(CHRYSOCYON BRACHYURUS) HELD EX SITU by Lauren E. Reiter

Home , Little Rock Zoo, Maned wolf, Nyctereutes

UNDERSTANDING POOR REPRODUCTION IN THE MANED WOLF (CHRYSOCYON BRACHYURUS) HELD EX SITU by

Lauren E. Reiter A Thesis Submitted to the Graduate Faculty of George Mason University in Partial Fulfillment of The Requirements for the Degree of Master of Arts in Interdisciplinary Studies Interdisciplinary Studies

Understanding Poor Reproduction in the Maned Wolf (Chrysocyon brachyurus) Held Ex Situ A thesis submitted in partial fulfillment of the requirements for the degree of Master of Arts in Interdisciplinary Studies at George Mason University

Lauren E. Reiter Bachelor of Arts Miami University, 2006

Director: Michael P. Gilmore, Professor New Century College

Summer Semester 2012 George Mason University Fairfax, VA

This work is licensed under a creative commons attribution-noderivs 3.0 unported license.

DEDICATION

This is dedicated to my parents, Mark and Colleen, and my three fur children, Tigger, Lucy and Gypsy.

iii

ACKNOWLEDGEMENTS

I would like to acknowledge the many friends, relatives and supporters who have made this happen.

I would like to thank my committee, Dr. Nucharin Songsasen, Dr. Elizabeth Freeman and Dr. Gilmore, for their endless patience, guidance and support. This work was possible thanks to the generosity and support of Dr. Nucharin Songsasen, sharing her data, knowledge and experience with me.

Many thanks go out to the Smithsonian Institution and Smithsonian Conservation Biology Institute for providing a well-equipped repository for conducting laboratory analyses and to Sarah Putman, Nicole Presley, Nicole Parker and Natalia Prado-Oviedo for teaching me how to use the technology available and assisting me with laboratory troubleshooting.

I would like to acknowledge my colleagues, Amy Johnson, Mayako Fujihara, Jennifer Nagashima, Stacie Castelda, Marieke Kester, Megan Brown and Tathiana Motheo, for their support and patience.

I would like to express my endless gratitude to all the zoological facilities and staff that helped me collect data: Smithsonian Conservation Biology Institute, National Zoological Park, Front Royal, VA; White Oak Conservation Center, Yulee, FL; Connecticut’s Beardsley Zoo, Bridgeport, CT; Houston Zoo, Houston, TX; Wildlife World Zoo, Litchfield Park, AZ; Pueblo Zoo, Pueblo, CO; Louisville Zoo, Louisville, KY; Buffalo Zoo, Buffalo, NY; Birmingham Zoo, Birmingham, AL; San Antonio Zoo, San Antonio, TX; Sedgwick County Zoo, Wichita, KS; Oklahoma City Zoo; Oklahoma City, OK; Omaha’s Henry Doorly Zoo and Aquarium, Omaha, NE; Little Rock Zoo, Little Rock, AR; and Dickerson Park Zoo, Springfield, MO.

Lastly, my research would not have been possible without the continuous support and confidence from my loving parents, Mark and Colleen, and the furry assistance from my three rescued dogs, Tigger, Lucy and Gypsy.

This research was made possible by the financial support from the Maned Wolf Species Survival Plan and Smithsonian’s Women’s Committee.

TABLE OF CONTENTS

Page List of Tables ...... viii List of Figures ...... ix List of Equations ...... x List of Abbreviations or Symbols ...... xi Abstract ...... xiv I. Introduction ...... 18 II. Literature Review ...... 24 Natural History of the Maned Wolf ...... 24 Physical Description ...... 24 Free-Ranging Maned Wolves ...... 26 The North American Population of Captive Maned Wolves ...... 28 Reproductive Endocrinology of the Maned Wolf ...... 31 Non-Invasive Hormone Monitoring ...... 35 Stress ...... 36 Nutritional Management ...... 39 Leptin ...... 40 III. Reproductive and Stress Hormone Monitoring in the Captive Maned Wolf (Chrysocyon brachyurus) ...... 44 Introduction ...... 44 Hypothesis ...... 46 Objectives ...... 46 Animals and Facilities ...... 47 Sample Collection ...... 48 Hormone Extraction from Feces ...... 48 Hormone Analysis ...... 49 Fecal Estrone Conjugate (EC) Enzyme Immunoassay ...... 49

Fecal Progestagen (Pg) Enzyme Immunoassay ...... 50 Fecal Cortisol (F) Enzyme Immunoassay ...... 51 High Performance Liquid Chromatography (HPLC) ...... 51 Statistical Analysis ...... 52 Characterization of Glucocorticoid Metabolites by HPLC ...... 54 Hormone Analysis ...... 54 Fecal Progestagen Hormone Analysis ...... 54 Fecal Cortisol Hormone Analysis ...... 55 Discussion ...... 56 IV. Non-invasive Assessment of Leptin Concentrations in the Captive Maned Wolf (Chrysocyon brachyurus) ...... 70 Introduction ...... 70 Objectives ...... 74 Animals and Facilities ...... 74 Urine Samples ...... 75 Blood Samples ...... 76 Leptin Radioimmunoassay ...... 76 Creatinine Enzyme Immunoassay ...... 76 Assay Validation ...... 77 Statistical Analysis ...... 78 Serum and Urinary Leptin Assay Validation ...... 79 Measuring Urinary Leptin Concentrations ...... 79 Effect of Sex ...... 79 Effect of Body Weight, Length and Body Mass Index ...... 79 Effect of Age ...... 80 Discussion ...... 80 V. Conclusion ...... 90 Appendices ...... 96 Appendix A ...... 97 Appendix B ...... 98 Appendix C ...... 99 Appendix D ...... 100 Appendix E ...... 101

Appendix F ...... 102 References ...... 103

vii

LIST OF TABLES

Table Page Table 3.1 Summary of female maned wolves sampled with corresponding study periods analyzed and reproductive outcome...... 62 Table 3.2 Fecal progestagen concentrations (µg/g feces) across reproductive phases in female maned wolves of various reproductive outcomes. Different letters within the same reproductive phase indicate differences (P < 0.05)...... 49 Table 3.3 Overall mean ± SEM fecal cortisol concentrations (µg/g feces) for three groups of female maned wolves across reproductive phases. Different letters within the same reproductive phase indicate differences (P < 0.05)...... 51 Table 4.2 Age, leptin concentration, weight, length and BMI measurements for six male maned wolves...... 88

viii

LIST OF FIGURES

Figure Page Figure 2.1 Photographs of a maned wolf housed at the Smithsonian Conservation Biology Institute reveal the canid’s long dark legs, orange-brown coat wth a black mane (A) and large ears and white throat (B). Photo credit: Lauren Reiter ...... 26 Figure 2.2 Longitudinal profiles of fecal gonadal (estrogen and progestagen) metabolites for (A) a pregnant maned wolf from 28 days before estrogen peak to parturition (Songsasen et al., 2006) and (B) a pseudo-pregnant maned wolf from 13 days before estrogen peak to the end of the luteal phase...... 17 Figure 2.3 Diagram illustrates the mechanisms by which the hypothalamic-pituitary- adrenal (HPA) axis responds to a stressor. Adapted from Irwin and Cole (2011)...... 20 Figure 3.1 A maned wolf fecal sample after it has been lyophilized and pulverized into a fine powder. Photo credit: Lauren Reiter ...... 46 Figure 3.2 HPLC profiles of immunoreactive glucocorticoids (cortisol and corticosterone) from female maned wolf feces...... 47 Figure 3.3 Frequency distribution with number of female maned wolves that gave birth for each month of the North American maned wolf breeding season (n = 18) ...... 48 Figure 3.4 Mean (± SEM) longitudinal profiles of fecal progestagen metabolites for female maned wolves that: (A) conceived and successfully raised pups (n = 9); (B) conceived, but lost pups early post-partum (n = 7); (C) became pseudo-pregnant (n = 11); and (D) were unpaired (n = 7). Each data point is a pool of 3-day means...... 50 Figure 3.5 Mean (± SEM) longitudinal profiles of fecal cortisol metabolites for female maned wolves that: (A) conceived and successfully raised pups (n = 10); (B) conceived but lost pups early post-partum (n = 8); and (C) became pseudo-pregnant (n = 12). Each data point is a pool of 3-day means...... 52 Figure 4.1 Parallelism curves resulting from RIA of serially diluted leptin standards and pooled serum and pooled urine. Based on the above graph, urine samples should be run at neat given the binding inhibition results observed at ~50%...... 69 Figure 4.2 Maned wolf urinary leptin recovery ...... 70 Figure 4.3 Mean leptin concentrations (ng/mg Cr) for six maned wolves where urine samples were collected more than 1 year apart from the same individuals (Age A = initial sample, Age B = subsequent sample)...... 72

LIST OF EQUATIONS

Equation Page Equation 1 BMI ...... 58

LIST OF ABBREVIATIONS AND SYMBOLS

Adrenocorticotropic Hormone ...... ACTH Analysis of Variance ...... ANOVA Antibody ...... Ab Antidiuretic Hormone ...... ADH Association of Zoos and Aquariums ...... AZA Autonomic Nervous System ...... ANS Body Condition Index ...... BCI Body Condition Score ...... BCS Body Length ...... L Body Mass Index ...... BMI Body Weight ...... W Central Nervous System ...... CNS Coefficient of Determination ...... R2 Coefficient of Variation ...... CV Condition Indices ...... CIs Corticotropin-Releasing Factor ...... CRF Cortisol ...... F Counts per minute of radiation ...... cpm Creatinine ...... Cr Date of birth ...... DOB Day ...... d Degrees Celsius ...... °C Disintegrations per minute of radiation ...... dpm Enzyme Immunoassay ...... EIA Enzyme-linked Immunosorbent Assay ...... ELISA Estrogen Conjugate ...... EC Extraction Efficiency ...... EE Fatness Index ...... FI Fecal Glucocorticoid Metabolites ...... FGM Female ...... ♀ Follicle Stimulating Hormone ...... FSH Global Positioning System ...... GPS Gonadotropin-Releasing Hormone ...... GnRH Gram ...... g Greater than ...... > High Performance Liquid Chromatography ...... HPLC

History ...... Hx Human Equivalent ...... HE Hypothalamic-Pituitary-Adrenal ...... HPA Index of Condition ...... C Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renovávei ...... IBAMA International Union for Conservation of Nature ...... IUCN Iodine-125 ...... 125I Least Significant Differences ...... LSD Less than ...... < Luteinizing Hormone ...... LH Male ...... ♂ Maned Wolf Species Survival Plan® ...... MWSSP Maximum ...... Max Mean ...... � Meter ...... m Microgram per gram ...... µg/g Microliter ...... µL Milligram per milliliter ...... mg/mL Milliliter ...... mL Minimum ...... Min Nanogram ...... ng Nanogram per gram ...... ng/g Nanogram per milligram of Creatinine ...... ng/mg Cr National Zoological Park ...... NZP Non-Specific Binding ...... NSB Obese ...... ob Pearson’s Correlation Coefficient ...... r Percent Binding ...... % B Phosphate buffered saline ...... PBS Picograms per milliliter ...... pg/mL Picograms per well ...... pg/well Plus or minus ...... ± Population and Habitat Viability Assessment ...... PHVA Population Management Plan ...... PMP Probability ...... P Progestagen ...... Pg Progesterone ...... P4 Quality Control ...... QC Radioimmunoassay ...... RIA Revolutions per minute ...... RPM Sample size ...... n Single Population Animal Records Keeping Software ...... SPARKS Smithsonian Conservation Biology Institute ...... SCBI Species Survival Plan® ...... SSP

xii

Square kilometers ...... km2 Standard Deviation ...... SD Standard Error ...... SE Standard Error of the Mean ...... SEM Statistical Package for Social Sciences ...... SPSS Taxon Advisory Group ...... TAG Tritium ...... 3H Years ...... yrs

xiii

ABSTRACT

UNDERSTANDING POOR REPRODUCTION IN THE MANED WOLF (CHRYSOCYON BRACHYURUS) HELD EX SITU

Lauren E. Reiter, MAIS

George Mason University, 2012

Thesis Director: Dr. Michael P. Gilmore

Maned wolves maintained in U.S. zoos have experienced limited reproductive success and high neonatal mortality. Sub-optimal health including, cystinuria, intestinal disorders and other health problems, coupled with stress associated with captive management have been indicated as potential causes of poor reproduction in the North American ex situ population. The objectives of this study were to: 1) utilize non-invasive endocrine monitoring to assess reproductive and stress hormones in female maned wolves of various reproductive histories and 2) develop and validate a urinary leptin assay as a biomarker for nutritional status in this species.

For the first objective, data were obtained during 39 reproductive cycles (n = 28 females), 30 that were of breeding females and the remaining cycles were of individuals housed alone. Of the 30 breeding cycles, 18 were pregnant, 10 of which resulted in births of healthy pups surviving to adulthood. The remaining 12-paired females resulted in

xiv pseudo-pregnancy. Fecal samples were collected from each animal 1-3 times per week during the breeding season (October-March) or until pups were produced, and stored at

-20°C until extraction and analyses of gonadal and adrenal hormone metabolites were carried out using validated enzyme immunoassays. Differences in mean fecal steroid metabolites were determined by analysis of variance (ANOVA) followed by multiple comparisons.

Mean progestagen concentrations during the pre-ovulatory phase did not differ significantly (P > 0.05) among groups; however, pregnant females (raised pups: 20.0 ±

3.0 µg/g feces, lost pups: 22.5 ± 3.0 µg/g feces) had higher (P < 0.05) progestagen concentrations than pseudo-pregnant (10.3 ± 1.3 µg/g feces) and unpaired (4.2 ± 0.5 µg/g feces) counterparts during the peri-ovulatory phase. During the luteal phase, the highest

(P < 0.05) mean progestagen metabolite concentration was found in pregnant females that raised pups (45.8 ± 2.3 µg/g feces), followed by pregnant wolves that lost pups (43.0

± 4.0 µg/g feces) and, then, pseudo-pregnant individuals (27.5 ± 1.2 µg/g feces). Profiles of singleton female maned wolves displayed low progestagen concentrations throughout the breeding season (mean ± SEM; 4.6 ± 0.3 µg/g feces) and the hormone level was lower (P < 0.05) than those of other groups. Mean fecal cortisol concentrations during the pre-ovulatory and peri-ovulatory phases did not differ significantly (P > 0.05) among groups; however, females that cared for their pups (1.8 ± 0.1 µg/g feces) had lower (P <

0.05) corticoid concentrations during the luteal phase than pseudo-pregnant females (3.3

± 0.2 µg/g feces) and individuals that experienced neonatal loss (3.1 ± 0.3 µg/g feces).

Furthermore, females who successfully reared their young excreted the lowest corticoid concentrations across all three reproductive phases of the cycle.

For the second objective, fresh urine (34 samples from 20 maned wolves) was opportunistically collected during the 2009-10 and 2009-11 breeding seasons.

Additionally, 10 serum samples from 10 individuals were collected during routine physical examinations. Urinary and serum leptin concentrations were assessed using multi-species leptin radioimmunoassay (RIA) kit. Curves of serially diluted pooled maned wolf serum and urine exhibited parallel displacement to that of the standard curve

(1.56-50 ng/mL) demonstrating immunoactivity of endogenous antigen similar to the assay standards. Net recovery of 1.56, 3.13, 6.25, 12.5, 25 and 50 ng leptin added to 200

µL neat urine was 1.12, 0.98, 1.72, 1.97, 3.86 and 8.46 ng, respectively (y = 0.31x + 0.51,

R2 = 0.98) and indicated a gross underestimation of hormone mass. Leptin concentrations were correlated with biological measures (i.e., age, sex, body weight, etc.) obtained opportunistically. Relationships between serum/urinary leptin and body condition were determined using correlation analyses. Sex did not influence (t17 = -0.52,

P > 0.05) urinary leptin concentration in the maned wolf (males: 2.30 ± 0.44 ng/mg Cr, range: 0.35-4.29 ng/mg Cr, n = 7; females: 2.63 ± 0.41 ng/mg Cr, range: 0.80-5.32 ng/mg

Cr, n = 12). Urinary leptin concentrations increased with age for both males and females

(t5 =-2.98; P = 0.03). Body length showed a marginally significant positive correlation with urinary leptin concentration (r3 = 0.73, P = 0.09, n = 5); however, no other significant correlations were found with the remaining biological measures.

xvi

This study demonstrated that: 1) ovulation only occurs in females paired with a male, indicating that the maned wolf is an induced ovulator; 2) pseudo-pregnant females, on average, excrete lower progestagen metabolites than pregnant individuals; 3) females that lost pups or did not become pregnant exhibit elevated corticoid excretion compared to individuals producing and raising pups; and 4) urinary leptin can be detected using a multi-species leptin RIA, although there appears to be a gross underestimation of hormone mass. These findings advance understanding about the causes of poor reproductive success in captive maned wolves, which is critical for developing an improved management strategy for ex situ populations. Future studies should focus on identifying factors contributing to and alleviating stress responses in captive maned wolves, which may then help improve pregnancy rates and reduce neonatal loss.

Additional studies are also required to further explore the usefulness of urinary leptin as a biomarker for assessing nutritional status and overall health of maned wolves kept ex situ.

xvii

I. INTRODUCTION

The maned wolf (Chrysocyon brachyurus), a neotropical canid, lives in habitats severely compromised by agricultural development. This ‘near threatened’ species once thrived and ranged throughout much of South America, but is now found only in Brazil,

Argentina, Bolivia and Paraguay (Rodden et al., 2004; 2008). The main threats to wild maned wolf populations include habitat loss, conflicts with humans, diseases transmitted from domestic dogs and road mortality (Rodden et al., 2004). Because of a rapidly declining and increasingly fragmented native habitat, it is important to sustain a ‘hedge’ collection of maned wolves ex situ. Maned wolves held ex situ are a ‘genetic repository’ for retaining all existing heterozygosity, which is especially important in the case of an unexpected catastrophic event impacting wild populations. Ex situ populations also serve as ambassadors for their wild counterparts and provide an opportunity to study the biology and physiology of this unique canid. Unfortunately, attempts to breed captive maned wolves have proven unsuccessful due to problems involving low pregnancy rates and high neonatal mortality (Maia and Gouveia, 2002; Cummings et al., 2007; Songsasen and Rodden, 2010).

