The Invasion Biology and Sociogenetics of Pharaoh Ants

2 mm Anna M.

Schmidt The invasion biology and sociogenetics of pharaoh ants

•The

Anna Mosegaard Schmidt invasion PhD thesis biology and oigntc of sociogenetics pharaoh ants

•PhD Thesis2010

Department of Biology University of Copenhagen August 2010 PhD thesis, A.M. Schmidt, August 2010 FACULTY OF SCIENCE UNIVERSITY OF COPENHAGEN

The invasion biology and sociogenetics of pharaoh ants

A dissertation submitted to the University of Copenhagen in accordance with the requirements of the degree of PhD at the faculty of Science, Department of Biology, to be defended publicly before a panel of examiners

Anna M. Schmidt

Advisory committee:

Ass. Prof. Jes S. Pedersen Prof. Jacobus J. Boomsma Ass. Prof. Patrizia d’Ettorre

Cover Front: newly eclosed Monomorium pharaonis gyne Back: Monomorium pharaonis colony fragment with workers gyne, queen and various brood stages All photos and illustraions in this thesis are by Anna M. Schmidt

Preface

This thesis is the result of a three‐year Ph.D. project carried out at the Centre for Social Evolution, Department of Biology, University of Copenhagen, Denmark, under the supervision of Associate Professor Jes S. Pedersen. The project was financed by the Centre for Social Evolution through a grant received from the Danish National Research Foundation. During my thesis work I also spent two months visiting Harvard University, Cambridge, Massachusetts, USA, hosted by Professor Naomi Pierce.

The thesis consists of three parts. First a general introduction in which I describe the theoretical background of my work, introduce the study species and describe the objectives and thesis structure. Then five chapters follow, four of which are written as research papers reporting the findings of my work, and the fifth is a short, primarily methodological, description of work that is still ongoing. Then follows a brief summary of the main findings of my work and suggested future directions.

Anna M. Schmidt Copenhagen, August 2010

Contents

Summary (English & Danish)………………………………………………………...... 7

A general introduction to the study system and questions ………………… 13

Looking at ants from an invasion biology perspective ….………………………… 14 Characteristic traits of introduced ants……………………………………..… 14 Pharaoh ant biology …………………………………………………………………………….…. 18 Production of new sexuals in the laboratory …..……………………….…. 19 Increased genetic diversity and the disease resistance hypothesis ………… 20 Effects of polyandry and pharaoh ants as polygynous models .…… 21

Objectives and thesis structure …..……...……………………………………………….……….…… 22 References .………………………………………………………………………….……………………….…… 24

Chapter 1 …….…………..……….……………………………………………………………………………. 30 Local scale population genetics of pharaoh ants (Monomorium pharaonis) in Thailand and the Thai‐Malay Peninsula

Chapter 2 ….…………..……….……………………………………………………………………………… 58 Frequent occurrence of Wolbachia in introduced populations of pharaoh ants

Chapter 3 …….…………..……….……………………………………………………………………………. 82 Deconstructing the effect of genetic diversity on disease resistance in a polygynous social insect

Chapter 4 …….…………..……….……………………………………………………………………………. 110 Queen‐worker caste ratio depends on colony size in the pharaoh ant (Monomorium pharaonis)

Chapter 5 …….…………..……….……………………………………………………………………………. 128 The pharaoh ants as a model organism: creating a heterogeneous stock and selecting on a social insect colony level trait

Summary and possible future directions …………………………………………………. 140

Acknowledgements ….…..……………………………………………………………………………... 146

Curriculum vitae ..……..………………………………………………………………………………..... 148

Pharaoh ant photos..……………………………………………………………………………..… 152

Summary

Social insect colonies perform a number of tasks affecting the environments they live in. Some unintentionally introduced species have attracted the attention of scientists and general public alike when causing a number of changes to the composition and functioning of ecosystems. Such “invaders” or “tramps”, though often considered negative influences, can also be seen as natural experiments, generating a number of questions in the fields of ecology and evolution. Pharaoh ants (Monomorium pharaonis) are very successful invaders of human habitation in most parts of the world. Individual pharaoh ants are small, their colonies are polygynous (have multiple queens), and consist of multiple interconnected nests that can spread to cover large areas through so‐called budding. Pharaoh ants appear to mate exclusively within their nests, indiscriminately inbreeding without a cost to colony performance. Combining these traits, and adding to them that numerous introductions of the species have resulted in genetically highly differentiated, low diversity colonies, makes pharaoh ants an interesting model system. During my PhD I have thus investigated the potential of pharaoh ants as models for questions of invasion biology and evolution in social insects. I have developed methods for establishing colonies of different genetic composition through controlled crossings of genetically different colonies, and established that measurable genetic as well as morphological variation exists between different laboratory lineages, thus building the foundation for future research on the species. In addition, I have started a selection experiment (still ongoing in collaboration with Dr. T. Linksvayer) using pharaoh ants, which is the first time artificial selection is attempted in an ant species. Pharaoh ants have been introduced multiple times from an as of yet unknown native range, but little is known about their local mode of spread and propagule pressure. To learn more, I investigated the population genetic structure and phylogeography of pharaoh ants on different geographical scales in Thailand and malaysia, a region where they are very common. Employing species‐specific microsatellite markers, I found a structure of multiple introductions and isolation of colonies even on relatively small geographical scales. Enemy release is a central concept in invasion biology, frequently proposed to play an important part in the success of introduced species. To investigate the enemy release hypothesis in Pharaoh ants, I screened introduced populations from several localities around the world for the presence of the presumed detrimental intracellular bacterium Wolbachia. Wolbachia was frequently present in introduced colonies, a finding which raises the question how harmful Wolbachia really is for pharaoh ants, as lack of this bacterium does not appear to be a prerequisite for successful introductions. Having many related individuals living at high densities, as found in social insect colonies, is expected to increase susceptibility to infection and within‐group disease transmission. Such potentially negative effects of sociality are hypothesized to be counteracted by increased within‐colony genetic variation. To test this hypothesis, knowing that pharaoh ant colonies generally have low levels of genetic diversity, I investigated the effect of genetic diversity on disease resistance in pharaoh ants by crossing colonies. I did so by creating different diversity groups, and infecting these with a generalist entomopathogenic fungus. The results showed variation between groups from different colonies, and also showed that higher levels of genetic diversity, be this inter‐or intra‐individual, did not necessarily result in increased

7 pathogen resistance in pharaoh ants. This finding is different from most previous research on the subject, and thus cautions against making general assumptions in the field. The success of an ant colony depends on the simultaneous presence of reproducing queens and non‐reproducing workers in a ratio that will maximize colony growth and reproduction. Though seemingly very basic, little is known about the effect of colony size on queen‐worker caste ratios in ants. Manipulating colony size through the creation of multiple queenless colonies, we found that smaller colonies produced more new queens relative to workers, and that those queens and workers also tended to be larger. These findings demonstrate high levels of plasticity in energy allocation towards female castes, and suggest that polygynous, budding species may adaptively adjust caste ratios to ensure rapid growth. The work in this thesis thus sheds some more light on pharaoh ants as introduced species, and takes the first crucial steps towards establishing it as a social insect model organism, while demonstrating some of the questions that may be addressed using this Monomorium‐model.

Resumé

Sociale insekters kolonier udfører en række forskellige opgaver, der påvirker miljøet omkring dem. Nogle utilsigtet introducerede arter har tiltrukket sig opmærksomhed ved at forårsage en række ændringer i økosystemers sammensætning og funktionsdygtighed. Sådanne ”invasive arter” eller ”vagabonder” anses ofte for værende negative påvirkninger, men kan samtidig også ses som en slags naturlige eksperimenter, idet de gennem deres tilstedeværelse genererer en masse spørgsmål indenfor forskningsområderne økologi og evolution. Faraomyrer (Monomorium pharaonis) er succesfulde i deres invasion af menneskehabitater i det meste af verden. Set som individer er faraomyrer små, deres kolonier er polygyne (d.v.s. der er adskillige dronninger i hver koloni), og kolonierne består af adskillige forbundne bo, der kan sprede sig ud til at dække store områder gennem såkaldt knopskydning. Faraomyrer parrer sig udelukkende i deres reder, og der forekommer således tilsyneladende indavl uden nogle fitnessmæssige konsekvenser for kolonierne. Kombinationen af disse træk og dertil lagt at adskillige introduktioner af arten har resulteret i genetisk højt differentierede, lavdiversitetskolonier gør, at faraomyrer er et interessant modelsystem. I løbet af min PhD har jeg således undersøgt faraomyrers potentiale som modelorganismer i forbindelse med invasionsbiologiske spørgsmål såvel som spørgsmål relateret til sociale insekters evolution. Jeg har udviklet metoder til at etablere kolonier af forskellig genetisk sammensætning gennem kontrollerede krydsninger af genetisk differentierede kolonier, og slået fast, at der er målbar genetisk såvel som morfologisk variation mellem forskellige laboratoriekolonier, derigennem byggende en basis for fremtidig forskning med denne art. Derudover har jeg startet et selektionseksperiment med faraomyrer (dette er endnu igangværende i samarbejde med Dr. T. Linksvayer), hvilket er det første forsøg nogensinde med kunstig selektion på en myreart. Faraomyrer er blevet introduceret en række gange fra endnu ukendte ophavsområder, men kun lidt vides om deres spredningsmekanismer og det introduktionstryk, de udøver på lokalt plan. For at undersøge dette nærmere har jeg analyseret faraomyrers populationsgenetiske struktur og fylogeografi på forskellige geografiske afstande i Thailand og Malaysia, et område, hvor forekomsten af faraomyrer er hyppig. Ved hjælp af artsspecifikke microsatellitmarkører fandt jeg ud af, at populationsstrukturen selv over meget korte geografiske afstande afspejler mange adskilte introduktioner og at de enkelte kolonier er isolerede genetisk set. Det, at nogle organismer ikke får følgeskab af deres naturlige fjender, når de introduceres til nye områder, er et centralt begreb indenfor invasionsbiologi, og påberåbes ofte som mulig forklaring, når introducerede arter er succesfulde. For at undersøge denne hypotese nærmere i forbindelse med faraomyrers succes, screenede jeg introducerede populationer fra en række forskellige lokaliteter fra hele verden for tilstedeværelse af en intracellulær bakterie (Wolbachia), der generelt menes at være skadelig for myrer. Wolbachia var ofte til stede i de screenede prøver, hvilket fik mig til at stille spørgsmålstegn ved, hvor skadelig Wolbachia er for faraomyrer, da det, at mangle bakterien, tilsyneladende ikke lader til at være en nødvendighed for at kunne opnå talrige succesrige introduktioner. Det at have mange nærtbeslægtede individer samlet på meget lille plads, som det er tilfældet i sociale insekters kolonier, forventes at øge modtagelighed overfor infektioner samt transmission af sygdom indenfor grupperne. Sådanne potentielle negative konsekvenser af et socialt live regnes for at blive modvirket af en øgning af genetisk diversitet indenfor kolonierne.

9 For at teste denne hypotese, vel vidende at faraomyrekolonier generelt har et lavt niveau af genetisk diversitet, undersøgte jeg effekten af genetisk diversitet på sygdomsmodstand blandt faraomyrer. Dette gjorde jeg ved at krydse forskellige kolonier, for derigennem at danne grupper med forskellig diversitet, og derefter inficerede jeg dem med en insektspecifik svamp. Resultaterne viste, at der var variation mellem grupperne og at de grupper, der havde højere genetisk diversitet, ikke nødvendigvis klarede sig bedre overfor svampen – dette uafhængigt af om diversiteten var at finde indenfor de enkelte individer, eller indenfor gruppen som hele. Dette fund skiller sig således ud fra størstedelen af den tidligere forskning på området, og kan således ses som en advarsel mod at lave generelle antagelser på dette område. En myrekolonis succes afhænger af samtidig tilstedeværelse af reproducerende dronninger og ikke‐reproducerende arbejdere i et forhold der vil maksimere koloniens vækst og samlede reproduktion. Skønt det kan lyde til at være basal viden, vides kun lidt om effekten af kolonistørrelse på dronninge‐arbejder kasteratio blandt myrer. Ved at manipulere kolonistørrelsen gennem dannelse af en række små, nye, dronningeløse kolonier, fandt vi frem til at mindre kolonier producerede relativt flere nye dronninger i forhold til arbejdere, og at både arbejdere og dronninger i disse kolonier også havde en tendens til at være større. Disse resultater demonstrerer at der er betydelig plasticitet i energiallokeringen i hun‐kasterne i en myrekoloni, og antyder at polygyne, knopskydende arter kan tilpasse deres kasteratio alt efter forholdende, de lever under, og derigennem sikre hurtig vækst. Det arbejde, jeg præsenterer i denne afhandling, kaster således mere lys over faraomyren som introduceret art, og tager samtidig de første, altafgørende skridt mod at etablere faraomyren som en social insekt modelorganisme, alt imens det demonstrerer nogle af den type spørgsmål, der kan adresseres ved brug af denne Monomorium‐model.

A general introduction to the study system and questions

Biological research has become highly interdisciplinary, with the optimum approach combining

“deep knowledge in one discipline and basic ‘fluency’ in several”, in the words ofthe National

Research Council’s Committee on a New Biology for the 21st Century (2009). In this thesis, I focus on the biology of the pharaoh ant (Monomorium pharaonis), a new model organism with great potential, to address questions from invasion biology and evolutionary genetics. The major advantages of this species over other ants are the ease with which it can be reared and bred in the laboratory, its relatively short life cycle and the genetically highly differentiated colonies it is possible to collect in the field. These qualities have allowed me to address classic questions with new methods, and to apply the experimental evolution paradigm to ants through artificial selection experiments for the first time (Chapter 5).

Pharaoh ants are considered a damaging introduced species, and I have therefore investigated their population genetics (Chapter 1), as well as the occurrence of the intracellular bacterium

Wolbachia in pharaoh ants and some of their congenerics (Chapter 2). I then took a strongly experimental approach, so far only available with this species, to quantify the effects of genetic diversity on disease resistance (Chapter 3). I also investigated how colony size affects the caste ratio, data that are both of fundamental interest (Tschinkel 2010) and helps us to understand how this introduced species grows and spreads (Chapter 4).

To verify the potential of pharaoh ants as model organisms for the study of sociogenetics, I designed and validated reliable methods for crossing and maintaining lab colonies over multiple generations in the laboratory, thereby providing a firm footing for experimental evolution and

13 artificial selection studies, which have never before been performed on ants. In addition to the application of experimental breeding to manipulate genetic diversity in Chapter 3, I have started two long‐term experiments: the creation of a heterogeneous stock, which provides excellent raw material for artificial selection experiments, as well as two generations of artificial selection on colony caste ratio. Due to the long‐term nature of the artificial selection experiment, the final outcome will not be known for a few more months, but the basics of the methods are presented here (Chapter 5).

Looking at ants from an invasion biology perspective

Introduced and invasive species are becoming increasingly conspicuous in most parts of the world, and range from domesticated plants and pets that have become pests, to wild species that have been inadvertently transported to new habitats. Many introduced species are currently causing great ecological and economic damage, such that there is a dire need for new methods for limiting their spread (Mack et al. 2000; Pimentel et al. 2001; Pimentel et al. 2005). A steadily increasing number of ant species are ranked among the most successful and costly invaders, some of which, such as the yellow crazy ant (Anoplolepis gracilipes), the little fire ant (Wasmannia auropunctata), and the red imported fire ant (Solenopsis invicta) are considered among the worst invasive species on the planet (Gotelli and Arnett 2000; Lowe et al. 2000; O'Dowd et al. 2003; Pimentel et al.

2005).

Characteristic traits of introduced ants

But what turns tiny ants into costly invaders? Most introduced/invasive ant species described so far appear to possess a suite of characteristics that could collectively be termed an “invasive ant

14 syndrome”: They are small (smaller than their native relatives, McGlynn (1999)), opportunistic in terms of food and habitat and they are polygynous (have multiple queens) and polydomous

(inhabit multiple nests). They typically reproduce by budding, in which a group of workers,brood and sometimes queens leave their natalnest or colony on foot and establish a new one. Their small size and mode of colony foundation has exquisitely pre‐adapted them to dispersal by humans and their populations genetic signatures will therefore usually be those of jump‐dispersal

(Suarez et al. 2001; Schmidt et al. 2010). Invasive ants also generally thrive in disturbed areas and may be tolerant of inbreeding. Lastly, invasive ants are frequently "supercolonial", meaning that they are not aggressive towards conspecific non‐nestmates, allowing colonies to effectively merge into large supercolonies that can cover thousands of kilometres in the most extreme cases

(Hölldobler and Wilson 1990; Passera 1994a; Giraud et al. 2002; Helanterä et al. 2009). The latter characteristic has caused a number of hypotheses to be generated to explain this possible change in social structure between native and invasive colonies (see e.g. Tsutsui et al. 2000; Giraud et al.

2002; Starks 2003).

Introduced ants are often portrayed as superior competitors, proficient in discovery as well as dominance of food sources (Human and Gordon 1996; Holway 1999; Holway et al. 2002). This clearly seems to be the case for some species, that are very well‐adapted to semi‐disturbed environments in which they are very abundant (LeBrun et al. 2007). Because the majority of introduced ants have been found in human modified environments, it has been argued that maybe the introduced species dominate not because they are superior competitors, but simply because the areas are no longer suitable for the local ant guilds (King and Tschinkel 2008a; King and Tschinkel 2008b; Tsuji 2010).

15 However, the question still remains what makes the introduced ants so well‐adapted to these places? Much research now suggests that two factors may be particularly important: 1) Pre‐ adaptation to life in “disturbed” or human‐modified environments and environments with high competition, meaning that they are not dependent on large, constructed nest structures, but rather live in ephemeral environments and quickly move if disturbed and that they are often fast at discovering food as well as recruiting nestmates to food sources (Feener 2000). 2) Unusual reproductive systems, such as clonal reproduction or inbreeding, potentially enabling rapid growth of colonies and/or ease of spread to new localities. For example, although normal as well as clonal reproduction occurs in the little fire ant, it appears that invasive populations are predominantly clonal, which may increase their invasion potential by removing the need to find a mate in new habitats (Queller 2005; Foucaud et al. 2007; Foucaud et al. 2009).

Budding, which is normal among introduced ants circumvents the dangerous mating flight and the slow initial establishment phase that lone queens otherwise go through. (Oster and Wilson 1978;

Fritz and Vander Meer 2003; Goodisman et al. 2007) (Fig. 1) Budding species thus trade off distance of dispersal with potential local dominance, as they gain a competitive edge in starting in greater numbers and potentially saturating habitats through slow spread, as has been seen in the different densities and close to lack of coexistence of the monogyne and polygyne social forms of the red imported fire ants in North America, where the monogyne form is outcompeted in zones of overlap (e.g. Fritz and Vander Meer 2003). Most previously described invasive ants have been found predominantly in tropical areas, and the yellow crazy ant (Anoplolepis gracilipes) has been probably the best known example of a species moving far beyond disturbed areas in its invasion of

Christmas Island (O'Dowd et al. 2003) and now also mainland Australia (Hoffmann and Saul 2010).

Fig. 1. A typical ant colony growth cycle, modified from Hölldobler & Wilson, (1990) and Tschinkel (1993). The black curve is based on the starting point for a solitarily founding ant species which goes through three phases: founding (the first shallow part of the curve) followed by a number of cycles of exponential growth and reproduction at the end of which it goes through a reduction in size once sexuals are produced and leave the colony (the subsequent waves). A budding ant species (red curve) can be seen as following more or less the same pattern,however, as it does not have solitary founding it gets a head start in terms of growth as multiple individuals are present from day one; starting size thus being equivalent of that of a several months old solitarily founding species, and having fast growth immediately. The nest of a budding species likewise goes through a phase of growth and, at a certain point, a decline as part of the nest buds off.

However, new types of invaders including Myrmica rubra and Lasius neglectus are now being identified in temperate regions (Seifert 2000; Groden et al. 2005; Groden 2006; Garnas et al. 2007;

Cremer et al. 2008), some species ‐ e.g. Pachycondyla chinensis (Guenard and Dunn 2010) ‐ have even been found in undisturbed temperate areas, indicating that the list of successful introduced and invasive ants will likely increase dramatically.

Identifying the native range and sources of introduced populations e.g. in the Argentine ant

(Linepithema humile) (Tsutsui et al. 2001), the red imported fire ant (Ross et al. 2007) and the little fire ant (Mikheyev and Mueller 2007) has enabled useful comparisons increasing understanding of

17 the introduction process as well as the introduced species themselves (Holway et al. 2002; Heller

2004; Suarez et al. 2008; Helanterä et al. 2009; Foucaud et al. 2010). Increasing the number of species for which the native range has been identified or for which population genetic tools are available, is therefore highly desirable in the work towards a better understanding of the evolutionary processes at play, as well as to enable well‐informed management (Ashley et al.

2003; Suarez and Tsutsui 2008; Suarez et al. 2010b).

Pharaoh ant biology

Although small and perhaps inconspicuous, pharaoh ants have received much attention in the literature due to their near omnipresence in and around human settlements (Wilson 1971;

Hölldobler and Wilson 1990; Passera 1994b; Suarez et al. 2010a; Wetterer 2010). They are among the most common introduced ant species, frequently spread through human mediated jump dispersal (Suarez et al. 2005; Schmidt et al. 2010) and often intercepted at customs (Lester 2005;

Suarez et al. 2005). Pharaoh ants are not thought to cause significant ecosystem damage, as they are mostly found in the immediate vicinity of houses.Their natural habitat and native range remain unknown and the species is likely the ant species with the longest introduction history

(Schmidt et al. 2010; Wetterer 2010). Considering their opportunism in nesting (CSIRO 2010), it is conceivable that pharaoh ants are pre‐adapted to environments much like those of human habitation.

What makes pharaoh ants particularly interesting and attractive as models in evolutionary biology studies is their dual properties of being introduced (meaning that their colonies can be seen as a

18 sort of “natural experiment“ (Sax et al. 2007)) and also that the species is among the few ant species that can be reared in the laboratory for many generations. Pharaoh ants have been introduced to most human‐inhabited areas and are commonly found as pest infestations in houses (Edwards 1986; Berndt and Eichler 1987) and fit within the classical “invasive ant syndrome”(Wheeler 1986; Berndt and Eichler 1987), and are highly polygynous. In addition they have effective foraging trails and like other Monomorium species they can engage in chemical warfare with competitors (Adams and Traniello 1981).

Fig. 2. Three young pharaoh ant queens on a background of brood and workers.

©AMS

Production of new reproductives in the laboratory

In pharaoh ants, new reproductives are produced at regular intervals as the queens senesce

(Petersen‐Braun 1975, AMS pers. obs.), and caste determination has been shown to depend on the social environment, i.e. on whether fertile queens are present in the colonies (Edwards 1987;

1991). Once queens in a colony are four weeks old the eggs they lay are bipotent, and production of new queens and males can be induced at any time simply by removing the old queens(Petersen‐Braun 1977). Workers are completely sterile, meaning that the only brood produced is that of the queens (Berndt and Eichler 1987). Because the developmental time of

19 sexuals is usually around 40‐45 days (Peacock 1950; Peacock and Baxter 1950; Petersen‐Braun

1975), this enables a generation time in the laboratory of as little as three to four months. Since all matings appear to be intranidal, i.e. within the nest, (both males and gynes are winged, but gynes are completely incapable of flying and males almost too), pharaoh ants can potentially be kept in the laboratory indefinitely (Peacock and Baxter 1949).

What makes pharaoh ants even more interesting, is that they appear to inbreed indiscriminately

(Berndt and Eichler 1987, as Hamilton also mused in his chapter on inbreeding and outbreeding in

Thornhil & Shield’s 1993 edited volume (Hamilton 1993)). In addition to this, different colonies, which can readily be collected in tropical countries, appear to be very highly genetically differentiated (Schmidt et al. 2010; Chapter 1). Interestingly, pharaoh ants colonies do not appear to suffer from inbreeding depression, and therefore appear unlikely do have single locus complementary sex determination (sl‐CDS), nor do they seem particularly susceptible to pathogens in their immediate surroundings (AMS pers. obs., Chapter 3). This makes them attractive subjects for testing the genetic diversity hypothesis on disease resistance (cf. below and

Chapter 3).

Increased genetic diversity and the disease resistance hypothesis

Genetic diversity within a group may play an important role in its ability to combat infection

(Hamilton 1987) and in social insects, increased genetic diversity has been hypothesised to enable a more effective division of labour, facilitating colony‐level homeostasis and survival (Crozier and

Page 1985; Oldroyd and Fewell 2007). An increase in within‐colony variation can be achieved through polyandry (i.e. species in which the queen mates with multiple males) or polygyny

20 (Crozier and Pamilo 1996). For this reason, an increased level of disease resistance has been hypothesized to be central in explaining especially how polyandry could evolve and be maintained

(Sherman et al. 1988; Keller and Reeve 1994; Boomsma and Ratnieks 1996; Sherman et al. 1998;

Hughes et al. 2008) because polyandry is otherwise thought to be costly.

Effects of polyandry and pharaoh ants as polygynous models

Several studies on social insects seem to conform to the hypothesis that increased genetic diversity within colonies is associated with increased disease resistance (e.g. Schmid‐Hempel and

Crozier 1999; Tarpy 2003; Hughes and Boomsma 2004; Tarpy and Seeley 2006; Seeley and Tarpy

2007; Reber et al. 2008). Presumably this is because colonies of higher genetic diversity may also harbour increased diversity in behavioural and/or innate physiological responses to parasites or pathogens (Bourke and Franks 1995; Baer and Schmid‐Hempel 2003; Hughes and Boomsma 2004).

Different patrilines in multiply‐mated or artificially inseminated honey bees, bumble bees, and ants have thus been shown to display different levels of resistance to pathogens or parasites.

Likewise, experimental groups composed of several different patrilines have been shown to have better survival when exposed to a pathogen (Baer and Schmid‐Hempel 2003; Tarpy 2003; Hughes and Boomsma 2004).

Given that pharaoh ant colonies are highly inbred (Berndt and Eichler 1987), we may ask whether they suffer from inbreeding depression with regards to disease resistance. Pharaoh ants are thus obvious candidates to test the generality of the increased genetic diversity hypothesis as the possibility to breed them in the laboratory makes manipulation of the genetic diversity of not only groups as a whole but also single individuals possible through controlled crosses (Chapter 3). In addition pharaoh ants are interesting in this context since they are polygynous and polygyny has

21 been proposed as an alternative means to maintaining high levels of genetic diversity within a colony, and is a more commonly occurring phenomenon in ants than polyandry (e.g. Hughes et al.

