Citation
Divergent Stress Coping Styles and Singing Behavior in the Short-Tailed Singing Mouse (Scotinomys teguina)

Material Information

Title:
Divergent Stress Coping Styles and Singing Behavior in the Short-Tailed Singing Mouse (Scotinomys teguina)
Creator:
Crino, Andrea
Place of Publication:
[Gainesville, Fla.]
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (58 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Zoology
Committee Chair:
Phelps, Steven M.
Committee Members:
Larkin, Iske
Reed, David L.
Brockmann, H. Jane Jane
Graduation Date:
8/9/2008

Subjects

Subjects / Keywords:
Animals ( jstor )
Corticosterone ( jstor )
Glucocorticoids ( jstor )
Hormones ( jstor )
Plasmas ( jstor )
Radioimmunoassay ( jstor )
Reactivity ( jstor )
Singing ( jstor )
Stress response ( jstor )
Vegetation ( jstor )
Zoology -- Dissertations, Academic -- UF
Genre:
Electronic Thesis or Dissertation
born-digital ( sobekcm )
Zoology thesis, M.S.

Notes

Abstract:
In response to stressors, vertebrates experience a suite of behavioral and physiological changes that elicit context-appropriate responses in order to maintain homeostasis. The acuity of the stress response and its associated behaviors vary greatly between individuals in many taxa and form suites of discrete characters described as stress coping styles. Stress coping styles have commonly been described in a proactive/reactive dichotomy. Proactive animal are behaviorally less wary and mount less acute physiological stress responses compared to reactive animals. It has been proposed that divergent stress coping styles give animals a survival or reproductive advantage in certain environmental contexts. I examined the prevalence of divergent stress coping styles in male short-tailed singing mice (Scotinomys teguina) from a national park used for agriculture (Park) and a relatively undisturbed nature reserve (Reserve) in 2006 and 2007. Additionally, I examine the relationship of divergent stress coping styles to singing behavior, a trait presumed to be important for reproductive success in the short-tailed singing mouse. In 2006, males from the Park displayed behavioral and physiological traits associated with the reactive stress coping style while males from the Reserve displayed the proactive stress coping style. Additionally, Reserve males were more likely to vocalize than males from the Park. In 2007 there were no differences in behavioral stress responses between sites, though the trends in physiological stress responses and singing behavior remained consistent. The proactive/reactive dichotomy does not appear to consistently explain behavioral and physiological variation in male short-tailed singing mice between sites. However, mice that expressed proactive stress coping behaviors were more likely to vocalize than reactive animals. This indicates that propensity to sing can be mapped on to the proactive/reactive stress coping style dichotomy. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2008.
Local:
Adviser: Phelps, Steven M.
Statement of Responsibility:
by Andrea Crino.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Crino, Andrea. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Classification:
LD1780 2008 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

1 DIVERGENT STRESS COPING STYLES AND SINGING BEHAVIOR IN THE SHORT TAILED SINGING MOUSE ( Scotinomys teguina) By ANDREA CRINO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Andrea Crino

PAGE 3

3 T o Drs. Paulette Bierzychudek, Robert Baylor, a nd Dan Blumstein in recognition of their contributions to my early enthusiasm for research and animal behavior

PAGE 4

4 ACKNOWLEDGMENTS I thank my chair, Dr. Steve Phelps, and supervisory committee for their mentoring and guidance that made this project possible. Additionally, I thank Drs. Polly Campbell and Alex O phir for their amazing professional and personal support. I also thank Bret Pasch, Jorge Pino, and Molly Philips of the Phelps lab for field assistance in 2006 and 2007. I thank Karen Patterson and Vitrel Sherif for their tireless pat ience and assistance with tedious logistics and paperwork. Finally, I thank my wonderful friends and fellow graduate students whose intellectual contributions helped to inspire this work and, more importantly, who helped make this a fun and enjoyable expe rience

