara BC, Normark BB, Chung JH & Jordal BH (2001) The evolution of agriculture in
beetles (Curculionidae : Scolytinae and Platypodinae). Evolution 55: 2011-2027.
Francke-Grosmann H (1956) Hautdrüsen als Träger der Pilzsymbiose bei Ambrosiakäfern. Z. Morph. u. Ökol.
Tiere 45: 275-308.
Francke-Grosmann H (1967) Ectosymbiosis in wood-inhabiting beetles. Symbiosis (Henry SM, ed), pp. 141-205.
Academic Press, New York.
Francke-Grosmann H (1975) Zur epizoischen und endozoischen Übertragung der symbiotischen Pilze des
Ambrosiakäfers Xyleborus saxeseni (Coleoptera: Scolitidae). Entomologica Germanica 1: 279-292.
Frank SA (1996) Host-symbiont conflict over the mixing of symbiotic lineages. Proceedings of the Royal Society
of London Series B-Biological Sciences 263: 339-344.
French JRJ & Roeper RA (1972) Interactions of ambrosia beetle, Xyleborus dispar (Coleoptera, Scolytidae), with
its symbiotic fungus Ambrosiella hartigii (Fungi imperfecti). Canadian Entomologist 104: 1635-1641.
Glass NL & Donaldson GC (1995) Development of primer sets designed for use with the PCR to amplify
conserved genes from filamentous ascomycetes. Appl. Environ. Microbiol. 61: 1323-1330.
Grafenhan T, Schroers HJ, Nirenberg HI & Seifert KA (2011) An overview of the taxonomy, phylogeny, and
typification of nectriaceous fungi in Cosmospora, Acremonium, Fusarium, Stilbella, and Volutella. Studies in
Mycology 68: 79-113.
111
Chapter 7
Grubbs KJ, Biedermann PHW, Suen G, Adams SM, Moeller JA, Klassen JL, Goodwin LA, Woyke T, Munk AC,
Bruce D, Detter C, Tapia R, Han CS & Currie CR (2011) The Complete Genome Sequence of Streptomyces cf.
griseus (XyelbKG-1), an Ambrosia Beetle-Associated Actinomycete. J. Bacteriol. 193: 2890-2891.
Haanstad JO & Norris DM (1985) Microbial symbiotes of the ambrosia beetle Xyletorinus politus. Microbial
Ecology 11: 267-276.
Harrington TC, Aghayeva DN & Fraedrich SW (2010) New combinations in Raffaelea, Ambrosiella, and
Hyalorhinocladiella, and four new species from the redbay ambrosia beetle, Xyleborus glabratus. Mycotaxon
111: 337-361.
Harrington TC & Fraedrich SW (2010) Quantification of Propagules of the Laurel Wilt Fungus and Other
Mycangial Fungi from the Redbay Ambrosia Beetle, Xyleborus glabratus. Phytopathology 100: 1118-1123.
Hölldobler B & Wilson EO (1990) The ants. The Belknap Press of Harvard University Press, Cambridge, MA.
Hulcr J, Mann R & Stelinski LL (2011) The Scent of a Partner: Ambrosia Beetles Are Attracted to Volatiles from
Their Fungal Symbionts. Journal of Chemical Ecology 37: 1374-1377.
Kajimura H & Hijii N (1992) Dymamics of the fungal symbionts in the gallery system and the mycangia of the
ambrosia beetle, Xylosandrus mutilatus (Blandford) (Coleoptera, Scolytidae). Ecological Research 7: 107-117.
Kim JJ, Ra JB, Son DS & Kim GH (2001) Fungi Colonizing Sapwood of Japanese Red Pine Logs in Storage.
Mycobiology 29: 205-209.
Kingsolver JG & Norris DM (1977) The interaction of Xyleborus ferrugineus (Fabr.) (Coleoptera: Scolytidae)
behavior and initial reproduction in relation to its symbiotic fungi. Annals of the Entomological Society of
America 70: 1-4.
Kirkendall LR, Kent DS & Raffa KF (1997) Interactions among males, females and offspring in bark and ambrosia
beetles: the significance of living in tunnels for the evolution of social behavior. The Evolution of Social Behavior
in Insects and Arachnids (Choe JC & Crespi BJ, eds), pp. 181-215. Cambridge University Press.
Kok LT & Norris DM (1972a) Comparative phospholipid compositions of adult female Xyleborus ferrugineus and
its mutualistic fungal ectosymbionts. Comparative Biochemistry and Physiology 42: 245-254.
Kok LT & Norris DM (1972b) Phospholipid composition of fungi mutualistic with Xyleborus ferrugineus.
Phytochemistry 11: 1449-1453.
Kok LT & Norris DM (1972c) Symbiotic interrelationships between microbes and ambrosia beetles 6. Aminoacid
composition of ectosymbiotic fungi of Xyleborus ferrugineus (Coleoptera, Scolytidae). Annals of the
Entomological Society of America 65: 598-602.
Kok LT & Norris DM (1973) Comparative sterol compositions of adult female Xyleborus ferrugineus and its
mutualistic fungal ectosymbionts. Comparative Biochemistry and Physiology 44: 499-505.
Leach JG, Hodson AC, Chilton SJP & Christensen CM (1940) Observations on two ambrosia beetles and their
associated fungi. Phytopathology 30: 227-236.
Martin MM (1987) Invertebrate-Microbial Interactions . Cornell University Press, Ithaca.
Moore GE (1973) Pathogenicity of three entomogenous fungi to the Southern Pine Beetle at various
temperatures and humidities. Environmental Entomology 2: 54-57.
Mueller UG (2002) Ant versus fungus versus mutualism: Ant-cultivar conflict and the deconstruction of the
attine ant-fungus symbiosis. Am. Nat. 160: S67-S98.
112
Chapter 7
Mueller UG, Gerardo NM, Aanen DK, Six DL & Schultz TR (2005) The evolution of agriculture in insects. Annual
Review of Ecology Evolution and Systematics 36: 563-595.
Mueller UG, Schultz TR, Currie CR, Adams RMM & Malloch D (2001) The origin of the attine ant-fungus
mutualism. Q. Rev. Biol. 76: 169-197.
Norris DM (1975) Chemical interdependence among Xyleborus spp. ambrosia beetles and their symbiotic
microbes. Material und Organismen 3: 479-788.
Norris DM (1979) The mutualistic fungi of Xyleborini beetles. Nutrition, Mutualism, and Commensalism (Batra
LR, ed), pp. 53-63. Allanheld, Osmun & Company, Montclair.
O'Donnell K & Cigelnik E (1997) Two Divergent Intragenomic rDNA ITS2 Types within a monophyletic lineage of
the fungus Fusarium are nonorthologous. Molecular Phylogenetics and Evolution 7: 103-116.
Peer K & Taborsky M (2005) Outbreeding depression, but no inbreeding depression in haplodiploid ambrosia
beetles with regular sibling mating. Evolution 59: 317-323.
Peer K & Taborsky M (2007) Delayed dispersal as a potential route to cooperative breeding in ambrosia beetles.
Behav. Ecol. Sociobiol. 61: 729-739.
Peterson SW, Bayer EM & Wicklow DT (2004) Penicillium thiersii, Penicillium angulare and Penicillium
decaturense, new species isolated from wood-decay fungi in North America and their phylogenetic placement
from multilocus DNA sequence analysis. Mycologia 96: 1280-1293.
Pinto-Tomas AA, Anderson MA, Suen G, Stevenson DM, Chu FST, Cleland WW, Weimer PJ & Currie CR (2009)
Symbiotic Nitrogen Fixation in the Fungus Gardens of Leaf-Cutter Ants. Science 326: 1120-1123.
Qi HY, Wang JG, Endoh R, Takeuchi Y, Tarno H & Futai K (2011) Pathogenicity of microorganisms isolated from
the oak platypodid, Platypus quercivorus (Murayama) (Coleoptera: Platypodidae). Applied Entomology and
Zoology 46: 201-210.
R Development Core Team (2008) R: A language and environment for statistical computing. Vienna, Austria, R
Foundation for Statistical Computing.
Rodrigues A, Bacci M, Mueller UG, Ortiz A & Pagnocca FC (2008) Microfungal "weeds" in the leafcutter ant
symbiosis. Microbial Ecology 56: 604-614.
Rodrigues A, Mueller UG, Ishak HD, Bacci Jr Mc & Pagnocca FC (2011) Ecology of microfungal communities in
gardens of fungus-growing ants (Hymenoptera: Formicidae): a year-long survey of three species of attine ants
in Central Texas. FEMS Microbiology Ecology 78: 244-255.
Roeper RA (1995) Patterns of mycetophagy in Michigan ambrosia beetles. Michigan Academian 27: 153-161.
Roeper RA & French JRJ (1981) Ambrosia fungi of the Western United States and Canada - beetle assocaitions
(Coleoptera: Scolytidae), Tree Hosts, and Distribution. Northwest Science 55: 305-309.
Roeper RA, Hazen CR, Helsel DK & Bunce MA (1980) Studies on Michigan ambrosia fungi. The Michigan Botanist
19: 69-73.
Rohlfs M & Churchill ACL (2011) Fungal secondary metabolites as modulators of interactions with insects and
other arthropods. Fungal Genetics and Biology 48: 23-34.
Saunders JL & Knoke JK (1967) Diets for rearing the ambrosia beetle Xyleborus ferrugineus (Fabricius) in vitro.
Science 15: 463.
Schneider I (1991) Some ecological aspects of the ambrosia-symbiosis. Anzeiger für Schadlingskunde
Pflanzenschutz Umweltschutz 64: 41-45.
113
Chapter 7
Schneider IA & Rudinsky JA (1969) Mycetangial glands and their seasonal changes in Gnathotrichus retusus and
G. sulcatus. Annals of the Entomological Society of America 62: 39-43.
Scott JJ, Oh DC, Yuceer MC, Klepzig KD, Clardy J & Currie CR (2008) Bacterial protection of beetle-fungus
mutualism. Science 322: 63.
Silva A, Rodrigues A, Bacci M, Pagnocca FC & Bueno OC (2006) Susceptibility of the ant-cultivated fungus
Leucoagaricus gongylophorus (Agaricales : Basidiomycota) towards microfungi. Mycopathologia 162: 115-119.
Six DL (2003) Bark beetle-fungus symbioses. Insect Symbiosis (Bourtzis K & Miller TA, eds), pp. 97-114. CRC
Press, Boca Raton.
Six DL & Bentz BJ (2007) Temperature determines symbiont abundance in a multipartite bark beetle-fungus
ectosymbiosis. Microbial Ecology 54: 112-118.
Six D, Doug Stone W, de Beer Z & Woolfolk S (2009) Ambrosiella beaveri sp. nov., Associated with an exotic
ambrosia beetle, Xylosandrus mutilatus (Coleoptera: Curculionidae, Scolytinae), in Mississippi, USA. Antonie
van Leeuwenhoek 96: 17-29.
Tamura K, Dudley J, Nei M & Kumar S (2007) MEGA4: Molecular Evolutionary Genetics Analysis (MEGA)
Software Version 4.0. Molecular Biology and Evolution 24: 1596-1599.
Tang AMC, Jeewon R & Hyde KD (2005) Succession of microfungal communities on decaying leaves of
Castanopsis fissa. Canadian Journal of Microbiology 51: 967-974.
Vilgalys R & Hester M (1990) Rapid genetic identification and mapping of enzymatically amplified ribosomal
DNA from several Cryptococcus species. J. Bacteriol. 172: 4238-4246.
Wheeler Q & Blackwell M (1984) Fungus-Insect Relationships . Columbia University Press , New York.
Wilding N, Collins NM, Hammond PM & Webber JF (1989) Insect-Fungus Interactions. Academic Press, London.
Zhang YC, Li C, Swenson DC, Gloer JB, Wicklow DT & DOWD PF (2003) Novel antiinsectan oxalicine alkaloids
from two undescribed fungicolous Penicillium spp. Organic Letters 5: 773-776.
Zimmermann G (1973) Die Pilzflora einiger im Holz lebender Borkenkäfer. Material und Organismen 8: 121-131.
114
Chapter 7
Graphs
proportion of samples
transmitted fungi
0.00
0.20
0.40
0.60
0.80
1.00
Raffaelea sulphurea
mycangium (N = 13)
Arthrinium spp.??
Raffaelea-like sp. B
Fusicolla acetilerea-like sp.
gut (N = 13)
Penicillium decaturense
fungi within gallery-classes
Paecilomyces variotii
foundress & eggs (N = 8)
foundress & larvae (N = 10)
only adults (N = 9)
other samples
larval bodies (N = 9)
fungus growing on adults (N = 4)
gallery dump (N = 8)
Fig. 1. Percentage of fungal species isolated relative to total number of samples taken. Isolations
from the spore-carrying organs (mycetangium, gut), of different gallery-classes, of whole larval
bodies, of the fungal layer growing on adult beetles, and of the gallery dump are displayed. N =
sample size of dissected adult beetles, galleries, or larvae.
115
Chapter 7
Fig.2. Relationship of different fungi with numbers of Xyleborinus saxesenii larvae during the
immature stage of colonies. During the stage with foundress, eggs, larvae and pupae the frequency
of Raffaelea sulphurea correlated positively with the numbers of larvae present (GEE: p = 0.035; LR:
r2 = 0.11). In contrast, the frequency of Paecilomyces variotii tended to correlate negatively with the
numbers of larvae present (GEE: p = 0.063; LR: r2 = 0.22). All other fungal species had no significant
relationship with offspring numbers. N = 10 galleries; for statistical details see Table 3.
Fig.3. Morphology of the ambrosial layer produced by Raffaelea sulfurea within laboratory galleries
of Xyleborinus saxesenii. Conidiophores with conidia are visible. Note the area on the left side where
several conidia are missing on top of the conidiophores – adult beetles have probably cropped them
off. The picture was taken using Scanning Electron Microscopy (SEM) with 400× magnification.
116
Chapter 7
Table 1. Characteristics of the six fungal species isolated from laboratory galleries of Xyleborinus saxesenii.
Order
Species
Ophiostomatales
Raffaelea
sulphurea
Raffaelealike sp. B
Arthrinium
sp.?
Fusicolla
acetilerealike sp.
Penicillium
decaturense
Paecilomyces
variotii
Sordariales
Hypocreales
Eurotiales
Medium
Location
CSMA > MA
BC > MT
CSMA > MA
BC < MT*
only on CSMA
ns
1
1
Fungal species dynamic
Influence on progeny
numbers
Dominance ( > 50% of
samples)
Proposed
association
positive (eggs*, larvae)
in gut,
during foundress/eggs
mutualist
not present during foundress/eggs
ns
-
?
ns.
ns
-
?
ns
in mycetangium
mutualist
foundress/eggs < adult progeny
2
foundress/larvae < adult progeny
ns
-
parasite (?)
not present during foundress/eggs
1,2
foundress/larvae > adult progeny
negative (larvae*)
on adult bodies,
in gallery dump
parasite
foundress/eggs > adult progeny
1
2
CSMA < MA
BC > MT
ns
ns
ns
BC < MT
1
foundress/eggs < foundress/larvae
1
foundress/larvae < adult progeny
2
foundress/larvae > adult progeny
2
2
Significant differences (p < 0.05) and trends (p < 0.01) indicated with * are shown, except were marked with ns (not significant).
Abbreviations: CSMA – cycloheximide-streptomycin-malt agar, MA – malt agar; BC – brood chamber, MT – main tunnel; “…>…” /”… <…” denote the direction of the significant difference in the
isolation frequency between the two media/locations/gallery classes.
1
2
Significant difference in the frequency of the fungus (p < 0.05)
Significant difference in the presence of the fungus (p < 0.05)
117
Chapter 7
Supplementary Information
Table S1. Factors influencing the abundance of fungi isolated from X. saxesenii. Separate (1) GEE models with
an exchangeable correlation structure of the response variable within a cluster (gallery-identity), for examining
differences between the frequency, and (2) GLM models for examining differences between the presence
(yes/no) of the isolated fungal species, influenced by culture medium (CSMA – cycloheximide-streptomycinmalt agar, MA – malt extract agar), location within the gallery (brood chamber or main tunnel), and stage of
gallery development. Model coefficients are reported as coeff.  se (standard error of the estimate), with the
group in brackets in the first row of the model as the reference category (coefficient set to zero). The influences
of main factors on the fungal frequencies are displayed as contrasts between classes to give a better
understanding. A positive contrast denotes that the mean of the second class is higher than the mean of the
first class; a negative contrast denotes the reverse. Significant relationships are highlighted in bold face.
Raffaelea sulphurea
coeff.  se
t
p
1.65  0.85
1.95
0.05
-2.6  0.37
-2.02  0.36
-1.28  1.05
-7.03
-5.62
-1.22
<0.001
<0.001
0.22
-2.41  1.09
-1.13  0.82
-2.2
-1.37
0.028
0.17
1.34  0.69
1.94
0.058
-1  0.64
-0.16  0.78
-1.57
-0.2
0.12
0.84
-0.63  0.79
-0.48  0.76
-0.8
-0.63
0.43
0.53
Intercept of frequency (CSMA, brood chamber,
foundress with eggs)
Contrast CSMA vs. MA
Contrast brood chamber vs. main tunnel
Contrast foundress with eggs vs. foundress with
larvae
Contrast foundress with eggs vs. only adult progeny
Contrast foundress with larvae vs. only adult
progeny
-11.7  23.3
-0.5
0.61
Intercept of frequency (CSMA, brood chamber,
foundress with eggs)
Contrast CSMA vs. MA
Contrast brood chamber vs. main tunnel
Contrast foundress with eggs vs. foundress with
larvae
Contrast foundress with eggs vs. only adult progeny
Contrast foundress with larvae vs. only adult
progeny
Intercept of presence (brood chamber, foundress
with eggs)
Contrast brood chamber vs. main tunnel
Contrast foundress with eggs vs. foundress with
larvae
Contrast foundress with eggs vs. only adult progeny
Contrast foundress with larvae vs. only adult
progeny
Artrinium sp.
Raffaelea-like sp. B
Only present on CSMA
-1.16
-5.66  4.88
-0.12
-3.8  32.9
0.25
0.91
-1.3  32.8
-2.45  1.69
-0.04
-1.45
0.97
0.15
Intercept of presence (brood chamber, foundress
with eggs)
Contrast brood chamber vs. main tunnel
Contrast foundress with eggs vs. foundress with
larvae
Contrast foundress with eggs vs. only adult progeny
Contrast foundress with larvae vs. only adult
progeny
-2.45  1.18
-2.08
0.04
-0.57  1.31
-0.16  1.52
-0.44
-0.1
0.67
0.92
-0.03  1.52
0.12  1.51
-0.02
0.08
0.98
0.94
Intercept of frequency (CSMA, brood chamber,
foundress with eggs)
Contrast CSMA vs. MA
Contrast brood chamber vs. main tunnel
Contrast foundress with eggs vs. foundress with
larvae
Contrast foundress with eggs vs. only adult progeny
Contrast foundress with larvae vs. only adult
progeny
-31.9  999
-0.00
1
Intercept of presence (brood chamber, foundress
with eggs)
Contrast brood chamber vs. main tunnel
Contrast foundress with eggs vs. foundress with
larvae
118
-3.02
0.003
-4.83  1.6
1.7
0.09
8.35  4.91
Only present during foundress with larvae
Only present during only adult progeny
-1.3
0.19
-2.26  1.74
0.45  0.69
0.65
0.52
-1.86
0.07
-1.47  0.79
Only present during foundress with larvae
Chapter 7
Fusicolla acetilerealike sp.
Penicillium
decaturense
Paecilomyces
variotii
Contrast foundress with eggs vs. only adult progeny
Contrast foundress with larvae vs. only adult
progeny
Only present during only adult progeny
1.13
0.26
1.37  1.21
Intercept of frequency (CSMA, brood chamber,
foundress with eggs)
Contrast CSMA vs. MA
Contrast brood chamber vs. main tunnel
Contrast foundress with eggs vs. foundress with
larvae
Contrast foundress with eggs vs. only adult progeny
Contrast foundress with larvae vs. only adult
progeny
-8.04  2.63
-3.05
<0.002
9.32  1.88
-1.67  0.51
-0.19  2.43
4.96
-3.24
-0.08
<0.001
0.001
0.94
4.05  2.52
4.17  1.33
1.61
3.13
0.11
0.002
Intercept of presence (brood chamber, foundress
with eggs)
Contrast brood chamber vs. main tunnel
Contrast foundress with eggs vs. foundress with
larvae
Contrast foundress with eggs vs. only adult progeny
Contrast foundress with larvae vs. only adult
progeny
0.39  0.67
0.58
0.56
-1.34  0.75
2.46  1.03
-1.8
2.38
0.08
0.02
0.19  0.82
-2.27  1.03
0.23
-2.2
0.82
0.03
Intercept of frequency (CSMA, brood chamber,
foundress with eggs)
Contrast CSMA vs. MA
Contrast brood chamber vs. main tunnel
Contrast foundress with eggs vs. foundress with
larvae
Contrast foundress with eggs vs. only adult progeny
Contrast foundress with larvae vs. only adult
progeny
-10.6  12.1
0.88
0.38
-11.9  8.66
-0.06  0.66
0.71  13.8
-1.38
-0.09
0.05
0.17
0.93
0.96
2.76  13.2
2.05  7.05
0.21
0.29
0.83
0.77
Intercept of presence (brood chamber, foundress
with eggs)
Contrast brood chamber vs. main tunnel
Contrast foundress with eggs vs. foundress with
larvae
Contrast foundress with eggs vs. only adult progeny
Contrast foundress with larvae vs. only adult
progeny
-2.37  1.17
-2.02
0.05
-0.83  0.89
0.59  1.4
-0.93
0.42
0.36
0.68
2.99  1.27
2.4  1.01
2.35
2.37
0.02
0.02
Intercept of frequency (CSMA, brood chamber,
foundress with eggs)
Contrast CSMA vs. MA
Contrast brood chamber vs. main tunnel
Contrast foundress with eggs vs. foundress with
larvae
Contrast foundress with eggs vs. only adult progeny
Contrast foundress with larvae vs. only adult
progeny
-21.8  999
-0.01
1
Intercept of presence (brood chamber, foundress
with eggs)
Contrast brood chamber vs. main tunnel
Contrast foundress with eggs vs. foundress with
larvae
Contrast foundress with eggs vs. only adult progeny
Contrast foundress with larvae vs. only adult
progeny
119
-0.23
0.82
-0.08  0.37
6.22
<0.001
2.39  0.39
Only present during foundress with larvae
Only present during only adult progeny
-3.25
0.001
-2.02  0.62
-20.87  999
-0.01
1
2.56
0.01
2.01  0.78
Only present during foundress with larvae
Only present during only adult progeny
-3.3
0.002
-2.53  0.77
Chapter 7
Table S2. The relationship between the most common fungal species and the number of X. saxesenii
offspring. Separate GEE models were performed with an exchangeable correlation structure of the
response variable within a cluster (gallery identity) to examine the potential influence of the
frequency of fungal species on offspring numbers, controlling for the influence of medium (CSMA –
cycloheximide-streptomycin-malt agar; MA – malt extract agar). The potential effects on egg and
larval numbers during particular periods of gallery development are shown (for graphical illustration
see fig.2). Model coefficients are reported as coeff.  se (standard error of the estimate), with the
group in brackets in the first row of the model as the reference category (coefficient set to zero). A
positive coefficient denotes a positive relationship; a negative coefficient denotes a negative
relationship.
coeff.  se
t
p
-0.01
-3.37
1.72
0.92
<0.001
0.09
Foundress with eggs, brood chamber
Raffaelea sulphurea
Intercept of frequency (CSMA)
Contrast CSMA vs. MA
Number of eggs
-0.14  1.45
-3.49  1.04
0.9  0.53
Fusicolla acetilerea-like sp.
Intercept of frequency (CSMA)
Contrast CSMA vs. MA
Number of eggs
-22.9  999
0
1
Only present on MA
-2.91  4.24 -0.69
0.49
Paecilomyces variotii
Not present
Foundress with larvae, brood chamber
Raffaelea sulphurea
Intercept of frequency (CSMA)
Contrast CSMA vs. MA
Number of larvae
-0.15  0.51
-2.49  0.61
0.05  0.02
-0.3
-4.08
2.11
0.77
<0.001
0.035
Fusicolla acetilerea-like sp.
Intercept of frequency (CSMA)
Contrast CSMA vs. MA
Number of larvae
-7.88  4.82
13.6  5.04
0.02  0.21
-1.63
2.69
0.1
0.1
0.007
0.92
Paecilomyces variotii
Intercept of frequency (CSMA)
Contrast CSMA vs. MA
Number of larvae
-0.86  0.49
-0.46  0.65
-0.06  0.03
-1.76
-0.71
-1.86
0.08
0.48
0.063
120
4
3
1
2
Species richness
5
6
Chapter 7
0
20
40
60
80
100
Number of samples
Fig. S1. Individual-based rarefaction curves of fungal communities associated with galleries
of X. saxesenii. Bars indicate 95% confidence intervals.
121
Chapter 7
122
Chapter 8
Patterns of functional enzyme activity show that larvae are the key to
successful fungus farming by ambrosia beetles
Henrik H. de Fine Licht1* and Peter H. W. Biedermann2
1
Section of Microbial Ecology, Department of Biology, Lund University, Solvegatan 37, SE-223 62 Lund, Sweden
Department of Behavioural Ecology, Institute of Ecology & Evolution, University of Berne, Baltzerstrasse 6, CH3012 Bern, Switzerland, Email: [email protected].
2
* Corresponding Author:
Henrik H. de Fine Licht, Section of Microbial Ecology, Department of Biology, Lund University, Solvegatan 37,
SE-223 62 Lund, Sweden, Email: [email protected], Tel: (+46) 46 222 3763.
Running title: Ambrosia symbiosis enzyme activity
Submitted to Frontiers in Zoology, in review
Abstract
Introduction: Fungus-farming by insects depends on the collaboration and division of labour of
multiple individuals. In wood-dwelling fungus-farming weevils, the so-called ambrosia beetles, larval
and adult stages have been recently shown to specialize in different social tasks. Wood is converted
into fungal food by the combined action of digging larvae (which create more space for the fungi to
grow), fungus tending adults and enzymatic hydrolysis by the cultivars. However, whether adults,
larvae and the mutualistic fungi are producing different enzymes in this process is unknown. Here we
characterize the enzyme profiles of all partners in the ambrosia beetle microbial symbiosis for the
first time. We measure 13 distinct plant cell-wall degrading enzymes in Xyleborinus saxesenii adults,
larvae and the fungus garden microbial consortium at different ages and locations within the nest.
