1. health and fitness
2. disease

DISS. ETH NO. 21417
HOW TO BUILD A HEALTHY COLONY:
AGE STRUCTURE AND IMMUNE STRATEGIES IN BUMBLEBEES
A dissertation submitted to
ETH ZURICH
For the degree of
Doctor of Sciences
Presented by
SÉVERINE DENISE BUECHEL
MSc., University of Lausanne
Date of birth: 9th of October 1981
Citizen of Starrkirch-Wil (SO)
Accepted on the recommendation of
Prof. Dr. Paul Schmid-Hempel, examiner
Prof. Dr. Sylvia Cremer, co-examiner
2013
TABLE OF CONTENTS
SUMMARY ................................................................................................................................. 2 ZUSAMMENFASSUNG ............................................................................................................. 4 1.1 GENERAL INTRODUCTION .............................................................................................. 6 Social life and parasite defence .......................................................................................... 6 Age polyethism with a note on worker longevity .............................................................. 7 The organization of bumblebees ........................................................................................ 8 How to build a healthy colony ........................................................................................... 9 1.2 THESIS OUTLINE .............................................................................................................. 11 2. COLONY PACE: A LIFE-HISTORY TRAIT AFFECTING SOCIAL INSECT
EPIDEMIOLOGY ...................................................................................................................... 16 3. WORKER LONGEVITY AND ITS RELATION TO COLONY RESISTANCE AGAINST
INFECTIOUS PARASITES IN A SOCIAL INSECT ............................................................... 32 4. WORKER LONGEVITY IN A PRIMITIVELY EUSOCIAL INSECT: IMMUNE
SENESCENCE DOES NOT AFFECT WORKER RESISTANCE TO AN ECOLOGICALLY
RELEVANT PARASITE. .......................................................................................................... 54 5. GENERAL DISCUSSION ..................................................................................................... 74 A hypothetical scenario: How parasite pressure may act as driving force for colony
social organization............................................................................................................ 75 ACKNOWLEDGMENTS ....................................... ERROR! BOOKMARK NOT DEFINED.
CURRICULUM VITAE ……………………....……………………………………………….84
1
SUMMARY
Longevity is an important aspect in any organism’s life history, from humans to bacteria. Tradeoffs with other traits and optimization of life-history strategies against an ecological background
(including parasites) will shape longevity patterns across individuals, populations, and species.
In social insects worker lifespan strongly affects colony success. Long-lived workers may be
favoured as they allow for large colony size, a general fitness measure in social insect colonies.
At the same time, long-lived and therefore old workers should pose an increased risk of parasite
transmission within the colony. This is because older workers are more likely to have encountered
a parasite, and moreover defence (immune) functions commonly decline with age. To counter this
epidemic risk, social insects may have evolved several adaptations related to worker age. Old
workers, for example, tend to be foragers and often have limited contact with younger workers
engaged in in-nest tasks. Bumblebees do not show such clear age-related changes in behaviour
and might be especially vulnerable to infection through old workers in the colony. In this thesis I
used the bumblebee, Bombus terrestris, and its trypanosomatid gut parasite, Crithidia bombi, to
test the role of worker age on the resistance to and transmission of this ecologically important
parasite.
In the first Chapter I demonstrate that long-lived workers indeed may be a risk for the colony
when exposed to parasites. Experimentally altering birth and death rates in infected worker
groups showed that a long-lifespan resulted in more infected workers. Because I controlled for
group size and differences in worker immunity, increased transmission should be a direct
consequence of the parasites’ residence in a long-lived host.
In the second Chapter I tested how worker resistance relates to longevity and colony size, yet
another trait that is positively related to parasite spread. I show that colonies with long-lived
workers were not only more resistant to experimental infection but long-lived workers were
typically also found in larger colonies. Increased individual resistance in colonies with long-lived
workers may reflect the high risk of parasite transmission associated with both worker longevity
2
and living at high densities. Individual resistance, however, may not necessarily be relevant to
decrease parasite transmission at old age, because immune functions (a component of resistance)
may decline with age. In the third Chapter I show that immune senescence does not affect
functional immunity in aged workers, but resistance to C.bombi is maintained. To conclude, a
‘healthy’ reproductive colony may on one hand sustain colony growth by producing long-lived
workers, but must compensate for the increased risk of having many old workers by investing in
functional immunity to avoid within- colony epidemics.
Based on my findings in this thesis, I suggest a scenario where parasite pressure positively
correlated to colony size and worker longevity acts as a selective force for the evolution towards
complex social structures. This hypothetical framework highlights the importance of parasitism in
determining colony structure and simultaneously sets the stage and calls for further investigations.
3
ZUSAMMENFASSUNG
Die individuelle Lebenserwartung ist ein wichtiger Aspekt der Lebensgeschichte. Dies gilt für alle
Organismen, von Bakterien bis zum Menschen. Unterschiedliche Investitionen in Langlebigkeit
oder andere Parameter der Lebensgeschichte und deren optimale Ausbalancierung im Lichte
verschiedener ökologischen Gegebenheiten (inklusive Parasitismus) wird die Evolution
unterschiedlicher Lebenserwartungen begünstigen. Sowohl zwischen verschiedenen Individuen
der gleichen Art als auch zwischen verschiedenen Populationen und Arten.
Bei sozialen Insekten beeinflusst die Lebenspanne einzelner Arbeiter stark den Erfolg der ganzen
Kolonie. Langlebige Individuen könnten vorteilhaft sein, da sie die Koloniegrösse positiv
beeinflussen können. Eine generell erstrebenswerte Eigenschaft in Hinsicht der Steigerung der
evolutionären Fitness. Zur gleichen Zeit jedoch stellen langlebigere Arbeiter ein erhebliches
Risiko bezüglich der Verbreitung von Parasiten innerhalb der Kolonie dar, da mit zunehmendem
Alter zum einen die Wahrscheinlichkeit einer Parasitenaufnahme steigt und zum anderen die
generelle Immunabwehr sinkt. Um dieses epidemische Risiko gering zu halten, könnten soziale
Insekten verschiedene Adaptationen hinsichtlich der Variation des Arbeiteralters evoluiert haben.
Zum Beispiel tendieren ältere Arbeiter eher dazu das Furagieren zu übernehmen und den Kontakt
zu den jüngeren Arbeitern, die sich eher im inneren des Stocks aufhalten und die Brut betreuen,
einzuschränken. Da Hummeln keine solchen deutlichen altersspezifischen Anpassungen zeigen
könnten sie besonders anfällige für die Risiken sein darstellen. In dieser Doktorarbeit habe ich
daher die Hummel, Bombus terrestris und ihren trypanosomen Darmparasiten Crithidia bombi
dazu verwendet um die Rolle des Arbeiteralters auf die Resistenz und die Transmission diese
öklogisch relevanten Parasiten zu untersuchen.
In einem ersten Kapitel demonstriere ich, dass langlebige Arbeiter in der Tat ein erhöhtes
Infektionsrisiko für die von Parasiten bedrohte Kolonien darstellen können. Ich habe
experimentell die Geburts- und Todesraten innerhalb von infizierten Arbeitern variiert und konnte
zeigen, dass eine längere Lebensspanne mehr infizierte Arbeiter zur Folge hatte. Nachdem ich für
4
die Gruppengrösse und Unterschiede in Arbeiterimmunität kontrolliert habe, sollte die erhöhte
Parasitenverbreitung eine direkte Folge der längeren Verweildauer der Parasiten in den
langlebigen Arbeitern sein.
In einem zweiten Kapitel habe ich dann die Frage untersucht, wie sich die Resistenz der Arbeiter
gegenüber Parasiten auf deren Langlebigkeit und die Koloniegrösse, beides wichtige Faktoren bei
der Parasitenausbreitung, auswirkt. Ich konnte zeigen, dass Kolonien mit langlebigeren Arbeitern
nicht nur resistenter gegenüber experimentellen Infektionen sind, sondern dass langlebigere
Arbeiter auch typischerweise in größeren Kolonien zu finden sind. Erhöhte individuelle Resistenz
in Kolonien mit langlebigeren Arbeitern könnte daher die direkte Folge von einem erhöhten
Risiko einer Parasitenübertragung durch sowohl erhöhte Arbeiterlebensspanne als auch Leben in
verdichteten Gruppen sein. Individuelle Immunität muss aber nicht zwangsläufig die
Parasitentransmission im höherem Alter positive beeinflussen, da die Immunität generell mit
zunehmendem Lebensalter abnimmt. In einem dritten Kapitel konnte ich zeigen, dass die
Immunseneszenz die funktionelle Immunität nicht beeinflusst, sondern dass die Resistenz
gegenüber C. bombi auch bei älteren Arbeitern gegeben ist. Zusammenfassend stelle ich fest, dass
eine „gesunde“ sich fortpflanzende Kolonie, ihr Koloniewachstum auf der einen Seite mittels der
Produktion von langlebigen Arbeitern sichern kann, während ein Investieren in funktionelle
Immunität auf der anderen Seite das epidemische Risiko gering halten kann.
Basierend auf meinen Ergebnissen entwerfe ich ein evolutionäres Szenario bei dem der
Parasitendruck gegenüber Koloniegrösse und Arbeiterlanglebigkeit als eine Selektive Kraft hin zu
der Evolution von höheren sozialen Strukturen sein kann. Dieses hypothetische Rahmenwerk
basierend auf den Ergebnissen der Kapitel meiner Doktorarbeit schlägt eine kritische Rolle von
Parasitismus in der Evolution von Koloniestrukturen vor. Gleichzeitig bereitet es den Weg, ja
geradezu verlangt nach nachfolgenden Untersuchungen.
5
1.1 GENERAL INTRODUCTION
Animals live in groups when the benefits of a social lifestyle outweigh its costs. Social insects,
the ants, bees, wasps, and termites, for example, live in some of the largest animal societies and
therefore are subject to a wide range of costs and benefits. A single colony can consist of several
million individuals (Janzen 1973). Social insect colonies populate virtually all continents of our
planet (Suarez et al. 2001), and E.O. Wilson (1971) estimated that 1015 individual ants live on the
earth at any time. This spectacular ecological and evolutionary success of social insects is
typically assigned to their social lifestyle (Wilson 1971), which may have
increased their
efficiency in brood care, foraging, and anti-predator defence. At the same time, a social life style
should offer amenable conditions for parasites (Schmid-Hempel 1998). In a world full of parasites,
the risk of living in groups is additionally increased when individuals live at high densities (Pie et
al. 2004), as density increases the probability of encountering pathogens because individuals are
bound to interact more frequently, which favours the transmission to naïve (uninfected) nest
mates (Otterstatter & Thomson 2007; Naug 2008). Furthermore, in typical social insect colonies
individuals are also closely related to one another (Hölldobler & Wilson 1990), which creates a
disproportionally homogenous genotypic environment, facilitating the spread of the parasite
through the colony still further (Shykoff & Schmid-Hempel 1991; Liersch & Schmid-Hempel
1998; Hughes & Boomsma 2004). The observation that antibacterial activity appears to increase
with the level of sociality (Stow et al. 2007) may reflect this basic dilemma.
Social life and parasite defence
To counter the increased risk of parasites associated with their social lifestyle, social insects,
along with other social species, have evolved adaptations that complement the individual immune
defence system and can efficiently contain an epidemic at the colony level (Cremer et al. 2007;
Wilson-Rich et al. 2009; Parker et al. 2011). These so-called social defences (social because they
are generally not available to solitary species), act at all steps of colony infection; from parasite
uptake, to parasite invasion, establishment and spread in the nest. To reduce the risk of primary
colony infection, for example, workers may simply avoid contact with locations or individuals
6
rich in parasites (a relatively simple, but effective behaviour that is also observed in solitary
species) (Orr et al. 1995; Moore 2002). Furthermore, upon infection, workers can be excluded
from the colony. Specialized guards, together with other workers, are known to attack or exclude
infected honeybee workers at the entrance of the hive (Drum & Rothenbuhler 1985). Within the
colony, the transmission of the parasite and its establishment in the nest may be controlled by
various defences mediated at the colony level (Cremer et al. 2007). In this context, a number of
sanitary behaviours have been described that might be of particular importance for preventing
parasite spread within the colony, including the removal of hazardous material and carcasses
(Julian 1999; Hart & Ratnieks 2002), removal and inactivation of infectious spores from either the
own body (self –grooming), or from the body of exposed nest mates (allo-grooming) (SchmidHempel 1998; Rosengaus et al. 1998; Reber et al. 2011) and the disinfection of brood with poison
produced by the ants themselves (Tragust et al. 2013). Anti-bacterial and anti-fungal compounds
as in ants (resin) (Chapuisat et al. 2007) and honeybees (mold) (Gilliam et al. 1988) are also
imported from outside the colony and have prophylactic effects. A recent study described the use
of a very low dose of a parasite to generate the effect of social vaccination (Konrad et al. 2012),
up-regulating immune function, and thus better protection of the colony from secondary
infections. However, as the activation of the immune system is costly (Moret 2000), social
vaccination may be adaptive only under high and persistent parasite pressure. Nevertheless, the
extraordinary abundance and complexity of behavioural adaptation are suggestive of the
importance of parasites during social evolution.
Age polyethism with a note on worker longevity
‘Age polyethism’ is a hallmark of many social insect colony organizations and refers to the
delegation of non-reproductive tasks in relation to age e.g. (Calderone 1995; Seid & Traniello
2006; Mersch et al. 2013). In its purest form, workers perform a specific task at a distinct age in a
defined spatial area. More specifically and with respect to diseases, young workers typically
perform tasks in disease-privileged areas of the colony, such as assisting the queen or brood in the
brood or queen chambers, whereas older workers are more likely to be foragers or engage in other
7
tasks at the periphery or outside the nest. Such a system may reduce interactions, with old
workers being socially more isolated, and as a consequence introduces a spatial and temporal
barrier to a spreading disease (Schmid-Hempel & Schmid-Hempel 1993).
Age-related changes in behaviour, among other things, raise the question of worker life span or
longevity. In general, long-lived workers should be selected for as they allow for large colony size
(Carey 2001). Larger colony size in turn usually correlates with efficiency and higher colony
reproductive output (Muller & Schmid-Hempel 1992; Palmer 2004; Lopez-Vaamonde et al. 2009).
At the same time, long-lived and therefore, on average, the presence of more older workers in the
colony, should impose an increased threat of parasites and parasite transmission to the colony.
This is because older workers are not only more likely to have encountered parasites but may also
be less protected as the immune system senesces over an animal's life (Doums et al. 2002; Moret
& Schmid-Hempel 2009; Stanley 2012). One part of this thesis asks how worker longevity affects
parasite transmission and whether age-based polyethism may mitigate this threat by restricting the
interactions of older, infection prone, workers.
The organization of bumblebees
Work in this thesis was done with bumblebees and their parasites. Bumblebee colonies are headed
by a single queen, typically singly mated (Schmid-Hempel & Schmid-Hempel 2000), and consist
of up to ca. one hundred workers. After the colony has been founded by the queen and established
with the emergence of the first brood, workers tend to the queen's eggs and engage in nonreproductive tasks such as brood care, foraging, and guarding. Bumblebees have, at best, only
partly evolved social defences against parasites. Examples are the (weak) tendency of
(trypanosome-) infected workers to leave the nest, a behaviour known also from ants (KornerNievergelt 2003), or the observation that workers parasitized by conopid parasitoid flies seek out
cold temperatures and thus leave the nest at night so as to suppress parasite development in their
body (Müller & Schmid-Hempel 1993). Prolonged survival in this case should add to colony
efficiency.
8
Bumblebees also show no clear-cut division of labour according to age (Cameron 1989; Jandt et
al. 2009) but task affinity based on body size is common. In particular, small workers are more
likely to perform hive tasks whereas bigger workers forage or guard the nest (Goulson et al. 2002;
Yerushalmi et al. 2006). In terms of social defences, bumblebees, therefore, may depend mainly
on a functional individual immune system to counter density effects in the face of parasites. Also,
immune senescence together with weak age-related adaptations should render bumblebee parasite
defences sensitive to a demographic shift towards old workers.
How to build a healthy colony
In this thesis I used the bumblebee Bombus terrestris L. and its trypanosme gut parasite Crithidia
bombi (Lipa & Triggiani 1988) to investigate the importance of worker age on the infection and
transmission of this ecologically relevant parasite. B. terrestris is a primitive eusocial insect.
‘Primitive’ in the sense that there is no clear-cut task division and reproduction is virtually
confined to queens. B. terrestris inhabits temperate zones of the Palaearctic and follows an
annual life cycle (with partly bivoltine cycles in warmer areas such as the Mediterranean). After
hibernation, mated queens found colonies in spring on their own, which subsequently grow in
number of workers until, in late summer, the colony eventually reproduces by producing sexuals
(males and gynes, the daughter queens) after which the colony collapses. C. bombi infects per os,
is picked up via contaminated flowers (Durrer & Schmid-Hempel 1994) and spreads through the
colony by infective cells shed in the faeces of workers. Costs to infection are mainly paid by
young queens that – if infected- often fail to found new colonies (in roughly half of the cases),
whereas increased mortality in workers was only observed under harsh conditions (i.e. under food
restriction). The parasite cannot live outside its host and thus should thus be under selection to
infect young queens, which are the only individuals going into and surviving hibernation.
From an evolutionary perspective, ‘healthy’ is a colony state that allows survival and
reproduction such as to leave as many daughter queens as possible that themselves found new
colonies in the next year. As daughter queen production correlates with colony size (Muller &
Schmid-Hempel 1992), 'health' should allow achieving large colony size and few infected
9
daughter queens. An insight of how worker lifespan in B. terrestris colonies may relate to colony
health is given in the following chapters. Each chapter will introduce the respective background
for the study and discuss the consequences of its findings. The General Discussion attempts to
summarize the thesis' results and synthesize them more conceptually.
10
1.2 THESIS OUTLINE
Chapter 1. General Introduction
Chapter 2. Colony pace: A life-history trait affecting social insect epidemiology.
For a given colony size, the colony may produce many short-lived workers or few long-lived
workers. In this Chapter I investigate whether differences in colony pace reflect a life-history
strategy against parasite transmission. I infected groups of bumblebees with the infectious gut
parasite Crithidia bombi and experimentally manipulated birth and death rates to mimic slow and
fast pace. I showed that fewer workers were infected in colonies with a higher turnover rate. Few
infected workers means less infective cells and a lowered probability of daughter queens being
infected. I therefore propose that a high worker turnover rate, hence many short-lived workers can
act as a strategy of defence against a spreading disease.
Chapter 3. Worker longevity and its relation to colony resistance against infectious parasites
in a social insect
Many short-lived or few long-lived workers may be of different ‘immune quality’. Here, we test a
quality/quantity tradeoff with respect to resistance to C. bombi. I first monitored the relationship
between worker lifespan and the number of workers a colony produced. In a second experimental,
I infected workers and measured the standing expression of bumblebee immune genes. I found
that immunity in bumblebees indeed is linked to longevity such that individuals from colonies
that produced long-lived workers were phenotypically better-protected. Colonies with short-lived
workers however were small colonies, which may be maintained by the success of short-lived
workers in preventing an epidemic within the nest.
Chapter 4. Worker longevity in a primitively eusocial insect: Immune senescence does not
affect worker resistance to an ecologically relevant parasite.
Colonies with long-lived workers inevitably have more old workers at any time. This Chapter
investigates the age effect on C. bombi resistance over a workers lifespan. I first tested age related
resistance, then I compared the resulting transmission potential to behavioural adaptations, such
as worker activity and distance to naive workers, which are likely to influence parasite spread. I
11
found that immune senescence may not necessarily affect resistance to C. bombi but that age
effects should be relevant for the spread of the parasite and may be mediated by behaviour rather
than by infection intensity.
Chapter 5. General discussion and a hypothetical scenario: How parasite pressure may act
as a driving force of colony social organization
I summarize my thesis findings and suggest a scenario where parasite pressure selects against
colony size and worker longevity selects for the evolution of complex social structure.
