Disease Risk in Wild Primate Populations: Host and Environmental Predictors, Immune Responses and Costs of Infection

by

Mercy Yvonne Akinyi

Department of Biology Duke University

Date: ______Approved:

______Susan Alberts, Supervisor

______Elizabeth Archie

______Charles Nunn

______Katharina Koelle

______Jenny Tung

Dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Biology in the Graduate School of Duke University

2017

i

v

ABSTRACT

Disease Risk in Wild Primate Populations: Host and Environmental Predictors, Immune Responses and Costs of Infection

by

Mercy Yvonne Akinyi

Department of Biology Duke University

Date: ______Approved:

______Susan Alberts, Supervisor

______Elizabeth Archie

______Charles Nunn

______Katharina Koelle

______Jenny Tung

An abstract of a dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Biology in the Graduate School of Duke University

2017

Copyright by Mercy Yvonne Akinyi 2017

Abstract

Disease risk in wild animal populations is driven by multiple factors, including host, parasite, and environmental traits, that facilitate the transmission of parasites and infection of hosts. Parasites inflict costs on their hosts that affect host fitness with downstream consequences on population structures and disease emergence patterns.

Most disease risk-related studies are conducted in captive animals, while few have focused on free-ranging populations because of the logistical challenges associated with long-term monitoring of the hosts and sample collection. Hence, data regarding disease dynamics in natural populations are scarce, which limits our understanding of the ecological and evolutionary context of disease dynamics. In this thesis, we investigate the forces driving disease risk in wild primates and the possible consequences of infection on these hosts.

We used longitudinal and cross-sectional data sets from wild primate populations in Kenya, Eastern Africa, to examine the following aims: 1) the effect of host behavior on hormones associated with disease risk, 2) environmental and host factors that predispose individuals to helminth infections, and 3) the immune responses and fitness costs associated with helminth infections. First, we investigated how two maturational milestones in wild male baboons—natal dispersal and rank attainment— were associated with variation in fecal hormone metabolites (glucocorticoids and testosterone). These two hormones are generally considered to be immunosuppressive

iv

and are often associated with high parasite loads. Within this analysis, we also investigated whether changes in the frequencies of behaviors (mating and agonistic encounters) were associated with adult dominance rank attainment. Second, we investigated multiple sources of variance in helminth burdens in a well-studied population of wild female baboons, including factors that contribute to both exposure and susceptibility (group size, social status, rainfall, temperature, age, and reproductive status). Third, we investigated how hematological indices and body mass index were associated with helminth burden.

In the first study, our results revealed that rank attainment is associated with an increase in fecal glucocorticoids (fGC) levels but not fecal testosterone (fT) levels: males that have achieved an adult rank have higher fGC than males that have not yet attained an adult rank. We also found that males win more agonistic encounters and acquire more reproductive opportunities after they have attained adult rank than before they have done so. The second study revealed that female baboons in Amboseli were infected with diverse helminth taxa, including both directly transmitted and indirectly transmitted helminths. In general, high parasite risk was linked to large group sizes, low rainfall conditions, old age, and pregnancy, although these predictors varied somewhat across helminth species. Fecal GC levels were not associated with any measures of helminth burden. The third study found that helminth burdens were positively associated with circulating lymphocyte counts and negatively associated with

v

neutrophil-lymphocyte ratios (NLR). We did not find any associations between helminth burdens and total WBC or eosinophil counts. Red blood cell indices were not predicted by our measures of helminth burden but instead varied with age class and sex.

Helminth burdens were also negatively correlated with body mass index (BMI).

Overall, the findings of this thesis are consistent with the hypothesis that host and environmental traits are important predictors of disease risk and infection in wild primate populations. In addition, our results suggest that wild primates mount immune responses to helminth burden and that helminth infections may have detrimental consequences on host body condition. Our work enhances the limited data on sources of disease variation and associated costs in wild populations. It also emphasizes the continued need for disease surveillance and health monitoring in wild populations.

vi

Dedication

This dissertation is dedicated to my beloved family and friends without whom I would not have achieved this important milestone.

“What lies behind us and what lies before us are small matters compared to what

lies within us. And when we bring what is within us out into the world, miracles

happen.” Ralph Waldo Emerson (1803-1882)

vii

Contents

Abstract ...... iv

List of Tables ...... xiii

List of Figures ...... xiv

Acknowledgements ...... xv

1. Introduction ...... 1

1.1 Background ...... 1

1.2 Behavior, hormones and disease risk ...... 2

1.3 Determinants of helminth infections ...... 4

1.4 Immune responses and costs of helminth infections ...... 7

1.5 Wild primates as a study model ...... 8

1.6 Thesis objectives ...... 10

2. Hormonal correlates of natal dispersal and rank attainment in wild male baboons. ... 13

2.1 Introduction ...... 13

2.1.1 Maturational milestones, glucocorticoids and testosterone ...... 14

2.1.2. Goals, hypotheses and predictions ...... 19

2.2 Methods ...... 20

2.2.1 Study population ...... 20

2.2.2 Study subjects, maturational milestones and behavioral data ...... 20

2.2.3 Fecal sample collection, hormone extraction, validation and analyses ...... 21

2.2.4 Data analysis ...... 22

viii

2.2.4.1 Analysis of variation in fGC and fT metabolites associated with natal dispersal and rank attainment ...... 22

2.2.4.2 Analysis of variation in mating and agonistic encounters before and after adult rank attainment...... 25

2.3 Results ...... 27

2.3.1 Analysis of variation in fGC and fT metabolites associated with natal dispersal and rank attainment...... 27

2.3.2 Analysis of variation in mating and aggression before and after rank attainment ...... 30

2.4 Discussion...... 32

3. Environmental and host-specific predictors of helminth burden in female baboons. .. 38

3.1 Introduction ...... 38

3.1.1 Environmental predictors of helminth burden: Social and ecological environments ...... 39

3.1.1.1 Group size ...... 39

3.1.1.2 Social status ...... 40

3.1.1.3 Rainfall and temperature ...... 41

3.1.2. Host traits as predictors of helminth burdens ...... 42

3.1.2.1 Host age...... 42

3.1.2.2 Reproductive state ...... 43

3.1.3 Role of glucocorticoids in susceptibility to helminth infections ...... 44

3.1.4 Aims of this study ...... 44

3.2 Methods ...... 48

3.2.1 Study population and site ...... 48

3.2.2 Fecal sample collection and parasitology analysis ...... 49 ix

3.2.3 Helminth species and life cycles ...... 50

3.2.4 Hormone assays ...... 52

3.3 Data analysis ...... 52

3.3.1 Hypothesis 1: Immune susceptibility and exposure drive helminth burdens through environmental and host-specific factors...... 53

3.3.2 Hypothesis 2: Elevated GCs are associated with suppressed immune function, hence individuals with higher fGC concentrations will have higher worm burdens...... 56

3.4 Results ...... 58

3.4.1 General patterns of parasite infection ...... 58

3.4.2 Environmental predictors of helminth burden: Social and ecological environments ...... 59

3.4.3 Host-specific predictors of helminth burden ...... 60

3.4.4 Association between Glucocorticoids and worm burden ...... 61

3.5 Discussion...... 63

3.5.1. Environmental predictors of helminth burden: Social and ecological environments ...... 64

3.5.1.1 Group size ...... 64

3.5.1.2 Social status ...... 66

3.5.1.3 Rainfall and temperature ...... 66

3.5.2 Host-specific predictors of helminth burden ...... 69

3.5.2.1 Age ...... 69

3.5.2.2 Reproductive state ...... 70

3.5.3 Host mechanisms that drive susceptibility to helminth infections...... 71

3.5.4 Conclusion...... 72 x

4. Effects of helminth infections in wild-caught olive baboons: immune responses, body condition and health...... 74

4.1 Introduction ...... 74

4.1.1 The effects of helminths on host immune function ...... 75

4.1.2 The effects of helminths on host health and body condition ...... 78

4.1.2.1 Body condition: body mass index ...... 80

4.1.2.2 Hematology: Red blood cell (RBC) indices ...... 83

4.1.3 Goals, hypothesis and predictions...... 83

4.2 Methods ...... 84

4.2.1 Study subjects and ethical review ...... 84

4.2.2 Sample collection...... 85

4.2.3 Hematology analysis ...... 86

4.3 Data analysis ...... 87

4.3.1 Hypothesis 1: Effects of helminths on host immune function...... 87

4.3.2 Hypothesis 2: Effects of helminths on host health and body condition...... 88

4.4 Results ...... 89

4.4.1 General epidemiology and heterogeneity of helminth burden by age class and sex ...... 89

4.4.2 Hypothesis 1: Effects of helminths on host immune function...... 90

4.4.3 Hypothesis 2: Effects of helminths on host health and body condition...... 93

4.5 Discussion...... 95

4.5.1 Immune responses to helminth infections ...... 95

4.5.2 Effects of helminth infection on health and body condition...... 100

xi

4.5.3 Conclusion...... 103

Appendix A: Amboseli baboons parasite guide ...... 104

Appendix B: Table S1 ...... 106

Appendix C: Table S2 ...... 109

Appendix D: Table S3 ...... 110

Appendix E: Table S4: ...... 111

Appendix F: Table S5 ...... 112

Appendix G: Table S6...... 113

Appendix G: Table S7...... 114

Appendix G: Table S8...... 115

References ...... 116

Biography ...... 138

xii

List of Tables

Table 1: Description of predictors of hormone levels ...... 24

Table 2: GLMM for fGC levels (fecal samples = 2963, males = 114) ...... 28

Table 3: GLMM for fT levels (fecal samples = 2963, males = 114) ...... 28

Table 4: Environment and host-trait predictors of helminth burden ...... 46

Table 5: Helminths of interest, life cycles and modes of transmission ...... 51

Table 6: Proportion of infected samples and infected individuals (prevalence) ...... 54

Table 7: Description of predictors of helminth burden ...... 56

Table 8: Environmental and host predictors of helminth burden (n=746, females=122) ... 62

Table 9: Summary of sample sizes in models used for statistical analysis...... 87

Table 10: Helminth species and prevalence ...... 90

Table 11: Immune responses to helminth burden ...... 91

Table 12: Effects of strongyle egg counts and richness on host health and body condition ...... 94

xiii

List of Figures

Figure 1: Average residual values of hormone concentrations as a function of age in male baboons...... 29

Figure 2: Numbers of consortships for each of 20 male baboons in the 6 months before and the 6 months after adult rank attainment...... 31

Figure 3: Counts of agonistic encounters for each of 20 male baboons in the 6 months before and the 6 months after adult rank attainment...... 31

Figure 4: Parasite richness distribution across individual samples ...... 59

Figure 5: Effect of rainfall by age interaction on log Trichuris abundance...... 61

Figure 6: Effect of log strongyle egg counts on lymphocytes...... 92

Figure 7: Effect of strongyle egg counts on NLR ...... 92

Figure 8: Effect of parasite richness on BMI ...... 93

xiv

Acknowledgements

My academic life for the past six years in graduate school to date has been greatly shaped by my advisor Dr. Susan Alberts. I am humbled and honored because she recognized the potential in me and encouraged me to apply for a PhD program at

Duke University. I have experienced overwhelming support and guidance from her as a mentor and I am still amazed at how she always has a way of turning lemons into lemonade. Her jovial and positive attitude have been extremely encouraging to me to the very end of this journey.

I had wonderful dissertation committee members who were equally supportive for the past few years. It has been an honor receiving guidance from you all. Thank you,

Drs. Elizabeth Archie, Charlie Nunn, Jenny Tung and Katharina Koelle.

I am grateful to all the members of the Alberts lab, past and present for their intellectual contribution and for ensuring that I felt at home at Duke: Jake G, Laurence,

Amanda, Emily M, Matthew, Emily L, Arielle, Courtney, Jake. M, Mathias, and Patrick. I am also grateful to the Tung lab members for the awesome years of fieldwork together and for their contributions during my stay at Duke. I thank BPCE and BEAM groups for constructive feedback for presentations and manuscript write ups. Thank you to all the undergraduates from Duke, Notre Dame, Princeton and IPR who contributed to the different data sets that were used in this thesis.

xv

I gratefully acknowledge the support I received from the Amboseli Baboon

Research Project. I am especially thankful to the project leaders (Drs. Altmann, Alberts,

Archie and Tung) and the field research team (Raphael, Serah, Kinyua, Gideon, Ben and

Longida) for the dedication to research spanning decades. I have enjoyed working with you all and I look forward to more years of collaborative research. I am equally grateful to Nkii, Munyoi and Alex for their support at the field site and to V. Somen and T.

Wango for their untiring assistance at the University of Nairobi. Karl Pinc has provided expertise in database design and management and I am very grateful for his seminal contributions to the development of BABASE, the Baboon Project database. I also thank

N. Learn, L. Maryott, and J. Gordon.

I acknowledge the various funding agencies that made this project possible. I was supported by the Franklin Mosher Baldwin Fellowship (LS.B. Leakey Foundation),

Nacey Maggioncalda Foundation and the National Evolutionary Synthesis Center. I also thank the following institutions for their contributions; Duke University, Princeton

University, the Chicago Zoological Society, and the Max Planck Institute for

Demographic Research, Kenya Wildlife Services, Institute of Primate Research, National

Museums of Kenya, National Council for Science and Technology, members of the

Amboseli-Longido pastoralist communities, Tortillis Camp, Ker & Downey Safaris, Air

Kenya, and Safarilink for their cooperation and assistance in Kenya.

xvi

I am grateful to my friends, and workmates at the Institute of primate research for their invaluable support in laboratory analysis and overall intellectual feedback. I am especially grateful to the animal sciences department staff, to the IPR administration,

Drs. Ngotho, Kariuki, Omondi, Chai and Ocholla, and Ruth, Waititu, Macharia, Fred,

Lucy, Mutura, Victor, and Peris.

I would like to thank my family and friends in Kenya (Lillian Aloo, Aggrey

Ojiambo, Charles Odeny, Amos Ndung’u, my extended family, book club friends) and in the United states (Francoise, Vinny, Danny, Henry, Bernard, Kelly, Stephanie, Amber,

John, Kristy and Travis, BSF, Summit and GCF community of friends) for their love and continued support over the years. Finally, I thank the Almighty God for good health, provision and the strength to faithfully run this race to the very end.

xvii

1. Introduction

1.1 Background

Kilpatrick & Altizer (2012) define disease ecology as the “study of host-pathogen interactions within the context of their environment and evolution”. Disease ecology aims to understand principles that influence disease transmission over time and space and the resulting effects on the affected animal populations (Kilpatrick & Altizer 2012). Disease risk is defined as “the probability of acquiring an infectious or a non-infectious disease” (Nunn

& Altizer, 2006). Within human and animal populations, individuals vary in their risk of disease; traits of the host, traits of the pathogens, and features of the environment all contribute to this variation. These factors are not mutually exclusive and may act in unison to influence disease risk by either increasing host susceptibility or facilitating host exposure to infective stages of parasites. The unifying theme of this dissertation is that we look at multiple host and environmental factors to explain variation in disease risk in wild primates in the context of their environment and evolution. We also investigate the association between helminth burden and host health and body condition.

In natural populations, parasites inflict costs on their hosts which not only affect host growth, reproduction and survival but also affect population structures and disease emergence patterns (Hudson and Dobson, 1992; Tompkins et al., 2011; Tompkins and

Begon, 1999). It is thus important to study predictors of disease within the context of

1

their environment. Despite the numerous studies done to understand how host, pathogen, and environmental factors interact to influence disease outcomes in individuals, there are still sources of variation that are yet to be identified. A key missing component of many studies, particularly in laboratory settings, is data on disease dynamics in natural populations; without such data, our understanding of the ecological and evolutionary context of disease dynamics is limited (Gulland, 1995; Wobeser, 2006).

The chapters in this thesis all aim to fill this gap by using longitudinal and cross- sectional data sets from wild primates to examine hosts and environmental factors that predispose individuals to infections and the immune responses and costs associated with infections.

1.2 Behavior, hormones and disease risk

Host behavior influences both exposure to parasites and host immune function and hence is associated with disease risk in hosts (Ezenwa et al., 2016a; Habig & Archie,

2015; Sapolsky, 2004). Increased exposure to parasites is facilitated by various host behaviors. For example, foraging increases exposure to infective stages of soil- transmitted helminths, affiliative behavior such as grooming facilitates exposure to ectoparasites and contagious parasites, mating is especially important for acquisition of sexually transmitted diseases, and migration can either assist or inhibit transmission of pathogens depending on host density (Ezenwa, 2016b; Freeland, 2009; Nunn & Altizer,

2006). Host behavior can also affect energetics and hormonal responses, which in turn

2

interfere with or enhance host immune competence. Migration, agonistic interactions and mate guarding require increased mobilization of host energy and hence deplete host resources required for reproductive success and maintenance of host immunity

(Bercovitch & Ziegler, 2002). Engaging in these behaviors may be perceived as stressful by the host and thus may be associated with changes in host physiology, specifically, activation of steroid hormones like glucocorticoids (GC) and testosterone (T) (Dixson,

2012; Nelson, 2011; Sapolsky et al., 2000). Short term production of GC is important for survival because it mobilizes energy needed to eliminate stressors, whereas chronic GC production is associated with increased disease risk (Bercovitch and Ziegler, 2002;

Sapolsky, 2004).

Glucocorticoids and T are thought to influence immune responses by altering host immunity through the reduced production of immune cells (e.g. T cells and monocytes) (Roberts, Buchanan, & Evans, 2004; Sapolsky et al., 2000; Zuk & McKean,

1996). Glucocorticoids also inhibit nuclear factor kappa B (NF-kappaB) signaling pathway which is important for activation of adaptive and innate immune responses against infections (Auphan et. al., 1995). Immunosuppressive effects of steroid hormones results in ineffective control of current infections and predispose the hosts to future infections (Elenkov & Chrousos, 2002; Martin, Weil, & Nelson, 2006; Sapolsky et al.,

2000). Several studies have associated glucocorticoids and testosterone levels with increased likelihood of infection by a range of parasites in various taxa (see discussions

3

in Roberts et al., 2004; Zuk & McKean, 1996) . High concentrations of these steroid hormones also directly facilitate growth and reproduction of parasites through mechanisms that are well known in some parasites and are yet to be understood in others (Klein, 2004). Protozoa, nematodes, trematodes and arthropod parasites have all been associated with increased steroid hormone levels in multiple animal species (Klein,

2004; Romano et al., 2015). Studies linking behavior, hormones, and disease risk are thus necessary to understanding variation in individual disease.

1.3 Determinants of helminth infections

In addition to host behavior, a range of variables, including environmental factors, additional host factors, and parasite factors, influence disease risk in hosts. I will discuss these factors with specific reference to helminths as the parasite of interest and primates as the hosts. Helminths are ubiquitous across all vertebrates and constitute 43% of the total parasites infecting free-living nonhuman primates (Nunn & Altizer, 2006).

Primates are infected by multiple genera of nematodes, trematodes and cestodes

(Cogswell, 2007). Most helminth infections appear asymptomatic but may occasionally be associated with detrimental effects, especially in individuals harboring exceptionally high helminth burdens (Girgis et al., 2013). Helminths can pose costs to their hosts and hence affect growth, reproduction and survival (Coop & Holmes, 1996; Hamilton & Zuk,

1982; Hudson & Dobson, 1992). It is therefore important to understand the factors that facilitate infection of hosts by these helminth species, including social factors (group

4

size, social status), environmental factors (rainfall, temperature, humidity, vegetation), host traits (age, reproductive state, sex, ) and pathogen traits (infectivity, virulence, pathogenicity, host specificity, life cycle) (Nunn & Altizer, 2006).

Factors predicting helminth infections operate through two main mechanisms to ensure successful infection of hosts and subsequent transmission of helminth infectious stages to additional hosts: 1) exposure of the host to infectious helminth stages and 2) host susceptibility. Exposure of the host is dependent on survival of infectious stages of helminths in the environment (in the case of directly transmitted helminths) or in the vectors (for indirectly transmitted helminths). In addition, hosts must encounter these infectious stages or vectors.

In general, most helminth infectious stages will not survive in extreme environments; therefore, rainfall and temperature are often considered to be important in determining host exposure to these infectious stages (Nunn & Altizer, 2006). Effects of ecology on helminth survival and reproduction are species-specific because of variation in life cycles and transmission modes between species. Ecological variables may also facilitate or limit the survival of vectors, which in turn determines whether hosts will be exposed to helminth infectious stages. Social environments such as group size and group ranging behavior, social networks, and social status (i.e., dominance rank) facilitate or reduce exposure of hosts to directly transmitted parasites (Habig & Archie,

2015; Kappeler, Cremer, & Nunn, 2015; Schmid-Hempel, 2017). Large group sizes, high

5

social status, high network proximity indices, and intensive group ranging are positively correlated with helminth burden (Cote & Poulin, 1995; Friant et al. 2016; Habig &

Archie, 2015; Kappeler, Cremer, & Nunn, 2015; Patterson & Ruckstuhl, 2013; Rifkin et al., 2012). Host traits such as age, reproductive state, sex, and body mass also affect the likelihood of exposure to parasites. Specifically, individuals who are older, pregnant, lactating, large bodied, and males are more likely to be exposed to more infectious stages because of increased intake of contaminated food (see discussion in Foo et al.,

2017; Nunn & Altizer, 2006; Vitone et al. 2004; Zuk & McKean, 1996).

Maintenance and transmission of parasites depend on the presence of susceptible, infected, and recovered hosts. Host susceptibility is a state whereby the host is unable to mount effective immune responses against infections; it depends on a myriad of factors. For instance, climatic factors such as amount of rainfall, temperature and humidity affect host nutrition and hence may indirectly affect disease status

(Pietrock & Marcogliese, 2003). Adequate rainfall is associated with increased food resources and sufficient water for host nutrition, resulting in good body conditions and host immunocompetence (Beldomenico et al., 2008; Beldomenico & Begon, 2010; Long,

KZ, Nanthakumar, 2004). On the other hand, extremes of ecological conditions are associated with energetic stress and impaired host immunity (Dowell, 2001; Hawley and

Altizer, 2011; Martin et al., 2008). Individuals who are old, high or low ranking, pregnant, lactating, and/or in poor body condition are likely to have impaired immune

6

systems and thus be incapable of mounting effective immune responses against parasites (see discussion in Nunn & Altizer, 2006, Table S1). Host susceptibility is thought to be highly linked to stress and GC levels. Mechanisms linking stress, GC levels, and disease susceptibility have been extensively studied in humans and in some primate populations but are still poorly understood (Kappeler, Cremer, & Nunn, 2015;

Sapolsky, 2004). Identifying the infection status of hosts is highly feasible for helminth infections in wild primate populations, because the required sampling is non-invasive

(Gillespie, 2006). However, it is relatively rare for studies to have information on more than one or a few potential predictors, thus limiting the simultaneous analysis of various predictors of helminth burden.

1.4 Immune responses and costs of helminth infections

Many studies performed in natural nonhuman primate populations have focused on surveillance of helminths without necessarily assessing immune responses or costs associated with parasitism (Gillespie, 2006; Nunn & Altizer, 2006). Immune responses mounted by hosts are important in determining the impact that parasites have on hosts and the probability of acquiring further infections (Llyod, 1995; Sandland and

Minchella 2003). Few cross-sectional or longitudinal studies have been conducted to evaluate immune responses and costs of helminth infection in wild primates (but see

Farah et al., 2003; Horii et al., 1982; Monteiro et al., 2010; Nguyen et al., 2015; Springer et al. 2015). Understanding patterns and consequences of helminth infections in natural

7

populations is important for characterizing factors that shape infectious disease emergence patterns and host population dynamics (Tompkins et al., 2011; Tompkins and

Begon, 1999).

Animals in the wild are exposed to a diversity of pathogens, which creates opportunities for co-infections. In fact, it is highly unlikely that animals in natural populations will only have a single parasite infection at a given time (Ezenwa & Jolles,

2011). Parasites may influence each other by facilitating or inhibiting each other’s proliferation and transmission (see discussion in Jolles et al., 2008). These interactions may affect host susceptibility, host availability, or disease severity through either resource competition or immune-mediated interactions (Graham, 2008; Pedersen &

Fenton, 2007). These interactions will determine whether costs of infection to the host will be alleviated or exacerbated (Ezeamama et al., 2005, 2008). Few studies have evaluated the effects of multiple helminth species on wild primate hosts, so quantifying the costs on host health and body condition will help us understand the selective pressures affecting host fitness in their environments.