Finally, we compare the enzyme profiles with other fungus-farming insects.
Results: We discovered that wood and fungus feeding larvae are important for successful wood
degradation and fungiculture: Cellulase activity was identified in whole-body extracts of both larval
and adult X. saxesenii, whereas xylanase activity was exclusively detected in larvae. These enzymes
apparently (pre-) digest plant cell-wall structures within larval guts, because larvae feed on fungusinfested wood, unlike adults that only feed on fungus. Furthermore, larval feces have been found to
be applied on the fungus gardens after gut passage, likely for further utilization by the mutualistic
fungi.
Conclusion: The enzymatic profile of the ambrosia symbiosis resembles other fungus-growing insect
systems. However, remarkably in ambrosia beetle societies adults and larvae do not compete for the
same food within their nests - in contrast, larvae increase colony fitness by facilitating wood
degradation and fungus cultivation. The differential enzyme profiles add an additional layer of
complexity to the division of behavioural tasks between life-stages already reported within the
primitively eusocial X. saxesenii. In general, enzymatic division of labour may be an unrecognized
more common feature of holometabolous insects, which dramatically reorganize their body during
metamorphosis. Larval physiology and enzymatic profile may provide novel evolutionary
opportunities to social insect societies capable of utilizing this resource.
Keywords
Symbiosis, Insoluble chromogenic enzyme substrates, xylomycetophagy, Xyloborinus saxesenii, insect
fungus farming, social evolution, enzymatic division of labor
123
Chapter 8
Introduction
Insects are the most abundant and diverse animal class on earth [1]. A key factor for their
enormous success are adaptations to novel environments and food sources by the help of symbiotic
microorganisms [2]. Insect hosts maintain prokaryotic, fungal, and bacterial associates in a variety of
ways, which help them in nutrient acquisition and recycling, environmental detoxification, and
defense against antagonists. By means of microbial symbionts insects are able to produce food out of
plant material low on insect-accessible nutrients, but rich in structural polysaccharides (cross-linking
glycans, cellulose, and lignin) and toxic metabolites [3]. In most instances this occurs internally inside
the insect host with the aid of an abundant microbial gut flora [4-6], however there are several
notable exceptions of insect lineages that cultivate microbes externally on plant material [7, 8].
These ectosymbionts are either kept in “gardens” and consumed directly by their insect hosts (e.g.
certain lineages of fungus-growing ants, termites, and ambrosia beetles; [9]), or contribute indirectly
by increasing the nutrient content of decomposing material (e.g. bark beetles; [10]), by degradation
of toxic plant compounds (e.g. terpenes by bark beetle associated fungi [11, 12]), or by provisioning
of extracellular enzymes which may facilitate insect ingestion or wood burrowing (e.g. wood wasps;
[5, 13]).
External symbionts of insects are typically filamentous fungi, yeasts and bacteria that may be
transported in mycetangia (also termed mycangia [14]), which are specialized organs for fungal spore
transmission that ensure successful re-establishment of the nutritional symbiosis after dispersal.
Mycetangia have evolved independently several times and are known from many different fungus
associated insects such as wood and phloem feeding beetles, gall midges and wood wasps [15-17].
The active care and maintenance of the fungal crops by the insect hosts after dispersal is, however,
rare. Only three insect lineages, notably the fungus-growing ants, termites and ambrosia beetles, are
truly farmers of their fungal crops. Within task sharing societies they not only propagate, but also
actively cultivate and sustainably harvest their crops in microbial gardens (i.e., advanced fungiculture
[9]).
Ambrosia beetle is an ecological term used for all weevils that farm fungi within tunnel
systems (galleries) in the wood of trees. Ambrosia farming is only found in Scolytinae and
Platypodinae and evolved repeatedly at least nine times from the phloem feeding habit without any
known reversal to non-farming [18, 19]. Ambrosia beetles usually dwell in recently dead wood that
seems almost free of microbes, where they initiate nest building by inoculating tunnel walls with
mutualistic fungi. The beetle-fungus relationship is often species (or genus) specific, with highly
selective transmission of the primary symbionts in mycetangia of dispersing beetles [20, 21]. These
so-called ambrosia fungi (usually species of the ascomycete genera Ambrosiella and Raffaelea) are
forming layers of conidiophores on the tunnel walls that produce nutrient rich conidiospores for
larval and adult beetle nutrition. Secondary symbionts, such as other filamentous fungi (e.g.
Fusarium sp., Graphium sp., Ophiostoma sp., Paecilomyces sp., Penicillium sp. [22]), yeasts (e.g.
Candida sp. [23]) and bacteria [24, 25], are also present within galleries and are often passively
vectored in small amounts attached to the surface of dispersing females [15, 20]. However, the
primary mutualistic ambrosia fungus is known for only a minority of the 3000 species worldwide [2628], and there has only been a single attempt to characterize the entire microbiome of an ambrosia
gallery [29]. Studies on the dynamics of filamentous fungi in xyleborine ambrosia beetle galleries
suggest that propagates of mutualistic ambrosia fungi (Ambrosiella and Raffaelea) are passively
spread on tunnel walls from the mycetangia or via beetle feces during the excavation by the gallery
founding female. This ensures that the mutualistic fungi dominate the gallery microbial flora initially
while eggs are laid and larvae develop [22, 30]. Later, when the first offspring mature, other saprobic
fungi (secondary symbionts like Penicillium sp. and Paecilomyces sp.) start to appear and increase in
frequency over time. These opportunistic fungi dominate the microbial gallery flora at the time when
the gallery is abandoned and all individuals disperse to found new galleries [22, 30]. In the ambrosia
beetle Xyleborinus saxesenii (Fruit-tree pinhole borer) their abundance negatively affects the number
of larvae and they are more commonly encountered within the entrance tunnel than within the
brood chambers and also dominate in the gallery dumps [30].
Larvae of X. saxesenii do not only feed on ambrosia fungi, like the adults and larvae of many
other ambrosia beetles, but feed xylomycetophagously on fungus infested wood [31]. In this way
they (a) create more space for the developing fungus to form fruiting bodies on the gallery walls, (b)
lower competition between group members by enlargement of the nest space, (c), reduce the
124
Chapter 8
growth of not identified molds, possibly by gregariously feeding on them [32] and (d) probably most
importantly benefit the growing ambrosia fungus, because the finely fragmented woody sawdust
within the larval feces is smeared on the gallery walls after defecation [32, 33]. However, nothing is
known about the mechanism or the enzymatic machinery whereby the ambrosia beetles in
collaboration with the consortium of symbiotic fungi degrade the surrounding wood. In larvae of a
related phloem feeding bark beetle (Phloeosinus bicolor) α-amylase-, invertase-, maltase-, lactase-,
and protease-activities were detected together with some hydrolytic activity on a substrate of crosslinking glycans (hemi-cellulose) but not on cellulose [34]. Similarly, in adults of the phloem feeding
Ips cembrae consistent activity against cross-linking glycans together with pectinase-, α-glucosidase-,
β-glucosidase-, α-galactosidase-, β-galactosidase-, trehalase-, tryptase-, peptitase-, and lipaseactivities were detected in the intestinal lumen [35]. Many of these enzymes are probably of
microbial origin, because bark beetles (i) carry yeasts and bacteria in their intestines [36-38] and (ii)
feed on phloem that is often infested by Ophiostomatoid fungi [16]. The bark beetle associated fungi
(e.g. the genera Ceratosytiopsis, Entomocorticium and Ophiostoma [39, 40]), in addition to associated
yeasts [12, 36, 39], and bacteria [41, 42] are capable of producing a variety of enzymes catalyzing (a)
protein/peptide degradation (endo-, exoproteases and peptidases), (b) polysaccharide/starch/sugar
degradation (glycoside-hydrolytic enzymes) and (c) fat/fatty acid degradation (lipases) [39, 40, 4349].
Here the activities of the major groups of plant cell-wall degrading enzymes: cellulases, endoxylanases, pectinases, and also endo-proteases and α-amylases in the ambrosia beetle system are
tested for the first time. We measured the enzyme activity in larval and adult female X. saxesenii and
of the whole microbiome from three distinct locations within laboratory galleries (gallery dump,
entrance tunnel, and brood chamber) and find enzymatic division of labor between adult beetles and
larvae. The latter possess a distinct cell-wall degrading endo-β-1,4-xylanase activity not found in
adults. This was done by a high-throughput screening method of 13 different enzymes in
combination with an artificial rearing technique of beetle galleries. Laboratory rearing enabled us to
precisely track the development of broods, and take enzyme measurements at particular time points:
(a) with abundant immature brood, (b) with both immature brood and mature offspring and (c) at
the end of gallery life with almost exclusively adult offspring.
Results
Of the 13 enzyme substrates screened in this study, six specific enzyme activities (endo-β-1,4glucanase, endo-β-1,3(4)-glucanase, endo-β-1,4-xylanase (xylan and arabinoxylan), endo-β-1,4mannanase, and endo-protease (casein)) were consistently detected in all samples (Fig. 1A, [see
Additional file 1]) and varied significantly between the three different gallery compartments:
entrance, brood chamber and gallery dump (log-likelihood ANOVA comparison of final mixed models
with reduced null models: likelihood-ratio3,5 = 14.1 – 50.4, p = < 0.0001 – 0.0009). For these six
enzyme activities the predictor variables: ‘gallery class’ and ‘total number of animals’ where not
significant (log-likelihood ANOVA comparison of final mixed models with reduced null models:
likelihood-ratio5,11 = 2.0 – 8.7, p = 0.1884 – 0.9169) when included as interaction terms with gallery
compartment (entrance, brood chamber or gallery dump). However, endo-β-1,4-xylanase activity
against the cross-linking glycans (xylan and arabinoxylan) and endo-β-1,4-mannanase against pectin
varied not only with gallery compartment but also with the age of the gallery (log-likelihood ANOVA
comparison of final mixed models with reduced null models: likelihood-ratio5,11 = 12.7 – 16.9, p =
0.0095 – 0.0472, Fig. 1A, [see Additional file 1]). The final reduced and most parsimonious models for
endo-β-1,4-xylanase and endo-β-1,4-mannanase enzyme activity therefore included both gallery
compartment and the interaction term of gallery compartment and gallery age as fixed factors,
whereas the final model of endo-β-1,4-glucanase, endo-β-1,3(4)-glucanase and endo-protease
(casein) only consisted of gallery compartment as the fixed factor. Enzyme activity against the
substrates xyloglucan, galactan, rhamnogalacturonan, debranched arabinan and amylose where
sporadic across the gallery compartments with many zero activities detected (Fig. 1A, [see Additional
file 1]), and these enzyme activities were therefore more appropriately analyzed separately for each
age cohort with a non-parametric Kruskal-Wallis test and the significance evaluated using a chisquare distribution [see Additional file 1]. In all samples no enzyme activity against the substrates
dextran and collagen were detected (Fig. 1A, [see Additional file 1]).
125
Chapter 8
The plant cell-wall degrading cellulases, endo-xylanases and pectinases had a consistently
higher activity in the gallery dump material compared to the entrance tunnel and the brood chamber
(Fig. 1A), whereas endo-protease activity against casein showed the opposite trend with the highest
enzyme activity in the entrance tunnel (Fig. 1A, [see Additional file 1]). Cellulolytic activity was similar
between the entrance tunnel and brood chamber across gallery ages, whereas endo-β-1,4-xylanase
(xylan and arabinoxylan) and endo-β-1,4-mannanase activity changed across age cohorts most
notably with an increase in enzyme activity in the gallery dump with age (Fig. 1A, [see Additional file
1]). For these three enzymes we also noted a consistent but non-significant trend of high activity in
the entrance tunnel compared to the brood chamber at age 45, similar activity at age 62 and the
opposite pattern at age 87 with the highest activity in the brood chamber compared to the entrance
tunnel (Fig. 1A, [see Additional file 1]). The increased enzyme activity of plant cell-wall degrading
enzymes in the gallery dump was also evident from the partial least square regression analysis
because these specific enzymes correlated (i.e. clustered) more closely to the gallery dump than
either the entrance tunnel or the brood chamber [see Additional file 1].
Screening for specific endo-β-1,4-glucanase and endo-protease (casein) activity did not reveal
any enzyme activity of adult and larval fungus-growing beetles (Fig. 2). In contrast, endo-β-1,3(4)glucanase (beta-glucan) activity were consistently present between 1st, 2nd/3rd instar larvae and
adults (Fig. 2). Endo-β-1,4-xylanase activity was detected in both 1st and even more so in 2nd/3rd instar
larvae, but not in adult beetles, in which not a single sample of the 14 x 3 beetles measured showed
any enzyme activity against xylan (Fig. 2).
Discussion
Glycoside hydrolases are very important enzymes in the fungus-growing ambrosia beetle symbiosis.
These enzymes aid in the degradation of the carbohydrate rich cell walls and thereby sustain the
conversion of wood that surrounds beetle galleries into fungal biomass. The usage of the
comprehensive set of plant cell-wall degrading enzymes generally produced by saprobic fungi [50-52]
in combination with mechanical preprocessing of the wood by beetle activities have allowed
ambrosia beetles to utilize their present niche deep inside recently dead wood inaccessible to other
non-symbiotic fungi and insects that only penetrate the outer bark layers [21]. By using a controlled
laboratory rearing method of X. saxesenii fungus-growing ambrosia beetles [53], the present study
provides a first glimpse of the intricate biochemical mechanisms whereby beetles with the help of
symbionts are able to exclusively live from wood.
As expected, plant cell-wall degrading cellulases, endo-xylanases and the pectinolytic endo-β1,4-mannanase dominate the enzymatic profile but also consistent endo-protease activity against
casein were detected at all measured time-points in all three gallery compartments (Fig. 1A). Taken
together the enzymatic profile of the microbial consortium of X. saxesenii ambrosia galleries
resembles common wood degrading ascomycete and basidiomycete fungi [50, 51, 54], highlighting
the universal similarity of enzymes required in the initial degradation of recently dead wood material.
However, the production of extracellular enzymes by filamentous fungi is highly dependent on the
growth medium and external conditions like temperature and moisture etc. [50], and it is therefore
extremely difficult if not impossible to obtain natural enzyme activity profiles under in-vitro
laboratory conditions because the actual micro-habitat experienced by microbes in nature cannot be
satisfactorily reproduced in the laboratory. Despite these caveats, the detailed enzymatic
measurements of laboratory reared and age-controlled beetle galleries containing all the naturally
vectored symbiotic microbes still provide an informative substitute for natural measurements of the
usually inaccessible ambrosia beetle galleries deep inside wood.
Ambrosia beetle galleries are not static environments because the composition of the
associated microbial consortium changes with gallery age [22, 30]. However, despite this gradual
turnover in the microbial flora, cellulase and endo-protease activity were remarkably similar at all
three stages of gallery development measured (Fig. 1A). In contrast, endo-xylanase activity
significantly increased in the brood chamber but decreased in the entrance tunnel with gallery age
(Fig. 1A). These distinct changes in endo-xylanase activity most likely reflect that the wood
surrounding the gallery walls is in different stages of degradation at the different gallery ages, which
thereby influence the set of hydrolytic enzymes required for efficient hydrolysis. However, we are
unable to distinguish whether these shifts in enzyme activity is due to changes in endo-xylanase
126
Chapter 8
production by the resident microbes, the succession of microbes in the galleries or beetle activities.
In addition, when comparing the endo-β-1,4-glucanase and endo-β-1,4-mannanase activity between
compartments within galleries, the entrance tunnel and brood chamber showed remarkably similar
enzymatic profiles whereas the gallery dump has much higher activity (Fig. 1A). The expelled sawdust material in the gallery dump material most likely represents a more accessible carbohydrate
resource for microbial hydrolysis than the galleries themselves and the observed enzyme activity is
likely due to opportunistic bacteria and fungi not necessarily associated with the beetle galleries.
Endo-symbionts play a crucial role in nutrient acquisition in many wood-feeding arthropods,
like termites or wood-boring beetles [2, 4, 49]. In bark and ambrosia beetles they seem of minor
importance because these beetles either feed on nutrient rich phloem, xylomycetophagously on
nutrient enriched fungus infested wood or mycetophagously only on fungal tissues. The gut flora of
ambrosia beetles has not been studied, but for bark beetles the species richness in larval and adult
guts is relatively low [37, 42]. The endo-symbiotic yeasts and bacteria in bark beetles have been
shown to detoxify poisonous wood compounds (e.g. tannins [11] and fix nitrogen [55]). However,
their role for wood-degradation appears rather small compared to the primary fungal symbionts that
are growing within the galleries of Ips and Dendroctonus beetles [37, 42]. For ambrosia beetles,
which only feed on fungi and not on woody material, an endogenous production of wood degrading
enzymes either by the beetles or associated endo-symbionts is therefore not expected. Hence, the
endo-β-1,3(4)-glucanase activity observed in whole-body extracts of X. saxesenii larvae and adults is
likely used to degrade the Ophiostomatoid fungal cultivars (Raffaelea sulfurea), which unusually for
fungi contain cellulose in their cell walls [27, 56]. However endogenous endo-β-1,3(4)-glucanase
production in arthropods is rare and has in certain cases been found to be horizontally acquired from
bacteria [57, 58], whereas the alternative explanation of acquisition and utilization of ingested fungal
enzymes is well known from several fungal-insect mutualisms (c.f. the acquired enzyme hypothesis
[13, 59], table 2). The enzymatic capabilities of ambrosia fungi are unknown, but other bark-beetle
associated Ophiostomatoid fungi are known to produce a variety of cellulolytic enzymes [44, 45, 48,
52, 60], although “blue-stain” fungi of which Ophiostoma sp. belong in general leave the cellulose
and cross-linking glycans mostly intact and instead utilize storage products in the living ray
parenchyma [27].
Unlike X. saxesenii adults, larvae of all species in the ambrosia beetle genus Xyleborinus are
xylomycetophagous and ingest both fungal tissue and particles of wood while feeding [31, 32].
Interestingly this difference in nutrition was also reflected by endo-β-1,4-xylanase activity observed
in whole-body extracts of larvae, but not of adults (Fig. 2). Again, like endo-β-1,3(4)-glucanase, this
endo-β-1,4-xylanase might be fungus derived, but that would imply that the fungus exclusively
produces endo-β-1,4-xylanase in the structures eaten by the larvae (e.g. xylanase activity in the
mycelial form of Ophiostoma ulmi is higher than in the yeast-like stage [45]) and not by the adults or
that the larvae but not the adults avoid internal proteolysis of this enzyme. On the other hand, a
handful of beetles has been shown to be capable of endogenous xylanase production, for example
larvae and adults of the wood-boring beetle Phaedon cochleariae [61] and the scolytid beetle
Hypothenemus hampei [Uniprot: E2J6M9]. Alternatively, endo-β-1,4-xylanase enzymes might be
produced by gut endo-symbionts that are either specific to the larvae (larval specific bacteria are
known from Ips and Dendroctonus bark beetles [37, 38]), or the endo-symbionts facultatively
produce and secrete endo-β-1,4-xylanases depending on context. Irrespective of enzymatic origin,
the breakdown of cross-linking glycans within the larval intestinal tract likely (i) has a positive
influence on larval nutrition and (ii) is enhanced by active mixing of small woody particles with endoβ-1,4-xylanase and endo-β-1,3(4)-glucanase.
Conclusions
Despite differences in the type of substrate used to cultivate symbiotic fungi a striking, but
perhaps not surprising, commonality between the major insect fungus-growing systems is the direct
or indirect use of similar fungal carbohydrate active enzymes to utilize recalcitrant plant material as a
stable source of food (Table 2). Plant cell-wall degrading xylanases, pectinases and perhaps to a
lesser degree cellulases in all cases dominate the enzymatic profiles, although inherent variation
between fungus-growing systems are certainly present at the level of specific enzymes.
Endogenously produced cellulase enzymes for example are not common among arthropods [62],
127
Chapter 8
which indicates that the provision of essential carbohydrate active fungal enzymes heavily shift the
cost/benefit ratio of fungus culturing in favour of symbiotic coevolution.
Feeding activity of X. saxesenii larvae not only benefits other group members by creating
more space for the ambrosia fungus to form ambrosial layers on the gallery walls, but here we show
that it also enhances wood degradation and nutrient cycling. Predigested larval feces, which
contains small woody particles and probably also enzymes, is smeared on gallery walls after
defecation [33]. The wood particles in this fecal inoculum apparently get further degraded and
nitrogenous excretions are recycled by the ambrosia fungi growing there [63]. This may in turn
explain the positive effect of larval numbers on group productivity in X. saxesenii [32], and
demonstrates a synergism between age groups that prevents competition for fungal food, because
adults and larvae feed differently and apparently use a complementary set of enzymes. The
differences in the enzymatic capabilities between X. saxesenii larvae and the adults are of high
interest for understanding the social system of this species. X. saxesenii is the only primitively
eusocial ambrosia beetle described and similarly to the highly eusocial ants, bees and termites
exhibit division of labor not only between the sexes, but most importantly also between larval and
adult offspring. Differential enzyme activity therefore adds an additional layer of complexity to the
behavioural division of labour between adults and larvae. Production of extra enzymes and nutrients
by larvae (and their trophallaxis to adults) has also been reported from other social insects, like ants
and wasps [64-66], and larvae of the leaf-cutting ant Acromyrmex subterraneous have even been
denoted the “digestive caste” of the colony based on the extensive enzymatic machinery detected in
their gut lumen [67]. We suggest that larvae in holometabolous insect societies may play a much
bigger role in resource utilization than is currently recognized.
Materials and methods
Biological samples
X. saxesenii adult females were collected in the Spilwald forest (560 m asl; 46°95’, 7°31’) close to
Berne, Switzerland in January 2010, by dissection of galleries from stumps of beech trees (Fagus
sylvatica) that had been cut about a year earlier. Individual adult females were brought to the
laboratory and sat up in rearing tubes as previously described [53], using a sterile nutrient enriched
beech saw-dust media solidified with agar in ~15 mL plastic tubes. Due to their obligate sib-mating
and vertical transmission of symbionts, emerging female offspring can be collected from the surface
of the media and immediately used for consecutive breeding by placing them onto new media.
Galleries used in this study are from the 5th laboratory generation. X. saxesenii galleries typically
consist of a straight entrance tunnel dug perpendicular into the media for about 2-5 cm where it
reaches a flat brood chamber of 2-3 cm2 and a height of 1 mm (Fig. 1B, S3). As long as there are
larvae present, this chamber is continuously expanded by their feeding activity. Thus larvae
constantly create fresh substrate and more space for the ambrosia fungus to form ambrosia layers
on the chamber walls, which are mainly grazed off by the adult beetles. Larval feces containing
woody pieces are smeared on gallery walls, apparently for further degradation and recycling of
nutrients by the fungus [33]. Major parts of the finely fragmented sawdust material and feces are,
however, disposed through the entrance in a pile of dump on the surface of the media (Fig. 1B).
In this study we collected samples from the gallery and the beetles and larvae themselves.
The gallery samples were obtained from three different locations and were collected at three
different time points and the specific enzyme activity immediately measured for 13 different plant
cell wall degrading enzymes (Table 1). Samples contained the entire microbiome (i.e. fungi, yeasts
and bacteria) (a) from the gallery dump, and (b) from wall material of the entrance tunnel and (c) the
brood chamber. These samples were taken at different time points by dissecting galleries either 45
days, 62 days and 87 days after gallery foundation, when also the number of adults, pupae, 1 st and
2nd/3rd instar larvae were counted. These sampling points roughly correspond to different stages in
the life-cycle of a gallery: at day 45 there are on average few adults and many 1 st and 2nd/3rd instar
larvae present in the gallery and the primary fungus in the microbiome at this stage is the Raffaelea
sulfurea symbiont [22, 30]; at day 62 there are more adults present on average but fewer larvae,
dispersal of adults is just starting and the microbiome has changed and is no longer completely
dominated by R. sulfurea but a mixture of several saprobes [22, 30]; and at day 87 almost all beetles
present are adults, many adults have left the gallery and the rest is close to leaving. At this stage
128
Chapter 8
gallery walls are heavily melanized and many different species of fungi and yeasts are present [22,
30].
Protein extraction
Total proteins were extracted by grinding 30 mg material (wet weight) of entrance tunnel, brood
chamber or gallery dump respectively in an Eppendorf tube with 260 µl ddH20 containing 0.1 %
Tween20 with a small plastic pestle. Samples were vortexed and centrifuged at 15.000g for 15 min at
4 °C. after which the supernatant was immediately measured for enzyme activity to minimize
internal proteolysis. Tween20 was added to the extraction water to keep enzymes in suspension. In
total 11, 15 and 8 galleries from day 45, 62 and 87, respectively, were used giving a total sample size
of 91 extracts from 34 galleries. 20 gallery dump samples had to be discarded because of insufficient
material.
In addition, total enzymes were extracted from adult beetles, 1st and 2nd/3rd instar larvae
from the gallery cohort dissected at day 45. All individuals were surface sterilized once in bleach and
once in 96% alcohol. Thereafter, either three adults, four 2nd/3rd instar larvae or twelve 1st larvae
were combined to normalize the amount of biological material to approximately 30 mg biomass and
grinded in 60 µl ddH20 containing 0.1 % Tween20.
Enzyme activity measurements
Enzyme activity was assayed with Azurine-Crosslinked (AZCL) polysaccharides that are purified
polysaccharides cross-linked with a blue dye to form a water insoluble substrate commercially
available from Megazyme© (Bray, Ireland) in the form of a powder (table 1). Assay plates were
prepared as previously described [68, 69] with a medium consisting of 1% agarose, 23 mM
phosphoric acid, 23 mM acetic acid and 23 mM boric acid, mixed and adjusted to pH = 6. The
medium was heated using a microwave to melt the agarose. When the medium had cooled to 65°C,
0.1 % weight/volume AZCL substrate wetted in 96% ethanol was added. The medium was then
poured into Petri dishes and allowed to solidify before ~4 mm2 wells were made with a cut off
pipette tip. 15 µl supernatant of each enzyme extract were applied per well and after 24 hours of
incubation at room temperature (ca. 21°C) in the dark the plates were photographed and later
analysed by measuring the area of the blue halo surrounding each well with image analysis software
(ImageJ ver. 1.37v, W. Rasband, http://rsb.info.nih.gov/ij/). The area surrounding the blue halo is a
quantitative measure of enzyme activity that can be compared between samples but does not
provide absolute values of enzyme activity [68, 70, 71]. A pilot study showed no activity of either
gallery, beetle or larval extracts against the substrates AZCL-pullulan, AZCL-chitosan, AZCL-curdlan,
and AZCL-pachyman (data not shown) and these substrates where therefore omitted from this
experiment (Table 1).
