Tribolium Castaneum)

Characterisation of variables associated with the flight behaviour of the rust-red flour beetle (Tribolium castaneum)

Komal Gurdasani Master of Biotechnology The University of Queensland

A thesis submitted for the degree of Doctor of Philosophy at The University of Queensland in 2017 School of Biological Sciences

ii Abstract

Tribolium castaneum Herbst (Coleoptera: Tenebrionidae) causes significant losses to stored grain products. This problem is aggravated by the development and spread of resistance to various grain protectants, especially the fumigant phosphine. Studying the dispersal behaviour of T. castaneum will help in understanding the spread of resistance. Active dispersal in T. castaneum occurs by both walking and flight. Flight enables dispersal across a wider geographical area and increases the probability of spreading resistance genes geographically. Dispersal of pests also has implications for understanding the ecology of resident populations and thus also for improving food security effectively. Therefore, a good understanding of T. castaneum dispersal by flight is critical to managing this pest and its resistance to phosphine.

The research reported here was designed to study factors that influence the flight of these long-lived beetles. Various techniques were employed, in both the laboratory and field, to investigate a broad range of factors that may influence T. castaneum flight behaviour. I also explored the use of nanoparticles as novel external markers for small insects, with an emphasis on their use in mark- recapture studies of T. castaneum. The ability to gauge the dispersal distance of these insects accurately would aid considerably in understanding the spatiotemporal dynamics (and gene flow) of this species across landscapes.

Chapter 2 investigated the flight propensity of T. castaneum at two different ages at the same time (old (~12 weeks) and young (~2 weeks post-emergence)). First generation laboratory cultures of field-collected beetles were used to avoid the known effects of protracted laboratory culturing. Flying beetles were captured individually using a customized trap, and their age, sex, weight, size, fat content, fecundity, polyandry and lipofuscin intensity (as an estimate of age) were determined. Beetles were also collected from the culture medium and considered to be resident (non-flyers). The assays indicated that flight propensity declined significantly with age as more young beetles took to flight than old ones. Flight was not affected by sex, body size or weight but lipofuscin intensity of old beetles was similar to young beetles in culture, so it is not an indicator of age. Since, its levels were only significantly higher in young flying females than young resident females it may be related to the individual’s metabolic status. Fecundity assessments indicate that the total fecundity was not related to age or flight as no significant differences were found between resident and flying beetles across and between all age groups.

Microsatellite markers were used to assess parentage of offspring. They confirmed that females of

iii this species are polyandrous in the field (as expected from their polyandry in laboratory cultures). The number of males inseminating females in the field ranged from one to four, and this held too for first generation laboratory cultures (which is reported in Chapter 3 as part of a larger study of fecundity, polyandry and genetic diversity).

The results from Chapters 2 & 3 were used to make predictions about the migration behaviour of T. castaneum in the field (Chapter 4). These predictions were tested by sampling beetles flying at various distances (2, 20 and 300 m) from a large field population (infesting cotton seed in storage). Beetles were also collected directly from the infested cotton seed (‘resident beetles’). The total fecundity of females until day 105, post emigration in all treatments was not statistically different. The females that emigrated are presumed to be relatively young and had mated multiple times as they had the potential to produce numerous offspring (mean = 307.3, range = 0 – 582) without having access to males. Difference in fat content was significant between flying beetles (both sexes) trapped at 2 m and flying beetles trapped at 20 m and 300 m from storage. However, no significant trends were observed between resident and flying beetles. The relationship of flight propensity with weight, body size, and lipofuscin intensity were similar to the trends documented in the laboratory flight experiment.

In chapter 5, I tested green fluorescent nanoparticles (CdTe/CdS quantum dots) as external markers on T. castaneum and other small insects for mark-recapture studies. Quantum dots (QDs) proved to be effective markers in the laboratory for at least 48 hours when beetles were kept without food. The culture medium seemed to be abrasive and so removed the particles. In the laboratory, QDs had no negative effects on the flight behaviour or survival of T. castaneum. In a field test, beetles treated with QDs were captured in good numbers at 20 m from the release point but the fluorescent signals from these recaptures were low, perhaps because the QDs were quenched by the release of stress chemicals (benzoquinones).

Four main conclusions are drawn from the research outlined above. (i) Flight propensity of T. castaneum is related to age, but is not affected by sex, weight, body size or fat content. (ii) Lipofuscin levels may be related to metabolic status rather than age. (iii) Females that fly have a high potential fecundity and considerable paternal genetic diversity in their spermathecae, as females are polyandrous before flight and this increases the chances of spreading phosphine resistance geographically. (iv) Quantum dots are non-toxic and safe as external markers in the laboratory, and further designing should make them effective in the field, so they can be used in large scale mark-recapture studies.

iv Declaration by author

This thesis is composed of my original work, and contains no material previously published or written by another person except where due reference has been made in the text. I have clearly stated the contribution by others to jointly-authored works that I have included in my thesis.

I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance, survey design, data analysis, significant technical procedures, professional editorial advice, and any other original research work used or reported in my thesis. The content of my thesis is the result of work I have carried out since the commencement of my research higher degree candidature and does not include a substantial part of work that has been submitted to qualify for the award of any other degree or diploma in any university or other tertiary institution. I have clearly stated which parts of my thesis, if any, have been submitted to qualify for another award.

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and, subject to the policy and procedures of The University of Queensland, the thesis be made available for research and study in accordance with the Copyright Act 1968 unless a period of embargo has been approved by the Dean of the Graduate School.

I acknowledge that copyright of all material contained in my thesis resides with the copyright holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright holder to reproduce material in this thesis.

v Publications during candidature Peer-reviewed papers

Rafter, M.A., McCulloch, G.A., Daglish, G.J., Gurdasani, K. &Walter, G.H. (2017). Polyandry, genetic diversity and fecundity of emigrating beetles: understanding new foci of infestation and selection. Journal of Pest Science. doi: 10.1007/s10340-017-0902-8

Publications included in this thesis

Contributor Statement of contribution Gurdasani, K. Molecular genetics (50%) Maintenance of laboratory cultures (100%) Data analysis (10%)

Rafter, M.A. Designed experiments (60%) Data analysis (60%) Wrote the paper (70%)

Daglish, G.J. Designed experiments (10%) Wrote the paper (10%)

McCulloch, G.A. Molecular genetics (50%) Designed experiments (20%) Data analysis (30%) Wrote the paper (10%)

Walter, G.H. Designed experiments (10%) Wrote the paper (10%)

vi Contributions by others to the thesis

Professor Gimme Walter, Dr. Michelle Rafter and Dr. Gregory Daglish contributed significantly to the conception, designing and development of this research work. Statistical guidance, critical review, troubleshooting and advice for data analyses and interpretation were also provided.

My colleague Dr. Graham McCulloch contributed significantly by providing training and guidance for molecular methods and computer analyses for polyandry detection in Tribolium castaneum and Rhyzopertha dominica. This work generated a jointly authored manuscript along with Dr. Michelle Rafter, Dr. Gregory Daglish and Dr. Gimme Walter as co-authors. This work has been listed as Chapter 3 of this thesis and is currently published in the Journal of Pest Science. My contribution was the characterization of polyandry in reference laboratory strains and newly established laboratory cultures of T. castaneum and Rhyzopertha dominica from field-collected insects. I analysed the number of paternal male contributors. The molecular skills acquired were then used to infer polyandry in field populations of T. castaneum as described in Chapter 4. For this reason and because the chapter provides a major connecting link between my studies of T. castaneum in the laboratory and field, it has been included in the main body of this thesis. A detailed contribution is listed as a declaration included in Chapter 3.

Dr. Li Li from the Australian Institute for Bioengineering and Nanotechnology contributed significantly by providing training in design, development and analyses of fluorescent nanoparticles (quantum dots). This work has been listed as a part of Chapter 5.

Dr. Richard Webb from the Centre for Microscopy and Microanalysis contributed by providing training for fluorescence microscopy for fluorescence imaging.

Statement of parts of the thesis submitted to qualify for the award of another degree

None

vii Acknowledgements

I would like to sincerely thank my Principal supervisor and mentor Dr. Gimme Walter for giving me this opportunity of a lifetime to complete a project under his guidance. I also owe him my sincere gratitude for always encouraging, guiding and supporting me in every phase of this journey. I would also like to thank him for the early morning coffee meetings that helped to develop my ecological knowledge and concepts (and also birds).

My heartfelt thanks to all my supervisors Dr. Gimme Walter, Dr. Michelle Rafter and Dr. Gregory Daglish for their valuable feedback and constructive comments on writing, experimental design and statistical analyses. Thank you Michelle for all the support and guidance and also for introducing little Damon amidst all the beetle experiments.

I would also, like to acknowledge the Walter lab members for their friendship, support and constructive feedback. A sincere thanks to all the postdoctoral fellows of the Walter lab, especially Dr. Graham McCulloch for providing training on molecular methods. Also, thanks to Sharon, Daisy, and James and to all those members who volunteered in the field and the laboratory. A sincere thanks to Cody for managing everything so efficiently and for the delicious cakes (viruses, Tribolium and what not).

I acknowledge The University of Queensland and the School of Biological Sciences for supporting me financially throughout this program. My deepest gratitude to the readers, Dr. Bronwen Cribb and Dr. Karyn Johnson for providing valuable comments during the PhD milestones.

I extend my gratefulness to Dr. Li Li, Dr. Richard Webb, Dr. Arjun Seth and Dr. Karine Mardon for their support in training and access of various laboratory equipment.

Last but not the least, I am very grateful to have the most supportive family members especially my parents and husband who have encouraged me at every stage of my PhD journey and life. I wouldn’t be here without their love and positivity. Special thanks to my dear husband Jugal, for being the always available volunteer and spending nights in the laboratory with me, I would have not completed this without you. You have been the pillar who always stood by my side no matter what the situation was. Thanks to Namu, Juhi, Didis and Bua for taking care of me emotionally. My heartfelt thanks to my friends Dr. Rokhsareh and Wana (also from the Walter lab) for their love and friendship during tough times. “A friend in need is a friend indeed”. viii

ix Keywords Tribolium castaneum, quantum dots (QDs), flight propensity, lipofuscin, polyandry, phosphine resistance.

Australian and New Zealand Standard Research Classifications (ANZSRC)

ANZSRC code: 050103 Invasive Species Ecology, 30% ANZSRC code: 060411 Population, Ecological and Evolutionary Genetics, 20% ANZSRC code: 070308 Crop and Pasture Protection (Pests, Diseases and Weeds), 50%

Fields of Research (FoR) Classification

FoR code: 0501, Ecological Applications, 30% FoR code: 0604, Genetics, 20% FoR code: 0703, Crop and Pasture Protection, 50%

Frontispiece.

A. Tribolium castaneum adults in a cotton seed shed, B. Marked elytron of callow Tribolium castaneum, C. Central release bin for mark-recapture, D. Funnel trap outside cotton seed shed, E. Cotton seed pile in a storage shed, F. Flight apparatus for laboratory flight test, G. Plastic cups with flour and adult offspring for fecundity tests.

xi Table of Contents

Abstract ...... iii

Contributions by others to the thesis ...... vii

Acknowledgements...... viii

Frontispiece...... xi

List of Figures ...... xvi

List of Tables ...... xx

List of Abbreviations used in the thesis ...... xxi

CHAPTER 1 ...... 1

General introduction ...... 1 1.1. Impact of insect pest infestations on food security ...... 2 1.2. Study Species: Tribolium castaneum Herbst (Coleoptera: Tenebrionidae) ...... 3 1.3. Role of fumigants and their limitations ...... 5 1.3.1. Phosphine ...... 6 1.4. Phosphine resistance – significance and management ...... 7 1.5. Characterization of Tribolium castaneum migrants – a step towards understanding its spatiotemporal dynamics ...... 9 1.6. Thesis structure and approaches to characterize flight in Tribolium castaneum ...... 11 1.6.1. Genetic studies ...... 11 1.6.2. Fecundity, morphometrics and fat content ...... 11 1.6.3. Markers and why to use them ...... 12 1.6.4. Markers used in this study ...... 12 1.6.5. Thesis structure...... 13

CHAPTER 2 ...... 15

Characterizing Tribolium castaneum adults that initiate flight in laboratory tests – generating predictions for the field...... 15 2.1. Introduction ...... 16 2.2. Materials and Methods ...... 18 2.2.1. Culturing field collected Tribolium castaneum...... 18 2.2.2. Flight apparatus ...... 18 2.2.3. Flight assay ...... 19

xii 2.2.4. Beetle characteristics ...... 20 2.2.5. Statistics ...... 21 2.3. Results ...... 22 2.3.1. Relation between age, sex and flight ...... 22 2.3.2. Time of flight initiation ...... 24 2.3.3. Relationship between age, weight and sex of flying and resident beetles...... 26 2.3.4. Survival rates and mean 90 day fecundity of flying and resident beetles ...... 27 2.3.5. Mean fat content ...... 29 2.3.6. Mophometrics ...... 30 2.3.7. Lipofuscin ...... 31 2.4. Discussion ...... 32

CHAPTER 3 ...... 35

Polyandry, genetic diversity and fecundity of emigrating beetles - understanding new foci of infestation and selection ...... 35 Declaration ...... 36 3.1. Introduction ...... 37 3.2. Materials and methods ...... 39 3.2.1. Field sample collections ...... 39 3.2.2. Survival and fecundity of field collected beetles ...... 39 3.2.3. Cultured field beetles for parentage analysis...... 40 3.2.4. Genotyping and paternity analysis ...... 41 3.2.5. Statistics ...... 42 3.3. Results ...... 43 3.3.1. Tribolium castaneum ...... 43 3.3.2. Rhyzopertha dominica ...... 45 3.4. Discussion ...... 51

CHAPTER 4 ...... 54

The dispersal flight of Tribolium castaneum – a field test of laboratory generated predictions ...... 54 4.1. Introduction ...... 55 4.2. Materials and Methods ...... 56 4.2.1 Field tests ...... 56 4.2.2. Measuring fecundity and survival ...... 57 4.2.3. Polyandry analysis ...... 58 4.2.4. Beetle characteristics ...... 58

xiii 4.2.5 Statistics ...... 59 4.3. Results ...... 59 4.3.1. Survival ...... 59 4.3.2. Fecundity ...... 61 4.3.3. Beetle weight and size ...... 64 4.3.4. Fat content ...... 65 4.3.5. Lipofuscin ...... 66 4.3.6. Polyandry ...... 67 4.4. Discussion ...... 69

CHAPTER 5 ...... 72

Nanoparticles as external markers for mark-recapture studies on Tribolium castaneum ...... 72 5.1. Introduction ...... 73 5.2. Materials and Methods ...... 74 5.2.1. Synthesis of hydrophilic CdTe/CdS QDs ...... 74 5.2.2. Marking insects with CdTe/CdS QDs ...... 75 5.2.3. Test for toxicity ...... 75 5.2.4. Test for stability ...... 76 5.2.5. Persistence of QDs on beetles in wheat ...... 76 5.2.6. Flight propensity in the laboratory ...... 76 5.2.7. Mark recapture in the field ...... 77 5.3. Results ...... 78 5.3.1. Distribution of QD markers on insects ...... 78 5.3.2. Test for toxicity ...... 81 5.3.3. Test for stability ...... 81 5.3.4. Persistence of QDs on beetles in wheat ...... 82 5.3.5. Flight propensity of beetles in the laboratory ...... 82 5.3.6. Mark recapture test in the field ...... 82 5.4. Discussion ...... 83

CHAPTER 6 ...... 86

General Discussion – Research findings, conclusions and future directions ...... 86 6.1. Introduction ...... 87 6.2. Research findings and predictions ...... 87 6.2.1. Characteristics of emigrant beetles (laboratory) ...... 87 6.2.2. Characteristics of emigrant beetles (field) ...... 89

xiv 6.2.3. Findings and comparisons of field with laboratory emigrant beetles ...... 90 6.2.4. General conclusions...... 91 6.3. Implications for interpreting the ecology of Tribolium castaneum ...... 92 6.4. Nanoparticles as external markers ...... 93 6.5. Future directions ...... 94

References ...... 96

Appendix ...... 120

xv List of Figures

Figure 2.1 Schematic diagram of the flight apparatus used in laboratory flight tests...... 19

Figure 2.2 The total number of males (black) and females (white) captured flying across the old, young and mixed-age beetle treatments in a) October 2015 (number of males vs. number of females P = 0.14), and b) September 2016 (number of males vs. number of females P = 0.65). The numbers above each bar represent the range in the total number of beetles that flew in each replicate of each treatment for 6 days (2015) and 9 days (2016) of the experiment. The black and white lines within the bars represent the median value...... 23

Figure 2.3 Pattern of flight of Tribolium castaneum across old, young and mixed-age treatments for day 1 of observation, in October 2015...... 24

Figure 2.4 The total number of males (black) and females (white) captured flying in 30 minute intervals prior to sunset. Data across all experimental days in October 2015 combined in the a) old, b) young and c) mixed-age treatments...... 25

Figure 2.5 Beetle mass (mg) and sex of old, young and mixed-age beetles. The y-axis represents beetle mass (mg) and x-axis the sex. Box plots a) old: flying (n = 21♀♀, 16♂♂) and b) resident beetles (n = 17♀♀, 16♂♂); c) young: flying (n = 42♀♀, 27♂♂) and d) resident beetles (n = 30♀♀, 38♂♂); and e) mixed-age: flying (n = 88♀♀, 35♂♂) and (f) resident beetles (n = 64♀♀, 60♂♂). The solid black line in the box represents the median, the whiskers the range, and the circles are outliers...... 26

Figure 2.6 Percent survival of Tribolium castaneum females in old, young and mixed-age treatments until day 90 from the day of capture. Blue lines represent resident females of old: n = 17, young: n = 30, and mixed-age: n = 29 and black lines the flying females of old: n = 21, young: n = 30, and mixed-age: n = 30. Each point on the line represents the percentage of surviving females at the respective 15 day interval. The comparisons old vs. young (residents and flyers) and old flyers vs. mixed-age flyers were significant (P < 0.05)...... 27

Figure 2.7 Mean fecundity of females (±2SE) of Tribolium castaneum females: a) flying (old: grey, n = 21; young: white, n = 30; mixed-age: cross pattern, n = 30), and b) resident (old: grey, n = 17; young: white, n = 30; mixed-age: cross pattern, n = 29) at 15 day time intervals over a 90 day period (P = 0.76 for total fecundity across all treatments)...... 28

xvi Figure 2.8 Mean fat content (±2SE) of Tribolum castaneum captured flying (flyers) and those sampled from the culture medium (residents), collected in September 2016. Treatments tested were old resident beetles (n = 28♀♀, 26♂♂), and young flying (n = 26♀♀, 28♂♂) & resident beetles (n = 28♀♀, 20♂♂). In each treatment the white and the grey bars represent females and males, respectively. Different letters above the relevant bars represent significant relationships...... 29

Figure 2.9 Mean lipofuscin intensity (arbitrary units, ±2SE) of young (captured flying and resident, n = 20) and old (resident, n = 20) Tribolium castaneum beetles of each sex. The number above each bar represents the number of individuals that had negligible fluorescence intensities (0-10). Significant relationships are represented by different letters above each bar. Lower case letters represent post-hoc pairwise comparisons between sexes within a treatment and upper case letters in circles indicate comparisons between treatments...... 32

Figure 3.1 The number of weeks in which progeny were produced by individual females against their total number of progeny produced within the 12 week observation period for: a) Tribolium castaneum collected from storage, b) T. castaneum trapped flying...... 44

Figure 3.2 The number of weeks in which progeny were produced by individual females against their total number of progeny produced within the 12 week observation period for: a) Rhyzopertha dominica collected from storage, b) R. dominica trapped flying...... 46

Figure 3.3 The total number of progeny produced by individual females plotted against the number of males contributing to those progeny for: a) Tribolium castaneum collected from storage (n = 26), b) T. castaneum trapped flying (n = 27), c) T. castaneum females collected from a new laboratory culture (n = 24), d) Rhyzopertha dominica collected from storage (n = 27), e) R. dominica females trapped flying (n = 30), and f) R. dominica females collected from a new laboratory culture (n = 18). The number of males contributing to progeny was estimated using 4-5 microsatellite markers to genotype individual females and 15 of her progeny using GERUD 2.0...... 49

Figure 3.4 The mean proportions (± SE on the right side of each bar) of contributions to paternity by each of four males (labelled a-d because the sequence in which they mated is unknown), in: a) Tribolium castaneum females collected from storage, b) T. castaneum trapped flying, c) T. castaneum females collected from a new laboratory culture, d) Rhyzopertha dominica females collected from storage, e) R. dominica females trapped flying, and f) R. dominica females collected from a new laboratory culture...... 50

xvii Figure 4.1 Percent survival of Tribolium castaneum females in each treatment until day 375 (±95%CI) from the day of capture. The treatments included beetles caught flying within the storage shed (n = 34), flying at a distance of 20 m from the shed (n = 30), flying at a distance of 300 m from the storage shed (n = 30) and resident within the pile of cotton seed (n = 26). P > 0.05, for all treatments...... 60

Figure 4.2 Percent survival, with time, of female beetles caught flying at a distance of 300 m from the shed in which cotton seeds were stored. At day 106 (±95%CI), 14 randomly selected replicates had a male beetle added to the container with the female. The other 11 females continued to be held without a male. The black solid line indicates mortality of those females held with males and the red dotted line the mortality of females held without males...... 60

Figure 4.3 Mean fecundity (±2SE) of Tribolium castaneum beetles, from four treatments, at 15-day intervals after capture. Beetles were caught flying within a cotton seed storage shed (n = 34), caught flying 20 m from the shed (n = 30), caught flying 300 m from the shed (n = 30) or were sampled from the seeds and considered to be resident (n = 26)...... 61

Figure 4.4 Mean fecundity at 15 day intervals from day 91-105 after capture. (±2SE): These are the same females in the same treatments as in Figure 4.3, but with some replicates exposed to a) further mating and b) some not exposed to males...... 63

Figure 4.5 Mean weight (mg) of beetles (±2SE) captured in flight at different distances from a large source of beetles from which, a sample was also taken (= resident on x-axis). Sample sizes are: flying within shed (n = 28 for each sex), flying outside shed (20 m, n = 28♀♀, n = 22♂♂ & 300 m, n = 23♀♀, n = 20♂♂), and resident (n = 26♀♀, n = 25♂♂). All treatments showed no statistical significance (P>0.05)...... 64

Figure 4.6 Mean fat content (mg) of beetles (±2SE) across four treatments. Males (grey bars) and females (white bars) of Tribolium castaneum were captured flying within the shed (n = 28 for each sex), flying outside (20 m) from the shed (n = 28♀♀, n = 22♂♂), flying 300 m from the shed (n=23♀♀, n = 20♂♂), and collected from the cotton pile within the shed as residents (n = 26♀♀, n = 25♂♂). Significant relationships are represented by different letters above each bar. Lower case letters denote males and uppercase letter relate to females, as comparison across sexes was not carried out...... 65

Figure 4.7 Mean lipofuscin intensities (arbitrary units, ±2SE) of Tribolium castaneum males (grey bars) and females (white bars) from three treatments. Beetles were collected from the pile of cotton

xviii seeds as resident (n = 20 ♀♀, n = 18 ♂♂) and captured flying within the shed (n = 20 for each sex) and 20 m from the cotton shed (n = 20 for each sex). The number above each bar represents the number of individuals that had negligible fluorescence intensities (i.e., between 0-10 units). Significantly different values are indicated by different letters above bars...... 66

Figure 4.8 Total number of progeny produced by individual female Tribolium castaneum until day 105 (after capture) plotted against the estimated number of fathers contributing to the progeny they produced between days 16 and 30 (n = 18 mothers and n = 15 progeny per mother in each treatment). The female beetles were captured in flight near a large population in a cotton seed storage or sampled from seeds: a) flying within shed, b) flying outside shed (20 m), c) flying outside shed (300 m), d) resident. The numbers of males were estimated by using four microsatellite markers...... 68

Figure 5.1 Fluorescence of quantum dots applied to Tribolium castaneum beetles. From left to right (a) unmarked T. castaneum ventral; (b, c) marked T. castaneum ventral; (d) marked elytron of callow adult; (e) unmarked dorsal surface of callow T. castaneum; and (f) marked callow dorsal. Visualisation was at an excitation level of λ = 420–495 nm. Magnifications used = 4X (a, b, e) and 10X (c, d, f) and exposure time 0.1–0.5 seconds...... 79

Figure 5.2 Fluorescence of quantum dots applied to Bactrocera tryoni. From left to right B. tryoni (a) unmarked head; (b) marked head; (c) unmarked wing; (d) marked wing; (e) unmarked leg; (f) marked leg. Visualisation was at an excitation level of λ = 420–495 nm. Magnification used = 4X and exposure time 0.1–0.5 seconds...... 80

Figure 5.3 Mean fluorescence intensity (arbitrary units, ±2SE) of the wash off obtained from Tribolium castaneum beetles 48 hours post treatment with CdTe/CdS quantum dots (QDs), recorded at λemission = 500nm. The bars represent marked beetles kept without wheat (grey), marked beetles held on wheat (maroon), and control beetles not marked with QDs (white dotted pattern)...... 81

Figure 5.4 Mean fluorescent intensity (arbitrary units, ±2SE) of the wash off obtained from Tribolium castaneum beetles marked with CdTe/CdS quantum dots (QDs): laboratory (beetles 48 hours post treatment beetles without wheat, grey bar), captured in the field (white bar) at λemission = 500 nm, and control beetles not marked with QDs (white dotted pattern)...... 83

xix List of Tables

Table 2.1 Morphometrics (in mm) of Tribolium castaneum from flying and resident males and females of old, young and mixed-age treatments...... 30

Table 3.1 Survival (%, sexes combined (see text)) of Tribolium castaneum six and twelve weeks after collection from the field (either in flight or from storage)...... 43

Table 3.2 Survival (%, sexes combined (see text)) of Rhyzopertha dominica six and twelve weeks after collection from the field (either in flight or from storage)...... 46

Table 3.3 Characteristics of five microsatellite markers used to infer paternity in field collected female beetles and in female beetles from new laboratory populations of Tribolium castaneum and

Rhyzopertha dominica. NA = number of alleles. HO = observed heterozygosity. HE = expected heterozygosity. PEXC = exclusion probability...... 48

Table 4.1 Comparison of fecundity at 0-15 days and 16-30 days after capture by multiple t-tests. . 62

Table 4.2 Generalised linear model (negative binomial regression) for effects of time on fecundity (offspring production) of females in all treatments ...... 62

Table 4.3 Two-way ANOVA table for morphometrics of male and female Tribolium castaneum beetles captured flying within a cotton storage shed, flying 20 m and 300 m from the cotton shed and collected as residents from the cotton seed*. No significant interactions were observed within and across sexes across all treatments for each body part measured...... 64

Table 4.4 Characteristics of four microsatellite markers used to infer paternity in female Tribolium castaneum beetles collected from a cotton shed (flyers and residents combined as one population)...... 67

xx List of Abbreviations used in the thesis

QDs Quantum dots PBS Phosphate buffered saline DLD Dihydrolipoamide dehydrogenase WHO World Health Organization CdTe/CdS Tellurium coated cadmium quantum dots EDTA Ethylene diamine tetraacetic acid DMD 4,8 dimethyl-decanal UV Ultraviolet

PH3 Phosphine

CdCl2 Cadmium chloride DAF Department of Agriculture Fisheries and Forestry FAO Food and Agriculture Organisation of the United Nations GRDC Grain Research and Development Corporation DNA Deoxyribo nucleic Acid PCR Polymerase chain reaction ANCOVA Analysis of covariance ANOVA Analysis of variance d.f. Degrees of freedom RH Relative humidity GLM Generalized linear model CI Confidence intervals

xxi

CHAPTER 1

General introduction

“To husband is to use with care, to keep, to save, to make last, to conserve. And so it appears that most and perhaps all of industrial agriculture's manifest failures are the result of an attempt to make the land produce without husbandry.” (Berry, 2009, p. 93)

1.1. Impact of insect pest infestations on food security

About one quarter to one third of all grain crops is lost each year worldwide after harvest (Emery, 2014). Low to middle income regions in Africa and Asia (for example, India and China, Mozambique and Nigeria) together account for most of this problem. The proportion of undernourished people in the world attests to the fact that grain loss is a serious threat, and data indicates that almost one in every six children in developing countries is undernourished (WHO, 2012). One of the major reasons for food shortage is pest infestation, especially in rural areas, and measures have been taken by some national governments in some of the most populous countries to curb undernourishment. For example, the “Food Security Act, 2013” has been implemented by the government of India to provide food grains to almost two thirds of its population at a subsidized rate, i.e., rice at 5.5¢ AUD per kg, wheat at 3.7¢ AUD per kg, coarse grain millet at 1.8¢ AUD per kg (Singh, 2013).

According to the Food and Agriculture Organization of the United Nations (FAO), “people will only be food secure when sufficient food is available, they have access to it, and it is well utilized” (FAO, 1996). A number of factors contribute to these general aspects that relate to food security, including poverty, inflation, corruption, lack of infrastructure, climate change, and damage by pests (FAO, 2000). The problem of food security prevails in spite of improvement in food storage facilities around the world, and one of the major problems is still wastage or deterioration as a result of the activity of pests. Insect pests damage most of the protein content (i.e. embryos) of the grains they attack, rendering them unsuitable, or, at least less nutritious for human consumption (Emery, 2014).

