Evolutionary Studies of Dawson's Burrowing

MOLECULAR ECOLOGY OF DAWSON’S BURROWING BEE Amegilla dawsoni (HYMENOPTERA: ANTHOPHORINI)

Maxine Beveridge (BSc Hons. PG Dip.)

School of Animal Biology

The University of Western Australia

This thesis is presented for the degree of Master of Science Of The University of Western Australia 2006.

SUMMARY

In the last two decades, the use of microsatellites has revolutionized the study of ecology and evolution. Microsatellites, or short tandem repeats (STRs), are stretches of DNA repeats, 1 to 5 nucleotides long, where the number of repeats varies between individuals. They are co-dominant, highly variable, neutral markers, and are inherited in a Mendelian fashion. Microsatellite loci were isolated from Dawson's burrowing bee, Amegilla dawsoni, a large, fast-flying solitary nesting bee endemic to the arid zone of Western Australia. Twelve polymorphic loci were found with an observed number of alleles ranging from two to 24 and observed heterozygosities between 0.17 and 0.85. These loci were used to examine two aspects of this bee’s molecular ecology; its population structure and mating system. The population structure of Dawson’s burrowing bee was examined using the 8 most variable microsatellite loci. Adult female bees were collected from 13 populations across the species range. The mean number of alleles per locus ranged from 4 to 38 and expected heterozygosity was uniformly high with a mean of 0.602. Pairwise comparisons of FST among all 13 populations ranged from 0.0071 to 0.0122 with only one significant estimate and an overall FST of 0.001. The entire sample collection was in Hardy-Weinberg equilibrium and there was no evidence of inbreeding with a mean

FIS of 0.010. The mating and nesting behaviour of this bee suggests that gene flow would be limited by monandry and the fact that almost 90% of females mate immediately on emergence. Nevertheless there is obviously sufficient gene flow to maintain panmixia and I suggest that this results from infrequent and unreliable rainfall in the species range, which causes the bees to congregate at limited food resources, allowing a small number of unmated females from one emergence site to come into contact with males from another population. In addition, when drought eliminates food resources near an emergence site, the whole population may move elsewhere, increasing gene flow across the species range. From behavioural data and observations, Dawson’s burrowing bee appears to be monandrous. The same microsatellite markers were used to analyse the genotypes of offspring from individual nests to determine the number of effective mates for each female. From these data it was determined that females almost certainly mate only once, which is consistent with male reproductive tactics that include protandry and intense male-male competition for access to virgin females. The molecular data were

also used to show that the nesting female is the mother of all her offspring and that brood parasitism is unlikely in this species. The data indicate that females make daughters at the beginning of the season followed by large sons in the middle, and then small sons at the end. Females often place one brood cell directly above another. The distribution of sex and morph in these doublets follows a pattern with most containing a female on the bottom and a minor male on the top, followed by almost equal numbers of female on top of female and minor male on top of major male. This pattern is likely favoured by emergence patterns, with males emerging before females and minor males emerging before major males. I suggest that although minor males have low reproductive success, their production may nonetheless be beneficial in that minor males open up emergence tunnels for their larger and reproductively more valuable siblings. In addition, minor males may represent the ‘best of a bad job’ provisioning tactic arising from changes in the costs to nesting females of gathering brood provisions over the course of the flight season. This thesis demonstrates that microsatellites can be used to answer many questions regarding the molecular ecology of a species from the behaviour of the bees on a population scale to the mating behaviour of individual bees and how they allocate resources for the next generation. Many other aspects of the bee’s ecology could also be examined now that suitable molecular markers exist.

TABLE OF CONTENTS

SUMMARY i TABLE OF CONTENTS iii ACKNOWLEDGEMENTS v PUBLICATIONS vi

CHAPTER 1: General Introduction 1.1. Molecular markers and their applications 2 1.2. Dawson’s burrowing bee – a natural history 3 1.3. Population genetics of Dawson’s burrowing bee 4 1.4. Genetic breeding system of Dawson’s burrowing bee 5 1.5. References 5

CHAPTER 2: Development of microsatellite loci for Dawson's burrowing bee (Amegilla dawsoni) and their cross-utility in other Amegilla species. 2.1. Abstract 10 2.2. Introduction 11 2.3. Materials and methods 11 2.4. Results and discussion 12 2.5. References 13

CHAPTER 3: Panmixia: an example from Dawson’s burrowing bee (Amegilla dawsoni) (Hymenoptera: Anthophorini) 3.1 Abstract 18 3.2 Introduction 19 3.3 Materials and methods 20 3.4 Results 22 3.5 Discussion 23 3.6 References 26

iii

CHAPTER 4: Genetic breeding system and investment patterns within nests of Dawson's burrowing bee (Amegilla dawsoni) (Hymenoptera: Anthophorini) 4.1. Abstract 34 4.2. Introduction 35 4.3. Materials and methods 37 4.4. Results 40 4.5. Discussion 44 4.6. References 46

CHAPTER 5: General Discussion 5.1. Microsatellites as a tool 57 5.2. Molecular markers of the future 58 5.3. References 60

ACKNOWLEDGEMENTS

Firstly, I would like to thank Professor Leigh Simmons for his help, support and encouragement over the past 4 years. I would also like to thank Professor John Alcock and his wife Sue for making those field trips to Carnarvon and beyond so much fun. Thanks must also go to Charlotta Kvarnemo for helping in the onerous task of excavating bee nests. I must also thank Melissa Bell for supplying two species of bee, Amegilla holmesi and Amegilla bombiformis for microsatellite evaluation and Kings Park and Botanical Gardens for generously allowed access to their sequencer.

In the School of Animal Biology, UWA, thanks must go to Jason Kennington for helping me wade through population genetics software packages and Mike Johnson for his comments on draft manuscripts. I must also thank Blair Parsons for producing Figure 3.1. I would also like to thank all the members of the Centre for Evolutionary Biology, in particular Dr Francisco García-González for all his help and support over the last 4 years and Renée Firman for keeping me entertained.

Finally, I would like to thank Andrew for his unwavering support and love.

PUBLICATIONS

This thesis is submitted as a series of discrete papers. The following papers are either published or in press.

I. Beveridge, M. & Simmons, L. W. (2004) Microsatellite loci for Dawson's burrowing bee (Amegilla dawsoni) and their cross-utility in other Amegilla species. Molecular Ecology Notes 4, 379-381.

II. Beveridge, M. & Simmons, L. W. (2006) Panmixia: an example from Dawson's burrowing bee (Amegilla dawsoni) (Hymenoptera: Anthophorini). Molecular Ecology 15, 951-957.

III. Beveridge, M., Simmons, L. W. & Alcock, J. (2006) Genetic breeding system and investment patterns within nests of Dawson's burrowing bee (Amegilla dawsoni) (Hymenoptera: Anthophorini). Molecular Ecology in press.

The first two papers were written in collaboration with Leigh Simmons and the third with the addition of John Alcock. In all papers, I was responsible for the experimental design, data collection, analysis of results and writing. Leigh Simmons contributed to the experimental design, analysis of results and writing. John Alcock contributed to the experimental design and writing. I was the main contributor in each paper (estimated as % contribution: I, MB = 95%, LWS = 5%; II, MB = 75%, LWS = 25%; III, MB = 75%, LWS = 20%, JA = 5%)

CHAPTER 1

GENERAL INTRODUCTION

1.1 Molecular markers and their applications

Over the last two decades, molecular markers have become invaluable in answering the questions raised by ecological and evolutionary studies (Rosenbaum & Deinard 1998). Several markers have been utilised including allozymes, restriction fragment length polymorphisms (RFLPs), multi-locus and single-locus minisatellites, and randomly amplified polymorphic DNA (RAPDs) (Queller et al. 1993). However, each of these methods has disadvantages and they have largely been superseded by microsatellites. These are highly variable, neutral, co-dominant markers inherited in a Mendelian fashion (Jarne & Lagoda 1996). They consist of stretches of DNA repeats, 1 to 5 nucleotides long, where the number of repeats varies between individuals. This variability is so high that when using several microsatellite loci, even closely related organisms can be individually identified. Since their introduction, microsatellites have been used in a wide range of applications, the first of which is in behavioural ecology. Male mating success was traditionally estimated by behavioural observations in the field but this has obvious limitations. In their study of pilot whales, Amos et al. (1993) used microsatellites to uncover a completely novel mating system which could not have been discovered by observation alone. There has been a long held view that the majority of bird species were monogamous but molecular markers have shown that 86% of passerine bird species show extra-pair paternity (Griffiths et al. 2002). Statistical techniques have also improved to the point where the number of potential fathers for a set of offspring can be estimated without their putative fathers’ genotypes being sampled (Bretman & Tregenza 2005). In the field of evolutionary biology, microsatellites continue to be an invaluable tool for analysing the current questions such as why females mate polyandrously (Simmons 2005). Molecular markers are the ideal choice for testing the genetic incompatibility hypothesis (Bretman et al. 2004; Jennions et al. 2004). Patterns of sperm use by multiply mated females can also be analysed (Simmons & Achmann 2000). The ability of microsatellites to distinguish between closely related individuals has also been extremely useful in the study of social insects and the evolution of sociality (Boomsma & Ratnieks 1996; Page 1986). The study of population genetics has been possible for many years using allozymes which provide the co-dominant information required to determine parameters

such as Hardy-Weinberg equilibrium, linkage disequilibrium and effective population size (Schlotterer & Pemberton 1998). Microsatellites also have co-dominant alleles but are much more variable and less likely to be under selection. In the area of conservation genetics, the study of endangered species, traditionally difficult due to lack of genetic variation, has now been made possible with the use of highly variable microsatellites (Whitehouse & Harley 2001). With the advent of automated fluorescent sequencers capable of high throughput, many loci and hundreds of samples can be analysed in a short space of time allowing the use of these markers to become more widespread. The applications mentioned here are just a few of those that utilise these versatile molecular markers. In this thesis I have used the most up-to-date methods for isolating microsatellites from a species of solitary bee (Zane et al. 2002) (Chapter 2). I then use these markers to examine the population genetic structure of the bee (Chapter 3), and to determine the genetic mating system and patterns of offspring sex allocation (Chapter 4).

1.2 Dawson’s burrowing bee – a natural history

Dawson’s burrowing bee (Amegilla dawsoni) is a large, fast-flying bee of the Anthophorini tribe and is endemic to the arid zone of Western Australia (Houston 1991). The females build solitary nests often in large aggregations in the hard-packed soils of clay pans. The species is univoltine with a flight season lasting only a few weeks in late winter and early spring. Nesting is carried out entirely by the female who digs a vertical tunnel 15-35cm deep. At the bottom of the tunnel, she then constructs a cell which is provisioned with pollen and nectar from nearby flowering plants, followed by oviposition. The cell is then sealed and another constructed. On completion of an average of 7 cells, the tunnel is filled with soil and the nest is abandoned. This species is unusual in that whilst female size follows a normal distribution, male size is distinctly dimorphic with large ‘major’ males, the same size as the females, and smaller ‘minor’ males. This male dimorphism was first observed in the adult bees by Houston (1991) and has since been observed in brood cell dimension (Tomkins et al. 2001) and mature larval weights (Alcock et al. 2005). This male size dimorphism gives rise to alternative male mating tactics (Alcock 1997a). Males emerge, on average, 15 days before females showing marked protandry (Alcock 1997b). Major males patrol the emergence area for newly emerging virgin females which they detect using cuticular

hydrocarbons (Simmons et al. 2003). When a major male detects an emerging female, he waits by the entrance of the emergence tunnel and defends it against other males. When the female eventually reaches the surface, competition between males is fierce and the largest male usually wins, mating with the female immediately, after which she appears to become unreceptive (Simmons et al. 2000). Small minor males patrol the emergence area only when the density of large males is low. When competition becomes too great, they adopt an alternative mating tactic of patrolling the periphery of the emergence area and the host flowering plants (Alcock 1997a). However, as only 13% of females leave the emergence area without mating, opportunities for these minor males to mate appear to be limited. In spite of this fact, minor males represent 80% of the males produced each season and the explanation for the overproduction of minor males remains an evolutionary puzzle (Alcock 1996a). Evidence to date suggests that provisioning cells becomes more difficult as the season progresses, with females becoming wing-worn, and competition for food increasing as the host plants produce less pollen and nectar (Tomkins et al. 2001). Due to these limitations females may be forced into making “the-best-of-a-bad- job” by producing large offspring at the beginning of the season when conditions are favourable and reverting to small offspring (minor males) at the end of the season when conditions are harsher (Alcock et al. 2005).

1.3 Population Genetics of Dawson’s burrowing bee

Genetic differentiation between geographic localities is shown, to some degree, in populations of most species (Avise 1994). In a comparative study of over 300 animal species, the more mobile species showed less population structure than sedentary organisms (Ward et al. 1992) but the complete lack of genetic structure or panmixia is, in fact, relatively rare (Hoarau et al. 2002; Ridgway et al. 2001). From previous studies of the mating and nesting behaviour of Dawson’s burrowing bee, it appears that nesting sites are used for many years. There are no data on whether males and females return to the place of emergence to breed although males do exhibit a high degree of site fidelity once they begin patrolling an emergence site (Alcock 1996b). If bees routinely use their natal nesting sites to breed then the level of genetic differentiation might be expected to be high with little gene flow between populations. This question is explored in Chapter 3.