Data analysis
Enzyme activity of the gallery data were ’log + 1’ transformed to improve normalization of the data.
Enzyme activity were analyzed separately for each substrate with a mixed linear model with the
factorial variables sample (three levels: ‘entrance’, ‘brood chamber’ and ‘gallery dump’) and the
interaction between sample and gallery age (three levels: ‘age45’, ‘age62’ and ‘age87’), sample and
gallery class (three levels: ‘adult beetles and immatures present’, ‘only immatures present’ and ‘no
beetles or larvae present’) and the continuous numerical variable ‘total number of animals in gallery
present’ as fixed effects. Because entrance, brood chamber and gallery dump samples from the same
gallery are not independent measures each gallery were assigned a code that was included as a
random factor in the model. Model estimation was performed with Maximum Likelihood using the
lme function implemented in R [72] and each variable evaluated by ANOVA analysis of log-likelihood
scores using a step-wise model reduction scheme. Specific means were compared with Tukey’s
multiple comparisons of the final model. The correlation of a particular enzyme activity and sample
location in the gallery were further analyzed using partial least square regression of a matrix
consisting of three x-variables (sample locations: entrance, brood chamber and gallery dump) and 13
y-variables (enzyme activity for each substrate screened) using the R package pls [73]. No statistical
analysis were performed on enzyme activities extracted from larvae or beetles because although
samples were approximately standardized to the same total biomass, much individual variation and
the inherent physiological difference between larval and adult morphology would render the result
very ambiguous.
129
Chapter 8
Authors’ contributions
HHDFL and PHWB designed the study. PHWB collected and maintained laboratory beetle galleries.
HHDFL and PHWB conducted enzyme measurements and HHDFL analyzed the data. HHDFL and
PHWB wrote the manuscript in collaboration.
Authors’ information
HHDFL is a postdoctoral fellow at the section for microbial ecology at Lund University, Sweden, and
studies the ecology and evolution of mutualistic systems. PHWB is Ph.D. student at the Department
of Behavioural Ecology at University of Berne, Switzerland. PHWB studies the evolutionary
mechanisms of social and mutualistic interactions.
Acknowledgements
The authors thank Marko Rohlfs, Anders Tunlid and Morten Schiøtt for helpful comments on a
previous version of this manuscript and Jacobus J. Boomsma for providing the AZCL reagents. HHdFL
gratefully acknowledges the M. P. Christiansen and wife foundation administered by the Danish
Mycological Society for a research travel grant. HHdFL is supported by a fellowship from the Danish
Research Council | Natural Sciences and PHWB partially by a DOC fellowship of the Austrian
Academy of Sciences and by a fellowship of the Roche Research Foundation.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
Mora C, Tittensor DP, Adl S, Simpson AGB, Worm B: How many species are there on earth
and in the ocean. PLoS Biol 2011, 9:e1001127.
Bourtzis K, Miller TA: Insect symbiosis. Boca Raton: CRC Press; 2003.
Hillis WE: Heartwood and tree exudates. New York: Springer; 1987.
Warnecke F, Luginbuhl P, Ivanova N, Ghassemian M, Richardson TH, Stege JT, Cayouette M,
McHardy AC, Djordjevic G, Aboushadi N, Sorek R, Tringe SG, Podar M, Martin HG, Kunin V,
Dalevi D, Madejska J, Kirton E, Platt D, Szeto E, Salamov A, Barry K, Mikhailova N, Kyrpides
NC, Matson EG, Ottesen EA, Zhang XN, Hernandez M, Murillo C, Acosta LG, et al:
Metagenomic and functional analysis of hindgut microbiota of a wood-feeding higher
termite. Nature 2007, 450:560-U517.
Martin MM: Invertebrate-microbial interactions. Ithaca: Cornell University Press; 1987.
Grunwald S, Pilhofer M, Holl W: Microbial associations in gut systems of wood- and barkinhabiting longhorned beetles [Coleoptera: Cerambycidae]. Syst Appl Microbiol 2010, 33:2534.
Anagnostou C, Dorsch M, Rohlfs M: Influence of dietary yeasts on Drosophila melanogaster
life-history traits. Entomol Exp Appl 2010, 136:1-11.
Hammond PM, Lawrence JF: Mycophagy in insects: A summary. In Insect-fungal
associations. New York: Oxford University Press; 2005: 275-283
Mueller UG, Gerardo NM, Aanen DK, Six DL, Schultz TR: The evolution of agriculture in
insects. Annu Rev Ecol Syst 2005, 36:563-595.
Ayres MP, Wilkens RT, Ruel JJ, Lombardero MJ, Vallery E: Nitrogen budgets of phloemfeeding bark beetles with and without symbiotic fungi. Ecology 2000, 81:2198-2210.
Hunt DWA, Borden JH: Conversion of Verbenols to verbenone by yeasts isolated from
Dendroctonus ponderosae (Coleoptera, Scolytidae). J Chem Ecol 1990, 16:1385-1397.
Dowd PF: Insect fungal symbionts - A promising source of detoxifying enzymes. J Ind
Microbiol 1992, 9:149-161.
Kukor JJ, Martin MM: Acquisition of digestive enzymes by Siricid woodwasps from their
fungal symbiont. Science 1983, 220:1161-1163.
Francke-Grosmann H: Hautdrüsen als Träger der Pilzsymbiose bei Ambrosiakäfern. Z
Morphol Tiere 1956, 45:275-308.
Francke-Grosmann H: Ectosymbiosis in wood-inhabiting beetles. In Symbiosis. Edited by
Henry SM. New York: Academic Press; 1967: 141-205
Six DL: Bark beetle-fungus symbioses. In Insect Symbiosis. Edited by Bourtzis K, Miller TA.
Boca Raton: CRC Press; 2003: 97-114
130
Chapter 8
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
Grebennikov VV, Leschen RAB: External exoskeletal cavities in Coleoptera and their possible
mycangial functions. Entomol Sci 2010, 13:81-98.
Farrell BD, Sequeira AS, O'Meara BC, Normark BB, Chung JH, Jordal BH: The evolution of
agriculture in beetles (Curculionidae : Scolytinae and Platypodinae). Evolution 2001,
55:2011-2027.
Hulcr J, Kolarik M, Kirkendall LR: A new record of fungus-beetle symbiosis in Scolytodes bark
beetles (Scolytinae, Curculionidae, Coleoptera). Symbiosis 2007, 43:151-159.
Beaver RA: Insect-fungus relationships in the bark and ambrosia beetles. In Insect-fungus
interactions. Edited by Wilding N, Collins NM, Hammond PM, Webber JF. London: Academic
Press; 1989: 121-143
Harrington TC: Ecology and evolution of mycophagous bark beetles and their fungal
partners. In Insect-fungal associations. New York: Oxford University Press; 2005: 257-295
Kajimura H, Hijii N: Dymamics of the fungal symbionts in the gallery system and the
mycangia of the ambrosia beetle, Xylosandrus mutilatus (Blandford) (Coleoptera,
Scolytidae). Ecol Res 1992, 7:107-117.
Kurtzman CP, Robnett CJ: Three new insect-associated species of the yeast genus Candida.
Can J Microbiol 1998, 44:965-973.
Haanstad JO, Norris DM: Microbial symbiotes of the ambrosia beetle Xyletorinus politus.
Microb Ecol 1985, 11:267-276.
Grubbs KJ, Biedermann PHW, Suen G, Adams SM, Moeller JA, Klassen JL, Goodwin LA, Woyke
T, Munk AC, Bruce D, et al: The complete genome sequence of Streptomyces cf. griseus
(XyelbKG-1), an Ambrosia beetle-associated Actinomycete. J Bacteriol 2011, 193:2890-2891.
Roeper RA, French JRJ: Ambrosia fungi of the Western United States and Canada - beetle
assocaitions (Coleoptera: Scolytidae), tree hosts, and distribution. Northwest Science 1981,
55:305-309.
Kirisits T: Fungal associates of European bark beetles with special emphasis on the
ophiostomatoid fungi. In Bark and Wood Boring Insects in Living Trees in Europe, a Synthesis.
Edited by Lieutier F, Keith RD, Battisti A, Gregoire JC, Evans HF. Dordrecht: Springer; 2004:
181-237
Alamouti SM, Tsui CKM, Breuil C: Multigene phylogeny of filamentous ambrosia fungi
associated with ambrosia and bark beetles. Mycol Res 2009, 113:822-835.
Endoh R, Suzuki M, Okada G, Takeuchi Y, Futai K: Fungus symbionts colonizing the galleries
of the Ambrosia beetle Platypus quercivorus. Microb Ecol 2011, 62:106-120.
Biedermann PHW, Klepzig KD, Ott E, Taborsky M, Six DL: Dynamics of filamentous fungi in
the ambrosia gardens of the primitively eusocial beetle Xyleborinus saxesenii Ratzeburg
(Scolytinae: Curculionidae). submitted.
Roeper RA: Patterns of mycetophagy in Michigan ambrosia beetles. Michigan Academian
1995, 27:153-161.
Biedermann PHW, Taborsky M: Larval helpers and age polyethism in ambrosia beetles. Proc
Natl Acad Sci USA 2011:1-6.
Hubbard HG: Some miscellaneous results of the work of the division of entomology. In US
department of agriculture bureau of entomology bulletin No 7. Edited by Howard LO.
Washington, DC.: US department of agriculture; 1897: 9-13.
Parkin EA: The digestive enzymes of some woodboring beetle larvae. J Exp Biol 1940,
17:364-377.
Balogun RA: Digestive enzymes of alimentary canal of larch bark beetle Ips cembrae Heer.
Comp Biochem Physiol 1969, 29:1267-1270.
Rivera FN, Gonzalez E, Gomez Z, Lopez N, Hernandez-Rodriguez C, Berkov A, Zuniga G: Gutassociated yeast in bark beetles of the genus Dendroctonus erichson (Coleoptera:
Curculionidae: Scolytinae). Biol J Linn Soc 2009, 98:325-342.
Delalibera I, Vasanthakumar A, Burwitz BJ, Schloss PD, Klepzig KD, Handelsman J, Raffa KF:
Composition of the bacterial community in the gut of the pine engraver, Ips pini (Say)
(Coleoptera) colonizing red pine. Symbiosis 2007, 43:97-104.
Vasanthakumar A, Delalibera I, Handelsman J, Klepzig KD, Schloss PD, Raffa KF:
Characterization of gut-associated bacteria in larvae and adults of the southern pine
beetle, Dendroctonus frontalis Zimmermann. Environ Entomol 2006, 35:1710-1717.
131
Chapter 8
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
Valiev A, Ogel ZB, Klepzig KD: Analysis of cellulase and polyphenol oxidase production by
southern pine beetle associated fungi. Symbiosis 2009, 49:37-42.
Geib SM, Filley TR, Hatcher PG, Hoover K, Carlson JE, Jimenez-Gasco MD, Nakagawa-Izumi A,
Sleighter RL, Tien M: Lignin degradation in wood-feeding insects. Proc Natl Acad Sci USA
2008, 105:12932-12937.
Schmidt O, Dietrichs HH: Zur Aktivität von Bakterien gegenüber Holzkomponenten. In
Organismen und Holz. Edited by Becker G, Liese W. Berlin: Duncker und Humblot; 1976: 91102
Delalibera I, Handelsman J, Raffa KF: Contrasts in cellulolytic activities of gut
microorganisms between the wood borer, Saperda vestita (Coleoptera: Cerambycidae),
and the bark beetles, Ips pini and Dendroctonus frontalis (Coleoptera: Curculionidae).
Environ Entomol 2005, 34:541-547.
Rosch R, Liese W, Berndt H: Studies on enzymes of blue-stain fungi .I. Cellulase-,
Polygalacturonase-, Pectinesterase- and Laccase-Activity. Arch Mikrobiol 1969, 67:28-50.
Przybyl K, Dahm H, Ciesielska A, Molinski K: Cellulolytic activity and virulence of Ophiostoma
ulmi and O. novo-ulmi isolates. Forest Pathol 2006, 36:58-67.
Binz T, Canevascini G: Xylanases from the Dutch elm disease pathogens Ophiostoma ulmi
and Ophiostoma novo-ulmi. Physiol Mol Plant P 1996, 49:159-175.
Binz T, Gremaud C, Canevascini G: Production and purification of an extracellular betagalactosidase from the Dutch elm disease fungus Ophiostoma novo-ulmi. Can J Microbiol
1997, 43:1011-1016.
Beckman CH: Production of Pectinase, Cellulases, and growth-promoting substance by
Ceratostomella Ulmi. Phytopathology 1956, 46:605-609.
Svaldi R, Elgersma DM: Further studies on the activity of cell wall degrading enzymes of
aggressive and non-aggressive isolates of Ophiostoma ulmi. Eur J Forest Pathol 1982, 12:2936.
Vega FE, Dowd PF: The role of yeasts as insect endosymbionts. In Insect-Fungal Assocations:
Ecology and Evolution. Edited by Vega FE, Blackwell M. New York: Oxford University Press;
2005: 211-243
Cooke RC, Rayner ADM: Ecology of saprotrophic fungi. Harlow: Longmann Group limited;
1984.
Dighton J: Nutrient cycling by saprotrophic fungi in terrestrial habitats. In The Mycota IV:
Environmental and microbial relationships. Edited by Esser K, Lemke PA. Berlin: Springer
Verlag; 1997: 287-300.
Nilsson T: Soft-rot fungi - decay patterns and enzyme production. In Organismen und Holz.
Edited by Becker G, Liese W. Berlin: Duncker und Humblot; 1976: 103-112.
Biedermann PHW, Klepzig KD, Taborsky M: Fungus cultivation by ambrosia beetles:
behavior and laboratory breeding success in three xyleborine species. Environ Entomol
2009, 38:1096-1105.
Dix NJ, Webster J: Fungal Ecology. London: Chapman & Hall; 1995.
Bridges JR: Nitrogen-fixing bacteria associated with bark beetles. Microb Ecol 1981, 7:131137.
Hoog GSD, Scheffer RJ: Ceratocystis versus Ophiostoma: A reappraisal. Mycologia 1984,
76:292-299.
Kikuchi T, Shibuya H, Jones JJ: Molecular and biochemical characterization of an endo-β-1,3glucanase from the pinewood nematode Bursaphelenchus xylophilus acquired by
horizontal gene transfer from bacteria. Biochem J 2005, 389:117-125.
Song JM, Nam K, Sun YU, Kang MH, Kim CG, Kwon ST, Lee J, Lee YH: Molecular and
biochemical characterizations of a novel arthropod endo-beta-1,3-glucanase from the
Antarctic springtail, Cryptopygus antarcticus, horizontally acquired from bacteria. Comp
Biochem Phys B 2010, 155:403-412.
Martin MM, Martin JS: Cellulose digestion in the midgut of the fungus-growing termite
Macrotermes natalensis: The role of acquired digestive enzymes. Science 1977, 199:14531455.
Tamerler C, Keshavarz T: Lipolytic enzyme production in batch and fed-batch cultures of
Ophiostoma piceae and Fusarium oxysporum. J Chem Tech Biotech 2000, 75:785-790.
132
Chapter 8
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
Girard C, Jouanin L: Molecular cloning of cDNAs encoding a range of digestive enzymes
from a phytophagous beetle, Phaedon cochleariae. Insect Biochem Molec 1999, 29:11291142.
Watanabe H, Tokuda G: Animal cellulases. Cell Mol Life Sci 2001, 58:1167-1178.
Norris DM: Chemical interdependence among Xyleborus spp. ambrosia beetles and their
symbiotic microbes. Mater Organismen 1975, 3:479-788.
Ishay J, Ikan R: Food exchange between adults and larvae in Vespa orientalis F. Anim Behav
1968, 16:298-303.
Hunt JH, Baker I, Baker HG: Similarity of amino-acids in nectar and larval saliva - the
nutritional basis for trophallaxis in social wasps. Evolution 1982, 36:1318-1322.
Hölldobler B, Wilson EO: The Ants. Cambridge: Harvard University Press; 1990.
Erthal MJ, Silva CP, Samuels RI: Digestive enzymes in larvae of the leaf cutting ant,
Acromyrmex subterraneus (Hymenoptera: Formicidae: Attini). J Insect Physiol 2007,
53:1101-1111.
De Fine Licht HH, Schiøtt M, Mueller UG, Boomsma JJ: Evolutionary transitions in enzyme
activity of ant fungus gardens. Evolution 2010, 64:2055-2069.
Schiøtt M, De Fine Licht HH, Lange L, Boomsma JJ: Towards a molecular understanding of
symbiont function: Identification of a fungal gene for the degradation of xylan in the
fungus gardens of leaf-cutting ants. BMC Microbiol 2008, 8.
Pedersen M, Hollensted M, Lange L, Andersen B: Screening for cellulose and hemicellulose
degrading enzymes from the fungal genus Ulocladium. Int Biodeter Biodegr 2009, 63:484489.
Kooij PW, Schiott M, Boomsma JJ, Licht HHD: Rapid shifts in Atta cephalotes fungus-garden
enzyme activity after a change in fungal substrate (Attini, Formicidae). Insec Soc 2011,
58:145-151.
R development core team: R: A language and environment for statistical computing. Vienna,
Austria; 2011.
Mevik BH, Wehrens R: The pls package: principal component and partial least squares
regression in R. J stat softw 2007, 18:1-24.
133
Chapter 8
Figures
Figure 1 Glycoside hydrolytic enzyme activity of X. saxesenii ambrosia beetle galleries. A. Enzyme
activity of 13 specific carbohydrate active enzymes presented as a heatmap with darker coloration
showing higher enzyme activity. Enzyme activity was measured when only immature brood was
present, when both immature and adult brood were present, and finally when only adult brood were
present (45, 62 and 87 days respectively, after gallery foundation by a single mated female). Enzymes
are divided into functional groups according to the plant cell structure functioning as substrate for
enzymatic hydrolysis. B. Picture of a X. saxesenii gallery in artificial media around day 45 after gallery
foundation. Note the three distinct compartments, where enzyme activities were measured:
entrance tunnel, brood chamber, and gallery dump. Many white larvae and a few light brown teneral
females are visible in the brood chamber and the lower part of the entrance tunnel.
134
Chapter 8
Figure 2 Endo-β-1,4-glucanase, endo-β-1,3(4)-glucanase, endo-β-1,4-xylanse and endo-protease
activity (mean±SE) of adult (n = 14×3 adults), large (2nd/3rd instar, n = 14×4 larvae) and small (1st
instar, n = 8×12 larvae) X. saxesenii larvae, respectively. Endo-β-1,3(4)-glucanase activity is present in
all three life stages, whereas endo-β-1,4-xylanase activity is not present in adult beetles and only
detected in large and small larvae.
Table 1. Insoluble chromogenic substrates used to test for enzyme activity and the specific type of
enzymes measured.
Substrate
Enzyme
Starch
AZCL-Amylose
α-amylase
Protein
AZCL-Casein
endo-protease
AZCL-Collagen
endo-protease
Pectin
AZCL-Debr. Arabinan
endo-α-1,5-arabinase
AZCL-Rhamnogalacturonan
rhamnogalacturonanase
AZCL-Galactomannan
endo-β-1,4-mannanase
AZCL-Galactan
endo-β-1,4-galactanase
Cellulose
AZCL-HE-Cellulose
cellulase (endo-β-1,4-glucanase)
AZCL-Barley β-Glucan
cellulase (endo-β-1,3-1,4-glucanase)
AZCL-Xyloglucan
endo-β-1,4-xyloglucanase
Cross-linking Glycans
AZCL-Xylan
endo-β-1,4-xylanase
AZCL-Arabinoxylan
endo-β-1,4-xylanase
AZCL-Dextran
endo-α-1,6-dextranase
AZCL = Azurine cross-linked polysaccharides (Megazyme©, Bray, Ireland).
135
Chapter 8
Table 2. Overview of highly derived, obligate nutritional symbioses between insects and fungi.
Coleoptera
Ambrosia
beetles
Curculionidae
Bark beetles1
Mutualistic
fungi
Ascomycota
(Ambrosiella,
Raffaelea,
Fusarium)
Ascomycota
(Ophiostoma,
Ceratocysti
opsis,
Grosmannia,
Ceratocystis)
Basidiomycota
(Entomocortici
um)
Age (Mya)
21–60
?
Agriculture
Mode of
cultivation
Mode of
nesting
Multi-species
community
Wood galleries
& chambers
Multi-species
community
Phloem
galleries &
chambers
Substrate for
fungi
Surrounding
wood
Mode of
agriculture2
Insect family
Enzymatic
profile
Fungus
garden (incl.
microbial
community)
Fungus
acquired
enzymes3
Mode of
feeding4
Adults
Larvae
Diptera
Hymenoptera
Gall midges
Wood wasps
Cecidomyiidae
Xiphydriidae,
Orussidae,
Anaxyelidae,
Siricidae
Ascomycota
(Endomyces)
Ascomycota
(Lasioptera,
Ramichlori
dium)
?
Isoptera
Fungus-growing
ants
Formicidae
Fungus-growing
termites
Termitidae
Basidiomycota
(Cerrena,
Stereum,
Amylostereum);
Ascomycota
(Daldinia
decipiens,
Entonaema
cinnabarina)
Basidiomycota
(Leucocoprinus,
Leucoagaricus
and the family
Pterulaceae)
Basidiomycota
(Termitomyces)
?
?
45–65
24–34
Monoculture
Monoculture
Monoculture
Monoculture
Monoculture
Wood tunnels
Plant galls
Wood tunnels
Subterranean
nests (occ.
mounds)
Subterranean
nests and
mounds
Surrounding
phloem (and
wood)
Surrounding
wood
Surrounding
plant tissue
Surrounding
wood
Collected plant
material
(dry leaf litter,
twigs, wood)
Advanced
Primitive
(possibly
advanced in
Dendroctonus
)
Primitive
?
Primitive
Collected plant
material
(twigs,
caterpillar feces,
leaf litter,
flowers, fruits,
fresh leaves)
Advanced
?
?
Wood
degrading
saprotrophism
(this study)
Wood
degrading
saprotrophism
Possible
(this study)
Possible
(e.g. in
Dendroc
tonus)
?
?
Present
Mycetophagy
Phloeomycetophagy
Phloeomycetophagy
No food
Plant sap
No food
Curculionidae
Ship-timber
beetles
Lymexylidae
Wood
degrading
saprotrophism
Saprotrophism
(saprobic and
biotrophic in
leaf-cutting
ants)
Present
Mycetophagy,
(plant material)
Mycetophagy
Advanced
Saprotrophism
(plant cell-wall
degrading)
Present
Mycetophagy,
(plant material)
Mycetophagy
Mycetophagy
XyloMycetophagy
Xylo(Xylomycetophagy
mycetophagy
mycetophagy5)
1
Here we only refer to bark beetles in nutritional symbioses with fungi.
2
Primitive fungiculture is defined only by dispersal and seeding of fungi; advanced fungiculture additionally involves active care of the
fungal crops (cf. Mueller et al. 2005).
3
Evidence for fungus acquired enzymes that are active in the insect gut or fecal exudates (Martin and Martin 1977; Kukor and Martin 1983).
4
Distinctions originating from the fungus-growing beetle literature: Mycetophagy = eating fungal mycelium, fruiting bodies or specific
fungal structures, Phloeomycethophagy = eating phloem and fungal biomass, Xylomycetophagy = eating xylem and fungal biomass.
5
Only in larvae of the genus Xyleborinus and probably Xylosandrus.
136
Chapter 8
SUPPORTING ONLINE MATERIAL
Patterns of functional enzyme activity in the fungus-growing ambrosia beetle mutualism reveal
that larvae may be key to successful wood degradation
Henrik H. de Fine Licht1 and Peter H. W. Biedermann2
1
Section of Microbial Ecology, Department of Biology, Lund University, Solvegatan 37, SE-223 62 Lund, Sweden,
Email: [email protected].
2
Department of Behavioural Ecology, Institute of Ecology & Evolution, University of Berne, Baltzerstrasse 6, CH3012 Bern, Switzerland, Email: [email protected].
Figure S1. X. saxesenii laboratory gallery enzyme activity (mean±SE) against 13 specific carbohydrate
substrates measured at 45, 62 and 87 days after gallery foundation by a single mated female. The
same data is presented as a heatmap in figure 1 in the manuscript. For ease of interpretation the
polysaccharide substrates are divided into five major groups based on enzyme activity: Cellulases –
cellulose, beta-glucan, xyloglucan; Cross-linking glycans – xylan, arabinoxylan, dextran; pectinases –
galactomannan, galactan, rhamnogalacturonan, debranched arabinan; Proteases – casein, collagen;
and Amylase – amylose. Different letters above horizontal columns indicate significant different
means in post-hoc Tukey’s tests. Ns = non-significant, k-w = Kruskal-Wallis test with a * denoting
significance at p < 0.05. Sample sizes for entrance tunnel, brood chamber, and gallery dump
respectively are: n = 16, n = 16, n = 11 at age 45, n = 11, n = 13, n = 5 at age 62, and n = 8, n = 8, n = 2
at age 87.
Figure S2.
Principal component analyses of gallery enzyme activity at age 45 (A), age 62 (B), and age 87 (C). X
loadings are gallery sampling points (brood chamber, entrance tunnel, and gallery dump); Y loadings
are specific enzyme activity: cellulos = HE-cellulose (cellulase), betaglu = β-glucan (cellulase / ßglucanase), xyloglu = xyloglucan (cellulase), xylan = endo-xylanase, axyl = arabinoxylan (endoxylanase), galman = galactomannan (endo-1,4-ß-mannanase), galac = galactan (endo-1,4-ß-galactanase),
debarab = debranched arabinan (endo-arabinase), casein = endo-protease, amyl = amylose (αamylase). The closer an enzyme substrate is to one of the three samples, the higher the particular
enzyme activity is in that sample.
