An Investigation Into Factors Affecting Breeding Success in The

An investigation into factors affecting breeding success in the

Tasmanian devil

(Sarcophilus harrisii)

Tracey Catherine Russell

Faculty of Science

School of Life and Environmental Science

The University of Sydney

Australia

A thesis submitted in fulfilment of the requirements for the degree

Doctor of Philosophy 2018 Faculty of Science The University of Sydney

Table of Contents

Table of Figures ...... viii

Table of Tables ...... x

Acknowledgements ...... xi

Chapter Acknowledgements ...... xii

Abbreviations ...... xv

An investigation into factors affecting breeding success in the Tasmanian devil

(Sarcophilus harrisii) ...... xvii

Abstract ...... xvii

1 Chapter One: Introduction and literature review ...... 1

1.1 Devil Life History ...... 4

1.2 Devil Facial Tumour Disease (DFTD) ...... 7

1.3 The impact of DFTD on the Tasmanian ecosystem ...... 12

1.4 Captive breeding, a measure to save endangered species ...... 12

1.5 Captive breeding and genetic diversity ...... 13

1.6 Thesis chapters ...... 14

1.6.1 Animal Husbandry and captive breeding in confined enclosures – chapter 2 ... 14

1.7 Devil reproduction ...... 16

1.7.1 Polyandry in a wild devil population – chapter 3 ...... 16

1.7.2 The major histocompatibility complex (MHC) and mate choice in captive devils

Faculty of Science The University of Sydney

– chapter 4 ...... 17

1.7.3 Investigations into devil endocrinology – chapter 5 ...... 19

1.8 References ...... 23

2 Chapter Two: An examination of factors associated with breeding success of

Tasmanian devils in captivity (2006 to 2012) ...... 44

2.1 Abstract ...... 44

2.2 Introduction ...... 45

2.3 Methods ...... 49

2.3.1 Insurance Population Establishment ...... 49

2.3.2 Insurance Population Breeding Success 2006-2012 ...... 50

2.3.3 Founder Evaluation ...... 51

2.3.4 Studbook Evaluation ...... 52

2.3.5 Housing ...... 53

2.3.6 Statistical Analysis ...... 53

2.4 Results...... 54

2.4.1 Insurance Population Breeding Success 2006-2012 ...... 54

2.4.2 Founder Evaluation ...... 57

2.4.3 Housing ...... 59

2.5 Discussion ...... 61

2.5.1 Breeding success ...... 62

2.5.2 Founder evaluation ...... 66

Faculty of Science The University of Sydney

2.5.3 Housing and stress ...... 67

2.5.4 Acknowledgements ...... 71

2.6 References ...... 72

3 Chapter Three: Multiple paternity and precocial breeding in wild Tasmanian devils (Sarcophilus harrisii) ...... 81

3.1 Abstract ...... 81

3.2 Introduction ...... 82

3.3 Materials and methods ...... 84

3.3.1 Genetic analyses ...... 84

3.3.2 Allele frequency, heterozygosity and linkage disequilibrium ...... 85

3.4 Results...... 86

3.4.1 Genetic analyses ...... 86

3.4.2 Paternity analyses ...... 89

3.4.3 Age ...... 90

3.5 Discussion ...... 91

3.6 Conclusion ...... 95

3.7 Acknowledgements ...... 96

3.8 References ...... 97

4 Chapter four: MHC diversity and female age underpin reproductive success in an Australian icon; the Tasmanian Devil ...... 107

4.1 Abstract ...... 108

4.2 Introduction ...... 109

iii

Faculty of Science The University of Sydney

4.3 Material and methods ...... 110

4.2.1 Study animals ...... 110

4.2.2 Genetic analyses ...... 111

4.2.3 Estimates of allele frequency, heterozygosity and linkage disequilibrium ...... 111

4.2.4 Statistical analyses ...... 111

4.3 Results...... 112

4.4 Discussion ...... 117

4.5 Conclusion ...... 118

4.6 References ...... 120

4.7 Acknowledgements ...... 127

4.8 Author Contributions ...... 127

5 Chapter Five: Exploring the role of endocrinology and behaviour in relation to the reproductive success of captive Tasmanian devils ...... 128

5.1 Abstract ...... 128

5.2 Introduction ...... 128

5.3 Materials and Methods ...... 132

5.3.1 Study animals ...... 132

5.3.2 Faecal samples ...... 133

5.3.3 Hormone analysis ...... 133

5.3.4 Behavioural observations ...... 134

5.3.5 Analysis ...... 135

Faculty of Science The University of Sydney

5.3.6 Statistical analysis ...... 135

5.4 Results...... 136

5.4.1 Progesterone profiles and timing of pairings ...... 136

5.4.2 Corticosterone ...... 138

5.4.3 Behavioural observations ...... 142

5.4.4 Breeding success/timing ...... 144

5.5 Discussion ...... 145

5.6 Conclusions...... 147

5.7 Acknowledgements ...... 147

5.8 References ...... 148

6 Chapter Six: Discussion and Conclusions ...... 154

6.1 Conclusion ...... 160

6.2 Future directions ...... 161

6.2.1 Mate choice and behavioural trials ...... 162

6.2.2 Hormone monitoring ...... 163

6.3 References ...... 165

7 Appendix One: The Devil’s Ark, the mark of success ...... 171

7.1 Abstract ...... 171

7.2 Introduction ...... 172

7.2.1 The Tasmanian Devil Insurance Program ...... 172

7.3 The future ...... 176

Faculty of Science The University of Sydney

7.4 References ...... 177

8 Appendix Two: Oncogenesis as a selective force: Adaptive evolution in the face of a transmissible cancer ...... 182

8.1 Abstract ...... 183

8.2 Introduction ...... 184

8.2.1 Host life-history traits, strategies and adaptation in response to parasites...... 185

8.2.2 Host life-history traits, strategies and adaptation in response to a transmissible cancer 187

8.2.3 Concluding remarks ...... 201

8.3 Acknowledgements ...... 202

8.4 References ...... 202

9 Appendix Three: Seasonality and breeding success of captive and wild

Tasmanian devils (Sarcophilus harrisii) ...... 218

9.1 Abstract ...... 219

9.2 Introduction ...... 220

9.3 Methodology ...... 227

9.3.1 Captive Breeding Population Data ...... 227

9.3.2 Wild Population Data ...... 228

9.3.3 Seasonal Fluctuations in Temperature and Daylight ...... 230

9.3.4 Statistics ...... 230

9.4 Results...... 231

9.4.1 Wild Population of Tasmanian Devils ...... 231

Faculty of Science The University of Sydney

9.4.2 Litter Size ...... 232

9.4.3 Timing of Seasonal Initiation of Breeding Activity ...... 233

9.4.4 Environmental conditions associated with the initiation of Breeding Activity233

9.5 Discussion ...... 236

9.5.1 Acknowledgements ...... 247

9.6 References ...... 248

10 Appendix Four: Genetic diversity, inbreeding and cancer ...... 256

10.1 Abstract ...... 257

10.2 Introduction ...... 258

10.3 Cancer aetiologies ...... 259

10.4 Genetic diversity, inbreeding and cancer in humans ...... 260

10.5 Genetic diversity, inbreeding and cancer in domestic and wild animals ...... 261

10.6 Genetic diversity, inbreeding and cancer development ...... 262

10.7 Conclusion ...... 264

10.7.1 Acknowledgements ...... 265

10.8 References ...... 265

vii

Faculty of Science The University of Sydney

Table of Figures

Figure 1.1 The distribution of DFTD over time...... 8

Figure 1.2 Association between stressors and overall translocation failure...... 20

Figure 2.1 a) Total number of adult female devils paired per year and b) produced a litter of 1-4 offspring between 2006-2012, demonstrating the shift from the breeding of wild-caught founders to captive born descendants over time...... 56

Figure 2.2 A representation of the proportion of male (n=71) and female (n=95) founders that produced no offspring and the over contribution of number of offspring the successful founders produced...... 59

Figure 2.3 A representation of the sex skew of overall founders and currently active founder lines for wild-caught devils...... 59

Figure 3.1 Weight (kg) ± S.E. for adult male and female devils from the Forestier remnant population. (Males 2009 n =1, 2010 n = 2, 2011 n = 12, Females 2010 n = 7,

2011 n = 9)...... 91

Figure 4.1 Relationship between female age and the success of reproduction in

Tasmanian devil ...... 114

Figure 4.2 Mode prediction of the interaction between the number of male and female heterozygous loci...... 116

Figure 5.1 The hormone profile of a female devil unsuccessfully paired with two male devils. Prog > baseline and Cort > baseline refer to all values for progesterone and corticosterone that were 1.5 s.d. above baseline. The receptive period (green) is when pairing should have occurred. Luteal phase (pink) coincides with a rise in progesterone and corticosterone...... 137

Figure 5.2 Progesterone and corticosterone profile for a successfully paired female devil...... 137

viii

Faculty of Science The University of Sydney

Figure 5.3 Relationship between body mass (kg) and corticosterone baseline levels in male devils...... 138

Figure 5.4 Relationship between body mass (kg) and corticosterone baseline levels in female devils...... 139

Figure 5.5 Corticosterone profile of a successfully paired male devil...... 140

Figure 5.6 Corticosterone profile of an unsuccessfully paired male devil...... 140

Figure 5.7 Corticosterone profile of a successfully paired female devil...... 141

Figure 5.8 Corticosterone profile of a female unsuccessfully paired with two male devils...... 141

Figure 9.1 Litter size distributions for wild and captive female devils...... 233

Figure 9.2 The estrous cycle timing for wild and captive devils...... 236

Faculty of Science The University of Sydney

Table of Tables

Table 2.1 Overall summary of the breeding success of the captive Tasmanian devil

Insurance Population (TDIP) (excluding FRE’s) between 2006-2012...... 55

Table 2.2 Number, age and average litter size of female Tasmanian devils recommended for breeding within the captive insurance program between 2007-2012...... 57

Table 3.1 Standard diversity indices and Hardy-Weinberg equilibrium...... 88

Table 4.1 Female – male pair combinations ...... 112

Table 4.2 Final results of the logistic mixed model...... 115

Table 5.1 Number of males paired with successful and unsuccessful females during their first, second and the third oestrous cycle...... 136

Table 7.1.1 The number of Tasmanian devil joeys born at Devil Ark 2011-2014 ...... 175

Table 9.1 The specific location and number of Tasmanian devils examined in this study. .... 228

Table 9.2 Summary of the mean values for the timing of the beginning of the first estrus

(week)...... 234

Faculty of Science The University of Sydney

Acknowledgements

It has been a privilege to work on such an iconic Australian animal. I would like to thank the following people for their role in facilitating my thesis completion. This was no mean feat and I can truly say that I could not have done it without the love, friendship and support of a lot of wonderful people.

Firstly thank you to my mum, who very sadly became ill and died while I was into the second year of this project. She always loved and believed in me, was a beautiful guide and loving friend, was always proud of me and supported my passion for science and wildlife research and would be very happy to see this project completed.

To Bea Ujvari, how can I thank you? You are my very dear friend and we have shared many, many laughs and a few tears too. Thank you for your kindness, your endless and willing generosity and help, your explanations, your patience and your optimism and encouragement which have meant so much.

Thank you Thomas Madsen, you have been such a tremendous help to me and I am so very grateful for your generosity of time, for sharing your endless knowledge, for your patience, insights, suggestions, explanations, and knowledge of all things statistical. You and Bea are an amazing and inspirational team.

Thank you Tamara Keeley for your incredible generosity, encouragement and collaboration and for answering endless questions with good humour and grace – I truly appreciate all of your input and time and for teaching me a lot. Thank you for being there with me throughout this journey.

To Rebecca Spindler thank you for allowing me to do this project and for all of your help and support. Thank you for your enthusiasm and for your can-do attitude in getting things

Faculty of Science The University of Sydney done and for offering help when needed. Thank you too for providing funding from the

Morris Animal Foundation grant.

Thank you to Amanda Lane for your help and never ending support and 2 am messages of encouragement. I look forward to working with you some more very soon.

Thanks to my lab buddies Liz, Belinda, Jian, Mette, Kat, Chen, YuanYuan, Ollie, and

Cassie, thanks for the laughs and for teaching me new techniques, and for chats and coffees and hugs. You are all a really good bunch of smart people and I have made some great and long lasting friends. I have learnt a lot from each and every one of you and I look forward to following some stellar careers in the future.

Thank you to Ros Bathgate for showing me kindness and support when I most needed it and for guiding me through some difficult times. Your help has been invaluable and much appreciated.

Thank you to my dear friends Sarah, Yvonne, Melissa, Belinda and to my family, who like me, look forward to putting this chapter behind us.

Chapter Acknowledgements

Appendix One: Devil Ark

I would like to acknowledge John Wiegel, the owner and founder of Devil Ark for providing information on Devil Ark and how it began as an idea, came to fruition and the specifications and workings of Devil Ark. I would also like to acknowledge Tim Faulkner and Liz Vella for information on the operations of Devil Ark and animals held at the institution.

xii

Faculty of Science The University of Sydney

Chapter Three: Multiple paternity

I would like to thank Dr. Amanda Lane for additionally confirming the multiple paternity results by the exclusion method and for helpful comments on this chapter.

Thank you to Dr. Katrina Morris for additional microsatellite results and comments on this chapter.

Thank you to Drs. Judy Clarke and Samantha Fox from DPIPWE for supplying additional information on the captured devils, their ages, weights and DFTD status. Thank you to Dr.

Shelly Lashich for comments on precocial breeding in the devil.

Thank you to DPIPWE staff for collecting and supplying the samples from these devils.

Chapter Five: Hormone analysis

Thank you to Dr. Beata Ujvari and Professor Thomas Madsen for guidance, ongoing advice, statistical and editorial support.

I would like to thank Dr. Rebecca Spindler for her supervision, advice and support during this project and for facilitating the completion of samples at Taronga Western Plains Zoo

(TPWZ).

I would like to acknowledge Dr. Tamara Keeley for providing corticosterone results for the male devils and for females from TPWZ, for organising the faecal collections from the zoological institutions and for training me in the procedures for the EIA’s and for how to process copious amounts of poo and answering endless questions with the utmost patience.

Thank you to Dr. Rebecca Hobbs and Roseanne Kirkup for reanalysing assays that were too concentrated and for running additional and completing incomplete progesterone and corticosterone assays.

Thank you to the keepers and staff at Taronga Zoo, Taronga Western Plains Zoo, Australia

Zoo, Healesville Sanctuary, DPIPWE institutions, Taroona, Cressy and Monarto Zoo for

xiii

Faculty of Science The University of Sydney collecting the faecal samples for this project and for providing behavioural observations of paired devils.

Thank you to Chen Wu and Kristy Playle for voluntary assistance in faecal preparation

(drying and sifting) and extractions, record keeping and for accompanying me on the

Dubbo field trips.

Chapter 4, Appendix 2 and Appendix 4 have been submitted for publication and have my contributions listed and signed off by lead or co-authors. Appendix 3 has been published and Chapter 3 has been prepared for publication to be submitted to Wildlife Research.

Therefore, there is some overlap in life history, DFTD occurrence and the initiation of the insurance population between chapters. Submitted chapters follow the format of the journal where they have been submitted to and therefore the formats are slightly different to the rest of the thesis.

Additional thanks

I would like to thank the Morris Animal Foundation for funding this project. In addition I would like to give my thanks and appreciation to all of the zoo keepers, zoo staff, ZAA and

DPIPWE staff who kindly collected samples or who gave me information for this project. I am very grateful. In particular I would like to thank Judy Clarke, Marissa Parrot and

Samantha Fox.

xiv

Faculty of Science The University of Sydney

Abbreviations

BP Before present

CBSG Captive Breeding Specialist Group

CTVT Canine Transmissible Venereal Tumour

DFTD Devil Facial Tumour Disease

DFT2 Devil Facial Tumour 2

DPIPWE Department of Primary Industries, Parks, Water and Environment

(Tasmania)

EIA Enzyme-immunoassay

FCM Faecal corticosterone metabolite

FGC Faecal glucocorticoid

FPM Faecal progesterone metabolite

GC Glucocorticoid

GCM Glucocorticoid metabolite

IUCN International Union for the Conservation of Nature

LH Life History

LRS Lifetime Reproductive Success

MEE Managed environmental enclosure

MHC Major Histocompatibility Complex

MK Mean Kinship

PCR Polymerase Chain Reaction

PHVA Population habitat and viability assessment

PVM Population viability model

PY Pouch young

Faculty of Science The University of Sydney

OR Olfactory Receptor

SNPs Single nucleotide polymorphisms

STTDP Save the Tasmanian Devil Program

TDIP Tasmanian devil insurance population

ZAA Zoo and Aquarium Association

xvi

Faculty of Science The University of Sydney

An investigation into factors affecting breeding success in the

Tasmanian devil (Sarcophilus harrisii)

Abstract

Wild populations of Tasmanian devils (Sarcophilus harrisii) were thought to be under the threat of extinction from devil facial tumour disease (DFTD). To ensure this species’ survival, an insurance population was developed in 2005 with the aim of saving the devils from extinction and to maintain 95% of the wild genetic diversity for 50 years. To achieve this aim breeding success needs to be optimal.

Within this PhD I explore factors influencing the breeding success within the Tasmania devil insurance population. The major findings are: i) in order to improve breeding success females should be bred at the onset of sexual maturity at age two. Pairing is recommended in the first oestrous period, and females should be paired with older males (Chapter 2), ii) multiple paternity has been documented for the first time and appears to be a common female reproductive strategy in devils. This could have positive implications for the captive breeding program by increasing genetic diversity within litters, iii) additionally, and for the first time, precocial male breeding has been documented in populations where older cohorts have succumbed to DFTD (Chapter 3), iv) I have demonstrated that disruptive selection on

MHC Class-I loci significantly enhances devil reproductive success (Chapter 4) and v)

xvii

Faculty of Science The University of Sydney through hormone analyses, I have confirmed, that the timing of pairing of devils by zoo keepers is in line with female receptivity, and that glucocorticoid levels do not appear to affect captive devil reproductive success (Chapter 5).

My thesis has explored the behavioural, animal husbandry and genetic factors influencing breeding success in the captive devil population and a wild one affected by DFTD. I highlight new life history traits and breeding strategies that could greatly enhance the success of the captive breeding program and which may well prove successful in other endangered species where breeding success is suboptimal.

xviii

Faculty of Science The University of Sydney

1 Chapter One: Introduction and literature review

Like the now extinct Thylacine (Thylacinus cynocephalus), the Tasmanian devil

(Sarcophilus harrisii) has been persecuted since European settlement and hunted relentlessly as a perceived threat to livestock (Guiler 1970a). Once roaming the entire

Australian continent, Tasmanian devils now occur only on the island of Tasmania (Johnson and Wroe 2003; Brown 2006). Their mainland demise was due to competition with dingos and humans in association with climate change. They were eradicated from the mainland

3000 - 4000 years before present (BP) (Wroe 2003). During the last two centuries the

Tasmanian devil, the world’s largest living carnivorous marsupial, has undergone significant population fluctuations (Guiler 1970a; Jones et al. 2003; Brüniche-Olsen et al.

2014). Presently however, the Tasmanian devil has been exposed to a novel threat in the past two decades: a transmissible cancer that threatens wild devils with extinction.

First reported in Tasmania’s north-east in 1996 Devil Facial Tumour Disease (DFTD) has caused population declines of up to 90% in some areas (Hawkins et al. 2006; Lachish et al.

2011; Epstein et al. 2016). DFTD is a contagious and fatal cancer spread by biting during social interactions which may have the capacity to render wild Tasmanian devils extinct within 15 -35 years (McCallum et al. 2007; Hamede et al. 2013). During the last 20 years

DFTD has spread rapidly and is affecting devils across much of their range. In 2005 the disease was present at over 51% of the devil’s range with absences only reported from the northwest and certain areas of the mid-north coast (Hawkins et al. 2006; Pyecroft et al.

2007). However, by 2018, over three quarters of Tasmania’s devil populations had been affected by this devastating disease (Lazenby et al. 2018; Figure 1.1).

Faculty of Science The University of Sydney

The loss of the Tasmanian devil, a keystone species and an apex predator, to extinction would be catastrophic to the Tasmanian ecosystem and would almost certainly promote the increase of introduced predators such as feral cats (Jones et al. 2007), induce the collapse of trophic cascades and alter Tasmania’s pristine ecosystem (Hollings et al. 2014; Hollings et al. 2016).

Due to the devastating effects of DFTD and the threat of losing the world’s largest carnivorous marsupial to extinction, an insurance population, sourced from disease free wild-caught devils, was established in 2005 with an aim of retaining 95% of the genetic diversity of the species (CBSG 2008; Hogg et al. 2016). The Tasmanian devil captive breeding program is the largest captive breeding project ever conducted in Australia.

Participation in the scheme involves institutions across six states and one territory and recently includes zoos in Denmark, the USA and New Zealand. It was originally thought that in the future, if DFTD could be eradicated, that devils from this program would be reintroduced into the remaining wild Tasmanian devil populations. At the time of writing captive devils are supplementing wild populations and are subject to regular monitoring

(STTDP 2018).

With mathematical models supporting the possibility of species extinction, it was proposed that DFTD had the capacity to lead to local extinctions within 15 - 35 years after its onset

(McCallum et al. 2007; McCallum et al. 2009; Hamede et al. 2012). However, these epidemiological models did not include or consider individual phenotypical difference, host mortality, susceptibility to infection or potential evolving adaptive responses to DFTD.

Recent studies have shown significant heterogeneities in several of the above parameters, suggesting that devils may yet overcome the threat of extinction from DFTD (Epstein et al.

2016; Pye et al. 2016; Ujvari et al. 2016; Russell et al. 2018).

Faculty of Science The University of Sydney

Tasmanian devils have persisted in DFTD affected areas for 20 years with no localised extinctions, thus supplementation of wild populations with captive bred devils has commenced. At the time of writing, 86 captive devils had been released back into the wild:

33 at Wukalina/Mt William (in the far north east of Tasmania) in 2017, 33 at Stony Head in the north in 2016, and 20 at Narawntapu National Park in 2015 (tassiedevil.com.au 2017).

Additionally, 39 animals were released onto the Forestier peninsular in 2015 (see Chapter

4) as well as further releases onto Maria Island (Thalman et al. 2016). Monitoring the survival of all released devils is ongoing.

Clearly, in order to have a successful population to release back into the wild, ensuring and maintaining the success of the captive breeding program is imperative. Therefore, it is essential to understand the factors underlying the life history and reproductive behaviour of the devils. The aims of my thesis were to use genetic, hormonal and behavioural information, in conjunction with published information regarding the life history of the devil in the wild, in order to investigate/decipher behavioural, phenotypic and genetic factors contributing to, and influencing, the reproductive fitness of the devils maintained in the captive Tasmanian Devil Insurance Population (TDIP).

Consequently, I will begin by describing the life history of Tasmanian devils in the wild, the nature of the disease ravaging the population, prior to a discussion of the factors influencing their reproductive success in captivity.

Faculty of Science The University of Sydney

1.1 Devil Life History

Tasmanian devils occupy a wide range of habitats, with sclerophyll forest and coastal scrub being favoured (Guiler 1970a; Hawkins et al. 2006). They are sexually dimorphic, with adult male body weight ranging between seven and 13 kg and female body weight between four and nine kg (Guiler 1970a). The life span of devils in the wild is approximately six years, although captive animals can live for over eight years.

Tasmanian devils are primarily nocturnal, non-territorial, solitary animals (Guiler 1970a), except when feeding (a maximum of 22 animals have been recorded feeding simultaneously on a cow carcass; Owen and Pemberton 2005). Both male and female devils have similar sized home ranges, although home range size is both density and resource dependent (Guiler 1970a; Pemberton 1990). Devils are highly mobile, travelling considerable distances at night in search of food. Males typically travel further than females

(Guiler 1970a), and some devils have been observed travelling up to 50 km/night (Hawkins et al. 2006; Lachish et al. 2011).

Devils have a very keen sense of smell and can detect food from up to a kilometre away

(Guiler, 1970a). In addition they are often alerted to a food source by the guttural vocalisations and screams (which is how they got the name of ‘devil’) resulting from competition over a food source (Owen and Pemberton 2005). Voracious feeders, devils can consume up to 40% of their body weight at a single feeding session (Pemberton and Renouf

1993) and have frequently been discovered by farmers ‘sleeping it off’ atop a cleaned fleece (S. Wilks pers comm).

Faculty of Science The University of Sydney

Devils have been recorded communicating in a variety of ways including vocalising, posturing and via chemical cues i.e. urination and ano-genital drag. They also use communal latrines which may also serve as a communication centre (Pemberton and

Renouf 1993).

Both sexes become sexually mature at an age of two years (Hawkins et al. 2006), with reproductive senescence in females often occurring by the age of five years (Kelly 1993;

Keeley et al. 2012). However, due to the significant demographic effects of DFTD, i.e. the high mortality in older devils, sexual maturity has been observed in ~50% of wild female devils at the age of one year (‘precocial breeding:’ Jones et al. 2008). This shift in age at sexual maturity has probably been facilitated by the increased availability of quality maternal dens and prey resulting from the DFTD related decline in devil numbers (Jones et al. 2008). Increased prey availability appears to have resulted in increased growth rates, as one year old females have been reported to reach a body mass of 6 kg, which is similar to that of older females (Lachish et al. 2009) and in excess of pre-DFTD mean juvenile body mass at one year.

Tasmanian devils have a synchronous breeding season, beginning in February/March and lasting for three to six weeks (Guiler 1970b; Hamede et al. 2009). Competition for females is fierce and males have been observed mating with numerous females (Kelly 2007).

Mating in devils can last up to 10 days, with males being very possessive of chosen females, keeping them away from other males by isolating them in mating dens and exhibiting mate guarding behaviours which include dragging the female around by the neck and preventing them from feeding (Kelly 2007).

Faculty of Science The University of Sydney

In spite of their synchronous breeding season some females may have a second, or rarely, a third, oestrus period. Breeding has been recorded in May (Hamede et al. 2009) in a second oestrus and as late in the year as August in the case of a third (Keeley et al. 2012).

Pregnancy lasts for approximately 18 days. Devils give birth to supernumerary young, and more than 20 may be born (Hughes 1982). As in all marsupials, the young are born in a very underdeveloped state (~ 0.18 – 0.29g) and have to make their way along a mucous trail from the birth canal to the pouch, where they complete their development (Flynn 1922;

Guiler 1970b). However, as female devils only have four teats, only the first four to attach will survive while the rest perish (Guiler 1970b). The young will remain attached to the teat for ~100 days. Female age appears to significantly affect female reproductive output.

Guiler (1970b) found correlations between litter size, mass and age, with younger, lighter animals having larger litters than older, larger females. Juveniles stay with the mother for approximately nine months, ensuring that they are under maternal care during the harsh

Tasmanian winter (Pemberton, 1990). The mother carries the young around in her pouch for the first three to four months. When the young are too large for the pouch but need to be transported, they ride on her back and hang on with their teeth, although they may be left alone in the natal den when the mother needs to feed (Pemberton, 1990). Dispersal of the young devils occurs at around nine months of age (usually between December and

February), and is male biased, with females often staying close to the natal site (Lachish et al. 2011). In the wild, the majority of pouch young survive to dispersal, although dispersing young suffer from high mortality rates (Guiler 1970a).

Devil offspring sex ratios are usually at parity, however in DFTD affected areas it has been found to be female biased (Lachish et al. 2009). In one example of why offspring are female biased, Trivers and Willard (1973), suggest that it is more energy intensive for females with compromised health to raise strong, healthy males

Faculty of Science The University of Sydney

Although the Tasmanian devil is an apex predator, they frequently feed on dead and dying animals, including road killed animals (both native and introduced), and weakened domestic stock (Guiler 1970a; Guiler 1970c). This scavenging behaviour results in carcass removal which reduces the risk of disease emerging from decaying animals.

The importance of devils in the Tasmanian ecosystem has been shown in recent studies where the presence of devils significantly restricted the distribution of introduced invasive predators such as feral cats (Hollings et al. 2014; Brüniche-Olsen et al. 2014; Hollings et al.

2016).

Although inter‐devil aggression (e.g. biting) has been observed during feeding, the majority of agonistic behaviour occurs during the breeding season with males and females frequently biting each other while mating (Hamede et al. 2009). Trapping studies show that up to 30% of devils carry injuries, either open wounds or healed scars with the majority of injuries occurring around the mouth (48%) or the rump (29%) (Pemberton and Renouf

1993; Hamede et al. 2008).

1.2 Devil Facial Tumour Disease (DFTD)

Devil Facial Tumour Disease (DFTD) is a fatal, transmissible cancer first observed in

Tasmanian devils in 1996 at Mount William National Park, N.E. Tasmania (Hawkins et al.

2006; Figure 1.1). DFTD is spread as an allograft amongst devils by biting (Pearse and

Swift 2006). A healthy animal is inoculated by DFTD when it bites into a tumour of an infected devil during an interaction such as mating or fighting over food (Hamede et al.

2013). Tumours often emerge in or around the mouth and death results within six to nine

Faculty of Science The University of Sydney months of the emergence of the first lesions due to metabolic failure, tumour related weakness and metastases (Pearse and Swift 2006; Pyecroft et al. 2007; Woods et al. 2015).

DFTD affects both males and females with equal frequency but is not vertically transferred from mother to offspring (Woods et al. 2015). Remarkably, DFTD often does not inhibit female reproduction. Similarly, males also show no reproductive inhibition and have comparable testosterone levels, sperm motility and testicular weights to that recorded in

DFTD free males (Keeley et al. 2012).

Figure 1.1 The distribution of DFTD can be seen to progress in an east-west direction over time in Tasmania (taken from Lazenby et al. 2018).

Faculty of Science The University of Sydney

Transmissible cancers appear to be rare. Few reports exist: apart from DFTD in devils, contagious cancers have been reported in dogs (Canis lupus familiaris) (Canine

Transmissible Venereal Tumour; CTVT; Murgia et al. 2006) in soft shell clams (Mya arenaria; Metzger et al. 2015) and Syrian hamsters (Mesocricetus auratus; Ostrander et al.

2016). In contrast to DFTD, CTVT is a non-fatal sexually transmitted cancer which has persisted without gross host mortality from 6,000 - 11,000 years ago (Murgia et al. 2006,

Murchison et al. 2014, Ostrander et al. 2016, Decker et al. 2015), while DFTD has decimated the wild Tasmanian devil population in only a few decades (Siddle and

Kaufman, 2015). However, both DFTD and CTVT may have emerged and proliferated in devils and dogs subjected to low genetic diversity (Belov 2011). Cytogenetic analyses show that DFTD originated from Schwann cells or Schwann cell precursors (Loh et al. 2006,

Murchison et al. 2010). DFTD tumour cells have a highly rearranged genome characterised by tumour-specific complex translocations and chromosomal rearrangements (Pearse and

Swift, 2006, Pearse et al. 2012, Deakin et al. 2012). The clonal nature of DFTD tumours has been supported by both large-scale genomic (Miller et al. 2011, Murchison et al. 2012), immunohistological (Loh et al. 2006) and genetic analyses (Siddle et al. 2007, Siddle et al.

2010).

Recently, a novel second variant of DFTD, described as DFT2, has been discovered with the original variant known as DFTD (Pye et al. 2016). DFT2 was found in five devils in southern Tasmania (Pye et al. 2016). As both DFTD and DFT2 result in similar tumours,

DFT2 could have evaded recognition and may therefore also have occurred unnoticed in other areas of Tasmania. The two DFTD variants differ by the presence of fragments of an

X chromosome in DFTD and remnants of a Y chromosome in DFT2, which suggests that

DFTD originated in a female, and DFT2 in a male devil (Pye et al. 2016). The recent

Faculty of Science The University of Sydney discovery of this novel variant of DFTD shows that this transmissible cancer has evolved at least twice in the devils.

Genetic diversity within the devil population is very low at both neutral markers

(microsatellite) as well as at crucial genes involved in immune function such as the Major

Histocompatibility Complex (MHC) genes (Jones et al. 2004; Woods et al. 2007; Siddle et al. 2010). Recent sequencing of the Tasmanian devil genome further demonstrates that devils possess limited genetic diversity (Miller et al. 2011; Hendricks et al. 2017). Despite their susceptibility to DFTD, Tasmanian devils have functional immune systems as they are able to produce antibodies in response to novel cellular antigens (Siddle et al. 2007; Woods et al. 2007). However, DFTD evades allorecognition as MHC Class I molecules are not expressed on the tumour cells’ surface (Siddle et al. 2010). The combination of the tumours’ ability to down regulate MHC expression and the low genetic diversity within the devil MHC have been proposed to be the two major components of the devil’s susceptibility to DFTD (Siddle et al. 2010; Woods et al. 2015).

Devils are showing evolutionary responses to the selective pressures of DFTD. Epstein et al. (2016) discovered loci under selection in genomic regions associated with immune defence and cancer risk (in humans) in devil populations that have been exposed to DFTD for > 20 years. Recently, a small number of devils within DFTD affected populations have shown tumour regression or complete recovery. Preliminary analysis has identified genes which may be involved in tumour suppression allowing the devils time to mount an immune response to DFTD (Pye et al. 2016; Wright et al. 2017). While the evidence so far is limited (with four animals showing tumour regression; Pye et al. 2016), these findings suggest that over time the devils may evolve (as did dogs afflicted with CTVT), and reach

Faculty of Science The University of Sydney an evolutionary stable strategy with DFTD akin to host-parasite systems (Hendricks et al.

2017; Russell et al. 2018).

Faculty of Science The University of Sydney

1.3 The impact of DFTD on the Tasmanian ecosystem

The Tasmanian Devil, being both a predator and a scavenger, is an integral keystone species that significantly contributes to the Tasmanian ecosystem by removing dead and dying animals from the landscape.There is legitimate concern that, if devils are lost from the wilds of Tasmania, feral cat numbers will rapidly increase with concomitant negative effects on other native mammals. Feral cat numbers have been shown to increase in areas where DFTD has depleted devil populations (Hollings et al. 2014), probably due to reduced intra-specific direct or indirect competition. This study also showed that in areas with severe DFTD induced reductions in devil numbers, there were significant declines in eastern quoll (Dasyurus viverrinus) numbers, most likely due to increased predation from feral cats. Thus, the depletion of devil numbers poses a significant threat to the integrity of the Tasmanian ecosystem, and sustained and intensive efforts are therefore merited to ensure their survival. Ongoing research is focused on the search for a cure.

1.4 Captive breeding, a measure to save endangered species

The establishment of captive breeding programs in order to save species on the brink of extinction has become a commonly used conservation measure. Unfortunately, these programs are usually established at later stages when limited numbers of breeding pairs of a given species are remaining in the wild. Decreased population size in the wild may therefore result in concomitant reduction in genetic diversity in animals used in captive breeding programs. However, not all captive breeding and reintroduction programs are successful and Armstrong and Seddon (2008) outline strategies that may improve success rates.

Faculty of Science The University of Sydney

The Australian captive devil insurance population is the largest captive breeding program ever established in this country (Hogg 2016). Forty nine wild sourced devils were originally collected in 2005 from areas at least 50 km from the DFTD disease front and held in quarantine in Tasmania for up to a year, with an additional 79 wild devils captured and added to the insurance population in 2008 (Hogg et al. 2016). Only newly weaned sub- adults with no signs of recent or healed penetrating injuries were used as founders (Jones et al. 2007). However, as reproductive success amongst captive devils remains suboptimal

(CBSG 2008; Hibbard 2010), with only 30 to 40% of females producing litters, it is crucial to investigate whether factors such as housing, stress or genetics may affect the breeding success of the devils in captivity. It is an aim of my thesis to understand why breeding success is not as successful as anticipated.

1.5 Captive breeding and genetic diversity

In order to maintain genetic diversity of captive populations, it is essential to avoid mating between closely related individuals. This is accomplished using studbooks which record all relevant life history information on every individual animal within a captive program including birth origin or place of capture, all births, parentage, transfers between facilities, deaths, and mean kinship along with unique identifying codes for each animal. However studbook data are often incomplete regarding the ancestry of founders (Ivy et al. 2009;

Alcaide et al. 2010). It would be beneficial to determine the relatedness of unknown founders wherever possible to minimise the risk of inbreeding (Glatston 2001), as in the past, in some captive programs, founders with unknown ancestry have been prevented from breeding (Willis 1993).

Faculty of Science The University of Sydney

Studbooks are used to calculate mean kinship (MK) and lower MK scores suggest higher genetic divergence. However, one of the crucial assumptions underlying the MK premise is that the founders are unrelated and that the pedigree of all animals is known. If this is not the case, MK analyses will be biased and will not accurately reflect the genetic diversity of the captive animals (Rudnick and Lacy 2008; Ballou et al. 2010). When founders are recruited from genetically depauperate, small populations within a restricted geographic area, as is the case with the Tasmanian devils, then assumptions that founders are unrelated are not realistic and can become problematic (Hogg et al. 2016). For example, Jones et al.

(2002) showed the risk of assuming no relationship amongst founders in populations of

Whooping Cranes (Grus americana). The studbook listed 88 founders however in 1941 a genetic bottleneck reduced the wild population to 13 individual cranes, thus highlighting the fact that assumptions made regarding the (un)relatedness of founders may often be unfulfilled. Using microsatellite markers, Ivy et al. (2009) were able to resolve the unknown ancestry for seven out of nine parma wallabies (Macropus parma), and discovered studbook errors concerning dam parentage.

The recent development of single nucleotide polymorphisms (SNPs) methodology makes these types of genetic markers a new and potentially powerful tool for genotyping devils.

1.6 Thesis chapters

1.6.1 Animal Husbandry and captive breeding in confined enclosures – chapter 2

Captive breeding of endangered species is a complex process and many species such as giant pandas (Ailuropoda melanoleuca), have less than optimal breeding success (Owen et al. 2004). Reproductive failure amongst captive animals can be multifactorial with causes ranging from stress (Wingfield and Sapolsky, 2003), age, reproductive behaviour, (Penfold

Faculty of Science The University of Sydney et al. 2014; Wachter et al. 2011), lack of, or inadequate breeding recommendations, inappropriate social structure, inadequate or incompatible mate choice opportunities, to inadequate enclosure design and size (Peel et al. 2005).

Tasmanian devils have been held in captivity for well over a century (Roberts 1915; Fleay

1935; Kelly 1993). In contrast to the present captive breeding program, the reason for keeping devils in captivity in the past was not focused on optimising captive devil reproductive success (Roberts 1915; Fleay 1935; Kelly 1993). The present captive

Tasmanian Devil Insurance Population (TDIP) was developed as a pre-emptive response to rapid declines in the wild population due to the effects of DFTD (Hogg et al. 2016).

Despite the devil’s long history of captive management, the breeding success of the insurance population remains suboptimal according to the recommendations and targets set out in the original population habitat and viability analysis (PHVA) report (CBSG, 2008) and by the Zoo and Aquarium Association (ZAA) Tasmanian devil management team

(Hibbard 2010).

The overall aim of this study was to evaluate potential factors associated with reproductive success in the Tasmanian devil captive breeding program. Historic records for the first seven years (2006 – 2012) of the captive breeding program were examined to elucidate any change in breeding rates and litter size over time. Differences in breeding success between wild caught founder animals and subsequent captive born descendants were examined along with the effects of enclosure design, size and number of devils per facility, to evaluate trends associated with housing structure or density in relation to pairing success.

Information provided by this study will contribute to the optimisation of the Tasmanian devil captive management and breeding programs for the future.

Faculty of Science The University of Sydney

1.7 Devil reproduction

1.7.1 Polyandry in a wild devil population – chapter 3

Multiple paternity has been recorded in litters in many vertebrates, including fish (e.g. Liu et al. 2013), amphibians (e.g. Hudson and Fu 2013), reptiles (e.g. Olsson et al. 1996), birds

(e.g. Griffith et al. 2002) and mammals (both eutherians; e.g. Thonhauser et al. 2013 and marsupials: e.g. Shimmin et al. 2000; Parrott et al. 2005). There have been several adaptive explanations for polyandry proposed based on direct and indirect female benefits. Direct benefits arise when a male provisions a females with resources e.g. food, shelter, parental input (see Neff and Pitcher, 2005) and can include increased transfer of nutrients during mating to females, as has been observed in numerous insect species (Arnqvist and Nilsson

2000). Extra-pair paternity sired broods in some birds have also been shown to result in increased provisioning rates and enhanced offspring survival (Townsend et al. 2010). In some mammals, infanticide has been shown to decrease in polyandrous species where several males interact socially with the same female such as in multi-male primate groups and in rodents where several male home ranges overlap those of a female (Klemme et al.

2007; Ebensperger and Blumstein 2007). Klemme et al. (2007) suggested that female multiple mating may enhance ovulation stimulation and hence may constitute yet another direct benefit of polyandry.

Polyandry has also been shown to provide females with indirect, i.e. genetic, benefits such as the production of more genetically diverse offspring, which can enhance female fitness especially in species where the offspring remain within their natal area (McLeod and

Marshall 2009). Female multiple mating has also been shown to result in increased offspring viability (e.g. Madsen et al. 1992). Moreover, by mating with more than one male, females can exploit post-copulatory strategies such as cryptic choice of male sperm

Faculty of Science The University of Sydney to minimise the risk and/or cost of fertilisation by genetically incompatible sperm (Olsson et al. 1996; Zeh and Zeh, 1996). However, as suggested by Jennions and Petrie (2000) direct benefits alone are unlikely to underpin female choice and indirect benefits are often inextricably linked with direct benefits.

While polyandry has been observed in marsupial species including the honey possum

(Tarsipes rostratus; Wooller et al. 2000), agile antechinus (Antechinus agilis; Shimmin et al. 2000; Kraaijeveld-Smit et al. 2002), brown antechinus (Antechinus stuartii; Holleley et al. 2006), red-tail phascogale (Phascogale calura; Foster et al. 2008), feather-tailed glider

(Acrobates pygmaeus; Parrot et al. 2005) and the spotted tailed quoll (Dasyurus maculatus;

Glen et al. 2009), it has not been determined nor documented in the Tasmanian devil. Using a combination of MHC-linked and neutral DNA microsatellites, the aim of my study presented in Chapter 4 was to investigate whether polyandry and multiple paternity occurred in litters within a wild population of Tasmanian devils from the Forestier

Peninsula.

1.7.2 The major histocompatibility complex (MHC) and mate choice in captive devils – chapter 4

The major histocompatibility complex (MHC) is one of the key molecular determinants of mate choice in numerous vertebrates (Kamiya et al. 2014). This genomic region encompasses the most variable set of genes, with up to 349 alleles described for a single locus (Robinson et al. 2000) and heterozygosity values exceeding those predicted by neutrality (Edwards and Hedrick 1998). Class I and Class II MHC molecules are responsible for the presentation of intra- and extra-cellular peptides to cytotoxic T cells and constitute a crucial part of the vertebrate immune system (Zinkernagel 1979; Klein 1986).

Faculty of Science The University of Sydney

Factors contributing to mate choice are extremely varied, with pre-copulatory (e.g. visual displays, body ornamentation, body size, behaviour), post-copulatory (e.g. sperm competition and cryptic female choice) and genetic mechanisms influencing reproductive success. Whilst the MHC is primarily involved in immune system function, its role and importance in mate choice has been documented in numerous studies (see Ziegler et al.

2010). MHC influences an organism’s individual odour through protein production which binds to volatile molecules (Tregenza and Wedell 2000) occurring in sweat, urine and gut microflora (Penn and Potts 1998). As MHC genes are closely linked to olfactory receptors

(OR) their tandem role in mate choice, kin recognition and hence avoidance via olfactory cues is likely (Zeigler et al. 2010).

In mammals females are often the choosey sex – largely due to their substantial investment in both gametes and as the primary carers of their offspring (Penn and Potts 1998). In polyandrous females, mate choice can be implemented either pre or post mating, with the latter being driven by female-male genetic compatibility (Tregenza and Wedell 2000).

MHC-based mate choice has been suggested to provide offspring with indirect genetic benefits via acquisition of “good genes” i.e. genetic elements that contribute to lifetime reproductive success regardless of an individual’s additional genotype (Olsson et al. 2005), producing offspring with optimal genetic compatibility (Kempenaers 2007) and/or achieving enhanced genetic diversity within litters (Ejsmond et al. 2014). In the present study (Chapter 5) I explore how MHC polymorphism might affect reproductive success in captive devils and hypothesise that female age and MHC polymorphism will be positively correlated to reproductive success in the Tasmanian devil.

Faculty of Science The University of Sydney

1.7.3 Investigations into devil endocrinology – chapter 5

Via the hypothalamic-pituitary-adrenal axis (HPA), stress, (both acute and chronic;

Dickens et al. 2010) activates the release of glucocorticoids from the adrenal glands via the sympathetic nervous system (Bonier et al. 2009; Figure 1-2). Chronic stress occurs when stress is relentless and continuous over time whereas acute stress is in response to an immediate stressor e.g. in response to a predator and normalises over time when the immediate threat has diminished (Dickens et al. 2010). While acute stress is vital for survival, and reproduction in some species (Fanson and Parrot 2015), chronic stress can have detrimental effects on the body where prolonged secretion of glucocorticoids can affect the immune system, including suppressed reproductive behaviour and physiology

(Sapolsky et al. 2000; Dickens et al. 2010).

Many processes involved with the capture and handling, securing, confining, transporting, changes in social structure or housing and relocating of wild animals constitute individual stressors, which alone can cause acute stress, or in combination, can lead to chronic stress

(Grandin 1997). When an animal experiences a chronic stress state, its ability to mount an acute stress response in reaction to a novel stressor can be suppressed (Dickens et al. 2010).

For example, in alpacas (Llama pacos) stress has been linked with abortion, premature birth and gastric ulceration (Anderson et al. 1999). Capture and handling can cause an almost immediate (within three minute) spike in blood glucocorticoid metabolites (GCM’s or

GC’s). This may be misinterpreted in reproductive studies as an indication of reduced reproductive fitness (Sheriff et al. 2010). Measuring stress via faecal glucocorticoid analyses (FGC) is a non-invasive practice and results are not influenced by the stress to an organism caused by its capture, handling, restraint and sample collection (Rothschild 2005;

Wilkening et al. 2016; Figure 1.2). FGC analyses have been shown to produce identical results as those derived from blood GCs’ indicating that the use of FGCs’ are a reliable tool

Faculty of Science The University of Sydney for gauging animal stress (Sheriff et al. 2010). Faecal GC’s have been successfully used to quantify stress in numerous species including jaguars (Panthera onca; Morato et al. 2004), river otters (Lontra canadensis; Rothschild et al. 2005), snow-shoe hares (Lepus americanus; Sheriff et al. 2009), bank voles (Myodes glareolus; Rogovin and Naidenko

2010) and rhesus macaques (Macaca mulatta; Brent et al. 2011).

Figure 1.2 Association between stressors and overall translocation failure. Abbreviation are as follows; SNS = sympathetic nervous system, HPA = hypothalamic-pituitary- adrenal axis, GC = glucocorticoids, FF = flight or fight response (taken from Dickens et al. 2010).

Faculty of Science The University of Sydney

Starvation, disease, reduced reproductive capacity, predation and dispersal are often cited as reasons for translocation failure. Dickens et al. (2010) suggest that chronic stress may be a major contributor to these scenarios.

Jones et al. (2005) studied the plasma cortisol levels in eight wild caught Tasmanian devils under captive conditions. Females were found to have higher cortisol levels than males for the duration of the study. These authors suggested that housing these naturally solitary animals in male/female pairs may have added to the higher female cortisol levels.

Similarly, in a study on brushtail possums, Trichosurus vulpecula, transferred from the wild into captivity, Baker et al. (1998) also found that females exhibited a higher and more prolonged stress response than males. However, some individual animals recover more quickly after a stressful event. For example, transportation of alpacas resulted in increased serum cortisol levels, however, levels for both males and females were back within normal parameters after a four hour recovery period (Anderson et al. 1999). Similarly, both male and female sheep subject to shearing showed significantly elevated cortisol and haematocrit levels but returned to baseline after 90 minutes (Hargreaves and Hutson 1990).

Additionally incorporating behavioural observations along with physiological results can reveal a more accurate assessment of animal stress (Grandin 1997).

In captive species sample collection for FGC measurement can be done as part of the normal routine husbandry (Stead-Richardson et al. 2010). Importantly, this technique has been used successfully to monitor progesterone and GC metabolites in order to assess reproductive cycles and stress in the common wombat (Vombatus ursinus) (Paris et al.

2002) and the greater bilby (Macrotis lagotis; Narayan et al. 2012).

Faculty of Science The University of Sydney

To ensure successful captive devil reproduction, it is also imperative that males be paired with females at the correct time in the reproductive cycle as signalled by phenotypic changes in the female such as the appearance of a fluid filled neck roll, nesting behaviour, retreating from keepers and conspecifics if communally housed and overall lethargy

(Hockley, 2011).

To further explore mechanisms relating to captive devil reproductive success, investigations into the timing of devil pairings using hormone metabolites and possible effects of corticosterone on the reproductive success of both male and female devils are presented and discussed in Chapter 5.

Faculty of Science The University of Sydney

1.8 References

Alcaide, A., Negro, J. J., Serrano, D., Antolín J. L., Casado, S. and Pomarol, M. (2010)

Captive breeding and reintroduction of the lesser kestrel Falco naumanni: a genetic analysis using microsatellites. Conservation Genetics 11, 331-338.

Anderson, D. E., Grubb, T. and Silveira, F. (1999) The effect of short duration transport on serum cortisol response in alpacas (Llama pacos). The Veterinary Journal 157, 189-191.

Armstrong, P. and Seddon, P. J. (2008) Directions in reintroduction biology. Trends in

Ecology and Evolution 23 (1), 20-25.

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

Ballou, J. D., Lees, C., Faust, L. J., Long, S., Lynch, C., Lackey, L. B. and Foose, T. J.

(2010) Demographic and genetic management of captive populations. In Wild Mammals in captivity. (Eds.) D. G. Kleiman, K.V. Thompson and C. K. Bauer. University of Chicago,

Chicago.

Baker, M. L., Gemmel, E. and Gemmel, R. T. (1998) Physiological changes in brushtail possums, Trichosurus vulpecula, transferred from the wild to captivity. The Journal of

Experimental Zoology 280, 203-212.

Belov, K. (2011) The role of the Major Histocompatibility Complex in the spread of contagious cancers. Mammalian Genome 22, 83-90.

Faculty of Science The University of Sydney

Bonier, F., Martin, P. R., Mooore, I. T. and Wingfield, J. C. (2009) Do baseline glucocorticoids predict fitness? Trends in Ecology and Evolution 24 (11), 634-642.

Brent, L. J. N., Semple, S., Dubuc, C., Heistermann, M. and MacLarnon, A. (2011) Social capital and physiological stress levels in free-ranging adult female rhesus macaques.

Physiology and Behavior 102, 76-83.

Brown, O. J. F. (2006) Tasmanian devil (Sarcophilus harrisii) extinction on the Australian mainland in the mid-Holocene: multicausality and ENSO intensification. Alcheringa 30,

49-57.

Brüniche-Olsen, A., Jones, M. E., Austin, J. J., Burridge, C. P. and Holland, B. R. (2014)

Extensive population decline in the Tasmanian devil predates European settlement and devil facial tumour disease. Biology Letters 10 (11), article 20140619.

CBSG (2008) Tasmanian devil population and habitat viability assessment final report.

IUCN/SSC Conservation Breeding Specialist Group, Apple Valley, MN.

Deakin, J. E., Bender, H. S., Pearse, A.-M.,Rens, W., O’Brien, P. C. M., , Ferguson-Smith,

M. A., Cheng, Y., Morris, K.,Taylor, R., Stuart, A., Belov, K., Amemiya, C. T.,Murchison,

E. P., Papenfuss, A. T., and Marshall Graves, J. A. (2012) Genomic restructuring in the

Tasmanian devil facial tumour: chromosome painting and gene mapping provide clues to evolution of a transmissible tumour. PLoS genetics 8, e1002483.

Faculty of Science The University of Sydney

Decker, B., Davis, B. W., Rimbault, M., Long, A. H., Karlins, E., Jagannathan, V., Reiman,

R., Parker, H. G., Drögemüller, C., Corneveaux, J. J., Chapman, E. S., Trent, J. M., Leeb,

T., Huentelman, M. J., Wayne, R. K., Karyadi, D. M. and Ostrander, E. A. (2015)

Comparison against 186 canid whole-genome sequences reveals survival strategies of an ancient clonally transmissible canine tumor. Genome Research 25, 1646-1655.

Dickens, M. J., Delehanty, D. J and Romero, L. M. (2010) Stress: an inevitable component of animal translocation. Biological Conservation 143, 1329-1341.

Ebensperger, L. A. and Blumstein, D. T. (2007) "Nonparental infanticide." in Rodent societies: an ecological & evolutionary perspective. (p. 267) J. O. Wolff and P. W.

Sherman (eds.). University of Chicago Press, Chicago.

Edwards, S. V. and Hedrick, P. W. (1998) Evolution and ecology of MHC molecules: From genomics to sexual selection. Trends in Ecology and Evolution 13 (8), 305-311.

Ejsmond, M. J., Radwan, J. and Wilson, A. B. (2014) Sexual selection and the evolutionary dynamics of the major histocompatibility complex. Proceedings of the Royal Society B,

281, 20141662 (doi:101098/rspb20141662).

Epstein, B., Jones, M., Hamede, R., Hendricks, S., McCallum, H., Murchison, E. P.,

Schönfeld, B., Wiench, C., Hohenlohe, P. and Storfer, A. (2016) Rapid evolutionary response to a transmissible cancer in Tasmanian devils. Nature Communications 7, 12684

(doi: 10.1038/ncomms12684).