1.4 Genetic Breeding System of Dawson’s burrowing bee

The study of the genetic breeding system of a species can help to explain how evolutionary processes have shaped its morphology, life history, and behaviour. Mating behaviour has been studied extensively in the insects where, despite potential costs to females, many species mate polyandrously (Arnqvist & Nilsson 2000). In the Hymenoptera, most species are polyandrous (Page 1986) but in the social Hymenoptera the opposite is the case, with monandry or very low levels of polyandry being the norm (Strassman 2001). The exceptions are 5 groups that exhibit very high levels of polyandry (leaf-cutting ants, seed-harvesting ants, army ants, honey bees and higher vespine wasps (Brown & Schmid-Hempel 2003; Kronauer et al. 2004). How polyandry has evolved in the social insects has been the focus of much recent research (Peters et al. 1999) but few studies have been made of the mating systems of the solitary Hymenoptera (Zavodna et al. 2005). In Dawson’s burrowing bee, observations of the mating behaviour suggest that females are monandrous as most mate immediately on emergence and become unreceptive shortly afterwards (Simmons et al. 2000). However, observational data can often be misleading, so another method should be employed to give a reliable result (Page 1986). In Chapter 4 I use the microsatellites developed in Chapter 2 to determine if this species is monandrous, and also explore levels of brood parasitism. Molecular markers can also be used to determine the sex of developing offspring, and in Chapter 4 I also look at the seasonal changes in investment patterns of individual nesting females to male and female offspring. Many aspects of the ecology of Dawson’s burrowing bee are still to be researched and the use of molecular markers makes some of these studies possible.

1.5. References

Alcock J (1996b) Site fidelity and homing ability of males of Dawson's burrowing bee (Amegilla dawsoni) (Apidae, Anthophorini). Journal of the Kansas Entomological Society 69, 182-190. Alcock J (1997a) Competition from large males and the alternative mating tactics of small males of Dawson's burrowing bee (Amegilla dawsoni) (Apidae, Apinae, Anthophorini). Journal of Insect Behavior 10, 99-113. Alcock J (1997b) Small males emerge earlier than large males in Dawson's burrowing bee (Amegilla dawsoni) (Hymenoptera: Anthophorini). Journal of Zoology 242, 453-462. Alcock J, Simmons LW, Beveridge M (2005) Seasonal change in offspring sex and size in Dawson's burrowing bees (Amegilla dawsoni) (Hymenoptera: Anthorphorini). Ecological Entomology 30, 247-254. Amos B, Schlotterer C, Tautz D (1993) Social structure of pilot whales revealed by analytical DNA profiling. Science 260, 670-672. Arnqvist G, Nilsson T (2000) The evolution of polyandry: multiple mating and female fitness in insects. Animal Behaviour 60, 145-164. Avise JC (1994) Molecular Markers, Natural History and Evolution, 1 edn. Chapman and Hall, New York. Boomsma JJ, Ratnieks FL (1996) Paternity in social Hymenoptera. Philosophical Transactions of the Royal Society of London B 351, 947-975. Bretman A, Tregenza T (2005) Measuring polyandry in wild populations: a case study using promiscuous crickets. Molecular Ecology 14, 2169-2179. Bretman A, Wedell N, Tregenza T (2004) Molecular evidence of post-copulatory inbreeding avoidance in the field cricket Gryllus bimaculatus. Proceedings of the Royal Society of London B 271, 159-164. Brown MJF, Schmid-Hempel P (2003) The evolution of female multiple mating in social Hymenoptera. Evolution 57, 2067-2081. Griffiths SC, Owens IPF, Thuman KA (2002) Extra pair paternity in birds: a review of interspecific variation and adaptive function. Molecular Ecology 11, 2195-2212. Hoarau G, Rijnsdorp AD, Van der Veer HW, Stam WT, Olsen JL (2002) Population structure of plaice (Pleuronectes platessa L.) in northern Europe: microsatellites revealed large-scale spatial and temporal homogeneity. Molecular Ecology 11, 1165-1176.

Houston TF (1991) Ecology and behaviour of the bee Amegilla (Asaropoda) dawsoni (Rayment) with notes on a related species (Hymenoptera: Anthophoridae). Records of the Western Australian Museum 15, 535-553. Jarne P, Lagoda PJL (1996) Microsatellites, from molecules to populations and back. Trends in Ecology and Evolution 11, 424-429. Jennions MD, Hunt J, Graham R, Brooks R (2004) No evidence for inbreeding avoidance through postcopulatory mechanisms in the black field cricket Teleogryllus commodus. Evolution 58, 92-97. Kronauer DJC, Schoning C, Pedersen JS, Boomsma JJ, Dadau J (2004) Extreme queen- mating frequency and colony fission in African army ants. Molecular Ecology 13, 2381-2388 Page RE (1986) Sperm utilization in social insects. Annual Review of Entomology 31, 297-320. Peters JM, Queller DC, Imperatriz-Fonseca VL, Roubik DW, Strassman JE (1999) Mate number, kin selection and social conflicts in stingless bees and honeybees. Proceedings of the Royal Society of London B 266, 379-384. Queller DC, Strassman JE, Hughes CR (1993) Microsatellites and Kinship. Trends in Ecology and Evolution 8, 285-288. Ridgway T, Hoegh-Guldberg O, Ayre DJ (2001) Panmixia in Pocillopora verrucosa from South Africa. Marine Biology 139, 175-181. Rosenbaum HC, Deinard AS (1998) Caution before claim: an overview of microsatellite analysis in ecology and evolutionary biology. In: Molecular Approaches to Ecology and Evolution (eds. DeSalle R, Schierwater B), pp. 87- 106. Birkhauser Verlag, Basel. Schlotterer C, Pemberton J (1998) The use of microsatellites for genetic analysis of natural populations - a critical review. In: Molecular Approaches to Ecology and Evolution (eds. DeSalle R, Schierwater B), pp. 71-86. Birkhauser Verlag, Basel. Simmons L, Alcock J, Reeder A (2003) The role of cuticular hydrocarbons in male attraction and repulsion by female Dawson's burrowing bee, Amegilla dawsoni. Animal Behaviour 66, 677-685. Simmons LW (2005) The Evolution of Polyandry: Sperm Competition, Sperm Selection, and Offspring Viability. Annual Review of Ecology, Evolution, and Systematics 36, 125-146.

Simmons LW, Achmann R (2000) Microsatellite analysis of sperm-use patterns in the bushcricket Requena Verticalis. Evolution 54, 942-952. Simmons LW, Tomkins JL, Alcock J (2000) Can minor males of Dawson's burrowing bee, Amegilla dawsoni (Hymenoptera: Anthophorini) compensate for reduced access to virgin females through sperm competition? Behavioral Ecology 11, 319-325. Strassman J (2001) The rarity of multiple mating by females in the social Hymenoptera. Insectes Sociaux 48, 1-13. Tomkins JL, Simmons LW, Alcock J (2001) Brood-provisioning strategies in Dawson's burrowing bee, Amegilla dawsoni (Hymenoptera: Anthophorini). Behavioural Ecology and Sociobiology 50, 81-89. Ward RD, Skibinski DOF, Woodwark M (1992) Protein heterozygosity, protein structure, and taxonomic differentiation. Evolutionary Biology 26, 73-159. Whitehouse AM, Harley EH (2001) Post-bottleneck genetic diversity of elephant populations in South Africa, revealed using microsatellite analysis. Molecular Ecology 10, 2139-2149. Zane L, Bargelloni L, Patarnello T (2002) Strategies for microsatellite isolation: a review. Molecular Ecology 11, 1-16. Zavodna M, Compton SG, Raja S, Gilmartin PM, van Damme JMM (2005) Do fig wasps produce mixed paternity clutches? Journal of Insect Behaviour 18, 351- 362.

CHAPTER 2

Development of microsatellite loci for Dawson's burrowing bee (Amegilla dawsoni) and their cross-utility in other Amegilla species.

2.1. Abstract

Microsatellite loci were isolated from Dawson's burrowing bee, Amegilla dawsoni, a species endemic to Western Australia. Twelve polymorphic loci were found with an observed number of alleles ranging from two to 24 and observed heterozygosities between 0.17 and 0.85. These 12 loci were tested for amplification in three additional species of Amegilla. The loci will be used for sex determination and the examination of mating frequency in this species.

2.2. Introduction

Dawson's burrowing bee Amegilla dawsoni is a large solitary nesting bee endemic to Western Australia (Houston 1991). Females nest in large aggregations. Male size is bimodal with large (major) males and small (minor) males adopting alternative male mating tactics (Alcock 1997). Major males patrol emergence sites and fight for access to newly emerging virgin females. Minor males adopt an alternative mating tactic of searching the periphery of the emergence site and flowering plants where females go to forage. Almost 90% of virgin females mate with majors immediately on emergence. Nevertheless 66-80% of the male population are minor males and several studies have attempted to explain why the dimorphism persists (Alcock 1996). One hypothesis suggests that minor males could gain fitness by mating with non-virgin females and engaging in sperm competition. In their behavioural observations, Simmons et al. (2000) found that females would not copulate more than once, suggesting that this species is monandrous. Behavioural data can be equivocal, and genetic data are required to draw firm conclusions regarding the genetic mating system of a species (Boomsma & Ratnieks 1996) hence the need for microsatellite loci.

2.3. Materials and methods

Microsatellite containing sequences were isolated using the FIASCO (Fast Isolation by AFLP of Sequences COntaining repeats) method (Zane et al. 2002). DNA was extracted from a single female A. dawsoni using the MasterPure™ Complete DNA purification kit (Epicentre) followed by digestion with MseI restriction enzyme (Invitrogen). AFLP adaptors (Invitrogen) were ligated onto the digested DNA and the digestion-ligation mixture diluted (1:10) before amplification with AFLP adaptor specific primer MseI-A (Invitrogen). This allowed amplfication of fragments flanked by MseI-A sites providing they were an appropriate size to be amplified by polymerase chain reaction (PCR). The partial genomic library was then enriched for microsatellite containing fragments as described by Glenn et al. (2000). Two mixtures of biotinylated oligos were used: a low temperature (55°C) set containing (AAC)8, (ATC)8, (AAG)8,

(AAT)8 and (ACT)8 (Invitrogen) and a medium temperature (65°C) set containing

(AC)12 and (AG)12 (Geneworks). The resulting hybrids were captured using agarose- avidin D beads (Vector Laboratories) and the DNA recovered by denaturation. Another

PCR using MseI-A primers was conducted to amplify the enriched DNA which was then cloned using a TOPO TA cloning kit (Invitrogen). 192 recombinant clones were picked from the trinucleotide enrichment and 96 from the dinucleotide enrichment. The insert in each clone was amplified by PCR using M13 forward and M13 reverse primers present in the vector and the resulting PCR products screened for the presence of microsatellites using a non-radioactive technique outlined in Glenn et al. (2000). The PCR products were also run on an agarose gel to estimate insert size. Following PCR clean-up using EXOSAP-IT (Amersham Pharmacia Biotech), 40 of the positive trinucleotide clones and 48 of the positive dinucleotide clones were sequenced using BigDye Terminator cycle sequencing kit (Applied Biosystems). Primers were designed for 10 of the trinucleotide clones and 6 of the dinucleotide clones using a web-based primer design service called Web Primer (2003). The forward primers were labelled with either 6-FAM or HEX (Geneworks) and tested for amplification and polymorphism using DNA from 48 female bees collected from Billabalong, Western Australia (S 27° 22.144', E 115° 52.998') in 2001. Each 10µl PCR reaction contained 1x GeneAmp PCR Buffer II (10mM Tris-HCl pH 8.3, 50mM KCl),

1.5mM MgCl2 (Applied Biosystems), 200µM of each dNTP (Invitrogen), 250nM of the forward primer (labelled and unlabelled in a ratio of 1:10) and 250nM of the reverse primer (Geneworks), 0.5units of AmpliTaq DNA polymerase (Applied Biosystems) and ~10ng of DNA. PCR amplification was performed in a PTC-0200 DNA Engine (Geneworks) with cycling conditions as follows: 94°C for 1minute, then 30 cycles of 94°C for 1 minute, 56°C for 1 minute and 72°C for 1 minute, and finally 72°C for 10 minutes. The PCR products were analysed on an ABI377 Sequencer and sized using Genescan-350 ROX internal size standard and Genescan software (version 3.1).

2.4. Results and Discussion

Of the 16 loci characterized, 12 were found to be polymorphic with between 2 and 24 alleles when tested with 48 A. dawsoni females. Primer sequences, allele size range and numbers of alleles for the 12 polymorphic loci are given in Table 2.1. The observed and expected heterozygosities for each locus were calculated using Microsatellite Analyzer v2.65 (Dieringer and Schlotterer 2003) and are shown in Table 1. Each locus was tested for Hardy-Weinberg equilibrium using Genepop v3.3 (Raymond and Rousset 1995), only Beedi11 deviated significantly (P=0.03).