Insect pest infestations lead to significant food and economic loss. Besides downgrading the physical and nutritional value of grains, infestation may even render the grain unfit for human consumption. Infestation belittles the efforts to grow, harvest, transport and store food grains. Infestations in grain are likely also to encourage mould growth, thus increasing the chances of toxic or allergenic residues accumulating in the grain. The most concerning allergenic responses are from the mycotoxin producing fungi (Zain, 2011). All of these effects impact on the economy as well, because insect and fungal infestation increases the possibilities of rejection by local or international customers (Baloch, 1999). Food lost through insect and fungal infestation leads to price rises and 2 leaves the poor undernourished. Small businesses and farmers are also impacted because consumers and traders are, to a large extent, intolerant of infested food products. Clearly, grain infestation affects the economy and health status of a major portion of the world’s human population.

1.2. Study Species: Tribolium castaneum Herbst (Coleoptera: Tenebrionidae)

Tribolium castaneum, the rust-red flour beetle, is a pest of stored grain and its products, including wheat flour and processed foods (Sokoloff, 1972). This species has a worldwide distribution and has been listed as the most common food pest of stored products by the United Nations (FAO, 1994). Despite their relatively small size (3-4 mm), these beetles pose a serious threat to worldwide grain reserves. The adult life span ranges from six months up to a year, and sometimes extends to longer than that (Good, 1936; Sokoloff, 1974). Both the adults and larvae damage stored grain and stored grain products, and particularly so under hot and humid conditions (e.g. monsoon conditions in India), as these provide a favourable environment for T. castaneum to grow and reproduce. They are also likely to be attracted by the pheromones of the residing infestation, i.e. other grain pests that act as primary colonizers (Trematerra et al., 2015). Growing T. castaneum populations generate metabolic heat and this results in the accumulation of moisture (Kumari et al., 2011), which, in turn, provides ideal conditions for fungi to proliferate and mycotoxins are thus deposited in stored products (Bosly & Kawanna, 2014). Kumari et al. (2011) characterized the microflora associated with T. castaneum; these include pathogenic bacteria (e.g., Pseudomonas aeruginosa and Staphylococcus aureus) and fungi (Aspergillus niger, Rhizopus oryzae and Fusarium spp.). Fungal infestation usually causes flour to turn greyish and mouldy with a displeasing pungent odour (GRDC, 2011). Besides that, the production of benzoquinones by T. castaneum adults (Senthilkumar et al., 2012; Niu et al., 2016) also affects the elasticity and viscosity of flour, turning it pink (Payne, 1925; Alexander & Barton, 1943; Kumari et al., 2011).

Tribolium castaneum has long had a reputation for being mostly sedentary and most experiments on movement patterns have been conducted by studying the walking behaviour of the adults (Good, 1933; White, 1981; Campbell & Arbogast, 2004; Campbell et al., 2010ab). Campbell et al. (2010a) studied the effects of fumigation (methyl bromide and sulfuryl fluoride) on T. castaneum movement by placing equidistant pitfall traps baited with aggregation pheromone (4,8 dimethyl-decanal) in flourmills for a period of 7 years. They reported that the numbers of individuals caught in traps immediately after fumigation were significantly lower than those caught before fumigation, and concluded by visual assessment that fumigation reduced the number of beetles captured as opposed to seasonal variations (Campbell et al., 2010a). They also found that T. castaneum numbers

3 increased before each fumigation event, which they suggested could be due to survival of the egg and pupal stages, due to places in which the insects hide not being exposed to the fumigant, like small crevices in walls, or because of presence of baseline founding population potentially unaffected by the fumigations (Campbell et al., 2010ab). Another study demonstrated that walking beetles were more attracted than flying beetles, to linted cotton seeds than to wheat, both in laboratory and field (Ahmad et al., 2012ab; Ahmad et al., 2013ab).

Good (1933) reported that T. castaneum can stay airborne for a maximum of only 20 seconds in the laboratory and many others subsequently accepted that these insects tend to disperse more by anthropogenic means (White, 1981; Campbell & Arbogast, 2004; Drury et al., 2009; Campbell et al., 2010a). But a field study using pheromone baited traps found that these beetles do fly and regularly fly long distances, as they were trapped regularly, and in numbers, at least 1 km from grain storage areas (Ridley et al., 2011). Ridley et al. (2011) also found that temperature plays a significant role in flight dispersal, as only extremely low numbers of T. castaneum individuals were trapped during the winter months in southern Queensland.

More recently, the field studies of Rafter et al. (2015) demonstrated that wind speed and direction are important features in the flight behaviour of T. castaneum. They found that T. castaneum generally flies downwind in the absence of any olfactory stimulus such as the aggregation pheromones of this species. They also found that the flight trajectory of T. castaneum was 1.2 m above ground and that there was negligible flight at wind speeds ≥3 m/s (Rafter et al., 2015). Various laboratory tests have been performed to link physiological factors like age, density, and phosphine resistance, to flight behaviour in T. castaneum. Strains that are resistant to phosphine have been found to be significantly less likely to fly than susceptible beetles (Malekpour et al., 2016) Studies on the effects of age, starvation, sex and rearing density on flight initiation of T. castaneum in the laboratory showed that these factors had no significant effects on flight initiation of T. castaneum (Perez-Mendoza et al., 2011a). However, a follow up study with wider age ranges of adult beetles demonstrated that flight initiation decreased with increasing age (Perez- Mendoza et al., 2011b). Likewise short periods of food deprivation led to increase in flight initiation especially in mated beetles than virgin individuals (Perez-Mendoza et al., 2011b).

Research is slowly progressing with respect to flight as a method of dispersal and migration in T. castaneum. Flight initiation is observed in these beetles even at relatively low population densities (Perez-Mendoza et al., 2011a), which leads to questions about the movement of T. castaneum beetles away from ample food resources. In other words, what triggers flight in these

4 beetles, even in the presence of favourable environmental conditions? Field studies that relate the physiological status and other intrinsic factors of these beetles to flight initiation and dispersal have not yet been conducted extensively. In my thesis therefore, I have focused on flight dispersal of T. castaneum and the physiological factors that might trigger flight. I have used conditions similar to those in the field to test the propensity for flight in these beetles in the laboratory. In particular, I quantified the relationship of various intrinsic factors to flight initiation in T. castaneum beetles, with emphasis on fat content, fecundity, lipofuscin intensity, size, weight, mating status and age.

I then used these test results to make predictions for field studies, in which beetles captured in flight were subjected to the same measurements mentioned above. These laboratory and field tests were designed to reveal the intrinsic factors that affect dispersal by flight in T. castaneum. A significant problem in studies of the flight of small insects is determining how far they fly and their directionality. Mark-recapture studies could contribute substantially in this area, but there are no really effective marking techniques that can be used for marking small insects in large numbers very quickly. I therefore investigated nanoparticles as potential external markers for mark recapture studies of flying beetles. The ultimate goal of this research is to contribute to an understanding of this species’ dispersal capability. As our understanding of the flight behaviour of this species advances, researchers will be in a better position to predict their movement and its consequences. This will also aid in the interpretation of resistance development across broad geographical regions, which will help to manage the resistance problem. Since pest dispersal also has implications for understanding the ecology of local insect populations and gene flow over a wider area so, the study of flight behaviour will also help improve integrated pest management strategies and food security.

1.3. Role of fumigants and their limitations

Fumigation is the primary method of controlling pests that damage food grains. It involves the use of chemicals that remain in a gaseous state under normal environmental conditions to kill insect pests by inducing metabolic changes (Reichmuth, 1994; Nath et al., 2011). Effective fumigation of grain requires a high level of airtightness to contain the gaseous fumigant (Winks & Ryan, 1992).

There are different types of fumigants, which can be used individually or in combination, depending on the degree and type of infestation (Kaur & Nayak, 2015). The main fumigants currently used for stored grain pests are phosphine and methyl bromide (Bullen, 2007). However, methyl bromide is now controlled by the Montreal protocol (Ozone Secretariat, 2012). High amounts of methyl bromide pose serious risks to health and the environment by contributing to ozone layer depletion

(Ozone Secretariat, 2012). Likewise, hydrogen cyanide and carbon disulphide are no longer used as fumigants, also because of their negative effects on the environment and health. Recently, sulfuryl fluoride has been used to control stored grain pests (Warrick, 2011).

Inappropriate use of fumigants has already led to insects adapting and thus becoming resistant to prescribed doses of pesticides (FAO, 1985). This challenges current pest control protocols and forces grain management authorities to increase pesticide doses until an alternative is introduced (Daglish & Collins, 1999; Nayak, 2013). This section focuses on the use of phosphine as a fumigant, with a detailed consideration of its advantages and limitations. This provides a prelude to considering that behavioural studies need to be performed extensively in the field, especially those related to the dispersal and flight behaviour of these beetles, if we are to understand the spread of phosphine resistance in this species.

1.3.1. Phosphine

Phosphine or hydrogen phosphide is a toxic and colourless gas. Phosphine is usually sold in the form of a metal phosphide (aluminium or magnesium) because the gas itself is highly unstable at room temperature. Therefore, chemicals like ammonium carbonate, urea, paraffin and ammonium carbide are added to control its release when in contact with moisture (Bond, 1989). The high toxicity of phosphine means it is an extremely effective fumigant, even at low concentrations, if used appropriately. To achieve such efficacy, pest insects need extended exposure to a normal dose if it is to kill them (>4 days), and to a prescribed concentration of phosphine at temperatures not lower than 50C (preferably 300 ppm above 250C for 7 days; or 250 ppm at 15 - 250C for 10 days) (GRDC, 2014). The gas itself leaves virtually no residues on food products, if any at all. Unlike methyl bromide it does not interact with ozone and is environmentally safe (if used in a confined space).

Phosphine is toxic to all aerobically respiring animals and is easily absorbed by porous materials. The presence of oxygen enhances the efficacy of phosphine. A complete loss of phosphine toxicity to the wheat weevil Sitophilus granarius in an environment devoid of oxygen was found in the laboratory (Bond et al., 1967). Hence, the toxicity of phosphine in aerobically respiring organisms, including T. castaneum, is linked to oxidative metabolism, and it does not affect the viability of food grains. Phosphine affects organisms through its action on particular neural, metabolic and redox pathways (Nath et al., 2011). In both vertebrates and invertebrates it is known to affect the G-

6 protein coupled acetylcholine pathways (neural), the electron transport chain (redox) and thus affects mitochondrial physiology and function (Nath et al., 2011).

Phosphine gas is regarded as a highly desirable fumigant because of its non-residual properties (Warrick, 2011). However, the development of resistance in insect pests has put the reputation of phosphine as a good fumigant under threat. The development of resistance is so profound that some populations of the pests (at least in Brazil and India) are almost 90% resistant to the prescribed doses of phosphine (Pimentel et al., 2007). For example, Opit et al. (2012) compared the LC99 values of phosphine for susceptible and resistant populations of both T. castaneum and Rhyzopertha dominica (F.) (also a stored grain pest) from Oklahoma, found that the most resistant populations of T. castaneum and R. dominica were 119-fold and 1500-fold (respectively) more resistant than susceptible populations, and that there has been a substantial increase in resistance over the past two decades (Opit et al., 2012).

1.4. Phosphine resistance – significance and management

Although insect pests can be effectively controlled by fumigants, the development of phosphine resistance is of particular significance with respect to limiting factors. Eggs and pupae are usually the hardest to kill, but with increased exposure to phosphine, mortality of eggs and pupae can be achieved (Manivannan, 2015). However, adults and larvae are easy to kill but resistant individuals may survive and pass on phosphine resistance to the next generation. Researchers have found evidence of resistance in the larval instars and adults of T. castaneum and R. dominica (Winks, 1984; Bell et al., 1977). Low levels of phosphine resistance in populations of R. dominica were initially noticed in 1971, but the real problem was the discovery of high-level resistance in 1997 (Collins et al., 2002). Two major genes rph1 and rph2 have been shown to confer different levels of phosphine resistance in T. castaneum (tc_rph1 and tc_rph2) and R. dominica (Jagadeesan et al., 2012; Schlipalius, 2002). The gene rph2/tc_rph2 is linked to a mitochondrial enzyme known as dihydrolipoamide dehydrogenase (DLD), which plays an important role in the metabolism of all eukaryotes. Similarly, a DLD linked gene (So_dld) has been found in Sitophilus oryzae, which is responsible for strong resistance against phosphine (Nguyen, et al., 2015).

Markers of the rph1 (rp6.79) and rph2 (rp5.11) resistant loci have been used to study the degree of resistance conferred by their homozygous and heterozygous states in R. dominica (Schlipalius et al., 2002; Schlipalius et al., 2008). Separately, these resistant genes produce a weak resistant phenotype (~50x by rph1 and ~12.5x by rph2), however rph1 in combination with rph2 in an individual, and

7 with both homozygous produces a phenotype far more resistant than just the sum (>250x) of individual contributions (Schlipalius et al., 2008).

The highest level of resistance in T. castaneum populations is the 100% resistance found in surveys from 36 locations in India (Rajendran, 1999). These field T. castaneum populations were resistant at an LD50 dose of 0.14 mg/L of phosphine, whilst the LD50 for a susceptible population was only 0.009 mg/L. Clearly, the fate of phosphine needs serious consideration.

The increase in frequency of beetles with resistant genes in the field is a major threat to current insect pest management strategies (Mau et al., 2012ab). Mau et al. (2012ab) provided evidence that evolution of phosphine resistance in outbreaks of R. dominica in Queensland, New South Wales and South Australia is confined to rph1 and rph2 genes, which is true for T. castaneum as well (Jagadeesan et al., 2012).

The resistance situation in these species is likely to get more serious, as indicated by recent genetic research on another grain beetle pest Cryptolestes ferrugineus. Jagadeesan et al. (2016) characterized the genes responsible for the extremely high levels of resistance observed in C. ferrugineus that have been recorded recently. They found that the two resistance genes in C. ferrugineus (Weak-R and Strong-R) were similar to those in R. dominica, that is the rph1 and rph2 genes, in terms of their degrees of dominance. However, when Weak-R and Strong-R were present together in an individual they acted synergistically to promote resistance to a very high level and this led to survival of nearly 15-30% of C. ferrugineus individuals at very high doses of phosphine exposure (0.2–3.0 mg/L). To explain such extreme levels of resistance they hypothesized that the presence of several additional dominant factors, at low frequencies, may be responsible for the very high level of resistance and that these might involve gene duplication, multiple mutations, or the presence of unique sequences, (Jagadeesan et al., 2016).

Developing an alternative for phosphine requires dedicated research but a replacement that has all the good qualities of phosphine has not yet been found. In the meantime it may be possible to adjust management strategies to reduce the selection pressures for phosphine resistance and, ideally, lead to a reduction in the frequency of resistance in the field. To this end, an appropriate understanding of the pest’s genetics, physiology, behaviour and ecology is essential.

The question is what aspects of the ecology of T. castaneum do we need to understand so as to contribute to the development of effective pest management strategies and extend the usefulness of

8 phosphine? We know little about these organisms outside of grain storages, and it is here that the insects move between storage and start new infestations. Perhaps some manage to survive the fumigation process as pupae and eggs, and disperse as adults. And this raises the general question of what drives dispersal in T. castaneum?

These insects have been found in bulk grain storage, cattle feedlots and cotton gins, sometimes at high densities. If food, mates and population density were to be considered as the causes of dispersal, it is not clear why T. castaneum adults would disperse from a readily available food source such as wheat grain in storages. Dispersal under such conditions may be attributable to intrinsic factors linked to their genotype, such as their age or physiological status, and such influences need to be identified. Although many laboratory studies have been conducted on dispersal by walking, in both the laboratory and field, few investigations focus on dispersal by flight in the field. Understanding movement patterns of these insects, including their seasonal flight patterns, average flight distance, and duration, propensity to fly, cues for flight initiation and movement to and from storage areas will help in understanding their dispersal in the field. Being able to predict movement in the field will help in monitoring storages, preventing infestations, and timing fumigations to achieve the maximum impact of phosphine. This would aid in deciding the dosage and time of fumigation.

1.5. Characterization of Tribolium castaneum migrants – a step towards understanding its spatiotemporal dynamics

Animal locomotion is defined in terms of the various methods adopted by animals to move from one place to the other. Movement in animals can be triggered by a variety of factors, for example seasonal changes, availability of resources, localisation of mates, and so on. Movement is also responsible for “population cohesion” and is a key in understanding an organism’s spatiotemporal behaviour and dynamics (Taylor & Taylor, 1983). However, studying the spatial and temporal aspects of the dynamics of an organism is not easy, but is crucial to understanding their ecology in the field. Movement is said to be the most direct, visible and measurable evidence of behaviour (Taylor & Taylor, 1983).

Movement in insects is a complex process, and has not been generalised into a general theory, mainly because of the vast diversity of intrinsic and extrinsic variables that seem to influence the behaviour of organisms and the range of ecological contexts that can be related to their movement. The cause of directionality and orientation has been subject to ongoing debate. J.W. Tutt (1902) described insect migration as ‘”spasmodic, irregular uncertain and undertaken solely on account of 9 the absolute necessities of the time”, however he could not discriminate different aspects of movement, and mainly he found no real distinction between dispersal and migration. Even the orientation of movement as a criterion for insect dispersal and migration has been disputed. For example, Williams (1930) stated that insect dispersal is highly directional. From the late 1930s to 1970s various researchers, including Hardy and Milne (1938), Freeman (1945), and Taylor (1974) found that the number and species of insects in flight was immeasurable and, hence, determining orientation is impossible.

Insect movement has been linked to a range of factors including competition and population density, reproductive cycles (the oogenesis flight syndrome), food availability, predation, temperature, humidity and other environmental variables (Rankin et al., 1986; Clobert et al., 2001). Competition and population density tend to be linked through the concept of the carrying capacity, defined by the constant K in the logistic equation. The relationships defined by this equation specify that a particular geographical area or locality can hold a certain density of population indefinitely, given constant resource availability. Individuals at higher population densities have no choice but to emigrate to survive. However, it fails to explain the dispersal that occurs even when population densities are low relative to an abundance of food resources, at least as seen in T. castaneum (Perez- Mendoza et al., 2011a). All these points indicate that flight is most likely controlled by a complex array of factors, both intrinsic and extrinsic.

Walter & Hengeveld (2014), among many others, have commented on the impracticality of the logistic equation, and its associated developments, because it treats each individual as identical and does not take into account the extensive range of species-specific adaptations that all organisms exhibit. This implies that understanding the movement of organisms and its ecological consequences requires an autecological approach in which the characterization of intrinsic factors and environmental variables is supported by quantitative and qualitative measurements of the genetics, physiology, behaviour and ecology.

Tribolium castaneum beetles, although small, are extremely robust insects. In particular, they can survive on limited food resources for almost their entire adult lifespan (Daglish, 2006). This helps explain their extraordinary success rate in infesting stored grains and building up high population densities. The insect is cosmopolitan and is known to spread by walking and flight, and is usually understood to move mainly when aided incidentally with human transport of grain (Good, 1933; Drury et al., 2009; Campbell, 2010a). Flight is a key tool for migration in many insects and T. castaneum is known to fly frequently enough and far enough, at least over distances as much as 1

10 km to result in genetic homogeneity over extensive areas (Ridley et al., 2011). Flight in T. castaneum has been mostly linked to mating or food resources, and not much is known about the factors that drive their movement.

1.6. Thesis structure and approaches to characterize flight in Tribolium castaneum

The central objective of the research reported in this thesis is to use an autecological approach to understand the intrinsic and extrinsic influences on the flight behaviour of T. castaneum. The following sub-sections list the ways in which I have undertaken to answer the key question of what leads T. castaneum adults to fly. These are followed by a brief description of the thesis structure.

1.6.1. Genetic studies

Since T. castaneum is known to be polyandrous (Pai & Yan, 2002), determining the level of multiple mating is crucial to understanding the gene flow and genetic variation across field populations (Bretman & Tregenza, 2005). Adult female T. castaneum is known to mate with multiple males and store the sperm in her spermatheca. Molecular techniques involving the use of microsatellite markers allow the level of polyandry to be estimated, and understanding parentage is helpful in interpreting gene flow. Microsatellites are short di- to tetra-nucleotides (2–4 base pairs) tandem repeats in the genome. The length of these repeats is variable among different regions of the DNA. Using microsatellites as genetic markers is similar to DNA fingerprinting, where the progeny is related to the parents by comparing specific genomic regions. Microsatellites are highly repeatable, co-dominant and polymorphic so provide good genetic markers for parentage analyses. Once a particular set of microsatellite markers has been identified it is easy to compare these regions with those of the progeny to decide the level of polyandry or multiple mating in a female. I have therefore used microsatellite markers identified by Demuth et al. (2007) to test the degree of polyandry among different test populations of T. castaneum.

1.6.2. Fecundity, morphometrics and fat content

Measuring fecundity of female beetles helps in quantifying the egg load that the female carries during flight and, incidentally, confirms whether a female has mated because virgins deposit no eggs (Sokoloff, 1972). Besides fecundity, the size, weight and fat content could be related to the propensity to fly (and, perhaps, to other features of flight).

1.6.3. Markers and why to use them

In biology, markers are objects, substances or molecules that are used to indicate the position, place, route, physiological state, age and so on, of an organism. Ideally a marker should be stable, biocompatible, non-invasive, specific, easily detectable and highly reproducible (Bigbee & Herberman, 2003). In entomology, markers are used both externally and internally to locate and study various insect species. For example, several inks and dyes have been used to mark insects to study their flight patterns (Walker & Wineriter, 1981).

Markers aid in the focus on a set of individuals or on an area to understand different particular problems related to the movement of organisms. One of my aims is to study T. castaneum beetles in the field, but a suitable external marker is needed, one that will allow numerous individuals to be easily and quickly marked and one that will persist on the organisms as they move around. I therefore tried marking beetles externally with fluorescent nanoparticles (quantum dots) and then utilizing this to study their flight patterns. Studying flight patterns will help determine their dispersal capacities and will provide insights and characteristics as to how far they fly and how they might distribute alleles for resistance.

The ability to estimate the age of beetles captured in the field would help in interpreting the behavioural patterns of adults of different ages. The question here is whether beetles of a particular age are more likely to initiate flight. An age estimate of beetles caught in the field would be another important aspect in understanding their flight patterns. This will play a major role in estimating the likelihood of grain infestation. Developing a quick test of age will help in correlating the degree of risk for cross-infestation with the age of beetles. Being able to determine the age would also help in correlating the physiological status of beetles with their behaviour.

1.6.4. Markers used in this study

(a) Tellurium coated cadmium nanoparticles: quantum dots

To mark beetles externally I used quantum dots (QDs), which are fluorescent nanoparticles with a diameter of ~2-10 nm. These nanometre scale particles are usually composed of repetitive units of one or more atoms. Their design affects their properties and mode of action. One of their unique properties is that they can absorb and emit light efficiently without getting photo-bleached (Zhu et al., 2012). In this way they remain stable and robust in comparison with organic dyes. They

12 have been successfully employed in vitro as well as in vivo as diagnostics for various diseases, like cancer (Gao & Dave, 2007).

(b) Lipofuscin as a possible marker for age or metabolic status

Lipofuscin is a non-degradable intra-lysosomal polymer like substance. Lipofuscin levels have been shown to increase with progressing age in many insects (Ettershank, 1983; Robson & Crozier, 2009; Terman et al., 2010). However, lipofuscin accumulation may also differ with different activities of an insect for example, flying houseflies had higher lipofuscin content as compared to low activity houseflies (Sohal & Donato, 1979). Lipofuscin intensity might change with physiological state, but physiological state of the insect may also provide clues to relate stress with their behaviour. For example, studying lipofuscin intensity of reproducing females may reveal the differences in their physiological state from that of unmated females. Therefore, I have studied and compared lipofuscin intensities of T. castaneum with respect to old and young age beetles in Chapter 2 and flying and resident beetles in Chapter 4.

1.6.5. Thesis structure

In Chapter 2, I tested beetles of known age in a laboratory flight experiment, to relate aspects of the insects’ age and their flight behaviour. In these tests a comparison was carried out between individuals caught flying and those collected from the food resource (which were considered to be resident beetles) to understand the potential fecundity, morphometrics, lipofuscin and fat content of individuals leaving the food resource. A comparison was made with resident beetles to see if these intrinsic factors are exclusive to flying individuals.

With respect to emigrating females, especially those flying long distances, little is known of their mating status and their level of polyandry. The more that females mate before emigrating, the greater the genetic diversity they carry in their spermathecae. This has been addressed in Chapter 3 where I contributed to characterizing the mating status and degree of polyandry in laboratory reared field populations of T. castaneum and R. dominica that had been in culture only for one generation.

Predictions generated from the experiments described in Chapters 2 and 3 are further tested by investigating the characteristics of field-captured T. castaneum for comparison with the characterisations performed in the laboratory these are described in Chapter 4. Along with

13 fecundity, size, lipofuscin and fat content, the mating status of the field caught females was also determined to estimate the degree of polyandry.

In Chapter 5, a novel marker, which uses the fluorescent ability of nanoparticles to mark the beetles, was investigated for its potential utility in mark-recapture studies in the field. Its ability as an external marker for mark-recapture studies of T. castaneum in the field was tested and its ability as a marker on various other small insects was also tested.

The final chapter, Chapter 6, is a general discussion. It lists and compares the outcomes of this research with previous studies and gives suggestions for future research.

CHAPTER 2

Characterizing Tribolium castaneum adults that initiate flight in laboratory tests – generating predictions for the field.

2.1. Introduction

Local movement and large-scale migration in organisms is key to their spatiotemporal dynamics (Walter & Hengeveld, 2014). Active movement in Tribolium castaneum is by walking or flight, but passive movement also occurs, and is mainly anthropogenic (Good, 1933; Campbell et al., 2010a; Ridley et al., 2011). Tribolium castaneum infests grain and processed food products globally and its feeding results in massive losses if unchecked (Good, 1936; Hagstrum & Gilbert, 1976; Boon & Ho, 1988; Campbell & Runnion, 2003). Phosphine fumigation is the treatment of choice, but these insects carry resistance genes against phosphine (Schlipalius et al., 2008; Jagadeesan et al., 2012). Thus, T. castaneum dispersal has implications for pest management.

Historically, authors have based their movement studies of T. castaneum mainly on their walking behaviour (Surtees, 1963; Hagstrum, 1973; Campbell et al., 2002, 2004). In this respect, studies have been conducted on the influence of age, sex, population density, quality and quantity of food resources, reproductive fitness and its genetic basis (Ziegler, 1976, 1977, 1978; Ritte & Lavie, 1977; Lavie & Ritte, 1978, 1980). That T. castaneum can, indeed, fly long distances, at least 1 km, has been demonstrated with pheromone traps (Ridley et al., 2011). Studies have been performed in the laboratory and field to determine flight distance, what influences flight direction, and the effect of various environmental factors on the flight of T. castaneum. Experiments investigating flight directionality discovered that T. castaneum do not initiate flight readily at wind speeds greater than 3 m/s and that the flight is directed downwind in the field (Rafter et al., 2015). Flight initiation was negligible in complete darkness, flight mostly occurred between 30-350C and there was no effect of relative humidity and light intensity (above a certain minimum) on flight initiation of T. castaneum in the laboratory (Perez-Mendoza et al., 2014). However, a subsequent study showed positive correlation between light intensity and flight initiation in T. castaneum (Drury et al., 2016). Food quality and temperature also significantly affected flight propensity in these latter tests, with flight initiation increasing at temperatures higher than 300C along with poor quality of food. However, flight did not occur at 400C and the beetles were observed to burrow in the culture medium when it got that hot, irrespective of the quality of food (Drury et al., 2016).

Few studies have investigated the physiological factors that influence flight in T. castaneum. Recently studies on the effects of phosphine resistance on movement have shown that resistant beetles did not initiate flight as readily as did beetles from susceptible laboratory strains and that resource location by walking was disoriented in resistant individuals (Malekpour et al., 2016). Rearing density, age, sex and short periods of starvation were found to have no significant effect on flight initiation in T. castaneum, but the presence of beetles of the opposite sex in the flight chamber 16 reduced their readiness to take flight (Perez-Mendoza et al., 2011a). These studies were performed within small flight chambers, and the division of beetles into age groups was limited. With respect to the long adult life of these insects (up to at least 12 months) (Sokoloff, 1974), treating 7-20 day old adults, as “old beetles” may not really represent what truly old beetles actually do. For this reason, a follow-up study was reported (Perez-Mendoza et al., 2011b) in which the effects of long- term food deprivation, age (up to 30 weeks), mating status, quality and availability of food on flight initiation of T. castaneum was tested. Older adults (26-30 weeks) did not initiate flight at all and flight initiation rates decreased with increasing age. Fecundity was also found to be related to age. Progeny production peaked in mature females (~8-17 weeks old) and then declined as the females aged. Food deprivation increased flight initiation and decreased oviposition while mating status was found to have no significant effect on flight initiation (Perez-Mendoza et al., 2011b). Again these studies were limited to small flight chambers (which may limit the natural behaviour of beetles) and with the tests performed on individual beetles except for those requiring the presence of other beetles.

A laboratory test of flight behaviour of T. castaneum, in a much larger arena, was therefore designed to provide further insight into the migratory behaviour of T. castaneum in relation to age, sex, fecundity, morphometrics, body weight, fat content and lipofuscin levels (to estimate age). The possibility of using lipofuscin as a marker of age was included because some studies have shown lipofuscin levels to be directly correlated with chronological age in ants, flesh flies and other organisms (Ettershank et al., 1983; Robson & Crozier, 2009).

