The influence of host social system on hostparasite evolutionary dynamics
Dissertation submitted for the degree of
Doctor of Natural Sciences
Presented by
Antoon Jacobus (Jaap) van Schaik
at the
Faculty
of Sciences
Department of Biology
Date of the oral examination: October 5, 2016
First supervisor: Martin Wikelski
Second supervisor: Gerald Kerth
Konstanzer Online-Publikations-System (KOPS)
URL: http://nbn-resolving.de/urn:nbn:de:bsz:352-0-370734
Towards an understanding of host-parasite evolutionary
dynamics in a social world using bats and ectoparasites
as a model system.
Supported by funding from:
Contents
Summary
1
Zusammenfassung
3
Chapter 1: General Introduction
Host-Parasite dynamics
Sociality and parasites
Simplifying the complexity: bat-ectoparasite model system
Research aims
Dissertation outline
8
12
15
19
24
26
Chapter 2: Host social organization and mating system shape parasite
transmission opportunities in three European bat species
28
Abstract
30
Introduction
31
Methods
36
Results
40
Discussion
46
Supplementary materials
52
Chapter 3: The effect of host social system on parasite population
genetic structure: comparative population genetics of two
ectoparasitic mites and their bat hosts
56
Abstract
58
Introduction
59
Methods
64
Results
71
Discussion
82
Supplementary materials
87
Chapter 4: Host and parasite life history interplay to yield divergent
population genetic structures in two ectoparasites living on the
same bat species
92
Abstract
94
Introduction
95
Methods
100
Results
107
Discussion
113
Supplementary materials
118
Chapter 5: Synthesis and general discussion
Host sociality and parasite infection dynamics
Host sociality and parasite population genetic structure
A brief aside on the evolution of virulence and parasite
induced restrictions to host sociality
Overall impact of host social system on host-parasite
evolutionary dynamics
Future directions
122
123
125
List of publications and selected abstracts
135
Bibliography
141
Author contributions
160
Acknowledgements
161
Curriculum vitae
163
128
131
132
Summary
Parasites are ubiquitous in nature, and increased parasite pressure is one of
the fundamental costs of sociality. In addition, sociality also affects hostparasite dynamics directly and indirectly at both ecological and evolutionary
timescales. At the ecological level, host social contact shapes parasite
infection intensity and transmission opportunities, while behavioural and
social avoidance mechanisms allow hosts to reduce overall parasite
pressure. On an evolutionary scale, the influence that host social system has
on parasite genetic structure and the reciprocal selection pressure imposed
by each species on the other can strongly influence the evolutionary outcome
of host-parasite interactions. Although the role of host social system in
shaping parasite infection dynamics is widely acknowledged, its additional
role in shaping parasite genetic structure, and thereby influencing the
evolutionary dynamics between the two interacting species, is comparatively
understudied.
In this dissertation, I use a multi-host multi-parasite comparative
framework to explore the effects of host social system, and its interplay with
parasite life history, on parasite infection dynamics and population genetic
structure. Host-parasite dynamics are shaped by myriad factors including the
abiotic and biotic environment, host and parasite life history and host social
system. This complexity can be considerably reduced when studying
permanent parasites, which do not leave their hosts, as the direct interaction
between the parasite and the environment is limited. Furthermore, by
investigating permanent parasites of closely related hosts with comparable
life histories, it becomes possible to directly assess the effect and interaction
between facets of host social system and specific parasite life history traits.
European bats of the genus Myotis, and two of their common,
permanent ectoparasites (wing-mites of the genus Spinturnix, flies of the
genus Basilia), offer an ideal model to study this interaction. All host species
share a similar life history, but differ in numerous aspects of their social
system, including female colony size, male sociality, mating system and the
degree of aggregation in winter. Likewise, while both permanent parasites of
their bat hosts, the parasites differ in several key life history traits, most
notably their reproductive rate and generation time. By using these species in
a comparative framework, these differences in host social system and
parasite life history can be exploited to attempt to elucidate the effect of
individual social and life history traits. Specifically, I 1) characterize the effect
that differences in host social system has on mite infection and transmission
1
dynamics, 2) compare the genetic structure of the mites of two bat species
to explore the effects of host social system traits on parasite genetic
structure, and 3) compare the genetic structure of a mite and fly species both
infecting a single host species, to explore the interaction between parasite
life history traits and host social system.
Regarding parasite infection and transmission dynamics (chapter 2), I
found substantial differences in mite infection intensity across three bat
hosts, which correlated closely with the degree of sociality (aggregation size)
of both host sexes. Additionally, the transmission rate of the different mite
species was strongly correlated to the degree of contact between the sexes
afforded by their differing mating systems. These differences at the
ecological level also had readily visible effects on mite population genetic
structure (chapter 3). In a comparison of population genetic structure
between Spinturnix bechsteini and S. myoti, I found large differences in the
level of genetic differentiation between, and degree of genetic turnover
within, individual mite populations. These differences again closely correlated
to those predicted based on host aggregation size and mating system.
Finally, in a comparison of mite and fly population genetic structure on a
single host (chapter 4), I showed that parasite reproductive rate and
generation time interact with seasonal changes in host social organization to
yield highly divergent genetic structures. In both investigations of parasite
population genetic structure, the differences observed between parasite
species were much larger than expected, and have significant implications
for the evolutionary potential and the evolutionary trajectory of the hostparasite interaction. These differences are additionally remarkable given the
narrow range of host social systems and parasite life histories that were
investigated, emphasizing the effect that even subtle differences in facets of
either host or parasite can have.
Taken together, these results provide a compelling example of the
influence of host social system in shaping host-parasite dynamics at both
ecological and evolutionary timescales. In doing so, they contribute to our
general understanding of the fundamental factors underlying the structure
and function of animal sociality. Notably, understanding host-parasite ecoevolutionary dynamics additionally has direct conservation applications as it
may aid in combatting emerging infectious diseases, as well as help
understand the effects of the numerous ways in which anthropogenic change
alters the social system and population dynamics of host animals.
2
Zusammenfassung
Die Allgegenwärtigkeit von Parasiten in jeder Form des sozialen
Zusammenlebens stellt einen der grundlegendsten Nachteile sozialer
Systeme dar. Enger Kontakt führt zwangsläufig auch zu einer größeren
Parasitenbelastung und somit zu höheren Kosten für die Wirtsart. Zusätzlich
beeinflusst soziales Verhalten die Ökologie und die Evolution der
Interaktionsdynamik zwischen Wirt und Parasit sowohl direkt als auch
indirekt. Auf ökologischer Ebene stellen Sozialkontakte eine Möglichkeit der
Übertragung für nicht mobile Parasiten dar, der die Wirtsart mit Anpassung
ihres Verhaltens sowie dem Vermeiden bestimmter Kontakte entgegenwirken
kann. Im evolutionären Maßstab gesehen können sowohl der Einfluss des
Sozialverhaltens der Wirtsart auf die genetische Struktur des Parasiten als
auch der gegenseitige Selektionsdruck zwischen Wirt und Parasit einen
enormen Einfluss auf das entwicklungsgeschichtliche Ergebnis von WirtParasit Interaktion haben. Obwohl die Rolle unterschiedlicher Sozialsysteme
von Wirtsorganismen als entscheidend für die Infektionsdynamik einer
zugehörigen Parasitenart anerkannt sind, wurde der Einfluss verschiedener
Sozialsysteme auf die genetische Beschaffenheit der entsprechenden
Parasiten-Populationen und die damit einhergehende Einwirkung auf die
evolutionäre Dynamik zwischen Wirt und Parasit jedoch bislang größtenteils
vernachlässigt.
In dieser Dissertation analysiere ich den Einfluss sozialer Wirtssysteme
auf den Lebenszyklus, die Infektionsdynamik und die genetische Struktur der
entsprechenden Parasitenarten mit Hilfe eines mehrere Wirts- sowie
Parasitenarten umfassenden Vergleichsrahmen. Die Interaktionsdynamik
zwischen Wirt und Parasit wird von unzähligen Faktoren beeinflusst, so zum
Beispiel von abiotischen und biotischen Umweltfaktoren, vom artspezifischen
Lebenszyklus von Wirt und Parasit, sowie vom Sozialsystem des Wirts.
Parasitenarten, die dauerhaft auf ihrer Wirtsart verbleiben, bieten sich daher
zum Studium solcher Systeme besonders an, denn der unmittelbare Kontakt
zwischen Parasit und Umwelt ist stark eingeschränkt und somit gestaltet sich
auch die Bewertung verschiedener Faktoren weniger kompliziert. Ein
Vergleich von verschiedener stationärer Parasiten welche eng verwandten
Wirtsarten mit ähnlichen Lebenszyklen befallen, erleichtert eine Analyse solch
komplexer Systeme zusätzlich. Die möglichen Rückschlüsse auf den Einfluss
verschiedenster Faktoren wie dem Sozialsystem des Wirts oder dem
Lebenszyklus des Parasiten kann wichtige Einblicke in die Grundlagen des
“Wettrüstens” zwischen Parasit und Wirt gewährt.
3
Die verschiedenen in Europa heimischen Fledermausarten der Gattung
Myotis sowie zwei ihrer stationären Ektoparasiten (Milben der Gattung
Spinturnix sowie Fledermausfliegen der Gattung Basilia) stellen ein ideales
Modellsystem für das Studium solcher Wechselwirkungen dar. Die
Fledermausarten dieser Gattung weisen einen ähnlichen Lebenszyklus auf,
aber unterscheiden sich bezüglich Koloniedichte der Weibchen,
Sozialverhalten der Männchen, Paarungssystem und Überwinterungsstrategie. Gleichzeitig unterscheiden sich die beiden im Fokus dieser Studie
stehenden Parasitengattungen maßgeblich in verschiedenen Aspekten ihres
Lebenszyklus und vor allem in der Reproduktionsrate sowie in der
Generationsdauer. Somit erlaubt dieses System eine Gegenüberstellung der
Sozialsysteme der Wirtsarten sowie der Auswirkungen unterschiedlicher
Lebenszyklen der Parasitenarten. Insbesondere drei Aspekte fanden
besondere Beachtung in der vorliegenden Arbeit: 1) der Einfluss der
unterschiedlichen Sozialsysteme der verschiedenen Wirtsarten auf
Infektions- und Übertragunsdynamik der parasitären Milben, 2) der Einfluss
der Unterschiede im Sozialverhalten der zwei Wirtsarten auf die genetische
Beschaffenheit ihrer Milbenpopulationen, 3) der direkter Vergleich zweier
unterschiedlicher Parasitenarten auf dem selben Wirt und die Auswirkung
ihrer verschiedenen Lebenszyklen auf ihre genetischen Populationsstrukturen.
In einen Vergleich der Belastung durch Parasiten sowie deren
Verbreitungsdynamik (Kapitel 2) konnte ich erhebliche zwischenartliche
Unterschiede in der Infektionsrate der drei Wirtsarten zeigen. Diese
korrelierten erheblich mit Unterschieden im Sozialverhalten von Weibchen
und Männchen. Besonders hatten das Maß an sozialen Kontakten zwischen
den Geschlechtern sowie das Paarungsverhalten einen erheblichen Einfluss
auf den Parasitenbefall der gesamten Population. Weiterhin zeigten sich,
abgesehen von solchen Einflüssen auf ökologischer Ebene, abhängig vom
Sozialverhalten der Wirtsart auch gravierende Unterschiede in der
genetischen Struktur verschiedener Milbenpopulationen auf der gleichen
Wirtsart (Kapitel 3). Bei einem Vergleich der genetischen Struktur der beiden
Parasitenarten Spinturnix bechsteini und S. myoti zeigten sich große
Unterschiede
bei
der
genetischen
Differenzierung
zwischen
unterschiedlichen Populationen sowie bei der Mutationsrate in einzelnen
Populationen. Diese Unterschiede korrelieren wiederum eng mit den auf
Grund von Koloniedichte und Paarungssystem der Wirtsart erwarteten
Auswirkungen. Mit einem Vergleich der populationsgenetischen Struktur von
Milben und Fledermausfliegen auf einer Wirtsart (Kapitel 4) konnte ich
schlussendlich zeigen, dass Reproduktionsrate sowie Generationsdauer der
Parasitenart stark mit jahreszeitlichen Veränderungen im Sozialverhalten der
Wirtsart zusammenhängen, um möglichst hohe genetische Diversität zu
4
gewährleisten. Die Analyse beider populationsgenetischer Strukturen zeigte
einen Unterschied zwischen den Arten, der weitaus höher war, als erwartet.
Dies ist ein wichtiger Aspekt in Hinsicht auf Evolutionäres Potential sowie
evolutionäre Zielsetzung von Wirt-Parasiten-Interaktionen. Gleichzeitig sind in
Anbetracht der begrenzten Variabilität sowohl der untersuchten WirtsSozialsysteme als auch der Lebenszyklen der Parasitenarten die
aufgedeckten Unterschiede bemerkenswert. Dementsprechend scheinen
schon geringste Unterschiede im Wirts- oder Parasitensystem einflussreich
zu sein.
Folglich können die Ergebnisse dieser Dissertation zu einem besseren
allgemeinen Verständnis fundamentaler Faktoren beitragen, welche Struktur
und Funktion sozialer Systeme beeinflussen. Besonders im Zusammenhang
mit der Bekämpfung neu auftretender Infektionskrankheiten kann das
Verständnis von öko-evolutionärer Wirt-Parasit Dynamik von großem Nutzen
sein und erläutern, wie anthropogene Veränderungen der Umwelt
Sozialsysteme sowie Populationsdynamik der Wirtsarten beeinflussen.
5
6
7
Chapter 1
General Introduction
Parasites are an enigma. To some people they are an unpleasant but
unavoidable fact of life. To others they are, like Victorian ankles, an
embarrassing topic to be avoided in polite conversation. But to a small
subset of human beings parasites are glorious creatures, no less a part of
Darwin’s tangled bank than organisms more acceptable to anthropocentric
idealization.
- Brooks & McLennan 1993
Parasitism is arguably the most common way of life (Price 1977). For hosts,
the threat of parasite infection is ever-present and constitutes a strong
selective force. Indeed, on an evolutionary scale, the reciprocal adaptation
between a host and parasite is thought to be one of the primary generators
of biodiversity (Hamilton 1982, Dybdahl and Storfer 2003). Although the
consensus has progressed considerably from the avirulence hypothesis,
which stated that parasites should always become less virulent as they adapt
and coevolve with a host (Smith 1904), towards a recognition that virulence is
often maintained at some non-zero level as parasite fitness is maximized
(Anderson and May 1982, Ewald 1983, Lenski and May 1994, Alizon et al.
2009, Alizon and Michalakis 2015), it remains remarkable that the vast
majority of host-parasite interactions go largely unnoticed as they do not
cause obvious costs or mass mortality. This observation becomes even more
8
striking when considering that parasite life-cycles are often much faster, and
population sizes much larger, than that of their hosts, thereby theoretically
giving them the advantage in an evolutionary arms race. Understanding the
factors that influence host-parasite evolutionary dynamics may help explain
this observation.
Increased risk of parasite exposure constitutes one of the initial and
most fundamental costs of sociality (Møller et al. 1993, Altizer et al. 2003,
Kappeler et al. 2015). Naturally, sociality also provides many benefits (e.g.
increased foraging efficiency, reduced thermoregulatory costs in dense
aggregations; Krause and Ruxton 2002), which may partially compensate for
the increased cost of parasite infection. Notably, in addition to these overall
benefits, sociality also affords a wealth of behavioral and social adaptations,
which can be used to directly reduce parasite pressure (the behavioral
immune system, cf. Schaller et al. 2015). At the population level, host
sociality will also affect parasite dynamics by affecting transmission among
host social units (i.e. metapopulation dynamics). Combined, these effects, in
combination with host and parasite life history, also shape a parasite’s
population genetic structure (Nadler 1995) and thereby its evolutionary
potential (Gandon and Michalakis 2002). Finally, host sociality may also
provide indirect benefits with regards to the evolutionary trajectory of the
interaction by limiting parasite local adaptation (Gandon et al. 2008), or
selecting for lower virulence (May and Anderson 1990). In sum, host sociality
has a large effect on host-parasite dynamics at both the ecological and
evolutionary level.
Despite widespread recognition and investigation of the direct effects
of sociality on parasite exposure risk and infection dynamics (e.g. Møller et
al. 1993, Clayton and Moore 1997, Nunn and Altizer 2006), empirical
investigations into the role of host social system in shaping parasite
population genetic structure and especially the resulting evolutionary
dynamics are comparatively rare (Poulin 2011). In this dissertation, I use a
comparative multi-host multi-parasite framework to explore how host social
9
system interacts with host and parasite life history to shape parasite
population genetic structure.
Terminology
Before getting started, it is important to define several terms as used in this
dissertation. For example, what precisely constitutes a parasite has been
extensively disputed (e.g. Price 1977, Windsor 1998, Clayton et al. 2015).
While some organisms are obviously parasitic, others (e.g. phytophagous
insect, brood parasitic birds, blood-sucking insects) are debatable. In other
cases, terms have been used synonymously by some (social organization /
social structure), while denoting different characteristics by others (compare:
Whitehead and Dufault 1999, Kappeler and van Schaik 2002). A complete list
of terms as they are used in this dissertation is provided in the glossary
below.
Glossary
Defense any strategy employed by a host to combat parasite infection. There are
three primary types of defense: physical barriers (e.g. skin, mucus, gut lining),
immune defense (e.g. inflammation, adaptive immune response) and behavioral
defense (e.g. grooming, use of latrines). These defense types can be employed in
three main defensive strategies:
Avoidance defense strategy that reduces of the risk of exposure to a parasite 1
Resistance defense strategy that reduces the parasite burden of an infected
host 1
Tolerance defense strategy that reduces the negative impact of infection on host
fitness. Unlike resistance, tolerance does not directly affect pathogen burden 1
Evolutionary potential the ability of either host or parasite to incorporate genotypes
able to overcome the weaponry put forward by the opposing species 2
Local adaptation the increased fitness of a population in its local habitat relative
to its fitness in a remote habitat. 2 Within the context of host-parasite interactions:
the increased fitness of a local population of a host or parasite with its sympatric
parasite or host population relative to its fitness with host or parasite populations
elsewhere
10
Infection the successful habitation of a host by a parasite
Intensity the number of individuals of a particular parasite species on a single
infected host 3
Prevalence the number of hosts infected with one or more individuals of a
particular parasite species (or taxonomic group) divided by the number of hosts
examined for that parasite species 3
Virulence the capacity of a parasite to inflict damage and reduction in host
fitness 4
Metapopulation spatial substructuring of a population into local breeding
populations (social systems) inhabiting discrete habitat patches where migration
among populations to at least some other patches is possible. Notably, local
dynamics within patches generally occurs at a much faster pace than overall
metapopulation dynamics (e.g. local extinction / recolonization) 5
Parasite an organism living in or on another organism: the host – feeding on it,
showing some degree of structural adaptation to it, and causing it some harm 6
While this definition is rather narrow, the themes and general principles discussed
here are also generally applicable to a wide range of parasite-like species,
including parasitoids (insects with free-living adult stages that lay eggs in hosts
which ultimately sterilize or kill the host) and pathogens (microorganisms that
cause disease in the host).
Permanent a parasite that passes all stages of its life cycle on the body of the
host 7
Mobile / non-mobile parasites that are able to disperse independently are
considered to be mobile parasites, whereas non-mobile parasites primary
depend on host movement for transmission 4
Horizontal transmission the transmission of parasites between hosts of the
same population and generation (i.e. not to own offspring: vertical
transmission)4. In cases where hosts form social units with limited interaction
between conspecifics outside of these groups, horizontal transmission may be
restricted to include only those events in which parasites are transmitted
between social units.
Social system (synonymous to social unit) the set of conspecific animals that
interact regularly and more so with each other than with members of other such
societies. Here I will use social system to refer to the general concept, and social
unit or colony to refer to individual societies. Can be described using three
characteristics: social organization, social structure and mating system 8
Social Organization the size, sexual composition and spatiotemporal cohesion
of a society 8
Social Structure the pattern of social interactions and the resulting relationships
among the members of a society 8
Mating system composed of a social and a genetic component. The social
mating system describes the subset of social interactions between mating
couples, whereas the genetic mating system describes the reproductive
consequences of mating interactions 8
11
Sensu
1
(Medzhitov et al. 2012)
2
(Gandon and Michalakis 2002)
3
(Bush et al. 1997)
4
(Schmid-Hempel 2011)
5
(Hanski and Gilpin 1997)
6
(Poulin 2011)
7
(Clayton and Moore 1997)
8
(Kappeler and van Schaik 2002)
Host-Parasite dynamics
The epidemiological triangle, classically applied in veterinary medicine, states
that three main components determine the persistence of an infectious
species or agent within a host population: host, parasite/pathogen and
environment
(McNew
1960,
Wobeser
2007).
This
framework
can
subsequently be extended to include the (co)evolutionary dynamics between
the interacting species (Vander Wal et al. 2014a). Despite this apparent
simplicity, myriad factors influence host-parasite infection and evolutionary
dynamics. A summary of these effects is given in Box 1.
Within this framework, the interaction between host and parasite can
be subdivided into three levels: First and most obviously, the immediate
infection dynamics at the level of the individual or populations. Second, the
evolutionary potential of both species as shaped by their population genetic
structure. Third, the ultimate evolution of adaptations (i.e. specialization),
notably including the directionality (i.e. evolution of virulence) of the resulting
co-evolutionary trajectory. Finally, in cases where there is spatiotemporal
variation in these three levels across discrete host-parasite populations, a
fourth emergent property will arise: behavioral and/or genetic differentiation
across populations (i.e. geographic mosaic of adaptation), which may impose
costs to dispersal and potentially lead to speciation. Although all properties
of the host, parasite and environment will influence these dynamics, I will
restrict the following discussion to the aspects directly influencing, and
influenced by, host social system.
12
Box 1: Host-parasite interaction and evolutionary dynamics
 At an individual level,
Likewise,
the
biotic
environment
adds
a
multitude of other species
inter-actions that must be
dealt with concurrently,
thereby often introducing
trade-offs and limitations
to adaptation as each
individual interaction must
be seen within the context
of all other interactions in
the local environment.
Individual
2
Host
4
Defense/Infection
3
3
2
Parasite
Selection pressure
Local scale evolutionary dynamics
The abiotic environment,
especially
climate
and
seasonality, plays a major
role
in
shaping
the
(temporal
changes
in)
resource
and
habitat
availability for species.
Environment
1
6
Pop genetic
structure
5
Pop genetic
structure
5
7
Adaptation
8
Broad scale
host-parasite
dynamics,
like all species interactions, are shaped by
three main factors: the two
interacting species and
their environment.
9
>
=
<
Adaptation
Local adaptation
Geographic mosaic
Diversity / Speciation
‚ Many characteristics of both host and parasite influence whether a parasite is able to
infect a given host. Some characteristics are largely static, such as genotype, general
life history and morphology, whereas others are more variable including health,
behavior, and notably the social interactions of the host.
ƒ ,„ Based on (the current state of) all of the aforementioned characteristics, hosts
can mount three basic types of defense: physical, immune and behavioral, which can
be employed in three primary defense strategies: avoidance, resistance or tolerance.
Correspondingly, a parasite’s ability to infect a host, and subsequently its reproductive
rate, transmission potential and virulence, will depend on whether its physical, immune
and behavioral traits can overcome the defense mounted by the host.
… Importantly, by shaping local population dynamics and transmission potential, the
life histories of both species as well as the social system of the host also shape the
population genetic structure (effective population size, genetic drift, gene flow) of the
parasite. Conversely, parasite life history and infection dynamics may also play a role in
shaping host genetic structure in the event that host mortality and reproduction are
substantially affected, but in most cases its effect will be negligible (dashed line).
13
† Besides shaping population genetic structure, parasite infection dynamics will also
determine the selection pressure exerted on both species by the opposing party. This
will often be much stronger for the parasites as the fitness of a parasite is entirely
dependent on successful infection, whereas host fitness is often influenced to a lesser
extent.
‡ ,ˆ The evolutionary potential of a local population is determined by the number of
new genotypes entering the population and is thus primary determined by their effective
population size and gene flow. Subsequently, evolutionary potential, in combination
with the strength of selection pressure, will determine the degree to which host and
parasite populations adapt to the local population of the other species. Finally,
depending on the relative selection pressure and evolutionary potential of the two
species, whichever species is adapting at a faster rate will be able to locally adapt to
the local population of the other. Vitally, the local adaptation of either host or parasite
will not necessarily result in increased costs for the other species, as increased
tolerance (host) and decreased virulence (parasite) can be adaptive under certain
conditions.
‰ The local adaptation of populations at a microgeographic scale will lead to heterogeneity at a macrogeographic scale. Such a geographic mosaic of coadapted
populations may decrease the fitness of individuals migrating between populations,
thereby potentially leading to increased differentiation and ultimately speciation. In this
context it is important to note that a significant proportion of differentiation at a
macrogeographic scale is not the result of adaptation, as isolation by distance and
genetic drift will also lead to increased spatial substructuring and heterogeneity.
References and further reading:

Abiotic: (Altizer et al. 2006, Real and Biek 2007, Lafferty 2009, Wolinska and King
2009, Nunn et al. 2014)
 Biotic: (Rigaud et al. 2010, Alizon et al. 2013, Hafer and Milinski 2016, van Velzen
et al. 2016)
‚ (Hart 1990, Côté and Poulin 1995, Loehle 1995, Poulin et al. 2011, Lion and
Gandon 2015, Boulinier et al. 2016)
ƒ (Parker et al. 2011, Cressler et al. 2015, Sears et al. 2015)
„ (Miller et al. 2005, Alizon et al. 2009, Raberg et al. 2009, Medzhitov et al. 2012,
Alizon and Michalakis 2015)
… (Nadler 1995, Sorci et al. 1997, Criscione et al. 2005, Barrett et al. 2008, MazéGuilmo et al. 2016)
† (Gandon et al. 1998, Barrett et al. 2008, Gandon and Nuismer 2009b)
‡ ,ˆ (Gandon and Michalakis 2002, Greischar and Koskella 2007, Hoeksema and
Forde 2008, Nuismer and Gandon 2008, Nuismer et al. 2010, Yeaman 2015)
‰ (Thompson 1994, Criscione 2008, Hoberg and Brooks 2008, Hoberg and Brooks
2015, Penczykowski et al. 2016)
14
Sociality and parasites
The following section provides an overview of how host social system, and
interactions between social system and environment as well as parasite life
history, influence host-parasite dynamics at each of the three levels of
interaction (i.e. immediate infection dynamics of the individual or local
population, genetic structure and evolutionary potential, directionality of the
evolutionary trajectory).
Sociality and infection dynamics
By far the most research has gone into how host social system affects
parasite infection and transmission dynamics. These effects can be roughly
subdivided into three categories.
First, sociality provides several opportunities for social behaviors
directly geared towards parasite defense (Altizer et al. 2003). Some, such as
social avoidance, mate choice and mate fidelity, act as avoidance strategies
that reduce the risk of exposure to parasites (Loehle 1995). Others, such as
allogrooming or caring for sick conspecifics, function by reducing the severity
of infection or parasite-induced mortality (Hart 1990).
Second, by shaping host spatiotemporal organization and contact
rates, social system also affects the distribution and transmission dynamics
of their parasites (Ezenwa 2004). The formation of discrete social units (i.e.
metapopulation dynamics; Hanski and Gilpin 1997) reduces overall contact
among social units and therefore also restricts parasite horizontal
transmission opportunities in proportion to the inter-social unit contact rate
(Nunn and Altizer 2006). Within social units, however, transmission is
facilitated and will increase as a function of intra-social unit contact rates
(Nunn et al. 2015). Indeed, the size and density of social aggregations have
been shown to correlate positively with parasite infection intensity across a
wide range of systems (Côté and Poulin 1995, Patterson and Ruckstuhl
2013). Similarly, the structure and strength of associations within the social
network have been shown to shape the transmission dynamics of contact
15
transmitted parasites in a wide variety of species (Godfrey et al. 2009, Zohdy
et al. 2012, Reynolds et al. 2015, Rimbach et al. 2015, Tung et al. 2015,
Springer et al. 2016, VanderWaal et al. 2016). Notably, these relationships
often do not hold for mobile parasites or for non-mobile parasites that can
persist off of the host or make use of vectors. In fact, the per capita infection
intensity of mobile parasites may often be lower in aggregations (i.e. dilution
effect; Rubenstein and Hohmann 1989, Krebs et al. 2014, Lehtonen and
Jaatinen 2016).
Third, a host’s social system influences, and is influenced by, many
other characteristics relevant to host-parasite dynamics. For example, host
social system is strongly influenced by climate and seasonality, often leading
to large variations in social system, and thus parasite transmission
opportunities, throughout the year (Altizer et al. 2006). How these changes
affect parasite dynamics will additionally depend on the life cycle and
persistence ability of the parasite in the “sub-optimal” seasons (PerezRodriguez et al. 2015). Social system also influences many aspects of host
health and susceptibility (i.e. reduction in immunological defense) to parasite
infection. Some of these influences are well established, such as the effect of
dominance hierarchy on health and hormone levels (MacIntosh et al. 2012),
whereas others, such as the effects of social isolation in a social species, are
only recently being recognized (Hawkley and Capitanio 2015).
Sociality and parasite population genetic structure
The wide array of effects of host social system on parasite infection
dynamics is similarly reflected in their influence on parasite population
genetic structure (Nadler 1995). Empirical studies of parasite population
genetic structure in relation to host sociality are comparatively rare, but have
become far more common in recent years (see Table S1 of Mazé-Guilmo et
al. 2016). Critically, almost none investigate the temporal turnover / stability
of local parasite populations, limiting their ability to quantify genetic drift and
gene flow of local parasite populations. Nevertheless, several general
conclusions can be drawn.
16
As previously described in relation to infection intensity, increases in
the size and density of host social aggregations will generally lead to a larger
local parasite population, and thus often leads to an increased effective
population size (Criscione and Blouin 2005). A large effective population size
reduces the influence of genetic drift, and generally also results in an
increased amount of standing genetic diversity in the local gene pool.
Likewise, as previously observed above with regard to parasite transmission
dynamics, host social system will strongly influence parasite gene flow
(Nadler 1995). Specifically, inter-social unit contact rates and dispersal will
shape parasite gene flow across the metapopulation (Mazé-Guilmo et al.
2016), whilst parasite gene flow within social units will be unrestricted or
follow patterns of intra-unit contact (Tung et al. 2015). In this context it is
important to note that these two facets are not necessarily linked, and thus
high levels of gene flow do not imply low levels of genetic drift and vice versa
(e.g. a local parasite population may have a very large effective population
size, but be almost entirely isolated from other populations). This is crucial as
the evolutionary potential of a local population is primarily a function of its
standing genetic diversity, gene flow and genetic drift (and to a lesser extent
mutation rate). Specifically, high levels of genetic diversity and gene flow and
relatively lower levels of genetic drift are suggested to result in maximal
evolutionary potential (Gandon and Michalakis 2002). Finally, as the ability to
locally adapt to the respective population of the other species is a function of
evolutionary potential and selection pressure (Gandon and Nuismer 2009b),
the influence of host social system on the evolutionary potential of its
parasites plays a vital role in shaping host-parasite evolutionary dynamics.
Sociality and virulence evolution
Perhaps the most poorly understood and underappreciated effect of host
sociality on host-parasite dynamics is its effect on the evolutionary trajectory
of virulence evolution (i.e. what happens if a parasite is able to locally adapt).
While perhaps counterintuitive, local adaptation of the parasite may
ultimately lead to lower overall costs for the host as optimal fitness of the
17
parasite is linked to both its virulence and its transmission (the transmissionvirulence trade-off hypothesis; Alizon et al. 2009, Alizon and Michalakis
2015). Theoretical models and experiments involving lower order hostparasite interactions (e.g. bacteria-phage) suggest that virulence is principally
influenced by two traits: host defense strategy and parasite transmission
dynamics. Specifically, host resistance (almost exclusively determined by
immune defense) is expected to lead to higher parasite virulence, while host
avoidance and tolerance defense strategies (including social and behavioral
adaptations) lead to the evolution of lower parasite virulence. Moreover,
parasite virulence is expected to be higher when parasite transmission is
primarily horizontal and lower when it is predominantly vertical (Boots et al.