2008). Using pharaoh ants as models it is thus possible to ask whether a positive effect of increased genetic diversity is generally to be expected in social insect colonies.

Objective and thesis structure

A large number of questions can be addressed when using social insects as a starting point, and the main objective of this thesis is to gain further insights into the invasion biology and population genetics of pharaoh ants as well as to address more fundamental general questions in the field of evolutionary biology while demonstrating and developing the different ways the pharaoh ant may be of great use as a model organism.

Chapter 1 investigates the population genetics of pharaoh ants in Thailand and Malaysia, employing microsatellite markers. We found that even in tropical areas where pharaoh ants can survive outside of houses, the spread of single colonies is very limited and no gene flow seems to occur at all. Rather the populations appear to be products of numerous introduction events.

Chapter 2 examines a large number of introduced pharaoh ant colonies as well as 11 other

Monomorium species for the occurrence of the intracellular bacterium Wolbachia, to test this as an indicator of enemy release. We found that a large number of colonies were infected in all areas sampled, leading us to question Wolbachia screenings’worth as indicators of enemy release as well as the presumed negative effect of Wolbachia in ants.

22 Chapter 3 challenges the genetic diversity‐disease resistance hypothesis in ants. Through experimental manipulation and sequential controlled crosses the genetic diversity of groups of individuals is manipulated to either increase the inter‐ or intraindividual diversity and these groups are subsequently exposed to a generalist entomopathogenic fungus, while asking: does increased genetic diversity in a group always confer an advantage to the group members? And, if so, is this irrespective of whether this diversity exists only at the level of the group or within every single member of the group? We found that colonies may not benefit from increase genetic diversity, and that experimental mixing of colonies may have a detrimental effect on group social interactions.

Chapter 4 investigates the role of the social environment on the allocation of resources in ant colonies, specifically whether the caste ratios (i.e. the ratio of new gynes to total number of female offspring produced) of colonies are affected by their size. We found a clear, negative correlation between caste ratio and colony size as smaller colonies produced relatively more gynes.

Chapter 5 shows the utility of pharaoh ants as laboratory organisms through two long term projects the creation of a heterogeneous stock, which can form the basis for many other experiments and the application of artificial selection on colony queen‐worker caste ratios. Both experiments have run over several years and are close to their end, in this chapter I will provide an outline of what has been done so far, including a bit of background and describing the methodologies used.

23 References:

Adams ES, Traniello JFA (1981) Chemical interference competition by Monomorium minimum (Hymenoptera, Formicidae). Oecologia 51:265‐270 Ashley MV, Willson MF, Pergams ORW, O'Dowd DJ, Gende SM, Brown JS (2003) Evolutionarily enlightened management. Biological Conservation 111:115‐123 Baer B, Schmid‐Hempel P (2003) Bumblebee workers from different sire groups vary in susceptibility to parasite infection. Ecol Lett 6:106‐110 Berndt KP, Eichler W (1987) Die Pharaoameise, Monomorium pharaonis (L.) (Hym., Myrmicidae). Mitt Zool Mus Berl 63:3 ‐186 Boomsma JJ, Ratnieks FLW (1996) Paternity in eusocial Hymenoptera. Philos Trans R Soc Lond B Biol Sci 351:947‐975 Bourke AFG, Franks NR (1995) Social Evolution in Ants. Princeton University Press, Princeton, New Jersey Council CoaNBftsCEtUSLtCBRNR (2009) A New Biology for the 21st Century. The National Academies Press Cremer S, Ugelvig LV, Drijfhout FP, Schlick‐Steiner BC, Steiner FM, Seifert B, Hughes DP, Schulz A, Petersen KS, Konrad H, Stauffer C, Kiran K, Espadaler X, d'Ettorre P, AktaÃ§ N, Eilenberg Jr, Jones GR, Nash DR, Pedersen JS, Boomsma JJ (2008) The evolution of invasiveness in garden ants. PLoS ONE 3:e3838 Crozier RH, Page RE (1985) On being the right size ‐ male contributions and multiplemating in social Hymenoptera. Behav Ecol Sociobiol 18:105‐115 Crozier RH, Pamilo P (1996) Evolution of Social Insect Colonies: Sex Allocation and Kin Selection. Oxford University Press, New York CSIRO (2010) Ants Down Under http://anic.ento.csiro.au/ants/biota_details.aspx?BiotaID=3866 Edwards JP (1986) The biology, economic importance, and control of the Pharaoh's Ant In: Bradleigh Vinson S (ed) Economic Impact and Control of Social Insects. Praeger Publishers, New York, pp 257‐271 Edwards JP (1987) Caste regulation in the pharaoh's ant Monomorium pharaonis: the influence of queens on the production of new sexual forms. Physiol Entomol 12:31‐39 Edwards JP (1991) Caste regulation in the pharaoh's ant Monomorium pharaonis: recognition and cannibalism of sexual brood by workers. Physiol Entomol 16:263‐271

24 Feener DJ (2000) Is the assembly of ant communities mediated by parasitoids? Oikos 90:79‐88 Foucaud J, Fournier D, Orivel J, Delabie JHC, Loiseau A, Le Breton J, Kergoat GJ, Estoup A (2007) Sex and clonality in the little fire ant. Molecular Biology and Evolution 24:2465‐2473 Foucaud J, ORIVEL J, FOURNIER D, DELABIE JHC, LOISEAU A, BRETON JL, CERDAN P, ESTOUP A (2009) Reproductive system, social organization, human disturbance and ecological dominance in native populations of the little fire ant, Wasmannia auropunctata. Mol Ecol 18:5059‐5073 Foucaud J, Orivel J, Loiseau A, Delabie JHC, Jourdan H, Konghouleux D, Vonshak M, Tindo M, Mercier J‐L, Fresneau D, Mikissa J‐B, McGlynn T, Mikheyev AS, Oettler J, Estoup A (2010) Worldwide invasion by the little fire ant: routes of introduction and eco‐evolutionary pathways. Evolutionary Applications 9999 Fritz GN, Vander Meer RK (2003) Sympatry of polygyne and monogyne colonies of the fire ant Solenopsis invicta (Hymenoptera : Formicidae). Ann Entomol Soc Am 96:86‐92 Garnas JR, Drummond FA, Groden E (2007) Intercolony aggression within and among local populations of the invasive ant, Myrmica rubra (Hymenoptera : Formicidae), in coastal Maine. Environmental Entomology 36:105‐113 Giraud T, Pedersen JS, Keller L (2002) Evolution of supercolonies: The Argentine ants of southern Europe. Proc Natl Acad Sci U S A 99:6075‐6079 Goodisman MAD, Sankovich KA, Kovacs JL (2007) Genetic and morphological variation over space and time in the invasive fire ant Solenopsis invicta. Biological Invasions 9:571‐584 Gotelli NJ, Arnett AE (2000) Biogeographic effects of red fire ant invasion. Ecol Lett 3:257‐261 Groden E (2006) Speculation: invasion success of Myrmica rubra in North America. In: XV Congress IUSSI, Washington, D.C., pp 120‐121 Groden E, Drummond FA, Garnas J, Franceour A (2005) Distribution of an invasive ant, Myrmica rubra (Hymenoptera : Formicidae), in Maine. J Econ Entomol 98:1774‐1784 Guenard B, Dunn RR (2010) A new (old), invasive ant in the hardwood forests of Eastern North America and its potentially widespread impacts. PLoS ONE 5 Hamilton WD (1987) Kinship, recognition, disease, and intelligence: constraints of social evolution. In: Ito Y, Brown JL, Kikkawa J (eds) Animal Societies: Theories and Facts Japan Scientific Societies Press, Tokyo, pp 81–102 Helanterä H, Strassmann JE, Carrillo J, Queller DC (2009) Unicolonial ants: where do they come from, what are they and where are they going? Trends Ecol Evol 24:341‐349

25 Heller NE (2004) Colony structure in introduced and native populations of the invasive Argentine ant, Linepithema humile. Insect Soc 51:378‐386 Hoffmann BD, Saul WC (2010) Yellow crazy ant (Anoplolepis gracilipes) invasions within undisturbed mainland Australian habitats: no support for biotic resistance hypothesis. Biological Invasions 12:3093‐3108 Holway DA (1999) Competitive mechanisms underlying the displacement of native ants by the invasive Argentine ant. Ecology 80:238‐251 Holway DA, Lach L, Suarez AV, Tsutsui ND, Case TJ (2002) The causes and consequences of ant invasions. Annu Rev Ecol Syst 33:181‐233 Hughes WOH, Boomsma JJ (2004) Genetic diversity and disease resistance in leaf‐cutting ant societies. Evolution 58:1251‐1260 Hughes WOH, Ratnieks FLW, Oldroyd BP (2008) Multiple paternity or multiple queens: two routes to greater intracolonial genetic diversity in the eusocial Hymenoptera. J Evol Biol 21:1090‐ 1095 Human KG, Gordon DM (1996) Exploitation and interference competition between the invasive Argentine ant, Linepithema humile, and native ant species. Oecologia 105:405‐412 Hölldobler B, Wilson EO (1990) The Ants. Springer, New York Keller L, Reeve HK (1994) Genetic variability, queen number, and polyandry in social Hymenoptera. Evolution 48:694‐704 King J, Tschinkel W (2008a) Experimental evidence that human impacts drive fire ant invasions and ecological change. Proc Natl Acad Sci USA 105:20339 ‐ 20343 King JR, Tschinkel WR (2008b) Experimental evidence that human impacts drive fire ant invasions and ecological change. Proc Natl Acad Sci U S A 105:20339‐20343 LeBrun EG, Tillberg CV, Suarez AV, Folgarait PJ, Smith CR, Holway DA (2007) An experimental study of competition between fire ants and Argentine ants in their native range. Ecology 88:63‐ 75 Lester PJ (2005) Determinants for the successful establishment of exotic ants in New Zealand. Diversity & Distributions 11:279‐288 Lowe S, Browne M, Boudjelas S., De Poorter M (2000) 100 of the world's worst invasive alien species a selection from the global invasive species database. In. The Invasive Species Specialist Group (ISSG) a specialist group of the Species Survival Commission (SSC) of the World Conservation Union (IUCN), pp 12

26 Mack RN, Simberloff D, Lonsdale WM, Evans H, Clout M, Bazzaz FA (2000) Biotic invasions: Causes, epidemiology, global consequences, and control. Ecological Applications 10:689‐710 McGlynn TP (1999) Non‐native ants are smaller than related native ants. Am Nat 154:690‐699 Mikheyev AS, Mueller UG (2007) Genetic relationships between native and introduced populations of the little fire ant Wasmannia auropunctata. Diversity Distrib 13:573‐579 O'Dowd DJ, Green PT, Lake PS (2003) Invasional 'meltdown' on an oceanic island. Ecol Lett 6:812‐ 817 Oldroyd BP, Fewell JH (2007) Genetic diversity promotes homeostasis in insect colonies. Trends Ecol Evol 22:408‐413 Oster GF, Wilson EO (1978) Caste and ecology in the social insects. vol 12. Princeton University Press, Princeton, New Jersey Passera L (1994a) Characteristics of tramp species. Exotic Ants, Biology, Impact, and Control of Introduced Species:23 ‐ 43 Passera L (1994b) Characteristics of tramp species. In: Williams DF (ed) Exotic Ants, Biology, Impact, and Control of Introduced Species. Westview Press, Boulder, Colorado, pp 23‐43 Peacock AD (1950) Studies in Pharaoh's ant, Monomorium pharaonis (L.). 4. Egg‐production. Entomol Mon Mag 86:294‐298 Peacock AD, Baxter AT (1949) Studies in Pharaoh's ant, Monomorium pharaonis (L.). 1. The rearing of artificial colonies. Entomol Mon Mag 85:256‐260 Peacock AD, Baxter AT (1950) Studies in Pharaoh's ant, Monomorium pharaonis (L.), 3: Life history. Entomol Mon Mag 86:171‐178 Petersen‐Braun M (1975) Investigations on social organization in Pharaohs ant, Monomorium pharaonis L (Hym Formicidae) .1. Regulation of brood cycle. Insect Soc 22:269‐291 Petersen‐Braun M (1977) Investigations on social organization of Pharaohs ant Monomorium pharaonis L (Hymenoptera, Formicidae) .2. Caste determination. Insect Soc 24:303‐318 Pimentel D, McNair S, Janecka J, Wightman J, Simmonds C, O'Connell C, Wong E, Russel L, Zern J, Aquino T, Tsomondo T (2001) Economic and environmental threats of alien plant, animal, and microbe invasions Agriculture, Ecosystems & Environment 84:1‐20 Pimentel D, Zuniga R, Morrison D (2005) Update on the environmental and economic costs associated with alien‐invasive species in the United States. Ecological Economics 52:273‐ 288 Queller D (2005) Evolutionary biology ‐ Males from Mars. Nature 435:1167‐1168

27 Reber A, Castella G, Christe P, Chapuisat M (2008) Experimentally increased group diversity improves disease resistance in an ant species. Ecol Lett 11:682‐689 Ross KG, Krieger MJB, Keller L, Shoemaker DD (2007) Genetic variation and structure in native populations of the fire ant Solenopsis invicta: evolutionary and demographic implications. Biological Journal of the Linnean Society 92:541‐560 Sax DF, Stachowicz JJ, Brown JH, Bruno JF, Dawson MN, Gaines SD, Grosberg RK, Hastings A, Holt RD, Mayfield MM, O'Connor MI, Rice WR (2007) Ecological and evolutionary insights from species invasions. Trends Ecol Evol 22:465‐471 Schmid‐Hempel P, Crozier RH (1999) Polyandry versus polygyny versus parasites. Philos Trans R Soc Lond B Biol Sci 354:507‐515 Schmidt AM, d'Ettorre P, Pedersen JS (2010) Low levels of nestmate discrimination despite high genetic differentiation in the invasive pharaoh ant. Front Zool 7:20 Seeley TD, Tarpy DR (2007) Queen promiscuity lowers disease within honeybee colonies. Proc R Soc B‐Biol Sci 274:67‐72 Seifert B (2000) Rapid range expansion in Lasius neglectus (Hymenoptera, Formicidae) ‐ an Asian invader swamps Europe. Deutsche Entomologische Zeitschrift 47:173‐179 Sherman PW, Seeley TD, Reeve HK (1988) Parasites, pathogens, and polyandry in social Hymenoptera. Am Nat 131:602‐610 Sherman PW, Seeley TD, Reeve HK (1998) Parasites, pathogens, and polyandry in honey bees. Am Nat 151:392‐396 Starks PT (2003) Selection for uniformity: xenophobia and invasion success. Trends Ecol Evol 18:159‐162 Suarez A, Holway D, Tsutsui N (2008) Genetics and behavior of a colonizing species: The invasive argentine ant. Am Nat 172:S72 ‐ S84 Suarez AV, Holway DA, Case TJ (2001) Patterns of spread in biological invasions dominated by long‐distance jump dispersal: Insights from Argentine ants. Proc Natl Acad Sci U S A 98:1095‐1100 Suarez AV, Holway DA, Ward PS (2005) The role of opportunity in the unintentional introduction of nonnative ants. PNAS:0506119102 Suarez AV, McGlynn TP, Tsutsui ND (2010a) Biogeographic and taxonomic patterns of introduced ants. In: Lach L, Parr CL, Abbott KL (eds) Ant Ecology. Oxford University Press, New York, pp 233‐244

28 Suarez AV, McGlynn TP, Tsutsui ND (2010b) Biogeographic and taxonomic patterns of introduced ants. Oxford University Press Suarez AV, Tsutsui ND (2008) The evolutionary consequences of biological invasions. Mol Ecol 17:351‐360 Tarpy DR (2003) Genetic diversity within honeybee colonies prevents severe infections and promotes colony growth. Proc R Soc Lond Ser B‐Biol Sci 270:99‐103 Tarpy DR, Seeley TD (2006) Lower disease infections in honeybee (Apis mellifera) colonies headed by polyandrous vs monandrous queens. Naturwissenschaften 93:195‐199 Tschinkel WR (1993) Sociometry and sociogenesis of colonies of the fire ant Solenopsis invicta during one annual cycle Ecological Monographs 63:425‐457 Tschinkel WR (2010) Back to basics: sociometry and sociogenesis of ant societies (Hymenoptera: Formicidae). Myrmecol News 14:49‐54 Tsuji K (2010) What brings peace to the world of ants (Hymenoptera: Formicidae)? Myrmecol News 13:130 ‐ 132 Tsutsui N, Suarez A, Holway D, Case T (2000) Reduced genetic variation and the success of an invasive species. Proc Natl Acad Sci USA 97:5948 ‐ 5953 Tsutsui ND, Suarez AV, Holway DA, Case TJ (2001) Relationships among native and introduced populations of the Argentine ant (Linepithema humile) and the source of introduced populations. Mol Ecol 10:2151‐2161 Wetterer JK (2010) Worldwide spread of the pharaoh ant, Monomorium pharaonis (Hymenoptera: Formicidae). Myrmecol News 13:115‐129 Wheeler DE (1986) Developmental and Physiological Determinants of Caste in Social Hymenoptera: Evolutionary Implications. Am Nat 128:13‐34 Wilson EO (1971) The Insect Societies. Harvard University Press, Cambridge, Massachusetts

Chapter 1

Local scale population genetics of pharaoh ants (Monomorium pharaonis) in Thailand and the Thai‐ Malay Peninsula

Anna M. Schmidt and Jes S. Pedersen

Local scale population genetics of pharaoh ants (Monomorium pharaonis) in Thailand and the Thai‐Malay Peninsula

Anna M. Schmidt1, Watana Sakchoowong2 and Jes S. Pedersen1

1Centre for Social Evolution, Department of Biology, University of Copenhagen, Universitetsparken 15, DK‐2100 Copenhagen, Denmark

2Forest Entomology and Microbiology Research Group, National Park, Wildlife and Plant Conservation Department, 61 Phaholyothin Road, Chatuchak, Bangkok 10900, Thailand

Pharaoh ants (Monomorium pharaonis) are widespread introduced ants and well‐know pests associated with human habitation in all areas of the world. Despite frequently being referred to as unicolonial or supercolony‐forming, the native range, natural habitat and sizes of pharaoh ant colonies are all unknown. We studied the population genetics of pharaoh ants at different geographical scales in Thailand and the Thai‐Malay Peninsula to infer their mode of spread and colony sizes. Pharaoh ants occur frequently in the region, into which they have likely been introduced. Using 17 species‐specific, polymorphic microsatellite markers we analysed data from 60 different colonies at distances spanning from 7 m to 1500 km and found no signs of isolation by distance or local gene flow. The colonies were very highly differentiated with an overall Fst of 0.631 ± 0.018 SE, except for a few colonies that clustered genetically. This indicates that the mode of spread even at very short distances is human mediated, and that colonies are unlikely to grow to inhabit areas spanning more than 50 m. The genetic composition of colonies and lack of isolation by distance further indicates a history of multiple introductions or exchanges on local and regional levels. The data thus leads us to conclude that high propagule pressure must be central to the success of pharaoh ants as an introduced species.

Key words: introduced species, microsatellite, genetic bottleneck, pharaoh ant, phylogeography, propagule pressure

32 Introduction

A steadily increasing number of ant species are described as invasive or tramp species due to their high levels of success and therefore often conspicuous and harmful presence in the areas to which they are introduced (Christian, 2001; O'Dowd et al., 2003; Sarty et al., 2007). Extensive surveys have been made of the population genetics and invasion histories of a few invasive ant species, primarily focusing on the red imported fire ant (Solenopsis invicta), the little fire ant (Wasmannia auropunctata) and the

Argentine ant (Linepithema humile) (Foucaud et al., 2009; Gotelli, Arnett, 2000; Holway et al., 1998;

Human, Gordon, 1996; Jourdan et al., 2002; Le Breton et al., 2004; Mikheyev, Mueller, 2007; Ross,

Shoemaker, 2008; Tsutsui, Suarez, 2003b; Vogel et al., 2010a).

Several hypotheses have been suggested to explain the success of invasive ants, most frequently invoked is that of a special social structure causing workers to be tolerant of non‐nestmates despite very low or no relatedness, which may be a preadaptation or a result of decreased genetic variation for recognition loci as a consequence of the introduction process (Giraud et al., 2002; Starks, 2003; Tsutsui et al., 2000). For a couple of species it has been suggested that their reproductive system may confer an advantage, as the little fire ant reproduces clonally, and the yellow crazy ant (Anoplolepis gracilipes) may also have an unusual reproductive sytem which preadapts it to an invasive lifestyle (Foucaud et al., 2009;

Thomas et al., 2010). Another possible explanation for the success of certain ant species is their ability to outcompete native species through superior interference‐ as well as exploitative competition skills; this has for example been suggested to be the case for the introduced fire ant and Argentine ant, as they both come from South American habitats in which there are high levels of competition (Feener, 2000). In addition, the size of the individual worker ants and the reproductive mode of the colonies, which is the budding off of part of a colony to establish new colonies, allowing them to slowly expand through

33 interconnected networks rather than engaging in risky mating flights, are both factors very likely to contribute to some species’ survival through the initial phase of the introduction process, resulting in a high propagule pressure (Holway et al., 2002; McGlynn, 1999).

Numerous ant species, however, have been introduced to areas outside their native range with highly varying success in becomming established, so in order to make general inferences regarding the traits that make introduced ants successful, comparisons between multiple species are needed (Helanterä et al., 2009; Suarez et al., 2010). Among the members of the group of introduced species, the pharaoh ant

(Monomorium pharaonis) may well be the species with the longest introduction history. Already described as an introduced species by Linneaus in 1758 (Wetterer, 2010), the pharaoh ant is now considered a ubiquitous species and a widespread household pest (Edwards, 1986; Edwards, Baker,

1981; Peacock et al., 1950). Pharaoh ants are “typical” tramp ants as they are small and opportunistic in their feeding as well as nesting habits, have polygynous colonies (i.e. with multiple reproducing queens), and the colonies reproduce by budding (Berndt, Eichler, 1987; Passera, 1994). In addition, pharaoh ants display low levels of intraspecific aggression (Schmidt et al., 2010), and, like the majority of other invasive ant species, they are considered to be unicolonial and form supercolonies (Helanterä et al.,

2009; Hölldobler, Wilson, 1990; Passera, 1994).

Pharaoh ants are among the most frequently intercepted ants at customs in the USA and New Zealand

(Lester, 2005; Suarez et al., 2005). Although it has been shown that introduced populations of pharaoh ants are likely unintentionally spread by humans and their colonies are highly genetically differentiated

(Schmidt et al., 2010), their native range, natural mode of spread, and the size of their colonies in natural or semi‐natural habitat are unknown. To better understand what makes pharaoh ants successful, we conducted a population genetic study investigating the occurrence and population genetic structure

34 of pharaoh ants in Thailand and the Thai‐Malay Peninsula where the species is common. We asked whether populations appeared to be the product of multiple separate introductions and whether there was any local gene flow and exchange of individuals between colonies, as well as how big a single colony would typically be.

Methods

Study material

Field work was conducted in central‐ and southern Thailand in 2008, during which 240 different localities were searched for pharaoh ants through a combination of stops at houses and gas stations along major roads and different size towns and searches at field sites in national park forests. Pharaoh ants’ native range and therefore also natural habitat have not been identified, and they are primarily known from houses or habitats modified by human activities. Thailand was chosen as an area to do extensive sampling because the species is common there (AMS, pers. obs.), also enabling sampling at different geographical scales. The area is also interesting because pharaoh ants have been hypothesized to be native to central Asia (reviewed in Wetterer, 2010) and have previously been found in fairly undisturbed habitats in Thailand. Our sampling was focused in and around buildings where we observed that several

Monomorium species were very common, whereas we had no luck finding Monomorium when searching forest localities, in branches and sifting through leaf litter. At all localities it was noted whether other ant species were present, and whether these were native to the area or introduced.

M. pharaonis workers or whole colonies were collected at over 100 localities of these 55 colony samples were chosen to represent different geographical scales and areas for further study. An additional five samples were obtained from Thailand and Malaysia thus combined representing sampling points of

35 distances spanning from less than 10 m to over 1500 km. (Sample ID and localities are given in Table 1).

The sampling covered longitudinal as well as latitudinal gradients combined with intense small scale sampling in a number of villages (Fig. 1). Samples were collected and stored in 96% EtOH until analysed.

Genetic analyses

The genetic variation and differentiation of colonies was examined based on 17 microsatellite loci for which primers were developed for M. pharaonis. Four primers had previously been developed (Schmidt et al., 2010) and an additional 13 were developed for this study, 11 of which were outsourced to

Ecogenics (Wagistrasse 27, CH‐8952, Zurich‐Schlieren, Switzerland) (Table 2). Primer development was through enrichment of target DNA for selected repeat motifs, subsequent screening and sequencing of positive clones and design of primers for suitable microsatellite loci. DNA was extracted from 10 or 20 workers from each colony (Table 1) by each specimen being crushed in 200 µl 5% Chelex solution, or extractions were made using Invisorb Spin Tissue mini kits (10 worker ants from each colony included).

The primers were run in separate PCR reactions in 20 µl reaction volumes with 1 µl extract as template, reaction buffer, 25 mM MgCl2, 0.5 mM GATC, 10 µM forward and reverse primers and 0.1 µl Taq gold polymerase. The PCR reaction conditions consisted of an initial denaturing step of 95 °C for 10 min, followed by 27 cycles of 95 °C for 30 s, 60 °C for 30 s, and 72 °C for 30 s, and finally an extension step at

72 °C for 10min. PCR products were run on 2% ethidium bromide agarose gels, and subsequently run on an Applied Biosystems 3130xl Genetic Analyzer automated sequencer with LIZ500 as internal standard. A multiplex protocol was developed to multiplex the primers in groups of 4 or 5 in the PCR and 8‐9 on the

ABI (Table 2). GeneMapper version 4.0 was used to score the alleles.