The utility of all twelve A. dawsoni loci was assessed in three other species of Amegilla. Ten specimens of A. paraclava (Brooks 1993) were collected from Mount Sandiman, Western Australia (S 24° 23.816', E 115° 25.440'). Two specimens of A. bombiformis were collected from Willoughby, New South Wales. Two specimens of A. holmesi were collected from Kurmond, one from Sydney Botanical Gardens, New South Wales and one from an unknown site. The results of the cross-amplification study are shown in Table 2.2. PCR conditions were identical to those for A. dawsoni. The species showing the highest number of amplifying loci and the highest number of alleles per loci was A. paraclava suggesting that the cross-utility will be greatest for this species.

2.5. References

Alcock, J (1996) Provisional rejection of three alternative hypotheses on the maintenance of a size dichotomy in males of Dawson's burrowing bee, Amegilla dawsoni (Apidae, Apinae, Anthophorini). Behavioral Ecology and Sociobiology, 39, 181-188. Alcock, J (1997) Competition from large males and the alternative mating tactics of small males of Dawson's burrowing bee (Amegilla dawsoni) (Apidae, Apinae, Anthophorini). Journal of Insect Behavior, 10, 99-113. Boomsma JJ, Ratnieks FLW (1996) Paternity in eusocial Hymenoptera. Philosophical Transactions of the Royal Society of London B, 351, 947-975.

Brooks RW (1993) A new Amegilla (Hymenoptera: Anthophoridae) from Western Australia. Records of the Western Australian Museum, 16, 279-282. Dieringer D, Schlotterer C (2003) Microsatellite Analyser (MSA): a platform independent analysis tool for large microsatellite data sets. Molecular Ecology Notes, 3, 167-169. Glenn TC, Cary T, Dust M, Hauswaldt S, Prince K, Ramsdell C, Shute I (2000) Microsatellite Isolation. http://www.uga.edu/srel/DNA_Lab/Msat_Easy_Isolation_2000.rtf Houston TF (1991) Ecology and behaviour of the bee Amegilla (Asaropoda) dawsoni (Rayment) with notes on a related species (Hymenoptera: Anthophoridae). Records of the Western Australian Museum, 15, 535-553.

Raymond M, Rousset F (1995) GENEPOP (Version 1.2): Population genetics software or exact tests and ecumenicism. Journal of Heredity, 86, 248-249. Simmons LW, Tomkins JL, Alcock J (2000) Can minor males of Dawson's burrowing bee, Amegilla dawsoni (Hymenoptera: Anthophorini) compensate for reduced access to virgin females through sperm competition? Behavioural Ecology, 11, 319-325. Web Primer (2003) Web Primer: DNA and Purpose Entry. http://genome- www2.stanford.edu/cgi-bin/SGD/web-primer Zane L, Bargelloni L, Patarnello T (2002) Strategies for microsatellite isolation: a review. Molecular Ecology, 11, 1-16.

Table 2.1. Characterization of microsatellite loci in Dawson's burrowing bee (Amegilla dawsoni). The annealing temperature (Tm) is in °C and MgCl2 concentration is in mM. The number of individuals tested (n), number of alleles (NA), allele size range in base pairs (bp), observed (HO) and expected

(HE) heterozygosity are listed for each locus.

GenBank Locus Repeat motif in clone accession Primer sequence (5'-3') n Tm [MgCl2] NA bp HO HE number

F:GGCAGTCACCATTTTGAAACG Beetri2 (GAA) CAA(GAA) AY521279 48 56 1.5 5 159-171 0.42 0.37 2 11 R:ACAGGCACGAATAACCGAGAT F:GGACCTTTTGTCGATTCA Beetri12 (ACT) AY521280 48 56 1.5 3 60-66 0.42 0.45 6 R:GAAGTCTAGAGGTAATGCCTG F:CGTGGGTAAATGGTTCAGGAA Beetri60 (TTC) GTCTTC AY521281 48 56 1.5 2 122-125 0.33 0.33 5 R:ATGCAGATCGCTAATACCGCA F:ACAAATGGGTGAAAGTGCGA Beetri94 (GAA) AAA(GAA) AY521282 48 56 1.5 2 167-179 0.17 0.15 4 3 R:TAACGTTTCAACGCTTGACG F:ATCTTCCGTGTAAATCGGCAG Beetri116 (GAA) AY521283 48 56 1.5 8 101-125 0.65 0.66 8 R:TACCTCTTTCCTCGGCATCAT F:ATGCAGATCGCTAATACCGCA Beetri132 GAAGAC(GAA) AY521275 48 56 1.5 2 121-124 0.33 0.33 5 R:TCGTGGGTAAATGGTTCAGGA F:TCGTAACGTGGCAATTCCAT Beetri143 (GAA) GAG(GAA) GTA(GAA) AY521276 48 56 1.5 3 193-199 0.46 0.48 2 4 5 R:ATCGCTCCAATAAGCAACAGC F:GCAGAGAAGCCATAATTTCCT Beetri152 (GAA) GCA(GAA) AY521277 48 56 1.5 2 152-155 0.42 0.49 2 6 R:CACTCTCGTCCATGATATTCG F:AACCTATATGCCTCGCGTCTC Beedi7 (GA) AY521272 48 56 1.5 6 153-163 0.69 0.63 14 R:CGACACATACGCCCCTTTCA F:GCAATTTTGTTCGACGCTTA Beedi11 (CT) AY521273 48 56 1.5 24 108-168 0.81 0.92 19 R:AAGCAACCTGCAAAGTGGAA F:CGTCGCGTGGAATATCATTT Beedi12 (GA) AY521274 48 56 1.5 12 116-142 0.85 0.72 11 R:ACCAACTGAATGCTGTCTGCT F:ATGAACAATGCAGCAACGC Beedi40 (CA) AY521278 48 56 1.5 5 166-178 0.58 0.60 12 R:AAAGGTGAGGGGATTCGATAA

Table 2.2. Cross-amplification of 12 Amegilla dawsoni microsatellite loci in three other species of Amegilla. The symbol (-) denotes no amplification product and n is the number of individuals screened.

Amegilla paraclava Amegilla bombiformis Amegilla holmesi (n=4) (n=10) (n=2)

Number of Allelic Number of Allelic Number of Allelic Locus alleles size range alleles size range alleles size range Beetri2 3 153-159 - 1 150 Beetri12 3 63-69 1 66 1 60 Beetri60 2 122-125 1 124 - Beetri94 1 179 1 179 - Beetri116 5 107-128 1 113 2 113-116 Beetri132 2 121-124 1 124 - Beetri143 3 196-205 1 188 4 205-217 Beetri152 1 152 1 158 1 149 Beedi7 5 155-163 - - Beedi11 5 114-136 1 112 1 112 Beedi12 10 126-154 4 122-132 2 124-126 Beedi40 8 166-188 1 166 1 170

CHAPTER 3

Panmixia: an example from Dawson’s burrowing bee (Amegilla dawsoni) (Hymenoptera: Anthophorini)

3.1. Abstract

Dawson’s burrowing bee is a large, fast-flying solitary nesting bee endemic to the arid zone of Western Australia. In this study the population structure of the species was examined with molecular markers. Using 8 microsatellite loci we genotyped 531 adult female bees collected from 13 populations of Dawson’s burrowing bee, Amegilla dawsoni, across the species range. The mean number of alleles per locus ranged from 4 to 38 and expected heterozygosity was uniformly high with a mean of 0.602. Pairwise comparisons of FST among all 13 populations ranged from 0.0071 to 0.0122 with only one significant estimate and an overall FST of 0.001. The entire sample collection was in Hardy-Weinberg equilibrium and there was no evidence of inbreeding with a mean

3.2. Introduction

The extent of genetic structure is dependent upon the relative strengths of the forces generating differentiation (mutation, genetic drift and natural selection) to those reducing differentiation (gene flow and natural selection) (Slatkin 1987). Methods for estimating gene flow are numerous and fit into two main categories; direct methods that monitor gene flow in the present, and indirect methods that infer gene flow that has occurred in the past. Some degree of genetic differentiation between geographic localities is exhibited in populations of most species (Avise 1994). A comparative summary of population structure for over 300 animal species showed that the more mobile insects and birds showed significantly lower levels of population structure than sedentary organisms (Ward et al. 1992). Nevertheless, many studies of birds and insects have shown significant population differentiation (Abbott & Double 2003; Paar et al. 2004; Shao et al. 2004; Widmer & Schmid-Hempel 1999). Panmixia, or the complete lack of any genetic differentiation across the entire range of a species, has been observed but is very rare (Hoarau et al. 2002; Ridgway et al. 2001). One potential example of panmixia was thought to be provided by the European eel, a species in which all adults migrate to the Sargasso Sea to reproduce, thus apparently comprising a single randomly mating population (Tesch 1977). However, using microsatellite genetic markers, global genetic differentiation has been detected (Wirth & Bernatchez 2001) and the hypothesis of panmixia has now been refuted. Dawson’s burrowing bee is a large, fast-flying bee of the Anthophorini tribe. It is endemic to the arid zone of Western Australia, well known for its limited and ephemeral rainfall. For a species dependent on flowering plants for its survival, this would seem to be an inhospitable environment. Nonetheless, it has been reported as being locally abundant with a widespread distribution (Houston 1991). The species has one flight season per year in late winter-early spring, the northern most populations emerge first and there is no temporal overlap with populations emerging at the southerly end of the distribution, over 700 km away (Houston 1991). The mating and nesting behaviour of this species is well documented. The females nest in solitary burrows in large aggregations in the hard packed soils of claypans. Freshly eclosed adult bees emerge from burrows dug in the previous years’ nesting season. Males emerge on average 15 days earlier than females (Alcock 1997b).

Males are dimorphic; large major males patrol the emergence area searching for emerging females, while small minor males can be found predominantly in the peripheral vegetation, or on flowering host plants some distance from the emergence site (Alcock 1997a). As females emerge, the males detect them, first visually from movement at the burrow entrance, and then via the female’s cuticular pheromones (Simmons et al. 2003). Aggressive interactions over emerging females are intense and the largest major males monopolise and mate with almost 90% of virgin females as they emerge (Alcock 1996a). Behavioural data suggest that females are monandrous (Simmons et al. 2000) and females change their pheromone blends after copulation, making themselves unattractive to searching males when they return to the emergence site to begin nesting (Simmons et al. 2003). The degree to which males and females of Dawson’s burrowing bee return to their natal nesting sites to breed is unknown, although once they have begun breeding, males do exhibit high levels of site fidelity and homing abilities (Alcock 1996b). Nesting sites do persist for many years and if bees routinely use natal nesting sites we might expect little gene flow and high levels of genetic divergence between populations. Here we use microsatellite markers to examine the genetic structure across 13 populations of Dawson’s burrowing bee from across the entire geographic range. Contrary to expectation, we find little evidence for genetic structure, suggesting that the ephemeral nature of resources required for breeding favours widespread dispersal, leading to panmixia across the species range.

3.3. Materials and methods

3.3.1. Sample collection

The bees were collected during the nesting season of July to August 2003, with the exception of the samples from Miaboolya and Billabalong, which were collected during the nesting season of 2001. Non-destructive sampling of female adult bees was carried out by catching each bee either as it emerged or whilst nesting. Bees were cooled on ice until torpid. A middle leg was then removed and stored in DNA storage buffer (4M NaCl, 0.25M EDTA pH 8.0, 20% DMSO) at room temperature until DNA extraction was performed. Thirteen populations were sampled that covered most of the

geographical range of the species, with distances between populations ranging from 1.5km to 707km (see Figure 3.1 and Table 3.1)

3.3.2. Microsatellite analysis

DNA was extracted from each bee leg using a salt based digestion buffer (100mM NaCl, 50mM Tris-HCl, 1% SDS, 100mM EDTA pH8.0) very similar to that used by Gemmell and Akiyama (1996). Following resuspension in 50µl TE buffer, each DNA sample was treated with 0.5units Riboshredder (Epicentre) to remove RNA. The following 8 primer pairs were chosen for the microsatellite screening based on the largest number of alleles detected (Beetri2, Beetri12, Beetri116, Beetri143, Beedi7, Beedi11, Beedi12 and Beedi40) (Beveridge & Simmons 2004). Each forward primer was labelled with a fluorescent dye; either 6-FAM (Geneworks), VIC, NED or ROX (Applied Biosystems). The primers were multiplexed so that all 4 trinucleotide repeat primers were run together in one polymerase chain reaction (PCR), Beedi7 and Beedi11 were run in another PCR and Beedi12 and Beedi40 were run in a third PCR. Each 10µl

PCR contained 1x PCR Buffer (20mM Tris-HCl pH 8.4, 50mM KCl), 1.5mM MgCl2 (Invitrogen), 200µM of each dNTP (Invitrogen), 250nM of each forward primer (labelled and unlabelled in a ratio of 1:10), 250nM of each reverse primer (Geneworks), 0.5units of Platinum Taq DNA polymerase (Invitrogen) and ~10ng of DNA. PCR amplification was performed in a PTC-0200 DNA Engine (Geneworks) with the following cycling conditions: 94°C for 1 minute, then 30 cycles of 94°C for 1 minute, 56°C for 1 minute and 72°C for 1 minute, and finally 72°C for 45 minutes. The PCR products were combined (1.5µl of each PCR) and analysed on an ABI3730 Sequencer and sized using Genescan-500 LIZ internal size standard and Genemapper software (version 3.0).