2004, Best et al. 2011), as horizontal transmission allows for parasites to
disperse more readily from terminal host individuals. Interestingly, in this
reference frame, the metapopulation structure often created by host sociality
would be more akin to vertical transmission as the parasite is strongly limited
from dispersing horizontally outside of its own social unit (Boots and Sasaki
1999, Bonds et al. 2005). Although empirical studies exploring the evolution
of parasite virulence in social host species are exceedingly rare, and it is also
beyond the scope of the work presented in this dissertation, it is nevertheless
important to keep in mind.
Synthesis and general notes
In summary, sociality not only fundamentally increases the risk of exposure
to parasites, but also can profoundly affect host-parasite infection and
evolutionary dynamics at all three levels. In the context of the two levels
investigated here, several outstanding questions remain.
Despite the wealth of knowledge and general understanding of the
effects of individual facets of host and parasite life history and host social
system, how these components interact to shape host-parasite infection
dynamics in individual systems remains complicated to predict. Moreover,
accurately predicting how these components interact to shape parasite gene
flow and genetic drift, and thus the evolutionary dynamics of the system, is
18
currently unfeasible. To make matters more complex, the sections above
have focused exclusively on effects directly pertaining to host sociality, but
as outlined in Box 1, many other parameters (e.g. the interaction between the
parasite and the environment) will also influence host-parasite dynamics.
To explore these questions requires a system where as many
confounding influences as possible can be eliminated, but with variation in
host social system, thus allowing for the characterization of the effects of
individual parameter changes in host sociality on both host-parasite infection
as well as evolutionary dynamics. Finally, by combining such studies from
across a range of host systems, it may become possible to develop more
comprehensive generalizations regarding the effects of host social system on
host-parasite interactions.
Simplifying the complexity: bat-ectoparasite model system
In this dissertation I characterize parasite infection dynamics and population
genetic structure in a comparative multi-host multi-parasite framework using
four European bat species all belonging to the genus Myotis (Box 2) and their
ectoparasites (Box 3) as a model system. This approach and model system
satisfies the requirements necessary to be an informative system for
exploring the effects of host social system described above, and has several
additional advantages.
The multi-host multi-parasite comparative framework is comprised of
two interrelated components. In the first, I aim to isolate the influence of host
social system by comparing the infection dynamics and population genetic
structure of a single parasite type across several hosts with very similar life
histories, but differing social systems. As both host and parasite life history
are similar across the investigated species pairs, I posit that any differences
in parasite infection dynamics or genetic structure should be primarily the
product of host social system. In the second, I aim to isolate the effects of
parasite life history within the context of a single host social system by
19
comparing the population genetic structure of two non-mobile parasites
living on the same host populations. As they infect the same host species,
they are afforded comparable direct transmission opportunities via host
social interactions. Therefore, any difference in population genetic structure
between the species should be the result of how different aspects of parasite
life history interplay with characteristics of host social system.
Box 2: The four Myotis species investigated
Myotis bechsteinii
©René Janssen
Myotis nattereri
©René Janssen
Myotis daubentonii
©René Janssen
Myotis myotis
©René Janssen
In this study, I examined four European bat species: Bechstein’s bat (Myotis
bechsteinii), Natterer’s bat (Myotis nattereri), Daubenton’s bat (Myotis daubentonii)
and the greater mouse-eared bat (Myotis myotis). All species have a comparable life
history composed of a three phase annual cycle (summer maternity colonies,
autumn mating and winter hibernation). They do, however, differ in several aspects
of their social system as summarized in the table below.
The primary host species investigated in this dissertation, Myotis bechsteinii, has
also been the most extensively studied in terms of behavior and social system
(reviewed in Kerth and van Schaik 2012). In this species, contact between
conspecifics outside of maternity colonies is severely limited, thus restricting
opportunities for parasite horizontal transmission. Additionally, they display several
(social) behaviors that further reduce parasite pressure such as (allo-)grooming
(Kerth et al. 2003a) and frequent roost switching including the potential avoidance of
parasitized roosts (Reckardt and Kerth 2007).
Contact among conspecifics increases in a gradient across the other three bat
species (M. nattereri < M. daubentonii < M. myotis; see table above). Furthermore,
as in M. bechsteinii, grooming is common in all species (Swift 1997, Giorgi et al.
2001), and roost switching is commonly observed in M. nattereri (Swift 1997) and M.
daubentonii (when roosting in tree cavities; Lučan 2006), but absent in M. myotis.
20
Bechstein's
bat
M. bechsteinii
Natterer's
bat
M. nattereri
Daubenton's
bat
M. daubentonii
Greater mouseeared bat
M. myotis
Ref
Social
organization
Summer
Colony size
♀: small (1050)
♀: small (10100), rarely
including some ♂
♀: small-medium (10200), sometimes
including ♂
♀: large (100-2000)
1
♂: generally
solitary
♂: solitary or bachelor
groups (1-50)
♀: fissionfusion and
roost switching
(tree cavity)
♀: fission-fusion
and roost
switching (tree
cavity)
♀,♂: fission-fusion and
roost switching
(tree cavity) or static
(cave/building)
♀: static (cave/
building)
♀,♂: solitary,
occasional
small clusters
(1-10)
♀,♂: solitary or
small clusters
(1-20)
♀,♂: solitary or small
clusters
(1-20)
♀,♂: solitary up to
large clusters
(1-1000)
1, 3
♀: strong
philopatry;
inter-colony
aggression
♀: strong
philopatry; rare
inter-colony
interaction
♀: strong philopatry;
inter-sex spatial
segregation
♀: philopatry; some
inter-colony interaction
outside of maternity
period
4, 5,
6, 7
Promiscuous;
swarming
Promiscuous;
swarming
Promiscuous; swarming
and local mating
Promiscuous;
swarming and
temporary harems
1, 8,
9
♂: solitary
♂: solitary
Spatio-temporal
cohesion
1, 2
♂: roost switching
(tree cavity) or static
(cave/building)
Winter
Aggregation
size
Social structure
Summer
Mating system
Autumn
References: 1: Krapp 2004, 2: Kerth 2008, 3: Wojechowski et al. 2007, 4: Kerth and van Schaik 2012,
5: Rivers et al. 2005 , 6: Angell et al. 2013 , 7: Castella et al. 2001 , 8: Encarnação and Reiners 2002,
9: Zahn and Dippel 1997
Sampling method
Several sampling approaches were used in this dissertation. For comparisons of
parasite population genetic structure, parasites were collected from female
maternity colonies in summer. In M. bechsteinii, female maternity colonies were
“caught” from their roosting sites: an established network of hundreds of bat boxes
across approximately twenty forest patches near Würzburg, Germany. Additionally,
at sites where the colonies are monitored on a daily basis, freshly deposited puparia
of the bat flies could be collected from the walls of the bat boxes. In M. myotis,
maternity colonies in churches across western Switzerland were caught as they
emerged in the evening. For comparisons of parasite transmission dynamics, bats
were caught at three large swarming sites, two in northern Bavaria and one in North
Rhine – Westphalia, in autumn. Naturally, such a sampling effort is only possible
with the help of many collaborators; please refer to the acknowledgements of the
respective chapters as well as the general acknowledgements for a complete
overview.
21
The investigated bat species provide an ideal model system for investigating
the influence of host social system on parasite dynamics, which compares
favorably to other mammalian systems. They have a very slow life history,
living upwards of 20 years, and giving birth to only a single young per year.
Thus, they have a very comparable life history to that of many other larger
mammal species such as ungulates and primates. Unlike these species,
however, the small body size and dense colony formation of bats allows for
comparatively rapid and comprehensive capture of entire host social units.
Furthermore, bats display a wide variety of social systems, with large
differences in group size and mating system (Box 2), thereby allowing for
comparative studies. They are also commonly infected by several
ectoparasite species (see below). Moreover, bats have been shown or
implicated to be reservoirs or vectors of many pathogens and emerging
infectious diseases, providing added direct relevance to understanding how
host sociality affects parasite infection dynamics in this system.
Like their bat hosts, wing mites (Spinturnix) and bat flies (Basilia), two
highly specialized bat ectoparasite taxa, are well suited as model species.
Both are obligate parasites, and thus the influence of environmental factors is
greatly minimized as their environment is essentially reduced to their host
and its immediate environment. Both are comparatively large (1-5 mm)
ectoparasites, which greatly facilitates sampling. They are broadly similar in
terms of life history but differ in several key aspects (Box 3).
22
Box 3: Wing mites (Spinturnix) and bat flies (Basilia)
Spinturnix bechsteinii
© GK & JvS
Basilia nana
adapted from Weiden 1954
Wing-mites and bat flies are both highly specialized obligate ectoparasites that
infect bats. Both are hematophagous, and constitute a low, but non-zero costs for
their hosts (Giorgi et al. 2001, Dick and Patterson 2006). The population dynamics of
both species closely follow the seasonal cycle of their host, generally being lowest
in early spring, growing throughout the summer maternity phase and subsequently
reducing again during the autumn mating season and winter hibernation. Similarly,
while both species are transmitted rapidly within host social units, both species
cannot disperse independently (but see flies below), thereby limiting inter-social unit
transmission.
In this study, I investigate four closely related mite species of the genus Spinturnix
(S. bechsteini, S. myoti, S. andregavinus and S. nattereri). These mites are broadly
host-specific, and live exclusively on the tail and wing membranes of their host.
They are oviviparous, with female mites giving birth to single adult-like
protonymphs, which rapidly develop into adults (Rudnick 1960). They never leave
the host and cannot survive off of the host for more than a few hours (Giorgi et al.
2004).
The bat fly investigated here, Basilia nana (Diptera: Nycteribiidae), is a wingless
ectoparasite that inhabits the pelage of bats (Dick & Patterson 2006, 2007). It
cannot disperse independently from its primary host, and female flies only briefly
leave their host approximately every 5-10 days to deposit a single terminal (3rdinstar) larva on the wall of the roost (Reckardt & Kerth 2006). These larvae
immediately develop into a puparium, and can emerge as fully developed imagoes
in ca. 3 weeks (Dick & Patterson 2007). Developed imagoes wait inside the puparia
and only emerge in the presence of a potential host (Reckardt & Kerth 2006).
Therefore, while not able to freely disperse, they may be transmitted between
colonies without direct host contact.
23
Research aims
In this dissertation I aim to contribute towards understanding the effects of
host social system on host-parasite infection and evolutionary dynamics. The
eventual goal being to understand how individual facets of host sociality
affect, and are restricted by, the interaction between a host and its parasites.
In this context, I address three main questions:
1) How do differences in bat social organization and mating system affect
mite infection and transmission dynamics?
2) How do differences in bat social system affect the population genetic
structure and evolutionary potential of their local mite populations?
3) How do differences in parasite life history traits interact with the same host
social system to affect parasite (mite and fly) population genetic structure
and evolutionary potential?
Predictions
Based on our current understanding of the study species, and host-parasite
interactions in general, several predictions regarding the effect that specific
aspects of host social system will have in this system can be made:
Colony / aggregation size: increased size and density of host aggregations
will increase parasite infection intensity and transmission between colony
members, while solitary individuals will show lower levels of parasite
prevalence and intensity due to a reduced exposure rate (Arneberg et al.
1998, Krasnov et al. 2002, Rifkin et al. 2012, Patterson and Ruckstuhl 2013).
Spatiotemporal cohesion: by reducing aggregation size, fission-fusion
dynamics may reduce parasite infection intensity. However, as the subunits
frequently interact, parasite transmission and prevalence are unlikely to be
24
affected. Roost switching will reduce the infection intensity of non-permanent
parasites that spend part of their life cycle in the roost but are unable to
disperse independently (e.g. bat flies) as hosts may not reuse parasitized
roosts within the infective period of the parasite (Reckardt and Kerth 2006).
Colony structure: female philopatry and active exclusion of non-colony
members will reduce/eliminate parasite transmission via dispersing or
prospecting host individuals (Kerth and van Schaik 2012, Boulinier et al.
2016).
Mating system: promiscuity will increase parasite transmission as multiple
sexual encounters increases the number of encounters with potentially
parasitized conspecifics. Additionally, through successive contact with
females from different maternity colonies, males may act as vectors for mite
transmission between colonies. Mating systems that result in longer contact
rates (e.g. temporary harems) are expected to further increase parasite
transmission (Cross et al. 2009).
25
Dissertation outline
To draw informative conclusions regarding the evolutionary consequences of
variation in individual facets of host social system and parasite life history, we
must first understand how these facets interact to shape current infection
dynamics. Therefore, in chapter 2, I investigate the effect of host social
organization and mating system on mite infection and transmission dynamics
across three bat species (Myotis nattereri, Myotis daubentonii and Myotis
myotis) during the autumn swarming season.
In chapter 3 I quantify the effects of these differences in parasite infection
and transmission dynamics by comparing the resulting population genetic
structure and evolutionary potential in two mite-bat pairs (Myotis bechsteinii /
Spinturnix bechsteini and Myotis myotis / Spinturnix myoti).
Next, in chapter 4 I compare the population genetic structure of two parasite
species (Spinturnix bechsteinii, Basilia nana) both infecting M. bechsteinii. As
both parasites concurrently infect the same host colonies, I posit that any
difference in population genetic structure between the species should be the
result of how different parasite life history traits interact with the same host
social system.
Finally, in chapter 5 I synthesize the findings from the previous chapters and
revisit the aims and research questions outlined above. Specifically, I
summarize the observed interaction between host social system and parasite
life history in shaping parasite infection and genetic structure, and discuss
the broader consequences and possible adaptive use of host social system
in the context of host-parasite (co)evolutionary dynamics.
26
27
Chapter 2
Host social organization and mating system
shape parasite transmission opportunities in
three European bat species
J. van Schaik, G. Kerth
Esperhöhle, Fränkische Schweiz, Germany
28
29
Abstract
For non-mobile parasites living on social hosts, infection dynamics are
strongly influenced by host life history and social system. We explore the
impact of host social systems on parasite population dynamics by comparing
the infection intensity and transmission opportunities of three mite species of
the genus Spinturnix across their three European bat hosts (Myotis
daubentonii, Myotis myotis, Myotis nattereri) during the bats’ autumn mating
season. Mites mainly reproduce in host maternity colonies in summer, but as
these colonies are closed, opportunities for inter-colony transmission are
limited to host interactions during the autumn mating season. The three
investigated hosts differ considerably in their social system, most notably in
maternity colony size, mating system and degree of male summer
aggregation. We observed marked differences in parasite infection during the
autumn mating period between the species, closely mirroring the predictions
made based on the social systems of the hosts. Increased host aggregation
sizes in summer yielded higher overall parasite prevalence and intensity, both
in male and female hosts.
Moreover, parasite levels in male hosts
differentially increased throughout the autumn mating season in concordance
with the degree of contact with female hosts afforded by the different mating
systems of the hosts. Critically, the observed host-specific differences have
important consequences for parasite population structure, and will thus
affect the coevolutionary dynamics between the interacting species.
Therefore, in order to accurately characterize host-parasite dynamics in hosts
with complex social systems, a holistic approach that investigates parasite
infection and transmission across all periods is warranted.
30
Introduction
Host-parasites interactions are omnipresent in biological communities
(Schmid-Hempel 2011). The epidemiology and population dynamics of
parasites are influenced by many abiotic and biotic factors, especially the
interaction of host and parasite life histories (Nadler 1995, Barrett et al. 2008).
For example, in non-mobile permanent parasites that spend their whole life
cycle on their hosts, infection intensity and diversity increases with increasing
host group size (Côté and Poulin 1995, Patterson and Ruckstuhl 2013) and
host density (Krasnov et al. 2002). Moreover, in parasites that are unable to
disperse independently of their host, horizontal transmission is dependent on
host spatiotemporal dynamics (Nunn and Altizer 2006). As a result, in many
permanent parasites, infections are biased towards one host sex and show
strong seasonal fluctuations as a function of the dynamics of host social
interactions (Krasnov et al. 2005, Altizer et al. 2006).
The degree of parasite dispersal between host social groups will have
strong consequences for the evolutionary dynamics between host and
parasite. For example, genetic analyses of parasites on several bird species
showed that parasite populations are largely undifferentiated among
breeding grounds as a result of inter-colony contact and parasite
transmission at communal wintering grounds or extensive prospecting by
juvenile host individuals (Gomez-Diaz et al. 2012, Levin and Parker 2013).
Parasite transmission between host populations are especially relevant from
an evolutionary perspective, as the potential for parasite local adaptation is
largely determined by the relative rate of gene-flow in host and parasite
(Gandon and Michalakis 2002), and the effective population size of both
(Gandon and Nuismer 2009a). Thus, when assessing host-parasite
coevolutionary dynamics, it is essential to not only characterize the phases of
the host’s annual cycle in which parasite infection levels are highest, but also
parasite phenology and transmission opportunities throughout the rest of the
annual cycle. The latter, however, has rarely been done so far.
31
European temperate zone bats offer an interesting system for
investigating the effects of host social system and seasonality on the
prevalence, intensity and transmission of permanent parasites. During
summer, female bats typically form large, dense maternity colonies to raise
their offspring (Kerth 2008). In addition to the large aggregation size, both
females and juveniles show reduced immunocompentence and grooming
compared to other times of the year (Christe et al. 2000). These conditions
result in high levels of parasite infection intensity and parasite population
growth in many bat colonies (Lučan 2006, Lourenço and Palmeirim 2008). In
contrast to female bats, males are often solitary throughout the summer,
although they may join female colonies or form male bachelor groups in
some species (Safi and Kerth 2007). As a result, unlike most other
mammalian systems (Zuk and McKean 1996, Schalk and Forbes 1997),
females of most temperate zone bats have higher levels of parasite infection
than males (e.g. Reckardt and Kerth 2009). Contact between the philopatric,
closed female maternity colonies, and between females and males, is very
rare during summer (Burland and Wilmer 2001, Kerth and van Schaik 2012).
This results in limited horizontal transmission (used here as parasite
exchange between different host colonies) opportunities for contacttransmitted parasites during this time (Bruyndonckx et al. 2009b, van Schaik
et al. 2015a).
In late summer and autumn, female maternity colonies disband and
bats meet to mate, thereby also allowing for parasite transmission. Mating in
temperate zone bat species is promiscuous and predominantly occurs either
in temporary harems (Zahn and Dippel 1997), locally with males in the
summer home range of the females (Senior et al. 2005), or during a behavior
known as swarming, in which bats show intense flight activity at the entrance
to underground sites used for hibernation in winter (Fenton 1969). During
these encounters, parasites may be directly transmitted between females (in
harems), or by males acting as a vector for parasite transmission between
females from different colonies (van Schaik et al. 2014). In winter, bats
hibernate either solitarily or in clusters (Dietz et al. 2007). In the latter case,
32
additional parasite transmission may take place during this time. However,
the reduced temperatures and torpor of the hosts correspondingly also
reduces the metabolism and activity of ectothermic parasites (Christe et al.
2007). Taken together, although parasite populations are largest during
summer, effective transmission and subsequent survival throughout autumn
and possibly winter is key to parasite persistence and dispersal within the
host meta-population. However, comparative studies of parasite population
dynamics and transmission on bats with different social organization and
mating systems in autumn are lacking.
In this context, we explore the transmission and temporal infection
dynamics of a common genus of ectoparasite (wing-mites of the genus
Spinturnix) on three common European bat species at swarming sites during
the autumn mating period. The three bat species: Daubenton’s bat (Myotis
daubentonii), the greater mouse-eared bat (Myotis myotis) and Natterer’s bat
(Myotis nattereri) all follow the same annual cycle described above, but differ
substantially in social system, as summarized in Table 2.1. Most notably,
they differ in social organization during summer (female maternity colony size
and male bachelor groups) and mating system (temporary harem formation
and promiscuous mating at swarming sites), providing a comparative
framework to explore the effects of these host characteristics on parasite
prevalence and distribution across the host sexes.
Wing-mites of the genus Spinturnix infect almost all European bat
species, and show a clear coevolutionary pattern with their bat hosts
(Bruyndonckx et al. 2009a). They are hematophagous and live exclusively on
the wing and tail membranes of their host (Evans 1968). Larval stages
develop inside the female mite, which gives birth to protonymphs that are
able to move and feed independently (Evans 1968). Thus mites never have to
leave the host. Reproduction occurs almost exclusively in summer maternity
colonies, with strong peaks in the presence of gravid female mites during
host pregnancy and lactation, and no gravid females observed at low
temperatures in winter (Lourenço and Palmeirim 2008).
33
Table 2.1 Overview of the differences in host social system and their
predicted effects on parasite transmission. * samples were taken from
lactating females only. § for all species male bats are sporadically also
observed in female maternity colonies. 1 (Krapp 2004), 2 (Lučan 2006), 3
(Postawa and Szubert-Kruszyńska 2014), 4 (Encarnação and Reiners 2012), 5
(Zahn and Dippel 1997)
Summer (Females)
Social Organization
Parasite prevalence
/ mean intensity
Summer (Males)
Social Organization
M. daubentonii
M. myotis
M. nattereri
Predicted effect
Small colonies
(10-100)1
Large colonies
(50-2000)1
Small colonies
(10-100)1
Larger colony size will
yield higher parasite
prevalence and intensity
100% /
10.3±1.1*2
100% /
17.9±1.22 3
-
Solitary or male
bachelor groups
Solitary §
Solitary §
Male groups will yield
higher parasite
prevalence and intensity
Swarming and
Temporary
harems 1, 5
Swarming 1
Temporary harems will
yield higher parasite
transmission than
swarming
§
Autumn
Mating system
Swarming and
Local mating 1, 4
Mite infection has been observed to decrease during autumn and winter in
M. daubentonii (Lučan 2006), although this pattern was not observed in
autumn in the North American bat Myotis lucifugus (Webber et al. 2015). This
reduction is presumably due a combination of reduced mite reproduction and
increased grooming activity and immunocompentence in their hosts. Mites
are able to distinguish between host species (Giorgi et al. 2004) and host sex
(Christe et al. 2007) and show a preference for juvenile hosts in summer
maternity colonies (Christe et al. 2000). Mite infection increases host
grooming activity and metabolism (Giorgi et al. 2001), and thus may impose a
substantial cost, especially to juvenile hosts in maternity colonies (Lourenço
and Palmeirim 2007). A study of mite population genetic structure across
maternity colonies in M. myotis and Myotis bechsteinii, the latter having a
similar social organization as M. nattereri (Rivers et al. 2005, Kerth and van
Schaik 2012), found contrasting levels of genetic diversity and genetic
differentiation between the maternity colonies of the two host species (van
34
Schaik et al. 2014). Nevertheless, in both M. myotis and M. bechsteinii,
substantial horizontal transmission of mites during the autumn mating season
was implicated but could not be examined.
In the present study, we investigated mite infection of bats captured at
three swarming sites in Germany throughout the bats’ autumn mating
season. As mite infection intensity reduces rapidly after the disbandment of
summer maternity colonies in autumn, and as there is substantial contact
between conspecifics in the bats during mating at this time, we expect most
successful mite horizontal transmission to occur during this period. Using
linear mixed effect models, we investigated the temporal dynamics of
infection throughout the mating season in each of the different host age and
sex classes. By comparing the results observed in each species, we aim to
explore the effect of differences in host social system on parasite infection
and opportunities for transmission. We hypothesize that differences in host
female social organization will affect overall mite prevalence and intensity
according to their summer maternity colony size, but that overall infection will
decrease throughout the season in all species. Additionally, we predict that
differences in host mating system and male social organization will affect the
temporal dynamics of infection across host sexes during the autumn
swarming season: with higher levels of mite prevalence and intensity in males
of M. daubentonii as a result of their summer bachelor groups; moderate
levels in M. myotis through extended contact with females in temporary
harems; and low levels in M. nattereri through exclusive contact to females at
swarming sites.
35
Methods
Study site and bat capture
The study was carried out at three large swarming sites in Germany (Table
2.2), during August and September in 2011 and 2012. Bats were caught
using mist-nets (Schönsteinhöhle and Esperhöhle) or a harp trap (Brunnen
Meyer) at the entrance of each site. In both years each site was sampled four
times (2011: 26 August - 29 September; 2012: 4 August - 27 September),
approximately every 7 (2011) to 14 (2012) days, depending on weather
conditions. This temporal range covers the peak swarming activity for all
three species (Piksa et al. 2011, van Schaik et al. 2015b).
Table 2.2 Overview of the three sampling locations, the number of individuals
of each species caught at each site and the total number of mites collected
per species.
Location
Brunnen Meyer
Coordinates
51.96, 7.37
Esperhöhle
49.76, 11.29
Schönsteinhöhle
49.81, 11.24
Total
Total number of mites sampled
Year
2011
2012
2011
2012
2011
2012
Sampling
events
M. daubentonii
4
111
4
111
4
6
4
36
4
14
4
38
24
316
414
36
M. myotis
1
3
181
245
52
43
525
1873
M. nattereri
175
123
31
39
252
137
757
241
For each individual bat, species, forearm length, mass, and age were
recorded. Age was classified as either adult or young of year (YOY) based on
the level of ossification of the epiphyseal joints (Kunz 1988). At the
Schönsteinhöhle and Esperhöhle, all captured bats were temporarily marked
on the thumbnail using permanent marker to exclude recaptured individuals.
At the Brunnen Meyer, M. nattereri and M. daubentonii had been marked for
a concurrent study (F. Meier, L. Grosche, unpublished data), thereby
excluding recaptures. During our study, bats of ten other species were also
caught (n=214); a summary of these captures (including mite prevalence and
intensity) is given in the supplementary materials (Table S2.1). Notably,
several species were caught in large numbers but were rarely or never
infected with mites (see discussion).
Captured bats were placed in separate capture bags to avoid
contamination of mites prior to sampling. Bat wing and tail membranes were
systematically inspected and all mites were collected and stored in 96%
alcohol.
Data analysis
Infection dynamics were investigated using two standard measures of
parasite infection: prevalence, the proportion of hosts infected with one or
more mites, and intensity, the number of mites on a single infected host
excluding hosts which are not parasitized (sensu (Bush et al. 1997)), using
generalized linear mixed effects models (package lme4 in R 2.14.1; (Bates et
al. 2014, R Development Core Team 2015)). First, for each species and each
37
measure (prevalence, intensity) we ran a separate model in which we
compared overall infection levels across the four host age and sex
combinations (adult male, adult female, YOY male, YOY female; henceforth
referred to as host classes), and a second set of models comparing the
overall temporal trend of infection pooled across all host classes. In these
models, host class or capture date (respectively) were used as fixed effects,
and capture site and year were included as random effects. Subsequently, to
investigate the change in infection level of each host class throughout the
swarming season, the models were run with both host class and capture
date as fixed effects and capture site and year as random effects.
In all cases, mite prevalence was modeled using logit-linked binomial
errors. Mite intensity was modeled using a poisson distribution (log-link). As
we intended to compare the temporal infection dynamics of male hosts to
those of females, adult female hosts were used as the intercept for all
models, and all other classes were evaluated relative to these estimates. To
assess model output and to evaluate differences among host classes, we
used the sim function (package arm; (Gelman et al. 2014)) to generate 95%
credibility intervals based on 2000 simulations of the posterior distribution.
Credibility intervals that did not include zero were considered to be
significant in the frequentist sense. For all models, the influence of the
random effects (sampling location and sampling year) is summarized in the
supplementary materials (Table S2.2).
As mite intensity (log-link) models were overdispersed, both with and
without day as an additional fixed effect, they were also analyzed using a
38
zero-truncated distribution (pospoisson: package VGAM; (Yee and Wild
2010)). In this model host class was included as a fixed effect with no
random effects included in the model. The additional models did not yield
qualitatively different results (Supplementary materials, Table S2.3).
39
Results
Capture overview
A total of 1598 bats of the three target species were caught, and 2528 mites
were collected from them (Table 2.2). We caught more male than female bats
in all three species, although male bias was stronger in M. daubentonii and
M. nattereri than in M. myotis (0.66, 0.69 and 0.56 respectively).
Approximately one third of the bats caught were identified as young of year
(0.30, 0.37 and 0.37 respectively).
Myotis daubentonii
In M. daubentonii, overall prevalence and intensity did not differ between
host classes (Table 2.3; Figure 2.1). Pooled across host classes, mite
prevalence and intensity all decreased over time during the swarming season
(Table 2.3). The temporal decrease in both infection parameters was
comparable across adult females, adult males and YOY males (Table 2.4;
Figure 2.2a, 2.3a). However, a stronger decrease in mite prevalence was
observed in YOY females relative to adult female hosts (Table 4).
Myotis myotis
In M. myotis, overall prevalence and intensity were all higher in adult females
than in both adult male and YOY male hosts (Table 2.3, Figure 2.1).
Prevalence levels between adult females and YOY females did not differ, but
mite intensity was lower in YOY females (Table 2.3). Pooled across host
classes, mite prevalence did not change temporally, but mite intensity
decreased throughout the swarming season (Table 2.3).
Temporal dynamics of mite prevalence and intensity did not differ
between adult females and both male and female YOY, with decreases in
both parameters throughout the mating season (Table 2.4; Figure 2.2b, 2.3b).
Conversely, mite prevalence and intensity increased over the swarming
season in adult male bats (Table 2.4). Indeed, by the end of swarming
40
season, prevalence and intensity levels in adult male hosts approached those
seen in the other host classes (Figure 2.2b, 2.3b).
Table 2.3 Estimated effect sizes and 95% credibility intervals for mite
prevalence (binomial; logit-scale) and intensity (poisson; log-scale) for each
host class pooled across all capture dates (top) and the overall temporal
trend pooled across all host classes (bottom) with capture location and year
as random effects. * denotes estimates where the 95% CI does not include
zero.
Prevalence
Species
Fixed effect
Host class
M. daubentonii
M. myotis
M. nattereri
Estimate
95% CI
Estimate
95% CI
Intercept (Ad F)
0.33
(-0.35, 0.99)
0.76
(0.56, 0.98)
Ad M
-0.13
(-0.71, 0.42)
0.16
(-0.08, 0.41)
YOY F
0.08
(-0.74, 0.87)
-0.17
(-0.54, 0.19)
YOY M
-0.48
(-1.19, 0.22)
0.11
(-0.19, 0.43)
Intercept (Ad F)
2.42
(1.82, 3.04)
1.55
(0.73, 2.33)
Ad M
-3.15*
(-3.82, -2.48)
-0.88*
(-1.05, -0.70)
YOY F
-0.42
(-1.37, 0.38)
-0.14*
(-0.26, -0.03)
YOY M
-1.86*
(-2.59, -1.15)
-0.57*
(-0.70, -0.44)
Intercept (Ad F)
-0.91
(-1.33, -0.50)
0.58
(0.34, 0.80)
Ad M
-2.18*
(-2.83, -1.54)
-0.17
(-0.66, 0.28)
YOY F
0.1
(-0.46, 0.69)
0.23
(-0.11, 0.57)
-0.17
(-0.64, 0.32)
-0.06
(-0.37, 0.27)
Intercept (1 Aug)
1.39
(0.76, 2.03)
1.28
(1.06, 1.50)
Day
-0.03*
(-0.05, -0.02)
-0.01*
(-0.02, -0.01)
Intercept (1 Aug)
0.52
(0.06, 0.96)
2.26
(2.03, 2.49)
Day
0.00
(-0.01, 0.01)
-0.02*
(-0.02, -0.02)
Intercept (1 Aug)
-1.08
(-1.90, -0.24)
1.13
(0.61, 1.64)
Day
-0.01
(-0.03, 0.01)
-0.01*
(-0.02, 0.00)
YOY M
Temporal
M. daubentonii
M. myotis
M. nattereri
Intensity
41
N
1.00
71
149
37
59
YOY M
a
YOY F
Figure 2.1 Overall (a) prevalence and (b) intensity of mites on bats captured
(mean ± SE), subdivided by host class. N denotes the number of sampled
individuals per host class.
147
184
84
110
143
332
88
194
Prevalence
0.75
0.50
YOY M
YOY F
Ad M
Ad F
YOY M
74
70
41
14
28
50
YOY M
YOY F
Ad M
60
YOY F
135
Ad M
28
Ad F
23
YOY M
84
M. nattereri
YOY F
41
YOY M
N
8
YOY F
b
M. myotis
Ad M
M. daubentonii
Ad M
Ad M
Ad F
0.00
Ad F
0.25
Intensity
6
4
M. daubentonii
Ad F
Ad F
2
M. myotis
42
M. nattereri
Myotis nattereri
Overall prevalence and intensity in M. nattereri were low (Figure 2.1), and did
not differ between adult females and YOY of both sexes (Table 2.3). Mite
prevalence on adult males was drastically lower (Table 2.3), but mite intensity
did not differ relative to the levels found in adult females. Similar to M.
myotis, when pooled across host classes, mite prevalence did not change
temporally, but mite intensity decreased throughout the swarming season
(Table 2.3).