Faculty of Science The University of Sydney

Fanson, K. V. and Parrott, M. L. (2015) The value of eutherian–marsupial comparisons for understanding the function of glucocorticoids in female mammal reproduction. Hormones and Behaviour 76, 41-47.

Fleay, D. (1935) Notes of the breeding of Tasmanian devils. Victorian Naturalist 52, 100-

105.

Flynn, T. T. (1922) XXVI.—Notes on certain reproductive phenomena in some Tasmanian

Marsupials. Annals and Magazine of Natural History: Series 9, 225-231.

Foster, W. K., Caton, W., Thomas, J., Cox, S. and Taggart, D. A. (2008) Timing of births and reproductive success in captive red-tailed phascogales, Phascogale calura. Journal of

Mammalogy 89 (5), 1136-1144.

Frankham, R. (2008) Genetic adaptation to captivity in species conservation programs.

Molecular Ecology 17 (1), 325-333.

Glatston, A. R. (2001) Relevance of studbook data to the successful captive management of grey mouse lemurs. International Journal of Primatology 22 (1), 57-69.

Glen, A. S., Cardoso, M. J., Dickman, C. R. and Firestone, K. B. (2009) Who’s your daddy? Paternity testing reveals promiscuity and multiple paternity in the carnivorous marsupial Dasyurus maculatus (Marsupialia: Dasyuridae). Biological Journal of the

Linnean Society 96, 1-7.

Faculty of Science The University of Sydney

Grandin, T. (1997) Assessment of stress during handling and transport. Journal of Animal

Science 75, 249-257.

Griffith, S. C., Owens, I. P. F. and Thuman, K. A. (2002) Extra pair paternity in birds: a review of interspecific variation and adaptive function. Molecular Ecology 11 (11), 2195-

2212.

Guiler, E. R. (1970a) Observations of the Tasmanian devil, Sarcophilus harrisii

(Marsupialia: Dasyuridae). I. Numbers, home range, movements, and food in two populations. Australian Journal of Zoology 18, 49-62.

Guiler, E. R. (1970b) Observations on the Tasmanian devil, Sarcophilus harrisii

(Marsupialia:Dasyuridae) II Reproduction, breeding, and growth of pouch young.

Australian Journal of Zoology 18, 63-70.

Guiler, E. R. (1970c) Tasmanian devils and agriculture. Tasmanian Journal of Agriculture

41 (2), 134-137.

Hamede, R. K., McCallum, H. and Jones, M. (2008) Seasonal, demographic and density- related patterns of contact between Tasmanian devils (Sarcophilus harrisii): Implications for transmission of devil facial tumour disease. Austral Ecology 33, 614-622.

Hamede, R. K., Bashford, J., McCallum, H. and Jones, M. (2009) Contact networks in a wild Tasmanian devil (Sacrophilus harrisii) population: using social network analysis to reveal seasonal variability in social behaviour and its implications for transmission of devil facial tumour disease. Ecology Letters 12, 1147-1157.

Faculty of Science The University of Sydney

Hamede, R. K., McCallum, H. and Jones, M. (2013) Biting injuries and transmission of

Tasmanian devil facial tumour disease. Journal of Animal Ecology 82, 182-190.

Hargreaves, A. L. and Hutson, G. D. (1990) The stress response in sheep during routine handling procedures. Applied Animal Behaviour Science 26, 83-90.

Hartmann-Furter, M. (2009). Breeding European wildcats (Felis silvestris silvestris,

Schreber 1777) in Species-specific enclosures for reintroduction in Germany. in Iberian lynx Ex situ conservation: an interdisciplinary approach. (p. 452).

Hawkins, C. E., Baars, C., Hesterman, H., Hocking, G. J., Jones, M. E., Lazenby, B.,

Mann, D., Mooney, N., Pemberton, D., Pyecroft, S., Restani, M. and Wiersma, J. (2006)

Emerging disease and population decline of an island endemic, the Tasmanian devil

Sarcophilus harrisii. Biological Conservation 131, 307-324.

Hendricks, S., Epstein, B., Schönfeld, B., Wiench, C., Hamede, R., Jones, M., Storfer, A. and Hohenlohe, P. (2017) Conservation implications of limited genetic diversity and population structure in Tasmanian devils (Sarcophilus harrisii). Conservation Genetics, 18

(4), 977-982.

Hesterman, H., Jones, S. M. and Schwarzenberger, F. (2008) Reproductive endocrinology of the largest dasyurids: Characterization of ovarian cycles by plasma and fecal steroid monitoring. Part I. The Tasmanian devil (Sarcophilus harrisii). General and Comparative

Endocrinology 155, 234-244.

Faculty of Science The University of Sydney

Hibbard, C. J., Hogg, C. J., Srb, C. and Hockley, J. (2010) Annual report for the DPIPWE-

ZAA Tasmanian devil insurance population. Zoo and Aquarium Association. pp 35.

Hockley, J. (ed.) (2011) DPIPWE/ZAA. Husbandry Guidelines for Tasmanian Devils,

Sarcophilus harrisii. DPIPWE, Tasmania, Australia.

Hogan, L. A., Lisle, A. T., Johnston, S. D. and Robertson, H. (2012) Non-invasive assessment of stress in captive numbats, Myrmecobius fasciatus (Mammalia: Marsupialia), using faecal cortisol measurement. General and Comparative Endocrinology 179, 376–

383.

Hogg, C. J., Lee, A. V., Srb, C. and Hibbard, C. (2016) Metapopulation management of an endangered species with limited genetic diversity in the presence of disease: the Tasmanian devil Sarcophilus harrisii. International Zoo Yearbook 51, 1-17.

Holleley, C. E., Dickman, C. R., Crowther, M. S. and Oldroyd, B. P. (2006) Size breeds success: multiple paternity, multivariate selection and male semelparity in a small marsupial, Antechinus stuartii. Molecular Ecology 15, 3439-3448.

Hollings, T., Jones, M., Mooney, N. and McCallum, H. (2014) Trophic Cascades

Following the Disease-Induced Decline of an Apex Predator, the Tasmanian Devil.

Conservation Biology, 28, 63-75.

Hollings, T., Jones, M., Mooney, N. and McCallum, H. (2016) Disease-induced decline of an apex predator drives invasive dominated states and threatens biodiversity. Ecology, 97,

394-405.

Faculty of Science The University of Sydney

Hudson, C. M. and Fu, J. (2013) Male-biased sexual size dimorphism, resource defense polygyny and multiple paternity in the Emei Moustache Toad (Leptobrachium boringii).

PLOS One 8 (6), 1-8.

Hughes, R. L. (1982) Reproduction in the Tasmanian devil Sarcophilus harrisii

(Dasyuridae Marsupialia). In M. Archer (ed.) Carnivorous Marsupials pp 49-63. Surrey

Beatty and Sons, Sydney.

Ivy, J. A., Miller, A., Lacy, R. C. and DeWoody, J. A. (2009) Methods and prospects for using molecular data in captive breeding programs: An empirical example using parma wallabies (Macropus parma). Journal of Heredity 100 (4), 441-454.

Johnson, C.S. and Wroe, S. (2003) Causes of extinction of vertebrates during the Holocene of mainland Australia: arrival of the dingo, or human impact? The Holocene 13, 1009-

1016.

Jones, K. L., Glenn, T. C., Lacy, R. C., Pierce, J. R., Unruh, N., Mirande, C. M. and

Chavez-Ramirez, F. (2002) Refining the Whooping Crane studbook by incorporating microsatellite DNA and leg-banding analysis. Conservation Biology 16 (3), 789-799.

Jones, M. E., Paetkau, D., Geffen, E. and Moritz, C. (2003) Microsatellites for the

Tasmanian devil (Sarcophilus laniarius). Molecular Ecology Notes 3, 277-279.

Faculty of Science The University of Sydney

Jones, M. E., Paetkau, D., Geffen, E. and Moritz, C. (2004) Genetic diversity and population structure of Tasmanian devils, the largest marsupial carnivore. Molecular

Ecology 13, 2197-2209.

Jones, S. M., Lockhart, T. J. and Rose, R. W. (2005) Adaptation of wild-caught Tasmanian devils (Sarcophilus harrisii) to captivity: evidence from physical parameters and plasma cortisol concentrations. Australian Journal of Zoology 53, 339-344.

Jones, M. E., Jarman, P. J., Lees, C. M., Hesterman, H., Hamede, R. K., Mooney, N. J.,

Mann, D., Pukk, C. E., Bergfeld, J. and McCallum, H. (2007) Conservation management of

Tasmanian devils in the context of an emerging, extinction – threatening disease: Devil facial tumor disease. EcoHealth 4, 326-337.

Jones, M. E., Cockburn, A., Hamede, R., Hawkins, C., Hesterman, H., Lachish, S., Mann,

D., McCallum, H. and Pemberton, D. (2008) Life-history change in disease-ravaged

Tasmanian devil populations. Proceedings of the National Academy of Sciences of the USA

105 (29), 10023-10027.

Keeley, T., O'Brien, J. K., Fanson, B. G., Masters, K. and McGreevey, P. D. (2012) The reproductive cycle of the Tasmanian devil (Sarcophilus harrisii) and factors associated with reproductive success in captivity. General and Comparative Endocrinology, 176 (2),

182-191.

Faculty of Science The University of Sydney

Kelly, A. (1993) Management of Tasmanian devils (Sarcophilus harrisii) at the Tasmanian

Wildlife Park In The Biology and Management of Australasian Carnivorous Marsupials.

(Eds. M. Roberts, J. Carnio, G. Crawshaw and M. Hutchins) pp. 79-84. (Metropolitan

Toronto Zoo: Toronto, Canada).

Kelly, A. (2007) Husbandry guidelines for the Tasmanian devil, Sarcophilus harrisii. In S.

Brice (ed.), www.zooaquarium.org.au. Sydney: Zoo and Aquarium Association 1-32.

Kempenaers, B. (2007) Mate choice and genetic quality, a review of the heterozygosity theory. Advances in the Study of Behavior, 37, 189–278.

Klein, J. (1986) Natural history of the Major Histocompatibility Complex. Wiley, New

York.

Kraaijveld-Smit, F. J. L., Ward, S. J. and Temple-Smith, P. D. (2002) Multiple paternity in a field population of a small carnivorous marsupial, the agile antechinus, Antechinus agilis.

Behavioral Ecology and Sociobiology 52, 84-91.

Lachish, S., McCallum, H. and Jones, M. (2009) Demography, disease and the devil: life- history changes in a disease-affected population of Tasmanian devils (Sarcophilus harrisii).

Journal of Animal Ecology 78, 427-436.

Lachish, S., McCallum, H., Mann, D., Pukk, C. E. and Jones, M. E. (2010) Evaluation of selective culling of infected individuals to control Tasmanian Facial Tumor Disease.

Conservation Biology 24 (3), 841-851.

Faculty of Science The University of Sydney

Lachish, S., Miller, K. J., Storfer, A., Goldizen, A. W. and Jones, M. E. (2011) Evidence that disease-induced population decline changes genetic structure and alters dispersal patterns in the Tasmanian devil. Heredity 106, 172-182.

Lazenby, B. T., Tobler, M. W., Brown, W. E., Hawkins, C. E., Hocking, G. J., Hume, F.,

Huxtable, S., Iles, P., Jones, M. E., Lawrence, C., Thalmann, S., Wise, P., Williams, H.,

Fox, S. and Pemberton, D. (2018) Density trends and demographic signals uncover the long-term impact of transmissible cancer in Tasmanian devils. Journal of Applied Ecology

55: (3), 1368-1379.

Liu, X., Tatarenkov, A., O’Rear, T. A., Moyle, P. B. and Avise, J. C. (2013). Molecular evidence for multiple paternity in a population of the viviparous Tule Perch Hysterocarpus traski. Journal of Heredity 104 (2), 217-222.

Loh, R., Hayes, D., Mahjoor, A., O'Hara, A., Pyecroft, S. and Raidal, S. (2006) The immunohistochemical characterization of devil facial tumor disease (DFTD) in the

Tasmanian Devil (Sarcophilus harrisii). Veterinary Pathology 43, 896-903.

McCallum, H., Tompkins, D. M., Jones, M., Lachish, S., Marvanek, S., Lazenby, B.,

Hocking, G., Wiersma, J. and Hawkins, C. (2007) Distribution and impacts of Tasmanian devil facial tumor disease. EcoHealth 4, 318-325.

McCallum, H., Jones, M., Hawkins, C., Hamede, R., Lachish, S., Sinn, D. L., Beeton, N. and Lazenby, B. (2009) Transmission dynamics of Tasmanian devil facial tumor disease may lead to disease-induced extinction. Ecology, 90 (12), 3379-3392.

Faculty of Science The University of Sydney

McLeod, L. and Marshall, D. J. (2009) Do genetic diversity effects drive the benefits associated with multiple mating? A test in a marine invertebrate. PloS ONE 4 (8) e 6347.

Madsen, T., Shire, R., Loman, J. and Håkansson, T. (1992) Why do female adders copulate so frequently? Nature 355, 440-441.

Metzger, M. J., Reinisch, C., Sherry, J. and Goff, S. P. (2015) Horizontal transmission of clonal cancer cells causes leukemia in soft-shell clams. Cell 161, 255–263.

Miller, W., Hayes, V. M., Ratan, A., Petersen, D. C., Wittekindt, N. E., Miller, J., Walenz,

B., Knight, J., Qi, J., Zhao, F., Wang, Q., Bedoya-Reina, O. C., Katiyar, N., Tomsho, L.

P., Kasson, L. M., Hardie, R. A., Woodbridge, P., Tindall, E. A., Bertelsen, M. F., Dixon,

D., Pyecroft, S., Helgen, K. M., Lesk, A. M.,Pringle, T. H., Patterson, N., Zhang, Y.,

Kreiss, A., Woods, G. M., Jones, M. E. and Schuster, S. C. (2011) Genetic diversity and population structure of the endangered marsupial Sarcophilus harrisii (Tasmanian devil).

Proceedings of the National Academy of Sciences 198,12348-12353.

Morato, R. G., Bueno, M. G., Malmheister, P., Vereschi, I. T.N. and Barnabe, R. C. (2004)

Changes in the fecal concentrations of cortisol and androgen metabolites in captive male

Jaguars (Panthera onca) in response to stress. Brazilian Journal of Medical and Biological

Research 37(12), 1903-1907.

Murchison, E. P., Tovar, C., Hsu, A., Bender, H. S., Kheradpour, P., Rebbeck, C. A.,

Obendorf, D., Conlan, C., Bahlo, M., Blizzard, C. A., Pyecroft, S., Kreiss, A., Kellis, M.,

Stark, A., Harkins, T. T., Graves, J. A. M., Woods, G. M., Hannon, G. J. and Papenfuss,

Faculty of Science The University of Sydney

A. T. (2010) The Tasmanian devil transcriptome reveals schwann cell origins of a clonally transmissible cancer. Science 327, 84-87.

Murchison, E. P., Schulz-Trieglaff, O. B., Ning, Z., Alexandrov, L. B., Bauer, M. J., Fu, B.,

Hims, M., Ding, Z., Ivakhno, S., Stewart, C., Ng, B. L., Wong, W., Aken, B., White, S.,

Alsop, A., Becq, J., Bignell, G. R., Cheetham, R. K., Cheng, W., Connor, T. R.,Cox, A. J.,

Feng, Z-P., Gu, Y., Grocock, R. J., Harris, S. R., Khrebtukova, I., Kingsbury, Z.,

Kowarsky, M., Kreiss, A., Luo, S., Marshall, J., McBride, D. J., Murray, L., Pearse, A-M.,

Raine, K., Rasolonjatovo, I., Shaw, R., Tedder, P., Tregidgo, C., Vilella, A. J., Wedge, D.

C., Woods, G. M., Gormley, N., Humphray, S., Schroth, G., Smith, G., Hall, K., Searle, S.

M. J., Carter, N. P., Papenfuss, A. T., Futreal, P. A., Campbell, P. J., Yang, F., Bentley, D.

R., Evers, D. J. and Stratton, M. R. (2012) Genome sequencing and analysis of the

Tasmanian devil and its transmissible cancer. Cell 148, 780-791.

Murchison, E. P., Wedge, D. C. Alexandrov, L. B., Fu, B., Martincorena, I., Ning, Z.,

Tubio, J. M. C., Werner, E. I., Allen, J., De Nardi, A. B., Donelan, E. M, Marino, G.,

Fassati, A., Campbell, P. J., Yang, F., Burt, A., Weiss, R. A., and Stratton, M. R. (2014)

Transmissible dog cancer genome reveals the origin and history of an ancient cell lineage.

Science 343, 437-440.

Murgia, C., Pritchard, J. K., Kim, S. Y., Fassati, A. and Weiss, R. A. (2006) Clonal origin and evolution of a transmissible cancer. Cell 126, 477-487.

Narayan, E., Hero J-M., Evans, N., Nicolson, V. and Mucci, A. (2012) Non-invasive evaluation of physiological stress hormone responses in a captive population of the greater bilby Macrotis lagotis. Endangered Species Research 18, 279-289.

Faculty of Science The University of Sydney

Neff, B. D. and Pitcher, T. E. (2005) Genetic quality and sexual selection: an integrated framework for good genes and compatible genes. Molecular Ecology 14, 19-38.

Olsson, M., Madsen, T., Wapstra, E., Silverin, B., Ujvari, B., and Wittzell, H. (2005) MHC, health, color, and reproductive success in sand lizards. Behavioural Ecology Sociobiology,

58, 289–294.

Ostrander, E. A., Davis, B. W. and Ostrander, G. K. (2016) Transmissible tumors: breaking the cancer paradigm. Trends in Genetics 32, 1-15.

Owen, D. and Pemberton, D. (2005) Tasmanian devil: a unique and threatened animal.

Allen & Unwin, Crows Nest, Sydney.

Paris, M. C. J., White, A., Reiss, A., West, M. and Schwarzenberger, F. (2002) Faecal progesterone metabolites and behavioural observations for the non-invasive assessment of oestrous cycles in the common wombat (Vombatus ursinus) and the southern hairy-nosed wombat (Lasiorhinus latifrons). Animal Reproduction Science 72, 245-257.

Parrot, M. L., Ward, S. J. and Taggart, D. A. (2005) Multiple paternity and communal maternal care in the feathertail glider (Acrobates pygmaeus). Australian Journal of Zoology

53, 79-85.

Pearse, A. M. and Swift, K. (2006) Transmission of devil facial-tumour disease. Nature

439, 549.

Faculty of Science The University of Sydney

Pearse, A.-M., Swift, K., Hodson, P., Hua, B., McCallum, H., Pyecroft, S., Taylor, R.,

Eldridge, M. D. B. and Belov, K. (2012) Evolution in a transmissible cancer: a study of the chromosomal changes in devil facial tumor (DFT) as it spreads through the wild Tasmanian devil population. Cancer Genetics 205, 101-112.

Peel, A. J., Vogelnest, L., Finnigan, M., Grossfeldt, L. and O’Brien, J. K. (2005) Non-

Invasive faecal hormone analysis and behavioral observations for monitoring stress responses in captive western lowland gorillas (Gorilla gorilla gorilla). Zoo Biology 24,

431-445.

Pemberton, D. and Renouf, D. (1993) A field study of communication and social behaviour of the Tasmanian devil at feeding sites. Australian Journal of Zoology 41, 507-526.

Penfold, L. M., Powell, D., Traylor - Holzer, K. and Asa, C. S. (2014) “Use it or lose it”:

Characerization, implication, and mitigation of female infertility in captive wildlife. Zoo

Biology 33, 20-28.

Penn, D. and Potts, W. (1998) How do Major Histocompatibility Complex genes influence odor and mating preferences? Advances in Immunology 69, 411-436.

Pye, R. J., Pemberton, D., Tovar, C., Tubioc, J., Dund, J. M. C., Dund, K. A., Fox, S.,

Darbya, J, Hayes, D., Knowles, G. W., Kreiss, A., Siddle, H. V. T., Swift, K., Lyons, A. B.,

Murchison, E. P. & Woods, G. M. (2016) A second transmissible cancer in Tasmanian devils. Proceedings of the Natural Academy of Science 113, 374-379.

Faculty of Science The University of Sydney

Pyecroft, S. B., Pearse, A., Loh, R., Swift, K., Belov, K., Fox, N., Noonan, E., Hayes, D.,

Hyatt, A., Wang, L., Boyle, D., Church, J., Middleton, D. and Moore, R. (2007) Towards a case definition for Devil Facial Tumour Disease: What is it? EcoHealth 4, 346-351.

Roberts, M.G. (1915) The keeping and breeding of Tasmanian devils (Sarcophilus harrisii). Proceedings of the Zoological Society of London 85, 575-581.

Robinson, J., Malik, A., Parham, P., Bodmer, J. G. and Marsh, S. G. E. (2000) IMGT/HLA

Database - a sequence database for the human major histocompatibility complex. Tissue

Antigens 55 (3), 280-287.

Rogovin, K. A. and Naidenko, S. V. (2010) Noninvasive Assessment of Stress in Bank

Voles (Myodes glareolus, Cricetidae, Rodentia) by Means of Enzyme Linked

Immunosorbent Assay (ELISA). Biology Bulletin 37(9), 959-964.

Rothschild, D. M., Serfass, T. L., Seddon, W. L., Hedge, L. and Fritz R. S. (2005) Using fecal glucocorticoids to assess stress levels in captive river otters. The Journal of Wildlife

Management. 72 (1), 138-142.

Rudnick, J. A. and Lacy, R. C. (2008) The impact of assumptions about founder relationships on the effectiveness of captive breeding strategies. Conservation Genetics 9,

1439-1450.

Russell, T., Madsen, T., Thomas, F., Raven, N., Hamede, R. and Ujvari, B. (2018)

Adaptive evolution in the face of a transmissible cancer. BioEssays 1700146.

Faculty of Science The University of Sydney

Sapolsky R. M., Romero, L. M. and Munck, A. U. (2000) How do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocrine Reviews 21(1), 55-89.

Schwarzenberger, F., Möstl, Palme, R. and Bamberg, E. (1996) Faecal steroid analysis for non-invasive monitoring of reproductive status in farm, wild and zoo animals. Animal

Reproduction Science 42, 515-526.

Sheriff, M. J., Curtis, V., Bosson, O., Krebs, C. J. and Boonstra, R. (2009) A non-invasive technique for analyzing fecal cortisol metabolites in snowshoe hares (Lepus americanus).

Journal of Comparative Physiology B-Biochemical Systemic and Environmental

Physiology 179, 305-313.

Sheriff, M. J., Krebs, C. J. and Boonstra, R. (2010) Assessing stress in animal populations:

Do fecal and plasma glucocorticoids tell the same story? General and Comparative

Endocrinology 166, 614–619.

Shimmin, G. A., Taggart, D. A. and Temple-Smith, P. D. (2000) Sperm competition and genetic diversity in the agile antechinus (Dasyuridae: Antechinus agilis). Journal of

Zoology 252, 343-350.

Siddle, H. V., Kreiss, A., Eldridge, M. D. B., Noonan, E., Clarke, C. J., Pyecroft, S.,

Woods, G. M. and Belov, K. (2007) Transmission of a fatal clonal tumor by biting occurs due to depleted MHC diversity in a threatened carnivorous marsupial. Proceedings of the

National Academy of Sciences of the USA 104(41), 16221-16226.

Faculty of Science The University of Sydney

Siddle, H. V., Marzec, J., Cheng, Y., Jones, M. and Belov, K. (2010) MHC gene copy number variation in Tasmanian devils: Implications for the spread of a contagious cancer.

Proceedings of the Royal Society B 277, 2001-2006.

Siddle, H. and Kaufman, J. (2015) Immunology of naturally transmissible tumours.

Immunology 144 (1), 11-20.

Souris, A-C., Kaczensky, P. and Julliard, R. (2007) Time budget, behavioral synchrony and body score development of a newly released Przewalski’s horse group Equus ferus przewalskii, in the Great Gobi B strictly protected area in SW Mongolia. Applied Animal

Behaviour Science 107, 307-321.

Stead-Richardson, E., Bradshaw, D., Friend, T. and Fletcher, T. (2010) Monitoring reproduction in the critically endangered marsupial, Gilbert’s potoroo (Potorous gilbertii):

Preliminary analysis of faecal oestradiol-17b, cortisol and progestogens. General and

Comparative Endocrinology 165, 155-162.

Thalmann, S., Peck, S., Wise, P., Potts, J. M., Clarke, J. and Richley, J. (2016)

Translocation of a top-order carnivore: tracking the initial survival, spatial movement, home-range establishment and habitat use of the Tasmanian devils on Maria Island.

Australian Mammalogy 38, 68-79.

Thonhauser, K. E., Thoβ, Musolf, K., Klaus, T. and Penn, D. J. (2014) Multiple paternity in wild house mice (Mus musculus musculus): effects on offspring genetic diversity and body mass. Ecology and Evolution 4 (2), 200-209.

Faculty of Science The University of Sydney

Townsend, A. K., Clark, A. B. and McGowan, K. J. (2010) Direct Benefits and Genetic

Costs of Extrapair Paternity for Female American Crows (Corvus brachyrhynchos). The

American Naturalist 175 (1), 1-9.

Tregenza, T. and Wedell, N. (2000) Genetic compatibility, mate choice and patterns of parentage: Invited Review. Molecular Ecology 9, 1013-1027.

Trivers, R.L. and Willard, D.E. (1973) Natural selection of parental ability to vary sex-ratio of offspring. Science 179, 90–92.

Ujvari, B., Hamede, R., Peck, S., Pemberton, D., Jones, M., Belov, K. and Madsen, T.

(2016) Immunoglobulin dynamics and cancer prevalence in Tasmanian devils (Sarcophilus harrissii). Scientific Reports 6, 25093 DOI: 10.1038/srep25093.

Wachter, B., Thalwitzer, S., Hofer, H., Lonzer, J., Hildebrandt, T. and Hermes, R. (2011)

Reproductive history and absence of predators are important determinants of reproductive fitness: the cheetah controversy revisited. Conservation Letters 4, 47–54.

Wilkening, J. L., Ray, C. and Varners, J. (2016) When can we measure stress noninvasively? Postdeposition effects on a fecal stress metric confound a multiregional assessment. Ecology and Evolution 6 (2), 502-513.

Willis, K. (1993) Use of animals with unknown ancestries in scientifically managed breeding programs. Zoo Biology 12, 161-172.

Faculty of Science The University of Sydney

Wimberger, K., Colleen, T., Downs, C. T. and Perrin, M.R. (2009) Two Unsuccessful

Reintroduction Attempts of Rock Hyraxes (Procavia capensis) into a Reserve in the

KwaZulu-Natal Province, South Africa. South African Journal of Wildlife Research

39,192-201.

Wingfield, J. C. and Sapolsky, R. M. (2003) Reproduction and resistance to stress: when and how. Journal of Neuroendocrinology 15, 711-724.

Woods, G. M., Kreiss, A., Belov, K., Siddle, H. V., Obendorf, D. L. and Muller, H. K.

(2007) The immune response of the Tasmanian devil (Sarcophilus harrisii) and Devil

Facial Tumour Disease. EcoHealth 4, 338-345.

Woods, G. M., Howson, L. J., Brown, G. K., Tovar, C., Kreiss, A., Corcoran, L. M. and

Lyons, A.B. (2015) Immunology of a Transmissible Cancer Spreading among Tasmanian

Devils. The Journal of Immunology 195, 23-29.

Wooller, R. D., Richardson, K. C., Garavanta, C. A. M., Saffer, V. M and Bryant, K. A.

(2000) Opportunistic breeding in the polyandrous honey possum Tarsipes rostratus.

Australian Journal of Zoology 48, 669-680.

Wright, B., Willet, C. E., Hamede, R., Jones, M., Belov, K. and Wade, C. M. (2017)

Variants in the host genome may inhibit tumour growth in devil facial tumours: evidence form genomewide association. Scientific Reports 7, 423 DOI: 10. 1038/s41598-017-00439-

Faculty of Science The University of Sydney

Wright, B. (2017) Genomic approaches to conserving the endangered Tasmanian devil.

PhD thesis. The University of Sydney. 267pp.

Ziegler, A., Santos, P. S. C., Kellermann, T. and Uchanska-Ziegler, B. (2010) Self/nonself perception, reproduction and the extended MHC. Self/Nonself 1(3), 176-191.

Zinkernagel, R. M. (1979) Associations between major histocompatibility antigens and susceptibility to disease. Annual Review of Microbiology 33, 201-213.

Faculty of Science The University of Sydney

2 Chapter Two: An examination of factors associated with

breeding success of Tasmanian devils in captivity (2006 to

2012)

T. Keeley1, 2 and T. Russell3

1School of Agriculture and Food Sciences, University of Queensland, Gatton, 4343

2Wildlife Reproductive Centre, Taronga Western Plains Zoo, Dubbo, 2830

3School of Veterinary Sciences, University of Sydney, Sydney, 2006

2.1 Abstract

The captive Tasmanian Devil Insurance Population was developed in 2005 in response to rapid declines in the wild population due to the transmissible cancer Devil Facial

Tumour Disease (DFTD). Founder members for the captive population were brought in from disease free areas at dispersal age. The first seven years (2006 – 2012) of the

Tasmanian devil insurance program has shown mixed reproductive success, the causes of which are likely multifactorial. An evaluation of studbook records and annual reports and recommendations for the period of 2006-2012 found that in those females provided with breeding opportunities, reproduction rates were variable and overall low (39%; range 26-49%). Despite adequate breeding opportunities, often multiple per breeding season, more female wild founders than males failed to produce offspring (28.4% versus

15.5%) during their reproductive lifespan. By the end of the 2012 breeding season,

Faculty of Science The University of Sydney

53.7% of female founder lines (51/95) and 80.2% of male founder lines (57/71) were still active, representing an overall loss of 34.9% of founder line retention. On average a female devil produced her first litter at 2.1 years of age, the average age of sexual maturity. The overall average litter size was 2.7 pouch young, with similar litter sizes produced by wild-caught founder and captive born females (2.8 versus 2.6 respectively).

Although a complete understanding of the factors reducing reproductive success in captivity remains elusive, there is evidence that female devils need to be provided with breeding opportunities at the age of sexual maturity (2 years old) to maximise the chance of breeding success.

2.2 Introduction

According to the IUCN red list (2015), there are currently 209 critically endangered,

481 endangered and 507 vulnerable mammalian species. The Tasmanian devil

(Sarcophilus harrisii), is one such species being listed as endangered in 2008 (IUCN

2015) due to the effects of a transmissible cancer, devil facial tumour disease (DFTD), and the population is currently listed as declining.

Due to the speed to which DFTD was decimating the wild Tasmanian devil population

(Hawkins et al. 2006; Pyecroft et al. 2007), an insurance program was initiated in 2005, in partnership between the Australian Zoo and Aquarium Association (ZAA), the Save the Tasmanian Devil Program (STTDP) and the Tasmanian Department of Primary

Industries, Parks, Water and the Environment (DPIPWE). The aim of the program is to maintain 95% of the wild Tasmanian devil genetic diversity within the captive population for 50 years (CBSG 2008). Currently, the intensively managed breeding component of the insurance population is maintained by over a dozen institutions

Faculty of Science The University of Sydney throughout six states (Tasmania, New South Wales, Queensland, Victoria, South

Australia and Western Australia) and the Australian Capital Territory in Australia, with a number of overseas zoos recently incorporated (Srb, 2015; STTDP 2016).

In threatened species recovery, captive breeding is becoming one of the main strategies to stave off extinction. To initiate a captive breeding program, ideally, unrelated, sexually mature, healthy founding animals are required. Founders are usually wild caught individuals of a species and when captured there is an underpinning assumption that they will be unrelated to each other (Lacy, 1989). However, unless founders can be sourced from geographically isolated populations, this assumption may not be met. This new population is then managed through a studbook where all details are recorded for all individuals: their origins, sex, age, identification numbers, breeding history and other life history traits. Using the studbook and breeding records, mean kinship (MK: the average relatedness of an animal in relation to other animals within the population;

Frankham et al. 2009), are calculated and breeding recommendations can be made based on this. Animals with the lowest mean kinship scores are the most desirable for breeding selection as they are the least related to any others (Frankham et al. 2009).

However, pedigree information cannot always be corroborated and errors of relatedness have been made in studbooks in a number of species e.g. Przewalski’s horses (Equus ferus Przewalski) and the Arabian Oryx (Oryx leucoryx) (Witzenberger and Hochkirch,

2011).

In order for a captive breeding program to be a success, high levels of genetic diversity must be maintained and breeding success of founding individuals must be achieved.

Genetic diversity can be lost through genetic drift and inbreeding depression (Ballou and Lacy, 1995) so supplementing captive populations with new founders is imperative

Faculty of Science The University of Sydney where possible. In order to maintain the highest levels of genetic diversity, each founder needs to produce a large and equivalent number of offspring to ensure that the founder alleles are retained within the captive population (Lacy, 1989). It is not only the founders that need to produce offspring as each founder line needs to be kept active through the breeding of each generation. Equalising founder lines and supplementing with new founders where possible is essential to the maintenance of genetic diversity as breeding from only captive bred individuals may produce genetic adaptations to captivity in as little as three generations in intensively managed systems (Araki et al.

2007). Both overrepresentation and underrepresentation of founders can lead to allelic loss and thus a loss of heterozygosity and genetic diversity and an increase in inbreeding

(Lacy, 1989).

Breeding of endangered species is a complex process and many programs have suboptimal breeding success including those for e.g. Cheetahs (Acinonyx jubatus), western lowland gorillas (Gorilla gorilla gorilla), giant pandas (Ailuropoda melanoleuca) (Owen et al., 2004) and eastern black rhinoceros (Dicerus bicornis michaeli) (Edwards et al., 2015), which can expound their conservation status.

Reproductive failure can be multifactorial and contributing factors could include: stress

(Wingfield and Sapolsky, 2003), lack of or inappropriate mate choice, enclosure design and size, social structure (Peel et al., 2005), age at first introduction to a potential mate, inadequate breeding recommendations, inconsistent consideration of a species’ reproductive biology (Penfold et al. 2014), and reproductive behaviour and prior experience (Wachter et al., 2011). Whilst many of these factors can be quantified, there is no one rule for every species.

Faculty of Science The University of Sydney

Although Tasmanian devils have been held and bred in captivity for over a century, breeding efforts have been extremely variable, likely due to the lack of need to breed this once common species (Roberts 1915; Fleay 1935; Kelly 1993). Originally believed to be monoestrous, female devils are polyestrous with the potential to undergo up to three oestrous cycles within a breeding season if pouch young are not produced (Guiler

1970; Keeley et al. 2012). Both captive and wild female devils have been recorded breeding as yearlings although this appears to be restricted to a small number of individuals that become precocial breeders, perhaps through the attainment of a critical body weight range (Lachish et al. 2009). Female devils are generally known to breed between the ages of two and four with reproductive senescence usually occurring by the age of five (Kelly 1993; Keeley et al. 2012).

The overall aims of this project are to evaluate potential factors associated with reproductive success in the Tasmanian devil captive breeding program. Specifically we will 1) evaluate historic records for the first seven years (2006 – 2012) of the insurance program to examine changes in breeding rates and litter size over time; 2) examine the potential effects of birth location (wild caught vs captive born) and age on reproductive success and litter size 3) evaluate the potential contribution of wild caught founders to the captive breeding program by evaluating founder line retention over time and 4) compare pen size and proximity of conspecifics between successful and unsuccessful females at four different institutions with the aim of elucidating potential factors associated with reproductive success in the Tasmanian devil. Information provided by this study will contribute to the optimisation of the Tasmanian devil captive management and breeding programs for the future.

Faculty of Science The University of Sydney

2.3 Methods

2.3.1 Insurance Population Establishment

The Tasmanian Devil Insurance Population (TDIP) was established predominately through three intakes of wild born Tasmanian devils, supplemented by a few

Australasian Species Management Program (ASMP) devils deemed appropriate for inclusion and wild devils brought into captivity opportunistically (typically pouch young hand-reared due to the euthanasia of the mother due to DFTD). The areas in which devils were captured were sites considered DFTD free at the time of the trapping to reduce the risk that the devils may have been exposed to DFTD prior to capture. The first group of wild Tasmanian devils (year of birth 2004) were removed from

Narawntapu National Park or Southport (Eastern Provenance; trapped February – March

2005; n = 13) or Temma West, Arthur River, and Granville Harbour (Western

Provenance; April – May 2005; n = 10) as 1 year old juveniles (just after dispersal) with no clinical signs of DFTD. The offspring (n = 6) from three litters of females with

DFTD captured in August 2004 from Woolnorth (Western Provenance), initially held as part of a vertical transmission study, were also added to the TDIP in 2005 upon receiving a clean bill of health. All of these animals were held and bred in quarantine facilities during their first breeding season (2006) before being transferred to mainland facilities to establish the TDIP.

The second group of wild Tasmanian devils (year of birth 2006) were captured from north-western Tasmania at the beginning of 2007 as yearling juveniles (n = 31). The third and final major intake of wild devils occurred in early 2008 with the intake of yearling juveniles (n = 62) from north-western Tasmania. Of these, some were sent to

Faculty of Science The University of Sydney intensively managed facilities on the mainland (n = 30) and the remainder (n = 32) were used to establish Free Ranged Enclosures (FRE) or Managed Environmental Enclosures

(MEE).

Until 2011, the TDIP included only mainland Australian facilities but this was expanded to include two Tasmanian DPIPWE facilities in 2011 and two privately owned

Tasmanian wildlife parks in 2012 (both historically holding and breeding non-TDIP devils).

2.3.2 Insurance Population Breeding Success 2006-2012

To evaluate the success of the TDIP during the first 7 years of the program, data regarding annual breeding recommendations and outcomes were gathered from the

Annual Report and Recommendations published between 2006 and 2012 by ZAA for all

14 institutions and the three purpose built quarantine facilities (first year of the program) participating in the TDIP during those years. Data evaluation was restricted to intensely managed facilities as data on Free Ranged Enclosures (FRE) or Managed

Environmental Enclosures (MEE) was limited and paternities (and sometimes maternities) of a large number of offspring are currently unknown. Information recorded and evaluated included the devil’s studbook number, name, sex, breeding recommendations, breeding success, litter size, litter sex ratio, and institution. The

Tasmanian devil studbook was used to provide dates of birth for breeding animals and to confirm information provided in the annual reports.

To assess the potential for male fertility to influence female devil reproductive success we determined whether or not unsuccessful females were paired with proven or

Faculty of Science The University of Sydney unproven males. A proven male was defined as one that had previously successfully sired viable pouch young. An unproven male was one that was offered mating opportunities but produced no confirmed pregnancies (i.e. no viable pouch young). In cases where records indicated that more than one male could be paired with the female and there was no confirmation whether all suggested males were trialled, these were designated as proven (if both males had produced viable young), unproven (if neither male had produced viable young) or unknown (if there was both a proven and unproven male breeding recommendation and no confirmation if one or both were tried).

2.3.3 Founder Evaluation

All records of wild caught Tasmanian devils (brought into captivity 2005 – 2010) that were listed in the studbook (Carla Srb, Zoos Victoria, 2015) for the species were databased and analysed to examine founder lines. In this study, we defined a “founder” as an individual that is wild born, with unknown wild parents, that either was brought into captivity with pouch young or had at least one opportunity to breed in captivity between the ages of two and four for females or two and six for males. Devils that were brought into captivity but never had an opportunity to be paired for mating were excluded as a founder. The number of litters per female, the male sire, the number of young and the sex of all offspring were recorded. Founder lines were traced for both males and females to determine how many were still active or had terminated. To be considered “active” a founder line had to have descendants that would be under the age of four for a female or six for a male, with the potential for breeding recommendations for subsequent years.

The Tasmanian devil studbook lists a large number of devils that never had the chance to contribute to the captive breeding program. Some of these animals were brought into

Faculty of Science The University of Sydney captivity for the purposes of breeding as part of the insurance population. Others are wild caught females, likely DFTD positive, with pouch young, none of which survived to contribute to the captive population, or males which died within a couple of months of capture. These animals were not included in the analysis. Females and males that died before having an opportunity to breed in captivity were also removed as potential founders for the analysis. Potential female and male founders that were housed in unmanaged FREs were excluded from the analysis due to the difficulty of ascertaining whether or not they had bred.

2.3.4 Studbook Evaluation

The studbook evaluation is limited by the accuracy of the record and the nature and amount of information made available to the studbook keeper over time. As such it is recognised that there may be a degree of inaccuracy within the records that cannot be accounted for. Initially, the Tasmanian Devil Insurance Population (TDIP) consisted of zoological institutions on the mainland of Australia. The devils housed in two of the

Tasmanian managed enclosures were not included in the breeding recommendations until 2011 and two further facilities were included in the breeding recommendations in

2012. In both 2008 and 2009 a female devil with a breeding recommendation died before being paired with a male, therefore these two animals were excluded from the dataset.

By the end of 2012, the TDIP population had grown to 515 devils (75% in intensively managed facilities) (Hogg et al. 2012). With the growth of the TDIP, and limitations on space, breeding recommendations were no longer provided to all adult female devils and an increase in multiple males paired with females meant paternity was not always known.

Faculty of Science The University of Sydney

2.3.5 Housing

After the breeding results for the most recent season were completed, a voluntary survey was sent to all nine mainland Australian institutions that were active participants in the captive breeding program (i.e. all mainland institutions with breeding recommendations throughout the survey period). The survey requested information from each institution on the layout, size (area, height of walls), construction material and composition of inclusions (e.g. den boxes, climbing structures) of enclosures used to house breeding

Tasmanian devils for captive breeding. The survey requested information regarding the location of occupancy of breeding females and all other devils (e.g. males, juveniles) at that institution during the 2011 breeding season with a request for additional years

(2008 to 2010) if available.

2.3.6 Statistical Analysis

All statistical analyses were performed using GenStat 16th Edition (VSN International

Ltd., UK). A linear mixed model was used to test the effects of origin of birth (wild- caught or captive born) and age of the dam on the number of pouch young produced. A liner mixed model was also used to test the effects of female age and the age difference between the male and female on reproductive success (confirmation of pouch young vs no pouch young). Data from individual female devils provided breeding opportunities for more than one year were assumed to be random variables for each year. A P value <

0.05 was considered to be significant. Data are presented as ± SEM, unless otherwise noted.

A two-tailed t-test was performed on the sex of pouch young born to wild caught females and a chi squared test on among year effect of breeding success over the seven year period.

Faculty of Science The University of Sydney

2.4 Results

2.4.1 Insurance Population Breeding Success 2006-2012

The breeding success (i.e. the number of females devils successfully producing pouch young subsequent to pairing) of the captive insurance population between the years of

2006 and 2012 was variable and low overall (Table 2.1, Figure 2.1) in comparison to recommendations made in the PHVA where baseline female breeding success was modeled at 47% for 2 year olds, 42% for 3 year olds and 22% for four year olds (CBSG,

2008): which was later revised upwards to 57% of 2-3 year olds and 35% of four year olds for genetic diversity targets to be met (Hibbard 2010). Although not included in

Table 2.1, breeding rates reported for the FREs in the 2011 and 2012 Annual Reports

(ZAA) were also low (58.1% and 44.2% respectively). Of the breeding recommendations provided to each institution by ZAA, high proportions (83 to 100%) were attempted (Table 2.1). The proportion of successful breeding females over the seven year period varied from 0.26 - 0.5 and this was not significantly different Χ2 =

4.305, d.f. = 6, p = 0.64 (Table 2.1). In four confirmed cases, the number of pouch young present at the first pouch check was greater than the number of pouch young at subsequent checks (2008 – one litter of four pouch young reduced to three pouch young,

2010 – two litters of four pouch young reduced to one and two pouch young, 2011 – one litter of two pouch young reduced to zero. As the first pouch check generally occurred at a minimum of two weeks post-partum, it is impossible to determine if the loss of one or more pouch young commonly occurred. The average pouch young litter size (Table 2.2) was not significantly different (P = 0.25) between wild caught founder and captive born females, and nor was it different if the sire was wild caught or captive

Faculty of Science The University of Sydney

born (P = 0.388). The average pouch young litter size associated with female age (two,

three and four years of age) was not significantly different (P = 0.553; Table 2.2).

The rates of successful production of a litter declined as the female increased in age

(two years old = 46%, three years old = 40.8%, four years old = 23.2%). Females that

had negative pouch checks after mating or pairing were on average significantly older

than females that had positive pouch checks (2.9 ± 0.07 yrs vs 2.6 ± 0.07 yrs, P =

0.003). The average age difference between paired males and females was greater in

successful pairings than unsuccessful pairings (males 0.50 ± 0.10 vs 0.20 ± 0.08 years

older, P = 0.018).

Table 2.1 Overall summary of the breeding success of the captive Tasmanian devil Insurance Population (TDIP) (excluding FRE’s) between 2006-2012. Attempted pairings included the introduction of one or more males sequentially during the

presumed oestrus period for 1 – 3 oestrous cycles. Total number of pouch young per year

includes the number of young present at the first pouch check (two weeks to two months post

parturition). Not all institutions were provided with breeding recommendations in every year of

participation. Purpose built, temporary holding facilities in Tasmania* (Recs** =

recommendations).

Number of

Institutions Number of Average

with TDIP Breeding Number of Percent of Percent of Total Pouch Young

Breeding Pair Recs pairings Number of Recs Recs Number of per breeding

Year Recs** (Females) attempted Successful Successful Attempted Pouch Young Female

2006 3* 21 21 8 38% 100% 18 2.3

2007 4 19 19 5 26% 100% 13 2.6

2008 8 38 37 13 34% 97% 33 2.5

2009 7 38 37 19 50% 97% 57 3.0

2010 9 42 40 17 40% 95% 51 3.0

2011 8 40 39 17 43% 98% 45 2.6

2012 11 45 43 19 42% 96% 49 2.6

Faculty of Science The University of Sydney

Figure 2.1 a) Total number of adult female devils paired per year and b) produced a litter of 1-4 offspring between 2006-2012, demonstrating the shift from the breeding of wild- caught founders to captive born descendants over time.

Of the females that did not have viable offspring, 51% were paired with males that had successfully produced offspring with at least one other female, 23% were paired with males that never produced offspring and 26% were paired with males of unknown fertility (e.g. records incomplete or non-specific). The number of unknown sires

Faculty of Science The University of Sydney

increased substantially in the latter years (2010 onward), making it difficult to

determine the success of individual males and increasing the difficulties of managing

the population in regards to male kinship and contributions.

Table 2.2 Number, age and average litter size of female Tasmanian devils recommended for breeding within the captive insurance program between 2007-2012. Captive Wild 1 Year 2 Year 3 Year 4 Year

Born Born old old old old

Females Females Females Females Females Females

Total Number with

Breeding

Recommendations 102 142 1 107 84 52

Total Litters Confirmed 39 59 1 51 34 12

Average Litter Size 2.6 ± 1.0 2.8 ± 1.1 2 2.7 ± 1.1 2.8 ± 0.9 2.4 ± 0.9

2.4.2 Founder Evaluation

For female founders, breeding recommendations were generally made for the first time

when the female was two years of age. First litters were produced at the age of one, two

or three for female founders. No first litters were recorded for any females of the age of

four. The average age at which a founder female that had bred produced her first litter

was 2.1 ± 0.1 years of age. The average litter size produced by a founder female during

her reproductive life was 2.8 ± 0.1 pouch young. Of the total 298 pouch young produced

by founder females (including wild caught females with pouch young sired in the wild),

there was no sex bias (p = 0.4), 151 were male and 136 were female.

Faculty of Science The University of Sydney

Of the 95 potential female founders, 27 (28.4%) never produced viable young (Figure

2). Of the 68 females which produced young, 17 (25%) of them no longer have viable founder lines as their offspring either died or failed to produce young of their own. A total of 14 female founders fit into this category and therefore only produced a single litter as a founder due to death preventing further contributions. A total of 19 (28%) females produced two litters, and nine (13%) females produced three litters.

Of the 71 potential male founders, 11 (15.5%) never produced viable young and an additional seven (9.9%) males not yet confirmed as sires were either young enough to have future breeding opportunities or were possible sires of litters where the female had access to multiple males during the breeding season (Figure 2). Of the 53 (74.6%) founder males confirmed to have offspring, three (6%) had founder lines that were no longer viable. Therefore as of 2012, 57 (80.3%) male founder lines were still active from the original 71 male founders (Figure 2.3).

For male founders, breeding recommendations were generally made for the first time when the male was 2 years of age. Males successfully produced pouch young from the ages of two to six years old. Of the males that sired pouch young, 16 (30.2%) sired a single litter, 22 (41.5%) sired two litters, 12 (22.6%) sired three litters and three (5.7%) sired four litters.

As of 2012, from the 166 founding devils, 27% (N = 45) never successfully bred and of the 121 that did breed, 17% (N = 20) of these founder lines are no longer active whereas

108 (65%) founder lines were still active (Figure 2.3).

Faculty of Science The University of Sydney

Figure 2.2 A representation of the proportion of male (n=71) and female (n=95) founders that produced no offspring and the over contribution of number of offspring the successful founders produced.

Figure 2.3 A representation of the sex skew of overall founders and currently active founder lines for wild-caught devils. (Key: Solid blue = male founder lines active, solid pink = female founder lines active, pink stripe = female founder lines terminated, blue stripe = male founder lines terminated).

2.4.3 Housing

Surveys requesting housing conformation and devil enclosure locations were sent to all nine institutions active in 2011. Four surveys were returned during the study period, limiting information on this subject. From the four surveys, it was confirmed that the size of the pens which housed successful breeding female devils ranged from 55m2 to

Faculty of Science The University of Sydney

150m2, all above the minimum required standards (50 m2; Hockley, 2013). For male devils, a minimum of one den was provided in each enclosure. Female devils were provided with two den options which included either two man-made dens (e.g. concrete square den and clay pipe within dirt mound) or a man-made den and the materials provided so that the devil could construct its own den (e.g. concrete square den and large, compacted dirt mound). All man-made dens were made of either wood or concrete, with either a natural (dirt) floor or floor constructed of the same material with a removable or hinged lid for management access. All enclosures had natural substrate

(e.g. dirt) and logs, branches or bushes for climbing or shelter and shade. All enclosures had a minimum of 1.2 m fence made of smooth material to prevent escape. Female devils were housed adjacent to a minimum of one to a maximum of three other devils

(typically male, sometimes a mix of male and female).

Cases of successful females being housed in adjacent enclosures occurred in two of the four institutions. The number of devils held within each facility varied between years at most institutions. The maximum number of adult females varied from two to 12 within a single institution with a minimum of an equal number of males. All breeding enclosures were off-exhibit and clustered within one structure or area. There appears to be no correlation between devil numbers and breeding success. In 2009 Facility one held 12 female and 15 male adult devils in addition to a number of juvenile devils and successfully bred 9 of the 12 females (75%). However Facility two held three adult females in the same year and only had a single female breed. Breeding success over years within each institution was variable and appears unlikely to be associated with enclosure size or absolute number of devils. The only institution that had a consistent breeding rate was Facility one, but only after the move of breeding females into new, purpose built devil enclosures for the 2009 breeding season. Since then (2009-2012)

Faculty of Science The University of Sydney they have consistently produced young from most of the females which had breeding recommendations (72 -89%; 9 - 11 females). These enclosures (7 x 20 m or 140m2 each), built in a line on the 20 m shared fence line, were used exclusively for breeding from 2009 onwards with female enclosures constructed with two den options, a concrete den with natural flooring and a dirt mound for building dens if desired.

Enclosure design (the amount of furniture, size and structure of dens) differed between facilities but all provided the minimum required by the recommended husbandry guidelines (Kelly 2007).

2.5 Discussion

Overall, the litter size of animals bred in captivity is smaller than that reported for wild devils. Litter production significantly declined with female age (Table 2.2). Within the part of the insurance population examined, there does not appear to be a difference associated with whether a female is wild-caught or captive born in relation to the number of joeys per litter produced. Females paired with older males had greater reproductive success than those paired with younger males or those of the same age. No sex bias was found amongst offspring born in captivity. Almost one third of potential female founders never produced viable young (Figure 2) and one quarter of those that did reproduce no longer have viable founder lines, whereas the majority of male founder lines were still active in 2012 (Figure 2.3). Successfully reproducing females were all housed in enclosures exceeding the size in recommended standards.

Faculty of Science The University of Sydney

2.5.1 Breeding success

A historic evaluation of litter size of wild devils prior to DFTD reported 3.4 ± 0.9 joeys per litter, is significantly different to litter size among wild caught captive females, with a higher proportion of litters with the maximal number of four joeys (63%) (Keeley et al. 2017). Another study, monitoring a population of devils affected by DFTD, reported similar breeding rates with a mean litter size of 3.4 joeys per litter and 66% of all dams carrying a full pouch of four joeys (Lachish et al. 2009). A further study within a wild

DFTD affected population revealed an average of 3.1 ± 0.2 joeys per litter with 22% of dams (n = 9) carrying four joeys and 67% with three joeys per litter (Russell, unpublished data). It is possible that in the wild, litter sizes are larger to compensate for a higher loss of young between birth and weaning (Guiler, 1970) whereas in captivity resources are abundant and survival rates are higher. A recent study evaluating

Tasmanian devil captive breeding rates found a larger disparity between litter sizes of wild-caught and captive born females (2.91 versus 2.41 joeys per litter respectively)

(Hogg et al. 2015). The primary differences between this study and that of Hogg et al.

(2015), is the inclusion of females housed in FRE and MEE enclosures and the inclusion of an additional year (2013) in the latter. As the devils in the FRE and MEE enclosures and records for the 2013 breeding season confirm that these consisted of approximately 80% captive females with an average of 2.6 pouch young produced per litter it is unknown why their numbers differ so greatly from those reported here. It is possible that there were some discrepancies between the data used between 2006 -2012 between the two studies.