Beetles tested were first generation insects bred from adults collected in the field, so they had not been exposed to unintentional selection imposed by prolonged laboratory culture, which reduces genetic variability and affects their physiology and behaviour (Hagstrum & Subramanyam, 2006; Ahmad et al., 2012b; Malekpour et al., 2016). Rearing was conducted in a structured way to ensure that beetles of different ages were available for flight tests at the same time. The individuals of different age that took flight were tested for their physiological status and were compared to beetles that were collected from the culture (and which were deemed to be non-flying or resident). The effects of age, sex, size, fecundity, weight, fat content and lipofuscin levels on the propensity of flight in T. castaneum were quantified. These tests were designed to answer the general question of what leads to flight in T. castaneum. The data derived from these tests thus allowed a set of predictions to be made as to which beetles in a population are more likely to initiate migratory flight. These predictions were used as reference for field tests reported in Chapter 4.

2.2. Materials and Methods

2.2.1. Culturing field collected Tribolium castaneum

Flight experiments were performed in spring in October 2015 and September 2016. Adult T. castaneum were collected from an on-farm grain storage facility near Dalby (27°11′0″S and 151°16′0″E) in 2015 and cotton seed shed near Moura (24° 35' 35.88" S and 149° 59' 58.07" E) in 2016, in Queensland, Australia. Field collected beetles were mass cultured by seeding 200 adults in a 5 L plastic bucket with 2/3rd wheat and 1/3rd organic wholemeal wheat flour (Kialla Pure Foods, Greenmount, Queensland, Australia) mixed together with 5% Torula yeast (Lotus, Carrum Downs, Victoria, Australia) in a controlled temperature room (250C; 52-60% RH). Mass culturing continued in this way to raise three treatments (n = 9 buckets, i.e., n = 3 replicates for each treatment) with F1 adults either old (~12 weeks post-emergence), young (~2 weeks post-emergence) or mixed-age beetles (approximately equal proportions of old and young), with all available for experiments simultaneously. Six buckets (5 L) were set up 5 months prior to the experiment to prepare three old- age F1 adult replicates and three mixed-age replicates. The old-age treatment was synchronized to contain only first-generation T. castaneum adults by transferring the adults regularly to fresh medium. The mixed-age treatment was not synchronized for age in this way. In summer 2016, one mixed-age culture bucket had high mortality of beetles, so it was not included in the experiment. Culturing of the young beetle treatment (three buckets, as above) was started 3 months after the old- age treatment. Four hundred of the field-collected T. castaneum adults (which were still alive at this point) were used in young replicates to start the culture, to maximize the egg production. The parental beetles were kept in the culture medium for 1 week only, to reduce the variability in the age window of the offspring.

Each replicate was cultured in a bucket and in each bucket there were ~1600-2000 beetles. This number was estimated by first weighing three batches of 100 adult beetles to get an average estimate of weight. All beetles in each bucket were then weighed together and their weight estimated by extrapolating from the average derived from the 100 beetles.

2.2.2. Flight apparatus

All three treatments (old, young and mixed-age) were set up inside a room subject to ambient temperature fluctuations (26 ± 30C) and each bucket was modified into a flight apparatus that served as a means for capturing beetles in flight. The flight apparatus included the culture buckets

18 as a base and mesh cages with cubic frames (BugDorm®, Australian Entomological Supplies, Sydney, Australia). A mesh cage was placed above each bucket, as a trap (Figure 2.1). These cages were modified to fit on the buckets by cutting a 15 cm circle from the bottom of each cage. The mesh around the hole was glued to a cardboard sheet with a 15 cm hole for support and this was placed on top of the bucket. Two small cones made of filter paper were placed on top of the flour- based medium in each culture to provide a stage from which the beetles could fly (Figure 2.1). Each cage had a pre-built opening extending into a mesh sleeve to provide access inside the cage for collection of beetles as they landed inside the cage.

Collection sleeve

Mesh cage trap

Circular hole leading into the cage for capture of beetles in flight Cardboard base to support the cage trap

Paper cones as raised platforms to aid flight initiation

Culture medium with Tribolium castaneum adults

Figure 2.1 Schematic diagram of the flight apparatus used in laboratory flight tests.

2.2.3. Flight assay

All nine buckets were observed for flight from 1500 h until dusk (~1817 h to 1822 h) for 6 days, during which temperature was monitored using a digital thermometer. Each beetle that flew from the culture bucket was collected and stored in a separate container. Equal numbers were subsampled randomly, after the flight observations had ceased, from those beetles that did not fly on each day of the experiment. This was achieved by sampling beetles manually, with a spoon, from three strata of the culture. The three strata comprised, roughly, the top, middle and bottom of the contents of the bucket, and beetles were collected from the central core of the bucket’s contents. The number of beetles to be collected from each stratum was determined by dividing the total

19 number of beetles that initiated flight from each particular bucket by three. Three people were involved in the observations that monitored flight activity across the replicates. All replicates were distributed so that one person observed three replicates from different treatments. These replicates were rotated each day to minimize against position effects.

2.2.4. Beetle characteristics

After obtaining a population of flying and non-flying beetles from each replicate, each beetle was weighed on the same day of collection and was kept individually on 20 g flour (in a 30 mL plastic cup) in a controlled temperature room (25±20C; 55-60% RH). Each beetle was held at 00C for 5 min to reduce its activity, then sexed and measured for body size (length and width of the thorax, head width and length of elytron). Its ventral side was inspected for the “sub-basal setiferous sex patch on the male prothoracic femora” (Halstead, 1963) under a compound microscope using NIS elements imaging software. Body measurements were performed on the captured image at 2.5X magnification by using the digital length measurement function in the NIS elements imaging software.

Beetles caught flying and those subsampled from the buckets (non-flyers) were assigned randomly to one of three groups and the following characterizations performed: fat content, fecundity of females, and measurement of lipofuscin concentration to estimate age. a) Fat content analysis

Fresh weight of the beetles was obtained on the each day of the collection after which the beetles were kept at -200C until analysis. Each was thawed and cleaned with a thin paintbrush to remove particles of flour and weighed. Their weight was measured on a Sartorius® analytical balance with a precision of 10-5 g. The beetles were dried at 60±50C for 48 h and weighed to record the dry weight after all the water had been lost to evaporation. For fat extraction, each specimen was placed individually in a 1.5 mL Eppendorf® tube (Eppendorf, Hamburg, Germany) and submerged in a 1 mL (1:1) mixture of diethyl ether and chloroform for 72 h to dissolve the fat (Nedved & Windsor, 1994; Östman, 2005; Knapp & Knappova, 2013). The tubes were sealed with Parafilm®M (VWR®, Radnor, Pennsylvania, USA) to prevent evaporation of the organic solvents. The specimens were then dried again for 48 h at 60±50C and weighed to record the lean dry weight. Fat content was calculated as the difference between the dry weight (weight measured after water loss) and lean dry weight (weight measured after fat loss by dissolution in organic solvents).

20 b) Fecundity analysis

Each female was placed, individually, in a 30 mL plastic cup on 20 g flour and yeast (5% by weight) at 25±20C and 55-60% RH. Parental females were transferred every 15 days onto fresh medium, for 90 days, after which they were stored in ethanol at -200C. The cups from each 15 day interval were examined visually without disturbing the medium for different stages of offspring at the top of the cup. Adult offspring started emerging ~6 weeks post seeding of parental females and the final counting of adult offspring was conducted 12-15 days post emergence of the first adults (emergence time for individual beetles was not recorded). Usually no pupae were observed beyond this time and, if present, were nearing emergence so were included in the progeny count. c) Lipofuscin intensity

Lipofuscin levels in individual beetle were determined using methods adapted from Ettershank et al. (1983). Lipofuscins were first extracted by grinding each beetle separately in liquid nitrogen for 60 s, then the crushed tissue was combined with spectroscopic grade chloroform: methanol (2:1; 1 mL) in a 1.5 mL safe-lock Eppendorf® tube. This was followed by sonication at 160 W for 3 min after which the tubes were centrifuged at 4000 g at 40C for 20 min. The supernatant was transferred to a fresh Eppendorf® tube and distilled de-ionized (0.5 mL) water was added followed by centrifugation for 20 min at 4000 g at 40C to remove any remaining particulate material and contaminating pteridines (Sheehy & Roberts 1991). Lipofuscin concentrations were then measured by taking a 0.5 mL sample of the subnatant, and analyzing it under a Tecan Infinite® 200 Pro fluorescence spectrophotometer (Tecan, Männedorf, Switzerland). The control for this measurement was a mixture of 2:1 of chloroform and methanol. Samples were excited with a wavelength of 340 nm, and the emission spectra recorded between 300 and 600 nm. The greatest intensity of the spectra (which corresponds to a λ emission = 410 nm) was recorded in arbitrary units. Spectral intensity at this wavelength is proportional to the lipofuscin concentration (Ettershank et al., 1983; Robson & Crozier, 2009).

2.2.5. Statistics

All data was analyzed using the R 2.15.1 statistical software package (R Development Core Team, 2013) and GraphPad Prism 7.0 (GraphPad software inclusive, 2016). Fisher’s chi square test was performed to test differences across all treatments with respect to age and flight. Log-rank and Gehan-Breslow-Wilcoxon tests were used to compare survival in different treatments and between

21 flyers and resident beetles. Two-way analysis of variance (ANOVA) was performed on weight, fecundity, fat content and lipofuscin of beetles in flight for comparison with non-flying or resident beetles with post hoc Tukey HSD to test differences between age groups. Morphometric data were tested by Mann Whitney unpaired non-parametric t-tests to indicate differences in head width, elytron length, thorax length and thorax width of flying females with respect to flying males and resident females across each age group. The Mann Whitney test was used because some of the data were not normally distributed.

2.3. Results

2.3.1. Relation between age, sex and flight

The old-age treatment yielded only 37 flying beetles, whereas 69 and 123 were retrieved respectively from the young and mixed-age treatments in October 2015 (Figure 2.2a). More females than males were captured flying in each treatment, (72% mixed-age, 56% old, 60% young) (Figure 2.2a), but this was not statistically significant (흌2 = 3.90, d.f. = 2, P = 0.14). There were no statistical differences in the numbers of males and females flying across the different age groups (old vs young: 흌2 = 0.17, d.f. = 1, P = 0.68; old vs mixed-age: 흌2 = 2.86, d.f. = 1, P = 0.09; young vs mixed-age: 흌2 = 2.30, d.f. = 1, P = 0.13).

In September 2016, young flying adults outnumbered both the old and mixed-age treatments (Figure 2.2b). The total number of young beetles flying was significantly higher than old beetles in both 2015 and 2016 (흌2 = 35.65, d.f. = 1, P < 0.0001). A total of 131 females and 74 males flew in the young age treatment, whereas the old and mixed-age treatments yielded only 19 beetles altogether (old ♂♂ = 4, old ♀♀ = 2; mixed-age ♂♂ = 5, mixed-age ♀♀ = 8). The ratio of young females to young males that flew during October 2015 did not differ significantly from the ratio captured flying during September 2016 (흌2 = 0.20, d.f. = 1, P = 0.65).

a) OCTOBER 2015 Males

Females

b) SEPTEMBER 2016

Total number of Total number beetles

Old Young Mixed-age

Treatment

2.3.2. Time of flight initiation

The initial observations on flight initiation in T. castaneum showed that the beetles flew sometime after 1500 h until dusk (~1822 h). Figure 2.3 depicts the total number of beetles flying on day 1 of the experiment (October 2015). This observation was made to make a decision on the timing to monitor the flight for the following days of the experiment.

10 Old

Young 8 Mixed-age Dusk

4 Total Total beetles in flight dayon 1 2

0 1500 1530 1600 1630 1700 1730 1800 1830 Time of capture (half-hour intervals in minutes)

Figure 2.3 Pattern of flight of Tribolium castaneum across old, young and mixed-age treatments for day 1 of observation, in October 2015.

Figure 2.4 depicts the time intervals during which most of the beetle flight occurred, totalled across all days (October 2015). The flight pattern was mostly crepuscular for most flight activity occurred prior to dusk (1822 h). Flight in all treatments declined at ~1800 h. Similar patterns of flight were observed in the flight test conducted in September 2016.

8 a) OLD Males Females 6

0 16 b) YOUNG

4 Total number of Total number beetles

0 25 c) MIXED-AGE

0 1500 1530 1600 1630 1700 1730 1800

Time of capture (half hour intervals)

2.3.3. Relationship between age, weight and sex of flying and resident beetles

There were no significant differences in the weight across flying and resident beetles within each sex across all treatments (Figure 2.5). A two way ANOVA indicates no significant difference between the flight, weight and sex of beetles (F = 0.15, d.f. = 5, P = 0.98).

ass (mg)

Beetle m Beetle

Sex

2.3.4. Survival rates and mean 90 day fecundity of flying and resident beetles

Young females had higher survival rate, mixed-age females between those of old and young, and old females had the lowest survival rate (Log-rank: 흌2 = 15.57, d.f. = 5, P = 0.008). Survival of old females was significantly lower than the young beetles both resident (Log-rank: 흌2 = 4.47, d.f. = 1, P = 0.034) and flying (Log-rank: 흌2 = 8.82, d.f. = 1, P = 0.003) (Figure 2.6). There were no significant differences in survival rates of young flying vs. resident (흌2 = 0.155, d.f. = 1, P = 0.693) and old flying vs. resident beetles (흌2 = 0.21, d.f. = 1, P = 0.649). The survival of mixed-age flying beetles was significantly higher than old flying beetles (흌2 = 5.89, d.f. = 1, P = 0.015). However, the survival rate of mixed-age resident beetles was not significantly different from either old (흌2 = 0.58, d.f. = 1, P = 0.445) or young (흌2 = 2.39, d.f. = 1, P = 0.122) resident beetles.

Percent survivalPercent

Days

Mean fecundity of both flying and resident female beetles from all age treatments was calculated over a 90 day period (Figure 2.7). There were no significant differences in the total fecundity of flying and resident females across all age groups (two-way ANOVA: F = 0.28, d.f. = 2, P = 0.760). Fecundity of the females at the 2nd oviposition interval (16-30 days after capture) seemed to be higher than the fecundity at the 1st interval (0-15 days after capture). But the interaction across treatments was not statistically significant (two-way ANOVA: F = 1.154, d.f. = 6, P = 0.331). However, multiple comparisons indicated that the mean fecundity of young flying females was significantly higher at the 16-30 day period than the 0-15 days period (M.D. = 44.33, P < 0.0001). This was not evident in flying females of any other age groups. No such trends were observed in resident females of all ages.

100 a) FLYING Old Young

75 Mixed-age

0 100 b) RESIDENT

75 Mean 90 90 Mean day fecundity

0 0-15 16-30 31-45 46-60 61-75 76-90 Time interval (15 days) Figure 2.7 Mean fecundity of females (±2SE) of Tribolium castaneum females: a) flying (old: grey, n = 21; young: white, n = 30; mixed-age: cross pattern, n = 30), and b) resident (old: grey, n = 17; young: white, n = 30; mixed-age: cross pattern, n = 29) at 15 day time intervals over a 90 day period (P = 0.76 for total fecundity across all treatments). 28

2.3.5. Mean fat content

Mean fat content of young and old age beetles collected in September 2016 (Figure 2.8) was calculated, because not enough beetles flew in 2015. Only resident beetles in the old age treatment could be tested because very low numbers of old beetles flew in 2016. Both resident and flying young age beetles were tested for fat content. Mixed-age beetles were not tested because they simply acted as a control treatment in the flight test.

Mean fat content of young beetles was higher than that of the old beetles. A two-way ANOVA with multiple comparisons indicated significant differences across the young resident beetles and the old resident beetles with the former having a higher fat content than resident old beetles (F = 38.1, d.f. = 1, P < 0.0003) within each sex. There were no significant differences in mean fat content across the young resident and young flying beetles within both the sexes (F = 0.027, d.f. = 1, P = 0.87). However, resident young males had significantly more fat than young flying females (M.D. = 0.768, P = 0.002).

0.4 Males Females

b bc 0.3 bc

c a 0.2 a

Mean fatMean content (mg) 0.1

0 Young (Flying) Young (Resident) Old (Resident)

Treatment

Figure 2.8 Mean fat content (±2SE) of Tribolum castaneum captured flying (flyers) and those sampled from the culture medium (residents), collected in September 2016. Treatments tested were old resident beetles (n = 28♀♀, 26♂♂), and young flying (n = 26♀♀, 28♂♂) & resident beetles (n = 28♀♀, 20♂♂). In each treatment the white and the grey bars represent females and males, respectively. Different letters above the relevant bars represent significant relationships.

2.3.6. Mophometrics

The head width, thorax length, thorax width and elytron length measuremnets of all age groups from both flying and resident beetles were analyzed by Mann Whitney U test because some of the data were not normally distributed and the sample sizes were variable. Table 2.1 respresents the mean of measurements (in mm) across all treatments in both flying and resident male and female T. castaneum. Overall there were no consistent statistically significant differences (except for males) observed across all age groups for all length measurements within sexes for flying and resident beetles (sexes and age groups were not compared). The elytron length of flying males was significantly longer than the resident males from all age groups (Table 2.1). The females showed no regular pattern of differences in any measurements. Where differences were observed it was the female flyers which had a longer elytron than the resident females.

Table 2.1 Morphometrics (in mm) of Tribolium castaneum from flying and resident males and females of old, young and mixed-age treatments.

Treatment Male Mann-Whitney test (Pexact) Flying Resident Old Mean SEM Mean SEM Head width 0.38 0.0048 0.38 0.0039 0.667 Thorax length 0.44 0.0082 0.43 0.0055 0.130 Thorax width 0.62 0.0071 0.61 0.0071 0.150 Elytron length 1.36 0.0166 1.31 0.0134 0.027** Young Head width 0.39 0.0061 0.39 0.0039 0.155 Thorax length 0.47 0.0134 0.44 0.0059 0.005** Thorax width 0.63 0.0068 0.62 0.0074 0.145 Elytron length 1.38 0.0172 1.33 0.0109 0.014** Mixed-age Head width 0.39 0.0039 0.38 0.0037 0.012** Thorax length 0.44 0.0069 0.43 0.0045 0.252 Thorax width 0.62 0.0050 0.58 0.0063 0.012** Elytron length 1.34 0.0119 1.30 0.0043 0.021** ** Represent significance, where P < 0.05; all measurements are in mm.

Table (continued on next page)

Treatment Female Mann-Whitney test (Pexact) Flying Resident Old Mean SEM Mean SEM Head width 0.40 0.0038 0.40 0.0052 0.410 Thorax length 0.46 0.0041 0.45 0.0064 0.130 Thorax width 0.64 0.0047 0.63 0.0075 0.052 Elytron length 1.42 0.0152 1.36 0.0182 0.030** Young Head width 0.40 0.0036 0.39 0.0045 0.312 Thorax length 0.46 0.0053 0.45 0.0064 0.146 Thorax width 0.64 0.0056 0.63 0.0058 0.264 Elytron length 1.41 0.0124 1.39 0.0108 0.129 Mixed-age Head width 0.39 0.0046 0.39 0.0039 0.801 Thorax length 0.45 0.0038 0.45 0.0060 0.734 Thorax width 0.60 0.0040 0.59 0.00733 0.422 Elytron length 1.40 0.0080 1.34 0.0158 0.0006** ** Represent significance, where P < 0.05; all measurements are in mm.

2.3.7. Lipofuscin

Young female flyers had the highest mean lipofuscin intensity of all treatments (two way ANOVA: F = 6.71, d.f. = 2, P = 0.0018). The interaction between lipofuscin intensity, age and sex was statistically significant (F1,73 = 11.05, P = 0.0014). Post hoc Tukey HSD tests of the mean lipofuscin intensities indicate that young flying females had a significantly higher mean lipofuscin intensity than those of young resident females (M.D.= 2538, P < 0.0001) as well as old ones (M.D. = 2225, P < 0.0001). Young flying females also had significantly higher lipofuscin intensity than young flying males (M.D. = 1343, P = 0.011). No significant differences were found across males in any of the treatments. Most of the resident beetles had negligible or undetectable levels of lipofuscin, so those individuals have been represented as a number above each bar in the Figure 2.9.

4000

B Males

Females 3000 q 1

2000 10 p A A A 1000 r A Mean lipofuscinMean intensity 8 s s 17 r 19 18 0 Young (Flying) Young (Resident) Old (Resident)

Treatment

2.4. Discussion

The main findings from the laboratory flight test are as follows: 1) flight in T. castaneum is mostly crepuscular (highly active between late afternoon and dusk), 2) flight propensity of T. castaneum is not affected by sex, body weight or size, but is significantly affected by age, 3) the total fecundity of females over the 90 day period was similar in all age groups, 4) young beetles have more fat than old beetles, and 5) lipofuscin intensities show no significant relation to age. This is the first study to investigate the relationship of physiological variables such as fat content, body weight and lipofuscin with respect to flight dispersal in T. castaneum.

Overall, the differences observed in flying and resident beetles were subtle and some observations are in accordance with previous findings on aspects of the flight behaviour of these beetles. The flight pattern in T. castaneum that was observed to be at its peak during late afternoon in this study (Figures 2.3 & 2.4) has been reported previously in the literature and is often referred to as crepuscular in the literature (Howe, 1952; Jones, 1967; Rafter et al., 2015; Malekpour et al., 2016). 32

No significant relationship was detected between flight, age and sex (Figures 2.2 & 2.4), significantly similar proportions of males and females flew in all age groups (2015 & 2016). The numbers of young beetles caught in flight in this study were higher than the number of old beetles flying, in both periods of observations (2015 and 2016). Previous findings demonstrate that flight initiation declines steadily with increasing age (age range 8-165 days) (Perez-Mendoza et al., 2011b). The general conclusion about age drawn from this study is that flight propensity in T. castaneum could be dependent on more than one variable at any given time that dispersal takes place.

Survival curves showed significant differences between the old and the young beetles, but there were no differences between flight take off and survival because resident beetles had similar survival rates to flying beetles of the same age (Figure 2.6). Fecundity of individual T. castaneum was not related to age or flight initiation. However, fecundity was significantly higher within young flying females during the16-30 day period post emigration, compared to the same females during the 0-15 days post emigration period, but this pattern was not significant in young resident females (Figure 2.7). Previous studies have demonstrated that young females (~14 days old) produce fewer progeny compared to relatively older females (~57-120 days old) (Perez-Mendoza et al., 2011b). This pattern was not observed in this study, as the overall fecundity of females of all ages was similar (Figure 2.7).

Fat content was higher in young beetles but this was not significantly related to flight initiation (Figure 2.8). The decrease in fat content with age can be linked to the decline in energy reserves, degeneration of cell organelles, or changes in respiratory enzymes that occur with increasing age (Sohal, 1985). The higher fat content in resident males as compared to flying females could be due to chance, because the resident beetles are likely to have higher variation than the beetles collected in flight and because the latter are effectively self selected. Lipofuscin intensities were generally higher in young flying beetles but were only significantly higher in young females that initiated flight (Figure 2.9). Most of the young and old resident samples did not have detectable levels of lipofuscin and those that were detectable were very low (Figure 2.9). This suggests that accumulation of lipofuscin is dependent on several physiological factors, for example, increase in free radicals, degeneration of organelles and lipid peroxidation (Benzi, 1990; Terman & Brunk, 1998). It is possible that flying females have a higher energetic cost, which is enhanced by their reproductive status, since T. castaneum females are known to be polyandrous (Pai & Bernasconi, 2008), and fecund when they initiate flight (as demonstrated in this test). So, flying beetles as compared their resident counterparts may have a higher respiratory and metabolic rate, and hence

33 the observed difference in lipofuscin. Studies on houseflies showed that high activity in these insects was associated with higher lipofuscin indices than in ones that had remained relatively inactive (Sohal & Donato, 1979). Thus, lipofuscin intensity can be used to infer cellular aging rather than chronological aging as the metabolic status of an individual changes with different activities (Benzi, 1990).

Flight propensity in T. castaneum is, therefore, not directly related to most of the physiological factors investigated in this study. However, young beetles do tend to fly more than old ones. There were no significant measures that define the physiological status of a flying individual of a certain age (at least until ~2-3 months). The dispersive nature, high fecundity, and long adult life of these insects have consequences for the management of T. castaneum (Good, 1936; Campbell & Runnion, 2003; Perez-Mendoza et al., 2011b; Jagadeesan et al., 2012; Ridley et al., 2011). Frequent “patch exploitation” of food grain (Campbell & Runnion, 2003) by active and passive dispersal in T. castaneum increases the vulnerability of current grain reserves even after treatment with phosphine and consequently leads to significant food loss. This in turn deteriorates the already existing problem of food security. The ability of these beetles to reproduce and survive in variable conditions would have been integral to the persistence of T. castaneum before the advent of anthropogenic grain reserves.

The laboratory assays presented here enable us to predict the following about T. castaneum populations in the field: 1) old beetles fly less than young ones, 2) flight propensity is not dependent on sex, body weight, body size, fat content, or lipofuscin levels, 3) young beetles have higher fat content than older beetles (>2 months), 4) the fecundity of flying and resident beetles is similar statistically and not affected by age, 4) lipofuscin levels appear to be related more to the metabolic status of an individual. These predictions are tested in Chapters 3 and 4, using methods developed here, on beetles sampled from the field.

CHAPTER 3

Polyandry, genetic diversity and fecundity of emigrating beetles - understanding new foci of infestation and selection

The citation of the paper is as follows:

Rafter, MA, McCulloch, GA, Daglish, GJ, Gurdasani, K, and Walter, GH (2017). Polyandry, genetic diversity and fecundity of emigrating beetles: understanding new foci of infestation and selection. Journal of Pest Science. doi: 10.1007/s10340-017-0902-8

Declaration

This chapter was developed in collaboration with my colleagues, which led to production of a peer- reviewed paper with one of my supervisors (Dr. Michelle Rafter) as the first author. This chapter is now published in the Journal of Pest Science. My contribution was the characterization of polyandry in reference laboratory strains and newly established laboratory cultures of T. castaneum and Rhyzopertha dominica from field-collected insects. I analysed the number of male contributors using Geneious 7.2.1 microsatellite plugin and ran tests on GERUD 2.0 to generate likelihood of male contributors. The molecular skills acquired were then used to infer polyandry in field populations of T. castaneum as described in Chapter 4. For this reason and because the chapter provides a major connecting link between my studies of T. castaneum in the laboratory and field, it has been included in the main body of this thesis. Formatting has been changed in accordance with the thesis requirements of The University of Queensland.

Author contributions

Michelle Rafter (M.R.), Graham McCulloch (G.M.), Gregory Daglish (G.D.) and Gimme Walter (G.W.) conceived and designed the field sampling work; M.R., G.M., G.D. and G.W. were involved in field sampling. M.R. conducted survival and offspring assessments in the laboratory of field collected material and Komal Gurdasani (K.G.) initiated and monitored the laboratory cultures. G.M. designed the molecular work, G.M. and K.G. performed the molecular work and analysed the data with input and further GLM analysis conducted by M.R. The first version of the manuscript was written by M.R. with input from G.M., G.D. and G.W. All co-authors have been involved in further development and finalisation of the manuscript.

3.1. Introduction

Polyandry attracts a lot of research interest, mainly because it is seen as a paradox; females should derive no benefit from mating with multiple males because mating is expected to be costly (Kokko & Mappes, 2012; Demont et al., 2014). Research has therefore focused on determining how polyandry impacts female fitness (Arnqvist & Nilsson, 2000). Polyandry is also seen to generate competition, so is relevant to interpreting several aspects related to evolutionary theory, mainly sexual selection through sperm competition (Simmons, 2005) and cryptic female choice (Eberhard, 1996). Investigations of polyandry are typically laboratory based and focused on the consequences for individual reproductive fitness. But polyandry is widespread enough (Taylor et al., 2014) to suggest its occurrence should not be seen as an evolutionary paradox (Arnqvist & Nilsson, 2000), but an unavoidable consequence of sexual reproduction (Pizzari & Wedell, 2013), with significant consequences at the population level (Lumley et al., 2015; Michalczyk et al., 2011; Rafter et al., 2017). We designed field-based tests to explore the level and impact of polyandry, and so revisit the paradox.

The movement of individuals between discrete resources has implications for the spatio-temporal dynamics and genetics of populations (Clobert et al., 2001; Bowler & Benton, 2005). Individuals that emigrate are purported to have different demographic properties to resident individuals in terms of morphology, behaviour, physiology and life history (Clobert et al., 2004). So when investigating the movement of alleles within and among populations (for example, those that confer insecticide resistance) over spatio-temporally discontinuous resources, it is important to characterise both dispersing and resident individuals that make up the populations (Lemel et al., 1997).