Maned wolves are monogamous and monestrus, experiencing one reproductive cycle per year (Kleiman, 1972; Dietz, 1984). Prime reproductive age for this species is between three and eight years of age. The breeding season spans between three and five

18 months occurring from March to June in Latin America and September to February in

North America (Maia and Gouveia, 2002). Following an approximate 65-day gestation, births occur any time between June and early September in the southern hemisphere and between mid- to late November and March in the northern hemisphere (Dietz, 1985;

Maia and Gouveia, 2002; Rodden et al., 2004).

Stress, as a result of sub-optimal housing and poor management conditions, is known to affect reproduction in zoo-managed species (Rivier and Rivest, 1991). This appears particularly prevalent in specialized carnivores such as leopard cats (Prionailurus bengalensis; Carlstead et al., 1993), clouded leopards (Neofelis nebulosa; Wielebnowski et al., 2002), cheetahs (Acinonyx jubatus; Marker-Kraus and Grisham, 1993; Jurke et al.,

1997; Terio et al., 2004), maned wolves (Kleiman, 1972; Montali and Kelly, 1989;

Songsasen et al., 2006), red wolves (Canis rufus; Rabon, Jr. and Waddell, 2010) and black-footed ferrets (Mustela nigripes; Poessel et al., 2011). In cheetahs, a negative relationship between adrenal and ovarian activity has been shown (Jurke et al., 1997).

Female cheetahs with high circulating cortisol concentrations experience less follicular activity and prolonged periods of anestrus. Additionally, cheetahs located near conspecifics may experience stress that has led to suppression of ovarian activity (Jurke et al., 1997). For male Pallas’ cats (Otocolobus manul) housed indoors, the artificial lighting simulated false photoperiods that delayed onset of the breeding season and decreased its length (Newell-Fugate et al., 2007). Additionally, management factors significantly influence the number of litters produced by small-sized felid species in

19 captivity, with those housed larger than a male-female pair resulting with poor reproductive success (Mellen, 1991).

Stress may have undesirable effects in pregnant females prior to parturition and post-partum, especially when individuals are housed in an unsatisfactory social or physical environment (Comizzoli et al., 2009; Poessel et al., 2011). In a monogamous species, such as maned wolves, pair separation can lead to maternal neglect or infanticide

(Thompson et al., 2010). In an attempt to provide appropriate housing and fulfill a species’ environmental needs, multiple dens are placed in a maned wolf’s enclosure.

Therefore, if a female experiences a perceived stressor, she has the choice and control to move her litter to another den, which discourages maternal infanticide (Thompson et al.,

2010). Thus, the degree of control an animal has on its environment greatly impacts its response to an aversive event (Mormède et al., 2007).

Stress associated with sub-optimal health can also result in poor reproductive success. Formulating optimal diets for captive maned wolves has been an ongoing challenge for maintaining this species ex situ. Free-ranging maned wolves are omnivores, ingesting approximately 51% plant and 49% animal matter, respectively

(Dietz, 1984; Motta-Junior et al., 1996). Furthermore, diet composition of wild maned wolves is seasonally variable (Dietz, 1984). In captivity, most animals are offered a more carnivorous diet (~65% animal protein) with less than ~20% plant-based foods annually.

This discrepancy between diets of zoo-housed and free-ranging maned wolves is hypothesized to be a contributing factor to sub-optimal health of this species.

Additionally, maned wolves have a very sensitive gastrointestinal tract and often respond

20 to minor dietary changes with diarrhea and poor body condition (Bush, 1980). Diets fed to captive maned wolves included foods originally formulated for domestic dogs.

However, it has been shown that there is species-specificity in digestibility (Childs-

Sanford and Angel, 2006). Specifically, maned wolves have lower digestibility of dry matter, energy and several minerals including copper, magnesium, iron and sodium than domestic dogs (Childs-Sanford and Angel, 2006). Although animal managers are careful in offering a well-balanced diet to captive maned wolves, the sub-optimal health observed in this species may be due to digestibility differences leading to an energy imbalance.

Maintaining adequate energy stores is critical for survival. Leptin, a cytokine, is secreted by large adipocytes of subcutaneous peripheral fat and regulates body weight and energy homeostasis, multiple hypothalamic-pituitary axes and a diverse array of other biological responses such as reproductive development and function (Van

Harmelen et al., 1998; Mantzoros, 2000; Henson and Castracane, 2000; Muehlenbein et al., 2003; Bouillanne et al., 2007; Greenstein and Wood, 2010). Leptin modulates timing of the first ovulation onset by signaling to the brain when a certain percentage of body fat has been reached for initiation of puberty (Mantzoros, 2000). Normal leptin levels are essential for sustaining ovulatory cycles and reproductive functions in animals and humans (Mantzoros, 2000). Serum leptin is positively correlated with body weight and body mass index in domestic dogs (Canis familiaris) and raccoons (Procyon lotor) indicating its usefulness as a marker for nutritional status in carnivores (Asikainen et al.,

2004; Shibata et al., 2005). Measuring leptin concentrations in the maned wolf can help

21 the Maned Wolf Species Survival Plan® (MWSSP) monitor nutritional status and, additionally, improve reproduction in the captive population.

The purpose of this research was to: 1) determine whether or not neonatal loss is linked to an adrenal/stress response, and 2) develop a biomarker for assessing nutritional and health status of captive maned wolves. The objectives were to utilize non-invasive fecal hormone monitoring to: 1) compare gonadal and adrenal hormone metabolites among females that gave birth and successfully raised pups to those that lost pups post- partum, pseudo-pregnant individuals and unpaired animals, 2) determine the relationship between gonadal and adrenal hormone metabolites, 3) develop and validate a noninvasive assay for measuring urinary leptin concentrations in maned wolves and 4) correlate urinary leptin with circulating leptin levels, age, sex, body weight, body condition and reproductive seasonality. The findings from this research will contribute to efforts in understanding causes of poor reproductive success in captive maned wolves.

Chapter Two describes the biology and physiology of the maned wolf, its status in the wild and the history of the North American captive maned wolf population. An examination of hormone monitoring, specifically gonadal and adrenal hormones, in zoo- managed maned wolves is explored in Chapter Three. Additionally, measuring adrenal activity in animals and using cortisol concentrations as a proxy for measuring stress are described. These techniques have been applied to captive maned wolves and a thorough analysis follows along with longitudinal profile characterization based on reproductive outcomes. The latter chapter focuses on leptin and the more recent studies focusing on animals and its potential for use as a biomarker. This is the first study to collect and

22 analyze leptin concentrations in the maned wolf. Finally, the thesis will culminate with a summary of the findings and recommendations for future studies.

II. LITERATURE REVIEW

Natural History of the Maned Wolf The maned wolf (Chrysocyon brachyurus) is the largest canid inhabiting the grasslands and scrub forest of central South America from the mouth of the Parnaiba River in north- eastern Brazil, south through the Chaco of Paraguay into Rio Grande do Sul State, Brazil, and west to the Pampas del Heath in Peru (Dietz, 1985; Rodden et al., 2004). Evidence has confirmed the presence of this species in Argentina with a reported wolf sighting in

1999 (Rodden et al., 2004). This has subsequently been corroborated by Salvatori et al.

(2004). The last account indicating the presence of maned wolves in northern Uruguay was in 1990. Unfortunately, no confirmations have been validated since, and it is possible that maned wolves have gone extinct in this country (Rodden et al., 2004).

However, the species still inhabits Argentina, Bolivia, Brazil, Paraguay and Peru with an estimated population totaling 23,000 individuals (de Paula et al., 2008; Songsasen and

Rodden, 2010).

Physical Description Maned wolves are sometimes described as a “fox on stilts” due to their fox-like appearance and character (Emmons et al., 2004; MWSSP, 2007). However, maned wolves are neither a fox nor a ‘true’ wolf, but rather exist in a separate genus,

Chrysocyon, which does not include any other canid species (Dietz, 1985; MWSSP,

2007). The maned wolf is the tallest canid standing approximately a meter tall at

24 shoulder height and weighing roughly 26 kilograms (Dietz, 1984; Emmons et al., 2004).

Maned wolves do not display considerable sexual dimorphism (MWSSP, 2007). Their pelage, with almost no observable individual variation, consists of rusty golden-red with black covering their legs, a white-tipped tail and a mane of long black-tipped hair running down their neck and shoulders (Fig. 2.1A). Their throat along with the inside of their large ears is white (Fig. 2.1B; Dietz, 1985; Emmons et al., 2004). Dietz (1984) reports significantly longer pinnae in males as compared to females. Another unique characteristic of the maned wolf is their gait. Their long legs are not well-equipped for running and are adapted for living in grasslands. Locomotion is achieved by moving the legs on the same side of the body together, rather than alternatively (MWSSP, 2007).

Their long legs, large ears, unique gait and pelage enable them to survive in the tall grasslands (Rodden et al., 2004).

Figure 2.1 Photographs of a maned wolf housed at the Smithsonian Conservation Biology Institute reveal the canid’s long dark legs, orange-brown coat with a black mane (A) and large ears and white throat (B). Photo credit: Lauren Reiter

Free-Ranging Maned Wolves Maned wolves live solitary lifestyles in low densities. Naturally, a mated pair (1♂.1♀) will defend a shared territory (Dietz, 1984). During the non-breeding season, the pair

26 lives separately, despite sharing the same home range (MWSSP, 2007). Individual paths intertwine with one another, but the pair rarely meets, except during the breeding season

(March-June) when mated pairs are often seen together and when pups are born (June-

August) (Rodden et al., 2004; MWSSP, 2007). Home range size varies among studies from 38.14 km2 (Melo et al., 2007) up to 80 km2 (Jácomo et al., 2009). This species is highly territorial using olfactory, visual and auditory mechanisms to communicate and mark territorial boundaries. Site-specific defecation spots, found on termite mounds, ant mounds, shrubs, trees and grass and any landmarks that present physical barriers, such as roads and rivers, define a wolf’s territory and is communicated via both visual and olfactory cues (Dietz, 1984). Additionally, maned wolves employ a roar-bark, as a means to communicate with their mate and to announce their location within their territory, which may serve to space individuals appropriately and avoid intraspecific aggression (Kleiman, 1972). Other vocalizations include whines and growls, which expresses the disposition of the individual (Kleiman, 1972; MWSSP, 2007).

Maned wolves are opportunistic omnivores and do not compete with humans for food (Emmons et al., 2004). Roughly half of their diet consists of plant matter and the other half comprises small animals (Jácomo et al., 2004; Childs-Sanford, 2005). Lobeira

(Solanum lycocarpum) or “Fruit of the Wolf” is a plant that produces a tomato-like fruit that is a major and consistent dietary component of free-ranging maned wolves living in

Brazil (Dietz, 1984; Allen, 1995; MWSSP, 2007). The increased presence of humans and their use of resources have impacted food sources consumed by the maned wolf. The lobeira fruit, for example, has been severely reduced due to habitat conversion of the

Cerrado for agriculture (Courtenay, 1994). Animal and plant matter consumption changes according to available resources, and these vary throughout the year (Motta-

Junior and Martins, 2002; Rodden et al., 2004). Small mammals and foliage are more often consumed during the dry season (June-September), while fruits and insects are more prevalent during the wet season (December-March) (Allen, 1995; Aragona and

Setz, 2001).

The maned wolf’s primary habitat comprises 80% of the Cerrado and is being severely reduced and replaced by pasture and agriculture (Trolle et al., 2007; Sollman et al., 2010). To date, only 20% of natural habitat remains. As the human population continues to increase, the demand for food and natural resources will further damage and fragment remaining core habitats and maned wolf territories (Karanth and Chellam,

2009).

The North American Population of Captive Maned Wolves Due to the continual decline in natural habitat and increasing threats to wild maned wolves, maintaining a captive population of this species is important. Ex situ conservation, aims to maintain sustainable populations, with zoos playing a stewardship role in creating reservoirs of wildlife that are managed to be as genetically close as possible to free-living counterparts (Comizzoli et al., 2009). The aim of management is to take a representative ‘snapshot’ of the species’ genetic diversity and preserve it (Lees and Wilcken, 2009). Zoo populations may also act as ‘insurance’ against catastrophic events or disasters that are capable of wiping out small populations occurring in the wild

(Wildt et al., 2001; Cummings et al., 2007). They provide opportunities for research such

28 as affording access to specific animals for studying behaviors, reproductive biology, health, diseases, nutrition and genetics (Hosey et al., 2009; Songsasen and Rodden,

2010). Finally, zoo populations serve as ambassadors for their wild relatives in their disappearing habitats. This is especially important for carnivores because they are highly vulnerable to environmental change and human persecution (Comizzoli et al., 2009).

Maned wolves were first imported from Brazil into the United States in 1965

(MWSSP, 2011). The Association of Zoos and Aquarium’s (AZA) Maned Wolf Species

Survival Plan® (MWSSP) population descended from 31 founders (MWSSP, 2011). The

AZA Canid and Hyaenid Taxon Advisory Group (TAG) have set the target Species

Survival Plan (SSP) population of maned wolves at 100 individuals (MWSSP, 2010).

The captive population was greatest in 1995 with a total of 97 individuals [48.49 (♂.♀)]

(M. Rodden, pers. comm.). Since 2004, the maned wolf population has gradually declined primarily due to poor reproductive success, as paired individuals have failed to reproduce and successfully rear pups (MWSSP, 2011). The current population consists of 92 individuals (46.46) housed across 30 institutions (M. Rodden, pers. comm.). One of the fundamental goals for the MWSSP program is to maintain a viable self-sustaining captive North American population. This includes maintaining gene diversity above

90%. However, due to poor reproduction (Maia and Gouveia, 2002), as well as a small and aging population, it is apparent that this goal is not achievable (M. Rodden, pers. comm.). Under the current situation, it is expected that this population can retain genetic diversity > 90% for the next five years, and only 28% for the next 100 years. Once gene diversity declines to below 90% of that in the founders, inbreeding depression could

29 significantly impact the viability of the population (Ebenhard, 1995; Lees and Wilcken,

2009).

Inbreeding has been linked to increased abnormalities, fatal mutant phenotypes, recessive deleterious mutations, decline in fitness, genetic diseases and lower fertility, survival and growth rates of individuals (Charlesworth and Willis, 2009; Rabon Jr. and

Waddell, 2010). Asa et al. (2007) report an inverse relationship between high levels of inbreeding with semen quality, as measured by percentages of motile sperm and sperm with normal morphology in the Mexican gray wolf (Canis lupus baileyi). Additionally,

Asa et al. (2007) conclude the level of defective sperm observed could depress fertility.

In a captive gray wolf (Canis lupus) population, Laikre and Ryman (1991) describe decreased juvenile weight, diminished reproductive success and reduced longevity as effects of inbreeding. Because low genetic diversity has been shown to impact animals’ health and reproductive fitness, there is an urgent need to improve reproductive capacity of captive maned wolves.

In addition to poor reproduction and an aging population, another challenge in maintaining a viable maned wolf population is sub-optimal health. Maned wolves are confronted with poor body conditions, chronically soft stools, rapid ingesta passage, gingivitis and cystinuria (Childs-Sanford and Angel, 2006). Captive breeding of maned wolves is additionally hindered by metabolic and digestive system disorders, as well as the lack of an optimal diet (Maia and Gouveia, 2002; Soler et al., 2005; Songsasen and

Rodden, 2010). As of 2010, a diet was developed specifically for maned wolves through

30 the partnership between the MWSSP and Land-O-Lakes/Mazuri. The newly formulated diet is currently being assessed (MWSSP, 2011).

Reproductive Endocrinology of the Maned Wolf The prime reproductive age for maned wolves is between three and eight years of age.

However, wolves as young as two years and as old as twelve years have produced pups

(MWSSP, 2010). As seasonal breeders, maned wolves are limited to only one opportunity for conception per season (Asa and Valdespino, 1998). Reproduction in maned wolves is affected by photoperiod (Velloso et al., 1998). Under natural photoperiods, maned wolves are seasonal breeders demonstrating gonadal activity from late September through February in North America and occurring from March to June in

Latin America (Dietz, 1985; Maia and Gouveia, 2002). This species is monogamous cohorting with its mate during the breeding season, despite living a solitary lifestyle for the remainder of the year (Kleiman, 1972). In captivity, breeding pairs [1.1 (♂.♀)] may be housed together (MWSSP, 2010). Following a gestation period of 60 to 65 days, litters produce three pups on average, although as few as one and as many as seven pups have been documented (Rodden et al., 1996; Maia and Gouveia, 2002).

Current knowledge on the canid reproductive cycle has been largely gleaned from studies on the domestic dog (Canis familiaris; Wildt et al., 1979; Concannon et al.,

1989). The reproductive cycle of canids can be divided into four distinct phases: proestrus, estrus, diestrus and anestrus. Proestrus, which is dominated by increased follicular activity, averages nine days in the domestic dog (Feldman and Nelson, 1987).