Captive devils have a larger number of small litters than wild devils (Keeley et al.

2017). It is possible that some of the litters were lost early (i.e. never confirmed) were small litters and the females may have chosen to discard their pouch young with hopes

Faculty of Science The University of Sydney of achieving a larger litter or more genetically fit offspring later in the season. In striped-face dunnart (Sminthopsis macroura) laboratory colonies, it was noted that small litters (one to two young) were never reared to weaning (Godfrey 1969; Woolley 1990).

It was also noted that 18 of 25 litters lost died at birth possibly due to handling, stress or inbreeding (Godfrey 1969). It was found that in the third breeding season of the colony, which was founded by three female and four male striped-face dunnarts, only two litters were born and no young survived, and irregular cycles were common, suggesting deleterious effects on reproduction associated with inbreeding (Godfrey 1969). Studies on the fat-tailed dunnart (Sminthopsis crassicaudata) showed similar results with pairings of close relatives (e.g. siblings, parent-offspring), which were often still successful but with reduced viability of the offspring (Bennett et al. 1990).

Unfortunately in devils it is not possible to confirm pregnancy or the loss of offspring at birth due to current management practices and lack of diagnostic tests, but it is possible that some of these factors contribute to their reduced reproductive success. If some of the founder devils were indeed siblings or half siblings, which is entirely possible due to the capture of wild founders from the same locations at the age of dispersal (Hogg et al.

2015), then this may have increased the rate of undetected inbreeding leading to an increased number of unviable offspring in subsequent generations.

Asymmetrical reproductive aging has been observed in species of rhinocerous and elephant, where most of the reproductive lifespan of the female is spent in pregnancy or lactation decreasing the number of oestrous cycles experienced and the potential deleterious effects of prolonged exposure to endogenous sex steroids during their lifetime (Hermes et al. 2004). In the lesser mouse lemur (Microcebus murinus) it was demonstrated that if seasonal reproduction is artificially manipulated using an accelerated photoperiodic regimen then longevity is affected but only in relation to

Faculty of Science The University of Sydney chronological age, not on the number of seasonal cycles or reproductive potential

(Perret 1997). If a species has a “fixed” number of oestrous cycles in which reproductive potential is optimal, then extension beyond this due to prolonged periods of non-mating may accelerate reproductive and chronological aging. For captive devils, a female may experience the same number of oestrous cycles in captivity (three in a year) that she would experience in the wild if successful every year (three over three years). This may have a negative effect on breeding success in subsequent years reducing her ability to become pregnant or produce viable young as she ages.

The primary age of first reproduction in the study females was two years of age. No record of first breeding at four years of age occurred for the females within this dataset.

Breeding recommendations were given for nine (data not shown) females in their 5th year of which four had produced a single litter of young previously, the others had not, despite adequate breeding opportunities with fertile males. None of these five year olds produced offspring; in fact, reports of five year old females producing offspring are rare and limited to females which produce several litters in their lifetime. This suggests that female devils may have decreased fertility related with an increase in age that is affected by their reproductive history. To improve the reproductive success of females, it is advisable that females are provided with breeding opportunities at the age of two years old and that breeding for the first time should not be delayed beyond this age. Delayed reproduction has been linked to reproductive fitness and failure (Penfold et al. 2014).

Interestingly, we found that female devils were more likely to be successful if paired with an older male rather than one of similar age or younger. A similar association has also been identified among island foxes (Urocyon littoralis) where female success was correlated with male age. Female foxes produced more offspring per litter when paired

Faculty of Science The University of Sydney with older males (Calkins et al. 2013). We would also suggest females be paired with older males as opposed to those younger or of the same age.

In another carnivore species, Wachter et al. (2011) found that breeding success, or lack thereof, in cheetahs was related to reproductive history and the age at which females were first bred, rather than a lack of genetic diversity, irregular hormone cycling or captive stress. Genital pathologies were observed in older and nulliparous females as opposed to younger, reproductive animals and the authors suggest that reproductive health in cheetahs is directly correlated with age and prior reproductive history

(Wachter et al. 2011). Breeding cheetahs in early adulthood is the most likely time to achieve breeding success in this species as it prevents asymmetric reproductive aging and prolongs the reproductive lifespan of the female (Penfold, et al. 2014; Wachter et al.

2011). Although there is limited information regarding reproductive pathologies in

Tasmanian devils, a previous study on a wild population found 16 individuals with abdominal or reproductive pathologies such as pouch tumours and ovarian cysts

(Hughes and Keeley 2014). A five year old, wild born female devil was confirmed with reproductive tract pathologies upon autopsy within the insurance population (T. Keeley, personal communication). This particular female had 2-3 oestrous cycles each year from two to four year of age, was paired with multiple males and never produced offspring. It is possible that exposure to an excess of endogenous hormones associated with multiple oestrous cycles per year without pregnancy and lactation may have contributed to her condition. It is unknown how frequent reproductive tract pathologies may occur in the devil but there is clearly a decrease in fecundity as the female ages (46.2% to 23.2% success from 2 to 4 years of age). This trend of decreased fecundity with age is also seen in other marsupials e.g. bandicoots (Price-Rees, 2012).

Faculty of Science The University of Sydney

Unlike devils, there are other dasyurid species involved in captive breeding programs where consistently high breeding rates are achieved. For example, the dibbler captive breeding program at Perth Zoo is part of a recovery plan for facilitating the reintroduction of dibblers to their natural habitat. During this program, an average of

85% of females paired have produced pouch young (Cathy Lambert, Perth Zoo personal communication; Lambert and Mills 2006). This success rate is far greater than observed in devils, although it has been noted that in the first year of captive dibbler reproduction a high level of offspring cannibalism occurred, presumably due to stress (Lambert and

Mills 2006). The husbandry was changed immediately, decreasing human activity, increasing the quantity of live food and increasing space allocations, resulting in a marked reduction in offspring canabalism (Lambert and Mills 2006). Such changes may also benefit other species such as the devil.

2.5.2 Founder evaluation

The breeding success of founder animals in the Tasmanian devil insurance program has been variable resulting in the loss of a large number of founder lines. Although male founder lines have persisted more than female founder lines (80% vs 53% still active respectively), the overall breeding success remains low and the loss of founder lines is of concern in a species with such limited genetic diversity.

When founders are recruited to establish or supplement a captive breeding program, it is under the assumption that they are unrelated (Rudnick and Lacy 2008). Breeding recommendations are made to pair unrelated individuals with the aim of minimising mean kinship (MK) whilst achieving reproductive success. When Tasmanian devils were caught at the age of dispersal (one year old), they were all assumed to be unrelated for the purposes of the captive breeding program, despite the fact that several devils

Faculty of Science The University of Sydney were caught at each trapping location and therefore had the potential of being full or half siblings (Hogg et al. 2015). Inbreeding avoidance is innate in most species and animals may avoid mating with close kin wherever possible. The recognition of kin has been shown in numerous species and some species will even distinguish along maternal or paternal lines (Lacy and Sherman, 1983). Kin recognition and mate choice are related to the genes of the Major Histocompatibility Complex, and it appears that closely related animals can avoid mating with each other through olfactory cues (Tregenza and

Wedell, 2000) identifying the others’ MHC type to determine compatibility or in the case of closely related kin, incompatibility (Ziegler et al. 2010). This could be one of the reasons why some paired wild-caught devils did not achieve mating success as their

MHC genotypes may have been too similar or not compatible (Chapter 4; Russell et al.

2018) or joeys conceived that were genetically incompatible may have been terminated in utero, at birth or shortly after parturition.

2.5.3 Housing and stress

It is probable that newly born devils that are lost between pouch checks, i.e. present for one and then absent at the next, or born but lost before the first pouch check, have been consumed by the mother, as newborn joeys have never been found in the enclosures.

This could be related to captivity related stressors and therefore future study is warranted. In captive dibblers, infanticide was observed and noted as a major cause of reproductive failure early in the establishment of the breeding program which was attributed to stress (Lambert and Mills, 2006). No pouch young bodies were found, and being present at initial pouch checks and then not at subsequent ones lends weight to the assumption that pouch young lost were cannibalised. Similarly, from 38 litters of captive red-tailed phascogale (Phascogale calura), only 22 (68%) were raised to weaning with the original birth complement (which ranged from one to eight pouch

Faculty of Science The University of Sydney young per litter), suggesting cannibalisation of young may be a result of stress (Foster et al. 2008). Shultz et al. (2006) found partly mutilated joeys in pouches of brush-tailed rock wallabies (Petrogale penicillata) confirming that pouch young losses can be attributed to the dam. Throughout that study, pouch young losses declined significantly, suggesting that purposeful improved husbandry practices, facilitated pouch young retention (Shultz et al. 2006). Wild ringtail possums have also been observed cannibalising their joeys when brought into captive situations after injury or misadventure (T. Russell, unpublished data) suggesting that this could be a stress response to handling or captivity.

It is possible that changes to the enclosures and management of devils at Facility one contributed to an increase and maintenance of higher breeding success. This may have been multifactorial and associated with changes in enclosure furniture, better suited dens and/or changes in substrate that may have better replicated wild conditions, thus decreasing stress. The effects of the captive environment on devil reproductive success was an area highlighted as in need of further research in the 2008 PHVA (CBSG 2008) and records to date confirm that future work in this area would be beneficial.

Whilst we did not find any evidence to suggest that enclosure size may be restrictive to breeding success, the continued success of Facility one in producing joeys is noteworthy after the development of new devil enclosures. Conspecific proximity was a factor affecting reproductive success in captive island foxes with models predicting higher litter sizes with increased distance to the next occupied enclosure (Calkins et al. 2013).

This was not evident in devils within this study as females would produce young with other female or male devils in the adjacent enclosures. However, these data were limited and this may be an area to examine further. Suppression of breeding behaviours in giant

Faculty of Science The University of Sydney pandas housed in sub-optimal enclosures was observed (Peng et al. 2007), however, when pandas were allowed into more spacious accommodations, their normal breeding behaviours resumed. In the clouded leopard (Neofelis nebulosa), it was found that enclosure height, in addition to adequate climbing structures, was crucial to minimise aggression between paired individuals and to increase breeding success (Wielebnowski et al. 2002). This demonstrates that enclosure size is important and must be coupled with features that cater to the species’ behavioural and physiological needs.

For the future development of new captive breeding programs for endangered species, it is recommended to ensure genotyping animals at the time of recruitment into the population (prior to breeding) so that their relatedness in relation to potential mates can be determined with confidence and this information can be incorporated into the studbook for future use and evaluation. Miller et al., (2015) observed disparities between studbook data and molecular results in their study on the captive management of greater bilbies (Macrotis lagotis). Studbook records were shown to have overestimated genetic diversity while calculations from the studbook underestimated inbreeding (Miller et al., 2015). It is very possible that the same may be true for the

Tasmanian devil. Without genetic corroboration of studbook pedigrees, genetic diversity can unknowingly be lost which would prove disadvantageous for genetically depauperate species like the Tasmanian devil. Attempts thus far have failed to confirm the relatedness of founder animals (Hogg et al. 2015) but with expansion of microsatellite numbers or improving technology and techniques (Wright et al. 2015;

Gooley et al. 2017) it would still be worthwhile to attempt to retrospectively examine founder relationships of the Tasmanian devil insurance population to determine if this may be a factor in the reduced reproductive success and smaller litter sizes confirmed by this study.

Faculty of Science The University of Sydney

With the complexity of enclosure size and design, transfer of animals between facilities for breeding, husbandry requirements, mate choice and limited genetic diversity, there are many interacting factors which may be contributing to suboptimal breeding success in the Tasmanian devil. This study has demonstrated that although breeding rates are overall low, they have become fairly consistent. To ensure that rates continue to be maintained or possibly increased it is important that the reproductive biology of the species is incorporated into the breeding recommendation planning process to ensure female devils are provided breeding opportunities at the optimal age of 2 years old and ideally paired with older males. Further examination of enclosure features and their potential impact on reproductive success is warranted. Perhaps as a first option, a facility could adopt as many features present in the enclosures in Facility One to see if these changes might improve breeding success. As DFTD is not vertically transmitted, additional founders are still available for incorporating into the insurance population which is continuing to feed into the founder base of this program. Unfortunately this may not always be the case therefore it is important to continue to investigate further factors which may influence reproductive success of this species to safeguard this species into the future.

The first seven years (2006 – 2012) of the Tasmanian devil insurance program has shown mixed reproductive success, the causes of which are likely multifactorial. The findings of our study and insights gained from our research and from others provide possible explanations for variable captive breeding success found in this species.

Faculty of Science The University of Sydney

2.5.4 Acknowledgements

We would like to acknowledge the efforts of K. Ritky who while an Honours student at the University of Sydney began collating information for this study and facilitated the housing survey.

We would like to acknowledge the hard work and dedication to all those who have been involved in the Tasmanian Devil Insurance Population breeding and management. The

Devil Insurance Population program would not have been possible without ongoing collaborations between the Australian Zoo and Aquarium Association and the

Tasmanian Department of Primary Industries, Parks, Water and the Environment. In particular Caroline Lees, Chris Hibbard and Carolyn Hogg from the Australian Zoo and

Aquarium Association, and Carla Srb of Healesville Sanctuary, studbook keeper provided dedicated time and effort over the years making breeding recommendations and closely monitoring the breeding outcomes and management of all the captive devils in the program. Valuable information on the housing and management of four institutions - Taronga Zoo, Taronga Western Plains Zoo, Monarto Zoo and Healesville

Sanctuary were provided by keeper and management staff. The Tasmanian devil keeper teams at all of the participating institutions are accredited for their knowledge, passion and contributions to breeding devils and assisting in a diverse suite of research initiatives. This project was granted research approval through the Tasmanian devil

Captive Research Advisory Group in 2011.

Faculty of Science The University of Sydney

2.6 References

Araki, H., Cooper, B. and Blouin, S. (2007) Genetic effects of captive breeding cause a rapid, cumulative fitness decline in the wild. Science 318: 101-103.

Ballou, J.D. and R.C. Lacy (1995) Identifying genetically important individuals for management of genetic diversity in pedigreed populations. Pages 76-111 in J.D. Ballou,

M. Gilpin, and T.J. Foose (eds.), Population Management for Survival & Recovery.

Analytical Methods and Strategies in Small Population Conservation. Columbia

University Press, New York.

Bennett, J. H., Breed, W. G., Hayman, D. L. and Hope, R.M. (1990) Reproductive and genetic studies with a laboratory colony of the dasyurid marsupial Sminthopsis crassicaudata. Australian Journal of Zoology 37, 207-222.

Calkins, E. S., Fuller, T. K., Asa, C. S., Sievert, P. R. and Coonan, T. J. (2013) Factors influencing reproductive success and litter size in captive island foxes. The Journal of

Wildlife Management 77(2), 346–351.

CBSG (2008) Tasmanian devil population and habitat viability assessment final report.

IUCN/SSC Conservation Breeding Specialist Group, Apple Valley, MN.

Faculty of Science The University of Sydney

Edwards, K. L., Walker, S . L., Dunham, A. E., Pilgrim, M., Okita-Ouma, B. and

Shultz, S. (2015) Low birth rates and reproductive skew limit the viability of Europe’s captive eastern black rhinoceros, Diceros bicornis michaeli. Biodiversity and

Conservation 24, 2831-2852.

Fleay, D. (1935) Notes of the breeding of Tasmanian devils. Victorian Naturalist 52,

100-105.

Foster, W. K., Caton, W., Thomas, J., Cox, S. and Taggart, D. A. (2008) Timing of birth and reproductive success in captive red-tailed phascogales, Phascogale calura. Journal of Mammalogy 89(5), 1136-1144.

Frankham, R., Ballou, J. D. and Briscoe, D. A. (2009) Introduction to Conservation

Genetics 2nd edition. Cambridge University Press. Cambridge UK.

Godfrey, G.K. (1969) Reproduction in a laboratory colony of the marsupial mouse

Sminthopsis larapinta (Marsupialia: Dasyuridae). Australian Journal of Zoology 17,

637-654.

Guiler, E.R. (1970) Observations on the Tasmanian devil, Sarcophilus harrisii

(Marsupialia: Dasyuridae) II. Reproduction, breeding and growth of pouch young.

Australian Journal of Zoology 18, 63-70.

Faculty of Science The University of Sydney

Hawkins, C. E., Baars, C., Hesterman, H., Hocking, G. J., Jones, M. E., Lazenby, B.,

Mann, D., Mooney, N., Pemberton, D., Pyecroft, S., Restani, M. and Wiersma, J. (2006)

Emerging disease and population decline of an island endemic, the Tasmanian devil

Sarcophilus harrisii. Biological Conservation 131, 307-324.

Hermes, R., Hildebrandt, T. B. and Goritz, F. (2004) Reproductive problems directly attributable to long-term captivity - asymmetric reproductive aging. Animal

Reproduction Science 82-83, 49-60

Hogg, C. J., Srb, C. and Hockley, J. (2012) Annual report for the DPIPWE-ZAA

Tasmanian devil insurance population. Zoo and Aquarium Association. pp 43 .

Hogg, C., Ivy, J.A., Srb, C., Hockley, J., Lees, C., Hibbard, C. and Jones, M. (2015)

Influence of genetic provenance and birth origin on productivity of the Tasmanian devil insurance population. Conservation Genetics 16, 1465-1473

Hughes, R. L., and Keeley, T. (2014) Population dynamics of Tasmanian devils in north-eastern Tasmania 1974 -1987. In 'The Australian Mammal Society Conference',

Melbourne, Australia

IUCN redlist www.iucnredlist.org accessed April 10, 2016.

Faculty of Science The University of Sydney

Keeley, T., McGreevy, P. M., Fanson, B. G., Masters, K. and O’Brien, J. K. (2012) The reproductive cycle of the Tasmanian devil (Sarcophilus harrisii) and factors associated with reproductive success in captivity. General and Comparative Endocrinology 176,

182-191

Keeley, T., Russell, T., Carmody, K., Kirk, G., Eastley, T., Britt-Lewis, A., Post, M.,

Burridge, M., Eccleston, S., Faulkner, T., Forge, T., Leonard, J. and Hughes, R.L.

(2017) Seasonality and breeding success of captive and wild Tasmanian devils

(Sarcophilus harrisii). Theriogenology 95, 33-41.

Kelly, A. (1993) Management of Tasmanian devils (Sarcophilus harrisii) at the

Tasmanian Wildlife Park In 'The Biology and Management of Australasian Carnivorous

Marsupials.' (Eds. M Roberts, J Carnio, G Crawshaw and M Hutchins) pp. 79-84.

(Metropolitan Toronto Zoo: Toronto, Canada).

Kelly, A. (2007) Husbandry guidelines for the Tasmanian devil, Sarcophilus harrisii.

Zoo and Aquarium Association, Sydney.

Lachish, S., McCallum, H. and Jones, M. (2009) Demography, disease and the devil: life-history changes in a disease-affected population of Tasmanian devils (Sarcophilus harrisii). Journal of Animal Ecology 78, 427-436.

Lacy, R. C. (1989) Analysis of founder representation in pedigrees: founder equivalents and founder genome equivalents. Zoo Biology 8, 111-123.

Faculty of Science The University of Sydney

Lacy, R. C. and Sherman, P. W. (1983) Kin recognition by phenotype matching. The

American Naturalist 121(4), 489-512.

Lambert, C., and Mills, H. R. (2006) Husbandry and breeding of the Dibbler

Parantechinus apicalis at Perth Zoo. International Zoo Yearbook 40, 290-301.

McCallum, H., Tompkins, D. M., Jones, M., Lachish, S., Marvanek, S., Lazenby, B.,

Hocking, G., Wiersma, J. and Hawkins, C. (2007) Distribution and impacts of

Tasmanian devil facial tumor disease. EcoHealth 4, 318-325.

Miller, E. J., Eldridge, M. D. B., Morris, K., Thomas, N. and Herbert, C. A. (2015)

Captive management and the maintenance of genetic diversity in a vulnerable marsupial, the greater bilby. Australian Mammalogy, 37, 170–181.

Owen, M. A., Swaisgood, R. R., Czekala, N. M., Steinman, K. and Lindburg, D. G.

(2004) Monitoring Stress in Captive Giant Pandas (Ailuropoda melanoleuca):

Behavioral and Hormonal Responses to Ambient Noise Zoo Biology 23, 147–164

(2004).

Peel, A. J., Vogelnest, L., Finnigan, M., Grossfeldt, L. and O'Brien, J. K. (2005) Non-

Invasive Fecal Hormone Analysis and Behavioral Observations for Monitoring Stress

Responses in Captive Western Lowland Gorillas (Gorilla gorilla gorilla) Zoo Biology

24, 431–445.

Faculty of Science The University of Sydney

Penfold, L.M., Powell, D., Traylor-Holzer, K. and Asa, C.S. (2014) "Use it or lose it":

Characterization, implications and mitigation of female infertility in captive wildlife.

Zoo Biology 33, 20-28

Peng, J., Jiang, Z., Qin, G., Huang, Q., Li, Y., Jiao, Z., Zhang, F., Li, Z., Zhang, J., Lu,

Y., Liu, X. and Liu, J. (2007) Impact of activity space on the reproductive behaviour of giant panda (Ailuropoda melanoleuca) in captivity. Applied Animal Behaviour Science

104, 151-161.

Perret, M. (1997) Change in photoperiodic cycle affects life span in a prosimian Primate

(Microcebus murinus). Journal of Biological Rhythms 12, 136-145.

Price-Rees, S. J., Congon, B. C. and Krockenberger, A. K. (2012) Size delays female senescence in a medium sized marsupial: The effects of maternal traits on annual fecundity in the northern brown bandicoot (Isoodon macrourus). Austral Ecology 37,

313-322.

Pyecroft, S. B., Pearse, A., Loh, R., Swift, K., Belov, K., Fox, N., Noonan, E., Hayes,

D., Hyatt, A., Wang, L., Boyle, D., Church, J., Middleton, D. and Moore, R. (2007)

Towards a case definition for Devil Facial Tumour Disease: What is it? EcoHealth 4,

346-351.

Roberts, M. G. (1915) The keeping and breeding of Tasmanian devils (Sarcophilus harrisii). Proceedings of the Zoological Society of London 85, 575-581.

Faculty of Science The University of Sydney

Rudnick, J. A. and Lacy, R. C. (2008) The impact of assumptions about founder relationships on the effectiveness of captive breeding strategies. Conservation Genetics

9, 1439-1450.

Russell, T., Ujvari, B., Spindler, R., Lane, A., Keeley, T., Srb, Thomas, F. C., Hogg,

C.J., Belov, K. and Madsen, T. (2018) MHC diversity and female age underpin reproductive success in an Australian icon; the Tasmanian Devil. Scientific Reports 8,

4175.

Russell, T., Ujvari, B., Keeley, T., Lane, A., Morris, K., Clarke, J., Spindler, S., Belov,

K. and Madsen, T. (in prep), refer to the chapter in this thesis. Multiple paternity and precocial breeding in males in a wild, east coast population of Tasmanian devils

(Sarcophilus harrisii).

Save the Tasmanian Devil Program (2016) www.tassiedevil.com.au

Shier, D. M. (2006) Effect of family support on the success of translocated black-tailed prairie dogs. Conservation Biology 20: 6, 1780–1790.

Shier, D. M. and Swaisgood, R. R. (2012) Fitness Costs of Neighborhood Disruption in

Translocations of a Solitary Mammal. Conservation Biology 26:1, 116-123.

Srb, C. (2015) Australasian Regional Tasmanian devil studbook: Sarcophilus harissii.

Faculty of Science The University of Sydney

Taggart, D. A., Schultz, D. J., Corrigan, T. C., Schultz, T. J., Stevens, M., Panther, D. and White, C. R. (2015) Australian Journal of Zoology 63, 383–397. Reintroduction methods and a review of mortality in the brush-tailed rock-wallaby, Grampians National

Park, Australia.

Tregenza, T. and Wedell, N. (2000) Genetic compatibility, mate choice and patterns of parentage: Invited Review. Molecular Ecology 9, 1013-1027.

Wachter, B., Thalwitzer, S., Hofer, H., Lonzer, J., Hildebrandt, T. and Hermes, R.

(2011) Reproductive history and absence of predators are important determinants of reproductive fitness: the cheetah controversy revisited. Conservation Letters 4, 47–54.

Wielebnowski, N., Fletchall, N., Carlstead, K., Busso, J. M., and Brown, J. L. (2002)

Noninvasive assessment of adrenal activity associated with husbandry and behavioural factors in the North American clouded leopard population. Zoo Biology 21, 77-98.

Wingfield, J. C. and Sapolsky, R. M. (2003) Reproduction and resistance to stress: when and how. Journal of Neuroendocrinology 15, 711-724.

Witzenberger, K. A. and Hochkirch, A. (2011) Ex situ conservation genetics: a review of molecular studies on the genetic consequences of captive breeding programmes for endangered animal species. Biodiversity and Conservation 20, 1843-1861.

Woods, G. M., Kreiss, A., Belov, K., Siddle, H. V., Obendorf, D., and Muller, H. K.

(2007) The immune response of the Tasmanian devil (Sarcophilus harrisii) and the

Devil Facial Tumour Disease. EcoHealth 4, 338-345.

Faculty of Science The University of Sydney

Woolley, P. (1990) Reproduction in Sminthopsis macroura (Marsupialia: Dasyuridae) I.

The Female. Australian Journal of Zoology 38, 187-205.

Woolley, P. A. (2015) The Julia Creek dunnart, Sminthopsis douglasi (Marsupialia:

Dasyuridae): breeding of a threatened species in captivity and in wild populations.

Australian Journal of Zoology 63, 411-423.

Wright, B., Morris, K., Grueber, C. E., Willet, C. E., Gooley, R., Hogg, C. J., O’Meally,

D., Hamede, R., Jones, M., Wade, C. and Belov, K. (2015). Development of a SNP- based assay for measuring genetic diversity in the Tasmanian devil insurance population. BMC Genomics 16,791.

Faculty of Science The University of Sydney

3 Chapter Three: Multiple paternity and precocial breeding in

wild Tasmanian devils (Sarcophilus harrisii)

3.1 Abstract

Polyandry (females mating with multiple males) is a common reproductive mode in mammals including many marsupials. Potential benefits of polyandry in female mammals include direct benefits, such as reduced risk of male infanticide, and indirect, such as increased genetic diversity and acquisition of ‘good genes’. In this study we investigated whether multiple paternity occurred in litters within a wild population of Tasmanian devils

(Sarcophilus harrisii) on the Forestier Peninsula on the east coast of Tasmania. Using a combination of 7 MHC-linked and 8 neutral microsatellite markers, paternity in nine litters, representing 28 offspring, were analysed. The results revealed multiple paternity in four of the nine litters and the first record of one year old male devils successfully siring offspring in the wild (precocial breeding). No effect of male body size and number of young sired was detected. This study provides the first record of multiple paternity in wild Tasmanian devil litters. To date, there are no data relating to the subsequent survival of devils from single and multiple sired litters therefore we do not know whether multiple paternity increases offspring survival in the wild. The results do, however, have implications for the

Tasmanian devil captive insurance program as introducing numerous males to females may increase the genetic diversity of litters and importantly, enhance the reproductive success of the captive devils.

Faculty of Science The University of Sydney

3.2 Introduction

Multiple paternity has been recorded in offspring in many vertebrate orders: fish (e.g. Liu et al. 2013), amphibians (e.g. Hudson and Fu 2013), reptiles (e.g. Olsson et al. 1996), birds

(e.g. Griffith et al. 2002) and mammals (both eutherians; e.g. Thonhauser et al. 2013 and marsupials: e.g. Parrott et al. 2005). Several adaptive explanations for polyandry have been proposed based on direct and indirect female benefits. Examples of direct benefits include male nutrient transfer to the female during copulation, as seen in several insect species

(Arnqvist and Nilsson, 2000). Extra-pair paternity sired broods in some birds have been shown to result in increased provisioning rates and enhanced offspring survival (Townsend et al. 2010). In some mammals, infanticide has been shown to decrease in polyandrous species where several males interact socially with the same female such as in multi-male group primates and in rodents where several male home ranges overlap those of a female

(Klemme et al. 2007; Ebensperger and Blumstein 2007). Klemme et al. (2007) suggested that female multiple mating may enhance ovulation stimulation and hence may constitute yet another direct benefit of polyandry.

Polyandry has also been shown to provide females with indirect, i.e. genetic, benefits such as the production of more genetically diverse offspring, which can enhance female fitness, especially so in species where the offspring remain within their natal area (McLeod and

Faculty of Science The University of Sydney et al. 1996; Zeh and Zeh, 1996). As suggested by Jennions and Petrie (2000) direct benefits alone are unlikely to underpin female choice: indirect benefits are often inextricably linked with direct benefits.

Tasmanian devils, Sarcophilus harrisii, Dasyuridae, the world’s largest living marsupial carnivores, exhibit sexual dimorphism. Adult males weighbetween 7 - 13 kg and adult females weigh between 4 - 9 kg. Sexual maturity is reached for both sexes at an age of two years (Hawkins et al. 2006). Male devils have large testes relative to body mass, a morphological trait often seen in species with intense sperm competition (Taggart et al.

2003). Successful breeding of one-year old females (precocial breeding) has been observed in populations where older cohorts have been lost due to Devil Facial Tumour Disease

(DFTD) (Jones 2008; Lachish 2009). In the wild, Tasmanian devils have a synchronous breeding season, beginning in February/March with the majority of mating occurring over a

3 - 6 week period (Guiler 1970; Hamede et al. 2008). However, as some females may have a second or, rarely, a third oestrus period, breeding has been recorded as late as in May

(Hesterman et al. 2008a; Hamede et al. 2008; Keeley et al. 2012a). Male competition for females is intense and mating can last from between one to ten days (average 3.4 ± 0.9 days) (Kelly, 2007; Hesterman et al. 2008a) with males being possessive of chosen females, isolating them in mating dens and exhibiting mate guarding behaviours

(Pemberton 1990; Kelly 2007). Oestrous lasts for an average of 32 (range 21-40) days

(Hesterman et al. 2008a) and the period from mating to birth is 14 - 21 days (mean 16.7 ±

0.8) (Hesterman et al. 2008a; Keeley et al. 2012a). As with many other sexually dimorphic dasyurids (see Taggart et al. 2003), Tasmanian devil females can mate prior to ovulation and store sperm in crypts within the isthmus of the oviduct for up to two weeks (Shimmin et al. 2000; Shimmin et al. 2002; Hesterman et al. 2008a) which is presumed to occur between the proestrous and luteal phases (Keeley et al. 2012a). Hesterman et al. (2008a;

Faculty of Science The University of Sydney

2008b) state that the time between mating and ovulation in the devil ranges from 3-9 days,

thus allowing time for mating with more than one male prior to ovulation.

Currently, Tasmanian devils are under threat of extinction from DFTD, a fatal,

transmissible cancer. Population reductions of up to 90% in much of Tasmania have been

recorded (McCallum et al. 2009; Woods et al. 2015). Genetic diversity within the devil

population is extremely low at both neutral markers (microsatellite) as well as at crucial

genes involved in immune function such as the Major Histocompatibility Complex (MHC)

genes (Woods et al. 2007; Siddle et al. 2010; Miller et al. 2011; Hendricks et al. 2017).

The aim of the present study was to investigate the presence of polyandry in female devil

litters obtained from a wild remnant population on the Forestier Peninsula in south-eastern

Tasmania using a combination of polymorphic microsatellite markers.

3.3 Materials and methods

The analyses in the present study are based on 53 wild devils ranging from one to six years of age that were captured and removed (16 males, 9 females and 28 joeys) from a 190km2 area on the Forestier Peninsula (42.9167° S, 147.9167° E) in May, 2011 as part of a depopulation event to eradicate DFTD from the peninsula.

3.3.1 Genetic analyses

Ear punch biopsies suspended in 90% ethanol were supplied for all devils by staff from the

Tasmanian Department of Primary Industries, Parks, Water and Environment (DPIPWE).

Devil DNA was extracted from ear biopsies using a Qiagen DNeasy® extraction kit. DNA

concentration was quantified on an Implen® nanophotometer.

Faculty of Science The University of Sydney

PCR amplification of 7 MHC – linked microsatellites loci developed by Cheng and Belov

(2012) and 8 neutral microsatellite markers developed by Jones et al. (2003) were used in the present study. Reaction volume was 15 µL comprising 1 x PCR Buffer (20mM Tris-

HCl) and 50 mM KCl, 2.0 mM MgCl2, 0.2 mM dNTP’s, 0.5 µM 5’-flourescent CAG primer, 0.05 µM CAG tag labeled primer, 0.5 µM unlabeled primer, 0.625 µg BSA, 1.5 U of Taq DNA polymerase and ~40 ng genomic DNA.

PCR reactions were performed in a BioRad MJ Mini Personal Thermocycler under the following conditions; 94oC for 3 minutes, followed by 10 x cycles of 94 oC for 30 seconds,

60 – 0.6 oC /cycle for 30 seconds, 72 oC for 30 seconds and then 30 x cycles of 94 oC for 30 seconds, 54 oC for 30 seconds, 72 oC for 30 seconds. The final step was 72 oC for 30 minutes followed by an infinite hold at 15 oC. PCR products were then examined on a 2% agarose gel. Amplified PCR products (1.5 µL) were analysed on an ABI 3130XL sequencer

(Applied Biosystems) using the size standard LIZ 500. Results were analysed and peaks were scored using Peak Scanner 2.0 (Applied Biosystems).

3.3.2 Allele frequency, heterozygosity and linkage disequilibrium

GENEPOP 4.2 (Raymond and Rousset, 1995) was used to generate observed and expected heterozygosity, and Arlequin 3.5 (Excoffier and Lischer, 2010) was used to calculate standard molecular diversity indexes: (allele frequencies and linkage disequilibrium).

These tests were separated for adults and offspring. The test for deviation from Hardy-

Weinberg equilibrium (H.W.E.) was conducted by using a Markov chain for all loci

(forecasted chain length 1 000 000, dememorisation steps 100 000). A significant p value of < 0.05 was applied after sequential Bonferroni correction.

Paternity was assigned using Cervus (Version 3.0.7) (Kalinowski et al. 2010). The Cervus program uses microsatellite data to assign paternity with the highest positive LOD score

(log-likelihood ratio) assigned indicates that the likelihood of paternity is high versus the

Faculty of Science The University of Sydney relative likelihood of an alternate male being the sire (Marshall et al. 1998). Further, from the two most likely males, a delta Δ score is given (defined as the logarithm of the ratio of likelihood ratios) (Marshall et al.1998). Confidence levels were set at the lower limit of

80% and the upper limit of 95%. The program run 10 000 iterations with an error rate of mis-scores set to 2% and the proportion of alleles typed at 0.91 and sires sampled set at

0.85 for this analysis. This parameter was set as it is acknowledged that some potential sires may not have been sampled, or may have died from DFTD prior to sampling. Only positive delta scores were used to infer paternity.

Paternity analysis was also scored independently, using the exclusion method, by two members of the research group, PhD candidate Tracey Russell and postdoctoral fellow Dr

Amanda Lane. Statistical analyses were conducted in JMP Pro 11.

3.4 Results

3.4.1 Genetic analyses

Within the adult devils sampled, four of the seven MHC linked loci (Sh102, Sh108, Sh110 and Sh111) showed increased levels of observed relative to expected heterozygosity (Table

3.1), whereas in the offspring only three of these four MHC linked loci (Sh102, Sh110 and

Sh111) displayed higher levels of observed relative to expected heterozygosity (Table 3.1).

This deviation from H.W.E. was significant in only one MHC linked locus (Sh108). One neutral locus, Sh2v, also deviated significantly from H.W.E. (Table 3.1). In three of the eight MHC-linked loci (Sh101, Sh105 and Sh106), lower observed heterozygosity than expected was found in the adult devils (Table 3.1); whereas in the offspring only two of the eight loci (Sh101 and Sh106) showed lower levels of observed relative that of expected heterozygosity (Table 3.1). MHC linked loci Sh108 could not be scored among the

Faculty of Science The University of Sydney offspring due to amplification failure (Table 3.1). Neutral locus Sh2l and MHC linked locus Sh105 were monomorphic in the devil offspring (Table 3.1).

Faculty of Science The University of Sydney

Table 3.1 Standard diversity indices and Hardy-Weinberg equilibrium. Exact test was conducted by using a Markov chain for all loci (forecasted chain length 1 000 000, dememorisation steps 100 000). * significant p < 0.05, ** significant p < 0.05 after sequential Bonferroni correction. Sh101-Sh111 (⁑) refers to the 7 MHC-linked loci and Sh2g-Sh5c refers to the 8 neutral microsatellite loci. Ho indicates observed heterozygosity, He indicates expected heterozygosity, H.W.E. p-values indicate deviation from Hardy-Weinberg Equilibrium.

Adults Offspring Locus No. No. Ho/He H.W.E. No. No. Ho/He H.W.E. gene of p-value gene of p-value copies alleles (s.d.) copies alleles (s.d.) 1Sh101 50 2 0.28/0.30 1 54 2 0.30/0.31 1 (0.00)

2Sh102 50 2 0.28/0.25 1 54 2 0.30/0.26 1 (0.00)

3Sh105 40 2 0.00/0.09 0.03* Monomorphic 4Sh106 46 3 0.35/0.58 0.01* 54 3 0.44/0.53 0.17 (0.00) 5Sh108 40 4 0.85/0.63 0.00** Monomorphic - due to lack of amplification 6Sh110 36 3 0.88/0.64 0.16 28 5 0.86/0.72 0.12 (0.00) 7Sh111 50 3 0.76/0.66 0.12 54 3 0.70/0.59 0.75 (0.00) 8Sh2g 50 3 0.68/0.54 0.44 54 2 0.15/0.14 1 (0.00)

9Sh6l 48 2 0.29/0.44 0.15 54 2 0.44/0.35 0.29 (0.00) 10Sh2v 48 4 0.17/0.55 0.00** 48 4 0.62/0.54 0.43 (0.00) 11Sh2l 22 2 0.09/0.09 1 Monomorphic 12Sh3a 18 2 0.44/0.52 1 24 2 0.50/0.51 1 (0.00)

13Sh2i 48 2 0.00/0.08 0.02* 46 2 0.04/0.04 1 (0.00)

14Sh3o 40 4 0.35/0.51 0.05* 24 2 0.08/0.34 0.03 (0.00) 15Sh5c 42 3 0.29/0.26 1 38 3 0.32/0.28 1 (0.00)

Mean 41.87 2.73 0.38 0.41 44.33 2.67 0.40/0.39 0.65 s.d. 10.01 0.79 0.29 0.21 12.47 0.99 0.25/0.20 0.00

Faculty of Science The University of Sydney

Linkage disequilibrium was high between MHC–linked microsatellites which was expected

due to their locality of the MHC genes on chromosome 4q (Lane et al. 2012) (Table 3.2).

MHC linked markers Sh101, Sh102, Sh105, Sh106, Sh110 and Sh111 are closely linked to

the MHC Class I loci Saha-UA, Saha-UB, Saha UC and Saha UD, whereas Sh108 is linked

to the non-antigen presenting MTCH1 and FGD2 genes (Table 3.2).

Table 3.2 Significant linkage disequilibrium (+ indicates significance = 0.05) for 9 adult females and16 adult male devils.

Locus Sh101 Sh102 Sh105 Sh106 Sh108 Sh110 Sh111 Sh2g Sh6l Sh2v Sh2l Sh3a Sh2i Sh3o Sh5c # Sh101 * - + + - - - - + ------Sh102 - * - - - - + ------+ Sh105 + - * + - - - - + - - - - - + Sh106 + - + * - - - + + + - - - - - Sh108 - - - - * + + + + ------Sh110 - - - - + * + + ------Sh111 - + - - + + * - - + + - - + - Sh2g - - - + + + - * + ------Sh6l + - + + + - - + * - + - - - - Sh2v - - - + - - + - - * + - + - - Sh2l ------+ - - + * - + - - Sh3a ------* - - - Sh2i ------+ + - * + - Sh3o ------+ * + Sh5c - + + ------*

3.4.2 Paternity analyses

The nine female devils produced 28 offspring (average litter size of 3.11 ± SE 0.6). One

joey was excluded from all analyses due to degraded DNA resulting in genotyping failure.

Due to common alleles within all of the devils across a number of loci, multiple paternity

could only be assigned to four of the nine litters with a confidence level of 95%. One of

the litters (n = 4 joeys) with multiple paternity was sired by three different males whereas

the remaining three multi-sired litters (n = 3, n = 3, n = 2 joeys) were sired by two different

males. While nine of the joeys were assigned a sire with 95% confidence, the exact sire of

three joeys from the multi-sired litters remained undetermined (Table 3.3).

Faculty of Science The University of Sydney

Table 3.3 Multiple paternity within four litters of wild Tasmanian devils. Allocation of sires (A-P) (n = 16) to offspring (n = 9) within the four litters showing multiple paternity. Where more than one letter is listed for the candidate sire, a sire could not be distinguished between the males listed by the corresponding letter. However, multiple paternity could still be established within the litter. Male weights at capture are also provided. Litter Offspring no. Allocated sires (A – Male weights kg) no. P) 1 1 A, H 6.1 (A), 5 (H) 1 2 F 4.8

1 3 F 4.8 1 4 F 4.8 2 1 G 5.2 2 2 A,B,C 6.1 (A), 5.9 (B), 6.7 (C) 2 3 A,B,C 6.1 (A), 5.9 (B), 6.7 (C) 3 1 I 6.8 3 2 P 8.1 3 3 I 6.8 4 1 F 4.8

4 2 G 5.2

Although the sample size is small, neither female weight nor litter size differed

significantly among the single and multiple sired litters (Wilcoxon rank scores: Z = 0.25, p

= 0.81 and Z = 0.29, p = 0.77; mean weight of females with single sired litters: 5.62 kg, SE:

±0.56, n = 5; mean weight of females with multiple sired litters: 5.90 kg SE: ±0.63, n = 4;

mean number of offspring in single sired litters: 3.2 SE: ±0.28; mean number of offspring

in multiple sired litters: 3.0 SE: ±0.32).

There was no correlation between number of offspring sired and male body weight

(Spearman rank correlation: R = -0.28, p = 0.30, n = 16) (Table 3.3).

3.4.3 Age

Of the 16 male devils, 13 were one year olds, two were two years old and one was three years old (average 1.0 ± S.D. 0.58). The age of the nine breeding females ranged from 1-3 years with four females aged one, four aged two and one aged three (average 1.67 ± S.D.

0.71). Of the nine offspring in which paternity could be assigned with confidence three of

Faculty of Science The University of Sydney the sires were one year old males (male F, G and I in table 3.3). Moreover, male A, B, C and H (Table 3.3) were also one year old males. These results confirm that one-year old males are able to successfully reproduce in the wild and hence constitute the first records precocial breeding in male devils.

Figure 3.1 Weight (kg) ± S.E. for adult male and female devils from the Forestier remnant population. (Males 2009 n =1, 2010 n = 2, 2011 n = 12, Females 2010 n = 7, 2011 n = 9).

3.5 Discussion

Allelic differences between parents and offspring within this dataset showed an increase

in numbers of alleles in offspring at the MHC-linked locus, Sh110, and a decrease in

numbers of alleles in the offspring at the neutral markers Sh2g and Sh3o. Similar to the

present study, Thonhauser et al. (2013) observed allelic increases in multiple sired litters

in house mice (Mus musculus) with no increase in heterozygosity. As Tasmanian devils

suffer from low levels of genetic MHC diversity (Siddle et al. 2010), any allelic increase

Faculty of Science The University of Sydney

in offspring MHC diversity is beneficial, as it may significantly contribute to

maintaining genetic diversity on this important genomic region. Over ten percent of

linkage disequilibrium tests involving neutral loci were significant. Although these

neutral microsatellites are assumed to be neutral, it is probable that they are in close

proximity to coding regions that could be under selection.

The very limited diversity of both MHC genes (Siddle et al. 2010; Morris et al. 2013) and

at neutral microsatellite loci (Jones et al. 2003) severely restricted the paternity assignments

in the present study. However, due to the low among-male genetic divergence it is possible

that some (or even all) of the five litters which were indeterminable, could have been multi-

sired, although siring by a single male cannot be discounted. Recently, additional

microsatellite markers have been developed for the Tasmanian devil (Gooley et al. 2017),

which when used for paternity determination, will increase the power to resolve questions

of paternity and heredity.

In spite of the relatively low variation of the molecular markers used in the present study, multiple paternity was observed in four out of the nine litters examined, the first time it has been recorded in wild Tasmanian devils. Recently, however, multiple paternity has also been identified from three free ranging captive devil facilities where females have had access to multiple males (T. Russell unpublished data), strongly suggesting that multiple matings and paternity may be common reproductive strategies employed by female devils.

Additionally, for the first time breeding of one year old males (precocial breeding) was observed, which had previously only been reported in female devils (Lachish et al. 2009).

There are a number of factors facilitating these behaviours including: the loss of older

cohorts to DFTD leading towards a younger age structure, and the depopulation activities

Faculty of Science The University of Sydney by DPIPWE staff which thus allowed young males access to females who would normally have been mate guarded by older/dominant males. Additionally, with fewer devils in the area due to the abovementioned factors, food availability would increase concurrent with reduced competition. Males could then achieve greater growth rates and body mass, allowing them to reach the critical weight required for mating success (Lachish et al. 2009).

Indeed, although the successfully reproducing young males were only one year old in this study, they have reached sexual maturity. Consequently, the demographic effects of DFTD appear to result in significant changes in devil life-history which may provide the species the potential to adapt to the devastating effects of this disease.

Multiple paternity is, however, not restricted to Tasmanian devils but has also been observed in numerous other marsupials, such as the honey possum (Tarsipes rostratus;

Wooller et al. 2000), agile antechinus (Antechinus agilis; Shimmin et al. 2000; Kraaijeveld-

Smit et al. 2002) brown antechinus (Antechinus stuartii; Holleley et al. 2006), swamp antechinus (Antechinus minimus; Sale et al. 2013); red-tailed phascogale (Phascogale calura; Foster et al. 2008), feathertail glider (Acrobates pygmaeus; Parrott et al. 2005) and the spotted tailed quoll (Dasyurus maculatus; Glen et al. 2009). Interestingly, the number of young per litter was found to correlate positively with the number of sires in the agile antechinus (Antechinus agilis): a possible explanation for this was suggested to be due to low numbers of male spermatozoa observed in the males of this species (Taggart and

Temple-Smith 1990a; Taggart and Temple-Smith 1990b; Taggart et al. 1999).

Additionally, due to the effective transport and storage of sperm in A. agilis, sperm quantity does not compromise fertilisation rates (Taggart et al. 1999), whereas sperm concentration in the devil was found to be variable and the motility low (Keeley et al. 2012b). Shimmin et al. (2000) discovered that paternity in A. agilis was overwhelmingly assigned to the second males who mated with females as opposed to the dominant males who had mated first,

Faculty of Science The University of Sydney proving that mate guarding does not guarantee paternity in this species. While the largest

(dominant) males may obtained more copulations with females, this did not equate to dominant paternity representation (Shimmin et al. 2000). This effect was also seen in this study, where smaller males sired offspring.

During the breeding season, male devils can lose up to 25% of their body weight (Jones et al. 2008) so it is likely that some of the more actively siring males in this study may have undergone significant weight loss prior to capture. However, in contrast to another carnivorous marsupial species, the spotted-tailed quoll (Dasyurus maculatus) (Glen et al.

2009), the size of the male devils did not affect male reproductive success. Similar results have been obtained in studies of brown antechinus where young were found to be equally sired by both large and small males (Shimmin et al. 2000; Holleley et al. 2006). Moreover, in two populations of eastern grey kangaroos (Macropus giganteus), a sexually dimorphic species, no reproductive skew was found favouring larger males, with some sires weighing almost half (33kg) the weight of others (62kg) (Rioux-Paquette, 2015).

As previously stated, several adaptive explanations for polyandry have been proposed based on direct and indirect female benefits. Since male devils that sire offspring do not participate in the upbringing of young devils, it is therefore unlikely that direct benefits underpin female mating behavior. In a recent study, however, Russell et al (2017) found that in captive devils, pairs with opposing numbers of heterozygous MHC linked microsatellites had a higher reproductive success compared to pairs with similar number of heterozygous alleles: lending support to a tentative suggestion that female devils are indeed exploiting post-copulatory strategies such as cryptic choice of male sperm to minimise the risk and/or cost of fertilisation by genetically incompatible sperm.

Faculty of Science The University of Sydney

Fisher et al. (2006) showed that in the brown antechinus (Antechinus stuartii), offspring born to polyandrous females had a threefold increase in survival to weaning compared to monoandrous females. Similar results were obtained in a study using mice where the post- natal survival of pups increased in multi-sired litters relative to those born from single sired litters (Thonhauser et al. (2013). Unfortunately, in the present study we do not have any data on the subsequent survival of the devils from single and multiple sired litters and hence do not know whether multiple paternity may increase offspring survival post- weaning in Tasmanian devils in the wild.

3.6 Conclusion

Whilst the conclusions in this studies were the result of examining a small population of devils and due to their genetic similarities, the results were not conclusive for all offspring, this is the first study to unambiguously show multiple paternity and male precocial breeding in wild Tasmanian devils. However, further research is certainly required to determine the frequency of multiple paternity in wild devil litters and also within the captive breeding program where some devils are housed in large free ranging enclosures with access to more than one potential mate. In order to increase the accuracy of paternity assignment, using a broader range of molecular markers in this species, with such low genetic diversity remains a priority. Finally, the results have implications for the

Tasmanian devil captive insurance program as introducing numerous males to females is likely to result in multiple paternity which may increase offspring genetic diversity and importantly enhance the reproductive success of the captive devils.

Faculty of Science The University of Sydney

3.7 Acknowledgements

This work was supported by a grant to R.S. from the Morris Animal Foundation. We would like to thank members of DPIPWE and the STTDP for supplying samples to T.R.

Faculty of Science The University of Sydney

3.8 References

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

Cheng, Y. and Belov, K. (2012) Isolation and characterisation of 11 MHC-linked microsatellite loci in the Tasmanian devil (Sarcophilus harrisii). Conservation Genetic

Resources 4, 463-465.

Ebensperger, L. A. and Blumstein, D. T. (2007). Functions of non-parental infanticide.

In Rodent Societies (eds. J. O. Wolff and P. W. Sherman), pp. 267–269. University of

Chicago Press, Chicago.

Excoffier, L. and Lischer, H. E. L. (2010) Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows.

Molecular Ecology Resources 10, 564-567.

Fisher, D. O., Double, M. C., Blomberg, S. P., Jennions, M. D. and Cockburn, A.

(2006) Post-mating sexual selection increases lifetime fitness of polyandrous females in the wild. Nature 444 (2), 89-92.

Foster, W. K., Caton, W., Thomas, J., Cox, S. and Taggart, D. A. (2008) Timing of births and reproductive success in captive red-tailed phascogales, Phascogale calura.

Journal of Mammalogy 89 (5), 1136-1144.

Faculty of Science The University of Sydney

Linnean Society 96, 1-7.

Gooley, R., Hogg, C. J., Belov, K. and Grueber, C. E. (2017) No evidence of inbreeding depression in a Tasmanian devil insurance population despite significant variation in inbreeding. Scientific Reports 7-1830. DOI:10.1038/s41598-017-02000-y

Griffith, S. C., Owens, I. P. F. and Thuman, K. A. (2002) Extra pair paternity in birds: a review of interspecific variation and adaptive function. Molecular Ecology 11 (11),

2195-2212.

Guiler, E. R. (1970) Observations on the Tasmanian devil, Sarcophilus harrisii

(Marsupialia:Dasyuridae) II Reproduction, breeding, and growth of pouch young.

Australian Journal of Zoology 18, 63-70.

Hamede, R.K., McCallum, H. and Jones, M. (2008) Seasonal, demographic and density related patterns of contact between Tasmanian devils (Sarcophilus harrisii):

Implications for transmission of devil facial tumour disease. Austral Ecology 33, 614-

622.

Hawkins, C. E., Baars, C., Hesterman, H., Hocking, G. J., Jones, M. E., Lazenby, B.,

Mann, D., Mooney, N., Pemberton, D., Pyecroft, S., Restani, M. and Wiersma, J. (2006)

Emerging disease and population decline of an island endemic, the Tasmanian devil

Sarcophilus harrisii. Biological Conservation 131, 307-324.

Faculty of Science The University of Sydney

Hendricks, S., Epstein, B., Schönfeld, B., Wiench, C., Hamede, R., Jones, M., Storfer,

A. and Hohenlohe, P. (2017) Conservation implications of limited genetic diversity and population structure in Tasmanian devils (Sarcophilus harrisii). Conservation

Genetics, 18: 4, 977-982.

Hesterman, H., Jones, S. M. and Schwarzenberger, F. (2008a) Reproductive endocrinology of the largest dasyurids: Characterization of ovarian cycles by plasma and fecal steroid monitoring. Part I. The Tasmanian devil (Sarcophilus harrisii).

General and Comparative Endocrinology 155, 234-244.

Hesterman, H., Jones, S. M. and Schwarzenberger, F. (2008b) Reproductive endocrinology of the largest Dasyurids: Characterization of ovarian cycles by plasma and fecal steroid monitoring. Part II. The spotted-tailed quoll (Dasyurus maculatus).

General and Comparative Endocrinology 155, 245-254.

Faculty of Science The University of Sydney

Hudson, C. M. and Fu, J. (2013) Male-biased sexual size dimorphism, resource defense polygyny and multiple paternity in the Emei Moustache Toad (Leptobrachium boringii).