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 8 ABSTRACT ................................ ................................ ................................ ................................ ..... 9 CHAPTER 1 PHYSIOLOGICAL AND BEHAVIORAL VARIATION IN STRESS RESPONSES IN VERTEBRATES ................................ ................................ ................................ .................... 11 2 PHYSIOLOGICAL AND MORPHOLOGICAL VARIATION BETWEEN TWO POPULATIONS OF SHORT TAILED SINGING MICE ( Scotinomys Teguina ) ................. 16 Introduction ................................ ................................ ................................ ............................. 16 Methods ................................ ................................ ................................ ................................ .. 17 Animal Trapping and Housing ................................ ................................ ........................ 17 Morphological and Condition Measurements ................................ ................................ 17 Vegetation Survey ................................ ................................ ................................ ........... 18 Results ................................ ................................ ................................ ................................ ..... 18 Morphological Summary ................................ ................................ ................................ 18 Parasite Prevalence ................................ ................................ ................................ .......... 19 Hematocrit ................................ ................................ ................................ ....................... 19 Vegetation Survey ................................ ................................ ................................ ........... 20 Discussion ................................ ................................ ................................ ............................... 20 3 BIOCHEMICAL AND PHYSIOLOGICAL VALIDATION OF CORTICOSTEROID RADIOIMMUNOASSAY FOR PLASMA AND FECAL SAMPLES IN THE SHORT TAILED SINGI NG MOUSE ( Scotinomys Teguina ) ................................ ............................. 26 Introduction ................................ ................................ ................................ ............................. 26 Methods ................................ ................................ ................................ ................................ .. 27 Phys iological Assay Validation of Plasma Corticosterone (CORT) ............................... 27 Biochemical Validation ................................ ................................ ................................ ... 28 Radioimmunoassay (RIA) ................................ ................................ ............................... 28 Results ................................ ................................ ................................ ................................ ..... 28 Physiological Assay Validation of Plasma CORT ................................ .......................... 28 Biochemical Valida tion ................................ ................................ ................................ ... 28 Conclusion ................................ ................................ ................................ .............................. 29

PAGE 6

6 4 DIVERGENT STRESS COPING STYLES IN MALE SHORT TAILED SINGING MICE IN TWO DISTINCT HABITATS ................................ ................................ ............... 32 Introduction ................................ ................................ ................................ ............................. 32 Methods ................................ ................................ ................................ ................................ .. 34 Animal Trapping and Site Descriptions ................................ ................................ .......... 34 Behavioral Stress Response: Open Field Assay ................................ .............................. 34 Hypothalamic Pituitary Adrenal (HPA) Activity and Reactivity: Fecal and Serum Corticos terone (CORT) ................................ ................................ ................................ 34 Radioimmunoassays (RIA) ................................ ................................ ............................. 35 Adrenocorticotropin Releasing Hormone (ACTH) Supplementation and Singing ......... 35 Results ................................ ................................ ................................ ................................ ..... 36 Behavioral Stress Assay: Open Field Behavior ................................ .............................. 36 HPA Activity and Reactivity: Fecal and Serum CORT ................................ .................. 37 Singing Behavior and Stress Responses ................................ ................................ .......... 38 Effect of ACTH on Singing Behavior ................................ ................................ ............. 39 Discussion ................................ ................................ ................................ ............................... 39 LIST OF REFERENCES ................................ ................................ ................................ ............... 54 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ......... 58

PAGE 7

7 LIST OF TABLES Table page 2 1 Mean morphological traits of adult males from the Reserve and the Park in 2006 and ferences between sites in 2006 and 2007. ................................ ................................ ................................ ............................ 25 4 1 Description of behaviors and activities measured in the open field behavior assay. ........ 53