This phase is characterized by vulvar swelling and sanguineous discharge and late in

31 proestrus, females will allow mounting attempts from the male (Velloso et al., 1998). In the domestic dog, it has been shown that estrogen concentration gradually increases and reaches the peak level at the end of this period (Concannon, 2009). The increase in circulating estrogen causes extensive vaginal epithelial cell proliferation and cornification

(Concannon, 2009). The onset of estrus coincides with the decline in estrogen level and the rise of circulating progesterone concentration. The duration of estrus is highly variable in the domestic dog, ranging from 1-2 weeks with an average of nine days

(Songsasen et al., 2006; Concannon, 2011). Irrespective of pregnancy, the estrus period is followed by diestrus, a luteal phase with an average duration of two months

(Songsasen and Wildt, 2007). Towards the end of diestrus, corpus luteum function declines causing females to enter a prolonged anestrous period that lasts two to ten months in the domestic dog. Anestrus is characterized by the lack of external signs of ovarian activity and nadir concentrations of circulating progesterone (Concannon, 2011).

Reproductive physiology of the maned wolf is largely similar to that of domestic dogs. For the maned wolf, average duration of proestrus is 10.6 days (Velloso et al.,

1998). By utilizing non-invasive hormone monitoring, it has been shown that estrogen metabolite concentrations increase during this period and reach peak level 1-2 days prior to the onset of breeding, then, decline to baseline on the day of lordosis or copulation

(Fig. 2.2; Songsasen et al., 2006). During estrus, females accept males as indicated by solicitous tail flagging, play invitations, sniffing, anogenital investigation and lordotic stances (Rodden et al., 1996; Velloso et al., 1998). Breeding is considered successful when a copulatory tie lasting up to several minutes has been achieved (Rodden et al.,

1996, 2004). During estrus, progestagen metabolite concentrations gradually increase and remain elevated in females housed with a male, regardless of breeding outcome. It has been shown that progestagen metabolite levels are significantly higher in pregnant females from proestrus to parturition (Songsasen et al., 2006). However, unlike the domestic dog, progestagen concentrations of a single female remains at a low level, indicating that the maned wolf may be an induced ovulator (Songsasen et al., 2006).

Following an average 65-day gestation period, progestagen concentrations drop at parturition (Dietz, 1984; Velloso et al., 1998; Maia and Gouveia, 2002). Diestrus ends with the onset of seasonal anestrus. In captive maned wolves, mutual avoidance and minimum levels of scent marking and vocalization are characteristic of anestrus (Dietz,

1985).

Figure 2.2 Longitudinal profiles of fecal gonadal (estrogen and progestagen) metabolites for (A) a pregnant maned wolf from 28 days before estrogen peak to parturition (Songsasen et al., 2006) and (B) a pseudo-pregnant maned wolf from 13 days before estrogen peak to the end of the luteal phase.

Non-Invasive Hormone Monitoring Non-invasive hormone monitoring is practiced in humans and many domestic species

(e.g. dogs, cats, and sheep). This nonintrusive technique has become useful, as it has proven to be a valuable means for assessing hormones in wildlife living ex situ and in situ. Fecal and urinary hormone analysis has been applied to captive and wild animals

(Lasley and Kirkpatrick, 1991; Millspaugh and Washburn, 2004). Fecal glucocorticoids have been used to assess adrenal activity. For example, in captive leopard cats, reduced exploration was determined to be a behavioral indicator signifying chronically elevated adrenocortical activity (Carlstead et al., 1993). Wielebnowski et al. (2002) found higher fecal corticoid concentrations in clouded leopards maintained on public display or adjacent to heterospecifics in comparison to individuals maintained off exhibit or in the absence of predators. Analysis of fecal and urinary gonadal hormone metabolites has permitted longitudinal monitoring of reproductive activity in various wildlife species living ex situ including bottlenose dolphins (Tursiops truncates; Biancani et al., 2009),

Asian and African elephants (Elephas maximus and Loxodonta africana; Brown, 2000), felids (Brown et al., 1994; Jurke et al., 1997; Wielebnowski et al., 2002; Newell-Fugate et al., 2007), black and white rhinoceros (Diceros bicomis and Ceratotherium simum;

Carlstead and Brown, 2005; Dorsey et al., 2010), western lowland gorillas (Gorilla gorilla gorilla; Clark et al., 2011), black-footed ferrets (Mustela nigripes; Poessel et al.,

2011), African wild dogs (Lycaon pictus; Monfort et al., 1998) and many others. This technology has proven useful in studying wildlife in their native habitat. Creel et al.

(2002) assessed fecal glucocorticoids in wild wolves (Canis lupus) and elk (Cervus elaphus) and concluded that snowmobile activity induced stress in these species.

Non-invasive hormone monitoring has also been used to study reproductive physiology of the maned wolf (see above). It has been shown that female maned wolves that have lost their neonates demonstrate lower progesterone levels during the second half of gestation as compared to female maned wolves that successfully raise their young

(Songsasen et al., 2006). It has been proposed that lower progestagen excretion in female maned wolves that lose their pups may be linked to an adrenal/stress response, which may be caused by suboptimal management. Preliminary analysis of fecal corticoid metabolites, when compared between females that successfully raised young to those females that lost their pups, showed that the latter group revealed significantly higher cortisol levels, especially during the periovulatory period (MWSSP, 2007). Thus, females that experience neonatal loss exhibit elevated cortisol concentrations that may be stress-induced.

Reproductive senescence has been eliminated as a cause of poor reproduction because individuals between ages ten and twelve have successfully reproduced (MWSSP,

2010; Songsasen and Rodden, 2010). MWSSP recommends maned wolves for breeding when individuals are at prime-breeding age (MWSSP, 2010).

Stress Stress, coupled with sub-optimal management, has been proposed as a contributing factor to low reproductive success and neonatal mortality in maned wolves. Only 38% of 166 recommended breeding pairs produced pups from 1996 through 2007 with 50% of pregnant maned wolves losing neonates (Songsasen and Rodden, 2010). Animals living in captivity experience a more static habitat, unlike their free-living counterparts

(Cummings et al., 2007). A top research priority for the MWSSP is to determine if chronic stress affects reproduction in maned wolves held ex situ.

No single, standard definition of stress exists nor does a single biochemical assay system exist to measure it (Möstl and Palme, 2002). Rivier and Rivest (1991) defined stress as any disruption of homeostasis. This can include, but is not limited to, energetic, immunological and psychological challenges (Padgett and Glaser, 2003; Hosey et al.,

2009). Stress, resulting from intense physical, physiological or immune challenges will activate the hypothalamus, pituitary and adrenal cortex (HPA) axis (Fig. 2.3).

Figure 2.3 Diagram illustrates the mechanisms by which the hypothalamic-pituitary-adrenal (HPA) axis responds to a stressor. Adapted from Irwin and Cole (2011).

A stressor initially acts upon the hypothalamus triggering the secretion of corticotropin-releasing hormone (CRH), which in turn stimulates the pituitary gland to produce adrenocorticotropic hormone (ACTH) (Fig. 2.3; Schatz and Palme, 2001). The release of ACTH, then, stimulates the adrenal cortex to secrete cortisol, a glucocorticoid hormone (Hosey et al., 2009). After cortisol has been released into the blood stream, it circulates back to the pituitary gland and hypothalamus where cortisol can bind and inhibit the stress response preventing more cortisol from being released. This negative feedback results in the return of cortisol to baseline levels.

Stressors are perceived differently among individuals and the degree to which the adrenal cortex responds varies with the type of stress (Liptrap, 1993). Stress may be acute or chronic, adaptive or maladaptive. Additionally, the type of stress intensity

(innocuous, aversive, noxious or extreme), duration and how the animal perceives them will determine its impact on biological functions (Hosey et al., 2009).

Short-term stress, also called acute stress, often promotes adaptation and increases survival (McEwen, 2004; Sheriff et al., 2011) and, thus, acute stress may be beneficial and increase fitness. An acute stressor results in the short-term elevation of glucocorticoid concentrations (Wikelski and Cooke, 2006; Sheriff et al., 2011) with the feedback mechanism operating efficiently and the system rapidly returning to normal

(Sheriff et al., 2011). Short-term stress is most often a form of good stress and benefits the organism and its survival.

The other form of stress is chronic and may result in long-term damage (McEwen,

2004). A chronic stressor produces weak feedback signals and the system remains active

38 for prolonged periods of time preventing the body from returning to and maintaining homeostasis (Wikelski and Cooke, 2006; Seaward, 2008; Sheriff et al., 2011).

Consequently, the effects of ‘chronic stress’ are known to have a wide variety of deleterious, potentially fatal physiological and immunological consequences (Carlstead,

1996; Sheriff et al., 2011). Chronic stress, as measured by persistent elevated levels of cortisol, can negatively impact reproduction, suppress immune function and lead to a number of other pathophysiological consequences (Stoskopf, 1983; Moberg, 2000;

McEwen, 2004; Seaward, 2008).

Extreme demands, such as chronic exposure to a threat, can impair the HPA axis and harm the body and cause reproductive problems (Nepomnaschy et al., 2004). The adaptive value of reproductive suppression mechanisms allows females to avoid mating during inopportune situations and instead focuses scarce resources on survival, improvement in overall condition, and investment in existing offspring (Sapolsky et al.,

2000). This process, modulated by various substances, acts to maintain the fitness of the individual rather than wasting energy on reproduction and any potential to further exhaust energy, especially if offspring are produced.

Nutritional Management It is well established that there is a link between nutritional resources and fertility.

Reproduction (pregnancy, parturition and lactation) requires enormous energy demands

(Songsasen and Rodden, 2010). However, formulation of appropriate diets for captive maned wolves has been a challenge. Historically, this species has been fed like a carnivore with raw meat comprising the entire diet (Childs-Sanford and Angel, 2006;

MWSSP, 2007). However, unlike felids, canids require a broader array of foods (Allen,

1995). Unfortunately, general diet composition of wild maned wolves has only been approximated since no nutrition studies have been conducted on wild maned wolves or their diets (Childs-Sanford and Angel, 2006). Presently, manufactured maned wolf diets are now formulated with canid diets (Childs-Sanford and Angel, 2006). North American facilities feed dog chow-based diets supplemented with a variety of fruits and vegetables in addition to small amounts of whole prey items (i.e. rats, mice and chicks) (Allen, 1995;

Songsasen and Rodden, 2010).

Maned wolves have a very sensitive gastrointestinal tract and often respond to minor management changes with soft stools or diarrhea (Childs-Sanford and Angel,

2006). In addition to poor body condition, maned wolves have a lower digestibility of dry matter, energy and several minerals including copper, magnesium, iron and sodium than domestic dogs and a rapid ingesta passage (Childs-Sanford and Angel, 2006).

Therefore, poor reproduction in maned wolves may be a result from an energy imbalance causing suboptimal deposit of body fat despite efforts to provide a well-balanced diet.

Leptin Leptin is an adipocyte hormone critical to the regulation of the state of equilibrium of energy (Moschos et al., 2002). It is a peptide hormone predominantly produced by white adipose tissue and a mediator in balancing food intake and energy expenditure and conservation (Greenstein and Wood, 2010). Since being discovered in 1994, studies have shown leptin influences reproduction in two ways: (1) timing of the onset of first ovulation and (2) maintaining reproductive function during adulthood (Capiro et al.,

2001; Henson and Castracane, 2006). Rising leptin levels are associated with initiation of puberty in animals and humans. Furthermore, normal leptin levels are required for maintaining normal reproductive function (Mantzoros, 2000). It has been suggested that leptin modulates the release of gonadotropin-releasing hormone (GnRH) from the hypothalamus, which in turn attenuates the secretion of follicle stimulating hormone

(FSH) and luteinizing hormone (LH), which regulates gonadal functions.

Leptin, while most commonly studied in humans and mice, is an endocrine hormone contributing to body weight regulation and energy metabolism (Muehlenbein et al., 2003; Greenstein and Wood, 2010). In humans, radioimmunoassays (RIA) have been developed and validated for measuring levels of serum leptin (Henson and Castracane,

2006). A positive correlation has been found between leptin concentrations in the blood and body fat content, indicating that leptin acts as an indicator for mass of fat storage

(Muehlenbein et al., 2003; Goodman, 2009), and circulating concentration of this protein hormone can reflect acute changes in nutritional state (Goodman, 2009).

During the past decade, several studies have been conducted to assess the relationship between leptin and various biological events in captive and free-ranging wildlife. Banks et al. (2001) measured serum leptin levels in wild and captive populations of baboons (Papio anubis, Papio hamadryas and Papio anubis/hamadryas hybrids) and found that captive baboons have levels roughly three times higher than wild baboons and about ten times higher leptin/weight ratio indicating that fat in captive baboons represents a higher proportion of body weight and body mass in wild baboons consists of less adipose tissue and more other tissues. Additionally, baboons demonstrate

41 a clear inverse relation between age and leptin levels (Banks et al., 2001). Spady et al.

(2009) measured serum leptin and reports leptin is highly predictive of percent body fat black bears (Ursus americanus); thereby recommending the use of leptin as a wildlife management tool to monitor adiposity in American black bears. Raccoons (Procyon lotor) have seasonal variations in circulating leptin concentrations with low concentrations from February to June and high concentrations from October to November

(Shibata et al., 2005) with higher concentrations during the latter period most likely linked to body fat accumulation in preparation for winter. Additionally, Shibata et al.

(2005) reported that serum leptin concentrations are significantly positively correlated with body weight, as well as body mass index (BMI) in raccoons. They also concluded that circulating leptin is a good marker of nutritional condition (Shibata et al., 2005). In

Iberian red deer (Cervus elaphus hispanicus), a positive relationship exists between plasma leptin concentrations during antler growth and final antler length with peak amplitude when males obtain the final antler size (Gasper-López et al., 2009). These results demonstrate that the relationship between leptin, body mass and body condition score (BCS) changes throughout the year for this seasonal breeder (Gasper-López et al.,

2009).

A leptin assay has been developed in the domestic dog and has been used to assess the level of this peptide hormone in other carnivores such as raccoon dogs

(Nyctereutes procyonoides; Nieminen et al., 2002), raccoons (Shibata et al., 2005) and mink (Mustela vison; Tauson and Forsberg, 2002). It has been shown that there are seasonal variations in leptin levels, and circulating hormone concentration is associated

42 with body weight and BMI in some carnivores. To date, there is no information on leptin levels in the maned wolf or how this peptide influences reproductive function in this species.

III. REPRODUCTIVE AND STRESS HORMONE MONITORING IN THE CAPTIVE MANED WOLF (CHRYSOCYON BRACHYURUS)

Introduction Ex situ conservation of threatened species is necessary in a world of ever expanding human population. Maintaining sustainable populations of animals is a fundamental goal of ex situ management and entails creating a reservoir of individuals as genetically close as possible to their conspecifics in the wild; thus, preserving genetic diversity (Comizzoli et al., 2009). Zoo populations act as a hedge against cataclysmic events that may extinguish populations in nature (Wildt et al., 2001; Cummings et al., 2007).

Additionally, captive animals also provide scientists with research opportunities to study behavior, reproductive biology, health, disease, nutrition and genetics (Hosey et al., 2009;

Songsasen and Rodden, 2010). They also serve as ambassadors for their conspecifics in the wild and for disappearing habitats. The main threats to wild maned wolf (Chrysocyon brachyurus) populations include habitat loss, conflicts with humans, diseases transmitted from domestic dogs (Canis familiaris) and road mortality (Rodden et al., 2004).

Carnivores are unique because they are highly vulnerable to environmental change and human persecution (Comizzoli et al., 2009). Consequently, it is critical to sustain captive populations of carnivore species, such as maned wolves. Yet, maintaining a self- sustaining captive maned wolf population is not without its challenges.

Zoo-managed maned wolves breed poorly despite efforts to optimize captive management. Causes contributing to poor reproduction may include, but are not limited to, inadequate diet, lack of physiological or environmental requirements, behavioral incompatibility and inbreeding depression (Snyder et al., 1996). In addition to poor reproductive success, ex situ maned wolves suffer high neonatal mortality. Maia and

Gouveia (2002) reported 1,532 births recorded from 1980 to 1998; however, only 44.3%, or 678, survived past year one. Of the 854 that did not survive past year one, parental incompetence comprised 66.1% of pup deaths followed by infectious disease, digestive system disorders, lesions and accidents and respiratory disorders (Maia and Gouveia,

2002). During the 2010-11 breeding season, 16 pups were born, of which 11 (69%) survived (MWSSP, 2011). Pup survival has shown signs of improvement, in part because the Maned Wolf Species Survival Plan® (MWSSP) has increased the number of recommended pairings. Nevertheless, maned wolves still exhibit poor reproductive success in captivity and the population experienced a significant decline from 93 animals in 2007 to 79 animals in August 2009.

Non-invasive hormone monitoring has provided insight in understanding poor reproductive success and has been successfully used to monitor the reproductive health of a number of populations of captive species, including maned wolves. Female maned wolves that experienced neonatal loss demonstrated lower progestagen concentrations during the second half of gestation as compared to female wolves that successfully raised their young (Songsasen et al., 2006). Lower progestagen excretion in female maned wolves that lost their pups may be linked to an adrenal-stress response caused by

45 undetermined variables within their environment. Preliminary analysis of fecal corticoid metabolites comparing females that successfully raised young with those females that lost their pups showed that the latter group experienced significantly higher cortisol levels, especially during the peri-ovulatory period (MWSSP, 2007), indicating that poor reproduction in this species may be due to stress associated with captive management.

Suboptimal housing and/or poor management conditions have been shown to cause stress and affect reproductive success in other zoo-managed species, particularly in specialized carnivores (Rivier and Rivest, 1991). Small and large felids including leopard cats (Prionailurus bengalensis; Carlstead et al., 1993), clouded leopards (Neofelis nebulosa; Wielebnowski et al., 2002) and cheetahs (Acinonyx jubatus; Jurke et al., 1997) are sensitive to stress. For instance, cheetahs exhibiting high circulating cortisol concentrations experience less follicular activity and prolonged periods of anestrus (Jurke et al., 1997). Additionally, reproductive success in small felids decreases when animals are housed in groups larger than a pair (1.1; Mellen 1991).