1.5 Wild primates as a study model

Research in wild animal populations is extremely important because it provides an opportunity to understand variables contributing to infection in the context of evolution and ecology (Bordes & Morand, 2011). Measuring physiological, behavioral, and social predictors of and responses to infectious disease by primates can be useful in

8

identifying the genesis of human responses to infectious diseases, as well as in understanding sources of variance in disease susceptibility (Nunn & Altizer, 2006).

Understanding disease variation is a pivotal topic for evolutionary medicine because disease-causing pathogens appear to have created some of the strongest signatures of selection so far identified in the human genome (Barreiro and Quintana-Murci 2010;

Bustamante et al., 2005; Fumagalli et al 2011; Olson & Varki, 2003).

Primates are also implicated in the transmission of zoonoses and emerging diseases, which are increasingly important challenges facing medical and public health practitioners (Taylor et al. 2001). Zoonoses are diseases that are transmitted between humans and animals, and emerging diseases are those that are appearing in the population for the first time or that have been present in a population but are increasing in incidence. The World Health Organization (WHO) reports that more than 75% of all infectious diseases affecting humans are of zoonotic origin, and two thirds of emerging disease pathogens are zoonotic (WHO, 2016). Infection by these pathogens may severely compromise health and in some cases cause death. Over the past few years, eradication efforts directed at zoonoses have focused on characterizing the disease-causing pathogens, understanding their transmission patterns, and developing and implementing management practices to eradicate these pathogens in the environment.

Another advantage of disease risk studies in nonhuman primates is that such studies provide a basic understanding of diseases affecting primates for the purposes of

9

conservation and diversity. Apart from habitat degradation and climate change, other causes of decline in primate species may include disease and parasitism. The

International Union for Conservation of Nature states that more than half of the world’s primates are endangered (IUCN, 2016). Developing conservation strategies to protecting these primates is crucial. Understanding the disease transmission and patterns in natural populations can contribute to the development of these strategies.

1.6 Thesis objectives

The goal of my doctoral thesis was to understand three major aspects of disease risk that are widely viewed as important but are still poorly understood: i) the effects of host behavior – in particular, behaviors associated with natal dispersal and the attainment of adult social status– on disease susceptibility, ii) the effects of environmental and host factors on helminth infections, and iii) the immune responses against helminth infections and associated costs.

Chapter 2 focuses on how host behavior contributes to variation in GC and T hormone levels (these hormones are associated with disease susceptibility). The initial aim of this study was to investigate the link between behavior, hormones, and helminth infections. We investigated the link between hormones and behavior but were unable to look at the relationship between host behavior and helminth infection because of challenges of sample collection for parasitology analysis. In this study, GC and T levels were considered as factors that may influence disease risk. In mammals, maturational

10

milestones such as puberty, rank attainment and natal dispersal, mark important transitions that are essential for the onset of reproductive activity. These transitions critically depend on a set of coordinated hormonal and socio-behavioral changes.

Studies that focus on the link between hormones and maturational milestones are uncommon in wild mammals because of the challenges of obtaining adequate sample sizes of maturing animals and of tracking the movements of dispersing animals. I studied members of a well-studied wild baboon population in the Amboseli region of

Kenya. I examined two maturational milestones in wild male baboons—natal dispersal and the attainment of adult social status (i.e., dominance rank)—and measured their association with variation in fecal testosterone (fT) and glucocorticoids (fGC). I also investigated changes in the frequencies of behaviors (mating and agonistic encounters) associated with adult dominance rank attainment.

Chapter 3 aims at investigating environmental and host predictors of disease risk

(helminth burden) in adult female baboons in the Amboseli population. Helminth infections are ubiquitous in most animals and affect host fitness. Helminth burdens vary across individuals as a function of varying susceptibilities and exposure rates. Several studies in primates have reported various combinations of factors influencing helminth burdens in individuals. However, there are mixed empirical results on which predictors are important. My study aimed at, 1) identifying the sources of variance in helminth burdens in adult female baboons and 2) identifying the mechanisms that drive

11

susceptibility to helminth infections. This study employed a multivariate model that examined the effects of multiple individual and environmental variables using a large data set. I also investigated the relationships between GC and helminth burdens in this population for the first time.

Chapter 4 focuses on measuring the immune responses and physiological costs experienced by hosts as a consequence of helminth infections. Parasites impose fitness costs on hosts by reducing energetic resources available for normal functioning; these costs, in turn, make individuals more susceptible to further disease (Brown 1987).

Immune responses elicited by hosts are also important in determining the impact the parasites have on hosts. I examined cellular immune responses associated with helminth infections, as well as the consequences of these infections for the hosts (wild-caught baboons). I capitalized on samples collected from baboons by the Institute of Primate

Research (IPR), Kenya during routine quarantine screening of wild caught baboons. IPR is a biomedical research center that uses wild caught primates trapped from areas of human-nonhuman primate conflict under the recommendation and approval of the

Kenya Wildlife Service. By taking advantage of the valuable and rare opportunity afforded by having access to a large set of animals immediately after they were trapped from the wild, I was able to investigate whether helminth infections were associated with circulating host immune cells, host physiological indices, physical condition and host immune status.

12

2. Hormonal correlates of natal dispersal and rank attainment in wild male baboons.

2.1 Introduction

In mammals, maturational milestones such as puberty, adult dominance rank attainment and natal dispersal mark important transitions associated with the onset of reproductive activity (Adkins-Regan, 2005; Jack et al., 2014; Pusey, 1987). These transitions critically depend on a set of coordinated hormonal and socio-behavioral changes (Finch and Rose, 1995). Further, they require the mobilization of resources from individuals’ energy stores, and physiological preparation for reproduction (see discussion in Belthoff and Dufty, 1995,1998; Nunes et al., 1999; Ronce and Clobert, 2012).

Steroid hormones, specifically glucocorticoids (GC) and testosterone (T) are presumed to play key roles in these processes because of their involvement in energy mobilization and reproduction (Dixson, 2012; Nelson, 2011; Sapolsky et al., 2000). In the present study, we examined two maturational milestones in wild male baboons: natal dispersal and adult dominance rank attainment. We investigated changes in frequencies of key behaviors (mating and agonistic encounters) associated with adult dominance rank attainment and we measured the association between the timing of these milestones and variation in fecal glucocorticoids (fGC) and testosterone (fT) (see Gesquiere et al., 2005;

Onyango et al., 2013b for similar analyses at puberty).

13

2.1.1 Maturational milestones, glucocorticoids and testosterone

Dispersal is nearly ubiquitous across the animal kingdom, enhancing gene flow between populations, limiting inbreeding, and allowing individuals to distribute themselves over available resources (Clutton-Brock, 1989; Greenwood, 1980; Johnson and Gaines, 1990; Handley and Perrin, 2007; Paradis et al., 1998). Dispersal represents a major maturational milestone for many mammals, because individuals often do not reproduce prior to dispersal (Greenwood, 1980; Pusey and Packer, 1987; Pusey, 1987).

Many mammals – including baboons, the subjects of this study – exhibit male-biased dispersal, so that social groups consist of families of female kin, with immigrant males who tend to be unrelated to other group members other than their own offspring

(Greenwood, 1980; Pusey, 1987; Clutton-Brock, 1989). Adult dominance rank attainment

— the stage at which a maturing male consistently begins to outrank at least one adult male in agonistic encounters — also represents an important maturational milestone because individuals (especially males) are often unable to acquire mating opportunities prior to adult rank attainment (Alberts and Altmann, 1995b; Alberts et al., 2006;

Charpentier et al., 2008; Beehner et al., 2009; Hamilton and Bulger, 1990; Van Noordwijk and Van Schaik, 2001).

Natal dispersal and adult rank attainment both represent a potent combination of social, reproductive, and energetic challenges, possibly among the greatest challenges that an animal will experience in its lifetime. For instance, dispersing primates may face

14

increased mortality risk as a result of predation, partly because they may often be alone until they join a new group (Alberts and Altmann, 1995; Cuarón, 1997; Fernandez-D uque, 2009; Isbell and Jack, 2009). Dispersing individuals may also experience nutritional stress resulting from lack of familiarity with existing resources in new areas

(Pusey and Packer, 1987). Some dispersing individuals travel long distances before they encounter a new group, potentially depleting energy stores during dispersal (see discussions in Dufty and Belthoff, 2001). Successful dispersal and rank attainment may also involve fighting with conspecifics, resulting in injuries or occasionally death (see discussion in Arlet et al., 2009; Isbell and Jack, 2009; Van Noordwijk and Van Schaik,

2001).

Hormones and socio-sexual behavior operate in a feedback loop, each responding to and contributing to the regulation of the other (Bercovitch and Ziegler,

2002; Nelson, 2011). For instance, a stressful event—an event that has the potential to perturb physiological homeostasis (Sapolsky, 2004)—leads to activation of the hypothalamic-pituitary-adrenal (HPA) axis and the adrenomedullary system, which in turn leads to downstream secretion of GCs (e.g., cortisol and corticosterone) and catecholamines (epinephrine and norepinephrine) (Elenkov and Chrousos, 2002;

Sapolsky et al., 2000). Secretion of these hormones in turn stimulates or suppresses various cardiovascular, metabolic, and immune system processes (e.g., heart rate, blood pressure, gluconeogenesis, lipolysis, inflammation), all of which may have immediate or

15

downstream consequences for behavior. A short-term release of GC during stressful encounters is adaptive and aids in survival. On the other hand, chronic release of GC is immunosuppressive and consequently leads to increased susceptibility to disease and infection (Sapolsky et al., 2000; Sapolsky, 2004; Selva et al., 2011).

Testosterone (T) is produced in the testes and regulated by the hypothalamic- pituitary-gonadal (HPG) axis (Hirschenhauser and Oliveira, 2006). Testosterone regulates the production of gametes, enhances anabolism, and maintains musculoskeletal performance (Adkins-Regan, 2005; Bardin, 1996; Bribiescas, 2001).

Testosterone has previously been associated with behaviors such as territoriality, mate guarding/consortship, aggression, and mating display (Beehner et al., 2006; Onyango et al., 2013a; Wickings and Dixson, 1992; Wingfield, et al., 1990). The ‘challenge hypothesis,’ designed to explain the link between T and competitive behavior in seasonal breeders, predicts that T levels rise in response to social challenges, particularly those associated with competition for mates, and specifically that T levels can be rapidly elevated when critically needed to promote aggressive competition in a reproductive context (Wingfield et al., 1990). Research in seasonally breeding birds has supported this hypothesis by showing that T is low in the absence of mating behavior or mate competition and elevated during aggressive episodes and mating (e.g. Teichroeb and

Sicotte, 2008). Studies in both seasonally and non-seasonally breeding nonhuman primate species have also supported the challenge hypothesis (chacma baboons: Beehner

16

et al., 2006; howler monkeys: Cristóbal-Azkarate et al., 2006; ring-tailed lemurs: Gould and Ziegler, 2007; moustached tamarin monkeys: Huck et al., 2005; tufted capuchin monkeys: Lynch et al., 2002; bonobos: Marshall and Hohmann, 2005; chimpanzees:

Muller and Wrangham, 2004; and red-fronted lemurs: Ostner et al., 2002). Testosterone is energetically costly to maintain at high levels because of its anabolic effects (e.g. muscle mass building). These energetic costs will lead to decreased investment in immune function and increased susceptibility to disease (Muehlenbein and Bribiescas,

2005; Zuk and McKean, 1996).

Studies that focus on the link between hormones and maturational milestones are uncommon in wild mammals because of the challenges of obtaining adequate sample sizes of maturing animals and of tracking the movements of dispersing animals.

Only a few studies have looked at the endocrine correlates of dispersal in mammals

(yellow baboons: Alberts et al., 1992; chacma baboons: Beehner et al. 2006, Bergman et al., 2005; spotted hyenas: Holekamp and Smale, 1998; white-faced capuchin monkeys:

Jack et al., 2014; chimpanzees: Kahlenberg et al. 2008; Belding’s ground squirrels: Nunes et al.,1999; Muriqui monkeys: Strier and Zeigler 2000; crested Macaques: Marty et al.

2017). Some studies of GC or T in relation to dispersal have compared hormone levels in immigrant males with those in resident males, whereas others have compared hormone levels in the same individuals before and after dispersal, or at different stages after immigration. For example, in spotted hyenas, immigrant males – those that have

17

immigrated within the previous 7 days – have higher circulating T than natal males even after controlling for the effects of age (Holekamp and Smale, 1998). Alberts et al. (1992) found unusually high GC and T levels in a particularly aggressive male baboon immediately after he immigrated into a new group. In chacma baboons, T levels appear to rise in anticipation of future mating opportunities and rank rises (Beehner et al., 2006;

Bergman et al., 2005). Fecal GC levels are higher in immigrant male chacma baboons in the month following immigration than they are in subsequent months of residency

(Bergman et al. 2005), and the same is qualitatively true in gray-cheeked mangabeys

(Arlet et al. 2009). In crested macaques, fGC levels are elevated the first few days post immigration after which they return to baseline (Marty et al. 2017). Among chimpanzees, recent immigrants to a community have higher GC levels than natal residents (Kahlenberg et al., 2008). However, a longitudinal study in black howler monkeys showed that immigrant males did not differ in their GC and T levels 2 weeks before and after immigration (Van Belle et al., 2009). In muriqui monkeys, cortisol levels did not differ between natal emigrant females and recent immigrant females (Strier and

Ziegler, 2000). Notably, only one study to our knowledge has looked at T variation around the time of adult rank attainment, in three primate species (chacma baboons, yellow baboons, and geladas: (Beehner et al., 2009). In that study, the acquisition of adult ranks in males occurs at the onset of adult T levels but before peak T levels (Beehner et al., 2009).

18

2.1.2. Goals, hypotheses and predictions

The goal of this study was to measure the hormonal correlates of natal dispersal and rank attainment in wild male baboons during the developmental period between puberty and full adulthood (6-10 years). We also looked at the behavioral changes associated with adult dominance rank attainment. Specifically, we tested the hypothesis that psychosocial and energetic demands on male baboons will be substantially higher after dispersal and/or rank attainment than before them. Animals may experience increased nutritional stress and other energetic and social demands as they adjust to their new environment after dispersal, and males are likely to experience substantially increased reproductive activity following rank attainment. This hypothesis yielded three predictions that we tested. 1) We predicted that individuals that had attained one or both of these milestones would have higher levels of fGC than those that had not. 2) We predicted that individuals that had attained one or both of these milestones would have higher levels of fT than those that had not as a consequence of greater mating and agonistic activity after milestone attainment. 3) In addition, because rank attainment appears to be pre-requisite for adult reproductive activity in male baboons, we predicted that individuals would engage in agonistic and mating encounters more frequently after attaining this milestone.

19

2.2 Methods

2.2.1 Study population

Our study focused on members of a well-studied wild baboon population in the

Amboseli region of Kenya. This baboon population is composed of yellow baboons,

Papio cynocephalus, that experience some admixture with neighboring populations of olive baboons, Papio anubis (Alberts and Altmann, 2001; Charpentier et al., 2012; Tung et al., 2008). The Amboseli baboon population has been studied for the last four decades; this research has produced longitudinal, individual-based behavioral and demographic data, as well as fecal samples used for the measurement of hormone metabolites.

2.2.2 Study subjects, maturational milestones and behavioral data

Demographic, behavioral, and life-history data (births, emigration, immigration and maturation events) are collected on a near-daily basis on known individuals in the study population (Kappeler et al., 2012) and are stored in BABASE, the Amboseli baboon project database (Pinc et al., 2016). The subjects of this study were 114 males aged 6 to 10 years; all subjects were born into study groups and known from birth, and hence their ages were known to within a few days. Males in this population attain puberty (assessed by testicular enlargement) at a median age of 5.4 years; 5% of males disperse before puberty, and 95% of males disperse after puberty, at a median natal dispersal age of 7.67 years (age range of post-pubertal natal dispersals is 6.81 - 13.42; Onyango et al., 2013b).

A third maturational milestone, adult dominance rank attainment, occurs when a

20

maturing male consistently begins to outrank at least one adult male in agonistic encounters. We consider the attainment of adult rank to signal the transition for males from subadulthood—a period of rapid growth but reproductive quiescence after puberty—to adulthood (Alberts and Altmann, 1995b; Altmann and Alberts, 2005). Adult rank attainment occurs at a median age of 7.45 years (range 6.2 to 8.7 years; Onyango et al., 2013b). Our study focused on the two post-pubertal maturational milestones: dispersal from a natal group to a non-natal group, and the attainment of adult dominance rank. Mating and agonistic encounters were collected during all-occurrences sampling of mate-guarding episodes and systematic representational sampling of agonistic interactions, and were recorded in our database (Alberts and Altmann, 2011).

Mate-guarding episodes involve mating as well as close, persistent following of an estrous female by an adult male, and are often termed ‘consortships’ in primates

(Hausfater, 1975; Onyango et al., 2013a). Agonistic encounters analyzed here were

‘decided’ dyadic interactions, i.e., those involving interactions between two males that yielded a clear winner and loser (Hausfater, 1975).

2.2.3 Fecal sample collection, hormone extraction, validation and analyses

Fresh fecal samples for hormone analysis were opportunistically collected from individually known males, labeled with the individual male’s ID and the date of sample collection, and stored in 95% ethanol. Protocols for extraction and measurement of fGC and fT metabolites are well-established and have been previously validated in previous 21

research on this population (Beehner et al., 2006; Gesquiere et al., 2005; Khan et al., 2002).

These samples were freeze dried, sifted, and weighed; they underwent methanol and solid phase extraction using Waters Oasis HLB cartridges. fGC and fT metabolites in the samples were measured in ng/g of dried feces using radio-immunoassays (RIA). RIA specific to fGC and fT metabolites have previously been validated for our population

(Beehner et al., 2009; Gesquiere et al., 2011a; Khan et al., 2002; Lynch et al., 2003). For fT assays, we have used different kits over the years, all of which produce highly correlated results. These kits have included the kit from Equate (SolidPhase, Portland, ME; See

Lynch et al., 2003 for validation), the kit from Diagnostic Systems Laboratories (Beckman

Coulter, Webster TX; Beehner et al., 2009; Gesquiere et al., 2011a), and the kit from

Pantex (Santa Monica, CA; Gesquiere et al., 2014). To render the fT concentrations directly comparable across kits, a correction factor was applied to the fT from the different kits (details in Gesquiere et al., 2011a).

2.2.4 Data analysis

2.2.4.1 Analysis of variation in fGC and fT metabolites associated with natal dispersal and rank attainment

Response variables and modeling approach

Our data set included measures of hormone levels from 2963 fecal samples collected from 114 unique subadult and adult males (aged 6-10 years) who were members of 7 different social groups between the years 2000 and 2013; most of these

22

individuals were represented in our data set by multiple fecal samples (mean number of fecal samples per individual=26, range 1 to 97). To test our predictions, we ran two general linear mixed models, one for each hormone. The response variables for the models were log-transformed fGC levels or log-transformed fT levels. Predictor variables are described below and in Table 1. To generate our mixed effect models, we used the glmmadmb package (Fournier et al., 2012) in R Statistical software, version

3.2.3 (R Foundation for Statistical Computing, 2014). Visual inspections of the data for all models indicated no violations of the assumptions regarding the normality and homogeneity of error variances.

Predictor variables

The identity of the individual was treated as a random effect to control for multiple samples from most individuals. Our fixed effect of interest was the maturation status of the individual at the time the fecal sample was collected. Individuals were assigned to one of the following categories for maturation status: neither (he had neither dispersed nor attained rank, N = 1422 fecal samples, 101 males), dispersed (he had dispersed but not attained rank, N = 114 fecal samples, 13 males), ranked (he had attained rank but had not dispersed, N = 211 fecal samples, 35 males), and both (he had dispersed and attained rank, N = 1216 fecal samples, 52 males). Age and season were included in all analyses as covariates because they are known to predict fGC and fT

23

hormone levels in male baboons. Specifically, fT levels in Amboseli males are stable prior to puberty between 2 and 5 years of age, increase through early adulthood to a peak at 9.5 years, and then gradually decline with age throughout adulthood (Altmann et al., 2010; Beehner et al., 2009; Onyango et al., 2013b). In addition, fGC levels are higher during the dry season (June-October) than the wet season (November-May), while fT levels show the reverse pattern with higher levels during the wet season (Alberts et al.,

2005; Altmann et al., 2002; Gesquiere et al., 2011a). We categorized each month of our study period as a dry month (June through October) or a wet month (November through May).

Table 1: Description of predictors of hormone levels

Predictor variable Description

a) Random effects Male ID Male Identity.

b) Fixed effects

Maturation status The maturation status of the individual at the time the fecal sample was collected (categorical). Statuses included: neither (he had neither dispersed nor attained rank), dispersed (he had dispersed but not attained rank), ranked (he had attained rank but had not dispersed), and both (he had dispersed and attained rank).

Age Age in years at sample collection (continuous).

Season Whether the sample was collected during a wet or dry season (wet, dry); binary.

24

Categorical analysis of post-dispersal fGC and fT

Finally, following Bergman et al. (2005), we tested the possibility that hormone levels would be elevated in the month immediately following dispersal but not during later post-dispersal months. Specifically, we grouped post-dispersal fecal samples into two classes, those that were collected within one month of dispersal, and those that were collected more than one month after dispersal (i.e., between 2 and 6 months post- dispersal). Then we identified subjects for whom we had fecal samples in both categories, and restricted the analysis to these males. We used the Wilcoxon test (a non- parametric equivalent of paired t-test) to examine the association between post-dispersal category and fGC, and between post-dispersal category and fT. This differed slightly from the approach taken by Bergman et al. (2005) because in their analysis, post- dispersal fGC levels did not vary across individuals, allowing them to disregard individual identity and perform an ANOVA. In our case, we had significant heterogeneity among individuals in post-dispersal fGC levels (F38,184=3.473, P<0.0001) and fT (F38,184=5.344, P<0.001). Thus, we opted for the paired Wilcoxon sign ranked test to control for these individual differences.

2.2.4.2 Analysis of variation in mating and agonistic encounters before and after adult rank attainment

Rank attainment and natal dispersal are frequently, but not always, attained within a few weeks or months of each other in our study population, making them highly associated (2 = 1809.6, p < 0.001, Cramer’s V=0.781; see also Alberts & Altmann 25

1995b, Onyango et al., 2013b). In order to examine behavioral changes solely associated with rank attainment, we examined only males that attained rank at least six months prior to dispersal. First, we retrieved data on agonistic encounters and consortships to measure differences in agonistic and mating behavior associated with adult rank attainment. This data set included behavioral data during the 6 months prior to and 6 months after rank attainment for 20 unique subadult and adult males (aged 6 to 10 years) who were members of 7 different social groups between the years 2000 and 2013.

Specifically, the data was restricted to males who (i) were present in a study group during the six months prior to and the six months following their adult rank attainment, and (ii) attained rank at least 6 months before natal dispersal. We were unable to examine behavioral changes solely associated with natal dispersal, because the individuals who dispersed before they attained adult rank either dispersed into non- study groups or attained rank within 6 months after dispersal.

We recorded the counts of wins, losses, sum of wins and losses and the total counts of consortships before and after rank attainment and did a Wilcoxon signed-rank

(paired) test to determine whether there were any significant changes in these behaviors prior to and after rank attainment.

26

2.3 Results

2.3.1 Analysis of variation in fGC and fT metabolites associated with natal dispersal and rank attainment

Fecal glucocorticoid (fGC) levels were significantly higher in males with a status of ‘ranked’ or ‘both’ than in males with a status of ‘neither’; in contrast, males with a status of ‘dispersed’ did not have higher fGC levels than males with a status of ‘neither’

(Table 2). Further, the effect estimate for ‘dispersed’ males was negative, indicating that the non-significance of this predictor was unlikely to be simply the result of limited statistical power. In combination, these results suggest that the significant estimate for males in the ‘both’ category is driven by males that had attained rank (with or without dispersing) rather than by males that had dispersed, and that rank attainment but not dispersal was associated with elevated fGC. In support of this inference, a visual comparison of fGC levels shows that individuals that had attained rank (i.e., those in the

‘both’ category plus those in the ‘ranked’ category) had higher fGC levels than those that had not (i.e., those in the ‘neither’ category plus those in the ‘dispersed’ category; Fig

1A). However, a parallel visual comparison for the dispersal case showed no effect of dispersal: those that had dispersed (i.e., those in the ‘both’ category plus the ‘dispersed’ category) had fGC levels that were indistinguishable from those that had not dispersed

(Figure 1B). These comparisons support our conclusion that rank attainment status has a robust association with fGC, while dispersal status does not, and that the significant estimate for males in the ‘both’ category is driven by males that have attained rank (with

27

or without dispersing). Consistent with previous analyses, fGC levels were higher in the dry season. In addition, we found that fGC levels increased with age over the age range studied here (6-10 years) (Table 2).