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Percent of parasitzed and nonparasitized males from the Reserve and the Park in 2006 and 2007. ................................ ................................ ................................ ................... 22 2 2 Percent hematocrit (red blood cells/whole blood) of males from the Reserve (n = 41) and Park (n = 30). ................................ ................................ ................................ .............. 23 2 3 Mean SEM maximum vegetation height at the Park and Reserve (n = 8 transects) ...... 24 3 1 Mean serum CORT titer SEM following zero, 3, 5, 10, and 20 minutes of restraint stress (n = 24, 8, 9, 22 respectively). ................................ ................................ ................. 30 3 2 Pa rallelism of log transformed fecal CORT serial dilutions and logit transformed percent bound with transformed standard curves.. ................................ ............................ 31 4 1 Mean SEM measures of open field behavior in 2006 with n = 29 for Reserve and n = 14 for Park ................................ ................................ ................................ ...................... 44 4 2 Mean SEM measures of open field behavior in 2007 with n = 17 for Reserve and n = 22 for Park. ................................ ................................ ................................ .................... 45 4 3 Mean fecal CORT S EM for 2006 and 2007 combined ................................ .................. 46 4 4 Mean fecal CORT (ng/g) SEM on the second and seventh day in captivity. ................. 47 4 5 Mean SEM CORT secretion in basal and stress (20 minute restraint) conditions. ....... 48 4 6 Percentage of males that sang at least once in a two hour period (callers) and males that did not sing (noncallers) ................................ ................................ .............................. 49 4 7 Singing rate in a two hour period in relation to anogenital distance. ................................ 50 4 8 Mea n SEM measures of open field behavior in 2006 and 2007 with n = 20 for Callers and n = 62 for Noncallers ................................ ................................ ...................... 51 4 9 ACTH treatment did not significantly change singing behavior compared to contr ol treatments (Wilcoxon signed rank, p = 0.21, n = 7 per treatment). ................................ ... 52

PAGE 9

9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science DIVERGENT STRESS COPING STYLES AND SINGING BEHAVIOR IN THE SHORT TAILED SINGING MOUSE ( Scotinomys teguina). By Andrea Crino August 2008 Chair: Steve Phelps Major: Zoology In response to stressors, vertebrates experience a s uite of behavioral and physiological changes that elicit context appropriate responses in order to maintain homeostasis. The acuity of the stress response and its associated behaviors vary greatly between individuals in many taxa and form suites of discre te characters described as stress coping styles. Stress coping styles have commonly been described in a proactive/reactive dichotomy. Proactive animal are behaviorally less wary and mount less acute physiological stress responses compared to reactive ani mals. It has been proposed that divergent stress coping styles give animals a survival or reproductive advantage in certain environmental contexts. I examined the prevalence of divergent stress coping styles in male short tailed singing mice ( Scotinomys teguina ) from a national park used for agriculture (Park) and a relatively undisturbed nature reserve (Reserve) in 2006 and 2007. Additionally, I examine the relationship of divergent stress coping styles to singing behavior, a trait presumed to be import ant for reproductive success in the short tailed singing mouse. In 2006, males from the Park displayed behavioral and physiological traits associated with the reactive stress coping style while males from the Reserve displayed the proactive stress coping style. Additionally, Reserve males were more likely to vocalize than males from the Park. In 2007 there were no differences in

PAGE 10

10 behavioral stress responses between sites, though the trends in physiological stress responses and singing behavior remained co nsistent. The proactive/reactive dichotomy does not appear to consistently explain behavioral and physiological variation in male short tailed singing mice between sites. However, mice that expressed proactive stress coping behaviors were more likely to vocalize than reactive animals. This indicates that propensity to sing can be mapped on to the proactive/reactive stress coping style dichotomy.