Both prevalence and intensity decreased over the swarming season in
adult females and both male and female YOY (Table 2.4, Figure 2.2c, 2.3c),
whereas they increased over time in adult male hosts (Figure 2.2c, 2.3c),
albeit insignificantly (Table 2.4).
Table 2.4 Estimated effect sizes and 95% credibility intervals for the
temporal change of mite prevalence (binomial; logit-scale) and intensity
(poisson; log-scale) per host class, with capture location and year as random
effects. * indicates estimates where the 95% CI does not include zero.
Prevalence
Species
M. daubentonii
M. myotis
M. nattereri
Intensity
Fixed effect
Intercept (Ad F * day)
Estimate
-0.02
95% CI
(-0.06, 0.02)
Estimate
-0.01
95% CI
(-0.03, 0.00)
Ad M * day
-0.01
(-0.05, 0.04)
0.00
(-0.01, 0.02)
YOY F * day
-0.11*
(-0.19, -0.02)
-0.01
(-0.04, 0.02)
YOY M * day
-0.04
(-0.10, 0.02)
0.00
(-0.03, 0.02)
Intercept (Ad F * day)
-0.05
(-0.09, 0.01)
-0.02
(-0.02, -0.01)
Ad M * day
0.11*
(0.06, 0.15)
0.02*
(0.00, 0.03)
YOY F * day
-0.01
(-0.07, 0.06)
0.00
(-0.01, 0.00)
YOY M * day
0.05
(-0.01, 0.09)
0.00
(0.00, 0.01)
Intercept (Ad F * day)
-0.03
(-0.06, 0.01)
-0.01
(-0.03, 0.01)
Ad M * day
0.05
(-0.01, 0.10)
0.01
(-0.05, 0.06)
YOY F * day
-0.02
(-0.08, 0.03)
-0.01
(-0.04, 0.02)
YOY M * day
0.01
(-0.03, 0.05)
0.01
(-0.02, 0.04)
43
Figure 2.2 Temporal dynamics of mite prevalence during the autumn mating
season in (a) M. daubentonii, (b) M. myotis, (c) M. nattereri.
a
1.00
Prevalence
M. daubentonii
0.75
0.50
0.25
0.00
0
b
20
40
60
1.00
Age/Sex class
Prevalence
M. myotis
0.75
Adult F
Adult M
0.50
YOY F
YOY M
0.25
0.00
c
0
20
40
60
0
20
40
60
1.00
Prevalence
M. nattereri
0.75
0.50
0.25
0.00
Day (1 = Aug 1)
44
Figure 2.3 Temporal dynamics of mite intensity during the autumn mating
season in (a) M. daubentonii, (b) M. myotis, (c) M. nattereri. Points indicate
average intensity per host class per sampling day.
a
5
Intensity
M. daubentonii
4
3
2
1
0
0
b
20
40
60
Intensity
M. myotis
10
Age/Sex class
Adult F
Adult M
YOY F
5
YOY M
0
c
0
20
0
20
40
60
5
Intensity
M. nattereri
4
3
2
1
0
Day (1 = Aug 1)
45
40
60
Discussion
Through a comparative framework across three European bat species and
their parasitic wing mites, we observed substantial differences in initial
parasite infection of all host classes and the infection dynamics of male hosts
between species as predicted based on the differences in their social
systems.
Overall mite prevalence and intensity
In female and YOY hosts, overall prevalence and intensity levels were highest
in M. myotis, intermediate in M. daubentonii and lowest in M. nattereri. In
male hosts prevalence and intensity was highest in M. daubentonii with lower
levels of infection in the other two species. The observed relative infection
rates of the hosts closely mirror the size of host summer aggregations, where
M. myotis have larger maternity colonies than the other two species, and
males of M. daubentonii form bachelor group aggregations while the males of
the other species are almost exclusively solitary during summer. Notably,
although overall prevalence and intensity values of females also broadly
correspond to the differences in body size between the species (M. myotis >
M. daubentonii and M.nattereri), this pattern does not hold for the males of
these species and across the other bat species also captured at the same
sites (compare in Supplementary materials; Table S2.1). Additionally, body
size does not account for the marked differences observed between sexes
within species, nor the different temporal dynamics of mite infection
observed in the three host species. Therefore, we conclude that host social
organization in summer, particularly aggregation size, shapes the overall level
of mite infection seen during the autumn mating phase across these species.
Temporal dynamics
Despite the difference in overall mite prevalence and intensity across
species, the overall temporal dynamics of mite infection of female and YOY
hosts (i.e. all host classes living in maternity colonies previous to our
46
sampling) were broadly similar between the investigated species. Overall
mite prevalence and intensity levels at the beginning of the autumn mating
season were already lower than peak infection levels observed in host
summer colonies (Lučan 2006, Encarnação et al. 2012, Postawa and
Szubert-Kruszyńska 2014). Additionally, mite intensity continued to decrease
throughout the investigated period in all species, and in M. daubentonii
overall prevalence also decreased. Presumably this decrease is the result of
the disbandment of host maternity colonies during our sampling period, an
increase in immunocompentence and grooming behavior of the hosts
(Christe et al. 2000), and the cessation of reproduction in the mites (Deunff
and Beaucournu 1981, Lourenço and Palmeirim 2008). Interestingly, in
contrast to the decrease observed in this study, no decrease in parasite load
was observed in Myotis lucifugus during the autumn mating season in
Canada (Webber et al. 2015). This was suggested to be a result of the cold
climate, which resulted in increased host aggregation during the mating
season and a comparatively short mating season due to the earlier onset of
hibernation (Webber et al. 2015), a pattern which is not observed in the three
species studied here. Taken together, these results highlight the role of the
abiotic environment in shaping host social organization and subsequently
host-parasite dynamics.
In contrast to female and YOY hosts, male temporal infection
dynamics did differ between species. In M. myotis adult male bats were
initially rarely infected, but by the end of the season, prevalence and intensity
of mites reached similar infection levels to those seen in adult female and
YOY hosts at this time, suggesting that substantial transmission is taking
place during the autumn mating period. This transmission can either occur
during contact between hosts at swarming sites, or in the temporary harems
formed during this period in which adult males roost together with several
females (either simultaneously or consecutively) for one or several days (Zahn
and Dippel 1997). In M. nattereri mite prevalence and intensity on male bats
was similarly low initially, but unlike the increase seen in M. myotis, only
trended upward throughout the autumn mating period. This is likely due to
47
the lower overall prevalence and intensity observed across all host classes in
M. nattereri, combined with the limited transmission opportunities afforded
by the short inter-individual contact during mating at swarming sites in
contrast to the relatively long contact time in harem groups in M. myotis.
Finally, in M. daubentonii, temporal infection dynamics of male bats did not
differ from those seen in female and YOY hosts, as predicted from their
aggregation in male bachelor groups throughout the summer. As all host
classes showed similar prevalence and intensity levels throughout the
season, we cannot infer that transmission of parasites is taking place as in
the other bat species, although it appears likely given that the majority of
individuals are infected, and individuals must come into contact for mating.
Altogether across the three systems, we find that the temporal dynamics of
mite infection strongly reflect the variation in host mating system as well as
the changes in host social organization during this period.
Host social system and parasite infection dynamics
The observed correlations between host social system and mite infection
dynamics have several important implications for parasite population
structure. In M. myotis, the formation of temporary harems during the mating
season is particularly likely to increase parasite transmission, as individuals
remain in contact for much longer than during the brief mating events at
swarming sites. In addition to resulting in higher infection levels of male bats,
S. myoti mites may also be transmitted directly from female to female
possibly originating from different colonies in harems where multiple females
roost with a male simultaneously (Zahn and Dippel 1997). Moreover, the high
overall prevalence of mites on all host classes is likely to facilitate further mite
transmission during winter. As a result, the mite meta-population should be
well mixed and its effective population size is expected to be much larger
than in the other investigated species. Support for this prediction is found in
an analysis of mite population genetic structure across six M. myotis
maternity colonies, which found high overall genetic diversity and very little
48
genetic differentiation between mites from different M. myotis colonies (van
Schaik et al. 2014).
In M. daubentonii, the presence of male bachelor groups increases the
number of host clusters available to mites during their reproductive period
(summer). This should increase the overall parasite persistence within the
local meta-population, making regional extinctions (see M. nattereri
discussion below) far less likely. Additionally, it effectively increases the
amount of transmission of mites during the bats’ mating season, as mites no
longer need to transfer exclusively to female hosts, but can additionally
persist on male hosts. Therefore, despite lower overall levels of mite infection
than in M. myotis, mite (genetic) population structure is likely to be similarly
well mixed and temporally stable across the mite meta-population.
Interestingly, mites have previously been shown to have a preference for
female hosts over male hosts in M. daubentonii (Christe et al. 2007), but this
preference was not reflected in the mite prevalence and intensities observed
here. However, we did find a slightly higher prevalence and intensity of mites
on adult hosts of both sexes over YOY hosts. This may indicate a preference
for adult hosts of either sex during the mating (and hibernation) phase, which
would be advantageous as these host individuals are more likely to survive
than YOY (Lentini et al. 2015).
Finally, in M. nattereri, we observed low overall level of mite infection,
and limited transmission during the swarming season. In combination with
the small female maternity colony size, we therefore anticipate survival of
mite populations from year to year in individual host populations to be much
lower than in the other investigated mite species. As a result, local
extinction/re-colonization events are expected to be frequent (Reckardt and
Kerth 2009). Such local extinctions in individual host summer maternity
colonies would further lower the overall prevalence and intensity of mites at
the onset of autumn swarming in the subsequent year. This may explain the
observed difference in prevalence and intensity between M. nattereri and M.
daubentonii despite their similar summer maternity colony sizes. Indeed, if
the density of host colonies in a particular area is low, and local mites
49
populations within several host maternity colonies go extinct, horizontal
transmission of mites at swarming sites may be insufficient to prevent
regional extinction of mites within the catchment area of those swarming
sites.
A comparison with Myotis bechsteinii, which has a similar social
system to M. nattereri (Dietz et al. 2007), shows that such a scenario may be
plausible. In this study we captured over 70 M. bechsteinii but we did not find
any to be infected with mites (see Table S2.1), despite the fact that such
mites are common in several nearby (±100 km) German regions
(Bruyndonckx et al. 2009b, van Schaik et al. 2014). This stochasticity in local
extinction patterns can also have strong consequences for parasite
population genetic structure in regions where the parasite does not go
regionally extinct. In M. bechsteinii, mite populations from different maternity
colonies were temporally unstable as a result of the limited transmission and
low number of founding individuals per host colony (Bruyndonckx et al.
2009b, van Schaik et al. 2014).
Conclusion
Taken together, our findings illustrate several important effects of host social
system on parasite population dynamics. As shown in many other vertebrate
systems, host group size and density during the reproductive period of both
host and parasite strongly influence parasite infection intensity (eg. Côté and
Poulin 1995, Rifkin et al. 2012, Patterson and Ruckstuhl 2013). However, as a
result of host social behavior restricting interactions between different host
social units, horizontal transfer of parasites between such aggregations may
be limited (Barrett et al. 2008). In hosts with strong seasonality, social
structure may subsequently change drastically through the year, as is the
case in temperate zone bats. These changes provide additional horizontal
transmission opportunities for permanent ectoparasites. This could happen
either via hosts of the opposite sex acting as vectors, or through direct
contact of conspecifics from different colonies as shown here for bats and in
other studies for migratory bird species with communal wintering grounds
50
(Gomez-Diaz et al. 2012). In addition, seasonal changes in the social
organization of hosts are often coupled with changes in host behavior (eg.
grooming rate) or physiology (eg. hibernation), which may also influence
parasite survival. Therefore, both the social organization of hosts during the
main reproductive season, as well as the transmission opportunities and
challenges to parasite survival during the “suboptimal” seasons will combine
to affect parasite population and genetic structure.
Notably, the resulting parasite population structure will ultimately
affect the coevolutionary dynamics between host and parasite (Boots et al.
2004, Gandon and Nuismer 2009a, Lion and Boots 2010). Therefore, in order
to understand both the proximate and ultimate dynamics of parasite infection
in hosts where the social organization changes throughout the year, a holistic
approach that includes infection during the main reproductive period, but
also the focal points of parasite transmission and the critical phases for
parasite survival, is required.
Acknowledgements
This study was funded by a grant from the Volkswagen Stiftung (Az 84 871).
We would like to thank Frauke Meier, Lena Grosche, Johannes Mohr,
Matthias Hammer, Leonie Baier, Stefan Greif, and many other volunteers for
their help in the field. We also thank Robin Abbey-Lee for her help with data
analysis, and Susanne Seltmann for her critical comments on the manuscript.
JvS is a member of the Max Planck Research School for Organismal Biology.
51
Supplementary Materials
Table S2.1 Overview of the other bat species caught at swarming sites
during this study, including their mite prevalence and intensities. Notably,
mites were entirely absent on M. bechsteinii despite moderately high sample
sizes.
Species
Sex - Age
N
Ninfected
Nparas
Prevalence
Intensity
B. barbastellus
M Adult
35
9
17
0.26
1.88
M YOY
3
1
2
0.33
2
F Adult
7
3
3
0.43
1
M Adult
3
1
1
0.33
1
M YOY
1
0
0
0
-
E. serotinus
M Adult
1
0
0
0
-
M. alcathoe
M adult
1
0
0
0
-
F adult
1
0
0
0
-
M Adult
45
0
0
0
-
M YOY
21
0
0
0
-
F Adult
2
0
0
0
-
F YOY
7
0
0
0
-
M Adult
6
1
2
0.17
2
M YOY
7
2
4
0.29
2
F Adult
5
3
4
0.60
1.33
M Adult
7
0
0
0
-
M YOY
1
0
0
0
-
F Adult
1
0
0
0
-
M Adult
9
2
5
0.22
2.5
M YOY
12
1
2
0.08
2
F Adult
7
2
4
0.29
2
F YOY
5
2
7
0.40
3.5
M Adult
12
1
4
0.08
4
M YOY
3
2
3
0.67
1.5
F Adult
2
0
0
0
-
F YOY
2
2
5
1.00
2.5
M Adult
6
0
0
0
-
M YOY
1
0
0
0
-
F Adult
1
0
0
0
-
E. nilssonii
M. bechsteinii
M. brandtii
M. dasycneme
M. mystacinus
P. auritus
P. pipistrellus
52
Table S2.2 Since mixed models of intensity were overdispersed when run
using a poisson fit, these models were reevaluated using a zero-truncated fit
(pospoisson VGLM, package VGAM; Yee and Wild 2010). In all tests,
direction and significance did not differ from the initial model.
Intensity - Overall
Intensity - Temporal
Species
Fixed effect
Estimate
P
Fixed effect
Estimate
P
M. daubentonii
Intercept (Ad F)
0.6
<0.01
Intercept (Ad F * day)
-0.02
0.02
Ad M
0.24
0.12
Ad M * day
0.01
0.42
YOY F
-0.29
0.24
YOY F * day
-0.03
0.14
YOY M
0.16
0.40
YOY M * day
0.00
0.91
Intercept (Ad F)
2.00
<0.01
Intercept (Ad F * day)
-0.02
<0.01
Ad M
-1.14
<0.01
Ad M * day
0.02
0.05
YOY F
-0.20
<0.01
YOY F * day
-0.01
0.12
YOY M
-0.67
<0.01
YOY M * day
0.00
0.32
Intercept (Ad F)
0.26
0.12
Intercept (Ad F * day)
-0.02
0.07
Ad M
-0.39
0.32
Ad M * day
0.00
0.96
YOY F
0.40
0.08
YOY F * day
-0.01
0.51
YOY M
-0.12
0.60
YOY M * day
0.02
0.42
M. myotis
M. nattereri
53
Table S2.3 Location and sampling year were included in all models as
random effects. Although no consistent patterns was evident, both location
and sampling year did contribute significantly to individual models as
summarized in the table below. In M. myotis, mite intensity differed across
sampling locations when the very small sample set from Brunnen Meyer (n =
4) was included, but did not differ when these were not considered.
Species
Random effect
M. daubentonii
location
overall per host class
year
M. myotis
Prevalence
Intensity
*
location
*
loc without BM
year
M. nattereri
*
location
year
M. daubentonii
location
year
overall temporal
M. myotis
location
loc without BM
year
M. nattereri
*
location
*
year
M. daubentonii
location
host class x temporal
year
M. myotis
location
loc without BM
year
M. nattereri
location
*
year
54
55
Chapter 3
The effect of host social system on parasite
population genetic structure: comparative
population genetics of two ectoparasitic mites
and their bat hosts
J. van Schaik, G. Kerth, N. Bruyndonckx and P. Christe
BMC Evolutionary Biology, 14:18, 2014
♂ Spinturnix bechstenii
56
57
Abstract
Background
The population genetic structure of a parasite, and consequently its ability to
adapt to a given host, is strongly linked to its own life history as well as the
life history of its host. While the effects of parasite life history on their
population genetic structure have received some attention, the effect of host
social system has remained largely unstudied. In this study, we investigated
the population genetic structure of two closely related parasitic mite species
(Spinturnix myoti and Spinturnix bechsteini) with very similar life histories.
Their respective hosts, the greater mouse-eared bat (Myotis myotis) and the
Bechstein’s bat (Myotis bechsteinii) have social systems that differ in several
substantial features, such as group size, mating system and dispersal
patterns.
Results
We found that the two mite species have strongly differing population genetic
structures. In S. myoti we found high levels of genetic diversity and very little
pairwise differentiation, whereas in S. bechsteini we observed much less
diversity, strongly differentiated populations and strong temporal turnover.
These differences are likely to be the result of the differences in genetic drift
and dispersal opportunities afforded to the two parasites by the different
social systems of their hosts.
Conclusions
Our results suggest that host social system can strongly influence parasite
population structure. As a result, the evolutionary potential of these two
parasites also differs strongly, thereby affecting the risk and evolutionary
pressure exerted by each parasite on its host.
58
Introduction
In the evolutionary arms race between hosts and their parasites, the
interacting species may use immunological, physiological and behavioural
adaptations (Loehle 1995, Ezenwa 2004). Hosts commonly use behavioural
adaptations to avoid exposure, or actively remove, parasites (Hart 1990). In
addition behavioural adaptations may also affect parasite population
structure, thereby potentially reducing parasite intensity (Loehle 1995), and
ultimately their evolutionary potential, classically defined as the ability to
incorporate genotypes able to outcompete those put forward by the
opponent (Gandon and Michalakis 2002). On the population level, a
parasite’s genetic structure is closely linked to the life histories of both
interacting species (Nadler 1995), as well as to the social system of the host
(Altizer et al. 2003). Specifically, the relative rate of dispersal, and thereby
gene flow (Gandon and Michalakis 2002), as well as the relative strength of
genetic drift (Gandon and Nuismer 2009b, Nuismer et al. 2010), are important
in determining host-parasite coadaptation dynamics. Therefore, comparisons
of population genetic structure across multiple host-parasite pairs where
hosts differ in key life-history and/or social system traits, will help elucidate
how these factors affect microevolutionary host-parasite dynamics.
Relative dispersal rates of hosts and parasites have been studied in a
number of systems. For instance, in the black-legged kittiwake and a
parasitic tick, parasite populations were found to be much more spatially
structured than those of their hosts (McCoy et al. 2005). This difference was
attributed to the fact that parasite dispersal was limited to the breeding
season of the hosts, and therefore dispersal of the host outside of this period
did not result in concurrent parasite dispersal (McCoy et al. 2005). In
contrast, in a comparison of dispersal rates between two shearwater species
and three parasitic lice, parasite gene flow was found to be much higher than
that of its host, which was attributed to the transmission of parasites at
communal wintering grounds where no host gene flow took place (GómezDíaz et al. 2007). These examples highlight the fact that relative dispersal is
dependent on the intricate interaction between the life histories of both
59
species. In cases where gene flow in host and parasite are comparable,
genetic drift can also substantially affect host-parasite coevolutionary
dynamics as it can impede adaptation; even in cases where standing genetic
variation is abundant in both species (Nuismer et al. 2010). This is especially
relevant for parasites of vertebrate hosts, where meta-population dynamics
often lead to patchily distributed parasite populations. As a result parasite
populations often experience extinction events and strong population
bottlenecks (e.g. Hafner et al. 2003). Unfortunately, comparative studies
investigating the role of host social system in shaping parasite population
genetic structure are rare.
In this study, we investigate the population genetic structure of two
ectoparasitic mite species that parasitize two closely related bat hosts with
differing social systems. The mite species (Spinturnix myoti and Spinturnix
bechsteini) are closely related and have indistinguishable life histories
(Rudnick 1960, Deunff et al. 2004). As a result, we expect any differences
observed in their microgeographic population genetic structure to be
primarily the result of differences in the social system of their hosts, the
greater mouse-eared bat (Myotis myotis) and the Bechstein’s bat (Myotis
bechsteinii).
The two host species share similar life-history traits but show
substantial differences in their social systems, which include their social
organization, social structure and mating system (Table 3.1). Both are longlived, monotocous, non-migratory European vespertilionid bat species that
follow the temperate cycle. In summer, they form exclusively female
maternity colonies, which however differ between the two species in size,
location, stability and degree of philopatry (Table 3.1). Females from different
maternity colonies of M. myotis sometimes visit other maternity colonies and
occasionally co-localize with maternity colonies of closely related Myotis
blythii, whereas females from M. bechsteinii maternity colonies do not
interact with one another or with other species. Males of M. myotis and M.
bechsteinii disperse from their natal colony and are solitary throughout the
summer (Zahn and Dippel 1997, Kerth and Morf 2004). In the autumn mating
60
61
High; one site (building/
cave) throughout summer
Free-hanging; solitary or
clustered
Roost fidelity
Hibernation
Mating system Temporary harems
(extensive contact) and/or
swarming (little contact)
High; but occasional
exchange of individuals
between colonies
Female natal
philopatry
Swarming (little contact)
In crevices
(mostly solitary)
Very low; frequent roost
switching (tree cavities)
and fission-fusion
dynamics
Very high; almost no
exchange of individuals
between colonies
Small (10–50)
Large (50–2000)
Myotis bechsteinii
Bechstein’s bat
Myotis myotis
Greater mouse-eared bat
Colony size
Social
organization
Castella et al. 2001
Kerth & van Schaik 2012
Krapp 2004
Reference
Temporary harems
increase parasite
transmission
Solitary roosting reduces
parasite transmission
Zahn & Dippel 1997
Kerth et al. 2003
Krapp 2004
Fission-fusion dynamics
Reckardt & Kerth 2007
may increase genetic drift Krapp 2004
because colonies split into
subgroups
Lower philopatry leads to
more parasite
transmission
Larger colonies lead to
less genetic drift
Predicted effect on
parasite population
genetic structure
Table 3.1 Overview of the key differences in social system between the
greater mouse-eared bat (Myotis myotis) and Bechstein’s bat (Myotis
bechsteinii), and the expected effects thereof on the population genetic
structure of their parasites.
season, male M. myotis form temporary harems in August and September
where several females roost in direct contact with a male for one or more
days. In contrast, M. bechsteinii mate at swarming sites where the sexes
meet very briefly during the night (Kerth et al. 2003b). In winter both species
hibernate at underground sites, but again differ in degree of aggregation and
body contact, where M. myotis may form large clusters whereas M.
bechsteinii roosts solitarily (Krapp 2004). Finally, the species differ in the
distance travelled between summer and winter roosts. M. myotis is
considered a regional migrant, easily travelling over 50 km, whereas M.
bechsteinii is considered sedentary, generally not travelling over 30 km
(Hutterer et al. 2005). Both species have recolonized the current study area
since the last glacial maximum. However, for M. myotis there is the possibility
of admixture as multiple glacial refugia have been identified (Iberia and Italy;
Ruedi et al. 2008), while all M. bechsteinii in Central Europe are believed to
originate from the Balkan region (Kerth et al. 2008). Both species are
parasitized by ectoparasitic mites of the genus Spinturnix, which typically
show strong cospeciation with their bat hosts (Bruyndonckx et al. 2009a).
Spinturnix wing mites live exclusively on membranes of bats,
reproduce sexually, are haematophagous and cannot survive off of their host
for more than a few hours (Giorgi et al. 2004). Therefore, while dispersal of
Spinturnix mites is extensive within bat maternity colonies, where female bats
live in close body contact, little (M. myotis) or no (M. bechsteinii) transmission
can occur between different maternity colonies in summer (Bruyndonckx et
al. 2009b). It is expected that horizontal transmission of mites between
colonies occurs during bat mating and hibernation periods. Bat maternity
colonies constitute optimal conditions for parasite reproduction, and
parasites show a strong preference for female and juvenile hosts (Christe et
al. 2007). Mite abundance subsequently strongly decreases in autumn and
throughout hibernation, and it is supposed that they are not able to
reproduce during this time (Lourenço and Palmeirim 2008). Mites are
believed to impose a substantial cost on their hosts during the maternity
period (Lourenço and Palmeirim 2007). For example, in M. myotis individuals
62
experienced increased oxygen consumption and weight loss when
experimentally infected (Giorgi et al. 2001). Mite prevalence and intensity is
much higher in M. myotis (intensity per female 12–20; Christe et al. 2000)
compared to M. bechsteinii (intensity per female 1–10; Reckardt and Kerth
2009). S. bechsteini is strictly host specific. S. myoti is found on M. myotis as
well as its sister species M. blythii (which has an identical social system;
Krapp 2004). In choice experiments S. myoti shows a clear preference for M.
myotis and it is presumed that M. myotis is the main host species (Christe et
al. 2003). A previous study of mtDNA sequence data in S. bechsteini found
strong spatial differentiation between bat colonies and a high temporal
turnover in the haplotypes found per bat colony, suggesting frequent local
extinction and recolonisation (Bruyndonckx et al. 2009b).
By comparing the genetic diversity, gene flow and genetic drift of the
selected parasite species, we aim to assess the consequences that the
differing social systems of the hosts have on the evolutionary potential of the
parasites. In accordance with Nadler (1995), we predict that the larger colony
size and increased contact rates among colonies of M. myotis will result in a
less genetically structured population in S. myoti, and allow for substantial
gene flow between colonies. In contrast, we expect the mite population of M.
bechsteinii
to
be
highly
sub-structured
due
to
reduced
dispersal
opportunities and strong genetic drift as a result of the smaller and
demographically more isolated host colonies.
63
Methods
Previously analysed samples
To compare host and parasite population genetic structure, we have
combined newly generated datasets for both parasites (S. myoti: mtDNA cytb
sequence data and 8 nucDNA microsatellites; S. bechsteini: 5 nucDNA
microsatellites), with previously generated datasets for one of the mite
species (S. bechsteini: mtDNA cytb) and both hosts (mtDNA and nucDNA),
supplemented for M. myotis with two additional colonies for this study. An
overview of the markers and samples previously analysed is given below as
well as in the supplemental materials (Table S3.1).
For M. myotis, four summer maternity colonies had been previously
genotyped
for
ten
microsatellites
and
sequenced
for
the
second
hypervariable domain (HVII) of the mtDNA control region (Castella et al.
2001). For M. bechsteinii, individuals from ten maternity colonies had been
genotyped for eight nucDNA microsatellites as well as two mitochondrial
microsatellites (Kerth et al. 2002b, Kerth et al. 2008). For S. bechsteini, mites
from thirteen bat maternity colonies had been previously sequenced for a
513 bp fragment of Cytochrome B (Bruyndonckx et al. 2009b).
Sample collection and DNA extraction
We sampled S. myoti in six M. myotis maternity colonies in Switzerland and
Northern Italy in August 2004 and 2005 (Figure 3.1, Table 3.2a). Colony size
estimates were made inside the roost and bats were caught either directly
inside the colonies’ roosts during the day, or upon emergence from the roost
entrance at night. Geographic distances between colonies ranged from 16 to
about 200 km, which is comparable to the range M. myotis is known to
disperse between summer and winter roosts (Hutterer et al. 2005). Twenty
mites per bat colony, each originating from a different bat, were used for the
genetic analyses.
64
Figure 3.1 Map of Central Europe showing the sampling locations for Myotis
myotis/Spinturnix myoti (circles), and sampling regions for Myotis
bechsteinii/Spinturnix bechsteini (triangles). For the latter pairing, the number
of sampled colonies within a region is indicated within the triangle because
all colonies within a region are located in close proximity to one another. All
colony and region names correspond to the abbreviations given in Table 3.2.
For each species, a picture of the typical roosting association is shown with
an inset of the studied mite species (picture credits M. bechsteinii: GK, M.
myotis: PC, mites: GK & JvS).
5"
9"
RP"
StU"
LF"
Cou"
Per"
Mei"
Sat"
Agl"
100 km
S. bechsteini samples were collected from 13 M. bechsteinii maternity
colonies in 2002 and in 2007 in two spatially distant regions (±200 km) in
Germany, Lower Franconia (LF) and Rhineland-Palatinate (RP; Figure 3.1). As
M. bechsteinii has not been recorded to disperse over 73 km (Hutterer et al.
2005) direct dispersal of either host or parasite between regions is not
expected. At the same time, within regions, inter-colony distances are again
65
comparable to the expected dispersal distance between summer and winter
roosts. We divided the available 18 sampling events from 13 colonies into
three main subgroups based on spatial and temporal characteristics (LF in
2002: 100 mites from 5 bat colonies; LF in 2007: 195 mites from 8 bat
colonies; RP in 2007: 107 mites from 5 bat colonies). The number of
sampling events in each subgroup is less than were analysed for mtDNA
(Bruyndonckx et al. 2009b) as all events with less than 10 samples were
discarded in order to ensure sufficient sample sizes for all nucDNA analyses.
Samples from 2002 were only used for the temporal analysis, whereas all
other comparisons were performed using only the samples from 2007. In all
cases, all bats within a (bat-box) roost were caught and sampled for mites,
and colony size estimates reflect the number of bats present in the colony.
All mites collected from a colony were used for the genetic analysis.
For both mite species, individuals were removed from the bat’s wing
and tail membrane using soft forceps and stored in 90% ethanol prior to DNA
extraction. Collected mite specimens were individually rehydrated for two
hours in 200 μl of sterile water before being crushed in liquid nitrogen. Total
DNA was isolated from each individual mite using a standard proteinase Kphenol chloroform method (Sambrook et al. 1989).
Table 3.2 (next page) Description of molecular variability in a) Spinturnix
myoti, b) Myotis myotis c) Spinturnix bechsteini d) Myotis bechsteinii. The
following parameters are recorded: latitude (Lat), longitude (Long), estimated
number of bats in the colony (colony size), number of individuals
sequenced/genotyped (n), total number of haplotypes (N), nucleotide
diversity (π), the mean number of microsatellite alleles per locus (NA), allelic
richness (K), observed (Ho) and expected (He) heterozygosities, the number of
private alleles per colony (Pall), and the inbreeding coefficient (Fis). For S.
myoti and S. bechsteini the number of individuals sequenced and genotyped
was the same for each colony.