36 Statistical analyses

The genetic differentiation (FST) between the colonies as well as measures of allelic richness (k´) and inbreeding at the colony level (FIS) were calculated using FSTAT 2.9.3.2 (Goudet, 2002) with significance testing based on 15,000 randomisations. Because the value of FST is affected by the allelic diversity at the marker loci applied, we further calculated the standardised F´ST (Hedrick, 2005) and the estimator Dest

(Jost, 2008) as alternative quantifications of the genetic differentiation, making comparisons with studies based on other marker loci possible (Heller, Siegismund, 2009).

The program BOTTLENECK 1.2.02 was used to detect signs of recent reduction in population through estimation of heterozygosity, as excess heterozygosity relative to that expected in equilibrium populations can be seen as an indicator of recent reductions in population size (Cornuet, Luikart, 1996;

Piry et al., 1999). The method does not require a native reference population which is an advantage for species like pharaoh ants with an unknown native range. A two‐phase model (TPM) for evolution of loci was applied, combining 95% single‐step mutations (SMM) and 5% multiple‐step mutations (IAM) as recommended by the authors (Piry et al., 1999). Wilcoxon’s tests were applied to evaluate the null hypothesis of no significant heterozygosity excess on average across loci separately for each populations.

Non‐spatial genetic mixture analysis was applied as implemented in BAPS 5.2 to cluster colonies that were likely to be genetically similar, be this through a common population history or because they represented the same supercolony (Pedersen et al., 2006; Vogel et al., 2010b). The maximum number of genetically divergent groups (K) was set to the number of colonies included in the data set (=60), and the analysis was repeated 10 times to ensure consistency of results between different runs. Genetically

37 divergent groups of colonies identified this way are interpreted as colony lineages. A Mantel test for possible correlation between genetic (FST) and geographical distances between colonies was performed in FSTAT (Goudet, 2002), based on 10,000 permutations. In addition, a principal component analysis

(PCA) of the microsatellite allele frequency data was performed in PCAGEN 1.3.1 (Goudet, 2000) as an exploratory technique to visualise colony or population associations possibly reflecting the phylogeographic history of the species.

Results

Pharaoh ants were very common in and around houses as we found them in ca. 50% of all houses searched. Houses appeared dominated by a small group of recurring introduced and invasive ant species including the Monomorium species M. pharaonis, M. destructor and M. floricola as well as the crazy ants

Anoplolepis gracilipes and Paratrechina longicornis. Of a total of 107 localities where pharaoh ants were found, one third had other Monomorium species present as well, these were other introduced species:

M. floricola or M. destructor; in one locality all three species occurred, in 19 M. floricola was present, and in 10 M. destructor was present. At no locality were the foraging trails of the different species seen in close proximity as they would usually be several meters apart.

Genetic diversity measures

Table 2 gives information on the microsatellite markers employed. The genetic diversity within colonies was generally low, as most colonies had several monomorphic loci. The expected heterozygosity (Hexp) was 0.21 ± 0.013 SE on average across colonies and the allelic richness (k’) was 1.52 ± 0.038 SE when adjusted to a minimum sample size of three individuals. As expected, these measures of the genetic diversity were highly correlated (Spearman r = 0.991; P < 0.0001). There was only a small, but statistically

38 significant, deviation from random mating overall within the colony samples (FIS = 0.051 ± 0.034 SE; P =

0.00067), in general confirming that each colony formed its own breeding population as seen in other invasive ant species (e.g. Pedersen et al., 2006). Furthermore, a few colonies displayed heterozygosity deficit indicating some level of inbreeding (cf. Table 1 and section on bottlenecks below).

Inter‐colonial differentiation

We consistently found very high and statistically significant values of inter‐colonial genetic differentiation (mean FST = 0.631 ± 0.018 SE; corresponding to F´ST = 0.802 and Dest = 0.462; n = 60; see

Supplemental File 1 for estimates of pairwise FST). The BAPS analyses yielded 58 genetic clusters as only two pairs of two colonies were grouped (Th13 + Th21 and Th136 + Th138; Table 1). The result (K = 58) had strong support by a posterior probability P = 1.000 in all 10 runs, and both cluster pairs had been collected at relatively close proximity being only seven and ca. 50 m apart, respectively, indicating that colonies can spread locally within this scale. However, otherwise the data give no indication of local exchange or intermixing of colonies. Numerous samples collected less than 50 meters apart were as highly differentiated as those collected over 500 km apart, indicating that the extent of local mixture of colonies is very limited, and that the species is likely moved around by humans extensively.

A Mantel test for association of geographical and genetic distances between colonies further confirms that there is no genetic isolation by distance between the colonies, as it did not show a significant correlation (Fig. 2; r = 0.001; P = 0.959). This was likewise reflected in the PCA plot (Fig. 3), on which the colonies coordinates did not reflect their geographic location. The four samples clustered on the far right along the PC1 axis, have thus been collected at localities quite distantly apart (cf. Table 1). Consequently, the colonies in the two groups containing the four colonies and the remaining 56 colonies were all significantly genetically divergent, indicating that there were 58 independent lineages represented in our

39 sample. Collapsing the samples clustering in the BAPS analyses caused only little change in the differentiation estimators as mean FST (n = 58) remained unchanged at 0.631 ± 0.018 SE; whereas

F´ST was increased slightly to 0.805 and Dest to 0.471. Separate analyses of six subgroups of geographically very close colonies (i.e. scale of a single village) likewise painted a picture of high levels of differentiation even at very small geographical scales (Table 3).

Detection of recent population bottlenecks

The program BOTTLENECK indicated a significant heterozygosity excess (ΔH/SD) in nine of the populations, and heterozygosity deficit in five, however, considering the potential age of these populations, it is not unlikely that the number of populations assigned as bottlenecked is an underestimate as the algorithm detects recent population bottlenecks only (Williamson‐Natesan, 2005).

Surprisingly, plotting heterozygosity (ΔH/SD) against allelic richness or expected heterozygosity shows a tendency towards a positive correlation between the two measurements, which is opposite the expectation for populations gone through recent reductions in population size. A simple linear regression is not statistically significant for the association with allelic richness (PASW; R = 0.229, P =

0.078), but significant in the case of ΔH/SD vs. arcsine transformed Hexp (PASW; R = 0.373, P = 0.003).

Discussion

Our findings show a history of multiple introductions of M. pharaonis into Thailand and no or very little gene flow between colonies even over very small distances. The sampled colonies are very highly genetically differentiated with FST = 0.631, and when analysed at the same four loci as those of Schmidt et al. (2010) Fst = 0.632, which is slightly lower than the estimate at global scale (Fst = 0.751), but still extremely high compared to those of other species. Even the differentiation among colonies within each

40 of six different groups representing very short distance sampling in neighbouring houses and villages

(Table 3) is comparable to the highest levels found in other sexually reproducing species (Heller,

Siegismund, 2009). Combined with the generally low levels of genetic diversity within colonies, this suggests that very little gene flow has taken place in the past, and possibly that new colonies are continuously being introduced though human mediated jump dispersal to human modified areas (King,

Tschinkel, 2008; Schmidt et al., 2010; Suarez et al., 2001).

Such a propagule pressure scenario fits well with the interception data from other countries (Lester,

2005; Suarez et al., 2005), and the strength of pharaoh ants as invaders may very well be found in the high propagule pressure they exert through numerous, continuous introductions. The habitats we found pharaoh ants in were decidedly created and disturbed by humans, most workers were found in trails on or in houses and colonies in walls, books, clothes etc., where no other ants were to be seen. This seems to be the type of environment to which they are adapted. Therefore pharaoh ants do not appear to face biotic resistance in the classical sense of the term (Sakai et al., 2001), by competing with the native ant guild, but rather the habitat they occur in seems to be dominated by a guild of introduced ant species, all to be found only in close proximity of houses. Recent research on the little fire ant likewise suggests multiple introductions as one of the factors determining its success (Mikheyev et al., 2008) as well as through being a specialist in disturbed habitats (Foucaud et al., 2009). This is likely the case for many invasive species (Suarez, Tsutsui, 2008), supporting the notion that natural disturbances may give rise to preadaptations that eventually lead to supercolony formation or unicoloniality (Tsuji, 2010) and that propagule pressure can play a central role.

The PCA plot indicated that a small group of colonies was genetically distinct from the main sample:

MpU55, MpU65, Th74 and Th183. Interestingly, the former three were the only colonies collected under

41 slightly more natural conditions, natural here meaning away from houses (Table 1). MpU55 and MpU65 were both collected in forests and the Th74 workers were found walking on a tree trunk ca. 5 m from a house, in a national park area. The last colony was found on an old roadside shed, previously used for a small business, but now abandoned. We do not know what had caused these colonies to be genetically distinct, but as MpU55, MpU65 and Th183 also have the highest allelic richness of the entire sample, it may be that they represent introductions from a different source in the unknown native range, or are in fact cryptic species.

The lack of detection of population bottlenecks in most cases is interesting, as judging from the biology of the species colonies must clearly have been bottlenecked. In contrast, there is a positive association of the heterozygosity deviation (ΔH/SD) from the BOTTLENECK analyses and either measure of colony genetic diversity (allelic richness or expected heterozygosity; Fig. 4), which is opposed to the expected negative correlation that would result from loss of alleles and simultaneous increase in heterozygosity excess in recently bottlenecked populations. This suggests that the bottleneck algorithm may not be applicable in a species with this particular kind of population dynamics, although the overall average heterozygosity deviation across colonies is slightly positive at 0.053± 0.056 SE.

The data thus tell a story of multiple introductions successfully leading to the establisment of new, genetically isolated colonies, thus creating islands of invasion that may be highly genetically differentiated despite being at very short geographical distances. Although pharaoh ants have displayed low levels of intraspecific aggression in the laboratory (Schmidt et al., 2010), indicating that colonies would have the potential to merge, should they meet, it appears that they rarely, if ever, do so. If the colonies do not exchange genetic material and will therefore be increasingly inbred, that must imply strong selection at the colony level, and leave little room for adaptation to environmental changes. This

42 also means that what makes pharaoh ants good invaders now may not be an evolutionarily stable strategy in the future. However, the large number of genetically divergent colonies may have a very high turnover, as it seems plausible that whole colonies may easily frequently be moved around due to their habit of moving nests periodically. Such moving can result in the introduction of new colonies and removal of old, which creates a basis for lineage selection at the local scale (Helanterä et al., 2009;

Nunney, 1999 ). This can be seen as a contrast to most other described introduced ants, which only appear to have relatively few lines from the native range represented in the introduced area (Foucaud et al., 2010; Giraud et al., 2002; Helanterä et al., 2009; Thomas et al., 2006; Tsutsui, Suarez, 2003a; Vogel et al., 2010a). One might therefore speculate whether the pharaoh ant has been able to evaded the detrimental long‐term consequences predicted for unicolonial species (Helanterä et al., 2009{Queller,

1998 #463)} through selection acting on many highly differentiated lineages, continuously being introduced.

43 Acknowledgements

We thank Daniel Kronauer for help and advise developing microsatellite primers and Sylvia Mathiasen for help in the lab. We thank The National Research Council of Thailand (NRCT) and the Department of

National Parks, Wildlife and Plant Conservation (DNP) for granting us permits to collect ants in Thailand.

David Hughes, Winanda Himaman, Chaweewan Hutacharern and David Lohman for help and advice on field work in Thailand. Weeyawat Jaitrong and Thailand Natural history Museum for collaboration in the field. Pitoon Kongnoo for assistance and translation in the field and Tana Thongrod for field assistance.

We thank Weeyawat Jaitrong, David Lohman, Timothy Linksvayer and Sze Hui Yek for kindly sending ants. The study was funded through a grant from The Danish National Research Foundation to the

Centre for Social Evolution.

44 References

Berndt KP, Eichler W (1987) Die Pharaoameise, Monomorium pharaonis (L.) (Hym., Myrmicidae). Mitteilungen aus dem Zoologischen Museum in Berlin 63, 3 ‐186. Christian CE (2001) Consequences of a biological invasion reveal the importance of mutualism for plant communities. Nature 413, 635‐639. Cornuet JM, Luikart G (1996) Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data. Genetics 144, 2001‐2014. Edwards JP (1986) The biology, economic importance, and control of the Pharaoh's Ant In: Economic Impact and Control of Social Insects (ed. Bradleigh Vinson S), pp. 257‐271. Praeger Publishers, New York. Edwards JP, Baker LF (1981) Distribution and importance of the Pharaoh's ant Monomorium pharaonis (L) in National Health Service Hospitals in England. Journal of Hospital Infection 2, 249‐254. Feener DJ (2000) Is the assembly of ant communities mediated by parasitoids? Oikos 90, 79‐88. Foucaud J, Orivel J, Fournier D, et al. (2009) Reproductive system, social organization, human disturbance and ecological dominance in native populations of the little fire ant, Wasmannia auropunctata. Molecular Ecology 18, 5059‐5073. Foucaud J, Orivel J, Loiseau A, et al. (2010) Worldwide invasion by the little fire ant: routes of introduction and eco‐evolutionary pathways. Evolutionary Applications 9999. Giraud T, Pedersen JS, Keller L (2002) Evolution of supercolonies: The Argentine ants of southern Europe. Proceedings Of The National Academy Of Sciences Of The United States Of America 99, 6075‐6079. Gotelli NJ, Arnett AE (2000) Biogeographic effects of red fire ant invasion. Ecology Letters 3, 257‐261. Goudet J (2000) PCA‐GEN, a program to perform Principal Component Analysis (PCA) on gene frequency data. Goudet J (2002) FSTAT, a program to estimate and test gene diversities and fixation indices. Institute of Ecology, University of Lausanne, Lausanne.

45 Hedrick PW (2005) A standardized genetic differentiation measure. Evolution 59, 1633‐1638. Helanterä H, Strassmann JE, Carrillo J, Queller DC (2009) Unicolonial ants: where do they come from, what are they and where are they going? Trends in Ecology & Evolution 24, 341‐ 349. Heller R, Siegismund HR (2009) Relationship between three measures of genetic differentiation

GST, DEST and GST: how wrong have we been? Molecular Ecology 18, 2080‐2083. Holway DA, Lach L, Suarez AV, Tsutsui ND, Case TJ (2002) The causes and consequences of ant invasions. Annual Review of Ecology and Systematics 33, 181‐233. Holway DA, Suarez AV, Case TJ (1998) Loss of intraspecific aggression in the success of a widespread invasive social insect. Science 282, 949‐952. Human KG, Gordon DM (1996) Exploitation and interference competition between the invasive Argentine ant, Linepithema humile, and native ant species. Oecologia 105, 405‐412. Hölldobler B, Wilson EO (1990) The Ants Springer, New York. Jost L (2008) GST and its relatives do not measure differentiation. Molecular Ecology 17, 4015‐ 4026. Jourdan H, de Larbogne LB, Chazeau J (2002) The recent introduction of the neotropical tramp ant Wasmannia auropunctata (Hymenoptera : Formicidae) into Vanuatu archipelago (Southwest Pacific). Sociobiology 40, 483‐509. King J, Tschinkel W (2008) Experimental evidence that human impacts drive fire ant invasions and ecological change. Proc Natl Acad Sci USA 105, 20339 ‐ 20343. Le Breton J, Delabie JHC, Chazeau J, Dejean A, Jourdan H (2004) Experimental evidence of large‐ scale unicoloniality in the tramp ant Wasmannia auropunctata (Roger). Journal of Insect Behavior 17, 263‐271. Lester PJ (2005) Determinants for the successful establishment of exotic ants in New Zealand. Diversity & Distributions 11, 279‐288. McGlynn TP (1999) Non‐native ants are smaller than related native ants. American Naturalist 154, 690‐699.

46 Mikheyev AS, Mueller UG (2007) Genetic relationships between native and introduced populations of the little fire ant Wasmannia auropunctata. Diversity and Distributions 13, 573‐579. Mikheyev AS, Tchingnoumba L, Henderson A, Alonso A (2008) Effect of propagule pressure on the establishment and spread of the little fire ant Wasmannia auropunctata in a Gabonese oilfield. Diversity and Distributions 14, 301‐306. Nunney L (1999 ) Lineage selection: natural selection for long‐term benefit. In: Levels of Selection in Evolution (ed. Keller L), pp. 238‐252. Princeton University Press, New Jersey. O'Dowd DJ, Green PT, Lake PS (2003) Invasional 'meltdown' on an oceanic island. Ecology Letters 6, 812‐817. Passera L (1994) Characteristics of tramp species. In: Exotic Ants, Biology, Impact, and Control of Introduced Species (ed. Williams DF), pp. 23‐43. Westview Press, Boulder, Colorado. Peacock AD, Hall DW, Smith IC, Goodfellow A (1950) The biology and control of the ant pest Monomorium pharaonis (L.). Dept. Agr. Scot.; Misc. Publ. no. 17: 1‐51. Pedersen JS, Krieger MJB, Vogel V, Giraud T, Keller L (2006) Native supercolonies of unrelated individuals in the invasive Argentine ant. Evolution 60, 782‐791. Piry S, Luikart G, Cornuet JM (1999) BOTTLENECK: A computer program for detecting recent reductions in the effective population size using allele frequency data. Journal Of Heredity 90, 502‐503. Ross KG, Shoemaker DD (2008) Estimation of the number of founders of an invasive pest insect population: the fire ant Solenopsis invicta in the USA. Proceedings of the Royal Society B‐ Biological Sciences 275, 2231‐2240. Sakai AK, Allendorf FW, Holt JS, et al. (2001) The population biology of invasive species. Annual Review of Ecology and Systematics 32, 305‐332. Sarty M, Abbott KL, Lester PJ (2007) Community level impacts of an ant invader and food mediated coexistence. Insectes Sociaux 54, 166‐173. Schmidt AM, d'Ettorre P, Pedersen JS (2010) Low levels of nestmate discrimination despite high genetic differentiation in the invasive pharaoh ant. Frontiers in Zoology 7, 20.

47 Starks PT (2003) Selection for uniformity: xenophobia and invasion success. Trends in Ecology & Evolution 18, 159‐162. Suarez AV, Holway DA, Case TJ (2001) Patterns of spread in biological invasions dominated by long‐distance jump dispersal: Insights from Argentine ants. Proceedings of the National Academy of Sciences of the United States of America 98, 1095‐1100. Suarez AV, Holway DA, Ward PS (2005) The role of opportunity in the unintentional introduction of nonnative ants. PNAS, 0506119102. Suarez AV, McGlynn TP, Tsutsui ND (2010) Biogeographic and taxonomic patterns of introduced ants Oxford University Press. Suarez AV, Tsutsui ND (2008) The evolutionary consequences of biological invasions. Molecular Ecology 17, 351‐360. Thomas ML, Becker K, Abbott K, Feldhaar H (2010) Supercolony mosaics: two different invasions by the yellow crazy ant, Anoplolepis gracilipes, on Christmas Island, Indian Ocean. Biological Invasions 12, 677‐687. Thomas ML, Payne‐Makrisa CM, Suarez AV, Tsutsui ND, Holway DA (2006) When supercolonies collide: territorial aggression in an invasive and unicolonial social insect. Molecular Ecology 15, 4303‐4315. Tsuji K (2010) What brings peace to the world of ants (Hymenoptera: Formicidae)? Myrmecological News 13, 130‐132. Tsutsui N, Suarez A (2003a) The colony structure and population biology of invasive ants. Conservation Biology 17, 48 ‐ 58. Tsutsui ND, Suarez AV (2003b) The colony structure and population biology of invasive ants. Conservation Biology 17, 48‐58. Tsutsui ND, Suarez AV, Holway DA, Case TJ (2000) Reduced genetic variation and the success of an invasive species. Proceedings of the National Academy of Sciences of the United States of America 97, 5948‐5953. Vogel V, Pedersen J, Giraud T, Krieger M, Keller L (2010a) The worldwide expansion of the Argentine ant. Diversity and Distributions 16, 170 ‐ 186.

48 Vogel V, Pedersen JS, Giraud T, Krieger MJB, Keller L (2010b) The worldwide expansion of the Argentine ant. Diversity and Distributions 16, 170‐186. Wetterer JK (2010) Worldwide spread of the pharaoh ant, Monomorium pharaonis (Hymenoptera: Formicidae). Myrmecological News 13, 115‐129. Williamson‐Natesan EG (2005) Comparison of methods for detecting bottlenecks from microsatellite loci. Conservation Genetics 6, 551‐562.

49 Table 1 Samples of Monomorium pharaonis. Colony ID is the abbreviations used in further analyses and figures. n is the number of samples amplifying for at least one locus. The total number of polymorphic loci was 17; P is the number of polymorphic loci for a given colony, k' is the average allelic richness, Hexp is the expected average heterozygosity for each colony; k’ and Hexp are adjusted to a sample size of three. ΔH/SD are averages of the heterozygosity deviation across polymorphic loci in BOTTLENECK based on a minimum of 3 individuals per locus.

Locality GPS Colony ID Latitude N Longitude E n P k’ Hexp ΔH/SD Locality description Th4a 12.2335 102.4007 20 12 1.84 0.34 0.31 Field station Th13 13.0017 102.1655 20 6 1.24 0.11 ‐0.28 Field station Th21a 13.0013 102.1653 10 3 1.20 0.09 0.19 Field station Th22 12.3654 102.0605 20 11 1.55 0.25 0.41 Large shop Th25 13.2517 102.1155 20 7 1.54 0.22 0.33 Gas station Th74a 12.5304 99.3801 10 8 1.59 0.25 0.49* Tree Th76 12.4758 99.2715 20 11 1.80 0.33 0.68* Building Th87 7.3312 99.4617 20 5 1.23 0.10 ‐0.14 Restaurant Th90 7.3301 99.4606 20 10 1.40 0.17 ‐0.16 Shop Th97 7.3319 99.4409 20 13 1.48 0.17 ‐0.59* House Th99 7.3324 99.4631 20 10 1.58 0.24 0.38 Shop/restaurant Th105 7.3308 99.4615 20 7 1.46 0.20 0.64 Shop Th116 7.3336 99.4255 20 4 1.20 0.09 0.78 Restaurant Th126 7.3325 99.4334 20 13 1.73 0.29 ‐0.01 Restaurant Th132 7.3322 99.3631 20 11 1.65 0.29 0.82* Restaurant Th133a 7.3316 99.3618 10 5 1.26 0.10 ‐0.79* Building Th136 7.3310 99.4411 20 10 1.43 0.18 ‐0.23 House Th138 7.3310 99.4410 20 10 1.47 0.19 ‐0.55 House Th139 7.3308 99.4413 20 6 1.33 0.15 0.57* House Th140 7.3307 99.4412 20 7 1.28 0.14 0.35 House/tea stand Th141 7.3307 99.4412 20 11 1.68 0.28 0.13 House Th142 7.3258 99.4341 20 9 1.47 0.19 0.03 House/mechanic Th143 7.3302 99.4341 20 3 1.21 0.09 ‐0.14 House Th151 7.3024 99.4550 20 6 1.26 0.10 ‐0.25 Shop Th163 7.4043 99.5216 20 8 1.26 0.10 ‐0.71* Building Th166a 7.4438 99.5429 20 9 1.41 0.18 0.06 Shed Th167 7.4436 99.5420 20 8 1.30 0.13 ‐0.07 House Th168 7.4403 99.5415 20 9 1.45 0.18 ‐0.30 House Th171 7.4358 99.5418 20 10 1.40 0.16 ‐0.55 House Th173 7.4408 99.5446 10 8 1.84 0.34 1.09* House Th182 7.4334 99.5446 20 9 1.26 0.10 ‐0.64 House Th183a 7.4332 99.5446 20 16 2.89 0.59 ‐0.75* Shed Th185 7.4332 99.5440 20 10 1.64 0.26 0.23 House Th187 7.4312 99.5345 20 11 1.60 0.27 0.28 House Th188 7.4323 99.5305 20 10 1.67 0.28 0.66* House/shop Th193 7.4405 99.5405 20 7 1.36 0.15 ‐0.04 House

50 Th194 7.4404 99.5407 20 8 1.37 0.15 0.11 House Th198 8.2036 100.1230 20 12 1.67 0.28 ‐0.05 Restaurant Th200 8.2125 100.1216 20 7 1.28 0.13 0.06 Shop Th204 8.2207 100.1426 20 12 1.66 0.26 ‐0.44 Shop Th209 8.2125 99.4759 20 5 1.25 0.11 0.12 Restaurant Th210 8.1754 99.2129 20 10 1.33 0.13 ‐0.79* House Th214 8.3007 98.3747 20 13 1.58 0.25 ‐0.15 Restaurant Th216 8.3059 98.3821 20 11 1.60 0.26 0.16 Restaurant Th217 8.3112 98.3824 20 6 1.28 0.12 0.15 Shop Th218 8.3120 98.3825 20 10 1.67 0.25 ‐0.17 Restaurant Th219 8.3243 98.3847 20 10 1.56 0.22 ‐0.06 Office building Th220 8.0155 98.5144 20 12 1.75 0.30 0.15 Shop Th223 8.0147 98.5203 20 10 1.54 0.25 0.66* Shop Th231 6.5112 99.4344 20 13 1.86 0.35 0.18 Shop Th232 6.5107 99.4355 20 6 1.18 0.08 ‐0.31 Shop Th233 6.5136 99.4624 20 7 1.27 0.11 ‐0.41 Shop Th234 6.5207 99.4651 20 8 1.38 0.16 ‐0.13 Restaurant Th235 6.5301 99.4709 20 10 1.43 0.18 ‐0.27 Large shop ThGrH 7.3324 99.4631 20 7 1.41 0.17 0.26 Field station U32 3.17 101.45 20 12 2.01 0.41 0.46* Field station U45 18.5014 98.5814 10 9 1.50 0.23 0.22 Hotel U55 12. 5848 102.1749 20 14 2.06 0.42 0.48* Forest U64 3.00 102.33 20 9 1.37 0.17 ‐0.10 Field station U65a 14.31 101.55 10 13 2.21 0.47 0.82* Forest

a indicates that P, k’ and Hexp given were based on fewer than 17 loci as four samples failed to amplify for one or more loci (Th74 and Th133 did not amplify Mph2, Th74 largely failed for Mp3, and Th133 for Mph9; U65 did not amplify Mp4 nor Mph26. Th21 largely failed for Mph2 and Mph26). * indicates that there was a significant heterozygosity excess or deficit in the BOTTLENECK analyses.

51 Table 2 Microsatellite loci analysed. Data on allele number and size range are for the total sample of 60 Monomorium pharaonis colonies analysed (n = 1140 individuals).