Table 2: GLMM for fGC levels (fecal samples = 2963, males = 114)

Variables Estimate SE Z value p value Direction Intercept 1.635 0.0401 39.72 <0.0001 Maturation status Dispersed -0.0195 0.0233 -0.84 0.404 Ranked 0.0995 0.0174 5.71 <0.0001 High fGC - ranked Both 0.0865 0.0162 5.35 <0.0001 High fGC - both Season (wet) -0.0162 0.0072 -2.24 0.025 High fGC - dry Age 0.0143 0.0058 2.47 0.014 High fGC - older

Fecal T levels were statistically indistinguishable among the four categories of maturation status for males (Table 3). In other words, neither attaining rank nor dispersing left any lasting detectable effect on fT levels in male baboons (Fig. 1C, 1D).

Consistent with previous analyses, fT levels were higher in the wet season. In addition, we found that fT levels increased with age over the age range studied here (6-10 years).

Table 3: GLMM for fT levels (fecal samples = 2963, males = 114)

Variables Estimate SE Z value p value Direction

Intercept 1.798 0.0517 34.76 <0.0001 Maturation status Dispersed 0.0268 0.0285 0.94 0.3481 Ranked -0.0041 0.0214 -0.19 0.8486 Both 0.0237 0.012 1.19 0.2343 Season (wet) 0.0364 0.0087 4.17 <0.0001 High T - wet Age 0.0244 0.0071 3.45 <0.0001 High T - older

28

Figure 1: Average residual values of hormone concentrations as a function of age in male baboons. Residuals were calculated from a model containing season as the only predictor and male identity as a random variable. Each value represents the mean + SE or mean - SE across males of the residuals of log transformed hormone concentrations for a 6-month period. Numbers above the data points represent the numbers of males sampled for each age period. A) Comparison of residual fGC levels for males who had attained rank (i.e., males in the ‘ranked’ plus the ‘both’ category) and males who had not attained rank (males in the ‘neither’ plus the ‘dispersed’ category). B) Comparison of residual fGC levels for males who had attained dispersal (males in the ‘dispersed’ plus the ‘both’ category) and males who had not attained dispersal (males in the ‘neither’ plus the ‘ranked’ category). C) Comparison of residual fT levels for males who had and had not attained rank (categories as in panel A). D) Comparison of residual fT levels for males who had and had not attained dispersal (categories as in panel B). The dashed lines in A and C represent median age at rank attainment (7.45 years). The dashed lines in B and D represent median age at dispersal (7.67 years).

29

Categorical analysis of post-dispersal fGC and fT

The paired Wilcoxon signed ranked tests revealed no differences in either fGC levels (W=73, p=0.2166, N=14) or in fT levels (W=65, p=0.4631, N=14) in the two post dispersal categories (one month post-dispersal, versus 2-6 months post-dispersal). These results indicate that individuals do not appear to experience elevated fGC or fT levels immediately following dispersal relative to later time periods.

2.3.2 Analysis of variation in mating and aggression before and after rank attainment

Males engaged in a significantly higher number of mate-guarding episodes

(consortships) in the 6 months after adult rank attainment than they did in the 6 months before rank attainment (Wilcoxon signed-ranks test, p=0.005, W=3, N=20 males, Figure

2). Males also won significantly more agonistic encounters in the 6 months after they had attained adult rank than in the 6 months before (Wilcoxon signed-ranks test, p<0.0001, W=0, N=20 males, Figure 3A. Neither the total number of agonistic encounters

(sum of wins and losses), nor the number of losses, were significantly different in the 6 months before and the 6 months after rank attainment (p=0.169, W=60.5, Figure 3B; p=0.084, W=138, Figure 3C).

30

Figure 2: Numbers of consortships for each of 20 male baboons in the 6 months before and the 6 months after adult rank attainment.

Figure 3: Counts of agonistic encounters for each of 20 male baboons in the 6 months before and the 6 months after adult rank attainment.

31

2.4 Discussion

We asked whether two important maturational milestones in male baboons

(natal dispersal and rank attainment) predict hormone levels in post-pubescent males between 6 and 10 years. Our most striking result is that male rank attainment was associated with an increase in fGC; males that had attained an adult rank have higher fGC than males that had not yet attained an adult rank. In male baboons, adult rank attainment is an important event, because it is frequently immediately followed by a rapid rise to high dominance rank (Alberts and Altmann, 1995b; Hamilton and Bulger,

1990). Because dominance rank is a major predictor of male mating and paternity success in baboons, as in many other primate species, attainment of adult dominance rank has important fitness consequences (Alberts, 2012; Alberts et al., 2006, 2003; Bulger,

1993; Cowlishaw and Dunbar, 1991). Our results suggest that once males have attained an adult rank they experience greater energetic demands and/or psychosocial stress than they did prior to attaining this milestone. This interpretation is consistent with data on behavioral differences associated with rank attainment. Specifically, our data indicate that males win significantly more agonistic encounters and acquire significantly more reproductive opportunities after they have attained adult rank than before they have done so (Figure 2 and Figure 3A). Previous work in this population has shown that age at rank attainment strongly predicts age at first consortship (which is a measure of age at first reproduction) in male baboons, with a median lag time between these two events of

32

2.5 months; first consortship rarely or never precedes attainment of adult rank (Alberts and Altmann, 1995b and unpublished).

Contrary to our expectations, natal dispersal did not predict fGC levels. We predicted that natal dispersal, like rank attainment, would involve psychosocial stress and energetic demands that are substantially higher than before dispersal, hence that fGC levels would be higher after dispersal. We posit that the lack of an effect of natal dispersal on fGC levels reflects the fact that individuals are well adapted to face the long-term challenges associated with dispersal, and that the short-term stress associated with dispersal is relatively ephemeral and thus difficult to detect unless hormone levels are sampled within hours or a few days of dispersal (see Alberts et al., 1992; Arlet et al.,

2009; Bergman et al., 2005). In addition, in this study most individuals dispersed to adjacent social groups which share the home ranges of their natal group. This may have removed key costs of dispersal, such as the lack of familiarity with resources and the long travel distances that characterize some dispersal events (see discussions in Pusey and Packer,1987; Dufty and Belthoff 2001). Importantly, this tendency to disperse into neighboring groups is a common pattern in primates, and has been reported in olive baboons, vervet monkeys, rhesus macaques, ring-tailed lemurs and gorillas (see review in Pusey and Packer, 1987).

The lack of a systematic effect of natal dispersal on fGC levels may also reflect the fact that males’ experiences of dispersal are quite heterogeneous. For instance, some

33

males may attain adult rank very soon after dispersal, so that an increase in fGC occurs in association with high levels of aggression. Thus, the high GC and T levels reported by

Alberts et al. (1992) in a single immigrant male were associated with unusually high levels of aggression directed by that male to other group members (see also Bergman et al., 2005). Other males, in contrast, may spend many months avoiding direct challenges to resident males. In support of the idea that males’ experiences upon immigration may be quite heterogeneous, Van Noordwijk and Van Schaik (1985) report two different types of immigrant males in long-tailed macaques: unobtrusive immigrants enter a new group low in the dominance hierarchy and form associations with females over a period of months before challenging high ranking males, while bluff immigrants attempt to take over the highest dominance rank immediately upon entry. Demographic factors in the new group, such as the relative numbers of adult males and adult females, may also be variable, and this variability will in turn influence aggressive and reproductive encounters. Thus, the variable nature of the dispersal experience – how far males travel, what demographic conditions they encounter in the new group, and how much they engage in aggression and reproduction after dispersal – may explain the absence of a clear pattern of change in hormone levels in post-dispersers compared to pre-dispersers.

Fecal T levels did not change systematically with either rank attainment or natal dispersal. Instead, our results show that fT levels increase steadily with age between ages 6 and 10 years, indicating that fT levels were driven more by age than by either of

34

the maturational milestones per se. Although we cannot rule out the possibility that our data set lacked the power to detect a subtle change in fT in response to these maturational milestones, we can say with confidence that the effects of attaining these milestones were certainly weaker than the effects of age or season. Male baboons experience a post-pubertal growth spurt between 6 and 8 years of age, during which they nearly double in body mass (Alberts and Altmann, 1995b; Altmann et al., 2010); this growth spurt is likely to be associated with the anabolic functions of T (Nelson, 2011).

Testosterone is an important predictor of growth during and after puberty in several primate species (chacma baboons and mangabeys: Bernstein et al., 2008; chimpanzees:

Copeland et al., 1985; white-faced capuchins: Jack et al., 2014). In our population as in other primates, fT is upregulated at the onset of reproductive maturity, and begins to decline with age starting at around 9.5 years of age (Altmann et al., 2010; Beehner et al.,

2009). Beehner et al. (2009) examined overall fT patterns and maturational milestones in both chacma and yellow baboons and found that male baboons generally attain adult rank after they achieve fT levels in the adult range, but before they achieve peak fT levels. This means that fT levels are already increasing in male baboons before they attain adult rank. Our results build on the study by Beehner et al. (2009) by demonstrating that after controlling for other predictors of fT levels, we did not find any effect of either maturational milestone on fT.

35

In conclusion, our results indicate that male baboons experience marked increases in fGC levels after rank attainment but not after dispersal, and that they experience no marked or persistent changes in fT levels associated with either adult rank attainment or natal dispersal. With respect to testosterone, our data combined with previous results in this population reveal that males experience a steady increase in fT beginning between 4 and 5 years of age, coincident with the onset of testicular enlargement, which continues throughout the subadult and early adult period, peaks around 9.5 years of age, and then begins to decline with age (Altmann et al., 2010;

Beehner et al., 2009; Gesquiere et al., 2005; Onyango et al., 2013). Fecal GC levels, too, increase with age across the age range studied here, but the attainment of adult dominance rank signals a marked change in these steroid hormones. In addition, our behavioral data indicate that the total numbers of both reproductive opportunities and agonistic wins that males experience are significantly higher after adult rank attainment than before. Increased mating opportunities will generally produce increased energetic demands, but an increased number of agonistic wins will presumably not produce increased psychosocial stress, especially if the total number of agonistic encounters remains unchanged. Thus, our results suggest that the elevation in fGC after adult rank attainment results from increased energetic demands on newly adult males rather than increased psychosocial stress.

36

Additional research will be needed to elucidate more fine-grained or short-lived effects of dispersal and adult rank attainment on steroid hormones in this as well as in other group-living species. For instance, fine-grained sampling has shown that male baboons that engage in more frequent mating exhibit higher T levels than those that engage in less mating (Onyango et al. 2013a). Similarly, fine-grained analyses of behavior with densely sampled hormone concentrations (i.e. daily or near-daily fecal samples) could potentially identify variation in fGC or fT associated with inter- individual differences in the attainment of maturation milestones. For example, individuals that travel longer distances should experience more energetic and psychosocial stress (and have higher GC levels) than those who disperse to neighboring areas, and individuals that experience more mating after rank attainment should experience higher GC levels than those who experience less. Near-daily repeated sampling of individuals during and immediately after the dispersal phase will be needed to reveal such inter-individual differences.

37

3. Environmental and host-specific predictors of helminth burden in female baboons.

3.1 Introduction

Parasites can impose important costs on mammalian hosts by affecting reproductive and mortality rates, and by influencing susceptibility to other pathogens

(Bethony et al., 2006; Ezenwa, 2016; Hudson and Dobson, 1995; Sheldon and Verhulst,

1996). Through these effects, parasites can also indirectly alter host population structure

(Nunn et al., 2004; Tompkins et al., 2011). Understanding the predictors of individual parasite risk is therefore important for both basic and applied purposes. Investigating the patterns and drivers of individual variation in parasite infection risk can help uncover the adaptive value of behavioral strategies and social structures, and can point to selection pressures and constraints that have shaped mammalian behavior and life history. At the same time, understanding patterns of parasitism can identify which individuals are most at risk and which are most likely to be sources of infection, informing management strategies in vulnerable populations.

In social mammals, including primates, three broad factors contribute to variation in parasitism: ecological and social environments, host traits (e.g., host health, behavior, sex, life history, immune status, genetics), and parasite traits (e.g., mode of transmission, lifespan). In primates, many studies have examined the effects of one or a

38

few of these factors on helminth infection, with mixed empirical results (see review in

Nunn and Altizer, 2006; Table S1). However, relatively few studies have evaluated more than two or three potential factors simultaneously because it is rare to have information on multiple predictors of parasitism in the same population (Table S1). Here we test several environmental and host predictors to gain an exceptionally complete understanding of the forces that explain inter-individual differences in helminth infection risk among females in a well-studied wild primate population.

3.1.1 Environmental predictors of helminth burden: Social and ecological environments

3.1.1.1 Group size

Group living can influence both host susceptibility and exposure to parasites. In terms of susceptibility, groupmates often compete for limited resources, which may lead to compromised host nutrition, and hence compromised immune function, among group members (Cote and Poulin, 1995; Kappeler et al., 2015; Koski and Scott, 2001). In terms of exposure, groupmates are important sources of parasite transmission, both through direct contact with infected hosts and because high population densities can increase the chance of contact with contaminated environmental substrates (Freeland,

2009; Nunn and Altizer, 2006). In support, meta-analyses in several taxa indicate that individuals in large social groups tend to carry higher burdens of both contagious parasites (e.g. nematodes) and vector-transmitted parasites than those in small groups,

39

and parasites occur at higher prevalence in large social groups compared to small groups (Cote and Poulin, 1995; Patterson and Ruckstuhl, 2013; Rifkin et al., 2012). In primates, positive correlations have generally been reported between helminth burden and group size with a few notable exceptions (Table S1).

3.1.1.2 Social status

Social status (dominance rank) may also influence individual parasite risk via mechanisms that link social status to exposure or susceptibility (Abbott et al., 2003;

Kappeler et al., 2015; Stuart and Strier, 1995). For example, increased access to food among high-ranking individuals may lead to increased intake of helminth infectious stages present in the soil. In addition, increased social contact through interactions such as aggression or grooming could also facilitate transmission of helminth infective stages

(eggs and larvae), thus increasing exposure (Mackintosh et al. 2012). Despite the potential for increased exposure, high ranking animals are often in better body condition and hence may be capable of mounting more effective immune responses to parasites than lower-ranking individuals (Nunn and Altizer, 2006). For instance, low ranking individuals may suffer more than high-ranking ones from malnutrition and chronic stress, both of which influence their susceptibility to infection (Sapolsky, 2004; Sapolsky et al., 2000). In contrast to these predictions, most studies in female primates fail to show any association between helminths and social status (Tables S1 and S2).

40

3.1.1.3 Rainfall and temperature

The survival and development of helminth infectious stages (eggs and larva) depends upon environmental factors such as temperature, humidity, and rainfall, with different helminth species (and/or their vectors) requiring different conditions (Pietrock and Marcogliese 2003; Nunn and Altizer, 2006). Most helminth species that infect primates will thrive in relatively warm and wet environments (Nunn and Altizer, 2006).

Adequate rainfall and warm temperatures may promote host immune competence if they are associated with high food availability and good nutrition for hosts (Coop and

Holmes, 1996; Coop and Kyriazakis, 1999). Hence, despite the increased presence of infective parasite stages during rainy seasons, hosts may be able to limit helminth burdens at these times because of enhanced immune systems (Dowell, 2001; Hawley and

Altizer, 2011; Martin et al., 2008). In contrast, harsh weather conditions (e.g., drought, excessive cold or heat) may limit parasites but may also be associated with increased host stress, and thus with increased disease susceptibility and reduced recovery rates

(Altizer et al., 2006; Mignatti et al., 2016). Drought conditions may lead to increased contamination of food and water resources if hosts aggregate around specific resources at such times, hence increasing exposure rate (Nunn and Altizer, 2006; Nunn et al. 2014).

In primates, most studies report increased helminth burdens in the wet season or during high rainfall (Table S1 and S2). The effects of ambient temperature have been much less

41

studied than the effects of rainfall, with positive trends reported by 5 studies (Tables S1 and S2).

3.1.2. Host traits as predictors of helminth burdens

3.1.2.1 Host age

As hosts age, two processes will affect infection rates: cumulative parasite encounter rates will increase, and age-related changes in immune regulation will occur

(Hudson and Dobson, 1995; Nunn and Altizer, 2006). Many helminths may live for several years (see Table 5), hence as animals age they may show increased burdens with age simply through accumulation (Wilson et al., 2002). Older animals may also experience down-regulation of important immune functions such as T-helper2 immunity, which is crucial for fighting helminth infections. Hence, positive correlations between age and helminth burdens may be driven by immune senescence (Wilson et al.,

2002, Hudson and Dobson, 1995; Nunn and Altizer, 2006). Conversely, helminth infections may decline with age because of acquired immunity against recurring parasites, such that older animals are able to clear infections more effectively than younger animals (MacIntosh et al., 2010; Wilson et al., 2002). Most prior studies that correlate host age and parasite risk in primates use age classes rather than individual ages, limiting the ability to detect senescence. Perhaps because of differences in how age is reported, the effects of age on helminth burdens are among the most variable in the literature, with most studies reporting increased burdens in older individuals or age

42

classes, five studies reporting the reverse, two studies reporting curvilinear relationships, and five reporting no relationship (Tables S1 and S2).

3.1.2.2 Reproductive state

The reproductive cycle of female mammals can be broadly categorized into three states: cycling, pregnancy, and lactation. These three states may have different relationships to parasite risk. Pregnancy and lactation, in particular, impose high energetic demands (Beasley et al., 2010; Coop and Kyriazakis, 1999; Houdijk et al., 2001;

Sakkas, 2011), and if cycling and lactating females have difficulty meeting these demands, they may be unable to mount effective immune responses against parasites

(Koski et al., 1999; Pedersen et al., 2002). The energetic demands of pregnancy are also thought to play a role in periparturient rise (PPR) — an increase in nematode egg counts near the end of pregnancy — in ewes and other domesticated ungulates (Beasley et al.,

2012, 2010; Cattadori et al., 2005; Houdijk et al., 2001). The mechanism responsible for

PPR may be increased foraging in late pregnancy in response to increasing energetic demands, with an associated increase in the risk of soil-transmitted nematodes

(Hutchings et al., 2000). In addition to high energetic demands, reproductive hormones associated with pregnancy and lactation (e.g. prolactin, progesterone, glucocorticoids) may also influence individual parasite risk via immunosuppressive effects (Klein, 2004,

2000). For example, prolactin has been positively correlated with increased nematode egg counts in ruminants (Barger, 1993; Houdijk et al., 2001; Rahman and Collins, 1992).

43

The effects of reproductive state on helminth burdens has been understudied in natural primate populations; with a few studies reporting an increase in helminth burdens during pregnancy and lactation relative to cycling (Tables S1 and S2).

3.1.3 Role of glucocorticoids in susceptibility to helminth infections

Steroid hormones such as glucocorticoids (GC) suppress antibody production and dampen inflammatory responses. Specifically, glucocorticoids depress the Th2 cytokine responses that are responsible for clearing helminth infections (see reviews in

Klein, 2004, 2000; Nava-Castro et al., 2011). Thus, if GC concentrations are elevated for extended periods, these high concentrations may result in increased susceptibility to infections (Bercovitch and Ziegler, 2002; Sapolsky, 2004). At the same time, host GC concentrations may rise in response to the energetic stress imposed by parasite depletion of host resources (Coop and Kyriazakis, 1999). Accordingly, elevated GC concentrations may either contribute to helminth infection or be a consequence of helminth infections, or both. Five studies have investigated the relationship between GC and helminth burden in wild female primates; two report a positive correlation and three report no correlation

(Tables S1 and S2).

3.1.4 Aims of this study

Two previous studies tested predictors of parasite infection risk in the Amboseli baboon population (Hausfater and Watson, 1976; Meade, 1983; Table S1). Both studies had small sample sizes, focused on only one social group, and employed bivariate

44

analyses that did not simultaneously account for multiple predictors of helminth infections (e.g., rainfall, age class, temperature, rank). We built on these previous studies by using a larger data set, examining multiple baboon social groups, carrying out a fine- grained analysis of individual age rather than age class, and employing a multivariate model that examined the effects of multiple individual and environmental variables. We also examined the relationship between GCs and parasite burdens for the first time in this population. Our analyses were guided by the results of previous studies in wild primate populations, described in Table 4. A holistic understanding of the factors that contribute to parasite risk in wild animal populations is important for elucidating both population process such as what drives disease transmission, as well as health and fitness consequences for individuals.

45

Table 4: Environment and host-trait predictors of helminth burden

Factor Immune susceptibility-driven Exposure rate-driven hypothesis Most frequently Results from this hypothesis report relationship study by helminth (Table S1) species

Group size  group size =  worms. Hosts  group size =  worms. Higher  group size=   group size=  worms in bigger groups have limited exposure rates in larger groups worms food resources and thus have because of social interactions and (T. trichiura, A. poorer body condition and increased fecal contamination. caucasica) compromised immune responses.

Social Status High rank =  worms. High High rank =  worms. High No association No association ranking hosts are in better ranking hosts have priority of body condition, resulting in access to food and thus increased better immune responses. exposure rates.

46

Rainfall  rain =  worms. Extremes of  rain =  worms. Extremes of  rain =  worms  rain =worms rainfall are associated with rainfall are not conducive to the nutritional stress (e.g., food survival of infectious stages: (T. trichiura, A. sources are limited in dry extremely high rainfall will caucasica) conditions and temporarily result in washing away of inundated during heavy rain), parasite infective stages, while compromising host immune extremely low rainfall results in responses. Hosts are in best a habitat too dry for infective condition with intermediate stages to thrive. Intermediate levels of rainfall that support levels of rainfall provide optimal growth of vegetation. sufficient rainfall for infective stages and allow for the continued existence of contamination in food and water.

Maximum  temp =  worms. Extremes  temp =  worms. Extremes of  temp =  worms  temp =worms temperature of temperature are associated temperature are not conducive to with thermal and nutritional the survival of infectious stages. (T. trichiura, stress, compromising host Strongyles, A. immune responses. caucasica, Parasite richness)

Age  age =  worms. Older hosts  age =  worms. Older hosts  age =  worms  age =  worms experience senescent declines have greater cumulative in immune responses. exposure. (T. trichiura, Strongyles, Parasite  age =  worms. Older hosts richness) can fight off infections because of acquired immunity.

Rep State  worms during lactation and  worms during lactation and Pregnant and Pregnant and lactating (pregnant, pregnancy. Increased energetic pregnancy. Increased energetic lactating individuals =  worms lactating, costs and reproductive demands lead to increased food individuals =  (A. caucasica, S.

47 cycling) hormones lead to intake and thus increased worms pigmentatus, Parasite

compromised immune exposure. richness) responses or facilitate parasite growth in the host.

3.2 Methods

3.2.1 Study population and site

Study subjects were adult female members of a natural population of baboons

0 0 (Papio cynocephalus) in the Amboseli basin in southern Kenya (2 40’S, 37 15’E, 1100m)

(Gesquiere et al., 2011). This population has been the subject of long-term, individual- based monitoring since 1971 (Alberts and Altmann, 2012). Systematic data collection on all study subjects includes life history events (births, maturation, migration, and death), behavioral observations, as well as measurements of the abiotic environment, including rainfall and temperature (Charpentier et al., 2008). We examined variation in helminth burdens in 122 distinct adult females (those that had attained menarche), who lived in several known social groups between March 2011 and September 2015. Female baboons remain in their natal group throughout their lives; thus, they are monitored from birth, and ages are known to within a few days; the age range of our study subjects was 4.25 years to 25.66 years.

The Amboseli ecosystem is comprised of a semi-arid short-grass savannah habitat (Alberts et al., 2005; Gesquiere et al., 2011, 2008). We maintain daily records of rainfall and minimum and maximum temperature. Rainfall is highly variable between and within years, and ranges between 150-550mm per year, with a mean of 348mm/year

(Alberts et al., 2005; Altmann et al., 2002). Average maximum temperatures observed each month ranged between 29.8C and 35.4C.