12
References
Calderone, N. W. 1995. Temporal division of labor in the honey bee, Apis melliferaA developmental
process or the result of environmental influences? Canadian Journal Of Zoology-Revue Canadienne
De Zoologie, 73, 1410–1416.
Cameron, S. A. 1989. Temporal Patterns of Division of Labor among Workers in the Primitively Eusocial
Bumble Bee, Bombus griseocollis (Hymenoptera: Apidae) 1). Ethology, 80, 137–151.
Carey, J. R. 2001. Demographic mechanisms for the evolution of long life in social insects. Experimental
Gerontology, 36, 713–722.
Chapuisat, M. M., Oppliger, A. A., Magliano, P. P. & Christe, P. P. 2007. Wood ants use resin to
protect themselves against pathogens. Proceedings. Biological sciences / The Royal Society, 274,
2013–2017.
Cremer, S., Armitage, S. A. O. & Schmid-Hempel, P. 2007. Social immunity. Current Biology, 17,
R693–R702.
Doums, C., Moret, Y., Benelli, E. & Schmid-Hempel, P. 2002. Senescence of immune defence in
Bombus workers. Ecological Entomology, 27, 138–144.
Drum, N. H. & Rothenbuhler, W. C. 1985 July 31. Differences in non-stinging aggressive responses of
worker honeybees to diseased and healthy bees in May and July - I. journal of apicultural research,
24, 184–187.
Durrer, S. & Schmid-Hempel, P. 1994. Shared use of flowers leads to horizontal pathogen transmission.
Proceedings Of The Royal Society Of London Series B-Biological Sciences, 258, 299–302.
Gilliam, M., Taber, S., III, Lorenz, B. J. & Prest, D. B. 1988. Factors affecting development of
chalkbrood disease in colonies of honey bees, Apis mellifera, fed pollen contaminated with
Ascosphaera apis. Journal Of Invertebrate Pathology, 52, 314–325.
Goulson, D., Peat, J., Stout, J. C., Tucker, J., Darvill, B., Derwent, L. C. & Hughes, W. O. H. 2002.
Can alloethism in workers of the bumblebee, Bombus terrestris, be explained in terms of foraging
efficiency? Animal Behaviour, 64, 123–130.
Hart, A. G. & Ratnieks, F. L. 2002. Waste management in the leaf-cutting ant Atta colombica. Behavioral
Ecology, 13, 224–231.
Hölldobler, B. & Wilson, E. O. 1990. The Ants. Harvard University Press.
Hughes, W. O. H. & Boomsma, J. J. 2004. Genetic diversity and disease resistance in leaf-cutting ant
societies. Evolution, 58, 1251–1260.
Jandt, J. M., Huang, E. & Dornhaus, A. 2009. Weak specialization of workers inside a bumble bee
(Bombus impatiens) nest. Behavioral Ecology and Sociobiology, 63, 1829–1836.
Janzen, D. H. 1973. Evolution of polygynous obligate acacia-ants in western Mexico. The Journal of
Animal Ecology, 727–750.
Julian, G. 1999. Undertaking specialization in the desert leaf-cutter ant Acromyrmex versicolor. Animal
Behaviour, 58, 437–442.
Konrad, M., Vyleta, M. L., Theis, F. J., Stock, M., Tragust, S., Klatt, M., Drescher, V., Marr, C.,
Ugelvig, L. V. & Cremer, S. 2012. Social transfer of pathogenic fungus promotes active
immunisation in ant colonies. PLoS Biology, 10, e1001300.
Korner-Nievergelt, P. 2003 July 28. Dynamics and consequences of immune sysytem activation in
13
individuals of Bombus Terrestris (L.) (Hymenoptera: Apidae). Diss. Swiss Federal Institute of
Technology Zurich,
Liersch, S. & Schmid-Hempel, P. 1998. Genetic variation within social insect colonies reduces parasite
load. Proceedings Of The Royal Society B-Biological Sciences, 265, 221–225.
Lipa, J. J. & Triggiani, O. 1988. Crithidia bombi sp. n. a flagellated parasite of a bumblebee Bombus
terrestris L. ( Hymenoptera, Apidae). Acta Protozoologica, 27, 287–290.
Lopez-Vaamonde, C., Raine, N. E., Koning, J. W., Brown, R M, Pereboom, J. J. M., Ings, T. C.,
Ramos-Rodriguez, O., Jordan, W. C. & Bourke, A. F. 2009. Lifetime reproductive success and
longevity of queens in an annual social insect. Journal Of Evolutionary Biology, 22, 983–996.
Mersch, D. P., Crespi, A. & Keller, L. 2013. Tracking Individuals Shows Spatial Fidelity Is a Key
Regulator of Ant Social Organization.
Moore, J. 2002. Parasites and the Behavior of Animals. Oxford University Press.
Moret, Y. 2000. Survival for Immunity: The Price of Immune System Activation for Bumblebee Workers.
Science, 290, 1166–1168.
Moret, Y. & Schmid-Hempel, P. 2009. Immune responses of bumblebee workers as a function of
individual and colony age: senescence versus plastic adjustment of the immune function. Oikos, 118,
371–378.
Muller, C. B. & Schmid-Hempel, P. 1992. Correlates of reproductive success among field colonies of
Bombus lucorum: the importance of growth and parasites. Ecological Entomology, 17, 343–353.
Müller, C. B. & Schmid-Hempel, P. 1993. Exploitation of cold temperature as defence against parasitoids
in bumblebees. Nature, 363, 65–67.
Naug, D. 2008. Structure of the social network and its influence on transmission dynamics in a honeybee
colony. Behavioral Ecology and Sociobiology, 62, 1719–1725.
Orr, M. R., Seike, S. H., Benson, W. W. & Gilbert, L. E. 1995. Flies suppress fire ants. Nature, 373,
292–293.
Otterstatter, M. C. & Thomson, J. D. 2007. Contact networks and transmission of an intestinal pathogen
in bumble bee (Bombus impatiens) colonies. Oecologia, 154, 411–421.
Palmer, T. M. 2004. Wars of attrition: colony size determines competitive outcomes in a guild of African
acacia ants. Animal Behaviour, 68, 993–1004.
Parker, B. J., Barribeau, S. M., Laughton, A. M., de Roode, J. C. & Gerardo, N. M. 2011. Nonimmunological defense in an evolutionary framework. Trends in ecology & evolution, 26, 242–248.
Pie, M. R., Rosengaus, R. B. & Traniello, J. 2004. Nest architecture, activity pattern, worker density and
the dynamics of disease transmission in social insects. Journal of Theoretical Biology, 226, 45–51.
Reber, A., Purcell, J., Buechel, S. D., Buri, P. & Chapuisat, M. 2011. The expression and impact of
antifungal grooming in ants. Journal Of Evolutionary Biology, 24, 954–964.
Rosengaus, R. B., Maxmen, A. B., Coates, L. E. & Traniello, J. F. A. 1998. Disease resistance: a benefit
of sociality in the dampwood termite Zootermopsis angusticollis (Isoptera: Termopsidae). Behavioral
Ecology and Sociobiology, 44, 125–134.
Schmid-Hempel, P. 1998. Parasites in Social Insects. Princeton University Press.
Schmid-Hempel, P. & Schmid-Hempel, R. 1993. Transmission of a pathogen in Bombus terrestris, with a
14
note on division of labour in social insects. Behavioral Ecology and Sociobiology, 33, 319–327.
Schmid-Hempel, R. & Schmid-Hempel, P. 2000. Female mating frequencies in Bombus spp. from
Central Europe. Insectes Sociaux, 47, 36–41.
Seid, M. A. & Traniello, J. F. A. 2006. Age-related repertoire expansion and division of labor in Pheidole
dentata (Hymenoptera: Formicidae): a new perspective on temporal polyethism and behavioral
plasticity in ants. Behavioral Ecology and Sociobiology, 60, 631–644.
Shykoff, J. A. & Schmid-Hempel, P. 1991. Parasites and the Advantage of Genetic Variability within
Social Insect Colonies. Proceedings Of The Royal Society B-Biological Sciences, 243, 55–58.
Stanley, D. 2012. Aging and immunosenescence in invertebrates. Invertebrate Survival Journal, 9, 102–
109.
Stow, A., Briscoe, D., Gillings, M., Holley, M., Smith, S., Leys, R., Silberbauer, T., Turnbull, C. &
Beattie, A. 2007. Antimicrobial defences increase with sociality in bees. Biology Letters, 3, 422–424.
Suarez, A. V., Holway, D. A. & Case, T. J. 2001. Patterns of spread in biological invasions dominated by
long-distance jump dispersal: Insights from Argentine ants. Proceedings of the National Academy of
Sciences, 98, 1095–1100.
Tragust, S., Mitteregger, B., Barone, V., Konrad, M., Ugelvig, L. V. & Cremer, S. 2013. Ants Disinfect
Fungus-Exposed Brood by Oral Uptake and Spread of Their Poison. Current Biology, 23, 76–82.
Wilson, E. O. 1971. The insect societies. Belknap Press.
Wilson-Rich, N., Spivak, M., Fefferman, N. H. & Starks, P. T. 2009. Genetic, Individual, and Group
Facilitation of Disease Resistance in Insect Societies. Annual Review of Entomology, 54, 405–423.
Yerushalmi, S., Bodenhaimer, S. & Bloch, G. 2006. Developmentally determined attenuation in circadian
rhythms links chronobiology to social organization in bees. Journal of Experimental Biology, 209,
1044–1051.
15
2. COLONY PACE: A LIFE-HISTORY TRAIT AFFECTING SOCIAL INSECT
EPIDEMIOLOGY
§
Abstract
In social insects, the worker turnover rate (its ‘pace’) shows considerable variation between
colonies. This has epidemiological consequences for parasites because in `fast-paced` colonies,
with short-lived workers, the time of residence in a given host will be reduced. Here we
investigate whether differences in colony pace reflect a life-history strategy against parasite
transmission. We infected bumblebees (Bombus terrestris) with the infectious gut parasite
Crithidia bombi and experimentally manipulated birth and death rates to mimic slow and fast
pace. Fewer workers and, importantly, fewer last generation workers were infected in colonies
with a higher turnover rate. This suggests increased fitness in “fast-paced colonies, as daughter
queens exposed to few infective cells are less likely to be infected and have a higher chance of
founding their own colonies. We therefore propose that a high worker turnover rate acts as a
strategy of defence against a spreading disease; parasitism therefore likely drives the variation in
worker lifespan commonly observed in social insects.
§
A version of this Chapter is in preparation as S. D. Buechel & P. Schmid-Hempel.
16
Introduction
The evolutionary transition from a solitary to a social lifestyle is considered a main reason for the
spectacular ecological success of social insects (Wilson 1971; Szathmáry & Smith 1995). At the
same time, social living may impose several disadvantages. Infectious parasites, for example, may
spread more easily when individuals live together at high densities, are closely related, and
interact frequently with one another - conditions that are characteristic for social insect colonies
(Shykoff & Schmid-Hempel 1991b; Schmid-Hempel 1998; Naug & Camazine 2002; Pie et al.
2004; Otterstatter & Thomson 2007). It is therefore conceivable that social insects should have
evolved a broader and more efficient immune system compared to their solitarily living
counterpart. This expectation is not confirmed by the available data, as social insects in fact seem
to have fewer immune genes than solitary insect species (Evans et al. 2006). However, social
insects may have alternative or complementary ways to defend themselves against diseases and
the transmission of pathogens. Among these non-immunological defences are those that are based
on altruistic behaviour (e.g. mutual grooming to remove parasite spores), collective decisionmaking (e.g. increase in nest temperature), and other collective actions that reduce parasite loads
or the impact of infection [reviewed in (Cremer et al. 2007)]. This so-called ‘social immunity’ can
effectively contain an infection at the colony level and so help to sustain colony growth, survival
and reproduction.
One aspect that has not been well covered so far is that social insects may also mitigate fitness
losses caused by parasitic infection by adjusting life-history parameters such as growth,
reproduction and survival (Moret & Schmid-Hempel 2004) see also review in (Agnew et al. 2000).
Life-history theory suggests that individuals should alter the timing and investment in these traits
to maximize their lifetime fitness in a given environment (Stearns 1992). The presence or absence
of a parasite arguably is a crucial environmental factor contributing to an organism’s optimal
pattern of resource allocation (Perrin et al. 1996; Sorci et al. 1996). A life-history strategy that
accelerates the host’s schedule of reproduction, for instance, should be favoured when parasites
shorten the remaining hosts lifespan and curtail it's reproductive capacity (Agnew et al. 2000;
17
Moret & Schmid-Hempel 2004). Evidence for a life-history response to parasitism maintained
throughout the evolution of sociality however, remains scarce.
Here we investigate a life-history parameter that is unique to social insects - the turnover of
individuals within a colony as given by worker lifespan and birth rate of new workers, defining a
colony pace. In most social insects, typically, only the queen reproduces, while the worker caste is
sterile (Hölldobler & Wilson 1990). Thus, worker turnover is akin to the turnover of cells in a
body, and an individual life takes a different meaning from that in most other organisms. With
this approach, the growth and reproductive success of a social insect colony depends on the effort
of its workforce, which is a function of the total number of worker present at any one time and is
thus given by the worker turnover rate. In bumblebees, only if a critical worker number is attained
does the queen reproduce and the colony thus reaches sexual maturity (Muller & Schmid-Hempel
1992).
In the bumblebee, Bombus terrestris, some colonies seem to have higher worker turnover rates
than others (Regula Schmid-Hempel pers. comm., and pers. obs.). In "fast-paced" colonies
workers die younger, but new workers are produced more frequently than in "slow-paced"
colonies - even under otherwise similar conditions (e.g. in the laboratory). For an infectious
parasite, an increased production of short-lived workers has epidemiological consequences, since
the expected time of residence in a given host will be reduced. Hence, transmission events must
become more frequent and, according to standard theory, short host lifespan should select for
rapid replication in parasites, which is typically associated with higher virulence and greater
infectivity.
To test this idea, we manipulated worker turnover rate in experimental colonies of B. terrestris to
measure its effect on the spread of the infectious parasite, Crithidia bombi. Moreover, we predict
that a decrease in residence time of infection and more frequent transmissions reduces the
genotypic diversity of the parasite by eliminating non-suitable strains (see (Yourth & SchmidHempel 2006). A reduction in parasite diversity at time of colony reproduction is important
18
because high strain diversity increases the likelihood of daughter queens becoming infected
(Ulrich et al. 2011). Eventually, at the end of a colony cycle, fast-paced colonies may thus have
fewer infected workers with fewer persisting strains, which may result in fewer daughter queens
being infected and thus higher colony founding success in the next generation.
Methods
Parasites and insects
The bumblebee (Bombus terrestris) is a eusocial insect with annual lifecycle. In spring, the singly
mated queens, after emerging from hibernation, found colonies that grow in numbers of workers
until the colony eventually reproduces in late summer; only the mated daughter queens undergo
hibernation to start next generation colonies. The trypanosomatid gut parasite, Crithidia bombi, is
common in most populations (Shykoff & Schmid-Hempel 1991a; Salathé & Schmid-Hempel
2011; Ruiz-González et al. 2012). It cannot live outside the host (Schmid-Hempel et al. 1999) and
hence underlies strong selective pressure to infect daughter queens - the only individuals that
survive the winter. But at least in our study area, only 10 -15% of spring queens ( daughter
queens) are infected with C. bombi. Colonies contract the infection either from other colonies via
shared use of flowers or via faeces deposited in the nest. The parasite is genetically highly diverse
(Schmid-Hempel & Reber Funk 2004; Salathé & Schmid-Hempel 2011). The time between
established infection and subsequent transmission ranges between two days for fast developing
genotypes and 10 days for slower strains (Schmid-Hempel & Schmid-Hempel 1993). C. bombi
increases worker mortality under stressful conditions and substantially reduces the colony
founding success of daughter queens (Brown et al. 2003).
Collection and culturing
B. terrestris queens emerging from hibernation were collected in spring 2010 from two distinct
populations in Switzerland (near Aesch BL; and Neunforn TG) and allowed to initiate colonies in
the laboratory. As soon as the first workers emerged, colonies were transferred to circular perlite
19
nests (Pomeroy & Plowright 1980). Seventeen uninfected colonies were reared to reach a
sufficient number of workers so as to serve as donor colonies for bees used in this experiment. All
bees were kept under standardized laboratory conditions (28± 2 °C, 60% RH) with constant red
light illumination and pollen and sugar water provided ad libitum.
The parasite, C. bombi was sampled from faeces of naturally infected queens that originated from
the same two populations in 2010. Single infective C. bombi cells were then isolated with a
fluorescence-activated cell-sorter (FACS) and maintained clonally in liquid medium (Salathé et al.
2012). The six strains used in this experiment had distinct multi-locus genotypes at five
polymorphic microsatellite loci and could therefore be easily distinguished by genetic markers.
Infection and experimental manipulation of the worker turnover rate
Workers from 17 donor colonies were assembled into experimental groups housed in boxes (with
two treatments each, n = 34 boxes in total), and exposed to high or low worker turnover rates. The
experiment started by placing four naïve (uninfected) workers and six workers each infected with
a distinct C. bombi strain (six strains in total per box) into a given box. Each experimental group
consisted of 10 workers at all times. A sketch of the experimental design is given in Figure 1.
(A) High worker turnover rate
Legend:
= infected
=10
= naïve (N =17)
Treatment start
Time (days)
= 10
(N =17)
(B) Low worker turnover rate
Figure 1. Experimental design of the study. We induced different worker turnover rates in initial groups
of 6 infected and 4 naïve workers housed in boxes (left, see legend). To subsequently create a high
worker turnover rate, two random workers from the group were exchanged against two naïve worker
(circular arrows) every three days. To create low worker turnover, workers were exchanged every 6
days only. The prevalence of infected workers in each group was measured on days 14, 30, and 38
(arrows). Six naïve workers, mimicking workers of the last cohort were added to the respective groups
at day 30.
20
For the infections, and to mimic a natural colony background, four (non-age controlled) workers
were selected at random from each donor colony (i.e. a total of 68 workers), starved for two hours
and presented with 10 µl inoculum containing 20'000 cells of a single C. bombi clone in medium
and sugar water (1:1 ratio). Workers that did not ingest the cocktail after 60 min were excluded
from the experiment to ensure primary infections and viability of the infectious cells. This
procedure was repeated six times for each donor colony (and for the six trains) always infecting
four workers with one out of six C. bombi clones (i.e. resulting in 24 workers per donor colony;
408 exposed workers in total). Bees were then kept individually to prevent cross infection and
eight days later their faeces were visually checked for the presence of C. bombi cells, confirming
their infection status. Subsequently, two (out of the four) successfully infected workers for each
of the six C. bombi strains were randomly assigned to one out of two experimental groups (low,
high turnover rate) and each experimental group was completed with four naïve workers from the
respective donor colony to yield the total of ten workers per experimental group (a colony x
turnover combination); each box thus had an infection prevalence of 60% of hosts. Four donor
colonies were found to be completely resistant to either one or two strains; the respective
experimental groups (both low and high turnover) thus started the experiment with an infection
proportion of 50% or 40%.
To create a high turnover rate, every third day, we randomly exchanged two workers from the
group against two randomly chosen naïve workers from the same donor colony. For the low
turnover rate, two randomly chosen naïve workers from the experimental group were exchanged
only every sixth day. This experimental protocol mimicked the “birth” and "death" rate in a B.
terrestris colony in the sense that workers are "reared" at different rates, get exposed to infections,
and are removed from the pool of infection at an arbitrary time due to uncorrelated, external
events. The time intervals were chosen based on the observed differences in the average worker
lifespan among colonies of B. terrestris. In the present case, average worker life spans of the
colony with the longest living worker are about twice the average life span of the colony with the
shortest living workers (Chapter 3). Furthermore, worker life span in the field is in the order of 4
21
weeks, which roughly corresponds to the low turnover treatment used here. We exchanged
workers from experimental groups following this treatment schedule for a total of 30 days and
sampled all workers on day 38. This schedule thus allowed workers to live up to the expected 4-6
weeks, which is typical for natural conditions (Goulson 2003). The proportion of infected workers
over time was monitored by visually checking the faeces of all 10 workers in the treatment groups
after 14, 30 and 38 days.