Hypothesis It was hypothesized that female maned wolves that failed to produce offspring, as defined by pups living longer than 30 days, have abnormal perturbed concentrations of gonadal hormones associated with persistent elevation of corticoids.

Objectives The findings from this research may provide insight into the underlying causes of poor reproductive success in captive maned wolves. This study has attempted to address the question of whether or not neonatal loss is related to an adrenal/stress response. The

46 objective of this study was to compare gonadal and adrenal hormones among female maned wolves of various reproductive outcomes.

Animals and Facilities The sample population consisted of female maned wolves within the Association of Zoos and Aquariums’ (AZA) MWSSP program that were within prime reproductive age, between three and eight years. Forty-two reproductive cycles from 29 individuals were collected for this study. However, three cycles were removed from analysis due to implantation of contraceptive devices. Thus, 39 reproductive cycles from 28 individuals were analyzed for gonadal and adrenal hormones (see Table 3.1). Individual maned wolves were located at 14 breeding facilities throughout North America: Smithsonian

Conservation Biology Institute (SCBI), National Zoological Park, Front Royal, VA;

White Oak Conservation Center (WOCC), Yulee, FL; Connecticut’s Beardsley Zoo,

Bridgeport, CT; Houston Zoo, Houston, TX; Wildlife World Zoo, Litchfield Park, AZ;

Pueblo Zoo, Pueblo, CO; Louisville Zoo, Louisville, KY; Buffalo Zoo, Buffalo, NY;

Birmingham Zoo, Birmingham, AL; San Antonio Zoo, San Antonio, TX; Sedgwick

County Zoo, Wichita, KS; Oklahoma City Zoo, Oklahoma City, OK; Omaha’s Henry

Doorly Zoo and Aquarium, Omaha, NE; and Little Rock Zoo, Little Rock, AR. All females for this project experienced the same breeding season, October though February.

The study monitored reproductive and stress hormones for the 2002-03, 2003-04, 2004-

05, 2009-10 and 2010-11 breeding seasons. Eighteen out of the 39 (46.2%) maned wolf reproductive cycles analyzed for this study resulted in pregnancies (n = 10 raised pups; n

= 8 lost pups). Nine profiles were collected on unpaired females. Lastly, 12 breeding

47 attempts concluded with no birth, but hormone profiles showed sustained elevations of progestin concentrations indicative of a pseudo-pregnancy.

Sample Collection Fecal samples (10 g) were collected three to five times per week from each maned wolf from October 1 until February 28, or until pups were produced. The fresh samples were placed in a plastic bag labeled with the animal’s name or studbook/identification number, facility name, and date, and stored at -20ºC until transported to the Smithsonian

Conservation Biology Institute (SCBI) for hormonal analysis (Brown et al., 2008).

Hormone Extraction from Feces Estrogen, progestagen and cortisol metabolites were extracted from fecal samples using the dry weight shaking extraction method, which has previously been validated for maned wolves (Songsasen et al., 2006). Frozen fecal samples were lyophilized (VirTis

Ultra Lyophilizer, SP Scientific, Gardiner, NY), and, then, pulverized into fine powder and stored frozen (-20ºC) until steroid extraction (Fig. 3.1). Well-mixed powdered feces

(0.2 g ± 0.02) was combined with 5 mL of 90% ethanol, vortexed so that fecal material was thoroughly mixed in the solution and vigorously shaken using a Multi Pulse vortexer

(Glas-Col, Terre Haute, IN) for 30 minutes (speed 60-70 RPM). Subsequently, the tubes were centrifuged at 2200 RPM for 20 minutes, the supernatant transferred into a glass tube and the pellet resuspended in an additional 5 mL of 90% ethanol, vortexed for 30 seconds and recentrifuged at 2200 RPM for 20 minutes. The supernatants were combined, dried under air and resuspended in 1 mL of 100% ethanol. Extractants were then sonicated (5 minutes), vortexed (30 seconds) and dried under air. Samples were

48 reconstituted in 1 mL of a phosphate buffer (0.2 M NaH2PO4, 0.2 M Na2HPO4, 0.15 M

NaCl; pH 7.0), vortexed (30 seconds), sonicated (15 minutes) and then, stored frozen

(-20ºC) until analysis (Songsasen et al., 2006; Brown et al., 2008).

To monitor extraction efficiency, 3H-progesterone was added to the fecal sample prior to extraction. Steroid extraction efficiency was determined by the recovery of tritiated progesterone. All efficiencies were >65%. CV was 0.13%. Samples were further diluted before analysis for estrogen (1:10-1:100), progestagen (1:500-1:40,000) and glucocorticoid (1:5-1:500) metabolites.

Hormone Analysis All samples were quantified for progestagen and cortisol metabolite concentrations by enzyme immunoassay (EIA), while samples collected during the 2009-10 breeding season also were assessed for estrogen metabolites (Munro and Stabenfeldt, 1984;

Gudermuth et al., 1998; Brown et al., 2008). Due to unidentified problems in the laboratory setting, the estrogen EIA became unreliable and unusable, thereby preventing estrogen metabolite analysis for all samples collected. Antibodies for estrogen

(polyclonal anti-estrone conjugate [EC] R522), pregnane (monoclonal pregnane CL425) and cortisol (R4866) metabolite analysis were obtained from Coralie Munro (University of California, Davis, CA).

Fecal Estrone Conjugate (EC) Enzyme Immunoassay Fecal estrogen concentrations from each sample were quantified in duplicate using a previously described EIA procedure (Munro and Stabenfeldt, 1984; Gudermuth et al.,

1998; Brown et al., 2008). The R522-2 cross-reacts with estrone-3-glucuronide (100%),

49 estrone-3-sulfate (66.6%), estrone (238%), estradiol-17β (7.8%), estradiol-3-glucuronide

(3.8%), estradiol-3-sulfate (3.3%), estradiol-17-sulfate (0.1%), estradiol-3-disulfate

(0.1%), ethinyl estradiol-17β (< 0.1%), estriol (< 0.1%), progesterone (< 0.1%), pregnanediol (< 0.1%), cortisol (< 0.1%), testosterone (< 0.1%) and androsterone (<

0.1%) (Munro et al., 1991). Serial dilutions of pooled fecal extracts produced displacement curves parallel to those of the appropriate standards. Linear regression analysis was performed on the recovery of added standard to pooled fecal extracts and yielded the equation y = 0.69x + 4.13 (R2 = 0.99). Assay sensitivity was 0.78 pg/well.

Intra- and inter-assay coefficients of variation were less than 10% and 8%, respectively.

Fecal Progestagen (Pg) Enzyme Immunoassay Each sample was quantified in duplicate using a previously described EIA procedure

(Munro and Stabenfeldt, 1984; Gudermuth et al., 1998; Brown et al., 2008). The EIA used a monoclonal antibody produced against 4-pregnen-11-ol-3,20-dione hemisuccinate:BSA (CL425, 1:10,000; C. Munro, University of California, Davis, CA).

The CL425 cross-reacts with various progestagen metabolites, including 4-pregnen-3,20- dione (100%), 4-pregnen-3α-ol-20-one (188%), 4-pregnen-3β-ol-20-one (172%), 4- pregnen-11α-ol-3,20-dione (147%), 5α-pregnan-3β-ol-20-one (94%), 5α-pregnan-3,20- dione (64%), 5β-pregnan-3β-ol-20-one (55%), 5β-pregnan-3,20-dione (8%), 4-pregnen-

11β-ol-3,20-dione (2.7%), 5β-pregnan-3α-ol-20-one (2.5%), 5β-pregnan-3α,20α-diol (<

0.1%), 5α-pregnan-3α,20β-diol (< 0.1%), 5β-pregnan-3,17-dione (< 0.1%) and 5β- pregnan-11β,21-diol-3,20-dione (< 0.1%) (Munro et al., 1991). Serial dilutions of pooled fecal extracts produced displacement curves parallel to those of the appropriate standards.

Linear regression analysis was performed on the recovery of added standard to pooled fecal extracts and yielded the equation y = 0.93x – 0.20 (R2 = 0.99). Assay sensitivity was 0.78 pg/well. Intra- and inter-assay coefficients of variation were less than 10% and

14.3%, respectively.

Fecal Cortisol (F) Enzyme Immunoassay The cortisol EIA used a polyclonal cortisol antibody generated against cortisol-3-CMO

(R4866, 1:20,000; C. Munro, University of California, Davis, CA). Fecal cortisol concentrations from each sample were quantified in duplicate using a previously described EIA procedure (Munro and Stabenfeldt, 1984; Gudermuth et al., 1998; Brown et al., 2008). The polyclonal antiserum (R4866) cross-reacts with cortisol (100%), prednisolone (9.9%), prednisone (6.3%), cortisone (5.0%), corticosterone (0.7%), deoxycorticosterone (0.3%), 21-deoxycortisone (0.5%) and 11-desoxycortisol (0.2%)

(Young et al., 2001; Young et al., 2004). Serial dilutions of pooled fecal extracts produced displacement curves parallel to those of the appropriate standards. Linear regression analysis was performed on the recovery of added standard to pooled fecal extracts and yielded the equation y = 1.01x + 4.98 (R2 = 0.99). Assay sensitivity was 3.9 pg/well at maximum binding. Intra- and inter-assay coefficients of variation were less than 10% and 9.9%, respectively.

High Performance Liquid Chromatography (HPLC) The numbers and relative proportions of immunoactive corticosteroid metabolites in maned wolf fecal extracts were determined using reverse phase high performance liquid chromatography (HPLC) (Reverse Phase Microsorb MV 100 C-18, 5 µm diameter

51 particle size; Varian, Woburn, MA). Six fecal extracts were combined, air dried, resuspended in 1 mL methanol, dried under air and stored at -20°C. The pooled extract was reconstituted with 0.5 mL phosphate buffered saline and filtered through a C-18 cartridge (SpiceTM, Analtech, Newark, DE) eluted with 5 mL methanol, air dried and spiked with known radiolabelled corticosteroids (3H-cortisol and 3H-corticosterone,

~2,500 dpm). Samples were then dried down and resuspended in 300 mL methanol.

Corticosteroid metabolites were separated using a gradient of 20-100% methanol in water within 80 minutes (1 mL/min flow rate, 1.0 mL fractions) for 55 µL portion of pooled fecal extract.

Co-elution profiles of the radiolabelled corticosteroids in the HPLC run were determined by adding 100 µL of each HPLC fraction to 3 mL of scintillation fluid

(Ultima Gold; Packard, Meriden, CT) and counted in a dual-label channel β scintillation counter (LS6500, Beckman, Fullerton, CA). Quench correction was automatically performed by the counter.

Each HPLC fraction was evaporated to dryness, reconstituted in 125 µL assay buffer and quantified for immunoactivity using cortisol EIA, as described above, and an established 125I corticosterone radioimmunoassay using a double-antibody 125I radioimmunoassay (RIA) (ICN Biomedicals, Inc., Costa Mesa, CA). Each fraction was assayed in singleton.

Statistical Analysis Standard descriptive statistics, including mean and standard error of the mean (SEM), were used to evaluate data. All statistical analyses were conducted using Microsoft Excel

2011 (Version 14.1.4, Microsoft Corporation, Seattle, WA) and IBM SPSS Statistics

(Version 19, SPSS Inc., Chicago, IL) software for the Macintosh. For all analyses, significance was assessed at the P = 0.05 level.

Data were reported as the mean ± SEM. Profiles of estrogen and progestagen metabolites were aligned to the day of the estrogen peak (Day 0). If estrogen was not measured, Day 0 was defined as the first day progestagen concentrations exceeded the baseline values plus 1.5 standard deviations (SD) or was calculated by subtracting 65 days from the day of pup birth (Brady and Ditton, 1979; Wasser et al., 1995; Songsasen et al., 2006). Fecal hormone levels across an individual’s profile were pooled into 3-day means (Songsasen et al., 2006). Longitudinal steroid metabolite profiles were divided into three reproductive stages: pre-ovulatory (days -31 to -10), peri-ovulatory (days -9 to

9) and luteal (days 10 to 70) using modified criteria of those described by Velloso et al.

(1998) and Walker et al. (2002). Baseline values for each individual animal were calculated by an iterative process, whereby high values (exceeding � + 1.5 [SD]) were excluded (Songsasen et al., 2006).

Mean daily fecal progestagen concentrations were compared among groups of: 1) pregnant females that produced and successfully raised pups (n = 10), 2) pregnant females that conceived but lost pups early post-partum (n = 8), 3) pseudo-pregnant females (n = 12), and 4) unpaired females (n = 9) for each phase of the reproductive cycle, whereas cortisol concentrations were compared for all paired females. Means ±

SEM were calculated from raw data of hormone metabolites during each reproductive phase for each class of reproductive outcome. Values exceeding � ± 2(SD) were

53 considered as outliers and excluded from the analysis. Consequently, reproductive hormone data was based on 27 cycles for paired females (n = 9 raised pups, n = 7 lost pups and n = 11 pseudo-pregnant females) and eight cycles for single maned wolves. All pregnant (n =18) and pseudo-pregnant (n = 12) cycles exhibited prolonged elevation of progestagen concentrations above baseline assumed to represent luteal activity.

Differences in fecal steroid metabolites between various stages of the breeding cycle among reproductive groups were determined by analysis of variance (ANOVA) followed by multiple comparisons. For all analyses, the Welch’s F-test was used because the assumption of homogeneity of variances was violated. Games-Howell post-hoc statistical tests were conducted for pair-wise comparisons.

Characterization of Glucocorticoid Metabolites by HPLC Evaluation of fecal extracts by HPLC revealed two immunoreactive peaks (fractions 35-

40 and 44-51; Fig. 3.2), one of which partially co-eluted with cortisol (fraction 37-43).

Hormone Analysis Females gave birth as early as mid-November and as late as the end of March. However, most births (67%) occurred in the months of December and January (Fig. 3.3).

Fecal Progestagen Hormone Analysis Fecal progestagen concentrations (µg/g feces) across reproductive phases in female maned wolves of various reproductive outcomes are provided in Table 3.2. Following with sexual receptivity, fecal progestagen metabolite concentrations rose steadily and remained elevated during pregnancy for females that successfully raised pups (Fig. 3.4A) and lost pups early post-partum (Fig. 3.4B).

We found no statistically significant differences between group means for fecal progestagen concentrations during the pre-ovulatory phase (F3,266 = 1.708, P = 0.17).

During the peri-ovulatory phase, there was a statistically significant (Welch’s F3,123.982 =

25.468, P < 0.001) difference between group means for fecal progestagen concentrations

(Table 3.2). No statistically significant difference was found between females that raised pups and those that lost pups (Games-Howell, P > 0.05). However, all pregnant females, those that raised pups and lost pups, had higher fecal progestagen concentrations in the peri-ovulatory phase than pseudo-pregnant bitches (Games-Howell, P < 0.05) and unpaired females (Games-Howell, P < 0.05). Lastly, pseudo-pregnant females had higher progestagen metabolite concentrations than unpaired females (Games-Howell, P <

0.05) in the peri-ovulatory phase of estrus.

Similar to the peri-ovulatory phase, there was a statistically significant difference

(Welch’s F3,364.898 = 238.865, P < 0.001) between group means for fecal progestagen concentrations during the luteal phase. Again, no statistically significant difference was found between pregnant females that raised pups and those that lost pups (Games-

Howell, P > 0.05). But, all pregnant females excreted higher progestagen concentrations than pseudo-pregnant (Games-Howell, P < 0.05) and unpaired (Games-Howell, P = 0.05) females. Finally, pseudo-pregnant females had higher fecal progestagen metabolite concentrations than unpaired females during the luteal phase (Games-Howell, P < 0.05).

Fecal Cortisol Hormone Analysis We found no statistically significant differences between group means for fecal cortisol concentrations during the pre-ovulatory phase (F2,234 = 2.057, P = 0.13), as well as the

55 peri-ovulatory phase (F2,272 = 2.953, P = 0.054). However, there was a statistically significant difference between group means (Table 3.3) for fecal cortisol concentrations during the luteal phase (Welch’s F2,345.167 = 28.315, P < 0.001). Pregnant maned wolves that successfully raised pups had lower fecal cortisol metabolite concentrations (Table

3.3) than pregnant females that lost pups and pseudo-pregnant females (Games-Howell, P

< 0.05) during the luteal phase. No differences were found between pseudo-pregnant females and those that suffered a neonatal loss (Games-Howell, P > 0.05) (Fig. 3.5).

Discussion The present study demonstrated that: 1) pseudo-pregnant females, on average, excreted lower fecal progestagen metabolites than pregnant bitches; 2) maned wolves are induced ovulators; and 3) females that conceived and raised pups excreted significantly lower fecal corticoid concentrations compared to females that became pseudo-pregnant, as well as pregnant females that experienced neonate loss.

The results of this study confirmed previously published observations on the reproductive physiology of the maned wolf (Wasser et al., 1995; Velloso et al., 1998;

Songsasen et al., 2006). Bestlemeyer (2000) reported most maned wolves entered estrus between October and December with peak estrogen levels 1-2 days prior to breeding.

Females in this study gave birth as early as mid-November and throughout late March with the majority of births occurring in December and January, all of which coincides with the specified breeding season in the Northern hemisphere and agrees with Maia and

Gouveia’s (2002) report of peak number of births occurring in December.

Females that conceived and raised pups excreted the highest mean progestagen concentrations during the luteal phase, which may be useful for predicting pregnancy.

Our findings agree with the findings of studies conducted by Songsasen et al. (2006) and

Velloso et al. (1998), which report lower mean progestagen concentrations in pseudo- pregnant bitches when compared to their pregnant counterparts using EIA and RIA. We concluded that fecal progestagen profiles might be used for pregnancy diagnosis.