This study focuses on two key insect pests of stored products, Tribolium castaneum (Herbst) (Coleoptera: Tenebrionidae) and Rhyzopertha dominica (F.) (Coleoptera: Bostrichidae). Tribolium castaneum is commonly used in studies of polyandry and sexual selection as it is easy to rear and highly polyandrous in the laboratory (e.g. Demont et al., 2014; Fedina & Lewis, 2008; Pai et al., 2007). Most studies have focused on reproductive success (Lewis et al., 2005), offspring fitness (Pai & Yan, 2002), sperm competition (Lewis & Austad, 1990; Lewis, 2004), cryptic female choice (Fedina, 2007; Fedina & Lewis, 2007) and the possible selective advantages of polyandry (Michalczyk et al., 2011; Grazer & Martin, 2012), and all of these studies have been conducted in the laboratory. The number of times that females mate evidently varies substantially across laboratory strains, with females of one particular strain never mating more than once and females of another mating up to 12 times within the hour under observation (Pai et al., 2007). It is also

37 evidently impacted by density and parasite load (Kerstes et al., 2013). The relevance of the interpretations offered in these studies is compromised somewhat by the lack of information about the level of polyandry in natural populations of T. castaneum (as discussed by Demont et al., 2014). Much less is known about the extent of polyandry in R. dominica, but confined pairs have been observed to mate several times (Thompson, 1966).

Tribolium castaneum has traditionally been considered to have limited flight ability (Good, 1933; Drury et al., 2009), but a recent study demonstrated significant levels of active movement across extensive agricultural landscapes (Ridley et al., 2011). Further, 88% of females trapped in flight had mated and were able to produce offspring for up to 20 weeks after capture (Ridley et al., 2011), but no resident individuals were included for comparison. By contrast, R. dominica adults are known as strong fliers, having often been caught in relatively large numbers at substantial distances from grain storage (Cogburn, 1988; Edde et al., 2006; Ridley et al., 2016), and across a diversity of habitats (Mahroof et al., 2010). Further, 95% of females caught whilst flying from a grain store had mated, as they produced offspring for up to 10 weeks after capture (Ridley et al., 2016).

To understand how these flying insects contribute to the spatio-temporal dynamics of each species, the mating status and potential reproductive output of those that disperse relative to those that remain within resources need to be quantified under field conditions (Lemel et al., 1997). Such species-specific information on polyandry would not only help interpret local structure in the population genetics of each species, but would have added value in dealing with applied issues such as insecticide resistance. To illustrate this, resistance to the fumigant phosphine (PH3) is a developing global issue within both T. castaneum and R. dominica, and in many other stored product pests (Price & Mills, 1988; Behalima et al., 2004; Pimentel et al., 2007; Taylor & Halliday, 1986; Opit et al., 2012; Jagadeesan et al., 2012). Understanding the extent of polyandry in field populations is crucial to interpreting the spatial spread of resistance and its dynamics within localities (where beetles settle). Multiple mating has been demonstrated to alter substantially the rate at which insecticide resistance spreads, whether within populations or spatially across discontinuous resources (Rafter et al., 2017; Blanco et al., 2010).

We therefore focused our investigations on field populations of T. castaneum and R. dominica, and primarily on reconstructing the mating history of dispersing (caught in flight) and resident (at the time of collection) beetles with molecular data from them and their offspring. The offspring tested were derived from the fecundity assessments from each captured female. Specifically, the number

38 of males contributing to a female’s progeny was estimated by screening mother–offspring arrays with microsatellite markers. These results also allowed tests of whether polyandry impacted upon offspring production. We included, for comparison, insects from cultures established recently from field-collected material, to establish whether mating frequency might change through selection pressures imposed during the culturing process.

3.2. Materials and methods

3.2.1. Field sample collections

We sampled beetles at two large grain depots in Central Queensland in Australia on three separate occasions, at Emerald (23°31′S 148°09′E) during April in 2013 and 2014, and 50 km north of Emerald at Capella (23°04′S 148°01′E) in January 2014.

Tribolium castaneum and R. dominica adults were intercepted in flight using pheromone traps, and others were simultaneously sieved from grain at the nearby storage facility. Commercially available T. castaneum and R. dominica pheromone lures (Insects Limited Inc., Indiana, USA) were attached to Lindgren funnel traps (Contech Inc., Victoria, Canada) that were deployed 1.5 m above the ground and at least 30 m from any grain storage facility or equipment (n = 16). At the bottom of each funnel trap, an easily removable collection jar was attached. Sampling commenced 2 h prior to and for 1 h after sunset (dusk), as this is when both species are most active (Wright & Morton, 1995; Rafter et al., 2015). Traps were monitored every 5 min during the sampling period, so that trapped specimens could be removed immediately to ensure no mating could occur in the collection jar. Each beetle was placed alone in a separate 7-ml plastic vial containing 7 g of whole wheat flour + 5% yeast for T. castaneum or 7 g of wheat (12% moisture content) for R. dominica and then transported to the laboratory.

3.2.2. Survival and fecundity of field collected beetles

Each beetle was monitored for 12 weeks by transferring it into a new diet cup containing 10 g of rearing medium (as defined above for each species) weekly. All diet cups were held for a number of weeks after the beetle had been removed (4 weeks for T. castaneum, 6 weeks for R. dominica because they differ from one another in generation time), at which point the adult offspring were counted and placed, by brood, in vials of ethanol for genetic analysis. Females were assumed not to have been mated at the time of their capture if they did not produce any offspring during the

12-week observation period, but only if they lived for at least 3 weeks in the laboratory. Field- collected beetles were sexed at the end of the 12-week observation period (or earlier if they died). Tribolium castaneum beetles were identified by the presence (male) or absence (female) of a sub- basal setiferous puncture on the ventral side of the anterior femora (Halstead, 1963). Rhyzopertha dominica beetles were sexed by squeezing each one until the tip of its genitalia appeared, to show either a pair of pointy ended parameres (male) or a pair of blunt ended styli (female) (Crombie, 1941).

3.2.3. Cultured field beetles for parentage analysis

New laboratory cultures of both T. castaneum and R. dominica were established with beetles collected from an on-farm storage facility in Dalby (27°11’S, 151°16’E), southeast Queensland, in January 2015. Separate cultures were also started with beetles from long-standing laboratory reference strains of T. castaneum (QTC4) and R. dominica (QRD14) (both phosphine susceptible). These strains had originally been established with beetles collected from storage facilities in Queensland: QTC4 in Brisbane (27°28’S, 158°02’E) in 1965 and QRD14 in Oakey (27°25’S, 151°43’E) in 1971 (Bengston et al., 1999; Schlipalius et al., 2008). All colonies (newly established laboratory cultures and long-standing reference cultures) were initiated simultaneously, so that the genetic analysis of polyandry would be on beetles from different sources, but reared under identical conditions.

The T. castaneum cultures were established with 200 adult insects in a 1-L glass jar containing a rearing medium of wheat flour (400 g) and yeast (20 g). These 200 adults were then sieved from the medium 2 weeks after starting the initial culture. Rhyzopertha dominica was cultured in similar manner, except that the culture was initiated with 600 adults in glass jars containing 400 g whole wheat at 12% moisture content. Progeny from the newly established laboratory cultures and long- standing reference cultures were collected 3 weeks after first adult emergence. The T. castaneum adults were removed from the culture jar and then sexed until 30 females were obtained. Likewise R. dominica adults were removed and sexed, but as the sexing technique can impact upon reproductive output (Sinclair, 1981), we used double the number of replicates (60) to maximise the likelihood that 30 females were obtained for analysis, and sexed the beetles after the experiment.

Each beetle removed from culture was placed individually in a diet cup containing 10 g of rearing medium and then held at 250C and 65% r.h. After 1 week, the diet cups were checked for dead beetles, and live beetles were transferred to a fresh diet cup. The same was done after the second

40 week, when all adults were labelled individually and stored in ethanol for genetic analysis. Each diet cup was held for 6 weeks after the founding female beetle had been removed, at which point the numbers of offspring produced over the 2-week period (and by now adult) were counted and those offspring from the second oviposition week were stored in ethanol for genetic analysis.

3.2.4. Genotyping and paternity analysis

Genomic DNA was extracted from T. castaneum and R. dominica females following the protocol of Ridley et al. (2016). Sixty field-collected females, 30 new laboratory culture females and 30 laboratory cultured beetles of each species were screened with microsatellite markers (T. castaneum: Demuth et al., 2007; R. dominica: Ridley et al., 2016) to identify the most polymorphic of the markers, and to determine microsatellite allele frequencies in the populations. Though the field-caught beetles were from three distinct sampling periods (across locations 30 km apart), they were treated as a single population as allele frequencies were statistically similar across these samples (and within species), and previous studies have demonstrated that T. castaneum and R. dominica populations are often genetically homogenous through time, and across much larger spatial scales than covered by the current study (Ridley et al., 2011; 2016). The five markers with the highest levels of polymorphism were subsequently used for parentage analyses.

Individual parent females, along with 15 of her Week 2 progeny, were genotyped using at least four of the five loci. Offspring from Week 2 of oviposition were selected as any initial sperm stratification resulting from multiple matings should have been entirely disrupted by this time (Lewis & Jutkiewicz, 1998). Microsatellite peaks were confirmed and binned manually using the Geneious 7.2.1 microsatellite plug-in (Kearse et al., 2012). The number of alleles per locus and allele frequencies, along with observed (HO) and expected (HE) heterozygosity, was calculated in GenAlEX 6.501 (Peakall & Smouse, 2006; 2012). Deviations from Hardy–Weinberg equilibrium and tests of linkage disequilibrium between loci were conducted in GENEPOP (Rousset, 2008). Null allele frequencies were estimated using FreeNA (Chapuis & Estoup, 2007).

We used the multilocus approach of GERUD 2.0 (Jones, 2008) to estimate the minimum number of fathers that are likely to have contributed to progeny arrays. This software does not allow for missing data, so only offspring that amplified at all loci (a minimum of 13 offspring per female) were included in the analysis (22 samples in total were discarded as they did not meet these requirements). The most likely paternal genotypes and the probable contribution of each paternal genotype were estimated, using the most likely minimum father combination in GERUD 2.0. When

41 more than one solution was possible, we used population allele frequencies to rank the solutions.

Exclusion probabilities for individual loci, and all loci combined, were also calculated using GERUD 2.0. We assessed the reliability of our polyandry analysis using the program GERUDSIM 2.0 (Jones, 2008). We ran 1000 simulations and determined the proportion of simulations that correctly assigned the true number of fathers (from 2 to 5) given the population allele frequencies of the markers used, and the observed skew in paternal contributions. We ran simulations with the mother’s genotype known, together with 15 offspring per female analysed.

3.2.5. Statistics

Survivorship of T. castaneum and R. dominica over the 12 week observation period in the laboratory were examined by Pearson’s chi-square test with Yate’s continuity correction in R (R Development Core Team), first by testing for differences in sampling date (with the Emerald 2014 trapping data omitted as no beetles were found in storage) and then by collection source (with the Emerald 2014 trapping data included). Sex ratio data were analysed by a goodness of fit chi- square analysis, on raw data, in R.

The mean fecundity of females collected from storage or trapped flying was analysed with a two- way ANOVA with sampling location (collected from storage and trapped flying) and the number of males contributing to progeny as factors (virgins removed from analysis). A linear regression of the number of weeks that females produced progeny against the total number of progeny produced was also conducted in R for each species. This was followed by an ANCOVA to compare the regression slopes of female beetles collected from storage and trapped during flight.

The numbers of males contributing to progeny of females collected from storage or trapped flying as well as the paternal contributions of these males were analysed by GLM with Poisson distribution (none of these tests was overdispersed).

3.3. Results

3.3.1. Tribolium castaneum a) Survival and fecundity

There was no significant difference in the survivorship across the field collected T. castaneum males and females when analysed for each sampling location independently (P > 0.05). Survival of beetles (sexes analysed together) was also not statistically different when collected from either Emerald in 2013 or Capella in 2014 (훘2 = 0.79, P = 0.37; Table 3.1). Survivorship of T. castaneum over 6–12 weeks was significantly higher in beetles trapped during flight than those collected from storage (훘 2 = 4.65, P = 0.03) (Table 3.1). The sex ratio of beetles collected from storage (50.7% ♀) or trapped in flight (56.1% ♀) did not deviate from unity (P > 0.05).

Almost all T. castaneum females had mated, whether collected from bulk storage (94%, n = 34), or trapped flying (97%, n = 32). Over the 12-week observation period, no significant difference was detected between the total number of progeny produced by T. castaneum females sampled from storage (286.1 ± 34.4) and those trapped flying (345.8 ± 44.4) (F1 = 0.03, P = 0.86). An analysis of covariance revealed a significant correlation between the number of successful oviposition weeks for each female (within the 12-week observation period) and the total number of progeny produced in beetles collected from storage (Figure 3.1a) as well as those trapped in flight (Figure 3.1b) (P < 0.001 in both cases). There was no significant difference in the slopes or intercepts of correlation of these two groups (P = 0.5369).

Table 3.1 Survival (%, sexes combined (see text)) of Tribolium castaneum six and twelve weeks after collection from the field (either in flight or from storage).

Species Location Source n Survival (%) T. castaneum 6 weeks 12 weeks Emerald 2013 Storage 33 77.1 62.9 Flight 31 80.6 74.2 Emerald 2014 Storage 0 - - Flight 18 100 88.9 Capella 2014 Storage 41 87.8 73.2 Flight 16 100 93.8

a) Collected from storage b) Trapped flying

Total Total progeny produced

Weeks of progeny production Figure 3.1 The number of weeks in which progeny were produced by individual females against their total number of progeny produced within the 12 week observation period for: a) Tribolium castaneum collected from storage, b) T. castaneum trapped flying. b) Paternity analysis

No loci exhibited significant departure from Hardy-Weinberg equilibrium or any evidence of null alleles, and no linkage disequilibrium was detected between any of the loci. Only a few alleles were located at each of the loci in the QTC4 laboratory strain (1-4 alleles/locus), so this strain was excluded from further analysis; the markers were not variable enough to detect polyandry. The five loci used in the analyses of the other beetles had a moderate number of alleles (field collected: 6.2 alleles/locus, new laboratory culture: 6.6 alleles/locus) but relatively high genetic diversity (field collected: average HE = 0.662, new laboratory culture: average HE = 0.679) (Table 3.3). Our power to detect multiple paternity was high in both the field caught beetles and those from a new laboratory culture, with allele frequencies giving a 5-locus exclusion probability (with one parent known) of 0.954 for field caught beetles and 0.973 for new laboratory culture beetles (Table 3.3). Likewise, simulation analyses done using GERUDSIM 2.0 suggested that contributions from multiple males would be detected 96.3% of the time in both of the field caught beetles and those from the newly established laboratory culture. The power to detect three fathers was lower in both field caught (65.6%) and newly established laboratory (64.0%) beetles, with four fathers detected only rarely (field caught: 16.0%; new laboratory: 18.4%).

Paternity analysis indicated that 90.6% of the 53 mated T. castaneum females produced offspring from more than one male, with the estimated number of fathers ranging from one to four (Figure 3.3a, b). No significant difference was detected in the number of males contributing to progeny in

44 females from storage (2.4 ± 0.1) or those trapped flying (2.4 ± 0.1) (GLM, Z = -0.1, P = 0.9) (Figure 3.3a, b). Similarly 91.7% of the 24 mated females in the new laboratory culture produced offspring from more than one male, with the estimated number of fathers ranging from one to four (mean: 2.7 ± 0.3; Figure 3.3c). The estimated number of fathers detected did not significantly impact the fecundity of females captured in the field (in storage or flight) (F3 = 0.9, P = 0.4).

The reproductive contributions of each of the fathers in multiply sired broods varied greatly within the field caught T. castaneum females as well as those from the newly established laboratory culture. In broods with two fathers, ‘male a’ sired on average 67.9% of the offspring in field caught beetles and 84.2% of the offspring in new laboratory cultured beetles (Figure 3.4 a-c), with both deviating significantly from the null expectation of 50% (field caught: GLM, Z = -6.7, P < 0.001; newly established laboratory culture GLM, Z = -6.7, P < 0.001). Likewise, in broods with three fathers a significant paternity skew in both the field caught (GLM, Z = -4.2, P < 0.001) and newly established laboratory culture (GLM, Z = -2.3, P = 0.02) beetles was evident, with ‘male a’ siring on average 50.5% of the offspring in field caught beetles and 50.0% of offspring in newly established laboratory beetles (Fig. 3.4 a-c). There was no significant difference in the relative contribution of males across beetles collected from storage and those caught in flight (two male paternity GLM, Z = -0.1, P = 1.0; three male paternity GLM, Z = 0.12, P = 0.9).

3.3.2. Rhyzopertha dominica a) Survival and fecundity

Survivorship of R. dominica was similar irrespective of sex (P > 0.05), sampling location (훘2 = 3.55, P = 0.06), or whether beetles were collected from storage or caught during flight (훘2 = 0.16, P = 0.68) (Table 3.2). The sex ratio of beetles collected from storage (42.3% ♀) or trapped in flight (60.3% ♀) did not deviate significantly from unity (P > 0.05).

Almost all R. dominica females (97%) had mated, whether collected from bulk storage (n = 38) or caught flying (n = 36). However, females trapped flying were significantly more fecund (333.5 ±

28.0) than those sampled from storage (179.7 ± 20.6) (F1 = 9.1, P = 0.004). An analysis of covariance revealed a significant correlation between the number of weeks for which each female oviposited (within the 12-week observation period) and the total number of progeny produced, in beetles collected from storage (Figure 3.2a) and those trapped in flight (Figure 3.2b) (P < 0.001 in both cases). Although the correlation between the number of offspring produced and the weeks of

45 progeny production was stronger in R. dominica trapped flying, it was not significantly different from that for females collected from storage (P = 0.08).

Table 3.2 Survival (%, sexes combined (see text)) of Rhyzopertha dominica six and twelve weeks after collection from the field (either in flight or from storage).

Species Location Source n Survival (%) R. dominica 6 weeks 12 weeks Emerald 2013 Storage 34 82.4 79.4 Flight 43 92.9 85.7 Emerald 2014 Storage 0 - - Flight 14 78.6 78.6 Capella 2014 Storage 57 87.9 69.0 Flight 3 66.7 0

a) Collected from storage b) Trapped flying

Total Total progeny produced

Weeks of progeny production

46 b) Paternity analysis

No loci exhibited significant departure from Hardy-Weinberg equilibrium or any evidence of null alleles, and no linkage disequilibrium was detected between any of the loci. There were too few alleles at all loci in the QRD14 laboratory strain (1-4 alleles/locus), to detect polyandry, so this strain was excluded from further analysis. The five loci used in analyses had a high number of alleles (field collected: 11.5 alleles / locus, field cultured: 14.0 alleles / locus) and genetic diversity

(field collected: average HE = 0.858, field cultured: average HE = 0.887; Table 2). Our power to detect multiple paternity was high in both the field caught and newly established laboratory culture R. dominica populations, with allele frequencies from adult females giving 5-locus exclusion probability (with one parent known) of 0.993 for field caught and 0.995 for new laboratory populations (Table 3.3). Likewise, simulation analyses in GERUDSIM 2.0 suggested that contributions from multiple males would be detected 99.8% of the time in field caught populations and 99.9% of the time in the new laboratory population. The power to detect three fathers or four fathers was likewise high in field caught (3 fathers: 99.3%, 4 fathers: 89.8%) and newly established laboratory (3 fathers: 99.6%, 4 fathers: 91.7%) populations.

Paternity analyses indicated 70.2% of the 57 mated R. dominica females genotyped produced offspring from more than one male, with the estimated number of fathers ranging from one to six (Figure 3.3 d, e). No significant difference was found in the number of males contributing to progeny in females from storage (2.1 ± 0.2) or those trapped flying (2.5 ± 0.2) (GLM, Z = -1.0, P = 0.3) (Figure 3.3d, e). Similarly 83.3% of the 18 mated females in the new laboratory culture also produced offspring from more than one male, with the estimated number of fathers ranging from one to four (mean: 2.3 ± 0.2; Figure 3.3f). The estimated number of fathers detected did not significantly impact the fecundity of females in the field samples (F3 = 0.4, P = 0.8).

As was the case for T. castaneum, the reproductive contributions of fathers in multiply sired R. dominica broods varied greatly in both the field caught and newly established laboratory beetles. In broods with two fathers, ‘male a’ sired on average 73.0% of the offspring in field caught beetles and 79.0% of the offspring in new laboratory beetles (Figure 3.4 d-f), deviating significantly from the null expectation of 50% (field caught: GLM, Z = -8.9, P < 0.001; new laboratory culture: GLM, Z = -5.5, P < 0.001). Likewise, in broods with three fathers significant paternity skew was clear in both the field caught (GLM, Z = -8.9, P < 0.001) and newly established laboratory culture (GLM, Z = - 2.53, P = 0.01) beetles with ‘male a’ siring 55.5% of the offspring in field caught beetles and 52.4% of offspring in new laboratory beetles (Figure 3.4d-f). No significant difference in relative

47 contribution of males was detected between beetles collected from storage and those caught in flight (two male paternity GLM, Z = 0.04, P = 1.0; three male paternity GLM, Z = 0.06, P = 1.0).

Table 3.3 Characteristics of five microsatellite markers used to infer paternity in field collected female beetles and in female beetles from new laboratory populations of Tribolium castaneum and Rhyzopertha dominica. NA = number of alleles. HO = observed heterozygosity. HE = expected heterozygosity. PEXC = exclusion probability.

Field caught New laboratory culture

Species Locus NA HO HE PEXC NA HO HE PEXC T. castaneum Tc-2.7 8 0.596 0.590 0.368 7 0.458 0.775 0.555 Tc-7.1 6 0.675 0.752 0.510 7 0.849 0.778 0.659 Tc-7.4 4 0.700 0.570 0.278 5 0.875 0.619 0.362 Tc-8.6 4 0.561 0.623 0.391 3 0.375 0.388 0.198 Tc-8.8 7 0.725 0.776 0.546 11 0.833 0.833 0.650 Average 5.8 0.651 0.662 0.954 6.6 0.678 0.679 0.973 R. dominica RL7 13 0.750 0.817 0.643 11 0.783 0.899 0.755 RL33 10 0.636 0.850 0.684 9 0.833 0.866 0.700 RL35 19 0.697 0.879 0.755 14 0.909 0.952 0.818 RL39 14 0.750 0.885 0.750 not used RL36 not used 12 0.667 0.830 0.645 Average 14 0.708 0.858 0.993 11.5 0.798 0.887 0.995

a) b) c)

a a a

a d) a e) a f)

a) b) c)

a a a

a d) a e) a f)

3.4. Discussion

Here we provide the first direct evidence that polyandry occurs in both T. castaneum and R. dominica populations in the field. Few field-sampled beetles were non-reproductive (6% T. castaneum (n = 66); 3% R. dominica (n = 74)) and most females had mated with more than one male (91% T. castaneum (n = 53); 70% R. dominica (n = 57)) (Figure. 3.3). The number of paternal males detected did not impact on female fecundity for either species (Figure 3.3). A difference in offspring production across dispersing and resident females in R. dominica was statistically significant, with females sampled from storage less fecund, but no such differences were evident in T. castaneum (Figure 3.3). Similar levels of polyandry were also observed in newly established laboratory cultures of both T. castaneum and R. dominica (Figure 3.3), however, comparisons with older laboratory cultures were not feasible because the genetic diversity of markers in these strains was too low.

No difference was detected in the mating frequency or fecundity of resident and dispersing T. castaneum in our data (Figures 3.1, 3.2, 3.3, 3.4). The low proportion of non-reproductive females in both storage and flying samples indicates that T. castaneum females mate soon after eclosion and prior to dispersal, with most of the females collected having mated multiple times prior to flight (Figure 3.3). This pattern of mating and fecundity in T. castaneum means that dispersers represent a random subset of individuals with regard to flight, mating frequency and fecundity, contrary to the expectations of Clobert et al. (2004). The lack of reproductive differentiation across the T. castaneum field samples could be related to the unusually extended duration of adult life in this species (Nilsson et al., 2002). But we do need to know when, after ecolsion, T. castaneum individuals are likely to disperse, and whether they disperse multiple times during their adult life. Survivorship of T. castaneum was significantly higher in beetles trapped during flight than those collected from storage (Table 3.1), which implies that flying beetles may be younger than beetles in storage.

Reproductive differences between resident and dispersing individuals were, by contrast, observed in R. dominica, with females collected from storage producing fewer offspring than those caught flying (Fig. 3.2), and a higher proportion of beetles had mated only once prior to flight (30%; n = 36). This species also has an extended adult period but, unlike T. castaneum, flight is purported to occur more frequently during the 2 weeks following eclosion (Barrer et al., 1993; Aslam et al., 1994; Dowdy, 1994) and the data presented here support this assertion. Further sampling incorporating age, physiology, phosphine resistance status and morphometric analyses (such as

51 weight and lipid content) are required to better characterise dispersing and resident individuals (for both species) better, as the differences between them may well be subtle (Clobert et al., 2004).

Population density is another factor that is expected to impact on mating systems (Kokko & Rankin, 2006), with females often reported to mate more frequently in the laboratory than they would in the wild (Sadek 2001; Bretman & Tregenza, 2005). Studies on Drosophila serrata (Malloch) have, however, estimated similar levels of polyandry in laboratory and field populations (Frentiu & Chenoweth, 2008), and that is the pattern reflected in our data (Figures 3.1, 3.2, 3.3). The levels of polyandry recorded in the field-collected insects and newly established laboratory cultures of T. castaneum were less than predicted based on laboratory assessments of multiple mating involving long-term laboratory reference strains of this species (Pai & Yan, 2003; Pai et al., 2007). However, not all copulations lead to fertilisation (Bretman & Tregenza, 2005; Turnell & Shaw, 2015) so direct comparisons across studies are not straightforward. The newly established laboratory colonies in our study were cultured for only a single generation before being used in tests. Laboratory reference strains used in previous laboratory assessments (e.g. Pai et al., 2005; Fedina & Lewis, 2007; Grazer & Martin, 2012) have been in culture for multiple generations [> 30 years in some cases (Wade, 1977)]. Culturing has been demonstrated to impact negatively upon movement, fecundity and development time in T. castaneum (Malekpour et al., 2016; Ahmad et al., 2012; White, 1984). Extended culturing is therefore likely to have imposed selection on aspects of the mating behaviour, including mating propensity, as well.

We detected significant paternity skew in both T. castaneum and R. dominica females that had mated multiple times (Figure 3.4). Previous investigations into the storage capacity of the spermatheca in T. castaneum found that only 33% more sperm was stored in doubly mated compared to singly mated females and the number of sperm remained constant across females that had experienced two or three matings (Lewis & Juthiewicz, 1998). From this, it can be inferred that the spermatheca is filled to two-thirds capacity after a single mating. When females mate multiple times, sperm within spermatheca has been shown to be completely mixed within 2 weeks of the last mating (Lewis & Juthiewicz, 1998). This is the point at which offspring in this study were collected for genotyping. The observed paternity skew is therefore not likely attributable to sperm stratification or sperm precedence (as suggested by Haubruge et al., 1997), but rather a consequence of the storage capacity of the spermatheca (Sinha, 1953; Arnaud et al., 2001a) and inter-mating interval (Arnaud et al., 2001b). The pattern of male paternity is similar across T. castaneum and R. dominica (and across study locations within each of the species) (Figure 3.4), which further indicates that pattern of paternity, in both of these species, is likely dictated by the

52 storage capacity of their spermathecae, the proportion of that capacity taken by the first male and the interval between mates.

The ecological consequences of multiple mating and sperm mixing in females before they undertake migratory flight is likely to have significant implications for: i) the success rates in the colonization of new sites (whether occupied by conspecifics or not), and ii) the geographical movement of alleles and their isolation (within the spermatheca) from any selection that might otherwise have been imposed. The results presented here demonstrate that both dispersing and resident females carry a diversity of male genetic material which means they are more likely to produce recombinant types that are better able to cope with unpredictable changes in the environment. That is, polyandry enhances these effects, to the extent that females, through polyandry, carry a much higher diversity of genetic material than would otherwise be the case. And that diversity should, in principle, increase the chances of successful establishment of the next generation in a new locality (whether occupied by conspecifics or not).

Our results provide a greater understanding not only of the movement of alleles within and among populations, but also of the diversity of alleles that females may carry when they do migrate and colonise new areas. This has implications for the spatial spread of resistance, including resistance to the fumigant phosphine. Multiple mating increases the probability that an individual female will (regardless of her phosphine resistance status) transfer phosphine resistance genes to new populations (given a high enough frequency of phosphine resistance in the population). In addition, mated females from highly resistant populations migrating into susceptible populations can re-mate and not only produce fully resistant offspring but also susceptible heterozygotes (Rafter et al., 2017). Polyandry also has implications for the genetic diversity across the offspring of resistant resident females that survive fumigation. That is, sperm stored within such females should be screened from the effects of phosphine (including those sperm from phosphine susceptible males with which she may have mated) (Rafter et al., 2017). Once fumigation has been completed, the multiple-mated females will produce progeny of mixed resistance genotypes and that has the potential to influence the dynamics of resistance alleles in the population, something that needs consideration in the development of models of phosphine resistance. These advantageous consequences of polyandry and the fact that most of the individual beetles samples in this study had mated multiple times provide justification as to why polyandry should not be viewed as an evolutionary paradox.

CHAPTER 4

The dispersal flight of Tribolium castaneum – a field test of laboratory generated predictions

54 4.1. Introduction

The study of movement is incomplete without understanding an organism’s behaviour in its natural environment. Environmental variables are far more complex in the field than under laboratory conditions and thus can have completely different effects on the physiology of organisms. Tribolium castaneum requires temperatures from 20 to 300C with high relative humidity for successful reproduction and development of offspring (Mukerji & Sinha, 1953; Howe, 1956; Singh & Prakash, 2015). Their ability to use a wide range of grain resources, processed food, and micro fungi (e.g., on cotton lint) (Good, 1933; Sokoloff, 1974; Ahmad et al., 2012ab; Ahmad et al., 2013ab) along with their high fecundity (Daglish, 2005) and polyandry (Chapter 3; Pai et al., 2007; Fedina & Lewis, 2008; Demont et al., 2014) are features that facilitate movement and colonization across extensive landscapes (Ridley et al., 2011).