On day 30, an additional six-naïve workers from the respective donor colonies were added to each
treatment group (n=30 for high and low turnover rate). These individually marked workers were
introduced to mimic the last generation of workers in the colony (late workers) that are
responsible for raising sexuals (daughter queens and males). At the same time, this cohort of
workers represents the parasite environment to which the emerging daughter queens are exposed.
Late workers were exposed to the treatment groups for eight days, which is about the time young
queens spend in the nest before they leave for the mating flight [see (Bourke 1997)]. As colonies
produce the last workers together with first males, which leave the colony before eventually
queens emerge, workers were no longer exchanged, once the last cohort was added. All workers
were sampled on day 38 and their gut tissue was genotyped to determine the number and identity
of C. bombi strains that successfully infected bees exposed to different worker turnover rates
(Figure 1). For groups were all workers provided faeces samples (n=20 for high and low turnover
rate) we additionally measured the average concentration of infective cells per treatment group
with a haemocytometer (Neubauer improved counting chamber).
Genotyping of infections
We dissected out the whole gut of all workers and extracted genomic DNA with a Qiagen
DNAeasy Blood & Tissue Kit (Qiagen, Hilden, Germany) following the protocol for purification
of total DNA from animal tissues. C. bombi infection was detected by the amplification and
visualization (1.5% agarose gel) of a part of the C. bombi Cyt b gene with Crithidia-specific
primers (CB- Cytb2-F and CB-Cytb2-B, described in (Schmid-Hempel & Tognazzo 2010). The C.
22
bombi strains present in successful infections were thereafter genotyped at the C. bombi
microsatellite markers Cri4, Cri2F10,Cri4G9, Cri16 and Cri1B6 (Schmid-Hempel & Reber Funk
2004) following the protocol in (Koch & Schmid-Hempel 2011) and the amplification product
was run on a MegaBACE sequencer (GE Healthcare, Glattburgg, Switzerland).
The
microsatellite peaks were scored twice with MegaBACE Fragment Profiler software v1.2 by an
observer blind to the treatment groups.
Analysis and statistics
Determinants for the number of infected workers over time were analysed using a generalized
linear-mixed model (GLMMs) with a binomial error structure and logit-link function. Because
workers did not provide a faeces sample every time the infection status was assessed, we used the
number of infected/ total number of sampled workers as independent response variable. In the full
model we included the experimental groups (high versus low turnover rate) and the three time
points at which we assessed the number of infected workers per group (14 days, 30 days, 38 days)
as fixed factors while the donor colony of each experimental group and the individual groups
were used as random factor. Note that including each individual group as random factor accounts
for repeated measurement over time (14 days, 30 days, 38 days). Model selection was done by
backwards elimination of non-significant terms (Crawley 1993). If interaction terms are not
presented, they did not have a significant effect. For every time point measured, we conducted
post hoc comparisons based on paired sampled t-tests on the model residuals and sequential
Bonferroni-correction (Holm 1979) to analyse differences in the number of infected workers
between groups with a high and low worker turnover rate. We used an analogous model to test the
influence of a high worker turnover rate on the prevalence of infected last cohort workers. The
model included the experimental groups (high versus low turnover rate) as fixed factor and the
donor colony of the experimental groups as single random factor, because the infection status of
last cohort workers was assessed only at day 38. Identical model parameters were applied to test
the number of persisting C. bombi strains in high versus low turnover groups, except that the
number of strains after 38 days were used as dependent and the number of strains at treatment
23
start as independent variable, because not all experimental groups started the experiment with six
infected workers and thus six C.bombi strains. The distribution of the C. bombi strains that
successfully infected workers exposed to a high versus low turnover rate was analysed using
Fishers exact test and the average infection load of workers in high versus low turnover groups
was tested with a Wilcoxon signed rank test for paired samples. All statistical analyses were
carried out in IBM SPSS Statistics version 19.0 or R statistical software version 2.15.2 (R Core
Team 2012). Model assumptions were met for all linear models presented.
Results
A high turnover rate in a group of bumblebee workers significantly reduced the spread of the
infectious parasite C. bombi, resulting in an average infection prevalence of Prev = 0.45, for
workers exposed to a high turnover rate, and Prev = 0.6 for workers exposed to a low turnover
rate (GLMMs; F
1,98
= 13.517, p < 0.0001). In 14 out of 17 paired experimental groups (high vs.
low turnover rate from the same donor colony), fewer workers were infected when exposed to a
high turnover rate. In two of the remaining pairs, there were fewer infected worker in the group
with the low turnover rate and in one pair workers were equally infected regardless of the
turnover rate they were exposed to. The infection was lost after 30 and 38 days in three groups
exposed to a high turnover rate, while in two groups with a low turnover rate all workers became
infected.
Even though the differences in infection proportion between experimental groups slightly varied
over time (Figure 2, GLMMs: F
2,98
= 3.329, p = 0.04), the high turnover rate reduced the
proportion of infected workers at all times measured (paired t-tests; 14 days: t16 = -2.234, p =
0.04; 30 days: t16 = -3.562, p = 0.003, 38 days: t16 = -2.569, p = 0.021).
24
Proportions of workers
Days
Figure 2. The proportion of infected workers in groups of B. terrestris workers with high and low
turnover rate (n=34), measured 14 days, 30 days and 38 days post-experimental infection at time ‘0’.
Shown are the estimated marginal means (±SE) from a model with turnover rate and time as fixed
factors, and colony and individual group as random factors (see text). Note that workers were
exchanged for 30 days while the final proportion of infected workers was measured at day 38.
Importantly, a higher worker turnover rate also resulted in fewer last cohort workers contracting
the infection (Figure 3, GLMMs: F
1,28
= 6.136, p = 0.02), and this difference in infection
prevalence between groups with different worker turnover rates was not due to different numbers
of C. bombi strains infecting either treatment (Figure 4, GLMMs: F
1,28
= 1.258, p > 0.1). We
further did not find that different strains infected workers with a high turnover rate when
compared to workers with a low turnover rate (Fishers exact test: n= 12, p > 0.1). Finally, the
average infection load in groups with a high turnover rate did not differ from the infection load of
workers exposed to a low turnover rate (paired Wilcoxon signed-rank test: Z= -1.274, n= 20, p >
0.1).
25
Proportion of infected last cohort workers
Low Pace (15)
High Pace (15)
Proportion of strains
Figure 3. The proportion of infected last cohort workers after they were exposed for 8 days to worker
groups with a high and low turnover rate (GLMMs: F 1,28 = 6.136, n= 30, p = 0.02). Bars represent the
estimated marginal means (±SE) derived from a model with e turnover rate as fixed factor, and colony
as random factor (see text).
High Pace (15)
Low Pace (15)
Figure 4. Percentage of different C. bombi strains (shades of grey, letters A-F) present in groups of B.
terrestris workers (n= 30) with a high and a low worker turnover rate, 38 days post-experimental
infection.
Discussion
In this study we aimed to understand the consequences of different worker turnover rates for a
spreading disease. We found that fewer workers and, most importantly, fewer workers that
mimicked the last cohort of workers were infected in groups with a higher worker turnover rate.
Few infective cells circulating in the colony therefore should result in fewer infected daughter
queens at the end of the colony cycle. Contrary to expectations, we did not find that, at the end of
the experiment, fewer parasite strains infected workers exposed to a high turnover rate, nor did we
find that different strains infected bumblebee workers exposed to different turnover rates. A high
turnover rate of workers did also not increase the infection intensity of C.bombi. Our results
26
therefore suggest that a high worker turnover rate can be a life-history strategy that protects the
colony against an infectious parasite.
The main virulence effect of C. bombi seems to be in the founding queens. C.bombi infection
greatly reduces the colony founding success in daughter queens of B. terrestris after they emerge
from hibernation the following spring (Brown et al. 2003). Furthermore, the number of C. bombi
strains circulating in the colony is an important factor predicting the infection rate of daughter
queens (Ulrich et al. 2011). A high turnover rate in our experiment did not reduce the number of
persisting strains but decreased the number of infective cells circulating in the colony. As
daughter queens can only be infected with exposure to a high dose (personal communication S.
Barribeau) a high worker turnover rate should result in fewer infected daughter queens. The
observed low infection prevalence among late workers moreover must be due to a short
transmission period of workers exposed to a high turnover rate.
In natural colonies, a high worker turnover rate would inevitable coincide with a demographic
shift towards younger workers. This may additionally lead to increased defence levels, since the
immune system of bumblebees declines with age (Doums et al. 2002; Moret & Schmid-Hempel
2009). Taken together, fast pace might prevent a parasite from spreading through a colony. In line
with results from a serial passage experiment (Yourth & Schmid-Hempel 2006), a high turnover
rate did not correlate with selection for increased virulence of C.bombi, if infection intensity can
be taken as a correlate. Note that C.bombi was exposed to the high turnover rate for more than
100 generations (Salathé et al. 2012).
Important with regard to worker turnover rate in Bombus is that the colony size early in the cycle
is a predictor of eventual colony size (worker number) at reproduction, which correlates closely
with reproductive success (Muller & Schmid-Hempel 1992). ‘Fast-paced’ colonies with an
increased death rate of workers, for this reason, should produce more workers over the colony
cycle to achieve similar sizes when compared to colonies with a low turnover rate. For given
resources or equal per capita workload hence, these more numerous workers reared in ‘fast27
paced’ colonies might be of lower quality (e.g. reduced immunity see Chapter 3) compared to
workers with a lower turnover rate. No evidence to this effect is as yet available in our system.
However, a comparable pattern of immune investment can be observed within species and among
castes of social insect. Short-lived male ants, for instance, have a lower immune response than the
longer-lived workers and queens (Baer et al. 2005).
While shifting the timing of reproduction requires collective decision-making in infected colonies
(Moret & Schmid-Hempel 2004), the worker turnover rate most likely depends on the queen. The
role of the queen as the pacemaker of the colony, as an aside, is an old idea going back to
Schneirla’s studies on army ants (Schneirla 1944) and was taken up repeatedly over the last
decades mainly focusing on the queen as the central force that initiates activity, coordinates work
or determines the timing of sexual reproduction (Reeve & Gamboa 1983; Premnath et al. 1995;
Jha et al. 2006; Souza & Prezoto 2011; Holland et al. 2013). Whether the observed increase in
worker turnover rate is the result of more efficient brood raising by workers, of how the queen
allocates her resources and hence might be a response to parasitism, or simply reflects the queen’s
high or low quality remains to be seen. For example, we would expect that for given resources a
queen produces either more short-lived workers at the expense of a reduced immune system, or
few long-lived workers that may have a good immune system to counter their prolonged period of
transmission.
Here we conclude that a high worker turnover rate could be a life-history strategy that protects a
bumblebee colony against the spread and transmission of an infectious parasite. This increase in
the turnover rate may be a direct response - plastic or selected - to parasitic infections, or may
indirectly be caused by factors that decrease worker longevity such as food shortage or queen
quality. Parasite-imposed selection for short-lived workers should in any case contribute towards
worker lifespan variation in social insects.
28
References
Agnew, P., Koella, J. C. & Michalakis, Y. 2000. Host life history responses to parasitism. Microbes and
infection / Institut Pasteur, 2, 891–896.
Baer, B., Krug, A., Boomsma, J. J. & Hughes, W. O. H. 2005. Examination of the immune responses of
males and workers of the leaf-cutting ant Acromyrmex echinatior and the effect of infection. Insectes
Sociaux, 52, 298–303.
Bourke, A. F. 1997. Sex ratios in bumble bees. Philosophical transactions of the Royal Society of London
Series B, Biological sciences, 352, 1921–1933.
Brown, M. J., Schmid-Hempel, R. & Schmid-Hempel, P. 2003. Strong context‐dependent virulence in a
host–parasite system: reconciling genetic evidence with theory. Journal Of Animal Ecology, 72, 994–
1002.
Crawley, M. 1993. GLIM for Ecologists. Wiley-Blackwell.
Cremer, S., Armitage, S. A. O. & Schmid-Hempel, P. 2007. Social immunity. Current Biology, 17,
R693–R702.
Doums, C., Moret, Y., Benelli, E. & Schmid-Hempel, P. 2002. Senescence of immune defence in
Bombus workers. Ecological Entomology, 27, 138–144.
Evans, J. D., Aronstein, K., Chen, Y. P., Hetru, C., Imler, J. L., Jiang, H., Kanost, M., Thompson, G.
J., Zou, Z. & Hultmark, D. 2006. Immune pathways and defence mechanisms in honey bees Apis
mellifera. Insect Molecular Biology, 15, 645–656.
Goulson, D. 2003. Bumblebees: Their Behaviour and Ecology. Oxford University Press,
Holland, J. G., Guidat, F. S. & Bourke, A. 2013. Queen control of a key life-history event in a eusocial
insect. Biology Letters, 9, 20130056–20130056.
Holm, S. 1979 March 30. A simple sequentially rejective multiple test procedure. Scandinavian Journal of
Statistics, 6, 65–70.
Hölldobler, B. & Wilson, E. O. 1990. The Ants. Harvard University Press.
Jha, S., Casey-Ford, R. G., Pedersen, J. S., Platt, T. G., Cervo, R., Queller, D. C. & Strassmann, J. E.
2006. The queen is not a pacemaker in the small-colony wasps Polistes instabilis and P-dominulus.
Animal Behaviour, 71, 1197–1203.
Koch, H. & Schmid-Hempel, P. 2011. Socially transmitted gut microbiota protect bumble bees against an
intestinal parasite. Proceedings Of The National Academy Of Sciences Of The United States Of
America, 108, 19288–19292.
Moret, Y. & Schmid-Hempel, P. 2004. Social life-history response to individual immune challenge of
workers of Bombus terrestris L.: a possible new cooperative phenomenon. Ecology Letters, 7, 146–152.
Moret, Y. & Schmid-Hempel, P. 2009. Immune responses of bumblebee workers as a function of
individual and colony age: senescence versus plastic adjustment of the immune function. Oikos, 118,
371–378.
Muller, C. B. & Schmid-Hempel, P. 1992. Correlates of reproductive success among field colonies of
Bombus lucorum: the importance of growth and parasites. Ecological Entomology, 17, 343–353.
Naug, D. & Camazine, S. 2002. The Role of Colony Organization on Pathogen Transmission in Social
Insects. Journal of Theoretical Biology, 215, 427–439.
29
Otterstatter, M. C. & Thomson, J. D. 2007. Contact networks and transmission of an intestinal pathogen
in bumble bee (Bombus impatiens) colonies. Oecologia, 154, 411–421.
Perrin, N., Christe, P. & Richner, H. 1996. On Host Life-History Response to Parasitism. Philosophical
transactions of the Royal Society of London Series B, Biological sciences, 75, 317.
Pie, M. R., Rosengaus, R. B. & Traniello, J. 2004. Nest architecture, activity pattern, worker density and
the dynamics of disease transmission in social insects. Journal of Theoretical Biology, 226, 45–51.
Pomeroy, N. & Plowright, R. C. 1980. Maintenance of bumble bee colonies in observation hives
(Hymenoptera: Apidae). The Canadian Entomologist, 112, 321–326.
Premnath, S., Sinha, A. & Gadagkar, R. 1995. Regulation of Worker Activity in a Primitively Eusocial
Wasp, Ropalidia-Marginata. Behavioral Ecology, 6, 117–123.
R Core Team. 2012. R: A language and environment for statistical computing. R Foundation for Statistical
Computing, Vienna,
Reeve, H. K. & Gamboa, G. J. 1983. Colony activity integration in primitively eusocial wasps: the role of
the queen (Polistes fuscatus, Hymenoptera: Vespidae). Behavioral Ecology and Sociobiology, 13, 63–
74.
Ruiz-González, M. X., Bryden, J., Moret, Y., Reber Funk, C., Schmid-Hempel, P. & Brown, M. J. F.
2012. Dynamic transmission, host quality and population structure in a multihost parasite of
bumblebees. Evolution, 66, 3053–3066.
Salathé, R. M. & Schmid-Hempel, P. 2011. The Genotypic Structure of a Multi-Host Bumblebee Parasite
Suggests a Role for Ecological Niche Overlap. PLOS ONE, 6, e22054.
Salathé, R., Tognazzo, M., Schmid-Hempel, R. & Schmid-Hempel, P. 2012. Probing Mixed-Genotype
Infections I: Extraction and Cloning of Infections from Hosts of the Trypanosomatid Crithidia bombi.
PLOS ONE, 7, e49046.
Schmid-Hempel, P. 1998. Parasites in Social Insects. Princeton University Press.
Schmid-Hempel, P. & Reber Funk, C. 2004. The distribution of genotypes of the trypanosome parasite,
Crithidia bombi, in populations of its host, Bombus terrestris. Parasitology, 129, 147–158.
Schmid-Hempel, P. & Schmid-Hempel, R. 1993. Transmission of a pathogen in Bombus terrestris, with a
note on division of labour in social insects. Behavioral Ecology and Sociobiology, 33, 319–327.
Schmid-Hempel, P., Puhr, K., Kruger, N., Reber, C. & Schmid-Hempel, R. 1999. Dynamic and genetic
consequences of variation in horizontal transmission for a microparasitic infection. Evolution, 426–434.
Schmid-Hempel, R. & Tognazzo, M. 2010. Molecular divergence defines two distinct lineages of
Crithidia bombi (Trypanosomatidae), parasites of bumblebees. The Journal of eukaryotic microbiology,
57, 337–345.
Schneirla, T. C. 1944 April 3. The Reproductive Functions of the Army ant Queen as Pace-Makers of the
Group Behavior Pattern. Journal of the New York Entomological Society, 52, 153–192.
Shykoff, J. A. & Schmid-Hempel, P. 1991a. Incidence and effects of four parasites in natural populations
of bumble bees in Switzerland. Apidologie, 22, 117–125.
Shykoff, J. A. & Schmid-Hempel, P. 1991b. Parasites and the Advantage of Genetic Variability within
Social Insect Colonies. Proceedings Of The Royal Society B-Biological Sciences, 243, 55–58.
Sorci, G., Clobert, J. & Michalakis, Y. 1996. Cost of Reproduction and Cost of Parasitism in the
Common Lizard, Lacerta vivipara. Philosophical transactions of the Royal Society of London Series B,
30
Biological sciences, 76, 121.
Souza, A. R. & Prezoto, F. 2011. Regulation of worker activity in the social wasp Polistes versicolor.
Insectes Sociaux, 59, 193–199.
Stearns, S. C. 1992. The Evolution of Life Histories. Oxford: OUP Oxford.
Szathmáry, E. & Smith, J. M. 1995. The major evolutionary transitions. Nature, 374, 227–232.
Ulrich, Y., Sadd, B. M. & Schmid-Hempel, P. 2011. Strain filtering and transmission of a mixed infection
in a social insect. Journal Of Evolutionary Biology, 24, 354–362.
Wilson, E. O. 1971. The insect societies. Belknap Press.
Yourth, C. & Schmid-Hempel, P. 2006. Serial passage of the parasite Crithidia bombi within a colony of
its host, Bombus terrestris, reduces success in unrelated hosts. Proceedings Of The Royal Society BBiological Sciences, 273, 655–659.
31
3. WORKER LONGEVITY AND ITS RELATION TO COLONY RESISTANCE
AGAINST INFECTIOUS PARASITES IN A SOCIAL INSECT§
Abstract
Social insects should face a classical trade-off when investing in their workforce, just as solitary
animals do when investing in their offspring. To sustain colony growth and reproduction,
resources may be invested in either many ‘low quality’ workers or few ‘high quality’ workers.