66
a)
mtDNA
S. myoti
Colony Name
Year
Lat
Long
n
N
Agliè (Agl)
2005
45.3671
7.767
20
14
Courtételle (Cou)
2004
47.341
7.3178
19
13
Meiringen (Mei)
2005
46.7296
8.1812
18
Perreux (Per)
2004
46.9479
6.8179
Satigny (Sat)
2005
46.2143
St-Ursanne (StU)
2005
47.3647
nucDNA microsatellites
π
NA
K
Ho
He
Pall
0.708
17.875
15.955
0.701
18.125
16.752
7
0.777
13.125
12.743
20
10
0.874
18
6.0357
20
8
0.536
7.1537
20
14
0.759
Fis
0.61
0.868
22
0.264
0.755
0.904
11
0.191
0.546
0.871
6
0.209
16.121
0.733
0.898
13
0.320
15.5
14.122
0.663
0.872
16
0.398
17.5
15.787
0.695
0.890
11
0.244
b)
mtDNA
M. myotis
Colony Name
Colony size
nucDNA microsatellites
n
N
π
n
NA
K
Ho
He
Pall
Fis
Agliè (Agl)
400
20
4
1.817
20
8.2
8.71
0.735
0.723
5
-0.017
Courtételle (Cou)
800
20
2
0.033
20
9
8.39
0.785
0.761
2
-0.031
80
20
2
0.062
20
9
7.43
0.795
0.772
4
-0.029
Perreux (Per)
200
20
3
0.161
20
9.9
8.16
0.79
0.789
9
-0.002
Satigny (Sat)
190
18
3
1.993
20
9.3
7
0.789
0.763
3
-0.034
St-Ursanne (StU)
200
20
1
0
20
9.7
9.03
0.733
0.763
6
0.039
Meiringen (Mei)
c)
mtDNA
S. bechsteini
Colony Name
Year
Lat
Long
n
Blutsee (BS)
2007
49.4331
9.4958
41
3
0.18
5.2
3.981
0.736
0.645
1
-0.130
Einsiedeln (ES)
2002
49.5407
9.5641
16
2
0.32
6.8
5.806
0.727
0.703
2
0
11
4
0.67
6.8
6.237
0.739
0.754
0
0.069
15
3
0.6
12
9.354
0.747
0.864
10
0.169
20
2
0.27
6.6
5.24
0.747
0.706
2
-0.032
25
3
0.46
8.2
6.133
0.677
0.750
2
0.118
22
6
0.77
10.2
7.822
0.757
0.843
2
0.125
21
5
0.49
7.4
5.9
0.632
0.743
3
0.175
20
1
0
5.2
4.375
0.627
0.631
0
0.033
23
1
0
5
4.453
0.753
0.677
1
-0.089
38
1
0
4.6
3.598
0.579
0.584
0
0.023
2007
Lower Frankonia (LF)
Guttenberg 2 (GB2)
2002
49.4443
9.5154
2007
Gramschatz 1(GS1)
2002
49.5409
9.5842
2007
Höchberg (HB)
2002
49.4705
9.5201
2007
Irtenberg 3 (IB3)
2002
49.4308
9.5016
Rhineland-Palatine (RP)
2007
N
nucDNA microsatellites
π
NA
K
Ho
He
Pall
Fis
Reutholz (RT)
2007
49.737
9.861
22
3
0.66
9.8
7.601
0.696
0.809
2
0.164
Steinbach (SB)
2007
49.4215
9.4442
21
1
0
4.2
3.833
0.657
0.572
1
-0.126
Altrich 4 (AL4)
2007
49.9636
6.8726
14
2
0.14
5.4
4.728
0.643
0.662
2
0.065
Bitburg (BI)
2007
49.9726
6.4793
22
3
0.44
7.4
5.735
0.609
0.753
1
0.216
Duppach (DU)
2007
50.2685
6.5507
24
6
0.72
8.2
6.549
0.670
0.772
2
0.154
Longuich (LO)
2007
49.7916
6.7514
25
3
0.67
6.6
5.01
0.701
0.710
1
0.033
Orenhofen (OH)
2007
49.9115
6.6781
22
1
0
2.6
2.255
0.345
0.352
0
0.043
67
d)
mt msats
Lower Frankonia (LF)
M. bechsteinii
Rhineland
-Palatine
(RP)
nucDNA microsatellites
Colony Name
Colon
y size
n
Blutsee (BS)
15-19
30
3.5
2.22
Einsiedeln (ES)
20-24
10
1
Guttenberg 2 (GB2)
35-39
70
Gramschatz 1 (GS1)
40-44
Höchberg (HB)
Na
K
n
Ho
Pal
NA
K
He
30
11.1
5.07
0.783
0.802
1
0.057
1
10
7.1
4.63
0.85
0.769
0
-0.052
2
1.55
70
12.6
4.95
0.801
0.825
2
0.041
93
1.5
1.08
93
13
4.83
0.828
0.81
4
-0.014
20-24
57
4.5
2.45
57
11.6
4.86
0.806
0.799
1
0.008
Irtenberg 3 (IB3)
25-29
16
1.5
1.35
16
9.5
4.99
0.844
0.799
2
-0.024
Reutholz (RT)
40-44
17
1
1
17
9.3
4.95
0.79
0.794
2
0.029
Steinbach (SB)
20-24
34
1
1
34
9.3
4.57
0.825
0.792
1
-0.007
Bitburg (BI)
40-44
9
1
1
9
6.8
4.85
0.819
0.795
3
0.028
Duppach (DU)
45-49
20
1.5
1.3
20
9.1
4.86
0.856
0.804
1
-0.039
l
Fis
Additional host samples
For the two newly analysed M. myotis maternity colonies (Satigny, StUrsanne), wing tissue punches from 20 bats were obtained with a sterile
biopsy punch of the wing membrane Ø 2 mm (Worthington-Wilmer and
Barratt 1996). For these colonies samples were taken concurrently with mite
samples, for the other bat maternity colonies samples were collected and
analysed previously (Castella et al. 2001). Samples were preserved,
extracted, and sequenced for the second hypervariable domain (HVII) of the
mtDNA control region and 10 microsatellites (A13, B11, B22, C113, E24, F19,
G9, H19, H29 and G30) as described in Castella et al. (2001).
Amplification
Amplification of Cytochrome b (cytb) for S. myoti was performed according
to the protocol described in Bruyndonckx et al. (2009a) using the primer pair
C1-J-2183 and C1-J-2797mod (Simons et al. 1994). Microsatellite loci for
both species were developed from an enriched genomic library of S. myoti.
For S. myoti, eight microsatellite loci were established (SM7, SM11, SM13,
SM17, SM18, SM19, SM51 and SM55), where for S. bechsteini only five
microsatellites could be successfully amplified (SM11, SM16, SM17, SM18
68
and SM35) [GenBank: JF288840- JF288850] (van Schaik et al. 2011).
Amplification protocol and multiplex configuration were as developed in van
Schaik et al. (2011).
Mite mtDNA genetic analysis
Haplotype diversity (H) and nucleotide diversity (π) were calculated using
Arlequin 3.5 (Excoffier et al. 2005). A statistical parsimony network was
computed using TCS 1.21 (Clement et al. 2000). An analysis of molecular
variance, performed in Arlequin 3.5, was used to examine the genetic
structure and estimate pairwise Φ-statistics among all populations. To test
whether Φ-statistics were sensitive to distances between haplotypes, we
performed the same analyses based only on haplotypic frequencies. Isolation
by distance in mite populations was tested through the correlation between
matrices of ΦST / (1- ΦST) values and log transformed geographic distance
using Mantel tests (10,000 permutations) in FSTAT 2.9.3 (Goudet 2002).
Geographic distances were calculated as the shortest linear distance
connecting the populations.
Mite microsatellite genetic analysis
For both mite species we checked for the presence of null alleles, allelic
dropout, and stuttering using MicroChecker 2.2.5 (van Oosterhout et al.
2004). We also checked for deviations from Hardy-Weinberg equilibrium and
linkage disequilibrium using Genepop on the web 4.0.10 (Rousset 2008). We
calculated number of alleles per locus, allelic richness, observed and
expected heterozygosity, FIS, and pairwise FST-values using Fstat 2.9.3
(Goudet 2002). Number of private alleles was calculated using Genalex 6
(Peakall and Smouse 2006). To assess the level of genetic differentiation of
mites within and between bat colonies, single-level and hierarchical AMOVAs
were performed in Arlequin 3.5 (Excoffier and Lischer 2010). Isolation by
distance was measured by comparing matrices of FST / (1- FST) and log
transformed geographic distance using Mantel tests (10,000 permutations) in
FSTAT 2.9.3 (Goudet 2002). Sequential Bonferroni corrections were used to
69
compute the critical significance levels for simultaneous statistical tests.
Pairwise F’ST and G”ST-values were calculated using Genodive 2.0b23
(Meirmans and Van Tienderen 2004).
To investigate temporal differentiation of mite populations within a
colony, mites from five colonies of M. bechsteinii were sampled twice (2002
and 2007) and compared using pairwise FST-values and a Mantel test in
FSTAT 2.9.3 (Goudet 2002).
Parasite and host comparison
Summary statistics, diversity indices and population differentiation were
calculated for both hosts from existing datasets and the two newly analysed
M. myotis colonies as described above for mites. To compare parasite and
host genetic distances we used a partial Mantel test (10000 permutations)
correcting for geographic distance. For comparison of parasite and host
mitochondrial genetic distance, we used ΦST / (1- ΦST), and nuclear genetic
distance was compared using FST / (1 - FST) (Rousset 1997). A structure
analysis was performed in STRUCTURE 2.3.3 (Pritchard et al. 2000) for all
species using only the nucDNA microsatellites, applying an admixture model
without location as a prior. For S. myoti and M. myotis, we selected a range
of K from 1 to 10, and for S. bechsteini and M. bechsteinii from 1 to 15. Five
iterations per K were run with a burn-in and run length of 200 000 and 1 000
000 repetitions respectively. The most probable K was inferred using the ΔK
method of Evanno et al. (Evanno et al. 2005). Finally, one-tailed Spearman’s
rank correlation tests were performed using R (R Development Core Team
2015) to test for a positive correlation between mite diversity indices (H, π, K,
HO, HE) and host colony size. Further supporting information can be found in
the supplementary materials (Figure S3.2, Figure S3.3)
70
Results
M. myotis genetic data
The additional host mitochondrial analysis revealed that all HV2 haplotypes
corresponded to previously published haplotypes (Castella et al. 2000,
Castella et al. 2001, Ruedi et al. 2008; Table 3.2b). As expected, M. myotis
bat colonies showed high mtDNA differentiation (ΦST = 0.56, p < 0.001) and,
despite being statistically significant, low nucDNA (FST = 0.015, p < 0.001)
differentiation, indicating strong female philopatry and male-biased dispersal.
Descriptive statistics for all colonies can be found in Table 3.2b; results are
similar to those previously published by Castella et al. (2001).
S. myoti mtDNA
Among the 117 S. myoti mites sequenced, 49 different cytb haplotypes were
detected. The 697 aligned nucleotides consisted of 49 variables sites, of
which 18 were parsimony-informative. All haplotypes were deposited in
GenBank under accession numbers [KJ174107-KJ174155].
The haplotype network of S. myoti presents two star-like patterns with
some haplotypes in between (Figure 3.2a). Two haplotypes were the most
represented: h1 was present in 12% of individuals and h2 in 16.2%. Nine
haplotypes were shared between bat colonies and 40 were private
haplotypes. The number of haplotypes per population ranged from 7 to 14
(mean = 11, Table 3.2a), and nucleotide diversity from 0.536 to 0.874 (mean
= 0.726; Table 3.2a). S. myoti populations showed low pairwise differentiation
(mean ΦST = 0.012; Table 3.3a), and only Satigny was significantly
differentiated from two other populations (Meiringen, Agliè). A mantel test
revealed no significant correlation between genetic and geographic distances
(r2 = 0.02, p = 0.64).
The 697 bp fragment analysed in S. myoti represents an overlap of
72% with the 513 bp fragment of cytb previously analysed for S. bechsteini
(Figure 3.2b; Bruyndonckx et al. 2009b). A comparison of the most common
haplotypes of each mite species showed 93% similarity between species.
71
The overall structure of the haplotype networks does differ in that S.
bechsteini has far fewer haplotypes (23 vs. 49) despite a larger sample size,
and does not exhibit two distinct haplotype clusters. This indicates that the
pattern observed in S. myoti may be the result of admixture from the two
glacial refugia of its host.
Figure 3.2 Haplotype network for a) the 49 haplotypes determined by
sequencing the cytochrome b (cytb) of 119 Spinturnix myoti, and b) the 23
haplotypes for cytochrome b (cytb) for the 402 Spinturnix bechsteini included
in this comparison, as previously analysed in Bruyndonkx et al. (2009b). The
size of circles is proportional to the number of individuals sharing the same
haplotype. For each species, the list of which haplotypes were found in each
colony is given as an inset.
a)
h31
h27
h41
h14
h69
h24
h43
h32
h25
h37
h2
h44
h68
h45
h6
h13
h11
h29
h36
h4
h7
h42
h28
h72
h30
h35
h38
h46
h8
h5
h64
Colony Name
mtDNA
N haplotypes
Agliè
h1, h2, h4, h9, h10, h11, h27, h35, h36,
h43, h44, h45, h64, h65
Courtételle
h1, h2, h5, h6, h12, h24, h25, h29, h41,
h42, h66, h67, h70
Meiringen
h5, h7, h8, h9, h26, h31, h42
Perreux
h2, h5, h12, h13, h16, h24, h34, h39, h40,
h42
Satigny
h1, h2, h32, h47, h55, h64, h68, h72
St-Ursanne
h1, h2, h12, h13, h14, h17, h28, h30, h37,
h38, h46, h49, h50, h69
h67
h47
h55
h1
h50
h66
h9
h49
h12
1
<5
<10
>10
individuals per
haplotypes
h16
h39
h10
h70
h34
h40
h65
h17
72
h26
b)
Colony Name mtDNA
N haplotypes
24
12
41
48
23
22
4
21
46
47
42
1
43
2
44
28
3
<5
<10
<25
<50
>50
5
29
61
6
individuals per
haplotypes
26
66
BS
1, 2, 3
ES
1, 2, 4, 24, 66
GB2
1, 2, 3, 22
GS1
1, 2, 6, 21, 23, 42, 43, 61
HB
1, 2, 4, 5, 6, 44
IB3
1, 3
RT
3, 6, 25
SB
2
AL4
2, 41
BI
12, 28, 41
DU
2, 4, 28, 29, 46, 47,
LO
1, 2, 48
OH
41
Table 3.3 Pairwise ΦST (below the diagonal) and FST-values (above the
diagonal) for a) Spinturnix myoti, b) Myotis myotis c) Spinturnix bechsteini d)
Myotis bechsteinii. For M. bechsteinii values below the diagonal are pairwise
FST-values of mitochondrial microsatellites. Asterisks indicate significant
differentiation (p < 0.05).
a) S. myoti
Mite ΦST \ Fst
Agliè
Courtételle
Meiringen
Perreux
Satigny
St-Ursanne
-
0.014*
0.018*
0.014*
0.026*
0.016*
Courtételle
-0.02
-
0.004
0.013
0.004
0.002
Meiringen
0.05
0.02
-
0.01
0.015*
0.004
Perreux
-0.01
-0.04
0.01
-
0.015
0.009
Satigny
0.11*
0.04
0.08*
0.05
-
0.012*
0.03
-0.01
0.03
-0.01
0.02
-
Agliè
St-Ursanne
b) M. myotis
Agliè
Courtételle
Meiringen
Perreux
Satigny
St-Ursanne
Agliè
-
0.043*
0.057*
0.034*
0.035*
0.036*
Courtételle
0.72*
-
0.016
-0.001
0.009
-0.003
Meiringen
0.71*
0.04
-
0.0004
0.007
0.014
Perreux
0.70*
0.15*
0.12*
-
0.006
0.003
Satigny
0.37*
0.21*
0.20*
0.18*
-
0.002
St-Ursanne
0.73*
0
0.05
0.18*
0.22*
-
Host ΦST \ FST
73
0.54*
0.73*
0.22*
0.88*
0.90*
0.62*
0.87*
0.83*
0.72*
ES
0.217*
0.51*
0.16*
0.75*
0.83*
0.33*
0.72*
0.61*
0.47*
GB2
0.255*
0.174*
0.36*
0.87*
0.91*
0.53*
0.110
0.77*
0.64*
GS1
0.202*
0.084*
0.131*
0.60*
0.69*
0.27*
0.55*
0.50*
0.40*
LF
HB
0.316*
0.184*
0.283*
0.159*
1.00*
0.66*
1.00*
0.94*
0.77*
IB3
0.126*
0.275*
0.307*
0.226*
0.348*
0.53*
1.00*
0.96*
0.83*
RT
0.153*
0.068*
0.116*
0.052*
0.171*
0.158*
0.66*
0.56*
0.45*
SB
0.351*
0.221*
0.315*
0.213*
0.299*
0.372*
0.233*
0.94*
0.78*
AL4
0.260*
0.130*
0.180*
0.133*
0.228*
0.322*
0.121*
0.250*
0.60*
BI
0.243*
0.123*
0.186*
0.130*
0.210*
0.284*
0.149*
0.244*
0.214*
-
RP
DU
0.185*
0.145*
0.186*
0.079*
0.204*
0.213*
0.113*
0.267*
0.173*
0.138*
LO
0.275*
0.177*
0.215*
0.098*
0.248*
0.301*
0.135*
0.239*
0.163*
0.181*
OH
0.406*
0.365*
0.421*
0.312*
0.422*
0.458*
0.309*
0.475*
0.314*
0.379*
Duppach
Longuich
0.59*
0.50*
0.29*
0.26*
0.47*
0.25*
0.25*
0.16*
0.62*
0.64*
0.70*
0.72*
0.31*
0.34*
0.61*
0.42*
0.52*
0.54*
0.41*
0.44*
0.29*
0.158*
-
0.382*
0.243*
Orenhofen
0.88*
0.76*
0.87*
0.62*
1.00*
1.00*
0.67*
1.00*
0.030
0.72*
0.63*
0.65*
-
Mite ΦST \ Fst
Blutsee
Einsiedeln
Guttenberg 2
Gramschatz 1
Höchberg
Irtenberg 3
Reutholz
Steinbach
Altrich 4
Bitburg
BS
d) M. bechsteinii
LF
Host mtFST \ FST
LF
Blutsee
BS
ES
GB2
GS1
RP
HB
IB3
RT
SB
BI
DU
-
0.023
0.004
0.005
0.008
0.008
0.011
0.006
0.072
0.016
Einsiedeln
0.333*
-
0.036*
0.032*
0.044
0.032
0.03
0.036*
0.5
0.356*
Guttenberg 2
0.331*
0.245*
-
0.007
0.010*
0.002
0.010*
0.006
0.448
0.445*
Gramschatz 1
0.364*
-0.001
0.252*
-
0.018*
0.005
0.004
0.004
0.488
0.351
Höchberg
0.242*
0.075
0.142*
0.087*
-
0.009
0.028*
0.01
0.375
0.341*
Irtenberg 3
0.139*
0.260*
0.239*
0.259*
0.148*
-
0.014
0.005
0.318
0.28
0.02
0.388*
0.373*
0.412*
0.304*
0.201*
-
0.014
0.072
0.211
0.146*
0.333*
0.274*
0.320*
0.195*
0.004
0.205*
-
0.333
0.312
0.011
0.028
0.004*
0.009*
0.017*
0.009*
0.015
0.012*
-
0.013
0.167*
0.042*
0.015*
0.013*
0.030*
0.022*
0.010*
0.019*
0.311*
-
Reutholz
Steinbach
RP
RP
LF
c) S. bechsteini
Bitburg
Duppach
Mite nucDNA
No linkage disequilibrium was found between loci in either mite species. All
six S. myoti populations showed highly significant deviations from HardyWeinberg equilibrium (p < 0.0001), as a result of heterozygote deficiency (FIS:
0.191-0.398). Similarly, S. bechsteini populations were not in HardyWeinberg equilibrium as a result of heterozygote deficiency (p < 0.0001), with
five exceptions (ES-2007, GB2-2007, HB-2007, IB3-2002, OH). Microchecker
indicated the possible presence of null alleles in all microsatellite loci for both
74
species, but found no evidence for allelic dropout or stuttering. As discussed
in the original marker description (van Schaik et al. 2011), we believe these
heterozygote deficiencies are not the result of null alleles for several reasons.
First, all markers show similarly heterozygote deviations indicating that null
alleles would have to be present in all markers across both species. Second,
mite populations may exhibit substantial inbreeding, as often observed in
obligate parasite species (e.g. Leo et al. 2005), as populations are often small
and largely isolated. Therefore, although the possible presence of null alleles
cannot be completely dismissed, we believe the markers to be biologically
informative.
S. myoti nucDNA
Overall genetic variation in S. myoti was very high with total number of alleles
per locus ranging from 13.13 to 18.13, and allelic richness ranging from 12.74
to 16.75 (Table 3.2a). A large proportion of these alleles were specific to one
bat colony, with the number of private alleles ranging from 6 in Meiringen, to
22 in Aglié (mean = 13.17, Table 3.2a). Between bat colonies, S. myoti
pairwise FST-values (Table 3.3a) showed very little differentiation (0.002 –
0.026), and significant pairwise differentiation was only found between Aglié
(separated from the other colonies by the alps) and all other colonies, and
between Satigny and two other colonies (Meiringen and St. Ursanne).
Pairwise F’ST- and G”ST-values were slightly higher (−0.003-0.241 and 0.0310.264 respectively) but did not differ strongly from the pairwise FST-values
(data not shown). An AMOVA analysis found the vast majority of variation
within mite populations (98.9%; Table 3.4). Unlike in the mtDNA sequence, a
significant, but very weak, correlation between genetic and geographic
distances (β = 0.005, r2 = 0.362, p = 0.027; Table S3.4) was found, but S.
myoti nuclear and mitochondrial genetic distance were nevertheless also
significantly correlated (r2 = 0.26, p = 0.049).
75
Table 3.4 Analysis of molecular variance for a) Spinturnix myoti and b)
Spinturnix bechsteini. For S. bechsteini a hierarchical AMOVA including
region as an additional factor is shown, with the values of a non-hierarchical
AMOVA given in parentheses. Asterisks indicate significant contributions.
Source of variation
Spinturnix myoti
Among colonies
Among individuals,
within colonies
Within individuals
Spinturnix bechsteini
Among regions
Among colonies,
within regions
Among individuals,
within colonies
Within individuals
d.f.
Sum of squares
Variance
components
Percentage
of variation
5
111
30.96
512.39
0.04
0.96
1.1*
26.0*
117
315.5
2.69
72.9
1
11 (12)
40.39
269.92 (309.64)
0.05
0.5 (0.52)
2.4*
22.5* (23.3*)
289
494.33 (518.55)
0.05 (0.08)
2.2* (3.4*)
302
487 (496)
1.61 (1.64)
72.9 (73.3)
S. bechsteini nucDNA
Overall genetic variation in S. bechsteini was lower than that of S. myoti with
2.6 to 12.0 alleles per locus (mean = 6.83), and a mean allelic richness of
5.52 (Table 3.2c). The number of private alleles was also lower (0–10, mean =
1.79). However, it is important to note that the number of S. bechsteini
populations was also much higher than in S. myoti thereby increasing the
chance that we had sampled rare but widespread alleles at multiple
locations.
Pairwise FST-values among populations were much higher in S.
bechsteini than in S. myoti (0.052-0.475, mean = 0.228; Table 3.3c) as well as
those of its host (Table 3.3d), and all S. bechsteini populations were
significantly differentiated from one another. Pairwise F’ST- and G”ST-values
showed even more exacerbated differentiation between mite populations
(0.333-0.929 and 0.334-0.93 respectively), but again closely followed the
pattern seen in the pairwise FST-values (data not shown). An AMOVA revealed
76
very similar patterns as in S. myoti, with the majority of the sampled variation
found within S. bechsteini populations (73.3%; Table 3.4). Nevertheless,
there was also significant differentiation among mite populations (p < 0.001).
Between the spatially distant regions of Lower Franconia (LF) and
Rhineland-Palatine (RP), 45% of the alleles were shared. Pairwise FST-values
of S. bechsteini between bat colonies from differing regions were very similar
to those between bat colonies from the same region (within LF = 0.215,
within RP = 0.235, between regions = 0.236). Indeed, a hierarchical AMOVA
including regions found only 2.4% of the variation between regions, with the
vast majority of variation found within S. bechsteini populations (72.9%) and
among bat colonies within regions (22.5%; Table 3.4). No correlation was
found between geographic and genetic distance (r2 = 0.027, p = 0.39; Table
S3.4).
A clear temporal genetic differentiation (FST = 0.085-0.254, mean =
0.147) in S. bechsteini was evident in all five bat colonies sampled for mites
in 2002 and 2007. This level of differentiation is comparable to the average
differentiation seen between all sampling events between different bat
colonies within one year (0.146 and 0.238 for 2002 and 2007 respectively).
Pairwise differentiation between the five colonies was not correlated between
2002 and 2007 (Spearman’s rank correlation: r = −0.360, p = 0.238),
indicating no stable sub-structuring within the region. Within individual bat
colonies, only 26.0-41.2% of alleles were conserved across sampling events,
also indicating strong temporal turnover of the S. bechsteini populations.
In all, our results for nucDNA closely mirror those of the previously
analysed mtDNA (Bruyndonckx et al. 2009b). It is therefore unsurprising that
a significant correlation between nuclear and mitochondrial FST-values can be
seen (r2 = 0.34, p = 0.001).
Comparison with host
We applied mantel test analysis to examine the correlation between the
genetic distances of hosts and parasites (Table 3.5). In neither of the hostparasite pairs were any of the pairwise ΦST or FST-values significantly
77
correlated between species after correcting for geographic distance. A plot
of the pairwise genetic distance (FST/1 – FST) between populations of hosts
and parasites reveals that this lack of correlation is due to large variation in
parasite pairwise differentiation at low levels of host genetic differentiation in
both species pairs (Figure 3.3a,b). Additionally, in M. bechsteinii and S.
bechsteini, no effect of geographic distance on host and parasite pairwise
genetic differentiation can be observed (Figure 3.3b).
Table 3.5 Correlations (partial mantel) between host and parasite genetic
differentiation corrected for geographic distance. For each pair the slope (β)
and the variance explained by the model (R2) are given. No correlations were
significant after correcting for distance.
Mite mtDNA - Host mtDNA
Mite mtDNA - Host nDNA
Mite nDNA - Host mtDNA
Mite nDNA - Host nDNA
S. myoti/M. myotis
β
R2
-2.59
0.139
-0.17
0.125
0.003
0.022
0.343
0.013
78
S. bechsteini/M. bechsteinii
β
R2
-0.329
0.115
-3.33
0.025
-0.101
0.097
-0.04
0.08
Figure 3.3 Pairwise genetic distance (FST/1 – FST) estimates between
populations of hosts and parasites for a) Myotis myotis and Spinturnix myoti,
and b) Myotis bechsteinii and Spinturnix bechsteini. For the latter pair, the
symbols indicate whether colonies were within the same region (closed
circles) or from different regions (open circles).
a)
0,030
S. myoti Fst / (1-Fst)
0,025
0,020
0,015
0,010
0,005
0,000
-0,01
0,00
0,01
0,02
0,03
0,04
0,05
0,06
0,07
M. myotis Fst / (1-Fst)
b)
0,7
S. bechsteini Fst / (1-Fst)
0,6
0,5
0,4
0,3
0,2
0,1
0,0
0,00
0,01
0,02
0,03
M. bechsteinii Fst / (1-Fst)
79
0,04
0,05
A STRUCTURE analysis revealed several interesting differences in both pairs
of host and parasite (Figure 3.4). In S. myoti no clear sub-structuring was
observed (Figure 3.4a), whereas in its host, M. myotis, the southernmost
colony (Aglié) appears to cluster separately (Figure 3.4b). This is concordant
with phylogeographic analyses, which have indicated that populations of M.
myotis south of the Alps originate from a different glacial refugium (Ruedi et
al. 2008), although this barrier is evidently not present for its mites. In M.
bechsteinii and S. bechsteini clear sub-structuring according to host colony
can be observed in the parasite (Figure 3.4c), while host colonies show no
evidence for population sub-structuring (Figure 3.4d) as is expected with an
outbreeding mating system such as swarming (Kerth et al. 2003b).
Finally, since host colony size between species appears to have a
large effect on the population genetic structure of its parasites, we used
Spearman’s rank correlation tests to investigate whether variation in host
colony size within each host was correlated with parasite diversity indices
(Table 3.6). Host colony size was not significant correlated with any of the
parasite genetic indices after Bonferroni correction, although there was a
positive trend between parasite allelic richness and host colony size in both
species pairs.
Table 3.6 Spearman rank correlations between host colony size and parasite
genetic parameters. For each pair the correlation coefficient (ρ) and the Pvalue are given. No correlations were significant after Bonferroni correction (p
< 0.01).
Parasite trait
Number of haplotypes
Nucleotide diversity
Expected heterozygosity
Observed heterozygosity
Allelic richness
ρ
0.75
-0.26
0.64
0.41
0.90
M. myotis
p
0.09
0.61
0.17
0.43
0.015
80
M. bechsteini
ρ
p
0.28
0.26
0.29
0.24
0.49
0.12
0.16
0.64
0.45
0.06
Figure 3.4 STRUCTURE results for a) Spinturnix myoti, b) Myotis myotis c)
Spinturnix bechsteini d) Myotis bechsteinii. Colony names correspond to the
abbreviations given in Tab. 2. Results are shown for the number of
subpopulations with the best ΔK and highest log likelihood (S. bechsteini K =
7, M. myotis K =3). In S. myoti and M. bechsteinii no discernable population
sub-structuring was found, with strongest support being found for K = 1. For
these species K = 2 is shown to illustrate the lack of structuring when
multiple subpopulations are assumed.
a) S. myoti (K=2)
Agl
Cou
Mei
Cou
Mei
Per
Sat
StU
Sat
StU
b) M. myotis (K=3)
Agl
Per
c) S. bechsteini (K=7)
BS
ES GB2 GS1
HB
IB3
RT
SB
BI
DU
d) M. bechsteinii (K=2)
BS ES
GB2
GS1
81
HB
IB3
RT SB BI DU
Discussion
The population genetic structure of parasite populations is strongly
influenced by both its own life history as well as the life history and social
system of its host. Here, we compared the population genetic structure of
both host and parasite in two closely related systems where parasite life
histories are nearly identical, but hosts differ substantially in social system.
We find that the population genetic structure of the two mite species differs
strongly as a result of the social system of their hosts, which is in agreement
with our predictions based on the differences in social system of their hosts.
Parasite population genetic structure
In S. myoti, overall mtDNA haplotype diversity (49 haplotypes in 120
individuals) and nuclear genetic diversity were high (mean 16.69 alleles per
locus in only 20 individuals per colony). This diversity is higher than that seen
in its host, M. myotis, and may be the result of admixture between parasites
originating from several glacial refugia or species, coupled with sufficiently
large host maternity colony size to prevent strong bottlenecks in the mites.
Between M. myotis colonies, we found very low genetic differentiation and no
evidence for population substructuring in the mites. Indeed the vast majority
of genetic diversity of S. myoti occurred within bat maternity colonies
suggesting a large amount of parasite exchange between hosts originating
from different maternity colonies outside of the maternity period, resulting in
a highly diverse and panmictic population. These results are concordant with
the observation of other empirical studies on host-parasite co-variation (e.g.
McCoy et al. 2005), that have found levels of parasite gene flow much higher
than originally predicted (Price 1977). Notably, the extensive mixing of
parasites outside of the breeding period has also been found in ectoparasitic
lice of birds, where the genetic structure of parasites was largely shaped by
the dispersal of lice at communal wintering sites (Gómez-Díaz et al. 2007).
The population genetic structure of S. bechsteini contrasted sharply
with that of S. myoti. Overall nuclear genetic diversity was lower, but still
82
higher than that of its host, M. bechsteinii. All S. bechsteini populations were
significantly differentiated from one another, and it was possible to assign
mites to subpopulations according to their host colony of origin. This is likely
to be due to strong genetic drift within host colonies as a result of their small
maternity colony size as well as their tendency to hibernate without body
contact with conspecifics. Both factors combine to drastically limit the
number mites present within a host colony when the colonies reform in
spring. Nevertheless, S. bechsteini populations within colonies were
temporally unstable (between sampling years mean FST = 0.147 for nucDNA)
suggesting that gene flow between colonies is still substantial. In agreement
with this finding, Bruyndonckx et al. (2009b) observed a large turnover in
haplotypes between sampling years. Despite this strong differentiation
among colonies and years at a local scale, differentiation of S. bechsteini
between colonies from spatially distinct regions was not higher than within
regions, suggesting that overall genetic diversity of the S. bechsteini
population at a regional level is stable. A comparable pattern of strong
population genetic substructure and high levels of genetic drift is also seen in
the parasites of other host taxa with similarly closed societies as in
Bechstein’s bats, e.g. in the ectoparasitic chewing lice of pocket gophers
(rev. in Hafner et al. 2003).
Relationship between parasite and host population structure
No correlation was found between host and parasite genetic differentiation in
either of the species. Additionally, within each of the two host species,
colony size did not significantly correlate with any of the parasite genetic
indices, although a trend of increasing allelic richness with increased colony
size was present in both species pairs.