Aspects of the dispersal of T. castaneum in the field can be understood by characterizing both environmental and physiological variables that promote or lead to initiation of flight in these beetles. While many studies have been conducted to characterize the abiotic and physiological factors that influence flight initiation in T. castaneum in the laboratory (Perez-Mendoza et al., 2011ab; Perez-Mendoza et al., 2014; Drury et al., 2016 Chapters 2 & 3), it is difficult to replicate these tests in the field. Flight assays in the field are hindered by low capture or re-capture rates (Bullock et al., 2012), as well as difficulties in reliably characterizing physiological variables such as age (Sokoloff, 1974), mating and insecticide resistance status. The ability to characterize the age of these beetles would be of particular importance as adults of this species are long lived (from 6 months up to more than a year) (Sokoloff, 1974). Environmental factors such as temperature, wind speed, light intensity and humidity cannot be controlled in the field and most of these have been shown to have a significant effect on the flight of these beetles (Perez-Mendoza et al., 2014; Rafter et al., 2015; Drury et al., 2016). The lack of stable external markers to track flight dispersal in these beetles by mark-recapture studies is another impediment to the investigation of flight in the field. Therefore, studying movement and tracking distances travelled and then relating these to the physiological status of an individual is complicated.

Characterizing dispersal is also vital because the effective management of T. castaneum relies mainly on phosphine fumigation. But with the advent of bulk grain storage and regular exposure to phosphine, these beetles have developed resistance to phosphine (Jagadeesan et al., 2012). This poses a significant problem, as currently there is no other way of controlling these pests effectively. Phosphine is an ideal fumigant because it leaves no residue on wheat and is safe for the

55 environment (Chaudhry, 2000; Nayak & Collins, 2008). The polyandrous behaviour of the female beetles (Chapter 3; Pai & Bernasconi, 2008) increases the chances of spreading resistance geographically, since T. castaneum is known to exploit fresh stockpiles of grains (Campbell & Runnion, 2002).

The laboratory flight initiation assays described in Chapter 2 found that emigrating T. castaneum beetles do not differ from resident beetles in any of the physiological factors tested, besides age. Chapter 3 quantified the fecundity and level of polyandry in field populations of T. castaneum by comparing resident with dispersing individuals at common sites and found that almost all females caught in flight had mated, with paternity analysis revealing that most females had mated with more than one male. Based on these results the following predictions regarding flight were made for the field populations. 1) Flight propensity is related to age, as more young beetles flew than the old ones (Chapter 2). 2) Flight propensity is not determined by beetle sex, weight or size. 3) Fecundity may be higher in beetles caught in flight, as indicated by Chapter 3 field sampling. 4) Females, regardless of where they are sampled, are likely to have mated with multiple males. 5) Fat content of young beetles is higher than that of old beetles (resident), but is not significantly related to flight propensity. 6) The lipofuscin content of flying beetles will be higher than that of resident beetles presumably due to the higher metabolic costs of flight.

Based on these predictions a field test was designed to study the flight dispersal of T. castaneum at different distances from a heavily infested cotton seed store to draw comparisons between the laboratory and the field beetles. Beetles were characterized in the same manner employed in the previous two chapters to test whether any variables were exclusively contributing to flight initiation in the field. The general question asked was what leads to dispersal in T. castaneum in the field? The variables studied in these tests were weight, body size, fat content, fecundity, polyandry, and the relation between lipofuscin intensity and flight status. The effect of the presence and absence of males on fecundity and survival of female beetles was also studied as an addition to the fecundity tests performed in Chapters 2 and 3.

4.2. Materials and Methods

4.2.1 Field tests

The field test was conducted at Moura (24° 35' 35.88" S and 149° 59' 58.07" E), Queensland, Australia in the grounds of a cotton gin. Cotton seed stripped of lint is stored in bulk for later use

56 and becomes heavily infested with T. castaneum. Lindgren funnel traps baited with 4,8- dimethyldecanal (DMD, the aggregation pheromone for T. castaneum and T. confusum, the latter of which does not occur in the study area) (Suzuki & Mori, 1983) were set up in and around the seed storage to intercept flying beetles. The traps were arranged randomly within the storage shed and in a radial pattern at about 20 m and 300 m around the storage shed. Beetles were also collected from cotton seeds and were called ‘resident’ beetles, because they showed no signs of moving out of the seeds at the time when these beetles fly (late afternoon to dusk (Rafter et al., 2015)). So, there were four treatments in total out of which three were based on distance from the source of beetles. Each captured beetle was held alone in a 30 mL plastic cup with 5 g flour and 5% Torula yeast for transport to the laboratory (about 3 days). In the laboratory the beetles were sexed, weighed, and females from each treatment were assigned for fecundity analysis. A15 g flour and 5% Torula yeast mixture was added to each of their cups. Additional female and male beetles were assigned for other measurements, including morphometrics, fat content, degree of polyandry and lipofuscin intensity, and thus to test the predictions outlined in the introduction. The following sub-sections describe the methods used in these tests.

4.2.2. Measuring fecundity and survival

Female beetles obtained from each treatment, namely flying within the shed (n = 34), flying outside the shed (20 m, n = 30 and 300 m, n = 30) and resident beetles (non-flying, n = 26) were kept individually on 20 g flour and Torula yeast medium (260C, 60% RH). Fecundity was recorded at each 15-day time interval, when each female was transferred to a new cup with culture medium. Adult female survival was also monitored until they perished. After about 6 weeks post seeding of parental females the cups were observed for the first adults. The final counting of adult offspring was conducted 12-15 days post emergence of the first adults. Usually no pupae were observed beyond this time and, if present, were nearing emergence so were included in the progeny count (adult emergence was not recorded). After 105 days, males (~2 week old adults) were introduced into one half of the cups with surviving females (randomly selected) from each treatment [one male per cup, flying within shed (n = 15), flying outside shed 20 m (n = 10), flying outside shed 300 m (n = 14) and resident (n = 9), with replicates having been reduced slightly by some mortality]. The rest were maintained as before and the fecundity of each one was also recorded every 15 days until all females had perished [flying within shed (n = 14), flying outside shed 20 m (n = 11), flying outside shed 300 m (n = 11) and resident (n = 8)]. This was done to study whether mating influences the survival and fecundity of old beetles.

57 4.2.3. Polyandry analysis

The level of polyandry was tested on 15 offspring from each parent beetle, with offspring taken from each of 18 females from each treatment in the fecundity tests at 16-30 days into that experiment (see section 4.2.1). Genomic DNA extraction was performed by using EDTA aided Tris-HCL (tris(hydroxymethyl)aminomethane) digestion of cell membranes followed by protein digestion using proteinase K using DNEasy kits (Qiagen Pty Ltd, Chadstone, Victoria, Australia). Four microsatellite loci were selected on the basis of polymorphisms reported in Queensland populations (Demuth et al., 2007, supplementary material; Ridley et al., 2011; Chapter 3; Appendix 3.1) and were amplified using the polymerase chain reaction (PCR) protocol (Demuth et al., 2007). The 5’ ends of forward primers were functionalized with M13 (-21) oligonucleotide adapter sequence. M13 primer labelled with fluorescent dyes (NED, VIC, FAM and PET) was added to each reaction to aid in detection of the DNA sequence. The final products of the PCR were cleaned by using an equal ratio of Exonuclease I and Antarctic Phosphatase (New England Biolabs, Ipswich, USA). These samples were pooled such that each fluorescent dye labelled only one locus in each pool (Ridley et al., 2016). The samples were analysed by ABI3730 (Macrogen Inc. Seoul, South Korea). DNA peaks were confirmed and binned manually using Geneious version 7, with the help of the microsatellite plugin, and GERUD 2.0 (Jones, 2005; Ridley et al., 2016, Kearse et al.,

2012). The observed (HO) and expected (HE) heterozygosity and deviations from Hardy-Weinberg equilibrium were analysed in GENEPOP (Rousset, 2008). Null allele frequencies were estimated using FreeNA (Chapuis and Estoup, 2007). GERUDSIM 2.0 (Jones, 2005) was used to assess the reliability of polyandry analysis (see section 3.2.4, Chapter 3).

4.2.4. Beetle characteristics

The methods employed for measurement of weight, morphometrics, fat content and lipofuscin levels of male and female beetles from all treatments were similar to those listed in Chapter 2, section 2.2.4. Lipofuscin analysis for the beetles flying at a distance of 300 m was not performed because not enough beetles were obtained from the field. Since, lipofuscin was found to be unrelated to beetle age, the lipofuscin intensities in this chapter were tested to check whether they may be related to flight initiation, i.e. whether lipofuscin intensities of flying and resident beetles differ.

58 4.2.5 Statistics

All data were analysed using the R 2.15.1 statistical software package (R Development Core Team, 2013) and GraphPad Prism 7.0 (GraphPad software inclusive, 2016). Survival was analysed by log- rank test. Fecundity was tested by two-way analysis of variance (ANOVA), analysis of covariance (ANCOVA) and generalized linear models (GLM) to test differences in fecundity until day 105 (dependent variable) with time (independent variable) and dispersion distances, i.e., treatments (factor). Individual time intervals were tested by pairwise multiple t-tests and generalized linear models. Two way ANOVA with sex as an independent variable was also used to test weight, morphometrics (where elytron length, thorax length, thorax width and head width were tested against sex and treatments), and lipofuscin intensities of beetles with post hoc Tukey HSD comparisons wherever significant results were obtained. One-way ANOVA was used to test differences in fat content between beetles flying at different distances within each sex. GERUD 2.0 was used to test the likelihood of contributing males. GERUDSIM 2.0 was used to run 1000 simulations to detect the reliability of polyandry analysis (see section 3.2.4). GENEPOP on the web and FreeNA were used to test allelic frequencies, linkage disequilibrium and null alleles.

4.3. Results

4.3.1. Survival

The survival rate of females recorded over successive 15-day time intervals until day 375 was not found to be significantly different across all the treatments (Log-rank test, χ2 = 1.497, d.f. = 3, P = 0.539) (Figure 4.1). The median survival of beetles captured flying inside shed, 20 m outside shed and 300 m outside shed was 180 days while that of resident beetles was 172.5 days. One half of the females from each treatment were introduced to one male individually at 106-120 days. The survival of females after introduction of males did not differ significantly across all the treatments (Log-rank test, χ2 = 4.084, d.f. = 3, P = 0.253). When survival of females with access to males was compared to their female counterparts that were not exposed to males, no significant differences were found in within shed flying (Log-rank test, χ2 = 0.128, d.f. = 1, P = 0.719), flying outside shed (20 m) (Log-rank test, χ2 = 1.042, d.f. = 1, P = 0.307) and resident females (Log-rank test, χ2 = 1.619, d.f. = 1, P = 0.203). However, the survival of females captured 300 m from the storage shed and provided with a male beetle at day 106, was significantly and negatively affected by those males. The females without males in this treatment survived significantly longer (median survival = 195 days) than those with males (median survival = 180 days) (Log-rank test, χ2 = 5.518, d.f. = 1, P = 0.019) (Figure 4.2).

Percent survivalPercent

Time interval (Days)

Figure 4.1 Percent survival of Tribolium castaneum females in each treatment until day 375 (±95%CI) from the day of capture. The treatments included beetles caught flying within the storage shed (n = 34), flying at a distance of 20 m from the shed (n = 30), flying at a distance of 300 m from the storage shed (n = 30) and resident within the pile of cotton seed (n = 26). P > 0.05, for all treatments.

Percent survivalPercent

Time interval (Days) Figure 4.2 Percent survival, with time, of female beetles caught flying at a distance of 300 m from the shed in which cotton seeds were stored. At day 106 (±95%CI), 14 randomly selected replicates had a male beetle added to the container with the female. The other 11 females continued to be held without a male. The black solid line indicates mortality of those females held with males and the red dotted line the mortality of females held without males.

60 4.3.2. Fecundity

The total fecundity across treatments until day 105 was not significantly different (ANOVA, F = 1.57, d.f. = 3, P = 0.20). However, the fecundity of all treatments in the 16-30 days interval after capture was significantly higher than that during the 0-15 days interval (Multiple t-tests: P<0.0001) (Table 4.1). In the laboratory test, by contrast, only the young flying females showed a significant increase in fecundity at 16-30 days after capture (Figure 2.7).

100 Flying within shed Flying outside shed (20m) Flying outside shed (300m)

75 Resident

50 Mean fecundity

0 0-15 16-30 31-45 46-60 61-75 76-90 91-105

Time interval (Days)

61 Table 4.1 Comparison of fecundity at 0-15 days and 16-30 days after capture by multiple t-tests.

Treatment Mean difference ±SE t ratio** d.f. Within shed flying 48.2 4.261 11.31 66

Flying outside shed (20 m) 41.9 6.845 6.121 54

Flying outside shed (300 m) 30.8 5.770 5.338 58

Resident 44.5 5.566 7.995 47 ** All values are significant, P < 0.05

An analysis of covariance (ANCOVA) revealed significant correlation between the oviposition intervals and the number of offspring produced by females in all treatments (F = 1.99, P = 0.008). Further analysis of data using a generalized linear model with negative binomial distribution (as the data were overdispersed) revealed that offspring production (after day 30) significantly decreased with increasing time until day 75 (GLM = R2 = 0.108, P < 0.0001) with no differences between treatments (P = 0.738) (Table 4.2).

Table 4.2 Generalised linear model (negative binomial regression) for effects of time on fecundity (offspring production) of females in all treatments.

Time interval Mean fecundity* Estimate Std. Error z value P Intercept 3.44275 0.11055 31.143 < 2e-16 **

16-30 73.4 0.83094 0.13125 5.906 2.44e-10 **

31-45 69.6 0.77709 0.13157 0.6636 3.50e-09 **

46-60 57.6 0.59200 0.13198 4.485 7.28e-06 **

61-75 42.8 0.29442 0.13282 2.217 0.0266 ** * all treatments combined; ** indicates significant values

62 In general, those females exposed to males from day 106 produced more offspring than females not exposed to males but this difference was not statistically significant (two way ANOVA: F = 1.35, d.f. = 3, P = 0.263) (Figure 4.4).

100 Flying within shed a) WITH ♂ Flying outside shed (20m)

80 Flying outside shed (300m) Resident

0 60

b) WITHOUT ♂ Mean fecundityMean

0 91-105 106-120 121-135 136-150 151-165 166-180 181-195 196-210 Time interval (Days)

63 4.3.3. Beetle weight and size

The mean weight of male and female beetles across treatments did not differ significantly (two way ANOVA: F = 1.822, d.f. = 3, P = 0.145) (Figure 4.5). Similarly, there were no significant differences in body size within and across the sexes in all treatments (Table 4.2).

2 Male

Female 1.5

0.5

Mean weightMean (mg)

0 Flying within shed Flying outside shed Flying outside shed Resident (20m) (300m)

Treatment Figure 4.5 Mean weight (mg) of beetles (±2SE) captured in flight at different distances from a large source of beetles from which, a sample was also taken (= resident on x-axis). Sample sizes are: flying within shed (n = 28 for each sex), flying outside shed (20 m, n = 28♀♀, n = 22♂♂ & 300 m, n = 23♀♀, n = 20♂♂), and resident (n = 26♀♀, n = 25♂♂). All treatments showed no statistical significance (P>0.05).

Body part F d.f. P Head width 1.838 3 0.141

Thorax length 2.533 3 0.058

Thorax width 1.943 3 0.124

Elytron length 0.157 3 0.925

*Sample size is the same as for the mean weight data (Figure 4.5)

64 4.3.4. Fat content

The fat content (in mg) of males and females captured flying within the shed was the highest (ANOVA (males) F = 8.029, d.f. = 3, P < 0.0001), (ANOVA (females): F = 3.026, d.f. = 3, P < 0.033). However, post hoc Tukey HSD comparisons indicated that fat content of males and females captured flying within shed, was not significantly higher than that of resident males (M.D = 0.025, P = 0.354) and females (M.D = 0.038, P = 0.361) (Figure 4.6). By contrast, the insects flying at some distance from the shed (20 m & 300 m) had a significantly lower fat content than insects caught flying within shed, but not different from that of the resident insects. For males: flying within shed vs. flying outside shed 20 m (M.D. = 0.585, P = 0.0006), and flying within shed vs. flying outside shed 300 m (M.D. = 0.069, P < 0.001). For females: flying within shed vs. flying outside shed 20 m (M.D. = 0.050, P = 0.017), and flying within shed vs. flying outside shed 300 m (M.D. = 0.051, P = 0.004).

0.2 Male Female

0.15 a AB

B ab B 0.1 b b

0.05

Mean fat contentMean (mg)

0 Flying within shed Flying outside shed Flying outside shed Resident (20m) (300m)

Treatment Figure 4.6 Mean fat content (mg) of beetles (±2SE) across four treatments. Males (grey bars) and females (white bars) of Tribolium castaneum were captured flying within the shed (n = 28 for each sex), flying outside (20 m) from the shed (n = 28♀♀, n = 22♂♂), flying 300 m from the shed (n=23♀♀, n = 20♂♂), and collected from the cotton pile within the shed as residents (n = 26♀♀, n = 25♂♂). Significant relationships are represented by different letters above each bar. Lower case letters denote males and uppercase letter relate to females, as comparison between sexes was not carried out.

65 4.3.5. Lipofuscin

The mean lipofuscin intensities (arbitrary units, λexcitation=340 nm, λemission=410 nm) measured in flying females and males were significantly higher than those in resident beetles, but with the exception of females caught flying at ~20 m distance (two-way ANOVA: F = 16.48, d.f. = 2, P < 0.0001) (Figure 4.7). Multiple comparisons by post hoc Tukey HSD indicate that resident males had significantly lower lipofuscin intensities than males captured flying within the shed (M.D. = 1251, P = 0.027) and those caught flying 20 m from the shed (M.D. = 2226, P<0.0001). Since resident females had no detectable levels of lipofuscin they were eliminated from the statistical comparisons. However, the lipofuscin intensity of females caught flying 20 m from the shed had significantly lower intensities than those flying within the shed (M.D. = 2201, P = 0.0001).

3500 a a Male 3000 1 Female a 2 2500

2000 intensity 8

1500 Mean

1000 b b 500 16 19 0 Flying within shed Flying outside shed (20m) Resident Treatment

Figure 4.7 Mean lipofuscin intensities (arbitrary units, ±2SE) of Tribolium castaneum males (grey bars) and females (white bars) from three treatments. Beetles were collected from the pile of cotton seeds as resident (n = 20 ♀♀, n = 18 ♂♂) and captured flying within the shed (n = 20 for each sex) and 20 m from the cotton shed (n = 20 for each sex). The number above each bar represents the number of individuals that had negligible fluorescence intensities (i.e., between 0-10 units). Significantly different values are indicated by different letters above bars.

66 4.3.6. Polyandry

Almost all females (94%) had mated multiple times and the number of males siring the offspring from the 2nd interval of oviposition ranged from one to four (Figure 4.8). None of the loci showed any presence of null alleles or any departure from Hardy-Weinberg equilibrium (Table 4.3). The observed and expected heterozygosity, as well as the allelic frequency was similar for both flying and resident beetles. Therefore, all treatments were considered as one population in the analyses and presented in Table 4.3 as such. Three out of the four loci showed relatively high number of alleles per locus while the average allelic frequency across all treatments was 7 alleles/locus (Table

4.3). The genetic diversity across all treatments was similar (flying within shed: average HE =

0.589, flying outside shed (20 m): HE = 0.608, flying outside shed (300 m): HE = 0.516, resident: HE = 0.546). The power to detect multiple paternity suggested by GERUDSIM 2.0 was as high as 90% in all treatments for two male contributors. This declined as the number of male contributors increased (3 males: 54.5%, 4 males: 7.5%).

Table 4.4 Characteristics of four microsatellite markers used to infer paternity in female Tribolium castaneum beetles collected from a cotton shed (flyers and residents combined as one population).

Locus NA HO HE PEXC Tc-2.7 8 0.681 0.689 0.501

Tc-7.1 8 0.681 0.744 0.571

Tc-7.4 4 0.264 0.242 0.129

Tc-8.8 8 0.764 0.629 0.355

Average 7 0.598 0.576 0.879

NA = number of alleles. HO = observed heterozygosity. HE = expected heterozygosity.

PEXC = exclusion probability (one parent known).

a) Flying within shed b) Flying outside shed (20m)

c) Flying outside shed (300m) d) Resident

Number ofNumber progeny

Estimated number of fathers contributing to the progeny Figure 4.8 Total number of progeny produced by individual female Tribolium castaneum until day 105 (after capture) plotted against the estimated number of fathers contributing to the progeny they produced between days 16 and 30 (n = 18 mothers and n = 15 progeny per mother in each treatment). The female beetles were captured in flight near a large population in a cotton seed storage or sampled from seeds: a) flying within shed, b) flying outside shed (20 m), c) flying outside shed (300 m), d) resident. The numbers of males were estimated by using four microsatellite markers.

68 4.4. Discussion

The main conclusions from the field study of T. castaneum are as follows: 1) flight propensity in dispersing beetles is not affected by weight and size of the beetles, 2) the overall fecundity and survival rate of beetles captured flying at different distances from the cotton shed and those sampled as residents were not significantly different, but offspring production declined significantly with time after day 30, 3) the fat content of males and females flying within the shed was significantly higher than that in those captured flying at 20 m and 300 m from the shed, 4) lipofuscin intensities of flying males and females (except for females captured flying at 20 m) were higher than the resident males and females, and 5) the estimated number of fathers siring the 16-30 day interval progeny ranged from one to four.

Generally the flight propensity of these beetles was not exclusively related to any particular variable, and was clearly unaffected by some including size and weight (Figure 4.5 & Table 4.2). This finding is consistent with the results from the laboratory experiment (Figure 2.5 & Table 2.1). The overall fecundity observed in these females over a period of more than one year shows their ability to reproduce at a relatively consistent (but diminishing) rate until they perish, and this must be a significant factor in terms of this species’ ability to spread across landscapes (Figures 4.3 & 4.4). Fecundity at 16-30 days after collection of females from the field was higher than their fecundity at 0-15 days after capture (Table 4.1). This increase in fecundity at 16-30 days after capture was also evident in the resident females and can be attributed to the uncertainty of knowing whether the resident females would have initiated flight in the future. The difference in fecundity soon after collection (0-15 days) and the subsequent interval (16-30 days) could, alternatively, have occurred because of the effects of their treatment in the laboratory, as female T. castaneum beetles are known to adjust their oviposition in response to changes in the quantity and quality of food available to them (Campbell & Runnion, 2002). By contrast, only the young female flyers from the laboratory test showed a significant increase in fecundity at day 16-30, and which was not observed in the young resident females (Figure 2.7).

Males were introduced to all treatments in the 106-120. The mean fecundity of females after introduction of males (from day 106) until death was not significantly different to that of females without access to males, across all treatments. However, the introduction of males affected the survival of females in one particular treatment, i.e., females caught flying 300 m from the cotton seed shed. When males were introduced to these females the survival rate reduced significantly relative to females from the same treatment (i.e., females flying at 300 m) that were not exposed to

69 males (Figure 4.2). This indicates that females mate readily irrespective of their age and that they might be stimulated to lay all their eggs before death. The phenomenon of reduced survival after exposure to one or more males has been observed in Sitophilus oryzae (Campbell, 2005).

The fat content of beetles captured flying within the shed was significantly higher that those flying at 20 m and 300 m, in both males and females (Figure 4.6). The decline of fat content with distance flown could be because of the energy spent in flight but it seems unlikely to have occurred so soon after taking flight (assuming that the beetles caught had only just left the stored cotton seeds). By contrast the fat content of young flying beetles in the laboratory was higher than that in the field- collected beetles (Figure 2.8). Since the laboratory beetles were reared on flour, the fat content could be more related to the different food resources on which the beetles had developed. Lipid content of field-collected Rhyzopertha dominica was found to be lower than that in laboratory strains of R. dominica (Perez-Mendoza et al., 1999).

Lipofuscin intensities of flying males and females were higher than those of resident beetles, with the exception of females caught flying 20 m from the shed (Figure 4.7). This trend of higher lipofuscin intensity in flying beetles was similar to the lipofuscin intensities detected in young flying beetles from the laboratory test (Figure 2.9). There were no peaks of lipofuscin in the resident females collected from the field, as the intensities were undetectable. Since lipofuscin was not found to have any correlation with age of beetles, as discussed in Chapter 2, in this study I have tested whether flight or dispersal distances are related to lipofuscin levels. The results obtained here again suggest that lipofuscin levels probably increase in response to metabolic stress and might be dependent on the physiological condition of an individual beetle (Fleming, 1992). Flying beetles may have higher metabolism, as has been shown in butterflies (Vande Velde & Van Dyck, 2013; Niitepõld & Boggs, 2015), house flies (Yan & Sohal, 2000), moths (Bartholomew & Casey, 1978) and birds (Nudds & Bryant, 2000). The lipofuscin results do, however require further validation by further characterization (including microscopic evidence), as conclusions on lipofuscin accumulation based on peaks obtained by spectrophotometer present only indirect evidence at best and can be unreliable (Brunk & Terman, 2002).

The polyandry and genetic diversity observed in these beetles was similar to that recorded in the T. castaneum beetles collected from the field and those obtained from newly established cultures of field beetles mentioned in Chapter 3. The numbers of males siring the brood from the 2nd interval of oviposition were the same, i.e. one to four (Figures 4.8 & 3.3). This finding has implications for understanding the development and spread of phosphine resistance and its management

70 geographically. Because females are capable of carrying sperm from several males, dispersal to new resources increases the chances of spreading resistance (Qazi et al., 1996; Lewis & Juthiewicz, 1998), as one mating with a resistant individual may be enough to introduce resistant progeny in a newly invaded site.

Overall, the predictions from the laboratory and field tests reported in Chapters 2 & 3 held true for the field beetles collected from the cotton seed shed. The general conclusions that can be drawn from this study are as follows: 1) no differences were evident in the weight and body size of the flying and resident beetles, 2) fecundity and survival of females in all treatments were similar, indicating that a higher proportion of young females flew or that the whole population was fairly young, 3) fat content was higher in beetles caught flying within the shed than those flying 20 m and 300 m from the shed, and this may be a result of the energetic costs involved in flight, although other explanations would have to be ruled out, 4) lipofuscin levels were higher in flying beetles as opposed to resident beetles, 5) females were highly polyandrous as the number of males contributing to the progeny from the 2nd oviposition interval of the captured females ranged from one to four. Thus, it can be said that flight dispersal in these insects is not affected by a single variable but is rather driven by a combination of factors some of which have been characterized in this study.

CHAPTER 5

Nanoparticles as external markers for mark-recapture studies on Tribolium castaneum

72 5.1. Introduction

Flight is central to the ecology of most insect species, but is extremely difficult to investigate (Bullock et al., 2002). Numerous tracking techniques have been tested in efforts to help quantify the direction and distance of insect flight. They include tagging with dyes and fluorescent dust, marking by mutilation, radioactive isotope marking, genetic marking and protein marking (Hagler & Jackson, 2001). The small size of insects and their often delicate structure pose a challenge to these traditional marking techniques, for example insects like thrips, grain beetles, whitefly and fruit fly that are not more than a few millimeters in length are particularly hard to mark. Also, it is difficult to mark enough individuals without damaging them, a serious challenge when recaptures in the field are low (Walker & Wineriter, 1981; Urquhart, 1987; Showers, 1997; Mahroof et al., 2010). For a marker to be effective it should be readily detectable, non-toxic, easily applicable on a large number of individuals simultaneously, and should have minimal or no impact on their behaviour (Southwood & Henderson, 2000). A clear need for an easily applicable marker for tiny insects exists, particularly if it can mark large numbers of individuals quickly and non intrusively.

Tribolium castaneum Herbst (Coleoptera: Tenebrionidae) is a serious cosmopolitan pest of stored products (Sokoloff, 1972). Understanding its movement propensity and capacity is crucial for managing food loss, because the development of resistance against the fumigant phosphine in these beetles has made it harder to manage infestations (Jagadeesan et al., 2012; Jagadeesan & Nayak, in press). The active dispersal of these beetles by flight also increases the chances of resistance spreading across wider areas (Ridley et al., 2011). Effective mark and recapture studies would therefore help to track their movement patterns and determine their associated behaviour.

The cuticle of adult T. castaneum darkens 2-3 days after eclosion and stays the same colour throughout the unusually long adult life span of these insects, which ranges from six months up to two years. This is extremely long relative to the very small size of these beetles i.e., 1-2 mm (Sokoloff, 1974). The cuticle is dry and waxy, and this complicates external marking (Sokoloff, 1972). Fluorescent dust has been used to mark small insects including stored grain beetles but it tends to fade with time, and transfer occurs with any contact, even to other individuals in a trap, thus yielding false positives in a study (Miller, 1993; Hagler & Jackson, 2001; Campbell et al., 2002; Mahroof et al., 2010).