48

3.2.2 Fecal sample collection and parasitology analysis

A total of 746 fresh fecal samples were collected opportunistically for both parasite and hormone analysis from our 122 subjects between March 2011 and

September 2015. Most individuals in our data set were represented by multiple fecal samples (mean of 8 samples per individual, range 1-25). Each fecal sample was homogenized then split into two portions: the portion for helminth analyses was stored in 10% buffered formalin, and the portion for hormone analyses was stored in 95% ethanol. Small fecal samples that could not be split were preserved only for helminth analysis, hence hormone analyses were unavailable for 37 of our 746 samples.

Helminth egg counts were carried out using sedimentation and floatation techniques adapted from Gillespie (2006). Prior to analysis, each sample was thoroughly homogenized to ensure uniform distribution of parasite eggs in the sample. To perform fecal flotation, we suspended 4 grams of sample in water in a 15ml centrifuge tube. The tube was centrifuged at 1500 rpm for 10 minutes and the supernatant was discarded before re-suspending the sample in saturated sugar solution (specific gravity ~1.27) until the sugar solution formed a bulging meniscus at the top of the tube. A coverslip was placed on the meniscus, and the tube was centrifuged at 1500 rpm for 10 minutes to collect parasite stages under the coverslip. The coverslip was placed on a microscope slide, and the procedure was repeated with the same sample using a second coverslip.

49

Both coverslips were examined systematically at 10X magnification to count all visible parasite egg morphotypes (Appendix A).

The sedimentation protocol was similar to the first steps of the floatation protocol. However, after the first centrifugation, the supernatant was discarded and the sample was reconstituted in a mixture of 750ul of water and 500ul of saturated sugar solution. Five drops of sediment were examined for parasite eggs by placing each drop under a coverslip and examining the coverslip at 10X magnification. Parasite identifications were based on parasite egg, larvae, cyst coloration, shape, contents and size (Gillespie, 2006). Floatation data were used to compute egg counts for T. trichiura, whereas both floatation and sedimentation data were used to measure presence/absence of all helminth species.

3.2.3 Helminth species and life cycles

Multiple genera of nematodes, trematodes and cestodes have been found to infect baboons (Hahn et al., 2003; Muriuki et al., 1998; Murray et al., 2000; Ryan et al.,

2012). Helminth species that have been previously reported in our study population and that we observed here include Trichuris trichiura (also known as whipworm), two strongyle species (genetic identifications indicate that these strongyles are

Oesophagostomum bifurcum and Trichostrongylus collubriformis; Archie, unpublished data),

Abbreviata caucasica, Streptopharagus pigmentatus, Enterobius vermicularis (also known as

50

pinworm), and Strongyloides fulleborni. Table 5 presents a summary of these species’ life histories.

Table 5: Helminths of interest, life cycles and modes of transmission

Helminth Life Mode of Transmission Time to egg Adult lifespan cycle (infective stages) shedding post infection

Trichuris trichiura Direct Ingestion of food or water 8-10 weeks 1-2 years contaminated with infective embryonated eggs.

Strongyles Direct Ingestion of infective 2-4 weeks 8-15 years (for third larvae stage. strongyles known to infect humans)

Strongyloides Direct Penetration through the 2-3 weeks 15 days fulleborni skin by infective larvae, ingestion or transmammary.

Enterobius Direct Ingestion of food or water 4 weeks 2 months vermicularis contaminated with infective embryonated eggs

Abbreviata Indirect Ingestion of cockroaches Unknown Unknown caucasica

Streptopharagus Indirect Ingestion of dung beetles Unknown Unknown pigmentatus

References: (Anderson, 2000; Audebert, Hoste, & Durette-Desset, 2002; Bethony et al., 2006; CDC., 2016; Cogswell, 2007; Crompton & Nesheim, 2002; Gems, 2000; Machida et al., 1978; Meade, 1983; Stephenson, Holland, & Cooper, 2000).

51

3.2.4 Hormone assays

Fecal samples were collected and stored in 95% ethanol and then placed in an evaporating cooling structure in our field site before they were transported to the laboratory. The samples were then freeze dried, sifted to remove vegetative matter, and weighed. The weighed fecal powder (0.2g) underwent extraction in 90% methanol followed by solid phase extraction using Waters Oasis HLB cartridges. Fecal GC (fGC) metabolites in the samples were then measured in nanograms per gram of dried feces

(ng/g) by radio-immunoassays (RIA), following a well-established protocol (e.g.

Gesquiere et al., 2011, 2008; Markham et al., 2015; Onyango et al., 2008). We used the

Corticosterone kit for Rats and Mice (ICN diagnostics, Costa Mesa, CA) which has previously been validated for our population (Gesquiere et al., 2005; Khan et al., 2002;

Lynch et al., 2003).

3.3 Data analysis

We structured our data analysis around two hypotheses, one concerning the effects of environmental factors and host traits on helminth burdens, and the second concerning the relationship between helminth burdens and fecal glucocorticoid (fGC) concentrations.

52

3.3.1 Hypothesis 1: Immune susceptibility and exposure drive helminth burdens through environmental and host-specific factors

We used three measures of helminth burden as our response variables. 1) We measured helminth abundance in each fecal sample (log number of eggs per gram of feces) for Trichuris trichiura, which was found in 92.5% of all the fecal samples. Parasite egg counts in fecal samples are commonly used as a measure of parasite infection intensity because they are a reliable indicator of the number and fecundity of adult worms in both domestic and wild animals (Cabaret and Gasnier, 1998; Pedersen and Saeed, 2000;

Seivwright et al., 2004; Stear et al., 1998, 1995). Because fecal egg counts were not normally distributed, we log-transformed our data (log base10 + 1) for analyses of T. trichiura. 2) We measured presence or absence of infection in each fecal sample in order to analyze the probability of infection via logistic regression, for Abbreviata caucasica,

Streptopharagus pigmentatus, and members of the family Strongylidae (i.e., strongyles).

Note that strongyles likely comprise multiple parasite species because species identity cannot be established from strongyle egg morphology. 3) Parasite richness was calculated as the number of different helminth species in each fecal sample. We used these measures of helminth burden to calculate the proportions of infected samples and hosts during the study period.

We built generalized linear mixed models (GLMMs) to analyze predictors of the most commonly observed helminths (Table 6); (i) parasite egg abundance (for T. trichiura), (ii) probability of infection (for strongyles, A. caucasica, and S. pigmentatus),

53

and (iii) parasite richness. We used a Gaussian error distribution for the T. trichiura abundance and the parasite richness analyses. For the analysis of probability of infection, we used a binomial error structure. Random and fixed effects of our GLMMs are described in Table 7.

Table 6: Proportion of infected samples and infected individuals (prevalence)

Macroparasite Percentage* of Percentage* of infected samples infected females (n=746) (n=122) Trichuris trichiura 92.5% (690) 96.7% (118) Strongyles 61.5% (459) 89% (103) Strongyloides fullerborni 1.9% (14) 10.7% (13) Enterobius vermicularis 6.6% (49) 22.1 % (27) Abbreviata caucasica 49.7% (371) 82% (100) Streptopharagus pigmentatus 27.7% (207) 63.1% (77) *Percentages were calculated by dividing the number of infected sample by the total number of samples collected or dividing the number of infected females by the total number of females sampled during the study period. Numbers in brackets represent the number of infected samples or the number of females associated with each helminth.

To test whether fecal samples collected during wetter periods had higher worm burdens than those collected during drier periods, we calculated the total amount of rainfall (in mm) that had fallen in the three months prior to the collection of a given fecal sample. To examine the effects of maximum temperature we calculated the average maximum daily temperature in the three months prior to the collection of the fecal sample; the minimum and maximum temperatures were positively correlated so we only included maximum temperatures in our models (r=0.8, p<0.0001). We chose a three month period because we speculated that it would be a good proxy for both food

54

availability for hosts as well as conducive environments for helminth infectious stages.

The 3-month metric was replaced by 1-month and 2-week measures of cumulative rainfall in subsequent analyses to test for the best rainfall metric that predicted parasite burdens. All metrics produced similar patterns of infections but the best model, based on Akaike Information Criterion (AIC) values, used the 3-month metric (Table S5). We also calculated an alternative measure, dubbed ‘number of dry months,’ that corresponded to the number of months (30-day periods) that had passed since the end of the last “wet” 30-day period; this rainfall measure, too, yielded very similar results to the 3 month total rainfall measure (see Tables S4 for the ‘number of dry months’ analyses).

We measured social status by calculating ordinal ranks based on dyadic agonistic encounters recorded during observations; ranks are assigned monthly to each individual in each study group (Hausfater, 1975). The highest ordinal rank status is 1 and the lowest rank status depends on the number of adult females in each social group. We also calculated proportional ranks (which approximates to the proportion of females that a subject dominates); proportional rank behaved nearly identically in our model to ordinal rank (Table S3).

55

Table 7: Description of predictors of helminth burden

Predictor Description a) Random effects

Individual ID The identity of the subject who contributed the sample.

Counter ID The identity of the person who counted parasites. b) Fixed effects

Group size Total number of all individuals in the group on the day the sample was collected.

Social status Ordinal rank of the subject who contributed the sample.

Rainfall Total amount of rainfall (in mm) in the three months before the sample was collected.

Temperature Average daily maximum temperature in degrees C during the three months before the sample was collected.

Age Age in years of subject when the sample was collected, known to within a few days for most subjects.

Reproductive state Whether the female was cycling, lactating or pregnant at sample collection.

3.3.2 Hypothesis 2: Elevated GCs are associated with suppressed immune function, hence individuals with higher fGC concentrations will have higher worm burdens.

Our dataset for this analysis included measures of fecal parasite burden and fGC levels from 709 fecal samples collected from 122 unique females. We have previously documented that fGC concentrations in female Amboseli baboons vary with a number of the predictors in our helminth models: fGC concentrations are higher in the dry

56

season, higher with higher ambient temperature, higher when females live in extremely large or small groups, and higher in pregnant than in lactating or cycling females

(Gesquiere et al., 2008; Markham et al., 2015). To deal with these features of GC concentrations – (i) their dual role as both a potential predictor of and a potential response to helminth burdens, and (ii) their known responsiveness to other predictors in our helminth models – we constructed a separate statistical model investigating the bivariate relationship between fGC concentrations and helminth burdens. We reasoned that this analysis would reveal whether fGCs alter susceptibility to helminth infections.

To test our predictions, we ran five general linear mixed models. The dependent variables were parasite abundance (in the model for T. trichiura), probability of infection

(in the models for strongyles, A. caucasica, and S. pigmentatus), and parasite richness. In each analysis, the independent variable was fGC concentration, and we had two random factors: the individual female ID and the ID of the person who counted the parasites.

To generate our mixed effect models, we used the glmmadmb package (Fournier et al., 2012) in R Statistical software, version 3.2.3 (R Foundation for Statistical

Computing, 2014) . Visual inspections using Q-Q plots confirmed that of residuals from our models approached normality. We tested for multicollinearity for model predictors using the ‘CAR’ package in R and we found no correlation between predictors (Variance

Inflation Factors were <2). Model fit was assessed using chi-square tests on the log- likelihood values to compare the fit of the best model (full model with predictors) to the

57

null model (model with random variables only). For all analyses significance was achieved at 0.05 p-values.

3.4 Results

3.4.1 General patterns of parasite infection

We identified six helminth species, which occurred at varying rates in our 746 samples and 122 females sampled. T. trichiura was the most common of all the species and was present in 92.5% of fecal samples and 96.7% of sampled females (Table 6). Three helminth species were fairly common and were observed in the individual samples and females as follows: Strongyles (samples: 61.5%, females: 89%), A. caucasica (samples:

49.7%, females: 82%) and S. pigmentatus (samples: 27.7%, females: 63.1%). Two species of helminths were relatively rare; E. vermicularis (samples: 6.6%, females: 22.1%) and S. fulleborni (samples: 1.9%, females: 10.7%). The low prevalence of these two species is probably linked to the short lifespan of adult female worms. Parasite richness ranged from 0-5 helminth species per individual sample, with a median of 2 species. Ninety- seven percent of samples harbored more than one helminth species (Figure 4). Two out of 122 females did not harbor any helminths.

58

Figure 4: Parasite richness distribution across individual samples

Number of individuals contributing to each number of helminth species reported on top of each bar.

3.4.2 Environmental predictors of helminth burden: Social and ecological environments

Among the social and ecological variables we tested, the size of a host’s social group and the climate the host had experienced in the previous three months were most important in predicting patterns of parasite infection. We found that hosts living in larger social groups exhibited higher T. trichiura egg counts (Table 8). Our analysis also revealed a marginally significant effect of group size on the probability of A. caucasica infection: hosts living in larger groups had a higher probability of infection compared to hosts in smaller social groups (Table 8). In terms of climate, and contrary to the most common results in the literature, hosts in low-rainfall periods exhibited higher T.

59

trichiura egg counts and were more likely to be infected by A. caucasica (Table 8).

Patterns of infection were also predicted by temperature; hosts exhibited lower T. trichiura egg counts, were less likely to infected by strongyles, and harbored fewer parasites during cool periods (Table 8). In addition, the probability of infection by A. caucasica was lower during cool periods, though this effect was marginally significant

(Table 8). Host dominance rank did not predict any measure of helminth burden and none of the environmental predictors was associated with the probability of acquiring S. pigmentatus infection (Tables 5 and S3).

3.4.3 Host-specific predictors of helminth burden

Among host-specific predictors, both female age and reproductive state explained patterns of helminth infection. In support of the idea that exposure and immuno-senescence increase helminth infection risk, older females exhibited higher T. trichiura egg counts and were more likely to be infected with strongyles, and hence often were infected with a greater diversity of parasites than younger females (Table 8).

Because the literature points to the possibility that this heightened vulnerability of older animals to helminth infection is mitigated in favorable environments (Hammalainen et al. 2015a), we re-ran all our models to include an age by rainfall interaction.

Interestingly, we observed a significant interaction effect between host age and rainfall for patterns of T. trichiura infection (Figure 5; n=746, estimate =-0.0001, SE=0.000, Z=-2.24, p<0.025). Specifically, infection patterns in the youngest females (age 4 to 9 years) were

60

less sensitive to variation in rainfall than females older than 9 years, suggesting that old females experience especially high parasite risk during dry periods. In terms of reproductive state, pregnant females were more likely to be infected with A. caucasica and S. pigmentatus and exhibited higher parasite richness than either cycling or lactating females (Table 8). Reproductive state was not a significant predictor for T. trichiura or strongyle burdens.

3.4.4 Association between Glucocorticoids and worm burden

Contrary to our predictions, we found no associations between fGC levels and any measure of parasite burden in our models (Table S6).

Figure 5: Effect of rainfall by age interaction on log Trichuris abundance.

The data points represent residual values of log Trichuris abundance for each fecal sample as a function of rainfall. For the purposes of visualizing the interaction, age class was used instead of age in years. Residuals were calculated from a model containing two random variables and all the predictors except rainfall and age.

61

Table 8: Environmental and host predictors of helminth burden (n=746, females=122)

62

3.5 Discussion

The forces driving inter-host variation in disease risk are complex, and include several environmental and host traits that shape both host exposure and susceptibility.

Understanding the relative importance of these predictors in wild primate populations is important, both for learning which selection pressures and constraints shape primate behavior and life histories, and for primate conservation and management. Parallel to prior studies, we found that female baboons in Amboseli were infected with diverse helminth taxa, including T.trichiura, A. caucasica, stongyle parasites, S.pigmentatus, E. vermicularis, and S. fulleborni (see also Hahn et al., 2003; Hausfater and Watson, 1976;

Meade, 1983). Inter-host differences in parasite infection were predicted by both environmental and host-related factors, including social group size, rainfall, temperature, host age, and reproductive state (Tables 5 and 6). These factors reflected underlying differences in both host exposure and susceptibility (Table 4). We are yet to determine whether the helminths we found have detrimental health effects in this baboon population. However in humans and other primates, heavy helminth burdens can lead to poor body condition, diarrhea, dysentery, anemia, ulcerations, intestinal obstruction, rectal prolapse and in extreme cases death (Bethony et al., 2006; Crompton and Nesheim, 2002; Kaur and Singh, 2009). Further, T. trichiura, S. fulleborni, and E. vermicularis species are of zoonotic importance and can be easily be transmitted to human populations that share watering points with the baboons in this study area (Kaur

63

and Singh, 2009; Muriuki et al., 1998; Stephenson et al., 2000). The nature and extent of detrimental effects on hosts are likely to depend on both environment and host factors that contribute to the intensity of exposure and the ability of the host to mount an effective immune response.

3.5.1. Environmental predictors of helminth burden: Social and ecological environments

3.5.1.1 Group size

We found that social group size, but not host social status, was an important predictor of helminth infection, as females in larger groups exhibited higher T. trichiura egg counts and were more likely to be infected with A.caucasica compared to females living in smaller groups. Our results are consistent with results from three meta- analyses that reported strong support for the idea that individual disease risk frequently increases with group size (Cote and Poulin, 1995; Patterson and Ruckstuhl, 2013; Rifkin et al., 2012). Positive relationships between group size and parasitism can be attributed to both exposure-mediated and susceptibility-mediated processes (Table 4). In terms of exposure, T. trichiura parasites are transmitted through fecal contamination in the environment; members of larger social groups live at higher experienced social densities and may encounter fecal contamination from group members more frequently than baboons living in smaller social groups. This phenomenon may explain the higher

Trichuris burdens among females in larger groups, but is puzzling with respect to the

64

higher infection risk from A. caucasica, which is transmitted by a vector. This result, though marginally significant, is surprising because A. caucasica has an indirect life cycle

(Table 5); hence, high infection risk cannot be attributed to increased exposure through contamination. We speculate that host susceptibility may be the key mechanism driving infection probability by A. caucasica.

Group size varied considerably in our study, with the smallest group having 13 individuals and the largest one 119 individuals. In terms of susceptibility-mediated processes, it is plausible that individuals in larger groups experienced much more intense competition for access to resources compared to those in smaller ones.

Competition for resources results in nutrition and social stress and can presumably increase host susceptibility to parasites if it results in compromised immune responses.

Amboseli females living in large groups spend a greater proportion of their time feeding and have longer inter-birth intervals than females in small groups, suggesting that there are nutritional costs associated with large group size (Altmann and Alberts, 2003;

Markham et al., 2015). Such costs may explain the effects of group size for both Trichuris and A. caucasica. More information on host-vector relationships —particularly host foraging behavior on cockroaches, the vectors implicated in transmission of A. caucasica, and on other potential vectors of this helminth— will be helpful in explaining these trends.

65

3.5.1.2 Social status

Like most studies in the literature, we found no association between female social status and female parasite risk. This finding is also consistent with previous work done in this population by Hausfater and Watson (1976) and Meade (1983), who reported no association between ova emissions (for both T. trichiura and Trichostrongylus sp.) and social status in females. In contrast, male primates frequently exhibit status- related differences in parasite risk, including in our own population (Clough et al., 2010;

Hausfater and Watson, 1976; Muehlenbein, 2006; Muehlenbein and Watts 2010; Arlet et al. 2015). In males, status-related differences in parasitism have been attributed to high stress in low-status animals (Sapolsky, 2004) and tradeoffs between male reproductive effort and immune responses (Martin et al., 2006; Sheldon and Verhulst, 1996). In female baboons, these mechanisms may operate differently because while male primates often compete for access to females during frequent, acute bouts, females compete for food resources in a protracted and non-acute manner.

3.5.1.3 Rainfall and temperature

We found that both rainfall and temperature predicted parasite burdens. Further, the relationships we observed for rainfall were more consistent with susceptibility rather than exposure-mediated processes (Table 4). Contrary to most prior studies (Table S1),

Amboseli baboons exhibited higher T. trichiura egg counts, greater A. caucasica infection risk, and higher parasite richness in dry periods as compared to rainy months. In most 66

nematodes, rain and high humidity enhance the survival of parasite infectious stages in the environment; hence, parasite exposure should be higher during wet periods as compared to dry periods (Altizer et al., 2006). However, rainfall directly affects food availability, which in turn affects host nutrition, downstream host immune response, and the host’s ability to mount effective immune responses against helminths (Dowell,

2001; Hawley and Altizer, 2011; Koski and Scott, 2001; Martin et al., 2008). During the dry season, Amboseli baboons feed on foods that require extensive processing for relatively small nutritional rewards, such as grass corms, whereas in the wet season they feed on more diverse and easily harvested foods, including young leaves, grass blades, grass seeds, fruits, and the leaves of shrubs (Alberts et al., 2005; Altmann, 1998; Post,

1981). In addition, during dry periods, baboons spend more time foraging and searching for food and water, thus increasing their energetic expenditure relative to periods with adequate rainfall (Alberts et al., 2005; Gesquiere et al., 2008). Thus, the dry season is associated with greater nutritional stress, probably resulting in poorer health, which in turn could contribute to higher helminth burdens (Bethony et al., 2006; Crompton and

Nesheim, 2002; Ezenwa, 2004; Kaur and Singh, 2009). In addition, negative correlations between rainfall and helminth burden could to some extent be accounted for by the

“wash away effect” whereby heavy rainfall washes away parasite infectious stages, thus decreasing host exposure (Meade, 1983; Poirotte et al., 2016).

67

Our results in regards to ambient temperature support the exposure-mediated mechanism more strongly than the susceptibility-mediated mechanism. Specifically, we found that parasite burdens were more likely to be high when animals were experiencing moderate daily maximum temperatures (i.e., at the low end of the range in our study, 29.8 C) than when the average maximum temperature was at the high end of our range (i.e., 35.4 C ). This is contrary to the idea that immune susceptibility drives parasite burden; animals in our study population experience thermal stress at the high end of our temperature range, leading to the prediction that high daily maximum temperatures would be associated with higher worm burdens (Gesquiere et al., 2008). In contrast to this prediction, T. trichiura egg counts, probability of infection for strongyles and A. caucasica, and parasite richness all decreased with increasing daily maximum temperatures. Free-living phases of nematode species such as eggs or larvae survive in broad temperature ranges, with different species having optimal environmental temperatures. Trichuris eggs require moderately warm temperature (20-30 C), moisture, and sandy or loamy soil to embryonate (Katz et al., 1989). We infer that the higher maximum daily temperatures in our study were not conducive for parasite development and hence resulted in decreased exposure rate.

68

3.5.2 Host-specific predictors of helminth burden

3.5.2.1 Age

Consistent with prior studies of primate parasite risk (Table S1), T. trichiura egg counts, strongyle infection risk, and parasite richness increased with age. These patterns represent a ‘type I’ age-intensity curve, which shows a positive correlation between age and helminth burden (Wilson et al. 2002). This profile can be attributed to increased exposure to parasites over an individual’s life time, host immuno-senescence, or both

(Nunn and Altizer, 2006; Wilson et al., 2002). Disentangling the effects of exposure and immuno-senescence on worm burden is challenging in wild populations. Adult T. trichiura are thought to live for 1-2 years (Cogswell, 2007; CDC, 2016; Anderson, 2000;

Gems 2000), though one study in humans reports a lifespan of 20 years (Winsberg et al.,

1975). Equally strikingly, some strongyles may live for 8-15 years (Table 5). The relatively long lives of these helminths would contribute to a pattern in which hosts accumulate these parasites as they age, leading to higher egg shedding at older ages. In addition, the interaction between host age and rainfall provides some evidence for changing immune responses with age. Specifically, middle-aged and older adults were more sensitive to low rainfall periods, experiencing especially high parasite burdens during dry periods compared to young adult females. In contrast to T. trichiura and strongyle infections, the risk of infection with S. pigmentatus and A. caucasica did not depend on age. These two helminth species are relatively short-lived and transmitted by arthropod vectors (e.g., dung beetles and cockroaches (Machida et al., 1978; Meade, 69

1983). Individuals of all age groups are equally likely to be exposed to the vectors that transmit these helminths; hence, host age may not increase the probability of acquiring an infection, especially compared to ecological factors that promote the survival of the vector in the environment.