Locus Primer sequences (5´‐3´) PCR Repeat motif GenBank No. of Size range (bp) multiplex accession no. alleles

Mp1 f: GCCAATGGTTTAATCCCTCA 1 (AG)3AA(AG)23 HM587312 21 190‐231 r: TCATACTGCGTGTGCCTTTC

Mp3 f: ACAAGGTAAGTCGCCACCAT 1 (GT)20 HM587313 15 116‐149 r: TCGTGATAATTCGCGATGAA

Mp12 f: TGGCCAAAAGTATCCAGGAG 1 (AC)8 HM587314 6 126‐139 r: TCGTCGAAAGTATCGAAGTAAAC

Mp13 f: CCCATTGAGATTGCGGCAT 1 (AC)10(GC)8ACA(AC)10 HM587315 15 253‐299 r: GCACAGGCACGTAACGATT

Mp4 f: CGGCAAATGCACAAACATTA 2 (GT)17 To be added 7 279‐303 r: CGTGAGGTCAAAAGTTCCGTA

Mp8 f: TTGTATGGGAAAGGCGAAAG 2 (GA)6GG(GA)12 To be added 10 97‐119 r: TAAGCCTCCTTGCACAATCC

Mph1 f: AACGTGTCGCCATTTAGAGG 4 (AC)14 To be added 10 148‐190 r: CGCGAATTTATGACGATCAG

Mph2 f: CGCACTAAAGCGCGAAAG 2 (TC)13 To be added 22 190‐236 r: ATCGTCGGTGTCCTCATTTC

Mph3 f: GACCGAGGAAGATTGAGGAAG 4 (GT)15 To be added 16 195‐236 r: GCACATGCATTATCACAAACG

Mph6 f: AGCGAGCACAAATGCATCC 4 (TC)9 To be added 12 87‐130 r: GACCCGGTGGCACACTTC

Mph9 f: TTCCTGCATCAAAAGTGCTG 2 (CT)3CC(CT)17 To be added 9 161‐203 r: ACAGTTGGCCGCATAAATTG

Mph1 f: GCGTTGCTATATCGCTTCCA 3 (TG)14 To be added 7 203‐229 CAGATATGCACGCATTAACGA

Mph1 f: GTGACAAATCGGACATCTCG 3 (TG)15 To be added 11 190‐218 r: TTAACGAAATGTAACGTTCAATAGAC

Mph1 f: TCAAGCCATTAGAAGCAAAGC 4 (AG)15 To be added 16 134‐177 r: GAGAAGGCGGCAAAGAAAC

Mph2 f: GGAGGCACGTTACTTGCTG 3 (TG)23 To be added 10 154‐178 r: CTGTCAAGTCGGATTACTGTGC

Mph2 f: CGCGAGGAGACGAACTACC 2 (GA)27 To be added 18 75‐115 r: TGCTCGTTTCTCGACGTATG

Mph2 f: CCTTTCACTTCAAATTTGTCCAG 3 (TC)27 To be added 18 91‐142 r: TCAGGTTTACATTTAAAGGATGG

52 Table 3 Groups of Monomorium pharaonis samples collected at short distances. n is the number of colonies included in any given group, the colony IDs of the samples included are as in Table 1. FST, F’ST and Dest were calculated separately for each group.

Group n Distance (km) F ST ±SE F ´ST Dest Samples included Khao Chong 8 ≤ 3.7 0.636 ± 0.042 0.796 0.438 Th136‐Th143 Na Yong 6* ≤ 1.6 0.688 ± 0.027 0.860 0.550 Th136‐Th143 Phatthalung 12 ≤ 1.6 0.607 ± 0.024 0.813 0.523 Th166a‐Th94 Pak Phanang 3 ≤ 3 0.521 ± 0.008 0.745 0.468 Th198‐Th204 Phang‐nga 5 ≤ 2.8 0.621 ± 0.039 0.839 0.574 Th214‐Th219 Satun 6 ≤ 4.1 0.615 ± 0.053 0.771 0.407 Th231‐Th235

* Na Yong had 7 sampling localities, but two of these, which were only 7 m apart, turned out to be the same colony (see text for explanation), and these have therefore been merged in the group analyses.

53 Fig. 1 Geographical positions of sampling localities for colonies of Monomorium pharaonis (n = 60) in Thailand and Malaysia

54 Fig. 2 Association of genetic and geographical distances between colonies of Monomorium pharaonis. A Mantel test yielded no significant correlation (r = 0.001; P = 0.97).

0.9

0.8 )

ST 0.7 (F

0.6

0.5 distance 0.4

0.3

Genetic 0.2

0.1

0 0.001 0.01 0.1 1 10 100 1000 10000 Geographical distance (km)

Fig. 3 Principal Component Analysis of microsatellite allele frequencies of Monomorium pharaonis colonies based on 14 microsatellite loci. Colony labels are as in Table 1. The explained variances are given in parentheses.

(10%) PC2

PC1 (16 %)

55 Fig. 4 Deviation in heterozygosity (ΔH/SD) from the BOTTLENECK analyses of colonies of Monomorium pharaonis plotted against (a) allelic richness (k’) and (b) expected heterozygosity (Hexp).

a b

Supplemental File 1 Estimates of pairwise FST between colonies of Monomorium pharaonis.

Chapter 2

Frequent occurrence of Wolbachia in introduced populations of pharaoh ants

Anna M. Schmidt and Jes S. Pedersen

Frequent occurrence of Wolbachia in introduced populations of pharaoh ants

Anna M. Schmidt and Jes S. Pedersen

Centre for Social Evolution, Department of Biology, University of Copenhagen, Universitetsparken 15, DK‐2100 Copenhagen, Denmark

An increasing number of highly successful ant species invade human habitation in most parts of the world. Which factors lead to their success is debated, and enemy release through the loss of the intracellular bacterium Wolbachia has been suggested as a possible contributing factor because Wolbachia has been found frequently in native, but is lacking or rare in introduced populations of the highly invasive fire ants (Solenopsis invicta) and Argentine ants (Linepithema humile). To test the generality of this hypothesis, we screened workers from 103 introduced pharaoh ant (Monomorium pharaonis) colonies as well as colonies representing 11 other invasive or non‐invasive Monomorium species for presence of Wolbachia using the wsp81F and wsp691R primers. A subset of samples was sequenced to investigate whether multiple strains occur in pharaoh ants, and whether strains are similar across Monomorium species or populations. We found Wolbachia in 34% of all pharaoh ant colonies. This relatively high frequency raises the question of whether Wolbachia is harmful to pharaoh ants, since release from the bacterium does not appear to be a prerequisite for successful invasion. Occurrence in the colonies of other Monomorium species ranged from 0 to 100%, possibly with some regional variation. The screened pharaoh ants only contained single strains, but several strains were present in the sequenced species and overlap in these possibly reflect a history of horizontal transmissions. Our results suggest that studies of introduced species may give new perspectives on the harmfulness and transmission modes of bacterial symbionts.

Key words: introduced species, intracellular bacteria, enemy release, Monomorium pharaonis, symbiont

60 Introduction

Wolbachia are maternally inherited intracellular alpha‐proteobacteria that are very common in arthropods and nematodes, and estimated to be found in 66% of all arthropod species

(Hilgenboecker et al., 2008; Jeyaprakash, Hoy, 2000; Werren, 1997; Werren et al., 1995), in which they are known to cause a number of different effects. Whereas the interactions between

Wolbachia and their nematodal hosts are known to be benign and mutualistic to the point that filarial nematodes will die if their Wolbachia are removed (Smith, Rajan, 2000), the effects on insects have generally been thought to be detrimental (Werren et al., 2008).

In insects Wolbachia can cause different kinds of reproductive parasitism to that will maximize the spread of Wolbachia in the host populations. These manipulations include cytoplasmic incompatibility (mating of infected males and uninfected females does not result in viable offspring), parthenogenesis, feminization and male killing (reviewed by Werren et al., 2008).

Due to this, Wolbachia is seen as a potential biocontrol agent, and has for example been shown to reduce the life span of Aedes aegypti, the major mosquito vector of the dengue virus

(McMeniman et al., 2009) or inhibit virus transmission (Bian et al., 2010) under laboratory conditions. However, there are also a number of examples of beneficial effects of Wolbachia on arthropod hosts, for example, it has been shown that Wolbachia confers virus protection in

Drosophila (Hedges et al., 2008) similarly to the protection pea aphids gain from parasitoids and heat shock when harbouring Regiella insecticola symbionts (Moran et al., 2005). Likewise

Wolbachia are common in Drosophila laboratory stocks, where they are not in all cases considered detrimental (Clark et al., 2005) and in certain parasitoid wasps they have been found necessary for oogenesis (Dedeine et al., 2005; Dedeine et al., 2001).

61 Wolbachia are also common in ants, found in approximately 50% of the species screened

(Wenseleers et al., 1998), but despite this frequency, it is unclear which effect, if any, they have on ants. So far studies on Formica ants have not found induction of parthenogenesis

(Wenseleers, Billen, 2000) or influence on sex ratios (Keller et al., 2001), but there may be a deleterious effect on colony function through a reduction in the relative number of produced sexuals (Wenseleers et al., 2002).

Enemy release, or the escape from coevolved natural enemies during the introduction process, has often been implicated in the explanation of the success of invasive species (Colautti et al.,

2004) and many examples exist of reduced number of pathogens and parasites in introduced animals (Torchin et al., 2003) as well as plants (Mitchell, Power, 2003). Introduced and native populations of the Argentine ant, Linepithema humile (Reuter et al., 2005; Tsutsui et al., 2003) and the red imported fire ant, Solenopsis invicta (Bouwma et al., 2006; Shoemaker et al., 2003;

Shoemaker et al., 2000) have been screened for Wolbachia. In both invasive ant species native populations contained Wolbachia, whereas the introduced populations did not, or only very rarely, suggesting that a loss of Wolbachia has conferred an advantage to the introduced populations, increasing their chances of success in the introduced ranges. Likewise introduced populations of the invasive garden ant Lasius neglectus had significantly lower occurrence of infection than its non‐invasive sister species La. turcicus (Cremer et al., 2008).

To test the generality of the hypothesis that invasive ant species have lost infections of

Wolbachia, which may have contributed to their invasion success, we studied the occurrence of

Wolbachia in the very widespread introduced pharaoh ants (Monomorium pharaonis) as well as in 11 other native or introduced Monomorium species. A further objective was to detect any

62 geographical patterns in the occurrence of Wolbachia that may indicate (1) the so‐far unknown native range of pharaoh ants or (2) the transmission mode of the symbiont within and among species of Monomorium.

Materials and metods

Study material

Samples of M. pharaonis and 11 other native or invasive Monomorium species were collected during fieldwork in 2004 (Ghana) and 2008 (Thailand) or obtained through colleagues between

2004 and 2009 (Table 1). The species studied were (with abbreviation in brackets): M. pharaonis

(Mp), M. algiricum (Malg), M. destructor (Md), M. emersoni (Mem), M. floricola (Mflo), M. hiten

(Mhit), M. intrudens (Mint), M. orientale (Mori), M. sp (Msp), M. sechellense (Msec), M. triviale

(Mtri), and M. viride (Mvir). Sampling of pharaoh ants was opportunistic to represent different localities around the world as well as to represent a smaller geographical scale in Thailand where the species is very frequent (Schmidt et al., in prep). Pharaoh ants have been hypothesized to be native to Southeast Asia (reviewed in Wetterer, 2010), where the species is found indoors as well as outdoors predominantly closely associated with human habitation. The laboratory colonies obtained had been reared in captivity for several years prior to sampling.

When possible, worker samples were stored in 96% EtOH until analysed, although some samples

(M. viride) were originally caught in pitfall traps and had been stored in propylene glycol for varying periods of time. We chose to screen two or more individuals per colony because pharaoh ant colonies are polygynous and it therefore cannot be assumed that all individuals in a colony have the same parents. The fidelity of transmission is unknown, although likely to be very

63 high, as it has been shown to be near 100% in the closely related red imported fire ant

(Solenopsis invicta) (Shoemaker et al., 2003).

Genetic analyses

DNA was extracted from 2‐10 workers from each colony using one of two methods (Table 1), either a 200 µl 5% Chelex solution in which the workers were crushed and subsequently boiled for 15 min or extractions were made using Invisorb® Spin Tissue Mini Kits according to the manufacturer's instructions, lysing samples for 24‐48 hours. Tests using Chelex extracted and kit extracted samples showed that older Chelex extracted samples were more prone to fail to amplify than kit extracted samples and longer lysis increased the number of positive samples. To reduce the risk of false negatives we therefore applied kit extractions where possible. The samples were screened for the presence of Wolbachia using primers wsp81F and wsp691R, specific for the cell surface protein‐encoding gene wsp, as described by Zhou et al. (1998). PCRs were run in 20 µl reaction volumes with 2 µl extract as template, reaction buffer, 2 µl 25 mM MgCl2, 2 µl 2 mM GATC , 1 µl forward and reverse primers and 0.2 µl Taq gold polymerase. The PCR reaction conditions consisted of an initial denaturing step of 94 °C for 1 min, followed by 35 cycles (or 45 cycles for M. viride & M. algericum samples) of 94 °C for 1 min, 55 °C for 1 min, and 72 °C for 1 min, and finally an extension step at 72 °C for

5‐10 min. A positive control from a Wolbachia infected ant as well as a negative control was included in each run. PCR products were run on 2% ethidium bromide agarose gels, and negative

PCRs of all Chelex extracted samples as well as the M. viride samples were repeated till a total of three negative screenings to minimize the likelihood of false negatives.

64 Of the samples amplifying wsp a subset were chosen, new PCRs were run on one individual per colony and the products cloned directly using a TOPO® TA Cloning® Kit (Invitrogen) following the manufacturer’s protocol. For 10 of the M. pharaonis colonies 16 clones per sample were chosen for sequencing in order to ascertain whether multiple Wolbachia infections existed in these, whereas two clones per individual were sequenced for an additional seven M. pharaonis colonies and likewise for six M. floricola, one M. destructor and five M. emersonii colonies. PCRs were run using the clones as template and universal M13 primers, samples purified using MSB®

HTS PCRapace plate purification, and shipped to Macrogen Korea for sequencing or cycle‐ sequenced using BigDye® Terminator Cycle Sequencing Kits (Applied Biosystems) following the manufacturer’s protocol, and run on an Applied Biosystems 3130xl Genetic Analyzer. Sequences were viewed and aligned using Sequencher 4.7 and BLAST applied to compare with GenBank sequences.

Results

Of the 103 pharaoh ant colonies screened, 35 (ca. 34 %) amplified wsp indicating infection with

Wolbachia (Fig. 1 and Table 1). There was no geographical pattern to be seen in the distribution of infected and non‐infected colonies as both were represented in all sampled continents. The native M. algericum and M. emersonii colonies screened were all infected, and so were three out of the four M. viride colonies, despite the fact that these were old samples that had been stored in propylene glycol. The M. viride colony for which no samples were positive, was one for which only 3 individuals were screened, and since these ants had been stored in propylene glycol it is not inconceivable that these were false negatives. Six of 23 M. destructor colonies

65 were infected (ca. 26 %) and 18 of 26 M. floricola colonies (ca. 70 %). None of the other species amplified wsp (Table 1 and Fig. 2).

Infections by multiple strains of Wolbachia appeared rare, and the wsp sequences were near‐ identical to those of several non‐hymenopteran host organisms or of American ant species.

None of the M. pharaonis individuals sequenced carried multiple strains. For any given individual the clones sequenced came out as identical (contigs had 98% minimum match percentage in when Blasted), and the samples contained one of two strains: that of Ephestia cautella (moth) or that of Malaya genurostis (mosquito), Table 1. Likewise, no multiple strain infections were found in the six M. floricola individuals screened; the strains found were similar to those of Ephestia cautella, Rhagolestis cerasi (cherry fruit fly strain wCer5), or Tripteroides bambusa (mosquito). Only one M. destructor sample was successfully cloned, and the two sequenced clones were both similar to Ixodes ricinus (sheep tick). The native North American M. emersonii were infected with either a strain similar to the Ixodes ricinus strain or one similar to those found in several American ant species, the Solenopsis invicta B strain. One of the screened

M. emersonii individuals (Mem_1, Table1) had two strains among its clones.

Discussion

Our results show that Wolbachia is common in introduced populations of pharaoh ants, occurring in colonies representing all of the geographical range sampled, and 34 % of the populations screened positive. This proportion is likely an underestimate of the prevalence, as we only had old low quality Chelex extractions or few individuals screened for a number of the

66 colonies, and polygynous species may be expected to have infected as well as uninfected queens in the same colony (Bouwma et al., 2006). This prevalence in introduced populations of the pharaoh ant is comparable to what is observed in the native range of the Argentine ant with 34‐

45% of the individuals and 37‐41% of the colonies screened positive, contrasting with infections only found in one of 21 introduced population (Reuter et al., 2005; Tsutsui et al., 2003).

Similarly, studies on the red imported fire ant and closely related species from the Solenopsis saevissima complex have shown either no or very little occurrence of Wolbachia in the introduced populations (Shoemaker et al., 2000). They further found that 6 of 9 native species were infected with substantial variation in population infection status among different geographical regions as well as social forms (Shoemaker et al., 2003), monogyne populations having higher infection frequencies. In a specific study of the red imported fire ant Bouwma et al. (2006) found Wolbachia in 2 of 10 introduced populations. In contrast, in the invasive garden ant 8 of 10 introduced populations were infected by Wolbachia although the prevalence at the individual level was only 24% and significantly lower than in native populations of its sister species (Cremer et al., 2008).

In the four Monomorium species known to be native, the prevalence of Wolbachia at the colony level varied between 0 and 100%. Interestingly, it appears that the native North American and

European species (M. emersoni, M. viride and M. algiricum) may all have 100% infection –the lower preservation quality of M. viride samples is very likely to have caused a single false negative – whereas the occurrence in the Asian native ants is not high. The two other known introduced species (M. destructor and M. floricola) had intermediate levels of occurrence (Fig.

2). This suggests that there is no close link between the occurrence of Wolbachia and whether populations are native or introduced in this ant genus.

Overall these findings indicate that Wolbachia may not be a relevant indicator of enemy release in pharaoh ants, nor useful as an indicator of the species’ native range. However, we cannot exclude the possibility that native populations of M. pharaonis may still deviate from introduced populations by having an even higher prevalence of Wolbachia. Stahlhut et al.’s (2006) study on an introduced wasp species showed frequent occurrence of Wolbachia in native as well as introduced populations, suggesting that this may be due to the recent invasion history of the species. Considering the very long invasion history of M. pharaonis (Wetterer, 2010) our findings indicate that Wolbacia may not be very harmful to pharaoh ant colonies, if it has any negative effect at all.

As it has been shown that the wsp sequence can be very variable (Baldo et al., 2006) a Multi Locus Sequence Typing approach is considered necessary to properly identify Wolbachia strains (Baldo, Werren, 2007), and no exact strain identities can thus be derived from our data.

However, it is clear that Wolbachia is variable in pharaoh ants, and it appears that some strains

(or wsp sequences) are identical to those found in another introduced Monomorium species (M. floricola) and that the sequences of another Monomorium species, M. destructor sampled in

Thailand, are very similar to those of M. emersonii, which is native to North America. Five of the six different wsp based strains in this study are very similar to those from other host taxa, indicating extensive horizontal transfer between species have taken place.

Considering our findings, it is tempting to speculate that, depending on the mode of between‐ species transfer, the strains infecting any given species may reflect their evolutionary history as well as that of their more recent geographical affinities and interactions. This could potentially

68 result in native species haboring an area‐specific set of strains, such as those found in American

M. emersonii as well as the American fire ants. Russell et al. (2009) found that strains could be specific to certain species or geographic areas in lycinid butterflies and ants, and that the New

World and Old World ants carried very different strains. How this would be reflected in introduced species is unknown, but various vectors for horizontal transfer, such as parasitoid wasps, have been suggested (Heath et al., 1999), and in a secondary acquisition scenario, strains in an introduced species could thus potentially be derived from both its native source area and from the area in which the species currently lives .

Noting the frequent occurrence and the variation of Wolbachia strains found in and among the

Monomorium species screened in this study, we think that much may be gained from a more detailed study of this or similar systems. Characterizing the bacteria present in an introduced species and employing Multi Locus Sequence Typing (MLST)(Baldo et al., 2006) or similar approaches to identify the bacterial community in a group of geographically widely dispersed congeneric invasive and non‐invasive species such as the Monomorium ants has potential to increase understanding of the phylogenetic and geographical relationships and modes of transfer of introduced species and their co‐occuring bacterial symbionts. In addition, pharaoh ants can be bred in the laboratory, and it would therefore be possible to make controlled crosses testing for any reproductive manipulations Wolbachia may cause (Werren et al., 2008), which could be an interesting avenue of future research.

69 Acknowledgements

We thank the National Research Council of Thailand (NRCT) and the Department of National

Parks, Wildlife and Plant Conservation (DNP) for granting us permits to collect ants in Thailand;

Watana Sackhoowong for collaboration in the field; David Hughes, Winanda Himaman,

Chaweewan Hutacharern and David Lohman for advice on field work in Thailand; Weeyawat

Jaitrong and Thailand Natural history Museum for collaboration in the field; Pitoon Kongnoo for assistance and translation in the field; and Tana Thongrod for field assistance. Furthermore, we thank a number of people for kindly sending us ants: Ehab Abouheif, Diane Allard, Emmanuel M.

Attua, Boris Baer, Anne‐Geneviève Bagnères, Howard Bell, Tone Birkemoe, Lisbeth W. Børgesen,

Cobblah–group Legon, Stephanie Dreier, Simon Dupont, Marie‐Julie Favé, Andre Francoeur, Tina

Frisch, Aleksandra Gliniewicz, Mikko Heini, Anders Hjort‐Hansen, Armin Ionescu, Weeyawat

Jaitrong, Todd Johnson, Yeo Kelo, Chow‐Yang Lee, Say‐Piau Lim, David Lohman, Boudanath

Maharajh, Ants Martin, Alexander S. Mikheyev, Jesse Czekanski‐Moir, David Oi, Fabien Ravery,

Vola Razakarivony, Watana Sakchoowong, Tsong Hong Su, Liselotte Sundström, Brian Taylor,

Walter R. Tschinkel, Kazuki Tsuji, Lumi Viljakainen, Merav Vonshak, Jaitrong Weeyawat and Seiki

Yamane. The study was funded through a grant from The Danish National Research Foundation to the Centre for Social Evolution.

70 References

Baldo L, Hotopp JCD, Jolley KA, et al. (2006) Multilocus sequence typing system for the endosymbiont Wolbachia pipientis. Applied and Environmental Microbiology 72, 7098‐ 7110. Baldo L, Werren JH (2007) Revisiting Wolbachia supergroup typing based on WSP: Spurious lineages and discordance with MLST. Current Microbiology 55, 81‐87. Bian G, Xu Y, Lu P, Xie Y, Xi Z (2010) The endosymbiotic bacterium Wolbachia induces resistance to dengue virus in Aedes aegypti. PLoS Pathogens 6, e1000833. Bouwma AM, Ahrens ME, DeHeer CJ, DeWayne Shoemaker D (2006) Distribution and prevalence of Wolbachia in introduced populations of the fire ant Solenopsis invicta. Insect Molecular Biology 15, 89‐93. Clark ME, Anderson CL, Cande J, Karr TL (2005) Widespread prevalence of Wolbachia in laboratory stocks and the implications for Drosophila research. Genetics 170, 1667‐1675. Colautti RI, Ricciardi A, Grigorovich IA, MacIsaac HJ (2004) Is invasion success explained by the enemy release hypothesis? Ecology Letters 7, 721‐733. Cremer S, Ugelvig LV, Drijfhout FP, et al. (2008) The evolution of invasiveness in garden ants. PLoS ONE 3, e3838. Dedeine F, Bouletreau M, Vavre F (2005) Wolbachia requirement for oogenesis: occurrence within the genus Asobara (Hymenoptera, Braconidae) and evidence for intraspecific variation in A. tabida. Heredity 95, 394‐400. Dedeine F, Vavre F, Fleury F, et al. (2001) Removing symbiotic Wolbachia bacteria specifically inhibits oogenesis in a parasitic wasp. Proceedings of the National Academy of Sciences of the United States of America 98, 6247‐6252. Heath BD, Butcher RDJ, Whitfield WGF, Hubbard SF (1999) Horizontal transfer of Wolbachia between phylogenetically distant insect species by a naturally occurring mechanism. Current Biology 9, 313‐316. Hedges LM, Brownlie JC, O'Neill SL, Johnson KN (2008) Wolbachia and virus protection in insects. Science 322, 702‐. Hilgenboecker K, Hammerstein P, Schlattmann P, Telschow A, Werren JH (2008) How many species are infected with Wolbachia? ‐ a statistical analysis of current data. FEMS Microbiology Letters 281, 215‐220.

71 Jeyaprakash A, Hoy MA (2000) Long PCR improves Wolbachia DNA amplification: wsp sequences found in 76% of sixty‐three arthropod species. Insect Molecular Biology 9, 393‐405. Keller L, Liautard C, Reuter M, et al. (2001) Sex ratio and Wolbachia infection in the ant Formica exsecta. Heredity 87, 227‐233. McMeniman CJ, Lane RV, Cass BN, et al. (2009) Stable introduction of a life‐shortening Wolbachia infection into the mosquito Aedes aegypti. Science 323, 141‐144. Mitchell CE, Power AG (2003) Release of invasive plants from fungal and viral pathogens. Nature 421, 625‐627. Moran NA, Russell JA, Koga R, Fukatsu T (2005) Evolutionary relationships of three new species of Enterobacteriaceae living as symbionts of aphids and other insects. Applied and Environmental Microbiology 71, 3302‐3310. Reuter M, Pedersen JS, Keller L (2005) Loss of Wolbachia infection during colonisation in the invasive Argentine ant Linepithema humile. Heredity 94, 364‐369. Russell JA, Goldman‐Huertas B, Moreau CS, et al. (2009) Specialization and geographic isolation among Wolbachia symbionts from ants and lycaenid butterflies. Evolution 63, 624‐640. Schmidt AM, Sackhoowong W, Pedersen J (in prep) Local scale population genetics of pharaoh ants (Monomorium pharaonis) in Thailand and the Thai‐Malay Peninsula. Shoemaker DD, Ahrens M, Sheill L, et al. (2003) Distribution and prevalence of Wolbachia infections in native populations of the fire ant Solenopsis invicta (Hymenoptera : Formicidae). Environmental Entomology 32, 1329‐1336. Shoemaker DD, Ross KG, Keller L, Vargo EL, Werren JH (2000) Wolbachia infections in native and introduced populations of fire ants (Solenopsis spp.). Insect Molecular Biology 9, 661‐ 673. Smith HL, Rajan TV (2000) Tetracycline inhibits development of the infective‐stage larvae of filarial nematodes in vitro. Experimental Parasitology 95, 265‐270. Stahlhut JK, Liebert AE, Starks PT, Dapporto L, Jaenike J (2006) Wolbachia in the invasive European paper wasp Polistes dominulus. Insectes Sociaux 53, 269‐273. Torchin ME, Lafferty KD, Dobson AP, McKenzie VJ, Kuris AM (2003) Introduced species and their missing parasites. Nature 421, 628‐630. Tsutsui ND, Kauppinen SN, Oyafuso AF, Grosberg RK (2003) The distribution and evolutionary history of Wolbachia infection in native and introduced populations of the invasive argentine ant (Linepithema humile). Molecular Ecology 12, 3057‐3068.