Progestagen values from a single sample would not be effective due to individual variations in metabolite concentrations; one female may excrete overall elevated values compared to another female that may yield lower concentrations. However, longitudinal hormone monitoring with frequent sampling (3-5 samples per week) for one month following breeding may predict whether a female is pregnant or not. Based upon the results from this study, we suggest that mean values < 30 µg/g feces indicate pseudo- pregnancy.

Fecal progestagen concentrations in unpaired females of prime reproductive age remained at baseline throughout the breeding season indicating a failure to ovulate.

Thus, ovulation only occurred in females paired with a male, indicating that the maned wolf is an induced ovulator. By understanding this mechanism, assisted reproductive technologies can be developed and methods improved for inducing ovulation and increasing reproductive success.

A preliminary report by the MWSSP (2007) describes significantly higher fecal corticoid metabolite concentrations during the peri-ovulatory phase in females that lost pups compared to females that raise pups. In addition to those findings, we showed that

57 with a larger sample size, females that became pseudo-pregnant or experienced neonate loss excreted significantly higher fecal corticoid concentrations during the luteal phase compared to pregnant females that raised pups.

Approximately half, 55%, of all pregnant maned wolves in the present study successfully raised pups. Based on the results from this study, pup mortality may be directly linked to increased adrenocortical activity, which may be in response to stressful events (e.g., suboptimal enclosure environment and health). Increased adrenal activity may be an indicator for predicting false pregnancies and pup loss in captive maned wolves. If so, this would contribute to cost-effective monitoring around parturition, as such advanced notice would permit management to prepare for whelping and, perhaps discourage infanticide (Wasser et al., 1995; Velloso et al., 1998). Stress-induced changes in the secretion of pituitary hormones including corticotropin-releasing hormone (CRH) have been linked to reproductive failure by inhibiting the secretion of gonadotropin- releasing hormone (GnRH), which stimulates luteinizing hormone (LH) secretion

(Moberg, 2000). Future research should focus on identifying ‘stressors’ and developing strategies toward alleviating stress.

The research supported our hypothesis, which stated, pregnant female maned wolves that lost pups early post-partum would present abnormal gonadal hormone concentrations and/or elevated concentrations of corticoids in comparison to females that successfully raised pups. Although mechanisms of stress-induced changes in progesterone concentrations are not well understood, Cooper et al. (1995) reported

58 cortisol and progesterone secretions in response to acute stress in steers and suggested that the ratio may offer a useful method for assessing stress.

Quantifying adrenal activity in female maned wolves beginning in mid-August throughout the beginning of April would prove to be more effective as some females bred before October. This would also provide more data for determining baseline corticoid concentrations prior to the onset of the reproductive period. Additionally, comprehensive behavioral analyses on females recommended for breeding should be considered.

Perhaps, some behaviors may predict pup loss prior to birth. For example, individual differences in response to stress may forecast which females are more likely to experience pup loss (Schmalz-Peixoto, 2003). Bakken et al. (1999) describe farmed silver fox (Vulpes vulpes) females that commit infanticide are more sensitive to stressors

(e.g., social competition) and respond faster to perceived stimuli, whereas non- infanticidal females expressed weaker responses. Bakken (1992; 1993) shows that females removed from visual access to conspecifics raised their cubs successfully.

Understanding the source for elevated corticoids should be examined closer, especially if the cause is due to suboptimal management. Schmalz-Peixoto (2003) reports that species with elaborate nutrition and environmental needs increase the likelihood for poor reproductive success in captive conditions (Donoghue and

Langenberg, 1994; Carlstead and Shepherdson, 1994; Lyons et al., 1997). Schmalz-

Peixoto (2003) reports significantly higher rates of neonatal mortality in species that prey on small to very small animals, as well as a significant effect of type of diet on infant mortality in captivity whereby insectivorous and omnivorous animals show the highest

59 rates of neonatal mortality in comparison to piscivorous, herbivorous and carnivorous animals.

According to Maia and Gouveia (2002), 67% of maned wolf pup deaths are caused by parental involvement such as inadequate maternal behavior including maternal cannibalism, infanticide and neglect. We remain unsure about cases of neonatal mortality and if maternal infanticide occurred or if pups were born unhealthy. Future efforts should focus on defining the time of pup mortality, as well as exploring infanticidal episodes by males. Pup death could be broken down, for example, according to longevity of offspring (e.g. < 1 day, 1-30 days and 30+ days). Further analysis on cause of death may also provide more insight.

A wide range of factors may impact reproductive success in captive maned wolves. Schmalz-Peixoto (2003) describes a strong positive correlation between transfers and relocations by females and the incidence of infant mortality in carnivores.

Subsequently, this warrants further investigation as this claim may significantly influence reproductive success in captive maned wolves. Other elements including enclosure size, climate, proximity to conspecifics and heterospecifics, type of rearing (i.e. parent/hand/foster), feeding schedule and enrichment program among others should be examined. To provide the best care and produce optimal reproductive rates, systematic data should continue to be collected, analyzed and shared.

Close examination of changes in absolute concentrations and hormonal ratios (i.e. progestagen/cortisol) may prove to be more informative. We recommend continuing this study and potentially expanding sample size via cooperation with international

60 institutions housing captive maned wolves. Additionally, hormone concentrations should be compared to environmental variables, as they may be sources contributing to stress and reduced progestagen excretion. Studies focusing on the maned wolf should continue to examine noninvasive means for diagnosing early pregnancy, perhaps through urinary steroid concentrations. But other considerations include examining behavioral inadequacies, effects of management practices by zoological institutions and others as potential factors influencing stress and reproductive outcomes in the maned wolf.

Wielebnowski (1998) suggests that non-invasive techniques to assess the adaptation of individuals to husbandry protocols can be very useful in reducing the impact of stress in reproduction. Metabolically monitoring stress and reproductive status is encouraged in the search for better understanding of the factors that may affect breeding in captive maned wolves.

Table 3.1 Summary of female maned wolves sampled with corresponding study periods analyzed and reproductive outcome.

Figure 3.1 A maned wolf fecal sample after it has been lyophilized and pulverized into a fine powder. Photo credit: Lauren Reiter

Figure 3.2 HPLC profiles of immunoreactive glucocorticoids (cortisol and corticosterone) from female maned wolf feces.

Figure 3.3 Frequency distribution with number of female maned wolves that gave birth for each month of the North American maned wolf breeding season (n = 18).

Table 3.2 Fecal progestagen concentrations (µg/g feces) across reproductive phases in female maned wolves of various reproductive outcomes. Different letters within the same reproductive phase indicate differences (P < 0.05).

Figure 3.4 Mean (± SEM) longitudinal profiles of fecal progestagen metabolites for female maned wolves that: (A) conceived and successfully raised pups (n =9); (B) conceived but lost pups early post-partum (n = 7); (C) became pseudo-pregnant (n = 11); and (D) were unpaired (n = 7). Each data point is a pool of 3-day means.

Table 3.3 Overall mean ± SEM fecal cortisol concentrations (µg/g feces) for three groups of female maned wolves across reproductive phases. Ranges were calculated from raw data. Different letters within the same reproductive phase indicate differences (P < 0.05).

Figure 3.5 Mean (± SEM) longitudinal profiles of fecal cortisol metabolites for female maned wolves that: (A) conceived and successfully raised pups (n =10); (B) conceived but lost pups early post-partum (n = 8); and (C) became pseudo-pregnant (n = 12). Each data point is a pool of 3-day means

IV. NON-INVASIVE ASSESSMENT OF LEPTIN CONCENTRATIONS IN THE CAPTIVE MANED WOLF (CHRYSOCYON BRACHYURUS)

Introduction The North American captive maned wolf (Chrysocyon brachyurus) population

(46♂.46♀) has fallen below the target number, 100, set by the Association of Zoos and

Aquariums’ (AZA) Canid and Hyaenid Taxon Advisory Group (TAG) due to poor reproductive success, high neonate mortality and suboptimal health (M. Rodden, pers. comm.). Health issues in the ex situ population is demonstrated through poor body conditions, chronically soft stools, rapid ingesta passage, gingivitis and cystinuria

(Childs-Sanford and Angel, 2006). Furthermore, captive maned wolves have exhibited metabolic and digestive system disorders (Maia and Gouveia, 2000; Soler et al., 2005;

Songsasen and Rodden, 2010). Lastly, creating an optimal diet for this species has proven difficult.

Historically, zoo-managed maned wolves have been fed a typical carnivore diet comprised of raw meat (Childs-Sanford and Angel, 2006; MWSSP, 2007). However, maned wolves are omnivorous with a diet consisting of ~50% plant materials and the remainder comprised of small mammals (Dietz, 1984; Motta-Junior et al., 1996;

Bestelmeyer, 2000; Songsasen and Rodden, 2010). Feeding maned wolves a diet high in animal protein has been linked to this species’ predisposition to cystinuria (Songsasen and Rodden, 2010). Cystinuria, an autosomal recessive disorder, which can lead to

70 complications including urinary tract infections and renal insufficiency, is highly prevalent in captive maned wolves (Childs-Sanford and Angel, 2006). Medical management and dietary modification along with prescribed supplements can aid in altering urine pH from acidic to basic and, therefore, reduce the likelihood of cysteine crystallization. Consequently, zoos in the United States have moved away from carnivore diets and currently provide maned wolves with dog chow-based diets supplemented with various fruits and vegetables in addition to small amounts of whole prey items (Allen, 1995; Songsasen and Rodden, 2010).

Determining the right diet is not the only gastrointestinal issue with maned wolves. For instance, they have a very sensitive gastrointestinal tract and often respond to minor management changes with soft stools or diarrhea (Childs-Sanford and Angel,

2006). Maned wolves are less able to digest dry matter and several minerals including copper, magnesium, iron and sodium than domestic dogs (Childs-Sanford and Angel,

2006), which could contribute to poor body condition. Furthermore, this canid has a rapid ingesta passage, which affects the amount of nutrient and intestinal microflora exposure and absorption (Childs-Sanford and Angel, 2006). All of these digestive problems can be detrimental to a self-sustaining maned wolf population because nutrition is a key determinant of reproductive capability (Mantzoros, 2000; Williams et al., 2002).

Reduced nutrition leads to the suppression of pulsatile luteinizing hormone (LH) release, which in turn terminates gonadal activity in several monogastric species (Nagatani et al.,

2000).

Leptin, a 16-kDa adipocyte-secreted protein, has been indicated as a biomarker for energy balance and nutritional status in many mammalian species, including humans

(Friedman, 2002). Leptin is predominantly produced by white adipose tissue and circulates in plasma in the free and bound form (Mantzoros, 2000). This peptide hormone plays a critical role in regulating body weight and energy homeostasis, as well as a diverse array of biological functions, including reproduction through the regulation of the hypothalamic-pituitary axis (Van Harmelen et al., 1998; Mantzoros, 2000; Henson and Castracane, 2000; Bouillanne et al., 2007). The level of circulating leptin is associated with food intake and nutritional status. Circulating leptin concentrations increase directly with fat mass and may act as an indicator for mass of fat storage, as well as reflecting acute changes in nutritional status (Nagatani et al., 2000; Muehlenbein et al.,

2003; Goodman, 2009). A positive correlation occurs between the release and synthesis of leptin and body mass index (BMI) or percentage of body fat (Garcia-Mayor et al.,

1997; Smith et al., 2002). High leptin concentrations are found in obese individuals compared to normal-weight individuals (Ahima et al., 1996; Garcia-Mayor et al., 1997).

In contrast, low leptin concentrations are found in underweight individuals (Garcia-

Mayor et al., 1997).

Leptin influences the timing of the onset of the first ovulation and maintenance of reproductive function during adulthood (Capiro et al., 2001; Henson and Castracane,

2006). Leptin modulates the release of gonadotropin-releasing hormones (GnRH) from the hypothalamus, which in turn regulates gonadal functions through the secretion of follicle stimulating hormones (FSH) and luteinizing hormones (Henson and Castracane,

2006). In studies of infertile rats, leptin fully restored gonadal function, accelerated puberty and reversed starvation-induced gonadal failure (Ahima et al., 1996; Garcia-

Mayor et al., 1997). Additionally, the onset of puberty is accelerated in juvenile mice and nonhuman primates administered with leptin (Mantzoros, 2000; Henson and

Castracane, 2000). Leptin also can diminish the stress response via lessening the stress- induced activity of the hypothalamic-pituitary-adrenal (HPA) axis (Roubos et al., 2012).

Leptin deficiency affects feedback efficiency of adrenal corticosteroids and, thus, leptin acts directly on the cortical cells to inhibit corticosteroid production (Roubos et al.,

2012).

During the past decade, there has been increased interest in utilizing leptin as a biomarker for assessing body condition in wildlife species. Banks et al. (2001) report captive baboons (Papio anubis) with serum leptin levels roughly three times higher than wild baboons (Papio anubis, Papio hamadryas and Papio anubis/hamadryas hybrids) suggesting wild baboons exhibit a lower percentage of their body weight comprised of adipose tissue. Buff et al. (2002) indicate concentrations of leptin increased with body condition score (BCS) in equids; BCS is a numerical scoring system used to assess the body condition or adiposity of animals (thin to overweight).

A leptin assay developed for the domestic dog (Canis familiaris) reveals plasma leptin is a reliable marker of adiposity in dogs regardless of breed, age and sex (Ishioka et al., 2007). Additionally, a positive relationship exists between plasma leptin concentration and body fat content in dogs (Sagawa et al., 2002). The canine leptin specific assay has also been used to assess levels of this peptide hormone in other

73 carnivores such as the raccoon (Procyon lotor; Shibata et al., 2005) where blood leptin is shown to be a good marker of nutritional status. Results from a multi-species leptin assay demonstrate that seasonal variations in leptin levels occur in the raccoon dog

(Nyctereutes procyonoides; Nieminen et al., 2002; Kitao et al., 2011) and mink (Mustela vison; Tauson and Forsberg, 2002), and circulating leptin concentrations are associated with body weight and BMI. To date, there are no data on leptin levels in the maned wolf, nor is there any information regarding how this peptide may influence reproductive function in this species.

Objectives The goals of the present study were two-fold: 1) to develop and validate a noninvasive assay for measuring urinary leptin concentrations in maned wolves, both female and male; and 2) to correlate urinary leptin concentrations with circulating leptin levels, age, sex, body weight and BMI. This information may prove valuable, enabling leptin to be used as a biomarker for nutritional condition and body adiposity in maned wolves.

Increasing our knowledge about leptin and how it relates with biological factors can help to improve management and reproductive success in this endangered species.

Animals and Facilities From 2009 to 2011, a total of 20 (8 ♂.12 ♀) maned wolves aged 3 to 10.5 years of age were included in the study (n = 34 urine samples). Study subjects were housed under comparable conditions, as set forth in the husbandry guidelines for maned wolves, at zoological facilities across North America. These institutions included the Smithsonian

Conservation Biology Institute (SCBI), Front Royal, Virginia (n = 9 (4.5) maned

74 wolves); Houston Zoo, Houston, Texas (n = 3 (1.2) maned wolves); Dickerson Park Zoo,

Springfield, Missouri (n = 5 (3.2) maned wolves); White Oak Conservation Center

(WOCC), Yulee, Florida (n = 2 (0.2) maned wolves) and Connecticut’s Beardsley Zoo,

Bridgeport, Connecticut (n = 1 (0.1) maned wolf).

Sex, age and institution number were determined using Maned Wolf Population

Analysis and Breeding Plans and Single Population Animal Records Keeping Software

(SPARKS) based on house name and studbook number labeled on samples. Body length, measured in meters (m) from the tip of the nose to tip of tail (see Appendix B), along with body mass, measured in kilograms (kg) were collected opportunistically in six male maned wolves during physical examinations. BMI was calculated based on body mass and length using the formula:

�� (��) �� = (��ℎ � )!

Urine Samples Fresh urine (~5 mL to 10 mL) was collected opportunistically via a collection tray, which was mounted on the fence in the animal’s enclosure, or during examination. Urine from each individual was collected in a separate tray and kept frozen at -20ºC until transported to SCBI for leptin analysis using a radioimmunoassay (see below). Additionally, nine urine specimens, from maned wolves that were banked at SCBI, were thawed and used in this study. These samples were collected between 2004 and 2007 (see Appendix A for complete dates).

Blood Samples If possible (i.e. during physical examinations), it was requested that blood samples be collected from the maned wolves in a red-top tube (with no anticoagulant) twice each year: once during mid-breeding season (November through January) and once during non-breeding season (May through August). Serum samples were collected from ten maned wolves. Each serum sample (n = 10) was centrifuged, and the serum obtained was stored at -20°C before being shipped to SCBI for leptin analysis.

Leptin Radioimmunoassay Urinary and serum leptin concentrations were assessed using a multi-species leptin radioimmunoassay (RIA) kit that utilized 125I-labeled human leptin and a multi-species leptin antiserum (Cat. # XL-85K, Millipore, St. Charles, MO). Samples were assayed as described by Millipore (2011). The antibody displayed broad cross reactivity to leptin molecules in many species including human leptin (100%), porcine leptin (67%), rat leptin (61%) and canine leptin (3%) (Millipore, 2011). Assay sensitivity was 0.801 ng/mL + 2SD Human Equivalent (HE). Inter-assay CV for two separate internal controls

(n = 2 assays for all samples) were 8.74 and 12.59% and intra-assay CV was < 10%.

The multi-species leptin RIA was validated by demonstrating: 1) parallelism between serial dilutions of maned wolf serum and urine extract and the standard curve and 2) recovery of exogenous hormone added to urine extract as described below.