PAGE 11

11 CHAPTER 1 PHYSIOLOGICAL AND BE HAVIORAL VARIATION I N STRESS RESPONSES I N VERTEBRATES In response to external s tressors, vertebrates mount a suite of behavioral and physiological stress responses to maintain or re establish homeostasis (Sapolosky et al. 2000). T he acuity of the stress response and its behavioral manifestations has been shown to vary greatly betwee n individuals in many taxa including rodents (De Boer et al. 2003), pigs (Ruis et al. 2001), birds (Marchetti and Drent 2000), fish (Dingemanse et al. 2007; Ward et al. 2004), and humans (Tyrka et al. 2006) These individual differences form suites of cor related behaviors and physiological responses known as s tress coping styles that are causally linked, predictable, and characteristic of a group or population of animals (Koolhaas et al. 1999). Coping styles have been characterized along a shy bold conti nuum and as a proactive/reactive dichotomy (Koolhaas et al. 1999). Boldness has broadly been described as the willingness to engage in risky behaviors in exchange for potential increases in reproductive or foraging success (Ward et al. 2004). Conversely, shy animals are defined as risk averse and engage in behaviors with lower gains, but may experience higher survival in some contexts (Ward et al. 2004). The proactive/reactive dichotomy describes similar behavioral types, but makes context specific behav ioral and physiological predictions. Behaviorally, proactive animals are highly aggressive toward conspecifics, readily develop rigid learned routines, and actively avoid or manipulate stressful stimuli (Koolhaas et al. 1999, Koolhaas et al. 2007). The r eactive stress coping style is characterized by low levels of conspecific aggression, flexible cue dependent learning routines, and passive behaviors such as immobility in response to stressful stimuli (Koolhaas et al. 1999, Koolhaas et al. 2007). In add ition to these stereotyped behaviors, proactive/reactive stress coping styles are characterized by distinct assemblages of physiological and neuroendocrine stress responses. In

PAGE 12

12 general, proactive animals show a high sympathetic response to stressors inclu ding an increase in blood pressure and heart rate and the production of catecholamines including both noradrenaline and adrenaline (verli et al. 2007). Conversely, reactive animals show higher parasympathetic reactivity to stressors than proactive animal s and have lower hypothalamic pituitary adrenal (HPA) axis activity and reactivity (verli et al. 2007). Therefore, proactive animals typically have lower levels of basal glucocorticoids (the main hormones involved in the stress response and the ultimate product of HPA axis activation) and lower stress induced glucocorticoid levels (verli et al. 2007). Summed together, these behavioral and physiological characteristics describe animals that differentially respond to environmental stressors and may exper ience differential survival and fitness under fluctuating environmental conditions such as population density, social stability, predator density, and resource availability (Korte et al. 2005). For example, in captive great tits ( Parus major ), reactive li foraging strategy when the location of food is experimentally manipulated. Conversely, been removed (Marchetti and Drent 2000). In wild populations of great tits, reactive individuals e an advantage during periods of resource stability because their social dominance allows them to dominate food sources (Marchetti and Drent 2000) Stress coping styles in rodents may provide similar survival advantages within an ecological context. In lab studies, proactive rodents aggressively defend their territory against conspecific intruders (de Boer et al. 2003) which could translate into higher fitness in densely populated environments (Korte et al. 2005).

PAGE 13

13 While proactive and reactive animals emplo y different strategies for coping with stressors, the ultimate result (minimizing further exposure to a stressor and, thus, restoring the animal to homeostasis) can be the same between these two phenotypes. This is exemplified in the shock prod burying te st used on laboratory rats (Koolhaas et al. 1999). In this assay, an animal is presented with an electrified prod in its home cage. When exploring the novel object, animals receive a mild but aversive shock. Rats that express the proactive phenotype wil l actively bury the electric prod with cage bedding and, thus, avoid further shocks. Conversely, reactive rats will hide in the corner of a cage and show immobility behavior. Though these two stress coping styles employ different behavioral responses, the ultimate outcome (avoiding further electrical shock) is identical (Koolhaas et al. 1999). Interestingly, these stress coping styles are highly context dependent. If a proactive animal is tested in the electric prod burying assay in a cage with no bed ding, it will employ the same behavioral response as a reactive animal (Koolhaas et al. 1999). In addition to the context specific advantages and expression of divergent coping styles, both phenotypes are also associated with context specific physiologica l and behavioral costs. Long term exposure to glucocorticoids has been associated with reduced immunocompetence, growth, and longevity (Cavigelli and McClintock 2003) providing a potential morbidity cost to the reactive stress phenotype. Likewise, proact ive animals may face mortality costs associated with their low wariness of novelty and propensity to engage in risky behaviors that increase their exposure to predators (Magnhagen and Staffan 2005). Different environmental contexts may generate tradeoffs in short and long term survival that influence the frequency and fitness of alternative stress phenotypes in certain populations or environments (Sih et al. 2004).