3.3.3. Data analysis

Mean number of alleles, allelic richness (a measure independent of sample size), number of private alleles and inbreeding coefficient FIS (Weir & Cockerman 1984) were calculated for each population using FSTAT 2.9.3.2 (Goudet 1995). Expected and observed heterozygosities were calculated using microsatellite analyser v3.15

(Dieringer & Schlotterer 2003). Using FSTAT, we also calculated pairwise FST values

between all pairs of populations and an overall FST value by bootstrapping over loci. In addition, we calculated the standardized genetic differentiation measure of G′ST (Hedrick 2005). With GENEPOP 3.4 (Raymond & Rousset 1995b) we tested deviations from Hardy-Weinberg expectations per locus, across populations and overall, using Fisher’s exact test and confidence intervals based on the Markov chain method. We also looked at linkage disequilibrium between all loci across all populations using Fisher’s method. In addition, we tested allelic differentiation across populations as described by Raymond and Rousset (1995a) and genotypic differentiation across populations using a log-likelihood (G) based exact test (Goudet et al. 1996). We analysed the data using spatial autocorrelation within the software program GenAlEx6 (Peakall & Smouse 2005) using distance classes of 20km and 50km.

3.4. Results

3.4.1. Overall genetic diversity

We sampled 531 female bees and detected between 4 and 38 alleles per locus over all populations. The mean number of alleles per locus was 7.8 and mean alleleic richness was 7.1 based on a minimum sample size of 28 diploid individuals (Table 3.2). Over all populations the mean expected heterozygosity was 0.602 and the mean observed heterozygosity was 0.596, indicating no inbreeding. This was in agreement with a very small overall mean FIS of 0.010. Only one locus deviated significantly from Hardy-Weinberg equilibrium with a P value of 0.043. This value does not survive standard Bonferroni correction for the 8 loci tested (Bonferroni P0.05=0.006). Over all loci and all populations the bees were in Hardy-Weinberg equilibrium (P = 0.336). We tested for linkage disequilibrium between 28 pairs of loci across all populations and found only one pair to be significant (Beetri116 and Beedi11).

3.4.2. Genetic distances between populations

Four populations had 1 private allele and 6 populations had 2 private alleles (see

Table 3.2). The overall FST was 0.001 with 95% confidence limits of -0.002 and 0.003 after bootstrapping over loci. The standardized genetic differentiation measure G′ST was

also very small at 0.001. Therefore, genetic differentiation between all populations did not differ significantly from zero. Of the 78 pairwise comparisons, only one was significant after 1560 permutations, with pairwise FST values ranging from -0.0071 to 0.0122. For the allelic differentiation test, P-values ranged from 0.067 to 0.942 across loci with an overall probability of 0.244 (Fisher’s method). The genotypic differentiation test yielded one significant P-value of 0.044 for locus Beetri2 with the other 7 loci ranging from 0.182 to 0.939. The overall P-value across loci was 0.223 (Fisher’s method). The results of the spatial autocorrelation analyses indicated that there was no detectable structure at distance classes of 20km or 50km.

3.5. Discussion

From our analysis of microsatellite variation, we found no genetic differentiation in Dawson’s burrowing bee across 13 populations covering almost the entire species range, suggesting that the species is probably panmictic. Using Wright’s FST (Weir &

Cockerman 1984), the standardized measure G′ST (Hedrick 2005), and allelic and genotypic probability tests (Raymond & Rousset 1995a) as measures of genetic differentiation, and spatial autocorrelation to look for variation with distance (Smouse & Peakall 1999), no genetic structuring could be detected. In addition, the results indicated no evidence of inbreeding, and the entire sample collection was in Hardy- Weinberg equilibrium. This result is in contrast to almost all other studies on population structure in bees, which have reported significant intraspecific genetic differentiation (Paar et al. 2004; Shao et al. 2004; Widmer & Schmid-Hempel 1999). Previous studies have focused almost exclusively on social species even though most bees are solitary (Wcislo & Cane 1996). For example, there have been population studies on the honeybee Apis mellifera (De la Rua et al. 2001; De la Rua et al. 2003) and the bumble bee Bombus ignitus (Shao et al. 2004) although these have focused on differentiation between island populations so they are not directly comparable to our data on populations occupying continuous habitat. Nevertheless, those studies that have examined continuous habitat over a similar geographical range to our study of Dawson’s burrowing bee, do find significant genetic differentiation. For example in their study of Apis dorsata in northeast India, Paar et al. (2004) found genetic differentiation even though the species is migratory and expected to show little population structure. Likewise, of the few

studies carried out on native Australian bees and wasps, moderate but significant population differentiation has been found in Trigonia sp. (Franck et al. 2004) and Ropalidia revolutionalis (Henshaw & Crozier 2004) respectively. These species both occupy a range of climate zones in Queensland which may provide the barriers to gene flow that would promote differentiation. In contrast, the habitat of Dawson’s burrowing bee is uniformly arid across its range which might promote rather than hinder panmixia. Similar to our findings with Dawson’s burrowing bee, a study of genetic differentiation in the bumble bee, Bombus terrestris, revealed remarkable homogeneity between populations across the entire European continent, yet there was genetic differentiation between island populations, and between islands and the continent (Estoup et al. 1996). This is an interesting comparison because it supports the notion that islands can provide barriers to gene flow that promote genetic differentiation (Johnson et al. 1994; Pudovskis et al. 2001). A more recent study of B. terrestris uncovered size homoplasy at some microsatellite loci, suggesting a more marked population structure on the continent than that originally reported (Viard et al. 1998). Size homoplasy could explain the lack of population structure in our study of A. dawsoni. However, the frequency of size homoplasy in A. dawsoni would need to be measured in order to determine its importance to our findings. A recent review by Estoup et al (2002) suggests that size homoplasy does not represent a significant problem for most population genetics analyses, and that the highly variable nature of microsatellite loci largely compensates for their homoplasious evolution. It could be argued that the populations of Dawson’s burrowing bee studied here constitute a single metapopulation. Anecdotal evidence suggests that populations, although they persist in certain locations for many years, can dwindle in numbers to a point where they go extinct. One of the key definitions for a metapopulation is that there are frequent extinctions and recolonizations (Hanski 1999). This generally promotes local genetic differentiation and leads to FST values significantly higher than zero (Wade & McCauley 1988). Only in certain circumstances, when the number of individuals colonizing a new population is much larger than the number of individuals moving between existing populations, does genetic differentiation decrease (Wade & McCauley 1988). Also, if a metapopulation has frequent extinction and recolonization events, then one would expect the level of heterozygosity in the individual populations, and in the metapopulation as a whole, to be reduced (Hedrick & Gilpin 1997). This was certainly not the case with A. dawsoni as we detected no heterozygote deficit. Frequent

extinction and recolonization also reduce the effective size of the population quite dramatically compared to the effective size of a corresponding panmictic population (Maruyama & Kimura 1980). Calculating the effective population size for a metapopulation is difficult however, because it is influenced by extinction and recolonization dynamics (Hedrick & Gilpin 1997). In addition, temporal data are generally required for an accurate estimate of the effective population size of a population and we do not have these data available. Nevertheless, given that FST was not significantly different from zero, and the levels of heterozygosity were high, it is highly unlikely that Dawson’s burrowing bee exists as a metapopulation. In comparison to other taxa, levels of genetic differentiation in the insects have been shown to be lower than other invertebrates (Ward et al. 1992). Levels of genetic differentiation in insects are comparable to those found in birds, indicating that mobility may have an important influence on population structure (Avise 1994). However, as mentioned above, even the most mobile species of bee, capable of long migrations, still exhibit significant population structure (Paar et al. 2004), so other aspects of the bee’s ecology must also be important. The mating and nesting behaviour of A. dawsoni is such that the opportunities for gene flow seem at first to be limited. Nearest populations of Dawson’s burrowing bee often appear to be separated by several to many kilometres. Both males and females emerge from the burrows of the previous year’s nests, with males emerging on average over 15 days earlier than females (Alcock 1997b). Little is known about the movement of males when they emerge but once they begin patrolling an emergence area for virgin females, males show remarkable site fidelity (Alcock 1996b). Females have been observed to mate only once (Simmons et al. 2000) so the species is probably monandrous. Almost 90% of females mate immediately upon emergence (Alcock 1996a), but again little is known about the proportion of females that return to their emergence area to nest. The main opportunity for gene flow appears to be the 10% of females that leave the emergence area unmated to go in search of flowering plants for food. It may be at these limited food resources that males and females from different emergence sites mate and genes are exchanged. Moreover, rainfall in the arid zone of northern Western Australia is highly stochastic and patchily distributed, so access to flowering host plants from a particular emergence site is not guaranteed year after year. Thus, if rain does not fall close to an emergence site there will be a deficit of resources at a suitable distance for females to

use for nest provisioning. In this case bees are likely to leave their natal emergence site and search for other suitable nesting sites closer to sources of pollen and nectar for provisioning offspring. Such behaviour would provide the opportunity for gene flow and would help explain the lack of population structure observed across the species range. We have anecdotal evidence that this may occur because during the nesting season of 2003, a particularly dry year when we collected the bulk of our samples, we collected from a large emergence site (Carnarvon Pistol Club) where no bees returned to nest, indicating that they had dispersed. A second site, Miaboolya, has been the focus of our research over the last decade (Alcock 1996a; Alcock et al. 2005) but in 2003 there were no nesting females here, although bees had earlier emerged in large numbers from this location. In conclusion, our data for Dawson’s burrowing bee seem to provide a rare example of panmixia over a considerable geographic range.

3.6. References

Abbott CL, Double MC (2003) Genetic structure, conservation genetics and evidence of speciation by range expansion in shy and white-capped albatrosses. Molecular Ecology 12, 2953-2962. Alcock J (1996a) The relation between male body size, fighting, and mating success in Dawson's burrowing bee, Amegilla dawsoni (Apidae, Apinae, Anthphorini). Journal of Zoology 239, 663-674. Alcock J (1996b) Site fidelity and homing ability of males of Dawson's burrowing bee (Amegilla dawsoni) (Apidae, Anthophorini). Journal of the Kansas Entomological Society 69, 182-190. Alcock J (1997a) Competition from large males and the alternative mating tactics of small males of Dawson's burrowing bee (Amegilla dawsoni) (Apidae, Apinae, Anthophorini). Journal of Insect Behavior 10, 99-113. Alcock J (1997b) Small males emerge earlier than large males in Dawson's burrowing bee (Amegilla dawsoni) (Hymenoptera: Anthophorini). Journal of Zoology 242, 453-462. Alcock J, Simmons LW, Beveridge M (2005) Seasonal change in offspring sex and size in Dawson's burrowing bees (Amegilla dawsoni) (Hymenoptera: Anthorphorini). Ecological Entomology 30, 247-254.

Avise JC (1994) Molecular Markers, Natural History and Evolution, 1 edn. Chapman and Hall, New York. Beveridge M, Simmons LW (2004) Microsatellite loci for Dawson's burrowing bee (Amegilla dawsoni) and their cross-utility in other Amegilla species. Molecular Ecology Notes 4, 379-381. De la Rua P, Galian J, Serrano J, Moritz RFA (2001) Genetic structure and distinctness of Apis mellifera L. populations from the Canary Islands. Molecular Ecology 10, 1733-1742. De la Rua P, Galian J, Serrano J, Moritz RFA (2003) Genetic structure of Balearic honeybee populations based on microsatellite polymorphism. Genetics Selection Evolution 35, 339-350. Dieringer D, Schlotterer C (2003) Microsatellite Analyser (MSA): a platform independent analysis tool for large microsatellite data sets. Molecular Ecology Notes 3, 167-169. Estoup A, Jarne P, Cornuet J-M (2002) Homoplasy and mutation model at microsatellite loci and their consequences for population genetics analysis. Molecular Ecology 11, 1591-1604. Estoup A, Solignac M, Cornuet J-M, Goudet J, Scholl A (1996) Genetic differentiation of continental and island populations of Bombus terrestris (Hymenoptera: Apidae) in Europe. Molecular Ecology 5, 19-31. Franck P, Cameron E, Good G, Rasplus J, Oldroyd B (2004) Nest architecture and genetic differentiation in a species complex of Australian stingless bees. Molecular Ecology 13, 2317-2331. Gemmel NJ, Akiyama S (1996) An efficient method for the extraction of DNA from vertebrate tissues. Trends in genetics 12, 338-339. Goudet J (1995) FSTAT (version 1.2): a computer program to calculate F-statistics. Journal of Heredity 86, 485-486. Goudet J, Raymond M, Meeus T de, Rousset F (1996) Testing differentiation in diploid populations. Genetics 144, 1933-1940. Hanski I (1999) Metapopulation Ecology. Oxford University Press, Oxford, UK. Hedrick PW (2005) A standardized genetic differentiation measure. Evolution 59, 1633- 1638.