PLoS One 8(6), 1-8.

Jennions, M. D. and Petrie, M. (2000) Why do females mate multiply? A review of the genetic benefits. Biology Letters 75, 21-64.

Jones, M. E., Paetkau, D., Geffen, E. and Moritz, C. (2003) Microsatellites for the

Tasmanian devil (Sarcophilus laniarius). Molecular Ecology Notes 3, 277-279.

Jones, M. E., Cockburn, A., Hamede, R., Hawkins, C., Hesterman, H., Lachish, S.,

Mann, D., McCallum, H. and Pemberton, D. (2008) Life-history change in disease- ravaged Tasmanian devil populations. PNAS 105 (29), 10023-10027.

Kalinowski, S., Taper, M. L. and Marshall, T. C. (2010) Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment. Molecular Ecology 16, 1099-1106.

100

Faculty of Science The University of Sydney

Keeley, T., O’Brien, J. K., Fanson, B. G., Masters, K. and McGreevy, P. D. (2012a) The reproductive cycle of the Tasmanian devil (Sarcophilus harrisii) and factors associated with reproductive success in captivity. General and Comparative Endocrinology 176,

182-191.

Keeley, T., McGreevy, P. D. and O’Brien, J.K. (2012b) The effects of season and devil facial tumour disease on the reproductive physiology of the male Tasmanian devil

(Sarcophilus harrisii). Reproduction, Fertility and Development 24, 999-1007.

Kelly, A. (2007). Husbandry guidelines for the Tasmanian devil, Sarcophilus harrisii.

In S. Brice (ed.), www.zooaquarium.org.au. Sydney Zoo and Aquarium Association pp.

1-32.

Klemme, I., Eccard, J. A. and Ylönen, H. (2007) Why do female bank voles,

Clethrionomys glareolus, mate multiply? Animal Behaviour 73, 623-628.

Kraaijveld-Smit, F. J. L., Ward, S. J. and Temple-Smith, P. D. (2002) Multiple paternity in a field population of a small carnivorous marsupial, the agile antechinus, Antechinus agilis. Behavioral Ecology and Sociobiology 52, 84-91.

Lachish, S., McCallum, H., and Jones, M. (2009) Demography, disease and the devil: life-history changes in a disease-affected population of Tasmanian devils (Sarcophilus harrisii). Journal of Animal Ecology 78, 427-436.

101

Faculty of Science The University of Sydney

Lane, A., Cheng, Y., Wright, B., Hamede, R., Levan, L., Jones, M., Ujvari, B. and

Belov, K. (2012) New insights into the role of MHC diversity in Devil Facial Tumour

Disease. PLoS One 7, 6, e36955.

Liu, X., Tatarenkov, A., O’Rear, T. A., Moyle, P. B. and Avise, J. C. (2013). Molecular evidence for multiple paternity in a population of the viviparous Tule Perch

Hysterocarpus traski. Journal of Heredity 104 (2), 217-222.

McCallum, H., Jones, M., Hawkins, C., Hamede, R., Lashich, S. (2009) Transmission dynamics of Tasmanian devil facial tumor disease may lead to disease-induced extinction. Ecology 90, 3379-3392.

McLeod, L. and Marshall, D. J. (2009) Do genetic diversity effects drive the benefits associated with multiple mating? A test in a marine invertebrate. PloS ONE 4 (8), e

6347.

Madsen, T., Shire, R., Loman, J. and Håkansson, T. (1992) Why do female adders copulate so frequently? Nature 355, 440-441.

Marshall, T. C., Slate, J., Kruuk, L. E. B. and Pemberton, J. M. (1998) Statistical confidence for likelihood based paternity inference in natural populations. Molecular

Ecology 7, 639-655.

102

Faculty of Science The University of Sydney

Miller, W., V. M. Hayes, A. Ratan, D. C. Petersen, N. E. Wittekindt, J. Miller, B.

Walenz, J. Knight, J. Qi, F. Zhao, Q. Wang, O. C. Bedoya-Reina, N. Katiyar, L. P

Tomsho, L. M. Kasson, R. A. Hardie, P. Woodbridge, E. A. Tindall, M. F. Bertelsen, D.

Dixon, S. Pyecroft, K. M. Helgen, A. M. Lesk, T. H. Pringle, N. Patterson, Y. Zhang, A.

Kreiss, G. M. Woods, M. E. Jones, and S. C.Schuster. 2011. Genetic diversity and population structure of the endangered marsupial Sarcophilus harrisii (Tasmanian devil). Proceedings of the National Academy of Sciences 198,12348-12353.

Morris, K., Austin, J. J. & Belov, K. (2013) Low major histocompatibility complex diversity in the Tasmanian devil predates European settlement and may explain susceptibility to disease epidemics Biology Letters 9, 20120900.

Olsson, M., Shine, R., Madsen, T., Gullberg, A. and Tegelstrom, H. (1996) Sperm selection by females. Nature 383, 585.

Olsson, M., Madsen, T., Wapstra, E., Silverin, B., Ujvari, B. and Wittzell, H. (2005)

MHC, health, color, and reproductive success in sand lizards. Behavioural Ecology

Sociobiology, 58, 289–294.

Parrot, M. L., Ward, S. J. and Taggart, D. A. (2005) Multiple paternity and communal maternal care in the feathertail glider (Acrobates pygmaeus). Australian Journal of

Zoology 53, 79-85.

Pemberton, D., 1990. Social organization and behaviour of the Tasmanian devil,

Sarcophilus harrisii, Zoology University of Tasmania, Hobart.Raymond, M. and

103

Faculty of Science The University of Sydney

Rousset, F. (1995) GENEPOP (version 1.2): population genetics software for exact tests and ecumenicism. Journal of Heredity 86, 248-249.

Rioux-Paquette, E., Garant, D., Martin, A. M., Coulson, G. and Festa-Bianchet, M.

(2015) Paternity in eastern grey kangaroos: moderate skew despite strong sexual dimorphism. Behavioral Ecology 26 (4), 1147-1155.

Russell, T., Lisovski, S., Olsson, M., Spindler, R., Lane, A., Keeley, T., Hibbard, C.,

Hogg, C., Thomas, F., Belov, K., Ujvari, B. and Madsen, T. (2018) MHC diversity and female age underpin reproductive success in an Australian icon; the Tasmanian Devil.

Scientific Reports 8, 4175.

Sale, M. G., Kraaijeveld-Smit F. J. L. and Arnould, J. P. Y. (2013) Multiple paternity in the swamp antechinus (Antechinus minimus) Australian Mammalogy 35, 227–230.

Save the Tasmanian Devil Program (STTDP) (2017) Accessed 12 June 2017.

Shimmin, G. A., Taggart, D. A. and Temple-Smith, P. D. (2000) Sperm competition and genetic diversity in the agile antechinus (Dasyuridae: Antechinus agilis). Journal of

Zoology 252, 343-350.

Shimmin, G. A., Taggart, D. A. and Temple-Smith, P. D. (2002) Mating behaviour in the agile antechinus Antechinus agilis (Marsupialia: Dasyuridae). Journal of Zoology

258, 39-48.

104

Faculty of Science The University of Sydney

Siddle, H., Marzec, J., Cheng, Y., Jones, M. and Belov, K. (2010) MHC gene copy number variation in Tasmanian devils: implications for the spread of a contagious cancer. Proceedings of the Royal Society B. 277, 2001-2006.

Taggart, D. A., Shimmin, G. A., Dickman, C. R. and Breed, W. G. (2003) Reproductive biology of carnivorous marsupials: clues to the likelihood of sperm competition. In

Jones, M. E., Dickman, C. R. and Archer, M. (eds.) Predators with pouches: the biology of carnivorous marsupials. CSIRO Publishing pp. 358-375. Victoria 3066. Australia.

Taggart, D. A. and Temple-Smith, P. D. (1990a) An unusual mode of progressive motility in spermatozoa from the dasyurid marsupial, Antechinus stuartii. Reproduction,

Fertility and Development -2, 107-114.

Taggart, D. A. and Temple-Smith, P. D. (1990b) Effects of breeding season and mating on total number and distribution of spermatozoa in the epididymis of the brown marsupial mouse, Antechinus stuartii. Journal of Reproduction and Fertility 88, 81-91.

Taggart, D. A. and Temple-Smith, P. D. (1991) Transport and storage of spermatozoa in the female reproductive tract of the brown marsupial mouse, Antechinus stuartii

(Dasyuridae). Journal of Reproduction and Fertility 93, 97-110.

Taggart, D. A., Shimmin, G. A., McCloud, P. and Temple-Smith, P. D. (1999) Timing of mating, sperm dynamics and ovulation in a wild population of agile antechinus

(Marsupialia: Dasyuridae) Biology of Reproduction 60, 283-289.

105

Faculty of Science The University of Sydney

Townsend, A. K., Clark, A. B. and Mcgowan, K. J. (2010). Direct benefits and genetic costs of extra pair paternity for female American crows (Corvus brachyrhynchos). The

American Naturalist 175, 1–9.

Woods, G.M., Kreiss, A., Belov, K., Siddle, H.V., Obendorf, D., and Muller, H. K. (2007)

The immune response of the Tasmanian devil (Sarcophilus harrisii) and the Devil Facial

Tumour Disease. EcoHealth 4, 338-345.

Woods, G. M., Howson, L. J., Brown, G. K., Tovar, C., Kreiss, A., Corcoran, L. M. and

Lyons, A. B. (2015) Immunology of a Transmissible Cancer Spreading among

Tasmanian Devils. The Journal of Immunology 195, 23-29.

Wooller, R. D., Richardson, K. C., Garavanta, C. A. M., Saffer, V. M. and Bryant, K. A.

(2000) Opportunistic breeding in the polyandrous honey possum, Tarsipes rostratus.

Australian Journal of Zoology, 48, 669-680.

Zeh, J. A. and Zeh, D. W. (1996) The evolution of polyandry I: intragenomic conflict and genetic incompatibility. Proceedings of the Royal Society of London B 263, 1711-

1717.

106

Faculty of Science The University of Sydney

4 Chapter four: MHC diversity and female age underpin

reproductive success in an Australian icon; the Tasmanian

Devil

Published in Scientific Reports 2018 8:4175, DOI:10.1038/s41598-018-20934-9

Tracey Russell1, Simeon Lisovski2, Mats Olsson3, Gregory Brown1, Rebecca Spindler4, Amanda Lane1, Tamara Keeley5, Chris Hibbard6, Carolyn J. Hogg1, Frédéric Thomas7, Katherine Belov1, Beata Ujvari8 and Thomas Madsen8,9* 1School of Life and Environmental Sciences, University of Sydney, Sydney, 2006, NSW, Australia

2University of California Davis, Department of Neurobiology Physiology and Behavior, Davis, CA 95616, USA

3Dept of Biological and Environmental Sciences University of Gothenburg, Box 463, SE 405 30 Gothenburg, Sweden

4Taronga Conservation Society Australia, Mosman, 2088, NSW, Australia

5School of Agriculture and Food Science, University of Queensland, Gatton, 4343, QLD, Australia

6Zoo and Aquarium Association Australasia, Mosman, 2088, NSW, Australia

7CREEC/ MIVEGEC, UMR IRD/CNRS/UM 5290, 911 Avenue Agropolis, BP 64501, 34394 Montpellier Cedex 5, France

8Centre for Integrative Ecology, School of Life and Environmental Science, Deakin University, Geelong, 3216, VIC, Australia

9School of Biological Sciences, University of Wollongong, 2522, NSW, Australia

*Corresponding author: Thomas Madsen email: [email protected]

107

Faculty of Science The University of Sydney

4.1 Abstract

Devil Facial Tumour Disease (DFTD), a highly contagious cancer, has decimated

Tasmanian devil (Sarcophilus harrisii) numbers in the wild. To ensure its long-term survival a captive breeding program was implemented but has not been as successful as envisaged since its launch in 2005. We therefore investigated the reproductive success of

65 captive devil pair combinations, of which 35 produced offspring (successful pairs) whereas the remaining 30 pairs, despite being observed mating, produced no offspring

(unsuccessful pairs). The devils were screened at six MHC Class I-linked microsatellite loci. Our analyses revealed that younger females had a higher probability of being successful than older females. In the successful pairs we also observed a higher difference in total number of heterozygous loci i.e. when one devil had a high total number of heterozygous loci, its partner had low numbers. Our results therefore suggest that devil reproductive success is subject to disruptive MHC selection which to our knowledge has never been recorded in any vertebrate. In order to enhance the success of the captive breeding program the results from the present study show the importance of using young

(2-year old) females as well as subjecting the devils to MHC genotyping.

Key words: MHC diversity, female age, reproductive success, Sarcophilus harrisii

108

Faculty of Science The University of Sydney

4.2 Introduction

The major histocompatibility complex (MHC) has been shown to be one of the key molecular determinants of mate choice in numerous vertebrates including humans1-7. Class

I and II MHC molecules are responsible in the processing and presentation of intra- and extra-cellular peptides derived from invading pathogens to cytotoxic T cells and helper T cells and, hence, constituting a crucial part of the vertebrate immune system8,9. Due to the ability to recognize and present peptides from a wide array of rapidly evolving pathogens, the MHC encompasses the most variable set of genes with heterozygosity values exceeding those predicted by neutrality10.

Pathogen-driven selection has been documented to result in two, not mutually exclusive, MHC responses: (i) selection for specific MHC alleles11-14 and/or (ii) selection for enhanced MHC polymorphism13-16. Although increased levels of MHC polymorphism enable wider recognition of pathogens, it might also lead to an inability to eliminate T cells reacting with self-peptide-MHC combinations17,18. Consequently, diversifying MHC selection might be counteracted by the deleterious effects caused by autoimmunity19.

Indeed, theoretical models as well as empirical studies have shown that optimal pathogen resistance often occurs at an optimal intermediate level of MHC polymorphism13,20-27.

MHC has also been shown to be involved in individual mate choice and providing offspring with indirect genetic benefits in at least three ways via (i) acquisition of “good genes” i.e. genetic elements that contribute to lifetime reproductive success regardless of an individual’s additional genotype28, and/or (ii) acquiring optimal genetic compatibility29 and/or (iii) achieving enhanced genetic diversity30.

Importantly, significant reduction in MHC diversity may not only affect resistance to pathogens but has also been shown to result in an increased risk of extinction due to inbreeding depression31. Moreover, reduced MHC polymorphism has been suggested to

109

Faculty of Science The University of Sydney have contributed to the emergence of a clonally transmissible cancer, Devil Facial Tumour

Disease (DFTD) in the world’s largest living marsupial carnivore: the Tasmanian devil

(Sarcophilus harrisii)32. DFTD was first reported in north-eastern Tasmania in 1996 but since then has caused severe reductions in devil numbers (> 70 %) questioning the long- term survival of this iconic species33,34. This devastating disease is spread among devils via biting during social interactions35. The devil’s immune system is unable to mount an effective immune response to the tumour as DFTD cells are able to avoid immune recognition by down-regulating MHC expression36,37. Metabolic failure, tumour related cachexia and metastases result in devil death within 6 to 9 months of the emergence of the first lesions38. In order to ensure that the Tasmanian devil will not face extinction a large scale captive breeding program commenced in 200539, but overall breeding success is still suboptimal40,41. In the present study we therefore explore how traits such as devil age and

MHC polymorphism affected reproductive success in the captive devil population.

4.3 Material and methods

4.2.1 Study animals

Data on pairings were obtained from the Tasmanian devil studbook69 and the insurance population annual reports59-65. Single female devils were paired with a single male when females showed signs of estrus such as a fluid filled neck roll, fat stores in the tail, reduced aggression and loss of appetite70. Consequently, the females were not given a choice of partner. The 37 females and 43 males used in the present study were kept across 11 institutions in Australia and the pairings were conducted from 2007 to 2012. All ear biopsies were collected by veterinarians and trained staff during the devil’s annual exams at the zoological institutions where they were housed and were hence carried out in accordance with relevant guidelines and regulations. The research was conducted under

University of Sydney animal ethics approval number N00/8-2011/1/5584.

110

Faculty of Science The University of Sydney

4.2.2 Genetic analyses

DNA was extracted from ear biopsies using a Qiagen DNeasy® extraction kit and subsequently genotyped at six MHC Class I linked microsatellite loci described by Cheng

& Belov71, Sh-MHCI101, Sh-MHCI102, Sh-MHCI105, Sh-MHCI106, Sh-MHCI107, and

Sh-MHCI108. For further details on size range and primer sequences see Cheng &

Belov71. Polymerase chain reactions (PCR) were carried out using the protocols of Cheng

& Belov71 and the PCR products were quality tested on a 2% agarose gel. The samples were analyzed at the Ramaciotti Centre (University of New South Wales, Australia) and microsatellite profiles of the individual devils were subsequently scored using Peak

Scanner 2.0 (Applied Biosystems 2012).

4.2.3 Estimates of allele frequency, heterozygosity and linkage disequilibrium

Analyses of dropout and the presence of null alleles were conducted using the software

Micro-Checker72. The number of alleles per locus together with observed and expected

73 heterozygosity were estimated using the software program ARLEQUIN 3.1 . Tests for deviations from Hardy–Weinberg equilibrium and linkage disequilibrium, as well as the analysis of genetic structure and differentiation of the two groups were carried out using

73 ARLEQUIN 3.1 (HWE parameters included: number of steps in Markov chain = 1,000,000, number of dememorisation steps = 100,000, number of permutations = 10,000, AMOVA parameters included 999 permutations and 3000 Markov steps). Sequential Bonferroni corrections were subsequently conducted on the Hardy-Weinberg equilibrium and the linkage disequilibrium tests. Within-pair genetic similarity was conducted by recording the number of shared/identical alleles in each of the six loci.

4.2.4 Statistical analyses

Logistic mixed models analyses were performed in SAS 9.4 using Proc GLIMMIX74,75 with institution and female ID number as random effects as 22 of the 36 females were

111

Faculty of Science The University of Sydney

paired on more than one occasion. Proc GLIMMIX uses restricted pseudolikelihoods and

the full model included male age and identification number, number of female

heterozygous loci, number of male heterozygous loci and number of similar alleles. The

final model included institution as random effect and absolute number of different

heterozygous loci and female age as fixed effects.

4.3 Results

Of a total of 65 captive devil pair combinations, 35 produced offspring whereas the

remaining 30 pair combinations did not produce any offspring (see Table 1 for a detailed

description of the pair combinations). The genetic diversity analyses did not reveal any

allele dropouts or null alleles.

Table 4.1 Female – male pair combinations

number of female Female Female number of male male male pair heterozygous age age loci (months) ID heterozygous loci (months) ID combination 4 48 FD1 1 48 MD1 successful 4 36 FD1 2 36 MD6 unsuccessful 4 36 FD1 3 36 MD21 unsuccessful 3 24 FD10 2 36 MD40 unsuccessful 4 36 FD11 1 60 MD8 successful 4 24 FD11 3 36 MD22 successful 6 48 FD12 1 60 MD8 successful 6 24 FD12 3 36 MD10 successful 3 24 FD13 4 24 MD15 unsuccessful 3 24 FD13 3 36 MD27 unsuccessful 4 36 FD14 2 24 MD30 unsuccessful 4 36 FD14 2 36 MD31 unsuccessful 4 36 FD14 3 24 MD37 unsuccessful 4 36 FD14 3 48 MD41 unsuccessful 4 24 FD15 1 24 MD32 successful 6 24 FD16 2 24 MD11 successful 6 48 FD16 3 48 MD43 successful 5 24 FD17 1 36 MD18 successful 5 36 FD17 2 36 MD31 successful 3 36 FD18 1 24 MD8 unsuccessful 3 24 FD18 4 48 MD15 successful

112

Faculty of Science The University of Sydney

1 24 FD19 4 36 MD5 successful 1 36 FD19 3 36 MD33 successful 5 24 FD2 2 36 MD29 successful 5 24 FD20 1 36 MD8 successful 5 48 FD20 1 36 MD18 unsuccessful 5 48 FD20 2 36 MD19 unsuccessful 2 24 FD21 3 24 MD34 successful 4 48 FD22 3 48 MD43 successful 5 24 FD23 0 24 MD36 successful 3 24 FD24 4 24 MD12 successful 1 36 FD25 1 60 MD1 unsuccessful 1 24 FD25 4 24 MD38 successful 1 36 FD25 3 36 MD39 unsuccessful 2 24 FD26 4 24 MD12 successful 5 36 FD27 4 48 MD3 unsuccessful 5 24 FD27 1 24 MD32 unsuccessful 1 24 FD28 4 36 MD17 successful 2 36 FD29 4 48 MD26 unsuccessful 2 48 FD29 2 48 MD31 unsuccessful 3 24 FD3 4 36 MD5 successful 5 36 FD30 2 24 MD40 successful 5 60 FD31 2 60 MD23 unsuccessful 3 36 FD32 6 48 MD16 unsuccessful 3 36 FD32 4 48 MD26 unsuccessful 3 36 FD33 2 36 MD4 successful 3 48 FD33 4 48 MD25 successful 3 24 FD34 4 36 MD3 successful 3 36 FD34 1 36 MD32 unsuccessful 1 24 FD35 4 24 MD12 successful 2 24 FD36 4 36 MD7 successful 3 24 FD37 1 24 MD8 unsuccessful 1 36 FD4 4 36 MD42 successful 4 48 FD5 3 36 MD13 successful 4 36 FD5 3 24 MD33 successful 3 48 FD6 4 48 MD2 unsuccessful 3 48 FD6 5 48 MD14 unsuccessful 3 48 FD6 5 48 MD35 unsuccessful 2 36 FD7 2 48 MD9 unsuccessful 2 24 FD7 2 36 MD20 successful 2 36 FD7 2 36 MD24 unsuccessful 5 24 FD8 4 36 MD38 successful 5 36 FD8 3 48 MD39 successful 3 60 FD9 2 60 MD23 unsuccessful 3 36 FD9 2 36 MD28 unsuccessful

113

Faculty of Science The University of Sydney

The full logistic mixed model revealed that male age, identification number, number of female heterozygous loci, number of male heterozygous loci and number of similar alleles did not affect devil reproductive success (P > 0.20), and those predictors were therefore subjected to backwards-elimination. The final model revealed that younger females had a higher probability of being successful than older females as indicated by a negative regression parameter estimate for female age = -0.1068 ± 0.04, SE, p = 0.012 (Table 2). In fact, the final analysis predicted a decrease in the chance of producing offspring from 0.68

(0.82-0.51, 95% confidence interval) for females at the age of 24 months to 0.19 (0.5-0.05) for females older than 60 months (Fig. 1).

Figure 4.1 Relationship between female age and the success of reproduction in Tasmanian devil

Points with error bars depict reproductive success (±95% CI) of pairs across female age.

Grey line shows the binomial model prediction of this relationship with the 95% confidence interval (grey polygon).

114

Faculty of Science The University of Sydney

Table 4.2 Final results of the logistic mixed model.

Parameter estimate Effect Estimate SE DF t Pr > |t| Intercept 1.9734 1.1183 62 1.76 0.0826 Difference in number of heterozygous 0.6204 0.2539 62 2.44 0.0174 loci Female age -0.0871 0.0316 62 -2.75 0.0077

Type III Tests of Fixed Effects

Den Effect Num DF F Pr > F DF Difference in number of heterozygous 1 62 5.97 0.017 loci Female age 1 62 7.58 0.008

As mentioned above, male age did not affect reproductive success. Interestingly, in

successful pairs we observed a higher difference in total number of heterozygous loci, i.e.

when one devil had a high total number of heterozygous loci, its partner had a low number,

as evident by a positive regression parameter estimate ( 0.8202 ± 0.3329, p = 0.0165).

Thus, pairs with opposing total numbers of heterozygous loci were found to have a higher

probability of being successful reproducers than pairs with similar numbers (Fig. 2).

115

Faculty of Science The University of Sydney

Figure 4.2 Mode prediction of the interaction between the number of male and female heterozygous loci. a) The color scale (within the data range) and contour lines (across the entire range of female and male heterozygous loci) indicate the probability of being successful according to the combination of the number of male and female heterozygous loci. b) Diagonal section of the model prediction shown in (a) across the optimal pair combinations with 95% confidence interval.

116

Faculty of Science The University of Sydney

4.4 Discussion

Prior to the emergence of DFTD female devils in the wild commenced reproduction at an age of two years and thereafter produced offspring for the next three years, becoming senescent at an age of five to six years33. The results from the present study, however, show that a decline in captive female reproductive success may already occur at an age of three years suggesting that some captive females might be affected by reproductive senescence at an earlier age than that recorded in the wild.

Previous studies have shown that devils are subjected to low genetic diversity at both neutral (microsatellite) and coding genomic regions as the MHC42-45. The low level of genetic polymorphism has been suggested to be caused by population bottlenecks during the Pleistocene and Holocene46,47 and by pathogens such as a canine-distemper-like disease in the early twentieth century44. The genetic diversity of the captive devils used in the present study was similar to that recorded in devils captured in the wild48. As we did not observe any overall significant difference in genetic diversity among the successful and the unsuccessful pair combinations we find it unlikely that the difference in reproductive success was caused by a concomitant discrepancy in genetic diversity.

In vertebrates such as sand lizards (Lacerta agilis)49, savannah sparrows

(Passerculus sandwichensis)50, southern dunlins (Calidris alpina schinzii)51, great tits

(Parus major)52 and the cynomolgus macaque (Macaca fascicularis)53 reproductive success has been negatively linked to male - female genetic similarity. However, the results from the present study suggest that within-pair genetic similarity did not affect devil reproductive success.

As mentioned above, both theoretical and empirical studies have shown that selection for an intermediate and optimal number of MHC alleles may result in both increased reproductive success and immune function13, 20-27. However, our results show

117

Faculty of Science The University of Sydney that pairs with higher difference in total number of heterozygous loci had an increased probability of being successful than pairs with similar numbers of heterozygous loci.

If devils have been under selection for an intermediate, optimal number of MHC alleles (e.g., 6 alleles), we would expect reproductive success in pairs partners with 3 heterozygous loci to be similar to that of pairs with 1 and 5 heterozygous partner loci. Our results therefore suggest that devil MHC may be subject to disruptive selection where extreme traits are reproductively favored54. Although similar MHC processes have been suggested to drive MHC evolution in allopatric taxa in different habitats55, it has to our knowledge, never been recorded to affect reproductive success.

We propose two, not mutually exclusive processes underpin the significant effect of selection for MHC-driven devil reproductive success: pre-copulatory and/or post- copulatory (cryptic) female choice of male sperm. As mentioned above MHC-based pre- copulatory mate choice, sometimes based on olfactory clues, has been recorded in wide range of vertebrates including humans1-7,56. MHC-based post-copulatory female cryptic choice of male sperm has also been documented in several vertebrates such as fish57, lizards49, birds58 and mammals59. As female devils in both the successful and unsuccessful pair combinations were observed mating60-66, we suggest that the significant difference in reproductive success between the two groups might be caused by post-copulatory cryptic female choice. In externally fertilizing fishes, the ovarian fluid released by the female may bias fertilization success towards males with a particular genotype67,68. However, if a similar mechanism may influence male fertilization success in internal fertilizers, such as mammals, is to our knowledge unknown.

4.5 Conclusion

118

Faculty of Science The University of Sydney

Our results suggest that the combination of female age (i.e. younger females having higher reproductive success compared to older females) and MHC diversity constitutes significant determinants of reproductive success in two groups of devils with different MHC traits and may provide an explanation for the low reproductive success so far recorded in captive breeding programs. In order to enhance the success of this iconic species, our results advocate (i) the use of young (2 year old) female devils, and (ii) that both male and female devils are subjected to MHC genotyping with pair combinations maximizing the total numbers of heterozygous loci at opposite ends of the heterozygosity scale in order to maximize breeding success.

119

Faculty of Science The University of Sydney

4.6 References

1. Bonneaud, C., Chastel, O., Federici, P, Westerdahl, H. & Sorci, G. Complex MHC-

based mate choice in a wild passerine. Proc. R. Soc. B 273, 1111–1116 (2006).

2. Griggio, M., Biard, C., Penn, D.J. & Hoi, H. Female house sparrows “count on” male genes:

experimental evidence for MHC-dependent mate preference in birds. BMC Evol. Biol. 11, 44

(2011).

3. Milinski, M. et al. Mate choice decisions of stickleback females predictably modified by

MHC peptide ligands. Proc. Natl. Acad. Sci. USA 102, 4414–4418 (2005).

4. Olsson, M. et al. Major histocompatibility complex and mate choice in sand lizards. Proc. R.

Soc. B 270, S254–S256 (2003).

5. Sin, Y.W. et al. MHC class II-assortative mate choice in European badgers Meles meles. Mol.

Ecol. 24, 3138–3150 (2015).

6. Wedekind, C. & Füri, S. Body odour preferences in men and women: do they aim for specific

MHC combinations or simply heterozygosity? Proc. R. Soc. B 264, 1471–1479 (1997).

7. Yamazaki, K. & Beauchamp, G.K. (2007). Genetic basis for MHC dependent mate choice.

Adv. Genet. 59, 129–145 (2007).

8. Klein, J. Natural history of the Major Histocompatibility Complex. (Wiley, New York, 1986).

9. Zinkernagel, R.M. Associations between major histocompatibility antigens and susceptibility

to disease. Ann. Rev. Microbiol. 33, 201-213 (1979).

10. Edwards, S.V. & Hedrick, P.W. Evolution and ecology of MHC molecules, from genomics

to sexual selection. Trends Ecol. Evol. 13, 305-311 (1998).

120

Faculty of Science The University of Sydney

11. Harf, R. & Sommer, S. Association between major histocompatibility complex class II DRB

alleles and parasite load in the hairy-footed gerbil, Gerbillus paeba, in southern Kalahari.

Mol. Ecol. 14, 85-91 (2005).

12. Lohm, J. et al. Experimental evidence for majorhistocompatibility complex-allele-specific

resistance to a bacterial infection. Proc. R. Soc. B 269, 2029-2033 (2002).

13. Madsen, T. & Ujvari, B. MHC Class I variation associates with parasite resistance and

longevity in tropical pythons. J. Evol. Biol. 19, 1973-1978 (2006).

14. Westerdahl, H.et al. Association between malaria and MHC genes in a migratory songbird.

Proc. R. Soc. B 272, 1511-1518 (2005).

15. Penn, D.J., Damjanovich, K. & Potts, W. MHC heterozygosity confers a selective advantage

against multiple-strain infections. Proc. Natl. Acad. Sci. USA 99, 11260-11264 (2002).

16. Thursz, M.R., Thomas, H.C., Greenwood, B.M. & Hill, A.V. Heterozygote advantage for

HLA class-II type in hepatitis B virus infection. Nature Genet. 17, 11-12 (1997).

17. Borghans, J.A., Noest, A.J. & De Boer, R.J. Thymic selection does not limit the individual

MHC diversity. Eur. J. Immunol. 33, 3353-3358 (2003).

18. Vidovic, D. & Matzinger, P. Unresponsiveness to a foreign antigen can be caused by self-

tolerance. Nature 336, 222-225 (1988).

19. Fernando, M.M.A. et al. Defining the role of the MHC in autoimmunity, a review and pooled

analysis. PLoS Genet. 4, e1000024 (2008).

20. Aeschlimann, P.B., Häberli, M.A., Reusch, T.B.H., Boehm, T. & Milinski, M. Female

sticklebacks Gasterosteus aculeatus use self-reference to optimize MHC allele number during

mate selection. Behav. Ecol. Sociobiol. 54, 119-126 (2003).

21. De Boer, R.J. & Perelson, A.S. How diverse should the immune system be? Proc. R. Soc. B

252, 171-175 (1993).

22. Kalbe, M. et al. Lifetime reproductive success is maximized with optimal MHC diversity.

Proc. R. Soc. B 276, 925-934 (2009).

121

Faculty of Science The University of Sydney

23. Milinski, M. The major histocompatibility complex, sexual selection and mate choice. Ann.

Rev. Ecol. Syst. 37, 159–186 (2006).

24. Nowak, M.A., Tarcy-Hornoch, K. & Austyn, J. The optimal number of major

histocompatibility complex molecules in an individual. Proc. Natl. Acad. Sci. USA 89,

10896-10899 (1992).

25. Reusch, T.B.H., Häberli, M.A., Aeschlimann, P.B. & Milinski, M. Female sticklebacks count

alleles in a strategy of sexual selection explaining MHC polymorphism. Nature 414, 300-302

(2001).

26. Wegner, K.M., Kalbe, M., Kurtz, J., Reusch, T.R.B. & Milinski, M. Parasite selection for

immunogenic optimality. Science 301, 1343 (2003).

27. Woelfing, B., Traulsen, A., Milinski, M. & Boehm, T. Does intra-individual major

histocompatibility complex diversity keep a golden mean? Phil. Trans. R. Soc. B 364, 117–

128 (2009).

28. Olsson, M. et al. MHC, health, color, and reproductive success in sand lizards. Behav. Ecol.

Sociobiol. 58, 289–294 (2005).

29. Kempenaers, B. Mate choice and genetic quality, a review of the heterozygosity theory. Adv.

Study Behav. 37, 189–278 (2007).

30. Ejsmond, M.J., Radwan, J. & Wilson, A.B. Sexual selection and the evolutionary dynamics

of the major histocompatibility complex. Proc. R. Soc. B 281, 20141662 (2014).

31. Madsen, T., Shine, R., Olsson, M. & Witzell, H. Restoration of an inbred adder population.

Nature 402, 34-35 (1999).

32. Belov, K. The role of the major histocompatibility complex in the spread of contagious

cancers. Mammal. Genome 22, 83–90 (2011).

33. Jones, M. et al. Life-history change in disease-ravaged Tasmanian devil populations. Proc.

Natl. Acad. Sci. USA 105, 10023–10027 (2008).

122

Faculty of Science The University of Sydney

34. Lachish, S., Miller, K. J., Storfer, A., Goldizen, A. W. & Jones, M. E. Evidence that

disease-induced population decline changes genetic structure and alters dispersal patterns

in the Tasmanian devil. Heredity 106, 172-182 (2011).

35. Hamede, R.K., McCallum, H. & Jones, M. Seasonal, demographic and density-related patterns

of contact between Tasmanian devils (Sarcophilus harrisii). Implications for transmission of

devil facial tumour disease. Austral Ecol. 33, 614-622 (2008).

36. Siddle, H.V. et al. Reversible epigenetic down-regulation of MHC molecules by devil facial

tumour disease illustrates immune escape by a contagious cancer. Proc. Natl. Acad. Sci. USA

110, 5103–5108 (2013).

37. Karu, N. et al. Discovery of biomarkers for Tasmanian devil cancer (DFTD) by metabolic

profiling of serum. J. Proteome Res. 15, 3827−3840 (2016).

38. Pearse, A-M. & Swift, K. Allograft theory transmission of devil facial-tumour disease. Nature

439, 549 (2006).

39. Hogg, C.J. et al. Influence of genetic provenance and birth origin on productivity of the

Tasmanian devil insurance population. Conserv. Genet. 16, 1465-1473 (2015).

40. Hockley, J. DPIPWE/ZAA Husbandry guidelines for Tasmanian devil, Sarcophilus harrisii.

Zoo and Aquarium Association, Australia pp 46. (2011).

41. Keeley, T., O’Brien, J.K., Fanson, B.G., Masters, K. & McGreevy, P.D. The reproductive

cycle of the Tasmanian devil Sarcophilus harrisii and factors associated with reproductive

success in captivity. Gen. Comp. Endocrinol. 176, 182-191 (2012).

42. Jones, M.E., Paetkau, D., Geffen, E. & Moritz, C. Genetic diversity and population structure

of Tasmanian devils, the largest marsupial carnivore. Mol. Ecol. 13, 2197–2209 (2004).

43. Epstein, B. et al. Rapid evolutionary response to a transmissible cancer in Tasmanian devils.

Nature Commun. 7, 12684 (2016).

123

Faculty of Science The University of Sydney

44. Morris, K., Austin, J.J. & Belov, K. Low major histocompatibility complex diversity in the

Tasmanian devil predates European settlement and may explain susceptibility to disease

epidemics. Biol. Lett. 9, 20120900 (2012).

45. Morris, K., Wright, B., Grueber, C.E., Hogg, C. & Belov, K. Lack of genetic diversity across

diverse immune genes in an endangered mammal, the Tasmanian devil Sarcophilus harrisii.

Mol. Ecol. 24, 3860-3872 (2015).

46. Brown, O. Tasmanian devil Sarcophilus harrisii extinction on the Australian mainland in the

Mid- Holocene, multicausality and ENSO intensification. Alcheringa 30, 49–57 (2006).

47. Bruniche-Olsen, A., Jones, M.E., Austin, J.J., Burridge, C.P. & Holland, B.R. Extensive

population decline in the Tasmanian devil predates European settlement and devil facial

tumour disease. Biol. Lett. 10, 20140619 (2014).

48. Lane, A. et al. New Insights into the role of MHC diversity in devil facial tumour disease.

PLoS One, 7, e36955 (2012).

49. Olsson, M., Shine, R., Madsen, T., Gullberg, A. & Tegelström, H. Sperm selection by

females. Nature 383, 585 (1996).

50. Freeman-Gallant, C.R., Meguerdichian, M., Wheelwright, N.T. & Sollecito, S.V. Social

pairing and female mating fidelity predicted by restriction fragment length polymorphism

similarity at the major histocompatibility complex in a songbird. Mol. Ecol. 12, 3077-3083

(2003).

51. Blomqvist, D., Angela Pauliny, A., Mikael Larsson, M. & Flodin, L-Å. Trapped in the

extinction vortex? Strong genetic effects in a declining vertebrate population. BMC Evol.

Biol. 10, 33 (2010).

52. García-Navas, V., Cáliz-Campal, C., Ferrer, E.S., Sanz, J.J. & Ortego, J. Heterozygosity at a

single locus explains a large proportion of variation in two fitness-related traits in great tits: a

general or a local effect? J. Evol. Biol. 27, 2807–2819 (2014).

124

Faculty of Science The University of Sydney

53. Aarnink, A., Mee, E.T., Savy, N., Congy-Jolivet, N., Rose. N.J. & Blancher, A. Deleterious

impact of feto-maternal MHC compatibility on the success of pregnancy in a macaque model.

Immunogenetics 66, 105–113 (2014).

54. Maynard Smith, J. Sympatric speciation. The American Naturalist 100 (916), 637-650 (1966).

55. Matthews, B., Harmon, L.J., M’Gonigle, L., Marchinko, K.B., & Schaschl, H. Sympatric and

allopatric divergence of MHC genes in threespine stickleback. PLoS ONE 5, e10948 (2010).

56. Kamiya, T., O'Dwyer, K., Westerdahl, H., Senior, A. & Nakagawa, S. A quantitative review

of MHC-based mating preference: the role of diversity and dissimilarity. Mol. Ecol. 23, 5151-

5163 (2014).

57. Yeates, S.E. et al. Atlantic salmon eggs favour sperm in competition that have similar major

histocompatibility alleles. Proc. R. Soc. B 276, 559–566 (2009).

58. Løvlie, H., Gillingham, M.A.F., Worley, K., Pizzari, T. & Richardson, D.S. Cryptic female

choice favours sperm from major histocompatibility complex-dissimilar males. Proc. R. Soc.

B 280, 0131296 (2013).

59. Schwensow, N., Eberle, M. & Sommer, S. Compatibility counts, MHC-associated mate

choice in a wild promiscuous primate. Proc. R. Soc. B 275, 555-564 (2008).

60. Lees, C. & Srb, C. Annual report for the DPIW-ARAZPA Tasmanian devil insurance

population. Zoo and Aquarium Association. pp 27 (2007).

61. Lees, C. & Srb, C. Annual report for the DPIW-ARAZPA Tasmanian devil insurance

population. Zoo and Aquarium Association. pp 49 (2008).

62. Hibbard, C.J., Ford. C. & Srb, C. Annual report for the DPIW-ZAA Tasmanian Devil

insurance population. Zoo and Aquarium Association. pp57 (2009).

63. Hibbard, C.J., Hogg, C.J., Srb, C. & Hockley, J. Annual report for the DPIPWE-ZAA

Tasmanian devil insurance population. Zoo and Aquarium Association. pp 35 (2010).

64. Hibbard, C.J., Hogg, C.J., Srb, C., & Hockley, J. Annual report for the DPIPWE-ZAA

Tasmanian devil insurance population. Zoo and Aquarium Association. pp 46 (2011).

125

Faculty of Science The University of Sydney

65. Hogg, C.J., Srb, C. & Hockley, J. Annual report for the DPIPWE-ZAA Tasmanian devil

insurance population. Zoo and Aquarium Association. pp 43 (2012).

66. Hogg, C.J., Srb, C. & Hockley, J. Annual Report for the DPIPWE-ZAA Tasmanian devil

insurance population. Zoo and Aquarium Association. pp 42 (2013).

67. Alonzo, S.H., Stiver, K.A. & Marsh-Rollo, S.E. (2016). Ovarian fluid allows directional

cryptic female choice despite external fertilization. Nature Commun. 7, 12452 (2012).

68. Gasparini, C., & Pilastro, A. Cryptic female preference for genetically unrelated males is

mediated by ovarian fluid in the guppy. Proc. R. Soc. B 278, 2495–2501 (2011).

69. Srb, C. (2014). Australasian regional Tasmanian devil studbook. (Healesville, Victoria,

2014).

70. Brice, S. Zoo aquarium association husbandry guidelines for Tasmanian devil, Sarcophilus

harrisii. Zoo and Aquarium Association, Australia pp 36 (2006).

71. Cheng, Y. & Belov, K. Isolation and characterisation of 11 MHC-linked microsatellite loci in

the Tasmanian devil Sarcophilus harrisii. Conserv. Genet. Res. 4, 463–465 (2012).

72. van Oosterhout, C., Hutchinson, W.F., Wills, D.P.M. & Shipley, P. Micro-checker: software

for identifying and correcting genotyping errors in microsatellite data. Mol. Ecol. Notes 4,

535–538 (2004).

73. Excoffier, L., Laval, G. & Schneider, S. ARLEQUIN Version 31, An Integrated Software

Package for Population Genetics Data Analysis Computational and Molecular Population

Genetics Lab CMPG, Institute of Zoology, University of Berne, Switzerland CMPG website

http//cmpgunibech/software/arlequin3 (2006).

74. SAS/STAT 9.22. User’s Guide The GLIMMIX Procedure. (SAS Institute Inc., Cary, NC

2010).

75. Wang, J., Xie, H., & Fisher, J.H. Multilevel models – Applications using SAS. De Gruyter

Higher Education Press, Berlin (2012).

126

Faculty of Science The University of Sydney

4.7 Acknowledgements

This work was supported by a Morris Animal Foundation grant to RS and ARC Discovery grant (DP140103260, DP110102731) to KB and the ANR EVOCAN, André Hoffmann and la Fondation MAVA to FT. We thank the Zoo and Aquarium Association and the Save the

Tasmanian Devil Program and all participating zoos and wildlife parks for access to samples and stud book records.

4.8 Author Contributions

T.R conducted the molecular analyses. S.L, G.B. and M.O. conducted the statistical analyses. T.R., T.M. and B.U. wrote the manuscript. S.L, G.B., M.O., R.S, A.L., T.K.,

C.H., C.J.H., F.T. and K.B. contributed to the scientific discussion and revision of the manuscript.

127

Faculty of Science The University of Sydney

5 Chapter Five: Exploring the role of endocrinology and behaviour in relation to the reproductive success of captive

Tasmanian devils

5.1 Abstract

In this study I investigated whether the reproductive success of captive Tasmanian devils is affected by hormone metabolites (progesterone and corticosterone) collected from faecal samples, and whether any behaviours, (as observed by zookeepers) underpin reproductive success. Progesterone levels were used to monitor the female reproductive cycle. Of a total of 42 pairings, 36 were judged to be correctly timed (i.e. within the correct range of when the female was in oestrous) however, of these correctly timed pairings, only eight resulted in the female producing offspring. Female devils had significantly higher corticosterone baselines than males. No difference was found in corticosterone baseline levels between successful and unsuccessfully paired males or in successful or unsuccessfully paired females. Similar to previous findings, these results do not suggest that corticosterone has any significant effect on the reproductive success of captive devils.

5.2 Introduction

Stress is a critical function of survival, causing internal (hormonal) and external

(physical/behavioural) reactions in all animals in the face of a real or perceived threat. Via the sympathetic nervous system, the hypothalamic-pituitary-adrenal axis (HPA) becomes activated (Bonier et al. 2009) where glucocorticoids, such as corticosterone, are released from the adrenal glands. The HPA axis regulates glucocorticoid secretions and maintains homeostasis. Corticosterone secretion increases in response to environmental perturbations

128

Faculty of Science The University of Sydney e.g. drought and in times of stress (Bonier et al. 2009). There are two main types of stress – chronic and acute (Dickens et al. 2010). Chronic stress occurs when the initial stressor persists over time or when an organism is barraged by constant, novel stressors (Dickens et al. 2010). While acute stress responses are vital for survival and reproduction (Fanson and

Parrot, 2015), chronic stress responses can have detrimental effects on the body. The prolonged secretion of glucocorticoids can affect the immune system including, suppression of reproductive behaviour (Sapolsky et al. 2000; Dickens et al. 2010).

The processes of capturing, securing and transporting wild animals all constitute potential individual stressors which alone or in combination may cause chronic stress (Grandin

1997). Acute stress has been shown to negatively affect reproduction in a wide range of animals, resulting in changes to behaviour and or physiology that may have a detrimental impact on organismal fitness (Wingfield and Sapolsky 2003; Van der Weyde et al. 2016).

In females, acute stress may result in disrupted ovulation and maturation of the uterus, causing the inability to support implantation and behavioural changes including changes to receptivity (Wingfield and Sapolsky 2003). In males, acute stress may also result in reduced fitness due to a lack of receptivity and proceptivity, erectile dysfunction and hormone inhibition (Wingfield and Sapolsky 2003).

Faecal hormone analysis is widely used to monitor stress and reproductive hormones in domestic, captive and wild animals. Due to the non-invasive nature of sample collection, results (glucocorticoids) are not influenced by handling stress. Faecal glucocorticoids have been successfully used to quantify stress in various species including jaguars (Panthera onca) (Morato et al. 2004), river otters (Lontra canadensis) (Rothschild et al. 2005), snowshoe hares (Lepus americanus) (Sheriff et al. 2009), bank voles (Myodes glareolus)

(Rogovin and Naidenko 2010), western lowland gorillas (Gorilla gorilla gorilla) (Peel et al. 2005) and rhesus macaques (Macaca mulatta) (Brent at al. 2011). Faecal progesterone metabolites have also been employed to monitor and assess reproductive cycles in

129

Faculty of Science The University of Sydney numerous marsupials including the common wombat (Vombatus ursinus), the southern hairy-nosed wombat (Lassiorhinus latifrons) (Paris et al. 2002), Gilbert’s potoroo (Potorus gilbertii), the honey possum (Tarsipes rostratus) (Bradshaw et al. 2004) and numbats

(Myrmecobius fasciatus) (Hogan et al. 2012).

Stress hormones and the reproductive cycles of animals are now widely monitored using faecal hormone metabolites due to their ease of collection with little to no disruption to the animals studied (Schwarzenberger et al. 1996), which is especially important with threatened and endangered species (Stead-Richardson et al. 2010). Levels of glucocorticoids present in faeces have been proven to be as consistent in many species as those isolated from plasma, thus providing a reliable representation of the physiological responses to stressors in animals (Sheriff et al. 2010).

However, the role of glucocorticoids in reproduction is ambiguous. For example, cyclic peaks of glucocorticoids can be crucial for normative reproductive function including aiding the signalling of receptivity during proestrous and ovulation stimulation (Fanson et al. 2014). In addition, Fanson and Parrott (2015) found no evidence that elevated glucocorticoids are directly responsible for reduced reproductive success in marsupials.

The Tasmanian devil is a polyovular, synchronous breeder, producing excessive oocytes that cannot be functionally supported (Guiler 1970). The breeding period peaks in March.

However, if pregnancy is not achieved oestrus will recur, and female devils have been documented having a second and even a third oestrus with births occurring in August and

(rarely) as late as October in the wild (Guiler 1970; Keeley et al. 2012). Litter size has been found to correlate with female devil age and body mass with younger, lighter females having a higher number of pouch young than that of older females (Guiler 1970). Within

130

Faculty of Science The University of Sydney the captive devil breeding program the prime breeding age for females has been found to occur in two year old females (Chapter two and five). Interestingly pseudopregnancy, where females with no embryos show physical signs of pregnancy – i.e. distended uteri and active mammary gland development yet no young are present, has been observed in devils

(Guiler 1970).

Extensive studies were undertaken into the reproductive cycle of the Tasmanian devil by

Hesterman (2008) and Keeley et al. (2012), with the latter dividing the oestrous cycle into three main phases - the follicular phase, the luteal phase and the intermediate phase. The follicular phase results in follicular development/growth, the secretion of oestradiol and the first of two progesterone peaks; and it is during this initial progesterone surge when females are in oestrus and thus receptive (Keeley et al. 2012). The luteal phase comprises the second and larger peak in progesterone intrinsic to ovulation and pregnancy. Following this is the intermediate phase which encompasses the end of the luteal phase until the beginning of the next proestrous phase within the period of one breeding season (Keeley et al. 2012). The timing of male introductions to females is crucial to reproductive success and zoo keepers use visual and behavioural cues to determine the onset of oestrous including appetite suppression of at least 50%, the observation of a fluid filled neck roll which develops in the female and is to protect her when being dragged around by the neck by the male when he is mate guarding, and nesting behaviours (Kelly 2007; Hockley 2011).

The pairing of devils has been reported to range from 1-15 days in duration with an average of 5 and 8 days for unsuccessful and successful pairings respectively (Keeley et al. 2012; this study). Not surprisingly, Keeley et al. (2012) found that the timing of pairing of devils is highly related to reproductive success, with pairings ending early in proestrus being less likely to produce offspring. Therefore, it is crucial to further explore the effects of timing of

131

Faculty of Science The University of Sydney the introduction of male devils to females in oestrus as well as monitoring stress levels associated with such pairings.

Using faecal hormone metabolites, the aims of this study were: i) to examine faecal progesterone metabolite profiles in conjunction with the timing of introductions to determine whether male devils were introduced to females at the correct time of female receptivity (i.e. at the onset of oestrus), ii) to assess whether elevated corticosterone levels affect reproductive success, and iii) to determine from keepers’ behavioural observations whether interactions between males and female devils can predict reproductive activity.

5.3 Materials and Methods

5.3.1 Study animals

Fifty two Tasmanian devils were included in this study, 27 males and 25 females housed at

7 Australian zoological institutions (New South Wales 2, Queensland 1, Tasmania 2 and

Victoria 2), representing 40 pair combinations. All pairing combinations were determined by the ZAA who are responsible for the management of the Tasmanian devil insurance population (Hibbard et al. 2011). Male devils were introduced to females by zoo keepers when females showed signs of oestrus – primarily a decrease in appetite by at least 50%

(Hockley 2011). Other signs include a fluid filled neck roll, nest building and a decrease in activity (Hockley 2011). The dates of pairings were recorded by zoo keepers and opportunistic behavioural observations were made on a scale of 1 – 5 (e.g. affiliation 5 or aggression and disinterest 1 between paired animals). Animals were separated if they showed no interest in each other or were overly aggressive.

132

Faculty of Science The University of Sydney

5.3.2 Faecal samples

Faecal samples were collected by zookeepers during their regular husbandry routines between 3-7 times per week (range of 30-70 samples per animal) over a six month period

(January – June) encompassing between one and three oestrous cycles per female devil.

Samples were labelled with the animal’s name and the date, placed in zip lock bags and frozen at -200C at the time of collection. Frozen samples were then thawed and oven dried for 12 hours at 650C. They were then lightly pounded with a hammer and sifted through a

2mm sieve to remove fur, feathers and bone. The dried samples were then weighed to 0.19

– 0.2 g, placed into glass scintillation vials, filled with 4 ml of 80% methanol and placed overnight onto a rotary mixer. Samples were then centrifuged for 20 minutes at 1000 g with the supernatant decanted for processing.

5.3.3 Hormone analysis

Hormone analysis of corticosterone and progesterone were conducted as per protocols in

Keeley et al. (2012) as follows: enzyme-immunoassays (EIAs) were used to determine hormone concentrations of faecal corticosterone metabolite (FCM) and faecal progesterone metabolite (FPM). Antibodies with broad spectrum cross reactivity, including those for metabolites of progesterone and corticosterone were used in the EIAs. An adrenal corticotropic hormone (ACTH) challenge was performed by Keeley et al. (2012) verifying the validity of these techniques for this species. Faecal samples were diluted in an assay buffer at the ratio of 1:5 to 1:10 v/v before hormone concentration analysis. The antiserum and horseradish peroxidase conjugated label for each EIA were provided by C. Munro,

University of California-Davis, Davis, USA. The progesterone EIA used a monoclonal antibody (progesterone CL425; 1:8000), progesterone horseradish peroxidase conjugated label (1:40,000) and progesterone standards (0.78–200 pg/50 µl; Sigma Aldrich Australia,

Ltd.). The corticosterone EIA used a polyclonal antibody (corticosterone CJM06;

133

Faculty of Science The University of Sydney

1:12,500), corticosterone horseradish peroxidase conjugated label (1:20,000) and corticosterone standards (4.88–2500 pg/50 µl; Steraloids Inc., USA). The cortisol EIA used a polyclonal antibody (cortisol R4866; 1:9000), cortisol horseradish peroxidase conjugated label (1:20,000) and cortisol standards (3.9–1000 pg/50 µl; Sigma Aldrich Australia, Ltd.).