PAGE 14

14 In addition to tradeoffs between short and long term survival, divergent stress coping sty les may also generate tradeoffs between survival and reproduction. For example, in food limited contexts, bold, highly aggressive female fishing spiders ( Dolomedes fimbriatus ) will cannibalize all courting male spiders and thus be unable to mate (Arnqvist and Henriksson 1997). Divergent stress coping styles may also influence reproduction by creating tradeoffs between sexually selected traits, such as display vocalizations, and predator or conspecific avoidance. In response to stressors, auditory signals have been shown to be modulated or withheld in Gulf toadfish (Remage Healey et al. 2006), tngara frogs (Ryan 1985), and katydids (Faure and Hoy 2000). Glucocorticoids have specifically been shown to modulate the prevalence of alarm vocalizations (Boinsk i et al. 1999, Blumstein et al. 2006, Swiergiel et al. 2007) and display vocalizations (Remage Healey et al 2006, Leary et al. 2006). Although the propensity to vocalize has not been differentially ascribed to divergent stress coping styles, it is likely that proactive animals that are less behaviorally wary and have lower levels of glucocortioids would engage in more display vocalization compared to reactive animals. If stress coping style influences display vocalizations, this characteristic could be a useful behavioral proxy to describe the ecological and evolutionary relevance of stress phenotypes in wild animals. While stress coping styles have been well described in the laboratory, far fewer studies have examined the occurrence of stress phenotypes in wild animals within natural contexts. Extending studies from the laboratory housed animals to wild animals will provide a more complete understanding of potential tradeoffs generated by stress coping styles. In addition, studying these coping styles o utside the laboratory will allow us to investigate the influence of

PAGE 15

15 ecological factors on stress coping style expression, the putative context specific reproductive advantages of divergent coping styles, and their long term evolutionary consequences. Her e, I document the prevalence of proactive/reactive stress coping styles and display vocalizations in two populations of short tailed singing mice ( Scotinomys teguina ) Muroidea: Cricetidae) in distinctly different habitats. I first describe the sites at wh ich the work was executed and document physiological and morphological differences in animals between sites. Next, I demonstrate the validity of fecal and serum corticosterone (CORT) assays used to measure the physiological stress response of singing mice Finally, I compare spontaneous and CORT induced singing behavior to behavioral stress responses (open field behavior), basal CORT titers (HPA activity), and stress induced CORT titers (HPA reactivity).

PAGE 16

16 CHAPTER 2 PHYSIOLOGICAL AND MO RPHOLOGICAL VARIATI ON BETWEEN TWO POPULATIONS OF SHORT TAILED SINGING MICE ( SCOTINOMYS TEGUINA ) Introduction Singing mice (Muroidea: Cricetidae) are small diurnal rodents that inhabit tropical cloud forests of Central America (Hooper and Carleton 1972). S inging mice derive their name from their ability to produce long frequency modulated vocalizations. While many rodent species produce ultrasonic vocalizations, singing mice vocalize between 10,000 and 40,000 Hz (partially within human auditory range) making them a unique a nd tractable species with which to study rodent vocalizations (Hooper and Carleton 1972, Phelps et al. unpublished data) Although the purpose of singing is not fully known evidence suggests the primary vocalization function s in mate attracti on and male male interactions (unpublished data) There are two species of singing mice in the genus Scotinomys: the short tailed singing mouse ( Scotinomys teguina) and the long tailed singing mouse ( Scotinomys xerampelinus ) Short tailed singing mice range from s outhern Mexico to western Panama and are fou nd at elevations between 1000 30 00 meters (Hooper and Carelton 1976 ). The long tailed singing mouse is found at elevations between 2150 3300 meters and ranging through Costa Rica and Panama (Hooper and Carleton 1976). Broadly, the two species are considered contiguously allopatric with small local areas of sympatry (Hooper and Carleton 1976). Despite being abundant and widely distributed, relatively little is known about the behavior and life histories of these charismatic species. I surveyed short tailed singing mice at two sites located near Boquete, Panama in 2006 and 2007. One site, the located in Jaramillo Arriba, 15 km north of Boqu ete, Panama. This reserve consists of approximately 100 hectares of secondary growth tropical cloud forest bordered on all sides by