3.5.2.2 Reproductive state

Reproductive state was a significant predictor of A. caucasica and S.pigmentatus infection and parasite richness. We predicted that pregnant and lactating females would have higher parasite burdens than cycling females because of the high energetic costs of these reproductive phases and changes in immune responses during pregnancy (Table1:

Beasley et al., 2010; Coop and Kyriazakis, 2001; Houdijk et al., 2001; Sakkas, 2011). We found support for this prediction for pregnancy, but not lactation. Prior research in

Amboseli has found conflicting results with respect to parasitism and reproductive state:

Hausfater and Watson (1976) found that cycling females exhibited higher parasite egg counts than females in other reproductive states; however this study did not differentiate helminth species. Consistent with our study, Meade (1983) found that pregnant females exhibited the highest parasite burdens. If energetic differences were driving patterns of infection, we would have expected both lactating and pregnant individuals to be affected. We speculate that these differences could be driven by steroid hormones which either influence host susceptibility or facilitate parasite growth, or both

(Escobedo et al., 2005; Romano et al., 2015). Pregnant females in this population

70

experience increased levels of progesterone, estrogen, and glucocorticoids relative to females in other reproductive states (Beehner et al., 2006). Steroid hormones are well known to influence both parasite growth and host immune competence in experimental set ups (see reviews in Hernández-Bello et al., 2010; Klein, 2004; Nava-Castro et al., 2011;

Romano et al., 2015). These effects however vary with the helminth species (Klein, 2004).

3.5.3 Host mechanisms that drive susceptibility to helminth infections

The role of GCs in suppressing the immune system (Sapolsky, 2004; Sapolsky et al., 2000; Webster Marketon and Glaser, 2008), coupled with their direct facilitation of growth and reproduction of parasites (Escobedo et al., 2005; Klein, 2004; Romano et al.,

2015), led us to predict that fGC concentrations would be correlated with higher helminth burdens. Contrary to our predictions, GC concentrations did not predict parasite burdens for any of the helminths in our data set. This result was surprising, considering that most of the predictors of parasite burden in this population (rainfall, temperature, age, and reproductive state) are associated with variation in fGC concentrations in females. Specifically, fGC concentrations increase with age (Altmann et al., 2010), during the dry season (Gesquiere et al., 2011), and in pregnant females

(Beehner et al., 2006; Gesquiere et al., 2008). However, group size has a complex association with fGC levels, such that individuals in the smallest and largest groups have higher fGC levels than females in intermediate-sized groups (Markham et al.,

2015). Our results were similar to those reported by Foerster et al. (2015) in female forest 71

guenons; these authors propose that GC levels obtained at sample collection may not be an accurate indicator of GC-associated immune responses to current parasite infections.

Further, it may be that other host mechanisms that drive susceptibility to helminth infection are more important than GC-associated immune responses. Some prior studies in primates have found positive correlations between GC concentrations and parasite risk; however, all of these studies were performed in males by Muehlenbein (2006),

Muehlenbein and Watts (2010) and Arlet et al., (2015) but see Clough et al. (2010). We posit that the effects of GCs on parasite risk may be more pronounced in males than in females as a result of additional immuno-modulatory effects of androgens

(Muehlenbein, 2006).

3.5.4 Conclusion

This study represents one of the most comprehensive investigations of how host environment and physiology predict parasite infection risk in a wild primate population. Our findings demonstrate that environmental and host factors are important predictors of parasite burden in female baboons. These predictors affect both exposure rate and/or host immune function, and they may be difficult to disentangle without additional information on host energetic state and immune phenotype. We hope that our study can serve as a basis for additional studies in this population and other primate populations. In particular, effects of ranging behavior, diet differences, and proximity measures (sociality indices) on parasite burden are yet to be examined in our study

72

population. Understanding predictors of helminth infection is key to understanding and to developing conservation and zoonotic prevention strategies.

73

4. Effects of helminth infections in wild-caught olive baboons: immune responses, body condition and health.

4.1 Introduction

Helminth infections are ubiquitous in natural vertebrate populations and common in domestic animals and humans as well. Often these infections are chronic and can result in long-term consequences for host fecundity and survival, although they are not generally fatal (Anthony et al., 2007; Bradley and Jackson, 2008). In domesticated animals and humans, helminth infections are associated with economic losses and are of public health importance, especially in developing countries (Bethony et al., 2006; CDC.,

2016; Charlier et al., 2014; Hotez et al., 2008). Costs inflicted by parasitic infections may include energetic costs, compromised host nutrition, reduced reproductive success, increased risk of predation, altered competitive ability, and compromised immune function (Coop and Holmes, 1996; Hamilton and Zuk, 1982; Hudson and Dobson, 1992).

Understanding patterns and consequences of helminth infections in natural populations is important for characterizing factors that shape infectious disease emergence patterns and host population dynamics (Ezenwa, 2016b; Ezenwa et al., 2010; Tompkins et al.,

2011; Tompkins and Begon, 1999). The immune responses and physiological costs experienced by hosts as a consequence of helminth infections are rarely studied in natural systems because such studies can present significant logistical challenges

(Budischak et al., 2012; Hudson and Dobson, 1995). Our study investigates the immune

74

responses associated with gastrointestinal worm infections and the consequences of these infections for measures of health in hosts (wild caught nonhuman primates).

Because nonhuman primates are closely related to humans, nonhuman primate research studies contribute to understanding disease etiology, progression, and transmission in humans. In addition, disease risk studies may also contribute to conservation strategies for wild nonhuman primates and management of zoonoses.

4.1.1 The effects of helminths on host immune function

Helminths are gastrointestinal worms that are classed broadly into three groups based mainly on their morphology: flatworms (cestodes and trematodes), round worms

(nematodes), and thorny threaded worms (acanthocephalins). Transmission mode varies across helminths: some may be acquired by hosts directly, through ingestion of infective stages (eggs or larvae), while others are transmitted indirectly via intermediate hosts

(Anderson, 2000; Katz et al., 1989). Immune responses elicited by hosts are important in determining the impact the parasites have on host fitness (Llyod, 1995; Sandland and

Minchella, 2003). These responses can operate at different levels including limiting parasite establishment, development, fecundity, and/or survival (Colditz, 2008; Llyod,

1995; Sheldon and Verhulst, 1996). The ability of the host to prevent and control infection is referred to as immunocompetence (Norris and Evans, 2000). Host immunocompetence depends on a myriad of factors such as age, sex, genotype,

75

nutrition, concurrent infections, stressors, adaptive immunity, and rate of parasite exposure (Colditz, 2008; Moreau and Chauvin, 2010).

Immune responses against helminth infections are mainly reported from controlled experimental studies and involve both innate and adaptive arms of the immune system. Immunological studies in laboratory rodents and humans have shown that the immune arm associated with helminth burdens is the T-helper 2 (Th2) arm of lymphocytes (Allen and Maizels, 2011). Th2 cytokines (e.g., interleukin-4 (IL-4), IL- 5, IL-

13, and IL-9) produce high serum levels of immunoglobulin E (IgE), stimulate mast cells and eosinophils, and increase gut motility (Girgis et al., 2013; Macdonald et al., 2002;

Maizels et al., 2004). Increased levels of Th2 cytokines result in increased eosinophil cells in the circulation (also referred to as eosinophilia). It is hypothesized that these cytokines are particularly strongly activated by helminths whose larval stages invade the intestinal tissue, enter the circulation, and travel to the other organs in the body before returning to the intestines (see discussion in Behm and Ovington, 2000). This invasion leads to mobilization of eosinophils to the blood and to affected tissues. In addition to Th2 responses, helminths also provoke production of T-regulatory cells. T- regulatory cells

(Tregs) are a subset of lymphocytic cells that down-regulate Th2 immune responses in cases of over-activation of this response resulting from high intensity or long duration of infections (Long and Nanthakumar, 2004). Treg responses are characterized by production of anti-inflammatory cytokines. In fact, chronic helminth infections are

76

thought to produce a stereotypical response in which T-cell production is biased towards both Th2 and Treg cells (termed a Th2/Treg response) (Jackson et al., 2009).

Ecologists use multiple monitoring techniques to assess immune function and host health status. In this study, we examine total white blood cell counts, lymphocyte counts, and eosinophil counts. Counts of circulating white blood cells (leucocytes) are a reliable measure of basal immune investment (Bordes and Morand, 2009). Differential counts of the various types of white blood cells (WBC) also gives additional information regarding the different arms of the immune system. Lymphocytes are representative of adaptive immunity whereas neutrophils, eosinophils, basophils, and monocytes have innate immune functions (Warrington et al., 2011). Other components used to measure immunocompetence include measuring plasma and serum proteins (Norris and Evans,

2000). A meta-analysis by Bordes and Morand (2009) in 33 mammal species showed that helminth richness was positively associated with overall circulating WBC counts and lymphocyte counts.

Host immune function can also be assessed by neutrophil-to-lymphocyte ratios

(NLRs), with higher ratios indicating increased physiological stress and hence decreased immunocompetence (Davis et al., 2008). Neutrophil-to-lymphocyte ratios are also associated with inflammatory disease risk, parasite infections, and are a product of natural genetic variation in humans (Buyukkaya et al., 2012; Núñez et al., 2008; Reich et al., 2009). In nonhuman primates, a study in semi-captive chimpanzees by Obanda et al.,

77

(2014) found higher NLR in older individuals and in those with high BMI values; in that study, high BMI values were associated with obesity and hence with relatively poor health, similar to the situation seen in many human populations. However, in wild animal populations, low BMI is associated with relatively poor health and high BMI with good health (e.g. Altmann et al., 2010).

Few studies to our knowledge have reported associations between immune responses and helminths in natural primate populations. Farah et al. (2003) assayed for specific immunoglobulins associated with schistosome infection in baboons and found that IgM antibodies were a good reflection of active infection and were highest in animals with the most severe pathology lesions. In tamarins, Monteiro et al., (2010) reported increased levels of gamma globulins (antibodies) in individuals co-infected with a helminth (Prosthenorchis sp) and trypanosomes. The same study also reported decreased gamma globulin levels for co-infections with two other helminths (Spiruridae and Trichostrongylidae). In Verreaux’s sifakas, Springer et al., (2015) reported no association between helminth burden (Onchocercidae sp., a filarial nematode) and NLR.

NLR was higher in older individuals compared to younger ones in this study.

4.1.2 The effects of helminths on host health and body condition

Parasites impose fitness costs on hosts by reducing energetic resources available for important functions such as reproduction, growth, and maintenance of host health

(i.e., survival) (Brown, 1987; Hotez et al., 2008). Through these effects on reproduction

78

and survival, helminths have the potential to affect host abundance and population structure (Gulland, 1995; Hudson and Dobson, 1998). Direct costs of parasitism result from competition between hosts and parasites for host resources, or from pathological effects inflicted by parasites on the host intestinal mucosa. Indirect costs occur when hosts are forced to allocate most of their energetic resources to immune function, limiting their investment in reproduction and growth (Pedersen and Fenton, 2007). In spite of their costs, infections by helminths often appear superficially to be asymptomatic in the host, with mortality occurring only in individuals with extremely high parasite loads or in immunocompromised individuals (Bethony et al., 2006; Girgis et al., 2013; Wilson et al., 2002).

Most studies in wild animals focus on prevalence and determinants of helminth infections, and rarely investigate helminth-related health costs, although a few studies have demonstrated positive relationships between helminth burden and host mortality rate in sheep, hares, and birds (see reviews in Hudson and Dobson, 1995; Tompkins and

Begon, 1999). In natural populations of nonhuman primates incidental cases of mortality associated with helminth infection have been reported (Kaur and Singh, 2009), but only one study has demonstrated direct effects of helminth infections on survival and reproduction: in geladas, infection with tapeworms is associated with elevated mortality and reduced reproductive success (Nguyen et al., 2015). In addition, the offspring of infected gelada females experience higher mortality than the offspring of uninfected

79

mothers (Nguyen et al., 2015). However, it is important to note that tapeworm infections in geladas are uncharacteristic of other helminth infections (which often appear asymptomatic) because they are associated with massive cysts. Two other studies in natural primates have also reported changes in behavior and physiology associated with helminth infections. A study in red colobus monkeys by Ghai et al., (2016) showed that individuals that tested positive for Trichuris species infections demonstrated higher rates of copulation, grooming, and resting behaviors compared to those who were negative.

Friant et al., (2016) conducted experimental field studies in red capped mangabeys and showed that following treatment with antiparasitic drugs, uninfected individuals spent more time foraging, less time resting and were more vigilant compared to when they were infected. In addition, cortisol levels were lower after antiparasitic treatment suggesting that hosts perceived helminth infections as a stressful state.

In this study, we examined two measures of host health and body condition and their relationship to helminth infections: body mass index and red blood cell (RBC) indices. Below, we briefly review the literature on each of these health and body condition metrics.

4.1.2.1 Body condition: body mass index

Helminth infections may result in poor body condition and even malnutrition because worms feed on essential nutrients, interfere with absorption of nutrients in the gut, and can reduce host appetite (Coop and Kyriazakis 2001; Coop and Holmes, 1996).

80

Examples of essential nutrients depleted by helminth infections include iron, amino acids, protein, copper, vitamin A and other micronutrients (Crompton and Nesheim,

2002). Malabsorption in the intestines occurs as a result of tissue destruction and inflammation on the intestinal mucosa caused by helminths (Crompton and Nesheim,

2002). In addition to these direct effects, poor body condition makes individuals more susceptible to disease, and thus individuals with existing helminth infections and poor body condition are vulnerable to further infections by additional parasites (Beldomenico et al., 2008; Beldomenico and Begon, 2010; Long and Nanthakumar, 2004). Studies in laboratory mice and in ruminants have shown that energy-restricted diets lead to defective immune responses against helminths and consequently to high worm burdens

(see discussions in Koski et al., 1999; Koski and Scott, 2001).

In nonhuman primates, studies investigating the relationship between helminth burden and body condition have often used body mass as a predictor of helminth burden. Horii et al. (1982) showed that helminth burden did not have adverse effects on body mass and Hämäläinen et al. (2015) found that that body mass was negatively correlated with helminth richness in the rainy season and positively correlated in the dry season, a pattern they attributed to a sex by season interaction. Additionally, a comparative meta-analysis by (Vitone et al. (2004) reported that larger-bodied arthropoid primates exhibited higher helminth richness than smaller ones. Larger

81

bodied animals are more likely to forage more and hence encounter more infective-stage helminths.

However, unless body mass is used in longitudinal analyses (i.e., unless changes in body mass over time are measured), body mass may not be a reliable measure of body condition because of its correlation with body size. Body condition indices such as body mass index (BMI) and other body scoring indices are better proxies for body condition than body mass alone. Specifically, BMI is defined as weight/height2 and is often used as an index for morbidity and mortality in human populations. BMI is thought to reflect the availability of energy reserves such as fat and protein (Peig and

Green, 2009). Only a few studies thus far have investigated the effects of helminth burdens in nonhuman primates using body condition indices. For instance, Rafalinirina et al., (2015) found a positive relationship between helminth burden and a body condition index (tail circumference index) in roufous mouse lemurs. Benavides et al.,

(2012) report a negative correlation in chacma baboons between parasite species richness and a body condition index (calculated from body mass, mean upper arm thickness, and mean skinfold thickness). Eley et al. (1989) found that olive baboons with poor quality diets had higher worm burdens and poorer body condition (measured qualitatively by appearance and quantitatively by subcutaneous fat thickness and body weight) than baboons with higher quality diets.

82

4.1.2.2 Hematology: Red blood cell (RBC) indices

Helminth infections can result in anaemia (i.e., reduced availability of RBCs and therefore reduced oxygen delivery to the host tissues) because of blood loss and malabsorption of important nutrients. In humans, infection of hosts by Ancylostoma species (hookworms) leads to anemia because these worms suck blood from the host intestinal mucosa and produce anticoagulants, which facilitate blood loss from mucosal blood vessels (Bethony et al., 2006). Trichuris and Strongyloides species cause inflammation and destruction of the intestinal mucosa, thus interfering with absorption of essential nutrients such as amino acids, copper, and protein, needed for hemoglobin formation (Budischak et al., 2012; Cogswell, 2007; Crompton and Nesheim, 2002). In wild primates, two studies have investigated the relationship between helminth infections and red blood cell indices: Monteiro et al., (2010) report that infection with

Prosthenorchis sp (an acanthocephalin) was associated with low packed cell volume

(PCV, or hematocrit) whereas infection with Spiruridae or Trichostrongylidae (nematodes) was not. Springer et al., (2015) report no associations between helminth burden

(Onchocercidae sp., a filarial nematode) and PCV in Verreaux’s sifakas.

4.1.3 Goals, hypothesis and predictions

Our study focused on identifying immune responses associated with helminth infections, and identifying the consequences of these infections for health and body condition of the hosts (wild primates). We tested the following hypotheses: 83

Hypothesis 1: Helminths and host immune function. Because helminth infections provoke hosts to mount immune responses, higher helminth burdens will be associated with higher levels of circulating host immune cells. Specifically, we predict that helminth burdens will be positively associated with a) total white blood cell counts

WBC counts (a measure of basal immune investment), b) with total lymphocyte counts

(a measure of adaptive immune function), and c) eosinophil counts (a measure of innate immune function and of Th2 cytokine responsiveness). We also predict that higher helminth burdens will be associated with lower neutrophil-to-lymphocyte ratios (NLRs).

Hypothesis 2: Helminths and host health and body condition. Because intestinal helminths contribute to blood loss, intestinal inflammation, and destruction of the mucosal lining of the intestines, higher helminth burdens will be associated with lower indices of host health and condition. In addition, individuals in poorer body condition are more susceptible to helminth infections compared to those with good body condition. We predict that higher helminth burdens will be associated with poorer host health and body condition as measured by lower body mass index (BMI) and lower red blood cell (RBC) indices.

4.2 Methods

4.2.1 Study subjects and ethical review

This study focused on 83 olive baboons (Papio anubis) that were trapped in the

Aberdares mountain range in the central region of Kenya, East Africa between March

84

and May 2014. These animals were trapped by the Institute of Primate Research (IPR) in

Kenya, a biomedical research center that is permitted to trap wild primates from areas of human-nonhuman primate conflict under the recommendation and approval of the

Kenya Wildlife Service. The animals were transported back to IPR on the outskirts of

Nairobi, and all sample collection and measurement was undertaken at IPR. All protocols and procedures used were reviewed and approved by of the Institutional

Review Committee of IPR (IRC/14/14) and of Duke University (A082-15-03).

4.2.2 Sample collection

The study subjects underwent routine quarantine screening of animals at IPR, which involves immobilization, physical examination, and collection of biological samples. Chemical restraint for quarantine screening was achieved by using a mixture of ketamine hydrochloride (Kepro, Holland) at a dosage of 10mg/kg and xylazine (Ekinos,

Argentina) at a dosage of 0.5mg/kg administered intramuscularly. All samples were collected within the first two weeks of the animals’ arrival at IPR, and included blood in

EDTA (anti-coagulant) tubes for hematology, blood in untreated tubes for serum retrieval, and feces in 10% buffered formalin for parasitology analysis (Table 9). Body mass, sex, crown to rump measures (length in cm from the cranium to the base of the tail) were also recorded (Table 9). Individuals were categorized into three age groups based on their body weights and dentition: juveniles, sub-adults and adults (no infants

85

were included in this study). Body condition was assessed by calculating body mass index (BMI) as weight/crown-rump length2 (see, e.g., Altmann et al., 1993).

4.2.3 Hematology analysis

Whole blood samples collected in EDTA anticoagulant were analyzed using an automated blood cell counter (AcT 5diff CP, Beckman Coulter Inc. France). We measured total red blood cell (RBC) counts and three RBC indices that are indicators of health and condition: Packed Cell Volume (PCV), Mean Corpuscular Volume (MCV),

Hemoglobin (HB). PCV is number of red blood cells in a set volume of blood, MCV is the average size of RBCs, and HB is the amount of hemoglobin in in a set of volume of blood (Budischak et al., 2012). These RBC indices are somewhat correlated (Budischak et al., 2012 and Table S7) and can be associated with one or more of the following clinical conditions; blood loss, reduced production of RBCs, or reduced levels of important proteins required by RBCs. We also measured total white blood cell (WBC) counts and

WBC differential counts (counts of specific WBC types: lymphocytes, neutrophils, eosinophils, basophils and monocytes).

86

Table 9: Summary of sample sizes in models used for statistical analysis

4.3 Data analysis

4.3.1 Hypothesis 1: Effects of helminths on host immune function.

We used generalized linear models (GLMs) to test the associations between circulating host immune cells and measures of helminth burden. We ran a separate model for each of three immune cell types (complete WBC counts, lymphocyte counts and eosinophil counts) and for NLR (see Table 9 for sample sizes) WBCs and lymphocytes were normally distributed but NLR was not normally distributed, hence

87

the data was log transformed (log+1 base 10). The eosinophil counts were overdispersed with at least twelve individuals having zero counts. We thus used a negative binomial analysis to examine the effects of helminth burden on eosinophil counts. In each case, the immune cell type was the response variable, and our main predictor variables were log strongyle egg count and parasite richness. We added age class and sex as covariates.

Adults are thought to have more mature immune systems than younger animals, and sex differences in immune cells may result from the influence of sex hormones (Nunn and Altizer, 2006).

4.3.2 Hypothesis 2: Effects of helminths on host health and body condition.

We also used GLMS to test the effect of helminth burden on RBC indices and on our measure of body condition (BMI). We ran a separate model for each RBC index

(total RBC counts, PCV, MCV and HB). In each model, the response variable was the

RBC index and our main predictor variables were log strongyle egg counts and parasite richness. As in our models of immune cell type, we added age class and sex as covariates, as all of these measures are known to influence our RBC indices in primates, including baboons (Akinyi et al., 2013; Harewood et al., 2000; McPherson, 2013; Obanda et al., 2014).

We used R statistical software, version 3.2.3 (R Foundation for Statistical

Computing, 2014) to run our models. Visual inspections of the data for all models indicated no violations of the assumptions regarding the normality and homogeneity of 88

error variances. We tested for multicollinearity for model predictors using the ‘CAR’ package in R and we found no correlation between predictors (Variance Inflation Factors were <2).

4.4 Results

4.4.1 General epidemiology and heterogeneity of helminth burden by age class and sex

We identified a total of six helminth morphotypes, corresponding to one unidentified strongyle morphoptype and five known species, with prevalence that varied from 6% to 100% (Table 10). All the study subjects harbored strongyles, though in varying intensities ranging from fecal egg counts of 2 to counts of 5,350 (Table 10).

Strongyles are a family of nematodes (Strongylidae), and their eggs are difficult to distinguish and hence we grouped them together. Other helminths occurred at different prevalence rates: Trichuris trichiura (49.4%), Strongyloides fullerboni (41%), Abbreviata caucasica (14.5%), Streptopharagus pigmentatus (33.7%) and Enterobius vermiculari (5%).

Based on the morphotypes we identified, the median individual parasite richness was 2, and the range was 1-5 parasites. Males harboured more parasite species than females

(W=79, p=0.001), and males harboured more T. trichiura eggs than females (W=40, p=<0.0001; note that to examine sex differences, we analyzed immature animals only, as there were no adult males in the study). Amongst females, we did not find any differences in helminth burdens between immatures, subadults, and adults (the absence of adult males meant that we did not perform this age-class analysis for males). 89

Table 10: Helminth species and prevalence

4.4.2 Hypothesis 1: Effects of helminths on host immune function.

We tested whether helminth burdens were associated with circulating host immune cells and found that strongyle egg counts were linked to lymphocyte cell counts in the predicted direction (Table 11, Figure 6). However, strongyle egg counts did not significantly predict total WBC or eosinophil counts, and parasite richness did not predict cell counts for any of the WBC types we measured (Table 11). There was no effect of age class or sex on immune cell indices. Means and ranges for all cell types are reported in Table S8 and S9). Strongyle egg counts, but not richness, predicted neutrophil-lymphocyte ratio (NLR): subjects with high strongyle egg counts had lower

NLR (Table 11, Figure 7).

90

Table 11: Immune responses to helminth burden

91

Figure 6: Effect of log strongyle egg counts on lymphocytes

Data points represent residual values of log lymphocyte counts for each individual as a function of strongyle egg counts. Residuals were calculated from a model containing all the predictors except strongyle egg counts (See Table 11).

Figure 7: Effect of strongyle egg counts on NLR

Each data point represents residuals values of log NLR for each individual as a function of strongyle egg count. Residuals were calculated from a model containing all the predictors except strongyle egg count (see Table 11).