Recently developed semiconductor quantum dots (QDs) have shown great potential for in vitro and in vivo labeling because of their unique optical properties (Gao, 2004; Nabiev, 2007; Geys, 2009;

73 Ekvall, 2013). QDs emit a strong fluorescent signal, have relatively high photo stability and high resistance to photo bleaching compared to traditional fluorescent dyes (Zhu et al., 2012). Moreover, QDs have broad excitation spectra and narrow emissions, allowing the fluorescent properties to be “tuned” by controlling their composition and size (Chan et al., 2002). These fluorescent characteristics of QDs make them easily detectable under a handheld UV source, a fluorescence microscope or confocal microscope (Zhu et al., 2012). Amino acid coated carboxyl quantum dots have been used successfully to track the swimming trajectory of Daphnia magna Straus (Cladocera: Daphniidae) in water currents (Ekvall et al., 2013). The QDs were also shown to have no negative effects on the behavioural responses of two types of aquatic insect larvae (mayfly and glass worm), along with D. magna.

In this chapter cadmium-tellurium (CdTe/CdS) QDs were employed to mark T. castaneum externally (Zhu et al., 2012). These QDs have shown positive results in cellular marking and in vivo marking of insect larvae (unpublished data). The aim was to assess the utility of CdTe/CdS QDs in marking T. castaneum, their stability when the beetles are in culture and flying in nature, and their toxicity to the insects. I also investigated the possibility of marking large numbers of insects quickly with QDs, and whether QDs remain on the insects for long enough to be useful in the field as markers. The ultimate aim is to develop a marker that will enhance the collection of data, by mark and recapture techniques, on the flight of T. castaneum. Two other small insects, Bactrocera tryoni Froggatt (Diptera: Tephritidae) and Plutella xylostella Linnaeus (Lepidoptera: Plutellidae), were also marked with QDs to test the generality of the ease of marking insects.

5.2. Materials and Methods

5.2.1. Synthesis of hydrophilic CdTe/CdS QDs

Hydrophilic CdTe/CdS QDs, stabilized by mercaptosuccinic acid (MSA), were synthesized by co- precipitation following hydrothermal treatment according to Zhu et al. (2012). Cadmium chloride solution (CdCl2, 0.04 M, 4 mL) was diluted to 50 mL and then tri-sodium citrate dihydrate (400 mg), Na2TeO3 (0.01 M, 4 mL), mercaptosuccinic acid (MSA, 100 mg) and sodium borohydride

(NaBH4, 50 mg) were added under vigorous stirring. When the solution turned green, it was loaded into a 35 ml Teflon-lined stainless steel autoclave. The autoclave reactor was heated to 1600C for 30 minutes in an oven, and then cooled to room temperature in a water bath. An equal volume of ethanol was added to the CdTe/CdS QDs aqueous solution, which led to precipitation of CdTe/CdS QDs that were collected by centrifugation (5000 rpm, 5 min). After decanting the supernatant, the

74 CdTe/CdS QDs were washed with ethanol three times. The CdTe/CdS QDs were finally dispersed in Milli-Q water and stored at 40C in the dark to avoid photo bleaching. The fluorescence spectra of the resulting QDs were recorded with a SpectraMax M5 Multi-mode microplate reader (Bio Tek Instruments, Vermont, USA) in the range 200–800 nm. The size of CdTe/CdS QDs was detected by transmission electron microscopy (TEM) (JEOL 1010 electron microscope, JEOL Ltd, Tokyo, Japan) operating at an acceleration voltage of 100 kV.

5.2.2. Marking insects with CdTe/CdS QDs

Marking with CdTe/CdS QDs was carried out by placing adult T. castaneum, B. tryoni and P. xylostella in three different Petri dishes (6 x 1.5 cm) with QDs (n = 10 insects per Petri dish). Bactrocera tryoni and P. xylostella were marked inside a cage because the insects could easily escape otherwise. They were left to float (T. castaneum) or walk (B. tryoni and P. xylostella) on the QD solution for 10 minutes and then placed in three new empty Petri dishes to dry for 1–2 hours. To assess the effectiveness with which the insects were marked, after the insects were fully dried (all insects alive) they were transferred individually to 30 mL empty plastic cups. The cups were placed on an ice block to immobilize insects after which they were visualized under an Olympus BX-61 light microscope (Olympus, Tokyo, Japan) fitted with fluorescence optics and a DP70 camera and Tecan i-control infinite 200Pro (Tecan, Männedorf, Switzerland) fluorescence spectrophotometer.

5.2.3. Test for toxicity

Tribolium castaneum adults are highly tolerant to starvation (Daglish, 2006), therefore marking beetles and placing them in a Petri dish without food was indicative of the effect of CdTe/CdS QDs on individual survival. Adult T. castaneum (n = 20, number of replicates = 3) were placed on 2 mL QD solution in a Petri dish, as described in section 5.2.2, for 24 h (to study the effects of increased exposure to QDs on the survival of beetles) at 12:12 photoperiod at 260C and 60% RH. The marked beetles were then placed in an empty Petri dish for 6 days and their survival was recorded every 24 hours. The control group for this assay contained unmarked beetles that had been placed on 2 mL of distilled water (the solvent for QDs) for 24 h (n = 20, number of replicates = 3) and otherwise treated in the same way as the test beetles.

75 5.2.4. Test for stability

Adult T. castaneum were marked as described in section 5.2.2 (n = 20, 2 mL QD solution) and then held in a Petri dish for 48 h at 26˚C and 60% RH. The control group contained 20 unmarked beetles, placed in 2 mL distilled water but were otherwise treated in the same way as the test insects. The control and marked beetles were placed separately into two empty Petri dishes. Individual beetles from the control and treatment Petri dishes were selected randomly and visualized 48 h post treatment under the Olympus BX61 light microscope (with fluorescence filters

λexcitation = 420–490 nm) until all 40 had been examined. All beetles, both marked and control were individually washed 48 h after marking with 50 μL distilled water in an Eppendorf® tube that was mixed on a vortex machine for 10 s. The liquid wash- off (50 μL) from each sample was analysed under the Tecan i-control fluorescence spectrophotometer (λexcitation = 275-360 nm, λemission = 500- 560 nm). The above steps were repeated for another set of 20 marked beetles 6 days after treatment.

5.2.5. Persistence of QDs on beetles in wheat

Adult beetles were marked with fresh CdTe/CdS QDs as described in section 5.2.2 (n = 40). After drying the QDs, the marked beetles were kept in Petri dishes with kibbled wheat grain (n = 30). The remaining 10 marked beetles were held individually in small plastic cups (30 mL) with kibbled wheat, to determine whether the presence of other individuals affects the persistence of the marker. The beetles were visualized under the microscope (same as above) and then washed with distilled water, which was then analysed under the fluorescence spectrophotometer (see section 5.2.4). The control groups (n = 40) consisted of unmarked beetles treated in the same way as the marked ones, but exposed to distilled water rather than CdTe/CdS QD solution.

5.2.6. Flight propensity in the laboratory

A flight bioassay of T. castaneum beetles was conducted to evaluate the effect of CdTe/CdS QDs on the flight propensity of the insects. The assay contained two controls and one treatment (n = 10 beetles for each). Treatment beetles were marked with 1 mL of CdTe/CdS QDs (section 5.2.2). Control B beetles were placed in 1 mL of distilled water and Control A comprised of untreated beetles, with no liquid medium added. After 10 minutes the control and marked beetles were moved to three new Petri dishes separately at 25˚C, 60% RH for 8–10 hours before testing.

76 A perspex push–pull airflow wind tunnel (120 × 60 × 50 cm length x width x height) at 260C and 60 % RH was used as the arena for testing flight propensity of these beetles. A Testo 405-V1 anemometer (Instrument Choice, Adelaide, Australia) was used to measure the wind speed at the center of the wind tunnel and held at a height of 5 cm from the base. The wind speed was kept constant at 1 m/s. Each beetle was tested individually by placing it on a thick cloth cone placed in a Petri dish at the center of the wind tunnel. Two pheromone Bullet LuresTM (Insects Limited Inc., Westfield, Indiana, USA), which contain the aggregation pheromone for T. castaneum (4,8-dimethyl decanal), were placed 50 cm upwind from the center (release point). Pheromone lures have been shown to attract beetles in a similar bioassay (Malekpour et al., 2016).

The test was started three hours before dusk because T. castaneum is mostly crepuscular (Howe, 1952; Jones, 1967; Rafter et al., 2015). Beetles in each treatment were assigned at random for flight testing and were placed on the cloth cone and observed for 10 minutes or until flight occurred (Malekpour et al., 2016).

5.2.7. Mark recapture in the field

A large number of T. castaneum adults was reared in 5 L buckets (n = 20) on flour, wheat grain and yeast. The culture medium (wheat and flour) was changed every three weeks. Beetles were separated from the culture medium by first removing the whole-wheat grain by sieving. The flour and beetle mixture was then placed in a jar with a paper ladder (piece of paper placed into the flour and folded over the lip of the jar). The jar was then placed in a bucket to trap the beetles that climbed the ladder and fell from the jar. This apparatus was covered and the beetles were allowed to accumulate in the bucket for 8–10 hours. The number of beetles was estimated by weighing three batches of 100 beetles and averaging their weight. This was extrapolated to the weight of all the beetles that entered the bucket. The beetles were marked by placing 2000-3000 beetles in batches of ~500 on the CdTe/CdS QDs (50 mL) in a glass Petri dish (20 x 2 cm). The beetles were left floating on the marker for 10 minutes and then placed in a plastic jar to dry.

The field experiment was performed at The University of Queensland’s Pinjarra Hills facility in Brisbane, Australia (27° 31' 46.78" S and 152° 55' 9.57" E), an area that does not host high densities of T. castaneum occurrence. The mean temperature, air pressure and wind speed at the site, from just before release of marked beetles until dusk (1430–1930 h for 3 days), were 33.0±20C, 1.0±2 bar and 1.0±2 m/s, respectively as measured by HOBO data loggers (Onset Computer Corporation,

77 Bourne, Massachusetts, USA). A galvanized release bin (16 L) was attached to a pole ~1.5 m above the ground. It had 12 holes (~4 cm diameter) and two circular cardboard pieces placed at two different heights within the bin with a 20 cm long cardboard cylinder to scatter the beetles and help them disperse from the release point. Around the central release bin, 21 Lindgren funnel traps (Contech Inc., Victoria, Canada) were arranged radially at equidistant intervals from each other and 20 m from the release point. Traps were suspended about 0.5 m above the ground. Each trap was baited with a single aggregation pheromone lure (see section 5.2.6). A funnel and plastic container were attached at the bottom of each trap to hold any trapped insects.

Freshly marked T. castaneum adults (~2400) were placed in the release bin on each of 3 consecutive days at 1500 h (AEST). Beetles were not removed from the release bin. Traps were monitored each day following beetle release and any captured beetles were transferred into individual cups and taken back to the laboratory. Captured beetles were then washed individually with distilled water, which was then analyzed using the Tecan i–control spectrophotometer to test for fluorescence.

5.3. Results

5.3.1. Distribution of QD markers on insects

Strong green fluorescence was clearly visible on the cuticle of T. castaneum (Figure 5.1), showing that the QDs do adhere to these insects mainly on the intersegmental spaces between the head, thorax and abdomen, at the base of the legs (Figures 5.1b & 5.1c), and also on the elytron of callow individuals (Figures 5.1d & 5.1f). Light green auto-fluorescence was observed from the eyes and mouthparts and the softer parts of the control beetles but the signal was comparably weak (Figures 5.1b & 5.1e). The cuticle otherwise showed no auto-fluorescence and the green of the QDs contrasted well against the dark red cuticle of the beetle (Figure 5.1).

Fluorescence microscopy images of marked B. tryoni and P. xylostella showed contrasting results as the marker adhered well to the hairy parts of B. tryoni (Figure 5.2) but there were no detectable traces of QDs on the moth P. xylostella.

78 a b

c d

e f

79 a b

c d

e f

80 5.3.2. Test for toxicity

The mortality results demonstrated that hydrophilic green CdTe/CdS QD had no obvious side effects on the beetles’ health, as marking T. castaneum beetles was not detrimental to them and no statistical difference was detectable in the mortality of the unmarked and marked beetles (Log–rank (Mantel-Cox) test 흌2 = 1.733, d.f. = 1, P = 0.188). Percentage mortality of control beetles was 8.33% (n = 60) while that of marked beetles was 16.67% (n = 60) at the end of six days.

5.3.3. Test for stability

The CdTe/CdS QD marker was stable on the exoskeleton of the beetles for at least 48 hours as the wash-off samples obtained from the beetles kept without wheat showed significant fluorescence peaks at 500 nm (ANOVA: F2, 57 = 15.32, P < 0.0001). The fluorescence intensity of the wash off obtained from the second wash and that from six days post treatment were significantly lower than that obtained form the first wash of the marked beetles 48 hours post treatment (ANOVA: F2, 57 = 15.38, P < 0.0001) (Figure 5.3). The microscopy results are listed in section 5.3.1 (see Figures 5.1a, 5.1b and 5.1c).

1300 1200 Without wheat 1100 On wheat 1000

Control 900 800 700 600 500

Mean intensityMean 400 300 200 100 0 Treatment

81 5.3.4. Persistence of QDs on beetles in wheat

Marked beetles held on wheat showed no fluorescence as no significant difference could be seen between the fluorescence intensity of the control and marked group on wheat (ANOVA: F2, 57 = 15.32, P < 0.0001, Tukey HSD: on wheat vs. control: MD = 11.89, P = 0.99) indicating that the marker is lost when the beetles are held on wheat grain (Figure 5.3). By contrast, marked beetles not exposed to wheat were still fluorescent (Tukey HSD: on wheat vs. without wheat: MD = 959.2, P < 0.0001).

5.3.5. Flight propensity of beetles in the laboratory

No significant difference was detected in the number of flying beetles between the marked and the unmarked control groups. A total of four beetles flew from the control group (total n = 10) and three flew from the control group B (total n = 10), with three also flying from the test group (total n = 10 marked beetles). The flights were mainly short flights, with the beetles landing within 10 cm of take-off location. Statistical tests are not permissible here because of low sample size and also because the aim of this study was to just test whether QDs affect flight.

5.3.6. Mark recapture test in the field

A total of 72 T. castaneum adults was caught (out of nearly 7000 that were released) in the funnel traps across three days. The wash-off obtained from these beetles showed very low fluorescence intensity peaks as compared to the laboratory samples (Figure 5.4). Most of the field-caught beetles showed negligible fluorescence, or no fluorescence under the microscope.

82 1200 Laboratory 1100 n = 20 Field 1000 Control 900 800 700 600 500

Mean intensityMean 400 300 200 100 n = 72 n = 20 0 Treatment

5.4. Discussion

Mark, release and recapture is a useful tool in obtaining information on an organism’s behaviour and movement patterns in nature, but it presents difficulties, especially with very small insects. Also, recaptures in the field tend to be low relative to the numbers released (Campbell et al., 2002; Bullock et al., 2002). Increasing the rate of recaptures is crucial, one way to do this is to release greater numbers of marked individuals, however, this requires a quick and reliable method for marking large numbers of individuals quickly.

This study demonstrated the following characteristics of the CdTe/CdS QDs tested. 1) Large numbers of T. castaneum can be easily and effectively marked with hydrophilic CdTe/CdS quantum dots, and that the marker is stable in the laboratory for at least 48 hours when beetles were kept without food. However, marked beetles kept on wheat did not retain the marker, presumably because of the abrasive effects of wheat on the cuticle of T. castaneum. 2) The QDs had no effect on the flight behaviour of marked T. castaneum in the laboratory, as tested in the wind tunnel. 3) CdTe/CdS QDs were not toxic to T. castaneum, so they provide a safe biomarker. 4) The field test indicated that that these quantum dots can be potentially used in field-based mark and recapture

83 studies, as beetles were successfully trapped. However, the marker presumably quenched because of the possible effects of benzoquinones released by beetles when stressed and further development of these QDs is required, as specified below.

Previous mark-recapture studies on T. castaneum and other grain beetles have been performed by using Day-Glo fluorescent powder within self-marking stations in a grain mill (Campbell et al., 2002). The authors reported that T. castaneum could not be marked evenly by the fluorescent powder as compared to other beetles and the recapture numbers of T. castaneum were low (Campbell et al., 2002). Another flour mill study, by Semeao et al. (2013), made use of the same fluorescent powder and a special coating on the traps (tangle trap sticky coating) to avoid transfer of the powder by beetles making contact with one another. About 4% of the marked beetles were recaptured. These were analysed under a UV lamp and a dissecting microscope was used for detection of the fluorescent powder (Semeao et al., 2013). These studies contributed to the understanding of how the beetles moved spatially within the mill and what conditions affected their movement. However, the cumbersome nature of the work required in analysing the recaptured individuals, plus the disadvantage of the powder in its getting transferred across beetles requires additions to the experimental setup. Therefore, a marker that can be easily detected from a large number of marked individuals at the same time is an advantage in simplifying mark-recapture studies and should help to save time and increase the amount of data obtained relative to effort in marking (by using automated machines to detect fluorescence from multiple samples, for e.g. spectrophotometer).

Wide ranges of studies have demonstrated the success of QDs as biomarkers for medical imaging and diagnostics (Gao et al., 2004; Biju et al., 2008; Smith et al., 2008), but very few studies exist on the utility or practicality of nanoparticles for use in mark-recapture studies of small organisms. Daphnia magna was marked externally with streptavidin functionalized CdSe/ZnS QDs (Lard et al., 2010). The authors successfully tracked D. magna in 2D for long periods without photo bleaching of the QDs. The study was, however, limited by the ability to detect QDs for periods longer than 2 days because D. magna frequently sheds its carapace. A successful demonstration of 3D tracking of D. magna has been made by using poly-L-lysine coated QDs for in vivo marking (Ekvall et al., 2013). The QDs had no negative effects on the organisms’ behaviour and the individuals could be easily detected under the camera in an aquarium.

The CdTe/CdS quantum dots tested on T. castaneum provide a promising approach to developing a practical marker for small insects because they did not affect the behaviour of the insects negatively

84 and were non-toxic, small, light, easily applicable, cost effective and photo stable. The QDs were stable and non-toxic to beetles under laboratory conditions when placed on beetles without food (Figures 5.1 & 5.3). This method will still be useful, in laboratory experiments not involving direct contact of the beetles with wheat or flour, for example in laboratory flight initiation tests, studying flight response to pheromones (olfactometer tests), or preliminary tests on flight initiation in the field considering the unavailability of food as a variable. This limitation of use explains why further development in design is needed. The loss in fluorescence observed in the field samples can be attributed to two main causes: 1) the large number of beetles that were marked together may have led to the release of large amount of benzoquinones which may have quenched the QDs, and 2) the harsh field conditions may have triggered benzoquinone release, with these compounds subsequently quenching the QDs. Although UV rays may have contributed to the quenching, these insects are mainly crepuscular, and so would have had little exposure. This is, however, a problem for studies on diurnal insects and to resolve this problem QDs can be functionalized with chemical groups such as organic ligands that can enhance the photo stability of QDs (Wang et al., 2012). Confirmation of the presence of CdTe/CdS QDs in the beetles collected in the traps in this field test requires an assay that can detect the presence of cadmium and tellurium in their elemental states in the wash-off obtained from the field samples. One possible method involves inductively coupled plasma–mass spectrometry, and these samples have been stored for such an analysis later.

The flexibility presented by QDs in terms of their range of colours, size and possibility of surface modifications to suit the needs of a particular insect system is a prime advantage. Further developments should be based on the surface functionalization of QDs to achieve photo-stability and improved adherence to the substrate, in this case the cuticle of T. castaneum. This may lead to better results in the field mark-recapture studies and the functionalized QDs may prove to be better external markers as compared to the QDs used in this study.

CHAPTER 6 General Discussion – Research findings, conclusions and future directions

86 6.1. Introduction

The findings presented in this thesis are an expansion to the current knowledge of flight behaviour of Tribolium castaneum. Some physiological characters that may have an effect on the dispersal by flight in T. castaneum were investigated on beetles in the laboratory and field. These research findings, and the predictions that can be made from them, are listed in section 6.2, and section 6.3 describes the implications of these findings for interpreting the ecology of T. castaneum. This is followed by the utility and practicality of using nanoparticles as external markers to aid in the tracking of flying beetles (section 6.4). The discussion ends with suggestions for future tests on the flight behaviour of T. castaneum and what it will mean to the management of phosphine resistance in this species (section 6.5).

6.2. Research findings and predictions

6.2.1. Characteristics of emigrant beetles (laboratory)

The results from Chapter 2 confirmed that most of the flight in T. castaneum occurred during late afternoon [often referred to as crepuscular in the literature (Howe, 1952; Jones, 1967; Rafter et al., 2015)] (Figures 2.3 & 2.4). Age was related to flight initiation, as more young beetles took to flight than old beetles (Figure 2.2). Sex, and weight had no significant effect on the flight propensity of T. castaneum and the percentage of males and females flying over two experimental periods (2015 & 2016) were similar to one another (Figures 2.2, 2.4, 2.5). The survival of young females (flying and resident) was significantly higher when compared to old females, whether flying or resident, over the 90-day period for which they were held after being caught in flight or sampled from the culture medium (Figure 2.6). A mixed-age treatment was included in these tests, to determine whether the beetles caught in flight had characteristics of the young beetles, and whether the resident beetles (i.e. the ones that did not fly) resembled the old beetles. Flying females from this mixed-age test had significantly higher survival than old females (both flying and resident) but survival of the mixed-age resident females was not significantly different from that of old or young resident beetles. This indicates that most of the flying beetles could be young, as originally expected (Figure 2.6).

Fecundity recorded over the 90-day period showed no significant differences and was relatively consistent across the different age groups of beetles and between flyers and residents (Figure 2.7). They produced a relatively small number of offspring, at regular intervals, for a long time. However, fecundity increased significantly in young flying females in the 16-30 day period after

87 capture, as compared to the fecundity at 0-15 days after emigration (Figure 2.7). Old flying females did not show this pattern, and neither did any of the resident beetles. This trend suggests three possibilities: (i) young females might be dumping their eggs before emigration, (ii) females had recently mated or, (iii) reduced fecundity of young beetles at 0-15 days after emigration could be because the females that took to flight were ≤ 2 weeks old. Females younger than 14 days have reduced fecundity as compared to relatively old females (Perez-Mendoza et al., 2011b). Most females captured in flight had already mated, so they are either not flying so young, or they mate repeatedly in the first two weeks of their adult life.

Although the mixed-age treatment comprised beetles of roughly equal proportions of young and old insects, yet higher numbers of beetles were captured in flight in this treatment relative to the number flying in the young and old beetle treatment (Figure 2.2). Also, the survival curve of flying mixed-age beetles was similar to that of the young beetles, whereas the survival of resident mixed- age females was between that of the old and young females (Figure 2.6). The survival rate of females from the old age treatment (>2 months) declined significantly within 90 days of their capture.

Young beetles had more fat than old beetles of both sexes but this difference was statistically significant only when comparing young and old resident beetles against one another. There was no significant difference in fat content between young resident and young flying beetles (Figure 2.8). Body size was not different across beetles of the same sex (in any of the test categories) and, most notably, there were no differences between flying and resident beetles (Table 2.1).

The lipofuscin intensities showed no relation to beetle age or weight and were perhaps more dependent on the metabolic status of the individuals rather than age (Figure 2.9), as the lipofuscin intensities associated with young flying beetles were higher than those of resident beetles, whether old or young. The highest lipofuscin intensities were those of young females (flying), which were significantly different from those of beetles in all treatments, and whether this included males or females. The significant difference between young male and female flyers could be related to the energetic costs of the females’ flight as well as their reproductive status, which may all have added up to an increase in metabolism. Energetic costs have been shown, for example, to increase during intersexual conflicts associated with reproduction in water striders (Watson et al., 1998).

Lipofuscin intensities reveal that the molecule is not related to the age of beetles, which is contrary to what has been reported in the literature for other organisms, including ants, flesh flies and blue

88 crabs (Ettershank et al., 1983; Ju et al., 1999; Robson & Crozier, 2009). Flying individuals of T. castaneum had higher lipofuscin intensities as compared to resident beetles. Lipofuscin accumulation is linked to lipid peroxidation and oxidative stress (Lippman, 1981). Studies on Lepidoptera and Diptera indicate that cellular aging is related to oxidative stress and high metabolic activity. For example, flying Musca domestica (Muscidae) had higher metabolic rates (which were positively correlated with lipofuscin levels) as compared to those that were confined in a small jar to reduce their flight activity (Sohal & Donato, 1979; Yan & Sohal, 2000). The amount of flight activity in Drosophila melanogaster (Drosophilidae) was positively correlated with oxidative stress and their cell membrane fatty acid composition was altered, which increased the chances of oxidative damage (Magwere et al., 2006), but lipofuscin levels were not measured. Studies on Speyeria mormonia (Nymphalidae) have also revealed that high flight activity is related to high resting metabolic state (Niitepõld & Boggs, 2015). Likewise there could be possible energetic costs relative to flight in T. castaneum, which may perhaps influence oxidative stress and ultimately lipofuscin accumulation.

The other possibility of a drastic difference in lipofuscin intensity could be because of the method by which lipofuscin is extracted. For example, the process of lipofuscin extraction has been questioned with particular reference to the blue crab study. The wild caught and tank raised blue crabs had a negative correlation of chronological age with lipofuscin indices (Crowley et al., 2014). The auto-fluorescence, which has been shown by spectroscopic measurement to result from the accumulation of lipofuscin and products of lipid peroxidation, could possibly result from the presence of proteins, vitamins and amino acid derivatives (Sheehy, 2008). In particular the observed auto-fluorescence in the extracts could have resulted from the light scattering due to 3D packing of proteins (Palmer, 2002). The absence of direct evidence that it is lipofuscin that is being measured means that the basing of aging solely on auto-fluorescence readings from extracts could be unreliable (Brunk & Terman, 2002). Therefore, there is a need of direct evidence on the relationship between auto-fluorescence from lipofuscin and age, which can be achieved by molecular tests and microscopy that can support the data from spectroscopic measurements of soluble lipofuscin.

6.2.2. Characteristics of emigrant beetles (field)

The results presented in Chapter 3 demonstrate that the field caught females were highly fecund and the number of matings had no effect on the fecundity (Figure 3.1). The survival of the field caught flying females was significantly higher than that of the field caught resident females (Table 3.1). Levels of polyandry recorded in field caught T. castaneum females and also in females from new

89 laboratory cultures was high. The numbers of males siring the offspring produced by females during their 2nd interval of oviposition (8-14 days after capture) were from one to four (Figure 3.3) and the reproductive contributions of each of the fathers in multiply sired broods varied greatly in both the field caught and new laboratory females. In both of these treatments the contributions from males deviated from the 50% null expectation of the Hardy-Weinberg equilibrium (Figure 3.4). Females from a long-established laboratory culture were also tested for polyandry but no conclusions could be drawn, as the microsatellite markers used were monomorphic. Findings from the comparisons drawn on fecundity and polyandry between females from the field and those from the new laboratory culture show that females are mostly mated before they emigrate. This indicates that mating occurs soon after eclosion, as the proportion of virgin females was only 6% while 91% of the females had mated several times.

These emigrating female T. castaneum beetles, due to their ability to store sperm in their spermathecae from multiple matings, carry a potential for progeny with a range of genetic diversity (Qazi et al., 1996; Lewis & Juthiewicz, 1998). Their dispersal provides a chance to increase the geographical spread of a range of genes, and so increases the chances of resistance genes being spread widely. However, the spread of resistant individuals themselves might not be as efficient as is the spread of susceptible beetles, as demonstrated in laboratory-based orientation studies performed on resistant and susceptible beetles (Malekpour et al., 2016). The resistant beetles do not fly as readily as susceptible ones do, and their orientation to resources is not as accurate (Malekpour et al., 2016).

6.2.3. Findings and comparisons of field with laboratory emigrant beetles

The findings from Chapters 2 and 3 were used to develop predictions about the characteristics of emigrating beetles for the field study that is reported in Chapter 4. Beetles were captured in flight at different distances from a large central beetle population in a cotton seed store, and some beetles were also collected from storage (and considered to be resident beetles). Laboratory measurements and tests were then repeated on the field-collected beetles to test whether the predictions held true in field populations.

The findings from Chapter 4 were quite similar to those from laboratory tests. The overall survival of females across 375 days was not significantly different across all treatments (Figure 4.1). Fecundity of females until day 105 was also consistent across treatments (Figure 4.3). The introduction, at day 106, of one male to each female in half of the surviving population of females

90 in each treatment did not affect the mean fecundity of females whether or not exposed to males (across all treatments) (Figure 4.4). But the introduction of male significantly affected the survival of females in one treatment (i.e., females caught at 300 m from the shed). In this treatment the survival of females that had access to males was significantly lower than those without males (Figure 4.2). This has been previously reported in Sitophilus oryzae where continuous exposure of female S. oryzae to one or more males decreased the survival rate significantly as opposed to females that had no access to males (Campbell, 2005). Presumably these females were, somehow, stimulated to deposit all their potential eggs and died soon after. This indicates that females of all ages mate readily and that mating has an effect on their reproductive physiology. The fecundity of females from all treatments at 16 days after emigration (but also including resident females) showed similar trends as the young females (flying) from the laboratory test, with an increase in fecundity over that achieved in the first 15 days (Figure 4.3 & Table 4.1). That resident females show the same pattern as emigrating ones suggests that females about to emigrate do not dump their eggs beforehand. Rather capture and the associated treatment seems to affect their reproductive physiology negatively.