Figure S3.
Picture of a X. saxesenii brood chamber (top) with 1st and 2nd/3rd instar larvae, pupae and teneral
females surrounded by lush mycelial growth of the symbiont Raffaelea sulfurea and close-up of two
adult females (bottom) inside a gallery with mites. Both pictures are form laboratory galleries and
the top picture is taken through the plastic tube containing the gallery. © P. H. W. Biedermann.
137
Chapter 8
Figure S1
138
Chapter 8
Figure S2A
Gallery age = 45
139
Chapter 8
Figure S2B
Gallery age = 62
140
Chapter 8
Figure S2C
Gallery age = 87
141
Chapter 8
Figure S3
142
Chapter 9
JOURNAL OF BACTERIOLOGY, June 2011, Vol. 193, No. 11
p. 2890-2891 0021-9193/11/$12.00 doi:10.1128/JB.00330-11
Copyright © 2011, American Society for Microbiology. All Rights Reserved.
Genome Sequence of Streptomyces griseus Strain XylebKG-1, an Ambrosia
Beetle-Associated Actinomycete
Kirk J. Grubbs,1,2 Peter H. W. Biedermann,1,3 Garret Suen,1,4 Sandra M. Adams,1,4 Joseph
A. Moeller,1 Jonathan L. Klassen,1 Lynne A. Goodwin,5,6 Tanja Woyke,5 A. Christine
Munk,5,6 David Bruce,5,6 Chris Detter,5,6 Roxanne Tapia,5,6 Cliff S. Han,5,6 and Cameron R.
Currie1,4*
1
Department of Bacteriology, University of Wisconsin—Madison, Madison, Wisconsin ; Cellular and Molecular Pathology Graduate
2
Program, University of Wisconsin—Madison, Madison, Wisconsin ; Institute of Ecology and Evolution, Division Behavioral Ecology,
3
University of Berne, Bern, Switzerland ; DOE Great Lakes Bioenergy Research Center, University of Wisconsin - Madison, Madison,
4
5
Wisconsin ; DOE Joint Genome Institute, Walnut Creek, California ; and Los Alamos National Laboratory,
6
Bioscience Division, Los Alamos, New Mexico
Received 3 March 2011/Accepted 25 March 2011
Streptomyces griseus strain XylebKG-1 is an insect-associated strain of the well-studied actinobacterial species S. griseus. Here, we
present the genome of XylebKG-1 and discuss its similarity to the genome of S. griseus subsp. griseus NBRC13350. XylebKG-1 was
isolated from the fungus-cultivating Xyleborinus sax-esenii system. Given its similarity to free-living S. griseus subsp. griseus
NBRC13350, comparative genom-ics will elucidate critical components of bacterial interactions with insects.
Streptomyces griseus is a soil bacterium known for its production of secondary metabolites, including streptomycin, the first
effective antibiotic for tuberculosis (19). Here, we present the
genome sequence of Streptomyces griseus strain XylebKG-1,
which, to our knowledge, is the first strain of S. griseus associated with an insect. XylebKG-1 was isolated from the ambrosia
beetle Xyleborinus saxeseni, which cultivates a fungus for food
(2). Actinobacteria-specific isolations from both beetles and their
fungal galleries resulted in isolation of XylebKG-1.
DNA from pure isolates was extracted using a bead-beating
protocol (10), and the genome was sequenced at the DOE Joint
Genome Institute (JGI). A noncontiguous finished genome of
XylebKG-1 was generated using a shotgun approach employing a
combination of Illumina (3) and 454 sequencing technologies
(15). An Illumina shotgun library (69,927,062 reads totaling ~5.3
Gbp) and two 454 GS (FLX Titanium) shotgun libraries (688,595
standard reads and 172,566 20-kbp paired-end reads totaling
~352 Mbp) were sequenced and assembled. The 454 data were
assembled using Newbler, version 2.3 (Roche), and Illumina
sequencing data were assembled with Velvet, version 0.7.63 (20).
All assemblies were integrated using parallel Phrap, version SPS D
4.24 (High Performance Software, LLC). Illumina data were used
to correct potential base errors and increase consensus quality
using the software program Polisher, developed at the JGI (Alla
Lapidus, unpublished). Possible misassembled regions were
corrected using gapResolution (Cliff Han, unpublished) or
Dupfinisher (11) or by sequencing cloned bridging PCR fragments
with subcloning. Gaps between contigs were closed by editing in
Consed (6-8), by PCR, and by Bubble PCR (J.-F.
Cheng, unpublished) primer walks. The total size of the genome is 8,727,768 bp, and the final assembly is based on 352.4
Mbp of 454 sequence (38 X coverage) and 5.3 Gbp of Illumina
sequence (257 X coverage).
The overall genome G+C content is 72.1%. Generation (http:
//compbio.ornl.gov/generation/), Glimmer (5), and Critica (version 1.05) (1) were used to predict a total of 7,265 candidate
protein-encoding gene models. RNAmmer (13) annotated six 16S
rRNAs and six 23S rRNAs. A tRNAscan-SE (14) search revealed 66
tRNAs corresponding to all 20 standard amino acids. Additional
PRIAM (4), KEGG (12), and COG (18) analyses were completed,
and the results can be accessed at http://genome.ornl
.gov/microbial/streACT1.
Evidence for XylebKG-1 as a strain of S. griseus is based on
similarity between the genomes of XylebKG-1 and the type strain
S. griseus subsp. griseus NBRC13350 (16). Both have similar
genome sizes (8.7 Mbp to 8.5 Mbp) and G+ C contents (72.1% to
72.2%), six rRNA operons, and 66 tRNAs. An average nucleotide
identity analysis conducted between both genomes using
Jspecies (version 1.2.1) (17) revealed a 98.98 ANIb value (91.68%
genome alignment) and a 98.95 ANIm value (94.13% genome
alignment), indi-cating a species level degree of similarity (9).
Given this similarity, XylebKG-1 represents a unique opportunity
to study genetic elements involved in Actinobacteria-insect associations.
Nucleotide sequence accession number. The genome sequence of Streptomyces griseus strain XylebKG-1 has been deposited in GenBank under accession no. ADFC00000000.
* Corresponding author. Mailing address: DOE Great Lakes Bioenergy
Research Center and Department of Bacteriology, 6155 MSB, 1550 Linden
Drive, University of Wisconsin - Madison, Madison, WI 53706. Phone:
(608) 265-8034. Fax: (608) 262-9865. E-mail: currie @bact.wisc.edu.
143
This work was funded by the DOE Great Lakes Bioenergy Research
Center (DOE BER Office of Science DE-FC02-07ER64494), supporting G.S.,
S.M.A., J.A.M., and C.R.C. This work was also funded by the National
Institute of Food and Agriculture, U.S. Department of Agriculture, under
identification number WISO1321, sup-
Chapter 9
porting K.J.G. P.H.W.B. was partly supported by a DOC fellowship
of the Austrian Academy of Sciences at the Department of
Behavioral Ecology, University of Bern. The work conducted by the
U.S. Department of Energy Joint Genome Institute is supported by
the Office of Science of the U.S. Department of Energy under
contract no. DE-AC02-05CH11231. Additional support was provided
by the National Science Foundation (MCB-0702025). Support for
J.L.K. comes from an NSERC postdoctoral fellowship.
9. Goris, J., et al. 2007. DNA-DNA hybridization values and their relationship to wholegenome sequence similarities. Int. J. Syst. Evol. Microbiol. 57:81-91.
10. Graff, A., and R. Conrad. 2005. Impact of flooding on soil bacterial communities
associated with poplar (Populus sp.) trees. FEMS Microbiol. Ecol.
53:401-415.
11. Han, C., and P. Chain. 2006. Finishing repetitive regions automatically with
Dupfinisher, p. 142-147. In H. Valafar (ed.), Proceedings of the 2006 International
Conference on Bioinformatics Computational Biology, BIOCOMP'06. CSREA Press, Las
Vegas, NV.
12. Kanehisa, M., S. Goto, M. Furumichi, M. Tanabe, and M. Hirakawa. 2010.
KEGG for representation and analysis of molecular networks involving diseases and
drugs. Nucleic Acids Res. 38:D355-D360.
13. Lagesen, K., et al. 2007. RNAmmer: consistent and rapid annotation of
ribosomal RNA genes. Nucleic Acids Res. 35:3100-3108.
14. Lowe, T. M., and S. R. Eddy. 1997. tRNAscan-SE: a program for improved
detection of transfer RNA genes in genomic sequence. Nucleic Acids Res.
25:0955-0964.
15. Margulies, M., et al. 2005. Genome sequencing in microfabricated high-density
picolitre reactors. Nature 437:376-380.
16. Ohnishi, Y., et al. 2008. Genome sequence of the streptomycin-producing
microorganism streptomyces griseus IFO 13350. J. Bacteriol. 190:4050-4060.
17. Richter, M., and R. Rossello-Mora 2009. Shifting the genomic gold standard for the
prokaryotic species definition. Proc. Natl. Acad. Sci. U. S. A. 106:
19126-19131.
18. Tatusov, R. L., M. Y. Galperin, D. A. Natale, and E. V. Koonin. 2000. The COG
database: a tool for genome-scale analysis of protein functions and evolution.
Nucleic Acids Res. 28:33-36.
19. Waksman, S. A. 1953. Streptomycin: background, isolation, properties, and
utilization. Science 118:259-266.
20. Zerbino, D. R., and E. Birney. 2008. Velvet: algorithms for de novo short read
assembly using de Bruijn graphs. Genome Res. 18:821-829.
REFERENCES
1. Badger, J. H., and G. J. Olsen. 1999. CRITICA: coding region identification tool invoking
comparative analysis. Mol. Biol. Evol. 16:512-524.
2. Batra, L. R. 1967. Ambrosia fungi: a taxonomic revision, and nutritional studies of
some species. Mycologia 59:976-1017.
3. Bennett, S. 2004. Solexa Ltd. Pharmacogenomics 5:433-438.
4. Claudel-Renard, C., C. Chevalet, T. Faraut, and D. Kahn. 2003. Enzyme-specific profiles
for genome annotation: PRIAM. Nucleic Acids Res. 31:
6633-6639.
5. Delcher, A. L., K. A. Bratke, E. C. Powers, and S. L. Salzberg. 2007. Identifying bacterial
genes and endosymbiont DNA with Glimmer. Bioinformat-ics 23:673-679.
6. Ewing, B., and P. Green. 1998. Base-calling of automated sequencer traces using
Phred. II. Error probabilities. Genome Res. 8:186-194.
7. Ewing, B., L. Hillier, M. C. Wendl, and P. Green. 1998. Base-calling of
automated sequencer traces using Phred. I. Accuracy assessment. Genome
Res. 8:175-185.
8. Gordon, D., C. Abajian, and P. Green. 1998. Consed: A graphical tool for sequence
finishing. Genome Res. 8:195-202.
144
Appendix 1
Fungal associates and their effects on the behaviours and success of the
ambrosia beetle Xyleborus affinis Eichhoff (Scolytinae: Curculionidae)
Peter H.W. Biedermann 1,2, Michael Taborsky1 and Diana L. Six3
1
Department of Behavioural Ecology, Institute of Ecology and Evolution, University of Bern, Baltzerstrasse
6, CH-3012 Bern, Switzerland
2
USDA Forest Service, Southern Research Station, 2500 Shreveport Hwy, Pineville, LA 71360, USA
3
Department of Ecosystem and Conservation Sciences, College of Forestry and Conservation, The
University of Montana, Missoula, MT 59812, USA
Corresponding author: Peter H.W. Biedermann, Baltzerstrasse 6, CH-3012 Bern, Switzerland; phone: 0041
31 631 3015; e-mail: [email protected]
Running title: Effects of fungi on ambrosia beetles
Manuscript in work
Note: All fungal species mentioned in this manuscript are not confirmed by molecular sequencing yet!
Abstract
True fungus gardens are found in fungus-farming ants, fungus-farming termites and ambrosia beetles
(Scolytinae, Platypodinae). Ant and termite fungus gardens consist of a complex community of fungi,
which are under control by the ants that aim to bias the biomass in favor of one cultivar. Fungus garden
communities of ambrosia beetles are largely unknown and studies on beetle behaviours in response to
different fungi are missing. The role of the different fungal associates for the beetles is unclear, but
observations of larval and adult behaviours in response to fungal garden communities can help to reveal
fungal roles and behavioural functions.
Using a laboratory breeding and observation technique we revealed a cooperatively breeding system in
the xyleborine ambrosia beetle Xyleborus affinis, which is characterized by delayed dispersal of adult
daughters, cooperative brood care by larvae and adults, and about half of the totipotent adult daughters
laying eggs within the natal nest. Most interesting, egg-laying females are more likely to engage in
cooperative behaviours. Furthermore, fungus gardens covering gallery walls composed of five different
filamentous fungi, of which a Raffaelea sp. A was predominant and likely served as the main food for
adults and larvae. Two species, Mucor sp. and Phaeoacremonium rubrigenum were most abundant in the
oldest gallery part close to the entrance, suggesting their parasitic role. Indeed, presence of P. rubrigenum
was also negatively correlated with the numbers of females nearby. Presence of fungi affected larval and
adult behaviours differently. Adult fungus cropping behaviour, which is feeding but may also serve
weeding, was more commonly observed in the presence of Unknown sp. In conclusion, species-diversity is
much lower in ambrosia gardens compared to gardens of fungus-farming ants. Behavioural control of
fungal weeds may be of minor importance in the ambrosia beetle mutualism.
145
Appendix 1
Introduction
Advanced fungus agriculture by insects has evolved once in ants, once in termites and eight times
independently in weevils (Farrell et al. 2001; Hulcr et al. 2007). Latter, combined in the ecological group
ambrosia beetles, are apparently especially prone for the evolution of fungus farming. They are regarded
to descend from phloem-feeding weevils, which shared their habitat with plant-pathogenic, endophytic
and saprobic fungal species (Klepzig and Six 2004). At the beginning, some of these fungi probably used
the beetles for getting dispersed to new plant material by producing sticky spores, as the insects have
target-oriented dispersal abilities. A nutritional dependence and fungus agriculture may have evolved in
the following (cf. the transmission-first model; Mueller et al. 2005). Advanced fungus agriculture by insects
today involves by definition (i) the insects’ obligate nutritional dependence on the fungi, (ii) adaptions by
the insect for transmission of fungal propagules, and (iii) the active management (farming) of the fungal
crops (Mueller et al. 2005).
The active management of fungal crops is dependent on sociality. Ants and termites had been
living in eusocial societies (characterized by reproductive division of labor, cooperative brood care, and
overlapping generations (Wilson 1971)) already at the origin of fungus agriculture. Ancestors of ambrosia
beetles, however, lived solitarily or gregariously, suggesting that active farming must have evolved in
association with social evolution (Mueller et al. 2005). One ambrosia beetle is known to be eusocial, living
within small family-groups that are composed by a reproducing queen and a few sterile workers
(Austroplatypus incompertus; Kent and Simpson 1992). At least one other ambrosia beetle, probably many
more, are primitively eusocial, living within family-groups with multiple reproductives, but no permanent
sterile worker caste (e.g. Xyleborinus saxesenii; 2012). The remainder of ambrosia beetles are either
subsocial (a single female cares for the brood) or communal (several reproductive females cooperate in
brood care and farming) (Kirkendall et al. 1997; Biedermann and Taborsky 2012). The level of sociality
typically relates to the longevity of the woody resource: The eusocial beetle breeds in living trees – an
infinite resource; the habitat of the others range from dying or dead large-diameter trees to small
diameter branches. Fungus agriculture is favoured by sociality because of the advantage of division of
labour, which is long known to strongly increase work efficiency of social groups (Smith 1776). Division of
labour is ubiquitous in the ant and termite farmers and has been recently also found in an ambrosia beetle
(Biedermann and Taborsky 2011). Xyleborinus saxesenii adult females engage in protection and
maintenance of the nest while larvae overtake gallery enlargement and some hygienic tasks. Larvae are
especially important because (i) they increase the wall surface for the fungus gardens to grow, (ii) they
suppress the spread of pathogenic fungi, (iii) their feces appears to be recycled by the fungus gardens and
(iv) unlike adults they produce xylanases (hemicellulases) for wood and fungus degradation (Biedermann
and Taborsky 2011; De Fine Licht and Biedermann 2012).
Ambrosia beetle galleries are of one of three types: (i) the brood-chamber type, (ii) the branching
tunnels type and (iii) the branching tunnels type with larval cradles (Biedermann and Taborsky 2012). If
these are settled for multiple generations (i.e., overtaken by the daughters) complex gallery patterns
result, characterized by tunnels (or brood-chambers) that branch several times and expand dozens of
centimeters (up to meters - drilled by beetles that are often less than 3 mm) into the wood (Schedl 1962;
Schneider 1987). The pattern is regarded important for social and beetle-fungus interactions: In branching
tunnel systems adults, brood and fungus gardens are spatially more separated than in communal brood
chambers (Biedermann and Taborsky 2011). Primitive eusociality, division of labour and fungus farming
has been recently reported from the brood-chamber type breeding species X. saxesenii (Biedermann and
Taborsky 2011), but the social system and fungus farming behaviours of species that construct branching
tunnels remain enigmatic. Promising model species are other species in the tribe Xyleborini, because this
group of beetles is ubiquitously predisposed for kin-selected sociality because of haplodiploid and
inbreeding within the natal nest (Peer and Taborsky 2004; 2005).
In general, healthy fungus gardens of ants, termites and beetles can be viewed as a complex
community of microbes that is dominated by the main cultivar fungus, which form the main part of the
garden biomass. In ambrosia beetles all developmental stages depend on the main ambrosia fungus as
their primary food source (Francke-Grosmann 1956; Batra 1963; Beaver 1989). Progeny complete their life
cycle in extensive gallery systems the beetles typically bore within the sap-wood of recently dead host
trees. Fungi are fed from a thin layer (= ambrosia gardens) covering the tunnel walls. Apart from bacteria
and yeasts (Haanstad and Norris 1985), this layer is composed of different filamentous fungal species
146
Appendix 1
simultaneously: (i) Auxiliary ambrosia fungi, which appear in associations with different host beetles and
(ii) the primary ambrosia fungus (usually a single species), which is highly beetle-specific (Batra 1966;
Beaver 1989) and cannot survive outside the symbiosis. Most described primary cultivars are from the
asexual, ophiostomatoid genera Ambrosiella and Raffaelea, which are not related, but convergently
evolved the typical ambrosial growth morphology (i.e. large and nutritional conidiospores, which are
grazed by the beetles; (Cassar and Blackwell 1996; Farrell et al. 2001; Rollins et al. 2001; Gebhardt et al.
2004). Characteristic for these species is furthermore the high sensitivity to even short periods of
desiccation (Zimmermann and Butin 1973). Therefore during beetle dispersal, when the adult is searching
for a new breeding substrate, spores of the primary ambrosia fungus get safely stored either within the
gut (Francke-Grosmann 1975) or more commonly within a specialized cuticular pouch, the mycetangium
(Francke-Grosmann 1956).
The fungal community of ambrosia gardens has been described for less than a handful of the 3500
ambrosia beetle species worldwide (Kajimura and Hijii 1992; Harrington and Fraedrich 2010; Endoh et al.
2011; Biedermann et al. 2012). In X. saxesenii only two fungal species are intentionally transmitted to new
galleries by the beetles, but after egg-laying and offspring development their relative abundance in the
fungus gardens decreases (Biedermann et al. 2012). Particular auxiliary species, like Penicillium sp. and
Paecilomyces sp., are associated with (i) old galleries, (ii) waste that is removed from the gallery and (iii)
dead individuals (Biedermann et al. 2012). An experiment showed that X. saxesenii larvae are able to
control the spread of these weeds to some extent (Biedermann and Taborsky 2011). It is unknown
however, if specialized behaviours play a role in this process, for example like fungus tending and weeding
in fungus-farming ants (Mueller et al. 2001). It is likely because several studies have reported an
immediate destruction of fungus gardens by auxiliary molds in response to the removal of beetles
(Schneider-Orelli 1913; Francke-Grosmann 1956; Norris 1993). Some auxiliary fungi may also have an
nutritional value for the beetles. Fusarium-like sp., for example, are common auxiliary species in nests of
fungus-growing ants and ambrosia beetles (Rodrigues et al. 2011), and several studies suggest for
particular species a slightly mutualistic function (Norris 1979; Qi et al. 2011; Biedermann et al. 2012). A
first step to determine whether a specific fungus has a mutualistic, commensal or pathogenic for the
beetles in the symbiosis is to document how fungus abundance relates to beetle numbers and their
behaviours. Up to now, however, no study has investigated if adult beetles and their larvae alter their
behaviours depending on the composition of the fungus gardens they experience.
Here we used artificial observation tubes that contained entire colonies of reproducing X. affinis beetles
and their fungus gardens (i) to determine the social system of an ambrosia beetle breeding in branchingtunnel galleries, which may constrain social interactions, (ii) to describe the fungal community of gardens
and relate it to beetle’s breeding success and behaviours. In particular we ask (a) whether offspring
produced in a gallery engage in alloparental brood care and fungus maintenance, (b) whether decisions of
adult females to help, to breed and to disperse relate to the number of potential beneficiaries, the
number of potential competitors and depend on the location within the nest (old vs. freshly excavated
gallery parts), and (c) whether reproduction affects the propensity to engage in cooperative behaviours.
Ambrosia beetle behaviours have been studied in detail for only a single species yet (X. saxesenii;
(Biedermann and Taborsky 2011) and they have not been related to certain gallery compartments and the
fungal flora, which is crucial for understanding behavioural functions in more details and the role of
different symbionts for the beetles. We should expect a primary mutualistic fungus in the genus
Ambrosiella or Raffaelea, as known from other ambrosia beetles (Harrington 2005), to dominate freshly
excavated gallery parts, where probably most developing brood is located. Feeding and brood care
behaviours might dominate there, whereas hygienic gallery maintenance, gallery protection and courtship
might be more frequent in the older entrance part of the nest because of higher expected abundances of
auxiliary weeds. Artificial observation tubes (Biedermann et al., 2009) allowed us targeted fungal analyses
of gallery compartments of different age, as well as the exact tracking of brood, adult numbers and their
behaviours.
147
Appendix 1
Materials & Methods
Study species
The sugar-cane borer Xyleborus affinis (Xyleborini, Scolytinae, Curcilionidae) is one of the most common
ambrosia beetles in tropical and sub-tropical zones worldwide. Best known for its damages in sugar-cane
plantations it also has been reported to attack more than 248 other woody plant species (Schedl 1962).
Depending on the host species, the available woody material and its resistance to degradation single
plants may be settled by the beetles for one or several generations. A gallery is always founded by a single
female that is attracted by the volatiles (mostly ethanol) of fairly stressed or recently dead plants. While
boring an entrance tunnel perpendicular into the wood she will intentionally inseminate its tunnel walls
with spores of an unknown ambrosia fungus (possibly Cephalosporium pallidum (Verrall 1943)) she carries
in her oral mycetangia (Francke-Grosmann and Schedl 1960; Schneider 1987). If successful, layers (=
gardens) of this fungus will appear on the walls and the foundress starts to lay eggs. By feeding solely on
the fungi larvae pass through three instars followed by pupation (Biedermann and Taborsky 2012). When
reaching adulthood many daughters will not disperse immediately but remain in the natal nest where they
probably help in gallery expansion, brood and fungus care, and may overtake breeding after the foundress
has died (Biedermann et al. 2011). Thus gallery tunnels with an initial length of about 20-30 cm after one
generation may expand over 6 m (within 4 years) by the work of several consecutive overlapping
generations (Schneider 1987). Staying and consecutive breeding within one gallery is possible because X.
affinis (i) produces its own fungal food and (ii) is an inbreeding species with regular brother-sister mating
and a haplodiploid sex determination system. Haplodiploidy enables founder females to assign optimal
brood sex-ratios (by laying unfertilized eggs that develop into males and fertilized eggs that develop into
females), which are strongly female biased. Ninety-five laboratory galleries have produced on average
19.4 ♀ (range 0 – 86) and 0.96 ♂ (range 0 - 6) offspring in the first generation (Biedermann et al. 2009),
and a field study has reported a mean sex ratio of ♀ : ♂ = 41.6 : 1 from 15 galleries (Schedl 1962). Eightyfive percent of laboratory galleries contain only a single male, which is hatching first and obviously capable
to fertilize a huge number of sisters (Roeper et al. 1980). Males are flightless and may disperse only by
food, what they do after offspring matured (Biedermann, unpublished data).
Beetle collection, laboratory breeding and phenology
Females used for breeding in this study were directly collected from oak logs in Pineville, LA, USA (123 ft
asl; 31°20’, 92°24’) in June 2007. Carried in sterile glass vials, they were immediately brought to the
laboratory and used for artificial rearing: Females were surface-sterilized (by washing them first for a few
seconds with 95% ethanol and then with distilled water) and afterwards singly placed on prepared sterile
artificial medium in separate glass tubes (18 mm diameter × 150 mm length; Bellco Glass, Vineland, NJ,
USA) that were covered by sterile plastic caps (Bellco Glass kap-uts, Vineland, NJ, USA). Artificial medium
had been rested for 5 days and consisted of an autoclaved mixture of 0.35 g streptomycin, 1 g Wesson’s
salt mixture, 5 g yeast, 5 g casein, 5 g starch, 10 g sucrose, 20 g agar, 75 g oak tree sawdust (freshly
grounded and oven dried), 2.5 ml wheat germ oil, 5 ml 95% ethanol and 500 ml deionized water (for
details on the preparation see Biedermann et al. (2009)). After introduction of the beetles, tubes were
stored at room temperature (∼23°C) in darkness (wrapped in paper, but light could shine on the
entrance). This way, females will start digging tunnels as if in wood, which they often build adjacent to the
glass of the tube. Behaviour of adults and brood can be observed in those tunnels when the paper is
removed (Biedermann et al. 2009). At 23°C a fungal layer and first eggs appear in successfully founded
galleries around 10 days after gallery foundation. The first adults hatch about a month later and females
remain in the natal nest for at least a week before they disperse by moving onto the surface of the media
and try to fly away (where they can be collected for consecutive breeding). The highest productivity is
reached around 60 days after gallery foundation when the first females start to disperse and offspring of
all stages are present together. However, in the laboratory more offspring develop typically for further 2030 days until the medium dries out and all individuals leave the gallery (Roeper et al. 1980; Biedermann et
al. 2011).