Creatinine Enzyme Immunoassay Creatinine (Cr) levels were determined for each urine sample to adjust for inter-sample differences in water concentration, as previously described (Munro et al., 1991). Samples with < 0.1 ng/mL Cr at 1:10 dilution were too dilute and consequently excluded from

76 analyses (n = 3 samples). Hormone mass in a urine sample was divided by Cr concentration and expressed as mass of hormone per milligram Cr (ng/mg Cr).

Assay Validation Parallelism was determined by comparing serially diluted pooled maned wolf urine and pooled serum samples to the standard curve. Blood and urine samples collected on the same day from males and females were used to make pooled samples. Pooled urine and serum samples were serially diluted nine times (neat to 1:512) and then assayed each aliquot for leptin concentration.

A sample pool was made from the urine of both male and female maned wolves.

Then, aliquots of the pooled sample were spiked with an equal amount from each standard, except the lowest. The spiked samples were run as unknowns in the assay. The amount of endogenous hormone present was determined by running the sample pool without any standard added, which was later subtracted from each spiked sample (Brown et al., 2008). The amount expected (concentration of standard spiked with/2) was plotted against the amount observed (concentration observed from assay results − background) and linear regression (y = mx + b) analysis was conducted. Slopes > or < 1 suggested an over or underestimation of hormone mass, respectively (Brown et al., 2008). Linear regression analysis (R2) was used to evaluate a relationship between standard and urinary leptin concentrations.

All samples for enzyme immunoassay (EIA) and radioimmunoassay (RIA) analysis were assessed in duplicate, and samples with a coefficient of variation (CV)

77 greater than 10% were excluded. Standard curves along with recovery and accuracy checks were prepared with each assay.

Statistical Analysis All statistical analyses were conducted using Microsoft Excel 2011 (Version 14.1.4,

Microsoft Corporation, Seattle, WA) and SPSS Statistics (Version 19, SPSS Inc.,

Chicago, IL) software for the Macintosh. For all analyses, significance was assessed at the P = 0.05 level.

Descriptive statistics were run on all variables. Outliers were defined as high values of urinary leptin concentration (exceeding � + 1.5 [SD]). Relationships were analyzed between urinary leptin concentrations and all variables. Linear regression analysis was performed on the leptin recovery. In the cases where a maned wolf contributed more than one urine sample, the most recent sample was used for analysis to avoid pseudo replication. Sex differences in urinary leptin were determined using an independent samples t-test. Correlation analyses were carried out to determine any relationship between urinary leptin with age, body weight and BMI. Lastly, a paired samples t-test was used to compare the effect of age on urinary leptin in 12 samples where identified maned wolves (n = 6) had multiple urine samples collected spanning over a minimum of at least one year. Serum leptin concentrations were not analyzed with respect to sex, age or body condition measures due to an insufficient number of samples.

All results were reported as mean ± standard error, unless otherwise noted and significance was determined at P < 0.05.

Serum and Urinary Leptin Assay Validation Serial dilutions of pooled serum and pooled urine displayed parallelism with the standard curve; therefore, demonstrating immunoactivity of endogenous antigens similar to the assay standards (Fig. 4.1). Net recovery of 1.56, 3.13, 6.25, 12.5, 25 and 50 ng leptin added to 200 µL neat urine was 1.12, 0.98, 1.72, 1.97, 3.86 and 8.46 ng, respectively (y =

0.31x + 0.51, R2 = 0.98; Fig. 4.2).

Measuring Urinary Leptin Concentrations Mean urinary leptin concentrations measured in the maned wolf were 2.70 ± 0.43 ng/mg

Cr. One urine sample (14.80 ng/mg Cr) provided by a male was determined to be an outlier and removed from further analyses. Subsequently, the mean concentration adjusted without the outlier was 2.31 ± 0.20 ng/mg Cr (n = 31 samples).

Effect of Sex Sex did not influence (t17 = -0.52, P > 0.05) urinary leptin concentration in the maned wolf (males: 2.30 ± 0.44 ng/mg Cr, range: 0.35-4.29 ng/mg Cr, n = 7; females: 2.63 ±

0.41 ng/mg Cr, range: 0.80-5.32 ng/mg Cr, n = 12).

Effect of Body Weight, Length and Body Mass Index Maned wolves, on average, weighed 26.1 ± 0.6 kg (Range: 20.5-30.5 kg) with males weighing significantly (t18 = 2.31, P = 0.03) more than females (males: 27.7 ± 0.9 kg, range: 24.0-31.0 kg, n = 8; females: 25.1 ± 0.7 kg, range: 20.5-29.5 kg, n = 12).

However, no significant correlation between weight and urinary leptin concentration was found (r17 = 0.21, P > 0.05).

Male maned wolves had a nose-to-tail length of 1.07 ± 0.03 m (range: 0.99-1.16 m, n = 6). Results indicated a marginally significant correlation between urinary leptin

79 concentration and body length (r3 = 0.73, P = 0.09, n = 5). Based on six male maned wolf samples (Table 4.1), mean BMI was 24.27 ± 0.80 (range: 21.78-27.51, n = 6). BMI was not related to (r4 = -0.26, P > 0.05) urinary leptin concentrations in these six maned wolves. However, when the outlier (2954) was removed, BMI was shown to have a marginally significant correlation with urinary leptin (r3 = -0.70, P = 0.09, n =5).

Effect of Age Preliminary findings suggest a positive association between age and urinary leptin concentration. A paired samples t-test revealed a positive correlation between age and urinary leptin concentrations (t5 = -2.98, P = 0.03), where all of the maned wolves had higher leptin concentrations in the second sample collected from them (Fig. 4.3) (see

Appendix C for raw data).

Discussion This investigation provides the first examination of leptin hormone concentrations in the maned wolf. We confirmed that leptin concentrations could be measured in maned wolf serum and urine. Using a multi-species leptin RIA kit, we showed serial dilutions of pooled serum and pooled urine displayed parallelism with the standard curve. However, the slope indicated serum and urine were possibly interfering with the accuracy of hormone quantification (Wasser et al., 1991). Thus while this RIA kit was able to detect serum and urine concentrations of leptin, it revealed a gross under estimation of hormone mass, which may have been caused by unknown hormones interfering with the RIA kit.

Similarly, Kitao et al. (2011) report relatively low serum leptin concentrations in raccoon dogs when using the multi-species leptin RIA kit, the same assay used in this study.

Based upon the results of these two studies, we do not recommend using the multi- species RIA kit for determining canine leptin concentrations. Leptin concentrations measured by the canine leptin-specific enzyme-linked immunosorbent assay (ELISA) kit, which uses dog leptin as a standard, yields concentrations 20-fold higher than the concentrations obtained by the multi-species leptin RIA kit (Kitao et al., 2011).

Consequently, it may be necessary to utilize the canine leptin-specific ELISA kit for determining baseline values of leptin concentrations in the maned wolf, despite the significantly higher cost, if other assays reveal an underestimation of hormone mass.

Preliminary findings suggested mean urinary leptin concentrations in the maned wolf to be 2.31 ± 0.20 ng/mg Cr. Based on the results of this study, there were no variations in leptin concentrations between males and females. Maned wolves exhibit no sexual dimorphism and this may explain why there is no effect of sex on leptin concentration. In spite of a lack of dimorphism in maned wolves, these results of the present study agree with previous studies on canids. Serum leptin concentrations measured by the canine leptin-specific ELISA kit reveal no significant differences between male and female raccoon dogs (Kitao et al., 2011). Similarly, Ishioka et al.

(2007) describe no significant difference in plasma leptin between male and female domestic dogs. Perhaps, sex does not affect leptin concentrations in canids, as it does in other species. For example, Buff et al. (2002) show greater serum concentrations in geldings and stallions than in mares. Thus, gender differences in leptin concentrations may be linked to a species-specific role.

We were unable to conclude any correlation between body weight and urinary leptin concentrations. It is possible that maned wolf leptin concentrations and body weight are positively correlated, but we were simply unable to detect this correlation due in part to a small sample size. Other canid species including the domestic dog and raccoon dog report a positive correlation between serum leptin concentrations and body weight (Jeusette et al., 2005; Kitao et al., 2011). We suggest repeating this study using different assays and utilizing a larger sample size of maned wolf subjects, increasing the quantity of opportunistic serum sample collection and implementing longitudinal urine sample collection.

Length positively correlated with urinary leptin concentrations in the maned wolf.

Generally, it would be expected that the longer the body length, as measured from tip of nose to tip of tail, would correlate with larger sized individuals (i.e. body mass). Leptin levels tend to rise as weight increases and, then, decrease as weight goes down (Jéquier,

2002; Oswal and Yeo, 2010). Therefore, an animal with greater body length would be expected to have higher leptin concentrations because they tend to weigh more.

Similar to our findings with body weight, we were unable to determine any significant correlation between BMI and urinary leptin concentration. Measuring BMI, while convenient, safe and of minimal cost, is imprecise and outdated (Shah and

Braverman, 2012). Furthermore, BMI is not a measurement of adiposity (Romero-Corral et al., 2008; Shah and Braverman, 2012). In domestic dogs, Nishii et al. (2010) describe significant correlation between plasma leptin concentration and body condition score

(BCS). The BCS or body condition index (BCI) is a numerical scoring system used to

82 assess the body condition of animals (Buff et al., 2002). Other studies have utilized BCS, which may prove to be more effective in the maned wolf. We do not suggest using BMI, but recommend developing a body condition scoring system specifically designed for the maned wolf and examining any relationship between the maned wolf specific BCS system with leptin concentrations.

In our study, we reported 14.80 ng/mg Cr for one maned wolf, which was the highest value among all subjects measured. This individual, also, exhibited the highest

BMI, which lead us to question whether BMI was an appropriate tool or if another measurement, such as BCS, may be more effective and, if so, would it reveal a positive correlation with leptin concentrations in the maned wolf. Perhaps, this warrants further study with a greater sample size.

Based upon a subset of samples we described a significant positive correlation between age and urinary leptin concentrations in the maned wolf. Young maned wolves exhibit lower body fat, as they are developing into mature individuals. It would be expected that mature maned wolves would present with greater body mass and only require resources (e.g., nutrients, energy, adiposity) to sustain their activity level. Lastly, higher leptin concentrations found in post-reproductive individuals may be attributed to less active lifestyles and metabolic changes. Age-dependent increases of serum leptin are reported in domestic dogs, as well (Nishii et al., 2010). Similar trends are shown in other species including humans (Baumgartner et al., 1999; Ram and Malathi, 2007) and horses

(Buff et al., 2002).

In summary, we showed that age and length, but not sex and body weight or BMI influence urinary leptin concentration in the maned wolf. Despite the small sample size, this study demonstrates that leptin monitoring may serve as a valuable management tool.

Measuring leptin concentrations twice monthly noninvasively may alert management to changes in body condition and permit assessments of nutritional status. While the multi- species leptin RIA was able to detect leptin in the maned wolf, the anti-human leptin antibody was not as specific for canine leptin and, thus, is not appropriate for measuring this peptide hormone in this species. The lack of some differences obtained in this study may be due to the gross under estimation of hormone mass from the multi-species assay.

Maned wolves are seasonal breeders and, as such, it would be expected that individuals would modify their winter energy expenditures and metabolic rate. We were unable to obtain seasonal samples from maned wolves. Consequently, future studies should assess seasonal fluctuations in leptin concentrations in males and females, as well as monitoring leptin concentrations in females throughout the reproductive cycle. Maned wolves that successfully reproduce and become pregnant would be expected to increase food intake and, therefore show an increase in body mass and fat stores, thus increasing leptin concentrations. Comparing leptin concentrations in pregnant and pseudo-pregnant captive maned wolves could help researchers learn more about leptin and its role in pregnancy in the maned wolf and, thereby, warrants further exploration.

Maned wolves require a diverse and palatable diet in captivity. Leptin monitoring can be used to measure nutritional status and act as a proxy to inform management whether nutritional requirements are not being met to avert prolonged periods of

84 inadequate nutrition, thereby preventing severe health problems. Vigilantly monitoring leptin concentrations along with body condition and other biological markers throughout an animal’s life will collectively provide a normal profile for that specific animal and help to alert management when significant changes occur. Leptin monitoring will provide insight into changes in the nutritional and metabolic state of individual maned wolves. Once baseline values have been determined in zoo-managed maned wolves, future studies should examine and compare leptin concentrations between maned wolves held ex situ to those managed in situ.

There is considerable variability in leptin concentrations among individuals, so it is advised that a more effective means of measuring body fatness in the maned wolf be developed or that current measurements are supplemented with additional biological measures (e.g. girth measurement and height) (see Appendix D for more examples of measurements). Consistently measuring urinary leptin concentrations in the maned wolf may help the Maned Wolf Species Survival Plan® (MWSSP) monitor nutritional status and, additionally, improve reproduction in the captive population. Further studies need to focus on which assays are best applicable to measuring leptin concentrations in the maned wolf before more complex and informative leptin studies are to be devised and executed in this species.

Figure 4.1 Parallelism curves resulting from RIA of serially diluted leptin standards and pooled serum and pooled urine. Based on the above graph, urine samples should be run at neat given the binding inhibition results observed at ~50%.

Figure 4.2 Maned wolf urinary leptin recovery.

Table 4.2 Age, leptin concentration, weight, length and BMI measurements for six male maned wolves.

Figure 4.3 Mean leptin concentrations (ng/mg Cr) for six maned wolves where urine samples were collected more than 1 year apart from the same individuals (Age A = initial sample, Age B = subsequent sample).

V. CONCLUSION

With the overall goal of improving understanding of the causes of poor reproduction in maned wolves (Chrysocyon brachyurus) maintained ex situ, this thesis compared reproductive and stress hormones among females with various reproductive outcomes.

Because nutritional and health status may have contributed to poor reproduction in this species, a noninvasive urinary leptin assay was validated in both male and female maned wolves and, then, values were correlated with several biological parameters.

The present study demonstrated that: 1) females that conceived and raised pups excreted significantly lower fecal glucocorticoid metabolite concentrations compared to females that became pseudo-pregnant, as well as pregnant females that experienced neonate loss; 2) pseudo-pregnant females, on average, excreted lower fecal progestagen metabolite concentrations than pregnant bitches; 3) maned wolves are induced ovulators; and 4) urinary leptin concentrations can be detected using a multi-species leptin radioimmunoassay (RIA), although there appeared to be a gross underestimation of hormone mass.

Although a preliminary report describes significantly higher fecal glucocorticoid metabolite concentrations during the peri-ovulatory phase in females that lost pups compared to females that raise pups (MWSSP, 2007), we showed that with a larger sample size, significantly higher fecal glucocorticoid concentrations were excreted during

90 the luteal phase by females that became pseudo-pregnant and females that lost pups, whereas pregnant females that raised pups excreted significantly lower corticoid concentrations.

Approximately half, 55%, of all pregnant maned wolves in the present study successfully raised pups. The data obtained from monitoring reproductive and stress hormones indicate that there is a relationship between pup mortality and elevated corticoids. Factors linked to elevated corticoid concentrations (e.g., suboptimal enclosure environment and health) may also have contributed to neonatal loss. Thus, future research should focus on identifying ‘stressors’ and developing strategies toward mitigating stress. Furthermore, the information that females raising pups excreted the highest mean progestagen metabolite concentrations during the luteal phase compared to females that lost pups and pseudo-pregnant individuals may be useful for predicting pregnancy and potential pup loss in captive maned wolves, which would contribute to cost-effective monitoring around parturition (Wasser et al., 1995). Such advanced notice permits management to prepare for whelping and, perhaps, discouraging infanticide

(Velloso et al., 1998).

Unpaired females, of prime reproductive age, failed to ovulate as indicated by sustained baseline fecal progestagen metabolite concentrations throughout the breeding season. Ovulation only occurred in females paired with a male, which indicated that the maned wolf is an induced ovulator. This study is the first to confirm that the maned wolf is an induced ovulator. By understanding this mechanism, assisted reproductive

91 technologies can be developed and methods improved for inducing ovulation and increasing reproductive success.

This study presents the first research to analyze leptin in the maned wolf. A multi-species leptin RIA kit was validated and, then, modified for measuring urinary leptin concentrations in maned wolves. Based upon the results of this study, sex did not influence urinary leptin concentration in the maned wolf, nor did weight or BMI. We did report a significant positive correlation between body length and urinary leptin concentrations, as well as age and urinary leptin concentration in the maned wolf. Maned wolves with longer body lengths would presumably have larger body mass and fat stores and, therefore, higher leptin concentrations as larger individuals require higher caloric intake to sustain their current physical state. Similarly, as maned wolves mature, it is expected that fat stores would increase. Young maned wolves are still developing and will exhibit lower body fat stores as compared to mature maned wolves that require greater body mass and fat stores to sustain their body weight and energy level.

Unfortunately, urinary leptin recovery revealed a gross underestimation of hormone mass. Therefore, the concentrations reported in this study are most likely relatively low due to the leptin RIA used and, in reality, may actually be significantly higher.

Consequently, we do not recommend using the multi-species leptin RIA for measuring maned wolf leptin. The next step should focus on evaluating other RIAs and enzyme- linked immunosorbent assays (ELISAs) for determination of leptin concentrations in serum and urine samples from maned wolves.

Establishing baseline leptin concentrations for each individual maned wolf held in captivity and general values applicable to the captive population would benefit management in monitoring nutritional status, body condition changes and other measures throughout the animal’s lifetime. Leptin has the potential to be used in different capacities, in addition to acting as a biomarker for measuring nutritional status.

Future studies should seek to evaluate leptin and its changes throughout the reproductive season, as well as comparing concentrations based on female reproductive status. These studies could provide insight into leptin’s influence in pregnant versus pseudo-pregnant and unpaired females, and expand our knowledge of leptin and its role throughout the reproductive process. A larger sample size coupled with a longer sampling period would enhance data results for determining leptin concentrations in the maned wolf. Leptin monitoring can be applied to non-reproductive individuals, as well.