Weight and body size were not related to the flight propensity of the beetles in all treatments (Figure 4.5 & Table 4.3). Also, the fat content of beetles across all treatments was similar and the males and females captured in flight just above the resource (2 m) carried significantly more fat than the long distance flyers (Figure 4.6). The lipofuscin intensities of flying beetles was generally higher than those of the resident beetles, and this reflects the situation with young flying beetles in the experiments reported in Chapter 2. However, this was not the case with female flyers caught at 20 m from the seed store, for these showed significantly lower lipofuscin intensities (Figure 4.7). The problems associated with this method have been discussed above, in section 6.2.1. Females from all treatments were multiply mated with the number of males siring the brood (from the 2nd oviposition interval, i.e., 16-30 days after capture) ranging from one to four (Figure 4.8).

6.2.4. General conclusions

Overall, the flight propensity of T. castaneum is statistically not related to the sex, weight or body size of these beetles, but declines with increasing age. The fecundity of females increased significantly in the interval 16-30 days after emigration, which again indicates that females are negatively affected by capture (because, again, the resident beetles showed the same pattern (Chapter 4)). The similarity in survival rates of females across all treatments and as found indicates that there was a higher proportion of young beetles flying in the field or perhaps the whole

91 population was fairly young. That the latter might be the case is suggested by the same pattern being observed in resident field-collected females, for these might have been about to take flight. Females mate multiple times before emigration and are polyandrous. Indeed 94% of the females used for polyandry analysis from all treatments had mated multiple times. With regard to age and flight in T. castaneum, very old beetles (i.e. 6 months or more, or 185-215 days in tests) did not initiate flight at all in the laboratory and flight declined with increasing age (Perez-Mendoza et al., 2011b). This was also reflected in the results reported in Chapter 2, which suggests that in this very long-lived species of beetles the age factor might start to affect flight only at a certain age, perhaps ≥3 months, and measures of fecundity at ≥5 months after eclosion. The fat content tests in the laboratory indicate that young beetles carry more fat than the old beetles and this is true for the short distance male flyers (2 m) observed in the field test. Since the overall fat content in the field beetles was lower than in the laboratory beetles it suggests that the fat content is dependent more on the food source of these insects. So a clear correlation between flight and age with fat content cannot be drawn. Likewise, the mean body weight and size of beetles observed in laboratory and field were more or less similar.

Overall, the factors investigated do not seem to have a drastic effect on flight, which implies that flight between resources is a regular feature in the life of these insects, and related to only few physiological and environmental factors. In general, it seems that it is mostly young insects that fly and that their flight activity is higher during late afternoon, and is further related to temperature (which must be above 250C) and wind (which must be low enough i.e., <3m/s) (Rafter et al., 2015).

6.3. Implications for interpreting the ecology of Tribolium castaneum

Dispersal has implications for gene flow generally and specifically for the geographical spread of phosphine resistance. Therefore, the influences of flight propensity in T. castaneum is important especially since these insects have, until recently been considered mostly not to fly (see Ridley et al., 2011). The effective management of phosphine resistance needs the knowledge of how these genes spread, as well as the geographical pattern of their spread. Since females are polyandrous and fecund most of their life, the ready availability of mates enhances the spread of resistance (Chapter 3). Their long adult lifespan, coupled with their flight abilities, also promotes gene flow spatially and temporally. Sperm mixing before migration could improve colonization of a new resource through the increased allelic diversity conferred by mating with multiple males. For example, if a female has mated with a resistant male and leaves the resource to exploit a fresh patch it would lead to the transfer of resistant genotypes spatially and the establishment of resistance if the new locality was one exposed to phosphine fumigation.

No differences in overall fecundity in the laboratory and field populations indicate that females are fecund throughout their lifespan. The fecundity of females, collected from the field, tended to decrease at about four months (106-120 days) but this seemed to be less related to age than the absence of mates. As soon as males were introduced their fecundity generally increased (but not always statistically significantly so) and the females continued to produce offspring until just before death, at which point fecundity was mostly negligible. This indicates that fecundity drops slowly but females are more or less fecund throughout almost all of their life span. This could explain the ability of these beetles to successfully infest resources in large numbers. The fact that the females survive and reproduce on cotton seed successfully with its associated micro fungi and actively disperse to new resources indicate the species’ abilities over a wide range of resources. The high reproductive output of T. castaneum, given time, and the ability of females to control the number of eggs laid depending on the quality and quantity of resources (Campbell & Runnion, 2003) explains how these beetles can survive and successfully reproduce on almost any favourable food resource, even cotton seed and its associated micro fungi (Ahmad et al., 2012ab). The evolutionary history of T. castaneum suggests that the beetles inhabited warm, humid environments and are generalist feeders. They have been known to scavenge detritus (Dawson, 1977) and are natural predators of various species of moths (Adeyemi, 1968). Besides, their resistance to starvation is another character that makes them robust pests (Daglish, 2006). These characters suggest that T. castaneum lived originally on rotting logs and tree bark before the advent of bulk grain storage (Dawson, 1977).

6.4. Nanoparticles as external markers

The green fluorescent cadmium tellurium (CdTe/CdS) nanoparticles (hydrophilic quantum dots or QDs) that were tested as external markers for mark-recapture studies of T. castaneum proved to be a safe biomarker which remained stable on beetles for 48 hours post marking, at least when beetles were kept without food (Figures 5.1-5.3). The QDs had no negative effects on flight and survival of T. castaneum. The ease of marking was demonstrated by the ability to mark large numbers of individuals simultaneously and in a short time. The mark-recapture of T. castaneum in the field showed that the marker was prone to quenching, probably as a result of stress chemicals (benzoquinones) released by the beetles when field temperatures reached 370C during the day (Figure 5.4). The absence of food, heat stress and the high density of beetles could well have triggered the release of benzoquinones and this needs to be tested further. Tribolium castaneum have been shown to release benzoquinones when stressed and as a defence mechanism to combat growth of microbes and fungi (Alexander & Barton, 1943; Loconti & Roth, 1953; Wirtz et al.,

93 1978; Fedina & Lewis, 2008; Lovett & Leger, 2016). To confirm that the low fluorescent intensity readings were because of quenching and not because the marker was absent from the beetles, further characterization of the wash-off obtained from the beetles recaptured from the field is needed. Future tests would involve elemental characterization of cadmium and tellurium in the wash-off.

The marking ability of QDs was also demonstrated on Bactrocera tryoni and Plutella xylostella with only B. tryoni able to take up the QDs, on its head and legs. This may indicate that the marking ability of QDs depends on the structure of the cuticle, but other application methods need to be tested on the moths. However, stability, toxicity and other such factors were not examined for both these insects. Future work on the use of QDs as external markers should focus on development of QDs with functional groups that promote the adherence of QDs on different cuticular structures. For example, a cuticle like that of diamondback moth is “dusty” and hydrophobic, so developing QDs with organic groups that help the particles bond with the cuticle as well as not reducing its fluorescence could be helpful. The smaller size of QDs and the ability to tune them to different colours makes them a desirable candidate for external as well as internal marking. This has already been demonstrated by studies on Daphnia magna (Daphniidae) where the zooplankton was successfully marked both externally and internally by functionalized QDs and could be tracked for 2D and 3D movement in large aquaria (Lard et al., 2010; Ekvall et al., 2013).

6.5. Future directions

Although the results presented in this thesis suggest it is relatively young beetles that fly, a marker that defines the age in these beetles accurately would be extremely useful in investigating the ecology of these insects in more detail. Research should be focused on markers that are true indicators of chronological age because the results presented here on lipofuscin show that this molecule is unlikely to be useful in this regard. Cuticular hydrocarbons could be a more reliable option of chronological age grading (Perez-Mendoza et al., 2004). Developing a reliable marker of age would have many advantages, as one would then be able to define patterns of movement in these beetles much more accurately. Such a marker would help gather much more direct evidence of the dispersal behaviour of different aged beetles exposed to different environmental conditions.

The ability to track resistance in the field would be a prime advantage for the management of these pests. Development of a practical external marker would help in tracking populations more precisely. This would enable marking of beetles and their subsequent successful recapture (by pheromone or grain traps), which would be the easiest way in which to track resistant insects

94 spatially and temporally. As the quantum dots presented in this study prove to be promising candidates for external marking it is now imperative to design and functionalize them in a way that would serve effective adherence to the beetles’ cuticle. Their colour tunability can be utilized and different populations of beetles could be marked and recaptured to understand dispersal patterns in the field. Using these markers to track resistance across landscapes would present an ideal arena for future developments in pest and resistance management.

95 References

Adeyemi, S.A.O. (1968). A laboratory study of competition between Tribolium castaneum (Hbst.) (Coleoptera, Tenebrionidae) and three moth species. Bulletin of Entomological Research. 58(1): 31-45.

Ahmad, F., Daglish, G.J., Ridley, A.W. & Walter, G.H. (2012a). Response of Tribolium castaneum Herbst (Coleoptera:Tenebrionidae) to olfactory cues from cotton seeds, the fungi associated with cotton seed, and cereals. Entomologia Experimentalis et Applicata. 145: 272–281.

Ahmad, F., Daglish, G.J., Ridley, A.W., Burrill, P.R. & Walter, G.H. (2013a). Short-range resource location by Tribolium castaneum Herbst (Coleoptera: Tenebrionidae) demonstrates a strong preference for fungi associated with cotton seed. Journal of Stored Products Research 52: 21-27.

Ahmad, F., Ridley, A., Daglish, G.J., Burrill, P.R. & Walter, G.H. (2013b). Response of Tribolium castaneum and Rhyzopertha dominica to various resources, near and far from grain storage. Journal of Applied Entomology 137: 773-781.

Ahmad, F., Walter, G.H. & Raghu, S. (2012b) Comparative performance of Tribolium castaneum (Herbst) (Coleoptera: Tenebrionidae) across populations, resource types and structural forms of those resources. Journal of Stored Products Research. 48: 73-80.

Alexander, P., & Barton, D.H.R. (1943). The excretion of ethylquinone by the flour beetle. Biochemical Journal. 37: 463-465.

Antony, C. & Jallon, J.M. (1982). The chemical basis for sex recognition in Drosophila melanogaster. Journal of Insect Physiology. 28: 873 – 880.

Arnaud, L., Haubruge, E. & Gage, M.J.G. (2001a) Morphology of Tribolium castaneum male genitalia and its possible role in sperm competition and cryptic female choice. Belgian Journal of Zoology. 131:111–115.

Arnaud, L., Gage, M.J.G. & Haubruge, E. (2001b) The dynamics of second and third-male fertilisation precedence in Tribolium castaneum. Entomologia Experimentalis et Applicata. 99:55–64.

Arnqvist, G. & Nilsson, T. (2000). The evolution of polyandry; multiple mating and female fitness in insects. Animal Behaviour. 60: 145-164.

Aslam, M., Hagstrum, D.W. & Dover, B.A. (1994) The effect of photoperiod on the flight activity and biology of Rhyzopertha dominica (Coleoptera: Bostrichidae). Journal of the Kansas Entomological Society. 67: 107-115.

Bailey, P.T., (2007). Pests of field crops and pastures: identification and control. CSIRO Publishing: Victoria, AU.

Baloch, U. K. (1999). Wheat: post-harvest operations. Food and Agriculture Organization of the United Nations: 10.

Barrer, P.M., Starick, N.T., Morton, R. & Jane Wright, E. (1993). Factors influencing initiation of flight by Rhyzopertha dominica (F.) (Coleoptera: Bostrichidae). Journal of Stored Products Research. 29: 1–5.

Bartholomew, G.A. & Casey, T.M. (1978). Oxygen-consumption of moths during rest, pre-flight warm-up, and flight in relation to body size and wing morphology. Journal of Experimental Biology. 76: 11–25.

Behalima, H., Chaudry, M.Q., Mills, K.A. & Price, N.R. (2004) Phosphine resistance in stored- product insects collected from various grain storage facilities in Morocco. Journal of Stored Products Research. 40: 241-249.

Bell, C.H., Hole, B.D. & Evans, P.H. (1977). The occurrence of resistance to phosphine in adult and egg stages of strains of Rhyzopertha dominica. Journal of Stored Products Research. 13: 91 – 94.

Bengston, M., Collins, P.J., Daglish, G.J., Hallman, V.L., Kopittke, R. & Pavic, P. (1999) Inheritance of phosphine resistance in Tribolium castaneum (Coleoptera: Tenebrionidae). Journal of Economic Entomology. 92: 17-20.

97 Benzi, G. (1990). Peroxidation, energy transduction and mitochondria during aging. John Libbey Eurotext, FR.

Berry, W. (2009). Bringing it to the Table: On Farming and Food. Counterpoint Press: Berkeley, CA.

Bigbee, W. & Herberman, R.B. (2003). Characteristics of the ideal tumor-marker. In: Kufe, D.W., et al. (Eds.). Holland-Frei Cancer Medicine (Sixth edition). BC Decker: Hamilton, ON.

Biju, V., Itoh, T., Anas, A., Sujith, A. & Ishikawa M. (2008). Semiconductor quantum dots and metal nanoparticles: syntheses, optical properties, and biological applications. Analytical and Bioanalytical Chemistry. 391: 2469–2495.

Blanco, C.A., Gould, F., Groot, A.T., Abel, C.A., Hernández, G., Perera, O.P. & Terán-Vargas, A.P. (2010). Offspring from sequential matings between Bacillus thuringiensis-resistant and Bacillus thuringiensis-susceptible Heliothis virescens moths (Lepidoptera: Nocuidae). Journal of Economic Entomology. 103: 861-868.

Bond, E. & Upitis, E. (1973). Response of three insects to sublethal doses of phosphine. Journal of Stored Products Research 8: 307-313.

Bond, E. J., Robinson, J. R. & Buckland, C. T. (1969). The toxic action of phosphine: absorption and symptoms of poisoning in insects. Journal of Stored Product Research 5: 289-298.

Bond, E.J. (1989). Manual of fumigation for insect control. Food and Agriculture Organization of the United Nations. Retrieved from http://www.fao.org/docrep/x5042e/x5042E00.htm#Contents.

Boon, K.S. & Ho, S.H. (1988) Factors influencing the post-fumigation reinfestation of Tribolium castaneum (Herbst) (Coleoptera: Tenebrionidae) in a rice warehouse. Journal of Stored Products Research 24: 87-90.

Bosly, H.A. & Kawanna, M.A. (2014). Fungi species and red flour beetle in stored wheat flour under Jazan region conditions. Toxicology and Industrial Health. 30(4): 304-310.

98 Bowler, D.E. & Benton, T.G. (2005). Causes and consequences of animal dispersal strategies: relating individual behaviour to spatial dynamics. Biological Reviews. 80: 205-225.

Bretman, A. & Tregenza, T. (2005). Measuring polyandry in wild populations: a case study using promiscuous crickets. Molecular Ecology. 14: 2169-2179.

Brunk U.T., & Terman, A., (2002). Lipofuscin: mechanisms of age related accumulation and influence on cell function. Free Radical Biology & Medicine. 33:611–619.

Bullen, K. (2007). Insect control in stored grain. Department of Primary Industries and Fisheries: Toowoomba, QLD. Retrieved from http://www.daff.qld.gov.au/__data/assets/pdf_file/0009/65817/Grain-Storage-Ins-Cont-Stored- Grain_.pdf

Bullock, J.M., Kenward, R.E., & Hails, R.S., (2002). Dispersal Ecology: 42nd Symposium of the British Ecological Society. Cambridge University Press, U.K. 24–49.

Campbell, J. & Arbogast, R.T. (2004). Stored-product insects in a flour mill: population dynamics and response to fumigation treatments. Entomologia Experimentalis et Applicata. 112: 217–225.

Campbell, J. & Runnion, C. (2003). Patch exploitation by female red flour beetles, Tribolium castaneum. Journal of Insect Science 3: 1-8.

Campbell, J. F., Toews, M. D., Arthur, F. H. & Arbogast, R.T. (2010a). Long-term monitoring of Tribolium castaneum in two flour mills: seasonal patterns and impact of fumigation. Journal of Economic Entomology 103: 991-1001.

Campbell, J. F., Toews, M. D., Arthur, F. H. & Arbogast, R.T. (2010b). Long-term monitoring of Tribolium castaneum in two flour mills: Rebound after fumigation. Journal of Economic Entomology 103: 991-1001.

Campbell, J.F. (2005). Fitness consequences of multiple mating on female Sitophilus oryzae L. (Coleoptera: Curculionidae). Behavior. 34(4). 833-843.

99 Campbell, J.F., Mullen, M.A. & Dowdy, A.K. (2002). Monitoring stored-product pests in food processing plants with pheromone trapping, contour mapping, and mark-recapture. Journal of Economic Entomology. 95(5): 1089-1101.

Chan, W.C., Maxwell, D.J., Gao, X.H., Bailey, R.E., Han, M. & Nie, S.M. (2002). Luminescent quantum dots for multiplexed biological detection and imaging. Current Opinion in Biotechnology. 13(1): 40-46.

Chapuis, M-P. & Estoup, A. (2007). Microsatellite null alleles and estimation of population differentiation. Molecular Biology and Evolution. 24: 621-631.

Clobert, J., Danchin, E., Dhondt, A.A. & Nichols, J.D. (Eds.) (2001). Dispersal. Oxford University Press. Oxford, UK.

Clobert, J., Ims, R. & Rousset, F. (2004). Causes, mechanisms and consequences of dispersal. Page 307–335. In: Hanski, I. &. Gaggiotti, O. E (Eds.) Ecology, genetics and evolution of metapopulations. Elsevier Academic, New York.

Cogburn, R.R. (1988). Detection, distribution and seasonal abundance of Sitotroga cerealella and Rhyzopertha dominica as indicated by pheromone-baited adhesive traps. Proceedings, XVIII International Congress of Entomology, 3 –9 July 1988. University of British Columbia, Vancouver, Canada.

Collins, P.J., Daglish, G.J., Bengston, M., Lambkin, T.M. & Pavic, H. (2002). Genetics of resistance to phosphine in Rhyzopertha dominica (Coleoptera: Bostrichidae). Journal of Economic Entomology. 95: 862-869.

Crombie, A.C. (1941). On oviposition, olfactory conditioning and host selection in Rhyzopertha dominica Fab. (Insecta, Coleoptera). Journal of Experimental Biology. 18: 62-79.

Crow, J.F. (1994). Advantages of sexual reproduction. Developmental Genetics. 15: 205-213.

Crowley, C.E., Gandy, R.L., Daly, K.L. & Van Vleet, E.S. (2014). Problems associated with a lipofuscin extraction method used to age blue crabs Callinectes sapidus cultured in Florida, USA. Aquatic Biology. 21(2): 85-92.

100

Daglish, G.J. (2005). Fecundity of wild Tribolium castaneum (Herbst) from southeast Queensland, Australia. Journal of Stored Products Research 41. 115-119.

Daglish, G.J. (2006). Survival and reproduction of Tribolium castaneum (Herbst), Rhyzopertha dominica (F.) and Sitophilus oryzae (L.) following periods of starvation. Journal of Stored Products Research. 42: 328–338.

Daglish, G.J. & Collins, P.J. (1999). Improving the relevance of assays for phosphine resistance. In: Jin, Z., Liang, Q., Liang, Y., Tan, X. & Guan, L. (Eds.). Proceedings of the seventh international working conference on stored-product protection, Beijing, China, 14–19th October, 1998. Sichuan Publishing House of Science and Technology: Chengdu, CN.

Dawson, P.S. (1977). Life history strategy and evolutionary history of Tribolium flour beetles. Evolution. 31(1): 226-229.

DeBrosse, S.D. & Kerr, D.S. (2012). Pyruvate dehydrogenase complex deficiencies. In: Lee-Jun C. Wong (Ed.). Mitochondrial disorders caused by nuclear genes. Springer: New York, NY.

Demont, M., Grazer, V.M., Michalczyk, L., Millard, A.L., Sbilordo, S.H., Emerson, B.C., Gage, M.J.G. & Martin, O.Y. (2014). Experimental removal of sexual selection reveals adaptations to polyandry in both sexes. Evolutionary Biology. 41: 62-70.

Demuth, J.P., Drury, D.W., Peters, M.L., Van Dyken, J.D., Priest, N.K. & Wade, M.J. (2007). Genome-wide survey of Tribolium castaneum microsatellites and description of 509 polymorphic markers. Molecular Ecology Notes. 7: 1189–1195.

Dingle, H. (2009). Chapter 167-Migration A2-Resh, V. H. In: Encyclopedia of Insects (Second Edition), Cardé, R.T. (Ed.). Academic Press: San Diego, USA: 628-633.

Dowdy, A.K. (1994). Flight initiation of lesser grain borer (Coleoptera: Bostrichidae) as influenced by temperature, humidity, and light. Journal of Economic Entomology. 87: 1714-1717.

101 Drury, D. W., Whitesell, M. E. & Wade, M. J. (2016). The effects of temperature, relative humidity, light, and resource quality on flight initiation in the red flour beetle, Tribolium castaneum. Entomologia Experimentalis et Applicata. 158(3): 269–274.

Drury, D.W., Siniard, A.L. & Wade, M.J. (2009). Genetic differentiation among wild populations of Tribolium castaneum estimated using microsatellite markers. Journal of Heredity. 100: 732-741.

Eberhard, W.G. (1996). Female control: Sexual Selection by Cryptic Female Choice. Princeton University Press. Princeton, New Jersey.

Edde, P.A., Phillips, T.W., Nansen, C. & Payton, M.E. (2006). Flight activity of the lesser grain borer, Rhyzopertha dominica F. (Coleoptera : Bostrichidae), in relation to weather. Environmental Entomology. 35: 616-624.

Ekvall, M.T., Bianco, G., Linse, S., Linke, H., Bäckman, J. & Hansson, L-A. (2013). Three- dimensional tracking of small aquatic organisms using fluorescent nanoparticles. PLOS ONE. 8(11): e78498. doi: 10.1371/journal.pone.0078498.

Emery, R. (2014). Insect pests of stored grain. Department of Agriculture and Food: Government of Western Australia. Retrieved from https://www.agric.wa.gov.au/pests/insect-pests-stored- grain?page=0%2C0.

Ettershank, G., Macdonnell, I. & Croft, R. (1983). The accumulation of age pigment by the ﬂeshﬂy Sarcophaga bullata Parker (Diptera: Sarcophagidae). Australian Journal of Zoology. 31: 131– 138.

FAO (1985). Pest control in stored produce. Prevention of post harvest food losses: a training manual. Retrieved from http://www.fao.org/docrep/x0039e/X0039E00.htm#Contents

FAO (1994). Insect pests of stored grains in hot climates. In Proctor D.L. (Editor). FAO Agricultural Services Bulletin. 109. Retrieved from http://www.fao.org/docrep/T1838E/T1838E1f.htm

FAO (1996). Rome declaration on world food security and world food summit plan of action. World Food Summit: Rome.

102

FAO (2000). The underlying causes of food insecurity. The elimination of food insecurity in the Horn of Africa: Final report. Retrieved from http://www.fao.org/docrep/003/x8406e/X8406e01.htm#P363_28423

Fedina, T.Y. (2007). Cryptic female choice during spermatophore transfer in Tribolium castaneum (Coleoptera : Tenebrionidae). Journal of Insect Physiology. 53: 93-98.

Fedina, T.Y. & Lewis, S.M. (2007). Female mate choice across mating stages and between sequential mates in flour beetles. Journal of Evolutionary Biology. 20: 2138-2143.

Fedina, T.Y. & Lewis, S.M. (2008) An integrative view of sexual selection in Tribolium flour beetles. Biological Reviews. 83: 151-171.

Fleming, J.E., Reveillaud, I. & Niedzwiecki, A. (1992). Role of oxidative stress in Drosophila aging. Mutation Research. 275(3-6): 267-279.

Freeman, J. A. (1945). Studies in the distribution of insects by aerial currents. The insect population of the air from ground level to 300 feet. Journal of Animal Ecology. 14: 128-154.

Frentiu, F.D. & Chenoweth, S.F. (2008). Polyandry and paternity skew in natural and experimental populations of Drosophila serrata. Molecular Ecology. 17: 1589-1596.

Gao, X. & Dave, S.R. (2007). Quantum dots for cancer molecular imaging. In: Warren C.W. Chan. (Ed.). Bio-Applications of Nanoparticles. Landes Bioscience: Austin, USA.

Gao, X.H., Cui, Y.Y, Levenson, R.M., Chung, L.W.K. & Nie, S.M. (2004). In vivo cancer targeting and imaging with semiconductor quantum dots. Nature Biotechnology. 22(8): 969-976.

Geys, J., De Vos, R., Nemery, B. & Hoet, P.H. (2009). In vitro translocation of quantum dots and influence of oxidative stress. American Journal of Physiology. 297(5): L903-911.

Good, N.E. (1933). Biology of the flour beetles, Tribolium confusum Duv. and T. ferrugineum Fab. Journal of Agricultural Research. 46: 327–334.

103 Good, N.E. (1936). The flour beetles of the genus Tribolium. USDA Technical Bulletin. 498: pp. 35.

GraphPad Prism version 7.00 for Mac. GraphPad Software. La Jolla, California US.

Grazer, V.M. & Martin, O.Y. (2012) Elevated temperature changes female costs and benefits of reproduction. Evolutionary Ecology. 26:625–637

GRDC (2011). Stored grain pests identification: The back pocket guide. Retrieved from https://www.grdc.com.au/~/media/8253D697BA6F4BF3AA5B5CBDFA7F4D2D.pdf

GRDC (2014). Grain storage: Grain fumigation guide. Retrieved from http://www.grdc.com.au/GRDC-FS-GrainFumigationGuide.

Gullan, P.J., & Cranston, P.S. (2014). The Insects: An Outline of Entomology (Fifth Edition). John Wiley & Sons, Incorporated, U.K. 493–525.

Hagler, J.R. & Jackson, C.G. (2001). Methods for marking insects: Current techniques and future prospects. Annual Review of Entomology. 46: 511 – 543.

Hagstrum, D.W. (1973). Infestation of flour by Tribolium castaneum: relationship between distribution of eggs and adults in a dispersing population. Annals of the Entomological Society of America. 66: 587–591.

Hagstrum, D.W. (2001). Immigration of insects into bins storing newly harvested wheat on 12 Kansas farms. Journal of Stored Product Research. 37: 221 – 229.

Hagstrum, D.W. & Gilbert, E.E. (1976). Emigration rate and age structure dynamics of Tribolium castaneum populations during growth phase of a colonization episode. Environmental Entomology. 5: 445–448.

Hagstrum, D.W., & Subramanyam, B. (2006). Fundamentals of Stored-Product Entomology. AACC International, Minnesota, USA.

104 Halstead, D.G. (1963). External sex differences in stored products Coleoptera. Bulletin of Entomological Research. 54: 199-234.

Hardy, A. C & Milne, P. S. (1938). Studies in the distribution of insects by aerial currents. Experiments in aerial tow-netting from kites. Journal of Animal Ecology. 71: 199-229.

Haubruge, E., Arnaud, L. & Mignon, J. (1997) The impact of sperm precedence in malathion resistance transmission in populations of the red flour beetle Tribolium castaneum (Herbst) (Coleoptera: Tenebrionidae). Journal of Stored Product Research. 33:143–146.

Howe, R. W. (1952). Entomological problems of food storage in Northern Nigeria. Bulletin of Entomological Research. 43: 111-144.

Howe, R.W. (1956). The effect of temperature and humidity on the rate of development and mortality of Tribolium castaneum (Herbst) (Coleoptera: Tenebrionidae). Annals of Applied Biology. 44(2): 356-368.

Jagadeesan R., Collins P.J., Daglish G.J., Ebert P.R. & Schlipalius, D.I. (2012) Phosphine resistance in the rust red flour beetle, Tribolium castaneum (Coleoptera: Tenebrionidae): inheritance, gene interactions and fitness costs. PloS One 7: 1- 12.

Jagadeesan, R., & Nayak, M.K. (in press). Phosphine resistance does not confer cross-resistance to sulfuryl fluoride in four major stored grain insect pests. Pest Management Science. doi: 10.1002/ps.4468

Jagadeesan, R., Fotheringham, A., Ebert, P.R. & Schlipalius, D.I. (2013). Rapid genome wide mapping of phosphine resistance loci by a simple regional averaging analysis in the red flour beetle, Tribolium castaneum. BioMed Central Genomics. 14: 1–12.

Jones, A. (2008). GERUD 2.0: a computer program for the reconstruction of parental genotypes from half-sib progeny arrays with known or unknown parents. Molecular Ecology Resources. 5: 708-711.

Jones, B. E. (1967). Tribolium castaneum (Herbst) (Coleoptera, Tenebrionidae) observed flying from an unusual habitat in Britain. Journal of Stored Products Research. 3(2): 185-187.

105

Ju, S.J., Secor, D.H. & Harvey, H.R. (1999). Use of extractable lipofuscin for age determination of blue crab Callinectes sapidus. Marine Ecology Progress Series. 185: 171−179.

Kaur, R. & Nayak, M.K. (2015). Developing effective fumigation protocols to manage strongly phosphine-resistant Cryptolestes ferrugineus (Coleoptera: Laemophloeidae). Pest Management Science. 71(9): 1297-1302.

Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton, S., Cooper, A., Markowitz, S. & Duran, C. (2012). Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics. 28: 1647- 1649.

Kennington, G.S. (1953). The effects of reduced atmospheric pressure on populations of Tribolium castaneum and Tribolium confusum. Physiological Zoology. 26(2): 179-203.