148
Appendix 1
Behavioural observations, gallery dissections and fungal isolations
For our study we used 23 successfully founded galleries. Eight of them were excluded after the
behavioural observations because they were used in another study. Tunnels were only partly visible in
another seven galleries, which were not used for the behavioural observations and just for the dissections
and fungal isolations. In summary, we observed the behaviours of individuals in 16 galleries, dissected and
isolated fungi of 15 galleries, and could relate behaviours with fungal isolations in eight of them.
We started our experiment at about the peak of gallery productivity, when our study galleries ranged
between 51 and 60 days of age (after gallery foundation), before females had started to disperse. In the
three days before the treatment we conducted daily behavioural observations of all visible individuals. For
each individual we noted the developmental stage and sex (larva, adult female, adult male), its location
within the gallery (main-tunnel, side-tunnel, brood-tunnel; see Fig. 1) and its behaviour at the moment of
observation. We decided between the larval behaviours allogrooming, cropping, cannibalism, locomotion,
being pushed by an adult female and inactivity, and the adult behaviours shuffling, blocking, digging,
allogrooming, self-grooming, cropping fungus, cannibalism, locomotion, pushing larva, inactivity and the
male mating (attempt) (for details see Table S1).
On the fourth day we brought galleries individually to a sterile bench, carefully broke the glass of the
tubes, and took 12 fungus samples each from the main-tunnel, the side-tunnel and the brood-tunnel with
a forceps that was flame-sterilized after each sample. Four samples were placed on malt agar plates (MA:
25 g malt extract, 20 g agar, 1 l deionized H2O), four samples on cycloheximide-streptomycin malt agar
plates (CSMA: 10 g malt extract, 15 g agar, 20 ml filter sterilized CSMA stock solution containing 2 mg of
cycloheximidin and 1 mg streptomycin, 1 L deionized H2O), and the last four on benomyl malt agar plates
(BMA: 10 g malt extract, 15 g agar, 1 ml filter sterilized BMA stock solution containing 1 mg benomyl
dissolved in dimethylsulfoxid, 1 L dionized H2O). Samples were spotted in the centre of the plates.
Afterwards we fully dissected the whole gallery system and counted all eggs, larvae, pupae, adult females
and males.
Identification of filamentous fungus isolates
Isolated fungi were initially placed into groups based on cultural colony characteristics (i.e. morphology
and color of mycelium and fruiting structures). Representative samples were used for DNA sequencing (for
details follow the protocol in Biedermann et al. (2012)).
Statistical analyses
Using all 12 samples taken from each location (main-, side-, brood-tunnel) of the laboratory gallery
(independent variable) we estimated the frequency and presence (yes/no) of each fungal species
(dependent variables) by controlling for medium (MA, CSMA, BMA; fixed factors) and gallery of origin
(random factor). We also analysed how number of eggs, larvae and adult females (dependent variable)
were affected by the location within the gallery (fixed factor) and the presence of the different fungi (fixed
factor) by controlling for gallery of origin (random factor). Differences in the frequency of male
observations between locations within the gallery were tested using Fisher’s exact test because of the
small number of males observed. In a third series of models we analysed whether frequencies of larval
and adult behaviours (dependent variable) were influenced by the location within the gallery (fixed
factor). Additionally, we first determined whether the frequency of major larval and adult behaviours
(dependent variable) were affected by the presence of the different fungi (fixed factor) by controlling for
location within the gallery (fixed factor) and gallery of origin (random factor) and second whether each
fungal frequency (dependent variable) was affected by the frequency of larval and adult fungus cropping
and adult shuffling behaviour (fixed factors) and gallery of origin (random factor). All these analyses were
done using generalized estimating equations (GEEs) in R (lmer; Version 2.12.1; R Development Core Team,
2008). GEEs are an extension of generalized linear models with an exchangeable correlation structure of
the response variable within a cluster, which allows for controlling the variation between observations
from a single gallery. This was necessary because of variation in sample sizes between galleries as a few
plates had to be excluded from the analyses (they did not yield any microbial growth). Finally, we used
GLMs to model how the proportion of egg-laying and dispersing females and females with developed
149
Appendix 1
ovaries (dependent variables) are affected by the number of brood and of other females (fixed factors)
within the nest.
Results
Overlapping generations and factors influencing reproductive division of labour
Eight experimental galleries at an age of about 60 days, shortly after the first generation of offspring had
reached adulthood and started to disperse, contained on average 33.6 (± 10 se) eggs, 7.1 (± 2.2) larvae,
5.4 (± 3) pupae, 0.3 (± 0.3) immature females, 6.8 (± 0.9) adult females and 0.9 (± 0.1) adult males. Of the
adult females a mean of 26.8% (± 9.7) had non-developed ovaries, 19.2% (± 7.9) had developed ovaries,
and 54.1% (± 13.3) were laying eggs (Table S9). All 14 dispersing adult females that were dissected had
either non-developed (N = 9) or developed ovaries (N = 5).
Egg-laying was unequally distributed among females and there was no fixed proportion of females laying
eggs per gallery: First, egg numbers were independent of the number of potential egg-layers (= females
with developed ovaries and egg-laying females) (GLM: p > 0.05) and second the proportion of egg-laying
females correlated negatively with number of females with non-developed ovaries (GLM: p = 0.007; Table
S7) and developed ovaries (p = 0.009). Proportion of females with developed ovaries tended to be
negatively correlated with the number of egg-layers (p=0.054), which might suggest that development of
ovaries is triggered by the opportunity to breed. The propensity of females to disperse was reduced with
increasing numbers of larvae dependent on brood care (p = 0.019) and females with non-developed
ovaries (p = 0.028), but was enhanced by increasing numbers of females with developed ovaries (p =
0.011). Interestingly, we found a trend for adult females with non-developed ovaries (N = 11) to engage
less likely in mutually beneficial behaviours (only allogrooming, cropping and shuffling were observed)
immediately before galleries were dissected, than females potentially laying eggs (Fisher’s exact test: p =
0.085, N = 21). Additionally, potential egg-layers tended to be found more frequently close to the brood
(i.e. in the side-tunnel) than in the main-tunnel compared to females with non-developed ovaries (Chi2test: χ2 = 2.94, df = 1, p = 0.087, N = 93)
Offspring and adult numbers and their behaviours in relation to the three gallery compartments
Eggs and pupae were only found in the brood-tunnels of the galleries. Larvae were most commonly found
in the brood-tunnels, followed by the side-tunnels (GEE: p = 0.03; Table S3, Fig 3), and the main tunnel (p
< 0.001). When present in the main-tunnel they only showed locomotion, which was also often observed
in the side-tunnel, but only rarely in the brood-tunnel (p < 0.001; Table S4). Larval allogrooming and
cropping were most commonly detected in the brood-tunnels (p < 0.05). Larval cannibalism was only
present at low rates in the side tunnel.
Adult females and males were least commonly observed in the brood-tunnels (p < 0.05; Table S3, Fig. 4).
Males tended to stay in the side tunnels (p = 0.064). The most frequent adult female behaviours were
shuffling and fungus cropping, with the last most common in the brood-tunnel (p = 0.001; Table S5, Fig. 4).
Inactivity tended to be was most commonly shown in the main-tunnel (p < 0.066). Blocking was, by
definition, only found in the main tunnel, whereas digging, cannibalism and pushing others was never
found there. All other female behaviours were equally common in all three gallery compartments (p >
0.05). Adult males spent their time mainly with cropping (27% of their time), locomotion (25%), inactivity
(16%), followed by attempting to mate (11%), and allogrooming (11%). Self-grooming, mating, being
pushed and shuffling were only rarely shown by males (< 4%).
Fungal isolations
Our isolations revealed five species of filamentous fungi associated with Xyleborus affinis. Two of these
fungi, Mucor sp. and Raffaelea sp., were only morphologically identified, because unfortunately they
could not be revived after cool storage before sequencing. Sequencing revealed the three fungi Fusarium
150
Appendix 1
sp., Phaeoacremonium rubrigenum and Unknown. As expected (Roeper and French 1981), an unknown
Raffaelea sp. was most commonly isolated from X. affinis. Detailed description of sequencing results by
Diana.
Overall, Raffaelea sp., P. rubrigenum and Unknown were more commonly detected on CSMA than MA
(GEEs: p = 0.017-0.001; Table S2), which shows that these three species are resistant to the antibiotic
Cycloheximide. Interestingly they grew also better on BMA than MA (p = 0.007-0.001). The opposite was
found in Mucor sp. and Fusarium sp., which were most frequently isolated from MA followed by BMA and
CSMA (p < 0.01).
Fungal diversity in relation to gallery compartment and age of gallery
The Raffaelea sp. was isolated from all compartments of all galleries, and was most commonly detected in
the brood-tunnel (GEEs: p = 0.003-0.001; Table S2, Fig 2 and S1), which strongly suggests its nutritional
role for X. affinis. Fusarium sp., Mucor sp. and Unknown were present in about 60% of all galleries.
Unknown and Fusarium sp. were isolated equally often from all compartments (p > 0.05), whereas Mucor
sp. and P. rubrigenum occurred more frequently near the entrance than the brood (p = 0.003-0.006).
Fusarium sp. and Mucor sp. were strongly associated with old galleries as well as dead females (i.e.,
isolation rate > 60%; Fig. S1). The other three fungi were detected in < 20% of samples from these
specimens (except Raffaelea sp., which was also present in > 80% of old galleries).
Influence of fungal species on beetle productivity and their farming behaviours
The regular presence of Raffaelea sp. in samples from all compartments of all galleries supports its
essential function for the beetles, but inhibited to test the influence of its presence on beetle productivity
and behaviours. It is likely to be the only fungus with a nutritional values for the brood, because egg and
larval numbers were both not affected by any of the other species (GEEs: p > 0.05; Table S3). In contrary,
adult female numbers were negatively affected by the presence of P. rubrigenum (p = 0.005). Adult
females increased cropping frequency in the presence of Unknown (p = 0.038; Table S6, Fig 5), increased
shuffling by decreasing inactivity levels in the presence of Fusarium sp. (p = 0.048-0.02) and increased
inactivity levels in the presence of P. rubrigenum (p = 0.002). None of the fungal species influenced the
activity of larvae, their cropping and shuffling behaviours (p > 0.05; data not shown).
Influence of larval and adult farming behaviours on the isolation frequency of the fungi
Raffaelea sp. were more commonly detected in galleries, which larvae had higher inactivity levels in the
three days before the isolations (p < 0.05; Table S6). Unknown was positively affected by the larval
cropping frequency (p = 0.004). Adult fungus cropping, shuffling or inactivity levels had no significant
effect on abundance of the five fungi. There was, however, a trend for a higher isolation frequency of
Mucor sp. and P. rubrigenum sp. from galleries, where adults showed higher inactivity rates in the days
before (p < 0.12).
Discussion
Here we report correlative evidence for ambrosia beetles to actively manage their fungus gardens, i.e.
behavioural changes in response to the fungal species composition of their ambrosia gardens. For each of
the three X. affinis gallery compartments – brood-tunnel, side-tunnel, and main-tunnel – we give an
overview of the number of larval and adult inhabitants, their respective behaviours and the diversity of
filamentous fungi. Ambrosia beetle behaviour normally cannot be observed within solid wood, so here we
rear beetles in artificial medium in glass tubes (Biedermann et al. 2009). That gave us the chance for the
first targeted fungal samplings in combination with behavioural observations.
151
Appendix 1
Observations and female dissections revealed primitive eusociality in X. affinis, which was characterized
by adult daughters remaining and helping, with only half of them reproducing in the natal gallery. The
number of egg-laying daughters correlated positively with the total number of adults present in the natal
nest. Interestingly, females that carried eggs in their ovaries tended to engage more frequently in social
behaviours immediately before the dissection. They were cropping their fungal gardens, for example,
which we found to be predominated by one ambrosia fungus, a Raffaelea sp.; it was invariably associated
with galleries of X. affinis. This species formed thick ambrosia layers on the walls of the side- and broodtunnels, which were fed upon by larvae and adults. Behavioural observations suggested Unknown to
function probably as a secondary food source and three other isolates – Fusarium sp., Phaeoacremonium
rubrigenum and Mucor sp. – to have negative effects for the beetles in the symbiosis. Raffaelea sp. was
typically associated with freshly excavated brood-tunnels were individuals preferably fed, whereas the
three proposed pathogens were predominantly found in the main-tunnels, old galleries and associated
with dead individuals. P. rubrigenum had negative effects on the number of adult females present, and
the other three either increased their shuffling (Fusarium sp.) or cropping intensity (Mucor sp., Unknown).
The social structure within branching-tunnel galleries of X. affinis
X. affinis galleries contained high numbers of individuals of all age groups at around day 60, which is
shortly after adult daughters start to disperse from galleries (Biedermann et al. 2011). Multiple egg-layers
were already present and on average about half of the adult daughters that had remained in the natal
nest had started to lay eggs. Although all adult females participate in social tasks in the natal nest, egglaying females were more inclined to do so. This higher investment of egg-laying individuals into brood
care than of non-reproducing helpers is typical for social groups that have not taken the transition to
eusociality. Reproducing females in the primitive eusocial Xyleborinus saxesenii, for example, typically
protect the gallery entrance against the introduction of predators, which is the most risky task in an
ambrosia beetle society (Biedermann and Taborsky 2011). By contrast, after the evolutionary transition to
eusociality, i.e. the evolution of a reproductive and a sterile worker caste, non-egg-layers (=workers) are
not only overtaking the most dangerous task but usually all the work that needs to be done (Wilson 1971;
Bourke 2011).
Larvae are predominantly found in the brood-tunnels, whereas adult females and males prefer to stay in
the side-tunnels. Males are waiting in the side tunnels for freshly emerging females to mate with.
Literature
Batra LR (1963) Ecology of ambrosia fungi and their dissemination by beetles. Trans. Kansas Acad. Sci. 66: 213-236.
Batra LR (1966) Ambrosia fungi: extent of specifity to ambrosia beetles. Science 153: 193-195.
Beaver RA (1989) Insect-fungus relationships in the bark and ambrosia beetles. Insect-fungus interactions (Wilding N,
Collins NM, Hammond PM & Webber JF, eds), pp. 121-143. Academic Press, London.
Biedermann PHW, Klepzig KD, Ott E, Taborsky M & Six DL (2012) Dynamics of filamentous fungi in the ambrosia
gardens of the primitively eusocial beetle Xyleborinus saxesenii Ratzeburg (Scolytinae: Curculionidae). in prep.
Biedermann PHW, Klepzig KD & Taborsky M (2009) Fungus cultivation by ambrosia beetles: behavior and laboratory
breeding success in three xyleborine species. Environmental Entomology 38: 1096-1105.
Biedermann PHW, Klepzig KD & Taborsky M (2011) Costs of delayed dispersal and alloparental care in the funguscultivating ambrosia beetle Xyleborus affinis Eichhoff (Scolytinae: Curculionidae). Behav. Ecol. Sociobiol. 65: 17531761.
Biedermann PHW & Taborsky M (2011) Larval helpers and age polyethism in ambrosia beetles. Proc. Natl Acad. Sci.
USA 108: 17064-17069.
152
Appendix 1
Biedermann PHW & Taborsky M (2012) Social fungus farming varies among ambrosia beetles. in prep.
Bourke AFG (2011) Principles of Social Evolution. Oxford University Press, Oxford.
Cassar S & Blackwell M (1996) Convergent origins of ambrosia fungi. Mycologia 88: 596-601.
De Fine Licht HH & Biedermann PHW (2012) Patterns of functional enzyme activity suggest that larvae are the key to
successful fungus farming by ambrosia beetles. Frontiers in Zoology in review.
Endoh R, Suzuki M, Okada G, Takeuchi Y & Futai K (2011) Fungus Symbionts Colonizing the Galleries of the Ambrosia
Beetle Platypus quercivorus. Microbial Ecology 62: 106-120.
Farrell BD, Sequeira AS, O'Meara BC, Normark BB, Chung JH & Jordal BH (2001) The evolution of agriculture in
beetles (Curculionidae : Scolytinae and Platypodinae). Evolution 55: 2011-2027.
Francke-Grosmann H (1956) Hautdrüsen als Träger der Pilzsymbiose bei Ambrosiakäfern. Z. Morph. u. Ökol. Tiere 45:
275-308.
Francke-Grosmann H (1975) Zur epizoischen und endozoischen Übertragung der symbiotischen Pilze des
Ambrosiakäfers Xyleborus saxeseni (Coleoptera: Scolitidae). Entomologica Germanica 1: 279-292.
Francke-Grosmann H & Schedl W (1960) Ein orales Übertragungsorgan der Nährpilze bei Xyleborus mascarensis Eichh
(Scolytidae). Naturwissenschaften 47: 405.
Gebhardt H, Begerow D & Oberwinkler F (2004) Identification of the ambrosia fungus of Xyleborus monographus and
X. dryographus. Mycol. Progr. 3: 95-102.
Haanstad JO & Norris DM (1985) Microbial symbiotes of the ambrosia beetle Xyletorinus politus. Microbial Ecology
11: 267-276.
Harrington TC (2005) Ecology and evolution of mycophagous bark beetles and their fungal partners. Insect-fungal
associationspp. 257-295. Oxford University Press, New York.
Harrington TC & Fraedrich SW (2010) Quantification of Propagules of the Laurel Wilt Fungus and Other Mycangial
Fungi from the Redbay Ambrosia Beetle, Xyleborus glabratus. Phytopathology 100: 1118-1123.
Hulcr J, Kolarik M & Kirkendall LR (2007) A new record of fungus-beetle symbiosis in Scolytodes bark beetles
(Scolytinae, Curculionidae, Coleoptera). Symbiosis 43: 151-159.
Kajimura H & Hijii N (1992) Dymamics of the fungal symbionts in the gallery system and the mycangia of the
ambrosia beetle, Xylosandrus mutilatus (Blandford) (Coleoptera, Scolytidae). Ecological Research 7: 107-117.
Kent DS & Simpson JA (1992) Eusociality in the beetle Austroplatypus incompertus (Coleoptera: Platypodidae).
Naturwissenschaften 79: 86-87.
Kirkendall LR, Kent DS & Raffa KF (1997) Interactions among males, females and offspring in bark and ambrosia
beetles: the significance of living in tunnels for the evolution of social behavior. The Evolution of Social Behavior in
Insects and Arachnids (Choe JC & Crespi BJ, eds), pp. 181-215. Cambridge University Press.
Klepzig KD & Six DL (2004) Bark beetle-fungal symbiosis: Context dependency in complex associations. Symbiosis 37:
189-205.
Mueller UG, Gerardo NM, Aanen DK, Six DL & Schultz TR (2005) The evolution of agriculture in insects. Annual
Review of Ecology Evolution and Systematics 36: 563-595.
Mueller UG, Schultz TR, Currie CR, Adams RMM & Malloch D (2001) The origin of the attine ant-fungus mutualism. Q.
Rev. Biol. 76: 169-197.
Norris DM (1979) The mutualistic fungi of Xyleborini beetles. Nutrition, Mutualism, and Commensalism (Batra LR,
ed), pp. 53-63. Allanheld, Osmun & Company, Montclair.
Norris DM (1993) Xyleborus ambrosia beetles - a symbiotic ideal extreme biofacies with evolved polyphagous
privileges at monophagous prices. Symbiosis 14: 229-236.
153
Appendix 1
Peer K & Taborsky M (2004) Female ambrosia beetles adjust their offspring sex ratio according to outbreeding
opportunities for their sons. J. Evol. Biol. 17: 257-264.
Peer K & Taborsky M (2005) Outbreeding depression, but no inbreeding depression in haplodiploid ambrosia beetles
with regular sibling mating. Evolution 59: 317-323.
Qi HY, Wang JG, Endoh R, Takeuchi Y, Tarno H & Futai K (2011) Pathogenicity of microorganisms isolated from the
oak platypodid, Platypus quercivorus (Murayama) (Coleoptera: Platypodidae). Applied Entomology and Zoology 46:
201-210.
Rodrigues A, Mueller UG, Ishak HD, Bacci Jr Mc & Pagnocca FC (2011) Ecology of microfungal communities in gardens
of fungus-growing ants (Hymenoptera: Formicidae): a year-long survey of three species of attine ants in Central
Texas. Fems Microbiology Ecology 78: 244-255.
Roeper R, Treeful LM, O'Brien KM, Foote RA & Bunce MA (1980) Life history of the ambrosia beetle Xyleborus affinis
(Coleoptera: Scolytidae) from in vitro culture. Great Lakes Entomologist 13: 141-144.
Roeper RA & French JRJ (1981) Ambrosia fungi of the Western United States and Canada - beetle assocaitions
(Coleoptera: Scolytidae), Tree Hosts, and Distribution. Northwest Science 55: 305-309.
Rollins F, Jones KG, Krokene P, Solheim H & Blackwell M (2001) Phylogeny of asexual fungi associated with bark and
ambrosia beetles. Mycologia 93: 991-996.
Schedl KE (1962) Scolytidae und Platypodidae Afrikas, II+III. Revista de Entomologia de Mocamique 5: 1-1352.
Schneider I (1987) Distribution, fungus-fransfer and gallery construction of the ambrosia beetle Xyleborus affinis in
comparison with X. mascarensis (Coleoptera, Scolytidae). Entomologia Generalis 12: 267-275.
Schneider-Orelli O (1913) Untersuchungen über den pilzzüchtenden Obstbaumborkenkäfer Xyleborus (Anisandrus)
dispar und seinen Nährpilz. Zentralblatt für Bakteriologie, Parasitenkunde, Infektionskrankheiten und Hygiene II 38:
25-110.
Smith A (1776) An Inquiry into the Nature and Causes of the Wealth of Nations. W.Strahan and T.Cadell, London.
Verrall AF (1943) Fungi associated with certain ambrosia beetles. Journal of Agricultural Research 66: 135-144.
Wilson EO (1971) The insect societies. Belknap Press of Harvard University Press, Cambridge.
Zimmermann G & Butin H (1973) Untersuchungen über die Hitze- und Trockenresistenz holzbewohnender Pilze.
Flora 162: 393-419.
154
Appendix 1
Graphs
Fig.1. Morphology of a X. affinis laboratory gallery in artificial medium at about 60 days after its
foundation. Note the species characteristic tunnel system with the four compartments: S – surface, MT –
main tunnel, ST – side tunnel, BT – brood tunnel. A white fungal layer and white larvae are visible within
the ST and BTs. Several adult beetles are sitting in the MT.
Fig. 2. Percentage of fungal species isolated relative to total number of samples taken per gallery
compartment. Mean ± SE; N = 12 samples per compartment; main-tunnel – 8 galleries, side – tunnel – 6
galleries, brood-tunnel – 7 galleries.
155
Appendix 1
Fig. 3. Behaviours of X. affinis larvae in the three gallery compartments. Mean ± se; different letters
denote significant differences (p < 0.05; for exact values see Table 3) between compartments. The mean
(± se) numbers of larvae observed in the three compartments are given in the box on the right.
Fig. 4. Behaviours of X. affinis adult females in the three gallery compartments. Mean ± se; different
letters denote significant differences (p < 0.05; for exact values see Table 4) between compartments. The
mean (± se) numbers of adult females observed in the three compartments are given in the box on the
right.
156
Appendix 1
Fig. 5. Effect of the presence of the five different fungi on the frequency of adult female cropping. Mean (±
se) frequencies, for fungus present or not per gallery, are given. Raffaelea sp. A was always present. * - p <
0.05 (for exact values see Table 5).
157
Appendix 1
Supplementary Information
Table S1. Ethogram of the observed behaviours of X. affinis larvae (L), females (F), and males (M)
(modified from Biedermann and Taborsky, 2011).
Behaviour
Digging
Cropping
Shown by
F
L, F, M
Shuffling
Cannibalism
F, M
L, F, M
Allogrooming
L, F, M
Blocking
F
Pushing others
Self-grooming
Inactive
Locomotion
F
F, M
L, F, M
L, F, M
Being pushed
Mating (attempt)
L, M
M
Definition
excavating new tunnels
grazing on the fungal layer covering gallery walls with
the maxillae and/or mandibles
moving frass and sawdust with the legs and elytra
feeding on a larva, pupa or adult beetle that is usually
dead
grooming an egg, larva, pupa or adult beetle with the
mouthparts (i.e., maxillae, labium)
staying inactive in the entrance tunnel and plugging it
with the body (abdomen directed to the outside)
shifting a larva or male within a tunnel
grooming oneself with the legs
being inactive without moving
creeping (L), or walking on the tibia with back-folded
tarsi (F, M)
getting shifted by an adult female
mounting a female or copulating with her
158
Mutual benefit
gallery extension
fungus care (?)
hygiene
hygiene (?)
brood care, hygiene
protection
protection
-
Appendix 1
Table S2. Influence of culture media and origin of sample (gallery compartment) upon the abundance of
fungi associated with Xyleborus affinis.
Fungal species
Parameter
coeff. ± se
z
p
Mucor sp.