Influence of host social system
Our results indicate that differences between hosts in colony size, mating
system, and in particular the degree of social interaction outside of the
summer maternity roosts have a large effect on parasite genetic structure. M.
83
bechsteinii has a remarkably closed social system, with very strong natal
philopatry and vastly limited social interaction outside of the summer
maternity season (Kerth et al. 2000, Kerth and van Schaik 2012). Previously,
it has also been shown that M. bechsteinii actively reduces the intensity of
another ectoparasite species that deposits its larval stages in the roosts via
roost-switching behaviour (bat flies; Reckardt and Kerth 2007). It thus seems
possible that the much more closed social system of M. bechsteinii, as
compared to M. myotis, has also evolved to restrict the infestation of other
parasites, such as wing mites, that depend on body contact between hosts.
In M. myotis, our data provide no evidence for such anti-parasite
behaviour. While this may be because there is an insufficient number of
alternative roosts to increase population subdivision into smaller colonies,
social interaction during other periods such as mating and hibernation is also
extensive, thereby further permitting parasite exchange.
Several other factors probably have also influenced the genetic
structure of both mite species, but these factors do not detract from the
influence of host social structure. For example, the phylogeographic history
of both mite species is quite different, and may explain the differences in the
number of haplotypes and nucleotide diversity found in the mtDNA
sequences (compare Figure 3.2a and Figure 3.2b), and likely also the
diversity
found
in
the
nucDNA
microsatellites.
Nevertheless,
these
macrogeographic differences between the two mite species cannot explain
the differences seen in population genetic structure on a microgeographic
scale. For example, almost all M. bechsteinii colonies harboured S.
bechsteini populations that were monotypic or had only two haplotypes as a
result of the strong winter bottleneck. This contrasts sharply with the
sampled populations of S. myoti, in which we found between seven and
fourteen haplotypes in only twenty samples per populations. Thus, although
the history of both species has certainly influenced the overall nucleotide and
genetic diversity observed, the differences in genetic structure observed on a
population level are still primarily the result of differences in host social
system. Another major difference between the two mite species is the
84
potential use of secondary host, with identical social system, in S. myoti.
Here too, the social systems of the hosts play a role, as the closed social
system of M. bechsteinii minimizes contact between conspecifics and
virtually eliminates contact with other species. Therefore, while the possible
use of additional hosts in S. myoti may increase its effective population size
and potentially also increase its dispersal opportunities, these factors can
also be (indirectly) attributed to the differing social systems of the hosts.
Consequences for parasite evolutionary potential
The observed differences in population genetic structure of the two mite
species should strongly influence their evolutionary potential. In S.
bechsteini, the influence of genetic drift strongly limits the likelihood that any
local adaptation to its host maternity colony is able to persist and spread. S.
myoti, in contrast, has a much more stable population genetic structure as
well as a higher rate of gene flow relative to its host, and may therefore be
able to locally adapt to its major host, M. myotis.
The observed difference in population genetic structure between the
mites analysed here also have broader consequences for general
investigations of parasites in relation to host social system. For example, they
suggest that parasite species with very similar life histories may have vastly
different evolutionary potentials depending on not only the life history but
also the social system of their hosts. As a result, we conclude that host social
system can be effective beyond reducing the overall chance of parasite
infection, by potentially also playing a role in limiting the evolutionary
potential of, often unavoidable, parasites. Therefore, comparative and
theoretical studies of host-parasite interactions that investigate the role of
host population size and community modularity (e.g. Caillaud et al. 2013) will
be critical not only in understanding parasite transmission, but also in
investigating the evolutionary potential of established parasite species.
85
Conclusions
In conclusion, our results suggest that host social system can strongly
influence parasite population genetic structure. Most notably, host maternity
colony size appears to strongly affect the genetic drift experienced by the
different parasite species and the social organization of the host outside of
the maternity period affects the opportunities for parasite exchange between
individuals from remote maternity colonies. We conclude that such
differences in host social system have consequences for both the direct
costs of parasites as well as the general threat of disease transmission.
Therefore, the concurrent genetic analysis of host and parasite allows for
inferences about the movements and social contacts of host species, as well
as broader conclusions regarding host-parasite dynamics.
Data availability
The haplotype sequences for S. myoti have been deposited in GenBank
under ascension numbers [KJ174107-KJ174155].
Acknowledgements
We would like to thank Michel Blanc, Pascal Roduit, Julien Oppliger, Peter
Zingg, Elena Patriarca, Paolo Debernardi, Karsten Reckardt, Isabelle Henry,
Markus Melber, Daniela Fleischmann, Manfred Weishaar, and many others
for their help during collection of mites. We are grateful to Thomas Broquet,
who provided useful comments on the analyses and interpretations, and to
Nathalie Tzaud, Sébastien Nusslé, Nelly DiMarco, Philippe Busso and Jari
Garbely
for
laboratory
facilities.
Funding
was
provided
by
grant
31003A_138187 from the Swiss National Science Foundation and by la
Fondation Herbette to PC, as well as the Julius Klaus Foundation and the
German Science Foundation (DFG KE 746/3-1 and KE 746/4-1) to GK. JvS is
funded by the Volkswagen Foundation, and the International Max Planck
Research School for Organismal Biology.
86
Supplementary materials
Table S3.1 Overview of the markers were used in this study and where a
complete description of their amplification conditions can be found.
Primers/markers
Data originally
Markers used
described in
published in
mtDNA
cytB (697bp)
Simons et al., 1994
this study
nucDNA
SM7, SM11, SM13, SM17,
SM18, SM19, SM51, SM55
van Schaik et al., 2011
this study
mtDNA
HV-2 (307bp)
Castella et al., 2001
Castella et al., 2001*
nucDNA
A13, B11, B22, C113, E24,
F19, G9, H19, H29, G30
Castella & Ruedi 2000
Castella et al., 2001*
mtDNA
cytB (513bp)
Bruyndonckx et al., 2009
Bruyndonckx et al., 2009
nucDNA
SM11, SM16, SM17, SM18,
SM35
van Schaik et al., 2011
this study
mtDNA
mtMicrosatellites (AT1, AT2)
Kerth et al., 2000
Kerth et al., 2000
nucDNA
B15, B22, B23, G30, P5, P8,
Kerth et al., 2002
Kerth et al., 2003
Species
S. myoti
M. myotis
S. bechsteini
M. bechsteinii
P20, paur3
* = plus two new colonies in this study
Figure S3.2 (next page) The program Structure was used to evaluate
population sub-structuring for each species. Below are the a) Log-likelihood
values for all tested K’s for each species, as well as b) the concordant DeltaK
values (as described in Evanno et al., 2005) used to determine the most likely
number of sub-populations.
87
a) Likelihood of all K values run (calculated in Structure Harvester):
S. myoti
S. bechsteini
M. myotis
M. bechsteinii
b) DeltaK values for all species (calculated using Structure Harvester):
S. myoti
M. myotis
S. bechsteini
M. bechsteinii
88
Figure S3.3 As can be seen in the graphs for S. bechsteini both K=2 and K=7
resulted in high DeltaK values, and K=10 (the actual number of populations
sampled) resulted in the highest overall Log-likelihood. Below, outputs for
K=2 and K=10 are provided (K=7 can be found in Figure 3.4).
S. bechsteini (K=2)
BS
ES GB2
GS1
HB
IB3
RT
SB
BI
DU
S. bechsteini (K=10)
BS
ES GB2
GS1
HB
IB3
RT
SB
BI
DU
Figure S3.4 Statistical correlations and graphs correlating nuclear genetic
distance and geographic distance (using FST / (1- FST) and Ln(distance)). For
both S. myoti and M. myotis genetic distance is significantly (positively)
correlated with geographic distance. No such relationship is found in S.
bechsteini or M. bechsteinii, nor is it found when only samples within LF are
analysed.
β
R2
S. myoti
0.005
0.362*
M. myotis
0.015
0.342*
S. bechsteini
-0.012
0.027
M. bechsteinii
0.001
0.026
89
S. myoti
M. myotis
S. bechsteini
M. bechsteinii
S. bechsteini (without RP)
M. bechsteinii (without RP)
90
91
Chapter 4
Host and parasite life history interplay to yield
divergent population genetic structures in two
ectoparasites living on the same bat species
J. van Schaik, D. Dekeukeleire, G. Kerth
Molecular Ecology, 24:10, 2015
Myotis bechsteinii at a swarming site, © René Janssen
92
93
Abstract
Host-parasite interactions are ubiquitous in nature. However, how parasite
population genetic structure is shaped by the interaction between host and
parasite life history remains understudied. Studies comparing multiple
parasites infecting a single host can be used to investigate how different
parasite life history traits interplay with host behaviour and life history. In this
study, we used 10 newly developed microsatellite loci to investigate the
genetic structure of the parasitic bat fly (Basilia nana). Its host, the
Bechstein’s bat (Myotis bechsteinii), has a social system and roosting
behaviour that restrict opportunities for parasite transmission. We compared
fly genetic structure to that of the host and another parasite, the wing-mite,
Spinturnix bechsteini. We found little spatial or temporal genetic structure in
B. nana, suggesting a large, stable population with frequent genetic
exchange between fly populations from different bat colonies. This contrasts
sharply with the genetic structure of the wing-mite, which is highly substructured between the same bat colonies as well as temporally unstable.
Our results suggest that although host and parasite life history interact to
yield similar transmission patterns in both parasite species, the level of gene
flow and eventual spatio-temporal genetic stability is differentially affected.
This can be explained by the differences in generation time and winter
survival between the flies and wing-mites. Our study thus exemplifies that the
population genetic structure of parasites on a single host can vary strongly
as a result of how their individual life history characteristics interact with host
behaviour and life history traits.
94
Introduction
In host-parasite interactions, initial infection risk, subsequent transmission,
and ultimately parasite population genetic structure are determined by how
host and parasite life histories interact (Nadler 1995, Barrett et al. 2008). As a
result, different parasites can respond contrastingly to the same life history
characteristics or behavioural adaptations of a host (Côté and Poulin 1995).
For example, host behaviours such as allogrooming may reduce infection
intensity of certain ectoparasites while increasing the transmission of other
directly transmitted parasites, such as nematodes (MacIntosh et al. 2012).
Elucidating how differing life history factors interact is important for
understanding host-parasite dynamics at both the local (population genetic
structure) and broad (coevolutionary patterns) scales (Criscione 2008), as
well as their effect on disease risk and spread (Cross et al. 2005, Nunn 2012).
In one of the most detailed investigations of parasite genetic structure
at a local scale, McCoy and colleagues (2003) found genetic differentiation in
a tick within host (bird) breeding cliffs, principally influenced by the spatial
organization of hosts as well as cliff topography. In the same system, local
adaptation of ticks was found between breeding cliffs located between 50
and 800 meters from one another (McCoy et al. 2002), demonstrating that
genetic differentiation found on such small spatial scales can indeed
influence host-parasite coadaptation. In another study of seabirds and their
parasites, three specialized louse species of three shearwater (sub-)species
showed no genetic differentiation across hosts, while a generalist flea
species showed strong differentiation across the same host (sub-)species
(Gómez-Díaz et al. 2007). These differences can be attributed to the fact that
lice remain on the host year-round thereby facilitating horizontal transmission
at communal wintering grounds of the host, whereas fleas only infect hosts
during the nesting phase and therefore horizontal transmission is restricted to
host nests in the immediate vicinity (Gómez-Díaz et al. 2007). Unfortunately,
such investigations of parasite population genetic structure are rare,
especially for parasites of hosts with social and/or behavioural anti-parasite
95
adaptations. Thus how differences in parasite life history interact with host
life history and social system to shape parasite population genetic structure
within host meta-populations remains understudied.
The Bechstein’s bat (Myotis bechsteinii) provides an interesting model
to investigate how host and parasite life history interact to shape parasite
local genetic structure. During the summer, this forest-living bat forms allfemale maternity colonies in tree cavities and bat boxes comprised of 10 to
50 adults and their dependent offspring (Kerth et al. 2002a). Female bats are
highly philopatric, and immigration into these colonies is extremely rare
(Kerth et al. 2002b). Male bats roost solitarily throughout the summer, also
switching roosts less frequently, but may occasionally utilize roosts that have
been previously occupied by female colonies (Kerth and Morf 2004). During
late summer and autumn both sexes meet at underground sites, that are
located within approximately 50 km of the summer habitat, in a lek-like
mating system known as swarming (Kerth et al. 2003b). During winter,
Bechstein’s bats hibernate, often singly, in deep crevices within these
underground sites (Baagøe 2001). As expected in such highly philopatric
female societies with outbreeding males, there is strong mitochondrial
differentiation (Kerth et al. 2000), but very little nuclear genetic structure
across colonies (Kerth et al. 2002a, Kerth et al. 2003b).
In addition to their closed societies, Bechstein’s bats show
behavioural adaptations aimed at reducing parasite pressure (Kerth and van
Schaik 2012). Most notably, Bechstein’s bats use frequent roost switching
behaviour and prefer novel roost sites (Reckardt and Kerth 2006, 2007) to
avoid the offspring of parasites that must temporarily leave the host to
reproduce. Female maternity colonies switch day-roost (bat boxes and tree
cavities) every 2-3 days on average, using up to 50 different day-roosts
throughout the summer (Reckardt and Kerth 2007). Additionally, maternity
colonies do not overlap in roost home-range with one another (Kerth et al.
2000). Moreover, individual bats groom themselves extensively, and
allogrooming within maternity colonies is common (Kerth et al. 2003a). Taken
together, the small colony size, severely limited inter-colony bat contact, and
96
behavioural adaptations strongly reduce the opportunities for the horizontal
transmission (defined here as the transmission of parasites from members of
one maternity colony to members of another colony or to solitary males) of
directly transmitted parasites (van Schaik et al. 2014). Despite these
adaptations, the Bechstein’s bat is commonly parasitized by two
ectoparasites: the bat fly (Basilia nana) and the wing-mite (Spinturnix
bechsteini). Both are highly specialized obligate ectoparasites primarily
infecting the Bechstein’s bat.
Bat
flies
(Diptera:
Nycteribiidae)
are
wingless,
obligatory,
hematophagous ectoparasites that inhabit the pelage of bats (Dick and
Patterson 2006, 2007). The bat fly, B. nana, cannot disperse independently
from its primary host, the Bechstein’s bat, and female flies only briefly leave
their host approximately every 5-10 days to deposit a single terminal (3rdinstar) larva on the wall of the roost (Reckardt and Kerth 2006). These larvae
immediately develop into a puparium, and can emerge as fully developed
imagoes in ca. 3 weeks (Dick and Patterson 2007). Developed imagoes wait
inside the puparia and only emerge in the presence of a potential host
(Reckardt and Kerth 2006). Due to the frequent roost switching observed in
the host, only 30% of bat fly B. nana puparia deposited in day-roosts
eventually emerge in the presence of a host (Reckardt and Kerth 2006).
Despite this temporary decoupling with the host, B. nana is remarkably
specific to Bechstein’s bats and is only sporadically found on other bat
species that roost in the same tree cavities, such as Natterer’s bat (Myotis
nattereri) and brown long-eared bat (Plecotus auritus) in our study area.
Critically, male Bechstein’s bats are also frequently infected by B. nana, and
may act as vectors for horizontal transmission by consecutively occupying
roosts also used by female Bechstein’s bat maternity colonies (Kerth and
Morf 2004, Schöner et al. 2010). Previous studies have shown that there are
at least two fly generations per year, the first coming directly after bats return
from hibernation and a second in late summer (Reckardt and Kerth 2006).
Bat fly puparia do not survive frost, and thus do not survive the winter in the
97
bat boxes (Reckardt and Kerth 2006); instead they survive the winter as
imagoes on the host.
Whether bat flies impose significant direct fitness costs on their hosts
is largely unknown, but costs are generally considered minor (Zahn and Rupp
2004, Dick and Patterson 2006). However, bat flies can serve as vectors for
more costly endoparasites, including Haemosporidia and Bartonella (Megali
et al. 2011, Billeter et al. 2012). Regardless of the exact cost of infection,
Bechstein’s bats employ several behaviours that limit B. nana transmission
and infection intensity (Reckardt and Kerth 2007), suggesting that infection
may be costly in some way. It is currently unknown to what extent these
behaviours also affect fly population genetic structure, and thus fly
evolutionary potential.
In several previous analyses, the effects of host life history and social
system on parasite local genetic structure of another common ectoparasite
of the Bechstein’s bat, the wing-mite Spinturnix bechsteini, have been
investigated (Bruyndonckx et al. 2009b, van Schaik et al. 2014). S. bechsteini
is highly host-specific, lives exclusively on the wing membranes and cannot
survive off of the host (Giorgi et al. 2004). Thus, unlike bat flies, wing-mites
are not affected by the roost-switching behaviour of their host. Wing-mites
reproduce sexually and are oviviparous, with female mites giving birth to
single adult-like protonymphs which rapidly develop into adults (Rudnick
1960). While able to disperse freely within bat maternity colonies,
transmission of wing-mites between colonies only occurs during bat mating
and hibernation periods (Bruyndonckx et al. 2009b). Male Bechstein’s bats
are only very rarely infected with wing-mites during summer (Reckardt and
Kerth 2009), presumably as they are groomed away. Wing-mites reproduce
faster than flies, and thus infection intensity within bat maternity colonies in
early autumn is much higher than that of flies (mean intensity per bat 3.61
and 0.56 respectively; Reckardt and Kerth 2009). Wing-mite populations
subsequently decrease sharply during late autumn and throughout
hibernation (Lourenço and Palmeirim 2008). Using mtDNA sequence data
(Bruyndonckx et al. 2009b) and nDNA microsatellites (van Schaik et al. 2014),
98
S. bechsteini was found to have high genetic differentiation between bat
colonies and a high temporal turnover within bat colonies, suggesting
frequent local extinction and re-colonisation.
Here we use 10 newly developed microsatellite loci to investigate the
population genetic structure of the ectoparasitic bat fly B. nana on a local
scale. Subsequently, we compare the population genetic structure of B. nana
with that of the previously investigated wing-mite S. bechsteini in the same
area, to investigate how differences in parasite life history interplay with host
social system and life history to shape population genetic structure. We
predict that fly genetic structure will be weaker than that of the previously
investigated wing-mite as a result of their longer generation time and higher
winter survival (parasite life history traits), as well as their increased
persistence and opportunity for horizontal transmission via male bats (host
life history). For the same reasons, we hypothesize that although roostswitching, allogrooming and limited inter-colony contact reduce fly infection
intensity, they do not cause strong temporal instability in population genetic
structure, as seen in S. bechsteini.
99
Methods
Study area and sampling
Flies were collected from 9 Bechstein’s bat maternity colonies living in
forests in the vicinity of Würzburg, Germany (Figure 4.1). All bat colonies
make regular use of bat boxes, and have been consistently monitored for up
to 20 years (Kerth and van Schaik 2012). Imagoes of flies were collected from
bats caught from their roosts in August and September. Additionally, fresh fly
puparia were collected from bat boxes for which constant monitoring could
show that Bechstein’s bats (and no other bat species) had recently occupied
the box (colonies BS, GB2, HB). All colonies additionally make use of natural
tree roosts to varying degrees, and overall roost switching is comparable
across roost types (Kerth and Melber 2009). No puparia could be collected
from these locations. Flies were removed from the bat’s fur using soft
forceps and stored in 90% ethanol prior to DNA extraction. All imago flies
were morphologically identified prior to genetic analysis (Theodor 1967,
Hutson 1984). Genomic DNA was isolated from individual flies using a
salting-out method (Miller et al. 1988). The abdomen of visibly gravid female
flies was removed prior to extraction, in order to avoid co-amplification of the
embryo’s DNA during the subsequent PCRs.
Microsatellite development
Genomic DNA was sent to GenoScreen (Lille, France; www.genoscreen.fr) for
the creation of a microsatellite library using high-throughput sequencing. This
yielded 416 potential microsatellite sequences; of which unlabelled primer
pairs were developed for 65 and initially tested for amplification and
polymorphism using pooled DNA from 8 individuals. All primer pairs were
tested with a range of 6 different annealing temperatures (54°C-64°C; Veriti
Thermal Cycler, Applied Biosystems). Each 10 µl PCR reaction contained 1 µl
of DNA template (10 ng), 5 µl of 2x QIAGEN Multiplex Master-Mix (Qiagen), 1
µl of 10 µM primer mix and 3 µl of distilled water. PCR products were
assessed on a 2% agarose gel using GelRed (Biotium Inc.), and 40 primer
pairs were subsequently ordered with an M13 - universal tail on the forward
100
Figure 4.1 Map showing the sampled bat colonies (circles), forest patches
(light grey), the river Main (dark grey) and the city of Würzburg (Bavaria,
Germany). All colony names correspond to the abbreviations given in Table
4.2.
primer (as in Schuelke 2000). PCR was preformed as before, but with a
second round of 10 amplification cycles with an annealing temperature of
53°C in the presence of the universal M13-fluorescent (FAM) primer. This
returned 16 primer pairs that yielded at least 3 alleles using the pooled DNA
101
sample, which were subsequently ordered with individual fluorescent labels.
To assess these loci for Hardy-Weinberg equilibrium and linkage, 50
individuals from the bat colonies BW and ES were genotyped. Analysis
performed in Genepop on the web 4.0.10 (Rousset 2008) found significant
deviations from Hardy-Weinberg equilibrium in 3 loci, as well as evidence for
linkage in 3 further loci, and thus these 6 loci were discarded. Therefore, we
established 10 polymorphic microsatellite loci in 3 multiplexes (Table 4.1),
which were used to genotype all individuals.
Microsatellite amplification
PCR amplification was performed using 3,5 µl reactions containing 2 µl 2x
QIAGEN Multiplex Master-Mix (Qiagen), 0.5 µl of 10 µM primer mix, 0.5 µl of
distilled water, and an evaporated DNA template. Each PCR was composed
of an initial denaturation at 95°C for 15 min; followed by 34 amplification
cycles (95°C for 60 s, 60°C for 60 s, and 72°C for 60 s), followed by a final
elongation of 72° C for 7 min. One µl of PCR product was added to 9 μL HiDi formamide and 0.07 μL GS-500 LIZ size-standard (Applied Biosystems)
and resolved in POP4 polymer in an ABI 3130 Genetic Analyser (Applied
Biosystems). Data was analysed using GENEMAPPER version 4.1.
Data analysis
In total, 332 flies from 9 bat maternity colonies and 16 sampling events
(spaced within and between years) were genotyped for all microsatellite loci
(Table 4.2).
All loci were checked for the presence of null alleles, allelic
dropout, and stuttering using MicroChecker 2.2.5 (van Oosterhout et al.
2004). All sampling events were checked for deviations from Hardy-Weinberg
equilibrium and linkage disequilibrium using Genepop on the web 4.0.10
(Rousset 2008).
102
Table 4.1 Characterization of the newly developed microsatellite loci for
Basilia nana. The following abbreviations are used: annealing temperature
(Ta), the multiplex mix in which each marker was included (Primer Mix) and
which fluorescent label was used (Label), the observed fragment length range
(Size range), the number of alleles observed (No. alleles), observed and
expected heterozygosity (Ho and He) and both primer sequences (F =
forward, R = reverse). All sequences have been deposited in the GenBank
under the accession numbers provided.
Locus
BN1
Repeat
motif
(CT)8
Ta
60
Primer Mix;
Label
B; FAM
Size range
(bp)
80-102
No. of
alleles
8
Ho
He
0.518
0.521
GenBank
Accession no.
KP974822
Primer Sequences (5'-3')
F: TGCCTAACACAAGGGGCTTA
R: GCAATTACATTAATGGCCTGA
BN6
(TC)7
60
A; HEX
103-121
9
0.541
0.591
KP974823
F: TTTCATACATCACAAAGCATCAA
R: ACTGATTGGAATGCTGGAGG
BN21
(AC)9
60
A; NED
120-136
9
0.618
0.614
KP974824
F: TCTGCCCTTTGCGTAGTTAG
R: TCATTTGTCATTCATCGTTGTG
BN25
(AC)7
60
B; FAM
134-166
12
0.502
0.558
KP974825
F: AACTTAGCTGGCTGCTGCTC
R: GCCGATGCTGGTATTTGTTT
BN32
(TTG)11
60
B; HEX
63-117
17
0.696
0.733
KP974826
F: GGCCGCTTGCTTATTTACTC
R: ACGACACGAATAGCAACACG
BN42
(TGT)7
60
B; HEX
126-156
11
0.566
0.594
KP974827
F: TGGATGTGTTGTTGTGGTTG
R: CACGAAAATCAGTTCCCGTT
BN55
(TTG)8
60
A; FAM
132-159
9
0.532
0.539
KP974828
F: TCAGCATCATGCAACTGTGA
R: CAACAGCATTTCAGCGAGTC
BN56
(TGT)6
60
C; HEX
149-155
3
0.248
0.238
KP974829
F: TGCTATGGTTACCACAATTTCG
R: ACTCGGCTTGCTTGTGATCT
BN64
(ACA)10
60
C; FAM
159-231
14
0.603
0.717
KP974830
F: CCAGCAACACGTTAAGCAGA
R: GCTGTCGGCCATCATTTATT
BN65
(GTT)9
60
A; FAM
206-233
9
0.459
0.497
KP974831
F: CGGATTCCGTTGTATTGCTT
R: CTAGCGTTCATCCGACACAA
Subsequently, in order to increase the number of colonies that could be
compared, as well as the intra-colony sample size, for the analysis of spatial
genetic structure, samples from 2002 and 2003 were pooled. A hierarchical
AMOVA comparing genetic variation between colonies and between years for
the two pooled years found only 0.006% of variance to be between 2002 and
2003 (supplementary materials, Table S4.1). All tests and comparisons were
103
also carried out using separate datasets for each year, but did not yield
different results (data not shown).
Number of alleles per locus, allelic richness, observed and expected
heterozygosity, FIS, and single-level AMOVAs were calculated using Genodive
2.0b25 (Meirmans and Van Tienderen 2004). Pairwise FST-values, as well as
the standardized measure G’ST (Hedrick 2005), were also calculated using
Genodive 2.0b25 (Meirmans and Van Tienderen 2004). The number of private
alleles per sampling event was calculated using Genalex 6 (Peakall and
Smouse 2006). Isolation by distance (IbD) was measured by comparing
matrices of FST / (1- FST) and log transformed geographic distance using
Mantel tests (10,000 permutations) in Genodive 2.0b25 (Meirmans and Van
Tienderen 2004). Additionally, fly genetic distance was compared to a matrix
of forest fragment ID (using the same method) to investigate the influence of
proximity within forest patches rather than geographic distance, to
investigate the possible transmission of flies between female Bechstein’s bat
maternity colonies via male Bechstein’s bats or other bat species throughout
the summer. Sequential Bonferroni corrections were used to compute the
critical significance levels for related statistical tests.
Temporal differentiation within a single bat maternity colony was
investigated in three colonies for which a sufficient sample was collected in
2002/3 and 2011 using pairwise FST-values and hierarchical AMOVAs
calculated in Genodive 2.0b25 (Meirmans and Van Tienderen 2004). A similar
analysis was carried out comparing the samples from 2002 and 2003 for the
four bat colonies for which both years were sampled (BS, BW, ES, SD).
In order to assess the influence of our destructive fly sampling on the
temporal differentiation found within colonies, we performed a Kendall’s rank
correlation between the number of flies sampled in the previous sample(s)
and the temporal FST-values. We found no effect of destructive sampling in
both 2003 (Kendall’s T < 0.001; p = 0.997) and 2011 (Kendall’s T = 0.280; p =
0.249).
104
Table 4.2 Overview of host colony information and basic genetic descriptors
of all fly samples used in this study. The following abbreviations are used: bat
colony latitude (Lat) and longitude (Long), estimated number of bats in the
colony (colony size), number of flies genotyped (n), mean number of
microsatellite alleles per locus (NA), allelic richness (K), observed (Ho) and
expected (He) heterozygosities, number of private alleles per colony (Pall), and
inbreeding coefficient (Fis).
Colony Name
Year
Lat
Long
Colony size
Blutsee (BS)
2002
49.725
9.833
15-19
Brandtwiese (BW)
K
Ho
He
3.639
0.543
0.567
Pall
0
0.034
Fis
13
2011
13
4
3.507
0.459
0.486
3
0.06
17
5.5
3.617
0.517
0.558
0
0.062
16
4
3.497
0.524
0.547
2
0.024
24
5.8
3.564
0.529
0.551
1
0.021
14
4.3
3.696
0.55
0.548
2
-0.005
2002
49.898
9.995
20-24
38
2011
2002
49.902
9.945
20-24
2003
Guttenberg 2 (GB2)
NA
4.9
2003
2003
Einsiedel (ES)
n
12
2003
27
49.745
9.865
35-39
42
5.1
3.639
0.571
0.577
2
-0.003
Gramschatz 1(GS1)
2011
2002
49.903
9.978
40-44
12
3.9
3.576
0.455
0.523
0
0.105
Gramschatz 2 (GS2)
2002
49.947
9.94
20-24
12
4.5
3.998
0.65
0.58
4
-0.122
Höchberg (HB)
2002
49.785
9.867
20-24
36
5.4
3.709
0.529
0.582
2
0.099
Steinbach (SB)
2002
49.704
9.745
20-24
9
3
2.931
0.461
0.434
0
-0.074
Stalldorf (SD)
2002
49.571
9.912
25-29
26
5.3
3.710
0.546
0.576
2
0.046
2003
21
Comparison with host and wing-mite population genetic structure
Existing genetic datasets are available for both the host, Myotis bechsteinii
(microsatellites, both nuclear and mitochondrial: Kerth et al. 2002b; Kerth and
van Schaik 2012), and for the ectoparasitic wing-mite, Spinturnix bechsteini
(nuclear microsatellites: van Schaik et al. 2014; and mtDNA cytb sequences:
Bruyndonckx et al. 2009b). For both species (host and wing-mite), the
datasets were adapted to match the colonies sampled for B. nana and
summary statistics and pairwise population differentiation were calculated as
described above.
105
To compare fly, host and wing-mite genetic distances we used a
partial Mantel test (10000 permutations) correcting for geographic distance.
Matrices of FST / (1 - FST) were used to compare the nuclear genetic distance
of all species, as well as the mitochondrial (microsatellite) genetic distance in
the host. For comparison to wing-mite mitochondrial genetic distance, we
used ΦST / (1- ΦST). A structure analysis comparing nuclear population genetic
structure was performed in structure 2.3.3 (Pritchard et al. 2000) for all
colonies for which data are available for all three species. For each species
we used an admixture model without providing location as a prior for a range
of potential subpopulations (K) from 1 to 10 (number of sampled colonies
included: 6). Five iterations were run for each K with a burn-in and run length
of 200000 and 1000000 repetitions respectively, and the most probable K
was inferred using the ΔK “Evanno method” (Evanno et al. 2005) using
Structure Harvester Web v0.6.93 (Earl and vonHoldt 2011). Additionally, data
were analysed using geneland 3.2.4 (Guillot et al. 2005), which has been
shown to be more robust with low sample sizes (Frantz et al. 2012). To
determine the number of subpopulations, 10 independent runs (1000000
MCMC iterations, thinning of 1000, uncertainty of spatial coordinates of
100m) were preformed where K was allowed to vary between 1 and 10.
106
Results
A Microchecker analysis did not find evidence for the presence of null alleles,
allelic dropout or stuttering when analysing over all populations within a year.
No linkage disequilibrium was found between loci. We found between 3 and
17 alleles per locus (mean = 10.1; Table 4.1).
Of the nine sampled B. nana populations, two showed highly
significant deviations from Hardy-Weinberg equilibrium due to heterozygote
deficiency (p < 0.0001; BW and SD), with highly significant deviations in two
(BN25, BN64) and five (BN6, BN25, BN37, BN42, BN65) markers
respectively. Additionally, two other populations were found to have
marginally significant heterozygote deficiencies (p<0.05; ES and HB),
stemming primarily from significant deviations in one (BN64) and two (BN6,
BN64) markers respectively. In other B. nana populations, single markers
showed deviations from Hardy-Weinberg in six cases. We believe that these
deficiencies are primarily the result of sampling of related individuals
(including puparia) within bat colonies (see also Dharmarajan et al. 2011).