Kerstes, A.G., Be´re´nos, C. & Martin, O.Y. (2013) Coevolving parasites and population size shape the evolution of mating behaviour. BMC Evolutionary Biology. 13:1471–2148.

Knapp, M. & Knappová, J. (2013). Measurement of body condition in a common carabid beetle, Poecilus cupreus: A comparison of fresh weight, dry weight, and fat content. Journal of Insect Science. 13(6): doi.org/10.1673/031.013.0601.

Kokko, H. & Mappes, J. (2012). Multiple mating by females is a natural outcome of a null model of male encounters. Entomologia Experimentalis et Applicata. 146: 26-37.

Kokko, H. & Rankin, D.J. (2006) Lonely hearts or sex in the city? Density-dependent effects in mating systems. Philosophical Transactions of the Royal Society B: Biological Sciences. 361: 319-334.

Kumari, P.C., Sivadasan, R. & Jose, A. Microflora associated with the red flour beetles Tribolium castaneum (Coleoptera, Tenebrionidae). Journal of Agricultural Technology. 7(6): 1625-1631.

Lard M., Bäckman, J., Yakovleva, M., Danielsson, B. & Hansson, L-A. (2010). Tracking the small with the smallest - using nanotechnology in tracking zooplankton. PLoS One. 5(10): 1–6.

106

Lavie, B. & Ritte, U. (1978). The relationship between dispersal behaviour and reproductive ﬁtness in the ﬂour beetle, Tribolium castaneum. Canadian Journal of Genetics and Cytology. 20: 589- 595.

Lavie, B. & Ritte, U. (1980). Correlated effects of the response to conditioned medium in the flour beetle, Tribolium castaneum. Researches on Population Ecology 21: 228-232. doi:10.1007/bf02513623.

Lemel, J-Y., Belichon, S., Clobert, J. & Hochberg, M.E. (1997). The evolution of dispersal in a two-patch system: Some consequences of differences between migrants and residents. Evolutionary Ecology. 11: 613-629.

Lewis, S.M. (2004). Multiple mating and repeated copulations: effects on male reproductive success in red flour beetles. Animal Behaviour. 67: 799-804.

Lewis, S.M. & Austad, S.N. (1990). Sources of intraspecific variation in sperm precedence in red flour beetles. The American Naturalist. 135: 351-359.

Lewis, S.M. & Juthiewicz, E. (1998). Sperm precedence and sperm storage in multiply mated red flour beetles. Behavioural Ecology and Sociobiology. 43: 365-369.

Lewis, S.M., Kobel, A., Fedina, T. & Beeman, R.W. (2005) Sperm stratification and paternity success in red flour beetles. Physiological Entomology. 30: 303-307.

Lippman, R.D. (1981). Developments in Aging Research: I. Free Radicals—A Primary Cause of Aging. Upsala Journal of Medical Sciences. 86(3):319-322. doi: .3109/03009738109179243.

Loconti, J.D., & Roth, L.M. (1953). Composition of the odorous secretion of Tribolium castaneum. Annals of the Entomological Society of America. 46: 281-289.

Lovett, B., & Leger, R.St., (2016). Genetics and Molecular Biology of Entomopathogenic Fungi- Volume 94 of Advances in Genetics. Academic Press: Massachusetts, USA.

107 Lumley, A.J., Michalczyk, L., Kitson, J.J.N., Spurgin, L.G., Morrison, C.A., Godwin, J.L., Dickinson, M.E., Martin, O.Y., Emerson, B.C., Chapman, T. & Gage, M.J.G. (2015). Sexual selection protects against extinction. Nature. doi:10.1038/nature14419.

Magwere, T., Pamplona, R., Miwa, S., Martinez-Diaz, P., Portero-Otin, M., Brand, M.D., et al. (2006). Flight activity, mortality rates, and lipoxidative damage in Drosophila. Journals of Gerontology. Series A-Biological Sciences Medical Sciences. 61(2):136–45.

Mahroof, R.M., Edde, P.A., Robertson, B.J., Puckette, A. & Phillips, T.W. (2010). Dispersal of Rhyzopertha dominica (Coleoptera: Bostrichidae) in different habitats. Environmental Entomology. 39: 930 – 938.

Mahroof, R.M., et al., (2010). Dispersal of Rhyzopertha dominica (Coleoptera: Bostrichidae) in different habitats. Environmental Entomology. 39(3): 930–938.

Malekpour, R., Rafter, M.A., Daglish G.J. & Walter, G.H. (2016). Influence of phosphine resistance genes on flight propensity and resource location in Tribolium castaneum (Herbst) (Coleoptera: Tenebrionidae): the landscape for selection. Biological Journal of the Linnaean Society. 119: 348-358.

Manivannan, S. (2015). Toxicity of phosphine on the developmental stages of rust-red flour beetle, Tribolium castaneum Herbst over a range of concentrations and exposures. Journal of Food Science and Technology. 52(10): 6810-6815.

Mau, Y.S., Collins, P.J., Daglish, G.J., Nayak, M.K., Pavic, H. & Ebert, P.R. (2012a). The rph1 gene is a common contributor to the evolution of phosphine resistance in independent field isolates of Rhyzopertha dominica. PLoS One. 7: e31541.

Mau, Y.S., Collins, P.J., Daglish, G.J., Nayak, M.K., Pavic, H. & Ebert, P.R. (2012b). The rph2 gene is responsible for high level resistance to phosphine in independent field strains of Rhyzopertha dominica. PLoS One. 7: e34027.

McArthur, M.C. & Sohal, R.S. (1982). Relationship between metabolic rate, aging, lipid peroxidation, and ﬂourescent age pigment in milkweed bug, Oncopeltus fasciatus (Hemiptera). Journal of Gerentology. 37: 268–274.

108

Michalczyk, L., Millard, A.L., Martin, O.Y., Lumley, A.J., Emerson, B.C., Chapman, T. & Gage, M.J.G. (2011) Inbreeding promotes female promiscuity. Science. 333:1739–1741.

Miller, L.R. (1993). Fluorescent dyes as markers in studies of foraging biology of termite colonies (Isoptera). Sociobiology. 23:127–34.

Nabiev, I., Mitchell, S., Davies, A., Williams, Y., Kelleher, D., Moore, R. & Gun'ko Y.K. (2007). Nonfunctionalized nanocrystals can exploit a cell's active transport machinery delivering them to specific nuclear and cytoplasmic compartments. Nano Letters. 7(11): 3452–3461.

Nath, N.S. Bhattacharya, I., Tuck, A.G., Schlipalius, D.I. & Ebert, P.R. (2011). Mechanisms of phosphine toxicity. Journal of Toxicology 2011: 1-9.

Nayak, M. (2013). Managing strongly phosphine resistant rusty grain beetle. Australian Grain. 23: 10–14.

Nguyen, T.T., Collins, P.J. & Ebert, P.R. (2015). Inheritance and characterization of strong resistance to phosphine in Sitophilus oryzae (L.), PLoS One 10: 1-14.

Niitepõld, K., & Boggs, C.L., (2015). Effects of increased Flight on the energetics and life history of the butterfly Speyeria mormonia. PLoS ONE 10(10): e0140104.

Nilsson, T., Fricke, C. & Arnqvist, G. (2002). Patterns of divergence in the effects of mating on female reproductive performance in flour beetles. Evolution. 56: 111-120.

Niu, Y., Hardy, G., Agarwal, M., Hua, L. & Ren, Y. (2016). Characterization of volatiles Tribolium castaneum (H.) in flour using solid phase microextraction–gas chromatography mass spectrometry (SPME–GCMS). Food Science and Human Wellness. 5(1): 24-29. doi.org/10.1016/j.fshw.2015.11.002.

Nudds, R.L. & Bryant, D.M. (2000). The energetic cost of short flights in birds. Journal of Experimental Biology. 203(10): 1561–72.

109 Opit, G.P., Phillips, T.W., Aikins, M.J. & Hasan, M.M. (2012). Phosphine resistance in Tribolium castaneum and Rhyzopertha dominica from stored wheat in Oklahoma. Entomological Society of America. 105: 1107 – 1114.

Ozone Secretariat (2012). Handbook for the Montreal protocol on substances that deplete the ozone layer. UNEP. UNON: Nairobi, KE.

Pai, A.T. & Yan, G.Y. (2002). Polyandry produces sexy sons at the cost of daughters in red flour beetles. Proceedings of the Royal Society B-Biological Sciences. 269: 361-368.

Pai, A.T. & Yan, G.T. (2003). Rapid female multiple mating in red flour beetles (Tribolium castaneum). Canadian Journal of Zoology. 81: 888-896.

Pai, A., Feil, S. & Yan, G. (2007). Variation in polyandry and its fitness consequences among populations of the red flour beetle, Tribolium castaneum. Evolutionary Ecology. 21: 687-702.

Pai, A. & Bernasconi, G. (2008). Polyandry and female control: the red flour beetle as a case study. Journal of Experimental Zoology. 310(B): 148-159.

Palmer D.N., Oswald M.J., Westlake V.J. & Kay G.W. (2002). The origin of fluorescence in the neuronal ceroid lipofuscinosis (Batten disease) and neuron cultures from affected sheep for studies of neurodegeneration. Archives of Gerontology and Geriatrics. 34:343–358.

Payne, N.M. (1925). Some effects of Tribolium on flour. Journal of Economic Entomology. 18(5): 735-744.

Peakall, R. & Smouse, P.E. (2006). GenAlEX 6: genetic analysis in Excel. Population genetic software for teaching and research. Molecular Ecology Notes. 6: 288-295.

Peakall, R. & Smouse, P.E. (2012). GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research - an update. Bioinformatics. 28: 2537-2539.

Perez-Mendoza, J., Campbell, J.F. & Throne, J.E. (2011a). Effects of rearing density, age, sex, and food deprivation on flight initiation of the red flour beetle (Coleoptera: Tenebrionidae). Journal of Economic Entomology 104: 443-451.

110

Perez-Mendoza, J., Campbell, J.F. & Throne, J.E. (2011b). Influence of age, mating status, sex, quantity of food, and long-term food deprivation on red flour beetle (Coleoptera: Tenebrionidae) flight initiation. Journal of Economic Entomology 104: 2078-2086.

Perez-Mendoza, J., Campbell, J.F. & Throne, J.E. (2014). Effect of abiotic factors on initiation of red flour beetle (Coleoptera: Tenebrionidae) flight. Journal of Economic Entomology 107: 469- 472.

Perez-Mendoza, J., Dover, B.A., Hagstrum, D.W. & Baker, J.E. (1998). Flight activity of Rhyzopertha dominica (Coleoptera: Bostrichidae) in response to feeding damage and accumulation of waste. Journal of Economic Entomology. 91: 1445–1448.

Perez-Mendoza, J., Dover, B.A., Hagstrum, D.W. & Hopkins, T.L. (1999a). Effect of crowding, food deprivation, and diet on flight initiation and lipid reserves of the lesser grain borer, Rhyzopertha dominica. Entomologia Experimentalis et Applicata. 91: 317–326.

Perez-Mendoza, J., Hagstrum, D.W., Dover, B.A., Hopkins, T.L. & Baker, J.E. (1999b). Flight response, body weight, and lipid content of Rhyzopertha dominica (F.) (Coleoptera: Bostrichidae) as influenced by strain, season and phenotype. Journal of Stored Products Research. 35(2):183-196. doi.org/10.1016/S0022-474X(98)00044-7.

Perez-Mendoza, J., Throne, J.E., Dowell, F.E. & Baker, J.E. (2004). Chronological age-grading of three species of stored-product beetles by using near-infrared spectroscopy. Journal of Economic Entomology. 97: 1159 – 1167.

Pimentel, M.A.G., Faroni, L.R.D.A., Silva, F.H., Batista, M.D. & Guedes, R.N.C. (2010). Spread of phosphine resistance among Brazilian populations of three species of stored product insects. Neotropical Entomology. 39: 101- 107.

Pimentel, M.A.G., Faroni, L.R.D’A., Totola, M.R. & Guedes, R.N.C. (2007). Phosphine resistance, respiration rate and fitness consequences in stored-product insects. Pest Management Science. 63: 876-881.

111 Pizzari, T. & Wedell, N. (2013). The polyandry revolution. Philosophical Transactions of the Royal Society B. 368: 1-5.

Price, L.A. & Mills, K.A. (1988). The toxicity of phosphine to the immature stages of resistant and susceptible strains of some common stored product beetles, and implications for their control. Journal of Stored Products Research. 24: 51-59.

Qazi, M.C.B., Herbeck, J.T. & Lewis, S.M. (1996) Mechanisms of sperm transfer and storage in the red flour beetle. Annals of the Entomological Society of America. 89: 892–897.

R Development Core Team. (2013) R: a language and environment for statistical computing. R Foundation for Statistical Computing: Vienna, Austria.

Rafter M.A., Ridley, A.W., Daglish, G.J., Burrill, P.R. & Walter G.H. (2015). Flight directionality of Tribolium castaneum soon after take-off under glasshouse and field conditions. Entomologia Experimentalis et Applicata. 156: 178-186.

Rafter, M.A., McCulloch, G.A., Daglish, G.J. & Walter, G.H. (2017). Progression of phosphine resistance in susceptible Tribolium castaneum (Herbst) populations under different immigration regimes and selection pressures. Evolutionary Applications. doi: 10.1111/eva.12493.

Rajendran, S. 1999. Phosphine resistance in stored grain insect pests in India. In: Jin, Z., Liang, Q., Liang, Y., Tan, X. & Guan, L. (Eds.). Stored product protection. Proceedings of the Seventh International Working Conference on Stored-product Protection, Beijing, China, 14–19 October 1998. Chengdu, Sichuan Publishing House of Science and Technology. 635–641.

Rankin, M. A., McAnelly, M. L. &. Bodenhamer, J. E. (1986). The oogenesis-flight syndrome revisited. In: Danthanarayana, W. (Ed.). Insect Flight, Dispersal and Migration. Springer- Verlag. Heidelberg, DE.

Reichmuth, C. (1994). Uptake of phosphine by stored-product pest insects during fumigation. Proceedings of the 6th international working conference on stored product protection, 17-23 April 1994, Canberra, Australia.1: 157-162.

112 Ridley, A.W., Hereward, J.P., Daglish, G.J., Raghu, S., Collins, P.J. & Walter, G.H. (2011). The spatiotemporal dynamics of Tribolium castaneum (Herbst): adult flight and gene flow. Molecular Ecology 20: 1635-1646. doi:10.1111/j.1365-294X.2011.05049.x.

Ridley, A.W., Hereward, J.P., Daglish, G.J., Raghu, S., McCulloch, G.A. & Walter, G.H. (2016) Flight of Rhyzopertha dominica (Coleoptera: Bostrichidae) – a spatio-temporal analysis with pheromone trapping and population genetics. Journal of Economic Entomology. 109: 2561- 2571.

Ridley, A.W., Seymour, M., Schlipalius, D.I., Rafter, M.A. & Collins, P.J. (2012). Sublethal exposure to phosphine decreases offspring production in strongly phosphine resistant female red flour beetles, Tribolium castaneum (Herbst). PLOS ONE. 7(12): e53356. doi:10.1371/journal.pone.0053356.

Ritte, U. & Lavie, B. (1977). The genetic basis of dispersal behavior in the flour beetle Tribolium castaneum. Canadian Journal of Genetics and Cytology 19: 717-722. doi:10.1139/g77-078.

Robson, S.K.A. & Crozier, R.H. (2009). An evaluation of two biochemical methods of age determination in insects (pteridines and lipofuscins) using the ant Polyrhachis sexpinosa Latrielle (Hymenoptera: Formicidae). Australian Journal of Entomology. 48: 102 – 106.

Rousset, F. (2008). GENEPOP’007: a complete re-implementation of the GENEPOP software for Windows and Linux. Molecular Ecology Resources. 8: 103-106.

Sadek, M.M. (2001). Polyandry in field-collected Spodoptera littoralis moths and laboratory assessment of the effects of male mating history. Entomologia Experimentalis et Applicata. 98:165-172.

Schlipalius, D.I., Chen, W., Collins, P.J., Nguyen, T., Reilly, P.E. & Ebert, P.R. (2008). Gene interactions constrain the course of evolution of phosphine resistance in the lesser grain borer, Rhyzopertha dominica. Heredity. 100: 506–516.

Schlipalius, D.I., Cheng, Q., Reilly, P.E.B., Collins, P.J., Ebert, P.R. (2002). Genetic linkage analysis of the lesser grain borer Rhyzopertha dominica identifies two loci that confer high-level resistance to the fumigant phosphine. Genetics.161(2): 773-782.

113

Schlipalius, D.I., Collins, P.J., Mau, Y. & Ebert, P.R. (2006). New tools for management of phosphine resistance. Outlooks Pest Management. 17: 51–56.

Schlipalius, D.I., Valmas, N., Tuck., A.G., Jagadeesan, R., Ma, L., Kaur, R., Goldinger, A., et al. (2012). A core metabolic enzyme mediates resistance to phosphine gas. Science Magazine. 338: 807–810.

Semeao, A.A., Campbell, J.F., Whitworth, J.R. & Sloderbeck, P.E. (2013). Movement of Tribolium castaneum within a flour mill. Journal of Stored Products Research. 54: 17-22. doi.org/10.1016/j.jspr.2013.03.004.

Senthilkumar, T., Jayas, D.S., White, N.D.G., Freund, M.S., Shafai, C. & Thomson, D.J. (2012). Characterization of volatile organic compounds released by granivorous insects in stored wheat. Journal of Stored Product Research. 48: 91–96.

Sheehy, M.R.J. (2008). Questioning the use of biochemical extraction to measure lipofuscin for age determination of crabs: Comment on Ju et al. (1999, 2001). Marine Ecology Progress Series. 353: 303–306.

Sheehy, M.R.J. & Ettershank, G. (1989). Extractable age pigment-like auto-ﬂourescence and its relationship to growth and age in the water-ﬂea Daphnia carinata King (Crustacea: Cladocera). Australian Journal of Zoology. 36: 611–625.

Sheehy, M.R.J. & Roberts, B.E. (1991). An alternative explanation for anomalies in ‘soluble lipofuscin’ ﬂourescence data from insects, crustaceans, and other aquatic species. Experimental Gerontology. 26: 495–509.

Showers, W.B. (1997). Migratory ecology of the black cutworm. Annual Review of Entomology. 42: 393–425.

Simmons, L.W. (2005). The evolution of polyandry: sperm competition, sperm selection, and offspring viability. Annual Review of Ecology, Evolution and Systematics. 36: 125–146.

114 Sinclair, E.R. (1981). Sexing live adult Rhyzopertha dominica (F.) (Coleoptera: Bostrichidae). Journal of Stored Products Research. 17: 143-145.

Sinclair, E.R. & Alder, E.J. (1984). Migration of stored insect pest from a small wheat bulk. Australian Journal of Experimental Agriculture and Animal Husbandry. 24: 260–266.

Singh, S. (2013). In: The Gazette of India. The National Food Security Act, 2013. (New Delhi, Ministry of Law & Justice). Retrieved from http://indiacode.nic.in/acts-in-pdf/202013.pdf.

Singh, S. & Prakash, S. (2015). Effect of temperature and humidity on the culture of Tribolium castaneum, Herbst (Coleoptera: Tenebrionidae) in the laboratory. International Journal of Scientific and Research Publications. 5(7): 1-6.

Sinha, R.N. (1953). The spermatheca in the flour beetle (Tribolium castaneum Herbst). Journal of the New York Entomological Society. 61:131–134.

Smith, A.M., Duan, H., Mohs, A.M., Nie, S.M. (2008). Bioconjugated quantum dots for in vivo molecular and cellular imaging. Advanced Drug Delivery Reviews. 60:1226–1240.

Sohal, R. S. (1985). Aging in insects. In: G. A. Kerhut & L. I. Gilbert (Eds.). Comprehensive insect physiology, biochemistry, and pharmacology. Pergamon: Oxford, UK.

Sohal, R.S., & Donato, H. (1979). Effect of experimental prolongation of life span on lipofuscin content and lysosomal enzyme activity in the brain of the houseﬂy, Musca domestica. Journal of Gerontology. 34: 489– 496.

Sokoloff, A. (1972). The Biology of Tribolium: with special emphasis on genetic aspects. 1: Clarendon Press: Oxford, UK.

Sokoloff, A., (1974). The Biology of Tribolium. Oxford University Press, U.K.

Southwood, T.R.E., (1978). Ecological methods with particular reference to the study of insect populations (Second Edition). John Wiley & Sons, Inc., New York, USA. 70–91.

115 Southwood, T.R.E., & Henderson, P. (2000). Ecological Methods (Third Edition). Blackwell Science: UK.

Surtees, G. (1963). Laboratory studies on dispersion behaviour of adult beetles in grain. III. - Tribolium castaneum (Hbst.) (Coleoptera, Tenebrionidae) and Cryptolestes ferrugineus (Steph.) (Coleoptera, Cucujidae). Bulletin of Entomological Research 54: 297-306. doi:10.1017/S0007485300048793.

Suzuki, T. & Mori, K. (1983). (4r, 8r)-(-)-4, 8-dimethyldecanal – the natural aggregation pheromone of the red flour beetle, Tribolium castaneum (Coleoptera, Tenebrionidae). Applied Entomology and Zoology. 18: 134–136.

Taylor, L.R. (1974). Insect migration, flight periodicity and boundary layer. Journal of Animal Ecology. 43: 228-235.

Taylor, L.R. & Taylor, R.A.J. (1983). The Ecology of Animal Movement. Clarendon Press. Oxford, UK.

Taylor, M.L., Price, T.A.R. & Wedell, N. (2014). Polyandry in nature: a global analysis. Trends in Ecology and Evolution. 29: 376-383.

Taylor, R.W.D. & Halliday, D. (1986). The geographical spread of resistance to phosphine by Coleopterous pests of stored products. In: British Crop Protection Conference–Pests and Diseases, Brighton Metropole, England, November 17-20, 1986. 1: 607-613.

Terman, A. & Brunk, U. T. (1998). Ceroid/lipofuscin formation in cultured human fibroblasts: The role of oxidative stress and lysosomal proteolysis. Mechanism of Ageing and Development. 104: 277–291.

Terman, A. & Brunk, U.T. (2004). Molecules in focus: Lipofuscin. The International Journal of Biochemistry and Cell Biology. 36: 1400 – 1404.

Terman, A., & Brunk, U.T. (1998). Lipofuscin: Mechanisms of formation and increase with age. APMIS.106(2): 265-276.

116 Terman, A., Kurz ,T., Navratil, M., Arriaga, E.A. & Brunk, U.T. (2010). Mitochondrial turnover and aging of long-lived postmitotic cells: The mitochondrial-lysosomal axis theory of aging. Antioxidants & Redox Signaling. 12: 503 – 535.

Thompson, V. (1966). The biology of the lesser grain borer Rhyzopertha dominica (Fab.). Bulletin of Grain Technology. 4: 163-168.

Trematerra P., Ianiro R., Athanassiou C.G. & Kavallieratos N.G., 2015 – Behavioral interactions of Sitophilus zeamais and Tribolium castaneum: the first colonizer matters. – Journal of Pest Science, 88 (3): 573-581.

Turnell, B.R. & Shaw, K.L. (2015) Polyandry and post copulatory sexual selection in a wild population. Molecular Ecology. 24: 6278-6288.

Tutt, J.W. (1902). The Migration and Dispersal of Insects. Elliott Stock. London, UK.

Tyler, F. & Tregenza, T. (2013). Why do so many flour beetle copulations fail? Entomologia Experimentalis et Applicata. 146:199–206.

Urquhart, F. (1987) The Monarch Butterfly: International Traveller. Nelson-Hall, Chicago, USA.

Vande Velde, L. & Van Dyck, H. (2013). Lipid economy, flight activity and reproductive behaviour in the speckled wood butterfly: on the energetic cost of territory holding. Oikos. 122(4): 555– 62.

Wade, M.J. (1977). An experimental study of group selection. Evolution. 31: 134-153.

Walker, T.J. & Wineriter, S.A. (1981). Marking techniques for recognizing individual insects. The Florida Entomologist. 64: 18–29.

Walter, G.H. & Hengeveld, R. (2014). Autecology: Organisms, Interactions and Environmental Dynamics. CRC Press, Taylor & Francis group. Boca Raton, USA.

117 Wang, J., Han, S., Ke, D. & Wang, R. (2012). Semiconductor quantum dots surface modification for potential cancer diagnostic and therapeutic applications. Journal of Nanomaterials. 1-8. doi:10.1155/2012/129041.

Warrick, C. (2011). Fumigating with phosphine, other fumigants and controlled atmospheres. Grains Research & Development Corporation, Stored Grain Project, Australia. www.grdc.com.au/uploads/documents/GRDC-Fumigating-withPhosphine-other-fumigants-and- controlled-atmospheres.pdf

Watson, P.J., Arnqvist, G. & Stallmann, R.R. (1998). Sexual conflict and the energetic costs of mating and mate choice in water striders. The American Naturalist. 151(1): 46-58.

Wellington, W.G. (1946). The effects of variations in atmospheric pressure upon insects. Canadian Journal of Research. 24d(2): 51-70.

White, G.G. (1981). Undetected infestation in wheat deliveries from farms. In: Williams P.A., & Amos T.G. (Eds.). First Australian Stored Grain Pest Control Conference. Commonwealth Scientific and Industrial Research Organisation: Melbourne, AU.

White, G.G. (1984). Variation between field and laboratory populations of Tribolium castaneum (Herbst) (Coleoptera: Tenebrionidae). Australian Journal of Ecology. 9: 153-155.

WHO (2012). Underweight in children. Global Health Observatory. Retrieved from http://www.who.int/gho/mdg/poverty_hunger/underweight/en/

Wickens, A.P. (2001). Aging and the free radical theory. Respiration Physiology. 128. 379–391.

Williams, C.B. (1930). Migration of Butterflies. Oliver & Boyd. Edinburgh, UK.

Williams, C.B. (1957). Insect migration. Annual Review of Entomology. 2:163-180.

Winks, R.G. (1984). The toxicity of phosphine to adults of Tribolium castaneum (Herbst): time as a dosage factor. Journal of Stored Products Research. 20: 45-56.

118 Winterbottom, D. C. (1922). Weevil in wheat and storage of grain in bags. A record of Australian experience during the war period (1915-1919). South Australian Government Printer: Adelaide, AU.

Wirtz, R.A., Taylor, S.L. & Semey, H.G. (1978). Concentrations of substituted p-benzoquinones and 1-pentadecene in the flour beetles Tribolium confusum J. Du Val and Tribolium castaneum (Herbst). Comparative Biochemistry and Physiology. 61 (1)B: 25-28.

Wright, E.J., & Morton, R. (1995). Daily flight activity of Trogoderma variabile (Coleoptera: Dermestidae) and Rhyzopertha dominica (Coleoptera: Bostrichidae). Journal of Stored Products Research. 31: 177–184.

Yan, L.J., & Sohal, R.S. (2000). Prevention of flight activity prolongs the life span of the housefly, Musca domestica, and attenuates the age-associated oxidative damage to specific mitochondrial proteins. Free Radical Biology and Medicine. 29(11): 1143–50.

Zain, M.E. (2011). Impact of mycotoxins on humans and animals. Journal of Saudi Chemical Society. 15: 129-144.Williams, C.B. (1957). Insect migration. Annual Review of Entomology. 2:163-180.

Zhu, Y., Li, Z., Chen, M., Cooper, H.M., Lu, G.Q. & Xu, Z.P. (2013). One-pot preparation of highly fluorescent cadmium telluride/cadmium sulfide quantum dots under neutral-pH condition for biological applications. Journal of Colloid and Interface Science. 390(1): 3-10. doi.org/10.1016/j.jcis.2012.08.003.

Ziegler, J.R. (1976). Evolution of the migration response: emigration by Tribolium and the influence of age. Evolution.30: 579–592.

Ziegler, J.R., (1977). Dispersal and reproduction in Tribolium: the inﬂuence of food level. Journal of Insect Physiology 23: 955-960.

Ziegler, J.R., (1978). Dispersal and reproduction in Tribolium: the inﬂuence of initial density. Environmental Entomology. 7: 149-156.

119 Appendix

The appendix provides supplementary information for the microsatellite markers used in the polyandry analyses of Tribolium castaneum in Chapters 3 & 4.

Appendix 3.1 Five primer sequences from the microsatellite markers of Tribolium castaneum Herbst (Demuth et al., 2007).

Name Linkage Forward Reverse Motif Length** group repeats* Tca-2.7 2 CTTACGACGGCT GCCCTCACAGGG TAT(8)- 230 GTGAAACA ATGAAATA TAT(6) Tca-7.1 7 TGTATTGCGTTT GACAAGGAATGT TAT (9) 239 TCTGCTCAA GGCGATCT Tca-7.4 7 CGTCTGTCTGGC CACAATTTAACA CAC (6) 197 TGTTTTGA CTTGGCACGTA Tca-8.6 8 TCCTGGACACAA GCGTGGGTCGGA GGTCA (5) 224 TCTCCCTAA TAGATATG Tca-8.8 8 TTTGTAGATTTG GGAATCCTGGAA TTA (20) 194 TATTTTGGCAAT TCTTAGAACG Table adapted from supplementary material from Demuth, et al. (2007). *Motif repeats have dashes in them to represent interruption between repeats **Length denotes the size of the fragment (bp) in the genome

120