Worker quality moreover may manifest in anything that improves survival. Here we use the
annual bumblebee Bombus terrestris and its gut parasite Crithidia bombi to study the proposed
quality/quantity trade-off. In particular we measure quality as a worker's capacity to respond to
immune challenge, which seems plausible as immunity plays an integral part in self- maintenance
and is inherently costly. We first measured variation in intrinsic worker lifespan (quality) and the
number of workers produced over the colony cycle (quantity). In a second experiment, we
infected workers with C. bombi and measured the standing expression of bumblebee immune
genes, to test for differential immune investment among colonies with different worker longevity.
We found that across bumblebee colonies, immunity indeed is linked to worker longevity, such
that individuals from colonies that produced longer-lived workers had lower C. bombi infection
intensity Colonies with short-lived workers tended to be small, which is usually associated with
low reproductive success. The standing variation in colony size and worker longevity may thus be
maintained by the relation of longevity to colony size and their linked consequences on individual
resistance and epidemic dynamics within the colony.
§
A version of this Chapter is in preparation as S. D Buechel, S.M. Barribeau & P. Schmid-Hempel.
32
Introduction
There is striking variation in individual longevity, within and between species, among most
multicellular organisms (Carey & Judge 2000; De Magalhaes & Costa 2009) including the social
insects (Calabi & Porter 1989; Tsuji et al. 1996; Chapuisat & Keller 2002; Omholt & Amdam
2004; Rueppell et al. 2009). Longevity is a life history trait that is under selection and
evolutionary and mechanistic theories of ageing (Medawar 1952; Williams 1966; Hamilton 1966;
Kirkwood 1977; Stearns 1992) can help to explain the observed variation. Resource allocation to
fecundity, growth, and maintenance (Stearns 1992; Roff 1993) against the background of
extrinsic mortality is thereby considered to be a key element of longevity. Increased investment
into growth and maintenance, for example, can be adaptive if the likelihood to survive to old age
is high. The organism then lives longer, grows larger and therefore produces more, or better
quality offspring as compared to short-lived, smaller parents. The reverse were true if the chances
to survive to old age is small (Stearns 1992). This principle of optimal resource investment not
only explains the range of life histories observed in solitary species but is also applicable to
lifespan variation among the non-reproductive worker caste of insect societies (Kramer &
Schaible 2013).
Analogous to the predicted strategies for solitary organisms, colonies subject to low extrinsic
mortality should invest into colony growth and maintenance and thus become long-lived, whereas
colonies that are subject to high mortality should invest in early reproduction. This is because, as
with body size for solitary species, colony size in social insects is positively associated with
reproductive output and competitiveness (Hölldobler & Wilson 1990; Adams 1990; Muller &
Schmid-Hempel 1992; Schmid-Hempel et al. 1993; Palmer 2004; Lopez-Vaamonde et al. 2009).
Additionally, and essential to our understanding of worker longevity, investment in growth and
maintenance of the colony may be seen as investment in the physiological maintenance and thus a
prolonged lifespan of workers. As workers provide most functions needed for colony maintenance
and colonies with long-lived workers are usually large colonies (Carey 2001), such an investment
could be selected for (Kramer & Schaible 2013). Moreover, it is likely that in most species the
33
queen is in control of important steps in the life cycle of the colony (Holland et al. 2013), and she
might therefore also control investment. Empirical data supports these predictions for the
reproductive caste (Keller & Genoud 1997; Keller 1998; Carey 2001) and suggests that in ants
increased worker lifespan tends to be associated with low levels of extrinsic mortality, and vice
versa (Tsuji et al. 1996; Chapuisat & Keller 2002; Hartmann & Heinze 2003; Rueppell et al.
2007). No such correlation, however, is as yet studied between worker longevity and another key
trait, colony size - defined as the number of workers present at any one time. In wasps, however,
there is a positive association between colony size and worker lifespan (O'Donnell & Jeanne
1992). Evidence in honeybees is equivocal as large colonies were reported to have short-lived
workers (see Rueppell et al. 2009).
During colony development, social insect colonies may face a classical trade-off for worker
quality versus worker quantity with repercussions on colony size and colony life span. The choice
may be to produce either few ‘high-quality’ workers (which then are long-lived) or many ‘lowquality’ workers (which are short-lived). In this context we define worker quality as relating to
anything that improves survival; in particular, we here measure "quality" as the worker's capacity
to resist infection by a common parasite and the ability to respond to immune challenges.
Although this covers just one aspect of "quality", it is a plausible and practical choice as
immunity is an integral part of self-maintenance and since an immune response is inherently
costly (Moret 2000; Sadd & Schmid-Hempel 2008). On the other hand, and from the perspective
of the epidemiology of an infectious disease spreading through the colony, short-lived workers
reduce the infectious period in any one host and thus the chances to encounter and transmit a
parasite suggesting that a short worker lifespan per se may be a colony defence strategy against a
spreading parasite (Chapter 2).
The bumblebee, Bombus terrestris, and its trypanosomatid gut parasite, Crithidia bombi, are a
model for host-parasite studies in evolutionary and ecological immunology. The extensive
toolbox available to study this host-parasite system and the presence of annual colonies that vary
in worker lifespan makes this system a good candidate to study the expected quality/quantity
34
trade-off with respect to investment into immune functions. C. bombi is a directly transmitted
trypanosomatid, most closely related to the Leishmania group (Schmid-Hempel & Tognazzo
2010), that induces severe fitness losses in young queens by reducing their likelihood of
successfully founding colonies (Brown et al. 2003). C. bombi infects per os, can be picked up via
contaminated flowers (Durrer & Schmid-Hempel 1994) and spread through the colony by
infectious cells shed in the feces of infected bees. Hence, cell count (infection intensity) in the gut
is a good proxy for the transmission success of the parasite as it correlates with cell numbers in
faeces (Sadd 2011).
Here we report the variation in intrinsic worker longevity (quality) and the number (quantity) of
workers produced over the colony cycle and assess how this colony level investment into worker
quality determines C. bombi infection susceptibility and the expression of 27 immunologically
important genes.
Materials and Methods
Experimental procedure
We did this experiment in two consecutive years using laboratory colonies started from spring
queens caught in 2011/2012, originating from two populations in Switzerland (near Aesch BL;
and Neunforn TG). Colonies were reared in circular perlite nests (Pomeroy & Plowright 1980)
and kept under standard laboratory conditions (28 ± 2 °C, 60% RH ) with pollen and sugar water
provided ad libitum. We carefully checked the queens for natural Crithidia bombi infections and
only used uninfected colonies. We started the experiment by collecting callow workers (young,
not yet sclerotized adults) on a daily basis when colonies reached a size of 12-15 workers; we
stopped collecting when the first males and gynes (daughter queens) were produced. Callows
were housed in boxes, in groups of two to six per colony and collection day, until the age of eight
days when we randomly assigned them to one of three treatments: (1) lifespan monitoring, (2) C.
bombi infection, and (3) immune gene expression. Bees that were used for the assessment of
variation in worker lifespan between colonies (treatment no. 1) were kept singly in boxes and
35
checked for their survival twice a day. For the screening of differential immune gene expression
(treatment no. 3), bees were snap frozen in liquid nitrogen at noon and stored at -80 ˚C for RNA
extraction. Note that we only extracted RNA of workers from the five colonies with the longest
and shortest-lived workers (N = 5 colonies each, N = 46 workers). With treatment no. 2, we
examined the correlation between resistance to C. bombi infection (measured as infection
intensity in the gut, 7 days post-infection) and mean worker lifespan in B. terrestris colonies. We
infected workers with a cocktail of six genetically distinct C. bombi strains. To infect, we starved
workers for two hours before presenting them with 10 µl of inoculum containing 18'000 cells in
medium and sugar water (1:1 ratio; 3'000 cells per strain). Bees that did not consume the
inoculum within 1 h were excluded from the study to ensure primary infections and viability of
the parasite cells. We individually kept the remaining (infected) bees for 7 days before we
assessed faeces cell counts (infection intensity reflecting parasite transmission success) and
sampled them for genetic analysis. Parasite strains used for infections were sourced from queens
caught in 2010 from the same two populations we used for this experiment. In addition, single
cells were isolated from the queens' faeces, cloned and maintained as cultures as described in
(Salathé et al. 2012). The same experimental procedure was re-done in 2012 with no gene
expression data collected, but callows were marked to monitor the total number of workers
produced in these colonies.
Quantification of infections
We dissected out the entire guts of workers to extract the DNA with a Qiagen DNeasy Blood &
Tissue Kit (Qiagen, Hilden, Germany) and amplified a partial sequence of the C. bombi Cyt b
gene (primers CB-Cytb2-F, CB-Cytb 2-B; (Schmid-Hempel & Tognazzo 2010) so as to visualize
the infection status of workers (1.5 % agarose gel). For infection quantification we run
quantitative real-time PCRs (qPCR) using Fast EvaGreen Master Mix (Biotium, Hayward, CA,
USA) with primers designed to amplify a fragment conserved in the C. bombi 18sRNA (Ulrich et
al. 2011). The reaction was set up in 10 µL reaction volume and included 2 µL eluted DNA 0.2
µM of each primer and 2 µL HOT FIREPol EvaGreen qPCR Mix. We used the thermal profile
36
described in (Ulrich et al. 2011). Each biological sample was run in technical duplicates with
negative controls (distilled H2O) and positive controls (bumblebee DNA). A dilution series of
DNA extracted from a known number of C. bombi cells was included as a standard to calculate
the relative infection intensities in workers, see Supplementary information figure S1. We
included the same dilution series for each qPCR run. This allowed us to standardize our
quantifications across plates.
RNA preparation and quantification of candidate genes
We used the protocol in Brunner et al. (2013) to measure the expression of candidate genes. In
brief, we extracted RNA (RNeasy Plus Mini kit, Qiagen) from 92 homogenized worker abdomen
(N= 46 workers from n=5 short-lived and long-lived colonies) using an Omni Bead Ruptor 24
Homogenizer at -4°C (OMNI International), confirmed RNA integrity for a subsample of n=12,
on a 2100 Bioanalyzer (Agilent Technologies) and evaluated quantity and purity of extracted
RNA with the Nanodrop 8000 (ThermoScientific). For six contaminated samples, with a low
260/280 nm or 260/230 nm ratios, we re-extracted RNA from the homogenized abdomen
preparation kept for this purpose. We reverse transcribed 0.4 µg RNA from each sample
(Quantitect reverse transcription kits, Qiagen) and amplified the cDNA product with intron
specific primers to confirm the absence of genomic DNA. We measured the expression of 27
genes (c.f. Table 1) on a Fluidigm 96.96 Dynamic Array (with EvaGreen DNA Binding Dye,
Biotium) with each biological sample measured over three technical replicates and the average of
each triplicate was used as the observed expression value (Ct).
We selected a set of candidate genes to cover known aspects of insect immunity, representing the
classical immune response cascades (recognition, signaling, effectors) of three main insect
immune pathways (Toll, Imd, JAK/STAT). Because the workers were not immune-challenged,
the expression profiles represent the non-induced, constitutive levels of gene activities. We
included genes relevant for viral infection, bacterial infection, metabolic activities, reactive
oxygen species, general stress response, melanisation and antimicrobial peptides; the included
37
genes were PGRP-S3, PGRP-LC, BGRP1, BGRP2, PGRP-LB, abaecin, apidaecin, defensin,
hymenoptaecin, argonaute, aubergine, relish, hopscotch, TEPA, catsup, yellow, punch,
prophenoloxidase, vitellogenin, apolipophorin III, cytochrome P450, serpin 27a, peroxiredoxin5,
jafrac, lysozyme3, transferrin, ferritin - as well as five reference genes (AK, PLA2, RPL13, Perit,
ef.1f.a2). We preferentially included genes that were published in previous studies and are
considered relevant for parasite defense (Riddell et al. 2009; Vermeulen et al. 2009; Li et al.
2010; Schlüns et al. 2010; Radyuk et al. 2010; Vogel et al. 2011; Erler et al. 2011; Riddell et al.
2011). We used primers published in (Brunner et al. 2013), except for vitellogenin (Li et al. 2010)
b and relish (Schlüns et al. 2010). Note that primers for, yellow, and PGRB-LB, apolipophorin III,
cytochrome P450 and prophenoloxidase were designed following the same protocol. NCBI
accession numbers and primer sequences for all genes can be found in the Supplementary
information Table S1.
Analysis and statistics
For each colony we first calculated the mean worker lifespan and infection intensity from the
experiment, as measured with qPCR, and tested the relation of resistance to C. bombi and
longevity of workers (adult life span after emergence, in days) using a linear regression model
with "infection intensity" (measured as total number of parasite cells per bee) included as
response variable. In a similar approach we regressed the total number of workers that each
colony produced against the mean worker lifespan to test whether colonies with short-lived
workers tend to produce more workers compared to colonies with long-lived workers. Linear
regressions were performed in IBM SPSS Statistics version 19.0, and data were log-transformed
to meet the normality criterion (Shapiro-Wilk-test). To analyze the expression of our candidate
genes in colonies that produced long-lived or short-lived workers, we used the Ct values (see
above) of candidate genes normalized against the geometric mean of the four most stable
reference genes (PLA2, RPL13, Perit, ef.1f.a2; determined with GeNorm, Biogazelle). We YeoJohnson-transformed the data to improve normality and performed a classical multivariate
analysis of variance (MANOVA) on all candidate genes including longevity groups (long-lived
38
workers and short-lived workers, n=5 colonies each) and colony identity as a fixed factor.
Different levels of expression in individual genes were explored with univariate variance analysis
(ANOVA). We additionally performed a linear discriminant analysis on all candidate genes to
determine different gene expressed among workers from colonies with long-lived and short-lived
workers. We analyzed these data with R (2.15.2, (R Core Team 2012).
Results
We found that adult worker lifespan varied greatly among B. terrestris colonies (N = 35 colonies,
N = 409 workers). Worker life span was, on average, 34.5 ± 5.44 days for the colony with
shortest-lived workers, and 82.0 ± 9.11 days for the colony with the longest-lived workers (Figure
Mean worker lifespan (Days ± S.E.)
1).
Colonies
Figure 1. The average worker lifespan (mean ± S.E. in days) in B. terrestris colonies (colony identity
indicated by letters) collected in 2010 and 2012 and reared under standard laboratory conditions.
Colonies marked with solid circles were subsequently used for the analysis of immune gene expression.
Small numbers below lines indicate sample sizes (number of workers monitored for their entire life
span).
39
Mean infection intensity (Cells per bee ± S.E.)
Mean worker lifespan (Days ± S.E.)
Figure 2. Relationship between colony mean worker life span (x-axis, in days) and intensity of C.
bombi infection (the number of parasite cells per bee as measured with qPCR, y-axis) in workers from
21 colonies (linear regression on ln-transformed data: [Mean infection intensity (cells/bee)] = 166029 18.796 * [mean worker lifespan in days]; F1,19= 5.297, r2= 0.218, P= 0.03; lines indicate ± 1 S.E.).
As hypothesized, we found that across colonies worker longevity correlates with resistance to
experimental infections by C. bombi with infection intensity inversely correlated to worker
longevity (Figure 2; linear regression: F1,19= 5.29, r2 = 0.218, P= 0.03), indicating that colonies
with long-lived workers also showed higher resistance. Colonies with long-lived workers also
generally differed in their constitutive immune gene expression from colonies with short-lived
workers (MANOVA for all genes combined: F27,87= 4.099, P< 0.001). More specifically, genes
involved in parasite recognition (PGRB-LB), melanisation (yellow, prophenoloxidase), and stress
response (cytochrome P450) were expressed at a higher level in colonies with long-lived workers.
The pattern was reversed - with higher expression in colonies with short-lived workers - for the
antimicrobial peptide abaecin, the antiviral gene aubergine, and the key protein vitellogenin
(Figure 3 & Table 1). A total 72% of workers moreover were correctly classified to the respective
longevity scheme in their colony based on scores of the linear discriminant analysis for the
expression levels of candidate genes. Contrary to our expectations from a hypothetical quality40
quantity trade-off, short-lived workers were produced at lower numbers compared to long-lived
workers as indicated by a positive relationship between the number of workers produced in a
colony and average worker longevity (linear regression: F1,11= 8.11, r2 = 0.425, P= 0.016, Figure
4).
10
Workers from colonies with:
Long-lived workers
Short-lived workers
46#
5
44#
46#
ΔCt
46#
46#
0
46#
46#
46#
45#
46#
46#
46#
-5
abaecin
aubergine
46#
46#
cytochrome
P450
PGRB-LB prophenoloxidase vitellogenin
yellow
Figure 3. Relative expression levels (see Methods) of 7 bumblebee immune genes that were
differentially expressed between colonies that produced short-lived workers (shaded boxes) and
colonies that produced long-lived workers (open boxes). Box plot showing median (line) 25-75%
quartiles (boxes) ranges (whiskers) and extreme values (circles). Sample sizes are added in small
numbers (workers from which expression level of the respective gene was assessed).
41
Table 1. ANOVA results for differential gene expression between colonies with short-lived workers versus long-lived
workers. Tested were 27 genes with known functions in insect immunity, metabolism and stress response. These genes
were involved in three main immune pathways of insects (Toll, Imd, Jak/stat) and normalized against 4 common
housekeeping genes of B. terrestris.
Class
Gene
Function
df
SS
F
P
expressiona
Receptor
PGRP-LB
recognition
87
2.996
12.244 <0.001
long-lived
PGRP-S3
recognition (Toll)
87
4.424
2.601
0.11
BGRP1
recognition (Toll)
87
0.289
0.136
0.71
BGRP2
recognition (Toll)
87
0.001
0.008
0.93
BGRP-LC
recognition (Imd)
87
0.001 0.004 0.91
relish
Signalling(Imd)
87
0.163
1.461
0.23
hopscotch
Signalling (Jak/stat)
87
0.000
0.024
0.88
abaecin
antimicrobial peptid
87
35.182
17.203
<0.001
apidaecin
antimicrobial peptid
87
0.438
0.363
0.55
defensin
antimicrobial peptid
87
1.575
0.572
0.45
hymenoptaecin
antimicrobial peptid
87
6.160
1.674
0.2
TEPA
Effector (Jak/stat)
87
3.357
1.285
0.26
lysozyme3
Effector, bacteriolytic
87
0.004
0.006
0.94
transferritin
irionbinding, antibacterial
87
3.649
3.282
0.07
ferritin
iron transportation
87
52.000
0.013
0.91
punch
enzyme melanin synthesis
87
0.059
1.104
0.3
catsup
enzyme melanin synthesis
87
0.024
0.053
0.82
yellow
enzyme melanin synthesis
87
0.783
21.419
<0.001
serpin27a
serine protease inhibitor
(PPO)
87
0.055
0.504
0.48
prophenoloxidase
enzyme, melanin synthesis
87
0.010
29.786
<0.001
long-lived
87
15.194
6.362
0.01
short-lived
87
1.621
2.754
0.1
Signalling
Effector
Melanisation
Antiviral
RNAi pathway genes,
antiviral
RNAi pathway genes,
antiviral
aubergine
argonaut
ROS
Metabolic
peroxiredoxin5
peroxiredoxin
87
0.000
0.000
0.99
jafrac
peroxiredoxin
87
15.290
1.010
0.32
endocrinological functions
87
2177.500
6.003
0.02
apolipophorin III
fat storage
87
45.800
0.197
0.66
cytochrome P450
stress response
87
891.000
6.982
<0.01
vitellogenin
b
a
short-lived
short-lived
long-lived
short-lived
long-lived
indicates significantly or marginally incresed level of expression in the respective colonies, long-lived = colonies that produced
long-lived workers, short-lived = colonies that produced short-lived workers.
b
vitellogenin has pleiotropic effects (see text)
42
Total number of workers produced
Mean worker lifespan (Days ± S.E.)
Figure 4. Relationship between colony mean worker life span (x-axis) and the total number of workers
produced (y-axis) in 15 colonies (linear regression on ln-transformed data: [Total worker number] = 30.405 + 2.666* [mean worker lifespan, in days]; r2 = 0.425, F1,19 = 5.339, P = 0.032; lines indicate ± 1
S.E.).