Hedrick PW, Gilpin ME (1997) Genetic effective size of a metapopulation. In: Metapopulation Biology: Ecology, Genetics, and Evolution (eds Hanski I, Gilpin ME), pp. 165-181. Academic Press, Inc., San Diego. Henshaw M, Crozier R (2004) Mating system and population structure of the primitively eusocial wasp Ropalidia revolutionalis: a model system for the evolution of complex societies. Molecular Ecology 13, 1943-1950. Hoarau G, Rijnsdorp AD, Van der Veer HW, Stam WT, Olsen JL (2002) Population structure of plaice (Pleuronectes platessa L.) in northern Europe: microsatellites revealed large-scale spatial and temporal homogeneity. Molecular Ecology 11, 1165-1176. Houston TF (1991) Ecology and behaviour of the bee Amegilla (Asaropoda) dawsoni (Rayment) with notes on a related species (Hymenoptera: Anthophoridae). Records of the Western Australian Museum 15, 535-553. Johnson MS, Watts RJ, Black R (1994) High levels of genetic subdivision in peripherally isolated populations of the atherinid fish Craterocephalus capreoli in the Houtman Abrolhos Islands, Western Australia. Marine Biology 119, 179- 184. Maruyama T, Kimura M (1980) Genetic variability and effective population size when local extinction and recolonization of subpopulations are frequent. Proceedings of the National Academy of Science USA 77, 6710-6714. Paar J, Oldroyd BP, Huettinger E, Kastberger G (2004) Genetic structure of an Apis dorsata population: the significance of migration and colony aggregation. Journal of Heredity 95, 119-126. Peakall R, Smouse P (2005) GenAlEx 6: Genetic analysis in Excel. Population genetic software for teaching and research. Australia National University, Canberra, Australia. http://www.anu.edu.au/BoZo/GenAlEx/ Pudovskis MS, Johnson MS, Black R (2001) Genetic divergence of peripherally disjunct populations of the gastropod Batillariella estuarina in the Houtman Abrolhos Islands, Western Australia. Molecular Ecology 10, 2605-2616. Raymond M, Rousset F (1995a) An exact test for population differentiation. Evolution 49, 1280-1283. Raymond M, Rousset F (1995b) GENEPOP (Version 1.2): Population genetics software for exact tests and ecumenicism. Journal of Heredity 86, 248-249.

Ridgway T, Hoegh-Guldberg O, Ayre DJ (2001) Panmixia in Pocillopora verrucosa from South Africa. Marine Biology 139, 175-181. Shao Z-Y, Mao H-X, Fu W-J, et al. (2004) Genetic structure of Asian populations of Bombus ignitus (Hymenoptera: Apidae). Journal of Heredity 95, 46-52. Simmons L, Alcock J, Reeder A (2003) The role of cuticular hydrocarbons in male attraction and repulsion by female Dawson's burrowing bee, Amegilla dawsoni. Animal Behaviour 66, 677-685. Simmons LW, Tomkins JL, Alcock J (2000) Can minor males of Dawson's burrowing bee, Amegilla dawsoni (Hymenoptera: Anthophorini) compensate for reduced access to virgin females through sperm competition? Behavioral Ecology 11, 319-325. Slatkin M (1987) Gene flow and the geographic structure of natural populations. Science 236, 787-792. Smouse P, Peakall R (1999) Spatial autocorrelation analysis of individual multiallele and multilocus genetic structure. Heredity 82, 561-573. Tesch E-W (1977) The Eel Chapman and Hall, London. Viard F, Franck P, Dubois M-P, Estoup A, Jarne P (1998) Variation of microsatellite size homoplasy across electromorphs, loci, and populations in three invertebrate species. Journal of Molecular Evolution 47, 42-51. Wade MJ, McCauley DE (1988) Extinction and recolonization: their effects on the genetic differentiation of local populations. Evolution 42, 995-1005. Ward RD, Skibinski DOF, Woodwark M (1992) Protein heterozygosity, protein structure, and taxonomic differentiation. Evolutionary Biology 26, 73-159. Wcislo WT, Cane JH (1996) Floral resource utilization by solitary bees (Hymenoptera: Apoidea) and exploitation of their stored foods by natural enemies. Annual Review of Entomology 41, 257-286. Weir BS, Cockerman CC (1984) Estimating F-statistics for the analysis of population structure. Evolution 38, 1358-1370. Widmer A, Schmid-Hempel P (1999) The population genetic structure of a large temperate pollinator species, Bombus pascuorum (Scopoli) (Hymenoptera: Apidae). Molecular Ecology 8, 387-398. Wirth T, Bernatchez L (2001) Genetic evidence against panmixia in the European eel. Nature 409, 1037-1040.

Figure 3.1. Sample collection localities for 13 populations of Dawson’s burrowing bee

Table 3.1. Geographical coordinates for collection localities and number of bees sampled

Coordinates (degrees) No of bees Site E S sampled 1 Billabalong Station 115.88330 27.36907 48 2 Miaboolya Road 113.64308 24.80198 46 3 Yuin Station 116.03400 27.98000 42 4 Ballinyoo Bridge 115.77845 27.52473 42 5 Onslow 115.05670 21.68275 42 6 Carnarvon Pistol Club 113.72452 24.91665 33 7 Kennedy Range roadside 115.18284 24.64858 41 8 Kennedy Range inland 115.19799 24.64717 41 9 Meeberrie Station 115.96974 26.96189 38 10 Babbage Island 113.63874 24.88273 29 11 Mt Sandiman Station 115.42400 24.39693 42 12 Landor Station 116.90263 25.12619 42 13 Gascoyne Junction Road 113.87124 24.83108 46

Table 3.2. Summary of genetic diversity in 13 populations of Dawson’s burrowing bee

No of Site A AR private HE HO FIS alleles 1 Billabalong Station 8.3 7.0 1 0.594 0.599 -0.009 2 Miaboolya Road 7.4 6.3 2 0.568 0.582 -0.025 3 Yuin Station 8.4 7.5 2 0.621 0.595 0.042 4 Ballinyoo Bridge 7.5 7.0 0 0.604 0.561 0.072 5 Onslow 8.0 7.1 0 0.607 0.574 0.054 6 Carnarvon Pistol Club 7.3 6.9 0 0.587 0.564 0.038 7 Kennedy Range roadside 7.9 7.3 1 0.612 0.610 0.004 8 Kennedy Range inland 7.1 6.5 2 0.624 0.625 -0.001 9 Meeberrie Station 7.9 7.2 1 0.590 0.582 0.014 10 Babbage Island 7.5 7.5 1 0.589 0.596 -0.011 11 Mt Sandiman Station 7.4 6.5 2 0.579 0.582 -0.005 12 Landor Station 8.5 7.6 2 0.625 0.634 -0.015 13 Gascoyne Junction Road 8.6 7.4 2 0.626 0.641 -0.025 Mean 7.8 7.1 1.2 0.602 0.596 0.010

± Standard Error ±0.1 ±0.1 ±0.2 ±0.005 ±0.007 ±0.009

A, mean number of alleles per locus; AR, allelic richness; number of private alleles per population; HE, mean expected heterozygosity per population; HO, mean observed heterozygosity per population; FIS, mean inbreeding coefficient per population

CHAPTER 4

Genetic breeding system and investment patterns within nests of Dawson's burrowing bee (Amegilla dawsoni) (Hymenoptera: Anthophorini)

4.1 Abstract

Dawson’s burrowing bee is a large solitary ground nesting bee endemic to the arid zone of Western Australia. In this study we use microsatellite markers to analyse the genotypes of offspring from individual nests to determine the number of effective mates for each female. From these data we have determined that females almost certainly mate only once which is consistent with male reproductive tactics that include protandry and intense male-male competition for access to virgin females. We also use the molecular data to show that the nesting female is the mother of all the offspring of her nest and that brood parasitism is unlikely in this species. The data indicate that females make daughters at the beginning of the season followed by large sons in the middle, and then small sons at the end. Females often place one brood cell directly above another. The distribution of sex and morph in these doublets follows a pattern with most containing a female on the bottom and a minor male on the top, followed by almost equal numbers of female on top of female and minor male on top of major male. This pattern is likely favoured by emergence patterns, with males emerging before females and minor males emerging before major males. We suggest that although minor males have low reproductive success, their production may nonetheless be beneficial in that minor males open up emergence tunnels for their larger and reproductively more valuable siblings. In addition, minor males may be a best of a bad job product arising from changes in the costs to nesting females of gathering brood provisions over the course of the flight season.

4.2 Introduction

Knowledge of the genetic mating system of a species can give valuable insight into the evolutionary processes that shape its behaviour, morphology, and life-history. Observational data on the mating behaviour of a species can be misleading however, and must be used in conjunction with other methods in order to be reliable (Page 1986). It was long thought that the majority of bird species were monogamous, but with the advent of molecular tools, the level of extra-pair paternity amongst surveyed passerine species has been estimated at 86% (Griffiths et al. 2002), and it is now widely accepted that observing the social bonds of a species cannot predict the genetic mating system. In the socially monogamous bird species studied to date, genetic monogamy has been found in less than 25% (Griffiths et al. 2002). Polyandry can be defined as a mating system in which females mate with more than one male within a single reproductive cycle, and monandry a mating system in which females mate with only one male (Thornhill & Alcock 1983). Mating behaviour has been studied extensively in insects, with polyandry appearing to be the norm, despite the potential costs to females from searching for and mating with multiple males (Arnqvist & Nilsson 2000). In the Hymenoptera both forms of mating behaviour are found (Thornhill & Alcock 1983). In social Hymenoptera however, the large majority of species studied have proved to be either monandrous, or to exhibit very low levels of polyandry (Strassmann 2001). The exceptions to this rule are limited to just 5 groups that exhibit very high levels of polyandry (leaf-cutting ants, seed-harvesting ants, army ants, honey bees and higher vespine wasps) (Brown & Schmid-Hempel 2003; Kronauer et al. 2004). How mating systems have evolved within the context of sociality is the focus of considerable research effort (Pedersen & Boomsma 1999b; Peters et al. 1999; Schmid-Hempel & Schmid-Hempel 2000). However, there have been very few studies of the mating systems of solitary Hymenoptera (Zavodna et al. 2005). Such studies are needed before we can explore the evolution of hymenopteran mating systems within a broad phylogenetic context. Here we determine the genetic mating system of a species of solitary bee, Amegilla dawsoni. Dawson’s burrowing bee is a solitary ground nesting bee belonging to the tribe Anthorphorini. Solitary aculeate Hymenoptera typically exhibit sexual dimorphism in body size with females being much larger than males (Stubblefield & Seger 1994). In Dawson’s burrowing bee, there are two size classes of males: large major males that are

the same size as females, and smaller minor males. This male size dimorphism is also found in a small number of other Hymenoptera (Danforth & Desjardins 1999; Danforth & Neff 1992). The male size dimorphism corresponds to a difference in mating tactics (Alcock 1997a). Large males patrol only the emergence area searching for emerging virgin females, which dig upwards to the surface from their underground cells after eclosion. Cuticular hydrocarbon profiles differ between males and females, and between virgin and nesting females, allowing the patrolling males to determine the sex and receptivity of the emerging bee (Simmons et al. 2003). When a large patrolling male detects the emergence of a virgin female, he waits by the entrance of the emergence tunnel and defends it against other males. When the female eventually emerges onto the surface, competition between males is fierce and the largest male generally wins the contest, mounting and mating with the female after which she appears to become unreceptive to further matings (Simmons et al. 2000). Small males sometimes use the strategy of patrolling the emergence area, but usually only when few large males are also patrolling. When male density and competition is high at the emergence site, smaller males employ an alternative mating tactic of patrolling the periphery of the emergence area and also the host plants that the bees visit to feed (Alcock 1997a). Any female that leaves the emergence area unmated may then be detected and mated by these small males. However, only 13% of females leave the emergence site unmated so that if females are indeed monandrous, minor males must have very low fitness. Yet minor males represent 80% of the male population and thus the maintenance of the male dimorphism poses an evolutionary puzzle (Alcock 1996a; Tomkins et al. 2001). One possible solution to this puzzle would be provided if females often mate with minor males after an initial mating with a major. Even though behavioural observations suggest that females are unwilling to mate more than once immediately upon emergence (Simmons et al. 2000), females might also mate with minor males away from the emergence site at some later date. To test this possibility, in this study we use microsatellite markers to determine the genetic mating system of this species. A recent study has shown that nest searching females frequently enter and remain within active nests until the owner returns from her provisioning trip. Regardless of size, owners almost always win any contest over residency that may ensue (Alcock et al. 2006). However, it is possible that nest searchers represent brood parasites

(Wuellner 1999), laying eggs in open brood chambers while the nesting female is away collecting provisions. The data provided by genotyping offspring derived from individual nests will allow us to estimate the frequency with which nest parasitism occurs. Finally, there is evidence to suggest that female Dawson’s burrowing bees might change the size of the offspring they produce through the season, making large offspring first followed by smaller offspring later (Tomkins et al. 2001). A cross-sectional survey of active nests suggested that, at the population level, female offspring are produced early in the season, females and major males in the middle of the season, and minor males late in the season (Alcock et al. 2005). Resource availability declines through the nesting season (Tomkins et al. 2001) and females may benefit from making the larger, more costly female and major male offspring earlier in the season when resources are most abundant (Peterson & Roitberg 2006; Torchio & Tepedino 1980). Thus, the observed male dimorphism might reflect a condition dependent strategy in which females make the best-of-a-bad-job in producing offspring of low fitness when resource limitations dictate (Alcock et al. 2005). It is unknown whether individual females can make continuous adjustments in the size and sex of their offspring through the season, or whether the patterns observed reflect fixed breeding tactics of females that nest at different times of the season. The molecular data collected in this study will allow us to analyse seasonal changes in the sex and size of offspring produced by individual females, and compare these results with previous data on population wide patterns (Alcock et al. 2005).