EIA’s were conducted as per Keeley et al. (2012). Samples were placed into 96 well microtitre plates and analysed in duplicate (50 µl/well). Standard curves were created and parallelism was determined between these and a series of faecal sample dilutions.

Coefficients of variation between and within assays was < 10%. Cross-reactivities for the corticosterone EIA antibody were: corticosterone 100%, desoxycorticosterone 14.25%, tetrahydrocorticosterone 0.9%, cortisol 0.23%, progesterone 2.65%, testosterone 0.64% and

<1% with all other steroids tested. The cross-reactivities of the progesterone EIA antibody were: progesterone 100%, 4-pregnen-3a-ol-20-one 188%, 4-pregnen-3-ol-20-one 172%, 4- pregnen-11a-ol-3,20-dione 147%, 5a-pregnen-3-ol-20-one 94%, 5a-pregnen-3a-ol-20-one

64%, 5a-pregnen-3,20-dione 55%, 5-pregnen-3-ol-20-one 12.5% and 5-pregnen-3,20-dione

8% and less than 0.1% with another nine steroids including oestradiol, corticosterone and testosterone. Both corticosterone and progesterone concentrations were reported in ng/g of faeces.

5.3.4 Behavioural observations

Participating zoos used in the present study were presented with a behavioural scoresheet for the paired devils to record during their routine observations of the paired animals. Two main categories were listed 1) aggression, with the sub headings of i) vocalisations ii) jaw locking iii) chasing and iv) attack and 2) affiliation, with the sub headings i) mate guarding ii) mate dragging iii) observed together and iv) observed mating. Keepers were also asked to provide an overall affiliation score of their opinions on the devils’ overall affinity over a sliding scale ranked from 1-5 where 1 = Very aggressive, no affiliation, 3 = Some

134

Faculty of Science The University of Sydney aggression, some affiliation, 5 = Very affiliative, no aggression seen. There was also room for keepers to make their own notes along with details such as the animals’ names, studbook number, dates the animals were paired, weight and which oestrous period was being observed.

5.3.5 Analysis

Baselines were established for all progesterone and corticosterone samples and all values

1.5 standard deviations above the mean baseline were considered to be significantly elevated as per Keeley et al. (2012). All progesterone and corticosterone results for females were graphed, along with the timing of male introductions and significant levels above baseline. Progesterone profiles were used to determine the time of receptivity (oestrus) and the luteal phase so that confirmation of the appropriate timing of pairings could be determined. No further analysis was conducted on progesterone patterns or levels.

5.3.6 Statistical analysis

Depending on the data both non-parametric and parametric tests were used to analyse the results. All tests were performed using JMP v 5.

135

Faculty of Science The University of Sydney

5.4 Results

5.4.1 Progesterone profiles and timing of pairings

Female baseline progesterone values ranged from 43.5 to 262.5 ng/g. Of the 42 pairings

one female with extremely low progesterone levels (paired with 3 males) and three females

with mistimed pairings were excluded, therefore the analyses are based on 36 pairings. All

of the remaining 36 pairings were judged to be correctly timed and within the correct range

of when the female was in oestrus (for an example of an incorrect and a correct pairing see

Figure 5.1 and Figure 5.2, respectively). Of these 36 pairings, eight were successful in

producing offspring (hereafter known as successful pairings) and in the remaining 28

pairings, females did not produce any offspring (hereafter know as unsuccessful pairings)

(Table 5.1). Seven of the successful females were paired in their first oestrous cycle and

one on the second (Table 5.1). The latter successful female that conceived on her second

oestrus cycle had been paired at first oestrous but the timing was deemed too early and the

devils showed no interest in each other (Figure 5.7). In the unsuccessful females 18 were

paired on their first oestrous cycle, 9 on the second and one on the third oestrous cycle

(Table 5.1).

Table 5.1 Number of males paired with successful and unsuccessful females during their first, second and the third oestrous cycle.

First oestrous First oestrous Second Second Third oestrous oestrous oestrous Paired with 1 male 2 males 1 male 2 males 1 male Successful female 7 1 Unsuccessful 12 6 5 4 1 female

136

Faculty of Science The University of Sydney

Figure 5.2 Progesterone and corticosterone profile for a successfully paired female devil. The time of pairing corresponds with the receptive period delineated by the smaller progesterone peaks during oestrus (black line over peak). The larger progesterone peaks represent parturition. The corticosterone profile closely matches the progesterone profile. Prog > baseline and Cort > baseline refer to all values for progesterone and corticosterone that were above baseline figures.

137

Faculty of Science The University of Sydney

5.4.2 Corticosterone

In female devils, the corticosterone baselines ranged from 6.8 ng/g to 262.5 ng/g and in the males the range was 15.1 ng/g to 139.8 ng/g. Female devils had significantly higher corticosterone baselines (1.5 s.d. above baseline) than males (Wilcoxon non-parametric test

Z = 2.10, p = 0.036, d. f. = 1). A negative correlation was found between male body weight and average baseline corticosterone (Spearman rank correlation R = -0.55, p = 0.021 n =

17; Figure 5.3). In females there is a tendency toward a negative correlation, however this was not significant (Spearman rank correlation R = -0.39, p = 0.10, n = 19; Figure 5.4).

Figure 5.3 Relationship between body mass (kg) and corticosterone baseline levels in male devils.

138

Faculty of Science The University of Sydney

Figure 5.4 Relationship between body mass (kg) and corticosterone baseline levels in female devils.

No difference in corticosterone baselines were observed between successful and unsuccessful males (Wilcoxon non-parametric test: Z = -1.37, p = 0.16, d.f. = 1, (mean corticosterone successful males: 31.72 ± S. E. 5.88; unsuccessful males 55.25 ± S. E. 8.92; the corticosterone baseline profiles of a successful male and an unsuccessful male are depicted in Figure 5.5 and Figure 5.6, respectively). Similar to the males, there was no difference between successful and unsuccessful female corticosterone baselines (Wilcoxon non-parametric test: Z = -0.05, p = 0.96, d.f. = 1; mean corticosterone successful females

72.96 ± S. E. 21.53; unsuccessful females 83.57 ± S. E. 16.78; the corticosterone baseline profiles of successful male and an unsuccessful

139

Faculty of Science The University of Sydney female.

Figure 5.5 Corticosterone profile of a successfully paired male devil. Above baseline corticosterone peaks at pairing time (no sample collected during actual pairing with female). This male was successfully paired with female 1 and unsuccessfully with female 2.

Figure 5.6 Corticosterone profile of an unsuccessfully paired male devil. This male was paired with one female.

140

Faculty of Science The University of Sydney

Figure 5.7 Corticosterone profile of a successfully paired female devil. This female devil was unsuccessfully paired with male 1 and successfully paired with male 2. Above baseline corticosterone peaks at pairing time (no sample collected during actual pairing with male).

Figure 5.8 Corticosterone profile of a female unsuccessfully paired with two male devils.

141

Faculty of Science The University of Sydney

5.4.3 Behavioural observations

Two females that were each paired with two different males showed high levels of aggression towards the males, vocalising and chasing them away. One of these females was paired in the non-receptive phase (Figure 5.1). One male showed immediate aggression towards a female and the pair were separated. Eight of the unsuccessfully paired devils (i.e. no offspring produced) were observed mating and had an affiliation score of 5. The unusual behaviour of one female mate guarding a male over a three day period was also observed.

Behavioural observations were available for 34 of 36 pairings however from the standardised questionnaire sent to devil keepers as outline above, there were responses relating to 25 pairs across five institutions. Some of these animals were paired over multiple oestrous periods (Table 5.4). Information was also gathered on the duration of pairings, which oestrus period the female was in and whether any young were produced from the pairings (six pairs produced joeys and 19 were unsuccessful) (Table 5.4). The remaining nine sets of observations were based on individual keepers’ notes, and while some contained animal behaviours, they were not standardised to the categories on the questionnaire so have been omitted from the following analysis.

142

Faculty of Science The University of Sydney

Table 5.2 The five institutions housing devils for pairing corresponding to the behavioural questionnaire.

Institution Number of Number of reproductively reproductively successful unsuccessful pairings pairings 1 2 3 2 0 15 3 2 3 4 1 2 5 1 2

The majority of aggressive behaviours were not physical (80%) i.e. vocalising or chasing

(Table 5.3). Mate guarding and mate dragging constituted almost 75% (73.82) of affiliative behaviours (Table 5.3).

Table 5.3 Summary of instances of aggressive and affiliative behaviours for 25 paired devils for the purpose of mating as scored by zoo keepers across five institutions.

Total no. of aggressive Successful Unsuccessful behaviours % % Vocalisation 26.3 34.7 Jaw locking 5.3 8.8 Chasing 18.4 28.2 Attacking 2.6 7.6

Total no. of affiliative

behaviours Mate guarding 39.5 30 Mate dragging 18.4 12.9 Observed together 47.4 40.7 Observed mating 36.8 4.12

143

Faculty of Science The University of Sydney

5.4.4 Breeding success/timing

There was no difference in the length of pairing (days) between reproductively successful pairs and unsuccessful pair (logistic regression: chi-square: 0.016, p = 0.89, df = 1). Nor was there an association between mean affiliation scores and breeding success (logistic regression: chis-square: 2.81, p = 0.09, df = 1. Not surprisingly, however, reproductive success was significantly different in cases with and without observed mating, (Fisher exact test p = 0.008). In institutions 1, 3, 4, and 5 no apparent difference in number of successful and unsuccessful parings was observed (Table 5.4.). The data from these four institutions were therefore pooled and compared to the reproductive success achieved at institution 2.

This analysis revealed a significant difference between the reproductive success of institution 1, 3, 4, and 5 relative to institution 2 (Fishers exact text p = 0.018; Table 5.4).

Table 5.4. Summary of pairing information for 25 pairs of Tasmanian devils paired for breeding purposes. Summary Successful No. pairs 25 Successful pairings (joeys produced) 6 Unsuccessful pairings 19 No. paired on 1st oestrus 19 5 No. paired on 2nd oestrus 13 1 No. paired on 3rd oestrus 2 No. paired on 1st and 2nd oestrus 10 No. paired on 2nd oestrus only 3 Length of pairing (range) in days 2-12 Average length of pairing (days) 6.27 ± 0.56 Length of pairing (range) in days (successful 2-12 pairs Average length of pairing (successful pairs) 6.33 ± 1.43 (days)

144

Faculty of Science The University of Sydney

5.5 Discussion

The progesterone analyses confirmed that the majority (93%) of pairings analysed in this study were correctly timed. In three pairings, the females were not in a receptive phase, and in the remaining pairing, the female had such low progesterone concentrations that this precluded assigning the receptive phase with confidence. Behavioural observations by zoo keepers revealed high levels of aggression in one of the former three females towards the two males that she was paired with and another of these three females was subjected to male aggression and the pair was separated. Observations by the zoo keepers revealed that twelve of the pairings were highly affiliative (scoring a 4 or a 5) and eight of the unsuccessful females were observed mating with males. The duration of mating was not quantified in this study, however in Julia Creek dunnarts (Sminthopsis douglasi), mating duration was correlated with reproductive success, where dunnarts that mated for at least 30 minutes were more reproductively successful (Woolley et al. 2015).

Although not yet confirmed, it has been suggested that in some cases captive females (i.e. the unsuccessful females whose hormone profiles mimic those of successful females) do conceive and even give birth but that the young are lost prior to the first pouch check, the possible factors underpinning this, if correct, are, however currently unknown (Keeley et al.

2012).

As mentioned above, seven of successful females were paired in their first oestrous cycle, and one on the second. This result suggests that pairing of females during their first oestrous cycle may increase the probability that the female will successfully reproduce.

Similar results were obtained in a study by Keeley et al. (2012), where almost half of the devils paired in their first cycle (44%) were successful, compared to 10% in their second,

145

Faculty of Science The University of Sydney and none in their third cycle. However, as the majority of the unsuccessful pairings (18 out of 28) were conducted at the female’s first cycle, it is highly unlikely that that the low reproductive success rate (8 of the 36 pairings) was caused by female infertility and hence this low rate may be caused by other factors such as male-female genetic compatibility as suggested in chapter 4.

This study demonstrated a negative correlation between corticosterone baseline levels and body mass in male devils. Thus, males with low body mass had higher levels of corticosterone compared to males with high body mass. As the devils were not housed in multi-male groups, which may cause stress to smaller males, the underpinning(s) of this correlation remains unknown. Similar results have, however, been observed in other vertebrates (Moore and Jessop 2003). Low body mass in male devils may indicate lower levels of energy stores (i.e. lipids) causing higher gluconeogenic activity which in turn may results in their higher corticosterone levels.

Females had significantly higher corticosterone levels than males although inter-animal hormone baselines were extremely variable. This effect has also been recorded in other marsupials (Narayan et al. 2012) and carnivorous mammals (Young et al. 2004).

Corticosterone in females mirrored progesterone and as can be seen in the hormone profiles

(e.g. Fig 6.1 and 6.2 which are indicative of all of the correctly timed female profiles), corticosterone peaks followed reproductive phases and were not associated with pairing.

However, in captive Gilbert’s potoroos, no differences were found between male and female glucocorticoids (Stead-Richardson et al. 2010). This suggests that the apparent sex- specific difference in corticosterone levels in devils requires further research.

The results from the present study showed that glucocorticoids do not appear to affect captive devil reproductive success concurring with results found by Stead-Richardson et al.

146

Faculty of Science The University of Sydney

(2010) in Gilbert’s potoroo and the conclusions reached by Fanson and Parrot (2015) regarding the lack of effect between glucocorticoids and successful reproduction in marsupials.

The results from this study show that the timing of pairing as judged by keepers is within the receptive periods for these female. Additionally, whether the female was in her first or subsequent reproductive cycle appears to have an impact on female reproductive success.

Moreover, the low reproductive success recorded at institution two suggests that additional factors, such as the housing or maintenance of the captive devils may significantly affect their reproductive success.

5.6 Conclusions

Captive breeding success in the Tasmanian devil is multifactorial. This study demonstrates that in order to improve breeding success female devils should be paired during their first oestrous period, and that the husbandry and management practice of the captive populations should be further investigated. The results from this study do not suggest that devil glucocorticoids have a significant effect on captive devil reproductive success.

5.7 Acknowledgements

I would like to thank all of the zoo staff from the participating institutions for their time and efforts into collecting behavioural information and samples over a long period of time so that this study could be conducted.

147

Faculty of Science The University of Sydney

5.8 References

Bonier, F., Martin, P. R., Moore, I. T. and Wingfield, J. C. (2009) Do baseline glucocorticoids predict fitness? Trends in Ecology and Evolution 24 (11), 634-642.

Bradley, A. J., McDonald, I. R., Lee, A. K., (1980) Stress and mortality in a small marsupial (Antechinus stuartii, Macleay). General and Comparative Endocrinology 40,

188–200.

Bradshaw, F. J., Stead-Richardson, E. J., Reeder, A. J., Oates, A. J. and Bradshaw, J. E.

(2004) Reproductive activity in captive female honey possums, Tarsipes rostratus, assessed by faecal steroid analysis. General and Comparative Endocrinology 138, 20-

31.

Brent, L. J. N., Semple, S., Dubuc, C., Heistermann, M. and MacLarnon, A. (2011)

Social capital and physiological stress levels in free-ranging adult female rhesus macaques. Physiology and Behavior 102, 76-83.

Dickens, M. J., Delehanty, D. J and Romero, L. M. (2010) Stress: an inevitable component of animal translocation. Biological Conservation 143, 1329-1341.

Fanson, K. V., Keeley, T. and Fanson, B. G. (2014) Cyclic changes in cortisol across the estrous cycle in parous and nulliparous Asian elephants. Endocrine Connections

3(57), 1-10.

148

Faculty of Science The University of Sydney

Fanson, K. V. and Parrott, M. L. (2015) The value of eutherian–marsupial comparisons for understanding the function of glucocorticoids in female mammal reproduction.

Hormones and Behaviour 76, 41-47.

Grandin, T. (1997) Assessment of stress during handling and transport. Journal of

Animal Science 75:249-257.

Guiler, E. R. (1970) Observations on the Tasmanian devil, Sarcophilus harrisii

(Marsupialia: Dasyuridae) II. Reproduction, breeding, and growth of pouch young.

Australian Journal of Zoology 18, 63-70.

Hesterman, H., Jones, S. M. and Schwarzenberger, F. (2008) Reproductive endocrinology of the largest dasyurids: Characterization of ovarian cycles by plasma and faecal steroid monitoring. Part I. The Tasmanian devil (Sarcophilus harrisii).

General and Comparative Endocrinology 155, 234-244.

Hibbard, C. J., Hogg, C. J., Srb, C. and Hockley, J. (2011) Annual report for the

DPIPWE-ZAA Tasmanian Devil Insurance Population. Zoo and Aquarium Association.

Pp 46.

Hockley, J. (ed.) (2011) DPIPWE/ZAA. Husbandry Guidelines for Tasmanian Devil,

Sarcophilus harrisii. DPIPWE, Tasmania, Australia.

Hogan, L. A., Lisle, A. T., Johnston, S. D. and Robertson, H. (2012) Non-invasive assessment of stress in captive numbats, Myrmecobius fasciatus (Mammalia:

149

Faculty of Science The University of Sydney

Marsupialia), using faecal cortisol measurement. General and Comparative

Endocrinology 179, 376–383.

Husak, J. F. and Moore, I. T. (2008) Stress hormones and mate choice. Trends in ecology and evolution 23(10), 532-534.

Keeley, T., O’Brien, J. K., Fanson, B. G., Masters, K. and McGreevy, P. D. (2012) The reproductive cycle of the Tasmanian devil (Sarcophilus harrisii) and factors associated with reproductive success in captivity. General and Comparative Endocrinology 176,

182-191.

Moore, I. T. and Jessop, T. S. (2003) Stress, reproduction, and adrenocortical modulation in amphibians and reptiles. Hormones and Behaviour 43, 39-47.

Morato, R. G., Bueno, M. G., Malmheister, P., Vereschi, I. T.N. and Barnabe, R. C.

(2004) Changes in the faecal concentrations of cortisol and androgen metabolites in captive male Jaguars (Panthera onca) in response to stress. Brazilian Journal of

Medical and Biological Research 37(12), 1903-1907.

Paris, M. C. J., White, A., Reiss, A., West, M. and Schwarzenberger, F. (2002) Faecal progesterone metabolites and behavioural observations for the non-invasive assessment

150

Faculty of Science The University of Sydney of oestrous cycles in the common wombat (Vombatus ursinus) and the southern hairy- nosed wombat (Lasiorhinus latifrons). Animal Reproduction Science 72, 245-257.

Peel, A. J., Vogelnest, L., Finnigan, M., Grossfeldt, L. and O’Brien, J. K. (2005) Non-

Invasive faecal hormone analysis and behavioral observations for monitoring stress responses in captive western lowland gorillas (Gorilla gorilla gorilla). Zoo Biology 24,

431-445.

Roberts, M. L., Buchanan, K. L., Bennett, A. T. D. and Evans, M. R. 2007, Mate choice in zebra finches: does corticosterone play a role?, Animal behaviour 74 (4), 921-929.

Rogovin, K. A. and Naidenko, S. V. (2010) Noninvasive assessment of stress in bank voles (Myodes glareolus, Cricetidae, Rodentia) by Means of Enzyme_Linked

Immunosorbent Assay (ELISA). Biology Bulletin 37(9), 959-964.

Rothschild, D. M., Serfass, T. L., Seddon, W. L., Hedge, L. and Fritz R. S. (2005)

Using faecal glucocorticoids to assess stress levels in captive river otters. The Journal of

Wildlife Management 72(1), 138-142.

Schwarzenberger, F., Möstl, Palme, R. and Bamberg, E. (1996) Faecal steroid analysis for non-invasive monitoring of reproductive status in farm, wild and zoo animals.

Animal Reproduction Science 42, 515-526.

151

Faculty of Science The University of Sydney

Sheriff, M. J., Curtis, V., Bosson, O., Krebs, C. J. and Boonstra, R. (2009) A non- invasive technique for analyzing faecal cortisol metabolites in snowshoe hares (Lepus americanus). Journal of Comparative Physiology B-Biochemical Systemic and

Environmental Physiology 179, 305-313.

Sheriff, M. J., Krebs, C. J. and Boonstra, R. 2010 Assessing stress in animals populations: Do faecal and plasma glucocorticoids tell the same story? General and

Comparative Endocrinology 166, 614-619.

Spencer, K. A., Buchanan, K. L., Goldsmith, A. R. and Catchpole, C. K. (2003) Songs as an honest signal of developmental stress in the zebra finch (Taeniopygia guttata).

Hormones and Behavior 44, 132-139.

Stead-Richardson, E., Bradshaw, D., Friend, T. and Fletcher, T. (2010) Monitoring reproduction in the critically endangered marsupial, Gilbert’s potoroo (Potorous gilbertii): Preliminary analysis of faecal oestradiol-17b, cortisol and progestogens.

General and Comparative Endocrinology 165, 155-162.

Van der Weyde, L. K., Martin, G. B. and Paris, M. C. J. (2016) Monitoring stress in captive and free-ranging African wild dogs (Lycaon pictus) using faecal glucocorticoid metabolites. General and Comparative Endocrinology 226, 50-55.

Wingfield, J. C. and Sapolsky, R. M. (2003) Reproduction and resistance to stress: when and how. Journal of Neuroendocrinology 15, 711-724.

152

Faculty of Science The University of Sydney

Woolley, P. A. (2015) The Julia Creek dunnart, Sminthopsis douglasi (Marsupialia

:Dasyuridae): breeding of a threatened species in captivity and in wild populations.

Australian Journal of Zoology 63, 411-423.

Young, K. M., Walker, S. L., Lanthier, C., Waddell, W. T., Monfort, S. L. and Brown,

J. L. (2004) Noninvasive monitoring of adrenocortical activity in carnivores by fecal glucocorticoid analyses. General and Comparative Endocrinology 137, 148-165.

153

Faculty of Science The University of Sydney

6 Chapter Six: Discussion and Conclusions

Captive breeding can be a crucial intervention when species face extinction in the wild

(Frankham et al. 2002). However, the value to conservation depends not only on the ability to successfully re-establish captive individuals back into their natural habitats, but for them to be able to survive and reproduce upon release.

The Tasmanian devil insurance program (TDIP) has successfully met many of the goals required for a captive breeding program. These successes include, as stipulated in the initial

Population and Habitat Viability Assessment PHVA in 2008, (which laid out plans for the

TDIP going forward -CBSG, 2008), a captive population exceeding 700 animals, the inclusion of overseas facilities and the release of animals back into the wild (Thalmann et al. 2016, STTDP 2017): all within eleven years since its inception in 2006. However, although the program has been hugely successful at meeting some of its goals, it has not met breeding targets set out within the original PHVA recommendations. These recommendations set targets for the percentage of females required to breed ( modelled at

47% for 2 year olds, 43% for 3 year olds and 22% for four year olds -see Chapter Two) in each year and age class in order to maintain a viable and genetically diverse population for

50 years (CBSG 2008; Hibbard 2010). In addition, although captive-bred animals have been released into the wild in Tasmania in numerous locations, including Stony Head and

Narawntapu, and managed colonies have been established on Maria Island and on the

Forestier Peninsula, there remains a need for an ongoing captive breeding program, as the threat to the devils from DFTD and the recently emerged novel transmissible cancer

(DFT2) is still present.

154

Faculty of Science The University of Sydney

This thesis examined factors affecting reproductive success in the captive population and in a wild DFTD affected remnant population of Tasmanian devils in an attempt to provide essential information to the captive management of the species.

The key findings from this research include: i) Reproductive success is suboptimal in the captive population. Although given ample opportunity to mate with previously successful males, healthy female devils that have reached reproductive age are not producing young. The factors underpinning reproductive success are complex. These include the age at which females are first given the opportunity to breed, (with the optimal age being two years Chapters 2, 4 and 5); as well as the age of the male, (with females having significantly more success producing pouch young when paired with older males than with similarly aged or younger ones Chapter 2) and;

ii) that polyandry is indeed a reproductive strategy used by devils in both wild and captive settings, and that litters are frequently sired by multiple males. Incorporating this knowledge into breeding recommendations as a mating strategy could contribute increased genetic diversity into the captive population. Additionally, in DFTD affected areas where older cohorts have succumbed to the disease, younger, one year old males are siring litters

(precocial breeding-Chapter 3), a phenomenon which had previously only been reported in female devils (Jones et al 2008; Lashich 2009), and; iii) that captive devils have higher breeding success when they have been paired with mates that have opposing heterozygosity at MHC loci i.e. when one partner has high heterozygosity, the other has low. These findings are the first reported instance of disruptive selection in a marsupial (Chapter 4;

Russell et al. 2018), and; iv) that the timing of pairing between devils, as determined by zookeepers, is usually accurate and does not contribute to reproductive failure. In addition, there does not appear to be an association between the stress hormone corticosterone and reproductive success in either female or male devils (Chapter 5).

155

Faculty of Science The University of Sydney

The data gathered by this large, ongoing program provide a unique and unprecedented opportunity to employ large, long-term datasets to investigate and discuss factors that may affect the fitness of devils within this captive arena.

To address the low breeding success in the captive population, in Chapter 2, annual reports were evaluated, along with breeding recommendations and studbook records from the program’s inception in 2006 until 2012. We discovered that a mean of 39 % of females within the program produced young (range 26-49%). Despite devil housing and enclosures being deemed adequate and breeding recommendations followed, more wild founding female devils than wild male founders failed to reproduce (28% versus 16%). An overall loss of 35% of wild founder lines was apparent by the end of the 2012 breeding season with

54% of female lines (51/95) and 80% of male founder lines (57/71) still active. The average age that females produced their first litter was 2.1 years, corresponding to the age that wild devils sexually mature (Guiler, 1970; Jones et al. 2008). The mean number of pouch young per litter was 2.7 with captive born devils having a mean of 2.6 and wild born devils 2.8 which is lower than their totally wild counterparts where the average litter size is 3.4

(Chapter 2).

We have found that breeding female devils at the age of two years, in line with the species’ reproductive biology (Keeley et al. 2012), is crucial to success as reproductive senescence can begin as early as three years (Keeley et al. 2017; Chapter 2; Chapter 4).

While the reproductive biology of this species has been investigated by Hesterman (2008) and Keeley et al. (2012), scope for further work remains. In Chapter two, we suggest that captive nulliparous females may be experiencing extra, artificially induced, oestrus cycles

156

Faculty of Science The University of Sydney e.g. up to three/year (due to lack of conception and extra cycling) compared to their wild counterparts, which is accelerating reproductive senescence. This trend is evident amongst some mammal species (Penfold et al. 2014). The trend of declining reproductive success with female age was also evident in Chapter four.

Pairing females with older males has also shown to be a significant factor in devil reproductive success (Chapter 2) and can be easily incorporated into future breeding recommendations. This has also been realised in other species, for example, in island foxes

(Urocyon littoralis), male age was attributed to female reproductive success with older males being more successful as sires than males younger than the vixen (Calkins et a.

2013), a fining similar to those reported in this thesis.

Enhancements to breeding success can be achieved if devils are paired according to genotype: we have shown that successful pairings occurred when devils with contrasting levels of MHC heterozygosity were paired (e.g. one with high heterozygosity pairing with one with low heterozygosity) (Milinski 2006; Chapter four) which we believe is the first example of disruptive selection recorded in a vertebrate. Disruptive selection describes a selection bias against the average individual in a population (Maynard Smith, 1966). My results show that reproductive success of the pairs increased when the sexes had opposing number of heterozygous loci (see Figure 4.2). These findings suggest that MHC genotyping can greatly enhance the success of the captive breeding program, and these results may also prove to be transferable to other species where breeding success is limited.

157

Faculty of Science The University of Sydney

At the instigation of the IUCN’s Conservation Breeding Specialist Group (CBSG), following a population viability analysis, decisions were made to incorporate additional large free range enclosures (FRE’s) into the management plan of the Tasmanian Devil

Insurance Program (TDIP) (CBSG 2008). FRE’s are now a major part of the TDIP with installations in Tasmania, Victoria and New South Wales. Devil Ark, in NSW, is one such operation, which in its maiden year of operation reported an 80% breeding success rate of all adult females given the opportunity to breed (John Weigel, personal communication,

Appendix one). In excess of one hundred new born joeys have been contributed to the

TDIP program from Devil Ark between 2011 and 2014 and is one of the program’s success stories. The FREs are large, ranging in size from 2 to 4 ha and animals live in mixed age and sex groups, allowing natural behaviours (including mate choice) to be exhibited and retained. With success in terms of cost reduction in relation to intensive housing (CBSG

2008), and reproductive success, it is suggested that this mode of operation be the archetype of choice for breeding Tasmanian devils into the future. Allowing the devils to free range with conspecifics in natural environments with social structures and the room to exhibit natural behaviours will benefit any animals selected for future release.

Maintaining accurate pedigrees is an ongoing issue for any FRE. With the confirmation of multiple paternity as a reproductive strategy in devils (Chapter 3), it is essential that paternity (and at times maternity) is determined with certainty. The current methodology using DNA microsatellite markers (as used in this thesis) was not always powerful enough to distinguish between similar genotypes due to the lack of genetic diversity amongst the devils (Jones et al. 2004; Siddle et al. 2010). However, next generation techniques as used by Wright et al. (2015), and additional microsatellite markers (Gooley et al. 2017) offer more robust methods of genotyping and parentage assignment.

158

Faculty of Science The University of Sydney

To further examine reproductive parameters in the devil, I have investigated the timing of male and female devil introductions by examining hormone profiles (Chapter 5).

Progesterone levels were examined in female devils to elucidate whether introductions between breeding pairs were correctly timed to oestrus and female receptivity.

Corticosterone levels were also evaluated for both sexes and compared between and within both males and females. The synchronous devil breeding season occurs from late February to March and both physical and behavioral cues signal keepers as to when the female is receptive and thus ready to mate (Kelly, 2007; Hockley, 2011).

The results showed that the majority of devils were paired at the correct time of female receptivity, so this would not appear to be a significant cause of reproductive failure. Since mating was observed in many of the unsuccessfully paired devils (i.e. no pouch young produced) and behaviorally high affiliation scores were assigned over the pairing period, other factors (e.g. genetic incompatibility or stress) rather than hormonal incompatibility and incorrect timing may be the cause of reproductive failure.

Significantly higher corticosterone levels were observed in female devils than males, which are explained as corticosterone profiles closely match the progesterone peaks occurring within the proestrous and luteal phases of the female reproductive system (Keeley, 2012).

At the level deemed significant for corticosterone (i.e. 1.5 s.d. above baseline) both male and female devils showed no difference between or within successful and unsuccessfully paired devils. However, there was a negative correlation with weight and corticosterone in male devils. Large body size in male animals often equates to dominance and breeding success (Husak and Moore, 2008) which can in turn relate to reduced stress due to less competition. However, as these animals were not housed multiply, male dominance should

159

Faculty of Science The University of Sydney not have been a factor. The majority of successful females in this study conceived during their first oestrus, which concurs with previous studies, although the data relating to second and third oestrus periods was limited (Keeley et al. 2012).

As with previous studies conducted on other marsupials (Fanson and Parrot 2015), my work has shown no correlation between corticosterone levels and captive devil reproductive success.

6.1 Conclusion

With over 1100 mammal species listed as endangered, vulnerable or threatened worldwide

(IUCN 2016), insurance populations will become more commonplace and captive species management will need to maximise breeding success. The results of my research will not only significantly benefit future devil researchers and have direct application in improving breeding rates amongst captive Tasmanian devils, but could also contribute to the protection of other endangered species.

The underpinnings of reproductive success are multifactorial and this thesis highlights some of these factors. My investigations into animal husbandry show continued reproductive success of devils kept in Free-Range Enclosures (FREs), an essential component of the aim of the captive breeding management to maintain the overall genetic diversity in the program.

For the first time, my study shows that devil litters are often sired by multiple males in the wild: therefore, by providing captive females with the opportunity to mate with several males, the genetic diversity of offspring, and the overall fitness of the population may

160

Faculty of Science The University of Sydney increase. My studies unambiguously demonstrate that disruptive selection on MHC Class I loci is a significant determinant of devil reproductive success. Consequently, my findings suggest that it would be highly beneficial for the captive devils to be MHC genotyped in order to enhance the success of the breeding program of this highly endangered and iconic species.

In conclusion, my thesis provides a blueprint of how datasets generated in the course of captive breeding programs can be used for understanding the life and natural history of endangered species, and how that knowledge can be applied to enhancing reproductive success and species survival.

6.2 Future directions

The findings from this thesis can be incorporated into captive management practices for this species and lead to further research.

As captive Tasmanian devils are now being released into the wild in Tasmania (Thalmann et al. 2016; STTDP, 2017), and are therefore in contact with DFTD, extensive monitoring of their survival and health is required. New technologies in animal monitoring (e.g. the use of drones) are becoming available which enable large areas over difficult terrains to be monitored efficiently (D. Saunders, Wildlife Drones, pers comm). Although some of the animals released from captivity have been vaccinated against DFTD (STTDP, 2017), current vaccine technologies have not been proven to be 100% effective (Hendricks et al.

2016) and there is currently no vaccine for DFT2 (K. Belov pers. comm.). Moreover, three vaccinated devils released at Narawntapu have developed tumours (STTDP 2017).

Furthermore, immunisation trials conducted in captivity so far only show incomplete

161

Faculty of Science The University of Sydney protection to DFTD (Kreiss et al. 2015; Tovar et al. 2017) and as yet there is no evidence confirming that field immunisation trials have had any prophylactic effects. In addition, their epidemiological consequences are unknown. Future studies are therefore essential to determine the efficiency and long-term impact of these DFTD immunisation trials.

Genotyping of wild devils surviving in areas where DFTD occurs could aid managers, who in turn could release captive devils with similar genotypes, thus improving their chances of surviving without developing DFTD. Releasing animals with disparate genotypes to those surviving in DFTD affected areas may add novel alleles into these populations with low genetic diversity but they may also, upon breeding, dilute resistant genotypes, thus refuelling DFTD virulence (Hendricks et al. 2016; Russell et al. 2018). This could further threaten the fitness of localised sustaining populations who have shown phenotypic plasticity and allelic variations as adaptations to their present-day environmental conditions. One of the recommendations of the Captive Breeding Study Group (2008) was for the program to breed resistant genotypes if they could be identified in any wild populations.

6.2.1 Mate choice and behavioural trials

More research could be conducted within the large free ranging facilities such as Devil Ark.

Captive MHC and mate choice trials could be undertaken at these large free range facilities where males of differing MHC types could be placed in separate pens on either side of target females and mate choice determined according to which male she chooses to spend the most time with as in trials with Y-mazes (e.g. Yamazaki et al. 1976; Radwan et al.

2008).

162

Faculty of Science The University of Sydney

To understand the social structures and interactions of these iconic animals, the free range enclosures would facilitate behavioural observation in real time via video cameras.

Research suggests (Chapter two) that stress could play a part in the loss of pouch young between birth and weaning as seen in other marsupial species (e.g. Shultz et al. 2006;

Foster et al. 2008; FitzGibbon, 2015) and along with mate choice, may prove to be restrictive in regard to reproductive successes. It would also be beneficial to monitor social behaviour and the behaviours of animals that are recently transferred to facilities as part of the TDIP breeding scheme to determine whether any agonistic or atypical behaviours occur. According to studbook records (Srb, 2015) a number of animals have died within six months of transfer between facilities throughout the insurance program and it would be beneficial to determine whether animals being transferred are suffering from acute stress.

6.2.2 Hormone monitoring

Monitoring hormones through faecal analysis in free ranging enclosures could be partnered with behavioural monitoring via video to determine whether there are any hierarchal or filial relationships amongst free ranging devils and also to determine whether animals undergo increased stress when conspecifics are moved or when new animals are added due to annual breeding recommendations and subsequent movement of animals as part of the

Tasmanian devil captive management plan. As this is a non-invasive practice, it can be done with minimal disruption to the animals and can be incorporated into routine husbandry duties. Additionally, hormone monitoring of animals that are to be moved for breeding purposes prior to and post move would be beneficial in determining whether the handling stress is detrimental to the animals: if so, practices could be implemented to minimise it.

163

Faculty of Science The University of Sydney

Additionally, the role of MHC II, which is also thought to play a significant role in mate choice (Sin et al. 2015), is yet to be explored in the devil. Ten MHC II linked microsatellites have been identified, awaiting final optimisation. Analysing the role of

MHC Class in mate choice and reproductive success could provide further information on the multiple factors influencing reproductive success, immune capacities and overall genetic diversity of the species.

Overall, the TDIP has been very successful in terms of the releases (and the subsequent breeding of these released animals) of captive animals back into the wilds of Tasmania and into free ranging managed situations i.e. Maria Island and the Forestier Peninsula

(STTDP 2017). Time will tell if the devils released into the wild can overcome DFTD threats and whether wild populations benefit from an influx of new animals released to genetically rescue the local populations (K. Belov pers. com). It is noteworthy that no long-term reduction in genetic diversity in DFTD affected populations has been recorded (Hendricks et al. 2017), most likely due to long range male-biased dispersal

(12-100km) (Lachish et al. 2011) and the polyandry shown in this study, which in combination can result in the maintenance of allelic richness.

Presently, the future of the devil since DFTD looks brighter than it ever has before, but due to the recent emergence of a novel transmissible cancer, DFT2, the captive breeding program will still have an important role in ensuring the survival of the world’s largest extant marsupial carnivore.

164

Faculty of Science The University of Sydney

6.3 References

Calkins, E. S., Fuller, T. K., Asa, C. S., Sievert, P. R. and Coonan, T. J. (2013) Factors influencing reproductive success and litter size in captive island foxes. The journal of wildlife management 77(2), 346-351.

CBSG (2008) Tasmanian devil population and habitat viability assessment final report.

IUCN/SSC Conservation Breeding Specialist Group, Apple Valley, MN.

FitzGibbon, S. I. (2015) Reproductive ecology of the northern brown bandicoot

(Isoodon macrourus) in habitat fragments of urban Brisbane. Australian Mammalogy

37, 253-259.

Foster, W. K., Caton, W., Thomas, J., Cox., S. and Taggart, D. A. (2008) Timing of birth and reproductive success in captive red-tailed phascogales, Phascogale calura. Journal of Mammalogy 89(5), 1136-1144.

Frankham, R., Ballou, J. D. and Briscoe, D. A. (2002) Introduction to Conservation

Genetics. Cambridge University Press, Cambridge, UK.

165

Faculty of Science The University of Sydney

Guiler, E. R. (1970) Observations on the Tasmanian devil, Sarcophilus harrisii

(Marsupialia:Dasyuridae) II Reproduction, breeding, and growth of pouch young.

Australian Journal of Zoology 18, 63-70.

Hendricks, S., Epstein, B., Schönfeld, B., Wiench, C., Hamede, R., Jones, M., Storfer,

A. and Hohenlohe, P. (2017) Conservation implications of limited genetic diversity and population structure in Tasmanian devils (Sarcophilus harrisii). Conservation Genetics,

18: 4, 977-982.

General and Comparative Endocrinology 155, 234-244.

Hockley, J. (ed.) (2011) DPIPWE/ZAA. Husbandry Guidelines for Tasmanian Devil,

Sarcophilus harrisii. DPIPWE, Tasmania, Australia.

Husak, J. F. and Moore, I. T. (2008) Stress hormones and mate choice. Trends in ecology and evolution 23(10), 532-534.

IUCN redlist www.iucnredlist.org Accessed April 10, 2016.

Jones, M. E., Paetkau, D., Geffen, E. and Moritz, C. (2004) Genetic diversity and population structure of Tasmanian devils, the largest marsupial carnivore. Molecular

Ecology 13:2197–2209.

166

Faculty of Science The University of Sydney

Jones, M. E., Cockburn, A., Hamede, R., Hawkins, C., Hesterman, H., Lachish, S.,

Mann, D., McCallum, H. and Pemberton, D. (2008) Life-history change in disease- ravaged Tasmanian devil populations. Proceedings of the National Academy of Sciences of the USA 105 (29), 10023-10027.

(2), 182-191.

Keeley, T., Russell, T., Carmody, K., Kirk, G., Eastley, T., Britt-Lewis, A., Post, M.,

Burridge, M., Eccleston, S., Faulkner, T., Forge, T., Leonard, J. and Hughes, R.L. (2017)

Seasonality and breeding success of captive and wild Tasmanian devils (Sarcophilus harrisii). Theriogenology 95, 33-41.

Kelly, A. (2007) Husbandry guidelines for the Tasmanian devil, Sarcophilus harrisii.

Zoo and Aquarium Association, Sydney.

Kreiss, A., Brown, G. K., Tovar, C., Lyons, A. B. and Woods, G. M. (2015) Evidence for induction of humoral and cytotoxic immune responses against devil facial tumor disease cells in Tasmanian devils (Sarcophilus harrisii) immunized with killed cell preparations.

Vaccine 33, 3016-25.

Lachish, S., Miller, K.J., Storfer, A., Goldizen, A.W., et al. (2011) Evidence that disease- induced population decline changes genetic structure and alters dispersal patterns in the

Tasmanian devil. Heredity 106, 172-82.

167

Faculty of Science The University of Sydney

Maynard Smith, J. (1966) Sympatric speciation. The American Naturalist 100 (916),

637-650.

Milinski, M. (2006) The Major Histocompatibility Complex, Sexual Selection, and

Mate Choice. Annual Review of Ecology, Evolution and Systematics 37, 159-186.

Penfold, L. M., Powell, D., Traylor-Holzer, K. and Asa, C. S. (2014) "Use it or lose it":

Characterization, implications and mitigation of female infertility in captive wildlife.

Zoo Biology 33, 20-28.

Radwan, J, Tkacz, A. and Kloch, A. (2008) MHC and preferences for male odour in the bank vole. Ethology 114, 827-833.

Russell, T., Madsen, T., Thomas, F., Raven, N., Hamede, R. and Ujvari, B. (2018)

Adaptive evolution in the face of a transmissible cancer. BioEssays 1700146.

Save the Tasmanian devil program (STTDP) tassiedevil.com.au accessed November 7,

2017.

Shultz, D. J., Whitehead, P. J. and Taggart, D. A. (2006) Review of surrogacy program for endangered Victorian brush-tailed rock wallaby (Petrogale penicillata) with special reference to animal husbandry and veterinary considerations. Journal of Zoo and

Wildlife Medicine 37(1), 33-39.

168

Faculty of Science The University of Sydney

Sin, Y. W., Annavi, G., Newman, C., Buesching, C., Burke, T., MacDonald, D. W. and

Dugdale, H. L. (2015) MHC class II-assortative mate choice in European badgers

(Meles meles). Molecular Ecology 24, 3138–3150.

Srb, C. (2015) Australasian Regional Tasmanian devil studbook: Sarcophilus harissii.

Thalmann, S., Peck, S., Wise, P., Potts, J. M., Clarke, J. and Richley, J. 2016

Translocation of a top-order carnivore: tracking the initial survival, spatial movement, home-range establishment and habitat use of the Tasmanian devils on Maria Island.

Australian Mammalogy 38, 68-79.

Tovar, C., Pye, R. J., Kreiss, A., Cheng, Y., Brown, G. K., Darbyl, J., Malley, R. C.,

Siddle, H. V. T., Skjødt, K., Kaufman, J., Silva, A., Morelli, A. B., Papenfuss, A. T.,

Corcoran, L. M., Murphy, J. M., Pearse, M. J., Belov, K., Lyons, A. B. and Woods, G.

M. (2017) Regression of devil facial tumour disease following immunotherapy in immunised Tasmanian devils. Scientific Reports 7, 43827.

Wright, B., Morris, K., Grueber, C. E., Willet, C. E., Gooley, R., Hogg, C., O’Meally,

D., Hamede, R., Jones, M., Wade, C. and Belov, K. (2015) Development of a SNP- based assay for measuring genetic diversity in the Tasmanian devil insurance population. BMC Genomics 16, 791.

169

Faculty of Science The University of Sydney

Yamazaki, K., Boyse, E. A., Mike, V., Thaler, H. T., Mathieson, B. J., Abbott, J., Boyse,

J., Zayas, Z. A. and Thomas, L. (1976) Control of mating preferences in mice by genes of major histocompatibility complex. Journal of Experimental Medicine 144, 1324-

1335.

170

Faculty of Science The University of Sydney

7 Appendix One: The Devil’s Ark, the mark of success

7.1 Abstract

In order to ensure that the Tasmanian devil (Sarcophilus harrisii) will not face extinction, a large scale captive breeding program commenced in 2005, but overall breeding success has unfortunately not been optimal (average <40% success rate). However, an exception is the captive breeding conducted on a 500 hectare, free range breeding facility of pristine forest at Barrington Tops (mid north NSW) - Devil Ark. Devil Ark is an integral player in the

Tasmanian Devil Insurance Program (TDIP) and was established by the Australian Reptile

Park in conjunction with the Save the Tasmanian Devil Program (STTDP) and the Zoo and

Aquarium Association (ZAA). The preliminary stages of Devil Ark were completed in

2010. In 2011, the site was operational and within its maiden year 80% of eligible females successfully produced young. This was followed up with a 78% success rate in 2012 which is exceptionally high for captive bred devils. Between 2011 and 2014, Devil Ark had contributed over 100 new born joeys into the program, and has one of the highest breeding success rates of all facilities. Devil Ark provides a unique example for threatened species recovery: the animals are housed in groups in spacious enclosures allowing them to exhibit natural behaviours and interact with conspecifics. Importantly, facilities such as Devil Ark also offer animals free ranging conditions that encourage wild type behaviours that may offer the animals an enhanced chance of success upon release. Devil Ark therefore represents an innovative and successful augmentation to the intensive holdings of devils in zoos and fauna parks, with the advantage of greatly improved cost-efficiency and conditions more conducive to the preservation of wild-type behaviours.

171

Faculty of Science The University of Sydney

7.2 Introduction

The world’s largest marsupial carnivore, the Tasmanian devil (Sarcophilus harrisii), is under the threat of extinction from Devil Facial Tumour Disease (DFTD). First discovered in 1996 at Mt William in Tasmania’s north east, this fatal contagious cancer has decimated wild devil populations by over 90% in some areas of Tasmania and has now reached more than 85% of the state (STTDP, 2014). The disease is progressively spreading to uninfected devil populations on an east to west trajectory and the majority of disease free devils can be found in the northwest of Tasmania (Hamede et al. 2013). The threat of DFTD to the survival of the wild devil populations has resulted in that they been listed as Endangered on the IUCN Red List of Threated Species (IUCN, 2014). To safeguard this species from futher declines in numbers an insurance program was developed in 2006 with the aim of conserving 95% of the wild genetic diversity in Tasmanian devils for the next 50 years

(STTDP, 2014).

In this chapter I provide a short overview of the Tasmanian devil captive breeding program, and then focus on the innovations that have been developed at Devil Ark, one of the largest and most successful breeding facilities within the program.

7.2.1 The Tasmanian Devil Insurance Program

Captive breeding programs are often based on population viability models (PVM’s; Traill et al. 2010). These models reflect the minimum number of animals required to maintain a genetically viable populations into perpetuity (or for a given time frame e.g. the next 50 years). The Tasmanian Devil Insurance Population (TDIP) originated in 2005. In 2008, a devil population and habitat viability assessment (PHVA) was conducted by the

International Union for the Conservation of Nature (IUCN) Captive Breeding Specialist

172

Faculty of Science The University of Sydney

Group (CBSG). As a result of the PHVA, the goals of the program was to achieve and mantain 500 breeding animals (CBSG, 2008; STTDP 2014). Wild caught disease free devils from various locations around Tasmania made up the founding captive population.

Unfortunately, the overall breeding success has not reached the targets set out in the

PHVA. However, an exception is the captive breeding conducted on a 500 hectare, free range breeding facility of pristine forest at Barrington Tops (mid north NSW), Devil Ark.

This free-range breeding facility is leading the way in Tasmanian devil captive breeding with exceptional breeding success and a unique, cost effective model facility for this species, where pedigrees and genetic polymorphism are scientifically managed through a partnership between a private institution and a scientific body and where free ranging animals can exhibit natural behaviours.

Devil Ark, managed by The Australian Reptile Park (ARP), is a large scale (500 ha) free range facility for the Tasmanian devil and is an integral part of the Tasmanian devil insurance program. Proposals for the Devil Ark model were presented at the PHVA workshop in 2008, where considerable discussion and broad informal support from participants was offered. Interest in the concept appeared to stem from the anticipated cost- effectiveness of a large captive population, and that more natural conditions would provide a range of benefits to the species – including the exhibition of natural behaviours and mate choice (CBSG, 2008). Maintaining behaviours displayed in the wild is paramount for a threatened species that may be released back into the wild and free range facilities lessen the risks of genetic selection for captive adaptation (Frankham, 2008).

The targeted conditions would need to be conducive to the preservation of wild-type social and ecological behaviour over time. The solution was for a modular facility, wherein

173

Faculty of Science The University of Sydney clusters of medium-sized enclosures (1-5ha) - each with a social contingent of Tasmanian devils, could be contained within a high-security perimeter fence.

Devil Ark is located on 500 hectares in northern New South Wales at Barrington Tops

32.0528° S, 151.4936° E, with an elevation of 1,500 m. The climate is mild with the average temperatures ranging from 7 – 12 oC in winter and 17 – 26 oC in summer

(www.environment.nsw.gov.au 2014). The climate at Barrington Tops is similar to that observed in Hobart in southern Tasmania (www.bom.gov.au, 2014).

Construction of the first stage was complete in late December 2010, consisting of four 2-4 ha breeding pens, two 1 ha crèche yards, and a bank of ten 100 m2 holding pens. An additional three breeding pens and two crèche yards for senescent devils were added during

2011. The four 2-4 ha pens were stocked in 2011, with equal number of males and females in either a 4 x 4 or 3 x 3 arrangement (John Wiegel unpublished data). Devil Ark began operation with 44 devils in total (devilark.org.au). An additional three breeding pens with the same configurations of devils was operational in 2012. In 2011, eight out of 10 (80%) females successfully bred, producing 23 young and in the following year 39 joeys were born to 14 of 18 (78%) paired females (Srb, 2015; John Weigel unpublished data). Thirty one joeys were born in 2013 and a further 22 in 2014 (Srb, 2015). The sex ratio between male and female joeys was close to parity in all years except 2014 where there were more females born than males (Table 7.1).

174

Faculty of Science The University of Sydney

Table 7.1.1 The number of Tasmanian devil joeys born at Devil Ark 2011-2014 (source: Tasmanian devil studbook 2015 (Srb 2015)) The fate of this *joey is unknown as is the detail of its unknown sex status.

Year Male Female Unknown sex 2011 11 12 2012 18 21 2013 14 17 2014 6 15 1*

As a result of social interaction, three injuries have been detected amongst mature devils –

all being minor. One death has been recorded, however the cause remains unknown (John

Wiegel unpublished data). In spite of this, there have been no recorded outward signs of

social dysfunction (or behavioural maladaptation) within the captive devil population (John

Wiegel, Tim Faulkener, personal communication). Consequently, Devil Ark has proven to

be a highly successful, cost effective model for threatened species recovery allowing devils

to exhibit natural behaviours in wild-type settings whilst maximising reproductive output.

A hurdle facing captive breeding programs with free ranging or multi stocked pens is that

of parentage. For the Tasmanian devil, a range of microsatellite markers have been used for

both paternity and maternity assignment of joeys. Microsatellites are short different sized

base-pair tandem repeating segments of DNA. However, due to their low genetic diversity

devil parental assignment is not always achievable e.g. microsatellite analysis proved to be

insufficient to distinguish potential sires of two males at 12 microsatellite loci (Russell

unpublished data). The combination of novel microsatellites (Gooley et al. (2017) and the

use of next generation sequencing techniques now being introduced and will most likley

remedy this situation in the future (Wright et al. 2015).

175

Faculty of Science The University of Sydney

7.3 The future

Ongoing expansion of Devil Ark is scheduled to continue until 2016, where after a stable population of 320 breeding devils will be housed in this facility (John Wiegel, unpublished data). The cost per devil has been estimated to decrease by an order of magnitude at Devil

Ark compared to the intensives at ARP further demonstrating the advantage to this mode of captive management (CBSG, 2008; John Wiegel, unpublished data). Beyond the intermediate aim of achieving and managing a population of 320 Tasmanian devils in a genetically-effective and cost-efficient manner, there is sufficient space on the property to more comprehensively meet the needs of STTDP/ZAA meta-population strategy, with the potential carrying capacity of 1,000 Tasmanian devils (John Wiegel, unpublished data).

Importantly, facilities such as Devil Ark that offer animals free ranging conditions that encourage wild type behaviours may offer the animals an enhanced chance of success upon release. For example, mathematical modelling proposed that long-term captive breeding has an accumulative effect on devil behaviour, and may potentially result in a higher probability of captive-raised animals being fatally struck by vehicles (Grueber et al 2017).

Tasmanian devils raised in a close to natural environment, such as Devil Ark may have a higher survival rate once returned to the wilds of Tasmania. In support of this, recently devils from the Ark have been succssfully released into the wild e.g. onto Maria Island

(Thalmann et al. 2016) and the Forestier Peninsula. Devil Ark therefore represents an innovative and successful augmentation to the intensive holdings of devils in zoos and fauna parks, with the advantage of greatly improved cost-efficiency and conditions more conducive to the preservation of wild-type behaviours.

176

Faculty of Science The University of Sydney

7.4 References

Belov, K. (2011) The role of the Major Histocompatibility Complex in the spread of contagious cancers. Mammalian Genome 22, 83-90.