Discussion
We tested a hypothesized quality vs. quantity trade-off, where "quality" was here measured as the
ability of workers to resist a common parasite and reduce infection intensity; this would also limit
the transmission success of the parasite. These measures were additionally highlighted by a gene
expression study. We found that across bumblebee colonies, resistance is indeed linked to worker
longevity, such that individuals from colonies that produced short-lived workers showed lower
resistance to C. bombi infection. But they also expressed higher levels of antimicrobial (abaecin)
and antiviral defence (aubergine) genes compared to individuals from colonies with long-lived
workers. The latter being phenotypically better protected against C. bombi, (showing lower
infection intensities) also expressed genes involved in parasite recognition (PGRB), stress
response (Cyt P450), and the general ‘non-specific’ melanisation response (PPO) at higher levels.
Although the detailed conclusions must remain preliminary for the time being, the differences in
gene expression patterns may indicate differences in how parasites are controlled - either early in
43
the defence cascade (e.g. by strongly expression recognition, in colonies with long-lived workers)
or later in the cascade (e.g. by anti-microbial effectors such as abaecin, in colonies with shortlived workers). Importantly and in contrast to predictions made based on the quality quantity
trade-off; short- lived workers generally were from colonies having lower workers numbers and,
hence, small colony size.
Colony size is a key life-history parameter in social insects, affecting a wide range of
characteristic, such as division of labour and social structure (e.g. in ants; (Holbrook et al. 2013),
and a correlate of parasite load, infection rate and transmission dynamics (Strassmann 1981;
Muller & Schmid-Hempel 1992; Schmid-Hempel 1995; 1998; Naug & Camazine 2002; Pie et al.
2004). Interestingly, body size of nurses is also positively correlated with the number of eggs laid
by the queen and the number of new workers emerging (Cnaani & Hefetz 1994), indicating that
large-sized workers (having received more investment as larvae) are also more productive for the
colony.
The positive relationship between worker longevity and colony size might be important to
understand differential investment in worker quality, e.g.in the context of parasite resistance. For
example, long-lived workers are more likely to live in large colonies and, therefore, may
encounter more parasites brought in by the many others, and have more opportunities to transmit
an infection to nest mates. On the other hand, we find them better defended individually.
Individual worker life span, individual resistance to infection, colony size and epidemic potential
could thus act together to control infectious diseases via demography and appropriate immune
responses. In fact, immune defences and social organisation may not always be independent
(Castella et al. 2010). Our previous work on worker turnover rate found that artificially increased
birth and death rate reduced parasite spread in groups of bumblebee workers Chapter 2.
Investment in worker defences - as it associates with worker longevity (Figure 2) - for this reason
should thus influence the respective epidemiological potential of a parasite (high with long lived
workers) in relation to worker longevity and colony size (Figure 5). With these considerations,
44
low quality colonies (queens) that produce only few poorly protected workers could be
Epidemiological resistance
Colony resistance
Colony resistance
maintained by their reduced epidemic potential.
Worker lifespan
Worker resistance
Epidemiological resistance
Worker lifespan
Figure 5. A sketch showing how colonies may vary in overall susceptibility to an infectious parasite
spreading within the colony as a function of worker life span (longevity). The sketch is based on the
current findings and those reported in Chapter 2. The solid lines illustrate the assumption that
"epidemiological resistance" decreases with worker longevity, i.e. an infectious parasite can spread
more easily within the colony. This is because long-lived workers allow longer infection times and
more transmission events; in addition, long-lived workers are found in larger colonies. These factors
make it more likely that a parasite can maintain itself and spread in the colony. The solid curves
illustrate the assumption that long-lived workers have higher individual resistance to infection, as found
in this study. This would increase the overall resistance of the colony as a whole - based on host
resistance per se. For illustrative purposes only, we assume that the resulting overall colony resistance is
a combination of individual resistance and colony-level "epidemiological resistance". To keep it simple,
and for the sake of argument, the two components will combine to yield maximum overall resistance
where the two processes (line and curve) intersect (point X). The two panels (a, b) demonstrate how
variation in worker life span affects this overall resistance In situation (a), individual worker resistance
may be the parameter most sensitive to changes in life spans (e.g. the solid curve changes to the dotted
curve) whereas the "epidemiological resistance" is not affected (line remains the same). If so, the new
point of maximum overall resistance will have moved to X1, predicting longer life span. In situation (b),
epidemiological resistance (via larger colony size, longer transmission windows) may be most sensitive
to changes in life spans (solid line changes to dotted line) whereas individual worker resistance remains
the same (solid curve). If so, worker lifespan should get shorter to yield maximum colony resistance
(point X1). Variation in worker life span can thus be driven by various processes that link to overall
resistance of a colony.
Based on the differences in immune gene expression, colonies with short-lived workers may
perhaps be prepared to defend themselves against different groups of parasites than colonies with
long-lived workers. For example, colonies with short-lived workers could defend themselves
primarily against rapidly spreading pathogens with immediate fitness detriments to colony growth
and reproduction such as viruses and bacteria. For example, honeybees viruses such as DWV
(Deformed Wing Virus) and ABPV (Acute Bee Paralysis Virus) account for the most prevalent
45
but mainly non-apparent infections, yet causing fatal effects for colonies under stressful
conditions (e.g. when co-infected with mites or under reduced food supply (reviewed in (Chen &
Siede 2007)). These viruses may also play a role in bumblebee disease (Genersch et al. 2006;
Singh et al. 2009). Colonies with short-lived workers indeed show higher expression for genes
involved in RNA-silencing (aubergine, argonaut) - a key antiviral defence - a pattern that perhaps
may reflect this threat. Additionally, these colonies had higher expression levels of vitellogenin, a
general yolk precursor, up-regulated together with transferritin (P = 0.07; in our study), mostly
needed when bees develop their ovaries (Koywiwattrakul & Sittipraneed 2008). In social insects
such as the honeybee, vitellogenin is a central pace maker and also acts as antioxidant to prolong
queen and worker longevity (Corona et al. 2007) and strongly affects social organization (Nelson
et al. 2007). Colonies with long-lived workers, by contrast, may be more generally protected.
Parasites might be detected and controlled at an early step of infection. The high expression of
PPO and yellow suggests effective melanisation response to foreign particles including wound
healing upon injury. Activation of the prophenoloxidase however can cause self-harm.
Considering that long-lived workers should have an increased risk of parasites and predators,
investing in general and costly defence might be relevant.
In conclusion we found that colonies with different worker longevity differed in immune
investment and colony characteristics. Overall, colonies with short-lived workers are small
colonies, which, at least in the absence of parasites, have low reproductive success (Hölldobler &
Wilson 1990; Adams 1990; Muller & Schmid-Hempel 1992; Schmid-Hempel et al. 1993; Palmer
2004). Variation in colony size and worker longevity, such as those observed in Figure 1,
therefore may be maintained by parasite diversity and its effect on colony size and success and
the drawbacks in terms of facilitating an epidemic within the nest.
46
References
Adams, E. S. 1990. Boundary Disputes in the Territorial Ant Azteca-Trigona - Effects of Asymmetries in
Colony Size. Animal Behaviour, 39, 321–328.
Brown, M. J., Schmid-Hempel, R. & Schmid-Hempel, P. 2003. Strong context‐dependent virulence in a
host–parasite system: reconciling genetic evidence with theory. Journal Of Animal Ecology, 72, 994–
1002.
Brunner, F. S., Schmid-Hempel, P. & Barribeau, S. M. 2013. Immune Gene Expression in Bombus
terrestris: Signatures of Infection Despite Strong Variation among Populations, Colonies, and Sister
Workers. PLOS ONE, 8, e68181.
Calabi, P. & Porter, S. D. 1989. Worker longevity in the fire ant Solenopsis invicta: Ergonomic
considerations of correlations between temperature, size and metabolic rates. Journal Of Insect
Physiology, 35, 643–649.
Carey, J. R. 2001. Demographic mechanisms for the evolution of long life in social insects. Experimental
Gerontology, 36, 713–722.
Carey, J. R. & Judge, D. S. 2000. Longevity records. Odense: Odense University Press.
Castella, G., Christe, P. & Chapuisat, M. 2010. Covariation between colony social structure and immune
defences of workers in the ant Formica selysi. Insectes Sociaux, 57, 233–238.
Chapuisat, M. & Keller, L. 2002. Division of labour influences the rate of ageing in weaver ant workers.
Proceedings Of The Royal Society B-Biological Sciences, 269, 909–913.
Chen, Y. P. & Siede, R. 2007. Honey bee viruses. Advances in virus research, 70, 33–80.
Cnaani, J. & Hefetz, A. 1994. The effect of workers size frequency distribution on colony development
inBombus terrestris. Insectes Sociaux, 41, 301–307.
Corona, M., Velarde, R. A., Remolina, S., Moran-Lauter, A., Wang, Y., Hughes, K. A. & Robinson, G.
E. 2007. Vitellogenin, juvenile hormone, insulin signaling, and queen honey bee longevity.
Proceedings Of The National Academy Of Sciences Of The United States Of America, 104, 7128–7133.
DE Magalhaes, J. P. & Costa, J. 2009. A database of vertebrate longevity records and their relation to
other life-history traits. Journal Of Evolutionary Biology, 22, 1770–1774.
Durrer, S. & Schmid-Hempel, P. 1994. Shared use of flowers leads to horizontal pathogen transmission.
Proceedings Of The Royal Society Of London Series B-Biological Sciences, 258, 299–302.
Erler, S., Popp, M. & Lattorff, H. M. G. 2011. Dynamics of Immune System Gene Expression upon
Bacterial Challenge and Wounding in a Social Insect (Bombus terrestris). PLOS ONE, 6, e18126.
Genersch, E., Yue, C., Fries, I. & de Miranda, J. R. 2006. Detection of Deformed wing virus, a honey
bee viral pathogen, in bumble bees (Bombus terrestris and Bombus pascuorum) with wing deformities.
Journal Of Invertebrate Pathology, 91, 61–63.
Hamilton, W. D. 1966. The moulding of senescence by natural selection. Journal of Theoretical Biology,
12, 12–45.
Hartmann, A. & Heinze, J. 2003. Lay egg slive longer: Division of labor and life span in a clonal ant
species. Evolution, 57, 2424–2429.
Holbrook, C. T., Eriksson, T. H., Overson, R. P., Gadau, J. & Fewell, J. H. 2013. Colony-size effects
on task organization in the harvester ant Pogonomyrmex californicus. Insectes Sociaux, 60, 191–201.
47
Holland, J. G., Guidat, F. S. & Bourke, A. 2013. Queen control of a key life-history event in a eusocial
insect. Biology Letters, 9, 20130056–20130056.
Hölldobler, B. & Wilson, E. O. 1990. The Ants. Harvard University Press.
Keller, L. 1998. Queen lifespan and colony characteristics in ants and termites. Insectes Sociaux, 45, 235–
246.
Keller, L. & Genoud, M. 1997. Extraordinary lifespans in ants: a test of evolutionary theories of ageing.
Nature, 389, 958–960.
Kirkwood, T. B. L. 1977. Evolution of ageing. Nature, 270, 301–304.
Koywiwattrakul, P. & Sittipraneed, S. 2008. Expression of Vitellogenin and Transferrin in Activated
Ovaries of Worker Honey Bees, Apis mellifera. Biochemical Genetics, 47, 19–26.
Kramer, B. H. & Schaible, R. 2013. Life Span Evolution in Eusocial Workers—A Theoretical Approach
to Understanding the Effects of Extrinsic Mortality in a Hierarchical System. PLOS ONE, 8, e61813.
Li, J., Huang, J., Cai, W., Zhao, Z., Peng, W. & Wu, J. 2010. The vitellogenin of the bumblebee,
Bombus hypocrita: studies on structural analysis of the cDNA and expression of the mRNA. Journal
of Comparative Physiology B, 180, 161–170.
Lopez-Vaamonde, C., Raine, N. E., Koning, J. W., Brown, R M, Pereboom, J. J. M., Ings, T. C.,
Ramos-Rodriguez, O., Jordan, W. C. & Bourke, A. F. 2009. Lifetime reproductive success and
longevity of queens in an annual social insect. Journal Of Evolutionary Biology, 22, 983–996.
Medawar, P. B. 1952. An Unsolved Problem of Biology. London: Lewis.
Moret, Y. 2000. Survival for Immunity: The Price of Immune System Activation for Bumblebee Workers.
Science, 290, 1166–1168.
Muller, C. B. & Schmid-Hempel, P. 1992. Correlates of reproductive success among field colonies of
Bombus lucorum: the importance of growth and parasites. Ecological Entomology, 17, 343–353.
Naug, D. & Camazine, S. 2002. The Role of Colony Organization on Pathogen Transmission in Social
Insects. Journal of Theoretical Biology, 215, 427–439.
Nelson, C. M., Ihle, K. E., Fondrk, M. K., Page, R. E. & Amdam, G. V. 2007. The Gene vitellogenin
Has Multiple Coordinating Effects on Social Organization. PLoS Biology, 5, e62.
O'Donnell, S. & Jeanne, R. L. 1992. The effects of colony characteristics on life span and foraging
behavior of individual wasps (Polybia occidentalis, Hymenoptera: Vespidae). Insectes Sociaux, 39,
73–80.
Omholt, S. W. & Amdam, G. V. 2004. Epigenetic Regulation of Aging in Honeybee Workers. Science of
Aging Knowledge Environment, 2004, pe28–pe28.
Palmer, T. M. 2004. Wars of attrition: colony size determines competitive outcomes in a guild of African
acacia ants. Animal Behaviour, 68, 993–1004.
Pie, M. R., Rosengaus, R. B. & Traniello, J. 2004. Nest architecture, activity pattern, worker density and
the dynamics of disease transmission in social insects. Journal of Theoretical Biology, 226, 45–51.
Pomeroy, N. & Plowright, R. C. 1980. Maintenance of bumble bee colonies in observation hives
(Hymenoptera: Apidae). The Canadian Entomologist, 112, 321–326.
R Core Team. 2012. R: A language and environment for statistical computing. R Foundation for Statistical
Computing, Vienna,
48
Radyuk, S. N., Michalak, K., Klichko, V. I., Benes, J. & Orr, W. C. 2010. Peroxiredoxin 5 modulates
immune response in Drosophila. Biochimica et Biophysica Acta (BBA) - General Subjects, 1800,
1153–1163.
Riddell, C. E., Sumner, S., Adams, S. & Mallon, E. B. 2011. Pathways to immunity: temporal dynamics
of the bumblebee (Bombus terrestris) immune response against a trypanosomal gut parasite. Insect
Molecular Biology, 20, 529–540.
Riddell, C., Adams, S., Schmid-Hempel, P. & Mallon, E. B. 2009. Differential Expression of Immune
Defences Is Associated with Specific Host-Parasite Interactions in Insects. PLOS ONE, 4, e7621.
Roff, D. A. 1993. The evolution of life histories. Chapman and Hall,
Rueppell, O. O., Bachelier, C. C., Fondrk, M. K. M. & Page, R. E. R. 2007. Regulation of life history
determines lifespan of worker honey bees (Apis mellifera L.). Experimental Gerontology, 42, 13–13.
Rueppell, O., Kaftanouglu, O. & Page, R. E., Jr. 2009. Honey bee (Apis mellifera) workers live longer in
small than in large colonies. Experimental Gerontology, 44, 447–452.
Sadd, B. M. 2011. Food-environment mediates the outcome of specific interactions between bumblebees
and its trypanosome parasite. Evolution, 65, 2995–3001.
Sadd, B. M. & Schmid-Hempel, P. 2008. PERSPECTIVE: Principles of ecological immunology.
Evolutionary Applications, 2, 113–121.
Salathé, R., Tognazzo, M., Schmid-Hempel, R. & Schmid-Hempel, P. 2012. Probing Mixed-Genotype
Infections I: Extraction and Cloning of Infections from Hosts of the Trypanosomatid Crithidia bombi.
PLOS ONE, 7, e49046.
Schlüns, H., Sadd, B. M., Schmid-Hempel, P. & Crozier, R. H. 2010. Infection with the trypanosome
Crithidia bombi and expression of immune-related genes in the bumblebee Bombus terrestris.
Philosophical Transactions of the Royal Society B: Biological Sciences, 34, 705–709.
Schmid-Hempel, P. 1995. Parasites and social insects. Apidologie, 26, 245–245.
Schmid-Hempel, P. 1998. Parasites in Social Insects. Princeton University Press.
Schmid-Hempel, P., Winston, M. L. & Ydenberg, R. C. 1993. Foraging of individual workers in relation
to colony state in the social Hymenoptera. Canadian entomologist, 125, 129–129.
Schmid-Hempel, R. & Tognazzo, M. 2010. Molecular divergence defines two distinct lineages of
Crithidia bombi (Trypanosomatidae), parasites of bumblebees. The Journal of eukaryotic microbiology,
57, 337–345.
Singh, R., Levitt, A. L., Rajotte, E. G., Holmes, E. C., Ostiguy, N., vanEngelsdorp, D., Lipkin, W. I.,
Depamphilis, C. W., Toth, A. L. & Cox-Foster, D. L. 2009. RNA Viruses in Hymenopteran
Pollinators: Evidence of Inter-Taxa Virus Transmission via Pollen and Potential Impact on Non-Apis
Hymenopteran Species. PLOS ONE, 5, –e14357.
Stearns, S. C. 1992. The Evolution of Life Histories. Oxford: OUP Oxford.
Strassmann, J. E. 1981. Parasitoids, Predators, and Group-Size in the Paper Wasp, Polistes-Exclamans.
Ecology, 62, 1225–1233.
Tsuji, K., Nakata, K. & Heinze, J. 1996. Lifespan and reproduction in a queenless ant.
Naturwissenschaften, 83, 577–578.
Ulrich, Y., Sadd, B. M. & Schmid-Hempel, P. 2011. Strain filtering and transmission of a mixed infection
in a social insect. Journal Of Evolutionary Biology, 24, 354–362.
49
Vermeulen, J., Pattyn, F., De Preter, K., Vercruysse, L., Derveaux, S., Mestdagh, P., Lefever, S.,
Hellemans, J., Speleman, F. & Vandesompele, J. 2009. External oligonucleotide standards enable
cross laboratory comparison and exchange of real-time quantitative PCR data. Nucleic Acids Research,
37, e138–e138.
Vogel, H., Altincicek, B., Glöckner, G. & Vilcinskas, A. 2011. A comprehensive transcriptome and
immune-gene repertoire of the lepidopteran model host Galleria mellonella. BMC Genomics, 12, 308.
Williams, G. C. 1966. Pleiotropy, Natural-Selection, and the Evolution of Senescence. Evolution, 11, 398–
411.