Intercept (BMA)
Contrast BMA vs. MA
Contrast BMA vs. CSMA
Contrast MA vs. CSMA
-2.16 ± 0.86
-0.23 ± 0.51
-1.56 ± 0.61
-1.34 ± 0.42
-2.2
-0.44
-2.59
-3.22
0.012
0.66
0.01
0.001
Intercept (main tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
-1.88 ± 0.74
-0.41 ± 0.38
-1.78 ± 0.46
-1.37 ± 0.46
-2.55
-1.08
-3.83
-3.1
0.011
0.28
<0.001
0.003
Intercept (BMA)
Contrast BMA vs. MA
Contrast BMA vs. CSMA
Contrast MA vs. CSMA
1.19 ± 0.38
-1.49 ± 0.35
0.08 ± 0.39
1.57 ± 0.3
3.1
-4.27
0.21
5.26
0.002
<0.001
0.83
<0.001
Intercept (main tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
0.36 ± 0.28
-0.23 ± 0.28
0.44 ± 0.29
0.67 ± 0.29
1.26
-0.82
1.55
2.35
0.21
0.41
0.12
0.019
Intercept (BMA)
Contrast BMA vs. MA
Contrast BMA vs. CSMA
Contrast MA vs. CSMA
-2.03 ± 0.73
1.25 ± 0.41
-1.38 ± 0.53
-2.63 ± 0.5
-2.76
3.04
-2.6
-5.24
0.006
0.002
0.009
<0.001
Intercept (main tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
-1.3 ± 0.63
-0.36 ± 0.36
-0.25 ± 0.35
0.11 ± 0.36
-2.07
-0.98
-0.7
0.3
0.039
0.33
0.49
0.76
Intercept (BMA)
Contrast BMA vs. MA
Contrast BMA vs. CSMA
Contrast MA vs. CSMA
-2.33 ± 0.52
-1.85 ± 0.68
-0.27 ± 0.63
1.59 ± 0.66
-4.52
-2.71
-0.42
2.39
<0.001
0.007
0.67
0.017
Intercept (main tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
-2.16 ± 0.41
-1.58 ± 0.61
-1.68 ± 0.61
-0.1 ± 0.76
-5.29
-2.59
-2.75
-0.13
<0.001
0.01
0.006
0.9
Intercept (BMA)
Contrast BMA vs. MA
Contrast BMA vs. CSMA
Contrast MA vs. CSMA
-4.67 ± 0.87
-5.37
Only present on BMA
3.11 ± 0.84
3.73
Only present on CSMA
<0.001
Raffaelea sp. A
Fusarium sp.
Phaeoacremonium rubrigenum
Unknown fungus
<0.001
Intercept (main tunnel)
-3.74 ± 0.63
-5.94
<0.001
Contrast main vs. side tunnel
0.07 ± 0.58
0.12
0.91
Contrast main vs. brood-tunnel
0.82 ± 0.5
1.63
0.1
Contrast side vs. brood-tunnel
0.75 ± 0.52
1.44
0.15
Separate GEE models with an exchangeable correlation structure of the response variable within a cluster (gallery-identity), for examining
differences in the isolation frequency between the culture media (BMA – benomyl-malt agar vs. MA – malt-extract agar vs. CSMA – cycloheximidestreptomycin-malt agar) and locations within the gallery (main tunnel vs. side tunnel vs. brood tunnel). Model coefficients are reported as coeff. ±
se (standard error of the estimate), with the group in brackets in the first row of the model as the reference category (coefficient set to zero). The
influences of independent variables on the fungal frequencies are displayed as contrasts between classes to give a better understanding. A
positive contrast denotes that the mean of the second class is higher than the mean of the first class; a negative contrast denotes the reverse.
159
Appendix 1
Fig. S1. Presence of fungal species in the three gallery compartments, old galleries and associated with
dead adult females. Twelve samples were taken from each compartment and from old galleries; four
samples from dead females. Presence was recorded as yes or no.
160
Appendix 1
Table S3. Influence of gallery compartment and presence of the different fungi upon the number of
inhabitants.
Offspring numbers
Parameter
coeff. ± se
z
p
Eggs
Intercept
Presence of Mucor sp.
Presence of Raffaelea sp. A
Presence of Fusarium sp.
Presence of P. rubrigenum
Presence of Unknown fungus
-15.2 ± 999
-1.07 ± 1.03
always present
19.03 ± 999
-1.18 ± 1.15
0.17 ± 0.69
0.0
-1.04
1
0.36
0
-1.02
0.25
1
0.36
0.82
Intercept (main tunnel)
Contrast main vs. side-tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
Presence of Mucor sp.
Presence of Raffaelea sp. A
Presence of Fusarium sp.
Presence of P. rubrigenum
Presence of Unknown fungus
-1.26 ± 0.86
2.57 ± 0.79
3.55 ± 0.86
0.98 ± 0.45
-0.64 ± 0.45
always present
-0.15 ± 0.35
0.07 ± 0.56
-0.04 ± 0.54
-1.46
3.24
4.15
2.18
-1.42
0.14
0.001
<0.001
0.03
0.16
-0.43
0.12
-0.08
0.67
0.91
0.94
Intercept (main tunnel)
Contrast main vs. side-tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
Presence of Mucor sp.
Presence of Raffaelea sp. A
Presence of Fusarium sp.
Presence of P. rubrigenum
Presence of Unknown fungus
2.8 ± 0.32
0.16 ± 0.18
-0.41 ± 0.21
-0.57 ± 0.21
-0.13 ± 0.23
always present
0.08 ± 0.22
-0.58 ± 0.21
-0.25 ± 0.24
8.77
0.9
-1.93
-2.71
-0.54
<0.001
0.37
0.054
0.007
0.59
0.39
-2.8
-1.02
0.69
0.005
0.31
Larvae
Adult ♀♀
Adult ♂♂
Fisher’s exact test (N = 17 observations)
Contrast main (N = 3) vs. side-tunnel (N = 12)
0.064
Contrast main (N = 3) vs. brood-tunnel (N = 2)
1
Contrast side (N = 12) vs. brood-tunnel (N = 2)
0.026
Separate GEE models with an exchangeable correlation structure of the response variable within a cluster (gallery-identity), for examining
differences in the number of eggs, larvae, adult females and males between the three gallery compartments and depending on the presence of
the five fungal species. Eggs were only found in the brood-tunnel. Model coefficients are reported as coeff. ± se (standard error of the estimate),
with the group in brackets in the first row of the model as the reference category (coefficient set to zero). The influences of independent variables
on the fungal frequencies are displayed as contrasts between classes to give a better understanding. A positive contrast denotes that the mean of
the second class is higher than the mean of the first class; a negative contrast denotes the reverse.
161
Appendix 1
Table S4. Influence of gallery compartment upon the behaviour of larvae.
Proportion of larvae
Allogrooming
Cropping
1
1
Cannibalism
Inactivity
1
1
Locomotion
Parameter
coeff. ± se
z
p
Intercept (side-tunnel)
Contrast side vs. brood-tunnel
-2.51 ± 0.53
1.23 ± 0.56
-4.76
2.2
<0.001
0.028
Intercept (side-tunnel)
Contrast side vs. brood-tunnel
-1.57 ± 0.39
1.02 ± 0.41
-4.06
2.47
<0.001
0.013
Intercept (side-tunnel)
Contrast side vs. brood-tunnel
-3.93 ± 1.01
-3.89
Only present in side-tunnel
<0.001
Intercept (side-tunnel)
Contrast side vs. brood-tunnel
-2.25 ± 0.68
0.79 ± 0.59
0.001
0.18
Intercept (brood-tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
-2.07 ± 0.37
-5.53
Only behaviour in main-tunnel
Only behaviour in main-tunnel
-2.17 ± 0.41
-5.26
1
-3.29
1.34
<0.001
<0.001
Intercept (brood-tunnel)
-4.61 ± 0.83
-5.56
<0.001
Contrast side vs. brood-tunnel
Only present in brood-tunnel
Separate GEE models with an exchangeable correlation structure of the response variable within a cluster (gallery-identity), for examining
differences in behavioural frequencies between the three gallery compartments. Model coefficients are reported as coeff. ± se (standard error of
the estimate), with the group in brackets in the first row of the model as the reference category (coefficient set to zero). The influences of
independent variables are displayed as contrasts between classes to give a better understanding. A positive contrast denotes that the mean of the
second class is higher than the mean of the first class; a negative contrast denotes the reverse.
1
Not present in main-tunnel.
Being pushed by adult
162
Appendix 1
Table S5. Influence of gallery compartment upon the behaviour of adult females.
Proportion of adult ♀♀
Parameter
coeff. ± se
z
p
Shuffling
Intercept (side-tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
-0.6 ± 0.14
-0.42 ± 0.22
-0.04 ± 0.31
0.46 ± 0.3
-4.14
-1.86
-0.12
1.52
<0.001
0.062
0.9
0.13
Blocking
Intercept (main-tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
-2.42 ± 1.32
-1.84
Only present in main-tunnel
Only present in main-tunnel
0.066
Digging
Intercept (side-tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
0.0 ± 1.41
0.0
Only present in side-tunnel
Only present in brood-tunnel
24.6 ± 999
0.0
1
Allogrooming
Intercept (side-tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
-1.5 ± 0.23
-0.11 ± 0.42
0.56 ± 0.48
0.67 ± 0.39
-6.5
-0.27
1.17
1.7
<0.001
0.79
0.24
0.089
Self-grooming
Intercept (side-tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
-1.83 ± 0.36
-0.51 ± 0.48
0.22 ± 1.2
0.73 ± 1.21
-5.09
-1.05
0.19
0.6
<0.001
0.3
0.85
0.55
Cropping fungus
Intercept (main-tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
-1.18 ± 0.29
0.19 ± 0.33
1.18 ± 0.36
0.99 ± 0.27
-4.12
0.57
3.23
3.57
<0.001
0.57
0.001
<0.001
Cannibalism
Intercept (brood-tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
-1.39 ± 0.79
-1.75
Only present in side-tunnel
Only present in brood-tunnel
-0.29 ± 1.14
-0.25
0.08
Inactivity
Intercept (side-tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
-1.13 ± 0.23
-0.89 ± 0.3
-0.77 ± 0.42
0.12 ± 0.43
-4.88
-3.0
-1.84
0.27
<0.001
0.003
0.066
0.78
Locomotion
Intercept (brood-tunnel)
Contrast main vs. side tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
0.0 ± 1.41
-0.23 ± 0.32
0.77 ± 1.44
1.0 ± 1.42
0
-0.7
0.53
0.7
1
0.48
0.59
0.48
1
0.8
Intercept (brood-tunnel)
-1.61 ± 1.1
-1.47
0.14
Contrast main vs. side tunnel
Only present in side-tunnel
Contrast main vs. brood-tunnel
Only present in brood-tunnel
Contrast side vs. brood-tunnel
-0.54 ± 1.19
-0.46
0.65
Separate GEE models with an exchangeable correlation structure of the response variable within a cluster (gallery-identity), for examining
differences in behavioural frequencies between the three gallery compartments. Model coefficients are reported as coeff. ± se (standard error of
the estimate), with the group in brackets in the first row of the model as the reference category (coefficient set to zero). The influences of
independent variables are displayed as contrasts between classes to give a better understanding. A positive contrast denotes that the mean of the
second class is higher than the mean of the first class; a negative contrast denotes the reverse.
Pushing others
163
Appendix 1
Table S6. Influence of gallery compartment and presence of the different fungi upon the frequency of
particular adult female behaviours.
Proportion of adult ♀♀
Parameter
coeff. ± se
z
p
Cropping
Intercept (main tunnel)
Contrast main vs. side-tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
Presence of Mucor sp.
Presence of Raffaelea sp. A
Presence of Fusarium sp.
Presence of P. rubrigenum
Presence of Unknown fungus
-2.77 ± 0.52
1.05 ± 0.45
1.49 ± 0.46
0.45 ± 0.42
0.68 ± 0.39
always present
0.22 ± 0.38
-0.41 ± 0.43
0.94 ± 0.46
-5.33
2.34
3.21
1.06
1.73
<0.001
0.019
0.001
0.29
0.084
0.59
-0.95
2.07
0.56
0.34
0.038
Intercept (main tunnel)
Contrast main vs. side-tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
Presence of Mucor sp.
Presence of Raffaelea sp. A
Presence of Fusarium sp.
Presence of P. rubrigenum
Presence of Unknown fungus
-0.9 ± 0.42
-0.63 ± 0.34
-0.68 ± 0.39
-0.04 ± 0.42
0.44 ± 0.37
always present
0.71 ± 0.36
-0.11 ± 0.31
0.27 ± 0.37
-2.13
-1.87
-1.74
-0.11
1.2
0.033
0.061
0.082
0.92
0.23
1.98
-0.35
0.72
0.048
0.72
0.47
Intercept (main tunnel)
Contrast main vs. side-tunnel
Contrast main vs. brood-tunnel
Contrast side vs. brood-tunnel
Presence of Mucor sp.
Presence of Raffaelea sp. A
Presence of Fusarium sp.
Presence of P. rubrigenum
Presence of Unknown fungus
-1.13 ± 0.79
0.07 ± 0.51
-0.52 ± 0.61
-0.59 ± 0.67
-0.47 ± 0.64
always present
-1.38 ± 0.6
1.79 ± 0.57
-0.51 ± 0.65
-1.43
0.14
-0.85
-0.88
-0.73
0.15
0.89
0.39
0.38
0.47
-2.32
3.14
-0.79
0.02
0.002
0.43
Shuffling
Inactive
Separate GEE models with an exchangeable correlation structure of the response variable within a cluster (gallery-identity), for examining
differences in behavioural frequencies between the three gallery compartments and the five fungal species. Model coefficients are reported as
coeff. ± se (standard error of the estimate), with the group in brackets in the first row of the model as the reference category (coefficient set to
zero). The influences of independent variables on the behavioural frequencies are displayed as contrasts between classes to give a better
understanding. A positive contrast denotes that the mean of the second class is higher than the mean of the first class; a negative contrast
denotes the reverse.
164
Appendix 1
Table S7. Influence of the frequency of specific larval and adult female behaviours upon the frequency of
the different fungal species.
Fungal abundances
Parameter
coeff. ± se
z
p
Mucor sp.
Intercept
Cropping larvae
Inactive larvae
-1.9 ± 0.56
0.49 ± 0.96
-1.1 ± 1.51
-3.4
0.51
-0.73
<0.001
0.61
0.47
Intercept
Cropping adult females
Inactive adult females
Shuffling adult females
-2.64 ± 0.98
1.48 ± 1.29
2.56 ± 1.4
-1.1 ± 1.23
-2.7
1.15
1.83
-0.9
0.007
0.25
0.067
0.37
Intercept
Cropping larvae
Inactive larvae
0.24 ± 0.29
-0.7 ± 0.65
2.3 ± 0.95
0.84
-1.08
2.42
0.4
0.28
0.016
Intercept
Cropping adult females
Inactive adult females
Shuffling adult females
-0.3 ± 0.61
0.42 ± 0.94
0.97 ± 1.09
0.91 ± 0.89
-0.5
0.45
0.89
1.03
0.62
0.66
0.37
0.3
Intercept
Cropping larvae
Inactive larvae
-1.08 ± 0.81
-0.45 ± 0.82
-2.74 ± 2.3
-1.34
-0.55
-1.19
0.18
0.58
0.23
Intercept
Cropping adult females
Inactive adult females
Shuffling adult females
-2.3 ± 1.23
0.68 ± 1.28
0.85 ± 2.02
0.91 ± 1.42
-1.88
0.53
0.42
0.64
0.06
0.6
0.67
0.52
Intercept
Cropping larvae
Inactive larvae
-3.51 ± 0.9
1.71 ± 1.48
-3.09 ± 4.78
-3.91
1.16
-0.65
<0.001
0.25
0.52
Intercept
Cropping adult females
Inactive adult females
Shuffling adult females
-4.04 ± 1.32
2.01 ± 1.97
3.18 ± 2.03
-0.08 ± 1.88
-3.06
1.02
1.57
-0.04
0.002
0.31
0.12
0.97
Intercept
Cropping larvae
Inactive larvae
-2.36 ± 0.48
2.36 ± 0.82
-4.36 ± 3.49
-4.91
2.86
-1.25
<0.001
0.004
0.21
Raffaelea sp.
Fusarium sp.
Phaeoacremonium rubrigenum
Unknown fungus
Intercept
-2.66 ± 1.06
-2.52
0.012
Cropping adult females
0.77 ± 1.54
0.5
0.62
Inactive adult females
-2.2 ± 2.46
-0.9
0.37
Shuffling adult females
1.67 ± 1.46
1.15
0.25
Separate GEE models with an exchangeable correlation structure of the response variable within a cluster (gallery-identity), for examining
differences in fungal frequencies per gallery depending on the frequency of inactivity and of potential fungus cleaning behaviours (cropping and
shuffling). Model coefficients are reported as coeff. ± se (standard error of the estimate).
165
Appendix 1
Table S8. Separate GLM models to examine factors influencing adult females to lay eggs, develop ovaries
or disperse.
Proportion of adult ♀♀
Parameter
coeff. ± se
z
p
Egg-laying
Intercept
Number of immature offspring*
Number of ♀♀, non-developed ovaries
Number of ♀♀, developed ovaries
Number of ♀♀*
2.9 ± 1.0
-0.02 ± 0.03
-0.82 ± 0.3
-0.57 ± 0.22
0.29 ± 0.47
2.89
-0.79
-2.72
-2.62
0.61
0.004
0.43
0.007
0.009
0.54
Developed ovaries
Intercept
Number of immature offspring*
Number of ♀♀, egg-laying
Number of ♀♀, non-developed ovaries
Number of ♀♀
-3.74 ± 2.47
0.02 ± 0.02
-0.5 ± 0.26
-0.59 ± 0.24
0.69 ± 0.29
-1.52
0.75
-1.93
-2.44
2.38
0.13
0.46
0.054
0.015
0.018
Intercept
0.49 ± 0.48
Number of eggs*
-0.01 ± 0.01
Number of larvae
-0.24 ± 0.1
Number of ♀♀, non-developed ovaries
-0.26 ± 0.12
Number of ♀♀, developed ovaries
0.48 ± 0.19
Number of ♀♀*
-0.1 ± 0.41
Model coefficients are reported as coeff. ± se (standard error of the estimate).
* Variable not in the final model after step-wise model reduction using ANOVA analysis of loglikelihood scores.
1.03
-0.86
-2.34
-2.19
2.53
-0.24
0.31
0.39
0.019
0.028
0.011
0.81
Dispersing
Table S9. Composition of X. affinis laboratory galleries about 60 days after gallery foundation.
Adult ♀♀
Immature stages
Gallery
A
Eggs
18
1st
instar
1
2nd/3rd
instar
4
B
42
1
-
-
-
1
-
-
4
2
C
20
3
-
1
-
-
-
-
3
5
D
22
8
3
1
2
-
2
6
1
14
E
53
8
5
-
-
1
2
2
5
1
F
19
-
1
5
-
1
3
1
3
8
G
92
5
-
-
-
1
-
1
4
3
H
3
-
-
12
-
1
4
-
3
-
Pupae
24
Teneral
♀♀
-
Adult
♂♂
1
non-developed
ovaries
7
developed
ovaries
3
egglaying
-
dispersing
3
166
Appendix 2
Mechanisms of fungus gardening in ambrosia beetles
Peter H.W. Biedermann1,2, Michael Taborsky1 and Cameron R. Currie2
1
Department of Behavioural Ecology, Institute of Ecology and Evolution, University of Bern, Baltzerstrasse
6, CH-3012 Bern, Switzerland
2
Department of Bacteriology, University of Wisconsin-Madison, Madison, USA
Corresponding author:
Peter H.W. Biedermann
Baltzerstrasse 6, CH-3012 Bern, Switzerland
phone: 0041 31 631 3015
e-mail: [email protected]
Manuscript in work
Introduction
The most elaborate of insect–fungus interactions is advanced fungus farming, in beetles called ambrosia
symbiosis. This relationship evolved repeatedly in beetles, at least eight times in weevils, and ambrosia
beetle is an ecological classification, not a phylogenetic group (Hulcr and Dunn 2011). In comparison to
bark beetles they differ in how they colonize trees and rely on fungi for food. Unlike bark beetles, which
live under the bark where they feed on fungus infested tree tissues, ambrosia beetles build tunnel
systems (galleries) deep within the sapwood, and they actively cultivate mutualistic fungi in “ambrosia
gardens” on the walls of these galleries as their sole food source (Francke-Grosmann 1967; Norris 1979;
Beaver 1989). Primary symbionts, so called ambrosia fungi in the two anamorph genera Ambrosiella and
Raffaelea (Ascomycota) are dominating these gardens (Beaver 1989; Farrell et al. 2001). Both genera
produce asexual fruiting structures (conidiophores and – spores) in the presence of the beetles. This so
called “ambrosial growth” forms large mats on the tunnel walls and is essential for beetle nutrition.
Convergent fruiting is also known from the cultivars of farming ants (= gongolydia; Bass and Cherrett
1996) and termites (= nodules ; Aanen et al. 2002), but in none of these systems it is fully understood how
this specialized growth is triggered by the insects.
Studies on Xyleborinus saxesenii and Xyleborus affinis (Chapter 1-6, Appendix 1) showed that in ambrosia
beetles larval and adult offspring of a single foundress cooperate in brood care, gallery maintenance, and
fungus gardening. The latter is termed cropping and is one of the most common behaviours of adult
females within galleries. Its role for the health and productivity of the fungus gardens is unknown. In a
descriptive study it had been mentioned, however, that pure cultures of Ambrosiella hartigii, the
mutualistic fungus of the ambrosia beetle Anisandrus dispar, produce fruiting structures, after they have
been in contact with adult females (see Fig. 1 from Batra 1967). This so-called ambrosial growth is
ubiquitous in ambrosia beetle galleries, but seems to occur only in the presence of the beetles.
In this study we aimed at (i) quantifying Batra’s (1967) results experimentally and (ii) testing their
general validity for ambrosia beetles. In particular we tested (a) if ambrosial growth can be induced by
adult females of four different ambrosia beetle species (Cnestus mutilatus, Anisandrus sayi, Xylosandrus
germanus, Xyleborinus saxesenii) in pure cultures of their mutualists, which we had isolated from their
mycetangia before. Additionally we also tested (b) if it can also be induced by larvae and pupae, and (c)
what are the potential underlying triggers for this phenomenon (e.g., mechanical disturbance, larval frass
and chemical substances).
167
Appendix 2
Material & Methods
Study species
Xyleborinus saxesenii and Xylosandrus germanus are found in the temperate zones worldwide and the
most common Xyleborini in Central Europe (Wood 1982). They differ in their social system and mode of
fungiculture. X. saxesenii is a primitively eusocial species with multiple developing generations within one
nest and larvae that feed on fungus infested wood (Biedermann and Taborsky 2011). X. germanus on the
other hand has only a single generation per nest and larvae that feed predominantly on fungus (Bischoff
2004). This may relate to the different natures of their ambrosia fungi, which are from two distantly
related clades of ascomycete fungi in which ambrosial growth convergently evolved as a result of beetle
feeding: Ambrosiella hartigii, the symbiont of X. germanus, belongs to the Ceratocystis clade, produces
larger fruiting structures and does not survive the winter in galleries (Zimmermann 1973). It is also
associated with Anisandrus dispar, which had been found by Batra to trigger the fruiting of A. hartigii on
plates in culture (Batra 1967). Raffaelea sulfurea, the symbiont of X. saxesenii, belongs to the Ophiostoma
clade, is cold tolerant and like all Ophiostoma fungi insensitive to the antibiotic cycloheximide, which is
produced by Streptomyces bacteria (Cassar and Blackwell 1996). This might be an adaption to grow in the
presence of such bacteria.
Experimental procedure
We caught adult females of the four different beetle species in ethanol baited traps in the surrounding of
Madison, USA during May 2009. Specimens were brought to the laboratory and surface sterilized. The
mycetangium from five individuals per species was dissected and containing fungal spores were cultured
on plates. Most of these isolations gave pure cultures and were stored in the refrigerator. The rest of the
beetles were put in rearing tubes on artificial medium to start laboratory colonies (Biedermann et al.
2009). This was only successful in X. saxesenii and X. germanus. Larvae, pupae, larval frass and medium (as
a control for larval frass) for the experiment were taken from these lab colonies.
Five days before the start of the experiment we set up pure cultures of the four different fungi and malt
agar plates. Thus, at the start of the experiment fungal mycelium had covered about 2/3 of the plate’s
diameter. Adult females (either from lab cultures or the field), pupae, larvae (when available) and two
potential excretions (larval frass and proline) were applied on pure cultures on agar plates of the four
different ambrosia fungi. Solitary individuals were directly placed onto the fungal mycelium and
subsequently covered with a small plastic cup. In all groups we had a control area, where nothing was
applied. In the X. saxesenii treatment we also controlled for the cup, and placed an empty cup onto the
medium. Sterile artificial medium was used as negative control for the larval frass. Distilled, sterile water
served as a negative control for proline, which was diluted in this water. Proline was tested because a
study on the fungal growth on artificial media suggested that the ambrosial growth form of the fungus in
the mycetangia is attributed to an extreme abundance of the free amino acid proline in the insect’s
haemolymph and body secretions (French and Roeper 1973).
Over a period of 13 days we compared the change of fungus growth to a control area where nothing was
applied. We took daily pictures of the plates and later analysed the amount of ambrosial growth or aerial
mycelium using image analysis software.
Results
Cnestus mutilatus. Ambrosiella beaver, the ambrosia fungus of Cnestus mutilatus produced ambrosial
growth that was randomly distributed on the plate and none of the treatments differed from the control
area where nothing was applied (Fig. 3).
Anisandrus sayi. The unknown ambrosia fungus of Anisandrus sayi did not produce any ambrosial growth
in culture on plates, although fruiting structures were visible within the gallery systems of this beetle in
artificial medium. We probably did not isolate the mutualistic ambrosia fungus of A. sayi.
Xylosandrus germanus. Ambrosiella hartigii, the ambrosia fungus of Xylosandrus germanus, produced
fruiting structures (ambrosial growth) in the presence of adult X. germanus females, pupae, larvae and
larval frass (p < 0.001; Fig. 2). In did not produce any ambrosial growth neither in the control treatment,
when proline or sterile artificial medium was applied nor when mycelium was mechanically disturbed with
a scalpel (p < 0.05). Interestingly, fruiting was also induced by the presence of adults of Anisandrus sayi – a
different ambrosia beetle, but only if the beetles were in direct contact with the fungus.
168
Appendix 2
Xyleborinus saxesenii. Raffaelea sulfurea, the ambrosia fungus of Xyleborinus saxesenii did not produce
any ambrosial growth on plates, but instead produced plenty of aerial mycelium. Analyses revealed larvae
(LME model: p < 0.001), larval frass (p < 0.05) and adult females (p < 0.05) to significantly induce the
growth of aerial mycelium in comparison to a control area, where nothing was applied (Fig. 4). Placing the
empty cup without larvae and adults on the fungus also tended to induce this growth (p = 0.06), however,
which may suggest that the changed gas content or air moisture triggered the fungal pleomorphism.
Taking the empty cup as new control group, only larvae further increased mycelium induction (p < 0.001).