72 Wenseleers T, Billen J (2000) No evidence for Wolbachia‐induced parthenogenesis in the social Hymenoptera. Journal of Evolutionary Biology 13, 277‐280. Wenseleers T, Ito F, Van Borm S, et al. (1998) Widespread occurrence of the micro‐organism Wolbachia in ants. Proceedings of the Royal Society of London Series B‐Biological Sciences 265, 1447‐1452. Wenseleers T, Sundstrom L, Billen J (2002) Deleterious Wolbachia in the ant Formica truncorum. Proceedings of the Royal Society of London Series B‐Biological Sciences 269, 623‐629. Werren JH (1997) Biology of Wolbachia. Annual Review of Entomology 42, 587‐609. Werren JH, Baldo L, Clark ME (2008) Wolbachia: master manipulators of invertebrate biology. Nat Rev Micro 6, 741‐751. Werren JH, Windsor D, Guo LR (1995) Distribution of Wolbachia among Neotropical Arthropods. Proceedings of the Royal Society of London Series B‐Biological Sciences 262, 197‐204. Wetterer JK (2010) Worldwide spread of the pharaoh ant, Monomorium pharaonis (Hymenoptera: Formicidae). Myrmecological News 13, 115‐129. Zhou WG, Rousset F, O'Neill S (1998) Phylogeny and PCR‐based classification of Wolbachia strains using wsp gene sequences. Proceedings of the Royal Society of London Series B‐ Biological Sciences 265, 509‐515.

73 Table 1 Samples of Monomorium pharaonis and other Monomorium sp. The colony ID is the abbreviations used in further analyses and figures. Extraction gives the type of extraction used for a given colony. Number is the sample size of individuals for each extraction method. Wsp “yes” means that at least one sample tested positive for Wolbachia. Cloned “yes” identifies the colonies from which the Wsp amplified fragments were cloned and Seq ID gives the initials of the host species in which a similar wsp sequence have been characterized and placed in GeneBank. Ec: Ephestia cautella; Mg: Malaya genurostis; Rc: Rhagolestis cerasi strain wCer5; Tb: Tripteroides bambusa; Ir: Ixodes ricinus and Am denotes a strain similar to that found in several American ant species. For samples collected during field work by the authors a more specific locality description is included. “Laboratory” samples had been kept in other laboratories as well as in Copenhagen for several years prior to this study. “House” is a private home and “Field” denotes various field localities, non‐urban or the natural habitats of native Monomorium species.

Species Colony Locality Extraction Number Wsp Cloned Seq ID Locality type M. pharaonis Th4a Trat, Thailand Chelex/kit 4/2 yes Field station M. pharaonis Th13 Chanthaburi, kit 2 Field station M. pharaonis Th21 Chanthaburi, Chelex/kit 4/2 Field station M. pharaonis Th22 Chanthaburi, kit 2 Large shop M. pharaonis Th25 Thailand kit 10 House M. pharaonis Th74 Thailand kit 2 Tree M. pharaonis Th76 Thailand kit 10 yes House M. pharaonis Th87 Thailand kit 2 Restaurant M. pharaonis Th93 Thailand kit 2 Tree M. pharaonis Th97 Thailand kit 2 yes House M. pharaonis Th99 Thailand kit 2 Shop/restaurant M. pharaonis Th136 Thailand kit 10 yes House M. pharaonis Th138 Thailand kit 10 yes House M. pharaonis Th139 Thailand kit 10 yes House M. pharaonis Th140 Thailand kit 10 yes House M. pharaonis Th141 Thailand kit 2 yes House M. pharaonis Th142 Thailand kit 10 yes House M. pharaonis Th143 Thailand kit 10 yes House M. pharaonis Th151 Thailand kit 2 Shop M. pharaonis Th163 Thailand kit 2 Building M. pharaonis Th167 Thailand kit 2 House M. pharaonis Th166a Thailand kit 2 Shed M. pharaonis Th168 Thailand Chelex/kit 4/2 House M. pharaonis Th173 Thailand kit 2 Shop M. pharaonis Th185 Thailand kit 2 House M. pharaonis Th187 Thailand kit 2 House

74 M. pharaonis Th188 Thailand kit 2 House/shop M. pharaonis Th193 Thailand kit 2 House M. pharaonis Th198 Thailand kit 2 Restaurant M. pharaonis Th204 Thailand Chelex/kit 4/2 yes Shop M. pharaonis Th210 Thailand kit 2 House M. pharaonis Th220 Thailand Chelex/kit 4/2 yes Shop M. pharaonis Th223 Thailand kit 2 Shop M. pharaonis ThGrH Thailand kit 2 Field station M. pharaonis Gh1 Legon, Ghana Chelex 10 House M. pharaonis Gh4 Fete, Ghana Chelex/kit 10/2 yes yes Mg House M. pharaonis Gh6 Asuansi, Ghana Chelex 10 House M. pharaonis Gh7 Biriwa, Ghana Chelex 10 yes yes Ec House M. pharaonis Gh8 Koforidua, Ghana Chelex 10 House M. pharaonis Gh9 CRIG, Ghana Chelex 10 House M. pharaonis Gh10 Legon, Ghana Chelex 10 yes yes Mg House M. pharaonis Gh11 Iturie, Ghana Chelex 10 yes yes Mg House M. pharaonis Gh12 CRIG, Ghana Chelex 10 yes yes Ec House M. pharaonis Gh13 Pokwasi, Ghana Chelex 10 yes yes Mg House M. pharaonis Gh14 Aburi, Ghana Chelex 10 House M. pharaonis Gh15 CRIG, Ghana Chelex 10 House M. pharaonis Gh16 Tafo, Ghana Chelex 10 House M. pharaonis Gh17 Tafo, Ghana Chelex 10 yes yes Ec House M. pharaonis T Tingbjerg, Denmark Chelex 10 House M. pharaonis I4 Copenhagen, Chelex 10 House M. pharaonis D Bonn, Germany Chelex 10 House M. pharaonis Z Zürich, Switzerland Chelex 10 yes yes Mg House M. pharaonis TW Taiwan Chelex 10 Laboratory M. pharaonis C Cameroon Chelex 10 yes yes Ec House M. pharaonis U1 Florida, USA Chelex 10 Laboratory M. pharaonis U2 Texas, USA Chelex 10 Laboratory M. pharaonis U3 Florida, USA Chelex 10 Laboratory M. pharaonis U4 Penang, Malaysia Chelex 10 Laboratory M. pharaonis U5 London, UK Chelex/kit 10/2 yes yes Ec Laboratory M. pharaonis U6 Debrecen, Hungary Chelex/kit 10/2 Laboratory M. pharaonis U7 Warsaw, Poland Chelex/kit 10/2 Laboratory M. pharaonis U8 Okinawa, Japan Chelex/kit 10/2 Laboratory M. pharaonis U9 Kyoto, Japan Chelex/kit 10/2 Laboratory M. pharaonis U10 Moscow, Russia Chelex/kit 10/2 House

75 M. pharaonis U11 Texas, USA Chelex/kit 10/2 House M. pharaonis U17 Lambto, Ivory Coast Chelex 10 House M. pharaonis U18 BCI, Panama Chelex 10 House M. pharaonis U19 Gamboa, Panama Chelex 6 House M. pharaonis U20 Montreal, Canada Chelex 10 House M. pharaonis MpU22 Copenhagen, Chelex 10 House M. pharaonis MpU23 Okinawa, Japan kit 4 Field M. pharaonis MpU24 Okinawa, Japan kit 4 Field M. pharaonis MpU25 Okinawa, Japan kit 4 Field M. pharaonis MpU26 Debrecen, Hungary Chelex 2 House M. pharaonis MpU27 Debrecen, Hungary Chelex 2 yes yes Ec House M. pharaonis MpU28 Debrecen, Hungary Chelex 2 House M. pharaonis MpU29 Debrecen, Hungary Chelex 2 yes yes Ec House M. pharaonis MpU31 Palau kit 4 House M. pharaonis MpU32 Malaysia kit 4 House M. pharaonis MpU33 Malaysia kit 4 House M. pharaonis MpU34 Malaysia kit 4 House M. pharaonis MpU35 Madagascar kit 4 House M. pharaonis MpU36 Lambto, Ivory Coast kit 4 House M. pharaonis MpU37 Florida, USA kit 4 House M. pharaonis MpU38 Tallinn, Estonia kit 4 yes yes Ec House M. pharaonis MpU39 Tallinn, Estonia kit 4 yes yes Ec House M. pharaonis MpU40 Tartu, Estonia kit 4 yes yes Ec House M. pharaonis MpU41 Tartu, Estonia kit 4 House M. pharaonis MpU42 Oslo, Norway kit 4 House M. pharaonis MpU43 Sandness, Norway kit 4 House M. pharaonis MpU44 Israel kit 4 House M. pharaonis MpU45 Chiang Mai, kit 4 yes Hotel M. pharaonis MpU46 Quebec, Canada kit 2 yes House M. pharaonis MpU47 Debrecen, Hungary kit 2 House M. pharaonis MpU48 Copenhagen, kit 2 House M. pharaonis MpU49 Helsinki, Finland kit 2 yes House M. pharaonis MpU50 Helsinki, Finland kit 2 yes House M. pharaonis MpU51 Helsinki, Finland kit 2 yes House M. pharaonis MpU52 Helsinki, Finland kit 2 yes House M. pharaonis MpU53 Vantaa, Finland kit 2 yes House M. pharaonis MpU55 Thailand kit 10 yes Forest M. pharaonis MpU65 Thailand kit 2 Forest

76 M. algiricum Malg_A Spain kit 10 yes Field M. algiricum Malg_C Spain kit 10 yes Field M. algiricum Malg_F Spain kit 10 yes Field M. destructor MdTh2 Thailand kit 2 Gas station M. destructor MdTh26 Thailand kit 6 yes Gas station M. destructor MdTh30 Thailand kit 2 Gas station M. destructor MdTh31 Thailand kit 6 Gas station M. destructor MdTh33 Thailand kit 2 Gas station M. destructor MdTh45 Thailand kit 2 yes Gas station M. destructor MdTh51 Thailand kit 2 Street/building M. destructor MdTh58 Thailand kit 2 Gas station M. destructor MdTh69 Thailand kit 2 Gas station M. destructor MdTh75 Thailand kit 2 Gas station M. destructor Md80 Thailand kit 2 Gas station M. destructor MdTh84 Thailand kit 2 Gas station M. destructor MdTh129 Thailand kit 2 Shop/restaurant M. destructor MdTh134b Thailand kit 2 Street M. destructor MdTh145b Thailand kit 2 House M. destructor MdTh180 Thailand kit 6 yes House M. destructor MdTh196 Thailand kit 2 Street/house M. destructor MdTh202 Thailand kit 2 House M. destructor MdTh213 Thailand kit 6 yes Gas station M. destructor MdTh221 Thailand kit 6 yes Shop/restaurant M. destructor MdTh225 Thailand kit 2 Restaurant M. destructor MdTh236 Thailand kit 2 yes yes Ir Restaurant M. destructor Md1 Malaysia kit 2 Laboratory M. emersoni Mem_1 Arizona, USA kit 4 yes yes Am Field M. emersoni Mem_2 Arizona, USA kit 4 yes yes Am Field M. emersoni Mem_3 Arizona, USA kit 4 yes Field M. emersoni Mem_4 Arizona, USA kit 4 yes Field M. emersoni Mem_5 Arizona, USA kit 4 yes yes Am Field M. emersoni Mem_6 Arizona, USA kit 4 yes yes Am Field M. emersoni Mem_7 Arizona, USA kit 4 yes yes Am Field M. emersoni Mem_8 Arizona, USA kit 4 yes Field M. floricola MfloTh2 Thailand kit 2 yes Gas station M. floricola MfloTh14 Thailand kit 2 House M. floricola MfloTh34 Thailand kit 2 Gas Station M. floricola MfloTh86 Thailand Chelex/kit 4/2 yes yes Tb Restaurant

77 M. floricola MfloTh88 Thailand Chelex/kit 4/2 yes yes Tb Restaurant M. floricola MfloTh102 Thailand kit 2 yes Restaurant M. floricola MfloTh103 Thailand kit 2 Building M. floricola MfloTh105 Thailand Chelex/kit 4/2 yes yes Rc Shop M. floricola MfloTh129 Thailand kit 2 yes Shop/restaurant M. floricola MfloTh134b Thailand kit 2 yes Restaurant M. floricola MfloTh145a Thailand kit 2 yes Shop/restaurant M. floricola MfloTh160 Thailand kit 2 Building M. floricola MfloTh167 Thailand kit 2 yes House M. floricola MfloTh173 Thailand kit 2 yes Building M. floricola MfloTh174 Thailand Chelex/kit 4/2 yes yes Ec House M. floricola MfloTh192 Thailand kit 2 yes yes House M. floricola MfloTh195 Thailand kit 2 House M. floricola MfloTh196 Thailand kit 2 yes Shop M. floricola MfloTh197 Thailand kit 2 Gas station M. floricola MfloTh202 Thailand kit 2 yes House M. floricola MfloTh229 Thailand kit 2 yes Restaurant M. floricola MfloTh236 Thailand Chelex/kit 4/2 yes Ir Pavillion M. floricola Mflo1 Malaysia kit 4 yes yes Rc Laboratory M. floricola Mflo2A Tokyo, Japan Chelex/kit 2/3 yes yes Tb Field M. floricola Mflo2B Tokyo, Japan Chelex/kit 2/3 Field M. floricola Mflo2C Tokyo, Japan Chelex/kit 2/3 Field M. hiten Mhit1A Tokyo, Japan Chelex/kit 3/3 Field M. hiten Mhit1B Tokyo, Japan Chelex/kit 3/3 Field M. intrudens Mint0 Tokyo, Japan Chelex/kit 3/3 Field M. intrudens Mint1 Tokyo, Japan Chelex/kit 3/2 Field M. intrudens Mint2 Tokyo, Japan Chelex/kit 3/2 Field M. intrudens Mint3 Tokyo, Japan Chelex/kit 3/2 Field M. intrudens Mint4 Tokyo, Japan Chelex/kit 3/2 Field M. intrudens Mint5 Tokyo, Japan Chelex/kit 3/2 Field M. orientale Mori1 Malaysia kit 4 Laboratory M. sp MspF1 Ile d’Orleon, France kit 4 Field M. sp MspF2 Ile d’Orleon, France kit 4 Field M. sechellense Msec1 Tokyo, Japan Chelex/kit 6/3 Field M. triviale Mtri1‐4* Tokyo, Japan Chelex/kit 8/8 Field M. triviale Mtri5‐9* Tokyo, Japan Chelex/kit 8/8 Field M. triviale Mtri10‐12* Tokyo, Japan Chelex/kit 6/6 Field M. triviale Mtri13 Tokyo, Japan Chelex/kit 2/2 Field

78 M. triviale Mtri14‐16* Tokyo, Japan Chelex/kit 6/6 Field M. triviale Mtri17‐19* Tokyo, Japan Chelex/kit 6/6 Field M. triviale Mtri20‐21* Tokyo, Japan Chelex/kit 6/4 Field M. triviale Mtri22‐23* Tokyo, Japan Chelex/kit 6/4 Field M. viride Mvir_1 Florida, USA kit 10 yes Field M. viride Mvir_5 Florida, USA kit 10 yes Field M. viride Mvir_6 Florida, USA kit 10 yes Field M. viride Mvir_7 Florida, USA kit 3 Field

* denotes a compilation of individuals used from multiple nests in the same population; 2‐3 individuals per nest were extracted using each method and screened subsequently.

79 Fig. 1 Geographical positions of M. pharaonis sampling localities. Localities with samples that did amplify Wolbachia are shown in red. Ca. 34% of the screened pharaoh ant populations (n = 103) were infected as indicated on the inserted pie chart.

Fig. 2 Proportion of Monomorium colonies screened positive (red) and non‐positive (white) for Wolbachia applying wsp primers. The number of colonies screened for each species is given in brackets, abbreviations for species names are as given in Matherials and Methods and the status of the different species are denoted: introduced (I), native (N) and unknown (U). See Table 1 for further information.

1 wsp

0.8 negative

0.6 and

0.4 positive

0.2

0 Proportion Mp Malg Md Mem Mflo Mhit Mint Mori Msp Msec Mtri Mvir (n=103) (n=3) (n=23) (n=8) (n=26) (n=2) (n=6) (n=1) (n=2) (n=1) (n=23) (n=4) I N I N I U UUU U N N

Chapter 3

Deconstructing the effect of genetic diversity on disease resistance in a polygynous social insect

Anna M. Schmidt, Tim A. Linksvayer, Jacobus J. Boomsma and Jes S. Pedersen

Submitted to Behavioral Ecology

Deconstructing the effect of genetic diversity on disease resistance in a polygynous social insect

Anna M. Schmidt, Timothy A. Linksvayer, Jacobus J. Boomsma and Jes S. Pedersen

Centre for Social Evolution, Department of Biology, University of Copenhagen, Universitetsparken 15, DK‐2100 Copenhagen, Denmark

Multiple queen‐mating has been shown to enhance disease resistance in insect societies because higher genetic diversity among nestmates improves collective immune defenses. However, it has remained ambiguous whether polygynous societies with large numbers of queens also benefit from increased genetic diversity. We used one of the very few ant species that can be reared across generations, the pharaoh ant Monomorium pharaonis, to create experimental colonies with two types of enhanced genetic diversity: 1. Mixed workers from three divergent inbred lineages representing the “polygyny‐equivalent” of multiple queen‐mating; 2. Uniform workers whose overall heterozygosity was increased by two subsequent generations of crossing between the same divergent inbred lineages. After standardized exposure to spores of the fungal pathogen Beauveria bassiana we found significant differences in worker survival among inbred lineages, indicating substantial genetic variation for resistance. Enhanced heterozygosity colonies had worker survival rates similar to the most resistant inbred lineage, but colonies with mixed workers from the three inbred lineages had lower worker survival. Workers did not show any infection‐ avoidance behavior and average larval survival benefited from increased heterozygosity but was reduced in mixed colonies independent of infection. Although we cannot exclude alternative explanations, our results suggest that polygyny and polyandry may not be directly comparable mechanisms for creating adaptive resistance towards pathogens. Rather, we hypothesize that life history differences between polyandrous and polygynous species mean that increasing genetic variation and its fitness enhancing effects rely on different mechanisms that also affect the nature of fitness benefits: costly additional matings in polyandrous species and crossing between colonies in highly structured polygynous populations.

Key words: Ants, breeding experiment, entomopathogenic fungi, Beauveria bassiana, Monomorium pharaonis

84 Introduction

Living in groups provides benefits of cooperation, but increases the risk of disease transmission

(Anderson and May 1985). Societies of ants, bees, wasps and termites are committed to advanced sociality and can thus be expected to experience considerable pressure from parasites and pathogens, because related individuals living in stable environments are attractive hosts to colonize (Boomsma et al. 2005; Schmid‐Hempel 1998). The benefits of social life that may offset these negative odds are thought to emerge from a combination of individual and social defenses that may surpass the effectiveness of solitary immune responses by a large margin (e.g. Boomsma et al. 2005; Cremer et al. 2007; Evans and Spivak 2010). Division of labor in insect societies has also allowed the evolution of novel defenses such as the use of antimicrobial compounds obtained during foraging (Bulmer et al. 2009; Chapuisat et al. 2007), secretions from specialized glands (e.g.

W.O.H. Hughes et al. 2008b), and domesticated symbionts (Little et al. 2006). These multiple lines of defense make social insect colonies difficult to invade, because they function as fortresses from the perspective of potential parasites (e.g. D.P. Hughes et al. 2008a; Wilson 1971).

Genetic diversity among nestmates may enhance a colony’s ability to combat infection (Hamilton

1987), as it is likely to increase the efficiency of division of labor and to facilitate colony‐level homeostasis (Crozier and Page 1985; Oldroyd and Fewell 2007). The “pure stands breed diseases” paradigm has been particularly influential in stimulating work to explain the evolution and maintenance of multiple queen‐mating (polyandry) in outbreeding hymenopteran societies headed by a single queen, as this is a costly, but powerful, mechanism for increasing intracolonial genetic variation (reviewed in Boomsma 1996; Crozier and Fjerdingstad 2001; Strassmann 2001).

85 Several studies with polyandrous social insects have shown that increased inter‐individual genetic diversity in colonies is indeed associated with increased disease resistance (e.g. W.O.H. Hughes and Boomsma 2004; Reber et al. 2008; Schmid‐Hempel and Crozier 1999; Tarpy 2003; Tarpy and

Seeley 2006). Examples of potential mechanisms through which genetic diversity may influence disease resistance range from genetically determined hygienic behaviors (Rothenbuhler 1964), via heritable variation for antimicrobial peptides (Decanini et al. 2007), to heritable size variation in worker metapleural glands (W.O.H. Hughes et al. 2010).

It is seldom made explicit that inter‐individual variation is only one way by which genetic diversity in eusocial colonies may be expressed. An alternative mechanism is individual heterozygosity, but only three studies have addressed this. Calleri et al. (2006) showed that groups of inbred termites suffered higher mortality when exposed to a fungal pathogen, and Ugelvig et al. (2010) showed that inbred Cardiocondyla ants are less good in detecting infections before they spread. However,

Gerloff et al. (2003) did not find negative effects of one generation of inbreeding in a bumblebee.

The explicit study of individual heterozygosity as a variable for disease resistance is particularly relevant when genetic diversity is generated by varying numbers of queens breeding in the same colony, as this type of colony structure often involves substantial degrees of inbreeding (Haag‐

Liautard et al. 2009), limited dispersal across occupied patches, and thus considerable genetic population structuring (Sundström et al. 2005). We therefore used the pharaoh ant, Monomorium pharaonis, as experimental model system as this global pest ant has extremely high genetic differentiation (measured as FST = 0.751 ± 0.006 SE) among its global populations (Schmidt et al.

2010), so that crossing and mixing of lineages can create substantial differences in intra‐ and inter‐ individual genetic diversity that can be exposed to controlled infection regimes.

In addition to shedding light on the relevance of two distinct types of genetic variation for colony level disease resistance, our study addresses the extent to which polyandry and polygyny are equivalent mechanisms for enhancing adaptive genetic diversity in insect societies, an issue that has remained controversial (Boomsma and Ratnieks 1996; W.O.H. Hughes et al. 2008c; Keller and

Reeve 1994; Schmid‐Hempel and Crozier 1999).

Materials and Methods

Study species

Pharaoh ants (Monomorium pharaonis) form large polygynous colonies and are commonly found as pests in houses. New queens and males mate within their natal nest (all queens are singly inseminated) and pharaoh ants can thus potentially be kept in the laboratory indefinitely (Berndt and Eichler 1987; Peacock and Baxter 1949). They only spread to new areas by walking or by being inadvertently transported via human‐mediated jump dispersal (Edwards and Baker 1981; Schmidt et al. 2010), and gene flow between colonies may therefore be rare, at least in the semi‐natural habitats associated with human settlement. Microsatellite markers have previously been employed to show that M. pharaonis colonies, both from laboratories and from the field, have very low levels of genetic diversity and are genetically highly differentiated between sites but without any correlation between genetic and geographic distance (Schmidt et al. 2010). For the present study we created experimental groups of workers and brood using three different source colonies (named U1, U3 and U5), each of them an inbred lineage originally collected in the US or

UK, and subsequently kept in laboratories for more than 10 years.

87 Controlled crosses to create a hybrid stock colony with enhanced heterozygosity

We created ants with high intra‐individual diversity through controlled crosses between the three inbred laboratory lineages (U1, U3 and U5). Controlled breeding is not possible with most ants

(see e.g. Cremer and Heinze 2003; Schrempf et al. 2006), so the approach that we report here is novel and has only been used in one other ant species (Ugelvig et al. 2010). The crosses were made over two generations (Fig. 1), first crossing colonies U5 (females) and U3 (males) and subsequently crossing the hybrid females with males from colony U1. This created a colony with high heterozygosity of the individual ants, while hardly affecting the inter‐individual genetic diversity (Fig. 1, see also below). The production of sexuals (virgin queens and males) for the crosses was induced by removing the queens from large colony fragments, waiting ca. 40 days for the workers to rear new sexual brood, and removing this sexual brood at the pupal stage, keeping males and females separate. Groups of 15‐20 females and 10‐15 males were then combined in 10 cm diameter Petri plates and mating often occurred immediately. Approximately one week later, the newly mated queens were supplemented with brood and workers from other colonies to boost growth rates of these incipient colonies, as the queens are practically incapable of founding new colonies by themselves (Peacock et al. 1955; author pers. obs.). The colonies were regularly checked for the first two months and any sexuals reared from the supplemental brood were removed before eclosing. The generation time was approximately 5 months.

Experimental groups

We chose to work with groups of ants as the survival of single M. pharaonis workers is generally very low (author pers. obs.) and small groups are biologically more relevant study units because trophallaxis, grooming, and other characteristic social interactions can occur naturally (W.O.H.