50
Supplementary Information
30.00
Plate1#
Plate2##
25.00
Plate3#
Plate#4#
Mean#CT#
20.00
15.00
10.00
5.00
20#
200#
2000#
20000#
Dilu.on#series#(#log10)#
Figure S1. The Ct values for a single logarithmic dilution series of a known concentration of C. bombi
cells from 4 qPCR runs performed on a ABI 7500 Real-Time PCR System (Applied Biosystems,
Rotkreuz, Switzerland)
51
52
Antiviral
Melanisation
Effector
XM_003400641
XM_003398481.1
argonaut
HM_142999
aubergine
XM_003392985
prophenoloxidase
XM_003393332
ferritin
serpin27a
XM_003401163
transferritin
XM_003399634.1
XM_003394052
lysozyme3
yellow
XM_003399699
TEPA
XM_003398173
XR_132450
hymenoptaecin
catsup
XM_003395924
defensin
XR_131852
XM_003402966
apidaecin
punch
XM_003394653
abaecin
XM_003401903
XM_003396463
BGRP-LC
XM_003399472
XM_003394713
BGRP2
hopscotch
XM_003397996
BGRP1
relish
XM_003401893
PGRP-S3
Signalling
XM_003400112
PGRP-LB
Receptor
NCBI asession
Gene
Class
AATTGCAAGATCAACCTGCC
GTCGCCCTTCTGCATATCTC
AGCGGCATAATACGTTGTGT
CCGATCATCCATTCGTATTC
GCG AAA GGT CCA TTT ACT CA
TTACCATGACGAGTCACCAA
ATTGCCAGGACACTTTCAAC
AAAGAATTGGACGCAAATGG
CAATTTCTTCACCGCATCCT
TATGGGCAAGAAGATTCGAC
GCGTTCTATGACCACCTGTT
TTCATCGTACTGGCTCTCTTCTG
GTCTGCCTTTGTCGCAAGAC
CCCGACTAATGTACCTGCCA
GCCACAATATGTGGAATCCT
CACAGACTGAAGCAGGTTGA
CAGCAGTAAAAATCCCCGAC
CAGCCACCTACGACAGATTT
TAACTCCCTTTGGAAACACG
AACGTGGAAGTCAAAGATGG
CGTGAAGGAGCTCATACCAT
TTTCCATGTTGCTCGCTTCG
Forward primer
CCTACCCAAAGACAAGGCAA
AAGATCGAACTGCTATCCGC
CCGAGGGATAGAAAGTCTCC
ACCTGCACTTGATATCCCTG
ACA CTG GTA TCC CAA CAA GC
ATGAGGAACCAAAGCATGAG
TACAAGCTGGAAACGGAAAC
CAGCGAACTGATGTCCAAGA
CCTCGTTATTTGGCTTGCAT
GTGTACATCGTTCACGCATC
TACAGGTTACTCCACAGCCC
AGCCGTAGTATTCTTCCACAGC
GACATTAGTCGCGTCTTCTTCG
GAAGGTGCGAATGTGTTGGA
ATGACCAGGGTTTGGTAATG
CATATGGGTAATTTGGTGCC
CAGCACGAATAAGTGAACATA
GTACATTCCGCTTGTGTCC
GGCGGTAAAATACTGAACGA
GCGAACGATGACTTGGTATT
CCAGGACTCATAGTGGCTGT
CGCGGTTTCCCTTTCGATATTAG
Reverse primer
80
175
190
329
164
110
355
211
259
131
219
212
85
139
131
141
353
156
101
249
206
200
Product size
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Schlüns et al. 2010
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Primer reference
Table S1. Details for primers of candidate genes. Primers with no reference were developed for this study according to the protocol in Brunner et al. 2013.
53
Housekeeping
Metabolic
AF_492888
XM_003401944
FN_391388
FN_391387
XM_003393291.1
ef1a
PLA2
RPL13
Perit
XM_003399563
AK
XM_003402572
cytochrome P450
XM_003402655
apolipophorin III
vitellogenin
XM_003401245
jafrac
b
XM_003394777
peroxiredoxin5
ROS
NCBI asession
Gene
Class
GGTGCACGATGTGTAACTGAC
GGTTTAACCAGCCAGCTAGAAA
TATCTTTCAATGCCCAGGAG
GCTGGTGACTCGAAGAACAATC
CTGGACTCTGGTGTCGGTAT
GAATGCGCAAAGAAGGTAGC
ATCAGGCTCAAACGAACATC
GTGACAAGCGAAGAGACTATTATG
CTCACTTCAGTCACTTGGCA
TCACACCAGGATGTTCCAAGAC
Forward primer
TCGCTCTTGAAATACGTAACGC
CTTCACAGGTCTTGGTGCAA
GTCGTAACAAATGTCATGCG
GGGTGGTTCAACACAATAACC
GTCTTTTGGTGGATGCTTGT
CGCTTTCCGCTCTTGTAATC
TTCGTTCACTTGTTGCTGAG
CGTGTTATCTGGCGTGAC
GCCAGCAGGACATACTTCTC
TTCTGCTCCGTGTTCTTTACCC
Reverse primer
141
83
129
74
129
313
269
154
290
146
Product size
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Brunner et al. 2013
Li et al. 2010
Brunner et al. 2013
Brunner et al. 2013
Primer reference
Table S1. (Continued) Details for primers of candidate genes. Primers with no reference were developed for this study according to the protocol in Brunner et al. 2013.
4. WORKER LONGEVITY IN A PRIMITIVELY EUSOCIAL INSECT: IMMUNE
SENESCENCE DOES NOT AFFECT WORKER RESISTANCE TO AN
ECOLOGICALLY RELEVANT PARASITE.§
Abstract
In insect societies, worker lifespan strongly affects colony success. Long-lived workers may be
favoured as they allow for large colony size, which typically correlates with higher reproductive
success. At the same time, long-lived and therefore, on average, older workers should pose an
increased risk of parasite transmission within the colony. This is because older workers are more
likely to have encountered a parasite, and defence (immune) functions commonly decline with
age. To counter this epidemic risk, social insects may have evolved adaptations related to worker
age. Old workers, for example, tend to be foragers and often have limited contact with younger
workers engaged in in-hive tasks. Our study organisms, the bumblebees, show little such agerelated changes in behaviour and might be especially vulnerable to infection through old workers
in the colony. Using the bumblebee, Bombus terrestris, and its trypanosomatid gut parasite,
Crithidia bombi, we experimentally tested the role of worker age for the resistance to this
ecologically important parasite. Contrary to expectations, we found that older workers were less
heavily infected, with fewer infective cells in the gut but nonetheless were also more likely to
transmit the parasite. Whereas this increased transmission could not be explained by different
strains infecting worker of different ages, it may be due behavioural changes. We conclude that
immune senescence may not necessarily affect resistance to this important parasite. Age effects,
however, may be relevant for the spread of the infection through the colony and may be mediated
by behaviour rather than by infection intensity. Adaptive demography in social insects for this
reason should be analysed, too, in the light of preventing or controlling disease epidemics.
§
A version of this Chapter is in preparation as S. D. Buechel , M. Arnoldini & P. Schmid-Hempel.
54
Introduction
Lifespan is a highly variable trait that differs among and within species (Keller & Genoud 1997;
Carey & Judge 2000; De Magalhaes & Costa 2009). Life history theory explains variation in
longevity by selection acting to shape individual life history such as to maximise Darwinian
fitness. A long life, for example, implies investment into body maintenance and should be
favoured only if the likelihood to later reproduce and thus to survive to an older age is high. If
instead adult survival were low, organisms should sacrifice prolonged lifespan for early
reproduction (Williams 1966; Stearns 1992).
Eusocial species such as the ants, bees, wasps and termites live in highly integrated societies that
often consist of one reproductive queen (and king in termites) and many sterile workers that
perform all non-reproductive tasks needed for the colony to survive and grow; this includes nest
construction, foraging, maintenance and caring for the queen and her brood (Hölldobler & Wilson
1990). Colony fitness, hence, depends crucially on worker performance (Bourke 1995). Among
other worker characteristics, variation in worker lifespan should therefore be selected so as to
contribute to the colony performance (Schmid-Hempel 1992; Moret & Schmid-Hempel 2009;
Armitage & Boomsma 2010; Helft et al. 2012). For example, long-lived workers might be
beneficial as they allow the colony to grow large and thus be better defended, be ergonomically
more efficient, and ultimately rear more males and daughter queens that contribute to the next
generation (Hölldobler & Wilson 1990; Adams 1990; Muller & Schmid-Hempel 1992; Carey
2001; Palmer 2004; Lopez-Vaamonde et al. 2009). But long-lived workers may also be costly - in
terms of their "production cost" (i.e. producing higher "quality" workers) as well as through
secondary effects. For example, we expect a secondary cost, as colonies with long-lived workers
might be more vulnerable to parasites. This is because colonies with long-lived workers will, on
one hand, have relatively more old workers than colonies with short-lived workers. Old workers
may be more likely to have encountered parasites and thus be infected, but may also be less
protected as the immune system undergoes senescence over an organisms life (Franceschi et al.
2000; Haussmann et al. 2005; Stanley 2012). Furthermore, increased foraging in bumblebees - as
55
is typical for older workers in general - correlates with a decreased immune response to an
implanted object (Konig & Schmid-Hempel 1995; Doums & Schmid-Hempel 2000); this age
effect may be mediated by the reduced number of haemocytes (immune cells involved in
encapsulation response) in old workers (Moret & Schmid-Hempel 2009). Similar patterns of a
decline in important immune function with age have been found in honeybees (Amdam et al.
2005; Schmid et al. 2008) and ants (Helft et al. 2012) - although as of yet there is no direct
evidence for diminishing resistance to real pathogenic infections with older age structure of the
colony. Nevertheless, it is conceivable that worker longevity should associate with higher
vulnerability of the colony to infection. On the other hand, long-lived workers would pose an
increased epidemic threat to the colony, both, by being more vulnerable individually and by
allowing for longer infection periods (e.g. leading to higher parasite loads and longer transmission
windows). Additionally, the resulting larger colony sizes also generate more opportunities for
transmission within the colony, perhaps assisted by higher within-colony densities and more
frequent behavioural interactions (Schmid-Hempel 1998; Naug & Camazine 2002; Pie et al. 2004;
Otterstatter & Thomson 2007).
Importantly, social insects have additionally evolved efficient behavioural defences based on
collective decision-making and altruistic behaviour (e.g. hygienic behaviour and mutual grooming
of diseased workers (Cremer et al. 2007; Parker et al. 2011). These social defences complement
the individual immune system, can counter the defence cost to large colonies, and may be another
key to the spectacular ecological dominance of social insects. Taken longevity and social
defences together, we might, for example, imagine that advanced insect societies mitigate the
threat posed by senescing workers through age-based polytheism (Schmid-Hempel & SchmidHempel 1993; Calderone 1995; Calderone & Page 1996; Seid & Traniello 2006; Camargo et al.
2007; Johnson 2008; Mersch et al. 2013). In such a system, young workers would perform tasks
in disease-privileged areas of the colony, such as assisting the queen or brood. Older, presumably
less protected, workers are more likely to be foragers or seek other tasks outside the colony,
56
where they would be more likely to encounter parasites but are less likely to directly transmit
them to important colony members.
In this paper we investigate the role of age-related resistance of workers to an ecologically
important infectious parasite. We use the primitively eusocial bumblebee Bombus terrestris and
its trypanosomatid gut parasite Crithidia bombi, a well-studied model for host – parasite studies in
evolutionary and ecological immunology. Their life cycles are simple: after hibernation mated
bumblebee queens found colonies in spring that grow in numbers of workers until sexuals (males
and gynes) are produced and the colony declines. C. bombi, in turn, is picked up per os via
contaminated flowers and spreads through the colony by infectious cells shed in the faeces of
workers (Durrer & Schmid-Hempel 1994). It has only one host in its life cycle, is not vectored,
but can infect virtually all species of Bombus.
Bumblebees are primitively eusocial insects that did not, or - as far as known - only partly
evolved social defences against parasites (Müller & Schmid-Hempel 1993; Korner-Nievergelt
2003). Workers also show no clear-cut division of labour according to age (Cameron 1989;
Cameron & Robinson 1990; O'Donnell et al. 2000; Jandt & Dornhaus 2009; Jandt et al. 2009) but
task affinity based on body size is common. In particular, small workers are more likely to
perform hive tasks whereas bigger workers forage or guard the nest (Goulson et al. 2002; Goulson
2003; Yerushalmi et al. 2006). With respect to infectious parasites, bumblebees should thus be
susceptible to a demographic shift towards old workers and we could expect that individual,
physiological immunity should be maintained at high levels throughout worker life (see (SchmidHempel 1992; Moret & Schmid-Hempel 2009; Armitage & Boomsma 2010; Helft et al. 2012).
We were thus interested in testing whether in this system worker age has an effect on elements
that affect the risk of an infection and spread of a parasite. In a first experiment we infected old
and young workers with C. bombi to test for age-related differences in resistance, infection load,
and parasite diversity and identity. We then grouped old and young infected workers with naïve,
uninfected nest mates to test the age specific potential to transmit the parasite. In a second
experiment we aimed to investigate differences in the behavioural repertoire of old and young
57
workers likely to influence their transmission potential. We video tracked worker groups that
included an infected young or old focal bee and analysed their respective activity and the distance
kept between workers.
Methods
Collection and culturing
We collected wild queens of B. terrestris in spring 2012 from two distinct populations in
Switzerland (near Aesch BL; and Neunforn TG) and allowed them to establish colonies in the lab.
As soon as the first workers emerged, we transferred the colonies to a circular perlite nests
(Pomeroy & Plowright 1980) and kept them under standardized laboratory conditions (28 ± 2 °C,
60% RH) with constant red light illumination and pollen and sugar water provided ad libitum.
Colonies were carefully checked for the presence of parasites in their faeces and only uninfected
colonies were used in these two experiments.
The parasite, C. bombi originated from naturally infected queens, sampled from the same two
populations in 2010 as from where the queens originated. Single infective cells were isolated and
maintained clonally in liquid medium (Salathé et al. 2012). The five strains used in this
experiment were shown to be equally infective in an earlier experiment (Chapter 2) and had
distinct multi-locus genotypes at three polymorphic microsatellite loci that allowed distinguishing
them by genetic markers.
Experiment 1: Age related resistance and the potential to transmit C. bombi to naïve workers
We marked young callow workers (not yet sclerotized adults that had emerged within the last 24
hours) daily in 21 colonies and randomly assigned them to two treatment groups - infected when
they were 4 days (young workers) or when 18 days old (old workers). We chose these particular
days to ensure that the infection did not interfere with worker development, that it approached a
natural setting (with adults living for 3-4 weeks and becoming infected at various times) as well
as for practical reasons (e.g. to ensure enough bees surviving to the end of the experiment at age
58
of 31 days (Chapter 3). At their respective age of infection workers were isolated and starved for
2 hrs and subsequently fed with a cocktail of 40’000 infectious C. bombi cells containing five
distinct parasite strains. Workers that did not ingest the cocktail after 1 h were excluded from the
experiment so as to ensure primary infections and viability of the cells. Workers were then kept
individually to prevent cross infection; 8 d later their faeces were visually checked for the
presence of C. bombi cells, confirming their infection status. For later use, "resistant" was defined
by the absence of infective cells in the faeces. To test the transmission potential of young and old
workers we grouped workers that were successfully infected at either young or old age with six
naïve (uninfected) workers from the respective colony. The artificial group thus consisted of 1
infected and 6 naïve workers and was housed in a plastic box (15 x 20 cm) for a period of 6 days,
which allowed transmission to naïve workers (as tested in a preliminary study; data not shown)
but prevented cross infection. Bees were then isolated for another 8 days, for the parasite to
establish and be readily detected with molecular tools using C. bombi specific primers (see below).
We also measured the infection intensity of young and old workers as the total number of
infective C. bombi cells in the gut (correction for body size did not change the results, as size was
anyway randomized across treatments), and we also genotyped the C. bombi strains that
successfully infected young and old workers.
Molecular analysis
We dissected out the whole gut of all workers and extracted genomic DNA following the protocol
for Blood & Tissue samples from Qiagen (Hilden, Germany). To visualise the infection status of
workers we then amplified a partial sequence of the C. bombi Cyt b gene (primers CB-Cytb2-F,
CB-Cytb2-F; (Schmid-Hempel & Tognazzo 2010) and run the amplification product on an 1.5 %
agarose gel. To quantify the infection intensity, we measured the number of C. bombi cells in the
guts of young and old workers using quantitative real time PCRs (qPCR) based on EvaGreen
(Biotium, Hayward, CA, USA) and primers that were designed to amplify a fragment conserved
in the C. bombi 18s RNA (Primers CriRTF2, CriRTR2 (Ulrich et al. 2011). The qPCRs were
performed on an ABI 7200 Fast Real-time PCR System (Applied Biosystems Rotkreuz
59
Switzerland) in 10 µl reaction volumes containing 2µL eluted DNA, 0.2 µL of each primer (10
µM) and 2µL HOT FIREPol EvaGreen qPCR Mix. We used the thermal profile described in
Ulrich et al. (2011). A dilution series of DNA from a known number of C. bombi cells was
included and served as a standard to calculate the relative infection intensities in the guts of young
and old workers. Including the same dilution series over all plates also allowed standardizing our
quantification across plates. Finally, we genotyped the C. bombi strains that successfully infected
young and old workers at the C. bombi microsatellite markers Cri4, Cri4G9 and Cri1B6 (SchmidHempel & Reber Funk 2004). For this, two multiplex reaction were performed on a ABI 3730xl
DNA sequencer (Applied Biosysytems, Rotkreuz, Switzerland) in 10 µL reaction volumes
including 2 µL eluted DNA in 2 µL 5 x reaction buffer, 0.5 µL dNTPS (2.5 mM), 0.05 µL GoTaq
polymerase (5U/µL) and (1) 0.1 µL of the primers Cri4 and Cri4G9, (2) 0.06 µL of the primer
Cri1B6. PCR conditions were the same as described in (Koch & Schmid-Hempel 2011). The
microsatellites were scored with Peakscanner software V1.0 (Applied Biosystems, Rotkreuz,
Switzerland) and only samples where strains could be clearly identified at all three markers were
included in the analysis
Experiment 2: Behavioural assay and transmission potential
We used the same colonies and an identical infection protocol to compare the transmission
potential of young and old workers (tested as described above) with their behavioural activities
and the physical distance they kept from naïve nest mates. These measures were found to be
relevant (Pie et al. 2004; Otterstatter & Thomson 2007) and seem plausible considering the
transmission mode of the parasite (infective cell shed in faeces and taken up per os). For the
experiment, we marked callows daily over 5 consecutive days and infected them at the age of 4,
respectively 18 days (as above). Workers where then housed with 3 naïve workers each in a box
(as above), as a group of 4 workers could technically be readily tracked (see below). After 6 days
- a time chosen to allow the parasitic infection to develop and hosts becoming highly infectious
for others - the group was recorded twice for 15 min under red light illumination. For the analysis
we chose recordings taken at similar daytimes. We did not analyse the first and last 2.5 min of
60
each recoding to eliminate the effects of possible differences in handling and timing for each box.
Custom software was developed by one of us (AM) using MATLAB (The Mathworks Inc.) to
analyze movies. The analysis proceeded in several steps. First, automated individual recognition
of bumblebees based on their grey values was performed. Second, recognition errors were
corrected manually. Third, the movement of individual bumblebees was tracked by connecting
bumblebees in the single images over time based on proximity. In a fourth step, this tracking was
checked and errors again corrected manually. These procedures resulted in data on activity and
distance maintained between the workers in the analysed movies.
Analyses and statistics
We tested the effect of worker age on resistance to C. bombi infection, defined as the absence of
parasite cells in faeces, using a linear mixed model with binomial error distribution. The model
used worker age (4 days, and 18 days at infection) as independent fixed variable and the identity
of the colony that workers originated from as random variable. The infection intensity of young
and old workers was measured as the number of parasite cells in the gut and was analysed with a
linear mixed model using worker age and colony identity as fixed and random variables,
respectively. The number of parasite cells in workers as determined with quantitative PCR was
log- transformed so that the residuals met the normality criterion. We tested for parasite diversity,
defined as the number of strains that persisted in young and old workers at time of checking, with
a nonparametric Mann-Whitney U-test, and for the identity of the strains that successfully
infected workers of different age (i.e. strain number) by Fisher's exact test. Finally, we tested the
age effect on the potential to transmit C. bombi to naïve nest mates as follows: the number of
infected workers that had been housed with young or old workers was the dependent variable in a
binomial logit link model with worker age as fixed and colony identity as random variable. To
correct for over-dispersed data we included an additional random variable at the level of the
individual observation (the number of data points), which allows to model extra variance in the
data. Behavioural data included the activity (total distance covered during observation) of the
focal bee (young or old infected worker), the "cohesiveness" of the group (average distance
61
between individual workers), and the average distance of the focal bee from naïve workers. These
measures were averaged over time intervals of 30 s (corresponding to 225 frames) and analysed
with linear models that used the worker age as fixed variable and time interval (20 intervals of 30
s), and the identity of the tracked group as random factor. Statistical analyses were carried out in
IBM SPSS Statistics version 19.0, or R statistical software version 2.15.2 (R Core Team 2012).
Note that all interaction terms were not significant and therefore removed from the models
presented here.