Overall genetic diversity was low with a mean of 4.64 alleles per locus per
population and a mean allelic richness of 3.59 (Table 4.2). We found a total
of 18 private alleles evenly distributed across colonies (range: 0-4, mean =
1.5; Table 4.2).
Spatial genetic structure
Pairwise FST-values of flies among bat colonies showed very little population
differentiation (range: -0.005 - 0.063, mean = 0.010; Table 4.3), with only flies
from BS being significantly differentiated from flies of two other bat colonies
(GS2 and SB). Pairwise G’ST-values gave a slightly larger range, but were not
qualitatively different (range: 0 - 0.138, mean = 0.033; Table 4.3). An AMOVA
found the vast majority of genetic variation within fly individuals (93%), with
low but significant differentiation among flies within colonies (6.4%) as well
as among colonies (0.6%; Table 4.4a). Similarly, a structure analysis was
unable to find any sub-structuring (Figure 4.2a), with the highest loglikelihood supporting a single population (lnP: -3268.440 ± 0.114; See
107
supplemental materials Figure S4.2 for all ln-P and ΔK values). Additionally, a
geneland analysis across all nine fly populations was also unable to detect
population sub-structuring in the fly, finding highest likelihood support for a
single population. No correlation was found between geographic and genetic
distance (r2 = 0.0, p = 0.537; Table 4.5), nor were flies from bat colonies living
within the same forest fragment more similar (p = 0.228).
Table 4.3 Pairwise FST-values (above the diagonal) and G”ST-values (below
the diagonal) for B. nana. Asterisks indicate significant differentiation after
correction.
G’ST \ FST
BS
BW
ES
GB2
GS1
GS2
HB
SB
SD
BS
--
0.002
0.003
-0.003
0.005
0.025*
0.010
0.063*
0.005
BW
0.007
0.003
-0.001
0.009
0.007
0.006
0.024
0.003
ES
0.010
0.009
--
-0.005
0.011
0.009
0.004
0.026
0.000
GB2
0.000
0.000
0.000
GS1
0.020
0.026
0.030
0.005
-0.002
0.008
-0.003
0.029
0.000
--
0.034
0.013
0.023
0.009
GS2
0.061
0.019
0.020
0.017
0.084
0.007
0.040
0.010
HB
0.027
0.015
0.012
0.000
0.041
0.021
0.035
0.002
SB
0.138
0.063
0.063
0.066
0.060
0.083
0.091
SD
0.015
0.008
0.001
0.003
0.028
0.026
0.008
--
-0.078
0.030
--
Temporal genetic structure
No temporal genetic differentiation (Mantel test, p = 0.330) of flies was found
in the three bat colonies sampled in 2002 or 2003 and 2011. The level of
differentiation within colonies across years (FST = 0.012) was slightly higher
than the average differentiation seen between different colonies sampled
within a single year (FST = 0.009). A similar analysis was carried out for the
four colonies where sufficient fly samples were collected in both 2002 and
2003 (BS, BW, ES, SD) yielding very similar results (mean within colony FST =
108
0.013, mean within year FST = 0.006). A hierarchical AMOVA supported this
result and found none of the genetic variance was partitioned across years (0.1%; Table 4.4b), while again finding the vast majority of variation within
individuals (93.9%).
Figure 4.2 STRUCTURE results for a) Basilia nana, b) Myotis bechsteinii and
c) Spinturnix bechsteini. Colony names correspond to the abbreviations in
Table 4.2. In B. nana and M. bechsteinii no discernable population substructuring was found (K=1). For these species K=2 is shown to illustrate the
lack of sub-structuring when multiple subpopulations are assumed. For S.
bechsteini, the best ΔK and highest log likelihood values were attained for
K=6 (the actual number of colonies sampled).
109
Table 4.4 a) Single-level analysis of molecular variance (AMOVA) in B. nana,
b) hierarchical AMOVA investigating temporal variance across all colonies
sampled in 2002/3 and 2011. Asterisks indicate significant contributions.
a)
Source of variation
Among colonies
Among individuals,
within colonies
Within individuals
8
Sum of
squares
32.277
252
261
763.085
694.500
d.f.
Variance
component
0.018
F-statistic
FST = 0.006
0.184
2.661
FIS = 0.065
FIT = 0.070
Percentage of
variation
0.6*
6.4*
93.0
b)
Source of variation
Among years
Among colonies,
within years
Among individuals,
within colonies
Within individuals
Sum of
squares
d.f.
Variance
component
Percentage of
variation
1
3.534
-0.004
F-statistic
FCT = -0.002
4
15.271
0.018
FSC = 0.006
0.6*
159
165
476.347
441.500
0.16
2.676
FIS = 0.056
FIT = 0.061
5.6*
93.9
-0.2
Comparison with host and wing-mite
We applied mantel tests to examine the correlation between the genetic
distances of host and parasites (Table 4.5). Fly and bat pairwise FST-values
(nuclear and mitochondrial) were not significantly correlated after correcting
for geographic distance. Fly nuclear genetic distance was however
significantly correlated with wing-mite nuclear genetic distance (r2 = 0.363, p
= 0.044), but not to wing-mite mitochondrial genetic distance (pairwise ΦSTvalues; r2 = 0.027, p = 0.300). Plots of pairwise genetic distance (FST/1 – FST)
between fly nDNA and both bat and wing-mite nDNA are provided in Figure
4.3.
A structure analysis for the six colonies sampled for all three species
was only able to detect population sub-structuring in the wing-mite (Figure
4.2) as had previously been show by van Schaik et al. (2014). Wing-mite
populations were highly differentiated and structure found evidence for 6
sub-populations, clustering wing-mites according to their native bat colony
(Figure 4.2c; see also van Schaik et al. 2014). In both the fly and bat no
110
structuring was detected, with the highest log-likelihood given for only one
population (Figure 4.2a-b; see supplemental materials Figure S4.2 for graphs
of ln-P and ΔK of all species).
Table 4.5 Correlation of fly genetic distance to geographic distance (IbD),
host nuclear and mitochondrial genetic distance and wing-mite nuclear and
mitochondrial genetic distance. All correlations to host and wing-mite genetic
distance were performed as partial mantel tests correcting for geographic
distance. For each correlation the slope (β), the variance explained by the
model (R2) and P-value (P) are given.
β
R2
P
Fly nDNA - Geographic distance
-0.0001
0
0.537
Fly nDNA - Host nDNA
-0.332
0.071
0.229
Fly nDNA - Host mtDNA
0
0.025
0.103
Fly nDNA - Mite nDNA
0.085
0.363
0.044
Fly nDNA - Mite mtDNA
0.001
0.027
0.300
111
Figure 4.3 Pairwise genetic distance (FST/1 – FST) estimates between
populations of hosts and parasites for a) fly nDNA vs. bat nDNA, b) fly nDNA
vs. wing-mite nDNA. There was no correlation between fly and host genetic
distance (β = -0.332; p = 0.229). Fly and wing-mite genetic distance were
significantly correlated (β = 0.085; p = 0.044).
0.07%
a)
0.06%
0.05%
Fly%FST/(1!FST)!!
0.04%
0.03%
0.02%
0.01%
0%
0%
0.005%
0.01%
0.015%
0.02%
!0.01%
b)
0.025%
0.03%
0.035%
0.04%
0.045%
0.05%
Bat%FST/(1!FST)!!
0.07%
0.06%
0.05%
Fly%FST/(1!FST)!!
0.04%
0.03%
0.02%
0.01%
0%
0%
!0.01%
0.05%
0.1%
0.15%
0.2%
Mite%FST/(1!FST)!!
112
0.25%
0.3%
0.35%
0.4%
Discussion
Spatio-temporal genetic structure
Using 10 nuclear microsatellite loci to investigate the population genetic
structure of the bat fly Basilia nana, we found almost no genetic
differentiation between the sampled host colonies. Temporally, the sampled
fly populations were similarly stable, as genetic differentiation over a 9 year
timespan was insignificant and comparable to the low differentiation seen
between different bat colonies within a single year. Genetic distance between
fly populations was neither correlated to geographic distance, nor to the
genetic distance of the host.
The observed population genetic structure (or lack thereof) can be
explained in light of both fly and bat life history. The lack of genetic
differentiation
across
host
colonies
suggests
extensive
horizontal
transmission of flies between host maternity colonies. Although horizontal
transmission via male Bechstein’s bats or perhaps even other bats is
possible, we did not observe increased genetic differentiation between forest
patches or over geographic distance as would be expected if transmission
occurred primarily via males or other species overlapping with multiple
adjacent female maternity colonies during summer. Instead, we suspect that
the focal point of transmission of flies from different host maternity colonies
and male bats occurs at underground sites during autumn and winter. This is
supported by the correlation seen between fly and mite genetic distance
across colonies as further outlined in the next paragraph. Given the possible
migration distances between summer and winter sites in M. bechsteini
(Hutterer et al. 2005), we expect all of the bat colonies in the study area can
potentially meet at communal mating/hibernation sites. As flies go through
only two generations in the summer (Reckardt and Kerth 2006), horizontal
transmission at communal sites in the other seasons thereby effectively
allows for gene flow within the catchment area of these sites every second
generation. This results in a large fly meta-population whose genetic pool
encompasses multiple bat colonies within the area, thereby overcoming the
limited population sizes within individual colonies.
113
Influence of host behavioural/social adaptations
By comparing the genetic structure of the fly to the previously analysed wingmite, the interaction between host and parasite life history can be more
extensively contrasted. We found a weak, but significant, correlation between
fly and wing-mite nuclear genetic differentiation across colonies (r2 = 0.363, p
= 0.044; figure 4.3). This suggests that fly and wing-mite horizontal
transmission between host colonies occurs in a similar fashion, i.e. primarily
during host mating and hibernation periods (van Schaik et al. 2014).
We argue that the large differences observed in the degree of genetic
differentiation in the two parasites are subsequently caused by differences in
the levels of gene flow and genetic drift in both species. During summer,
frequent roost switching with a strong preference for previously unoccupied
roosts (Reckardt and Kerth 2007) significantly reduces the number of fly
puparia that are able to successfully infect host colonies. Additionally, as
Bechstein’s bat maternity colonies do not overlap in roost use, this does not
allow for direct transmission of flies between female maternity colonies.
However, presumably as a result of roost transmission, flies are able to
persist on male Bechstein’s bats throughout the summer, increasing the
transmission between female maternity colonies via male bats. Coupled with
their low number of generations per year, these life history characteristics
yield low genetic drift and high gene flow, resulting in a temporally stable and
undifferentiated meta-population. On the other hand, while wing-mites are
not affected by roost switching behaviour of the host, their survival and gene
flow are strongly affected by the small colony size and extremely limited
inter-colony interaction of female bats, and their inability to persist on male
bats. Moreover, the higher rate of reproduction in the mites results in
substantial inbreeding within colonies and allele frequency differences
between colonies. These characteristics interact to yield strong genetic
differentiation of mites between individual host maternity colonies as well as
high
temporal
instability
including
frequent
local
extinction
events
(Bruyndonckx et al. 2009b, Reckardt and Kerth 2009, van Schaik et al. 2014).
As a consequence, the levels of gene flow and genetic drift, and thus
114
population genetic structures, of the two parasite species are highly
divergent despite the fact that the interaction between host and parasite life
histories broadly leads to similar patterns of transmission.
Interestingly, a comparative study across two wing-mite species
illustrates that small changes in host and parasite life history can also lead to
a genetically stable population structure in wing-mites. In Spinturnix myoti, a
closely related wing-mite species found on the greater mouse-eared bat
(Myotis myotis), a genetically diverse and unstructured population was found
(van Schaik et al. 2014), very similar to the one observed in B. nana in this
study. Greater mouse-eared bat summer maternity colony sizes are much
larger than those of Bechstein’s bats, and they frequently interact with
another bat species (lesser mouse-eared bat; Myotis blythii) also infected by
S. myoti, and often hibernate in large clusters (Güttinger et al. 2001). As a
result, S. myoti does not suffer from the same winter bottleneck, and the
initial spring population in each summer maternity colony is much larger than
in S. bechsteini (van Schaik et al. 2014). Additionally, during the autumn
mating period, greater mouse-eared bats may also form temporary harems
where direct contact between females from different maternity colonies may
take place. In combination, this yields a wing-mite population that is
genetically highly diverse and panmictic.
Comparison with other host-parasite systems
Several investigations of parasite genetic structure have found large
differences due to differing parasite dispersal abilities. In a comparison of
three parasites (two lice and one fly) of the Galapagos Hawk (Buteo
galapagoensis), Whiteman and colleagues (2007) found differing levels of
inter-island genetic differentiation in the parasites closely reflecting their
ability to disperse. While the fly closely mirrored the lack of genetic structure
observed in the host, the lice showed moderate and strong differentiation as
a result of their differing dispersal methods (horizontal and vertical
transmission respectively). This is very similar to the results found in the
current study, although the strong difference between the genetic structures
115
of different parasites observed in our study is not the result of different
degrees of vertical transmission but rather due to differences in the amount
of genetic drift induced by the interaction between parasite life history and
host social system. In another study, two feather lice species of the
Australian Magpie, Toon & Hughes (2008) were found to have strongly
contrasting phylogeographic patterns over the range of the host. While one
feather louse species closely followed the phylogeographic structure
observed in the host, the other feather louse showed a contrasting pattern.
This contrast was attributed to differences in dispersal ability between the
lice, with the latter species being able to potentially disperse via long-range
male host dispersal, or via hitchhiking on hippoboscid flies (Toon and Hughes
2008). Here too, strong similarities can be found with the current study, as
male vectors appear to strongly influence the phylogeographic signal of two
otherwise similarly transmitted parasites.
Conclusion
The two parasites of the Bechstein’s bat investigated here show highly
divergent spatial genetic structures despite relying on comparatively similar
transmission opportunities, presumably due to differences in the amount of
gene flow and genetic drift. Concordantly, we find that the temporal genetic
stability of the two parasite populations within our single host metapopulation
is
similarly
divergent.
Taken
together,
the
few
studies
investigating the population genetic structure of multiple parasites on a single
host (Gómez-Díaz et al. 2007, Whiteman et al. 2007, Toon and Hughes 2008;
this study) have illustrated that parasite population genetic structure can
differ strongly as a result of both parasite life history characteristics (dispersal
ability, reproductive rate), host life history (inter-population contact, male
vectors, colony size), and even inter-parasite interactions (parasite acting as
a vector for another parasite). Therefore, in order to make accurate
predictions regarding the genetic structure, and ultimately evolutionary
potential, of a parasite, the life histories of both interacting species must be
well characterized and their potential interactions evaluated.
116
Data accessibility
Nucleotide sequences of the 10 new microsatellites generated for this study
were deposited in GenBank under Accession Numbers KP974822KP974831. Microsatellite data for all flies analysed have been submitted to
DataDryad: doi:10.5061/dryad.5dd67
Acknowledgements
We would like to thank Daniela Fleischmann, Markus Melber, Karsten
Reckardt and many others for their help collecting samples; Ina Römer, Alain
Frantz and Sébastien Puechmaille for their help in the lab and with data
analysis; and Pascal Hablützel, Serena Dool and five anonymous referees for
their useful comments on the manuscript. This study was carried out under
license from the nature conservancy department of Lower Franconia. Our
research was funded by two grants from the Volkswagen Stiftung to JvS (Az
84 871) and GK (Az 84 959). JvS is a member of the Max Planck Research
School for Organismal Biology.
117
Supplementary materials
Table S4.1 A hierarchical analysis of molecular variance (AMOVA) comparing
all samples from 2002 and 2003. As only 0.006% of variance was found
among years, we pooled samples from both years when analysing spatial
patterns of genetic differentiation to increase sample size and overall number
of compared colonies.
Source of
variation
Among years
d.f.
1
Sum of
squares
7.761
Variance
component
0.02
F-statistic
0.006
Percentage of
variation
0.6*
Among colonies,
within years
6
25.055
0.02
0.006
0.6*
Among individuals,
within colonies
170
564.130
0.173
0.055
5.4*
Within individuals
178
529.000
2.972
0.067
93.3
Figure S4.2 Below are the log-likelihood values and corresponding DeltaK
values (as described in Evanno et al. 2005) for all K’s tested using structure
(K= 1-10; 5 iterations per K). All graphs were produced using Structure
Harvester (Earl and vonHoldt 2011). In B. nana and M. bechsteinii, no
evidence for population sub-structuring was found (K=1 has the highest loglikelihood), whereas in S. bechsteini the number of recovered subpopulations (K=6) corresponds to the number of populations sampled.
Basilia nana (fly)
118
Myotis bechsteinii (bat)
Spinturnix bechsteini (wing-mite)
119
120
121
Chapter 5
Synthesis and general discussion
A friend of mine recently expressed her dismay at “lowly” parasites. I beg to
differ – if anything, parasites, including bat flies, are incredible examples of
evolution at its best, organisms capable of both adapting to life in the most
hostile of environments (the very substrate you live on wants you dead!) and
resisting diseases that live inside your body.
- Piotr Naskrecki
Social hosts face a wide diversity of parasites and pathogens at all times,
requiring an equally wide arsenal of defense mechanisms. The aim of this
dissertation is to contribute towards understanding how facets of host social
system interact with host and parasite life history to shape host-parasite
infection and evolutionary dynamics. Specifically, I focused on exploring this
topic within a multi-host, multi-parasite comparative system of European bat
species and their mite and fly ectoparasites. Using this framework, I
addressed three main research questions regarding the influence of bat
social system on mite infection dynamics and mite and fly population genetic
structure (see research aims; chapter 1).
In this chapter, I synthesize the effects observed in chapters 2-4 and
discuss their relevance at the two investigated levels of host-parasite
dynamics (local infection dynamics, population genetic structure and
122
evolutionary potential). Subsequently, I briefly explore the possible
consequences of host sociality at the third level (the directionality of the
evolutionary trajectory), and discuss the possible restrictions that parasites
impose on host social system. Finally, I integrate these effects into a general
summary of the overall impact of host sociality on host-parasite dynamics,
and provide an outlook regarding future directions in host-parasite
evolutionary research.
Host sociality and parasite infection dynamics
In chapter 2 I provide an unambiguous example of how host social
organization and mating system influence parasite infection and transmission
dynamics. Across three hosts with varying group size in summer and
differing degrees of contact during mating in autumn, I observe marked
differences in parasite infection intensity throughout the autumn mating
period closely mirroring the predictions made based on the social systems of
the hosts.
At the onset of the autumn mating period, mite prevalence and
infection intensity was highest in females and juveniles of M. myotis, with
lower infection levels in M. daubentonii and M. nattereri, faithfully reflecting
differences in host aggregation size in summer. Males of M. myotis and M.
nattereri were rarely infected at the onset of the mating season, matching
their generally solitary nature during summer. In M. daubentonii, the
formation of male bachelor groups during summer had a clear effect on mite
prevalence and infection intensity as M. daubentonii of all host age and sex
classes were similarly infected. Additionally, the gradual disbandment of
maternity colonies in autumn was clearly visible in all species, as overall mite
prevalence and intensity decreased throughout autumn in each, further
highlighting the role of host social organization in shaping parasite infection
dynamics.
123
Regarding parasite transmission, a similarly clear effect of host mating
system was found in males of M. myotis and M. nattereri. In M. myotis, where
there is lengthy contact (1 or more days) between the sexes in temporary
harems, mite infection of male bats rose to levels equivalent to those of the
other host classes by the end of the season. Meanwhile, in M. nattereri,
where contact between the sexes during mating is comparatively short, male
bats still experienced lower levels of parasite prevalence and intensity at the
end of the autumn mating season. As a final note, despite catching dozens of
M. bechsteinii, our focal species in the other chapters, we did not find any
mites on this species. Nor were mites found on approximately 200
Bechstein’s bats sampled at ten other sites and on single occasions during
the first year of sampling (data not shown). This suggests that mite infection
levels in this host are likely even lower, or that mites may even go locally
extinct in certain host populations.
In general, the effects of host social system on parasite infection
dynamics are comparatively well characterized, as shown by the accuracy of
the predictions in this study. Nevertheless, it remains important to examine
these effects, as accurate interpretation of parasite population genetic
structure can only be achieved when these patterns are known. For example,
at the level of population genetic structure, the genetic differentiation of
parasite populations may either be the result of 1) low parasite transmission
between populations, 2) strong local adaptation of the local parasite
populations or 3) may simply arise due to strong genetic drift within
populations. Therefore, understanding infection and transmission dynamics
at the ecological level remains paramount to accurately interpreting parasite
population genetic structure and the resulting evolutionary dynamics of the
host-parasite interaction. Naturally, other genetic indices will also help reveal
which of these scenarios is most likely, but these estimates are often inexact
at the low sample sizes and marker numbers achievable for most parasite
species. This is especially relevant in host species where social system is
complex and aggregation and contact rates vary throughout the year (e.g.
seasonal fluctuation), particularly when these changes are coupled with
124
major changes in host physiology (e.g. hibernation). Although often
overlooked (Altizer et al. 2006), in these cases, understanding the concurrent
dynamics of parasite infection, and notably how parasites persist during
“sub-optimal” periods, becomes even more critical.
Host sociality and parasite population genetic structure
In the other two chapters I explore how host social system and parasite life
history interact to shape parasite population genetic structure.
In chapter 3 I show that differences in host social system result in
contrasting population genetic structures in two bat-mite pairs (M. myotis
and S. myoti, M. bechsteinii and S. bechsteini). S. myoti populations showed
high levels of genetic diversity and very little pairwise differentiation, whereas
those of S. bechsteini were much less diverse, strongly differentiated and
showed strong temporal turnover. As in chapter 2, the contrasting patterns
observed in parasite structure can be interpreted based on the predicted
effect of several aspects of host social system that differ between the two
host species. Specifically, M. myotis, relative to M. bechsteinii, forms much
larger summer maternity colonies and there is more inter-group contact
during all seasons. In S. myoti, this yields a large and panmictic population
structure, which is both spatially and temporally stable. In S. bechsteini, mite
populations in individual host colonies undergo a yearly winter bottleneck
that severely limits effective population size and leads to the temporal
instability of mite populations (including frequent local extinction). As mite
horizontal transmission occurs primary in autumn before this bottleneck, this
additionally leads to strong genetic differentiation among colonies due to
genetic drift.
These results corroborate the results from chapter 2 and illustrate that
the difference in overall mite infection and mite transmission, as shaped by
the social system of their hosts, also result in marked differences in parasite
population genetic structure. Significantly, as the relative rate of gene flow
125
and genetic drift between the host and parasite were also opposite in the two
species pairs investigated, these differences are predicted to lead to a higher
evolutionary potential of the mite in one instance (M. myotis / S. myoti) and of
the bat in the other (M. bechsteinii / S. bechsteini).
Naturally, parasite population genetic structure is also strongly
influenced by the life history of the parasite (Barrett et al. 2008). Therefore, in
chapter 4, I explore how different parasite life histories interact with the
same host social system by comparing the population structure of the mite
S. bechsteini characterized in chapter 3 to that of the bat fly, B. nana. In
contrast to S. bechsteini (described above), B. nana showed little spatial or
temporal genetic structure, suggesting a large, stable population with
frequent genetic exchange between fly populations from different bat
maternity colonies. These results suggest that although the transmission
patterns were similar in both parasite species, the level of gene flow and
eventual spatio-temporal genetic stability differs due to the reproductive
mode and life-history strategy of the parasites. Specifically, female S.
bechsteinii give birth to protonymphs directly on the host, while B. nana
deposits puparia on the walls of the roost. Coupled with the frequent roost
switching behavior of the host, which strongly affects the ability of B. nana to
successfully infect the host (Reckardt and Kerth 2007), this yields much
faster population growth in S. bechsteinii than in B. nana. However, as the
flies that do successfully infect hosts live longer and flies only go through 2-3
generations isolated in host colonies before having the opportunity to
horizontally transmit at autumn swarming sites, this ironically yields to a
much higher effective population size and a more temporally stable genetic
structure in B. nana as the entire region effectively becomes one population
rather than a sub-structured metapopulation. Similar to the previous study,
these differences in parasite population genetic structure again yield
contrasting evolutionary potentials, and hence ability to locally adapt to the
host, in the two parasite species.
As a whole the results presented in these two chapters show that
parasite population genetic structure can be interpreted based on 1) the
126
predicted effects of individual parameters of host social system and 2) their
interaction with parasite life history. However, despite this apparent success,
how these parameters interact and the relative importance of each remain
difficult to predict. For example, although the highly differentiated and
temporally unstable population genetic structure of S. bechsteini can be
interpreted in light of its hosts small summer maternity colony size, it would
not have been possible to estimate the severity of the annual bottleneck or
the stark contrast between the population genetic structure of S. bechsteini
and S. myoti without exploring it empirically. Analogously, as the second
study shows, the interaction between host social system and parasite life
history is sensitive to very small differences in either, and certain aspects of
host behavior may be highly relevant for one parasite species, but not for
another. Being able to accurately gauge the importance of these effects is
critical for coevolutionary outcomes, as theoretical models predict that even
slight differences in parasite gene flow or drift can strongly alter parasite
evolutionary potential and local adaptation dynamics (Gandon and Nuismer
2009b). Moreover, this sensitivity may also explain why many meta-analyses
that aim to uncover general trends between host and parasite characteristics
(and genetic structure) find little to no effect (Poulin and Forbes 2012, MazéGuilmo et al. 2016).
These differences are even more striking when considering the limited
diversity of host social systems and range of parasite types investigated. The
comparative framework used here was specifically intended to eliminate
possible confounding effects of differences in host life history, environmental
factors, and large differences in parasite life history (e.g. generalist vs.
specialist, mobile vs. non-mobile). Given our poor predictive ability despite
these simplifications, it is therefore evident that accurate estimations of
parasite population genetic structure and its consequences for evolutionary
potential based solely on the predicted effects of all host social and parasite
life
history
characteristics
currently
remains
effectively
impossible.
Nevertheless, comparative investigations exploring these interactions are by
no means pointless, as it will only be possible to uncover generalities
127
regarding the relative importance, and interaction of individual effects,
through such studies.
A brief aside on the evolution of virulence and parasite
induced restrictions to host sociality
Although beyond the scope of the work presented in this dissertation, a
discussion of the effect of host sociality on host-parasite dynamics would be
incomplete without briefly considering its likely impact at the third level of
interaction (i.e. what happens when parasites are able to locally adapt?), as
well as the possible restrictions that parasite pressure impose on host social
system.
Theoretical observations suggest that the formation of tight-knit social
units may play an instrumental role in shaping the evolutionary trajectory of
parasite virulence. By subdividing the overall population into distinct social
units (i.e. creation of metapopulation), the transmission between individuals
within such units becomes analogous to vertical transmission, which is
expected to lead to lower virulence (Boots and Sasaki 1999). This is
supported by work in a moth-virus system, where virulence was indeed
shown to be lower in spatially structured populations (Boots and Mealor
2007). Interestingly, within social units, the increased social contact within
groups may similarly select for reduction of parasite virulence (Bonds et al.
2005). This is because virulence is predicted to be highest in cases where
host defense is achieved through resistance, as opposed to tolerance or
avoidance (Raberg et al. 2009). As resistance and avoidance mechanisms
become ineffective above a certain parasite prevalence threshold (Bonds et
al. 2005), tolerance mechanisms will often be favored in social hosts. In sum,
it therefore appears that host sociality per se may often favor the evolution of
reduced virulence in associated parasites, though the more subtle
contribution of individual aspects of host social system remain unexplored.
128
The obvious question remains whether parasite pressure ultimately
limits the breadth of social systems available to a host species, or whether a
host’s social system is adaptively shaped to minimize parasite pressure. In
1976, Freeland boldly stated that parasites and disease have had, and have,
an impact in determining many of the characteristics of primate sociality
(Freeland 1976). He subsequently outlined 16 hypotheses regarding specific
ways in which parasites and disease may influence host sociality in both
primates and mammals in general, concluding that behaviors that aid in
avoiding infection are numerous, but whether these mechanisms resulted
from targeted selection for disease avoidance remains to be tested (Freeland
1976). Forty years later, documented examples of host social and behavioral
avoidance mechanisms are abundant in all social taxa (e.g. Møller et al. 1993,
Moore 2002). Likewise, the effects that individual facets of host social system
have on parasite infection dynamics are broadly understood (Poulin and
Forbes 2012). However, evidence that parasites actively influence host
sociality remains rare. Nevertheless, recently there is an increased
recognition that the interaction between host ‘behavioral immune system’
and parasites is strongly reciprocal, and that hosts may indeed actively adapt
based on this feedback (Ezenwa et al. 2016). Perhaps the most convincing
evidence for restriction of host social system by parasites comes from a
recent work on Grant’s gazelle and its parasites (Ezenwa and Snider 2016). In
this system, Ezenwa and Snider found a reciprocal feedback between the
maintenance of male territories used to attract mates and parasite load.
Specifically, parasites were found to be a cost of territoriality, while
concurrently these parasites dampened the behaviors required for territory
maintenance, suggesting that the inter- and intra-individual variability in
territory defense (with its obvious direct implications on mating system) is
strongly influenced by parasite pressure (Ezenwa and Snider 2016).
Thus far, investigations on the interplay between host sociality and
parasite pressure, and the possible resulting restrictions to host social
system, have focused primarily on the direct effects of sociality on parasite
infection levels at an ecological level. Significantly, the possible role of
129
adaptive alterations and restrictions of host social system may extend to the
evolutionary timescale and its interaction with parasite virulence as outlined
in the first paragraph of this section (Ezenwa et al. 2016). In this extension,
the feedback between parasite pressure and host social system at an
evolutionary scale may restrict host sociality to those systems in which the
host’s primary parasite threats are either locally maladapted, or in which
parasite adaptation leads to the evolution of lower virulence levels. This
addition implies that the ‘behavioral immune system’ may act at both the
ecological and evolutionary timescales, and could be critical in reducing the
virulence of parasites and pathogens that would otherwise have the potential
to cause significantly more damage to their hosts. It goes without saying that
such adaptive constraints of host social system will only occur in situations
where parasites impose significant selective pressure on hosts, which
additionally cannot be compensated for by more flexible immediate
adaptations such as increased investment in immune response, or grooming
and avoidance behaviors (Friant et al. 2016, Villa et al. 2016).
Providing evidence for this hypothesis and establishing a causal link
between parasite infection and constraints to host social system on an
evolutionary scale is obviously complicated by several factors. For one, the
selective pressures that led to current social system will, in many cases, be
absent (the ghosts of parasites past, cf. Mooring et al. 2006). Additionally, it
is nearly impossible to disentangle the effects of parasites and disease from
the other factors that influence or constrain host social system, such as
predation pressure (Kappeler and van Schaik 2002) and the defensibility of
resources
and
mates
(Clutton-Brock
1989,
Clutton-Brock
2009).
Nevertheless, targeted comparative studies certainly have the potential to
further our understanding of host-parasite eco-evolutionary dynamics (e.g.
examples in Table 1 of Ezenwa et al. 2016). Ultimately, regardless of whether
these consequences have been actively selected for based on parasite and
pathogen pressure, or whether they are merely convenient byproducts, their
influence on host-parasite evolutionary dynamics remains.
130
Overall impact of host sociality on host-parasite evolutionary
dynamics
To summarize across the three levels of host-parasite interaction, several
important effects of host social system on host-parasite dynamics can be
recognized. At the individual level, sociality inherently increases the risk of
parasite exposure (Kappeler et al. 2015), as avoidance mechanisms are only
effective if the entire social unit is able to avoid infection. I illustrate this in
chapter 2, where increased host social aggregation and increased contact
during mating both lead to increased parasite infection intensity and
transmission. Conversely, the benefits of sociality increase the opportunity
for a wide range of behavioral and social adaptations, as well as
compensatory trade-offs, which increase the viability of tolerance as an
effective defense mechanism. In addition, as tolerance and behavioral /
social adaptations can be comparatively fast, plastic and effective against a
broader range of parasites as compared to immunological and physical
defenses (Medzhitov et al. 2012, Sears et al. 2015), their increased use may
reduce overall defense costs.
At the level of local genetic population structure, the impact of host
social system is again substantial. In chapters 3 and 4 almost completely
opposite parasite population genetic structures are observed, despite the
narrow range of parasite life histories and host social systems investigated
here.
As
population
genetic
structure
of
both
interacting
species
consequently determines local evolutionary dynamics, these striking
differences in parasite genetic structure are likely to subsequently lead to
similarly divergent evolutionary trajectories. These predicted consequences
for parasite local adaptation can theoretically be tested using transplant
experiments, although the feasibility of such experiments in the current
system may be complicated by difficulties in parasite marking (unpublished
data).