88 Hughes et al. 2002). The experimental groups consisted of 18 workers and 12 second to third instar larvae. The ant colonies were kept in boxes containing multiple tubes in which they keep their brood. We sampled from these nest tubes to obtain workers of intermediate age, which was feasible because younger workers take care of the brood and older workers are mostly foragers

(Berndt and Eichler 1987).

Low genetic diversity groups were created from each of the original lineages U1, U3 and U5

(inbred lineages; Fig. 2). We further used these inbred lineages to create high inter‐individual genetic diversity groups by mixing individuals from these three colonies (mixed groups; Fig. 2).

Since pharaoh ants are highly polygynous and generally display low levels of intraspecific aggression, one can normally combine worker ants from different colonies without inducing antagonistic interactions (Schmidt et al. 2010). We used the high heterozygosity stock colony to create the high heterozygosity experimental groups (crossed groups, Fig. 2). Previous work

(Schmidt et al. 2010) has shown that the three inbred laboratory lineages are genetically highly divergent (FST = 0.601; P < 0. 0001), having low levels of genetic diversity with an average allelic richness across four polymorphic microsatellite loci of only 2.1 ± 0.2 SD (based on standard samples of 30 workers). Our breeding and mixing design therefore meant that the allelic richness became about twice as high, both in the mixed (4.3 ± 0.7) and crossed (4.1 ± 1.1) groups, as in the original inbred U1, U3 and U5 lineages. Furthermore, whereas intra‐individual genetic diversity was similar in the U1, U3, U5 and mixed groups (average heterozygosity across loci,

H = 0.322 ± 0.109 SD), this had become almost twice as high (0.619 ± 0.051) in the crossed groups

(estimation based on analysis in FSTAT 2.9.3.2; Goudet 2002).

89 Each experimental group was made by carefully transferring six workers and four larvae at a time to 10 cm diameter Petri plates (Fig. 2), resulting in a uniform procedure for setting up all groups of

18 workers and 12 larvae. Larvae were transferred first, because these exchange food with the workers and their presence may thus reduce the stress level of newly added workers (Berndt and

Eichler 1987). A total of 36 groups (18 treatment groups which were exposed to the fungal pathogen (see below), and 18 controls which were not exposed) were made for each of the three low diversity colonies (U1, U3 and U5), as well as for each of the mixed and crossed group combinations.

Pathogen exposure

After transfer to Petri plates, the ants were given a water tube plugged with cotton wool as well as a small piece of paper to hide under. The treatment groups were exposed to conidiospores from the entomopathogenic fungus Beauveria bassiana (strain KVL04‐33) which were added to the Petri plates on pieces of filter paper. We chose B. bassiana as the infecting agent because it is a generalist entomopathogen, has a world‐wide distribution, is frequently used for biocontrol, and has very characteristic conidiospore balls, making infected individuals easily identifiable (Hajek and

Leger 1994; Schmid‐Hempel 1998). To the best of our knowledge no specific natural pathogens have previously been identified for pharaoh ants, but as they are scavengers and eat dead insects, it is likely that they would naturally come across B. bassiana when foraging.

To simulate natural exposure to spores, as through a contaminated food source, we placed the ant’s normal food (dead, frozen mealworms, boiled egg yolk, boiled liver and honey) on spore‐ soaked filter paper. Filter paper was prepared by soaking 4 cm2 pieces in high concentration spore

90 suspension by dipping them all directly into a water‐filled agar plate on which the fungus had been cultured. The filter paper was left to dry, and then transferred to the Petri plates with the ants. To further increase the exposure, a small portion of fungal hyphae and spores was further added on top of each piece of filter paper. Although this method of exposure may appear somewhat crude, we found it necessary to simulate natural exposure by providing an environment with very high spore concentrations, as pilot experiments showed that pharaoh ants are very resistant to smaller, albeit still high doses. The control groups had filter paper without spores added.

The Petri plates were placed in fluon coated trays and left undisturbed in 26°C climate cabinets for a week, after which the number of live and dead workers were counted. This time‐span was chosen because it takes several days for the fungus to kill the ants. Dead ants were removed and surface sterilized by dipping them in 70% alcohol, dH2O, 10% NaClO and three subsequent changes of dH2O. At this time the number of surviving larvae was also counted. No dead larvae were surface sterilized as they are routinely cannibalized by the workers. During the experiment the humidity in the plates was not controlled externally as the ants seek out relatively high levels of humidity and often nest close to or in the openings of water tubes. This behavior also provides an environment that facilitates germination of conidiospores. The germination rate of conidiospores was checked separately on agar plates kept under similar conditions and found to be 100%. After an additional six days in high humidity boxes the dead ants were inspected for fungal growth under a stereomicroscope. B. bassiana was assessed as being the cause of death by the presence of hyphae and characteristic conidiospore balls on the cuticle.

91 Data analysis

There were some worker escapes from the groups during the experiments, which meant that some groups lost some workers and others gained some. These exchanges only occurred between groups of the same colony category and treatment, and for the mixed groups only a few individuals were involved, rendering any re‐sorting according to group of origin highly unlikely; we therefore chose to include all experimental groups that still contained a total of 15‐21 workers

(including both dead and live workers) at the end of the experiment to ensure comparable sizes of experimental colonies in our analyses. A total of 138 experimental groups were analyzed using the proportion of workers alive after one week to control for variation in worker number per experimental colony. The data were analyzed in R version 2.10.1 (2009) using a Generalized Linear

Model (glm function) on the proportions of live ants or the number of live brood, applying group and fungal treatment as fixed effects and quasibinomial and poisson error structures for the proportion and count data, respectively. The effects of treatment, group category and the interaction of these terms were further tested using analyses of deviance with F or chi‐square tests.

Results

All experimental colony categories were adversely affected by the Beauveria bassiana treatment and suffered increased worker mortality relative to controls (glm on averages for the different group categories; F1,132 = 9.96, P < 0.002; Fig. 3a). The groups differed in the proportion of workers surviving (F4,133 = 10.23, P < 0.001; Fig. 3a), and although there were significant effects of group

92 category, there was no interaction between group category and treatment (F4,128 = 1.36, P = 0.25).

When using glms to compare the average survival of workers across group categories, the crossed groups had higher survival than the inbred U1 and mixed groups (P < 0.02), but not significantly higher than the inbred U3 and U5 groups. The mixed groups had significantly higher survival than the U1 groups, which had lower average survival than all other groups (P < 0.001, Fig. 3a). The inbred U5 groups had significantly higher survival than all other group categories, except for the crossed groups for which the difference was not significant. For larval survival there was no effect of fungal treatment (P = 0.188, Fig. 3b), but a significant effect of group category (P < 0.001, Fig.

3b), with mixed groups having significantly fewer larvae surviving than all other groups (P < 0.001), and the crossed groups having significantly more larvae surviving than the other groups except U1

(P = 0.008). There was no significant interaction between group category and treatment (P =

0.052), although the U1 larvae did appear to be adversely affected by the fungal treatment (Fig.

3b).

No obvious pathogen avoidance behavior was observed during the experiment. Although no systematic behavioral observations were made, it was striking that workers could be regularly observed walking over the spore infected filter paper rather than changing their course. In addition 5‐10% of the colonies nested under the spore infected filter paper.

We found B. bassiana conidiospore balls on dead ants in 75% of the treatment groups when checked after six days, whereas none of the dead ants from the controls had any B. bassiana growth (Fisher’s exact test, P < 0.0001). Of the 178 ants that died in the fungus treated groups,

101 had conidiospores (i.e. 57% total, varying between 27 and 75% of the dead individuals in the

93 different groups). The values are likely underestimates because fungal growth may take longer to become visible, and sporulation is expected to increase over time (Sun et al. 2002). Overall this confirms that the mortality differences found were largely caused by the fungus.

Discussion

We found a decrease in worker survival in all group categories when exposed to Beauveria bassiana, but no effect of fungal treatment on larval survival. Furthermore, there was considerable variation in worker survival (Fig. 3a) across the different inbred control groups not exposed to the pathogen. For example, lineage U5 had considerably higher survival than lineage U1 (Fig. 3a). This suggests that the inbred lineages may vary in general survival rates and likely in their abilities to resist a pathogen such as B. bassiana. Similarly in other organisms such as Daphnia (Altermatt and

Ebert 2008; Ebert et al. 1998) and Dropsophila (Lazzaro et al. 2006) much inter‐population variation for disease resistance exists. Comparisons of worker survival between the different inbred lineages, the mixed, and the crossed groups show that the crossed groups did significantly better than the U1 lineage and the mixed groups, although the U5 lineage appeared to have equivalent levels of survival (Fig. 3a).

Worker survival in the mixed and crossed groups was approximately as expected if the three inbred lineages contributed to these in an additive way, weighted according to the pedigree (i.e.

U1 contributing 50% and U3 and U5 contributing 25% each; Fig. 3a). This suggests that group disease resistance depends on the presence of specific alleles found in the colony, regardless how

94 the alleles are distributed within and between colony members. However, the crossed groups had higher worker survival than the mixed groups and they appeared to do slightly better than expected from additive contributions of the inbred lineages. This indicates that heterozygosity per se provides survival benefits after exposure to infection. However, we cannot exclude that survival in mixed groups was somewhat affected by subtle forms of within‐group antagonism due to nestmate discrimination behavior (d'Ettorre and Lenoir 2010). Although we did not see any aggression while setting up the small experimental colonies, we did observe some transient aggression shortly after we combined very large numbers of individuals from the inbred lineages in other experiments (author pers. obs.). The observation that larvae in mixed groups were significantly less likely to survive (Fig. 3b) also indicates that there may have been some longer‐ term agonism in these groups. This means that the observed worker and larval mortality in the mixed groups may be an overestimate relative to what it would be in an established colony where nestmate discrimination would not be an issue. However, this does not diminish the result that enhanced heterozygosity resulted in the same level of survival or provided a survival benefit relative to inbred controls.

Reber et al. (2008) showed that experimental mixing of workers from different monogynous

(single queen) colonies of Formica selysi increased the mean resistance of group members when exposed to another pathogenic fungus (Metarhizium anisopilae), but, interestingly, they found that survival in mixed control groups was reduced, consistent with our result. A possible way to exclude antagonistic interactions as a factor could be by rearing worker pupae separately from their colonies to minimize the chance that they acquire colony specific odors and subsequently display discrimination behaviors (Errard et al. 2006; Lenoir et al. 1999; Vander Meer et al. 1998).

95 Our finding of increased larval mortality in mixed groups irrespective of infection may also be due to mergers of different colonies inducing some overall reduction in cohesion of social behavior.

Previous studies suggested that successful colonies may have co‐adapted behavioral complexes, with favorably‐interacting worker, brood, and queen phenotypes (Foitzik et al. 2003; Linksvayer

2007; Linksvayer 2008). It is therefore conceivable that adaptive social cohesion for which selection may have taken place in the different inbred lineages may not be optimally expressed when members of different inbred lineages are combined. This could imply that interactions that enhance social immunity (Cremer et al. 2007) among colony members become impaired, perhaps even to a point that natural fusion of genetically differentiated pharaoh ant colonies is selected against. However, a disruption of co‐adapted gene complexes may be expected to affect the social cohesion of crossed lineages even more (Linksvayer 2008), but here we observed consistently positive effects of crossing on worker survival.

Whereas other studies have shown the occurrence of clear behavioral responses to pathogens, e.g. ants removing and covering up spores (Jaccoud et al. 1999) or termites actively avoiding spores or fungus‐killed individuals (Kramm et al. 1982; Mburu et al. 2009), we did not observe any such responses. In fact, pharaoh ant workers often walked directly through or nested under the filter paper with the spores. If there are behavioral responses of pharaoh ants to B. bassiana, these must therefore be more subtle, post‐exposure responses involving e.g. mutual grooming rather than avoidance. As much research has illustrated the importance of social context and behavior in disease resistance (Calleri et al. 2006; Rosengaus et al. 1998; Traniello et al. 2002;

Wilson‐Rich et al. 2009; Yanagawa et al. 2008), investigating the behavioral interactions of ants in

96 different categories of groups such as the ones that we created would be an interesting way to obtain a better understanding of the underlying mechanisms involved.

Our study shows that increasing the genetic variation in a group of ants does not always lead to increased survival, and that it matters whether genetic diversity is increased within or across nestmates. Apparently, groups with increased inter‐individual diversity may not function optimally even in species where they might be expected to do so because workers can easily be transferred between nests. As natural mixing of pharaoh ant colonies appears to be rare (author unpubl. data), this suggests that gene flow between lineages is highly restricted, consistent with the high

FST values found previously, irrespective of geographic distance (Schmidt et al. 2010). Pharaoh ants can thus be seen as somewhat extreme representatives of polygynous ant species, being inbred, readopting newly mated queens into their colonies, and having only very limited dispersal. Results like this should make us reluctant to consider genetic diversity generated by multiple queen‐ mating and polygyny as being two sides of the same coin (Boomsma and Ratnieks 1996; W.O.H.

Hughes et al. 2008c; Keller and Reeve 1994; Schmid‐Hempel and Crozier 1999). Rather, our results suggest that some form of enhanced colony‐level genetic diversity may often, though not necessarily always, be favorable for disease resistance. It may also be that, if increasing intracolonial genetic diversity, the largest benefits are to be expected from increasing the category of diversity that is most “difficult” to obtain. For outbreeding eusocial insects with single‐queen colonies, this benefit would be obtained through costly additional matings to increase inter‐ individual genetic diversity of nestmates (Baer and Schmid‐Hempel 2003). However, for highly polygynous ants with consistent inbreeding, a benefit may be obtained through otherwise rare hybridization with a genetically distinct other lineage. These different mechanisms of increasing

97 genetic variation have fundamental differences that may affect the nature of the benefits: additional matings come from males from the same outbred population whereas crosses between highly structured polygynous populations involve potentially divergent lineages. This also means that crosses between lineages may have initial advantages due to increased heterozygosity (hybrid vigor) but later generations may experience fitness costs due to the breakup of coadapted complexes (outbreeding depression). It is thus conceivable that increased levels of heterozygosity do not confer a long‐term advantage to colonies and hybridization is selected against.

Even though parasite resistance advantages can be shown to apply in manipulation experiments, this does not imply that these advantages are large under field conditions. The bumblebees of

Baer and Schmid Hempel (2003) have retained their single queen‐mating habit in spite of the documented colony‐level costs in disease resistance. We infer that pharaoh ants have retained their inbred, highly structured populations because colony‐level selection may have caused some colonies to have high levels of resistance and colonies may generally have other fitness advantages that surpass the costs emanating from vulnerability towards disease.

98 Acknowledgements

We thank David Oi and Howard Bell for kindly providing ant lineages, Sophie Armitage and

Annette Bruun Jensen for advice on rearing fungal cultures, Annette Bruun Jensen for providing these cultures, and Luke Holman and Volker Nehring for help with the statistical analyses. David P.

Hughes gave valuable comments on a previous version of this manuscript. Our study was supported by The Danish National Research Foundation (AMS, JJB, JSP) and an EU Marie Curie

International Incoming Fellowship (TAL).

99 References

Altermatt F, Ebert D. 2008. Genetic diversity of Daphnia magna populations enhances resistance to parasites. Ecol. Lett. 11:918‐928. Anderson RM, May RM. 1985. Vaccination and herd‐immunity to infectious diseases. Nature 318:323‐329. Baer B, Schmid‐Hempel P. 2003. Bumblebee workers from different sire groups vary in susceptibility to parasite infection. Ecol. Lett. 6:106‐110. Berndt KP, Eichler W. 1987. Die Pharaoameise, Monomorium pharaonis (L.) (Hym., Myrmicidae). Mitt. Zool. Mus. Berl. 63:3 ‐186. Boomsma JJ. 1996. Split sex ratios and queen‐male conflict over sperm allocation. Proc. R. Soc. Lond. Ser. B‐Biol. Sci. 263:697‐704. Boomsma JJ, Ratnieks FLW. 1996. Paternity in eusocial Hymenoptera. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 351:947‐975. Boomsma JJ, Schmid‐Hempel P, Hughes WOH. 2005. Life histories and parasite pressure across the major groups of social insects. In: Fellowes MDE, Holloway GJ, Rolff J, editors. Insect Evolutionary Ecology. Wallingford, CABI publishing. p. 139‐175. Bulmer MS, Bachelet I, Raman R, Rosengaus RB, Sasisekharan R. 2009. Targeting an antimicrobial effector function in insect immunity as a pest control strategy. Proc. Natl. Acad. Sci. U.S.A. 106:12652‐12657. Calleri DV, Reid EM, Rosengaus RB, Vargo EL, Traniello JFA. 2006. Inbreeding and disease resistance in a social insect: effects of heterozygosity on immunocompetence in the termite Zootermopsis angusticollis. Proc. R. Soc. B‐Biol. Sci. 273:2633‐2640. Chapuisat M, Oppliger A, Magliano P, Christe P. 2007. Wood ants use resin to protect themselves against pathogens. Proc. R. Soc. B‐Biol. Sci. 274:2013‐2017. Cremer S, Armitage SAO, Schmid‐Hempel P. 2007. Social immunity. Curr. Biol. 17:R693‐ R702.

100 Cremer S, Heinze J. 2003. Stress grows wings: Environmental induction of winged dispersal males in Cardiocondyla ants. Curr. Biol. 13:219‐223. Crozier RH, Fjerdingstad EJ. 2001. Polyandry in social Hymenoptera ‐ disunity in diversity? Ann. Zool. Fenn. 38:267‐285. Crozier RH, Page RE. 1985. On being the right size ‐ male contributions and multiplemating in social Hymenoptera. Behav. Ecol. Sociobiol. 18:105‐115. d'Ettorre P, Lenoir A. 2010. Nestmate Recognition. In: Lach L, Parr CL, Abbott KL, editors. Ant Ecology. New York, Oxford University Press. p. 197‐209. Decanini LI, Collins AM, Evans JD. 2007. Variation and heritability in immune gene expression by diseased honeybees. J. Hered. 98:195‐201. Ebert D, Zschokke‐Rohringer CD, Carius HJ. 1998. Within‐ and between‐population variation for resistance of Daphnia magna to the bacterial endoparasite Pasteuria ramosa. Proc. R. Soc. Lond. Ser. B‐Biol. Sci. 265:2127‐2134. Edwards JP, Baker LF. 1981. Distribution and importance of the Pharaoh's ant Monomorium pharaonis (L) in National Health Service Hospitals in England. J. Hosp. Infect. 2:249‐254. Errard C, Hefetz A, Jaisson P. 2006. Social discrimination tuning in ants: template formation and chemical similarity. Behav. Ecol. Sociobiol. 59:353‐363. Evans JD, Spivak M. 2010. Socialized medicine: Individual and communal disease barriers in honey bees. J. Invertebr. Pathol. 103:S62‐S72. Foitzik S, Stratz M, Heinze J. 2003. Ecology, life history and resource allocation in the ant, Leptothorax nylanderi. J. Evol. Biol. 16:670‐680. Gerloff CU, Ottmer BK, Schmid‐Hempel P. 2003. Effects of inbreeding on immune response and body size in a social insect, Bombus terrestris. Funct. Ecol. 17:582‐ 589. Goudet J. 2002. FSTAT, a program to estimate and test gene diversities and fixation indices. 2.9.3.2 ed. Lausanne: Institute of Ecology, University of Lausanne.

101 Hajek AE, Leger RJS. 1994. Interactions between fungal pathogens and insect hosts. Annu. Rev. Entomol. 39:293‐322. Hamilton WD. 1987. Kinship, recognition, disease, and intelligence: constraints of social evolution. In: Ito Y, Brown JL, Kikkawa J, editors. Animal Societies: Theories and Facts. Tokyo, Japan Scientific Societies Press. p. 81–102. Hughes DP, Pierce NE, Boomsma JJ. 2008a. Social insect symbionts: evolution in homeostatic fortresses. Trends Ecol. Evol. 23:672‐677. Hughes WOH, Boomsma JJ. 2004. Genetic diversity and disease resistance in leaf‐cutting ant societies. Evolution 58:1251‐1260. Hughes WOH, Bot ANM, Boomsma JJ. 2010. Caste‐specific expression of genetic variation in the size of antibiotic‐producing glands of leaf‐cutting ants. Proc. R. Soc. B‐Biol. Sci. 277:609‐615. Hughes WOH, Eilenberg J, Boomsma JJ. 2002. Trade‐offs in group living: transmission and disease resistance in leaf‐cutting ants. Proc. R. Soc. B‐Biol. Sci. 269:1811‐1819. Hughes WOH, Pagliarini R, Madsen HB, Dijkstra MB, Boomsma JJ. 2008b. Antimicrobial defense shows an abrupt evolutionary transition in the fungus‐growing ants. Evolution 62:1252‐1257. Hughes WOH, Ratnieks FLW, Oldroyd BP. 2008c. Multiple paternity or multiple queens: two routes to greater intracolonial genetic diversity in the eusocial Hymenoptera. J. Evol. Biol. 21:1090‐1095. Haag‐Liautard C, Vitikainen E, Keller L, Sundstrom L. 2009. Fitness and the level of homozygosity in a social insect. J. Evol. Biol. 22:134‐142. Jaccoud DB, Hughes WOH, Jackson CW. 1999. The epizootiology of a Metarhizium infection in mini‐nests of the leaf‐cutting ant Atta sexdens rubropilosa. Entomol. Exp. Appl. 93:51‐61. Keller L, Reeve HK. 1994. Genetic variability, queen number, and polyandry in social Hymenoptera. Evolution 48:694‐704.

102 Kramm KR, West DF, Rockenbach PG. 1982. Termite pathogens: transfer of the entomopathogen Metarhizium anisopliae between Reticulitermes sp. termites. J. Invertebr. Pathol. 40:1‐6. Lazzaro BP, Sackton TB, Clark AG. 2006. Genetic variation in Drosophila melanogaster resistance to infection: A comparison across bacteria. Genetics 174:1539‐1554. Lenoir A, Fresneau D, Errard C, Hefetz A. 1999. Individuality and colonial identity in ants: the emergence of the social representation concept. In: Detrain C, Deneubourg JL, Pasteels JM, editors. Information processing in social insects. 1 ed. Basel, Birkhäuser. p. 219‐237. Linksvayer TA. 2007. Ant species differences determined by epistasis between brood and worker genomes. PLoS ONE 2:e994. Linksvayer TA. 2008. Queen‐worker‐brood coadaptation rather than conflict may drive colony resource allocation in the ant Temnothorax curvispinosus. Behav. Ecol. Sociobiol. 62:647‐657. Little AEF, Murakami T, Mueller UG, Currie CR. 2006. Defending against parasites: fungus‐ growing ants combine specialized behaviours and microbial symbionts to protect their fungus gardens. Biol. Lett. 2:12‐16. Mburu DM, Ochola L, Maniania NK, Njagi PGN, Gitonga LM, Ndung'u MW, Wanjoya AK, Hassanali A. 2009. Relationship between virulence and repellency of entomopathogenic isolates of Metarhizium anisopliae and Beauveria bassiana to the termite Macrotermes michaelseni. J. Insect Physiol. 55:774‐780. Oldroyd BP, Fewell JH. 2007. Genetic diversity promotes homeostasis in insect colonies. Trends Ecol. Evol. 22:408‐413. Peacock AD, Baxter AT. 1949. Studies in Pharaoh's ant, Monomorium pharaonis (L.). 1. The rearing of artificial colonies. Entomol.Mon. Mag. 85:256‐260. Peacock AD, Sudd JH, Baxter AT. 1955. Studies in Pharaoh's ant, Monomorium pharaonis (L.). 11. Colony foundation. Entomol.Mon. Mag. 91:125‐129.

103 R Development Core Team. 2009. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3‐ 900051‐07‐0, URL http://www.R‐project.org. Reber A, Castella G, Christe P, Chapuisat M. 2008. Experimentally increased group diversity improves disease resistance in an ant species. Ecol. Lett. 11:682‐689. Rosengaus RB, Maxmen AB, Coates LE, Traniello JFA. 1998. Disease resistance: a benefit of sociality in the dampwood termite Zootermopsis angusticollis (Isoptera : Termopsidae). Behav. Ecol. Sociobiol. 44:125‐134. Rothenbuhler WC. 1964. Behaviour genetics of nest cleaning in honey bees. I. Responses of four inbred lines to disease‐killed brood. Anim. Behav. 12:578‐583. Schmid‐Hempel P. 1998. Parasites in Social Insects. Princeton: Princeton University Press. Schmid‐Hempel P, Crozier RH. 1999. Polyandry versus polygyny versus parasites. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 354:507‐515. Schmidt AM, d'Ettorre P, Pedersen JS. 2010. Low levels of nestmate discrimination despite high genetic differentiation in the invasive pharaoh ant. Front. Zool. 7:20. Schrempf A, Aron S, Heinze J. 2006. Sex determination and inbreeding depression in an ant with regular sib‐mating. Heredity 97:75‐80. Strassmann J. 2001. The rarity of multiple mating by females in the social Hymenoptera. Insect. Soc. 48:1‐13. Sun JZ, Fuxa JR, Henderson G. 2002. Sporulation of Metarhizium anisopliae and Beauveria bassiana on Coptotermes formosanus and in vitro. J. Invertebr. Pathol. 81:78‐85. Sundström L, Seppa P, Pamilo P. 2005. Genetic population structure and dispersal patterns in Formica ants ‐ a review. Ann. Zool. Fenn. 42:163‐177. Tarpy DR. 2003. Genetic diversity within honeybee colonies prevents severe infections and promotes colony growth. Proc. R. Soc. Lond. Ser. B‐Biol. Sci. 270:99‐103. Tarpy DR, Seeley TD. 2006. Lower disease infections in honeybee (Apis mellifera) colonies headed by polyandrous vs monandrous queens. Naturwissenschaften 93:195‐199.