PAGE 17

17 agricultural lands. Because this reserve is privately owned, there is very little anthropogenic disturbance within its bou ndaries. The second site, Volcn Baru National Park (Park) is a publicly owned and operated national park located 15 km west of Boquete, Panama with an altitudinal range between 1,840 3478 meters While dominated by secondary growth cloud forest, this p ark is heavily fragmented due to extensive agriculture and grazing. These sites are located approximately 12 linear kilometers apart, separated by lowland habitat (975 m), and divided by the Chiriqui River. Preliminary genetic analyses recovered a sing le mitochondrial haplotype for animals from both sites suggesting that they constitute one genetically indistinct population (unpublished data). Her e I provide morphological data for male short tailed singing mice from both sites. I also compared condit ion of animals between sites by quantifying hematocrit (t he percentage of red blood cells to whole blood ) and parasite prevalence. Finally, I quantify average vegetation height to examine habitat differences between the two sites Methods Animal T rapping a nd Housing I trapped mice from May to July 2006 and May to June 2007 using Sherman live traps baited with peanut butter and oats. I clipped no more than two toes with no more than one toe taken from each foot in a unique pattern to identify individual an imals. Mice were housed in the laboratory in wire mesh cages and fed an ad libitum diet of dry kitten food and sunflower seeds. Morphological and Condition Measurements After capture, I assessed repro ductive condition, parasite prevalence and measured morphological traits including head width (width of head directly rostral to ears), hind foot ( from back of heel to tip of middle toe excluding nail), half hind foot ( from back of heel to hallux ), tail

PAGE 18

18 length (posterior margin of anus to tail tip), ano gen ital distance (anus to genital projection), mass, and residual mass ( mass/hind foot length). Singing mice carry several ectoparasites including unidentified species of lice, fleas, and ear mites (personal observation). I categorized parasite prevalence based on the presence or absence of any visible ectoparasites. To quantify hematocrit, I collected 100 150 using microhematcrit tubes. I centrifuged blood at 3000 rpm for four minutes and measured hematocrit as the percent of red blood cells to whole blood. Vegetation Survey To examine vegetation differences betwee n sites, I quantified mean maximum vegetation height along eight 20 meter transects at both sites. Excluding trees, I measured maximum vegetation height as the highest vegetation stem at each meter mark for every meter on each 20 meter transect. I chose the transaction location and compass direction randomly within areas where I captured mice. Results Morphological S ummary Interestingly, despite their genetic similarity, I n oted overall differences in male body size between sites in 2006 and 2007 (Tabl e 2 1). There were significant site by year interactions for the measurements of hind foot length half hind foot length head width, ano genital distance, absolute mass, and residual mass (p < 0.05 for all) For this reason, morphological traits were an alyzed separately for 2006 and 2007. In 2006, male mice from the Reserve were significantly smaller than males from the Park in measures of tail length, hind foot length head width, absolute mass, and body condition (Table 2 1 ). Conversely, in 2006 Rese rve males had significantly larger ano genital distances than Park ma les (Table 2 1, p < 0.01 F = 12.40). In