Finally, although not investigated in this system, the potential influence
and feedback between host social system and parasite infection and
131
virulence at the evolutionary level may be vastly underappreciated. Notably,
how host social system affects dynamics at this third level has momentous
consequences for our interpretation of the effects of host social system at
the level of the local genetic structure. In the case that host substructuring
into social units indeed leads to the evolutionary reduction of virulence, this
may mean that the consequences of local adaptation at second level are
exactly opposite to those currently accepted (i.e. local adaptation of the
parasite will lead to increased costs for the host).
In sum, the effects of host social system on parasite infection
intensity, transmission dynamics and population genetic structure observed
here clearly illustrate that host sociality strongly shapes host-parasite
evolutionary dynamics. Therefore, host sociality should be considered an
instrumental, albeit perhaps partially unintentional, component of host
defense against parasite and pathogen pressure.
Future directions
As demonstrated by the many unknowns and suppositions in the sections
above, our understanding of precisely how social hosts affect, and are
affected by, parasites on an evolutionary scale is still in its infancy.
The increased applicability of advanced genetic and genomic tools to
non-model species (and smaller organisms) has already greatly increased our
ability to investigate parasite population genetics, as exemplified by the
exponential increase in studies over the past few years (Mazé-Guilmo et al.
2016). A continuation of this trend, with special emphasis on comparative
studies characterizing parasite genetic structure across a diversity of host
systems, will help separate the effects of specific facets of host social
system and explore (the variation in) their interactions. Likewise, an increased
integration of local dynamics (landscape genetics) and broad scale patterns
and history (phylogeography) will help uncover to what extent local
132
adaptation generates and maintains differentiation at a broader scale (Rissler
2016).
As noted above, perhaps the biggest gap in our knowledge pertains to
our understanding of how host sociality affects local adaptation and the
evolution of virulence in parasites. Here too, advances in comparative
genomics, with a special emphasis on emerging parasite and pathogen
dynamics across a range of host social systems, will give us some insight
into how host sociality shapes the evolutionary outcome of these
interactions.
Such
studies
are
naturally
contingent
upon
prior
characterization of infection dynamics and local population genetic structure,
as performed in this dissertation. As such, the bat-ectoparasite model
presented here provides an ideal foundation to be expanded and continued
upon in this direction.
Finally, the integration of these insights into applied disease ecology
and conservation approaches will greatly improve the efficacy of these efforts
(Vander Wal et al. 2014a, Vander Wal et al. 2014b, Forbes et al. 2015). This
will be critical given the wide range of anthropogenic changes to ecosystems
(e.g. population reduction, fragmentation, introduction of exotic species,
climate change) that will affect species interactions (Lampert et al. 2016) and
host-parasite dynamics in particular (Longdon et al. 2015). As a result,
understanding how host sociality is regulated by, and itself affects, parasites
and pathogens will be instrumental to applied conservation efforts and the
control of disease across both animals and humans (Streicker et al. 2012).
133
134
List of publications
and selected abstracts
In addition to the chapters presented in this dissertation, I have (co-)authored
several other publications during my time as a PhD candidate concerning
parasites, bat behaviour and conservation. These are listed below, including
abstracts of the most pertinent publications.
2011 Isolation and characterisation of microsatellite loci for two species
of Spinturnicid bat wing mites (Spinturnix myoti and Spinturnix
bechsteini)
J. van Schaik, N. Bruyndonckx, G. Kerth, P. Christe
Acarologia 51:127-131
Part of MSc research and preamble to the work presented in this dissertation
Abstract
To investigate the potential for host-parasite coadaptation
between bats and their wing mites, we developed microsatellite loci for two
species of Spinturnix mites. For Spinturnix myoti, parasite of Myotis myotis,
we were able to develop nine polymorphic loci and screened them in 100
mites from five bat colonies. For S. bechsteini, parasite of M. bechsteinii, we
developed five polymorphic loci, which were also screened in 100 mites from
five bat colonies. In both species, all markers were highly polymorphic (22-46
and 6-23 alleles per locus respectively). The majority of markers for both
species exhibited departure from Hardy-Weinberg proportions (8 of 9 and 3
of 5, respectively). One marker pair in S. myoti showed evidence for linkage
disequilibrium. As the observed departures from Hardy-Weinberg proportions
are most likely a consequence of the biology of the mites, the described
microsatellite loci should be useful in studying population genetics and hostparasite dynamics of Spinturnix myoti and Spinturnix bechsteini in relation to
their bat hosts.
135
2012 Causes and consequences of living in closed societies: lessons
from a long-term socio-genetic study on Bechstein's bats
G. Kerth, J. van Schaik
Molecular Ecology 21:633-646
Invited Review
Abstract
Understanding the ecological, behavioural and genetic factors
influencing animal social systems is crucial to investigating the evolution of
sociality. Despite the recent advances in population genetic methods and the
analysis of social interactions, long-term studies exploring the causes and
consequences of social systems in wild mammals are rare. Here, we provide
a synthesis of 15 years of data on the Bechstein’s bat (Myotis bechsteinii), a
species that raises its young in closed societies of 10–45 females living
together for their entire lives and where immigration is virtually absent. We
discuss the potential causes and consequences of living in closed societies,
based on the available data on Bechstein’s bat and other species with similar
social systems. Using a combination of observational and genetic data on
the bats together with genetic data on an ecto-parasite, we suggest that
closed societies in Bechstein’s bats are likely caused by a combination of
benefits from cooperation with familiar colony members and parasite
pressure. Consequences of this peculiar social system include increased
sensitivity to demographic fluctuations and limits to dispersal during colony
foundation, which have broad implications for conservation. We also hope to
illustrate by synthesizing the results of this long-term study the diversity of
tools that can be applied to hypothesize about the factors influencing a
species’ social system. We are convinced that with the expansion of the
number of social mammals for which comparably detailed socio-genetic
long-term data are available, future comparative studies will provide deeper
insights into the evolution of closed societies.
136
2015 Bats Swarm Where They Hibernate: Compositional Similarity
between Autumn Swarming and Winter Hibernation Assemblages
at Five Underground Sites
J. van Schaik, R. Janssen, T. Bosch, A.-J. Haarsma,
J. J. A. Dekker, B. Kranstauber
PLoS ONE 10:e0130850
Analysis based on previous bat conservation work in Netherlands
Abstract
During autumn in the temperate zone of both the new and old
world, bats of many species assemble at underground sites in a behaviour
known as swarming. Autumn swarming behaviour is thought to primarily
serve as a promiscuous mating system, but may also be related to the
localization and assessment of hibernacula. Bats subsequently make use of
the same underground sites during winter hibernation, however it is currently
unknown if the assemblages that make use of a site are comparable across
swarming and hibernation seasons. Our purpose was to characterize the bat
assemblages found at five underground sites during both the swarming and
the hibernation season and compare the assemblages found during the two
seasons both across sites and within species. We found that the relative
abundance of individual species per site, as well as the relative proportion of
a species that makes use of each site, were both significantly correlated
between the swarming and hibernation seasons. These results suggest that
swarming may indeed play a role in the localization of suitable hibernation
sites. Additionally, these findings have important conservation implications,
as this correlation can be used to improve monitoring of underground sites
and predict the importance of certain sites for rare and cryptic bat species.
137
accepted
Swarming behaviour, catchment area and seasonal
movement patterns of the Bechstein’s bats: implications for
conservation
D. Dekeukeleire, R. Janssen, A.-J. Haarsma, T. Bosch, J. van Schaik
Acta Chiropterologica
Analysis of Bechstein’s bat swarming behaviour
Abstract
Bats display marked seasonality throughout the temperate zone
and use different habitats during different parts of the year. Unfortunately,
detailed information regarding seasonal distribution and movements is often
lacking, thereby hampering the development of adequate conservation
measures. In this study we used radio telemetry to track females of the
endangered Bechstein’s bat from autumn swarming sites to their summer
maternity colony ranges. We were able to tag 22 individuals, 18 of which
were subsequently recovered at 9 roost sites up to 20.58 km away. Females
from multiple colonies visited the same swarming site on a single night.
Concurrently, we recovered females from a single maternity colony at
different swarming sites on the same night. The catchment area of the
investigated swarming sites measured 27.05 km2, and was notably skewed
to the northwest. Tagged bats were recovered in forest fragments ranging in
size from 5.42 to 128.98 ha. Notably, all but one of the recovered roosts were
found in forests that have been continuously wooded since at least 1775.
Surveys during the summer at these sites confirmed the presence of
maternity colonies at six out of seven locations that could be investigated.
Our study exemplifies that characterizing the seasonal movement patterns,
and identifying the habitats critical to the conservation of an elusive species
throughout its annual cycle, can facilitate concerted conservation efforts.
138
Other Contributions
Dekeukeleire, D., R. Janssen, and J. van Schaik. 2013. Frequent melanism in
Geoffroy’s bat (Myotis emarginatus, Geoffroy 1806). Hystrix, 24:197198.
Kerth, G., D. Fleischmann, J. van Schaik, and M. Melber. 2013. From
behaviour and genetics to nature conservation: 20 years research on
Bechstein’s bats. Pages 35-50 in M. Dietz, editor. Populationsökologie
und Habitatansprüche der Bechsteinfledermaus Myotis bechsteinii.
Zarbock GmbH & Co. KG, Frankfurt.
Mohr J., C. Koch von Helversen, J. van Schaik, F. Mayer, S. P. Ripperger, D.
Josic, C. Strätz. accepted. Eine neue Fledermausart für Bayern – die
Nymphenfledermaus (Myotis alcathoe, Helversen & Heller 2001).
Nyctalus, 18:3-4
139
140
Bibliography
Alizon, S., J. C. de Roode, and Y. Michalakis. 2013. Multiple infections and
the evolution of virulence. Ecology Letters 16:556-567.
Alizon, S., A. Hurford, N. Mideo, and M. Van Baalen. 2009. Virulence
evolution and the trade-off hypothesis: history, current state of affairs
and the future. Journal of Evolutionary Biology 22:245-259.
Alizon, S., and Y. Michalakis. 2015. Adaptive virulence evolution: the good
old fitness-based approach. Trends in Ecology & Evolution 30:248254.
Altizer, S., A. Dobson, P. Hosseini, P. Hudson, M. Pascual, and P. Rohani.
2006. Seasonality and the dynamics of infectious diseases. Ecology
Letters 9:467-484.
Altizer, S., C. L. Nunn, P. H. Thrall, J. L. Gittleman, J. Antonovics, A. A.
Cunningham, A. P. Dobson, V. Ezenwa, K. E. Jones, A. B. Pedersen,
M. Poss, and J. R. C. Pulliam. 2003. Social Organization and Parasite
Risk in Mammals: Integrating Theory and Empirical Studies. Annual
Review of Ecology, Evolution, and Systematics 34:517-547.
Anderson, R. M., and R. M. May. 1982. Coevolution of hosts and parasites.
Parasitology 85:411-426.
Angell, R. L., R. K. Butlin, and J. D. Altringham. 2013. Sexual segregation and
flexible mating patterns in temperate bats. PLoS ONE 8:e54194.
Arneberg, P., A. Skorping, B. Grenfell, and A. F. Read. 1998. Host densities
as determinants of abundance in parasite communities. Proceedings
of the Royal Society of London B: Biological Sciences 265:1283-1289.
Baagøe, H. J. 2001. Myotis bechsteinii - Bechsteinfledermaus. Pages 443471 in F. Krapp, editor. Handbuch der Säugetiere Europas. AULAVerlag GmbH, Wiebelsheim.
Barrett, L. G., P. H. Thrall, J. J. Burdon, and C. C. Linde. 2008. Life history
determines genetic structure and evolutionary potential of hostparasite interactions. Trends in Ecology & Evolution 23:678-685.
Bates, D., M. Maechler, B. Bolker, and S. Walker. 2014. lme4: Linear mixedeffects models using Eigen and S4.
141
Best, A., S. Webb, A. White, and M. Boots. 2011. Host resistance and
coevolution in spatially structured populations. Proceedings of the
Royal Society B: Biological Sciences 278:2216-2222.
Billeter, S., D. Hayman, A. Peel, K. Baker, J. Wood, A. Cunningham, R. SuuIre, K. Dittmar, and M. Kosoy. 2012. Bartonella species in bat flies
(Diptera: Nycteribiidae) from western Africa. Parasitology 139:324-329.
Bonds, M. H., D. C. Keenan, A. J. Leidner, and P. Rohani. 2005. Higher
disease prevalence can induce greater sociality: a game theoretic
coevolutionary model. Evolution 59:1859-1866.
Boots, M., P. J. Hudson, and A. Sasaki. 2004. Large shifts in pathogen
virulence relate to host population structure. Science 303:842-844.
Boots, M., and M. Mealor. 2007. Local interactions select for lower pathogen
infectivity. Science 315:1284-1286.
Boots, M., and A. Sasaki. 1999. ‘Small worlds’ and the evolution of virulence:
infection occurs locally and at a distance. Proceedings of the Royal
Society of London B: Biological Sciences 266:1933-1938.
Boulinier, T., S. Kada, A. Ponchon, M. Dupraz, M. Dietrich, A. Gamble, V.
Bourret, O. Duriez, R. Bazire, J. Tornos, T. Tveraa, T. Chambert, R.
Garnier, and K. D. McCoy. 2016. Migration, Prospecting, Dispersal?
What Host Movement Matters for Infectious Agent Circulation?
Integrative and Comparative Biology.
Brooks, D. R., and D. A. McLennan. 1993. Parascript: parasites and the
language of evolution. Smithsonian Institution Press, Washington D.C.
Bruyndonckx, N., S. Dubey, M. Ruedi, and P. Christe. 2009a. Molecular
cophylogenetic relationships between European bats and their
ectoparasitic mites (Acari, Spinturnicidae). Molecular Phylogenetics
and Evolution 51:227-237.
Bruyndonckx, N., I. Henry, P. Christe, and G. Kerth. 2009b. Spatio temporal
population genetic structure of the parasitic mite Spinturnix bechsteini
is shaped by its own demography and the social system of its bat
host. Molecular Ecology 18:3581-3592.
Burland, T. M., and J. W. Wilmer. 2001. Seeing in the dark: molecular
approaches to the study of bat populations. Biological Reviews of the
Cambridge Philosophical Society 76:389-409.
Bush, A. O., K. D. Lafferty, J. M. Lotz, and A. W. Shostak. 1997. Parasitology
meets ecology on its own terms: Margolis et al. revisited. Journal of
Parasitology 83:575-583.
142
Caillaud, D., M. E. Craft, and L. A. Meyers. 2013. Epidemiological effects of
group size variation in social species. Journal of The Royal Society
Interface 10.
Castella, V., M. Ruedi, and L. Excoffier. 2001. Contrasted patterns of
mitochondrial and nuclear structure among nursery colonies of the bat
Myotis myotis. Journal of Evolutionary Biology 14:708-720.
Castella, V., M. Ruedi, L. Excoffier, C. Ibanez, R. Arlettaz, and J. Hausser.
2000. Is the Gibraltar Strait a barrier to gene flow for the bat Myotis
myotis (Chiroptera: Vespertilionidae)? Molecular Ecology 9:1761-1772.
Christe, P., R. Arlettaz, and P. Vogel. 2000. Variation in intensity of a parasitic
mite (Spinturnix myoti) in relation to the reproductive cycle and
immunocompetence of its bat host (Myotis myotis). Ecology Letters
3:207-212.
Christe, P., M. S. Giorgi, P. Vogel, and R. Arlettaz. 2003. Differential speciesspecific ectoparasitic mite intensities in two intimately coexisting
sibling bat species: resource-mediated host attractiveness or parasite
specialization? Journal of Animal Ecology 72:866-872.
Christe, P., O. Glaziot, G. Evanno, N. Bruyndonckx, G. Devevey, G. Yannic,
P. Patthey, A. Maeder, P. Vogel, and R. Arlettaz. 2007. Host sex and
ectoparasite choice: preference for, and higher survival on female
hosts. Journal of Animal Ecology 76:703-710.
Clayton, D. H., S. E. Bush, and K. P. Johnson. 2015. Coevolution of life on
hosts: integrating ecology and history. University of Chicago Press.
Clayton, D. H., and J. Moore. 1997. Host-parasite evolution: general
principles and avian models. Oxford University Press (OUP).
Clement, M., D. Posada, and K. A. Crandall. 2000. TCS: a computer program
to estimate gene genealogies. Molecular Ecology 9:1657-1659.
Clutton-Brock, J. 1989. Review lecture: mammalian mating systems.
Proceedings of the Royal Society B: Biological Sciences 236:339-372.
Clutton-Brock, T. 2009. Structure and function in mammalian societies.
Philosophical Transactions of the Royal Society B: Biological Sciences
364:3229-3242.
Côté, I. M., and R. Poulin. 1995. Parasitism and group size in social animals:
a meta-analysis. Behavioral Ecology 6:159-165.
Cressler, C. E., A. L. Graham, and T. Day. 2015. Evolution of hosts paying
manifold costs of defence. Proceedings of the Royal Society B:
Biological Sciences 282:20150065.
143
Criscione, C. D. 2008. Parasite co-structure: broad and local scale
approaches. Parasite 15:439-443.
Criscione, C. D., and M. S. Blouin. 2005. Effective sizes of macroparasite
populations: a conceptual model. Trends in Parasitology 21:212-217.
Criscione, C. D., R. Poulin, and M. S. Blouin. 2005. Molecular ecology of
parasites: elucidating ecological and microevolutionary processes.
Molecular Ecology 14:2247-2257.
Cross, P. C., J. Drewe, V. Patrek, G. Pearce, M. D. Samuel, and R. J.
Delahay. 2009. Wildlife population structure and parasite transmission:
implications for disease management. Pages 9-29 Management of
Disease in Wild Mammals. Springer.
Cross, P. C., J. O. Lloyd-Smith, P. L. F. Johnson, and W. M. Getz. 2005.
Duelling timescales of host movement and disease recovery determine
invasion of disease in structured populations. Ecology Letters 8:587595.
Deunff, J., and J. C. Beaucournu. 1981. Phenology and Variations of
Dermecos in Some Species of Spinturnicidae (Acarina, Mesostigmata).
Annales De Parasitologie Humaine Et Comparee 56:203-224.
Deunff, J., G. Walter, A. Bellido, and M. Volleth. 2004. Description of a cryptic
species, Spinturnix bechsteini n. sp. (Acari: Mesostigmata,
Spinturnicidae), parasite of Myotis bechsteinii (Kuhl, 1817) (Chiroptera:
Vespertilionidae) by using ecoethology of host bats and statistical
methods. Journal of Medical Entomology 41:826-832.
Dharmarajan, G., J. C. Beasley, and O. E. Rhodes, Jr. 2011. Heterozygote
deficiencies in parasite populations: an evaluation of interrelated
hypotheses in the raccoon tick, Ixodes texanus. Heredity 106:253-260.
Dick, C. W., and B. D. Patterson. 2006. Bat flies: obligate ectoparasites of
bats. Pages 179-194 in S. Morand, B. R. Krasnov, and R. Poulin,
editors. Micromammals and Macroparasites. Springer, Tokyo.
Dick, C. W., and B. D. Patterson. 2007. Against all odds: explaining high host
specificity in dispersal-prone parasites. International Journal for
Parasitology 37:871-876.
Dietz, C., O. von Helversen, and D. Nill. 2007. Handbuch der Fledermäuse
Europas und Nordwestafrikas. Franckh-Kosmos Verlags GmbH & Co.
KG, Stuttgart.
Dybdahl, M. F., and A. Storfer. 2003. Parasite local adaptation: Red Queen
versus Suicide King. Trends in Ecology & Evolution 18:523-530.
144
Earl, D. A., and B. M. vonHoldt. 2011. STRUCTURE HARVESTER: a website
and program for visualizing STRUCTURE output and implementing the
Evanno method. Conservation genetics resources 4:359-361.
Encarnação, J., and T. E. Reiners. 2012. Mating at summer sites: indications
from parentage analysis and roosting behaviour of Daubenton’s bats
(Myotis daubentonii). Conservation Genetics 13:1161-1165.
Encarnação, J. A., D. Baulechner, and N. I. Becker. 2012. Seasonal
Variations of Wing Mite Infestations in Male Daubenton's Bats (Myotis
daubentonii) in Comparison to Female and Juvenile Bats. Acta
Chiropterologica 14:153-159.
Evanno, G., S. Regnaut, and J. Goudet. 2005. Detecting the number of
clusters of individuals using the software STRUCTURE: a simulation
study. Molecular Ecology 14:2611-2620.
Evans, G. O. 1968. The external morphology of the post-embryonic
developmental stages of Spinturnix myoti kol. (Acari: Mesostigmata).
Acarologia 10:589-608.
Ewald, P. W. 1983. Host-parasite relations, vectors, and the evolution of
disease severity. Annual Review of Ecology and Systematics 14:465485.
Excoffier, L., G. Laval, and S. Schneider. 2005. Arlequin (version 3.0): An
integrated software package for population genetics data analysis.
Evolutionary Bioinformatics 1:47-50.
Excoffier, L., and H. E. L. Lischer. 2010. Arlequin suite ver 3.5: a new series
of programs to perform population genetics analyses under Linux and
Windows. Molecular Ecology Resources 10:564-567.
Ezenwa, V. O. 2004. Host social behavior and parasitic infection: a
multifactorial approach. Behavioral Ecology 15:446-454.
Ezenwa, V. O., E. A. Archie, M. E. Craft, D. M. Hawley, L. B. Martin, J. Moore,
and L. White. 2016. Host behaviour-parasite feedback: an essential
link between animal behaviour and disease ecology. Proceedings of
the Royal Society B: Biological Sciences 283.
Ezenwa, V. O., and M. H. Snider. 2016. Reciprocal relationships between
behaviour and parasites suggest that negative feedback may drive
flexibility in male reproductive behaviour. Proceedings of the Royal
Society B: Biological Sciences 283.
Fenton, M. B. 1969. Summer activity of Myotis lucifugus (Chiroptera:
Vespertilionidae) at hibernacula in Ontario and Quebec. Canadian
Journal of Zoology 47:597-602.
145
Forbes, K. M., H. Henttonen, V. Hirvela-Koski, A. Kipar, T. Mappes, P. Stuart,
and O. Huitu. 2015. Food provisioning alters infection dynamics in
populations of a wild rodent. Proceedings of the Royal Society B:
Biological Sciences 282.
Frantz, A. C., S. Bertouille, M. C. Eloy, A. Licoppe, F. Chaumont, and M. C.
Flamand. 2012. Comparative landscape genetic analyses show a
Belgian motorway to be a gene flow barrier for red deer (Cervus
elaphus), but not wild boars (Sus scrofa). Molecular Ecology 21:34453457.
Freeland, W. 1976. Pathogens and the evolution of primate sociality.
Biotropica 8:12-24.
Friant, S., T. E. Ziegler, and T. L. Goldberg. 2016. Changes in physiological
stress and behaviour in semi-free-ranging red-capped mangabeys
(Cercocebus torquatus) following antiparasitic treatment. Page
20161201 in Proceedings of the Royal Society B: Biological Sciences.
The Royal Society.
Gandon, S., A. Buckling, E. Decaestecker, and T. Day. 2008. Host-parasite
coevolution and patterns of adaptation across time and space. Journal
of Evolutionary Biology 21:1861-1866.
Gandon, S., D. Ebert, I. Olivieri, and Y. Michalakis. 1998. Differential
adaptation in spacially heterogeneous environments and host-parasite
coevolution. Pages 325-342 Genetic Structure and Local Adaptation
in Natural Insect Populations. Springer.
Gandon, S., and Y. Michalakis. 2002. Local adaptation, evolutionary potential
and host-parasite coevolution: interactions between migration,
mutation, population size and generation time. Journal of Evolutionary
Biology 15:451-462.
Gandon, S., and S. L. Nuismer. 2009a. Interactions between genetic drift,
gene flow, and selection mosaics drive parasite local adaptation. The
American Naturalist 173:212-224.
Gandon, S., and S. L. Nuismer. 2009b. Interactions between genetic drift,
gene flow, and selection mosaics drive parasite local adaptation. The
American Naturalist 173:212-224.
Gelman, A., Y.-S. Su, M. Yajima, J. Hill, M. G. Pittau, J. Kerman, T. Zheng,
and V. Dorie. 2014. Data Analysis Using Regression and
Multilevel/Hierarchical Models.
Giorgi, M. S., R. Arlettaz, P. Christe, and P. Vogel. 2001. The energetic
grooming costs imposed by a parasitic mite (Spinturnix myoti) upon its
bat host (Myotis myotis). Proceedings of the Royal Society B:
Biological Sciences 268:2071-2075.
146
Giorgi, M. S., R. Arlettaz, F. Guillaume, S. Nussle, C. Ossola, P. Vogel, and P.
Christe. 2004. Causal mechanisms underlying host specificity in bat
ectoparasites. Oecologia 138:648-654.
Godfrey, S. S., C. M. Bull, R. James, and K. Murray. 2009. Network structure
and parasite transmission in a group living lizard, the gidgee skink,
Egernia stokesii. Behavioral Ecology and Sociobiology 63:1045-1056.
Gómez-Díaz, E., J. González-Solís, M. A. Peinado, and R. D. M. Page. 2007.
Lack of host-dependent genetic structure in ectoparasites of
Calonectris shearwaters. Molecular Ecology 16:5204-5215.
Gomez-Diaz, E., J. A. Morris-Pocock, J. Gonzalez-Solis, and K. D. McCoy.
2012. Trans-oceanic host dispersal explains high seabird tick diversity
on Cape Verde islands. Biology Letters 8:616-619.
Goudet, J. 2002. FSTAT 2.9.3: a program to estimate and test gene
diversities and fixation indices (updated from Goudet 1995). Lausanne,
Switzerland.
Greischar, M. A., and B. Koskella. 2007. A synthesis of experimental work on
parasite local adaptation. Ecology Letters 10:418-434.
Guillot, G., F. Mortier, and A. Estoup. 2005. GENELAND: a computer
package for landscape genetics. Molecular Ecology Notes 5:712-715.
Güttinger, R., A. Zahn, F. Krapp, and W. Schober. 2001. Myotis myotis Großes Mausohr. Pages 111-121 in F. Krapp, editor. Handbuch der
Säugetiere Europas. AULA-Verlag GmbH, Wiebelsheim.
Hafer, N., and M. Milinski. 2016. Inter- and intraspecific conflicts between
parasites over host manipulation. Proceedings of the Royal Society B:
Biological Sciences 283:20152870.
Hafner, M. S., J. W. Demastes, T. A. Spradling, and D. L. Reed. 2003.
Cophylogeny between pocket gophers and chewing lice. Pages 195220 in R. D. M. Page, editor. Tangled Trees: Phylogeny, Cospeciation,
and Coevolution. University of Chicago Press.
Hamilton, W. 1982. Pathogens as causes of genetic diversity in their host
populations. Pages 269-296
Population Biology of Infectious
Diseases. Springer.
Hanski, I., and M. E. Gilpin. 1997. Metapopulation Biology. Academic Press.
Hart, B. L. 1990. Behavioral adaptations to pathogens and parasites: five
strategies. Neuroscience & Biobehavioral Reviews 14:273-294.
Hawkley, L. C., and J. P. Capitanio. 2015. Perceived social isolation,
evolutionary fitness and health outcomes: a lifespan approach.
147
Philosophical Transactions of the Royal Society B: Biological Sciences
370:20140114.
Hedrick, P. W. 2005. A standardized genetic differentiation measure.
Evolution 59:1633-1638.
Hoberg, E. P., and D. R. Brooks. 2008. A macroevolutionary mosaic: episodic
host-switching, geographical colonization and diversification in
complex host-parasite systems. Journal of Biogeography 35:15331550.
Hoberg, E. P., and D. R. Brooks. 2015. Evolution in action: climate change,
biodiversity dynamics and emerging infectious disease. Philosophical
Transactions of the Royal Society B: Biological Sciences 370.
Hoeksema, J. D., and S. E. Forde. 2008. A meta-analysis of factors affecting
local adaptation between interacting species. The American Naturalist
171:275-290.
Hutson, A. M. 1984. Keds, flat-flies and bat-flies. Diptera, Hippoboscidae and
Nycteribiidae. Handbooks for the Identification of British Insects, Royal
Entomological Society, UK 10.
Hutterer, R., T. Ivanova, C. Meyer-Cords, and L. Rodrigues. 2005. Bat
migrations in Europe: a review of banding data and literature. Federal
Agency for Nature Conservation, Bonn.
Kappeler, P. M., S. Cremer, and C. L. Nunn. 2015. Sociality and health:
impacts of sociality on disease susceptibility and transmission in
animal and human societies. Philosophical Transactions of the Royal
Society B: Biological Sciences 370.
Kappeler, P. M., and C. P. van Schaik. 2002. Evolution of primate social
systems. International Journal of Primatology 23:707-740.
Kerth, G. 2008. Causes and consequences of sociality in bats. Bioscience
58:737-746.
Kerth, G., B. Almasi, N. Ribi, D. Thiel, and S. Lüpold. 2003a. Social
interactions among wild female Bechstein's bats (Myotis bechsteinii)
living in a maternity colony. acta ethologica 5:107-114.
Kerth, G., A. Kiefer, C. Trappmann, and M. Weishaar. 2003b. High gene
diversity at swarming sites suggest hot spots for gene flow in the
endangered Bechstein's bat. Conservation Genetics 4:491-499.
Kerth, G., F. Mayer, and B. Konig. 2000. Mitochondrial DNA (mtDNA) reveals
that female Bechstein's bats live in closed societies. Molecular
Ecology 9:793-800.
148
Kerth, G., F. Mayer, and E. Petit. 2002a. Extreme sex-biased dispersal in the
communally breeding, nonmigratory Bechstein's bat (Myotis
bechsteinii). Molecular Ecology 11:1491-1498.
Kerth, G., and M. Melber. 2009. Species-specific barrier effects of a
motorway on the habitat use of two threatened forest-living bat
species. Biological Conservation 142:270-279.
Kerth, G., and L. Morf. 2004. Behavioral and genetic data suggest that
Bechstein's bat predominately mate outside the breeding habitat.
Ethology 110:987-999.
Kerth, G., B. Petrov, A. Conti, D. Anastasov, M. Weishaar, S. Gazaryan, J.
Jaquiery, B. Konig, N. Perrin, and N. Bruyndonckx. 2008. Communally
breeding Bechstein's bats have a stable social system that is
independent from the postglacial history and location of the
populations. Molecular Ecology 17:2368-2381.
Kerth, G., K. Safi, and B. Koenig. 2002b. Mean colony relatedness is a poor
predictor of colony structure and female philopatry in the communally
breeding Bechstein's bat (Myotis bechsteinii). Behavioral Ecology and
Sociobiology 52:203-210.
Kerth, G., and J. van Schaik. 2012. Causes and consequences of living in
closed societies: lessons from a long-term socio-genetic study on
Bechstein's bats. Molecular Ecology 21:633-646.
Krapp, F. 2004. Handbuch der Säugetiere Europas. Bd. 4/II: Fledertiere
(Chiroptera) II, Vespertilionidae 2, Mollosidae, Nycteridae. Aula Verlag,
Wiebelsheim.
Krasnov, B., I. Khokhlova, and G. Shenbrot. 2002. The effect of host density
on ectoparasite distribution: an example of a rodent parasitized by
fleas. Ecology 83:164-175.
Krasnov, B. R., S. Morand, H. Hawlena, I. S. Khokhlova, and G. I. Shenbrot.
2005. Sex-biased parasitism, seasonality and sexual size dimorphism
in desert rodents. Oecologia 146:209-217.
Krause, J., and G. D. Ruxton. 2002. Living in Groups. Oxford University
Press, Oxford.
Krebs, B. L., T. K. Anderson, T. L. Goldberg, G. L. Hamer, U. D. Kitron, C. M.
Newman, M. O. Ruiz, E. D. Walker, and J. D. Brawn. 2014. Host group
formation decreases exposure to vector-borne disease: a field
experiment in a 'hotspot' of West Nile virus transmission. Proceedings
of the Royal Society B: Biological Sciences 281:20141586.
Lafferty, K. D. 2009. The ecology of climate change and infectious diseases.