Brook, B. W., Bradshaw, C. J. A., Traill, L. W. and Frankham, R. (2011) Minimum viable population size: not magic, but necessary. Trends in Ecology and Evolution 26

(12), 619-620.

Bureau of Meteorology www.bom.gov.au accessed September 9, 2014.

Devil Ark www.devilark.org.au accessed September 9, 2014.

Frankham, R. (2008) Genetic adaptation to captivity in species conservation programs.

Molecular Ecology 17, 325- 333.

Guiler, E. R. (1970) Observations on the Tasmanian devil, Sarcophilus harrisii

(Marsupialia: Dasyuridae) II Reproduction, breeding, and growth of pouch young.

Australian Journal of Zoology 18, 63-70.

177

Faculty of Science The University of Sydney

Grueber, C., Reid-Wainscoat, E., Fox, S., Belov, K., Shier, D., Hogg, C. and Pemberton,

D. (2017) Increasing generations in captivity is associated with increased vulnerability of Tasmanian devils to vehicle strike following release to the wild. Scientific Reports 7,

1-7.

Hamede, R. K., McCallum, H. and Jones, M. (2013) Biting injuries and transmission of

Tasmanian devil facial tumour disease. Journal of Animal Ecology 82, 182-190.

Hawkins, C. E., Baars, C., Hesterman, H., Hocking, G. J., Jones, M. E., Lazenby, B.,

Mann, D.,Mooney, N., Pemberton, D., Pyecroft, S., Restani, M. and Wiersma, J. (2006)

Emerging disease and population decline of an island endemic, the Tasmanian devil

Sarcophilus harrisii. Biological Conservation 131, 307-324.

Hollings, T., Jones, M., Mooney, N. and McCallum, H. (2014) Trophic cascades following the disease-induced decline of an apex predator, the Tasmanian devil.

Conservation Biology 28 (1), 63-75.

IUCN redlist www.iucnredlist.org accessed September 9, 2014.

178

Faculty of Science The University of Sydney

Jones, M. E., Cockburn, A., Hamede, R., Hawkins, C., Hesterman, H., Lachish, S.,

Mann, D., McCallum, H.and Pemberton, D. (2008) Life-history change in disease- ravaged Tasmanian devil populations. PNAS 105 (29), 10023-10027.

182-191.

Kelly, A. (2007) Husbandry guidelines for the Tasmanian devil, Sarcophilus harrisii. In,

S. Brice (ed.) www.zooaquarium.org.au Sydney: Zoo and Aquarium Association 1-32.

Kreiss, A., Cheng, Y., Kimble, F., Wells, B., Donovan, S., Belov, K. and Woods, G. M.

(2011) Allorecognition in the Tasmanian devil (Sarcophilus harrisii), an endangered marsupialspecies with limited genetic diversity. PLoS One 6 (7), e22402.

Vaccine 33 (26), 3016-3025.

McCallum, H. and Jones, M. (2011) Sins of omission and sins of commission: St

Thomas Aquinas and the devil. Australian Zoologist 35 (2), 307-314.

Murchison (2010) The Tasmanian devil transcriptome reveals Schwann cell origins of a clonally transmissible cancer. Science 327, 84-87.

179

Faculty of Science The University of Sydney

Pearse, A. M. and Swift, K. (2006) Transmission of devil facial-tumour disease. Nature

439, 549.

Pemberton, D. and Renouf, D. (1993) A field study of communication and social behaviour of the Tasmanian devil at feeding sites. Australian Journal of Zoology 4, 507-

526.

Pyecroft, S. B., Pearse, A., Loh, R., Swift, K., Belov, K., Fox, N., Noonan, E., Hayes,

D., Hyatt, A.,Wang, L., Boyle, D., Church, J., Middleton, D. and Moore, R. (2007)

Towards a case definition for Devil Facial Tumour Disease: What is it? EcoHealth 4,

346-351.

Save the Tasmanian Devil Program (2014) www.tassiedevil.com.au

Siddle, H. V. and Kaufman, J. (2013) A tale of two tumours: Comparison of the immune escape strategies of contagious cancers. Molecular Immunology 55, 190-193.

Srb, C. (2014) Australasian Regional Tasmanian devil studbook: Sarcophilus harissii.

Srb, C. (2015) Australasian Regional Tasmanian devil studbook: Sarcophilus harissii.

Thalmann, S., Peck, S., Wise, P., Potts, J. M., Clarke, J. and Richley, J. 2016

Translocation of a top-order carnivore: tracking the initial survival, spatial movement, home-range establishment and habitat use of the Tasmanian devils on Maria Island.

Australian Mammalogy 38, 68-79.

180

Faculty of Science The University of Sydney

Traill, L. W., Brook, B. W., Frankham, R. R. and Bradshaw, C. J. A. (2011) Pragmatic population viability targets in a rapidly changing world. Biological Conservation 143,

28-34.

Wright, B., Morris, K., Grueber, C. E., Willet, C. E., Gooley, R., Hogg, C., O’Meally,

D., Hamede, R., Jones, M., Wade, C. and Belov, K. (2015) Development of a SNP- based assay for measuring genetic diversity in the Tasmanian devil insurance population. BMC Genomics 16, 791.

181

Faculty of Science The University of Sydney

8 Appendix Two: Oncogenesis as a selective force: Adaptive

evolution in the face of a transmissible cancer

Published BioEssays 2018, 1700146 DOI.10.1002/bies.201700146

Tracey Russell1, Thomas Madsen2, Frédéric Thomas3, Nynke Raven2, Rodrigo

Hamede2,4, Beata Ujvari2,4*

1School of Life and Environmental Sciences, The University of Sydney, Sydney, NSW,

2006, Australia

2Centre for Integrative Ecology, School of Life and Environmental Sciences, Deakin

University, Waurn Ponds, Victoria, 3218, Australia

3CREEC/ MIVEGEC, UMR IRD/CNRS/UM 5290, 911 Avenue Agropolis, BP 64501,

34394 Montpellier Cedex 5, France

4School of Biological Sciences, University of Tasmania, Private Bag 55, Hobart,

Tasmania, 7001, Australia.

*Corresponding author

182

Faculty of Science The University of Sydney

8.1 Abstract

Similar to parasites, malignant cells exploit the host for energy, resources and protection, thereby impairing host health and fitness. Despite cancer being widespread in the animal kingdom, its impact on life history traits and strategies have rarely been documented. Devil facial tumour disease (DFTD), a transmissible cancer, afflicting

Tasmanian devils (Sarcophilus harrisii), provides an ideal model system to monitor the impact of cancer on host life-history, and to elucidate the evolutionary arms-race between malignant cells and their hosts. In the current review we first provide an overview of parasite induced host life history (LH) adaptations, then focus on the phenotypic and genetic evolution of Tasmanian devils in the face of DFTD. We conclude that, akin to parasites, cancer can directly and indirectly affect devil LH traits and trigger host evolutionary responses. Consequently, it is important to consider oncogenic processes as selective force in wildlife.

Keywords: adaptation, evolutionary arms-race, life-history traits, reproductive strategies, cancer immunotherapy, Tasmanian devil, devil facial tumour disease

183

Faculty of Science The University of Sydney

8.2 Introduction

Host-pathogen coevolution, first described by Haldane in 1949 [1], may often result in strong selective pressure on both the host and the pathogen. The process is driven by the reciprocal evolution of (i) host resistance (i.e. genetic, physiological and phenotypic) that reduces the negative effects of pathogen establishment, survival, and/or development within the host and (ii) pathogen response to host resistance. Coevolution is therefore an antagonistic process, and the relative selective strength is determined by the level of host resistance and pathogen virulence. The dynamic interactions between the host and its pathogen consequently provide outstanding models for studying evolutionary change [2,3]. Importantly, pathogen infections often result in intense selective pressure making it possible to document the evolution of host resistance over a few generations [4]. There are numerous examples of rapid evolutionary responses in taxonomically diverse host groups such as bacteria [5], plants [6], invertebrates [7], and vertebrates [8].

Similar to other pathogens and parasites (the two terms used interchangeably throughout the text), malignant cells (with or without underlying infections) exploit the host for energy, resources and protection, thereby impairing host health and fitness.

Congruently, cancer has been proposed to be a developing species behaving akin to parasites [9], and therefore imposes significant selective pressure on host adaptive responses (reviewed in [10]). Despite cancer being widespread in the animal kingdom

[11], the impact of cancer on life history traits and strategies have rarely been documented in the wild (reviewed in [10]). An omission mostly stemming from (i) cancer being perceived as a post-reproductive disease, and therefore the evolution of

184

Faculty of Science The University of Sydney adaptive responses against it being unlikely and (ii) the difficulty of long-term monitoring of malignancies in wildlife due to the lack of non-invasive diagnostic markers. Since cancer (including premalignant lesions and progressive tumours) can increase the likelihood of predation and hence reduce organismal fitness [12,13], similar to parasites, natural selection should favour the evolution of adaptations to overcome the fitness reducing impacts of cancer [14]. Evidence indeed exists, although scarce, showing significant life history changes in response to cancer [10]. For example, tumours caused by herpes virus in green turtles (Chelonia mydas) disrupt the turtle’s biologic functions and activity (foraging, diving) and impair predator evasion and feeding [15]. Haemic neoplasia in blue mussels (Mytilus trossulus) causes reduced phagocytic capacity of haemocytes leading to reduced immune function and ultimately to increased mortality [16]. Moreover, laboratory experiments showed that Drosophila females with cancer oviposit earlier than healthy ones [17]. The emergence and spread of a transmissible cancer, devil facial tumour disease (DFTD), within Tasmanian devil populations, provides an ideal model system to monitor the long-term impact of cancer on devil fitness, and to investigate host responses in the evolutionary arms-race of malignant cells and their hosts. In the current review we first provide an overview of parasite induced host life history adaptations, then focus on the phenotypic and genetic evolution of Tasmanian devils in the face of DFTD. Throughout the text we distingusih between the underlying causal mechanisms, being evolutionary change (i.e., change in allele frequencies) and conditional responses to environmental change (e.g. life history responses to increased prey availability).

8.2.1 Host life-history traits, strategies and adaptation in response to parasites

By imposing significant selective pressures on the host, parasites play an important role in the evolution of host life-history (LH) traits, i.e. growth, reproductive strategies and

185

Faculty of Science The University of Sydney longevity [18-20]. To curtail the fitness-reducing impact of parasites, selection favours hosts that are either able to avoid the source of the pathology [21], or prevent the progression of the parasite once infected. If however, parasite development and proliferation are not preventable, the remaining option for the host is to alleviate the fitness costs by increased levels of tolerance to the infection [14].

Parasites use up important host resources that would otherwise be allocated to maintenance, growth and/or reproduction [22,23]. Inter-individual variation in the ability to respond to parasite infections may not only generate concomitant inter- individual variation in LH traits such as fecundity and survival [19,20] but also ‘drive’ behavioural variations affecting parasite encounter and transmission rates between conspecifics [24-27]. For example, great tits (Parus major) infected with avian malaria

(Plasmodium and Leucocytozoon) showed different behavioural patterns, i.e. increased alertness, enhanced problem solving, higher risk-averseness compared to uninfected birds [26].

Host responses to parasites include both behavioural responses such as avoidance of infected conspecifics [24-27], state-dependent dispersal [28] and physiological/immunological responses [22,23]. If, however, the host is unable to resist infection, in order to compensate for the negative fitness costs of parasites, host life- history traits may undergo flexible and adaptive responses [18,29]. For example, by employing faster LH strategies such as reproducing at an earlier age or by increasing reproductive effort the host may partially compensate for the parasite-induced reduction in fitness [30-36].

Apart from the direct impact of parasites on a given host, parasite-induced adaptive mechanisms and life history trait responses may be translated to the next generation via parental programming and genetic/epigenetic inheritance (i.e. parental effects) [37-39].

For example, due to the metabolic demand of harmful parasites parents may not be able

186

Faculty of Science The University of Sydney to allocate sufficient energy to reproduction and therefore produce smaller offspring

[40,41]. Additionally, parental stress may also impact reproductive cell development, i.e. sperm phenotype, and hence alter post-zygotic development and performance [42-

46]. Finally, apart from genetic inheritance of immune genotypes, hormonal and physiological parental adjustments can also be translated to the next generation via epigenetic modifications, resulting in offspring pre-adapted to their parents’ parasite community [47]. Ultimately, transgenerational responses to parasites result in evolutionary changes in the host population, such as changes in allele frequencies, behavioural patterns as well as shifts in reproductive strategies (e.g. early maturation seen in the marine gastropod (Zeacumantus subcarinatus) in areas with high prevalence of castrating trematodes) [48,49].

8.2.2 Host life-history traits, strategies and adaptation in response to a transmissible cancer

8.2.2.1 Tasmanian devil facial tumour disease

Once roaming the entire Australian continent, Tasmanian devils (Sarcophilus harrisii), due to environmental changes and anthropogenic interactions (competition with dingos and humans) have been eradicated from the mainland and during the last ~3,000 years only surviving on the island of Tasmania [50]. Presently, however, devils have become exposed to a novel threat, a transmissible cancer, devil facial tumour disease (DFTD)

[51,52].

First reported at Mount William National Park in north-eastern Tasmania in 1996,

DFTD has caused declines in devil numbers of up to 90% in some areas of Tasmania

[51,53] and as a result the species has been listed as Endangered by the International

Union for Conservation of Nature (IUCN red list) [54]. DFTD is a clonally

187

Faculty of Science The University of Sydney transmissible and fatal cancer transmitted by biting during social interactions (e.g. mating and fighting over food) [55]. DFTD is transmitted as an allograft [56] and tumours often emerge in or around the mouth (Fig 1). The particular ecology and natural history of the devils, being primarily scavengers and having no extant predators, allows the persistence of devils with DFTD. Due to metabolic failure, tumour related weakness and metastases, death results within six to twelve months after the emergence of the first lesions [57].

During the last 20 years DFTD has spread rapidly across Tasmania; in 2005 the disease was present at over approximately 50% of the devil’s range [51] but by 2014 over three quarters of Tasmania’s devil populations had been affected by this devastating disease

[58]. Importantly, in 2016, a novel clonal transmissible cancer (described as DFT2) was discovered, suggesting that during the last decades the two transmissible cancers

(henceforth referred to as DFTD and DFT2) have evolved independently on at least two occasions [59]. The sudden emergence of DFT2 also raises the questions whether previous transmissible cancers have impacted the species [60,61], and whether future variants of the disease or novel transmissible cancers will appear [62].

In addition to the emergence of a novel transmissible cancer, molecular studies of

DFTD (including analyses of karyotypes, methylomes and telomeres) have shown high cancer evolutionary plasticity and adaptability [60,61,63-67]. The combination of the

DFTD’s ability to evolve novel traits (e.g. phenotypic plasticity, epigenetic profiles and karyotypes (reviewed in [60,61]) and the concurrent significant reduction in devil numbers questions the long-term survival of this iconic, keystone species. This scenario was supported by mathematical modelling suggesting that DFTD had the capacity to lead to local extinctions within 15-35 years after the onset of the epidemic outbreak [68-

70]. However, these epidemiological models were deterministic and did not consider individual differences in phenotypic plasticity, host mortality, susceptibility to infection

188

Faculty of Science The University of Sydney or the evolution of adaptive responses to DFTD. Recent studies show significant heterogeneities in several of these parameters, suggesting that devils may be able to overcome the extinction threat of DFTD [71-73]. In this review we therefore discuss

Tasmanian devil genetic and phenotypic traits as well as life history adaptations that could provide the natural “tool box” for the species to overcome the extinction threat of the disease and hence ensure the survival of this iconic marsupial.

8.2.2.2 Devils show genetic adaptations and phenotypic plasticity to DFTD

Previous studies have shown that devils are characterised by low genetic diversity at both neutral (microsatellite) and functional immune genes [74-77]. Despite this, other studies have shown that Tasmanian devils have a functional immune system capable of mounting immune responses to novel antigens [72,78-81]. The low levels of genetic polymorphism have been suggested to have been caused by population declines during the last glacial maximum (i.e. approximately 25,000 years ago) and due to an increased

‘El Niño–Southern Oscillation’ activity during the mid-Holocene (3,000 – 5,000 years ago) [82]. Given that dingos were never introduced to Tasmania, and the aboriginal population density was relatively low during the Holocene [83], climate change has been suggested to be the primary cause of population declines in Tasmania prior to

European settlement [82]. Devil populations in Tasmania have been proposed to have undergone population fluctuations during the 19th and 20th centuries [84,85], due to hunting by early European settlers and due to potential exposure to pathogens, such as a canine-distemper-like disease [74]. The negative impact of human activity over the last

200 years may, however, been ameliorated by the introduction of novel prey such as sheep and during the second half of 20th century by the increased access to road killed prey.

189

Faculty of Science The University of Sydney

The emergence of DFTD has been attributed to the low level of genetic diversity at the non-self-recognising Major Histocompatibility Complex (MHC) immune genes [86].

However, the recent spread of DFTD to areas with genetically disparate devil populations has revealed that DFTD successfully crosses histocompatibility barriers that would normally prevent allograft acceptance demonstrating that devil MHC polymorphism does not affect tumour cell transmission [87]. Rather, DFTD is actively evading the host immune system by down-regulating genes involved in the antigen- processing pathways, resulting in the concomitant loss of cell surface expression of

MHC Class I molecules, and escape of immune recognition [87].

Four recent studies have, however, showed signs of potential genetic/phenotypic adaptations to DFTD. The first study demonstrated that devils with higher relative natural antibody (IgM) levels compared to specific antibody (IgG) levels appear to be less susceptible to the cancer and could potentially be one of the first steps of host evolutionary responses combating the deleterious effects of DFTD [73]. Despite devils showing a significant age-specific decline in immune function [73], the second study showed that devils are capable of naturally mounting immune responses to DFTD resulting in tumour regression [72]. Using genome–wide analyses two additional studies on devils have identified several genomic regions linked to immune function and/or to cancer in humans (allele frequency changes at loci PAX3, TLL1[88], and CRBN,

MCAM, Cbl, THY1, USP2, C1QTNF5 [71]), further suggesting that devils’ are evolving immune-modulated responses to the cancer [71,88]. Considering the devils’ short generation time (2 years) Epstein et al. [71] proposed that the potential genomic response could have evolved within only a few generations. Similar rapid adaptation has been observed in the least killifish (Heterandria formosa) where fish acquired resistance to cadmium within 6 generations [89]. Therefore, genetic adaptations (i.e. allele frequency variation [71,88]) and phenotypic plasticity (i.e. variation in expression levels

190

Faculty of Science The University of Sydney of immunoglobulins [73]) in devils strongly suggest rapid evolutionary responses due to the extreme selection imposed by DFTD [71,88].

Phenotypic plasticity, expression of different phenotypes from a given genotype can confer rapid responses to stressors such as pathogens [90]. For example, the two studies showing immune responses to DFTD [72,73] indicate that devils are potentially using the available building blocks of their immune system, and by altering their temporal regulation and expression, they are able to reduce the impact of the cancer. The emergence of adaptive immune phenotypes does not require changes (e.g. mutations, emergence of novel alleles), but rather relies on the varying usage of the standing genetic material. Therefore, phenotypic plasticity may provide even a genetically depauperate organism, like the devil, with an adaptive potential to combat novel diseases [91].

Persistent challenges by pathogens (including transmissible cancers), will eventually select for increased frequency of resistance-conferring alleles, or even for the emergence of resistant genotypes that provide transgenerational protection to the pathogens. For example, dogs carry the signature of selective responses to another naturally occurring transmissible cancer, canine venereal tumour disease (CTVT)

[92,93]. This transmissible cancer is considered to be the oldest known somatic cell line and has been proposed to have appeared between 6,000 - 11,000 years ago (exact date is still debated) in a post-domestication canid [92-95]. Although CTVT impairs the reproduction of dogs at the time of infection, the disease is rarely fatal, strongly suggesting that the dog’s immune system has evolved to recognise and suppress tumour growth and eventually achieving complete tumour regression [96].

Since transmissible cancer cell lines evolve on the evolutionary landscape of the host genomes [60,61,97], a sufficiently high prevalence of a transmissible cancer can result in a tug-of-war akin to host-parasite interactions between the selfish malignant cell lines

191

Faculty of Science The University of Sydney and host genomes (sensu “Red-Queen” dynamics [23,98-100]). A recent study by

Decker et al. [92] demonstrated such host adaptation to CTVT cells where potential resistance conferring mutations were observed in the dog genome in self-antigen presentation (MHC), immune surveillance, genome integrity maintenance and cell apoptosis regulation pathways. The aetiology of CTVT (initial progression followed by regression) in combination with the observed genetic mutations in the host genome indicate that this transmissible cancer has reached an evolutionary equilibrium with its host over the thousands of years of co-evolution. The results outlined above, as well as recent evidence showing rapid adaptation to DFTD in devils (as discussed above) [71-

73] suggest that, over evolutionary time, the devils and DFTD may indeed reach a similar evolutionary equilibrium as recorded in dogs and CTVT. However, it is important to note that the high transmission frequency and the frequency dependent transmission pattern of DFTD may prolong the evolutionary arms-race between DFTD and its host.

8.2.2.3 Devils show life history adaptation in response to DFTD

Reproductive plasticity - changes in age of first reproduction - conditional responses to environmental change

Prior to the appearance of DFTD the life span of devils in the wild was approximately

5-6 years, although in captivity devils may live for more than 8 years. Both sexes became sexually mature at the age of 2 years [85] and devils were able to reproduce on

3 to 4 occasions throughout their lifetime [85,101]. As DFTD is primarily transmitted during the mating season, when interaction between conspecifics is the highest[102], presently the majority of devils only survive to reproduce once [103]. Due to the significant demographic effects of DFTD, i.e. the high mortality in older devils [104], sexual maturity has been observed in about 50% of wild female devils at the age of one

192

Faculty of Science The University of Sydney year (precocial breeding [103]) and recently, a study showed that some one year old males were also able to breed in DFTD affected populations [105]. These results strongly suggest that DFTD is causing rapid and significant shifts in devil life history processes.

These shifts in reaching sexual maturity may be facilitated by increased availability of quality maternal dens as well as increased prey availability brought on by DFTD decimating devil numbers [103]. The latter being supported by the fact that when devil numbers decline the growth rate of sub-adults increases [104]. Devils need to reach a critical body mass to be able to reproduce and the fact that not all one-year old females are able to breed [103] shows that food availability and the mothers’ health status (being infected or not infected by DFTD) will influence the extents of precocial breeding. An interesting question is how precocial breeding will affect DFTD epidemiology.

Breeding earlier may lead to an earlier transmission of the disease, fuelling and maintaining the epidemic cycle hence questioning whether precocial breeding may be able to compensate for the negative demographic effects caused by the disease.

Reproductive plasticity – the key is in the numbers

Pregnancy in Tasmanian devils lasts for approximately 18 days. Devils are supernumerary and in excess of 30 young may be born. As in all marsupials, the young are born in a very underdeveloped state (body size ~ 5 mm) and make their way from the birth canal to the pouch along a mucus trail, created by the mother licking a path to the pouch. However, as females only have four teats, four young can attach and have a chance to survive whilst the rest perish [85].

Although reproducing such a high number of embryos/may seem an excess of resources, supernumeracy may actually be an evolutionary masterstroke. Producing a large number of embryos may provide an evolutionary palette to enable selection to

193

Faculty of Science The University of Sydney wean out the weak and to ensure that the mother invests her resources in the fittest offspring [106-108]. Despite being an apex predator, devils also frequently feed on dead and dying animals [84] therefore the saliva contains numerous pathogens, such as bacteria [109], viruses, parasites (nematodes and cestodes [110]) and potentially DFTD cells. The necessity of moving into the pouch along the saliva trail may therefore expose the embryos to pathogens and thus initiating and enhancing their immune function (as observed in early-onset sepsis in humans [111]). Importantly, the saliva may also contain immune molecules/antibodies protecting the neonates along their journey to the pouch. For example, in humans, saliva contains numerous defence proteins that are involved in both innate and acquired immunity [112,113]. In addition, high saliva IgA levels have been shown to result in decreased risk of cancer mortality in humans [113].

Furthermore, in dogs and rats, saliva was found to be bactericidal against pathogens implicated in neonatal septicaemia [114,115]. Similar to other mammals, immune molecules in devil saliva may protect the neonates from pathogens and increase their immunological capacities (including tumour suppression). Although, transgenerational immune priming (including immunoglobulin transfer via milk and prenatally via the yolk sac placenta) of marsupial offspring is well known [116] the saliva trail may, however, be an additional maternal strategy to enhance fitness.

Due to the lack of diagnostic markers, the exact route and time of DFTD transmission is still unclear. Hamede et al. [55] proposed that cancer cell transmission primarily occurs through wounds and injuries inside the mouth of healthy devils that have bitten into the tumours of sick animals. Since social interaction increases when devils reach sexual maturity, exchange of tumour cells have been proposed to occur mostly in adults.

However, there is a possibility that mothers affected by DFTD may expose their undeveloped pouch young to cancer cells by either licking or via the mucous trail which

194

Faculty of Science The University of Sydney may result in (i) selection of offspring with enhanced tumour suppression capacities, or

(ii) prime their immune system to recognise malignant cells later on in their life, or (iii) provide them with protective molecules such as IgA which may reduce tumour development.

Sexually immature devils are rarely observed with tumours [103]. This may, however, be an artefact of the potential latent period of DFTD; if the devils are infected just after weaning, when most of the fights and bites occur, clinical signs of the disease are not likely to appear until devils reach an age of one to two years. This is supported by the fact that the longest latent period in an asymptomatic animal taken from the wild developed tumours after 11 months in captivity (Hamede pers. com). Moreover, the low prevalence of DFTD in young devils may be attributed to juveniles rarely engaging in fights that result in bite wounds necessary for the transmission of DFTD [55,102,117].

Additionally, young devils may potentially be protected by the expression of tumour suppressor genes that are required for proper chromosomal segregation in developing embryos [118-120]. Due to the unique reproduction of marsupials (completing development in the pouch [121]) the activity of these essential genes for early development may be sustained while the developing young occupy the pouch and thus protect them from cancer development. A final potential explanation for the appearance of DFTD in sexually mature devils could be due to a shift in resource allocation to mating and reproduction at the cost of continued tumour suppression. The latter is supported by the decrease in peripheral immune molecule expression levels with advancing age [73] showing that devils in the wild undergo immunosenescence. A recent study showed that in captivity two year old devils had lower immune responses compared to 1 year old devils [122]. Whether the observed age-specific appearance of

DFTD is an artefact of latency or the result of ongoing protection by tumour suppressor mechanisms during development, behavioural differences (lack of injurious contacts

195

Faculty of Science The University of Sydney that result in disease transmission) and/or life history trade-offs remains unanswered until a diagnostic test becomes available.

Reproductive plasticity - multiple paternity and genetic diversity

Genetic diversity is essential for adaptive capacities, providing organisms with the potential of successfully responding to intrinsic and extrinsic challenges. Loss of genetic diversity can lead to inbreeding depression i.e. decreased growth rate, fertility, fecundity and offspring viability [123] as well as in increased vulnerability to pathogens [124]. By allowing enhanced admixture of the genepool, multiple paternity (polyandry) may increase the possibility of maintaining genetic diversity [125].

Male competition for females is fierce in Tasmanian devils, and mating and mate guarding can last up to 15 days [85,126]. However, it has recently been shown that, along with many other marsupials, devil litters are often sired by multiple males [105].

The combination of devils having relatively low genetic diversity, and the significant increase in DFTD induced mortality may therefore further reduce devil genetic diversity

[53]. However, no long-term reduction in genetic diversity in DFTD affected populations has been recorded [76], most likely due to long range male-biased dispersal

(12-100km) [53] and polyandry which in combination result in the maintenance of allelic richness.

Reproductive plasticity – adaptive sex allocation

Female devils with DFTD have been shown to produce significantly more female offspring compared to healthy females [53]. In their classic paper Trivers and Willard

[127] suggested that if maternal condition affects the breeding success of male offspring more than that of female offspring, females in poor condition should bias their offspring sex ratio towards females. The facultative adjustment of sex ratios may therefore be an

196

Faculty of Science The University of Sydney adaptive response to the decreased maternal condition brought on by the physiological and metabolic demands and impacts of DFTD. Consequently, by investing more in daughters than sons may provide higher fitness returns to DFTD affected females [128].

However, since DFTD has decimated the number of devils in the wild resulting in increased prey/food availability, this extra supply of food could improve maternal condition, even in DFTD affected animals, and therefore ultimately return the offspring sex ratio to parity.

8.2.2.4 The tug of war between sexual and natural selection in Tasmanian devils

Due to the significant reduction of devil fitness, natural selection should favour individuals capable of discriminating between infectious and non-infectious conspecifics and hence reduce the spread of the disease. However, as DFTD is frequently transmitted during sexual encounters [102,129], such an adaptive strategy would also signficantly reduce devil mating opportunities with concomitant fitness reductions. Moreover, DFTD is also transmitted during fights over prey such as roadkill, which further increases the risk of disease transmission [102,117]. The disease initiated tug-of war between sexual and natural selection may create an even more complex evolutionary scenario. A recent study [129] showed that due to their dominant and more aggressive behaviour infected devils may actually have higher life time reproductive success compared to non-infected devils. Despite the opposing consequences of natural and sexual selection on devil evolution, depending on the trade- off between avoiding infected individuals and reduced reproductive success, a less aggressive behavior in both sexual and feeding encounters could become fixed and form an evolutionary stable strategy (ESS sensu [130]) and ultimately reduce the spread of

DFTD.

197

Faculty of Science The University of Sydney

8.2.2.5 Co-adaptation of DFTD and devils

Since DFTD negatively affects devil fitness by reducing survival, selection would be expected to evoke host evolutionary responses. As described above, devils indeed appear to respond to DFTD by genetic (allele frequency changes) and phenotypic adaptations (altered expression of immunoglobulins). Parallel to the host, DFTD also may show signs of adaptations. Although no direct evidence for co-evolution between host and DFTD has yet been found, chromosomal [63,65,66], genomic [131] and epigenetic [64,66] studies demonstrate that DFTD is not a static, but rather an evolving clonal cell line. Notably, different tumour variants have been found to have differing epidemic patterns; tetraploid tumours showed reduced force of infection and no demographic effects at a particular population (West Pencil Pine) in north-western

Tasmania [132]. Once diploid tumours replaced the tetraploid variants, disease prevalence increased and devil populations declined, suggesting competition between tumour lineages driven by selection to maximize transmission [132].

In the evolutionary arms race, the cancer appears to be running faster than its host, but devils are rapidly adapting to the disease. Despite the low genetic diversity of devils

[76] the species still has sufficient genetic diversity to provide material for selection to increase the allele frequency of certain cancer associated genotypes and phenotypes that ultimately may provide an increased resistance and /or tolerance to DFTD [71,73]. In most host-parasite systems, where the host’s immune system is responding to a non-self invader, the low frequency of alleles providing resistance could be caused by previous evolutionary responses to either the same parasite or to other similar pathogens

[99,100]. However, DFTD being an altered self-cell line, additional explanations may apply: (i) the genes/phenotypes involved in self / non-self recognition and control of cell growth may also control autoimmune disease development (e.g. immunoglobulins [73]), and/or (ii) devils may have previously been affected by other transmissible cancers. The

198

Faculty of Science The University of Sydney recent appearance of a novel transmissible cancer cell line, DFT2 [59], certainly indicates the high susceptibility of the species to invasive malignant cells. These factors may therefore provide an explanation of the present low frequency of resistance alleles and phenotypes in wild devils.

8.2.2.6 Implications for the conservation of the species

Devils have co-existed with DFTD for > 20 years and therefore we propose that disease management and vaccine development may benefit from considering the devils’ natural defensive immune and tumour suppressor mechanisms. Selection on DFTD has already increased the allele frequency of certain genotypes and phenotypes that may provide protection against tumour development. Therefore, when developing management strategies, future work could consider these evolutionary processes, and place special attention on wild populations that have evolved with DFTD with the aim of supporting the maintenance and increasing the frequency of particular geno-/phenotypes that allow devils to confront the effects of the disease. Furthermore, while immunisation trials conducted in captivity so far only show incomplete protection to DFTD [133,134] there is no evidence confirming that field immunisation trials have had any prophylactive effects and their epidemiological consequences are unknown. Future studies are essential to determine the efficiency and long-term impact of DFTD immunization trials. Notably, nearly all effective human vaccines rely on the proof of concept provided by the immune system generating a protective response [135]. According to the fundamental tenet of vaccinology “the best way to develop an effective vaccine is to design a candidate that mimics infection and induces responses akin to natural immunity” [135]. Combining the evolutionary ecology and immunological knowledge acquired over the last 10 years with vaccine development may aid in exploiting the

199

Faculty of Science The University of Sydney natural immunity of devils. Evolution and the devils’ immune system have already responded to DFTD and given that adaptive processes in the devil-DFTD system are underway, applying these evolutionary ecological principles to the management of

DFTD may help the long-term conservation prospects of the species.

Although this review raises the hope that evolution and natural adaptation hold the key to the survival of the species, further research is essential to validate, confirm and determine the type of adaptations to DFTD and allow its use in future conservation and management of Tasmanian devil populations. Our views highlight the importance of taking Darwinian principles into consideration in the management of endangered species.

8.2.2.7 Lessons learnt from transmissible cancers such as DFTD may be applicable to human cancer immunotherapy

The current resurgence of cancer immunotherapy, harnessing the immune system to combat cancer, raises the hope for the “beginning of the end of cancer” [136]. However, immunotherapies vary in their pathophysiology, suggesting that their effectiveness may be patient and/or cancer specific [137-139]. The Tasmanian devil research further indicates the importance of host responses determining cancer and therapy outcomes.

Furthermore, human immunotherapy has, so far primarily focused on the use of IgG antibodies that target specific tumour associated antigens [138] and has largely overlooked the potential importance of IgM isotypes. The latter has been shown to provide extended tumour immunosurveillance as well as anti-tumour cytotoxic activity in other organisms [140], and IgM antibody therapy has been shown to reduce neuroblastoma and melanoma in humans [141,142]. IgM antibodies detect cancer precursor stages (i.e., autoantigens), bind to various tumour-antigens as well as induce apoptosis of malignant cells [143]. As seen in DFTD, devils with increased IgM relative

200

Faculty of Science The University of Sydney to IgG expression levels had significantly lower DFTD prevalence, suggesting that

IgM/IgG ratios may play an important role in determining devil susceptibility to DFTD.

Rare human studies show similar associations, immunotherapy based on IgM such as

PAT-SM6 indicates a promising avenue for the treatment of melanoma [142]. These results therefore highlight the importance of IgM in combating both animal and human cancers, and may therefore provide novel immunotherapeutic directions. Finally, since both DFTD and human cancers use the same strategy of downregulating tumour MHC expression levels [144], profiling analyses of classical and non-classical MHC molecule expression in both human and devil cancers may provide essential information for future immunotherapy, given their fundamental role in the modulation of immune cell interactions [138].

8.2.3 Concluding remarks

As demonstrated by the evolutionary adaptations of Tasmanian devils to DFTD, akin to parasites, cancer can directly and indirectly affect LH traits and trigger host evolutionary responses. The case study of Tasmanian devils proves that it is important to consider oncogenic phenomena as a selective force altering life-history traits and reproductive strategies in wildlife [145]. Clearly the development of non-invasive diagnostic markers, more data and research are needed to fully understand the evolution of life-history traits in a cancer context. Malignant transformation, including transmissible and infection-induced ones, are part of multicellular life, therefore incorporating them into evolutionary, ecological and species conservation studies will have a substantial impact on understanding the evolutionary ecology of our ecosystems

[145].

201

Faculty of Science The University of Sydney

8.3 Acknowledgements

This work was supported by two Eric Guiler Tasmanian devil Research Grants through the Save the Tasmanian devil appeal of the University of Tasmania Foundation, an

Australian Academy of Science, FASIC Early Career Fellowship to BU, an

International Associated Laboratory Project France/Australia to FT & BU, and a Deakin

Faculty of Science Engineering & Built Environment Research Grant Scheme 2017

(SEBE-RGS17-UJVARI). FT is also supported by the ANR EVOCAN, André

Hoffmann and la Fondation MAVA. RH is supported by a National Science Foundation grant (NSF DEB-1316549) and an Australian Research Council DECRA Fellowship

(DE170101116). We are grateful to two anonymous reviewers and to A/Prof Athena

Aktipis for their valuable comments on our manuscript.

8.4 References

1 Haldane JBS. 2006. Disease and Evolution. Malaria: Genetic and Evolutionary

Aspects. Boston, MA: Springer US. p 175-87.

2 Decaestecker E, Gaba S, Raeymaekers JAM, Stoks R, et al. 2007. Host-parasite

/`Red Queen/' dynamics archived in pond sediment. Nature 450: 870-3.

3 May RM, Anderson RM. 1983. Epidemiology and genetics in the coevolution of parasites and hosts. Proceedings of the Royal Society of London Series B, Biological sciences 219: 281-313.

4 Best SM, Kerr PJ. 2000. Coevolution of host and virus: The pathogenesis of virulent and attenuated strains of Myxoma virus in resistant and susceptible European rabbits.

Virology 267: 36-48.

202

Faculty of Science The University of Sydney

5 Weitz JS, Hartman H, Levin SA. 2005. Coevolutionary arms races between bacteria and bacteriophage. Proceedings of the National Academy of Sciences of the United

States of America 102: 9535-40.

6 Dodds PN, Rathjen JP. 2010. Plant immunity: towards an integrated view of plant– pathogen interactions. Nat Rev Genet 11: 539-48.

7 Ebert D. 2008. Host–parasite coevolution: Insights from the Daphnia–parasite model system. Current Opinion in Microbiology 11: 290-301.

8 Kerr PJ. 2012. Myxomatosis in Australia and Europe: A model for emerging infectious diseases. Antiviral Research 93: 387-415.

9 Duesberg P, Mandrioli D, McCormack A, Nicholson JM. 2011. Is carcinogenesis a form of speciation? Cell cycle (Georgetown, Tex) 10: 2100-14.

10 Ujvari B, Beckmann C, Biro PA, Arnal A, et al. 2016. Cancer and life- history traits: lessons from host-parasite interactions. Parasitology 143: 533-41.

11 Madsen T, Arnal A, Vittecoq M, Bernex F, et al. 2017. Cancer in the animal kingdom. In B. U, Roche B, Thomas F, eds; Ecology and Evolution of Cancer:

Academic Press. p 11-46.

12 Martineau D, Lemberger K, Dallaire A, Labelle P, et al. 2002. Cancer in wildlife, a case study: beluga from the St. Lawrence estuary, Québec, Canada.

Environmental Health Perspectives 110: 285-92.

13 McAloose D, Newton AL. 2009. Wildlife cancer: a conservation perspective.

Nat Rev Cancer 9: 517-26.

14 Thomas F, Guégan J-F, Renaud F. 2009. Ecology and Evolution of

Parasitism.: Oxford University.

15 Page-Karjian A, Norton TM, Krimer P, Groner M, et al. 2014. Factors influencing survivorship of rehabilitating green sea turtles (Chelonia mydas) with

203

Faculty of Science The University of Sydney fibropapillomatosis. Journal of zoo and wildlife medicine : official publication of the

American Association of Zoo Veterinarians 45: 507-19.

16 Carballal MJ, Barber BJ, Iglesias D, Villalba A. 2015. Neoplastic diseases of marine bivalves. Journal of invertebrate pathology 131: 83-106.

17 Arnal A, Jacqueline C, Ujvari B, Leger L, et al. 2017. Cancer brings forward oviposition in the fly Drosophila melanogaster. Ecology and evolution 7: 272-6.

18 Michalakis Y, Hochberg ME. 1994. Parasitic effects on host life-history traits: a review of recent studies. Parasite (Paris, France) 1: 291-4.

19 Sorci G, Boulinier T, Gauthier-Clerc M, Faivre B. 2009. The evolutionary ecology of the immune response.; Ecology and evolution of parasitism. p 5-17.

20 Thomas F, Guegan J, Michalakis Y, Renaud F. 2000. Parasites and host life-history traits: implications for community ecology and species co-existence.

International journal for parasitology 30: 669-74.

21 Hart BL. 1994. Behavioural defense against parasites: interaction with parasite invasiveness. Parasitology 109 Suppl: S139-51.

22 Poulin R. 2007. Evolutionary Ecology of Parasites. Princeton: Princeton

University Press.

23 Schmid-Hempel P. 2011. Evolutionary Parasitology: The Integrated Study of

Infections, Immunology, Ecology, and Genetics. Oxford: Oxford University Press.

24 Boyer N, Reale D, Marmet J, Pisanu B, et al. 2010. Personality, space use and tick load in an introduced population of Siberian chipmunks Tamias sibiricus. J

Anim Ecol 79: 538-47.

25 Bull CM, Godfrey SS, Gordon DM. 2012. Social networks and the spread of

Salmonella in a sleepy lizard population. Molecular Ecology 21: 4386-92.

204

Faculty of Science The University of Sydney

26 Dunn JC, Cole EF, Quinn JL. 2011. Personality and parasites: sex-dependent associations between avian malaria infection and multiple behavioural traits. Behavioral

Ecology and Sociobiology 65: 1459-71.

27 Seaman B, Briffa M. 2015. Parasites and personality in periwinkles (Littorina littorea): Infection status is associated with mean-level boldness but not repeatability.

Behavioural Processes 115: 132-4.

28 Iritani R, Iwasa Y. 2014. Parasite infection drives the evolution of state- dependent dispersal of the host. Theoretical Population Biology 92: 1-13.

29 Hochberg ME, Michalakis Y, De Meeus T. 1992. Parasitism as a constraint on the rate of life-history evolution. Journal of Evolutionary Biology 5: 491-504.

30 Forbes MRL. 1993. Parasitism and host reproductive effort. Oikos 67: 444-50.

31 Minchella DJ, Loverde PT. 1981. A Cost of Increased Early Reproductive

Effort in the Snail Biomphalaria glabrata. The American Naturalist 118: 876-81.

32 Sorci G, Clobert J, Michalakis Y. 1996. Cost of reproduction and cost of parasitism in the common lizard (Lacerta vivipara). . Oikos 76: 121-30.

33 Polak M, Starmer WT. 1998. Parasite–induced risk of mortality elevates reproductive effort in male Drosophila. Proceedings of the Royal Society of London

Series B: Biological Sciences 265: 2197.

34 Adamo SA. 1999. Evidence for adaptive changes in egg laying in crickets exposed to bacteria and parasites. Animal Behaviour 57: 117-24.

35 Vézilier J, Nicot A, Gandon S, Rivero A. 2015. Plasmodium infection brings forward mosquito oviposition. Biology Letters 11.

36 Agnew P, C. Koella J, Michalakis Y. 2000. Host life history responses to parasitism. Microbes and Infection 2: 891-6.

205

Faculty of Science The University of Sydney

37 Mousseau TA, Uller T, Wapstra E, Badyaev AV. 2009. Evolution of maternal effects: past and present. Philosophical Transactions of the Royal Society B:

Biological Sciences 364: 1035-8.

38 Wolf JB, Wade MJ. 2009. What are maternal effects (and what are they not)?

Philosophical Transactions of the Royal Society B: Biological Sciences 364: 1107-15.

39 Poulin R, Thomas F. 2008. Epigenetic effects of infection on the phenotype of host offspring: parasites reaching across host generations. Oikos 117: 331-5.

40 Gallizzi K, Alloitteau O, Harrang E, Richner H. 2008. Fleas, parental care, and transgenerational effects on tick load in the great tit. Behavioral Ecology 19: 1225.

41 Hakkarainen H, Huhta E, Koskela E, Mappes T, et al. 2006. Eimeria- parasites are associated with a lowered mother's and offspring's body condition in island and mainland populations of the bank vole. Parasitology 134: 23-31.

42 Crean AJ, Dwyer JM, Marshall DJ. 2012. Fertilization is not a new beginning: The relationship between sperm longevity and offspring performance. PLOS

ONE 7: e49167.

43 Bromfield JJ, Schjenken JE, Chin PY, Care AS, et al. 2014. Maternal tract factors contribute to paternal seminal fluid impact on metabolic phenotype in offspring.

Proceedings of the National Academy of Sciences of the United States of America 111:

2200-5.

44 Rando OJ. 2012. Daddy Issues: Paternal Effects on Phenotype. Cell 151: 702-

45 Zajitschek S, Hotzy C, Zajitschek F, Immler S. 2014. Short-term variation in sperm competition causes sperm-mediated epigenetic effects on early offspring performance in the zebrafish. Proceedings of the Royal Society B: Biological Sciences

281.

206

Faculty of Science The University of Sydney

46 Crean AJ, Dwyer JM, Marshall DJ. 2013. Adaptive paternal effects?

Experimental evidence that the paternal environment affects offspring performance.

Ecology 94: 2575-82.

47 Sorci G, Clobert J. 1995. Effects of maternal parasite load on offspring life- history traits in the common lizard (Lacerta vivipara). Journal of Evolutionary Biology

8: 711-23.

48 Fredensborg BL, Poulin R. 2006. Parasitism shaping host life-history evolution: adaptive responses in a marine gastropod to infection by trematodes. Journal of Animal Ecology 75: 44-53.

49 Lafferty KD. 1993. The marine snail, Cerithidea californica, matures at smaller sizes where parasitism is high. Oikos 68: 3-11.

50 Brown OJF. 2006. Tasmanian devil (Sarcophilus harrisii) extinction on the

Australian mainland in the mid-Holocene: multicausality and ENSO intensification.

Alcheringa: An Australasian Journal of Palaeontology 30: 49-57.

51 Hawkins CE, Baars C, Hesterman H, Hocking GJ, et al. 2006. Emerging disease and population decline of an island endemic, the Tasmanian devil Sarcophilus harrisii. Biological Conservation 131: 307-24.

52 McCallum H, Jones M. 2006. To lose both would look like carelessness...

Tasmanian Devil Facial Tumour Disease. Unsolved Mystery. Public Library of Science. p 1671 – 4.

53 Lachish S, Miller KJ, Storfer A, Goldizen AW, et al. 2011. Evidence that disease-induced population decline changes genetic structure and alters dispersal patterns in the Tasmanian devil. Heredity 106: 172-82.

54 IUCN. 2001. Red list of threatened species. Version 3.1. IUCN, Gland,

Switzerland. . Available from wwwiucnredlistorg (accessed date xxxx).

207

Faculty of Science The University of Sydney

55 Hamede RK, McCallum H, Jones M. 2013. Biting injuries and transmission of Tasmanian devil facial tumour disease. Journal of Animal Ecology 82: 182-90.

56 Pearse AM, Swift K. 2006. Allograft theory: transmission of devil facial- tumour disease. Nature 439: 549.

57 Hamede R, Lachish S, Belov K, Woods G, et al. 2012. Reduced Effect of

Tasmanian Devil Facial Tumor Disease at the Disease Front. Conservation Biology 26:

124-34.

58 STDP. 2017. http://www.tassiedevil.com.au/tasdevil.nsf/Mapping-the- disease/A140AACCA1B1F6B0CA2576CB0011BD2C. Save the Tasmanian devil

Program.

59 Pye RJ, Pemberton D, Tovar C, Tubioc JMC, et al. 2016. A second transmissible cancer in Tasmanian devils. Proceedings of the Natural Academy of

Sciences 113: 374-9.

60 Ujvari B, Gatenby RA, Thomas F. 2016. The evolutionary ecology of transmissible cancers. Infection, Genetics and Evolution 39: 293-303.

61 Ujvari B, Gatenby RA, Thomas F. 2017. Transmissible cancer, the evolution of inter-individual metastasis. In Ujvari B, Roche B, Thomas F, eds; Ecology and evolution of cancer: Elsevier. p 290.

62 Darwin C. 2009. The origin of species: by means of natural selection, or the preservation of favoured races in the struggle for life Cambridge: Cambridge University

Press.

63 Deakin JE, Bender HS, Pearse A-M, Rens W, et al. 2012. Genomic restructuring in the Tasmanian devil facial tumour: chromosome painting and gene mapping provide clues to evolution of a Transmissible tumour. PLOS Genetics 8: e1002483.

208

Faculty of Science The University of Sydney

64 Ingles ED, Deakin JE. 2015. Global DNA Methylation patterns on marsupial and devil facial tumour chromosomes. Molecular Cytogenetics 8: 74.

65 Pearse A-M, Swift K, Hodson P, Hua B, et al. 2012. Evolution in a transmissible cancer: a study of the chromosomal changes in devil facial tumor (DFT) as it spreads through the wild Tasmanian devil population. Cancer Genetics 205: 101-

12.

66 Ujvari B, Pearse A-M, Peck S, Harmsen C, et al. 2012. Evolution of a contagious cancer: epigenetic variation in Devil Facial Tumour Disease. Proceedings of the Royal Society B: Biological Sciences 280.

67 Ujvari B, Pearse A-M, Taylor R, Pyecroft S, et al. 2012. Telomere

Dynamics and Homeostasis in a Transmissible Cancer. PLOS ONE 7: e44085.

68 Hamede R, Bashford J, Jones M, McCallum H. 2012. Simulating devil facial tumour disease outbreaks across empirically derived contact networks. Journal of

Applied Ecology 49: 447-56.

69 McCallum H, Jones M, Hawkins C, Hamede R, et al. 2009. Transmission dynamics of Tasmanian devil facial tumor disease may lead to disease-induced extinction. Ecology 90: 3379-92.

70 McCallum H, Tompkins DM, Jones M, Lachish S, et al. 2007. Distribution and Impacts of Tasmanian Devil Facial Tumor Disease. EcoHealth 4: 318.

71 Epstein B, Jones M, Hamede R, Hendricks S, et al. 2016. Rapid evolutionary response to a transmissible cancer in Tasmanian devils. Nature

Communications 7: 12684.

72 Pye R, Hamede R, Siddle HV, Caldwell A, et al. 2016. Demonstration of immune responses against devil facial tumour disease in wild Tasmanian devils. Biology

Letters 12: 28120799.

209

Faculty of Science The University of Sydney

73 Ujvari B, Hamede R, Peck S, Pemberton D, et al. 2016. Immunoglubolin dynamics and cancer prevalence in Tasmanian devils (Sarcophilus harrisii). Scientific

Reports 6: 25093.

74 Morris K, Austin JJ, Belov K. 2013. Low major histocompatibility complex diversity in the Tasmanian devil predates European settlement and may explain susceptibility to disease epidemics. Biology letters 9: 20120900.

75 Morris KM, Wright B, Grueber CE, Hogg C, et al. 2015. Lack of genetic diversity across diverse immune genes in an endangered mammal, the Tasmanian devil

(Sarcophilus harrisii). Molecular Ecology 24: 3860-72.

76 Hendricks S, Epstein B, Schönfeld B, Wiench C, et al. 2017. Conservation implications of limited genetic diversity and population structure in Tasmanian devils

(Sarcophilus harrisii). Conservation Genetics: 1-6.

77 Jones ME, Paetkau D, Geffen ELI, Moritz C. 2004. Genetic diversity and population structure of Tasmanian devils, the largest marsupial carnivore. Molecular

Ecology 13: 2197-209.

78 Brown GK, Kreiss A, Lyons AB, Woods GM. 2011. Natural killer cell mediated cytotoxic responses in the tasmanian devil. Plos One 6: (9): e24475.

79 Van Der Kraan LE, Wong ESW, Lo N, Ujvari B, et al. 2013. Identification of natural killer cell receptor genes in the genome of the marsupial Tasmanian devil

(Sarcophilus harrisii). Immunogenetics 65: 25-35.

80 Ujvari B, Belov K. 2015. Characterization of antibody V segment diversity in the Tasmanian devil (Sarcophilus harrisii). Veterinary Immunology and

Immunopathology 167: 156-65.

81 Woods GM, Howson LJ, Brown GK, Tovar C, et al. 2015. Immunology of a transmissible cancer spreading among Tasmanian devils. The Journal of Immunology

195: 23-9.

210

Faculty of Science The University of Sydney

82 Brüniche-Olsen A, Jones ME, Austin JJ, Burridge CP, et al. 2014.

Extensive population decline in the Tasmanian devil predates European settlement and devil facial tumour disease. Biology Letters 10: 20140619. .

83 Lourandos H. 1997. Continent of hunter–gatherers: new perspectives in

Australian prehistory. Cambridge, UK: Cambridge University Press.

84 Guiler E. 1970. Obsevations on the Tasmanian Devil, Sarcophilus harrisii

(Marsupialia : Dasyuridae) I. Numbers, home, range, movements and food in two populations. Australian Journal of Zoology 18: 49-62.

85 Guiler E. 1970. Observations on the Tasmanian Devil, Sarcophilus harrisii

(Marsupialia : Dasyuridae) II. Reproduction, breeding and growth of pouch young.

Australian Journal of Zoology 18: 63-70.

86 Belov K. 2012. Contagious cancer: Lessons from the devil and the dog.

BioEssays 34: 285-92.

87 Siddle HV, Kreiss A, Tovar C, Yuen CK, et al. 2013. Reversible epigenetic down-regulation of MHC molecules by devil facial tumour disease illustrates immune escape by a contagious cancer. Proceedings of the National Academy of Sciences 110:

5103–8.

88 Wright B, Willet CE, Hamede R, Jones M, et al. 2017. Variants in the host genome may inhibit tumour growth in devil facial tumours: evidence from genomewide association. Scientific Reports 7: 423.

89 Xie L, Klerks PL. 2003. Responses to selection for cadmium resistance in the least killifish, Heterandria formosa. Environmental Toxicology and Chemistry 22: 313-

20.

90 Siddle KJ, Quintana-Murci L. 2014. The Red Queen's long race: human adaptation to pathogen pressure. Current Opinion in Genetics & Development 29: 31-8.