Discussion
In this study we show that some ambrosia beetles are able to induce growth or fruiting in their mutualistic
ambrosia fungus. Only one fungus, Ambrosiella beaveri fruited independent of the presence and activity
of the beetles.
The induction of aerial mycelium by larvae in X. saxesenii and ambrosial growth (i.e., nutritional fruiting
bodies) by larvae, pupae and adult females in X. germanus is an extraordinarily important result for my
thesis, because it shows for the first time that immatures (larvae, pupae) and adults are able to increase
the productivity of their fungus. Hence, the presence of immatures and delayed dispersing adults is likely
to increase the productivity of the fungus, and thus the amount of food available for the whole colony.
It remains unknown what substance(s) and/or behaviour(s) induce the fungal pleomorphism in X.
saxesenii and X. germanus. The different gas content or air humidity within our cups, may resemble aerial
conditions within the tunnels. Unknown X. saxesenii larval secretions or excretions seem to play an
additional role. The induction by non-moving pupae in X. germanus strongly suggests that chemical
substances produced either by the insects or other microbes are involved. Ambrosial growth in galleries is
covered with unstudied bacteria and many microbes can be isolated from the body surface and mouth
parts of adult beetles and larvae (own unpublished data). Most importantly, (i) the application of
unidentified yeasts associated with the ambrosia beetle Xyleborus dispar triggered the ambrosial growth
in its ambrosia fungus in culture (Peklo and Satava 1950), and (ii) the bark beetle associate Leptographium
procerum, a relative of the ambrosia fungi, was strongly induced to produce asexual fruiting structures in
the presence of several bacteria (e.g. Pantoea sp., Pseudomonas sp.; Adams et al. 2009). A positive role of
yeasts in the gardens of ambrosia beetles has been frequently claimed (Webb 1945; Francke-Grosmann
1963; Francke-Grosmann 1967; Beaver 1989), but never tested. In summary, there are strong indications
that symbiotic microbes might be involved in the fruiting of the ambrosia fungi, which would be a unique
new function of mutualistic microbes in fungus cultivating insects.
Literature
Aanen DK, Eggleton P, Rouland-Lefévre C, Guldberg-Frøslev T, Boomsma JJ & Rosendahl S (2002) The evolution of
fungus-growing termites and their mutualistic fungal symbionts. PNAS 99: 14887-14892.
Adams AS, Currie CR, Cardoza Y, Klepzig KD & Raffa KF (2009) Effects of symbiotic bacteria and tree chemistry on the
growth and reproduction of bark beetle fungal symbionts. Canadian Journal of Forest Research-Revue Canadienne de
Recherche Forestiere 39: 1133-1147.
Bass M & Cherrett JM (1996) Leaf-cutting ants (Formicidae, Attini) prune their fungus to increase and direct its
productivity. Functional Ecology 10: 55-61.
Batra LR (1967) Ambrosia fungi - A taxonomic revision and nutritional studies of some species. Mycologia 59: 9761017.
Beaver RA (1989) Insect-fungus relationships in the bark and ambrosia beetles. Insect-fungus interactions (Wilding N,
Collins NM, Hammond PM & Webber JF, eds), pp. 121-143. Academic Press, London.
Biedermann PHW, Klepzig KD & Taborsky M (2009) Fungus cultivation by ambrosia beetles: behavior and laboratory
breeding success in three xyleborine species. Environmental Entomology 38: 1096-1105.
Biedermann PHW & Taborsky M (2011) Larval helpers and age polyethism in ambrosia beetles. Proc. Natl Acad. Sci.
USA 108: 17064-17069.
169
Appendix 2
Bischoff LL (2004) The social structure of the haplodiploid bark beetle, Xylosandrus germanus. Diploma Thesis,
Zoological Institute, University of Berne, Switzerland.
Cassar S & Blackwell M (1996) Convergent origins of ambrosia fungi. Mycologia 88: 596-601.
Farrell BD, Sequeira AS, O'Meara BC, Normark BB, Chung JH & Jordal BH (2001) The evolution of agriculture in
beetles (Curculionidae : Scolytinae and Platypodinae). Evolution 55: 2011-2027.
Francke-Grosmann H (1963) Some new aspects in forest entomology. Annual Review of Entomology 8: 415-438.
Francke-Grosmann H (1967) Ectosymbiosis in wood-inhabiting beetles. Symbiosis (Henry SM, ed), pp. 141-205.
Academic Press, New York.
French JRJ & Roeper RA (1973) Patterns of nitrogen utilization between the ambrosia beetle Xyleborus dispar and its
symbiotic fungus. Journal of Insect Physiology 19: 593-605.
Hulcr J & Dunn RR (2011) The sudden emergence of pathogenicity in insect-fungus symbioses threatens naive forest
ecosystems. Proceedings of the Royal Society B: Biological Sciences.
Norris DM (1979) The mutualistic fungi of Xyleborini beetles. Nutrition, Mutualism, and Commensalism (Batra LR,
ed), pp. 53-63. Allanheld, Osmun & Company, Montclair.
Peklo J & Satava J (1950) Fixation of Free Nitrogen by Insects. Experientia 6: 190-192.
Webb S (1945) Australian ambrosia fungi. Proceedings of the Royal Society of Victoria 57: 57-79.
Wood SL (1982) The bark and ambrosia beetles of North and Central America (Coleoptera: Scolytidae), a taxonomic
monograph. Great Basin Naturalist Memoirs 6: 1-1359.
Zimmermann G (1973) Vergleichende ökologisch-physiologische Untersuchungen an Ambrosiapilzen, Assoziierten
Bläuepilzen und Luftbläuepilzen. Doctoral thesis, Georg-August University Göttingen, Germany.
Graphs
Fig. 1. An agar plate with a pure culture of Ambrosiella hartigii, the mutualistic fungus of Anisandrus
dispar (the mycelial growth form is black). White fruiting structures were induced at locations where adult
females had been crawling before (Figure from Batra 1967).
170
Appendix 2
Fig. 2. Induction of ambrosial growth in Ambrosiella hartigii, the ambrosia fungus of Xylosandrus germanus. A Graph showing the induction of spores and spore-like
structures by different treatments involving different stages of X. germanus and adult females of Anisandrus sayi. Significant differences to the control are marked by ***- p
< 0.001 (LME models); number of tested samples are given in brackets. B Example of the experimental procedure. Individuals and certain substances were confined to /
applied on pure cultures of the ambrosia fungus A. hartigii. Reaction of the fungus is displayed six days after the treatment. Note that mycelial growth is black, while
ambrosial growth is white.
171
Appendix 2
Fig. 3. Ambrosial growth in Ambrosiella beaveri, the ambrosia fungus of Cnestus mutilatus. A Graph showing the number of spores and spore-like structures depending on
adult females and different concentrations of proline (and associated controls) applied on the fungus. There are no significant differences to the control: p > 0.05 (LME
models); number of tested samples are given in brackets. B Picture of A. beaveri and the results of the treatment after six days. Note the dark mycelial growth and the white
ambrosial growth.
172
Appendix 2
Fig. 4. Growth of aerial mycelium in Raffaelea sulphurea, the ambrosia fungus of Xyleborinus saxesenii. A Quantity of aerial mycelium induced by larvae, pupae and adult
female X. saxesenii and other treatments. Significant differences to the control are marked by ***- p < 0.001, * - p < 0.05, (*) – p < 0.1 (LME models); number of tested
samples are given in brackets. B Picture of Raffaelea sulphurea six days after the treatment. Note the darker mycelial growth and the whitish ambrosial growth.
173
Appendix 2
174
Summary & Conclusion
Summary & Conclusion
The most famous farming insects are, without doubt, the attine ants. Rightly so, but they are
not the only ones...
The ancient and highly evolved mutualisms between insects and fungi are a textbook example of
symbiosis (Mueller et al. 2005). Ants, termites and ambrosia beetles independently evolved the
ability to grow fungi for food about 40–60 million years before the origin of human agriculture
(Mueller and Gerardo 2002). Fungus farming evolved once in ants and termites and at least nine
times in beetles (Farrell et al. 2001; Hulcr et al. 2007), with no known reversal to non-farming. The
evolution of fungiculture seems to be always accompanied by a large radiation and a huge ecological
success (Farrell et al. 2001; Mueller et al. 2005). Like in humans, farming by insects can only be
managed in cooperative societies, in which planting, protecting, cultivating and harvesting of the
crops are shared by several individuals. Thus, it is not surprising that ambrosia beetles evolved the
highest sociality among beetles, as one ambrosia beetle is regarded eusocial (Kent and Simpson
1992) and others at least sub-social (Kirkendall et al. 1997).
In this Ph.D. project I aimed to contribute a puzzle piece to the understanding of social evolution and
evolution of symbioses in nature. In particular, I investigated the social system and the mode of
fungiculture focusing on two exemplary species of the ambrosia beetle tribe Xyleborini, the native
fruit-tree pinhole borer Xyleborinus saxesenii and the American sugarcane shot-hole borer Xyleborus
affinis. I assume that they stand for two fundamentally different social systems and modes of fungus
agriculture, because X. saxesenii inhabits cave-like galleries, whereas X. affinis, like most other
ambrosia beetles, digs out branching tunnel systems.
New insights into the social system of xyleborine ambrosia beetles
Until recently, behaviours of ambrosia beetles within their galleries had not been studied, because
they live a hidden life inside the wood, where they are virtually impossible to observe. The only
report on eusociality in ambrosia beetles was not based on behavioural data, but reproductive roles
were inferred by destructive sampling of active nests and the finding that only one Austroplatypus
incompertus female per colony had developed ovaries (Kent and Simpson 1992). An artificial rearing
and observation technique (Saunders and Knoke 1967) was rediscovered by our group (Peer &
Taborsky, 2004; 2005), however, which I was able to adapt for laboratory rearing of several
xyleborine ambrosia-beetle species (Biedermann et al. 2009). Among those, Xyleborinus saxesenii
and Xyleborus affinis were supposed to have a high potential for advanced sociality and fungus
gardening, because (i) field studies had shown that dozens of larvae and adults gregariously live in
galleries that are sometimes settled by several offspring generations (Hosking 1972; Schneider 1987)
and (ii) correlative data suggested that X. saxesenii adult daughters delay dispersal from the natal
nest depending on the number of brood – possibly depending on care (Peer and Taborsky 2007).
The first observation with our laboratory colonies in tubes was unique for holometabolous insects:
larval and adult offspring of a single foundress cooperate in brood care, gallery maintenance, and
fungus gardening, showing a clear division of labour between larval and adult colony members
(Chapter 1 and 6, Appendix 1). Larvae of both species participate in brood care, and in X. saxesenii
larvae also enlarge the gallery and participate in gallery hygiene. The cooperative effort of adult
females in the colony and the timing of their dispersal depend on the number of sibling depending on
brood care (larvae and pupae) and on the number of adult workers in both species. This suggests that
175
Summary & Conclusion
staying and helping in the nest – for gaining indirect fitness – is triggered by demands of brood
dependent on care.
Insects with a metamorphosis between the larval and adult stage (Holometabola; e.g., bees, wasps,
ants, beetles) dramatically reorganize their morphology during ontogenesis, because the two stages
represent distinct developmental and evolvable modules (Yang 2001). In some ambrosia beetles, like
X. saxesenii and some platypodine species, this has apparently led to larvae that are predisposed by
their body morphology and the frequent renewal of mandibles by moulting to assume certain tasks,
like balling of frass and gallery enlargement (Chapter 1). Additionally, X. saxesenii larvae seem to be
capable of producing wood-degrading enzymes (Hemicellulases), which are not found in the bodies
of adults (Chapter 8). By contrast, larvae of solely mycetophagous ambrosia beetles, like X. affinis,
play a minor role for gallery productivity, because the only cooperative behaviour they engage in is
allogrooming (Chapter 6, Appendix 1). The role of solely mycetophagous larvae in the induction of
ambrosial growth has not been tested yet, however (Appendix 2). In summary, larvae in most
ambrosia-beetle species are free to move within the natal nest, and are not confined to small areas
or brood cells like those of most hymenopteran social societies (Wilson 1971; Hölldobler and Wilson
1990). This, in combination with different capabilities, predisposed ambrosia beetles for division of
labour between larval and adult stages. Importance and specific roles of larvae in the galleries appear
to vary between species (Chapter 6).
Males also take part in cooperative behaviours (X. saxesenii; Chapter 1 and 4), which suggests that
relatedness asymmetries caused by haplodiploidy, which would favour female-biased help, are
probably offset by inbreeding in this species (comparable to thrips; Hamilton 1972; Chapman et al.
2000). Nevertheless, because of strong local male competition, there are only one or two males on
ten to eighty females per gallery, and thus their help is of minor importance. The only exception are
male-only galleries (founded by unfertilized females) that develop normally and can produce as much
offspring as normal galleries (Chapter 4).
Adult female offspring exhibit philopatric periods of at least 7 days (X. affinis) / 17 days (X. saxesenii;
Chapter 2 and 5), and some even stay all their life in the natal nest (X. saxesenii: Chapter 3, Peer and
Taborsky 2007). Removal and independent breeding of staying X. affinis females showed that they
are totipotent and fully capable of founding their own nest (Chapter 5). Ovary dissections revealed
that one (X. saxesenii field galleries: Chapter 2) to two quarters (X. affinis laboratory galleries;
Appendix 1) of staying females also lay eggs in the natal nest during their philopatric period (Chapter
2 and 5). These egg-laying females (i) engaged more often in social behaviours than non-egg-laying
females (X. affinis: Appendix 1) and (ii) they appeared to be the only ones overtaking the most
dangerous blocking behaviour (X. saxesenii: Chapter 1), which exposes females to predation and
parasites. Numbers of egg-layers did neither correlate with the number of staying adult females nor
with the number of eggs, which suggests that egg numbers are adjusted to fungus productivity.
Interestingly, delayed dispersal imposes fitness costs on independent reproduction, because X. affinis
females that disperse after their philopatric period produced fewer eggs in their brood when bred
independently than females removed from the gallery before their philopatric period, suggesting
that delayed dispersal comes at a cost to X. affinis females (Chapter 5). This cost probably results
from reproduction and alloparental care in the natal nest during philopatry. Hence, we could reject
the hypothesis that females engage in maturation feeding (i.e., store up body reserves) during this
period.
Artificial selection on early and late dispersal in X. saxesenii resulted within six generations in a
significantly different timing of dispersal between the two lines (Chapter 3). Late dispersal was
correlated with a higher number of offspring produced by the mother and more adults staying within
the nest, which suggests that delayed dispersal is linked with a cooperative strategy. Indeed, adult
female activity and the frequency of cooperative nest protection are higher in the late-dispersing
strain than the early-dispersing strain. The cooperative strategy of late dispersers is, however,
associated with a lower success of independent nest foundation (Chapter 3).
In summary, X. saxesenii and X. affinis are typical cooperative breeders, characterized by overlapping
generations, cooperative brood care, and totipotent workers. Their galleries are relatively long-lived
in comparison to other subsocial ambrosia beetles (e.g., Anisandrus dispar, Xylosandrus germanus;
176
Summary & Conclusion
Schneider-Orelli 1913; Bischoff 2004), and thus several females in consecutive offspring generations
can sometimes continue breeding in the same protected nest. In the temperate X. saxesenii, this is
possible because larval and pupal stages produced in fall stay together with adults in the natal nest
during the winter and adults may restart egg-laying in the following spring. Females in the temperate
X. germanus also delay dispersal, but have only one breeding cycle per gallery; they overwinter
exclusively as adults (Heidenreich 1960; 1964). It is possible that cold-tolerance of immature X.
saxesenii has facilitated the evolution of cooperative breeding / facultative eusociality in this species
(cf. facultative eusociality in halictid bees; Eickwort et al. 1996; Field et al. 2010). In conclusion, the
high degree of sociality in X. saxesenii and X. affinis (and likely many other xyleborine ambrosia
beetles) seems to result from a combination of four major factors (for details see Chapter 1 and 6): (i)
parental care as a preadaptation for the evolution of sociality in the ancestors of modern ambrosia
beetles; (ii) very high relatedness within families due to haplodiploidy and inbreeding; (iii) a
proliferating, monopolisable nutritional resource providing ample food for many individuals that
needs to be tended and protected; and (iv) high costs of dispersal due to the difficulties of finding a
suitable host tree, of nest foundation, and a successful start of fungiculture, which render predispersal cooperation particularly worthwhile. Current knowledge suggests that all ambrosia beetles
are at least sub-social, which seems necessary for successful fungus gardening, protection of the nest
and brood care. Cooperative breeding may evolve in species, which galleries may persist for several
consecutive offspring generations. This depends (a) on the generation time, which is not more than
30 days under optimal conditions in the majority of xyleborine ambrosia beetles, and (b) on the
longevity of the breeding substrate (i.e., competition with other ambrosia beetles, the timing of
beetle attack in the dying process of a tree, diameter of branches/stems, wood type). Totipotency of
all females is, however, expected to be maintained in all species settling dying or dead trees that are
able to disperse if the productivity of the nest cannot be guaranteed any longer. In X. saxesenii, for
example, many galleries need to be left after a single generation, despite the fact that some galleries
are productive for several offspring cycles. The obligatorily eusocial ambrosia beetle Austroplatypus
incompertus settles living tress – a habitat with almost endless productivity. Indeed, single galleries
can be productive for at least 36 years, although productivity is apparently very low because beetles
need six years to reach adulthood (Kent and Simpson 1992).
New insights into the feedback between sociality and fungus-culturing
in xyleborine ambrosia beetles
Division of labour and sociality in ambrosia beetles evolved multiple times in association with
fungiculture, and every step towards higher sociality possibly facilitated the beetle-fungus
mutualism, or vice versa. This contrasts with fungus-farming ants and termites, which already lived in
eusocial groups at the origin of fungiculture (Mueller and Gerardo 2002; Mueller et al. 2005), and
makes ambrosia beetles a unique model system to study the evolution of sociality in relation to
fungiculture (Mueller et al. 2005; Biedermann and Taborsky 2011). But why should fungiculture profit
from sociality and division of labour? First, farmed fungi can profit from a group of individuals that
jointly excavate a gallery, protect the gardens, and maintain optimal growing conditions by regulating
humidity, for example by blocking (Chapter1), and weeding the gardens. The latter has not been fully
demonstrated in ambrosia beetles, but in an experiment we showed that X. saxesenii larvae can
reduce the growth of mould (Chapter 1), and correlative behavioural studies showed that certain
hygienic behaviours of adult female X. affinis (e.g. shuffling, fungus cropping) are more commonly
shown in the presence of fungal pathogens (Appendix 1).
The strongest evidence for a positive feedback between sociality and fungus-farming, however, is our
finding that in some species fruiting of ambrosia fungi can be induced, apparently simply by the
physical contact of the fungus with either larvae, pupae, or adult females (see below, Appendix 2).
The resulting fruiting structures apparently serve as the dominant food for the whole colony. Hence,
the more immatures and adults are present within a gallery, the stronger they induce ambrosial
177
Summary & Conclusion
growth and the better is the food supply for the developing brood. This suggests that daughters that
delay dispersal enhance gallery productivity simply by their presence.
New insights into the xyleborine ambrosia beetle-fungus mutualism
Ambrosia beetles are close relatives of bark
beetles, but they differ in their way to
colonize trees and their dependence on fungi
for food. Unlike bark beetles, which live
under the bark where they feed on fungusinfested tree tissues, ambrosia beetles build
tunnel systems (galleries) deep within the
sapwood, and they actively cultivate
mutualistic fungi in “ambrosia gardens” on
the walls of these galleries as their sole food
source (Francke-Grosmann 1967; Norris
1979; Beaver 1989). Primary symbionts, in
the two anamorph genera Ambrosiella and
Raffaelea (Ascomycota), dominate these gardens (Beaver 1989; Farrell et al. 2001). Interestingly,
both genera originate from two different clades of fungal plant pathogens (Ceratocystis and
Ophiostoma; Cassar and Blackwell 1996; Harrington et al. 2010), but they show convergent
morphological adaptations for the symbiosis with beetles. They are strongly polymorphic and change
their mycelial growth either in a yeast-like form within their host beetle mycetangia, or produce
asexual fruiting structures (conidiophores and conidiospores) in the presence of the beetles. This
ambrosial growth forms large mats on the tunnel walls (Fig. 1) and is essential for beetle nutrition.
Convergent fruiting is also known from the cultivars of farming ants (= gongolydia; Bass and Cherrett
1996) and termites (= nodules; Aanen et al. 2002), but in none of these systems it is fully understood
how the insect trigger this specialized growth. Adult female beetles induce ambrosial growth in pure
cultures of the fungus on plates via an unknown mechanism (in Xyleborus dispar; Batra and Michie
1963; French and Roeper 1972), and one of my studies showed that also larvae and pupae can trigger
ambrosial growth (in Xylosandrus germanus; Appendix 2). The induction by non-moving pupae
strongly suggests that chemical substances produced either by the insects or other microbes are
involved. Ambrosia layers within galleries are covered with unstudied bacteria (see Fig. 1) and yeasts,
bacteria and fungi can be isolated from the body surface and mouth-parts of adult beetles and larvae
(own unpubl. data). A positive role of yeasts in the gardens of ambrosia beetles has been frequently
claimed (Webb 1945; Francke-Grosmann 1963; 1967; Beaver 1989), but never tested. In summary,
there are strong indications that symbiotic microbes might be involved in the fruiting of the ambrosia
fungi, which would be a unique new function of mutualistic microbes in fungus-cultivating insects.
Originally thought to just involve the beetles and ambrosia fungi, fungus gardens of ambrosia beetles
are now considered to inhabit multiple other microbial associates (Haanstad and Norris 1985;
Kajimura and Hijii 1992), like other filamentous fungi (e.g. Fusarium sp., Penicillium sp., Paecilomyces
sp.; Francke-Grosmann 1967; Beaver 1989; Six et al. 2011), yeasts and bacteria (Haanstad and Norris
1985; Canganella et al. 1994; Endoh et al. 2008; Ninomiya et al. 2010; Chapter 9). A community of
microbes with different enzymatic capabilities may have advantages to detoxify and degrade wood
compounds synergistically and may profit from nutrient exchange (Shifrine and Phaff 1956;
Canganella et al. 1994; Adams et al. 2008; Chapter 8). Additionally, antibiotics-producing symbionts
(e.g. Streptomycetes species) are known from several other insects and defend their hosts against
pathogens (Currie et al. 1999; Kaltenpoth et al. 2005; Martin 2009). Pathogens and parasites are
probably also a threat to ambrosia gardens. However, although fresh wood is very attractive for
many wood-degrading microbes, beetle fungus gardens are well protected against environmental
microbes within tunnels of almost sterile wood. In contrast, fungus-farming ants and termites need
178
Summary & Conclusion
to (i) sequester their gardens from the surrounding soil that is heavily contaminated with microbes
and developed, and (ii) have advanced microbial defences against microbes that they constantly
bring into the nest with the substrate they collect to feed their fungi (Mueller and Gerardo 2002;
Mueller et al. 2005). Nevertheless, my studies showed that several pathogenic moulds (Paecilomyces
sp., Mucor sp., etc.) also co-exist in ambrosia gardens (Chapter 7, Appendix 1), and that both larvae
and adults are able to hinder their spread within their galleries in unknown ways (Chapter 2), which
may also involve other microbes. Antibiotics-producing bacteria inhibit the spread of an antagonistic
fungus in farming ants (Currie et al. 1999; Currie and Stuart 2001) and other bark beetles (Cardoza et
al. 2006; Scott et al. 2008). Additionally, insect-associated yeasts have been shown to curtail growth
of antagonistic fungi (Liu et al. 2007; Adams et al. 2008; Rodrigues et al. 2009). Such a defence by
microbes has the advantage that, unlike the insects, microbes can potentially evolve at the same rate
as pathogens, enabling mutualistic insect-microbe systems to respond more rapidly to the
emergence of novel disease genotypes (Currie et al. 1999; Denison et al. 2003; Adams et al. 2009).
Literature
Aanen DK, Eggleton P, Rouland-Lefévre C, Guldberg-Frøslev T, Boomsma JJ & Rosendahl S (2002) The evolution
of fungus-growing termites and their mutualistic fungal symbionts. PNAS 99: 14887-14892.
Adams AS, Currie CR, Cardoza Y, Klepzig KD & Raffa KF (2009) Effects of symbiotic bacteria and tree chemistry
on the growth and reproduction of bark beetle fungal symbionts. Canadian Journal of Forest Research-Revue
Canadienne de Recherche Forestiere 39: 1133-1147.
Adams AS, Six DL, Adams SM & Holben WE (2008) In vitro interactions between yeasts and bacteria and the
fungal symbionts of the mountain pine beetle (Dendroctonus ponderosae). Microbial Ecology 56: 460-466.
Bass M & Cherrett JM (1996) Leaf-cutting ants (Formicidae, Attini) prune their fungus to increase and direct its
productivity. Functional Ecology 10: 55-61.
Batra LR & Michie MD (1963) Pleomorphism in some ambrosia and related fungi. Transactions of the Kansas
Academy of Science 66: 470-481.
Beaver RA (1989) Insect-fungus relationships in the bark and ambrosia beetles. Insect-fungus interactions
(Wilding N, Collins NM, Hammond PM & Webber JF, eds), pp. 121-143. Academic Press, London.
Biedermann PHW & Taborsky M (2011) Larval helpers and age polyethism in ambrosia beetles. Proc. Natl Acad.
Sci. USA 108: 17064-17069.
Bischoff LL (2004) The social structure of the haplodiploid bark beetle, Xylosandrus germanus. Diploma Thesis,
Zoological Institute, University of Berne, Switzerland.
Canganella F, Paparatti B & Natali V (1994) Microbial Species Isolated from the Bark Beetle Anisandrus-Dispar F.
Microbiological Research 149: 123-128.
Cardoza YJ, Klepzig KD & Raffa KF (2006) Bacteria in oral secretions of an endophytic insect inhibit antagonistic
fungi. Ecological Entomology 31: 636-645.
Cassar S & Blackwell M (1996) Convergent origins of ambrosia fungi. Mycologia 88: 596-601.
Chapman TW, Crespi BJ, Kranz BD & Schwarz MP (2000) High relatedness and inbreeding at the origin of
eusociality in gall-inducing thrips. PNAS 97: 1648-1650.