92

4.4.3 Hypothesis 2: Effects of helminths on host health and body condition

Parasite richness, but not strongyle egg counts, significantly predicted body mass index (BMI), with higher parasite richness associated with lower BMI (Table 12, Figure

8). Contrary to our predictions, neither strongyle egg counts nor parasite richness were predictors of any of our RBC indices (Table 12). RBC counts, PCV and Hb levels differed with age class, with immatures (juveniles and subadults combined) having higher levels than adults. RBC and PCV levels also differed with sex such that males had lower counts than females. Age class predicted all measures of body condition, with juveniles and subadults having lower indices compared to adults (Table 12).

Figure 8: Effect of parasite richness on BMI

Each data point represents residuals values of BMI for each individual as a function of parasite richness. Residuals were calculated from a model containing all the predictors except parasite richness. (see Table 12).

93

Table 12: Effects of strongyle egg counts and richness on host health and body condition

94

4.5 Discussion

Helminth infections are ubiquitous in natural populations and hence play a potentially large role in influencing host fitness through the immune responses they provoke and the costs they may impose on host health and body condition (Sandland and Minchella 2003). Here, we investigated whether helminth burdens were associated with various circulating immune cells and measures of host health and body condition.

Overall, we found that subjects in this study were infected with several helminth species including strongyles, T.trichiura, A. caucasica, S.pigmentatus, E. vermicularis, and S. fulleborni. We found that higher helminth burdens were associated with higher lymphocyte counts, but not higher eosinophil counts or total WBC counts. Neutrophil- to-lymphocyte ratios (NLR) were also predicted by strongyle egg counts. Host body mass index (BMI) was predicted by parasite richness and host haematological indices were not predicted by our measures of helminth burden but instead varied with age class and sex. We discuss each of these findings below.

4.5.1 Immune responses to helminth infections

To investigate the immune responses associated with helminth burden, we looked at cells which represented both adaptive and innate immune system.

Lymphocytes, which constitute 20-30% of total WBCs, are mainly involved in adaptive arm of the immune system and respond to helminth infections by promoting either 1)

Th2 responses, which provoke production of cytokines and antibodies and by

95

stimulating effector cells such as eosinophils and mast cells, or 2) T-reg responses, which suppress Th2 immunity (Allen and Maizels, 2011; Maizels et al., 2004). Host immune responses are crucial for limiting parasite establishment, development, and fecundity, and for enhancing expulsion of adult worms from the intestines (Rowe et al., 2008).

For most of our analysis, we focused on immune responses to strongyles, because they were the most prevalent in this population, and on parasite richness because of the diverse set of species that we identified. We found that strongyle counts were positively associated with lymphocyte counts, suggesting that helminth burden provoked adaptive immune responses. Our result was in agreement with a study in wild buffalos that found a positive correlation between lymphocyte counts and egg counts of Cooperia sp. (a strongyle affecting buffalos) (Budischak et al., 2012). In contrast, a study by Rowe et al. (2008) demonstrated that lymphocyte burdens were negatively associated with Haemonchus sp. fecundity (also a strongyle species). These contrasting studies point to differences between parasite species in the host immune responses they provoke. Unfortunately, in this study, we were unable to distinguish between the strongyle species infecting our subjects to determine whether different strongyle species provoked different immune responses.

Our result that lymphocyte counts increased with strongyle egg counts in this study could have resulted from heightened Th2 or Treg responses in hosts. Th2 responses are especially helpful for fighting helminth infections in the short term,

96

whereas Treg responses are important for fighting heavy or chronic helminth infections

(Long and Nanthakumar, 2004). T regulatory responses protect animals against the severe inflammation and pathology that could result from an uncontrolled Th2 response, in a process known as immunomodulation (Jackson et al., 2009). Because wild primates are chronically infected by helminths, we propose that Treg cells are mainly responsible for the adaptive immune responses we reported. Unfortunately, we did not look at any Treg responses (e.g. levels of IL-10, TGF) in these primates but propose that this can be done in future studies. We looked at IL-5, a Th2 response, but the serum levels were low in most individuals (data not presented here).

The lack of any relationship between parasite richness and lymphocyte counts in our study did not mirror results from the meta-analysis by Bordes and Morand (2011), which reported that increased lymphocytes were associated with taxa facing higher helminth richness. Bordes and Morand (2011) also reported increased WBC counts with helminth richness, which we did not find in our study. Our results point to individuals adopting a more specific immune response (lymphocytes) to fighting helminth infections as opposed to investment on basal immunity (WBC count).

We also examined whether innate immune responses (measured by eosinophil counts) were associated with helminth burdens. Eosinophils constitute 1-5% of total circulating WBCs and are mainly stimulated by Th2 lymphocytes during helminth infections (Behm and Ovington, 2000). Helminth-induced eosinophilia has been reported

97

in laboratory mice, humans and in domestic animals (Behm and Ovington, 2000; Maizels et al., 2004). In contrast, studies in livestock and wild buffalos have shown negative associations between eosinophils and strongyles (Callaby et al., 2014; Ezenwa et al.,

2010; Jolles et al., 2008). The authors of these studies speculate that this negative relationship could be driven by resistance to infections, such that resistant individuals have high eosinophil counts and low worm burdens. Contrary to both of these observed relationships (helminth-induced eosinophilia in some species, negative associations between eosinophils and strongyles in others), strongyle egg counts were not associated with eosinophil count in our study subjects. This result is surprising because strongyle infections are characterised by larval stages which penetrate the mucosa of the large intestines, and eosinophils are known to migrate to infected tissues to kill larval stages of helminths (Allen and Maizels, 2011; Anderson, 2000; Cogswell, 2007). Our speculation that Treg responses are activated in our study subjects in response to high strongyle egg counts would be consistent with the lack of association between strongyle egg counts and eosinophil counts, because Tregs dampen Th2 responses such as eosinophil production. Similar to Bordes and Morand (2011) we did not find any association between parasite richness and eosinophil counts, which indicates that the number of species infecting an individual is not important in determining specific immune responses to helminths.

98

We also looked at Neutrophil-to-lymphocyte ratios (NLR) because they have been previously associated with various factors including, age, infection, physiological stress and inflammatory disease risk (Davis et al., 2008; Obanda et al., 2014). In line with our prediction, we found that younger subjects and those with high strongyle egg counts exhibited lower NLR (fewer neutrophils compared to lymphocytes). Interpreting these ratios in the light of our previous findings of increased lymphocytes with strongyle abundance, we posit that age and helminth infections are the main drivers of

NLR in this population of baboons. Similar results have been reported in one other study in Verreaux sifakas, by Springer et al., (2015) who found high NLR in older individuals but did not find any association between between Onchocercidae species (a filarial nematode) and NLR. Springer et al., (2015) speculate that NLR may be affected by increased production of epinephrine during capture, which leads to downstream mobilization of lymphocytes, hence masking effects of parasite infections on NLR.

Because these individuals in our study were sampled within the first month of arrival from the wild, it is possible that stress could have driven increased production of both helminths and lymphocytes. Some studies have shown association of stress hormones and helminth burden in other populations (Arlet et al., 2015; Clough et al., 2010;

Muehlenbein, 2006; Muehlenbein and Watts, 2010). It is difficult to tease apart the effects of stress and infection status on NLR ratios.

99

An important challenge for analyzing immune responses in natural populations is that multiple factors influence host immune responses at the time of sample collection.

For example, host genotypes differ, such that some individuals have resistant genotypes compared to others and hence respond to infections differently (Bordes and Morand,

2009; Clough et al., 2010). Social and environmental variables (Semple et al., 2002), behavior (e.g. mating promiscuity, Nunn et al., 2000), and active infections by other macro or microparasites (Ezenwa, 2016b) also affect host immune responses. Ideal studies would involve screening individuals for multiple groups of parasites (e.g. fungi, virus, bacteria, and arthropods) and performing an integrative analysis on host immune responses. We screened subjects for two microparasites, Herpes Papio Virus 2 and

Hepatocystis kochi, but we lacked statistical power to test for associations between helminth burden, immune responses and microparasite burdens. Additionally, more specific immune measures can be employed such as testing for a wide array of cytokines that may be associated with helminths.

4.5.2 Effects of helminth infection on health and body condition

We analyzed RBC indices as a measure of host health because these indices reflect host ability to acquire, distribute and utilize oxygen required for most tissue functions. We found that neither strongyle burden nor parasite richness affected host hematological indices. This was contrary to our expectations because studies in

100

ruminants and humans have reported reduced RBC indices in individuals with high strongyle egg counts (Bethony et al., 2006; Budischak et al., 2012; Callaby et al., 2014).

Further, studies in wild and captive primates report weight loss, diarrhea, irritation of the large intestines, and formation of abdominal adhesions in cases of heavy infection by strongyles (Cogswell, 2007; Kaur and Singh, 2009). We expected to observe reduced hematology indices in our study subjects resulting from irritation of the intestinal mucosa and utilization of host resources required for production of red blood cells.

Our results parallel Monteiro et al. (2010), who investigated effects of

Trichostrongylus sp. in tamarins and who did not find any association between helminth burden and host hematology indices. They posit that individuals who have the heaviest infections may die relatively early, leaving as survivors those with lighter infections (see also Jolles et al., 2008). This hypothesis could apply to the subjects we studied, and points to the limitation of using cross-sectional data for analysis of health indices. An alternative hypotheses is that haematological variables may be more sensitive to factors such as age and sex than to helminth abundance and diversity. Age and sex were significant predictors of most hematology indices in our subjects, with adults and males having lower indices. A third hypothesis is that effects of helminth burden are specific to each strongyle species, such that aggregate strongyle counts mask species-specific costs associated with different species of strongyles. For instance, Budischak et al., (2012) did not find any effects of strongyles on host hematology when they used aggregate

101

strongyle counts, but reported negative correlations when they repeated their anlayses using species specific strongyle data. Future analysis using species specific data might be useful to tease out these differences.

Mass-for-length measurements such as body mass index are a common measure of body condition in humans and some wild populations (Altmann et al. 1993; Peig and

Green, 2009; Obanda et al., 2014). In line with our predictions and in agreement with other studies in primates, we found that higher parasite richness was associated with lower BMI values, but strongyle abundance was not. Benavides et al., (2012) report a negative correlation between parasite species richness and a body condition index in chacma baboons, and Eley et al., (1989) reported that olive baboons with poor quality diets had high worm burdens and poor body condition. Helminth infections drive poor body conditions because worms feed on essential host nutrients, interfere with absorption of nutrients in the gut, reduce host appetite and force hosts to allocate their limited energetic resources to fighting infections (Blackwell et al., 2010; Coop and

Holmes, 1996; Coop and Kyriazakis, 1999; Cooper et al., 2012). At least three of the six helminth species (Strongyloides sp., Oesophagostomum sp. (strongyle) and T. trichiura) we studied have been linked to weight loss and poor growth rates in captive primates hence explaining our findings of helminth richness on body condition. In contrast to these results, Rafalinirina et al., (2015) found a positive relationship between helminth burden and a body condition index in roufous mouse lemurs. This surprising result highlights

102

the challenges of inferring general relationships between helminths and body condition from single studies, and emphasizes the usefulness of meta-analyses when sufficient data become available. However, even with multiple results showing a detrimental effect of parasties on host body condition, it is difficult to claim a causal link between helminth burden and host body condition because infection can lead to poor condition and vice versa (Beldomenico et al., 2008).

4.5.3 Conclusion

This study highlights the immune responses associated with helminth burdens in baboons and demonstrates that individuals infected with high helminth burdens or multiple species of helminths are prone to poor body conditions. Our study adds to the limited data on disease surveillance and health monitoring in wild populations thus providing the missing link between laboratory and field studies. It also serves as a baseline for future research on responses to pathogen infections and potential outcomes in natural animal populations.

103

Appendix A: Amboseli baboons parasite guide

104

105

Appendix B: Table S1

Associations of different combinations of factors with helminth infection from results of 32 studies done in wild primates.

106

107

Key: (+)- positive association, (-) = negative association, none=no association,

108 NA=predictor not in the study, P=pregnant,

L=Lactating, C=cycling, NL= non-lactating, *= effect observed in at least one parasite one parasite species.

Because our study focused on predictors of helminth burdens in females, we excluded from this table four studies that examined only males (Arlet et al. 2015, Gillespie et al. 2013, Muehelbein et al 2006, 2010). We report results on female trends for the studies that analyzed males and females separately, otherwise we report general trends reported in both sexes.

Appendix C: Table S2

Summary of Studies done in primates

Social Reproductive Group Rainfall Effects status state Age size /Season Temp GCs Number of studies 11 8 21 10 21 9 5 Number of + effects 1 P/L > C = 4 9 4 8 5 1 Number of - effects 1 C>L = 1 6 3 6 2 1 Number of no effects 9 2 4 3 4 2 3 Other effects 1 2 3

109

Appendix D: Table S3

Environmental and host predictors of helminth burden using proportional ranks (n=746, females=122)

110

Appendix E: Table S4:

Environmental and host predictors of helminth burden using number of dry months

111

Appendix F: Table S5

Summary results of alternative rainfall models

Helminth Cumulative AIC Rainfall effect burden rainfall period T. trichiura 3 months 1501.5  rain =worms I month 1540  rain =worms 2 weeks 1569.6  rain =worms

Strongyles 3 months 949.5 None 1 month 946.4  rain =worms 2 weeks 952.1  rain =worms

A. caucasica 3 months 945.9  rain =worms 1 month 970  rain =worms 2 weeks 983  rain =worms

S. pigmentatus 3 months 834 None 1 month 828.5 None 2 weeks 830 None

Parasite richness 3 months 2033 None 1 month 2039.3  rain =worms 2 weeks 2050.7 None

112

Appendix G: Table S6

fGC as predictor of parasite burdens (n=709, females=122)

113

Appendix G: Table S7

Correlation between RBC indices

RBC index RBC counts PCV MCV HB

RBC counts -

PCV R= 0.87, p<0.0001 -

MCV R=-0.20, p=0.012 R=0.21, p=0.054 -

HB R=0.82, p<0.0001 R=0.92, p>0.0001 R=0.20, p=0.06 -

114

Appendix G: Table S8

15

1

References Abbott, D. H., Keverne, E. B., Bercovitch, F. B., Shively, C. A., Mendoza, S. P., Saltzman, W., … Sapolsky, R. M. (2003). Are subordinates always stressed? A comparative analysis of rank differences in cortisol levels among primates. Hormones and Behavior, 43(1), 67–82.

Adkins-Regan, E. (2005). Hormones and animal social behavior. Princeton: Princeton University Press.

Akinyi, M. Y., Tung, J., Jeneby, M., Patel, N. B., Altmann, J., & Alberts, S. C. (2013). Role of grooming in reducing tick load in wild baboons (Papio cynocephalus). Animal Behaviour, 85, 559–568.

Alberts, S. ., & Altmann, J. (2011). Monitoring Guide for the Amboseli Baboon Research Project: Protocols for long-term monitoring and data collection.

Alberts, S. C. (2012). Magnitude and sources of variation in male reproductive performance. In J. Mitani, J. Call, P. Kappeler, R. Palombit, & J. B. Silk (Eds.), Evolution of Primate Societies. Chicago: University of Chicago Press.

Alberts, S. C., & Altmann, J. (1995). Balancing costs and opportunities: Dispersal in male baboons. American Naturalist, 145(2), 279–306.

Alberts, S. C., & Altmann, J. (1995). maturity of age at reproductive determinants and activation : Preparation in male baboons. Behavioral Ecology and Sociobiology, 36(6), 397–406.

Alberts, S. C., & Altmann, J. (2001). Immigration and Hybridization Patterns of Yellow and Anubis Baboons In and Around Amboseli , Kenya. American Journal of Primatology, 154(February), 139–154.

Alberts, S. C., & Altmann, J. (2012). Long term field studies of primates. (P. Kappeler & D. P. Watts, Eds.). Springer-Verlag.

Alberts, S. C., Archie, E. A., Gesquiere, L. R., Altmann, J., Vaupel, J. W., & Christensen, K. (2014). The Male-Female Health-Survival Paradox: A Comparative Perspective on Sex Differences in Aging and Mortality. The National Academies Press, 339–363.

Alberts, S. C., Buchan, J. C., & Altmann, J. (2006). Sexual selection in wild baboons: from mating opportunities to paternity success. Animal Behaviour, 72(5), 1177–1196.

Alberts, S. C., Hollister-Smith, J. A., Mututua, R., Sayialel, S. N., Muruthi, P. M., Warutere, J. K., & Altmann, J. (2005). Seasonality and long term change in a savanna environment. In D. K. Brockman & C. P. Van Schaik (Eds.), Seasonality in Primates: Studies of Living and Extint Human and Non-Human Primates (p. 571). Cambridge: Cambridge University Press.

116

Alberts, S. C., Sapolsky, R. M., & Altmann, J. (1992). Behavioral, endocrine, and immunological correlates of immigration by an aggressive male into a natural primate group. Hormones and Behavior, 26(2), 167–178.

Alberts, S. C., Watts, H. E., & Altmann, J. (2003). Queuing and queue-jumping: long-term patterns of reproductive skew in male savannah baboons, Papio cynocephalus. Animal Behaviour, 65(4), 821–840.

Allen, J. E., & Maizels, R. M. (2011). Diversity and dialogue in immunity to helminths. Nature Reviews. Immunology, 11(6), 375–88.

Altizer, S., Dobson, A., Hosseini, P., Hudson, P., Pascual, M., & Rohani, P. (2006). Seasonality and the dynamics of infectious diseases. Ecology Letters, 9(4), 467–484.

Altmann, J., & Alberts, S. C. (2003). Intraspecific variability in fertility and offpsring survival in a non-human primate: behavioral control of ecological and social sources. In K. W. Watcher & R. A. Bulatao (Eds.), Offspring: Human Fertility Behavior in a Biodemographic Perspective (pp. 140–169). Washington D.C.: National Academy Press.

Altmann, J., & Alberts, S. C. (2005). Growth rates in a wild primate population : ecological influences and maternal effects. Behavioral Ecology and Sociobiology, 57, 490–501.

Altmann, J., Alberts, S. C., Altmann, S. A., & Roy, S. B. (2002). Dramatic change in local climate patterns in the Amboseli basin , Kenya. African Journal of Ecology, 40, 248–251.

Altmann, J., Gesquiere, L., Galbany, J., Onyango, P. O., & Alberts, S. C. (2010). Life history context of reproductive aging in a wild primate model. Annals of the New York Academy of Sciences, 1204, 127–138.

Altmann, J., Schoeller, D., Altmann, S. A., Muruthi, P., & Sapolsky, R. M. (1993). Body size and fatness of free living baboons reflect food availability and activity levels. American Journal of Primatology, 30(2), 149–161.

Altmann, S. (1998). Foraging for survival: Yearling baboons in Africa. Chicago: University of Chicago Press.

Anderson, R. C. (2000). Nematode parsites of vertebrates: Their development and transmission (2nd ed.). UK: CABI Publishing.

Anthony, R. M., Rutitzky, L. I., Urban Jr., J. F., Stadecker, M. J., & Gause, W. C. (2007). Protective immune mechanisms in helminth infection. Nature Reviews Immonulogy, 7(12), 975–987. Arlet, M. E., Chapman, C. A., Isbell, L. A., Molleman, F., Mänd, R., Hõrak, P., & Carey, J. R. (2015). Social and Ecological Correlates of Parasitic Infections in Adult Male Gray-Cheeked Mangabeys (Lophocebus albigena). International Journal of Primatology, 36(5), 967–986.

117

Arlet, M. E., Grote, M. N., Molleman, F., Isbell, L. A., & Carey, J. R. (2009). Reproductive tactics influence cortisol levels in individual male gray-cheeked mangabeys (Lophocebus albigena). Hormones and Behavior, 55(1), 210–216.

Audebert, F., Hoste, H., & Durette-Desset, M. C. (2002). Life cycle of Trichostrongylus retortaeformis in its natural host, the rabbit (Oryctolagus cuniculus). Journal of Helminthology, 76(3), 189–192.

Auphan, N., DiDonato, J. A., Rosette, C., Helmberg, A., & Karin, M. (1995). Immunosuppression by glucocorticoids: Inhibition of NF-kappaB activity through induction of IkappaB synthesis. Science (New York, N.Y.), 270(5234), 286.

Babayan, S. A., Allen, J. E., Bradley, J. E., Geuking, M. B., Graham, A. L., Grencis, R. K., … Pedersen, A. B. (2011). Wild immunology: Converging on the real world. Annals of the New York Academy of Sciences, 1236(1), 17–29.

Bardin, C. W. (1996). The anabolic action of testosterone. The New England Journal of Medicine, 266(9), 52–53.

Barger, I. A. (1993). Influence of sex and reproductive status on suceptibility of ruminants to nematode parasitism. International Journal for Parasitology, 23(4), 463–469.

Barreiro, L. B., & Quintana-Murci, L. (2010). From evolutionary genetics to human immunology: how selection shapes host defence genes. Nature Reviews. Genetics, 11, 17–30.

Beasley, A. M., Kahn, L. P., & Windon, R. G. (2010). The periparturient relaxation of immunity in Merino ewes infected with Trichostrongylus colubriformis: Parasitological and immunological responses. Veterinary Parasitology, 168(1–2), 60–70.

Beasley, A. M., Kahn, L. P., & Windon, R. G. (2012). The influence of reproductive physiology and nutrient supply on the periparturient relaxation of immunity to the gastrointestinal nematode Trichostrongylus colubriformis in Merino ewes. Veterinary Parasitology, 188(3–4), 306–324.

Beehner, J. C., Bergman, T. J., Cheney, D. L., Seyfarth, R. M., & Whitten, P. L. (2006). Testosterone predicts future dominance rank and mating activity among male chacma baboons. Behavioral Ecology and Sociobiology, 59(4), 469–479.

Beehner, J. C., Gesquiere, L., Seyfarth, R. M., Cheney, D. L., Alberts, S. C., & Altmann, J. (2009). Testosterone related to age and life-history stages in male baboons and geladas. Hormones and Behavior, 56(4), 472–480.

Beehner, J. C., Nguyen, N., Wango, E. O., Alberts, S. C., & Altmann, J. (2006). The endocrinology

118

of pregnancy and fetal loss in wild baboons. Hormones and Behavior, 49(5), 688–699.

Behm, C. A., & Ovington, K. S. (2000). The role of eosinophils in parasitic helminth infections: Insights from genetically modified mice. Parasitology Today, 16(5), 202–209.

Beldomenico, P. M., & Begon, M. (2010). Disease spread, susceptibility and infection intensity: vicious circles? Trends in Ecology and Evolution, 25(1), 21–27.

Beldomenico, P. M., Telfer, S., Gebert, S., Lukomski, L., Bennett, M., & Begon, M. (2008). Poor condition and infection: a vicious circle in natural populations. Proceedings. Biological Sciences / The Royal Society, 275(1644), 1753–9.

Belle, S. Van, Estrada, A., Ziegler, T. E., & Strier, K. B. (2009). Social and hormonal mechanisms underlying male reproductive strategies in black howler monkeys (Alouatta pigra). Hormones and Behavior, 56(4), 355–363.

Belthoff, J. R., & Dufty, Jr, A. M. (1995). Locomotor activity levels and the dispersal of western screech-owls,Otus kennicottii. Animal Behaviour, 50(2), 558–561.

Belthoff, J. R., & Dufty Jr, a M. (1998). Corticosterone, body condition and locomotor activity: a model for\rnatal dispersal in birds. Animal Behaviour., 55(2), 405.

Benavides, J. A., Huchard, E., Pettorelli, N., King, A. J., Brown, M. E., Archer, C. E., … Cowlishaw, G. (2012). From parasite encounter to infection: Multiple-scale drivers of parasite richness in a wild social primate population. American Journal of Physical Anthropology, 147(1), 52–63.

Bercovitch, F. B. (1999). Behavioral ecology and socioendocrinology of reproductive maturation in cercopithecine monkeys. Comparative Primate Socioecology, 298–320.

Bercovitch, F. B., & Ziegler, T. E. (2002). Current Topics In Primate Socioendocrinology. Annual Review of Anthropology, 31(1), 45–67.

Bergman, T. J., Beehner, J. C., Cheney, D. L., Seyfarth, R. M., & Whitten, P. L. (2005). Correlates of stress in free-ranging male chacma baboons, Papio hamadryas ursinus. Animal Behaviour, 70(3), 703–713.