Results
To begin with, both, old (proportion infected = 0.7) and young workers (proportion infected =
0.6) were equally likely to contract the experimental infection (GLMMs: F1,184 =1.862, p > 0.1,
Figure 1). Upon primary infection though, old workers were more efficient in clearing the
infection as indicated by a reduced number of parasite cells measured in their gut at 8 d postinfection than younger workers (number of cells old workers: 6.35 x103 ± 1.26 x103 versus young
workers 3.21 x103 ± 0.55 x103; LMM: F=
1,20
= 7.189, p = 0.01, Figure 2). Yet, low infection
intensity in old workers did not directly result in a low transmission potential to the surrounding
naïve nest mates. To the contrary, more workers were infected in groups with old focal workers
compared to groups with young focal workers (proportion of naïve workers infected = 0.21 ± 0.05
S.E. for old workers, and 0.12 ± 0.035 for young workers; GLMMs: z = 2.022, p = 0.04, Figure 3).
This increased risk to transmit the parasite was not explained by differences in parasite diversity,
as the number of strains did not differ according to age (old workers: 1.92 ± 0.199 S.E. versus
young 1.96 ± 0.204; Mann-Whitney U test: p = 0.9), nor could it be assigned to different parasites
strains having infected workers of young or old age (Fishers exact test: p = 0.9). Rather,
behavioural differences may have accounted for the increased transmission.
62
Proportion of infected bees (Mean ± 95% CI)
!",#
!"+#
!"*#
!")#
!"(#
75
115
!"'#
!"&#
!"%#
!"$#
!"!#
Young (age 4 days)
Old (age 18 days)
Figure 1. The proportion of young (open bar) and old workers (filled bar) with established infections
upon experimental exposure, defined as the presence or absence of C. bombi cells in their faeces 8 days
post-infection (GLMMs; F1,184 =1.862, p > 0.1). Lines represent 95% C.I., small numbers indicate
sample sizes (number of workers).
± SE)
)"
Mean infection intensity (Cells per bee
*"
x105
+"
("
'"
35
&"
%"
28
$"
#"
!"
Young (age 4 days)
Old (age 18 days)
Figure 2. The number of parasite cells in the gut of young (open bars) and old workers (filled bar), as
measured with qPCR, 13 days post-infection (LMM; F= 1,20 = 7.189, p = 0.01). Bar plot showing mean
number of cells according to worker age, S.E. (lines) and sample sizes (workers, small numbers).
63
Proportion infected bees in groups (Mean ± SE)
0.30
0.25
0.20
0.15
38
0.10
38
0.05
0.00
Young (age 4 days)
Old (age 18 days)
Figure 3. The proportion of naïve workers that were infected when grouped 6 days with young infected
worker (open bar) and old infected workers (filled bar), (GLMMs; z = 2.022, p = 0.04). Groups
contained a total of 7 workers and the infection status was determined using C. bombi specific primers.
Bars represent mean infections of groups, and lines show S.E. Sample sizes (workers) are given in small
numbers.
$!!"#
,!"#
Proportion of strains
+!"#
*!"#
-#
)!"#
.#
/#
(!"#
'!"#
0#
23
24
Young (age 4 days)
Old (age 18 days)
1#
&!"#
%!"#
$!"#
!"#
Figure 4. Percentage of different C. bombi strains (shades of grey, letters A-F) present in young
workers (open bar) or old workers (filled bars), (Fishers exact test; p = 0.9). Strains were typed 13 days
post- infection.
64
We found that old workers showed higher levels of activity (distance in cm old workers: 24.68 ±
1.02 S.E. and young workers: 16.82 ± 0.67; LMM: F1,22 = 6.109, p = 0.02, Figure 5a) which may
have led to a less cohesive group structure (group with old focal worker 5.70 ± 0.09 S.E. versus
young focal worker: 4.7 ± 0.09; LMM: F1,22 = 4.871, p = 0.03, Figure 5b) and a tendency to keep
more distance to naïve workers (old workers: 16.75 ± 0.29 S.E. and young workers: 13.98 ± 0.33;
LMM: F1,22 = 3.499, p = 0.07, Figure 5c).
65
(a)
35
Distance (cm)
30
25
12
20
15
12
10
5
0
Young focal worker
Old focal worker
(b)
7
Group cohesiveness (cm)
6
12
5
4
12
3
2
1
0
Young focal worker
Old focal worker
(c)
20
18
Distance focal to all (cm)
16
12
14
12
12
10
8
6
4
2
0
Young focal worker
Old focal worker
Figure 5. Measures of activity and proximity of young and old infected focal workers, grouped together
with 6 naïve workers from their respective colony (n= 12 colonies for both groups). Data from halfautomated video tracking was obtained for: (a) the total distance (in cm) the focal workers walked
during the 10 min of tracking ("activity measure", old workers: LMM; F1,22 = 6.109, p = 0.02, Figure
5a), (b) the cohesiveness of the group measured as the mean pair wise distances (in cm) of all workers
in groups (LMM; F1,22 = 4.871, p = 0.03, Figure 5b), and (c) the distance focal workers kept from the
surrounding naïve workers (LMM; F1,22 = 3.499, p = 0.07, Figure 5c). The analysis is based on
averages from of 30 s intervals (see text). Bars represent the estimated marginal means (± S.E.) derived
from a model using worker age as fixed factor and the time intervals as well as the video identity as
random factor (see text).
66
Discussion
Contrary to expectation, we found that older bumblebee workers seemed to have cleared the
infection better as compared to young workers. Despite a lower number of infectious cells in the
gut and thus in their faeces, old workers nevertheless infected more naïve workers than did their
younger sisters. This high potential to transmit the parasite could not be explained by parasite
diversity nor could it be assigned to different parasites strains infecting young or old workers.
Rather, increased transmission may have resulted from age-related differences in the behavioural
repertoire of bumblebees, for example, increased levels of activity in old workers infected with C.
bombi.
At first sight, bumblebees therefore maintain functional immunity against this parasite as they age.
Low infection intensity in old workers however, does not primarily contradict the observation that
the immune system of bumblebees declines with age, but rather suggests that at least against C.
bombi infections, immune senescence may be less relevant. One possible, although not yet tested
and therefore hypothetical scenario may be that old workers are protected against C. bombi by
means of the microbial gut community (Koch & Schmid-Hempel 2011), whose composition may
differ from that of young workers. Also not all immune functions necessarily always decline with
worker age. In the case of prophenoloxidase, an enzyme relevant for melanisation response,
higher levels were measured in older ants and honey bees (Schmid et al. 2008; Armitage &
Boomsma 2010).
We may hypothesize that resistance to C. bombi infection is also explained by selection on
worker longevity acting at the colony level. Long-lived workers in the colony would allow
achieving big colony sizes (Carey 2001), which ultimately results in increased colony fitness
(Muller & Schmid-Hempel 1992; Schmid-Hempel et al. 1993; Lopez-Vaamonde et al. 2009).
That old bumblebees nevertheless infected more naïve workers may at first be an intriguing
disadvantage, as it would allow a parasite spreading through the colony more easily. In the
present case, and considering the virulence pattern of C. bombi, however, the virulence costs of an
infection are mainly paid by the young, daughter queens that if infected, are less likely to found
67
their own colony next year (Brown et al. 2003). Interestingly, young queens seem also more
resistant to C. bombi infection than workers are and can only be infected upon ingestion of a high
number of cells (S. Barribeau, pers. comm.). Low infection intensity in older workers may thus in
the first place be relevant to reduce the risk of infecting daughter queens, which emerge only at
the end of colony cycle when many older workers are present. On one hand, therefore, and
besides other constraints, the appearance of sexual brood late in the colony cycle (as also
determined by the development time for the sexual brood) could also function as temporal buffer
against the presence of younger workers and thus a high number of infectious cells that circulate
in the colony. We could not test here whether the above-mentioned changed behaviour of older
workers would also mean increased contacts with daughter queens of the colony and, on balance,
would thus still lead to more transmission. As to the males, these are produced along with last
workers before young queens emerge (Goulson 2003). Males may be especially prone to infection
(Vainio et al. 2003; Gerloff et al. 2003; Baer et al. 2005) but leave the colony only few days after
emergence before shedding infective cells in the colony. It thus remains to further investigate
whether colony demography in B. terrestris - with many older and less infected workers towards
the end - may also be understood as adaptation to the threat posed by infectious parasites for an
annual life cycle of the colony. Alternatively, and in contrast to our finding, old bumblebee
workers were often reported to be inactive (Jandt & Dornhaus 2009) which is also true, for
example, for ants (Cole 1992; Pinter-Wollman et al. 2012). At least in some bumblebee colonies,
workers tend to shift their spatial zone of activity further away from the colony centre as they age
(Jandt & Dornhaus 2009). In our experimental setup, increased activity in old workers may be a
result of them seeking distance from their nest mates. Groups with old infected workers indeed
showed lower cohesiveness and old infected workers tended to keep more distance to all other
workers in the groups compared to younger workers (P= 0.07).
In this paper we aimed to test the effect of worker age resistance to an ecologically relevant
parasite in the light of worker longevity in primitively eusocial insects. We conclude that immune
senescence may not be relevant against C. bombi infection in bumblebees. Age effects, however,
68
can become relevant for the spread of a disease in the colony, but this may be mediated by
behavioural changes rather than infection loads per se. Adaptive demography in social insects
should thus be analysed, too, in the light of preventing or controlling disease epidemics.
69
References
Adams, E. S. 1990. Boundary Disputes in the Territorial Ant Azteca-Trigona - Effects of Asymmetries in
Colony Size. Animal Behaviour, 39, 321–328.
Amdam, G. V., Aase, A. L. T. O., Seehuus, S.-C., Kim Fondrk, M., Norberg, K. & Hartfelder, K.
2005. Social reversal of immunosenescence in honey bee workers. Experimental Gerontology, 40,
939–947.
Armitage, S. A. O. & Boomsma, J. J. 2010. The effects of age and social interactions on innate immunity
in a leaf-cutting ant. Philosophical transactions of the Royal Society of London Series B, Biological
sciences, 56, 780–787.
Baer, B., Krug, A., Boomsma, J. J. & Hughes, W. O. H. 2005. Examination of the immune responses of
males and workers of the leaf-cutting ant Acromyrmex echinatior and the effect of infection. Insectes
Sociaux, 52, 298–303.
Bourke, A. 1995. Social Evolution in Ants. Princeton University Press,
Brown, M. J., Schmid-Hempel, R. & Schmid-Hempel, P. 2003. Strong context‐dependent virulence in a
host–parasite system: reconciling genetic evidence with theory. Journal Of Animal Ecology, 72, 994–
1002.
Calderone, N. W. 1995. Temporal division of labor in the honey bee, Apis melliferaA developmental
process or the result of environmental influences? Canadian Journal Of Zoology-Revue Canadienne
De Zoologie, 73, 1410–1416.
Calderone, N. W. & Page, R. E., Jr. 1996. Temporal polyethism and behavioural canalization in the
honey bee,Apis mellifera. Animal Behaviour, 51, 631–643.
Camargo, R. S., Forti, L. C., Lopes, J., Andrade, A. & Ottati, A. 2007. Age polyethism in the
leaf‐cutting ant Acromyrmex subterraneus brunneus Forel, 1911 (Hym., Formicidae). Journal of
Applied Entomology, 131, 139–145.
Cameron, S. A. 1989. Temporal Patterns of Division of Labor among Workers in the Primitively Eusocial
Bumble Bee, Bombus griseocollis (Hymenoptera: Apidae) 1). Ethology, 80, 137–151.
Cameron, S. A. & Robinson, G. E. 1990. Juvenile hormone does not affect division of labor in bumble
bee colonies (Hymenoptera: Apidae). Annals of the Entomological Society of America, 83, 626–631.
Carey, J. R. 2001. Demographic mechanisms for the evolution of long life in social insects. Experimental
Gerontology, 36, 713–722.
Carey, J. R. & Judge, D. S. 2000. Longevity records. Odense: Odense University Press.
Cole, B. J. 1992. Short-term activity cycles in ants: age-related changes in tempo and colony synchrony.
Behavioral Ecology and Sociobiology, 31, 181–187.
Cremer, S., Armitage, S. A. O. & Schmid-Hempel, P. 2007. Social immunity. Current Biology, 17,
R693–R702.
De Magalhaes, J. P. & Costa, J. 2009. A database of vertebrate longevity records and their relation to
other life-history traits. Journal Of Evolutionary Biology, 22, 1770–1774.
Doums, C. & Schmid-Hempel, P. 2000. Immunocompetence in workers of a social insect, Bombus
terrestris L., in relation to foraging activity and parasitic infection. Canadian Journal Of ZoologyRevue Canadienne De Zoologie, 78, 1060–1066.
Durrer, S. & Schmid-Hempel, P. 1994. Shared use of flowers leads to horizontal pathogen transmission.
70
Proceedings Of The Royal Society Of London Series B-Biological Sciences, 258, 299–302.
Franceschi, C., Bonafè, M. & Valensin, S. 2000. Human immunosenescence: the prevailing of innate
immunity, the failing of clonotypic immunity, and the filling of immunological space. Vaccine, 18,
1717–1720.
Gerloff, C. U., Ottmer, B. K. & Schmid-Hempel, P. 2003. Effects of inbreeding on immune response and
body size in a social insect, Bombus terrestris. Functional Ecology, 17, 582–589.
Goulson, D. 2003 April 1. Bumblebees: Their Behaviour and Ecology. Oxford University Press,
Goulson, D., Peat, J., Stout, J. C., Tucker, J., Darvill, B., Derwent, L. C. & Hughes, W. O. H. 2002.
Can alloethism in workers of the bumblebee, Bombus terrestris, be explained in terms of foraging
efficiency? Animal Behaviour, 64, 123–130.
Haussmann, M. F., Winkler, D. W., Huntington, C. E., Vleck, D., Sanneman, C. E., Hanley, D. &
Vleck, C. M. 2005. Cell-mediated immunosenescence in birds. Oecologia, 145, 269–274.
Helft, F., Tirard, C. & Doums, C. 2012. Effects of division of labour on immunity in workers of the ant
Cataglyphis cursor. Insectes Sociaux, 59, 333–340.
Hölldobler, B. & Wilson, E. O. 1990. The Ants. Harvard University Press.
Jandt, J. M. & Dornhaus, A. 2009. Spatial organization and division of labour in the bumblebee Bombus
impatiens. Animal Behaviour, 77, 641–651.
Jandt, J. M., Huang, E. & Dornhaus, A. 2009. Weak specialization of workers inside a bumble bee
(Bombus impatiens) nest. Behavioral Ecology and Sociobiology, 63, 1829–1836.
Johnson, B. R. 2008. Within-nest temporal polyethism in the honey bee. Behavioral Ecology and
Sociobiology, 62, 777–784.
Keller, L. & Genoud, M. 1997. Extraordinary lifespans in ants: a test of evolutionary theories of ageing.
Nature, 389, 958–960.
Koch, H. & Schmid-Hempel, P. 2011. Socially transmitted gut microbiota protect bumble bees against an
intestinal parasite. Proceedings Of The National Academy Of Sciences Of The United States Of
America, 108, 19288–19292.
Konig, C. & Schmid-Hempel, P. 1995. Foraging activity and immunocompetence in workers of the
bumble bee, Bombus terrestris L. Proceedings. Biological sciences / The Royal Society, 260, 225–227.
Korner-Nievergelt, P. 2003 July 28. Dynamics and consequences of immune sysytem activation in
individuals of Bombus Terrestris (L.) (Hymenoptera: Apidae). Diss. Swiss Federal Institute of
TEchnology Zurich,
Lopez-Vaamonde, C., Raine, N. E., Koning, J. W., Brown, R M, Pereboom, J. J. M., Ings, T. C.,
Ramos-Rodriguez, O., Jordan, W. C. & Bourke, A. F. 2009. Lifetime reproductive success and
longevity of queens in an annual social insect. Journal Of Evolutionary Biology, 22, 983–996.
Mersch, D. P., Crespi, A. & Keller, L. 2013. Tracking Individuals Shows Spatial Fidelity Is a Key
Regulator of Ant Social Organization.
Moret, Y. & Schmid-Hempel, P. 2009. Immune responses of bumblebee workers as a function of
individual and colony age: senescence versus plastic adjustment of the immune function. Oikos, 118,
371–378.
Muller, C. B. & Schmid-Hempel, P. 1992. Correlates of reproductive success among field colonies of
Bombus lucorum: the importance of growth and parasites. Ecological Entomology, 17, 343–353.
71
Müller, C. B. & Schmid-Hempel, P. 1993. Exploitation of cold temperature as defence against parasitoids
in bumblebees. Nature, 363, 65–67.
Naug, D. & Camazine, S. 2002. The Role of Colony Organization on Pathogen Transmission in Social
Insects. Journal of Theoretical Biology, 215, 427–439.
O'Donnell, S., Reichardt, M. & Foster, R. 2000. Individual and colony factors in bumble bee division of
labor (Bombus bifarius nearcticus Handl; Hymenoptera, Apidae). Insectes Sociaux, 47, 164–170.
Otterstatter, M. C. & Thomson, J. D. 2007. Contact networks and transmission of an intestinal pathogen
in bumble bee (Bombus impatiens) colonies. Oecologia, 154, 411–421.
Palmer, T. M. 2004. Wars of attrition: colony size determines competitive outcomes in a guild of African
acacia ants. Animal Behaviour, 68, 993–1004.
Parker, B. J., Barribeau, S. M., Laughton, A. M., de Roode, J. C. & Gerardo, N. M. 2011. Nonimmunological defense in an evolutionary framework. Trends in ecology & evolution, 26, 242–248.
Pie, M. R., Rosengaus, R. B. & Traniello, J. 2004. Nest architecture, activity pattern, worker density and
the dynamics of disease transmission in social insects. Journal of Theoretical Biology, 226, 45–51.
Pinter-Wollman, N., Hubler, J., Holley, J.-A., Franks, N. R. & Dornhaus, A. 2012. How is activity
distributed among and within tasks in Temnothorax ants? Behavioral Ecology and Sociobiology, 66,
1407–1420.
Pomeroy, N. & Plowright, R. C. 1980. Maintenance of bumble bee colonies in observation hives
(Hymenoptera: Apidae). The Canadian Entomologist, 112, 321–326.
R Core Team. 2012. R: A language and environment for statistical computing. R Foundation for Statistical
Computing, Vienna,
Salathé, R., Tognazzo, M., Schmid-Hempel, R. & Schmid-Hempel, P. 2012. Probing Mixed-Genotype
Infections I: Extraction and Cloning of Infections from Hosts of the Trypanosomatid Crithidia bombi.
PLOS ONE, 7, e49046.
Schmid, M. R. M., Brockmann, A. A., Pirk, C. W. W. C., Stanley, D. W. D. & Tautz, J. J. 2008. Adult
honeybees (Apis mellifera L.) abandon hemocytic, but not phenoloxidase-based immunity. Journal Of
Insect Physiology, 54, 6–6.
Schmid-Hempel, P. 1992. Worker castes and adaptive demography. Journal Of Evolutionary Biology, 5,
1–12.
Schmid-Hempel, P. 1998. Parasites in Social Insects. Princeton University Press.
Schmid-Hempel, P. & Reber Funk, C. 2004. The distribution of genotypes of the trypanosome parasite,
Crithidia bombi, in populations of its host, Bombus terrestris. Parasitology, 129, 147–158.
Schmid-Hempel, P. & Schmid-Hempel, R. 1993. Transmission of a pathogen in Bombus terrestris, with a
note on division of labour in social insects. Behavioral Ecology and Sociobiology, 33, 319–327.
Schmid-Hempel, P., Winston, M. L. & Ydenberg, R. C. 1993. Foraging of individual workers in relation
to colony state in the social Hymenoptera. Canadian entomologist, 125, 129–129.
Schmid-Hempel, R. & Tognazzo, M. 2010. Molecular divergence defines two distinct lineages of
Crithidia bombi (Trypanosomatidae), parasites of bumblebees. The Journal of eukaryotic microbiology,
57, 337–345.