4.3 Materials and methods

4.3.1 Sample collection

This study was conducted from 5th August to 8th October 2003 at a large nesting aggregation of Dawson’s burrowing bee just to the side of Gascoyne Junction Road about 25km east of Carnarvon, Western Australia. On August 5th and 6th, a large number of active nests were labelled and the female occupant of the nest was captured, marked and released. The bees were captured using the method described in Alcock et al. (2005). On 12th and 13th August, the occupant of the nest was checked to see if it was the same individual as that marked on 5th or 6th August. On 13th August, for those

nests where this was the case, the bee was collected and the nest plugged with a small piece of sponge to prevent it from being re-occupied. A middle leg of the bee was removed and stored in DNA storage buffer (4M NaCl, 0.25M EDTA pH 8.0, 20% DMSO) at room temperature until DNA extraction was performed. The rest of the bee was also retained and frozen. The sponge was subsequently removed from the nest entrance and the tunnel filled with plaster-of-Paris. Because brood cells are capped by the bee, the plaster filled only the main tunnel directly above the cells. From 6th to 8th October, using the plaster-of-Paris cast as a guide, 31 nests were excavated and 208 offspring were removed. The excavated prepupae were weighed and the head was removed and placed in DNA storage buffer. As a female occupies the nest for several weeks she provisions a series of brood cells that are laid down sequentially. The female digs a vertical tunnel to a depth of c. 15-20 cm before turning to dig in a roughly horizontal direction (see Fig.4 in Houston 1991). The first brood cell is dug downward at the end of the tunnel, which is then extended for each subsequent brood cell. Females will, on occasion, place one cell directly above another, in this way making doublets (Houston 1991). The position of each prepupa in a female’s nesting sequence was recorded for all but two of the nests. The female and her offspring were then subjected to microsatellite analysis for three purposes: to sex the offspring, to determine the genotype of each offspring and thus determine the number of effective mates for each female, and to determine whether any nests were subject to brood parasitism.

4.3.2 Microsatellite analysis

DNA was extracted from each female bee leg and prepupa head using a salt based digestion buffer (100mM NaCl, 50mM Tris-HCl, 1% SDS, 100mM EDTA pH8.0) very similar to that used by Gemmell and Akiyama (1996). Following resuspension in 50µl TE buffer, each DNA sample was treated with 0.5units Riboshredder (Epicentre) to remove RNA. The following 8 primer pairs were chosen for the microsatellite screening based on the largest number of alleles detected, 4 containing trinucleotide repeats (Beetri2, Beetri12, Beetri116, Beetri143), and 4 containing dinucleotide repeats (Beedi7, Beedi11, Beedi12 and Beedi40) (Beveridge & Simmons 2004). Each forward primer was labelled with a fluorescent dye; either 6-

FAM (Geneworks), VIC, NED or ROX (Applied Biosystems). The primers were multiplexed so that all 4 trinucleotide repeat primers were run together in one polymerase chain reaction (PCR). In addition, Beedi7 and Beedi11 were run in another PCR and Beedi12 and Beedi40 were run in a third PCR. Each 10µl PCR contained 1x

PCR Buffer (20mM Tris-HCl pH 8.4, 50mM KCl), 1.5mM MgCl2 (Invitrogen), 200µM of each dNTP (Invitrogen), 250nM of each forward primer (labelled and unlabelled in a ratio of 1:10), 250nM of each reverse primer (Geneworks), 0.5units of Platinum Taq DNA polymerase (Invitrogen) and ~10ng of DNA. PCR amplification was performed in a PTC-0200 DNA Engine (Geneworks) with the following cycling conditions: 94°C for 1 minute, then 30 cycles of 94°C for 1 minute, 56°C for 1 minute and 72°C for 1 minute, and finally 72°C for 45 minutes. The PCR products were combined (1.5µl of each PCR) and analysed on an ABI3730 Sequencer and sized using Genescan-500 LIZ internal size standard and Genemapper software (version 3.7).

4.3.3 Data analysis

The sex of each offspring was determined as follows. Each individual was scored as heterozygous or not at each of the eight microsatellite loci. Any individuals that were heterozygous at one or more loci were classified as female. Individuals that were not heterozygous at any loci were assumed to be hemizygous and therefore male. Using the product of the observed homozygosity of the eight microsatellite loci (Beveridge & Simmons 2004), the probability of a female bee being homozygous at all eight loci was 0.014. So, for a sample of 208 offspring we may expect to have misclassified 2 or 3 individuals, an error rate unlikely to affect our conclusions. The genotypes of each female and her offspring were analysed using GERUD 1.0 (Jones 2001). This program determines the number of effective mates for a given progeny array with one known parent. We also used the program PARENTAGE (Emery et al. 2001), which infers the probable number of fathers using population allele frequencies. In addition, we also analysed the data using the software program MATESOFT v1.0 which is specifically designed for haplodiploid species (Moilanen et al. 2004). This program can infer the genotypes of fathers, assign offspring to patrilines and calculate full mating frequency statistics including average paternity skew, average pedigree effective mate number of females and the combined average weighted non- identification error.

The genotypes of all the females were also used to calculate an inbreeding coefficient for the sample. We then used the inferred genotypes of the putative males to calculate the relatedness (r) between each female and her putative mate using the software program RELATEDNESS 5.0 (Queller & Goodnight 1989).

4.4 Results

4.4.1 Genetic mating system

For each nest, the number of offspring varied from 3 to 11 with a mean and SE of 6.71 ± 0.39. However, for determining the number of effective mates, only female offspring could be used because male offspring are fatherless and therefore give no information on the genotype of their mother’s mate(s). For 3 families, only one offspring was female, which was insufficient for determining the number of effective mates in these cases. Of the remaining 28 families, the numbers of female offspring ranged from 2 to 6 with a mean and SE of 3.61 ± 0.25. We first estimated the number of potential fathers by simple allele counting. This involved identifying the maternal alleles and disregarding them. Homozygotes at a particular locus were assumed to have one maternal and one paternal allele and heterozygotes with the same genotype as the mother were taken to have one contributing paternal allele. Alleles were then counted and the number from the locus with the most non-maternal alleles was taken as an estimate of the number of fathers present. This estimate was 1 for each of the 28 families (see Table 4.1). Second, we used the software program GERUD version 1.0, which is more sophisticated than allele counting because it uses multiple loci simultaneously. After removing the maternal alleles from offspring genotypes, it then simulates all possible paternal genotypes and calculates the combinations of these genotypes that yield the fewest possible males that could have contributed to the offspring genotypes observed. For the 28 families analysed, 22 yielded 1 as the number of effective mates using 8 microsatellite loci. For the remaining families, 5 of these gave 1 as the minimum number of mates and 2 as a maximum number due to shared genotypes between the mother and potential father(s) at one of the loci. For one family, there were shared genotypes at 2 loci giving a minimum number of one father and a maximum number of 4 paternal genotypes (see Table 4.1).

We also used PARENTAGE to infer the probable number of fathers of each family using population allele frequencies and initial probabilities based on what we already knew about the parentage of the offspring. Each maternal genotype was specified and initial probabilities for the number of possible fathers were set, based on the number of offspring present in each family. The maternity share was set with a low probability of there being more than one mother (gamma distribution with a shape parameter of 1 and a mean of 0.25). The paternity share was set using a gamma distribution with a shape parameter of 1 and a mean of 0.005. Trials were run using means of 0.02 and 0.002 to simulate both a higher number and lower number of fathers but these did not change the outcome, so we used 0.005 for the final analysis. To allow for a mutation rate that agrees with levels observed for microsatellite markers (Weber & Wong 1993), a mutation rate was set using a gamma distribution with a shape parameter of 2 and a mean of 0.001. This allowed for 95% of the mutation rates to lie between 0.00014 and 0.0028 mutations per generation. In the output for 28 families, only 2 families showed a number of mutations above 0 and these were both extremely low (0.0004 and 0.0006), indicating that the genotypes had been scored correctly and that mutations were very rare. We ran 5000 iterations for each family (using a burn-in of 5000 and a thinning interval of 400). The results of the analysis are shown in Table 4.1 and indicate that the effective number of mates for these 28 females was 1. Finally, we used MATESOFT 1.0, a software program specifically designed to analyse the mating systems of haplo-diploid species (Moilanen et al. 2004). This program implements procedures developed to estimate effective mating frequency and degree of paternity skew including corrections for non-detection and non-sampling errors (Pedersen & Boomsma 1999a). We used the same 28 families used in previous calculations which gave a corrected average paternity skew (c-bar) of 0.9985, an observed proportion of double mated queens (Dobs) of 0.0357 and an average pedigree effective mate number (me,p)of 1.0030. The errors of these estimates can be calculated by jackknifing over groups for standard error and bootstrapping by groups for confidence limits (Moilanen et al. 2004). However, this is recommended only for data sets that have a minimum of 5 groups in each category of detected single and double matings which is not the case with this data set. For all 31 families, the genotypes of the male offspring were consistent with the genotype of the mother in each case. There was no evidence, therefore, that other

females sneaked into an active nest during the nesting season to lay eggs in open brood chambers to parasitize the resident female. We also calculated the inbreeding coefficient, F, for the sample of 31 females using Microsatellite analyser (Dieringer & Schlotterer 2003) and found that the overall value across the 8 loci was -0.0029 (SE ± 0.0555) showing no evidence of inbreeding. From the genotypes of the offspring of each family we inferred the genotype of the putative male. We then calculated the relatedness between each female and her putative mate using RELATEDNESS 5.0 (Queller & Goodnight 1989). The average relatedness over all 31 pairwise comparisons was 0.0846 with 95% CL of 0.0898. We can conclude from this that the females were unrelated to their putative mates.

4.4.2 Investment sex and morph ratio

From a total of 31 excavated nests, we recovered 208 offspring, of which 104 were male and 104 were female, giving a population offspring sex ratio (M/F) of 1.0. The frequency distribution for male and female offspring prepupal weights is shown in Figure 4.1.1 and 4.1.2 respectively. As observed in previous studies of adults (Alcock 1999), female prepupal weight was normally distributed while male prepupal weight was bimodally distributed. Based on the distribution of male prepupal weights in Fig. 1b, a weight of 0.9g was used as the threshold below which individuals were classified as minors and above which they were classified as majors. Of the 104 males, 43 were majors, 56 were minors, and 5 could not be determined as these prepupae were ruptured during excavation, so that a weight could not be recorded. We analysed individual variation in the production of the two sexes, and the production of major and minor sons, using a general linear model with binomial error distributions and logit link functions. For sex ratio, the number of sons was used as the dependent variable with total offspring as the binomial denominator. For morph ratio, the number of major males was used as the dependent variable with total number of sons as the binomial denominator. There was no significant variation in sex ratio

2 produced by individual females (χ1 = 0.432, P = 0.511) or in the proportion of major

2 males to minor males (χ1 = 0.105, P = 0.746) . The average sex ratio (M/F) across 31 nests was 1.24 (SE ± 0.16) and the average major to minor ratio was 0.78 (SE ± 0.14).

We therefore pooled data across nests in order to analyse changes in sex and morph production through the nesting sequence. Of the 31 nests excavated, the sequence in which the broods were made was available for 29 nests. We again used a general linear model with sequence position as the predictor variable, the frequency of sons as the dependent variable, and the total number of offspring as the binomial denominator. There was a significant increase in

2 male production through the nesting sequence (χ1 = 6.989, P = 0.008) (Figure 4.2.1). Likewise, we examined changes in the male morph ratio through the nesting sequence, using the frequency of minor males as the dependent variable, and the total number of males as the binomial denominator. There was a significant increase in minor male

2 production through the nesting sequence (χ1 = 10.19, P = 0.0014) (Figure 4.2.2). Thus, females invest first in daughters, then sons, and among sons they invest first in major sons and then minor sons. Females often produced doublets, placing one brood directly above another. Out of the 29 nests with sequence information there were 26 nests containing 61 doublets, which equates to 62% of the total number of offspring produced by the study population. The frequency at which each combination of female and male (major and minor morph) occurred in each position was scored for the total number of doublets present. There was significant heterogeneity in the combinations of offspring contained in doublets (P = 0.0043 ± 0.0002; Monte Carlo R x C contingency test). The largest number of doublets had a female in the bottom cell and a minor male on top. This was followed by almost equal numbers of ‘female over female’ and ‘minor male over major male’ combinations. There were 6 occasions when a minor male was found over another minor male, and all other combinations were at a very low frequency or not found at all (see Figure 4.3). When we just looked at the first doublet and the last doublet to be made in a nest (22 nests had two or more doublets), the female over female combination changed from 11 to 3 and the minor male over major male changed from 3 to 9. This change was significant using a Monte Carlo R x C contingency table test (P = 0.0154 ± 0.0003).