Bernstein, R. M., Leigh, S. R., Donovan, S. M., & Monaco, M. H. (2008). Hormonal correlates of ontogeny in baboons (Papio hamadryas anubis) and Mangabeys (Cercocebus atys). American Journal of Physical Anthropology, 136(2), 156–168.

Bethony, J., Brooker, S., Albonico, M., Geiger, S. M., Loukas, A., Diemert, D., & Hotez, P. J. (2006). Soil-transmitted helminth infections: ascariasis, trichuriasis, and hookworm. Lancet, 367(9521), 1521–1532.

119

Biard, C., Monceau, K., Motreuil, S., & Moreau, J. (2015). Interpreting immunological indices: The importance of taking parasite community into account. An example in blackbirds Turdus merula. Methods in Ecology and Evolution, 6(8), 960–972.

Blackwell, A. D., Snodgrass, J. J., Madimenos, F. C., & Sugiyama, L. S. (2010). Life history, immune function, and intestinal helminths: Trade-offs among immunoglobulin E, C- reactive protein, and growth in an Amazonian population. American Journal of Human Biology, 22(6), 836–848.

Bordes, F., & Morand, S. (2009). Coevolution between multiple helminth infestations and basal immune investment in mammals: Cumulative effects of polyparasitism? Parasitology Research, 106(1), 33–37.

Bordes, F., & Morand, S. (2011). The impact of multiple infections on wild animal hosts: a review. Infection Ecology & Epidemiology, 1(0), 1–10.

Bradley, J. E., & Jackson, J. a. (2008). Measuring immune system variation to help understand host-pathogen community dynamics. Parasitology, 135(7), 807–823.

Bribiescas, R. G. (2001). Reproductive Ecology and Life History of the Human Male. American Journal Of Physical Anthropology, 44(January 2001), 148–176.

Brown, P, .J. (1987). Microparsaites and Macroparasites. Cultural Anthropology, 2(1), 155–171.

Budischak, S. A., Jolles, A. E., & Ezenwa, V. O. (2012). Direct and indirect costs of co-infection in the wild: Linking gastrointestinal parasite communities, host hematology, and immune function. International Journal for Parasitology: Parasites and Wildlife, 1, 2–12.

Bulger, J. B. (1993). Dominance Rank and Access to Estrous Females in Male Savanna Baboons. Behaviour, 127(1), 67–103.

Bustamante, C. D., Fledel-Alon, A., Williamson, S., Nielsen, R., Hubisz, M. T., Glanowski, S., … Clark, A. G. (2005). Natural selection on protein-coding genes in the human genome. Nature, 437, 1153–1157.

Buyukkaya, E., Karakas, M. F., Karakas, E., Akcay, a. B., Kurt, M., Tanboga, I. H., & Sen, N. (2012). Correlation of Neutrophil to Lymphocyte Ratio With the Presence and Severity of Metabolic Syndrome. Clinical and Applied Thrombosis/Hemostasis, 20(2), 159–163.

Cabaret, J., & Gasnier, N. (1998). Faecal egg counts are representative of digestive-tract strongyle worm burdens in sheep and goats. Parasite (Paris, France), 5(2), 137–42.

Callaby, R., Hanotte, O., Conradie VAN Wyk, I., Kiara, H., Toye, P., Mbole-Kariuki, M. N., …

120

Kruuk, L. E. B. (2014). Variation and covariation in strongyle infection in East African shorthorn zebu calves. Parasitology, 142, 1–13.

Cattadori, I. ., Boag, B., Bjornstad, O. ., Cornell, S. ., & Hudson, P. . (2005). Peak shift and epidemiology in a seasonal host-nematode system. Proceedings of the Royal Society B: Biological Sciences, 272(1568), 1163–1169.

CDC. (2016). Centers for Disease Control and Prevention. Retrieved from https://www.cdc.gov/

Chapman, C. A., Saj, T. L., & Snaith, T. V. (2007). Temporal dynamics of nutrition, parasitism, and stress in colobus monkeys: Implications for population regulation and conservation. American Journal of Physical Anthropology, 134(2), 240–250.

Chapman, C. A., Speirs, M. L., Hodder, S. A. M., & Rothman, J. M. (2010). Colobus monkey parasite infections in wet and dry habitats: Implications for climate change. African Journal of Ecology, 48(2), 555–558.

Chapman, C. A., Wasserman, M. D., Gillespie, T. R., Speirs, M. L., Lawes, M. J., Saj, T. L., & Ziegler, T. . (2006). Do Food Availability, Parasitism, and Stress Have Synergistic Effects on Red Colobus Populations Living in Forest Fragments? American Journal of Physical Anthropology, 131, 525–534.

Charlier, J., van der Voort, M., Kenyon, F., Skuce, P., & Vercruysse, J. (2014). Chasing helminths and their economic impact on farmed ruminants. Trends in Parasitology, 30(7), 361–367.

Charpentier, M. J. E., Fontaine, M. C., Cherel, E., Renoult, J. P., Jenkins, T., Benoit, L., … Tung, J. (2012). Genetic structure in a dynamic baboon hybrid zone corroborates behavioural observations in a hybrid population. Molecular Ecology, (21), 715–731.

Charpentier, M. J. E., Tung, J., Altmann, J., & Alberts, S. C. (2008). Age at maturity in wild baboons: Genetic, environmental and demographic influences. Molecular Ecology, 17(8), 2026–2040.

Clough, D., Heistermann, M., & Kappeler, P. M. (2010). Host intrinsic determinants and potential consequences of parasite infection in free-ranging red-fronted lemurs (Eulemur fulvus rufus). American Journal of Physical Anthropology, 142(3), 441–452.

Clutton-Brock, T. H. (1989). Female transfer and inbreeding avoidance in social mammals. Nature.

Cogswell, F. (2007). Parasites of non-human primates. In D. Baker (Ed.), Flynn’s parasites of laboratory animals (2nd Editio, pp. 693–743). UK.

Colditz, I. G. (2008). Six costs of immunity to gastrointestinal nematode infections. Parasite

121

Immunology, 30(2), 63–70.

Coop, R. L., & Holmes, P. H. (1996). Nutrition and parasite interaction. International Journal for Parasitology, 26(8–9), 951–962.

Coop, R. L., & Kyriazakis, I. (1999). Nutrition–parasite interaction. Veterinary Parasitology, 84(3), 187–204.

Coop, R. L., & Kyriazakis, I. (2001). Influence of host nutrition on the development and consequences of nematode parasitism in ruminants. Trends in Parasitology, 17(7), 325–330.

Cooper, N., Griffin, R., Franz, M., Omotayo, M., & Nunn, C. L. (2012). Phylogenetic host specificity and understanding parasite sharing in primates. Ecology Letters, 15, 1370–1377.

Copeland, K. C., Eichberg, J. W., Parker, R. C., & Bartke, A. (1985). Puberty in the chimpanzee: Somatomedin-C and its relationship to somatic growth and steroid hormone concentrations. The Journal of Clinical Endocrinology and Metabolism, 60(6), 1154–1160.

Cote, I. M., & Poulin, R. (1995). Parasitism and Group-size In Social Animals: A Meta-analysis. Behav Ecol, 6(2), 159–165.

Cowlishaw, G., & Dunbar, R. I. M. (1991). Dominance rank and mating success in male primates. Animal Behaviour, 41(6), 1045–1056.

Cristóbal-Azkarate, J., Chavira, R., Boeck, L., Rodr ́ıguez-Luna, E., & Vea`1, J. J. (2006). Testosterone levels of free-ranging resident mantled howler monkey males in relation to the number and density of solitary males: A test of the challenge hypothesis. Hormones and Behavior, 49(2), 261–267.

Crompton, D. W. T., & Nesheim, M. C. (2002). Nutritional Impact of Intestinal Helminthiasis During the Human Life cycle. Annual Review of Nutrition, 22(1), 35–59.

Cuarón, A. D. (1997). Conspecific aggression and predation: Costs for a solitary mantled howler monkey. Folia Primatologica, 68(2), 100–105.

Davis, A. K., Maney, D. L., & Maerz, J. C. (2008). The use of leukocyte profiles to measure stress in vertebrates: A review for ecologists. Functional Ecology, 22(5), 760–772.

Dixson, A. F. (2012). Primate sexuality: Comparative studies of the prosimians, monkeys, apes, and humans (Second). Oxford: Oxford University Press.

Dowell, S. F. (2001). Seasonal variation in host susceptibility and cycles of certain infectious diseases. Emerging Infectious Diseases, 7(3), 369–374.

122

Dufty, A. M., & Belthoff, J. R. (2001). Proximate mechanisms of natal dispersal: The role of body condition and hormones. In J. Clobert, E. Danchin, A. A. Dhondt, & D. J. Nicholas (Eds.), Dispersal (pp. 217–229). New York: Oxford University Press.

Elenkov, I. J., & Chrousos, G. P. (2002). Stress hormones, proinflammatory and antiinflammatory cytokines, and autoimmunity. Annals of the New York Academy of Sciences, 966(April), 290– 303.

Eley, R. M., Strum, S. C., Muchemi, G., & Reid, G. D. F. (1989). Nutrition, body condition, activity patterns, and parasitism of free-ranging troops of olive baboons (Papio anubis) in Kenya. American Journal of Primatology, 18(3), 209–219.

Escobedo, G., Roberts, C. W., Carrero, J. C., & Morales-Montor, J. (2005). Parasite regulation by host hormones: An old mechanism of host exploitation? Trends in Parasitology, 21(12), 588– 593.

Ezeamama, A. E., Friedman, J. F., Olveda, R. M., Acosta, L. P., Kurtis, J. D., Mor, V., & McGarvey, S. T. (2005). Functional significance of low-intensity polyparasite helminth infections in anemia. Journal of Infectious Diseases, 192(12), 2160–2170.

Ezeamama, A. E., McGarvey, S. T., Acosta, L. P., Zierier, S., Manalo, D. L., Wu, H., … Friedman, J. F. (2008). The synergistic effect of concomitant schistosomiasis, hookworm, and trichuris infections on children’s anemia burden. PLoS Neglected Tropical Diseases, 2(6).

Ezenwa, V. O. (2004). Interactions among host diet, nutritional status and gastrointestinal parasite infection in wild bovids. International Journal for Parasitology, 34(4), 535–542.

Ezenwa, V. O. (2016b). Helminth-microparasite co-infection in wildlife: lessons from ruminants, rodents, and rabbits. Parasite Immunology, (April), 527–534.

Ezenwa, V. O., Archie, E. a, Craft, M. E., Hawley, D. M., Martin, L. B., Moore, J., … White, L. (2016a). Host behaviour – parasite feedback: an essential link between animal behaviour and disease ecology. Proceedings of the Royal Society B, 283, 20153078.

Ezenwa, V. O., Etienne, R. S., Luikart, G., Beja-Pereira, A., & Jolles, A. E. (2010). Hidden consequences of living in a wormy world: nematode‐induced immune suppression facilitates tuberculosis invasion in African buffalo. The American Naturalist, 176(5), 613–624.

Ezenwa, V. O., & Jolles, A. E. (2011). From host immunity to pathogen invasion: The effects of helminth coinfection on the dynamics of microparasites. Integrative and Comparative Biology, 51(4), 540–551. Farah, I., Börjesson, A., Kariuki, T., Yole, D., Suleman, M., Hau, J., & Carlsson, H.-E. (2003). Morbidity and immune response to natural schistosomiasis in baboons ( Papio anubis). Parasitology Research, 91(4), 344–8.

123

Fernandez-Duque. (2009). Natal dispersal in monogamous owl monkeys (Aotus azarai) of the Argentinean Chaco. Behaviour, 146(4), 583–606.

Finch, C. E., & Rose, M. (1995). The quarterly of of life history evolution, 70(1), 1–52.

Foerster, S., Kithome, K., Cords, M., & Monfort, S. L. (2015). Social status and helminth infections in female forest guenons (Cercopithecus mitis). American Journal of Physical Anthropology, 158(1), 55–66.

Foo, Y. Z., Nakagawa, S., Rhodes, G., & Simmons, L. W. (2016). The effects of sex hormones on immune function: A meta-analysis. Biological Reviews, 551–571.

Fournier, D. A., Skaug, H. J., Ancheta, J., Ianelli, J., Magnusson, A., Maunder, M., … Sibert, J. (2012). AD Model Builder: Using Automatic Differentiation for Statistical Inference of Highly Parameterized Complex Nonlinear Models. Optim Methods Softw, 27, 233–249.

Freeland, A. W. J. (2009). Pathogens and the Evolution of Primate Sociality. Biotropica, 8(1), 12–24.

Friant, S., Ziegler, T. E., & Goldberg, T. L. (2016). Primate reinfection with gastrointestinal parasites: Behavioural and physiological predictors of parasite acquisition. Animal Behaviour, 117, 105–113.

Fumagalli, M., Sironi, M., Pozzoli, U., Ferrer-Admettla, A., Pattini, L., & Nielsen, R. (2011). Signatures of environmental genetic adaptation pinpoint pathogens as the main selective pressure through human evolution. PLoS Genetics, 7(11).

Gaetano, T. J., Danzy, J., Mtshali, M. S., Theron, N., Schmitt, C. A., Grobler, J. P., … Turner, T. R. (2014). Mapping Correlates of Parasitism in Wild South African Vervet Monkeys ( Chlorocebus aethiops ). South African Journal of Wildlife Research, 44(1), 56–70.

Gems, D. (2000). Longevity and ageing in parasitic and free-living nematodes. Biogerontology, 1(4), 289–307.

Gesquiere, L. R., Altmann, J., Khan, M. Z., Couret, J., Yu, J. C., Endres, C. S., … Wango, E. O. (2005). Coming of age: Steroid hormones of wild immature baboons (Papio cynocephalus). American Journal of Primatology, 67(1), 83–100.

Gesquiere, L. R., Khan, M., Shek, L., Wango, T. L., Wango, E. O., Alberts, S. C., & Altmann, J. (2008). Coping with a challenging environment: Effects of seasonal variability and reproductive status on glucocorticoid concentrations of female baboons (Papio cynocephalus). Hormones and Behavior, 54(3), 410–416.

Gesquiere, L. R., Onyango, P. O., Alberts, S. C., & Altmann, J. (2011). Endocrinology of year-

124

round reproduction in a highly seasonal habitat: Environmental variability in testosterone and glucocorticoids in baboon males. American Journal of Physical Anthropology, 144(2), 169– 176.

Gesquiere, L. R., Ziegler, T. E., Chen, P. A., Epstein, K. A., Alberts, S. C., & Altmann, J. (2014). General and Comparative Endocrinology Measuring fecal testosterone in females and fecal estrogens in males : Comparison of RIA and LC / MS / MS methods for wild baboons ( Papio cynocephalus ). General and Comparative Endocrinology, 204, 141–149.

Ghai, R. R., Fugère, V., Chapman, C. A., Goldberg, T. L., & Davies, T. J. (2015). Sickness behaviour associated with non-lethal infections in wild primates. Proceedings. Biological Sciences / The Royal Society, 282(1814), 20151436-.

Gillespie, T. R. (2006). Noninvasive assessment of gastrointestinal parasite infections in free- ranging primates. International Journal of Primatology, 27(4), 1129–1143.

Gillespie, T. R., Greiner, E. C., & Chapman, C. A. (2005). Gastrointestinal parasites of the colobus monkeys of Uganda. The Journal of Parasitology, 91(3), 569–573.

Gillespie, T. R., Lonsdorf, E. V., Canfield, E. P., Meyer, D. J., Nadler, Y., Raphael, J., … Travis, D. A. (2010). Demographic and ecological effects on patterns of parasitism in eastern chimpanzees (Pan troglodytes schweinfurthii) in Gombe National Park, Tanzania. American Journal of Physical Anthropology, 143(4), 534–544.

Girgis, N. M., Gundra, U. M., & Loke, P. (2013). Immune Regulation during Helminth Infections. PLoS Pathogens, 9(4), 11–13.

Gotoh, S. (2000). Regional differences in the infection of wild Japanese macaques by gastrointestinal helminth parasites. Primates, 41(July), 291–298.

Gould, L., & Ziegler, T. E. (2007). Variation in fecal testosterone levels, inter-male aggression, dominance rank and age during mating and post-mating periods in wild adult male ring- tailed lemurs (Lemur catta). American Journal of Primatology, 69, 1325–1339.

Graham, A. L. (2008). Ecological rules governing helminth-microparasite coinfection. Proceedings of the National Academy of Sciences of the United States of America, 105, 566–570.

Greenwood, B. Y. P. J. (1980). Mating Systems , Philopatry and Dispersal in Birds and Mammals, (1960), 1140–1162.

Gulland, F. M. D. (1995). Impact of infectious diseases on wild animal populations: a review. In B. T. Grenfell & A. P. Dobson (Eds.), Ecology of infectious diseases in natural populations (pp. 20– 51). Cambridge: Cambridge University Press.

125

Habig, B., & Archie, A. (2015). Social status, immune response and parasitism in males: a meta- analysis. Philosophical Transactions of the Royal Society B, 370, 20140109.

Hahn, N. E., Proulx, D., Muruthi, P. M., Alberts, S., & Altmann, J. (2003). Gastrointestinal parasites in free-ranging Kenyan baboons (Papio cynocephalus and P.anubis). International Journal of Primatology, 24(2), 271–279.

Hämäläinen, A., Raharivololona, B., Ravoniarimbinina, P., & Kraus, C. (2015). Host sex and age influence endoparasite burdens in the gray mouse lemur. Frontiers in Zoology, 12(1), 25. Hamilton, W. D., & Zuk, M. (1982). Heritable true fitness and bright birds: A role for parasites? Science, 218(ii), 384–387.

Hamilton, W. J., & Bulger, J. B. (1990). Natal male baboon rank rises and successful challenges to resident alpha males. Behavioral Ecology and Sociobiology, 26(5), 357–362.

Harewood, W. J., Gillin, a, Hennessy, a, Armitstead, J., Horvath, J. S., & Tiller, D. J. (2000). The effects of the menstrual cycle, pregnancy and early lactation on haematology and plasma biochemistry in the baboon (Papio hamadryas). Journal of Medical Primatology, 29(6), 415–20.

Hausfater, G. (1975). Dominance and reproduction in baboons (Papio cynocephalus). A quantitative analysis. Contrib. Primatology, 7, 1–150.

Hausfater, G., & Watson, D. F. (1976). Social and reproductive correlates of parasite ova emissions by baboons. Nature, 262, 688–689.

Hawley, D. M., & Altizer, S. M. (2011). Disease ecology meets ecological immunology: Understanding the links between organismal immunity and infection dynamics in natural populations. Functional Ecology, 25(1), 48–60.

Hernández-Bello, R., Escobedo, G., Guzmán, C., Ibarra-Coronado, E. G., López-Griego, L., & Morales-Montor, J. (2010). Immunoendocrine host-parasite interactions during helminth infections: From the basic knowledge to its possible therapeutic applications: Review Article. Parasite Immunology, 32(9–10), 633–643.

Hirschenhauser, K., & Oliveira, R. F. (2006). Social modulation of androgens in male vertebrates: Meta-analyses of the challenge hypothesis. Animal Behaviour, 71(2), 265–277.

Holekamp, K. E., & Smale, L. (1998). Dispersal status influences hormones and behavior in the male spotted hyena. Hormones and Behavior, 33(3), 205–216.

Horii, Y., Imada, I., Yanagida, T., Usui, M., & Mori, A. (1982). Parasite changes and their influence on the body weight of Japanese monkeys (Macaca fuscata fuscata) of the Koshima troop. Primates, 23(3), 416–431.

126

Hotez, P. J., Brindley, P. J., Bethony, J. M., King, C. H., Pearce, E. J., & Jacobson, J. (2008). Helminth infections: the great neglected tropical diseases. Journal of Clinical Investigation, 118(4).

Houdijk, J. G., Jessop, N. S., & Kyriazakis, I. (2001). Nutrient partitioning between reproductive and immune functions in animals. The Proceedings of the Nutrition Society, 60, 515–525.

Huck, M., Petra, L., Heymann, E. W., & Heistermann, M. (2005). Characterization and Social Correlates of Fecal Testosterone and Cortisol Excretion in Wild Male Saguinus mystax. Internation Journal of Primatology, 26(1).

Hudson, P., & Dobson, A. P. (1995). Macroparasites: observed patterns. In B. T. Grenfell & A. . Dobson (Eds.), Ecology of infectious diseases in natural populations (pp. 144–176). Cambridge: Cambridge University Press.

Hudson, P. J., & Dobson, A. P. (1992). Do Parasites make Prey Vulnerable to Predation ? Red Grouse and Parasites. Journal of Animal Ecology, 61(3), 681–692.

Huffman, M. A., Gotoh, S., Turner, L. A., Hamai, M., & Yoshida, K. (1997). Seasonal trends in intestinal nematode infection and medicinal plant use among chimpanzees in the Mahale Mountains, Tanzania. Primates, 38(2), 111–125.

Hutchings, M. R., Kyriazakis, I., Papachristou, T. G., Gordon, I. J., & Jackson, F. (2000). The herbivores’ dilemma: trade-offs between nutrition and parasitism in foraging decisions. Oecologia, 124(2), 242–251.

Isbell, L. A., & Jack, K. M. (2009). Dispersal in primates: advancing an individualized approach. Behaviour, 146(4), 429–436.

IUCN. (2016). The IUCN Red List of Threatened Species. http://www.iucnredlist.org

Jack, K. M., Schoof, V. A. M., Sheller, C. R., Rich, C. I., Klingelhofer, P. P., Ziegler, T. E., & Fedigan, L. M. (2014). Hormonal correlates of male life history stages in wild white- faced capuchin monkeys (Cebus capucinus). General Comparative Endocrinology, 195(4), 58–67.

Jackson, J.A., Friberg, I.M., Little, S., Bradley, J.E., 2009. Review series on helminths, immune modulation and the hygiene hypothesis: Immunity against helminths and immunological phenomena in modern human populations: Coevolutionary legacies? Immunology 126, 18– 27.

Johnson, M. L., & Gaines, M. S. (1990). Evolution of dispersal - theoretical models and empirical tests using birds and mammals. Annual Review of Ecology, Evolution, and Systematics, 21, 449– 480.

127

Jolles, A. E., Ezenwa, V. O., Etienne, R. S., Turner, W. C., & Olff, H. (2008). Interactions between macroparasites and microparasites drive infection patterns in free-ranging African buffalo. Ecology, 89, 2239–2250.

Kahlenberg, S. M., Thompson, M. E., Muller, M. N., & Wrangham, R. W. (2008). Immigration costs for female chimpanzees and male protection as an immigrant counterstrategy to intrasexual aggression. Animal Behaviour, 76(5), 1497–1509.

Kappeler, P. ., & Watts, D. P. (2012). Long-term field studies of primates. (P. M. Kappeler & D. P. Watts, Eds.). New York: Springer Science and Business Media.

Kappeler, P. M., Cremer, S., & Nunn, C. L. (2015). Sociality and health: impacts of sociality on disease susceptibility and transmission in animal and human societies. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 370, 20140116.

Katz, M., Despommier, D., & Gwadz, R. W. (1989). Parasitic Diseases (2nd Editio). New York: Springer-Verlag.

Kaur, T., & Singh, J. (2009). Primate-parasitic zoonoses and anthropozoonoses: a literature review. In M. Huffman & C. Chapman (Eds.), Primate parasite ecology (pp. 199–230). Cambridge: Cambridge University Press.

Khan, M. Z., Altmann, J., Isani, S. S., & Yu, J. (2002). A matter of time : evaluating the storage of fecal samples for steroid analysis. General and Comparative Endocrinology, 128, 57–64.

Kilpatrick, A. M., & Altizer, S. (2012). Disease ecology. Nature Education Knowledge, 3, 55.

Klein, S. L. (2000). Hormones and mating system affect sex and species differences in immune function among vertebrates. Behavioural Processes, 51(1–3), 149–166.

Klein, S. L. (2004). Hormonal and immunological mechanisms mediating sex differences in parasite infection. Parasite Immunology, 26(6–7), 247–264.

Koski, K. G., & Scott, M. E. (2001). Gastrointestinal nematodes, nutrition and immunity: Breaking the Negative Spiral. Annual Review of Nutrition, 21(1), 297–321.

Koski, K. G., Su, Z., & Scott, M. E. (1999). Energy deficits suppress both systemic and gut immunity during infection. Biochem Biophys Res Commun, 264(3), 796–801.