211

Faculty of Science The University of Sydney

91 Woolhouse MEJ, Webster JP, Domingo E, Charlesworth B, et al. 2002.

Biological and biomedical implications of the co-evolution of pathogens and their hosts.

Nat Genet 32: 569-77.

92 Decker B, Davis BW, Rimbault M, Long AH, et al. 2015. Comparison against 186 canid whole-genome sequences reveals survival strategies of an ancient clonally transmissible canine tumor. Genome Research 25: 1646-55.

93 Ostrander EA, Davis BW, Ostrander GK. 2016. Transmissible tumors: breaking the cancer paradigm. Trends in Genetics 32: 1-15.

94 Murgia C, Pritchard JK, Kim SY, Fassati A, et al. 2006. Clonal origin and evolution of a transmissible cancer. Cell 126: 477-87.

95 Murchison EP, Wedge DC, Alexandrov LB, Fu B, et al. 2014.

Transmissible dog cancer genome reveals the origin and history of an ancient cell lineage. Science 343: 437-40.

96 Yang TJ. 1988. Immunobiology of a spontaneously regressive tumor, the canine transmissible venereal sarcoma (review). Anticancer research 8: 93-5.

97 Haig D. 2015. Maternal–fetal conflict, genomic imprinting and mammalian vulnerabilities to cancer. Philosophical Transactions of the Royal Society of London B:

Biological Sciences 370: 20140178.

98 van Valen L. 1973. A new evolutionary law. Evolutionary Theory 1: 1-30.

99 Lythgoe KA, Read AF. 1998. Catching the Red Queen? The advice of the

Rose. Trends in ecology & evolution 13: 473-4.

100 Salathé M, Kouyos RD, Bonhoeffer S. 2008. The state of affairs in the kingdom of the Red Queen. Trends in ecology & evolution 23: 439-45.

101 Keeley T, O’Brien JK, Fanson BG, Masters K, et al. 2012. The reproductive cycle of the Tasmanian devil (Sarcophilus harrisii) and factors associated with

212

Faculty of Science The University of Sydney reproductive success in captivity. General and Comparative Endocrinology 176: 182-

91.

102 Hamede RK, McCallum H, Jones M. 2008. Seasonal, demographic and density-related patterns of contact between Tasmanian devils (Sarcophilus harrisii):

Implications for transmission of devil facial tumour disease. Austral Ecology 33: 614-

22.

103 Jones ME, Cockburn A, Hamede R, Hawkins C, et al. 2008. Life-history change in disease-ravaged Tasmanian devil populations. Proceedings of the National

Academy of Sciences 105: 10023-7.

104 Lachish S, McCallum H, Jones M. 2009. Demography, disease and the devil: life-history changes in a disease-affected population of Tasmanian devils (Sarcophilus harrisii). Journal of Animal Ecology 78: 427-36.

105 Russell T. 2016. An investigation into factors affecting breeding success in the

Tasmanian devil (Sarcophilus harrisii). University of Sydney.

106 Stockley P, Macdonald DW. 1998. Why do female common shrews produce so many offspring? Oikos 83: 560-6.

107 Forbes LS, Mock DW. 1998. Parental optimism and progeny choice: when is screening for offspring quality affordable? Journal of Theoretical Biology 192: 3-14.

108 Stearns SC. 2013. The evolution of sex and its consequences: Birkhäuser.

109 Cheng Y, Fox S, Pemberton D, Hogg C, et al. 2015. The Tasmanian devil microbiome—implications for conservation and management. Microbiome 3: 76.

110 Kreiss A. 2017. Overview of Tasmanian Devils. Veterinary Manual: Merck.

111 Simonsen KA, Anderson-Berry AL, Delair SF, Davies HD. 2014. Early-

Onset Neonatal Sepsis. Clinical Microbiology Reviews 27: 21-47.

112 Brandtzaeg PER. 2007. Do salivary antibodies reliably reflect both mucosal and systemic immunity? Annals of the New York Academy of Sciences 1098: 288-311.

213

Faculty of Science The University of Sydney

113 Phillips AC, Carroll D, Drayson MT, Der G. 2015. Salivary

Immunoglobulin A Secretion Rate Is Negatively Associated with Cancer Mortality: The

West of Scotland Twenty-07 Study. PLOS ONE 10: e0145083.

114 Hart BL, Korinek E, Brennan P. 1987. Postcopulatory genital grooming in male rats: prevention of sexually transmitted infections. Physiology & behavior 41: 321-

115 Hart BL, Powell KL. 1990. Antibacterial properties of saliva: role in maternal periparturient grooming and in licking wounds. Physiology & behavior 48: 383-6.

116 Old JM, Deane EM. 2000. Development of the immune system and immunological protection in marsupial pouch young. Developmental & Comparative

Immunology 24: 445-54.

117 Pemberton D, Renouf D. 1993. A field-study of communication and social- behavior of the Tasmanian devil at feeding sites. Australian Journal of Zoology 41:

507-26.

118 Shimizu T, Ho L-L, Lai Z-C. 2008. The mob as tumor suppressor gene is essential for early development and regulates tissue growth in drosophila. Genetics 178:

957-65.

119 Kuang X, Liu C, Fang J, Ma W, et al. 2016. The tumor suppressor gene lkb1 is essential for glucose homeostasis during zebrafish early development. FEBS Letters

590: 2076-85.

120 Levy O. 2007. Innate immunity of the newborn: basic mechanisms and clinical correlates. Nat Rev Immunol 7: 379-90.

121 Sharman GB. 1959. Marsupial Reproduction. In Keast A, Crocker RL,

Christian CS, eds; Biogeography and Ecology in Australia. Dordrecht: Springer

Netherlands. p 332-68.

214

Faculty of Science The University of Sydney

122 Cheng Y, Heasman K, Peck S, Peel E, et al. 2017. Significant decline in anticancer immune capacity during puberty in the Tasmanian devil. Scientific Reports 7:

44716.

123 Keller LF, Waller DM. 2002. Inbreeding effects in wild populations. Trends in ecology & evolution 17: 230-41.

124 O'Brien SJ, Evermann JF. 1988. Interactive influence of infectious disease and genetic diversity in natural populations. Trends in ecology & evolution 3: 254-9.

125 Taylor ML, Price TAR, Wedell N. 2014. Polyandry in nature: a global analysis. Trends in ecology & evolution 29: 376-83.

126 Jones M, Dickman C, Archer M. 2003. Predators with pouches. the biology of carnivorous marsupials: CSIRO Publishing.

127 Trivers RL, Willard DE. 1973. Natural selection of parental ability to vary the sex ratio of offspring. Science 179: 90-2.

128 Sheldon BC. 1999. Sex allocation: at the females' whim. Current Biology 9:

R487-R9.

129 Wells K, Hamede RK, Kerlin DH, Storfer A, et al. 2017. Infection of the fittest: devil facial tumour disease has greatest effect on individuals with highest reproductive output. Ecology Letters 20: 770-8.

130 Smith JM, Price GR. 1973. The Logic of Animal Conflict. Nature 246: 15-8.

131 Murchison Elizabeth P, Schulz-Trieglaff Ole B, Ning Z, Alexandrov

Ludmil B, et al. 2012. Genome sequencing and analysis of the Tasmanian devil and its transmissible cancer. Cell 148: 780-91.

132 Hamede RK, Pearse A-M, Swift K, Barmuta LA, et al. 2015. Transmissible cancer in Tasmanian devils: localized lineage replacement and host population response. Proceedings of the Royal Society of London B: Biological Sciences 282:

20151468.

215

Faculty of Science The University of Sydney

133 Kreiss A, Brown GK, Tovar C, Lyons AB, et al. 2015. Evidence for induction of humoral and cytotoxic immune responses against devil facial tumor disease cells in Tasmanian devils (Sarcophilus harrisii) immunized with killed cell preparations.

Vaccine 33: 3016-25.

134 Tovar C, Pye RJ, Kreiss A, Cheng Y, et al. 2017. Regression of devil facial tumour disease following immunotherapy in immunised Tasmanian devils. Scientific

Reports 7: 43827.

135 Johnston MI, Fauci AS. 2011. HIV Vaccine development — improving on natural immunity. New England Journal of Medicine 365: 873-5.

136 Farkona S, Diamandis EP, Blasutig IM. 2016. Cancer immunotherapy: the beginning of the end of cancer? BMC Medicine 14: 73.

137 Bayer V, Amaya B, Baniewicz D, Callahan C, et al. 2017. Cancer immunotherapy: an evidence-based overview and implications for practice. Clinical journal of oncology nursing 21: 13-21.

138 Emens LA, Ascierto PA, Darcy PK, Demaria S, et al. 2017. Cancer immunotherapy: Opportunities and challenges in the rapidly evolving clinical landscape. European Journal of Cancer 81: 116-29.

139 Schadendorf D, Hodi FS, Robert C, Weber JS, et al. 2015. Pooled Analysis of Long-Term Survival Data From Phase II and Phase III Trials of Ipilimumab in

Unresectable or Metastatic Melanoma. J Clin Oncol 33: 1889-94.

140 Schwartz-Albiez R, Monteiro RC, Rodriguez M, Binder CJ, et al. 2009.

Natural antibodies, intravenous immunoglobulin and their role in autoimmunity, cancer and inflammation. Clinical and Experimental Immunology 158: 43-50.

141 Erttmann R. 2008. Treatment of neuroblastoma with human natural antibodies. Autoimmunity Reviews 7: 496-500.

216

Faculty of Science The University of Sydney

142 Hensel F, Eckstein M, Rosenwald A, Brändlein S. 2013. Early development of PAT-SM6 for the treatment of melanoma. Melanoma Research 23: 264-75.

143 Toubi E, Shoenfeld Y. 2007. Protective autoimmunity in cancer (review).

Oncology reports 17: 245-51.

144 Fassati A, Mitchison NA. 2010. Testing the theory of immune selection in cancers that break the rules of transplantation. Cancer Immunology, Immunotherapy 59:

643-51.

145 Thomas F, Jacqueline C, Tissot T, Henard M, et al. 2017. The importance of cancer cells for animal evolutionary ecology. Nature ecology & evolution 1: 1592-5.

217

Faculty of Science The University of Sydney

9 Appendix Three: Seasonality and breeding success of captive and wild Tasmanian devils (Sarcophilus harrisii)

Published Theriogenology 95 (2017) 33-41

Keeley, T.1, Russell, T.2, Masters, K.2, Kirk, G.3, Schaap, D.4, Eastley, T.5, Britt-Lewis,

A.6, Post, M.7, Hockley, J.4, Burridge, M.8, Eccleston, S.9, Faulkner, T.10, Forge, T.11,

Leonard, J.12, and Hughes, R. L.1

1 School of Agriculture and Food Science, University of Queensland, Gatton

2 School of Life and Environmental Science, University of Sydney, Sydney

3 Taronga Western Plains Zoo, Dubbo

4 Department of Primary Industries, Parks, Water and the Environment, Tasmania

5 Healesville Wildlife Sanctuary, Healesville

6 Taronga Zoo, Sydney

7 Monarto Zoo, Monarto

8 Dreamworld, Coomera

9 Currumbin Wildlife Sanctuary, Currumbin

10 Australian Reptile Park, Gosford

11 Australia Zoo, Beerwah

12 Ballarat Wildlife Park, Ballarat

218

Faculty of Science The University of Sydney

9.1 Abstract

The synchrony and timing of reproductive events are crucially important factors to maximize individual and offspring survival, especially in seasonal environments.

Despite this, very few studies have compared captive to wild populations of seasonally breeding species occurring at different latitudes to evaluate potential environmental factors influencing seasonality and breeding success in captivity. To increase our understanding of the physiological basis of seasonality in Tasmanian devils, records from captive facilities and results from recent reproductive physiology studies on captive animals were compared to historic records from a wild population study (1975 to 1987). Tasmania, the natural habitat of the Tasmanian devil, has a cool, temperate climate, longer days and an increased rate of day length change compared to the mainland captive breeding facilities. As such, environmental factors including weekly maximum temperature, day length and rate of day length change were examined to determine correlations to seasonality and breeding success. Overall, breeding activity began approximately 2 weeks sooner in the captive population than the wild population

(week 5.7 ± 0.6 versus week 7.7 ± 0.5 for devils entering into oestrus during the first two week phase; n = 26 and n = 23 respectively). The birth peak breadth, the period in which 80% of all females entered into their first oestrus, was more extended in the captive than the wild population (6 versus 4 weeks respectively). Rate of day length change was not a predictor of the timing of reproductive activity (P > 0.05) but absolute day length was (P < 0.01). If the timing of reproductive activity is considered against day length rather than date, both the captive and wild population displayed similar distributions of the timing of breeding (13. 0 ± 0.7 hr versus 13. 0 ± 0.7 hr respectively;

P < 0.01) confirming day length as a proximal cue eliciting a physiological response to

219

Faculty of Science The University of Sydney trigger seasonal reproductive activity regardless of location or the rate of day length change. Breeding success of wild devils was high with 75% (127 of 170) either pregnant (n = 44) or in the early stages of lactation (n = 83) when examined. Only 41%

(51 of 123) of captive female devils successfully produced pouch young despite positive behavioural signs or confirmed mating during pairing. Litter size was larger (3.4 ± 0.9 versus 2.8 ± 1.1 joeys per litter) in wild females, who also produced a significantly higher proportion of full litters of 4 joeys (63% versus 33%)(P < 0.01). Temperature had no relationship to the timing of reproductive activity (P >0.05) but was higher for the captive than the wild population (27.7 ± 4.7 oC versus 22.3 ± 2.7 oC respectively) and for the captive population small litter size (n = 1 joey) was correlated to elevated mean maximum temperatures (P < 0.05). The drivers for reproductive success in captive

Tasmanian devils are likely multifactorial, but our results suggest that elevated temperatures associated with shifts in breeding activity and geographical location should be examined further.

9.2 Introduction

The Tasmanian devil (Sarcophilus harrisii) is a dasyurid (carnivorous marsupial) that is endemic to the island state of Tasmania, Australia. It is a scavenger and nocturnal, the largest (6 to 10 kg) of the current extant species classified as dasyurid (Pemberton,

1990; Pemberton and Renouf, 1993). They are considered to be solitary, only congregating to communally feed on large carcasses or to breed (Pemberton and

Renouf, 1993). Although considered historically abundant during the middle and late

20th century, the devil has suffered a significant population decline over the last decade due to the fatal, transmissible facial cancer, commonly known as Devil Facial Tumour

Disease (DFTD) (Hawkins et al., 2006; Lachish et al., 2007). DFTD is transmitted

220

Faculty of Science The University of Sydney through contact, specifically biting during agnostic encounters over carcass feeding or during mating (Pearse and Swift, 2006; Pemberton and Renouf, 1993). The devil lacks an appropriate immune response to the foreign cells and the development of predominantly facial tumours ensures followed by metastases or starvation through the loss of tissue, teeth and bone mass in the nasal-oral region, preventing food consumption (Loh et al., 2006; Pyecroft et al., 2007; Siddle et al., 2007; Woods et al.,

2007). Currently it is estimated that more than 80% of the natural population has disappeared due to DFTD (http://www.tassiedevil.com.au) and as a result the

Tasmanian devils is currently the classified as endangered by the IUCN (IUCN, 2008).

An Insurance Population of Tasmanian devils was initiated with the first intake of 30 juvenile devils (11 males, 19 females) in early 2005. As little was known about DFTD at this time, these devils remained in quarantine for more than a year to ensure they were cancer free before being transferred to one of the first four mainland Australia institutions to become the initial founders of the Insurance Population. The current

Insurance Populations includes captive breeding efforts of intensively managed captive breeding at more than a dozen mainland Australian and Tasmanian facilities and at free- range enclosures (FREs) on mainland Australia, Tasmania and on Maria Island off the coast of Tasmania (Thalmann et al., 2016). Devils have been held and bred in captivity for over a century, yet captive breeding rates of the first 6 years of the Insurance

Population (2007-2012) averaged only 38% (per communication, Australia Zoo and

Aquarium Association), with the cause of reproductive failure of most of the unsuccessful pairings unknown (Keeley et al., 2012; Roberts, 1915). Despite low overall captive breeding success, the Insurance Population continues to increase as a result of captive breeding, increased number of wild founders and the dedication and hard work of animal care staff, researchers, volunteers and veterinarians across

221

Faculty of Science The University of Sydney

Australia under the direction of the Australian Zoo and Aquarium Association (ZAA) and the Department of Primary Industries, Water and the Environment Tasmania

(DPIPWE).

Breeding rates reported for wild Tasmanian devils prior to the effects of DFTD vary

(Green, 1967; Guiler, 1970, 1978; Hughes and Keeley, In Preparation). Possible reasons for the variability in reported wild breeding rates may include natural variations in population densities and food resources throughout Tasmania, differences in the timing of field work for population studies (e.g. if occurring between weaning and breeding, rates may be reported as artificially low), and variability in trapping and evaluation methodology (Guiler, 1982). For example the breeding success rate of female devils at

Granville Harbour between 1966 and 1975 was only 58.9% with an average of 2.67 pouch young per litter (Guiler, 1978). Trapping predominately occurred in the later months of the year (e.g. November) and it is unknown if low success rates and lower litter size is a result of reduced breeding success or an increased loss of pouch young prior to weaning (Guiler, 1978). As well, at this time of the year, it would also be very difficult to distinguish adult female devils from females that were coming into their second year in life and had yet to breed, potentially artificially decreasing the estimated breeding rate. Accurate methods to age a devil through tooth eruption and wear had yet to be established (Pemberton, 1990) and it appears that only body weight was used to distinguish age class (Guiler, 1978). Another study used the combination of body weight and pouch development as determinates of age class for female devils as immature or juvenile females have a shallow, small, undeveloped pouch (Hughes and

Keeley, in preparation). This study found that 75% of evaluated female devils considered sexually mature were pregnant or lactating at the time of evaluation (Hughes and Keeley, In Preparation) which is likely still an underestimation as some females

222

Faculty of Science The University of Sydney examined had not yet come into oestrus. Despite the effects of DFTD, breeding success in the wild in areas where DFTD is present have been reported as high (81 - 100%)

(Lachish et al., 2009) which continues to beg the question as to why captive breeding rates are low?

The lifetime reproductive potential of the female Tasmanian devil is limited due to a combination of seasonal breeding and reproductive senescence typically by the age of five. Initially believed to be monoestrous, we now know that the female devil is facultative polyoestrous (Guiler, 1970; Keeley et al., 2012). Within a breeding season, a female devil has the potential of undergoing up to three oestrous cycles, at intervals of approximately two months, if conception is not achieved or if pouch young are lost at or shortly after birth (Keeley et al., 2012). Although female devils have been recorded as breeding successfully as young as 1 year of age, this is limited to the small number of animals that exceed the minimum body weight threshold that is likely needed to become a precocial breeder (Lachish et al., 2009). Generally female devils breed between the ages of 2 to 4 years, which equates to the production of a maximum of 3 litters per lifetime (although rare, females have been recorded to exceed this and produce a fourth litter in their fifth year of life). Despite giving birth to supernumerary young, a maximum of only four embryos can survive due to the presence of only four teats in the pouch (Hughes, 1982).

Tasmanian devil reproductive physiology is unique and remarkable. Female devils are presumed to have sperm storage crypts within their oviducts, storing sperm in crypts of the isthmus region of the oviduct similar to other dasyurids such as the brown antechinus (Antechinus stuartii), for a period of approximately one week prior to ovulation (Keeley et al., 2012; Selwood and McCallum, 1987). The spermatozoa of the

223

Faculty of Science The University of Sydney

Tasmanian devil are long compared to eutherian species, with an overall length of 253

µm (head 11.1 µm, midpiece 34.4 µm, and tail 207.4 µm)(Hughes, 1965) and follow a sinusoidal mode of motility due to dorsal-ventral flattening of the midpiece (Keeley et al., 2011). Compared to most eutherian species, sperm concentrations are low with an average of only 1.33 ± 0.85 x 106 spermatozoa per male being able to be extracted by post-mortem gamete rescue from the epididymides (Keeley et al., 2011). Low sperm numbers available for ejaculation are further exacerbated by low overall percentage of sperm motility in the cauda epididymis (29.4 ± 16.8 %) (Keeley et al., 2011). Despite this, fertilization rates are relatively high (80%) (Hughes, 1982) suggesting that spermatozoa are efficient in traversing the female reproductive tract and fertilizing the oocytes, similar to other dasyurids (Taggart, 1994). Superovulation occurs with approximately 40 oocytes (up to 114 oocytes in a single ovulation event recorded) quickly entering the oviduct (Hughes, 1982; Hughes and Keeley, 2013). Similar to other marsupial species, fertilization must occur in the ampulla of the oviduct as penetration will be blocked by the deposit of tertiary investments, the shell membrane and mucoid coat, during the rapid decent of the oocyte to the uterus (Taggart, 1994). The gestation of the Tasmanian devil is short, 12 to 13 days, with implantation occurring during the last 2 to 2.5 days prior to the birth of embryonic young (Hughes, 1982; Hughes and

Keeley, 2013; Keeley et al., 2012). Similarities in progesterone and prostaglandin secretion during the short gestation and the luteal phase, suggest an autonomous ovary and presumption of pregnancy regardless of mating success (Keeley and Dehnhard

2011, Keeley, McGreevy et al. 2012). Data from the laproscopic removal and evaluation of uteri from wild Tasmanian devils confirms superovulation, high fertilization rates and production of supernumerary young yet captive devil breeding rates remain suboptimal

(Hughes 1982, Keeley, McGreevy et al. 2012, Hughes and Keeley 2013).

224

Faculty of Science The University of Sydney

Although we know the devil is a seasonal breeder, we know little about the physiological mechanisms that drive this process in this species. Length of day or

“photoperiod” is the primary factor involved in regulating the timing of reproduction in most seasonal species. Melatonin is an indole amine hormone produced by the pineal gland during the dark phase of the day with variations in production associated with day length (Hazlerigg and Simonneaux, 2015). Melatonin plays a role in regulating the production of GnRH and the processing of thyroid hormones in the hypothalamus, a part of the primary pathway of the neuroendocrine control of reproduction (for more information and a full review, please refer to (Hazlerigg and Simonneaux, 2015)).

Changes in melatonin production are the primary trigger for seasonal breeders, synchronizing reproductive activity within the population as well as optimizing the timing of offspring births. Overall, gestation and lactation duration and the timing of the optimal period of food resources are related to the timing of reproductive activity for seasonal breeders and dictate the physiological response to either an increase or decrease in melatonin production beyond a critical threshold. A short-day breeder is defined as a species which becomes reproductively active as the days are getting shorter

(autumn) and are in anoestrus in spring and summer. A long-day breeder is defined as a species which becomes reproductively active as the days are getting longer (spring, summer) and are typically in anoestrus in winter.

Mammals use environmental signals to time reproduction to take advantage of the most favourable time of year to maximize individual fitness and offspring survival. For many species, this means timing births to occur in warm seasons and when energy resources are abundant. As marsupials have short gestations and give birth to altricial, embryonic young, the optimal timing for breeding has different parameters. For marsupials, it is the timing of late lactation and weaning that must coincide with optimal resource

225

Faculty of Science The University of Sydney abundance as this represents both the period of greatest maternal investment and access to the necessary energy requirements for offspring survival. Photoperiod and the rate of day length change have been confirmed to be key cues for timing seasonal reproduction in marsupials (Dickman, 1985; McAllan et al., 2006; McAllan et al., 1991). The

Tasmanian devil is classified as a short-day breeder with the onset of oestrous cyclicity associated with late summer and autumn (Guiler, 1970; Keeley et al., 2012) but the specific physiology associated with the timing of reproductive activity has yet to be examined.

For many wildlife species, captive breeding programs are global. This means that animals are often expected to cycle and breed not only in artificial, captive conditions but in environments that often differ in latitude, longitude and altitude to their native habitat. This often means differences in absolute day length, seasonal amplitudes of day length, temperature and humidity. For example, Tasmanian devils naturally occur in the island state of Tasmania which is colder and more southerly than any other Australian state. Historic wild devil breeding records confirm most breeding and births occur over a short period in March-April, but our initial examination of the data from captive devils suggest a more extended and earlier breeding season (Hughes and Keeley, In

Preparation; Keeley et al., 2012). Seasonal breeding facilitates births and weaning to occur at the optimal time for food abundance, environmental conditions and ultimately offspring survival. As food resources are rarely a factor limiting or controlling the timing of reproduction in a captive setting, photoperiod is likely the primary driver of seasonality in captive animals. There is limited information available on the relationship of altered day length, rate of change in day length, and temperature to the timing and success of oestrous cycles and mating in captive environments outside of native habitats

(Marneweck et al., 2013; Rademaker and Cerqueira, 2006; Woolley, 1991; Zerbe et al.,

226

Faculty of Science The University of Sydney

2012). This review aims to examine these factors by using data gathered during previous wild and captive population research studies to gain a better understanding of seasonality and environmental factors affecting breeding success in female Tasmanian devils.

9.3 Methodology

The data for this project has been collated from historic breeding season records in addition to retrospective data from completed or unpublished research studies on captive and wild Tasmanian devils and evaluated against location specific environmental parameters.

9.3.1 Captive Breeding Population Data

Captive breeding records for adult female devils housed at 9 mainland Australia institutions and 2 Tasmanian institutions were collated to represent a range of geographical locations (Table 1) over a 6 year timeframe (2007 to 2102). Records for one year old female devils were only included if they successful produced young or displayed typical signs of oestrus and were observed to be mated (n = 3). Otherwise, female devils were between 2 and 4 years of age (n = 84 individuals) and total records

(n = 123) included 1 to 3 years of data for each female. Data from captive female devils was sourced from a previous research study (Keeley et al., 2012), an unpublished research study (of co-authors T. Keeley and T. Russell) or were obtained directly from each institution to increase study animal numbers where possible. Not all records from every year, for every institution were available for this study. Data from captive female devils included, devil’s name and studbook number, year of birth, body weight (where available), institutional location, dates of the beginning of the oestrous cycle

227

Faculty of Science The University of Sydney

(determined by signs of behavioural oestrus which was followed by attempted mating or

associated changes in faecal progesterone concentrations), and production of young (0

to 4). As the date of the beginning of oestrus for each female may be subjected to

human error, data for any oestrus cycles which were not considered reliable were

excluded from the study.

Table 9.1 The specific location and number of Tasmanian devils examined in this study.

Devils housed in areas within a short distance (less than 200 km) of each other were analysed together to maximize numbers per Australian state where needed. Maximum daily temperatures were measured at each given weather station and averaged to provide weekly maximum temperatures for evaluation.

Numb er of Weather Locality Record Years Location Station Latitude Longitud Tasmania (Wild s e Population) 169 1974 to 1987 Avoca District 91022 - 41.78 147.72 Southern 17 2007 to Australia Zoo 40284 - 26.86 152.96 Queensland 2012 Dreamworld 40764 - 27.86 153.31 Currumbin Wildlife 40764 - 28.14 153.49 Victoria 23 2009 to HealesviSanctuarylle 2012 Wildlife 86383 - 37.65 145.52 BallaratSanctuary Wildlife Sanctuary 86383 - 37.56 14.87 South Australia 18 2008 to 2012 Monarto Zoo 24584 - 35.12 139.14 New South 29 2008 to 2011 Taronga Western Wales (Inland) Plains Zoo 65070 - 32.22 148.58 New South 28 2008 to Taronga Zoo 66037 - 33.87 151.21 Wales (Coastal) 2011 Australian Reptile Park 66037 - 33.43 151.34

9.3.2 Wild Population Data

Data from a 13 year study (1975 to 1987) by co-author Dr. Leon Hughes on wild

Tasmanian devils trapped in north-eastern Tasmania, in the Avoca district, were used to

compare natural breeding dynamics to those of captive breeding facilities (Hughes and

228

Faculty of Science The University of Sydney

Keeley, In Preparation). Data from wild Tasmanian devils included, body weight, reproductive status (lactating, pregnant, non-pregnant, pre-ovulatory, post-ovulatory, or pubescent), pouch size (small and undeveloped or large and active), trapping date and pouch young size (head length, weight). Female devils under 4 kg body weight with small, undeveloped pouches were excluded from consideration as these were considered immature devils (approximately 1 year of age). Devils between the weights of 4 – 4.9 kg were included only if lactating or confirmed pregnant. All devils of a body weight of

5 kg and above were estimated to be sexually mature and were included regardless of reproductive status and pouch size. At the time of the study, methods to accurately age a devil using tooth eruption and wear had not yet been established (Pemberton, 1990) and therefore a combination of body weight and pouch development was used to distinguish immature from mature female devils.

The primary aim of the population study at the time was to evaluate gestation and embryonic development in the Tasmanian devil (Hughes, 1982). As such, trapping occurred primarily between late February and late May (Hughes and Keeley, In

Preparation). Any live trapped female devil that was not lactating was subjected to a laparotomy to determine reproductive status. If she was confirmed to be pre-ovulatory, post-ovulatory, pregnant, recently post-partum, or lactating then the date of birth was estimated using an embryonic development timetable or pouch young growth rate

(Hughes, 1982; Hughes and Keeley, 2013, In Preparation). From this, the beginning of the oestrous cycle, the initiation of oestrus, was estimated to be an average of 26 days prior to birth based on previous research on the length of the oestrous cycle of the

Tasmanian devil (Keeley et al., 2012). For both wild and captive devils, the day on which oestrus started was designated a week number for evaluation. Due to the method of evaluating the dates of oestrus of captive and wild devils, it is hoped that evaluating

229

Faculty of Science The University of Sydney this on a week basis should provide the data with an allowance of a few days either side of the specific calendar date to help reduce error due to biases from collection and calculation data.

Birth peak breadth (BPB) (Zerbe et al., 2012) was calculated to determine the shortest time frame in which that majority of the first oestrous cycles began for the captive and wild population. We used a BPB of 80%, determined to be the shortest number of weeks in which 80% of all oestrous cycles began for each locality.

9.3.3 Seasonal Fluctuations in Temperature and Daylight

Sunset and sunrise times and daily maximum temperatures were acquired for each study location (or the closest weather station with available data) for each year as needed from

Geoscience Australia (http://www.ga.gov.au) and the Australian Bureau of Meteorology

(http://www.bom.gov.au). Sunset and sunrise times were used to calculate individual day lengths and then rate of change of day length (or rate of change in photoperiod) was calculated by subtracting each successive calculation for the day length from the preceding day length to provide a value representing the difference in photoperiod from one day to the next. All values for temperature, day length and rate of day length change were averaged for each week of the breeding season (1 January to 30 June) to reduce bias associated with extreme daily temperature fluctuations. Day length did not varying between years for each location therefore a single year’s information was used for all animals at each location.

9.3.4 Statistics

All statistical analyses were performed using GenStat 16th Edition (VSN International

Ltd., UK). A general linear model was used to test the effects of day length, rate of day length change, temperature, body weight class (3.0 to 4.9 kg versus 5 kg +), age and

230

Faculty of Science The University of Sydney location on the timing (week) of the onset of oestrus. Data from female devils across multiple years were assumed to be independent. As the timing of the onset of oestrus may have an impact on reproductive success, we looked for a relationship between the litter size and the timing of oestrus. Logistic regressions were performed using the following variables: location (captive versus wild), litter size (1, 2, 3, or 4 pouch young), week of oestrus, and average weekly maximum temperature. P < 0.05 was considered to be significant. Data are presented as ± SD, unless otherwise noted.

9.4 Results

9.4.1 Wild Population of Tasmanian Devils

The records from the wild population study provided information for 170 adult female devils (n=167 individuals; n= 3 sampled in two separate years). Additional records from female devils with a body weight under 4.9 kg with a small, under-developed pouch were considered pre-pubertal and not included in this study. A total of 41 females between 4 – 4.9 kg were found to be reproductively active (pregnant or lactating) confirming sexual maturity and therefore were included for analysis in this study.

Of the 170 individual records, 36 (21%) female devils were not pregnant nor had small pouch young. Of these 36, six were noted to have cystic uteri or cystic ovaries or a combination of both and one was confirmed to be at least 5 years of age (due to a capture record 4 years prior) and was therefore likely post-reproductive. Another 17 females showed signs of latent lactation suggesting the females were in the final stages of weaning young from the previous year and therefore had yet to enter into oestrus.

The remaining non-reproductive females (n = 18) had no indications to the possible cause of the lack of pregnancy or lactation. As the timing of oestrus could not be

231

Faculty of Science The University of Sydney calculated for females that were not pregnant nor carried small pouch young, these devils were not included in any further evaluations.

Of the 134 reproductively active, wild female devils, 83 were in early lactation with 1 to

4 pouch young and 44 were confirmed pregnant or within 3 days post-parturition through laproscopic evaluation. Another seven were confirmed pre-ovulatory, within a week of ovulation and therefore were included for evaluation.

9.4.2 Litter Size

Of the records from captive female devils, 43 of 123 (35%) successfully produced pouch young (2.9 ± 1.0 joeys per litter) during their first oestrous cycle. Of the remaining 80 females, 66 (83%) were confirmed to have a second oestrous cycle, of those 16 females (20%) were confirmed to have a third oestrous cycle. A total of 6 females (6 of 66; 9% of 2nd oestrous cycles or 6 of 80; 8% of females unsuccessful on the 1st oestrous cycle) successfully produced pouch young on the second oestrous cycle and 2 females (2 of 16; 13% of 3rd oestrous cycles or 2 of 80; 3% females unsuccessful on previous cycles) produced pouch young on the third oestrous cycle. For all litters produced by captive female devils regardless of oestrous cycle number (n = 51 of 123;

41%), litter size was 2.8 ± 1.1 joeys per litter. Of the 83 wild females observed with pouch young, litter size (3.4 ± 0.9 joeys per litter) was overall larger than those of captive female devils (P > 0.05). Wild devils had a higher percentage of litters with the maximum number of joeys (4) than captive females (Fig. 9.1).

232

Faculty of Science The University of Sydney

Figure 9.1.1 Litter size distributions for wild and captive female devils.

* Denotes statistical significance in difference between the percentage of total litters that include 1 to 4 pouch young between captive and wild female devils (P < 0.05).

9.4.3 Timing of Seasonal Initiation of Breeding Activity

Body weight class (3 to 4.9 kg versus 5 kg +) in wild devils had no relationship to the timing of the breeding activity but in captive devils, smaller females came into their first oestrus later than larger females (10.0 ± 3.6 versus 8.1 ± 2.0 weeks respectively; P <

0.05). For captive females (as age data was not available for wild devils), body size was linked to age as smaller females tended to be younger females (2 years of age). No weight data was available for the 1 year old devils but these three devils all came into oestrus late in the season (week 14 – 19).

9.4.4 Environmental conditions associated with the initiation of Breeding Activity

Day length and rate of day length change varied with geographical location (Table 2).

The earliest date for the initiation of the first oestrous cycle in a captive female devil was the 23 January (2010), with a total of 8 females entering into oestrus during the last quarter of January (24 to 31 January) and another 17 females during the first quarter of

233

Faculty of Science The University of Sydney

February (1 to 7 February). This included females from all localities with a range of

ages 2 to 4 years. Breeding success of this first wave of females for this oestrous cycle

was 27% (7/26). Conversely, the earliest date for the initiation of the first oestrous cycle

in a wild female devil was 4 February (1981), the only female coming into oestrus

during the first quarter of February. Four wild female devils entered into oestrous during

the second quarter of February (8 to 14 February) and another 18 during the third

quarter of February (15 to 21 February). Overall, reproductive activity was initiated 2

weeks earlier (Table 2) in captive than wild devils with the first two week phase of

devils entering into oestrus during week 5.7 ± 0.6 or week 7.7 ± 0.5 for captive and wild

devils respectively (P < 0.05) but the day length at which this occurred was the same

(13.8 ± 0.3 hrs and 13.8 ± 0.2 hrs respectively, P < 0.01). In the wild, reproductive

activity was more synchronized than in captivity when evaluated by date (week) and the

BPB was 4 and 6 weeks respectively (Fig. 2a). Figure 2b demonstrates that seasonality

in female devils has a strong correlation to day length (P < 0.01), with both the captive

and wild populations demonstrating a similar distribution of females entering into

oestrus if considered against day length, standardized for the weekly day length

intervals found in Tasmania, the devil’s natural habitat. Rate of day length change was

not a predictor of the timing of the onset of breeding activity in female devils (P > 0.1).

Table 9.2 Summary of the mean values for the timing of the beginning of the first estrus (week).

Rate of day length change (min/d), day length (hr) and maximum temperature (oC). All factors

were evaluated as weekly averages or the week in which the beginning of estrus was confirmed or

calculated to occur. Data is ± SD. * Denotes statistical significance in difference (P < 0.01)

between the captive and wild populations.

Rate of Day

Number of Length Change Maximum

New South Wales (Inland) 29Records 9.1 ± 3.4 1.8(Minutes/Day) ± 0.2 Day 12.8Length ± 0.7 (Hr) 29.7Temperature ± 3.7

Locality Week (oC)

234

Faculty of Science The University of Sydney

Southern Queensland 17 9.0 ± 2.0 1.6 ± 0.1 12.7 ± 0.4 29.1 ± 1.2 South Australia 18 6.9 ± 1.5 2.0 ± 0.2 13.5 ± 0.3 30.0 ± 5.0 New South Wales (Coastal)

Victoria 2328 8.310.2 ± ± 2.2 3.8 2.32.0 ± 0.20.1 13.212.6 ± 0.60.9 26.825.5 ± 3.93.1 Tasmania (Wild) 134 10.2 ± 2.2* 2.8 ± 0.1* 13.0 ± 0.7 22.3 ± 2.7* Captive Overall 115 8.8 ± 3.0* 2.0 ± 0.3* 12.9 ± 0.7 28.1 ± 4.0* Tasmania (Wild) - First 2

Captiveweek phase Overall - First 2 23 7.7 ± 0.5** 2.6 ± 0.1** 13.8 ± 0.2 25.0 ± 1.6** week phase 24 5.7 ± 0.6** 1.8 ± 0.2** 13.7 ± 0.2 32.3 ± 4.7**

The latest date for the initiation of the first oestrous cycle in a captive female devil was

9 May (2008) with another 3 females entering into their first oestrous cycle of the year

between 29 April and 7 May. All of the captive female devils which entered into their

first oestrous cycle late in the season (n = 10), week 14 or later (29 March or later),

were either 1 or 2 years of age, this being their first year of sexual maturity. Four of

these (2 years of age) for which body weights were available were all under 4 kg body

weight at the beginning of the breeding season. For wild females (n = 11) entering into

their first oestrous cycle of the year during a similar timeframe (29 March or later), age

could not be determined but these females had an average body weight of 5.8 ± 1.1 kg.

Mean maximum daily temperature was higher for the captive population than for the

wild population (Table 2; P < 0.01). There was no correlation with maximum weekly

temperature and the timing of the onset of the breeding season. For captive female

devils, litters sizes of only 1 pouch young were correlated with higher weekly

temperatures than those with litter sizes of 4 pouch young (P < 0.05). No relationship

between temperature and litter size was observed in the wild population (P > 0.1).

235

Faculty of Science The University of Sydney

Figure 9.2 The estrous cycle timing for wild and captive devils. The timing, date by week intervals, of the beginning of the estrous cycle for wild (n =

134) and captive (n = 115) devils. Lines indicate the period over which 80% of the birth peak breadth occurs. 2b.The timing, by day length, of the beginning of the estrous cycle for wild and captive devils. The weekly day length intervals are standardized to those which naturally occur in Tasmania.

9.5 Discussion

For global captive breeding programs for threatened and endangered wildlife, the gold standard is to try to replicate natural conditions, including environmental and social context, within the restrictions of a captive environment to ensure the retention of natural behaviours and to optimize animal health and breeding success. To do this, a thorough understanding of the species within their natural environment is essential. This study examined differences in environmental conditions between wild devils in their natural habitat of Tasmania to those held in captive breeding centres throughout

Australia to increase our understanding of location (latitude and longitude) specific factors associated with seasonality and breeding success.

Photoperiod and rate of change of day length and their ability to control seasonality has been examined in a number of marsupials, including some of the smaller dasyurids

236

Faculty of Science The University of Sydney

(Dickman, 1985; McAllan et al., 2006; McAllan et al., 1991). All antechinus are semelparous with males dying shortly after the brief, remarkably synchronous breeding season (Bradley, 2003). The timing of this highly synchronized breeding season has been confirmed to be controlled by a combination of endogenous circannual rhythm, photoperiod and rate of change of photoperiod (Dickman et al., 1987). The results from our study demonstrate that in the wild, most Tasmanian devils breed over a 4 week period in late summer, early autumn. When the timing of the first oestrous cycle for females within the captive population is considered against day length instead of date, the results show a similar distribution as the wild population. This confirms that seasonality in female devils is strongly associated with day length, during the period of decreasing day length. This physiological response suggests that it is likely that changes in melatonin production initiates this response and triggers reproductive activity.

Captive devils housed at latitudes which are lower (geographically more northern to

Tasmania), exhibit a phase shift and extension of the first oestrous cycle window. For the highly synchronized breeding season of the brown antechinus, it has been shown that a phase delay of the natural photoperiod by two months delays the timing of the breeding season in accordance of the delay (McAllan et al., 1991). The authors hypothesize that it is the rate of day length change that plays a critical role in the control of reproduction in the antechinus (McAllan et al., 1991). Conversely, our results suggest that perhaps the progression to a critical day length, with a specific direction of day length change, is important in the control of reproduction in devils but is not necessarily rate dependant. The rate of change of day length is dependent on the geographical location, specifically altitude and latitude, and therefore each rate of change of day length would also be related to the very specific day length of that geographical location. If day length remains static and the rate of change decreases, this can shift or

237

Faculty of Science The University of Sydney extend the breeding activity. We have demonstrated that differences in magnitude of the rate of change of day length did not change the absolute day length at which reproductive activity occurred for devils at differing geographical locations, but extended the period of time over which this occurred. This extension of the window in which the first oestrous cycle occurs is due to an extension in the number of days over which the same day lengths occur naturally in Tasmania. An expansion of the period over which reproductive activity occurs may lead to a less synchronous breeding season. For the devil, this is unlikely to be a concern in regards to synchronizing male and female reproduction as male devils produce sperm throughout the year (Keeley et al., 2011). However, for the antechinus where males die shortly after breeding, the period of breeding must be well synchronised and occur over a short period; lengthening the period in which females enter oestrus may decrease breeding success if males have exhausted their sperm reserves or are already disappearing from the population.

With different antechinus species inhabiting the same geographical area, there is also a differential response to rate of day length change associated with body weight, with larger species breeding earlier, ensuring the isolation of reproductive activity between sympatric species (McAllan et al., 2006). Although we did not observe a correlation with body weight and the first wave of females to enter into oestrus, we did note that in captive devils, small females tended to breed later in the season, supporting the hypothesis that body weight thresholds may also factor into individual seasonal variation. For antechinus it is likely it is a very specific, population dependant relationship between rate of change in day length and absolute day length, in addition endogenous circannual rhythm and body size, that synchronizes and times reproductive activity (Dickman, 1985; McAllan et al., 2006; McAllan and Geiser, 2006). For devils, the rate of change of day length does not appear to be a factor in the timing of

238

Faculty of Science The University of Sydney reproductive activity and we have not yet seen evidence which suggests an independent endogenous circannual rhythm.

Latitude has been shown to have a pronounced effect on the timing of reproductive patterns in the opossum (Didelphis) (Rademaker and Cerqueira, 2006). When examining results from multiple studies over a 14o latitude change (30o to 44o N) in the

Didelphis virginiana, mean litter size decreased as the location neared the equator and the duration of the breeding season shifted towards an earlier onset (Rademaker and

Cerqueira, 2006). These results parallel the trend seen with devils with decreased litter size and earlier onset of the breeding season in captive devils compared to their more southerly wild counterparts over a similar range in latitude change. Didelphis albiventris and Didelphis aurita are also found across a large latitude range (5 – 34o S and 19 -27o

S respectively) but had less consistent trends through locations possibly due to a closer proximity to the equator overall (Rademaker and Cerqueira, 2006). Overall across didelphid species, it appears that the duration of the breeding season decreases with an increase in latitude, with some of those closest to the equator, displaying year-round breeding (Rademaker and Cerqueira, 2006). This is likely due to decreased variation in day length and temperature, the closer to the equator a species inhabits, which also may decrease the potential seasonal effects of food resources (Rademaker and Cerqueira,

2006). These authors suggest that higher latitudes increase environmental restraints, decreasing the breeding season or shifting it, to time weaning with periods of food abundance. As such the female must optimize their fecundity by increasing mean litter size to ensure as many offspring survive as possible, which is extremely important at the extremes of the ranges where adult survival across years is low (Rademaker and

Cerqueira, 2006). This could also be a possible alternative hypothesis for devils, to explain the disparity between average litter size between the captive and wild population. If survival rates for both adults and young are lower in the wild, this may

239

Faculty of Science The University of Sydney favour larger litter sizes to maximise the chance of genetic perpetuation. In seasonal environments, the timing and synchrony of reproduction can be considered crucially important factors for fitness and survival.

Whilst we try to simulate captive conditions to those of the wild as best as possible, including habitat and diet, there are obvious constraints associated with captivity, including limitations on housing design and availability. Species such as the dibbler

(Parantechinus apicalis) are monoestrous with females coming into oestrus over a short period in the Austral late summer, early autumn (Mills et al., 2012). Dibblers are naturally found on one of three islands off the western coast of Australia near Jurien

Bay (-30.297 S, 115.042 E) about 200 km north of Perth (-31.952 S, 115.859 E) or in the Fitzgerald River National Park (-33.948 S, 119.615 E) approximately 500 km southeast of Perth (Mills et al., 2012). An early study showed that 3 male and 3 female island dibblers translocated to Melbourne (-37.814 S, 144.963 E) prior to becoming sexually mature, housed under natural daylight conditions of the area for three years, all achieved sexual maturity (producing sperm or completing oestrous cycles confirmed by changes in body weight and vaginal cytology) but the timing of reproductive activity within and between sexes was overall asynchronous and all attempts to breed were unsuccessful (Woolley, 1991). It is possible that translocation prior to sexual maturity in this species was enough to alter their photoperiod history, further compounding the disruption in photoperiod cues associated with a different rate of photoperiod change and absolute day length to that of their natural habitat. In antechinus, if exogenous melatonin is administered before the antechinus reaches sexual maturity, desynchronization of the breeding season occurs (McAllan et al., 2002). Research describing early attempts to breed dibblers in captivity at Perth Zoo acknowledge the potential role photoperiod may have in initiating and timing seasonality of both male and female animals and suggest this may have been a factor in previous failed attempts

240

Faculty of Science The University of Sydney to breed this species in captivity, outside of its natural habitat and as such, decided to use photoperiod similar to their home range in their breeding program (Lambert and

Mills, 2006). This species in particular may be hypersensitive to changes in photoperiod. Photoperiodic responses of mainland and island female dibblers naturally differ, exhibiting slight differences in the timing of their oestrous cycles and gestation length (Mills et al., 2012). Mainland dibblers, which are on average heavier (~20g) than island dibblers, have an earlier onset of oestrus (by approximately 2 weeks) and longer gestation (38 vs 45 days respectively)(Mills et al., 2012). Overall, mating occurs between late February and late April, with spermatogenesis confirmed to encompass this period, February (occasionally late January) until May (Mills and Bencini, 2000;

Mills et al., 2012). The captive breeding program at Perth Zoo is part of the specie’s recovery plan, facilitating the reintroduction of Dibblers to their natural habitat and has been able to do so with an average of 85% of females paired producing pouch young over the history of the program (Cathy Lambert, Perth Zoo) (Lambert and Mills, 2006).

Currently, Perth Zoo is the only facility that contributes to the captive breeding program, so it is unknown if it is possible to replicate this success in other geographical locations.

Temperature and rainfall may also be proximate cues for the timing of seasonal reproduction but because of the lack of predictability and the potential for sharp differences over short distances (especially rainfall), these factors often lack strong predictive values (Hazlerigg and Simonneaux, 2015). In extreme years, where overall temperatures over extended periods are higher than average, combined with lower than average rainfall, this may cue forthcoming drought conditions which may have a negative impact (Hazlerigg and Simonneaux, 2015). However, the timing needed to have a significant effect in a marsupial may differ from eutherians as the primary

241

Faculty of Science The University of Sydney maternal investment is in lactation not pregnancy therefore the occurrence of oestrous cycles may not be inhibited but the successful production of offspring may be. The timing and occurrence of an oestrous cycle and the successful production and retention of pouch young in a dasyurid is likely influenced by not only photoperiod but a combination of other proximate cues related to perceived energy availability. Captivity ensures consistent food resources but environmental cues may still trigger innate perceptions that less favourable conditions are developing (e.g. drought) or alternatively higher temperatures may induce stress (through elevated pressure on thermoregulatory physiology), both potentially detrimental to breeding success.

The overall success rate of the captive devils from the records used in this study was only 41% which was similar to the overall success rate of the captive breeding program

(38%) over the period from which records were obtained (Australian Zoo and Aquarium

Association records). The first wave of females to come into oestrus in captivity had only a 27% success rate of producing viable young despite 12 of the 26 being 2 years of age (4 of 12 produced young), the prime age of breeding for a female Tasmanian devil.

Within our dataset, the majority of pouch young were produced during the first oestrous cycle (84%). There is anecdotal evidence that reproductive potential declines with age if successful reproduction is not achieved in the previous year (unpublished observations).

If critical day length is a driver for timing reproductive activity in female devils and thereby shifts the beginning of the breeding season by a few weeks in captivity due to an earlier obtainment of the appropriate day length, this also causes the breeding season to occur over periods of maximum temperature that are significantly higher than those of their natural environment. The island state of Tasmania has a cool temperate climate providing cooler temperatures than the mainland states of Australia due to is geographical location off the south-east coast of Australia. It is possible that higher average temperatures and heat waves, often experienced at least once per season in

242

Faculty of Science The University of Sydney many parts of mainland Australia may have a negative impact on the production of pouch young in captive devils. If heat is perceived to be associated with drought and the potential for decreased food resource or causes stress, some devils may defer reproduction until the next year. Periods of drought, common in Australia, have been shown to have a detrimental effect on breeding success of marsupials (Parrott et al.,

2007; Renfree and Shaw, 2001; Tyndale-Biscoe, 2001). For a species such as the antechinus, it doesn’t have the luxury of delaying reproduction until the next breeding season as males have only a single opportunity to breed due to its semelparous nature. If environmental conditions are significant enough to cause a decline in breeding success, this can have catastrophic effects on the population. Drought conditions have been shown to decrease both number of female agile antechinus successfully producing pouch young as well as litter size (Parrott et al., 2007). As the energetics of pregnancy in a marsupial are low due to short gestations and the birth of embryonic young, adverse environmental conditions may not supress the oestrous cycle or pregnancy but may increase the loss of pouch young at or soon after birth due to energetic demands of lactation (Tyndale-Biscoe, 2001). This may be other potential factor associated with reduced litter size in captive devils housed in mainland Australia.

Reproductive failure in Tasmanian devils is likely to occur at one of two critical time points during the breeding season. The first critical stage is at conception, failure most likely related to either mate incompatibility resulting in a lack of mating or male infertility. Despite low genetic diversity, breeding rates reported for wild devil populations are generally good suggesting that unlike other species like the cheetah, decreased male fertility associated with reduced genetic diversity is unlikely in the devil

(Crosier et al., 2007; Hughes, 1982; Jones et al., 2003, 2004;

243

Faculty of Science The University of Sydney

Lachish et al., 2009). A recent study of captive devils showed that 61% of unsuccessful breeding attempts were with male devils with proven fertility (production of pouch young) (Keeley et al., 2012) suggesting that male infertility is unlikely to be the primary factor in failed pairings. Mate incompatibility is difficult to determine if obvious signs of aggression are not observed. Although mating is not always observed in captivity due to the devil’s nocturnal nature and a lack of infrared video surveillance at most facilities, pairings are considered “good” if affiliative behaviours are observed (eg. denning together) or the male displays mate-guarding. Devils are separated when aggression is noted but a lack of aggression does not necessary confirm an optimal pairing. A recent examination of MHC class I balancing selection between paired individuals found genotypic, genic differentiation and genetic structure differed between successful and unsuccessful pairings regardless of a lack of aggression in the latter

(Russell et al., Submitted). It is unknown at which level this would decrease the success of pairings, at conception, during embryonic development or at birth (maternal selection) but may be one factor contributing to low breeding success in captive devils.

Reproductive failure could also occur at or shortly after birth. Conception rates are high in Tasmanian devils (75% of females from the examined wild population were pregnant or had small pouch young) and even if a large percent of embryos fail to complete maturation, the large number of eggs ovulated (on average 40-60 per ovulation) increases the chances that joeys are born (Hughes, 1982). In many of the captive devil pairings, mating was observed yet the female failed to produce young. Pouch checks are generally conducted at least 2-3 weeks after predicted birth, therefore for unsuccessful females, it is unknown if joeys are born but lost at or shortly after birth. If devils are presumed to have an autonomous ovary, then there is a physiological presumption of pregnancy regardless of mating success; this also means that “birthing” behaviour occurs independent of conception and we have yet to find a pregnancy marker for this

244

Faculty of Science The University of Sydney species (Keeley and Dehnhard, 2011; Keeley et al., 2012). If loss at or shortly after birth is the stage at which most breeding attempts fail, we have yet to be able to identify social or environmental factors that are causing this to occur and therefore have been unable to improve overall captive breeding success. It is possible that a shift in the timing of birthing events or geographical location to hotter weather may influence the retention of some or all of the joeys born. Our results confirm that captive devils are exposed to hotter temperatures than their native Tasmania and temperature has a negative correlation to the production of young and litter size.