104 Team RDC. 2009. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; R Development Core Team. Traniello JFA, Rosengaus RB, Savoie K. 2002. The development of immunity in a social insect: Evidence for the group facilitation of disease resistance. Proc. Natl. Acad. Sci. U. S. A. 99:6838‐6842. Ugelvig LV, Kronauer DJC, Schrempf A, Heinze J, Cremer S. 2010. Rapid anti‐pathogen response in ant societies relies on high genetic diversity. Proc. R. Soc. B:Biol. Sci.: doi:10.1098/rspb.2010.0644 Vander Meer RK, Breed MD, Winston ML, Espilie KE. eds. 1998. Pheromone Communication in Social Insects: Ants, Wasps, Bees, and Termites. Boulder, Colorado: Westview Press. Wilson‐Rich N, Spivak M, Fefferman NH, Starks PT. 2009. Genetic, individual, and group facilitation of disease resistance in insect societies. Annu. Rev. Entomol. 54:405‐ 423. Wilson EO. 1971. The Insect Societies. Cambridge, Massachusetts: Harvard University Press. Yanagawa A, Yokohari F, Shimizu S. 2008. Defense mechanism of the termite, Coptotermes formosanus Shiraki, to entomopathogenic fungi. J. Invertebr. Pathol. 97:165‐170.

105 Fig. 1 The crossing scheme applied to obtain individuals of high genetic diversity in Monomorium pharaonis. The bars represent putative alleles at any marker locus; as males are haploid these are represented by one bar only. After two crossings (generations) the individuals of the F2 generation had contributions from all three parental lineages

106 Fig. 2 Experimental setup and genetic diversity of the different treatment groups of Monomorium pharaonis. Three highly genetically differentiated, low diversity inbred source lineages (U5, U3 and U1; average allelic richness, k’ = 2.1 ± 0.2 SD; average heterozygosity, H = 0.322 ± 0.109 SD, cf. Methods) were used singly, mixed or crossed thereby creating groups of different levels of intra‐ or inter‐individual level of genetic diversity (mixed groups, k’ = 4.3 ± 0.7 SD, H = 0.322 ± 0.109 SD; crossed groups, k’ = 4.1 ± 1.1 SD, H = 0.619 ± 0.051, cf. Methods). Each worker in the figure represents six and each larva in the figure represents four individuals. The bars of different colors represent alleles of different origin as given in Fig. 1

107 Fig. 3 Proportion of Monomorium pharaonis worker ants (a) and number of larvae (b) alive after one week. Open and filled symbols indicate controls and Beauveria bassiana treatments respectively. The B. bassiana treated groups had significantly lower worker ant survival. Number of replicates (without/with fungal treatment): U1, n = 18/11; U3, n = 16/18; U5, n = 12/9; mixed, n = 12/17; crossed, n = 9/16. Significant differences between group categories are indicated by different lower case letters (a‐d) on the graphs.

108

109

Chapter 4

Queen‐worker caste ratio depends on colony size in the pharaoh ant (Monomorium pharaonis)

Anna M. Schmidt, Tim A. Linksvayer, Jacobus J. Boomsma and Jes S. Pedersen

Submitted to Insectes Sociaux

110

111

Queen‐worker caste ratio depends on colony size in the pharaoh ant (Monomorium pharaonis)

Anna M. Schmidt1§, Timothy A. Linksvayer1, Jacobus J. Boomsma1 and Jes S. Pedersen1

Centre for Social Evolution, Department of Biology, University of Copenhagen, Universitetsparken 15, DK‐2100 Copenhagen, Denmark

The success of an ant colony depends on the simultaneous presence of reproducing queens and non‐reproducing workers in a ratio that will maximize colony growth and reproduction. Despite its presumably crucial role, queen‐worker caste ratios (the ratio of adult queens to workers) and the factors affecting this variable remain scarcely studied. Maintaining polygynous pharaoh ant (Monomorium pharaonis) colonies in the laboratory has provided us with the opportunity to experimentally manipulate colony size, one of the key factors that can be expected to affect colony level queen‐worker caste ratios and body size of eclosing workers, gynes and males. We found that smaller colonies produced more new queens relative to workers, and that these queens and workers both tended to be larger. However, colony size had no effect on the size of males or on the sex ratio. Furthermore, for the first time in a social insect, we confirmed the general life‐history prediction by Smith and Fretwell (1974) that offspring number varies more than offspring size. Our findings document a high level of plasticity in energy allocation towards female castes and suggest that polygynous species with budding colonies adaptively adjust caste ratios to ensure rapid growth.

Key words: Caste, colony size, ergonomics, resource allocation, polygyny

112 Introduction

Ant colonies are typically founded by a single queen and go through different stages of resource allocation to optimize growth and reproduction. During the founding stage, queens rear the first batch of workers (e.g. Oster and Wilson, 1978), after which the colony enters the ergonomic stage, characterized by rapid exponential growth. Once the colony has reached a certain worker number, it enters the reproductive stage, during which excess resources are periodically invested in new reproductives (males and gynes) rather than more workers (Oster and Wilson,

1978; Hölldobler and Wilson, 1990; Bourke and Franks, 1995). However, there are also many polygynous ant species that have variable numbers of queens per colony and recurrently readopt newly fertilized daughter queens (Glancey and Lofgren, 1988; Hölldobler and Wilson,

1990; Crozier and Pamilo, 1996).

Although both queen and worker numbers vary in polygynous species, systematic studies of queen‐worker caste ratio allocation have rarely been done, likely due to logistic difficulties usually associated with such colony level studies. Instead, studies have focused on phenotypic plasticity in allocation to different worker castes. Several studies have emphasized the importance of specific environmental components in determining allocation towards workers and soldiers as a function of competition (Passera et al., 1996) or food availability (McGlynn and

Owen, 2002). In recent years, other studies have shown that caste determination may have a significant genetic component (Jaffe et al., 2007; Hughes and Boomsma, 2008; Schwander et al.,

2008). Such studies also often indicate a strong effect of the social environment, i.e. of the workers rearing the new brood and actively influencing resource allocation during larval development (reviewed by e.g. Anderson et al., 2008; Schwander et al., 2010).

113

Pharaoh ants (Monomorium pharaonis) are perhaps one of only few ant species that can be mass‐reared in the laboratory for many generations, because the species is highly polygynous

(Wheeler, 1986), produces new broods of reproductives at intervals of down to only a few months, and has intranidal mating (i.e. mating in the nest without dispersal of either sex). New queens thus become reproductively active and contribute to colony growth shortly after eclosing. In addition, the species exclusively disperses by colony budding of worker groups and brood that either contain queens or rapidly rear a new cohort of queens and males (Berndt and

Eichler, 1987). Finally, pharaoh ants have genetically highly structured populations, but can nonetheless be freely mixed in laboratory cultures without any significant aggression between unrelated workers or queens (Schmidt et al., in press). This implies that both queen number and worker number in pharaoh ants vary continuously, providing unique opportunities for studying queen‐worker caste allocation under experimentally controlled conditions, i.e. allocation towards egg‐laying potential on one hand, and “somatic” support by completely sterile workers on the other.

The objective of our study was to experimentally create queenless colonies of three size classes, provide them with a constant amount of brood relative to workers, and measure the queen‐ worker caste ratio, body sizes and sex ratio of the brood that these colonies reared. This design was meant to provide insight into the extent to which colony level selection may have produced phenotypically plastic responses that are likely to maximize colony survival across a range of different colony sizes, consistent with earlier correlative evidence in the leafcutter ant, Atta cephalotes (Wilson, 1983) and the red imported fire ant, Solenopsis invicta (Porter and

Tschinkel, 1985). In addition, this experimental set‐up further allowed us to test the basic life

114 history principle formulated by Smith and Fretwell (1974), that offspring number is expected to vary more than offspring size. The rationale of this idea is that offspring size should normally be under balancing (optimizing) selection, so that different parental resource levels translate primarily into varying offspring number. To our knowledge, this hypothesis has never been tested in social insects.

Methods

Study species

Pharaoh ants (Monomorium pharaonis) are commonly found as massive pest infestations in houses (Edwards, 1986; Berndt and Eichler, 1987). There does not appear to be any seasonality in their reproductive cycles as new sexuals are produced every 3‐8 months (Petersen‐Braun,

1975, AMS pers. obs.), which may be due to their tropical origin. Caste determination has been shown to depend on the social environment, i.e. on whether fertile queens are present in the colonies (Edwards, 1987; 1991), and on the age of queens (Petersen‐Braun, 1977). As in the

Argentine ant (Linepithema humile) and the red imported fire ant, pharaoh ant workers have been shown to be able to discriminate and kill sexual larvae and thereby potentially control colony caste ratios (Vargo and Fletcher, 1986; Edwards, 1991; Vargo and Passera, 1991;

Klobuchar and Deslippe, 2002). Eggs laid by queens older than four weeks are bipotent

(Petersen‐Braun, 1977), meaning that they can develop into either gynes or workers. Eggs are always queen‐laid, as workers have no ovaries and are thus completely sterile (Berndt and

Eichler, 1987), and experimental removal of queens from a colony results in the almost immediate rearing of new sexuals. New queens and males mate within their natal nest so that pharaoh ants can potentially be kept in the laboratory indefinitely (Peacock and Baxter, 1949;

115 Berndt and Eichler, 1987). Pharaoh ants generally display only low levels of intraspecific aggression, which means that it is usually possible to combine worker ants from different colonies without agonistic interactions (Schmidt et al., in press).

Experimental setup

Brood and workers from 10 genetically different laboratory colonies were mixed to form one large queenless experimental colony. Three different experimental colony sizes (small, medium and large) were subsequently created by transferring either ca. 1 ml, 2 ml, or 3 ml brood and workers from the experimental colony to 15 cm diameter Petri plates using a metal measuring spoon. This approach provided sub‐colonies of approximately the same worker‐brood composition, but did not enable us to determine the exact numbers of individuals in each colony. Water tubes plugged with cotton wool and food ad libitum (a diet consisting of cooked liver, boiled egg yolk, honey, mealworms and almonds) were added, and colonies were kept in climate rooms at 26‐28°C. Starting 25 days after the creation of the experimental colonies, all pupae produced were continuously removed and the number of offspring workers, gynes and males counted every 4‐5 days for the next 13‐18 days until no brood was left. To avoid matings and enable verification of caste identifications, gynes and males collected from each colony were reared separately in 10 cm diameter Petri plates, which were also supplied with food and water ad libitum. A few adult workers were added to the gyne and male plates to facilitate eclosure of the pupae (Edwards, 1986).

Once eclosed, the new workers, gynes, and males were placed in 95% EtOH and subsequently the head widths of individuals from each caste and experimental colony were measured at 50x magnification using a Leica MZ125 microscope with attached Leica DFC420 camera and analyzing

116 data using Leica Application Suite. Head width was measured directly above the eyes, and chosen as proxy for overall body size, as head width is known to correlate strongly with overall size of ants (Hölldobler and Wilson, 1990). Fifteen workers and gynes were measured from each colony and, when possible, 15 males as well. The number of males measured varied from 4‐15 as male pupae appear to be particularly fragile and have high pre‐eclosure mortality, so that a large number of the male pupae did not survive till eclosure. We do not know whether our handling had any role in causing the mortality, so we chose to calculate sex ratios based on the number of pupae that were originally removed from the colonies.

Statistics

Queen‐worker caste ratios were calculated as the ratio of gynes produced to the total number of females (gynes + workers) produced. The data were analyzed in R version 2.10.1 (2009) using a Generalized Linear Model (GLM function) on the queen‐worker caste ratios of colonies with group size category (small, medium or large) or colony productivity (measured as the total number of individuals reared by the workers of a given colony) as fixed effect and a quasibinomial error structure. Logit regressions were used to predict changes in queen‐worker caste ratio relative to experimental colony size or total offspring production, using also an analysis of deviance to compare the three colony size categories. In addition, we used a GLM with a binomial error structure to analyze queen‐worker caste ratios as a function of colony productivity as fixed effect, i.e. offspring number adjusted for differences in the mean fresh weight of workers, males and gynes: unpublished data for workers 0.247 mg (n = 714), gynes

1.359 mg (n = 247), males 0.955 mg (n = 235). Sex ratios were calculated as the ratio of gynes to the total number of males and gynes produced, and the data were analyzed by the same approach as for the queen‐worker caste ratio data, using a binomial error structure in the GLM.

117 Linear Mixed Effect models (LME function) were applied to the individual size measurement data using group size category (small, medium, large) or productivity as a random effect and colony ID as fixed effect. The group size effect analyses were also combined with analyses of deviance.

Results

The colonies produced between 52‐157 gynes, 15‐71 males and 196‐1280 workers and the total number of individuals ranged from 263‐1436. Queen‐worker caste ratios varied from 0.076 to

0.238 (gynes/(gynes+workers); median = 0.155, interquartile range = 0.048 (Fig. 1). Sex ratios were female biased at 0.60‐0.85 (gyne/(gynes+males); median = 0.739, interquartile range =

0.028. Our analyses showed a significant effect of colony size category (small, medium, large) on caste ratio (GLM, analysis of deviance, P < 0.001) as size change from large to medium caused an increase in queen‐worker caste ratio by a factor of 1.21 (t 36 = 2.09, P = 0.04) and large to small caused an increase by a factor of 1.79 (t 36 = 5.54, P < 0.001). We also found a significant negative correlation between queen‐worker caste ratio and total number of offspring reared, as the queen‐worker caste ratio decreased by a factor of 0.37 for every unit of increase in offspring number (Fig. 1a; t 37 = –8.93, P < 0.001). Likewise a significant correlation was found between queen‐worker caste ratio and total productivity measured as adjusted fresh weight (Fig. 1b; z37 =

4.55, P < 0.001). Expressed as ratios rather than proportions, the small colonies had queen‐ worker caste ratios of approximately 1:3 (number of queens to number of workers) whereas the large colonies had queen‐worker caste ratios of around 1:12 (right vertical axis in Fig. 1).

118 Colony size category had a small, but statistically significant effect on mean gynes size (F2,36 =

3.2, P = 0.050) as the medium and small groups produced larger gynes (large to medium: t 36 =

2.27, P = 0.029; large to small: t 36 = 2.12, P = 0.040) and, although not significant overall (F2,36 =

2.33, P = 0.111), it appeared that worker size was also slightly affected as the small groups produced slightly larger workers than the large groups (large to medium: t 36 = 1.68, P = 0.100; large to small: t 36 = 2.01, P = 0.051). Colony size category had no effect on the size of males (F2,36

= 1.40, P = 0.36) nor on the sex ratio produced (analysis of deviance, P = 0.33). There was no significant effect of the total number of offspring produced on the size of any of the castes (Fig.

2; workers: t 36 = –0.80, P = 0.43; gynes: t 36 = –1.18, P = 0.24; males: t 36 = –1.78, P = 0.083), nor on the sex ratio of the brood produced (t37 = 0.72, P = 0.48).

Discussion

Although many polygynous ants reproduce by colony budding (Hölldobler and Wilson, 1990;

Bourke and Franks, 1995), the pharaoh ant breeding system has taken this mode of reproduction to extremes because worker groups that bud off from the parental nest do not need to contain queens, and often rear them from bipotent female brood (Petersen‐Braun,

1977; Berndt and Eichler, 1987). This implies that flexible caste‐ratio responses likely have evolved in response to selection for maximizing colony growth. As constraints on either queen or worker production appear to be negligible in pharaoh ants, our data show the kind of plasticity colony level selection has resulted in when it comes to the effect of group size on queen‐worker caste ratio. It further allows an assessment of the degree to which such selection regime has affected how individual body size varies relative to variation on offspring number.

119 In ants where queens and males disperse, reproduction normally occurs when the marginal fitness returns from further worker production have fallen below a certain threshold (Tschinkel,

1993; Crozier and Pamilo, 1996; Bourke and Ratnieks, 1999). However, when sexuals do not disperse, but remain in the colony (either as new queens or as sperm stored by these queens), the overall ergonomic principles for queen‐worker caste allocation should be more straightforward because intra‐colony conflict has disappeared (Nonacs, 1988; Pamilo, 1991;

Nonacs, 1993): workers are completely sterile and queens can be produced and culled at almost any time when this serves the collective interest. Our results clearly show that under such circumstances there is no single optimal proportion of individuals reared as gynes versus workers, but instead we see a graded response producing four times as many gynes in small relative to large experimental colonies (Fig. 1). This high level of plasticity likely enables colonies to respond rapidly to the environmental disturbances that naturally tend to create colonies of similar small size as our experimental colonies (Buczkowski and Bennett, 2009, AMS, pers. obs.).

We hypothesize that the observed plasticity of the queen‐worker caste ratio is adaptive as it prioritizes egg‐laying potential in the smallest colony fragments and foraging potential in larger fragments where enough queens are present so that eggs are no longer a limiting factor for colony growth.

The colonies used to create the mixed experimental colony had all been established from colony fragments at least one year prior to the experiment, which meant that they should be considered mature in the sense of having reached a rather large size and a balanced queen‐ worker caste ratio. None of the source colonies were producing sexuals at the time, so we assumed that all diploid brood in the experimental colonies that we created was developmentally bipotent (i.e. able to develop into both gynes and workers). The queen‐worker

120 caste ratios found in our experimental colonies should therefore reflect the typical allocation of

“colony buds” as our experimental colony sizes were small compared to the mature source colonies used to create the overall stock. Because we mixed ants and brood from a number of genetically different colonies when creating our experimental colonies, we expect that our results reflect a general average relationship between colony size and caste ratio in pharaoh ants. However, this does not preclude that variation around this average may exist across different genetic lineages of pharaoh ants, as preliminary data from non‐mixed colonies seem to indicate (AMS and TAL pers. obs.).

The observed effect of colony size on gyne and worker body size provides a test of the Smith and Smith & Fretwell (1974) model predicting that in non‐social animals offspring number should be much more variable than offspring size. Our results indicate that this expectation was largely met as gyne size varied by 2.01 % around the overall mean and worker size by 2.37 %, whereas gyne and worker numbers varied by 25.4 and 43.8 %, respectively. This difference of about an order of magnitude suggests that group selection on collective performance per se is not sufficient to create more variation in body size. This may be because there is only a single way of being a good egg layer in pharaoh ants where new colonies are only produced by budding, in contrast to other polygynous ants where trade‐offs from alternative dispersal tactics has led to the evolution of a bimodal queen caste (Ruppell and Heinze, 1999).

Body size of gynes and workers varied with experimental colony size, but male body size and sex ratio remained constant across the three colony size categories. This indicates that, with exclusive intranidal mating and no dispersal, male size and number hardly represent an independent fitness component. All that apparently matters is that male number increases

121 proportionally with gyne number (hence a constant sex ratio), so that gynes can be inseminated at a constant rate to initiate their reproductive roles. This suggests that there has been no selection for phenotypic plasticity in male function, neither in terms of body size, nor in relative numbers per gyne to mate with.

Acknowledgements

We thank Christa Funch Jensen, Isabel Højgaard Rasmussen, Markus Drag, Mathilde Lerche‐

Jørgensen, Nathia Hass Brandtberg and Signe Lolle for help sorting, feeding, and measuring ants.

The study was supported by The Danish National Research Foundation (AMS, JJB, JSP) and an EU

Marie Curie International Incoming Fellowship (TAL).

122 References

Anderson, K. E., Linksvayer, T. A. and Smith, C. R. 2008. The causes and consequences of genetic caste determination in ants (Hymenoptera: Formicidae). Myrmecol. News 11: 119‐132 Berndt, K. P. and Eichler, W. 1987. Die Pharaoameise, Monomorium pharaonis (L.) (Hym., Myrmicidae). Mitt. Zool. Mus. Berl. 63: 3 ‐186 Bourke, A. F. G. and Franks, N. R. 1995. Social Evolution in Ants. Princeton University Press, Princeton, New Jersey. 529 pp. Bourke, A. F. G. and Ratnieks, F. L. W. 1999. Kin conflict over caste determination in social hymenoptera. Behav. Ecol. Sociobiol. 46: 287‐297 Buczkowski, G. and Bennett, G. 2009. Colony budding and its effects on food allocation in the highly polygynous ant, Monomorium pharaonis. Ethology 115: 1091‐1099 Crozier, R. H. and Pamilo, P. 1996. Evolution of Social Insect Colonies: Sex Allocation and Kin Selection. Oxford University Press, New York. 306 pp. Edwards, J. P. 1986. The Biology, Economic Importance, and Control of the Pharaoh's Ant In: S. Bradleigh Vinson (ed) Economic Impact and Control of Social Insects. Praeger Publishers, New York, pp 257‐271 Edwards, J. P. 1987. Caste regulation in the pharaoh's ant Monomorium pharaonis: the influence of queens on the production of new sexual forms. Physiological Entomology 12: 31‐39 Edwards, J. P. 1991. Caste regulation in the pharaoh's ant Monomorium pharaonis: recognition and cannibalism of sexual brood by workers. Physiological Entomology 16: 263‐271 Glancey, B. M. and Lofgren, C. S. 1988. Adoption of newly‐mated queens; a mechanism for proliferation and perpetuation of polygynous red imported fire ants, Solenopsis invicta, Buren. Florida Entomologist 71: 581‐587 Hughes, W. O. H. and Boomsma, J. 2008. Genetic royal cheats in leaf‐cutting ant societies. Proc. Natl. Acad. Sci. U. S. A. 105: 5150‐5153 Hölldobler, B. and Wilson, E. O. 1990. The Ants. Springer, New York. 732 pp. Jaffe, R., Kronauer, D. J. C., Kraus, F. B., Boomsma, J. J. and Moritz, R. F. A. 2007. Worker caste determination in the army ant Eciton burchellii. Biol. Lett. 3: 513‐516

123 Klobuchar, E. A. and Deslippe, R. J. 2002. A queen pheromone induces workers to kill sexual larvae in colonies of the red imported fire ant (Solenopsis invicta). Naturwissenschaften 89: 302‐304 McGlynn, T. P. and Owen, J. P. 2002. Food supplementation alters caste allocation in a natural population of Pheidole flavens, a dimorphic leaf‐litter dwelling ant. Insect. Soc. 49: 8‐ 14 Nonacs, P. 1988. Queen number in colonies of social Hymenoptera as kin‐selected adaptation. Evolution 42: 566‐580 Nonacs, P. 1993. The effects of polygyny and colony life history on optimal sex investment. In: L. Keller (ed) Queen Number and sociality in Incects. Oxford University Press, New York, pp 110‐131 Oster, G. F. and Wilson, E. O. 1978. Caste and ecology in the social insects, vol 12. Princeton University Press, Princeton, New Jersey. 352 pp. Pamilo, P. 1991. Evolution Of Colony Characteristics In Social Insects .1. Sex Allocation. Am. Nat. 137: 83‐107 Passera, L., Roncin, E., Kaufmann, B. and Keller, L. 1996. Increased soldier production in ant colonies exposed to intraspecific competition. Nature 379: 630‐631 Peacock, A. D. and Baxter, A. T. 1949. Studies in Pharaoh's ant, Monomorium pharaonis (L.). 1. The rearing of artificial colonies. Entomologists' Monthly Magazine 85: 256‐260 Petersen‐Braun, M. 1975. Investigations on social organization in Pharaohs ant, Monomorium pharaonis L (Hym Formicidae) .1. Regulation of brood cycle. Insect. Soc. 22: 269‐291 Petersen‐Braun, M. 1977. Investigations on social organization of Pharaohs ant Monomorium pharaonis L (Hymenoptera, Formicidae) .2. Caste determination. Insect. Soc. 24: 303‐ 318 Porter, S. D. and Tschinkel, W. R. 1985. Fire ant polymorphism (Hymenoptera: Formicidae): factors affecting worker size. Ann. Entomol. Soc. Am. 78: 381‐386 R Development Core Team (2009). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3‐900051‐07‐0, http://www.R‐project.org. Ruppell, O. and Heinze, J. 1999. Alternative reproductive tactics in females: the case of size polymorphism in winged ant queens. Insect. Soc. 46: 6‐17

124 Schmidt, A. M., d'Ettorre, P. and Pedersen, J. S. Low levels of nestmate discrimination despite high genetic differentiation in the invasive pharaoh ant. In press. Front. Zool. Schwander, T., Humbert, J.‐Y., Brent, C. S., Cahan, S. H., Chapuis, L., Renai, E. and Keller, L. 2008. Maternal effect on female caste determination in a social insect. Curr. Biol. 18: 265‐ 269 Schwander, T., Lo, N., Beekman, M., Oldroyd, B. P. and Keller, L. 2010. Nature versus nurture in social insect caste differentiation. Trends Ecol. Evol. 25: 275‐282 Smith, C. C. and Fretwell, S. D. 1974. Optimal balance between size and number of offspring. Am. Nat. 108: 499‐506 Tschinkel, W. R. 1993. Sociometry and sociogenesis of colonies of the fire ant Solenopsis invicta during one annual cycle Ecol. Monogr. 63: 425‐457 Vargo, E. L. and Fletcher, D. J. C. 1986. Evidence of pheromonal queen control over the production of male and female sexuals in the fire ant, Solenopsis invicta. J. Comp. Physiol. A‐Sens. Neural Behav. Physiol. 159: 741‐749 Vargo, E. L. and Passera, L. 1991. Pheromonal and behavioral queen control over the production of gynes in the Argentine ant Iridomyrmex humilis (Mayr). Behav. Ecol. Sociobiol. 28: 161‐169 Wheeler, D. E. 1986. Developmental and Physiological Determinants of Caste in Social Hymenoptera: Evolutionary Implications. Am. Nat. 128: 13‐34 Wilson, E. O. 1983. Caste and division of labor in leaf‐cutter ants (Hymenoptera, Formicidae, Atta). 4. Colony ontogeny of Atta cephalotes. Behav. Ecol. Sociobiol. 14: 55‐60

125 Fig. 1 Queen‐worker caste ratio (ratio of gynes produced relative to total production of females) as a function of colony productivity, expressed as (a) total number of offspring and (b) total fresh weight of offspring (n = 39 colonies). The gyne:worker ratios corresponding to the proportion of gynes are given on the right‐hand side.

126 Fig.2 Relationship between colony productivity (total number of offspring reared) in a given colony and the size (head width) of (a) workers, (b) gynes and (c) males (n = 39 colonies).