For example, urinary leptin can be used as an indicator for health status (i.e. responses to dietary changes, irritable bowel disease, etc.). Creating a specific Body Condition Score

(BCS) system for maned wolves based on specific measurements (i.e. nose to base-of- tail, girth, neck, etc.) may aid in establishing a more precise monitoring system for assessing the well-being of zoo-housed maned wolves.

We, as animal management staff, need to examine and try to understand what maned wolves, as a species and as individuals, perceive to be threats. Stimulus levels, proximity of conspecifics and heterospecifics, noise levels, enrichment and training programs, type of facility (i.e. research, breeding or display), individual animal temperaments, enclosure size, enclosure type, nutritional status, climate, etc. are factors

93 that should be measured and applied when quantifying reproductive and adrenal hormones.

Relatively few individual maned wolves are located at one single facility, adding variability when monitoring stress. Housing conditions are not the same across zoos and may influence an animal. Cases where an animal has been recently transferred per breeding recommendations or heavy construction took place during the breeding season should be taken into consideration in addition to examining and even quantifying the amount of control and choice an animal has in its environment.

Study design can be improved with minor adjustments. Since copulation may not always be observed, documenting affiliative behaviors and corresponding dates would be advantageous and valuable for data analyses. For future studies, a package including a short questionnaire (e.g., Appendix E), sample inventory and necessary supplies for sample collection would overcome discrepancies and increase efficiency, overall.

In summary, maned wolves continue to reproduce poorly in captivity and exhibit persistent health problems. It is a challenge in attempting to understand and isolate potential threats contributing to increased adrenal activity in this species, but we need to examine internal and external factors as potential stressors in captive maned wolves before we can improve reproductive success. Females that raise pups experience lower cortisol concentrations, overall. Exploring biotic and abiotic factors that impact pairs recommended for breeding may enlighten us to potential husbandry issues and monitoring leptin concentrations noninvasively can be used as a biomarker for assessing status in this species. Longitudinal monitoring of fecal gonadal and adrenal hormones

94 along with urinary leptin concentrations together will provide a bigger picture into how an animal perceives and responds to their environment.

APPENDICES

APPENDIX A

Urinary leptin concentrations (ng/mg Cr) and morphometric measures for 20 (8.12) maned wolves.

APPENDIX B

Maned wolf body length measurements were based on values from tip of nose to tip of tail as shown below.

NOSE-TO-TAIL

APPENDIX C

Mean urinary leptin concentrations (ng/mg Cr) for six maned wolves where urine samples were collected more than 1 year apart from the same individuals (Age A = initial sample, Age B = subsequent sample).

APPENDIX D

Specific morphometric measures suggested for collection in future studies on maned wolves.

100

APPENDIX E

Short questionnaire to be filled out in future studies by institution holding female maned wolves participating in reproductive study.

101

APPENDIX F

National Zoological Park – Institutional Animal Care and Use Committee (IACUC) proposal approval.

Smithsonian

National Zoological Park

MEMORANDUM

DATE: June 13, 2011

TO: Nucharin Songsasen

FROM: Dr. William McShea Chair, NZP Institutional Animal Care and Use Committee Pierre Comizzoli Designated Reviewer

SUBJECT: NZP-IACUC Action on Proposal #11-20

I am pleased to notify you that your proposal, entitled “Relationship between leptin production, body condition and reproductive success in maned wolves” has been approved by the NZP Institutional Animal Care and Use Committee (NZP-IACUC).

Approval for this submission is granted with the understanding that no changes will be made in the animal handling and care protocols specified in your proposal and subsequent response without prior consultation with, and approval from, the NZP- IACUC. This approval is valid for three (3) years from the date of this memorandum. If your project will go beyond three years, you must request a renewal of the ACUC approval when this one expires. This approval will expire on 6/13/14.

As a part of ongoing review of approved IACUC activities, we require that you provide, as mandated, a brief annual update on the progress of your animal handling and care protocols prior to one year from the date of this letter (6/13/12). If in the course of your research you experience unexpected morbidity or mortality, you are required to contact the IACUC Chair and provide a detailed accounting of the circumstances surrounding these events, within one month of the event.

Please be advised that it is your responsibility to obtain all necessary permits and/or authorizations that may be required to conduct this research or to import or export any biological samples. Please contact Laura Morse, Registrar of the Smithsonian National Zoological Park, to clarify permit issues.

We hope that your research is a success.

102

REFERENCES

103

REFERENCES

Ahima, R. S., Prabakaran, D., Mantzoros, C., Qu, D., Lowell, B., Maratos-Flier, E., Flier, J. S. 1996. Role of leptin in the neuroendocrine response to fasting. Nature. 382:250–2.

Allen, M. E. (1995). Maned wolf nutritional management. In: Husbandry Manual for the Maned Wolf Chrysocyon brachyurus. Fletchall, N. B., Rodden, M., Taylor, S. (editors). American Association of Zoos and Aquariums.

Aragona, M., Setz, E. Z. F. 2001. Diet of the maned wolf, Chrysocyon brachurus) (Mammalia: Canidae), during wet and dry seasons at Ibitipoca State Park, Brazil. J Zool Lond 254:131-6.

Asa, C. S., Miller, P., Agnew, M., Rebolledo, J. A. R., Lindsey, S., Callahan, M., Bauman, K. 2007. Relationship of inbreeding to sperm quality and reproductive success in Mexican gray wolves (Canis lupus baileyii). Anim Conserv 10:326-31.

Asa, C. S., Valdespino, C. 1998. Canid reproductive biology: an integration of proximate mechanisms and ultimate causes. Amer Zool 38:251-9.

Asikainen, J., Mustonen, A. M., Hyvarinen, H., Nieminen, P. 2004. Seasonal physiology of wild raccoon dog (Nyctereutes procyonoides). Zool Sci 21:385-91.

Bakken, M. 1992. The relationship between open field activity, competition capacity and first year reproductive success among farmed silver fox cubs (Vulpes vulpes). Norw J Agric Sci Suppl(9):520-8.

Bakken, M. 1993. The relation between competition capacity and reproduction in farmed silver foxes vixens, Vulpes vulpes. J Anim Breed Genet 110:147-55.

Bakken, M., Moe, R. O., Smith, A. J., Selle, G. E. 1999. Effects of environmental stressors on deep body temperature and activity levels in silver fox vixens (Vulpes vulpes). Appl Anim Behav Sci 64:141-51.

Banks, W. A., Phillips-Conroy, J. E., Jolly, C. J., Morley, J. E. 2001. Serum leptin levels in wild and captive populations of baboons (Papio): implications for the ancestral role of leptin. J Clin Endocrinol Metab 86(9):4315-20.

104

Baumgartner, R. N., Waters, D. L., Morley, J. E., Patrick, P., Montoya, G. D., Gary, P. J. 1999. Age-related changes in sex hormones affect the sex difference in serum leptin independently of changes in body fat. Metab 48(3):378-84.

Bestelmeyer, S. V. 2000. Solitary, Reproductive and Parental Behavior of Maned Wolves (Chrysocyon brachyurus)[dissertation]. [Fort Collins (CO)]: Colorado State University. 203 p.

Biancani, B., Da Dalt, L., Lacave, G., Romagnoli, S., Gabai, G. 2009. Measuring fecal progestogens as a tool to monitor reproductive activity in captive female bottlenose dolphins (Tursiops truncatus). Theriogenology 72:1282-92.

Bouillanne, O., Golmard, J-L., Coussieu, C., Noël, M., Durand, D., Piette, F., Nivet- Antoine, V. 2007. Leptin a new biological marker for evaluating malnutrition in elderly patients. Eur J Clin Nutr 61:647-54.

Brady, C. A., Ditton, M. K. 1979. Management and breeding of maned wolves at the National Zoological Park, Washington. Int Zoo Yrbk 19:171-6.

Brown, J. L. 2008. Wildlife endocrinology manual. CRC Endocrine Research Laboratory. Smithsonian National Zoological Park, Conservation and Research Center (internal publication). 105 p.

Brown, J. L. 2000. Reproductive endocrine monitoring of elephants: an essential tool for assisting captive management. Zoo Biol 19:347-67.

Brown, J. L., Wasser, S. K., Wildt, D. E., Graham, L. H. 1994. Comparative aspects of steroid hormone metabolism and ovarian activity in felids, measured noninvasively in feces. Biol Reprod 51(4):776-86.

Buff, P. R., Dodds, A. C., Morrison, C. D., Whitley, N. C., McFadin, E. L., Daniel, J. A., Dijane, J., Keisler, D. H. 2002. Leptin in horses: Tissue localization and relationship between peripheral concentrations of leptin and body condition. J Anim Sci 80(11):2942-9.

Bush, M. B. 1980. Medical management of maned wolves (Chrysocyon brachyurus). Ann Proc Amer Assoc Zoo Vets 1980:131-4.

Capiro, M., Fabbrini, E., Isidori, A. M., Aversa, A., Fabbri, A. 2001. Leptin in reproduction. Trends Endocrinol Metab 12(2):65-72.

Carlstead, K. 1996. Effects of captivity on the behavior of wild mammals. In: Kleiman, D. G., Allen, M. E., Thompson, K. V., Lumpkid, S. (editors). Wild Mammals in Captivity. Chicago: Univ of Chicago Pr. p 317-33.

105

Carlstead, K., Brown, J. 2005. Relationships between patterns of fecal corticoid excretion and behavior, reproduction, and environmental factors in captive black (Diceros bicornis) and white (Ceratotherium simum) rhinoceros. Zoo Biol 24:215-32.

Carlstead, K., Brown, J. L., Seidensticker, J. 1993. Behavioral and adrenocortical responses to environmental changes in leopard cats (Felis bengalensis). Zoo Biol 12(4):321-31.

Carlstead, K., Shepherdson, D. 1994. Effects of environmental enrichment on reproduction. Zoo Biol 13:447-58.

Charlesworth, D., Willis, J. H. 2009. The genetics of inbreeding depression. Nature 10:783-96.

Childs-Sanford, S. E., Angel, C. R. 2006. Transit time and digestibility of two experimental diets in the maned wolf (Chrysocyon brachyurus) and domestic dog (Canis lupus). Zoo Biol 25:369-81.

Clark, F. E., Fitzpatrick, M., Hartley, A., King, A. J., Lee, t., Routh, A., Walker, S. L., George, K. 2011. Relationship between behavior, adrenal activity, and environment in zoo-housed western lowland gorillas (Gorilla gorilla gorilla). Zoo Biol 30:1-16.

Comizzoli, P., Crosier, A. E., Songsasen, N., Gunther, M. S., Howard, J. G., Wildt, D. E. 2009. Advances in reproductive science for wild carnivore conservation. Reprod Domest Anim 44(Suppl 2):47-52.

Concannon, P. W. 2011. Reproductive cycles of the domestic bitch. Anim Reprod Sci 124(3-4):200-10.

Concannon, P. W., Castracane, V. D., Temple, M., Montanez, A. 2009. Endocrine control of ovarian function in dogs and other carnivores. Anim Reprod 6(1):172-93.

Concannon, P. W., McCann, J. P., Temple, M. 1989. Biology and endocrinology of ovulation, pregnancy and parturition in the dog. J Reprod Fertil Suppl 39:3-25.

Cooper, C., Evans, A. C. O., Cook, S., Rawlings, N. C. 1995. Cortisol, progesterone and β-endorphin response to stress in calves. Can J Anim Sci 95:197-201.

Courtenay, O. 1994. Conservation of the Maned Wolf: fruitful relationships in a changing environment. Canid News 2.

106

Creel, S., Fox, J. E., Hardy, A., Sands, J., Garrott, B., Peterson, R. O. 2002. Snowmobile Activity and Glucocorticoid Stress Responses in Wolves and Elk. Conserv Biol 16(3):809-14.

Cummings, D., Brown, J. L., Rodden, M. D., Songsasen, N. 2007. Behavioral and physiologic responses to environmental enrichment in the maned wolf (Chrysocyon brachyurus). Zoo Biol 26:331-43. de Paula, R. C., Medici, P., Morato, R. G. 2008. Plano de ação para conservação do Lobo-Guará: analise de viabilidade populacional e de habitat. Brasilia, Brazil: Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis (IBAMA).

Dietz, J. M. 1985. Chrysocyon brachyurus. Mammalian Species 234:1-4.

Dietz J. M. 1984. Ecology and social organization of the maned wolf, Chrysocyon brachyurus. Smithsonian Cont Zool 392:1-51.

Donoghue, S., Langenberg, J. 1994. Clinical nutrition of exotic pets. Aust Vet J 71:337- 41.

Dorsey, C., Dennis, P., Guagnano, G., Wood, T., Brown, J. L. 2010. Decreased baseline fecal glucocorticoid concentrations associated with skin and oral lesions in black rhinoceros (Diceros bicornis). J Zoo Wildl Med 41(4):616-25.

Ebenhard, T. 1995. Conservation breeding as a tool for saving animal species from extinction. Trends Ecol Evol 10:438-43.

Emmons, L., Deem, S. L., Rodden, M., Maldonado, J. 2004. The secret wolf. Zoogoer 33(6).

Feldman, E. C., Nelson, R. W. 1987. Canine and feline endocrinology and reproduction. Philadelphia: Saunders.

Friedman, J. M. 2002. The Function of Leptin in Nutrition, Weight, and Physiology. Nutr Rev 60(10):S1-14.

Garcia-Mayor, R. V., Amalia Andrade, M., Rios, M., Lage, M., Dieguez, C., Casanueva, F. F. 1997. Serum Leptin Levels in Normal Children: Relationship to Age, Gender, Body Mass Index, Pituitary-Gonadal Hormones, and Pubertal Stage. J Clin Endocrinol Metab 82(9):2849-55.

107

Gasper-López, E., Casabiell, J., Estevez, J. A. Landete-Castillejos, T., De La Cruz, L. F., Gallego, L., García, A. J. 2009. Seasonal changes in plasma leptin concentration related to antler cycle in Iberian red deer stags. J Comp Physiol B 179:617-22.

Goodman, H. M. 2009. Basic medical endocrinology. 4th ed. New York: Academic Pr. 309 p.

Greenstein, B., Wood, D. 2010. The endocrine system at a glance. 2nd ed. Malden: Blackwell Pub. 120 p.

Gudermuth, D. F., Concannon, P. W., Daels, P. F., Lasley, B. L. 1998. Pregnancy- specific elevations in fecal concentrations of estradiol, testosterone and progesterone in the domestic dog (Canis familiaris). Theriogenology 50:237-48.

Henson, M. C., Castracane, V. D. 2000. Leptin in Pregnancy. Biol Reprod 63:1219-28.

Hosey, G., Melfi, V., Pankhurst, S. 2009. Zoo animals: Behaviour, management, and welfare. New York: Oxford Univ Pr. 661 p.

Ishioka, K., Hosoya, K., Kitagawa, H., Shibata, H., Honjoh, T., Kimura, K., Saito, M. 2007. Plasma leptin concentration in dogs: effects of body condition score, age, gender and breeds. Res Vet Sci 82(1):11-5.

IUCN. 2008. 2008 IUCN red list of threatened species. Gland, Switzerland, and Cambridge, UK: IUCN. Available at http://www.iucnredlist.org (accessed 5 May 2010).

Jácomo, A. T. A., Kashivakura, C. K., Ferro, C., Furtado, M. M., Astetes, S. P., Tôrres, N. M., Sollman, R., Silveira, L. 2009. Maned wolf home-range and spatial organization in the Brazilian Grasslands. J Mammal 90:150-7.

Jácomo, A. T. A., Silveira, L., Diniz-Filho, J. A. F. 2004. Niche separation between the maned wolf (Chrysocyon brachyurus), the crab- eating fox (Dusicyon thous) and the hoary fox (Dusicyon vetulus) in central Brazil. J Zool Lond 262:99–106.

Jéquier, E. 2002. Leptin Signaling, Adiposity, and Energy Balance. Ann NY Acad Sci 967:379-88.

Jeusette, I. C., Detilleux, J., Shibata, H., Saito, M., Honjoh, T., Delobel, A., Istasse, L., Dietz, M. 2005. Effects of chronic obesity and weight loss on plasma ghrelin and leptin concentrations in dogs. Res Vet Sci 79(2):169-75.

Jurke, M. H., Czekala, N. M., Lindburg, D. G., Millard, S. E. 1997. Fecal Corticoid Metabolites Measurement in the Cheetah (Acinonyx jubatus). Zoo Biol 16:133-47.

108

Karanth, K. U., Chellam, R. 2009. Carnivore conservation at the crossroads. Oryx 43:1- 2.Kawate, N., Inaba, T., Mori, J. 1993. Effects of cortisol on the amounts of estradiol-17β and progesterone secreted and the number of luteinizing hormone receptors in cultured bovine granulosa cells. Anim Reprod Sci 32(1-2):15-25.

Kitao, N., Fukui, D., Shibata, H., Saito, M., Osborne, P. G., Hasimoto, M. 2011. Seasonality and Fasting Effect in Raccoon Dog Nyctereutes procyonoides Serum Leptin Levels Determined by Canine Leptin-Specific Enzyme-Linked Immunosorbent Assay. J Exp Zool 315:84-9.

Kleiman, D. G. 1972. Social behavior of the maned wolf (Chrysocyon brachyurus) and bush dog (Speothos venaticus): a study in contrast. J Mammal 53(4):791-806.

Laikre, L., Ryman, N. 1991. Inbreeding Depression in a Captive Wolf (Canis lupis) Population. Conserv Biol 5:33-40.

Lasley, B. L., Kirkpatrick, J. F. 1991. Monitoring Ovarian Function in Captive and Free- Ranging Wildlife by Means of Urinary and Fecal Steroids. J Zoo Wildl Med 22(1):23-31.