Population studies suggest that Tasmanian devils in the wild are more likely to produce pouch young every year within their reproductive lifespan than captive devils. Due to the association of day length with seasonality, female devils in the wild are less likely to have three oestrous cycles if breeding is unsuccessful as this would occur too late in the season to facilitate weaning prior to the next breeding season (Hughes, 1982). Any reports of “out of season” breeding of wild Tasmanian devils are likely the result of the second oestrous cycle and in some cases the timing was likely over estimated due to a lack of understanding of female reproductive physiology and growth rates (Green,

1967). The ability for some captive devils to enter into a third oestrous cycle may be related to the slower rate of day length change compared to Tasmania combined with an earlier start of the first oestrous cycle. The photoperiodic cues that would end the breeding season in the wild would occur later at lower latitudes. Regardless, individual variability and physiology still limits the number of devils which undergo subsequent oestrous cycles if earlier breeding is unsuccessful.

Records indicate that some captive female devils (n = 14) did not undergo a second oestrous cycle despite being unsuccessful during the first oestrous cycle. For most of these females we could not confirm the presence or absence of further oestrous cycles due to the limitations of the record keeping systems and a lack of hormone data, so there

245

Faculty of Science The University of Sydney is a chance that some of these females did cycle again, only it was not detected or recorded by animal care staff. We can only confirm that three of the 2 year old devils only had a single oestrous cycle during their first breeding season as these animals were monitored for faecal progesterone concentrations throughout the breeding season

(Keeley et al., 2012). All three of these females were small in body weight (less than 4 kg) suggesting that body condition may be a factor associated with the potential number of oestrous cycles a female can have within one breeding season.

How much flexibility does the Tasmanian devil have in its life history? It appears to be not as much as we had hoped. Captive breeding efforts are continuously improving, yet captive breeding rates remain much lower than those observed in the wild. Although we cannot rule out cryptic female mate choice as a factor, it is unlikely to account for all of the unsuccessful pairings. Anecdotal reports suggest that captive breeding within facilities in Tasmania can also be variable, so although temperature may be a factor in mainland Australia, there is still likely multifactorial aspects of captivity which are impacting breeding success that we have yet to elucidate.

With wildlife species, it is not always possible to collect blood samples frequently enough to monitor hormones such as melatonin to gain information about seasonal reproductive physiology, especially for endangered species for which handling is restricted. With this study, we demonstrate that comparison of historic breeding records between captive and wild devils was sufficient to gain valuable information on the seasonal regulation of reproduction in this large dasyurid species. Species which exhibit seasonal patterns of reproduction may benefit from the study and evaluation of ecological factors in play at the onset of the breeding season or onset of sexual maturity to better understand the relationship between an animal and its environment.

Photoperiod is a predictive cue for seasonal reproduction in the Tasmanian devil and alters the timing of reproduction in accordance with geographical location.

246

Faculty of Science The University of Sydney

9.5.1 Acknowledgements

We would like to acknowledge the hard work and dedication to all those who have been involved in the Tasmanian devil insurance population care, breeding and management.

In particular Caroline Lees, Chris Hibbard and Carolyn Hogg from the Australian Zoo and Aquarium Association, and Carla Srb of Zoos Victoria, studbook keeper.

Instrumental animal care providers and fellow researchers, in addition to co-authors of this manuscript, are also acknowledged for their contributions to this research; Steve

Kleinig and Todd Jenkinson of Taronga Western Plains, Nick De Vos of Taronga Zoo,

Vikki Quinn of Currumbin Wildlife Sanctuary, Sara Washford of Monarto Zoo, Natasha

Rose of Ballarat Wildlife Park, and Dr. Marissa Parrot of Zoos Victoria. A big thanks to

Dr. Steve Johnston of the University of Queensland for comments on the manuscript.

247

Faculty of Science The University of Sydney

9.6 References

Bradley, A.J., 2003. Stress, hormones, and mortality in small carnivorous marsupials, in: Jones, M., Dickman, C., Archer, M. (Eds.), Predators with pouches: the biology of carnivorous marsupials. CSIRO Press, Sydney, pp. 250-263.

Crosier, A.E., Marker, L., Howard, J.G., Pukazhenthi, B., Henghali, J.N., Wildt, D.E.,

2007. Ejaculate traits in the Namibian cheetah (Acinonyx jubatus): influence of age, season and captivity. Reproduction, Fertility and Development 19, 370-382.

Dickman, C.R., 1985. Effects of photoperiod and endogenous control on timing of reproduction in the marsupial genus Antechinus. Journal of Zoology London 206, 509-

524.

Dickman, C.R., Hayssen, V., King, D.H., 1987. Effects of seasonal reversal of photoperiod on the reproductive rhythm of a small marsuipal Journal of Zoology

London 213, 766-768.

Green, R.H., 1967. Notes on the devil (Sarcophilus harrisii) and quoll (Dasyurus viverrinus) in North-eastern Tasmania. Records of the Queen Victoria Museum

Launceston 27, 1-13.

Guiler, E.R., 1970. Observations on the Tasmanian devil, Sarcophilus harrisii

(Marsupialia: Dasyuridae) II. Reproduction, breeding and growth of pouch young.

Australian Journal of Zoology 18, 63-70.

248

Faculty of Science The University of Sydney

Guiler, E.R., 1978. Observations on the Tasmanian devil, Sarcophilus harrisii

(Dasyuridae: Marsupialia) at Granville Harbour, 1966-1975. Papers and Proceedings of the Royal Society of Tasmania 112, 161-188.

Guiler, E.R., 1982. Temporal and spatial distrubution of the Tasmanian devil,

Sarcophilus harrisii (Dasyuridae: Marsupialia). Papers and Proceedings of the Royal

Society of Tasmania 116, 153-163.

Hawkins, C.E., Baars, C., Hesterman, H., Hocking, G.J., Jones, M.E., Lazenby, B.,

Mann, D., Mooney, N., Pemberton, D., Pyecroft, S., Restani, M., Wiersma, J., 2006.

Emerging disease and population decline of an island endemic, the Tasmanian devil

Sarcophilus harrisii. Biological Conservation 131, 307-324.

Hazlerigg, D., Simonneaux, V., 2015. Seasonal regulation of reproduction in mammals, in: Plant, T.M., Zeleznick, A.J. (Eds.), Knobil and Neill's Physiology of Reproduction,

4th ed. Elsevier Inc., London, pp. 1575-1603.

Hughes, R.L., 1965. Comparative morphology of spermatozoa from five marsupial families. Australian Journal of Zoology 13, 533-543.

Hughes, R.L., 1982. Reproduction in the Tasmanian devil Sarcophilus harrisii

(Dasyuridae, Marsupialia), in: Archer, M. (Ed.), Carnivorous Marsupials. Surrey Beatty and Sons, Sydney, pp. 49-63.

Hughes, R.L., Keeley, T., 2013. The reproductive biology and embryonic development of the Tasmanian devil, The Australian Mammal Society Conference, Sydney,

Australia.

249

Faculty of Science The University of Sydney

Hughes, R.L., Keeley, T., In Preparation. Population dynamics of Tasmanian devils,

Sarcophilus harrisii, in north-eastern Tasmania between 1974 -1987. Australian

Mammalogy.

IUCN, 2008. IUCN Red List of Threatened Species. IUCN Species Survival

Commission, Gland, Switzerland.

Jones, M.E., Paetkau, D., Geffen, E., Moritz, C., 2003. Microsatellites for the

Tasmanian devil (Sarcophilus laniarius). Molecular Ecology Notes 3, 277-279.

Jones, M.E., Paetkau, D., Geffen, E., Moritz, C., 2004. Genetic diversity and population structure of Tasmanian devils, the largest marsupial carnivore. Molecular Ecology 13,

2197- 2209.

Keeley, T., Dehnhard, M., 2011. Hormonal evidence of the autonomous ovary in the

Tasmanian devil (Sarcophilus harrisii), International Society for Wildlife

Endocrinology, Toronto, Canada.

250

Faculty of Science The University of Sydney

Keeley, T., McGreevy, P.M., Fanson, B.G., Masters, K., O’Brien, J.K., 2012. The reproductive cycle of the Tasmanian devil (Sarcophilus harrisii) and factors associated with reproductive success in captivity. General and Comparative Endocrinology 176,

182-191.

Keeley, T., McGreevy, P.M., O'Brien, J.K., 2011. Characterization and short-term storage of Tasmanian devil spermatozoa collected post-mortem. Theriogenology 76,

705-714.

Lachish, S., Jones, M., McCallum, H., 2007. The impact of the disease on the survival and population growth rate of the Tasmanian devil. Journal of Animal Ecology 76, 926-

936.

Lachish, S., McCallum, H., Jones, M., 2009. Demography, disease and the devil: life- history changes in a disease-affected population of Tasmanian devils (Sarcophilus harrisii). Journal of Animal Ecology 78, 427-436.

Lambert, C., Mills, H.R., 2006. Husbandry and breeding of the Dibbler Parantechinus apicalisat Perth Zoo. International Zoo Yearbook 40, 290-301.

Loh, R., Bergfeld, J., Hayes, D., O'Hara, A., Pyecroft, S., Raidal, S., Sharpe, R., 2006.

The pathology of Devil Facial Tumor Disease (DFTD) in Tasmanian devils

(Sarcophilus harrisii). Veterinary Pathology 43, 890-895.

Marneweck, D.G., Ganswindt, A., Rhodes, S., Bellem, A., Bryant, J., Wielebnowski,

N., Dalerum, F., 2013. Reproductive endocrinology of zoo-housed aardwolves. Acta

Theriol 58, 223-232.

251

Faculty of Science The University of Sydney

McAllan, B.M., Dickman, C.R., Crowther, M.S., 2006. Photoperiod as a reproductive cue in the marsupial genus Antechinus: ecological and evolutionary consequences.

Biological Journal of the Linnean Society 87, 365-379.

McAllan, B.M., Geiser, F., 2006. Photoperiod and the timing of reproduction in

Antechinus flavipes (Dayuridae: Marsupialia). Mammalian Biology 71, 129-138.

McAllan, B.M., Joss, J.M.P., Firth, B.T., 1991. Phase delay of the natural photoperiod alters reproductive timing in the marsupial Antechinus stuartii. Journal of Zoology

London 225, 633- 646.

McAllan, B.M., Westman, W., Joss, J.M.P., 2002. The seasonal reproductive cycle of a marsupial, Antechinus stuartii: effects of oral administration of melatonin. General and

Comparative Endocrinology 128, 82-90.

Mills, H.R., Bencini, R., 2000. New evidence for facultative male die-off in island populations of dibblers, Parantechinus apicalis. Australian Journal of Zoology 48,

501-510.

Mills, H.R., Bradshaw, F.J., Lambert, C., Bradshaw, S.D., Bencini, R., 2012.

Reproduction in the marsupial dibbler, Parantechinus apicalis; differences between island and mainland populations. General and Comparative Endocrinology 178.

Parrott, M., Ward, S.J., Temple-Smith, P., Selwood, L., 2007. Effects of drought on weight, survival and breeding success of agile antechinus (Antechinus agilis), dusky

252

Faculty of Science The University of Sydney antechinus (A. swainsonii) and bush rats (Rattus fuscipes). Wildlife Research 34, 437-

442.

Pearse, A.M., Swift, K., 2006. Transmission of devil facial-tumour disease. Nature 439,

549.

Pemberton, D., 1990. Social organization and behaviour of the Tasmanian devil,

Sarcophilus harrisii, Zoology University of Tasmania, Hobart.

Pemberton, D., Renouf, D., 1993. A field study of communication and social behaviour of the Tasmanian devil at feeding sites. Australian Journal of Zoology 41, 507-526.

Pyecroft, S., Pearse, A., Loh, R., Swift, K., Belov, K., Fox, N., Noonan, E., Hayes, D.,

Hyatt, A., Wang, L., Boyle, D., Church, J., Middleton, D., Moore, R., 2007. Towards a case definition for devil facial tumour disease: What is it? EcoHealth 4, 346-351.

Rademaker, V., Cerqueira, R., 2006. Variation in the latitudinal reproductive patterns of the genus Didelphis (Didelphimorphia:Didelphidae). Austral Ecology 31, 337-342.

Renfree, M.B., Shaw, G., 2001. Reproduction in Monotremes and Marsupials.

Encyclopedia of Life Sciences, 1-5.

Roberts, M.G., 1915. The keeping and breeding of Tasmanian devils (Sarcophilus harrisii). Proceedings of the Zoological Society of London 85, 575-581.

253

Faculty of Science The University of Sydney

Selwood, L., McCallum, F., 1987. Relationship between longevity of spermatozoa after insemination and the percentage of normal embryos in brown marsupial mice

(Antechinus stuartii). Journal of Reproduction and Fertility 79, 495-503.

Siddle, H.V., Kreiss, A., Eldridge, M.D.B., Noonan, E., Clarke, C.J., Pyecroft, S.,

Woods, G.M., Belov, K., 2007. Transmission of a fatal clonal tumor by biting occurs due to depleted MHC diversity in a threatened carnivorous marsupial. Proceedings of the National Academy of Science 104, 16221-16226.

Taggart, D.A., 1994. A comparison of sperm and embryo transport in the female reproductive tract of marsupial and eutherian mammals. Reproduction, Fertility and

Development 6, 451-472.

Thalmann, S., Peck, S., Wise, P., Potts, J. M., Clarke, J. and Richley, J. 2016

Translocation of a top-order carnivore: tracking the initial survival, spatial movement, home-range establishment and habitat use of the Tasmanian devils on Maria Island.

Australian Mammalogy 38, 68-79.

Tyndale-Biscoe, C.H., 2001. Australian marsupials - to cherish and to hold.

Reproduction, Fertility and Development 13, 477-485.

Woods, G.M., Kreiss, A., Belov, K., Siddle, H.V., Obendorf, D., Muller, H.K., 2007.

The immune response of the Tasmanian devil (Sarcophilus harrisii) and the Devil

Facial Tumour Disease. EcoHealth 4, 338-345.

254

Faculty of Science The University of Sydney

Woolley, P., 1991. Reproductive pattern of captive Boullanger Island dibblers,

Parantechinus apicalis (Marsupialia: Dasyuridae). Wildlife Research 18, 157-163.

Zerbe, P., Clauss, M., Codron, D., Lackey, L.B., Rensch, E., Streich, J.W., Hatt, J.,

Muller, D.W.H., 2012. Reproductive seasonality in captive wild ruminants: implications for biogeographical adaptation, photoperiodic control, and life history. Biological

Reviews of the Cambridge Philosophical Society 87, 965-990.

255

Faculty of Science The University of Sydney

10 Appendix Four: Genetic diversity, inbreeding and cancer

Published Proceedings of the Royal Society B-Biological Sciences 2018, 285: 1875, 20172589

Beata Ujvari1,2, Marcel Klaassen1, Nynke Raven1, Tracey Russell3, Marion Vittecoq4,

Rodrigo Hamede1,2, Frédéric Thomas5, Thomas Madsen1,6

1Centre for Integrative Ecology, School of Life and Environmental Sciences, Deakin

University, Waurn Ponds, Victoria, 3216, Australia

2School of Biological Sciences, University of Tasmania, Private Bag 55, Hobart,

Tasmania, 7001, Australia

3School of Life and Environmental Sciences, University of Sydney, Sydney, NSW

2006, Australia

4Institut de Recherche de la Tour du Valat, le Sambuc, 13200 Arles, France

5CREEC/ MIVEGEC, UMR IRD/CNRS/UM 5290, 911 Avenue Agropolis, BP 64501,

34394 Montpellier Cedex 5, France

6School of Biological Sciences, University of Wollongong, 2522, NSW, Australia

256

Faculty of Science The University of Sydney

10.1 Abstract

Genetic diversity is essential for adaptive capacities, providing organisms with the potential of successfully responding to intrinsic and extrinsic challenges. Although a clear reciprocal link between genetic diversity and vulnerability to parasites and pathogens has been established across taxa, the impact of loss of genetic diversity and inbreeding on the emergence and progression of non-communicable diseases, such as cancer, have been overlooked. Here we provide an overview of such associations and show that low genetic diversity and inbreeding associate with an increased risk of cancer in both humans and animals. Cancer being a multifaceted disease, loss of genetic diversity can directly (via accumulation of oncogenic homozygous mutations) and indirectly (via increased susceptibility to oncogenic pathogens) impact cancer emergence and prevalence. The observed link between reduced genetic diversity and cancer in wildlife may further imperil the long-term survival of numerous endangered species, highlighting the need to consider the impact of cancer in conservation biology.

Finally, the somewhat incongruent data originating from human studies suggest that the effect of reduced genetic diversity and inbreeding on cancer development is multifactorial and may be tumour specific. Further studies are therefore crucial in order to elucidate the underpinnings of the interactions between genetic diversity, inbreeding and cancer.

Key words: genetic diversity, inbreeding, oncogenic mutations, oncogenic pathogens, human and wildlife cancer

257

Faculty of Science The University of Sydney

10.2 Introduction

Genetic diversity provides populations with the ability to respond to challenges, such as parasites/pathogens, predators and environmental perturbations. Attenuation of genetic diversity has been linked to increased risk of inbreeding depression resulting in decreased growth rate, fertility, fecundity and offspring viability [1-9] as well as in increased vulnerability to pathogens [10-12]. Loss of genetic diversity therefore has a negative impact organismal fitness, and limit a population’s ability to respond to threats in both the long and short term (for review see [13]). Akin to parasites, malignant transformations, that emerge either due to environmental challenges, infections and/or host genotype either in isolation or via the interaction between genotype and environment, exploit the host for energy and resources, and thereby impair host fitness and pose as a significant selective force [14-16]. Indeed, recent studies have proposed that malignant cells should be regarded as a developing species that behave in a manner akin to parasites [17]. Consequently, multicellular hosts that have the genetic toolkit to recognise and control cancer causing infections and malignant cell proliferation, will have a significant fitness advantage over those that lack such mechanisms. Although a clear reciprocal link between genetic diversity and vulnerability to parasites and pathogens has been widely acknowledged across taxa, so far the vast majority of studies have overlooked how reduced genetic diversity and inbreeding may influence the appearance and progression of non-communicable diseases, such as cancer. Here we discuss how genetic diversity and inbreeding may contribute to increased risk of cancer development and progression in humans and animals.

258

Faculty of Science The University of Sydney

10.3 Cancer aetiologies

Cancer, the uncontrolled division of malignant cells, is a ubiquitous disease of metazoans [18] and has been proposed to have appeared with one of the major transitions of life i.e. the transition from unicellularity to multicellularity [19].

Fossilized bones, mummified tissue data and phylogenetic analyses of oncogenic pathogens show that malignant transformations have been afflicting human and animal populations for eons (reviewed in [20]).

Cancer is a multifactorial disease, although a small proportion of human cancers (5-

10%) originates from inherited mutations, the majority of cases (90-95%) can be attributed to acquired mutations due to environment and/or lifestyle [20]. The majority of familial human cancers have been proposed to root from high-penetrance genetic variants or polymorphisms [21]. However, cancer predisposition by rare, high- penetrance alleles (e.g. TP53, RB1) have also been observed in animal malignancies

[22]. Human lifestyle, however, is the major cause of cancer development and almost

25–30% of all cancer-related human deaths are due to tobacco and/or 30–35% are linked to diet (reviewed in [23]).

Several of the factors resulting in increased cancer prevalence in humans such as smoking, alcohol and diet are highly unlikely to cause cancer in animals (but see [24,

25]), whereas stress [26-28], infections (reviewed in [29]), and exposure to environmental carcinogens have been found to increase cancer prevalence in other vertebrates, such as the brown bullhead (Ameiurus nebulosus) [30], California sea lion

(Zalophus californianus) [31] and beluga whales (Delphinapterus leucas) [32].

Infections are the direct or indirect underlying factors of a substantial proportion of both human and animal cancers [33]. Pathogens altering cellular regulatory mechanisms

259

Faculty of Science The University of Sydney

(e.g. apoptosis, cell-cycle arrest) and cell proliferation rates break down cellular barriers and directly contribute to neoplasm formation. Inflammatory responses initiated by infections may also increase mutation rates and alter proliferation signals, and hence indirectly initiate malignant transformations (reviewed in [33, 34]). Apart from viruses, the most frequent sources of infection-induced cancers are protozoans (e.g. Plasmodium falciparum) [35], bacteria (e.g. Helicobacter pylori) [36, 37] and trematodes (e.g.

Schistosoma haemotobium) [36, 38] have all been shown to directly or indirectly cause malignancies. Although rare, contagious cancers without underlying infectious aetiologies do occur in the wild and eight naturally occurring transmissible cancers, one lineage in dogs [39], two lineages in Tasmanian devils (Sarcophilus harrisii) [40, 41] and five lineages in bivalves [42], have so far been recorded.

10.4 Genetic diversity, inbreeding and cancer in humans

Several reports provide evidence that low genetic diversity and inbreeding may increase cancer risk and that cancer may have a recessive basis in humans [43-45]. For example, thyroid cancer has been found to be associated with significantly higher levels of inbreeding as well as a higher number and longer runs of homozygosity (ROH) [46] and acute leukaemia have been found to be linked to low levels of genetic diversity and inbreeding [47]. Moreover, extended germline homozygosity has been shown to result in an increased risk of lung cancer [48] and homozygosity of the MTHFR gene has been found to be associated with an increased risk of breast cancer [49].

Genome-wide association studies have also found a significant association between recessive alleles/inbreeding and cancer such as Hodgkin lymphoma [50]. Based on the same methodology, two studies observed that inbreeding and ROH resulted in an increased risk of colorectal cancer [51, 52], whereas a third study could not find such an

260

Faculty of Science The University of Sydney association [53]. Similar disconsonant results have been reported from studies focusing on countries with high close-kin unions such as the United Arab Emirates and Qatar, with up to 54% consanguinity prevalence [52, 53]. The two studies showed that reduced genetic diversity and inbreeding was associated with a reduced risk of breast, skin, thyroid and female genital cancers but an increased risk of developing leukaemia, lymphoma, colorectal and prostate cancer [54, 55]. The incongruous results observed in some human studies suggest that the effect of genetic diversity and inbreeding on cancer development may be tumour specific.

10.5 Genetic diversity, inbreeding and cancer in domestic and wild animals

Strong artificial selection and small founder population size during domestication of animals have had the unintentional effect of diminishing genetic diversity, and resulted in the accumulation of deleterious genetic variants. For example, despite their exceptional phenotypic diversity both domestic dogs and cats have significantly lower genetic diversity compared to their wild conspecifics, and/or their wild ancestors [56-

62]. Apart from additional factors, such as anthropogenically induced longer lifespan and altered environment (e.g. diet and exposure to tobacco smoke), the loss of genetic diversity has been linked to the observed relatively high cancer prevalence in both cats and dogs [63-66] Data originating from the histopathology analyses of >30,000 malignant neoplastic cases of cats and dogs revelead skin being the most frequently affected tissue in both species, and purebred dogs being more prone to develop neoplasms in general [65]. The latter finding has been further supported by a survey from Italy that showed an almost 2-fold higher incidence rate of malignant tumours in both purebred cats and dogs compared to mixed breeds [66].

261

Faculty of Science The University of Sydney

Despite neoplasia being recorded in most metazoans [67], and being common in domesticated animals, it has generally been assumed to be rare in the wild. In our view this is most likely due to the fact that cancer prevalence in wildlife is extremely difficult to identify and reports are highly scattered in the scientific literature, and hence challenging to access [18]. In some fish populations cancer prevalence can actually reach 100% and being caused by contagious agents, pollution, inbreeding or the combination of all these factors [18, 68]. Moreover, the high cancer prevalence (26%) recorded in some populations of California sea lions (Zalophus californianus) has been suggested to be caused by a herpesvirus and/or persistent organic pollutants, but a high prevalence of urogenital carcinoma has been linked to loss of genetic diversity at a single locus, the heparanase 2 gene (HPSE2) [69]. Additionally, two recent studies have observed a link between low genetic diversity and high cancer prevalence (>50%) in Santa Catalina Island foxes (Urocyon littoralis catalinae) [70, 71] and the South

African Cape mountain zebra (Equus zebra zebra)[72-74].

10.6 Genetic diversity, inbreeding and cancer development

Cancer being a multifaceted disease, loss of genetic diversity and inbreeding can impact cancer emergence both directly and indirectly. Reduction of population size, cultural traditions promoting consanguineous marriages, and natural selection purging favouring certain haplotypes contribute to an increased likelihood of a reduction in genetic diversity which may result in higher frequency of long stretches of ROH regions [75,

76]. ROH harbour disproportionately more deleterious homozygotes than other parts of the genome [75], and the presence of identical pathogenic variants of both alleles have been shown to result in recessive disorders [44, 77]. Reduced genetic diversity magnifies the impact of deleterious homozygous mutations [75] and genomic studies

262

Faculty of Science The University of Sydney suggest that homozygosity of some germline low-penetrance cancer genes act as significant contributing factors to the development of human oesophageal [78], oral

[79], lung [80, 81], bladder [82], acute lymphocytic leukemia [83] and breast cancers

[54, 84, 85].

Apart from the direct role of cancer increasing homozygous genomic regions, a general reduction in genetic diversity can also contribute to the development of tumours via infectious agents such as viruses (e.g. [29, 33, 86]). Loss of genetic diversity at important immune gene loci such as the Major Histocompatibility Complex (MHC),

Toll-like receptors (TLRs) and type I interferons (IFN‑a and IFN‑b) [12, 87], will increase the risk of pathogen infections that either directly or indirectly initiate malignant transformations. For example, hepatitis C virus (HCV), one of the most common chronic blood-borne infections, results in chronic hepatitis in approximately

80% of infected patients, and leads to death in up to 5% of these patients from hepatocellular carcinoma (HCC) or liver cancer [88]. A complex interplay between host genetics, immunology and viral factors has been proposed to determine the outcome of HCV infection [88-91]. Ethnic background, immune gene polymorphism as well as the presence of specific alleles (e.g. interleukin 28B, inhibitory natural killer cell receptors and MHC classes I and II, and variants of interferon (IFN)L3-IFNL4, etc.) have been identified as key elements of HCV clearance, and consequent disease progression [88-91].

Helicobacter pylori infections, an underlying factor of gastric cancer, provide an excellent example of how the host genotype may indirectly contribute to initiating of malignant transformations. H. pylori affects at least 50% of humans worldwide, and hence owns the uncoveted title of being “the most common single chronic bacterial infection in the world” [92]. A twin study showed that both host genetic and environmental factors (“rearing environment”) influence the acquisition of H. pylori

263

Faculty of Science The University of Sydney infection [93]. Importantly, host proinflammatory genetic makeup appears to have a major contribution to the pathogenesis of gastric cancer. Individuals with proinflammatory genotypes (IL-1B-511*T carriers/IL-1RN*2 homozygotes) have an increased risk for gastric carcinoma. The carriers of the specific genotypes generate heightened inflammatory response to H. pylori infection, that ultimately creates a chronically inflamed environment with elevated oxidative/genotoxic stress (due to hypochlorhydria) and eventually initiate a proneoplastic drive [92, 94].

Reduced genetic diversity may also increase susceptibility of endangered wildlife species to pathogens and its associated cancers such as papillomatosis and carcinomatosis syndrome in western barred bandicoots (Perameles bougainville) [95] and viral papilloma and squamous cell carcinomas in snow leopards (Uncia uncia) [96].

The metastatic urogenital carcinoma of California sea lions and the high cancer prevalence observed in inbred Santa Catalina Island foxes and South African Cape mountain zebra strongly suggest an association between loss of genetic diversity and cancer development in wildlife.

10.7 Conclusion

As mentioned above, maintenance of genetic diversity is fundamental for adaptive capacities and provides organisms with an ability to successfully respond to challenges caused by parasites/pathogens [97], habitat fragmentation [3, 98] and global climate change [99, 100]. In contrast to parasites and pathogens cancer has so far been largely overlooked as a significant determinant of wildlife fitness. The present review, however, suggests that low genetic diversity and inbreeding may elevate cancer development in wildlife, further imperilling the long-term survival of the numerous

264

Faculty of Science The University of Sydney species presently suffering from low genetic diversity. Our review hence demonstrates the need to consider the effects of cancer in conservation biology.

The results originating from human studies indicate that the effects of genetic diversity and inbreeding on the development of a complex disease, such as cancer, may be tumour specific. Importantly, by reducing immune function, and thereby increasing the vulnerability to cancer causing parasite/pathogen infections, overall loss of genetic diversity and inbreeding may therefore constitute a significant underpinning of cancer development in humans as well as in other organisms [101, 102]. Finally, the link between low genetic diversity/inbreeding and cancer may be just as arduous as the disease itself, and further studies are therefore urgently needed to decipher the underpinnings of such associations.

10.7.1 Acknowledgements

This work was supported by two Eric Guiler Tasmanian devil Research Grants through the Save the Tasmanian devil appeal of the University of Tasmania Foundation, by the

ANR (Blanc project EVOCAN) by the CNRS (INEE), by an International Associated

Laboratory Project France/Australia, by an Australian Academy of Science, FASIC

Early Career Fellowship to BU, by Deakin University’s CRGS Grant. We are also grateful to Mr André Hoffmann, MAVA foundation, Tour du Valat, for his continuous support of our research.

10.8 References

1 Keller, L.F. & Waller, D.M. 2002 Inbreeding effects in wild populations. Trends Ecol.

Evol. 17, 230-241. (doi:10.1016/S0169-5347(02)02489-8).

2 Crnokrak, P. & Roff, D.A. 1999 Inbreeding depression in the wild. Heredity 83, 260-

270.

265

Faculty of Science The University of Sydney

3 Madsen, T., Shine, R., Olsson, M. & Wittzell, H. 1999 Conservation biology:

Restoration of an inbred adder population. Nature 402, 34-35. (doi:10.1038/46941).

4 Madsen, T., Stille, B. & Shine, R. 1996 Inbreeding depression in an isolated population of adders Vipera berus. Biol. Conserv. 75, 113-118. (doi:10.1016/0006-

3207(95)00067-4).

5 Madsen, T., Ujvari, B. & Olsson, M. 2004 Novel genes continue to enhance population growth in adders (Vipera berus). Biol. Conserv. 120, 145-147.

(doi:10.1016/j.biocon.2004.01.022).

6 Saccheri, I., Kuussaari, M., Kankare, M., Vikman, P., Fortelius, W. & Hanski, I. 1998

Inbreeding and extinction in a butterfly metapopulation. Nature 392, 491-494.

(doi:10.1038/33136).

7 Slate, J., Kruuk, L.E., Marshall, T.C., Pemberton, J.M. & Clutton-Brock, T.H. 2000

Inbreeding depression influences lifetime breeding success in a wild population of red deer (Cervus elaphus). Proc. R. Soc. B 267, 1657-1662. (doi:10.1098/rspb.2000.1192).

8 Huisman, J., Kruuk, L.E.B., Ellis, P.A., Clutton-Brock, T. & Pemberton, J.M. 2016

Inbreeding depression across the lifespan in a wild mammal population. Proc. Natl.

Acad. Sci. USA 113, 3585-3590. (doi:10.1073/pnas.1518046113).

9 Westemeier, R.L. et al. 1998 Tracking the long-term decline and recovery of an isolated population. Science 282, 1695-1698. (doi: 10.1126/science.282.5394.1695).

10 O'Brien, S.J. & Evermann, J.F. 1988 Interactive influence of infectious disease and genetic diversity in natural populations. Trends Ecol. Evol. 3, 254-259.

(doi:10.1016/0169-5347(88)90058-4).

11 Townsend, A.K., Clark, A.B., McGowan, K.J., Miller, A.D. & Buckles, E.L.

2010 Condition, innate immunity and disease mortality of inbred crows. Proc. R. Soc. B

277, 2875-2883. (doi:10.1098/rspb.2010.0480).

266

Faculty of Science The University of Sydney

12 Ujvari, B. & Belov, K. 2011 Major Histocompatibility Complex (MHC) markers in conservation biology. Int. J. Mol. Sci. 12, 5168-5186.

(doi:10.3390/ijms12085168).

13 Altizer, S., Harvell, D. & Friedle, E. 2003 Rapid evolutionary dynamics and disease threats to biodiversity. Trends Ecol. Evol. 18, 589-596.

(doi:10.1016/j.tree.2003.08.013).

14 Ujvari, B. et al. 2016 Cancer and life-history traits: lessons from host-parasite interactions. Parasitology 143, 533-541. (doi:10.1017/s0031182016000147).

15 Vittecoq, M. et al. 2013 Cancer: a missing link in ecosystem functioning?

Trends Ecol. Evol. 28, 628-635. (doi:10.1016/j.tree.2013.07.005).

16 Thomas, F. et al. 2017 The importance of cancer cell communities for animal evolutionary ecology. Nat. Ecol. Evol. (doi:10.1038/s41559-017-0343-z).

17 Duesberg, P., Mandrioli, D., McCormack, A. & Nicholson, J.M. 2011 Is carcinogenesis a form of speciation? Cell Cycle 10, 2100-2114.

(doi:10.4161/cc.10.13.16352).

18 Madsen, T. et al. 2017 Cancer prevalence and etiology in wild and captive animals. In Ecology and Evolution of Cancer (eds. B. Ujvari, B. Roche & F. Thomas) pp. 11-46, Academic Press.

19 Szathmary, E. & Smith, J.M. 1995 The major evolutionary transitions. Nature

374, 227-232. (doi:10.1038/374227a0).

20 Ewald, P.W. 2017 Ancient cancers and infection-induced oncogenesis. Int. J.

Paleopath. (doi:10.1016/j.ijpp.2017.08.007).

21 Balmain, A., Gray, J. & Ponder, B. 2003 The genetics and genomics of cancer.

Nat. Genet. 33 Suppl, 238-244. (doi:10.1038/ng1107).

267

Faculty of Science The University of Sydney

22 Gregory-Bryson, E., Bartlett, E., Kiupel, M., Hayes, S. & Yuzbasiyan-Gurkan,

V. 2010 Canine and human gastrointestinal stromal tumors display similar mutations in c-KIT exon 11. BMC Cancer 10, 559. (doi:10.1186/1471-2407-10-559).

23 Anand, P. et al. 2008 Cancer is a preventable disease that requires major lifestyle changes. Pharmaceut. Res. 25, 2097-2116. (doi:10.1007/s11095-008-9661-9).

24 Reif, J.S. & Cohen, D. 1971 The environmental distribution of canine respiratory tract neoplasms. Arch. Environ. Health 22, 136-140.

(doi:10.1080/00039896.1971.10665823).

25 Reif, J.S., Dunn, K., Ogilvie, G.K. & Harris, C.K. 1992 Passive smoking and canine lung cancer risk. Am. J. Epidemiol. 135, 234-239.

26 Thaker, P.H. et al. 2006 Chronic stress promotes tumor growth and angiogenesis in a mouse model of ovarian carcinoma. Nature Med. 12, 939-944.

(doi:10.1038/nm1447).

27 Le, C.P. et al. 2016 Chronic stress in mice remodels lymph vasculature to promote tumour cell dissemination. Nature Comm. 7, 10634.

(doi:10.1038/ncomms10634).

28 Cannas, S. et al. 2016 Stress and cancer in dogs: Comparison between a population of dogs diagnosed with vancer and a control population - A pilot study. Mac.

Vet. Rev. 39, 201-208. (doi.org/10.1515/macvetrev-2016-0088).

29 de Martel, C. & Franceschi, S. 2009 Infections and cancer: Established associations and new hypotheses. Crit. Rev. Oncol/Hematol. 70, 183-194.

(doi:10.1016/j.critrevonc.2008.07.021).

30 Baumann, P.C., LeBlanc, D.R., Blazer, V., Meier, J.R., Hurley, S.T. & Kiryu,

Y. 2008 Prevalence of tumors in brown bullhead from threelakes in Southeastern

Massachusetts, 2002. U.S. Geological Survey Scientific Investigations Report 2008-

5198.

268

Faculty of Science The University of Sydney

31 Gulland, F.M.D., Trupkiewicz, J.G., Spraker, T.R. & Lowenstine, L.J. 1996

Metastatic carcinoma of probable transitional cell origin in 66 free-living Califronia sea lions (Zalophus californianus), 1979 to 1994. J. Wildl. Diseases 32, 250-258.

(doi:10.7589/0090-3558-32.2.250).

32 Martineau, D., Lemberger, K., Dallaire, A., Labelle, P., Lipscomb, T.P.,

Michel, P. & Mikaelian, I. 2002 Cancer in wildlife, a case study: beluga from the St.

Lawrence estuary, Québec, Canada. Environ. Health Perspectives 110, 285-292.

33 Ewald, P.W. & Swain Ewald, H.A. 2015 Infection and cancer in multicellular organisms. Phil. Trans. R. Soc. B. 370. (doi:10.1098/rstb.2014.0224).

34 Ewald, P.W. & Swain Ewald, H.A. 2012 Infection, mutation, and cancer evolution. J. Mol. Med. 90, 535-541. (doi:10.1007/s00109-012-0891-2).

35 Molyneux, E.M. et al. 2012 Burkitt's lymphoma. Lancet 379, 1234-1244.

(doi:10.1016/s0140-6736(11)61177-x).

36 Ewald, P.W. & Swain Ewald, H.A. 2014 Joint infectious causation of human cancers. Adv. Parasitol. 84, 1-26. (doi:10.1016/b978-0-12-800099-1.00001-6).

37 Mager, D.L. 2006 Bacteria and cancer: cause, coincidence or cure? A review.

J. Transl. Med. 4, 14. (doi:10.1186/1479-5876-4-14).

38 Mostafa, M.H., Sheweita, S.A. & O'Connor, P.J. 1999 Relationship between schistosomiasis and bladder cancer. Clin. Microbiol. Rev. 12, 97-111.

39 Murgia, C., Pritchard, J.K., Kim, S.Y., Fassati, A. & Weiss, R.A. Clonal origin and evolution of a transmissible cancer. Cell 126, 477-487.

(doi:10.1016/j.cell.2006.05.051).

40 Pearse, A.M. & Swift, K. 2006 Allograft theory: Transmission of devil facial- tumour disease. Nature 439, 549. (doi:10.1038/439549a).

41 Pye, R.J., Pemberton, D., Tovar, C., Tubio, J.M.C., Dun, K.A., Fox, S., Darby,

J., Hayes, D., Knowles, G.W., Kreiss, A., et al. 2016 A second transmissible cancer in

269

Faculty of Science The University of Sydney

Tasmanian devils. Proc. Natl. Acad. Sci. USA 113, 374-379.

(doi:10.1073/pnas.1519691113).

42 Metzger, M.J. et al. 2016 Widespread transmission of independent cancer lineages within multiple bivalve species. Nature 534, 705-709.

(doi:10.1038/nature18599).

43 Rudan, I. et al. 2003 Inbreeding and risk of late onset complex disease. J. Med.

Genetics 40, 925-932.

44 Assié, G., LaFramboise, T., Platzer, P. & Eng, C. 2008 Frequency of germline genomic homozygosity associated with cancer cases. JAMA 299, 1437-1445.

(doi:10.1001/jama.299.12.1437).

45 Lebel, R.R. & Gallagher, W.B. 1989 Wisconsin consanguinity studies. II:

Familial adenocarcinomatosis. Am. J. Med. Genetics 33, 1-6.

(doi:10.1002/ajmg.1320330102).

46 Thomsen, H. et al. 2016 Runs of homozygosity and inbreeding in thyroid cancer. BMC Cancer 16, 227. (doi:10.1186/s12885-016-2264-7).

47 Feldman, J.G., Lee, S.L. & Seligman, B. 1976 Occurrence of acute leukemia in females in a genetically isolated population. Cancer 38, 2548-2550.

48 Orloff, M.S., Zhang, L., Bebek, G. & Eng, C. 2012 Integrative genomic analysis reveals extended germline homozygosity with lung cancer risk in the PLCO cohort. PloS One 7, e31975. (doi:10.1371/journal.pone.0031975).

49 Deligezer, U., Akisik, E.E. & Dalay, N. 2005 Homozygosity at the C677T of the MTHFR gene is associated with increased breast cancer risk in the Turkish population. In Vivo 19, 889-893.

50 Thomsen, H. et al. 2016 Evidence of Inbreeding in Hodgkin Lymphoma. PloS

One 11, e0154259. (doi:10.1371/journal.pone.0154259).

270

Faculty of Science The University of Sydney

51 Bacolod, M.D., Schemmann, G.S., Giardina, S.F., Paty, P., Notterman, D.A. &

Barany, F. 2009 Emerging aradigms in cancer genetics: Some important findings from high-density single nucleotide polymorphism array studies. Cancer Res. 69, 723-727.

(doi:10.1158/0008-5472.CAN-08-3543).

52 Bacolod, M.D. et al. 2008 The signatures of autozygosity among patients with colorectal cancer. Cancer Res. 68, 2610-2621. (doi:10.1158/0008-5472.CAN-07-5250).

53 Spain, S.L., Cazier, J.B., Houlston, R., Carvajal-Carmona, L. & Tomlinson, I.

2009 Colorectal cancer risk is not associated with increased levels of homozygosity in a population from the United Kingdom. Cancer Res. 69, 7422-7429. (doi:10.1158/0008-

5472.can-09-0659).

54 Denic, S., Frampton, C. & Nicholls, M.G. 2007 Risk of cancer in an inbred population. Cancer Detect. Prev. 31, 263-269. (doi:10.1016/j.cdp.2007.07.006).

55 Bener, A. et al. 2009 Impact of consanguinity on cancer in a highly endogamous population. Asian Pacific J. Cancer Prev. APJCP 10, 35-40.

56 Boyko, A.R. 2011 The domestic dog: man's best friend in the genomic era.

Genome Biol. 12, 216. (doi:10.1186/gb-2011-12-2-216).

57 Freedman, A.H. et al. 2014 Genome sequencing highlights the dynamic early history of dogs. PLoS Genetics 10, e1004016. (doi:10.1371/journal.pgen.1004016).

58 Hu, Y. et al. 2014 Earliest evidence for commensal processes of cat domestication. Proc. Natl. Acad. Sci. USA 111, 116-120.

(doi:10.1073/pnas.1311439110).

59 Lindblad-Toh, K. et al. 2005 Genome sequence, comparative analysis and haplotype structure of the domestic dog. Nature 438, 803-819.

(doi:10.1038/nature04338).

271

Faculty of Science The University of Sydney

60 Marsden, C.D. et al. 2016 Bottlenecks and selective sweeps during domestication have increased deleterious genetic variation in dogs. Proc. Natl. Acad.

Sci USA 113, 152-157. (doi:10.1073/pnas.1512501113).

61 Montague, M.J. et al. 2014 Comparative analysis of the domestic cat genome reveals genetic signatures underlying feline biology and domestication. Proc. Natl.

Acad. Sci. USA 111, 17230-17235. (doi:10.1073/pnas.1410083111).

62 Vonholdt, B.M. et al. 2010 Genome-wide SNP and haplotype analyses reveal a rich history underlying dog domestication. Nature 464, 898-902.

(doi:10.1038/nature08837).

63 Dorn, C.R. & Schneider, R. 1976 Inbreeding and canine mammary cancer: a retrospective study. J. Natl. Cancer Inst. 57, 545-548.

64 Dorn, C.R., Taylor, D.O., Frye, F.L. & Hibbard, H.H. 1968 Survey of animal neoplasm in Alameda and Contra Costa counties, California. I. Methodology and desciption of cases. J. Natl. Cancer Inst. 40, 295-305.

65 Dorn, C.R., Taylor, D.O., Schneider, R., Hibbard, H.H. & Klauber, M.R. 1968

Survey of animal neoplasms in Alameda and Contra Costa Counties, California. II.

Cancer morbidity in dogs and cats from Alameda County. J. Natl Cancer Inst. 40, 307-

318.

66 Vascellari, M., Baioni, E., Ru, G., Carminato, A. & Mutinelli, F. 2009 Animal tumour registry of two provinces in northern Italy: incidence of spontaneous tumours in dogs and cats. BMC Vet. Res. 5, 39. (doi:10.1186/1746-6148-5-39).

67 Leroi, A.M., Koufopanou, V. & Burt, A. 2003 Cancer selection. Nature Rev.

Cancer 3, 226-231. (doi:10.1038/nrc1016).

68 Coffee, L.L., Casey, J.W. & Bowser, P.R. 2013 Pathology of tumors in fish associated with retroviruses. Vat. Pathol 50, 390-403.

(doi:10.1177/0300985813480529).

272

Faculty of Science The University of Sydney

69 Browning, H.M., Gulland, F.M., Hammond, J.A., Colegrove, K.M. & Hall,

A.J. 2015 Common cancer in a wild animal: the California sea lion (Zalophus californianus) as an emerging model for carcinogenesis. Phil. Trans. R. Soc. B 370.

(doi:10.1098/rstb.2014.0228).

70 Funk, W.C. et al. 2016 Adaptive divergence despite strong genetic drift: genomic analysis of the evolutionary mechanisms causing genetic differentiation in the island fox (Urocyon littoralis). Mol. Ecol. 25, 2176-2194. (doi:10.1111/mec.13605).

71 Vickers, T.W. et al. 2015 Pathology and epidemiology of ceruminous gland tumors among endangered Santa Catalina Island foxes (Urocyon littoralis catalinae) in the Channel Islands, USA. PloS One 10, e0143211.

(doi:10.1371/journal.pone.0143211).

72 Marais, H.J. & Page, P.C. 2011 Treatment of equine sarcoid in seven Cape

Mountain zebra (Equus zebra zebra). J. Wildl. Diseases 47, 917-924.

73 Sasidharan, S.P., Ludwig, A., Harper, C., Moodley, Y., Bertschinger, H.J. &

Guthrie, A.J. 2011 Comparative genetics of sarcoid tumour-affected and non-affected mountain zebra (Equus zebra) populations. S. Afr. J. Wildl. Res. 41, 36-49.

(doi:10.3957/056.041.0017).

74 Marais, H.J., Nel, P., Bertschinger, H.J., Schoeman, J.P. & Zimmerman, D.

2007 Prevalence and body distribution of sarcoids in South African Cape mountain zebra (Equus zebra zebra). J. S. Afr. Vet. Assoc. 78, 145-148.

75 Szpiech, Z.A. at al. 2013 Long runs of homozygosity are enriched for deleterious variation. Am. J. Hum. Genet. 93, 90-102. (doi:10.1016/j.ajhg.2013.05.003).

76 Shen, H. & Jin, G. 2013 Human genome epidemiology, progress and future. J.

Biomed. Res. 27, 167-169. (doi:10.7555/JBR.27.20130040).

273

Faculty of Science The University of Sydney

77 Hosking, F.J. et al. 2010 Genome-wide homozygosity signatures and childhood acute lymphoblastic leukemia risk. Blood 115, 4472-4477

(doi:10.1182/blood-2009-09-244483).

78 Gao, C.M. et al. 2002 Glutathione-S-transferases M1 (GSTM1) and GSTT1 genotype, smoking, consumption of alcohol and tea and risk of esophageal and stomach cancers: a case-control study of a high-incidence area in Jiangsu Province, China.

Cancer Lett. 188, 95-102.

79 Buch, S.C., Notani, P.N. & Bhisey, R.A. 2002 Polymorphism at GSTM1,

GSTM3 and GSTT1 gene loci and susceptibility to oral cancer in an Indian population.

Carcinogenesis 23, 803-807.

80 Benhamou, S. et al. 2002 Meta- and pooled analyses of the effects of glutathione S-transferase M1 polymorphisms and smoking on lung cancer risk.

Carcinogenesis 23, 1343-1350.

81 Cascorbi, I., Brockmoller, J., Mrozikiewicz, P.M., Bauer, S., Loddenkemper,

R. & Roots, I. 1996 Homozygous rapid arylamine N-acetyltransferase (NAT2) genotype as a susceptibility factor for lung cancer. Cancer Res. 56, 3961-3966.

82 Engel, L.S. et al. 2002 Pooled analysis and meta-analysis of glutathione S- transferase M1 and bladder cancer: a HuGE review. Am. J. Epidemiol. 156, 95-109.

83 Gutierrez-Camino, A., Martin-Guerrero, I. & Garcia-Orad, A. 2017 Genetic susceptibility in childhood acute lymphoblastic leukemia. Med. Oncol. 34, 179.

(doi:10.1007/s12032-017-1038-7).

84 Weber, B.L. & Nathanson, K.L. 2000 Low penetrance genes associated with increased risk for breast cancer. Eur. J. Cancer 36, 1193-1199.

85 Ambrosone, C.B. et al. 2003 CYP17 genetic polymorphism, breast cancer, and breast cancer risk factors. Breast cancer Res. 5, R45-51.

274

Faculty of Science The University of Sydney

86 Ewald, P.W. & Swain Ewald, H.A. 2013 Toward a general evolutionary theory of oncogenesis. Evol. Appl. 6, 70-81. (doi:10.1111/eva.12023).

87 Acevedo-Whitehouse, K. & Cunningham, A.A. 2006 Is MHC enough for understanding wildlife immunogenetics? Trends Ecol. Evol. 21, 433-438.

(doi:10.1016/j.tree.2006.05.010).

88 Campo, D.S. at al. 2016 Increased mitochondrial genetic diversity in persons infected with hepatitis C virus. Cell. Mol. Gasrtoenterol. Hepatol. 2, 676-684.

(doi:10.1016/j.jcmgh.2016.05.012).

89 Bengsch, B., Thimme, R. & Blum, H.E. 2009 Role of host genetic factors in the outcome of hepatitis C virus infection. Viruses 1, 104-125. (doi:10.3390/v1020104).

90 Matsuura, K. & Tanaka, Y. 2016 Host genetic variants influencing the clinical course of hepatitis C virus infection. J. Med. Virol. 88, 185-195.

(doi:10.1002/jmv.24334).

91 Schmidt, J., Thimme, R. & Neumann-Haefelin, C. 2011 Host genetics in immune-mediated hepatitis C virus clearance. Biomark. Med. 5, 155-169.

(doi:10.2217/bmm.11.19).

92 Snaith, A. & El-Omar, E.M. 2008 Helicobacter pylori: host genetics and disease outcomes. Expert Rev. Gastroenterol. Hepatol. 2, 577-585.

(doi:10.1586/17474124.2.4.577)

93 Malaty, H.M., Engstrand, L., Pedersen, N.L. & Graham, D.Y. 1994

Helicobacter pylori infection: genetic and environmental influences. A study of twins.

Ann. Int. Med. 120, 982-986.

94 Figueiredo, C. et al. 2002 Helicobacter pylori and interleukin 1 genotyping: an opportunity to identify high-risk individuals for gastric carcinoma. J. Natl Cancer Inst.

94, 1680-1687. (doi:10.1093/jnci/94.22.1680).

275

Faculty of Science The University of Sydney

95 Smith, S. & Hughes, J. 2007 Microsatellite and mitochondrial DNA variation defines island genetic reservoirs for reintroductions of an endangered Australian marsupial, Perameles bougainville. Conser. Genet. 9, 547. (doi:10.1007/s10592-007-

9368-1).

96 Joslin, J.O. et al. 2000 Viral papilloma and squamous cell carcinomas in snow leopards (Uncia uncia). In: American Association of Zoo Veterinarians (AAZV) and

International Association for Aquatic Animal Medicine (IAAAM) Joint Conference

(New Orleans, USA).

97 Spurgin, L.G. & Richardson, D.S. 2010 How pathogens drive genetic diversity: MHC, mechanisms and misunderstandings. Proc. R. Soc. B. 277, 979-988.

(doi:10.1098/rspb.2009.2084).

98 Újvári, B., Madsen, T., Kotenko, T., Olsson, M., Shine, R. & Wittzell, H. 2002

Low genetic diversity threatens imminent extinction for the Hungarian meadow viper

(Vipera ursinii rakosiensis). Biol. Conserv. 105, 127-130. (doi:10.1016/S0006-

3207(01)00176-8).

99 Hoffmann, A.A. & Sgro, C.M. 2011 Climate change and evolutionary adaptation. Nature 470, 479-485. (doi:10.1038/nature09670).

100 Pauls, S.U., Nowak, C., Bálint, M. & Pfenninger, M. 2013 The impact of global climate change on genetic diversity within populations and species. Mol. Ecol.

22, 925-946. (doi:10.1111/mec.12152).

101 Jacqueline, C. et al. 2017 Infections and cancer: the “fifty shades of immunity” hypothesis. BMC Cancer 17, 257. (doi:10.1186/s12885-017-3234-4).

102 Gomez-Raya, L., Amoss, M.S., Da, Y., Beattie, C.W., Ash, O. & Rauw, W.M.

2009 Role of selection and inbreeding on the incidence of cutaneous malignant melanoma in Sinclair swine. J. Anim. Breed Genet. 126, 242-249. (doi:10.1111/j.1439-

0388.2008.00779.x).

276

Faculty of Science The University of Sydney

103 Hawkins, C.E. et al. 2006 Emerging disease and population decline of an island endemic, the Tasmanian devil Sarcophilus harrisii. Biol. Conserv. 131, 307-324.

(doi:10.1016/j.biocon.2006.04.010).

104 Lachish, S., Miller, K.J., Storfer, A., Goldizen, A.W. & Jones, M.E. 2011 Evidence that disease-induced population decline changes genetic structure and alters dispersal patterns in the Tasmanian devil. Heredity 106, 172-182. (doi:10.1038/hdy.2010.17).

105 IUCN. 2001 Red list of threatened species. Version 3.1. IUCN, Gland,

Switzerland. . Available from www.iucnredlist.org.

106 Belov, K. 2012 Contagious cancer: Lessons from the devil and the dog.

BioEssays 34, 285-292. (doi:10.1002/bies.201100161).

277