0.48 width (mm) width (mm) 0.46

0.44 Worker head Worker head

0.42

0.66 b

0.64 width (mm) width (mm)

0.62 Gyne head Gyne head 0.60

0.58

0.70 c

0.65 width (mm) width (mm)

0.60 head head

Male Male 0.55

0.50

127

Chapter 5

The pharaoh ants as a model organism: creating a heterogeneous stock and selecting on a social insect colony level trait

Anna M. Schmidt

128

129

The pharaoh ants as a model organism: creating a heterogeneous stock and selecting on a social insect colony level trait

The pharaoh ant is one of only few ant species that can be kept in laboratories, induced to produce sexuals in the laboratory and for which it is also possible to make controlled crosses. Bearing this potential in mind, I set out to perfect a crossing protocol and do sequential, multi‐ generation crosses to obtain a genetically diverse, i.e. heterogeneous, stock (Hansen, Spuhler, 1984). From this resource it has been ‐ and will be ‐ possible to address a number of hypothesis on evolution in general, and social evolution in particular, in novel ways. One example is the effect of genetic diversity on groups of individuals (cf. Chapter 3 this thesis); another can be the heritability of traits, as may be investigated through artificial selection experiments. Although the practices of selective breeding and artificial selection experiments have been extensively used in model organisms such as mice, rats and fruit flies (e.g. Fuller et al., 2005; Hansen, Spuhler, 1984; Kraaijeveld, Godfray, 1997), this is the first time multi‐generation selective breeding has been done on an ant species. This chapter briefly outlines the methods I have refined for making crosses in the laboratory as well as a selection experiment I am currently running in collaboration with Dr. Tim Linksvayer.

130 Introduction and study species

Artificial selection has long been employed as a valuable tool for demonstrating and understanding the heritability and evolution of traits (Fuller et al., 2005; Hill, Caballero, 1992).

Previous work and pilot studies have shown that pharaoh ants are highly differentiated genetically as well as morphologically and likely behaviourally (Schmidt et al., (2010); AMS unpublished data; see also Fig. 1). These properties and the relative ease with which pharaoh ants can be kept and mated in the laboratory, makes the species a prime candidate for the first attempt to perform artificial selection on ant colonies, as well as other experiments manipulating the genetic composition of individuals or colonies.

Pharaoh ants (Monomorium pharaonis) are introduced ants commonly found as pests in houses

(Berndt, Eichler, 1987; Edwards, 1986; Schmidt et al., 2010). There does not appear to be any seasonality in their reproductive cycles and new sexuals are produced every 3‐8 months

(Petersen‐Braun, 1975, AMS pers. obs.). Caste determination has been shown to depend on the social environment, i.e. on whether fertile queens are present in the colonies (Edwards, 1987;

1991), and on the age of queens (Petersen‐Braun, 1977), and pharaoh ant workers are able to discriminate and kill sexual larvae and thereby potentially control colony caste ratios (Edwards,

1991). Eggs laid by queens older than four weeks are bipotent (Petersen‐Braun, 1977), meaning that they can develop into either gynes or workers. Eggs are always queen‐laid, as workers are completely sterile (Berndt, Eichler, 1987), and experimental removal of queens from a colony results in the almost immediate rearing of new sexuals. New queens and males mate within their natal nest, meaning that pharaoh ants can potentially be kept in the laboratory indefinitely

(Berndt, Eichler, 1987; Peacock, Baxter, 1949).

131 Pharaoh ants should be kept at at 26‐28°C, supplied with water tubes plugged with cotton wool and food ad libitum (a diet consisting of cooked liver, boiled egg yolk, honey, mealworms and almonds). In order to obtain sexuals or measure the caste ratio of a colony all old queens are removed from either a fragment of the colony or the whole colony.

Fig. 1. Variation in the size (measured as headwidth, top) and weight (bottom) of workers produced in six different colonies of pharaoh ants. These were same size colonies that had been kept in laboratories for at least two years. Bars indicate 95% CI from PASW 18 (AMS and TAL unpublished data). The measurements are based on average mass and head width of 15‐50 worker pupae measured within the last approximately 24 hours before eclosion.

132 Creating a heterogeneous stock

Genetic variation is needed for selection to take place and artificial selection experiments should therefore take their starting point in a genetically diverse stock, from which individuals displaying different phenotypes can be chosen each generation to parent the next generation

(Falconer, Mackay, 1996). In many model systems, however, the stocks (or lines) kept in laboratories may be highly inbred strains, and therefore have only low levels of variation, limiting the potential response to selection. In such cases, the first step is the creation of a genetically heterogeneous stock through sequential crossings of a number of the inbred laboratory lineages. This scheme was applied to the pharaoh ants, starting with eight highly divergent, low genetic diversity lineages (Schmidt et al., 2010), which were combined following the approach of Hansen and Spuhler (1984), albeit with the modification that a fully crossed design was employed in the first generation and in subsequent generations crosses were made to make all combinations possible in order to maintain a high number of colonies. Fig. 2 illustrates part of the crossing scheme, representing the equivalent of Hansen and Spuhler’s crosses (1984), see legend for further details. Ants from the F2 generation were used for selection experiments to be able to conduct these within a couple of years.

Selection on caste ratio and perspectives

As the developmental fate of female (diploid) hymenopteran brood in most known cases is determined primarily or exclusively on their environment (food they receive from the workers

(Hölldobler, Wilson, 1990)), selecting on caste ratio is interesting as it is a colony level trait,

133

Fig. 2. Crossing scheme employed, methodology adapted from Hansen and Spuhler (1984), with the modification that F1 was made based on full factorial crosses of the eight original colonies. The depicted crosses thus only represent a fraction of the crosses made in each generation since as many combinations as possible were made maintain a high number of colonies. One gyne and one male represent each colony in the P (parental) generation, colour‐coded according to colony. As the crosses progress the colours in the filial generations symbolize the mixing of genes from the parents. Each cross was made from 10‐20 gynes and 4‐15 males (example bottom right).

not only reflecting the genotype of the focal individual, but also those of the individuals with whom it interacts – in the form of the other colony members, i.e. indirect genetic effects (IGE,

(Anderson et al., 2008; Linksvayer, 2006)). Fig. 3 represents a simplified schematic of this, the focal individual’s phenotype z is affected directly by it’s own genotype and environment (a and

134 e) as well as indirectly by those of the individuals with which it interacts (a’ and e’) (extended version in McGlothin et al. (2010)).

Fig. 3. Direct genotypic and environmental effects (a and e) as well as the indirect genetic effects of the phenotypes of its nestmates (z’), affect the phenotype of the focal individual (top).

To increase our understanding of the importance of the IGE of the extensive interaction networks characterizing social hymenoptera (Johnson, Linksvayer, 2010) and specifically that related to differential investment in queens and workers (Schwander et al., 2010), I started a selection experiment on queen‐worker caste ratio in collaboration with Dr. Tim Linksvayer, as this is an essential colony‐level trait potentially strongly influencing colony fitness. The experiment includes selection for high and low caste ratios in parallel (Fig. 4.).

Selection for high and low lines is standardized so that two high lines and two low lines each containing approx. 36 colonies are maintained in each generation. Caste ratios are calculated as the number of queens relative to the total number of females reared, and found by counting all individuals reared in all colonies starting 23 days after de‐queening. During the experiment measurements are made of the size of the workers, males and gynes when possible, and the

135

Fig. 4. Theoretical schematic of first round of selection. A normally distributed range of the measured trait, caste ratios in this case, are measured (top) and the colonies producing the highest and lowest caste ratios are used in separate groups of crossings to create the next generation (red arrows). If a response to selection occurs, the means of the new distributions are expected to differ from the original in the direction of selection after a number of generations (bottom).

occurrence of intercastes is noted as this phenomenon is occasionally found in pharaoh ants

(Hall, Smith, 1953; Peeters, 1991). The colonies in each group were created based on controlled randomized combinations of sexuals from sets of 12 colonies having the highest or lowest caste ratios. So far, two generations of selection have been processed and we see no response to selection, as this is often the case in the first couple of generations (Falconer, Mackay, 1996), the third ‐and likely final‐ generation will be censused this fall. Combining results from this with behavioural observations and experiments such as cross‐fostering may help shed light on the combined roles of nature and nurture in pharaoh ants, and in the longer term, building from/on

136 this contributing towards a better understanding of the molecular basis of social life (Robinson

et al., 2005).

References

Anderson KE, Linksvayer TA, Smith CR (2008) The causes and consequences of genetic caste determination in ants (Hymenoptera: Formicidae). Myrmecological News 11, 119‐132. Berndt KP, Eichler W (1987) Die Pharaoameise, Monomorium pharaonis (L.) (Hym., Myrmicidae). Mitteilungen aus dem Zoologischen Museum in Berlin 63, 3 ‐186. Edwards JP (1986) The biology, economic importance, and control of the Pharaoh's Ant In: Economic Impact and Control of Social Insects (ed. Bradleigh Vinson S), pp. 257‐271. Praeger Publishers, New York. Edwards JP (1987) Caste regulation in the pharaoh's ant Monomorium pharaonis: the influence of queens on the production of new sexual forms. Physiological Entomology 12, 31‐39. Edwards JP (1991) Caste regulation in the pharaoh's ant Monomorium pharaonis: recognition and cannibalism of sexual brood by workers. Physiological Entomology 16, 263‐271. Falconer DS, Mackay TFC (1996) Introduction to Quantitative Genetics, 4 edn. Longman Group, Essex. Fuller RC, Baer CF, Travis J (2005) How and when selection experiments might actually be useful. Integrative and Comparative Biology 45, 391‐404. Hall DW, Smith IC (1953) Atypical Forms of the Wingless Worker and the Winged Female in Monomorium pharaonis (L.). (Hymenoptera: Formicidae). Evolution 7, 127‐ 135.

137 Hansen C, Spuhler K (1984) Development of the national institutes of health genetically heterogeneous rat stock. Alcoholism: Clinical and Experimental Research 8, 477‐ 479. Hill WG, Caballero A (1992) Artificial selection experiments. Annual Review of Ecology and Systematics 23, 287‐310. Hölldobler B, Wilson EO (1990) The Ants Springer, New York. Johnson BR, Linksvayer TA (2010) Deconstructing the superorganism: social physiology, groundplans, and sociogenomics. Quarterly Review of Biology 85, 57‐79. Kraaijeveld AR, Godfray HCJ (1997) Trade‐off between parasitoid resistance and larval competitive ability in Drosophila melanogaster. Nature 389, 278‐280. Linksvayer TA (2006) Direct, maternal, and sibsocial genetic effects on individual and colony traits in an ant. Evolution 60, 2552‐2561. McGlothlin JW, Moore AJ, Wolf JB, Brodie Iii ED (2010) Interacting phenotypes and the evolutionary process. III. social evolution. Evolution, no‐no. Peacock AD, Baxter AT (1949) Studies in Pharaoh's ant, Monomorium pharaonis (L.). 1. The rearing of artificial colonies. Entomol. Mon. Mag. 85, 256‐260. Peeters CP (1991) Ergatoid queens and intercastes in ants: Two distinct adult forms which look morphologically intermediate between workers and winged queens. Insectes Sociaux 38, 1‐15. Petersen‐Braun M (1975) Investigations on social organization in Pharaohs ant, Monomorium pharaonis L (Hym Formicidae) .1. Regulation of brood cycle. Insectes Sociaux 22, 269‐291. Petersen‐Braun M (1977) Investigations on social organization of Pharaohs ant Monomorium pharaonis L (Hymenoptera, Formicidae) .2. Caste determination. Insectes Sociaux 24, 303‐318. Robinson GE, Grozinger CM, Whitfield CW (2005) Sociogenomics: Social life in molecular terms. Nature Reviews Genetics 6, 257‐U216.

138 Schmidt AM, d'Ettorre P, Pedersen JS (2010) Low levels of nestmate discrimination despite high genetic differentiation in the invasive pharaoh ant. Frontiers in Zoology 7, 20. Schwander T, Lo N, Beekman M, Oldroyd BP, Keller L (2010) Nature versus nurture in social insect caste differentiation. Trends in Ecology & Evolution 25, 275‐282.

139

Summary and possible future directions

140

141

Summary and possible future directions

The work on this thesis has shed light on different aspects of pharaoh ant biology.

The pharaoh ant appears to be a very successful tramp ant species, seemingly benefitting from frequent human mediated dispersal, and interestingly apparently not experiencing any gene flow between colonies in the introduced areas where the survey took place. This is interesting because introduced or invasive ants are often characterized as merging supercolony forming, unicolonial species (Helanterä et al., 2009; Holway et al., 1998; Passera, 1994) from which one might expect the exact opposite: large, more or less panmictic population with free exchange of individuals at distances reaching far beyond those of normal colonies, something that has e.g. been seen in the most extreme case in the supercolonies formed by Argentine ants in Southern

Europe (Giraud et al., 2002). Previous investigations on pharaoh ants do indicate that the species can be tolerant of completely unrelated non‐nestmates (Schmidt et al., 2010), which might lead one to expect that two or more colonies in very close proximity might merge, but this does not appear to be the case.

What appears to be happening is that colonies may be frequently moved and that some areas, such as the one in Thailand where we conducted field work, experience very high propagule pressures and introduction numbers far exceeding those reported for other introduced ant species, for which only few separate introductions or introduction routes seem to have been the basis for the current introduced populations (Foucaud et al., 2010; Giraud et al., 2002).

142 The lack of exchange of individuals between colonies and relatively low level of genetic variation within colonies despite there being multiple queens is also interesting from an inbreeding and disease resistance perspective, as theory would predict diverse colonies to be more resistant (Sherman et al., 1998) and inbreeding depression might be expected

(Charlesworth, Charlesworth, 1987). I was interested in examining whether this apparent lack of diversity would therefore come at a price of reduced immunocompetence or “social immunity” for colonies (Cremer et al., 2007). Our results indicate that such a tradeoff may not exist, as increasing genetic diversity did not necessarily increase survival of colonies, which could lead one to speculate that lineage selection may be taking place and/or that coadapted gene complexes may in fact be broken up if crossings are made between colonies that genetic diversity may not be of great importance in this species if exposed to the kind of generalist entomopathogen we applied. Further experiments focusing on the behavioural interactions within groups of different genetic composition would be an interesting next step to test some of these hypotheses.

The finding of fairly frequent occurrence of Wolbachia in introduced populations of pharaoh ants naturally generates the question of whether Wolbachia is even really harmful to them. An exciting and obvious next step to shed some light on this phenomenon would be to for example do controlled crosses different strains of ants with and without Wolbachia infection; such crosses would enable identification of any reproductive effects or incompatabilities

(Werren et al., 2008) there may exist.

Lastly, the clear effect of colony size on caste ratios implies a high level of plasticity in response to the social environment, likely reflecting adaptations ensuring optimal survival and growth for any given colony size. It will be very interesting to see whether it is also possible to select for a social trait such as caste ratio, which, if successful, would likely be through the

143 selection of queens that produce workers with certain thresholds for raising new workers vs. queens.

144 References

Charlesworth D, Charlesworth B (1987) Inbreeding depression and its evolutionary consequences. Annual Review of Ecology and Systematics 18, 237‐268. Cremer S, Armitage SAO, Schmid‐Hempel P (2007) Social immunity. Current Biology 17, R693‐ R702. Foucaud J, Orivel J, Loiseau A, et al. (2010) Worldwide invasion by the little fire ant: routes of introduction and eco‐evolutionary pathways. Evolutionary Applications 3, 363‐374. Giraud T, Pedersen JS, Keller L (2002) Evolution of supercolonies: The Argentine ants of southern Europe. Proceedings Of The National Academy Of Sciences Of The United States Of America 99, 6075‐6079. Helanterä H, Strassmann JE, Carrillo J, Queller DC (2009) Unicolonial ants: where do they come from, what are they and where are they going? Trends in Ecology & Evolution 24, 341‐ 349. Holway DA, Suarez AV, Case TJ (1998) Loss of intraspecific aggression in the success of a widespread invasive social insect. Science 282, 949‐952. Passera L (1994) Characteristics of tramp species. In: Exotic Ants, Biology, Impact, and Control of Introduced Species (ed. Williams DF), pp. 23‐43. Westview Press, Boulder, Colorado. Schmidt AM, d'Ettorre P, Pedersen JS (2010) Low levels of nestmate discrimination despite high genetic differentiation in the invasive pharaoh ant. Frontiers in Zoology 7, 20. Sherman PW, Seeley TD, Reeve HK (1998) Parasites, pathogens, and polyandry in honey bees. American Naturalist 151, 392‐396. Werren JH, Baldo L, Clark ME (2008) Wolbachia: master manipulators of invertebrate biology. Nat Rev Micro 6, 741‐751.

145

Acknowledgements

There are many people I would like to thank….

The members of my advisory committee, Jes Søe Pedersen for advise on pop genetic analyses, Koos Boomsma for teaching me where one can cut and add while writing papers and your inspiring ability to see the interesting aspects of many different research areas, Patrizia d’Ettorre for a bit of your perspective on the world of science.

Tim Linksvayer for collaboration in the lab through many, many, many hours of various forms of ant sorting.

David Nash for always being willing to answer questions on stats, wacom or whatnot and for setting a great example in your use of various graphics and illustrations in your work.

David Hughes for many inspiring conversations.

To my office mates Volker Nehring, Elisa Bresciani and Jens Broch for being good company! …especially Volker for also being my co‐interior designer and to Luke Holman for being chatty, willing to correct my non‐British‐English, and for giving me feedback on the intro for my thesis and many other things! ‐ both of you of thanks for patiently putting up with my silly R questions 

All of my colleagues in and around the CSE that have made work possible, interesting and inspiring and fun.

Sylvia Mathiasen, both in and out of the DNA lab ‐ I dare not think how many liters of egg yolk and honey and kilos of liver you must bought & brought for the ants by now! Daniel Kronauer for DNA lab and msat development guidance. And Bettina Markussen for sorting out a lot of paperwork, bills and salaries for helpers. Sophie Armitage for your input and help with the fungi and infected ants! Thank you Susanne den Boer, Sandra Breum Andersen, Nicky Maria Bos, Matthias Fürst, Lisi Stafflinger, Henrik Hjarvard de fine Licht, Luke Holman, Volker Nehring, Sanne Nygård, Morten Schiøtt, Andras Tartally, Line Vej Ugelvig, Svjetlana Vojvodic, Irina Levinsky, Anna‐Sophie Stensgård, Susanne Fritz, Aniek Ivens, Marlene Stürup, Tim Linksvayer, Nana Hesler, Erica Ahlenfeldt, Sze Hui Yek, Jelle van Zweden, Nathalie Stroymeyt, Daniel Kronauer, Sophie Armitage, Luigi Pontieri…. for the good times we’ve had; CSE and the department would not have been the same without you.

146

I would also like to thank Naomi Pierce for her hospitality giving me the chance to be part of her inspiring lab for a little while, and thank you the group: Marco Archetti, Chris Baker, Angelica Cibrian‐Jaramillo, Mark Cornwall, Rod Eastwood, Benjamin Goldman‐Huertas, David Hughes, Ada Kalizewska, Daniel Kronauer, Petra Kubikova, Gabriel Miller, Manus Patten, Jon Sanders and Wenfei Tong; thank you all, I had a great and inspiring time at MCZ!

For making my studies possible by sending ants to me one way or the other: Ehab Abouheif, Diane Allard, Emmanuel M. Attua, Boris Baer, Tone Birkemoe, Anne‐ Geneviève Bagnères, Howard Bell, Lisbeth W. Børgesen, Cobblah–group Legon, Simon Dupont, Marie‐Julie Favé, Andre Francoeur, Tina Frisch, Aleksandra Gliniewicz, Mikko Heini, Anders Hjort‐Hansen, Armin Ionescu, Weeyawat Jaitrong, Todd Johnson, Yeo Kelo, Pitoon Kongnoo, Chow‐Yang Lee, Say‐Piau Lim, Timothy Linksvayer, Dave Lohman, Boudanath Maharajh, Ants Martin, Alexander S. Mikheyev, Jesse Czekanski‐Moir, David Oi, Fabien Ravery, Vola Razakarivony, Watana Sakchoowong, Andrew Suarez, Tsong Hong Su, Liselotte Sundström, Brian Taylor, Walter R. Tschinkel, Kazuki Tsuji, Lumi Viljakainen, Merav Vonshak, Jaitrong Weeyawat, Seiki Yamane, Sze Hui Yek

Thank you to the student helpers without whose presence and care the numerous ant colonies I have been playing around with would never have survived the past two years and who have also helped through some of the most labor‐intensive sorting and measuring, thank you Camilla, Christa, Isabel, Knud‐Brian, Kristian, Luigi, Maria, Markus, Mathilde, Nathia, Signe and Søren.

For making my field work experience in Thailand possible, pleasant and fruitful: Winanda Himaman, Chaweewan Hutacharern, Dave Lohman, Weeyawat Jaitrong, Watana Sakchoowong for collaboration, Pitoon Kongnoo (Toon) for help in the field, Tana Thongrod (Jun)

Last, but definitely not least, I would like to thank those of my friends who do not work on ants or other peculiar things at bio, my family, the people I love, for pretending that what I do is interesting and maybe even makes sense when you really think it is strange; thank you for your unwavering love and moral support. Mavsen, favsen, Rune, Trine, August, Cornelius, Lasse, Nina & Marius. Jeg elsker jer.

147

Curriculum vitae

Personal data

Name Anna Mosegaard Schmidt

Work address Centre for Social Evolution Department of Biology University of Copenhagen Universitetsparken 15, DK‐2100 Copenhagen Denmark E‐mail [email protected] Telephone 0045 23455022

Date of birth January 24th, 1979. Place of birth Copenhagen, Denmark Nationality Danish citizen.

Education

2007‐present Ph.D. student, University of Copenhagen, Denmark.

2002 ‐ 2006 M.Sc. in Biology at the University of Copenhagen.

1999 ‐ 2002 B.Sc. in Biology at the University of Copenhagen.

Scientific positions

2006 ‐ 2007 Research assistant at the Centre for Social Evolution (CSE), University of Copenhagen.

2003 Student assistant at the Danish Pest Infestation Laboratory (August – September)

148

Published manuscript not part of this thesis

Schmidt, A.M., d'Ettorre, P. and Pedersen, J.S. (2010). Low levels of nestmate discrimination despite high genetic differentiation in the invasive pharaoh ant. Frontiers in Zoology. 7, 20.

Conference and meeting attendance

2009 XII Congress of the European Society for Evolutionary Biology (ESEB), Turin, Italy, August 2009. Poster presentation: ‘Effect of genetic diversity on individual and social disease resistance in an insect society’ 2009 Nordforsk closing symposium. Reykjavik, Iceland, August 2009. 2009 15th Annual European Meeting of PhD Students in Evolutionary Biology (EMPSEB 2009), Schoorl, the Netherlands, August 2009. Oral presentation: ‘Effect of genetic diversity on disease resistance in an insect society’ 2008 EM IUSSI (European meeting, International Union for the Study of Social Insects), La Roche‐en‐Ardenne, Belgium. Oral presentation: ‘Invasive or native? Population genetics of pharaoh ants in Central‐ and Southern Thailand’ 2008 ICE (International Congress of Entomology), Durban, South Africa. Oral presentation: ‘Occurrence of Wolbachia in introduced populations of pharaoh ants: A case of enemy release?’ 2007 NordForsk meeting on The Biology of Social Insects. Tartu, Estonia, November‐ December 2007. Oral presentation: ‘Local and global dispersal and diversity of pharaoh ants (Monomorium pharaonis) – a widespread invasive species’ 2007 XI Congress of the European Society for Evolutionary Biology (ESEB). Uppsala, Sweden, August 2007. 2007 ISOBIS annual meeting, University of Copenhagen, June 2007. Oral presentation: ‘Evolutionary changes in invasive ants ‐ pharaoh ants as model’ 2006 IUSSI (International Union for the Study of Social Insects), Washington, D.C., USA. Oral presentation: ‘Global dispersal and genetic diversity in the invasive pharaoh ant, Monomorium pharaonis’ 2004 INSECTS (INtegrated Studies on Social Evolution) network, closing symposium, Helsingør, Denmark. Poster presentation: ‘Dispersal of the pharaoh ant (Monomorium pharaonis) ‐ an invasive species’

Peer reviewing experience

Insectes Sociaux Naturwissenschaften The Thai Journal of Agricultural Science

149

Teaching experience

2009 Teaching Assistant, Invasion Biology, University of Copenhagen, graduate level course. Responsible for two lectures.

2008 Teaching Assistant, Invasion Biology, University of Copenhagen, graduate level course.

2006 Supervision of B.Sc. project by Erica Juel Ahrenfeldt ‘Variation in morphology and behaviour between different lineages of the Pharaoh ant Monomorium pharaonis’ (co‐supervised with J.S. Pedersen).

Field work experience and external stay

2009 Two month stay at Harvard University, Cambridge, USA, hosted by Naomi Pierce.

2008 Field work for my Ph.D., collecting primarily house infecting Monomorium ants in Central and Southern Thailand (5 weeks).

2007 Field assistant for CSE researcher Patrizia d’Ettorre, collecting Camponotus ants in Italy (8days).

2007 Field assistant for CSE researcher David Hughes in Trang, Thailand, investigating interactions between the entomopathogenic Cordyceps fungi and Camponotus ants (1 month).

2005 Field assistant for CSE researchers Sophie Armitage and Michael Poulsen in Gamboa, Panama, working on leaf‐cutting ants (1 month).

2004 Field work in Ghana for my master’s project on pharaoh ants (Monomorium pharaonis), collecting ants in Southern Ghana (8days).

2003 Field assistant for CSE researcher David Nash on Læsø, Denmark, on the MacMan project, mark‐recapture of Maculinea butterflies (10 days).

150

PhD courses and workshops attended

2010 Evolutionary Genetic Approaches to Study Social Evolution, University of Copenhagen, Denmark

2008 Pathogens in Social Insects, Uppsala, Sweden

2008 Grant Writing Course, Tvärminna, Finland

2008 The Biology of Social Insects, University of Copenhagen, Denmark

2007 Insect Pathology, University of Copenhagen (Life), Denmark

2007 The Ant Course, Portal, Arizona (California Academy of Sciences), USA

2007 Evolutionary Biology of Communication, University of Copenhagen, Denmark

2007 Molecular marker analysis of plant population structure and processes, University of Copenhagen (Life), Denmark

Languages

Danish mother tongue, fluent in English, basic‐intermediate German, rudimentary Russian.

151 Monomorium in the lab

Beauveria growing from dead worker

Monomorium mug shot; ‐for head width measurements

Queen, intercaste and worker Colony with gynes and workers

Worker pupae at different ages, the darkest is about to eclose

152 Mating of newly reared sexuals

Colony with newly produced gynes and gyne pupae after old queens were removed

Mating from one of the controlled crosses

153 A typical field site in Thailand

Large M. pharaonis colony found under laundry pile

Surrounding, highly disturbed area

154 Worker ant carrying a large sexual larva

155