Ecology 90:888-900.
149
Lampert, A., A. Hastings, and J. Drake. 2016. Stability and distribution of
predator-prey systems: local and regional mechanisms and patterns.
Ecology Letters:n/a-n/a.
Lehtonen, J., and K. Jaatinen. 2016. Safety in numbers: the dilution effect
and other drivers of group life in the face of danger. Behavioral
Ecology and Sociobiology.
Lenski, R. E., and R. M. May. 1994. The evolution of virulence in parasites
and pathogens: reconciliation between two competing hypotheses.
Journal of Theoretical Biology 169:253-265.
Lentini, P. E., T. J. Bird, S. R. Griffiths, L. N. Godinho, and B. A. Wintle. 2015.
A global synthesis of survival estimates for microbats. Biology Letters
11:20150371.
Leo, N. P., J. M. Hughes, X. Yang, S. K. Poudel, W. G. Brogdon, and S. C.
Barker. 2005. The head and body lice of humans are genetically
distinct (Insecta: Phthiraptera, Pediculidae): evidence from double
infestations. Heredity 95:34-40.
Levin, I. I., and P. G. Parker. 2013. Comparative host–parasite population
genetic structures: obligate fly ectoparasites on Galapagos seabirds.
Parasitology 140:1061-1069.
Lion, S., and M. Boots. 2010. Are parasites ‘‘prudent’’ in space? Ecology
Letters 13:1245-1255.
Lion, S., and S. Gandon. 2015. Evolution of spatially structured host-parasite
interactions. Journal of Evolutionary Biology 28:10-28.
Loehle, C. 1995. Social barriers to pathogen transmission in wild animal
populations. Ecology 76:326-335.
Longdon, B., J. D. Hadfield, J. P. Day, S. C. Smith, J. E. McGonigle, R.
Cogni, C. Cao, and F. M. Jiggins. 2015. The causes and
consequences of changes in virulence following pathogen host shifts.
PLoS Pathogens 11:e1004728.
Lourenço, S. I., and J. M. Palmeirim. 2007. Can mite parasitism affect the
condition of bat hosts? Implications for the social structure of colonial
bats. Journal of Zoology 273:161-168.
Lourenço, S. I., and J. M. Palmeirim. 2008. Which factors regulate the
reproduction of ectoparasites of temperate-zone cave-dwelling bats?
Parasitology Research 104:127-134.
Lučan, R. K. 2006. Relationships between the parasitic mite Spinturnix
andegavinus (Acari: Spinturnicidae) and its bat host, Myotis
daubentonii (Chiroptera: Vespertilionidae): seasonal, sex-and age150
related variation in infestation and possible impact of the parasite on
the host condition and roosting behaviour. Folia Parasitologica 53:147.
MacIntosh, A. J., A. Jacobs, C. Garcia, K. Shimizu, K. Mouri, M. A. Huffman,
and A. D. Hernandez. 2012. Monkeys in the middle: parasite
transmission through the social network of a wild primate. PLoS ONE
7:e51144.
May, R. M., and R. M. Anderson. 1990. Parasite—host coevolution.
Parasitology 100:S89-S101.
Mazé-Guilmo, E., S. Blanchet, K. D. McCoy, G. Loot, and P. Thrall. 2016.
Host dispersal as the driver of parasite genetic structure: a paradigm
lost? Ecology Letters 19:336-347.
McCoy, K. D., T. Boulinier, S. Schjorring, and Y. Michalakis. 2002. Local
adaptation of the ectoparasite Ixodes uriae to its seabird host.
Evolutionary Ecology Research 4:441-456.
McCoy, K. D., T. Boulinier, and C. Tirard. 2005. Comparative host-parasite
population structures: disentangling prospecting and dispersal in the
black-legged kittiwake Rissa tridactyla. Molecular Ecology 14:28252838.
McCoy, K. D., C. Tirard, and Y. Michalakis. 2003. Spatial genetic structure of
the ectoparasite Ixodes uriae within breeding cliffs of its colonial
seabird host. Heredity 91:422-429.
McNew, G. 1960. The nature, origin, and evolution of parasitism. Pages 1969 Plant pathology: an advanced treatise.
Medzhitov, R., D. S. Schneider, and M. P. Soares. 2012. Disease tolerance as
a defense strategy. Science 335:936-941.
Megali, A., G. Yannic, and P. Christe. 2011. Disease in the dark: molecular
characterization of Polychromophilus murinus in temperate zone bats
revealed a worldwide distribution of this malaria-like disease.
Molecular Ecology 20:1039-1048.
Meirmans, P. G., and P. H. Van Tienderen. 2004. GENOTYPE and
GENODIVE: two programs for the analysis of genetic diversity of
asexual organisms. Molecular Ecology Notes 4:792-794.
Miller, M. R., A. White, and M. Boots. 2005. The evolution of host resistance:
tolerance and control as distinct strategies. Journal of Theoretical
Biology 236:198-207.
Miller, S., D. Dykes, and H. Polesky. 1988. A simple salting out procedure for
extracting DNA from human nucleated cells. Nucleic acids research
16:1215.
151
Møller, A. P., R. Dufva, and K. Allander. 1993. Parasites and the evolution of
host social behavior. Advances in the Study of Behavior 22:6040560402.
Moore, J. 2002. Parasites and the behavior of animals. Oxford University
Press.
Mooring, M. S., B. L. Hart, T. A. Fitzpatrick, D. D. Reisig, T. T. Nishihira, I. C.
Fraser, and J. E. Benjamin. 2006. Grooming in desert bighorn sheep
(Ovis canadensis mexicana) and the ghost of parasites past.
Behavioral Ecology 17:364-371.
Nadler, S. A. 1995. Microevoltion and the genetic structure of parasite
populations. Journal of Parasitology 81:395-403.
Nuismer, S. L., and S. Gandon. 2008. Moving beyond common-garden and
transplant designs: insight into the causes of local adaptation in
species interactions. The American Naturalist 171:658-668.
Nuismer, S. L., R. Gomulkiewicz, and B. J. Ridenhour. 2010. When is
correlation coevolution? The American Naturalist 175:525-537.
Nunn, C. L. 2012. Primate disease ecology in comparative and theoretical
perspective. American Journal of Primatology 74:497-509.
Nunn, C. L., and S. Altizer. 2006. Infectious Diseases in Primates: Behavior,
Ecology and Evolution. Oxford University Press, Oxford, UK.
Nunn, C. L., F. Jordan, C. M. McCabe, J. L. Verdolin, and J. H. Fewell. 2015.
Infectious disease and group size: more than just a numbers game.
Philosophical Transactions of the Royal Society B: Biological Sciences
370.
Nunn, C. L., P. H. Thrall, and P. M. Kappeler. 2014. Shared resources and
disease dynamics in spatially structured populations. Ecological
Modelling 272:198-207.
Parker, B. J., S. M. Barribeau, A. M. Laughton, J. C. de Roode, and N. M.
Gerardo. 2011. Non-immunological defense in an evolutionary
framework. Trends in Ecology & Evolution 26:242-248.
Patterson, J. E., and K. E. Ruckstuhl. 2013. Parasite infection and host group
size: a meta-analytical review. Parasitology 140:803-813.
Peakall, R., and P. E. Smouse. 2006. GENALEX 6: genetic analysis in Excel.
Population genetic software for teaching and research. Molecular
Ecology Notes 6:288-295.
152
Penczykowski, R. M., A.-L. Laine, and B. Koskella. 2016. Understanding the
ecology and evolution of host-parasite interactions across scales.
Evolutionary Applications 9:37-52.
Perez-Rodriguez, A., I. de la Hera, S. Bensch, and J. Perez-Tris. 2015.
Evolution of seasonal transmission patterns in avian blood-borne
parasites. International Journal for Parasitology 45:605-611.
Piksa, K., W. Bogdanowicz, and A. Tereba. 2011. Swarming of Bats at
Different
Elevations
in
the
Carpathian
Mountains.
Acta
Chiropterologica 13:113-122.
Postawa, T., and A. Szubert-Kruszyńska. 2014. Is parasite load dependent
on host aggregation size? The case of the greater mouse-eared bat
Myotis myotis (Mammalia: Chiroptera) and its parasitic mite Spinturnix
myoti (Acari: Gamasida). Parasitology Research 113:1803-1811.
Poulin, R. 2011. Evolutionary ecology of parasites. 3rd edition. Princeton
university press, Princeton, NJ.
Poulin, R., and M. R. Forbes. 2012. Meta-analysis and research on host–
parasite interactions: past and future. Evolutionary Ecology 26:11691185.
Poulin, R., B. R. Krasnov, D. Mouillot, and D. W. Thieltges. 2011. The
comparative ecology and biogeography of parasites. Philosophical
Transactions of the Royal Society B: Biological Sciences 366:23792390.
Price, P. W. 1977. General concepts on evolutionary biology of parasites.
Evolution 31:405-420.
Pritchard, J. K., M. Stephens, and P. Donnelly. 2000. Inference of population
structure using multilocus genotype data. Genetics 155:945-959.
R Development Core Team. 2015. R: A language and environment for
statisical computing. R Foundation for Statisical Computing, Vienna,
Austria.
Raberg, L., A. L. Graham, and A. F. Read. 2009. Decomposing health:
tolerance and resistance to parasites in animals. Philosophical
Transactions of the Royal Society B: Biological Sciences 364:37-49.
Real, L. A., and R. Biek. 2007. Spatial dynamics and genetics of infectious
diseases on heterogeneous landscapes. Journal of The Royal Society
Interface 4:935-948.
Reckardt, K., and G. Kerth. 2006. The reproductive success of the parasitic
bat fly Basilia nana (Diptera : Nycteribiidae) is affected by the low roost
153
fidelity of its host, the Bechstein's bat (Myotis bechsteinii).
Parasitology Research 98:237-243.
Reckardt, K., and G. Kerth. 2007. Roost selection and roost switching of
female Bechstein's bats (Myotis bechsteinii) as a strategy of parasite
avoidance. Oecologia 154:581-588.
Reckardt, K., and G. Kerth. 2009. Does the mode of transmission between
hosts affect the host choice strategies of parasites? Implications from
a field study on bat fly and wing mite infestation of Bechstein's bats.
Oikos 118:183-190.
Reynolds, J. J., B. T. Hirsch, S. D. Gehrt, and M. E. Craft. 2015. Raccoon
contact networks predict seasonal susceptibility to rabies outbreaks
and limitations of vaccination. Journal of Animal Ecology.
Rifkin, J. L., C. L. Nunn, and L. Z. Garamszegi. 2012. Do animals living in
larger groups experience greater parasitism? A meta-analysis. The
American Naturalist 180:70-82.
Rigaud, T., M. J. Perrot-Minnot, and M. J. Brown. 2010. Parasite and host
assemblages: embracing the reality will improve our knowledge of
parasite transmission and virulence. Proceedings of the Royal Society
B: Biological Sciences 277:3693-3702.
Rimbach, R., D. Bisanzio, N. Galvis, A. Link, A. Di Fiore, and T. R. Gillespie.
2015. Brown spider monkeys (Ateles hybridus): a model for
differentiating the role of social networks and physical contact on
parasite transmission dynamics. Philosophical Transactions of the
Royal Society B: Biological Sciences 370.
Rissler, L. J. 2016. Union of phylogeography and landscape genetics.
Proceedings of the National Academy of Sciences 113:8079-8086.
Rivers, N. M., R. K. Butlin, and J. D. Altringham. 2005. Genetic population
structure of Natterer's bats explained by mating at swarming sites and
philopatry. Molecular Ecology 14:4299-4312.
Rousset, F. 1997. Genetic differentiation and estimation of gene flow from Fstatistics under isolation by distance. Genetics 145:1219-1228.
Rousset, F. 2008. GENEPOP ' 007: a complete re-implementation of the
GENEPOP software for Windows and Linux. Molecular Ecology
Resources 8:103-106.
Rubenstein, D. I., and M. E. Hohmann. 1989. Parasites and Social-Behavior
of Island Feral Horses. Oikos 55:312-320.
154
Rudnick, A. 1960. A revision of the mites of the family Spinturnicidae
(Acarina). University of California Publications in Entomology 17:157284.
Ruedi, M., S. Walter, M. C. Fischer, D. Scaravelli, L. Excoffier, and G. Heckel.
2008. Italy as a major ice age refuge area for the bat Myotis myotis
(Chiroptera : Vespertilionidae) in Europe. Molecular Ecology 17:18011814.
Safi, K., and G. Kerth. 2007. Comparative analyses suggest that information
transfer promoted sociality in male bats in the temperate zone. The
American Naturalist 170:465-472.
Sambrook, J. E., F. Fritsch, and T. Manitatis. 1989. Molecular cloning: a
laboratory manual. Cold Spring Harbor Laboratory Press 2 ed.:New
York.
Schalk, G., and M. R. Forbes. 1997. Male biases in parasitism of mammals:
Effects of study type, host age, and parasite taxon. Oikos 78:67-74.
Schaller, M., D. R. Murray, and A. Bangerter. 2015. Implications of the
behavioural immune system for social behaviour and human health in
the modern world. Philosophical Transactions of the Royal Society B:
Biological Sciences 370:20140105.
Schmid-Hempel, P. 2011. Evolutionary parasitology: the integrated study of
infections, immunology, ecology, and genetics. Oxford University
Press New York.
Schöner, C. R., M. G. Schöner, and G. Kerth. 2010. Similar is not the same:
social calls of conspecifics are more effective in attracting wild bats to
day roosts than those of other bat species. Behavioral Ecology and
Sociobiology 64:2053-2063.
Schuelke, M. 2000. An economic method for the fluorescent labeling of PCR
fragments. Nature Biotechnology 18:233-234.
Sears, B. F., P. W. Snyder, and J. R. Rohr. 2015. Host life history and hostparasite syntopy predict behavioural resistance and tolerance of
parasites. Journal of Animal Ecology 84:625-636.
Senior, P., R. K. Butlin, and J. D. Altringham. 2005. Sex and segregation in
temperate bats. Proceedings of the Royal Society B: Biological
Sciences 272:2467-2473.
Simons, C., F. Frati, A. Beckenbach, B. Crespi, H. Liu, and P. Floors. 1994.
Evolution, weighting, and phylogenetic utility of mitochondrial gene
sequences and a compilation of conserved polymerase chain reaction
primers. Annals of the Entomological Society of America 87:651-701.
155
Smith, T. 1904. Some problems in the life history of pathogenic
microorganisms. Science:817-832.
Sorci, G., A. P. Moller, and T. Boulinier. 1997. Genetics of host-parasite
interactions. Trends in Ecology & Evolution 12:196-200.
Springer, A., A. Mellmann, C. Fichtel, and P. M. Kappeler. 2016. Social
structure and Escherichia coli sharing in a group-living wild primate,
Verreaux's sifaka. BMC Ecology 16:6.
Streicker, D. G., S. Recuenco, W. Valderrama, J. Gomez Benavides, I.
Vargas, V. Pacheco, R. E. Condori Condori, J. Montgomery, C. E.
Rupprecht, P. Rohani, and S. Altizer. 2012. Ecological and
anthropogenic drivers of rabies exposure in vampire bats: implications
for transmission and control. Proceedings of the Royal Society B:
Biological Sciences 279:3384-3392.
Swift, S. M. 1997. Roosting and foraging behaviour of Natterer's bats (Myotis
nattereri) close to the northern border of their distribution. Journal of
Zoology 242:375-384.
Theodor, O. 1967. An Illustrated Catalogue of the Rothschild Collection of
Nycteribiidae (Diptera) in the British Museum (Natural History).
Publication 655. London: Trustees of the British Musem (Natural
History) 8:5.
Thompson, J. N. 1994. The Coevolutionary Process. University of Chicago
Press.
Toon, A., and J. M. Hughes. 2008. Are lice good proxies for host history? A
comparative analysis of the Australian magpie, Gymnorhina tibicen,
and two species of feather louse. Heredity 101:127-135.
Tung, J., L. B. Barreiro, M. B. Burns, J. C. Grenier, J. Lynch, L. E. Grieneisen,
J. Altmann, S. C. Alberts, R. Blekhman, and E. A. Archie. 2015. Social
networks predict gut microbiome composition in wild baboons. Elife 4.
van Oosterhout, C., W. F. Hutchinson, D. P. M. Wills, and P. Shipley. 2004.
Microchecker: Software for identifying and correcting genotyping
errors in microsatellite data. Molecular Ecology Notes 4:535-538.
van Schaik, J., N. Bruyndonckx, G. Kerth, and P. Christe. 2011. Isolation and
characterisation of microsatellite loci for two species of Spinturnicid
bat wing mites (Spinturnix myoti and Spinturnix bechsteini). Acarologia
51:127-131.
van Schaik, J., D. Dekeukeleire, and G. Kerth. 2015a. Host and parasite life
history interplay to yield divergent population genetic structures in two
ectoparasites living on the same bat species. Molecular Ecology
24:2324-2335.
156
van Schaik, J., R. Janssen, T. Bosch, A.-J. Haarsma, J. J. A. Dekker, and B.
Kranstauber. 2015b. Bats Swarm Where They Hibernate:
Compositional Similarity between Autumn Swarming and Winter
Hibernation Assemblages at Five Underground Sites. PLoS ONE
10:e0130850.
van Schaik, J., G. Kerth, N. Bruyndonckx, and P. Christe. 2014. The effect of
host social system on parasite population genetic structure:
comparative population genetics of two ectoparasitic mites and their
bat hosts. BMC evolutionary biology 14:18.
van Velzen, E., S. Pérez-Vila, and R. S. Etienne. 2016. The Role of WithinHost Competition for Coexistence in Multiparasitoid-Host Systems.
The American Naturalist 187:48-59.
Vander Wal, E., D. Garant, S. Calme, C. A. Chapman, M. Festa-Bianchet, V.
Millien, S. Rioux-Paquette, and F. Pelletier. 2014a. Applying
evolutionary concepts to wildlife disease ecology and management.
Evolutionary Applications 7:856-868.
Vander Wal, E., D. Garant, and F. Pelletier. 2014b. Evolutionary perspectives
on wildlife disease: concepts and applications. Evolutionary
Applications 7:715-722.
VanderWaal, K. L., V. Obanda, G. P. Omondi, B. McCowan, H. Wang, H.
Fushing, and L. A. Isbell. 2016. The “strength of weak ties” and
helminth parasitism in giraffe social networks. Behavioral
Ecology:arw035.
Villa, S. M., G. B. Goodman, J. S. Ruff, and D. H. Clayton. 2016. Does
allopreening
control
avian
ectoparasites?
Biology
Letters
12:20160362.
Webber, Q. M., Z. J. Czenze, and C. K. Willis. 2015. Host demographic
predicts ectoparasite dynamics for a colonial host during prehibernation mating. Parasitology 142:1260-1269.
Whitehead, H., and S. Dufault. 1999. Techniques for Analyzing Vertebrate
Social Structure Using Identified Individuals: Review. Advances in the
Study of Behavior 28:33.
Whiteman, N. K., R. T. Kimball, and P. G. Parker. 2007. Co-phylogeography
and comparative population genetics of the threatened Galápagos
hawk and three ectoparasite species: ecology shapes population
histories within parasite communities. Molecular Ecology 16:47594773.
Windsor, D. A. 1998. Controversies in parasitology, Most of the species on
Earth are parasites. International Journal for Parasitology 28:19391941.
157
Wobeser, G. A. 2007. Disease in wild animals: Investigation and
Management. 2nd edition. Springer, Berlin.
Wojciechowski, M. S., M. Jefimow, and E. Tegowska. 2007. Environmental
conditions, rather than season, determine torpor use and temperature
selection in large mouse-eared bats (Myotis myotis). Comparative
Biochemistry and Physiology: Part A 147:828-840.
Wolinska, J., and K. C. King. 2009. Environment can alter selection in hostparasite interactions. Trends in Parasitology 25:236-244.
Worthington-Wilmer, J., and E. Barratt. 1996. A non-lethal method of tissue
sampling for genetic studies of Chiropterans. Bat Research News
37:1-3.
Yeaman, S. 2015. Local Adaptation by Alleles of Small Effect*. The American
Naturalist:S000-S000.
Yee, T. W., and C. J. Wild. 2010. VGAM: Vector Generalized Linear and
Additive Models.
Zahn, A., and B. Dippel. 1997. Male roosting habits and mating behaviour of
Myotis myotis. Journal of Zoology 243:659-674.
Zahn, A., and D. Rupp. 2004. Ectoparasite load in European vespertilionid
bats. Journal of Zoology 262:383-391.
Zohdy, S., A. D. Kemp, L. A. Durden, P. C. Wright, and J. Jernvall. 2012.
Mapping the social network: tracking lice in a wild primate
(Microcebus rufus) population to infer social contacts and vector
potential. BMC Ecology 12:4.
Zuk, M., and K. A. McKean. 1996. Sex differences in parasite infections:
patterns and processes. International Journal for Parasitology
26:1009-1023.
158
159
Author contributions
The ideas presented in this dissertation have been developed by me and
extensively discussed and refined with Gerald Kerth. All chapters and
‘general parts’ of this dissertation were written by me and commented on by
co-authors. An itemized list of contributions per chapter can be found below.
Chapter 2: JvS designed the study, performed the fieldwork, and analyzed
the data. JvS wrote the paper with input and discussion from GK.
Chapter 3: All authors designed the research and participated in sample
collection; JvS and NB performed the molecular genotyping, sequencing and
data analysis; and JvS wrote the paper with input from all coauthors.
Chapter 4: The study was designed by JvS and GK; field sampling was
managed by GK and JvS; DNA extraction and genotyping was conducted by
DD; data analysis was performed by JvS and DD. The manuscript was
written by JvS with input from all authors.
This work would not have been possible without the aid of many additional
scientists who aided in collecting samples in the field or provided other
assistance. Please refer to the acknowledgement statements of the individual
chapters for details.
160
Acknowledgements
I am indebted to a great many people who helped make this dissertation
possible.
Scientific research would not be possible without funding. Therefore, I would
first sincerely like to thank the Volkswagen Foundation and the IMPRS for
Organismal Biology for giving me the opportunity to carry out my work.
Similarly, funding would not help much without a great support staff to make
it through the bureaucracy and paperwork. A special thanks to Dr. Grewe,
Daniel, Mäggi, Carmen, Ilse and the entire Seewiesen administration team for
facilitating and allowing me to focus on the fun stuff instead.
Dear Gerald, I could not have wished for a better mentor, friend and
supervisor. Thank you for your endless patience, despite my occasional
stubbornness, side-project distractions and disregard for deadlines. Thanks
also for your keen insight, helpful comments, fruitful discussions, many
phone calls and overall support! Even from a distance, it’s been an absolute
pleasure working with you.
This project would also not have been possible with Bart and his entire
department who supported me in Seewiesen. I would also like to thank Bart,
as well as Philippe and Karen for their wise words and guidance as advisory
committee members, and Martin, for agreeing to be my supervisor in
Konstanz.
Next, a big thanks to Daan, for your truly exemplary MSc project, your
profound patience in dealing with fly morphology and tricky primers, and for
all the other interesting discussions and collaborations.
161
Before going any further, I should thank the many people that helped lay the
foundation for this work. Especially the Dutch bat crew: René, Bart, A-J, Thijs
and Jan, for introducing me to the dark side and continuing to collaborate
together on all sorts of adventures ever since. Thanks also to my thematic
predecessors, Nadia and Karsten, who provided the foundation necessary
for my work and left hundreds of assorted parasite samples in the freezers
for me to work with.
I am eternally grateful to the countless people who assisted me in the field.
The Würzburg crew: Markus, Dani and the army of students who have helped
collect parasites over the years. Frauke, Lena and Johannes for their
immense hospitality in Münster and Forchheim, and for sharing their
incredible field sites and knowledge with me. Finally, to Matthias for help
obtaining permits to make all my adventures legal.
After fieldwork comes lab work, and here too, I am indebted to many people.
Particularly to Jari, Sylvia (and the whole Seewiesen lab team) and Ina for
their instruction, advice and support in all things molecular.
A big thanks to all my friends, who helped in all sorts of various ways, from
help with data analysis to proof-reading to brainstorming to simply being
great friends: Leo, Stefan, Robin, Jacky, Uli, Lisa, Bryson, Ryan, the entire
Seewiesen United and 3M/ESPE football teams, Caro, Michael, Veronika,
Sébastien, and the many others, you know who you are!
And of course to Susi, for being a friend like no other.
Finally, to my family, for making it seem like the most natural thing in the
world to follow your passion and become a biologist.
Thank you, Dankjewel, Danke
162
Curriculum vitae
Antoon Jacobus (Jaap) van Schaik
Date of birth:
Current affiliation:
E-mail:
21.11.1986
Max Planck Institute for Ornithology, Germany
University of Greifswald, Germany
[email protected]
Education
2010 - 2016 PhD
Max Planck Institute for Ornithology, Seewiesen, Germany
“The influence of host social system on parasite genetic
population structure.”
2007 - 2009 MSc (cum laude)
University of Groningen, The Netherlands
Top Master program Evolutionary Ecology
2004-2007
Bsc (with honors)
University College Utrecht, The Netherlands
Specialization in Molecular and Developmental Biology
2004
High School (honors arts distinction)
Durham School of the Arts, North Carolina, USA
Academic Awards and Scholarships
2014
Awarded 1 year PhD Scholarship; International Max Planck
Research School for Organismal Biology
2010
Awarded 3,5 year PhD Scholarship; Volkswagen Stiftung.
2010
Best Student Talk; 14th International Bat Research Meeting,
Prague, CZ
163
Scientific Activities
Representation
2011 - 2012 Student representative on the Board of the International Max
Planck Research School for Organismal Biology
Course participation and organization
2010 - 2016 Attended: Scientific Integrity and Ethics, Scientific Illustration,
Scientific Presentation, Scientific Writing, Bird Ringing, Spatial
statistics in R
2012
Organized: Joint symposium between the International Max
Planck Research Schools for Organismal Biology and
Evolutionary Biology
Teaching
2009 - 2012 Co-organizer and instructor at annual field course on bat
capture and handling
2011 - 2015 Annual lecture on Conservation in the Benelux, University of
Greifswald
2013
Supervision of MSc student
2016
Field course in Bat biology
Public outreach
2012
Volunteer at “Tag der offnen Tür” at MPIO Seewiesen
2014
Public lecture in series “Organismen als Lebensraum”,
University of Greifswald
Peer review
Publons ID: 752323
2010 Referee for: Acta Chiropterologica, Animal Behaviour,
Behavioral Ecology, Behavioral Ecology and Sociobiology,
Conservation Genetics, Journal of Mammalogy, Molecular
Ecology, Journal of Heredity
Scientific Skills
Languages
English (native), Dutch (native), German (B2), Spanish (A2)
Practical
Field skills: Bat capture and handling, Bird ringing, General natural
history
164
Laboratory skills: DNA extraction, Sequencing and alignment,
Microsatellite development and scoring
Computer
Basic skills: Microsoft Office, Adobe Photoshop and Illustrator
Statistical analysis: R (basic statistics, general linear modeling,
similarity indices), Population genetics software (Genemapper,
Genodive, Structure, Genepop, fstat, arlequin, CodonCode Aligner)
Publications
ORCID ID: 0000-0003-4825-7676
ResearchGate: https://www.researchgate.net/profile/Jaap_van_Schaik
Google Scholar: https://scholar.google.de/citations?user=acnwIxQAAAAJ
Publications submitted for peer review
van Schaik J, Kerth G (submitted) Host social organization and mating
system shape parasite transmission opportunities in three European
bat species
Peer reviewed publications
Dekeukeleire D, Janssen R, Haarsma A-J, Bosch T, van Schaik J (2016)
Detecting bat maternity colonies in a highly fragmented landscape via
radio tracking from autumn swarming sites. Acta Chiropterologica, in
press
Mohr J, Koch von Helversen C, van Schaik J, Mayer F, Ripperger SP, Josic
D, Strätz C (accepted) Eine neue Fledermausart für Bayern – die
Nymphenfledermaus (Myotis alcathoe, Helversen & Heller 2001).
Nyctalus, 18:3-4
van Schaik J, Janssen R, Bosch T, Haarsma A-J, Dekker JJA, Kranstauber B
(2015) Bats swarm where they hibernate: compositional similarity
between autumn swarming and winter hibernation assemblages at five
underground sites. PloS one 10.7
van Schaik J, Dekeukeleire D, Kerth G (2015) Host and parasite life history
interplay to yield divergent population genetic structures in two
ectoparasites living on the same bat species. Molecular Ecology
DOI: 10.1111/mec.13171
165
van Schaik J, Kerth G, Bruyndonckx N, Christe P (2014) The effect of host
social system on parasite population genetic structure: comparative
population genetics of two ectoparasitic mites and their bat hosts.
BMC evolutionary biology 14, 18.
Dekeukeleire D, Janssen R, van Schaik J (2013) Frequent melanism in
Geoffroy’s bat (Myotis emarginatus, Geoffroy 1806). Hystrix, the Italian
Journal of Mammalogy 24, 197-198.
Kerth G, Fleischmann D, van Schaik J, Melber M (2013) From behaviour and
genetics to nature conservation: 20 years research on Bechstein’s
bats.
In:
Populationsökologie
und
Habitatansprüche
der
Bechsteinfledermaus Myotis bechsteinii (ed. Dietz M), pp. 35-50.
Zarbock GmbH & Co. KG, Frankfurt.
Kerth G, van Schaik J (2012) Causes and consequences of living in closed
societies: lessons from a long-term socio-genetic study on
Bechstein's bats. Molecular Ecology 21, 633-646.
van Schaik J, Bruyndonckx N, Kerth G, Christe P (2011) Isolation and
characterisation of microsatellite loci for two species of Spinturnicid
bat wing mites (Spinturnix myoti and Spinturnix bechsteini). Acarologia
51, 127-131.
Coyer JA, Hoarau G, van Schaik J, Luijckx P, Olsen JL (2011) Trans‐Pacific
and trans‐Arctic pathways of the intertidal macroalga Fucus distichus
L. reveal multiple glacial refugia and colonizations from the North
Pacific to the North Atlantic. Journal of Biogeography 38, 756-771.
Non peer reviewed publications
van der Plas ALD, van Schaik J (2010) Determinatie tabel tot de boom- en
bodemwantsen van Nederland. Jeugdbondsuitgeverij. ISBN: 978-905107-044-6
[translation: key to the stink bugs of Netherlands]
Haarsma, A-J, van Schaik J, Bosch T, Janssen R (2015). The Bat Catcher's
Manual: Manual for assessment of reproductive status, age and
health in European bats. Electronic publication: www.vleermuizen
vangen.nl
Conference contributions
2010
International Berlin Bat Meeting – Talk
“A comparison of genetic population structure in host and
parasite”
2010
IMPRS Grand Challenges Symposium – Talk
“Might mites be a mighty source of information”
2011
Meeting of German Bat Researchers – Poster
166
“Bat-Mite genetic population structure”
2011
Fachtagung Bechsteinsfledermaus - Talk
“Interaktion der Bechsteinfledermaus mit ihren Parasiten”
2011
European Society for Evolutionary Biology – Poster
“A comparison of genetic population structure in two hostparasite systems (Myotis bechsteinii - Spinturnix bechsteini and
Myotis myotis - Spinturnix myoti)”
2011
Volkswagen Symposium Evolutionary Biology – Poster
“A comparison of genetic population structure in two hostparasite systems (Myotis bechsteinii - Spinturnix bechsteini and
Myotis myotis - Spinturnix myoti)”
2013
International Bat Research Meeting – Talk
“Comparative Population Genetics of Bechstein’s Bat and Two
of its Ectoparasites”
2013
Volkswagen Symposium Evolutionary Biology – Talk
“Comparative Population Genetics of Bechstein’s Bat and Two
of its Ectoparasites”
2013
Göttinger Freilandtage – Talk
“Can a closed host social system reduce the evolutionary
potential of parasites? Insights from the Bechstein’s bat and
two of its parasites”
2014
IMPRS Grand Challenges Symposium – Talk
“Host-Parasite interactions in a social world”
2015
International Berlin Bat Meeting – Talk
“Small differences in host and parasite life history can strongly
affect parasite population genetic structure: lessons from two
bat hosts and three ectoparasites”
2016
Summer school on dispersal (Univ of Greifswald) – Talk
“Dispersal at the mercy of your host: dispersal and gene flow in
non-mobile parasites”
167