Seid, M. A. & Traniello, J. F. A. 2006. Age-related repertoire expansion and division of labor in Pheidole
dentata (Hymenoptera: Formicidae): a new perspective on temporal polyethism and behavioral
72
plasticity in ants. Behavioral Ecology and Sociobiology, 60, 631–644.
Stanley, D. 2012. Aging and immunosenescence in invertebrates. Invertebrate Survival Journal, 9, 102–
109.
Stearns, S. C. 1992. The Evolution of Life Histories. Oxford: OUP Oxford.
Ulrich, Y., Sadd, B. M. & Schmid-Hempel, P. 2011. Strain filtering and transmission of a mixed infection
in a social insect. Journal Of Evolutionary Biology, 24, 354–362.
Vainio, L., Hakkarainen, H., Rantala, M. J. & Sorvari, J. 2003. Individual variation in immune function
in the ant Formica exsecta; effects of the nest, body size and sex. Evolutionary Ecology, 18, 75–84.
Williams, G. C. 1966. Pleiotropy, Natural-Selection, and the Evolution of Senescence. Evolution, 11, 398–
411.
Yerushalmi, S., Bodenhaimer, S. & Bloch, G. 2006. Developmentally determined attenuation in circadian
rhythms links chronobiology to social organization in bees. Journal of Experimental Biology, 209,
1044–1051.
73
5. GENERAL DISCUSSION
In this thesis I made use of a primitively eusocial bumblebee and its infectious gut parasite C.
bombi to investigate the importance of worker longevity on the infection and transmission of this
ecologically relevant parasite. In particular I tested how life span variation at the colony level and
the underlying colony demography relates to the within colony epidemics and the potential threat
to the colony's survival and reproduction. I thereby considered the epidemic effect of long versus
short worker lifespans and tested how lifespan may be associated with physiological immunity.
The role of senescence was thereafter assessed with respect to immunity and for the potential to
transmit a disease. A ‘healthy’, reproductive colony in any case should maximize colony size,
while keeping a within-colony epidemic low.
As a starting point, long-lived, and thus old workers, may be a risk for the colony exposed to
parasites. This is because older workers have an increased probability to have already
encountered a parasite brought into the colony by nest mates. Furthermore, longer lifespan means
long transmission periods that may additionally render the colony vulnerable to the spread of
diseases (Chapter 2). Individual resistance as a possible response may mitigate this risk if
positively related to worker longevity (Chapter 3, 4); it should also be important to sustain colony
growth in bumblebees (Chapter 2). (Table 1 summarizes these findings). In more detail, and in an
epidemiological context, lifespan (here the time an adult worker spends working for the colony)
determines the time of residence of a given parasite in a host individual (e.g. leading to higher
infection intensities) and the length of time the host can transmit the disease. In Chapter 2, I
investigated these patterns.
A long lifespan of workers resulted in more other infected workers in the group. By keeping
group size constant and experimentally altering worker lifespan and birth rate, I controlled for
density effects and the immunological background of workers. This high epidemic potential of
long-lived workers would become even more relevant at high densities, yet another characteristic
that facilitates parasite spread (Naug & Camazine 2002; Pie et al. 2004). Such a positive
74
relationship between colony size and worker longevity was proposed before (O'Donnell & Jeanne
1992; Carey 2001) and may suggest selection for increased individual immunity in long-lived
workers. This was shown by workers from colonies with long-lived workers being more resistant
to experimental infections and their colonies also having larger size (Chapter3). But how about
variation in individual resistance? Resistance as a plastic, density dependent response was shown
in this system before (Moret & Schmid-Hempel 2009; Ruiz-Gonzalez et al. 2009). Moreover,
immune responses were often shown to change over an individuals life and especially showing
senescence in later life (Doums et al. 2002; Moret & Schmid-Hempel 2009) The increased
immune investment associated with long-lived workers therefore may not be relevant to decrease
parasite transmission in colonies with long-lived workers. In Chapter 4, I found the opposite.
Immune senescence did not affect functional immunity at old age but resistance to C.bombi was
upheld in older workers. This finding highlights the importance for old workers to be well
protected. Nevertheless, in Chapter 4, I also found that older workers were more likely to transmit
the parasite. This overall effect must therefore be due to changes beyond the individual immune
defences. A likely candidate is changed behaviours.
Table 1. Summary of epidemic characteristics of colonies with long-lived workers and colonies with short-lived
workers.
Effect
Long-lived workers
Short-lived workers
Chapter 2
Transmission potential
High
Low
Chapter 3
Resistance
High
Low
Chapter 3
Colony size (Density)
High
Low
Chapter 4
Effect of Immune senescence
Low
Low
A hypothetical scenario: How parasite pressure may act as driving force for colony social
organization
Social insects evolved from having simple, small societies to having large colonies with complex
colony structure and an efficient system for division of labour. These transitions were associated
with changes in the mating system, the evolution of caste/colony organisation, and are typically
75
considered to have evolved for ergonomic benefits (Oster & Wilson 1978), that is, increasing the
economic benefits for the colony. Evolution along this trajectory was favoured by kinship effects
e.g. (Hughes et al. 2008). However, the role of parasites for the evolution of social insect colony
organisation is probably often underestimated. Based on the results and insights presented in this
thesis, I here suggest a hypothetical scenario of how parasite pressure can act as a selective force
for the evolution of colony organization.
To start with, a few preconditions are necessary for this scenario to be plausible. These can be
summarized as follows:
(1) It is assumed that social evolution went from small to large colonies, from singly mated to
multiply mated queens, from monogyny to polygyny, and from simple colony structure to
complex structure, eventually with the presence of defined castes (Bourke 1999; Hughes et al.
2008; Boomsma 2009). (2) Social evolution offered conditions for increased longevity as a longer
working life is the main contribution of individuals towards their own (inclusive) success (Keller
& Genoud 1997; Keller 1998; Carey 2001). (3) Long-lived workers may also add to larger colony
size as discussed in (Carey 2001). Empirical evidence is given in Chapter 2 and by (O'Donnell &
Jeanne 1992). (4) Parasites are ubiquitous and have important selective effects on social insects.
This argument was first summarized by P. (Schmid-Hempel 1998) and is supported by en ever
increasing literature e.g. (Boomsma et al. 2005). (5) The risk of contracting an infection from the
outside increases with colony size. This is shown, for example, for bumblebees (Muller &
Schmid-Hempel 1992), social spiders (Hieber & Uetz 1990) and colonial birds (Brown & Brown
1986). (6) Parasite transmission is more frequent in large colonies (Naug & Camazine 2002; Pie
et al. 2004; Otterstatter & Thomson 2007) and among long-lived workers (7) (discussed in
Chapter 2).
76
The scenario
Given the above-mentioned preconditions, the evolution towards more complex social systems
may have been driven by parasite pressure according to the following steps: (visualized in Figure
1).
$"
&"
Colony size
$%"
#"
!"
'"
iii) No/limited worker reproduction
ii) Overlapping generations
'%"
i) Nest as locus
Origin of eusociality
Short-lived
Critical lifespan
Long-lived
Individual worker lifespan
Figure 1. A sketch showing how colony size, worker longevity, and parasite pressure may have
influenced the evolutionary path towards more complex social structure. Firstly, several steps (i-iii,
trajectory A) mark the origin of sociality to the evolution of complex colonies and may have led to an
increase in worker longevity and colony size. At a certain point (B) the colony reaches the critical
epidemic potential in terms of size and worker longevity. The benefits of colony size at this point may
no longer outweigh the cost of increased parasite transmission and the higher chances of sustaining an
epidemic (shaded grey areas). From there, social insect evolution may have followed different paths as
dictated, among other things by the critical density of workers within a colony. Above this density,
parasite transmission is too high and the colony risks serious systemic effects of an infection; hence,
regions C, D, E are not available for evolution. Critical density is affected by colony size and is
inversely correlated with worker lifespan, as more workers are present in the colony at any one time.
From point B, evolution could maintain colonies at critical colony size and worker lifespan (option B).
Alternatively, selection for larger colony size could persist, which increases efficiency and thus
reproductive output. But selection for large colony size may have decreased worker longevity (C'&C)
resulting in a high worker turnover rate, or it may have favoured additional, behavioural adaptations (D)
to counter the additive epidemic potential imposed by colony size and worker longevity, leading to
large complex colonies. Evolution towards small colony size (E’) seems unlikely, as this would entail a
loss of reproductive capacity. In all, parasite pressure could generate a range of worker longevities and
colony sizes, depending in the nature of the selective forces and the relationship of colony size with
worker longevity and epidemic risks.
77
(a) Along the evolutionary pathway from the origin of sociality to advanced systems, the
reproductive division of labour, may first have put selection on worker longevity towards longerlived worker and, consequently, larger colony size (A in Figure 1: trajectory from the origin
towards point B).
The steps that perhaps prolonged worker longevity have to some extent been envisioned in (Carey
2001) and, in brief, include:
(i) Nest as locus: The nest originated as a protected microclimate for a singly mated female,
her brood and food provisions. As this female spent longer time in the protected nest,
mortality rates due to predation were reduced, and there was selection for decreased rate of
senescence. The female thus may have become longer lived (Williams 1966; Stearns 1992)
and more likely to be alive when the brood emerged to form a social group. Along with
selection on the female (queen), protection, shelter and food provisioning, may also have
improved larvae and adult survival and resulted in offspring staying longer in the nest. This
in turn allowed for further selection to decrease senescence rate in adults; thus, as workers
evolved, they became longer-lived, too.
(ii) Overlapping generations: Females may have reared their own brood in the sheltered
nest, occasionally providing their sister’s brood. This adaptation likely allowed females to
spend on the average more time in the nest. Extrinsic mortality once more may have been
decreased.
(iii) Reproductive division of labour: The transition to sterile worker castes according to
(Bourke 1999) and (Rodriguez-Serrano et al. 2012) may be a consequence of increased
colony size and competition over reproduction. As the number of individuals increases
chances to be part of the reproductive caste decrease. Loss of reproduction may have saved
resources for body maintenance and repair, prolonging individual life span at the proximate
level (Stearns 1992). Colony size increases as sisters rear brood more efficiently.
78
(b) At a certain point, evolution may reach a critical point (B in Figure 1). This is because the
"epidemic potential", i,e. the probability of contracting an infectious parasite and to sustain an
epidemic within the colony increases with colony size and worker longevity. This is augmented,
for instance, when the individual immune defence capacity reaches its physiological maximum. A
colony reaching this point may have several evolutionary options. It could, for example, maintain
a critical colony size and worker lifespan (point B, Figure 1) Alternatively, selection for larger
colony size could continue, which increases efficiency and thus reproductive output. If so,
selection for large colony size could have decreased worker longevity (C’&C in Figure 1), or may
have favoured additional adaptations (e.g. changed behaviours) to compensate for the increased
risk of an epidemic (D in Figure 1). Selection for long-lived workers at the expense of colony size
is unlikely, as reproductive output in social insect colonies should largely depend on the
workforce present at time of reproduction (e.g. workers that rear brood) (Muller & SchmidHempel 1992). Scenario E and E’ (Figure 1) are thus not very realistic (c.f. Figure 1).
(c) A plausible way to sustain colony growth while not having too long-lived workers may be to
increase birth rate and thus worker turnover rate (C’& C in Figure 1). In Chapter 2, I
demonstrated that many short-lived workers might effectively protect against a spreading disease.
Colonies with many short-lived workers may even reach a higher critical colony size; this,
however, depends on the relative epidemic potential of factors colony size and worker longevity
(Chapter 3). Above the critical density however (C’, Figure 1), further anti-parasite behaviour
may have evolved due to a possible cost associated with increased worker production (low
individual resistance, Chapter 3).
(d) Strong selection for increased colony size and thus long-lived workers (O'Donnell & Jeanne
1992; Carey 2001) may have favoured additional anti-parasite adaptations to overcome the
cumulative effects epidemic effects of density and worker lifespan (D, Fig 1.) These behavioural
adaptations may include:
79
(i) Changes in the mating system (polygyny, polyandry) that increase genetic diversity and
thus decrease the potential for parasite spread (Shykoff & Schmid-Hempel 1991; Liersch &
Schmid-Hempel 1998; Baer & Schmid-Hempel 1999; Tarpy 2003; Hughes & Boomsma
2004; Tarpy & Seeley 2006). Genetic heterogeneity may also have allowed for more
elaborate caste structure (colony organization) in these colonies. There is indeed evidence
hat task affiliation can have a genetic component (see (Page et al. 1995) and references
therein).
(ii) Behavioural adaptations that reduce the uptake, establishment and spread of the parasite
in the colony (e.g. avoidance behaviour, removal and inactivation of infectious spores, nest
hygiene and immunity transfer by social interactions (Cremer et al. 2007; Walker &
Hughes 2009; Bos et al. 2011; Konrad et al. 2012; Tragust et al. 2013). Age related
behaviour moreover may have introduced heterogeneity in an otherwise homogenous
structure (e.g. as in the conveyer belt model (Schmid-Hempel 1998). Overall, the
transmission of a parasite may become more difficult as colony structure gains in
complexity (Naug & Camazine 2002).
Together this scenario suggests that parasite pressure against colony size and worker longevity
could act as a selective force for the evolution towards complex social structure, eventually
leading to caste formation, with ergonomic benefits following suite rather than be the prime or
only source of selection. A comparative study on colony size worker longevity and the degree of
complexity among social insect species would allow further assessing the applicability of this idea.
However, while this is an idealized sketch and not all lineages may have followed this path, this
scenario integrates the research I did in this thesis and recalls the importance of parasitism on
colony structure, patterns and processes that are worthwhile for further investigation.
80
References
Baer, B. & Schmid-Hempel, P. 1999. Experimental variation in polyandry affects parasite loads and
fitness in a bumble-bee. Nature, 397, 151–154.
Boomsma, J. J. 2009. Lifetime monogamy and the evolution of eusociality. Philosophical Transactions of
the Royal Society B: Biological Sciences, 364, 3191–3207.
Boomsma, J. J., Schmid-Hempel, P. & Hughes, W. O. H. 2005. Life Histories and Parasite Pressure
Across the Major Groups of Social Insects. In: Insect evolutionary ecology (Fellowes MDE, Holloway
G J, Rolff J. Eds) CABI publishing, Wallingford, UK, 211, 139.
Bos, N., Lefevre, T., Jensen, A. B. & D’Ettorre, P. 2011. Sick ants become unsociable. Journal Of
Evolutionary Biology, 25, 342–351.
Bourke. 1999. Colony size, social complexity and reproductive conflict in social insects. Journal Of
Evolutionary Biology, 12, 245–257.
Brown, C. R. & Brown, M. B. 1986. Ectoparasitism as a Cost of Coloniality in Cliff Swallows (Hirundo
Pyrrhonota). Ecology, 67, 1206.
Carey, J. R. 2001. Demographic mechanisms for the evolution of long life in social insects. Experimental
Gerontology, 36, 713–722.
Cremer, S., Armitage, S. A. O. & Schmid-Hempel, P. 2007. Social immunity. Current Biology, 17,
R693–R702.
Doums, C., Moret, Y., Benelli, E. & Schmid-Hempel, P. 2002. Senescence of immune defence in
Bombus workers. Ecological Entomology, 27, 138–144.
Hieber, C. S. & Uetz, G. W. 1990. Colony size and parasitoid load in two species of colonial Metepeira
spiders from Mexico (Araneae: Araneidae). Oecologia, 82, 145–150.
Hughes, W. O. H. & Boomsma, J. J. 2004. Genetic diversity and disease resistance in leaf-cutting ant
societies. Evolution, 58, 1251–1260.
Hughes, W. O. H., Oldroyd, B. P., Beekman, M. & Ratnieks, F. L. W. 2008. Ancestral monogamy
shows kin selection is key to the evolution of eusociality. Science, 320, 1213–1216.
Keller, L. 1998. Queen lifespan and colony characteristics in ants and termites. Insectes Sociaux, 45, 235–
246.
Keller, L. & Genoud, M. 1997. Extraordinary lifespans in ants: a test of evolutionary theories of ageing.
Nature, 389, 958–960.
Konrad, M., Vyleta, M. L., Theis, F. J., Stock, M., Tragust, S., Klatt, M., Drescher, V., Marr, C.,
Ugelvig, L. V. & Cremer, S. 2012. Social transfer of pathogenic fungus promotes active
immunisation in ant colonies. PLoS Biology, 10, e1001300.
Liersch, S. & Schmid-Hempel, P. 1998. Genetic variation within social insect colonies reduces parasite
load. Proceedings Of The Royal Society B-Biological Sciences, 265, 221–225.
Moret, Y. & Schmid-Hempel, P. 2009. Immune responses of bumblebee workers as a function of
individual and colony age: senescence versus plastic adjustment of the immune function. Oikos, 118,
371–378.
Muller, C. B. & Schmid-Hempel, P. 1992. Correlates of reproductive success among field colonies of
Bombus lucorum: the importance of growth and parasites. Ecological Entomology, 17, 343–353.
81
Naug, D. & Camazine, S. 2002. The Role of Colony Organization on Pathogen Transmission in Social
Insects. Journal of Theoretical Biology, 215, 427–439.
O'Donnell, S. & Jeanne, R. L. 1992. The effects of colony characteristics on life span and foraging
behavior of individual wasps (Polybia occidentalis, Hymenoptera: Vespidae). Insectes Sociaux, 39,
73–80.
Oster, G. F. & Wilson, E. O. 1978. Caste and Ecology in the Social Insects. Princeton University Press,
Princeton.
Otterstatter, M. C. & Thomson, J. D. 2007. Contact networks and transmission of an intestinal pathogen
in bumble bee (Bombus impatiens) colonies. Oecologia, 154, 411–421.
Page, R. E., Jr, Robinson, G. E., Fondrk, M. K. & Nasr, M. E. 1995. Effects of worker genotypic
diversity on honey bee colony development and behavior (Apis mellifera L.). Behavioral Ecology and
Sociobiology, 36, 387–396.
Pie, M. R., Rosengaus, R. B. & Traniello, J. 2004. Nest architecture, activity pattern, worker density and
the dynamics of disease transmission in social insects. Journal of Theoretical Biology, 226, 45–51.
Rodriguez-Serrano, E., Inostroza-Michael, O., Avaria-Llautureo, J. & Hernandez, C. E. 2012. Colony
Size Evolution and the Origin of Eusociality in Corbiculate Bees (Hymenoptera: Apinae). PLOS ONE,
7, e40838.
Ruiz-Gonzalez, M. X., Moret, Y. & Brown, M. J. F. 2009. Rapid induction of immune density-dependent
prophylaxis in adult social insects. Biology Letters, 5, 781–783.
Schmid-Hempel, P. 1998. Parasites in Social Insects. Princeton University Press.
Shykoff, J. A. & Schmid-Hempel, P. 1991. Parasites and the Advantage of Genetic Variability within
Social Insect Colonies. Proceedings Of The Royal Society B-Biological Sciences, 243, 55–58.
Stearns, S. C. 1992. The Evolution of Life Histories. Oxford: OUP Oxford.
Tarpy, D. R. & Seeley, T. D. 2006. Lower disease infections in honeybee (Apis mellifera) colonies headed
by polyandrous vs monandrous queens. Naturwissenschaften, 93, 195–199.
Tarpy, D. R. D. 2003. Genetic diversity within honeybee colonies prevents severe infections and promotes
colony growth. Proceedings. Biological sciences / The Royal Society, 270, 99–103.
Tragust, S., Mitteregger, B., Barone, V., Konrad, M., Ugelvig, L. V. & Cremer, S. 2013. Ants Disinfect
Fungus-Exposed Brood by Oral Uptake and Spread of Their Poison. Current Biology, 23, 76–82.
Walker, T. N. & Hughes, W. O. H. 2009. Adaptive social immunity in leaf-cutting ants. Biology Letters, 5,
446–448.
Williams, G. C. 1966. Pleiotropy, Natural-Selection, and the Evolution of Senescence. Evolution,
11, 398–411.
82