Lees, C. M., Wilcken, J. 2009. Sustaining the Ark: the challenges faced by zoos in maintaining viable populations. Int Zoo Yrbk 43(1):6-18.

Liptrap, R. M. 1993. Stress and reproduction in domestic animals. Ann NY Acad Sci 697:275-83.

Lyons, J., Young, R. J., Deag, J. M. 1997. The effects of physical characteristics of the environment and feeding regime on the behavior of captive felids. Zoo Biol 16:71- 83.

Maia, O. B., Gouveia, A. M. G. 2002. Birth and mortality of maned wolves, Chrysocyon brachyurus (Illiger, 1811) in captivity. Braz J Biol 62:25–32.

Maned Wolf Species Survival Plan®. 2011. 2011 Population analysis and breeding plan: Maned wolf (Chrysocyon brachyurus) Species Survival Plan® [Personal Email]. Accessed 15 Jul 2011.

Maned Wolf Species Survival Plan®. 2010. 2010 Population analysis and breeding plan: Maned wolf (Chrysocyon brachyurus) Species Survival Plan® [Personal Email]. Accessed 4 Oct 2010.

Maned Wolf Species Survival Plan®. 2007. Maned wolf husbandry manual. Silver Spring, Association of Zoos and Aquariums.

109

Mantzoros, C. S. 2000. Role of Leptin in reproduction. Ann NY Acad Sci 900:174-83.

Marker-Kraus, L., Grisham, J. 1993. Captive breeding of cheetahs in North American zoos (1987-1991). Zoo Biol 12:5-18.

McEwen, B. S. 2004. Allostasis and allostatic overload and relevance to the pathophysiology of psychiatric disorders. Ann NY Acad Sci 1032:1-7.

Mellen, J. D. 1991. Factors influencing reproductive success in small captive exotic felids (Felis spp.): A multiple regression analysis. Zoo Biol 10(2):95-110.

Melo, L. F. B., Sábato, M. A. L., Magni, E. M. V., Young, R. J., Coelho, C. M. 2007. Secret lives of maned wolves (Chysocyon brachyurus Illiger 1815): as revealed by GPS tracking collars. J Zool Lond 271:27-36.

Millipore. 2011. Multi-Species Leptin RIA Kit 250 Tubes (Cat. # XL-85K). St Charles: Millipore Technical Publications. 12 p.

Millspaugh, J. J., Washburn, B. E. 2004. Use of fecal glucocorticoid metabolites measures in conservation biology research: considerations for application and interpretation. Gen Comp Endocrinol 138:189-99.

Moberg, G. P. 2000. Biological response to stress: implications for animal welfare. In: Moberg, G. P., Mench, J. A. (editors). The biology of animal stress: basic principles and implications for animal welfare. New York: CABI Publishing. p 1-22.

Monfort, S. L., Mashburn, K. L., Brewer, B. A., Creel, S. R. 1998. Evaluating adrenal activity in African wild dogs (Lycaon pictus) by fecal corticosteroid analysis. J Zoo Wildl Med 29(2):129-33.

Montali, R. J., Kelly, K. 1989. Pathologic Survey and Review of Diseases of Captive Maned Wolves (Chrysocyon brachyurus) and Bush Dogs (Speothos venaticus). Verh ber Erkrg Zootiere 31:35-43.

Mormède, P., Andanson, S., Aupérin, B., Beerda, B., Guémené, D., Malmkvist, J., Manteca, X., Manteuffel, G., Prunet, P., van Reenen, C. G., Richard, S., Veissier, I. 2007. Exploration of the hypothalamic-pituitary-adrenal function as a tool to evaluate animal welfare. Physiol Behav 92:317-39.

Moschos, S., Chan, J. L., Mantzoros, C. S. 2002. Leptin and reproduction: a review. Amer Soc Reprod Med 77(3):433-44.

110

Möstl, E., Palme, R. 2002. Hormones as indicators of stress. Domest Anim Endocrinol 23: 67-74.

Motta-Junior, J. C., Martins, K. 2002. The frugivorous diet of the maned wolf, Chrysocyon brachyurus, in Brazil: ecology and conservation. In: Levey, D. J., Silva, W. R., Galetti, M. (editors). Seed Dispersal and Frugivory: Ecology, Evolution and Conservation. New York: CABI Publishing. p 291-304.

Motta-Junior, J. C., Talamoni, S. A., Lombardi, J. A., Simokomaki, K. 1996. Diet of the maned wolf, Chrysocyon brachyurus, in central Brazil. J Zool 240:277–84.

Muehlenbein, M. P., Campbell, B. C., Richards, R. J., Svec, F., Falkenstein, K. P., Murchison, M. A., Myers, L. 2003. Leptin, body composition, adrenal and gonadal hormones among captive male baboons. J Med Primatol 32(6):320-4.

Munro, C. J., Stabenfeldt, G. 1984. Development of a microtiter plate enzyme immunoassay for the determination of progesterone. J Endocrinol 101:41-9.

Munro, C. J., Stabenfeldt, G. H., Cragun, J. R., Addiego, L. A., Overstreet, J. W., Lasley, B. L. 1991. Relationship of serum estradiol and progesterone concentrations to the excretion profiles of their major urinary metabolites as measured by enzyme immunoassay and radioimmunoassay. Clin Chem 37(6):838-44.

Nagatani, S., Zeng, Y., Keisler, D. H., Foster, D. L., Jaffe, C. A. 2000. Leptin Regulates Pulsatile Luteinizing Hormone and Growth Hormone Secretion in Sheep. Endocrinol 141(11):3965-75.

Nepomnaschy, P. A., Welch, K., McConnell, D., Strassmann, B. I., England, B. G. 2004. Stress and female reproductive function: a study of daily variations in cortisol, gonadotrophins, and gonadal steroids in a rural Mayan population. Am J Hum Biol 16(5):523-32.

Newell-Fugate, A., Kennedy-Stoskopf, S., Brown, J. L., Levine, J. F., Swanson, W. F. 2007. Seminal and Endocrine Characteristics of Male Pallas’ Cats (Otocolobus manul) Maintained Under Artificial Lighting With Simulated Natural Photoperiods. Zoo Biol 26:187-99.

Nieminen, P., Mustonen, A. M., Asikainen, J., Hyvärinen, H. 2002. Seasonal weight regulation of the raccoon dog (Nyctereutes procyonoides): interactions between melatonin, leptin, ghrelin, and growth hormone. J Biol Rhythms 17:155-63.

Nishii, N., Yamasaki, M., Takasu, M., Honjoh, T., Shibata, H., Otsuka, Y., Takashima, S., Ohba, Y., Kitagawa, H. 2010. Plasma Leptin Concentration in Dogs with Diabetes Mellitus. J Vet Med Sci 72(6):809-11.

111

Oswal, A., Yeo, G. 2010. Signaling Mechanisms, and Role in the Pathogenesis of Obesity. Diabetes 18:221-9.

Padgett, D. A., Glaser, R. 2003. How stress influences the immune response. Trends Immunol 24:444–8.

Poessel, S. A., Biggins, D. E., Santymire, R. M., Livieri, T. M., Crooks, K. R., Angeloni, L. 2011. Environmental enrichment affects adrenocortical stress responses in the endangered black-footed ferret. Gen Comp Endocrinol 172:526-33.

Rabon, Jr., D. R., Waddell, W. 2010. Effects of Inbreeding on Reproductive Success, Performance, Litter Size, and Survival in Captive Red Wolves (Canis rufus). Zoo Biol 29:36-49.

Ram, M. R., Malathi, R. 2007. Possible correlation of leptin with body fat distribution and adiposity: Evaluation of serum leptin in South Indian population. Reprod Med Biol 6(2):117-25.

Rivier, C., Rivest, S. 1991. Effect of stress on the activity of the hypothalamic-pituitary- gonadal axis: peripheral and central mechanisms. Biol Reprod 45:523-32.

Rodden, M., Rodrigues , F., Bestelmeyer, S. 2008. Chrysocyon brachyurus. In: IUCN 2010. IUCN Red List of Threatened Species. Version 2010.4. . Downloaded on 05 November 2010.

Rodden, M., Rodrigues, F., Bestelmeyer, S. 2004. Maned wolf (Chrysocyon brachyurus). In: Sillero-Zubiri, C., Hoffman, M., Macdonald, D. E. editors. Canids: Foxes, Wolves, Jackals and Dogs. Status Survey and Conservation Action Plan. IUCN/SSC Canid Specialist Group: Gland, Switzerland and Cambridge.

Rodden, M. D., Sorenson, L. G., Sherr, A., Kleiman, D. G. 1996. Use of behavioral measures to assess reproductive status in maned wolves (Chrysocyon brachyurus). Zoo Biol 15(6):565-85.

Romero-Corral, A., Somers, V. K., Sierra-Johnson, J., Thomas, R. J., Collazo-Clavell, M. L., Korinek, J., Allison, T. G., Batsis, J. A., Sert-Kuniyoshi, F. H., Lopez-Jimenez, F. 2008. Accuracy of body mass index in diagnosing obesity in the adult general population. Int J Obes 32(6):959-66.

Roubos, E. W., Dahmen, M., Kozicz, T., Xu, L. 2012. Leptin and the hypothalamo- pituitary-adrenal stress axis. Gen Comp Endocrinol 177(1):28-36.

112

Sagawa, M. M., Nakadomo, F., Honjoh, T., Ishioka, K., Saito, M. 2002. Correlation between plasma leptin concentration and body fat content in dogs. Am J Vet Res 63(1):7-10.

Salvatori, V., Soler, L., Caceres, C., Perez, P., Gori, M., Fleita, A., Cuello, P. 2004. Estimating presence of maned wolf in northern Argentina from local knowledge: preliminary results. Memorias: Manejo de Fauna Silvestre en Amazonia y Latinoamérica p 288-91.

Sapolsky, R. M., Romero, L. M., Munck, A. U. 2000. How do glucocorticoids influence stress response? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocrinol Rev 21:55-89.

Schatz, S., Palme, R. 2001. Measurement of faecal cortisol metabolites in cats and dogs: a non-invasive method for evaluating adrenocortical function. Vet Res Commun 25:271-87.

Schmalz-Peixoto, K.E. 2003. Factors Affecting Breeding in Captive Carnivora [dissertation]. [Oxford (United Kingdom)]: University of Oxford. 212 p.

Seaward, B. L. 2008. Physiology of Stress. In: Managing Stress: Principles and Strategies for Health and Wellbeing with Art of Peace. 6th ed. Sudbury: Jones & Bartlett. 600 p.

Shah, N. R., Braverman, E. R. 2012. Measuring Adiposity in Patients: The Utility of Body Mass Index (BMI), Percent Body Fat, and Leptin. PLoS ONE 7(4):e33308. doi:10.1371/journal.pone.0033308

Sheriff, M. J., Dantzer, B., Delehanty, B., Palme, R., Boonstra, R. 2011. Measuring stress in wildlife: techniques for quantifying glucocorticoids. Oecologia 166(4):869-87.

Shibata, H., Akahane, R., Honjoh, T., Asano, M., Mominoki, K., Fujii, K., Suzuki, M., Ohtaishi, N., Ishioka, K., Ahmed, M., Soliman, M., Kimura, K., Saito, M. 2005. Seasonal changes in serum leptin of the feral raccoon (Procyon lotor) determined by canine-leptin specific ELISA. J Exp Zool 303A:527-33.

Smith, G. D., Jackson, L. M., Foster, D. L. 2002. Leptin Regulation of Reproductive Function and Fertility. Theriogenology 57:73-86.

Snyder, N. F. R., Derrickson, S. R., Beissinger, S. R., Wiley, J. W., Smith, T. B., Toone, W. D., Miller, B. 1996. Limitations of Captive Breeding in Endangered Species Recovery. Conserv Biol 10(2):338-48.

113

Soler, L., Carenton, J. M., Birochio, D., Orozco, M. M., Rosso, M., Carpinetto, M., Fortabat, S., Steiman, L., Diaz, M. L., de la Cruz Pino, M., Calvimontes, V., Kina, A. M., Jimenez, G., Prates, P., Solis, G. 2005. Problems and recommendations for the conservation of the maned wolf in Argentina: Results from the first workshop of Chrysocyon brachyurus in Argentina and surrounding countries: Conservation in situ and ex situ. 12-14 Sept 2002 Mburucuya, Corrientes, Argentina. Endangered Species Update 22(1):3-9.

Sollmann, R., Furtado, M. M., Jácomo, A. T. A., Tôrres, N. M., Silveira, L. 2010. Maned wolf survival rate in central Brazil. J Zool 282:207-13.

Songsasen, N., Rodden M. D. 2010. The role of the Species Survival Plan in Maned Wolf Chrysocyon brachyurus conservation. Int Zoo Yrbk 44:1-13.

Songsasen, N., Rodden, M., Brown, J. L., Wildt, D. E. 2006. Patterns of fecal gonadal hormone metabolites in the maned wolf (Chrysocyon brachyurus). Theriogenology 66:1743-50.

Songsasen, N., Widlt, D. E. 2007. Oocyte biology and challenges in developing in vitro maturation systems in the domestic dog. Anim Reprod Sci 98:2-22.

Spady, T. J., Harlow, H. J., Butterstein, G., Durrant, B. 2009. Leptin as a surrogate indicator of body fat in the American black bear. Ursus 20(2):120-30.

Stoskopf, M. K. 1983. The physiological effects of psychological stress. Zoo Biol 2:179- 90.

Tauson, A. H., Forsberg, M. 2002. Body-weight changes are clearly reflected in plasma concentrations of leptin in female mink (Mustela vison). Br J Nutr 87:101-5.

Terio, K., Marker, L., Munson, L. 2004. Evidence for chronic stress in captive but not free-ranging cheetahs (Acinonyx jubatus) based on adrenal morphology and function. J Wildl Dis 40:259-66.

Thompson, K. V., Baker, A. J., Baker, A. M. 2010. Chapter 29: Parental Care and Behavioral Development in Captive Mammals. In: Kleiman, D. G., Thompson, K. V., Baer, C. K. (editors). Wild Mammals in Captivity: Principles and Techniques for Zoo Management. 2nd ed. Chicago: Univ Chicago Pr. p 367-85.

Trolle, M., Noss, A. J., Lima, E. S., Dalponte, J. 2007. Camera-trap studies of maned wolf density in the Cerrado and the Pantanal of Brazil. Biodivers Conserv 16:1197– 1204.

114

Van Harmelan, V., Reynisdottir, S., Eriksson, P., Thörne, A., Hoffstedt, J., Lönnqvist, F., Arner, P. 1998. Leptin secretion from subcutaneous and visceral adipose tissue in women. Diabetes 47(6):913-7.

Velloso, A., Wasser, S. K., Monfort, S. L., Dietz, J. M. 1998. Longitudinal fecal steroid excretion in maned wolves (Chrysocyon brachyurus). Gen Comp Endocrinol 112:96–107.

Walker, S. L., Waddell, W. T., Goodrowe, K. L. 2002. Reproductive endocrine patterns in captive female and male red wolves (Canis rufus) assessed by fecal and serum analysis. J Reprod Fertil Suppl 44:37-44.

Wasser, S. K., Monfort, S. L., Wildt, D. E. 1991. Rapid extraction of faecal steroids for measuring reproductive cyclicity and early pregnancy in free-ranging yellow baboons (Papio cynocephalus cynocephalus). J Reprod Fertil 92:415-23.

Wasser, S. K., Velloso, A. D. L., Rodden, M. D. 1995. Using fecal steroids to evaluate reproductive function in female maned wolves. J Wildl Manage 59(4):889-94.

Wielebnowski, N. 1998. Contributions of behavioral studies to captive management and breeding of rare and endangered mammals. In: Caro, T. (editor). Behavioral Ecology and Conservation Biology. New York: Oxford Univ Pr. p 130–162.

Wielebnowski, N., Fletchall, N., Carlstead, K., Busso, J., Brown, J. 2002. Noninvasive assessment of adrenal activity associated with husbandry and behavioral factors in the North American clouded leopard population. Zoo Biol 21:77-98.

Wikelski, M., Cooke, S. J. 2006. Conservation physiology. Trends Ecol Evol 21(1):38- 46.

Wildt, D. E., Brown, J. L., Howard, J. G. 2001. Role of reproductive sciences in carnivore conservation. In: Gittleman, J. L., Funk, S. M., Macdonald, D. W., Wayne, R. K. (editors). Carnivore Conservation. New York: Cambridge Univ Pr. p 359-71.

Wildt, D. E., Panko, W. B., Chakraborty, P. K., Seager, S. W. J. 1979. Relationship of serum estrone, estradiol-17 and progesterone to LH, sexual behavior and time of ovulation in the bitch. Biol Reprod 20:648-58.

Williams, G. L., Amstalden, M., Garcia, M. R., Stanko, R. L., Nizielski, S. E., Morrison, C. D., Keisler, D. H. 2002. Leptin and its role in the central regulation of reproduction in cattle. Domest Anim Endocrinol 23:339-49.

115

Young, K. M., Brown, J. L., Goodrowe, K. L. 2001. Characterization of reproductive cycles and adrenal activity in the black-footed ferret (Mustela nigripes) by fecal hormone analysis. Zoo Biol 20:517-36.

Young, K. M., Walker, S. L., Lanthier, C., Waddell, W. T., Monfort, S. L., Brown, J. L. 2004. Noninvasive monitoring of adrenocortical activity in carnivores by fecal glucocorticoid analyses. Gen Comp Endocrinol 137:148-65.

116

CURRICULUM VITAE

Lauren E. Reiter graduated from Washington-Lee High School, Arlington, Virginia, in 2002. She received her Bachelor of Arts from Miami University, Oxford, Ohio, in 2006.

117