4.5 Discussion

Using four different estimates for number of effective mates, we conclude that female Dawson’s burrowing bees almost certainly mate only once or use the sperm from only one male to fertilise their offspring. The genetic data clearly indicate that the broods in a single nest have only one father. This result is consistent with our behavioural observations of these bees and with previous experiments in which female bees refused to mate more than once even when given opportunities to do so (Simmons et al. 2000). It is also consistent with previous findings that female bees change their cuticular pheromones after mating to repel further male advances (Simmons et al. 2003). The genetic data also indicate that the offspring in each nest have only one mother, so there is no evidence of brood parasitism by female nest searchers. Male Dawson’s burrowing bees tend to emerge on average 15 days earlier than females, showing marked protandry (Alcock 1997b). This pattern is typical for solitary bees and wasps and for other insect groups such as butterflies, mayflies and mosquitoes (Thornhill & Alcock 1983). Protandry is highly correlated with monandry in females throughout the solitary Hymenoptera (Thornhill & Alcock 1983), and Dawson’s burrowing bee is consistent with this pattern. In addition, this species exhibits intense intraspecific competition between males for access to unmated females (Alcock 1997a), and this too supports the hypothesis that protandry is an adaptive reproductive strategy adopted by males to acquire matings with newly emergent receptive females (Wiklund & Fagerström 1977). Dawson’s burrowing bee, in common with other Hymenoptera, can control both the sex and size of their offspring via haplodiploid sex determination, and the amount of resource provided to the brood respectively. In addition to the typical sexual dimorphism of large females and small males, female Dawson’s burrowing bees also produce large males. This male dimorphism has been observed in previous studies of brood cell volumes (Tomkins et al. 2001) and mature larval weights (Alcock et al. 2005). In this study, we also show the same unimodal distribution for females and bimodal distribution for males, this time at the prepupal stage. In a previous study, it was also observed that there is population wide seasonal variation in female allocation patterns with daughters produced first, followed by daughters and major sons in the middle of the season, and minor sons at the end. This pattern is correlated with a decline in resource availability and increase in foraging trip duration (Tomkins et al.

2001), and a decrease in survival rates of foraging females (Alcock et al. 2005). In this study, we find the same patterns within individual nests in a different population sampled in a different season. This leads us to believe that temporal changes in the production of offspring classes are likely to be intrinsic to the species and not dependent on seasonal or geographical differences in resource availability. But why do females produce minor males at all, given their low prospects of mating at the emergence site? During nesting and the construction of brood chambers, one brood pot is often constructed directly on top of another (Houston 1991). The data presented here indicate that the distribution of sex and morph within these doublets is not random. Instead, large offspring, whether female or male, tend to occur in the bottom cell and either a female or a minor male usually occupies the top cell. Likewise, in certain twig-nesting bees and wasps, males are always placed in the outer cells of these linear nests and females in the inner cells (Krombein 1967). This arrangement is likely an adaptation favoured by protandry, otherwise emerging males would destroy their sisters as they excavate their way out of the nest on emergence (Thornhill & Alcock 1983). A similar arrangement appears to be the case here, with the addition that minor males emerge before major males (Alcock 1997b). It has been observed that during excavation of Dawson’s burrowing bee nests, occasionally a fully eclosed bee is found trapped in the midst of emergence and has died in the process. Therefore, the addition of a minor male on top of a reproductively more important major male or female may be adaptive, allowing the minor male to emerge first and dig a tunnel to allow the larger offspring to emerge unhindered. Such an argument, although appealing, is unlikely to explain fully the production of large numbers of minor male offspring. Recently Peterson and Roitberg (2006) proposed a model of maternal investment in which the parental fitness functions for male and female offspring differed. The models predict that if male and female fitness functions intersect, females should invest in the sex with lower average fitness return when the cost of collecting resources reduces the total amount of investment available per offspring (Peterson & Roitberg 2006). This approach is readily extended to include three offspring classes, as occur in Dawson’s burrowing bee (Fig. 4.4), Although the fitness functions depicted in Fig. 4.4 are hypothetical, we know that males show a steep increase in mating success with increasing body size (Alcock 1996b), that female reproductive success is relatively insensitive to the range of body sizes found in natural populations (Alcock et al. 2006), and that minor male fitness is likely to be very

low because few females are not mated by patrolling major males (Alcock 1996a). The model predicts that females should produce minor males when provisioning costs are high because the fitness return from minor sons, though low, exceeds that from producing daughters or major sons. Thus, the production of minor sons may represent a conditional provisioning strategy. The concept of intersecting fitness functions is not new in the analysis of alternative male morphologies. For example, in models of status dependent selection, individuals adopt the breeding tactic that maximises their fitness given their status. The tactic adopted may well have lower overall fitness, but because of intersecting fitness functions, it provides the individual with its maximum fitness return given the constraint of its status (Gross 1996; Hunt & Simmons 2001). Future work on the alternative sizes and behaviours of Dawson’s burrowing bee needs to quantify the precise nature of the fitness functions for the three offspring classes.

4.6 References

Alcock J (1996a) Provisional rejection of three alternative hypotheses on the maintenance of a size dichotomy in males of Dawson's burrowing bee, Amegilla dawsoni (Apidae, Apinae, Anthophorini). Behavioural Ecology and Sociobiology 39, 181-188. Alcock J (1996b) The relation between male body size, fighting, and mating success in Dawson's burrowing bee, Amegilla dawsoni (Apidae, Apinae, Anthphorini). Journal of Zoology 239, 663-674. Alcock J (1997a) Competition from large males and the alternative mating tactics of small males of Dawson's burrowing bee (Amegilla dawsoni) (Apidae, Apinae, Anthophorini). Journal of Insect Behavior 10, 99-113. Alcock J (1997b) Small males emerge earlier than large males in Dawson's burrowing bee (Amegilla dawsoni) (Hymenoptera: Anthophorini). Journal of Zoology 242, 453-462. Alcock J (1999) The nesting behaviour of Dawson's burrowing bee, Amegilla dawsoni (Hymenoptera: Anthophorini), and the production of offspring of different sizes. Journal of Insect Behaviour 12, 363-384.

Alcock J, Simmons LW, Beveridge M (2005) Seasonal change in offspring sex and size in Dawson's burrowing bees (Amegilla dawsoni) (Hymenoptera: Anthorphorini). Ecological Entomology 30, 247-254. Alcock J, Simmons LW, Beveridge M (2006) Does variation in female body size affect nesting success in Dawson's burrowing bee, Amegilla dawsoni (Apidae: Anthophorini)? Arnqvist G, Nilsson T (2000) The evolution of polyandry: multiple mating and female fitness in insects. Animal Behaviour 60, 145-164. Beveridge M, Simmons LW (2004) Microsatellite loci for Dawson's burrowing bee (Amegilla dawsoni) and their cross-utility in other Amegilla species. Molecular Ecology Notes 4, 379-381. Brown MJF, Schmid-Hempel P (2003) The evolution of female multiple mating in social Hymenoptera. Evolution 57, 2067-2081. Danforth BN, Desjardins CA (1999) Male dimorphism in Perdita portalis (Hymenoptera, Andrenidae) has arisen from preexisting allometric patterns. Insectes Sociaux 46, 18-28. Danforth BN, Neff JL (1992) Male polymorphism and polyethism in Perdita texana (Hymenoptera: Andrenidae). Annals of the Entomological Society of America 85, 616-626. Dieringer D, Schlotterer C (2003) Microsatellite Analyser (MSA): a platform independent analysis tool for large microsatellite data sets. Molecular Ecology Notes 3, 167-169. Emery A, Wilson I, Craig S, Boyle P, Noble L (2001) Assignment of paternity groups without access to parental genotypes: multiple mating and developmental plasticity in squid. Molecular Ecology 10, 1265-1278. Gemmel NJ, Akiyama S (1996) An efficient method for the extraction of DNA from vertebrate tissues. Trends in Genetics 12, 338-339. Griffiths SC, Owens IPF, Thuman KA (2002) Extra pair paternity in birds: a review of interspecific variation and adaptive function. Molecular Ecology 11, 2195-2212. Gross MR (1996) Alternative reproductive strategies and tactics: diversity within sexes. Trends in Ecology and Evolution 11, 92-98. Houston TF (1991) Ecology and behaviour of the bee Amegilla (Asaropoda) dawsoni (Rayment) with notes on a related species (Hymenoptera: Anthophoridae). Records of the Western Australian Museum 15, 535-553.

Hunt J, Simmons L (2001) Status-dependent selection in the dimorphic beetle Onthophagus taurus. Proceedings of the Royal Society of London Series B - Biological Sciences 268, 2409-2414. Jones A G (2001) GERUD1.0: a computer program for the reconstruction of parental genotypes from progeny arrays using multilocus DNA data. Molecular Ecology Notes 1, 215-218. Krombein KV (1967) Trap-nesting wasps and bees: life histories, nests, and associations. Smithsonian Press, Washington, D. C. Kronauer DJC, Schoning C, Pedersen JS, Boomsma JJ, Dadau J (2004) Extreme queen- mating frequency and colony fission in African army ants. Molecular Ecology 13, 2381-2388 Moilanen A, Sundström L, Pedersen JS (2004) MATESOFT: a program for deducing parental genotypes and estimating mating system statistics in haplodiploid species. Molecular Ecology Notes 4, 795-797. Page RE (1986) Sperm utilization in social insects. Annual Review of Entomology 31, 297-320. Pedersen JS, Boomsma JJ (1999a) Multiple paternity in social Hymenoptera: estimating the effective mate number in single-double mating populations. Molecular Ecology 8, 577-587. Pedersen JS, Boomsma JJ (1999b) Positive association of queen number and queen- mating frequency in Myrmica ants: a challenge to the genetic-variability hypotheses. Behavioral Ecology and Sociobiology 45, 185-193. Peters JM, Queller DC, Imperatriz-Fonseca VL, Roubik DW, Strassmann JE (1999) Mate number, kin selection and social conflicts in stingless bees and honeybees. Proceedings of the Royal Society of London B 266, 379-384. Peterson JH, Roitberg BD (2006) Impacts of flight distance on sex ratio and resource allocation to offspring in the leafcutter bee, Megachile rotundata. Behavioral Ecology and Sociobiology 59, 589-596. Queller DC, Goodnight KF (1989) Estimating relatedness using genetic markers. Evolution 43, 258-275. Schmid-Hempel R, Schmid-Hempel P (2000) Female mating frequencies in Bombus spp. from Central Europe. Insectes Sociaux 47, 36-41.

Simmons L, Alcock J, Reeder A (2003) The role of cuticular hydrocarbons in male attraction and repulsion by female Dawson's burrowing bee, Amegilla dawsoni. Animal Behaviour 66, 677-685. Simmons LW, Tomkins JL, Alcock J (2000) Can minor males of Dawson's burrowing bee, Amegilla dawsoni (Hymenoptera: Anthophorini) compensate for reduced access to virgin females through sperm competition? Behavioral Ecology 11, 319-325. Strassmann J (2001) The rarity of multiple mating by females in the social Hymenoptera. Insectes Sociaux 48, 1-13. Stubblefield JW, Seger J (1994) Sexual dimorphism in the Hymenoptera. In: The Differences Between the Sexes (eds. Short RV, Balaban E), pp. 71-103. Cambridge University Press, Cambridge. Thornhill R, Alcock J (1983) The Evolution of Insect Mating Systems Harvard University Press, Cambridge, Massachusetts. Tomkins JL, Simmons LW, Alcock J (2001) Brood-provisioning strategies in Dawson's burrowing bee, Amegilla dawsoni (Hymenoptera: Anthophorini). Behavioural Ecology and Sociobiology 50, 81-89. Torchio PF, Tepedino VJ (1980) Sex ratio, body size and seasonality in a solitary bee, Osmia lignaria propinqua Cresson (Hymenoptera: Megachilidae). Evolution 34, 993-1003. Weber JL, Wong C (1993) Mutations of human short tandem repeats. Human Molecular Genetics 2, 1123-1128. Wiklund C, Fagerström T (1977) Why do males emerge before females? A hypothesis to explain the incidence of protandry in butterflies. Oecologia 31, 153-158. Wuellner CT (1999) Alternative reproductive strategies of a gregarious ground-nesting bee, Dieunomia triangulifera (Hymenoptera: Halictidae). Journal of Insect Behaviour 12, 845-863. Zavodna M, Compton SG, Raja S, Gilmartin PM, van Damme JMM (2005) Do fig wasps produce mixed paternity clutches? Journal of Insect Behaviour 18, 351- 362.

Figure 4.1 Frequency distributions of male (Fig. 4.1.1) and female (Fig. 4.1.2) prepupae of Dawson’s burrowing bee.

30 Males

0 .4 .6 .8 1 1.2 1.4 1.6 1.8

16 Females

0 .95 1 1.05 1.1 1.15 1.2 1.25 1.3 1.35 1.4 1.45 Weight of prepupae (g)

Figure 4.2.1 Frequency distribution (right hand y axis) of males (top of figure) and females (bottom of figure) through the nest sequence for 29 nests of Dawson’s burrowing bee and the predicted probability of being male where male = 1 and female = 0 (left hand y axis).