Currie CR, Scott JA, Summerbell RC & Malloch D (1999) Fungus-growing ants use antibiotic-producing bacteria
to control garden parasites. Nature 398: 701-704.
179
Summary & Conclusion
Currie CR & Stuart AE (2001) Weeding and grooming of pathogens in agriculture by ants. Proceedings of the
Royal Society of London Series B-Biological Sciences 268: 1033-1039.
Denison RF, Kiers ET & West SA (2003) Darwinian agriculture: When can humans find solutions beyond the
reach of natural selection? Q. Rev. Biol. 78: 145-168.
Eickwort GC, Eickwort JM, Gordon J & Eickwort MA (1996) Solitary behavior in a high altitude population of the
social sweat bee Halictus rubicundus (Hymenoptera: Halictidae). Behav. Ecol. Sociobiol. 38: 227-233.
Endoh R, Suzuki M & Benno Y (2008) Ambrosiozyma akamigamensis sp. nov. and A. neoplatypodis sp. nov., two
new ascomycetous yeasts from ambrosia beetle galleries. Antonie van Leeuwenhoek 94: 365-376.
Farrell BD, Sequeira AS, O'Meara BC, Normark BB, Chung JH & Jordal BH (2001) The evolution of agriculture in
beetles (Curculionidae : Scolytinae and Platypodinae). Evolution 55: 2011-2027.
Field J, Paxton RJ, Soro A & Bridge C (2010) Cryptic Plasticity Underlies a Major Evolutionary Transition. Current
Biology 20: 2028-2031.
Francke-Grosmann H (1956) Hautdrüsen als Träger der Pilzsymbiose bei Ambrosiakäfern. Z. Morph. u. Ökol.
Tiere 45: 275-308.
Francke-Grosmann H (1963) Some new aspects in forest entomology. Annual Review of Entomology 8: 415-438.
Francke-Grosmann H (1967) Ectosymbiosis in wood-inhabiting beetles. Symbiosis (Henry SM, ed), pp. 141-205.
Academic Press, New York.
French JRJ & Roeper RA (1972) Interactions of ambrosia beetle, Xyleborus dispar (Coleoptera, Scolytidae), with
its symbiotic fungus Ambrosiella hartigii (Fungi imperfecti). Canadian Entomologist 104: 1635-1641.
Haanstad JO & Norris DM (1985) Microbial symbiotes of the ambrosia beetle Xyletorinus politus. Microbial
Ecology 11: 267-276.
Hamilton WD (1972) Altruism and related phenomena, mainly in social insects. Ann. Rev. Ecol. Syst. 3: 193-232.
Harrington TC, Aghayeva DN & Fraedrich SW (2010) New combinations in Raffaelea, Ambrosiella, and
Hyalorhinocladiella, and four new species from the redbay ambrosia beetle, Xyleborus glabratus. Mycotaxon
111: 337-361.
Heidenreich E (1960) Primärbefall durch Xylosandrus germanus an Jungeichen. Anzeiger fur Schadlingskunde
Pflanzenschutz Umweltschutz 23: 5-10.
Heidenreich E (1964) Ökologische Bedingungen für Primärbefall durch Xylosandrus germanus. J. Appl. Entom.
54: 131-140.
Hölldobler B & Wilson EO (1990) The ants. The Belknap Press of Harvard University Press, Cambridge, MA.
Hosking GB (1972) Xyleborus saxeseni, its life-history and flight behaviour in New Zealand. N. Z. J. Forest Science
3: 37-53.
Hulcr J, Kolarik M & Kirkendall LR (2007) A new record of fungus-beetle symbiosis in Scolytodes bark beetles
(Scolytinae, Curculionidae, Coleoptera). Symbiosis 43: 151-159.
Kajimura H & Hijii N (1992) Dymamics of the fungal symbionts in the gallery system and the mycangia of the
ambrosia beetle, Xylosandrus mutilatus (Blandford) (Coleoptera, Scolytidae). Ecological Research 7: 107-117.
Kaltenpoth M, Gottler W, Herzner G & Strohm E (2005) Symbiotic Bacteria Protect Wasp Larvae from Fungal
Infestation. Current Biology 15: 475-479.
180
Summary & Conclusion
Kent DS & Simpson JA (1992) Eusociality in the beetle Austroplatypus incompertus (Coleoptera: Platypodidae).
Naturwissenschaften 79: 86-87.
Kirkendall LR, Kent DS & Raffa KF (1997) Interactions among males, females and offspring in bark and ambrosia
beetles: the significance of living in tunnels for the evolution of social behavior. The Evolution of Social Behavior
in Insects and Arachnids (Choe JC & Crespi BJ, eds), pp. 181-215. Cambridge University Press.
Liu X, Wang J, Gou P, Mao C, Zhu ZR & Li H (2007) In vitro inhibition of postharvest pathogens of fruit and
control of gray mold of strawberry and green mold of citrus by aureobasidin A. International Journal of Food
Microbiology 119: 223-229.
Martin K (2009) Actinobacteria as mutualists: general healthcare for insects? Trends in Microbiology 17: 529535.
Mueller UG & Gerardo N (2002) Fungus-farming insects: Multiple origins and diverse evolutionary histories.
Proceedings of the National Academy of Sciences of the United States of America 99: 15247-15249.
Mueller UG, Gerardo NM, Aanen DK, Six DL & Schultz TR (2005) The evolution of agriculture in insects. Annual
Review of Ecology Evolution and Systematics 36: 563-595.
Ninomiya S, Mikata K, Nakagiri A, Nakase T & Kawasaki H (2010) Pichia porticicola sp. nov., a novel
ascomycetous yeast related to Pichia acaciae isolated from galleries of ambrosia beetles in Japan. The Journal
of General and Applied Microbiology 56: 281-286.
Norris DM (1979) The mutualistic fungi of Xyleborini beetles. Nutrition, Mutualism, and Commensalism (Batra
LR, ed), pp. 53-63. Allanheld, Osmun & Company, Montclair.
Rodrigues A, Cable RN, Mueller UG, Bacci M & Pagnocca FC (2009) Antagonistic interactions between garden
yeasts and microfungal garden pathogens of leaf-cutting ants. Antonie Van Leeuwenhoek International Journal
of General and Molecular Microbiology 96: 331-342.
Schneider I (1987) Distribution, fungus-fransfer and gallery construction of the ambrosia beetle Xyleborus
affinis in comparison with X. mascarensis (Coleoptera, Scolytidae). Entomologia Generalis 12: 267-275.
Schneider-Orelli O (1913) Untersuchungen über den pilzzüchtenden Obstbaumborkenkäfer Xyleborus
(Anisandrus) dispar und seinen Nährpilz. Zentralblatt für Bakteriologie, Parasitenkunde, Infektionskrankheiten
und Hygiene II 38: 25-110.
Scott JJ, Oh DC, Yuceer MC, Klepzig KD, Clardy J & Currie CR (2008) Bacterial protection of beetle-fungus
mutualism. Science 322: 63.
Shifrine M & Phaff HJ (1956) The Association of Yeasts with Certain Bark Beetles. Mycologia 48: 41-55.
Six DL (2003) Bark beetle-fungus symbioses. Insect Symbiosis (Bourtzis K & Miller TA, eds), pp. 97-114. CRC
Press, Boca Raton.
Six D, Poulsen M, Hansen A, Wingfield M, Roux J, Eggleton P, Slippers B & Paine T (2011) Anthropogenic effects
on interaction outcomes: examples from insect-microbial symbioses in forest and savanna ecosystems.
Symbiosis1-21.
Webb S (1945) Australian ambrosia fungi. Proceedings of the Royal Society of Victoria 57: 57-79.
Wilson EO (1971) The insect societies. Belknap Press of Harvard University Press, Cambridge.
181
Summary & Conclusion
182
Acknowledgements
Acknowledgements
Dankeschön. Merci. Thank you.
Eine Lebensabschnitt – 7 Jahre am Hasli – gehen für mich nun zu Ende!
Michael, vielen Dank für deine Ermunterung, nach Bern zu kommen, und das Sozialverhalten der
Ambrosiakäfer zu erforschen. Ich habe durch dein „Riegersburg-Seminar“ nicht nur eine Passion für
Ambroiakäfer und eine Nische für meine weitere Karriere als Wissenschaftler entdeckt, sondern habe auch – 2
Jahre später – meine baldige Frau Tabea kennengelernt. Danke, dass du mich immer fachlich, aber auch
finanziell, voll und ganz unterstützt hast. Als Betreuer hast du mir eine klare Richtung für meine Forschung
gewiesen, mir aber gelichzeitig viel Freiheit bei der Durchfürung gelassen. Das war genau richtig für mich. Mit
deinem breiten Fachwissen hast du immer die treffenden Fragen gestellt und neue Lösungswege aufgezeigt.
Auch wenn du mir keine finanzierte Doktorandenstelle anbieten konntest, hast du mich immer ermuntert,
selbst Forschungsanträge zu schreiben. Ich war gezwungen, selbst aktiv zu werden, Projekte selbst zu planen,
und mir Netzwerke mit anderen Wissenschaftlern aufzubauen. Obwohl das Schreiben eigener Anträge mich
sehr gefordert hat, und ich mir damals oft eine fixe Stelle gewünscht hätte, bin ich im Nachhinein sehr froh
über die lehrreichen Erfahrungen, die ich in dieser Zeit machen konnte. Mit dem über die Jahre aufgebauten
Rüstzeug fühle ich mich jetzt sehr gut auf eine unabhängige wissenschaftliche Karriere vorbereitet. Ich danke
dir ausserdem, dass du mich dabei unterstützt, auch nach dem Abschied aus Bern mit dem Modellsystem
„Ambrosiakäfer“ weiterzuarbeiten. Danke für die tolle und unglaublich lehrreiche Zeit am Hasli!
Während dieser Arbeit habe ich grosses Entgegenkommen von mehreren Personen am Institut für
Forstentomologie der BOKU Wien erfahren. Christian, vielen Dank für deine sofortige Zusage, meine
Doktorarbeit und den Antrag bei der ÖAW (Österreichischen Akademie der Wissenschaften) zu unterstützen.
Du hast mir Anfang 2008 mit dem gemeinsamen Durcharbeiten und „Zuspitzen“ den Antrag für mein DOCStipendium gerettet! Ich bin mir sicher, dass ich die 2 jährige Finanzierung ohne deine Hilfe nicht bekommen
hätte. Dein Wohlwollen und dein Glaube an meine wissenschaftlichen Fähigkeiten haben mir immer sehr viel
Selbstvertrauen gegeben. Thomas, dir möchte ich vor allem für die Vermittlung der Kontakte in die USA
danken. Wo würde ich heute ohne die Erfahrungen und das Wissen, das ich mir in den USA angeeignet habe,
stehen?! Rudi und Axel, euch danke ich für die spannenden Einblicke in die angewandte Borkenkäferforschung.
Am gesamten Institut für Forstentomologie habe ich immer sehr viel Menschlichkeit und ehrliche Freundschaft
erlebt – danke den Führungspersonen Axel, Rudi und Christian für dieses tolle Klima.
In the first year of my Ph.D.-studies I got the chance to visit the Southern Research Station (Louisiana, USA) for
a few months where I learned how to isolate and work with fungi. My deepest thanks go to Kier for enabling
this collaboration. I am so grateful that you invited me, even though you did not know me personally. I profited
so much from your experience in working with fungi, and your financial support for visiting the labs of Diana,
Cameron, Ken and Ulrich. The scientific input from these people was absolutely essential for the success of my
Ph.D. I also want to thank Stacy and Chi for giving me social support during this visit in Pineville. I want to thank
you two, Dan, Bill and Michael for the hospitality. Discussions with Dr. Moser, Brian and Will were a big joy. I
got to meet Diana and Eric in Missoula, which resulted in an exciting collaboration on fungus identifications.
Thank you, Diana, for sharing your knowledge on fungi with me. Eric, thanks for the adventures during free
time. Cameron, thank you so much for your invitation to Madison and the months I could spend in your lab.
The insights you gave me into ant-fungus interactions were a great inspiration. I enjoyed meeting Frank, Kirk,
Garret and Ken, and became friends with Sandye, Aaron, Michael and Jiri. Thank you all for helpful discussions.
183
Acknowledgements
Ulrich, your passion on insect-fungus symbioses was immediately contagious. It was a great experience to go
out in the field and dig up ant Atta-nest together with you. I profited a lot from the many discussions we had in
Austin and elsewhere and inputs you gave me were crucial for several of my experiments, and your comments
facilitated our PNAS publication. In Austin I also became friends with Barrett – thank you so much for the
wonderful beetle drawings (on my wall and on the cover of this thesis) and the passion and joy you radiate.
Thanks also to Prof. Heinz Richner for chairing my Ph.D.-defense and to Prof. Koos Boomsma who kindly agreed
to be my external referee.
Alex, Markus und Arne danke für eure Freundschaft und für den Spass, den wir gemeinsam hatten – sei es bei
Partys, bei Wanderungen in der Natur, unseren „Badminton-Männerabenden“ oder den vielen lustigen
Spieleabenden. Besonders geschätzt habe ich eure Hingabe bei wissenschaftlichen Diskussionen. Auch Babsi,
Thomas und Stefan danke ich für die gemeinsamen Erlebnisse. Cori, Michael und Andrea, euch danke ich für
die nette Gemeinschaft im Büro und den netten Gesprächen. Marcel, Florian und Claudia, danke für eure
Freundschaft. Ihr wart wichtige Gesprächspartner für mich, nicht nur bei beruflichen, sondern auch bei privaten
Themen und die gemeinsamen Unternehmungen bleiben mir in schöner Erinnerung. Schliesslich möchte ich all
den vielen herzlichen Menschen am Hasli und in der Baltzerstrasse danken, in deren Gemeinschaft ich mich
aufgehoben gefühlt habe.
Ich danke meinen Eltern, ohne die ich nicht der wäre, der ich bin. Ihr habt mich in meiner Leidenschaft für die
Natur immer unterstützt und mir auch bei Schwierigkeiten geholfen, dranzubleiben. „Bua“, dir danke ich für die
Besuche und die gemeinsamen Unternehmungen in der Schweiz und, dass ich mich immer auf dich verlassen
kann. Jörg und Michael, für eure langjährige Verbundenheit. Auch wenn wir uns alle vier einmal lange nicht
gesehen haben, ist es doch dann schnell wieder so nah wie bei den gemeinsamen Schulferien am Mitterberg.
Meinen langjährigen Freunden in Österreich – Gernot, Franz und Martin, danke ich für eure Verbundenheit
auch auf die Distanz, die Gespräche über Skype und gemeinsamen Unternehmungen. Tabea, dir danke ich, für
deine seelische wie auch fachliche Unterstützung. Viele der Texte in dieser Doktorarbeit hast du mehrmals
korrigiert und mit deinen Sprachkenntnissen weitreichend verbessert. Du bist die wichtige Stütze und starke
Frau an meiner Seite!
184
Contributions & Funding
Contributions & Funding
Chapter 1
P.H.W.B. and M.T. designed research; P.H.W.B. performed research; P.H.W.B. analyzed data;
and P.H.W.B. and M.T. wrote the paper.
Chapter 2
P.H.W.B., K.P. and M.T. designed research; P.H.W.B. and K.T. performed research; P.H.W.B.
analyzed data; and P.H.W.B. and M.T. wrote the paper.
Chapter 3
P.H.W.B. and M.T. designed research; P.H.W.B. performed research; P.H.W.B. analyzed data;
and P.H.W.B. and M.T. wrote the paper.
Chapter 4
P.H.W.B. designed research, performed research, analyzed data and wrote the paper.
Chapter 5
P.H.W.B. designed research; P.H.W.B. performed research; P.H.W.B. analyzed data; and
P.H.W.B., K.D.K. and M.T. wrote the paper.
Chapter 6
P.H.W.B. designed research; P.H.W.B. performed research; P.H.W.B. analyzed data; and
P.H.W.B. and M.T. wrote the paper.
Chapter 7
P.H.W.B. and K.D.K. designed research; P.H.W.B. and D.L.S. performed research; P.H.W.B.
analyzed data; and P.H.W.B., M.T. and D.L.S. wrote the paper.
Chapter 8
P.H.W.B. and H.H.d.F.L. designed research; P.H.W.B. and H.H.d.F.L. performed research;
H.H.d.F.L. analyzed data; and P.H.W.B. and H.H.d.F.L. wrote the paper.
Chapter 9
P.H.W.B. provided research material.
Appendix 1
P.H.W.B. designed research; P.H.W.B. and D.L.S. performed research; P.H.W.B. analyzed data;
and P.H.W.B., M.T. and D.L.S. wrote the paper.
Appendix 2
P.H.W.B. and C.R.C. designed research; P.H.W.B. performed research; P.H.W.B. analyzed data;
and P.H.W.B. wrote the paper.
This project was funded by the following organizations:
•
Austrian Academy of Science (ÖAW)
•
Roche Research Foundation
•
US-Department of Agriculture (USDA)
•
Department of Bacteriology, University of Wisconsin-Madison
•
Phil. nat. Faculty, University of Bern
•
Swiss Zoological Society (SZS)
•
Conférence universitaire de Suisse occidentale (CUSO)
•
Ethological Society
•
International Union for the Study of Social Insects (IUSSI)
185
186
187
188
Curriculum Vitae
Curriculum Vitae
Name:
Peter Hans Wilhelm Biedermann
Date of birth:
21 June 1981 in Leoben, Austria
Nationality:
Austrian
Address:
Lorystrasse 12, CH-3008 Bern, Switzerland
tel.: +41 31 534 9019
e-mail: [email protected]
st
EDUCATION
1/2009-
Ph.D. candidate
Thesis on “The Evolution of Cooperation in Ambrosia beetles” Institute of Ecology & Evolution,
University of Bern, Switzerland; Supervisor: Prof. Dr. Michael Taborsky
7/2007
Master degree
Master of Science in Ecology and Evolution, awarded with distinction (“summa cum laude”)
4/2005-7/2007
Study of Ecology and Evolution
Thesis on “Social behaviour of Xyleborinus saxesenii”
Institute of Zoology, University of Bern, Switzerland; Supervisor: Prof. Dr. Michael Taborsky
1/2005
Bachelor degree
Bachelor in Biology; Thesis on “Hidden leks in the yellow browed warbler Phylloscopus
inornatus”; Institute of Zoology, University of Graz, Austria; Supervisor: Prof. Dr. Heiner Römer
2002-2005
Study of Biology/Zoology
Major in Ethology and Neurobiology; University of Graz, University of Vienna, Austria
2000 -2001
Study of Physics
University of Graz, Austria
1999
School leaving examination
(Matura) awarded with distinction, BRG Leoben, Austria
FURTHER EXPERIENCES
4-8/2009
Visiting researcher “Microbial communities of Ambrosia beetles”
Department of Bacteriology, UW Madison/WI, USA; Supervisor: Ass.-Prof. Dr. Cameron Currie
8-11/2004
Internship on “Sexual selection in blue tits”
Max Planck Institute for Ornithology, Seewiesen, Germany; Supervisors: Prof. Dr. Bart
Kempenaers
4-8/2004
Project on “Drone (Apis mellifera) – swallow interactions”
Institute of Zoology, University of Graz, Austria. supervisor: Prof. Dr. Karl Crailsheim
2-3/2004
Field assistant “Neurophysiology of katydids in relation to predation by bats”
Smithsonian Tropical Research Institute, BCI, Panama; Supervisor: Prof. Dr. Heiner Römer
4-8/2003
Project on “Breeding-ecology of Yellow-browed Warblers” in Mongolia
University of Göttingen, Germany, funded by the DAAD; supervisor: Prof. Dr. M. Mühlenberg
8-10/2002
Field assistant Thunder Cape Bird Observatory
ThunderBay/ON, Canada; banding of songbirds and raptors, bird observations
6-8/2001 & 2002 Field assistant “Sex differences in Sandhill-cranes”
Intern. Crane Foundation, Baraboo/WI, USA; banding, radio-tracking cranes, prairie restoration
189
Curriculum Vitae
5/2001
Field assistant for whale and dolphin research
Tethys organization, Mediterranean sea
2000-2005
Bird monitoring for “Atlas of breeding-birds of Graz”
Museum of Natural History (Joanneum), Graz, Austria
POSITIONS HELD
1/2009-
Ph.D. student with Prof. Michael Taborsky
Institute of Ecology & Evolution, University of Bern, Switzerland
8-12/2007
Visiting researcher with Dr. Kier Klepzig
Southern Research Station, USDA Forest Service, Pineville/LA, USA
INVITED TALKS AND SEMINARS
2011
- Max-Planck Institute for Chemical Ecology, Jena, Germany
2010
- Department of Environmental Sciences, ETH Zurich, Switzerland
- Institute for Zoology, University of Regensburg, Germany
- Seminar in Ecology & Evolution, University of Neuchatel, Switzerland
- Zoological Colloquium, University of Graz, Austria
2007
- Section of Integrative Biology, University of Texas, Austin, USA
- Department of Bacteriology, University of Wisconsin, Madison, USA
TALKS AT MEETINGS AND CONFERENCES
2011
- Genetics of Bark Beetles and Associated Microorganisms (IUFRO), Sopron, Hungary
- European Society of Evolutionary Biology (ESEB), Tübingen, Germany (Poster)
- Deutsche Gesellschaft für allgemeine und angewandte Entomologie (DGaaE), Berlin, D
- Austrian Entomological Society (öEG), Graz, Austria
- Biology2011, Zurich, Switzerland
2010
- International Union for the Study of Social Insects (IUSSI), Kopenhagen, Denmark
- COST meeting, Diversity of symbioses in arthropods, Bad Bevensen, Germany (Poster)
- Biology2010, Neuchatel, Switzerland
2009
- International Union for the Study of Social Insects (IUSSI), Frauenchiemsee, Germany*
* Awarded with the Kutter-prize
- European Society of Evolutionary Biology (ESEB), Torino, Italy (Poster)
- 57th Annual Meeting Entomological Society of America, Indianapolis, USA (Virtual Poster)
2008
- Behavioural Biology, Dijon, France
- Ethological Society, Regensburg, Germany
2007
- East Texas Forest Entomology Seminar (ETFES07), Nacogdotches/TX, USA
- International Union of Forest Research (IUFRO07), Vienna, Austria
- D-Day, Lausanne, Switzerland (Poster)
- Meeting of PhD Students in Evolutionary Biology (EMPSEB13), Lund, Sweden
- Xylobionten-Meeting, Bern, Switzerland
- DZG and Ethological Society, Göttingen, Germany
- Biology2007, Zurich, Switzerland (Poster)
2006
- Ethological Society, Bielefeld, Germany (Poster)
2005
- International Union for the Study of Social Insects (IUSSI), Halle, Germany
- Zoological Colloquium, Graz, Austria
190
Curriculum Vitae
List of Publications
P U B L I C A T I O N S (peer reviewed)
(1) Biedermann PHW (2006) Hidden leks in the Yellow-browed Warbler (Phylloscopus inornatus)? Investigations from the Khan Khentey Reserve (Mongolia). Acrocephalus 27: 233-247.
(2) Biedermann PHW & Kärcher MH (2008) Weather-dependent activity and flying height of Barn Swallows
(Hirundo rustica) and House Martins (Delichon urbica) in southwestern Styria. Egretta 50: 76-81.
(3) Kärcher MH, Biedermann PHW, Hrassnigg N & Crailsheim K (2008) Predator-prey interaction between
drones A. m. carnica and swallows Hirundo rustica or Delichon urbica. Apidology 39(3): 302-309.
(4) Delhey K, Peters A, Biedermann PHW & Kempenaers B (2008) Optical properties of the uropygial gland
secretion: no evidence for UV cosmetics in birds. Naturwissenschaften 95(10): 939-946.
(5) Biedermann PHW, Klepzig KR & Taborsky M (2009) Fungus cultivation by ambrosia beetles: Behavior
and laboratory breeding success in three Xyleborine species. Environmental Entomology 38(4): 10961105.
(6) Biedermann PHW (2010) Observations on sex ratio and behavior of males in Xyleborinus saxesenii
Ratzeburg (Scolytinae, Coleoptera). In: Cognato AI, Knížek M (Eds) Sixty years of discovering scolytine
and platypodine diversity: A tribute to Stephen L. Wood. Zookeys 56: 253-267.
(7) Grubbs K.J., Biedermann P.H.W., Suen G., Adams S.M., Moeller J.A., Klassen J.L., Goodwinm L.A.,
Woyke T., Munk A.C., Bruce D., Detter C., Tapia R., Han C.S. & Currie C.R. (2011): Complete Genome
Sequence of Streptomyces cf. griseus (XyelbKG-1 1), an Ambrosia Beetle-Associated Actinomycete.
Journal of Bacteriology 193(11): 2890–2891.
(8) Biedermann PHW, Klepzig KD & Taborsky M (2011): Costs of delayed dispersal and alloparental care in
the fungus-cultivating ambrosia beetle Xyleborus affinis Eichhoff (Scolytinae: Curculionidae). Behavioral
Ecology & Sociobiology 65:1753–1761.
(9) Biedermann PHW & Taborsky M (2011): Larval helpers and age polyethism in ambrosia beetles.
Proceedings of the National Academy of Science USA 108(41): 17064-17069.
(10) Biedermann PHW, Peer K & Taborsky M (2011): Female dispersal and reproduction in the ambrosia
beetle Xyleborinus saxesenii Ratzeburg (Coleoptera; Scolytinae). Mitteilungen der Deutschen
Gesellschaft für allgemeine und angewandte Entomologie 18: in press.
Submitted
(11) De Fine Licht HH & Biedermann PHW (submitted): Patterns of functional enzyme activity in Xyleborinus
saxesenii fungus-growing ambrosia beetles. Frontiers in Zoology, in review.
OTHERS
(1) Biedermann PHW (2003) Die Kraniche der Welt. Zool. Newsletter 2; Landesmuseum Joanneum Graz.
191

Meet our staff: Victoria, Anais, Ines, Jessica and Kim

椎木静寧 - Shiigi Shizune

William Morris Gallery Schools engagement project brief

The 2 JBs - What is PEDAS?

Tours - Art Dealers Association of Canada

PDN Magazine - Paulette Tavormina

ANNA BARLOW

John MacLean | CV |

DRAFT OF A GALLERY WEBSITE “More interesting

September Fine Pri - Hastings Valley Fine Art Association Inc.