Lawson Handley, L. J., & Perrin, N. (2007). Advances in our understanding of mammalian sex- biased dispersal. Molecular Ecology, 16(8), 1559–1578.

Lilly, A. A., Mehlman, P. T., & Doran, D. (2002). Intestinal parasites in gorillas, chimpanzees, and

128

humans at Mondika research site, dzanga-ndoki national park, Central African Republic. International Journal of Primatology, 23(3), 555–573.

Llyod, S. (1995). Enviornmental influences on host immunity. In B. T. Grenfell & P. Dobson, A (Eds.), Ecology of infectious diseases in natural populations (pp. 327–361). Cambridge.

Long, KZ, Nanthakumar, N. (2004). Energetic and nutritional regulation of the adaptive immune response and trade-offs in ecological immunology. American Journal of Human Biology, 16, 499–507.

Lynch, J. W., Khan, M. Z., Altmann, J., Njahira, M. N., & Rubenstein, N. (2003). Concentrations of four fecal steroids in wild baboons : short-term storage conditions and consequences for data interpretation. General and Comparative Endocrinology, 132, 264–271.

Lynch, J. W., Ziegler, T. E., & Strier, K. B. (2002). Individual and seasonal variation in fecal testosterone and cortisol levels of wild male tufted capuchin monkeys, Cebus apella nigritus. Hormones and Behavior, 41, 275–287.

Macdonald, A. S., Araujo, M. I., Edward, J., & Pearce, E. J. (2002). Immunology of Parasitic Helminth Infections Immunology of Parasitic Helminth Infections. Infection and Immunity, 70(2), 427–433.

Machida, M., Araski, J., Koyoma, Z., Kumada, M., Horii, Y., Mada, I., … Tiba, T. (1978). The life cycle of Streptopharagus pigmentatus (Nematoda, Spiruroidea) from the Japanese monkey. Zoology, 4, 1–9.

MacIntosh, A. J. J., Hernandez, A. D., & Huffman, M. A. (2010). Host age, sex, and reproductive seasonality affect nematode parasitism in wild Japanese macaques. Primates, 51(4), 353–364.

MacIntosh, A. J. J., Jacobs, A., Garcia, C., Shimizu, K., Mouri, K., Huffman, M. A., & Hernandez, A. D. (2012). Monkeys in the Middle: Parasite Transmission through the Social Network of a Wild Primate. PLoS ONE, 7(12), 15–21.

Maizels, R. M., Balic, A., Allen, J. E., Taylor, M. D., & Allen, J. E. (2004). Helminth parasites : masters of regulation . Immunol Rev Helminth parasites – masters of regulation. Immunology Reviews, 201, 89–116.

Markham, A. C., Gesquiere, L. R., Alberts, S. C., & Altmann, J. (2015). Optimal group size in a highly social mammal. Proceedings of the National Academy of Sciences of the United States of America, 112(48), 14882–7.

Marshall, A. J., & Hohmann, G. (2005). Urinary Testosterone Levels of Wild Male Bonobos ( Pan paniscus ) in the Lomako Forest , Democratic Republic of Congo. American Journal of Primatology, 92(September 2003), 87–92.

129

Martin, L. B., Weil, Z. M., & Nelson, R. J. (2006). Refining approaches and diversifying directions in ecoimmunology. Integrative and Comparative Biology, 46(6), 1030–1039.

Martin, L. B., Weil, Z. M., & Nelson, R. J. (2008). Seasonal changes in vertebrate immune activity: mediation by physiological trade-offs. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 363(1490), 321–339.

Martinez-mota, R. (2015). The effects of habitat disturbance, host traits, and host physiology on patterns of gastrointestinal parasite infection in black howler monkeys (alouatta pigra). University of Illinois at Urbana.

Marty, P. R., Hodges, K., Heistermann, M., Agil, M., & Engelhardt, A. (2017). Is social dispersal stressful? A study in male crested macaques (Macaca nigra). Hormones and Behavior, 87, 62– 68.

Mcgrew, W. C., Tutin, C., & Collins, D. (1989). Intestinal Parasites of Sympatric Pan-Troglodytes and Papio Spp At 2 Sites. American Journal of Primatology, 155, 147–155.

McPherson, F. J. (2013). Normal Blood Parameters, Common Diseases and Parasites Affecting Captive Non-human Primates. Journal of Primatology, 2(2), 112

Meade, B. J. (1983). Host-parasite dynamics among Amboseli baboons (Papio Cynocephalus). Virgina Polytechnic Institute and State University.

Mignatti, A., Boag, B., & Cattadori, I. M. (2016). Host immunity shapes the impact of climate changes on the dynamics of parasite infections. Proceedings of the National Academy of Sciences of the United States of America, 1–6.

Monteiro, R. V., Dietz, J. M., & Jansen, A. M. (2010). The impact of concomitant infections by Trypanosoma cruzi and intestinal helminths on the health of wild golden and golden- headed lion tamarins. Research in Veterinary Science, 89(1), 27–35.

Moreau, E., & Chauvin, A. (2010). Immunity against helminths: Interactions with the host and the intercurrent infections. Journal of Biomedicine and Biotechnology, 2010.

Muehlenbein, M. P. (2006). Intestinal parasite infections and fecal steroid levels in wild chimpanzees. American Journal of Physical Anthropology, 130(4), 546–550.

Muehlenbein, M. P., & Bribiescas, R. G. (2005). Testosterone-mediated immune functions and male life histories. American Journal of Human Biology, 17(5), 527–558.

Muehlenbein, M. P., & Watts, D. P. (2010). The costs of dominance: testosterone, cortisol and intestinal parasites in wild male chimpanzees. BioPsychoSocial Medicine, 4(1), 21.

130

Müller-Graf, C. D. M., Collins, D. A., & Woolhouse, M. E. J. (1996). Intestinal parasite burden in five troops of olive baboons (Papio cynocephalus anubis) in Gombe Stream National Park, Tanzania. Parasitology, 112(5), 489–497.

Muller, M. N., & Wrangham, R. W. (2004). Dominance, aggression and testosterone in wild chimpanzees: A test of the “challenge hypothesis.” Animal Behaviour, 67(1), 113–123.

Muriuki, S. M. K., Murugu, R. K., Munene, E., Karere, G. M., & Chai, D. C. (1998). Some gastro- intestinal parasites of zoonotic (public health) importance commonly observed in old world non-human primates in Kenya. Acta Tropica, 71(1), 73–82.

Murray, S., Stem, C., Boudreau, B., & Goodall, J. (2000). Intestinal Parasites of Baboons ( Papio cynocephalus anubis ) and Chimpanzees ( Pan troglodytes ) in Gombe National Park Published by : American Association of Zoo Veterinarians Your use of the JSTOR archive indicates your acceptance of JSTOR ’ s Terms an. Journal of Zoo and Wildlife Medicine, 31(2), 176–178. Nava-Castro, K., Muñiz-Hernández, S., Hernández-Bello, R., & Morales-Montor, J. (2011). The neuroimmunoendocrine network during worm helminth infections. ISJ, 8, 143–152.

Nelson, R. (2011). An introduction to Behavioral Endocrinology (Fourth). Sunderland: Sinauer Associates, Inc.

Nguyen, N., Fashing, P. J., Boyd, D. A., Barry, T. S., Burke, R. J., Goodale, C. B., … Knauf, S. (2015). Fitness impacts of tapeworm parasitism on wild gelada monkeys at Guassa, Ethiopia. American Journal of Primatology, 77(5), 579–594.

Norris, K., & Evans, M. R. (2000). Ecological immunology : life history trade-offs and immune defense in birds. Behavioral Ecology, 11(1), 19–26.

Nunes, S., Muecke, E., Anthony, J. A., & Batterbee, A. S. (1999). Endocrine and Energetic Mediation of Play Behavior in Free-Living Belding ’ s Ground Squirrels. Hormones and Behavior, 165, 153–165.

Núñez, J., Núñez, E., Bodí, V., Sanchis, J., Miñana, G., Mainar, L., … Llàcer, A. (2008). Usefulness of the neutrophil to lymphocyte ratio in predicting long-term mortality in ST segment elevation myocardial infarction. The American Journal of Cardiology, 101(6), 747–752.

Nunn, C. L. (2002). A comparative study of leukocyte counts and disease risk in primates. Evolution; International Journal of Organic Evolution, 56(1), 177–190.

Nunn, C. L., & Altizer, S. (2006). Infectious diseases in primates: behavior, ecology and evolution. Oxford: Oxford University Press.

131

Nunn, C. L., Altizer, S., Sechrest, W., Jones, K. E., Barton, R. a, & Gittleman, J. L. (2004). Parasites and the evolutionary diversification of primate clades. The American Naturalist, 164 Suppl(november), S90–S103.

Nunn, C. L., Gittleman, J. L., & Antonovics, J. (2000). Promiscuity and the Primate Immune System. Science, 290(5494), 1168–1170.

Nunn, C. L., Thrall, P. H., & Kappeler, P. M. (2014). Shared resources and disease dynamics in spatially structured populations. Ecological Modelling, 272, 198–207.

Obanda, V., Omondi, G. P., & Chiyo, P. I. (2014). The influence of body mass index, age and sex on inflammatory disease risk in semi-captive chimpanzees. PLoS ONE, 9(8), 37–41.

Olson, M. V, & Varki, A. (2003). Sequencing the chimpanzee genome: insights into human evolution and disease. Nature Reviews. Genetics, 4, 20–28. Onyango, P. O., Gesquiere, L. R., Altmann, J., & Alberts, S. C. (2013). Puberty and dispersal in a wild primate population. Hormones and Behavior, 64(2), 240–249.

Onyango, P. O., Gesquiere, L. R., Altmann, J., & Alberts, S. C. (2014). Testosterone positively associated with both male mating effort and paternal behavior in savannah baboons (Papio cynocephalus). Hormones and Behavior, 63(3), 430–436.

Onyango, P. O., Gesquiere, L. R., Wango, E. O., Alberts, S. C., & Altmann, J. (2008). Persistence of maternal effects in baboons: Mother’s dominance rank at son’s conception predicts stress hormone levels in subadult males. Hormones and Behavior, 54(2), 319–324.

Ostner, J., Kappeler, P. M., & Heistermann, M. (2002). Seasonal variation and social correlates of androgen excretion in male redfronted lemurs ( Eulemur fulvus rufus ), 485–495.

Paradis, E., Baillie, S. R., Sutherland, W. J., & Gregory, R. D. (1998). Patterns of natal and breeding dispersal in birds. Journal of Animal Ecology, 67(4), 518–536.

Parr, A., Fedigan, M., & Kutz, J. (2013). Predictors of Parasitism in Wild White-Faced Capuchins (Cebus capucinus). International Journal of Primatology, 34(6), 1137–1152.

Patterson, J. E. H., & Ruckstuhl, K. E. (2013). Parasite infection and host group size: a meta- analytical review. Parasitology, 1–11.

Pedersen, A. B., & Fenton, A. (2007). Emphasizing the ecology in parasite community ecology. Trends in Ecology and Evolution, 22, 133–139.

Pedersen, S., & Saeed, I. (2000). Experimental infection of pigs with three dose levels of Trichuris suis. Parasite (Paris, France), 7(4), 275–281.

132

Pedersen, S., Saeed, I., Michaelsen, K. F., Friis, H., & Murrell, K. D. (2002). Impact of protein energy malnutrition on Trichuris suis infection in pigs concomitantly infected with Ascaris suum. Parasitology, 124(Pt 5), 561–568.

Peig, J., & Green, A. J. (2009). New perspectives for estimating body condition from mass/length data: The scaled mass index as an alternative method. Oikos, 118(12), 1883–1891.

Pietrock, M., & Marcogliese, D. J. (2003). Free-living endohelminth stages: At the mercy of environmental conditions. Trends in Parasitology, 19(7), 293–299.

Pinc, K. O., Altmann, J., Alberts, S. C., & Gorgon, J. B. (2016). Technical Specifications for the Amboseli Baboon Project Data Management System.

Poirotte, C., Basset, D., Willaume, E., Makaba, F., Kappeler, P. M., & Charpentier, M. J. E. (2016). Environmental and individual determinants of parasite richness across seasons in a free- ranging population of Mandrills (Mandrillus sphinx). American Journal of Physical Anthropology, 159(3), 442–456.

Post, D. G. (1981). Activity patterns of yellow baboons (Papio cynocephalus) in the Amboseli national park, Kenya. Animal Behaviour, 29, 357–374.

Pusey, A. (1987). Sex-biased dispersal and inbreeding avoidance in birds and mammals. Trends in Ecology and Evolution, 2(10), 295–299.

Pusey, A. E., & Packer, C. (1987). The Evolution of Sex-Biased Dispersal in Lions. Behaviour, 101(4), 275–310.

Rafalinirina, H. A., Aivelo, T., Wright, C. P., & Randrianasy, J. (2015). Comparison of parasitic infections and body condition in rufous mouse lemurs (microcebus rufus) at Ranomafana National Park, southeast Madagascar. Madagascar Conservation and Development, 10(2), 60– 66.

Rahman, W. A., & Collins, G. H. (1992). An association of faecal egg counts and prolactin concentrations in sera of periparturient Angora goats. Veterinary Parasitology, 43(1–2), 85–91.

Reich, D., Nalls, M. A., Kao, W. H. L., Akylbekova, E. L., Tandon, A., Patterson, N., … Wilson, J. G. (2009). Reduced neutrophil count in people of African descent is due to a regulatory variant in the Duffy antigen receptor for chemokines gene. PLoS Genetics, 5(1).

Rifkin, J. L., Nunn, C. L., & Garamszegi, L. Z. (2012). Do animals living in larger groups experience greater parasitism? A meta-analysis. The American Naturalist, 180(1), 70–82.

Roberts, M. L., Buchanan, K. L., & Evans, M. R. (2004). Testing the immunocompetence handicap hypothesis: A review of the evidence. Animal Behaviour, 68(2), 227–239.

133

Romano, C., Jiménez, P., Miranda-Brito, C., & Valdez, R. A. (2015). Parasites and steroid hormones: corticosteroid and sex steroid synthesis, their role in the parasite physiology and development. Frontiers in Neuroscience, 9(June), 224.

Ronce, O., & Clobert, J. (2012). Dispersal Ecology and Evolution. In J. Clobert, M. Baguette, T. Benton, & J. Bullock (Eds.) (pp. 119–138). Oxford: Oxford University Press.

Rothman, J. M., Pell, A. N., & Bowmanf, D. D. (2008). Host-Parasite Ecology of the Helminths in Mountain Gorillas. Journal of Parasitology, 94(4), 834–840.

Rowe, A., McMaster, K., Emery, D., & Sangster, N. (2008). Haemonchus contortus infection in sheep: Parasite fecundity correlates with worm size and host lymphocyte counts. Veterinary Parasitology, 153(3–4), 285–293.

Ryan, S. J., Brashares, J. S., Walsh, C., Milbers, K., Kilroy, C., & Chapman, C. a. (2012). A survey of gastrointestinal parasites of olive baboons (Papio anubis) in human settlement areas of Mole National Park, Ghana. The Journal of Parasitology, 98(4), 885–8.

Sakkas, P. (2011). Nutritional sensitivity of periparturient breakdown of immunity to gastrointestinal nematode parasites in mammals.

Sandland, G. J., & Minchella, D. J. (2003). Costs of immune defense: An enigma wrapped in an environmental cloak? Trends in Parasitology, 19(12), 571–574.

Sapolsky, R. M. (2004). Social Status and Health in Humans and Other Animals. Annual Review of Anthropology, 33(1), 393–418.

Sapolsky, R. M., Romero, L. M., & Munck, a. U. (2000). How Do Glucocorticoids Influence Stress Responses ? Preparative Actions *. Endocrine Reviews, 21(April), 55–89.

Schmid-Hempel, P. (2017). The Parasites and Their Social hosts. Trends in Parasitology, In press.

Seivwright, L. J., Redpath, S. M., Mougeot, F., Watt, L., & Hudson, P. J. (2004). Faecal egg counts provide a reliable measure of Trichostrongylus tenuis intensities in free-living red grouse Lagopus lagopus scoticus. Journal of Helminthology, 78(1), 69–76.

Selva, N., Cortés-Avizanda, A., Lemus, J. a, Blanco, G., Mueller, T., Heinrich, B., & Donázar, J. a. (2011). Stress associated with group living in a long-lived bird. Biology Letters, 7(4), 608–610.

Semple, S., Cowlishaw, G., & Bennett, P. M. (2002). Immune system evolution among anthropoid primates: parasites, injuries and predators. Proceedings of the Royal Society B-Biological Sciences, 269(1495), 1031–1037.

134

Setchell, J. M., Bedjabaga, I. Ben, Goossens, B., Reed, P., Wickings, E. J., & Knapp, L. A. (2007). Parasite prevalence, abundance, and diversity in a semi-free-ranging colony of Mandrillus sphinx. International Journal of Primatology, 28(6), 1345–1362.

Sheldon, B. C., & Verhulst, S. (1996). Ecological immunology - costly parasite defenses and trade- offs in evolutionary ecology. Trends in Ecology & Evolution, 11(96), 317–321.

Snaith, T. V., Chapman, C. A., Rothman, J. M., & Wasserman, M. D. (2008). Bigger groups have fewer parasites and similar cortisol levels: A multi-group analysis in red colobus monkeys. American Journal of Primatology, 70(11), 1072–1080.

Springer, A., Fichtel, C., Calvignac-Spencer, S., Leendertz, F. H., & Kappeler, P. M. (2015). Hemoparasites in a wild primate: Infection patterns suggest interaction of Plasmodium and Babesia in a lemur species. International Journal for Parasitology: Parasites and Wildlife, 4(3), 385–395

Springer, A., & Kappeler, P. M. (2016). Intestinal parasite communities of six sympatric lemur species at Kirindy Forest, Madagascar. Primate Biology, 3(2), 51–63.

Stear, M. J., Bairden, K., Bishop, S. C., Gettinby, G., McKellar, Q. A., Park, M., … Wallace, D. S. (1998). The proceses influencing the distribution of parasitic nematodes among naturally infected lambs. Parasitology, 117, 165–171.

Stear, M. J., Bishop, S. C., Duncan, J. L., McKellar, Q. A., & Murray, M. (1995). The repeatability of faecal egg counts, peripheral eosinophil counts, and plasma pepsinogen concentrations during deliberate infections with Ostertagia circumcincta. International Journal for Parasitology, 25(3), 375–380.

Stephenson, L. S., Holland, C. V., & Cooper, E. S. (2000). The public health significance of Trichuris trichiura. Parasitology, 121(2000), S73–S95.

Strier, K., & Ziegler, T. (2000). Lack of pubertal influences on female dispersal in muriqui monkeys, Brachyteles arachnoides. Animal Behaviour, 59(4), 849–860.

Stuart, M. D., Greenspan, L. L., Glander, K. E., & Clarke, M. R. (1990). A coprological survey of parasites of wild mantled howling monkeys, Alouatta palliata palliata. Journal of Wildlife Diseases, 26(4), 547–549.

Stuart, M. D., & Strier, K. B. (1995). Primates and parasites:A case for multidisciplinary approach. International Journal of Primatology, 16(4), 577–593.

Taylor, L. H., Latham, S. M., & Woolhouse, M. E. (2001). Risk factors for human disease emergence. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 356(1411), 983–9.

135

Teichroeb, J. A., & Sicotte, P. (2008). Social correlates of fecal testosterone in male ursine colobus monkeys (Colobus vellerosus): The effect of male reproductive competition in aseasonal breeders. Hormones and Behavior, 54(3), 417–423.

Tompkins, D. M., & Begon, M. (1999). Parasites can regulate wildlife populations. Parasitology Today, 15(8), 311–313.

Tompkins, D. M., Dunn, A. M., Smith, M. J., & Telfer, S. (2011). Wildlife diseases: From individuals to ecosystems. Journal of Animal Ecology, 80(1), 19–38.

Trejo-Macías, G., Estrada, A., & Mosqueda Cabrera, M. Á. (2007). Survey of helminth parasites in populations of Alouatta palliata mexicana and A. pigra in continuous and in fragmented habitat in southern Mexico. International Journal of Primatology, 28(4), 931–945.

Tung, J., Charpentier, M. J. E., Garfield, D. A., Altmann, J., & Alberts, S. C. (2008). Genetic evidence reveals temporal change in hybridization patterns in a wild baboon population. Molecular Ecology, 17, 1998–2011.

Van Noordwijk, M. A., & Van Schaik, C. P. (1985). Male migration and rank acquisition in wild long-tailed macaques (Macaca fascicularis). Animal Behaviour, 33(3), 849–861.

Van Noordwijk, M. A., & Van Schaik, C. P. (2001). Career Moves : Transfer and Rank Challenge Decisions by Male Long-Tailed Macaques. Behaviour, 138(3), 359–395.

Vitone, N. D., Altizer, S., & Nunn, C. L. (2004). Body size, diet and sociality influence the species richness of parasitic worms in anthropoid primates. Evolutionary Ecology Research, 6(2), 183– 199.

Warrington, R., Watson, W., Kim, H. L., & Antonetti, F. R. (2011). An introduction to immunology and immunopathology. Allergy, Asthma & Clinical Immunology, 7(Suppl 1), S1.

Webster Marketon, J. I., & Glaser, R. (2008). Stress hormones and immune function. Cellular Immunology, 252(1–2), 16–26.

WHO. (2016). Zoonoses. http://www.who.int/zoonoses/vph/en/

Wickings, E. J., & Dixson, A. F. (1992). Testicular function, secondary sexual development, and social status in male mandrills (Mandrillus sphinx). Physiology and Behavior, 52(5), 909–916.

Wilson, K., Bjornstad, O. N., Dobson, A. P., Merler, S., Poglayen, G., Randolph, S. E., … Skorping, A. (2002). Heterogeneities in macroparasite infections: patterns and processes. In P. Hudson, A. Rizzoli, B. T. Grenfell, H. Heersterbeek, & A. P. Dobson (Eds.), The Ecology of wildlife diseases (pp. 6–44). New York: Oxford University Press.

136

Wingfield, J. C., Hegner, R. E., Dufty, Jr., A. M., & Ball, G. F. (1990). The “Challenge Hypothesis”: Theoretical Implications for Patterns of Testosterone Secretion, Mating Systems, and Breeding Strategies. The American Naturalist, 136(6), 829.

Winsberg, G. R., Sonnenschein, E., Dyer, a R., Schnadig, V., & Bonilla, E. (1975). Prevalence of intestinal parasites in Latino residents of Chicago. American Journal of Epidemiology, 102(6), 526–32.

Wobeser, G. A. (2006). Essentials of Disease in Wild Animals. Oxford: Blackwell Publishing Proffessional.

Zuk, M., & McKean, K. a. (1996). Sex differences in parasitic infections: patterns and processes. International Journal of Parasitology, 26(10), 1009–1024.

137

Biography

Mercy Yvonne Akinyi was born in 1981 in Nairobi, Kenya, Eastern Africa. She grew up in Kenya and attended the University of Nairobi from 2000 to 2005 graduating at the top of her class with a Bachelors in Veterinary Medicine (BVM). She then went on to undertake her Master of Science in Medical Physiology (Msc.) degree at the same university from 2006 to 2010. While undertaking her degree, Mercy joined the Institute of Primate Research (IPR) as a veterinarian and an assistant research scientist. She did her master’s thesis research in collaboration with the Amboseli Baboon Research Project and IPR. The results from this thesis were published in Animal Behavior; Akinyi et al.

2013 in a manuscript titled “The role of grooming in reducing tick load in wild baboons”. While at IPR she received the Government of Kenya, National Council Of

Science And Technology, Science, Technology And Innovations (St&I) Grant. During her period at Duke University as a graduate student, she has received the following scholarships; Franklin-Mosher Baldwin fellowship award from the L.S.B. Leakey

Foundation, Nacey Fellowship research award for field studies, Graduate Student

Fellowship in Evolutionary Medicine, National Evolutionary Synthesis Center and the

Ray J. Tysor Graduate Fellowship, Duke University. She also received the James B. Duke

Fellowship and was a member of the Society of Duke Fellows (SDF). Mercy was

Awarded an honorable mention for the Founders Memorial Poster Award at the Animal

Behavior Society Meeting, New Mexico, in June 2012.

138

139