Frequency of males

Frequency of females

Figure 4.2.2 Frequency distribution (right hand y axis) of minor males (top of figure) and major males (bottom of figure) through the nest sequence for 29 nests of Dawson’s burrowing bee and the predicted probability of being minor where minor = 1 and major = 0 (left hand y axis).

Frequency of minors

Frequency of majors

Figure 4.3 Frequency of doublets containing female, major male and minor male in all combinations for 61 doublets found in 26 nests of Dawson’s burrowing bee.

Top cell Female 20 Minor male Major male

0 123 Female Major male Minor male

Bottom cell

Figure 4.4 Hypothetical fitness functions describing the increase in fitness returns from provisioning daughters (solid bold curve), major sons (dashed bold curve), and minor sons (dashed curve). Major sons and daughters show a steep increase in mating success with increasing provisions that translate into offspring size. The optimal investments in minor sons (mi), females (F) and major sons (Mj) are located at the maximum rate of gain for that offspring class. Given enough resources females should invest in major sons because they offer the greatest fitness returns. As resources become limited they should switch to investment in daughters because they offer greater returns than small major sons. As resources become more limited they should switch to producing minor sons who offer greater returns than small daughters. The functions as depicted are consistent with the observation that major males are on average larger than females, and that males of intermediate size are never seen in the population because females producing this size class of offspring have greater returns if they produce females.

Fitness return per offspring

mi F Mj

Maternal provisioning (offspring size)

Table 4.1 Number of effective mates for 28 female Dawson’s burrowing bees estimated using PARENTAGE, GERUD, and allele counting. For the PARENTAGE estimate, P (mode) is the observed proportion of samples with the modal number of fathers from 5000 simulations. The GERUD estimate is the minimum and maximum number of fathers estimated from the data. For allele counting, the estimate is based on the locus with the greatest number of non-maternal alleles.

PARENTAGE GERUD Allele Female Mean 95% CL Mode P (mode) Min-Max counting 1 1.001 0.001 1 0.999 1 1 2 1.001 0.001 1 0.999 1 1 3 1.000 0.000 1 1.000 1 1 4 1.000 0.000 1 1.000 1 1 5 1.002 0.001 1 0.998 1 1 6 1.000 0.000 1 1.000 1 1 7 1.000 0.001 1 1.000 1 1 8 1.000 0.000 1 1.000 1 1 9 1.003 0.001 1 0.997 1-2 1 10 1.000 0.000 1 1.000 1 1 11 1.001 0.001 1 0.999 1 1 12 1.000 0.000 1 1.000 1 1 13 1.000 0.000 1 1.000 1 1 14 1.000 0.000 1 1.000 1 1 15 1.002 0.001 1 0.998 1-2 1 16 1.000 0.001 1 1.000 1 1 17 1.000 0.000 1 1.000 1 1 18 1.002 0.001 1 0.998 1 1 19 1.000 0.000 1 1.000 1 1 20 1.001 0.001 1 0.999 1 1 21 1.000 0.000 1 1.000 1 1 22 1.000 0.000 1 1.000 1-2 1 23 1.000 0.000 1 1.000 1 1 24 1.001 0.001 1 0.999 1-4 1 25 1.000 0.000 1 1.000 1 1 26 1.001 0.001 1 0.999 1 1 27 1.000 0.000 1 1.000 1-2 1 28 1.000 0.000 1 1.000 1-2 1

CHAPTER 5

GENERAL DISCUSSION

5.1 Microsatellites as a tool

It is clear from the results obtained in this thesis that microsatellites are an invaluable tool for evolutionary studies of a species’ behaviour and ecology. In Chapter 2 I demonstrated the use of the most up-to-date method for isolating microsatellites involving an enrichment step (Zane et al. 2002). This step simplifies and shortens the process of isolating large numbers of microsatellites which then need only to be characterized and optimized. I also demonstrated the usefulness of these loci in three other species of Amegilla (Table 2.2.). Microsatellites are co-dominant, highly variable molecular markers and are widely used to study population genetics (Schlotterer & Pemberton 1998). Other molecular markers can also be used for this type of study, including allozymes (Van Bortel et al. 2003) and amplified fragment length polymorphisms (Takami et al. 2004). In a recent comparison of allozymes and microsatellites to measure gene flow in a garter snake, the two markers produced very similar results (Bittner & King 2003). The only difference was in the amount of effort involved with 11 allozyme loci screened (2 – 10 alleles per locus) compared to 4 loci screened with microsatellites (21 – 36 alleles per locus). In another comparison of 4 neutral molecular markers; allozymes, co-dominant random amplified polymorphic DNAs (RAPDs), microsatellites and dominant RAPDs, patterns of structure were comparable between the markers, although microsatellites were found to be the most effective for detecting structure at the lowest level of population subdivision (Ross et al. 1999). In Chapter 3 I used eight of the most variable microsatellite loci developed in Chapter 2 to examine the population genetics of Amegilla dawsoni. The mating and nesting behaviour of this species suggests that gene flow would be restricted by probable monandry and the fact that almost 90% of females mate immediately on emergence. However, analysis of the microsatellite data showed there to be no genetic structure between the populations and that gene flow was sufficient to maintain panmixia. This might be explained by the fact that rainfall in the species range is infrequent and unreliable which could cause bees from different populations to congregate at limited resources. The small number of unmated females from one population could then come into contact with males from another population. In addition, drought could eliminate food resources from an emergence site causing the entire population to move elsewhere, increasing gene flow across the species range.

Due to their highly variable nature, microsatellites are also used extensively in kinship studies because, with enough loci, even highly related individuals can be distinguished (Queller et al. 1993). This utility was highlighted in Chapter 4 where I determined the number of effective mates for a female bee by analysing the genotypes of her offspring. In the case of Dawson’s burrowing bee the number of effective mates was estimated at one. The data were also used to show that all the genotypes in a nest were consistent with the nesting female, providing no evidence for brood parasitism by nest searching females that frequent nesting aggregations of this species. In a previous study, microsatellites were used to sex the immature offspring of a population of nesting bees using the haplo-diploid nature of Hymenopterans (Alcock et al. 2005). In Chapter 4, this method was used again to determine the sex of offspring in order to examine the pattern of investment in males (major and minor) and females throughout a season in individual nests. The pattern was comparable with the previous study where nesting females made large females first, followed by major males and then minor males. This investment pattern was consistent in two different populations and from season to season and so is unlikely to depend on seasonal or geographical differences in resource availability.

5.2 Molecular markers of the future

Although microsatellites have proved to be the best tool to answer the questions posed in this thesis, they do have certain limitations. Mutations, insertions, or deletions in the primer binding site lead to non-amplification and the presence of null alleles. If they remain unidentified, null alleles can lead to an apparent surplus of homozygotes in a population. They can also lead to errors in parentage determination (Dakin & Avise 2004). A simple redesign of the PCR primer is usually all that is required. Another limitation that is more difficult to solve is that of a species in which few microsatellites can be found (Fagerberg et al. 2001). For example, using the same method as described here in Chapter 2, I isolated 19 microsatellite loci for the dung beetle species Onthophagus taurus. Only 7 of these loci were polymorphic with the number of alleles ranging from 2 to 4 which is insufficient variability to make them useful for evolutionary studies such as those reported in this thesis (Beveridge, unpublished). The use of microsatellites to infer phylogenetic relationships among taxa has been criticised due to size homoplasy and genetic distance calculations based on

assumptions about the mutational model for microsatellites which may not be true (Rosenbaum & Deinard 1998). Size homoplasy occurs when microsatellite electromorphs (PCR products of the same size) arise from independent mutational events (Estoup et al. 1995). These alleles are not identical by descent (IBD) and if they remain undetected, can lead to inaccurate measures of genetic distance. The sequencing of mitochondrial DNA has proved far more useful for phylogenetic analyses (Villalba et al. 2002). To overcome some of these limitations, new molecular markers are now becoming more widespread. Single nucleotide polymorphisms (SNPs) are point mutations in coding and non-coding regions of the genome that evolve in a manner that can be described by a simple mutation model (Morin et al. 2004). They are abundant and widespread with SNPs occurring every 300 to 1000 base pairs in many species compared to every 5000 to 50,000 base pairs for microsatellites. The process of finding SNPs within a genome involves sequencing genome segments from multiple individuals, where the screening of 75 to 100 genome segments should yield more than 50 independent SNPs. There are several methods for SNP genotyping including the more traditional gel electrophoresis and newer methods such as microarrays and fluorescence polarization which require additional laboratory equipment. SNPs have been used in population genetics to estimate genetic variation, genetic distances and effective population sizes (Morin et al. 2004). However, they are less useful for parentage and relatedness analyses where simulations have estimated that even 100 SNPs might not be sufficient to distinguish relationships other than parent-offspring pairs (Glaubitz et al. 2003). In addition to new markers, there are also more applications for existing markers. The last decade has seen the widespread use of molecular markers to study quantitative traits. These are traits under the control of many genes which often show continuous variation within and among populations (Erickson et al. 2004). The evolution of life history, behavioural and morphological traits is thought to be the result of evolution at many loci and so, the genetic analysis of these traits may contribute to our understanding of evolutionary processes. The use of genetic markers to examine genetic architecture is called quantitative trait locus (QTL) analysis. Microsatellites and amplified fragment length polymorphisms (AFLPs) are the markers most commonly used for QTL analysis but SNPs and expressed sequence tag polymorphisms (ESTs) are becoming more popular as the cost of sequencing is reduced. EST-based markers

utilize sequence differences within genes and so can be targeted to genes of interest (Erickson et al. 2004). So, with more molecular markers available and a greater number of applications than ever, the study of the ecology and evolution of a species can be far more detailed than was thought possible just a decade ago.

5.3 References

Alcock J, Simmons LW, Beveridge M (2005) Seasonal change in offspring sex and size in Dawson's burrowing bees (Amegilla dawsoni) (Hymenoptera: Anthorphorini). Ecological Entomology 30, 247-254. Bittner TD, King RB (2003) Gene flow and melanism in garter snakes revisited: a comparison of molecular markers and island vs. coalescent models. Biological Journal of the Linnean Society 79, 389-399. Dakin EE, Avise JC (2004) Microsatellite null alleles in parentage analysis. Heredity 93, 504-509. Erickson DL, Fenster CB, Stenoien HK, Price D (2004) Quantitative trait locus analyses and the study of evolutionary process. Molecular Ecology 13, 2505-2522. Estoup A, Tailliez C, Cornuet J-M, Solignac M (1995) Size homoplasy and mutational processes of interrupted microsatellites in two bee species, Apis mellifera and Bombus terrestris (Apidae). Molecular Biology and Evolution 12, 1074-1084. Fagerberg AJ, Fulton RE, Black IV WC (2001) Microsatellite loci are not abundant in all arthropod genomes: analysis in the hard tick, Ixodes scapularis and the yellow fever mosquito, Aedes aegypti. Insect Molecular Ecology 10, 225-236. Glaubitz JC, Rhodes JR E, Dewoody JA (2003) Prospects for inferring pairwise relationships with single nucleotide polymorphisms. Molecular Ecology 12, 1039-1047. Morin PA, Luikart G, Wayne RK, group Sw (2004) SNPs in ecology, evolution and conservation. Trends in Ecology and Evolution 19, 208-216. Queller DC, Strassman JE, Hughes CR (1993) Microsatellites and Kinship. Trends in Ecology and Evolution 8, 285-288. Rosenbaum HC, Deinard AS (1998) Caution before claim: an overview of microsatellite analysis in ecology and evolutionary biology. In: Molecular

Approaches to Ecology and Evolution (eds. DeSalle R, Schierwater B), pp. 87- 106. Birkhauser Verlag, Basel. Ross KG, Shoemaker DD, Krieger MJB, DeHeer CJ, Keller L (1999) Assessing genetic structure with multiple classes of molecular markers: A case study involving the introduced fire ant Solenopsis invicta. Molecular Biology and Evolution 16, 525- 543. Schlotterer C, Pemberton J (1998) The use of microsatellites for genetic analysis of natural populations - a critical review. In: Molecular Approaches to Ecology and Evolution (eds. DeSalle R, Schierwater B), pp. 71-86. Birkhauser Verlag, Basel. Takami Y, Koshio C, Ishii M, et al. (2004) Genetic diversity and structure of urban populations of Pieris butterflies assessed using amplified fragment length polymorphism. Molecular Ecology 13, 245-258. Van Bortel W, Trung HD, Roelants P, Backeljau T, Coosemans M (2003) Population genetic structure of the malarial vector Anopheles minimus A in Vietnam. Heredity 91, 487-493. Villalba S, Lobo JM, Martin-Piera F, Zardoya R (2002) Phylogenetic relationships of Iberian dung beetles (coleoptera: Scarabaeinae): insights on the evolution of nesting behaviour. Journal of Molecular Evolution 55, 116-126. Zane L, Bargelloni L, Patarnello T (2002) Strategies for microsatellite isolation: a review. Molecular Ecology 11, 1-16.