PAGE 19

19 2007, males from the Reserve were significantly smaller than Park males in tail length only (Tabl e 2 1 p = 0.01, F = 6.74). There were no signi ficant differences between Reserve and Park males in all other morphological measurements includ ing testes mass in 2007 (Table 2 1 ). Within the Park, males differed between years. In 2007 males from the Park were significantly smaller (p < 0.01, F = 22. 43), half hind foot (p < 0.01, F = 17.40), head width (p < 0.01, F = 19.93), and weight (p = 0.04, F = 4.48). However, in 2007 Park males had significantly larger ano genital distances than Park males in 2006 (p = 0.04, F = 4.82). Males from the Reserve w ere not significantly different in morphological measurements between 2006 and 2007 (0.84 > p > 0.18) for all characters except for having significantly smaller ano genital distances in 2007 (p = 0.01, F = 6.78). Parasite Prevalence There were no signifi cant differences in parasite load between males from the Reserve and the Park in 2006 and 2007 2 = 1.38). There were more parasitized males from the 2 = 3.77 Fig ure 1 1 ) Hematocrit There was no significant site by year interaction in percent hematocrit between Reserve and Park males for 2006 or 2007 (p = 0.26, F = 1.27) Analyzing years together and combining adult and sub adult males, I found that males from t he Park had significantly higher percent hematocrit than males from the Reserve (p = 0.04, F = 4.25, n = 30 and 41 respectively). Similarly, comparing only adult males from 2006 and 2007, males from the Park had a non significant trend for higher mean per cent hematocrit than males from the Reserve (p = 0.06, F = 3.55, n = 27 and 34 respectively).

PAGE 20

20 Vegetation Survey I found no significant difference in mean maximum vegetation height between the two sites (Figure 1 2; p = 0.21, F = 1.61, n = 8 for each site) I also found no significant difference in vegetation heterogeneity (measured by comparing the standard deviation of mean vegetation height) between the two sites (p = 0.19, F = 1.89). Although I did not further assess s ite differences in vegetation or habitat, based on the distinct and obvious visual differences, it is probable that additional measurements would reveal significant site differences. Discussion Here I have provided demographic and morphological data for male short tailed singing mice f rom two sites in Boquete, Panama. Interestingly, in 2006 I found that males from the Park a re larger than males from the Reserve in several morphological traits, including tail length, hind foot, absolute mass, and head width, but had smaller ano gential distances. In 2007, Park males were of comparabl e size to Reserve males in all morphological measures with the exception of having longer tails. In addition to size differences, more males from the Reserve had ectoparasites than males from the Park in b oth 2006 and 2007. Despite obvious visual differences in habitat, I found no difference in mean vegetation height between the Park and the Reserve. It seems likely that these habitat differences could be quantitatively described in a more biologically m eaningful way by measuring habitat brokenness or floral diversity as examples. Habitat differences likely influence other ecological factors such as predator and conspecific density and resource availability. These ecological factors could be easily measu red and, if labile, could contribute to the observed variation in morphological data between sites in 2006 but not 2007. Variation in ecological factors could also explain the anomalous trend in 2006 for Park males to be morphologically larger than Rese rve males in all measurements except ano genital

PAGE 21

21 distance in which they were significantly smaller. The overall larger body size may reflect a higher density of insect prey localized around the agriculture areas in the Park. Ano gential distance is often used as a proxy for sexual maturity (Ashby et al. 1997, Pocock et al. 2002). It is possible that, despite putatively larger food availability, the males I trapped from the Park were less sexually mature than males from the Reserve. The majority of the a nimals surveyed from the Park were captured in fringe habitat bordering agricultural areas. If these fringe areas in the Park are marginal in respect to cover for predator avoidance or breeding habitat, it is possible that the mice I surveyed are younger habitat. However, if this were true, I would also expect to observe a similar trend in body size differences between sites in 2007. Instead, I observed no difference in body size between sites in all measurements except tail length. Alternatively, the anomaly between size trends for body size and ano genital distance between sites in 2006 could be attributed to endocrine disrupting agents such as pesticides. Pesticides and other chemicals ar e used extensively in the agricultural areas at the Park (personal observation). Some of these pesticides, such as Atrazine, have been demonstrated to retard sexual development in rats (Stoker et al. 2000) and male Japanese quail (Wilhelms et al. 2005). If pesticide use is influencing sexual development in the mice at the Park, the inconsistent application of these chemicals (dependent on agriculture schedules) could explain the opposite trends in body size and ano gential distance between males from the Park and Reserve in 2006 and the absence of these trends in 2007.

PAGE 22

22 Figure 2 1. Percent of parasitzed and nonparasitized males from the Reserve and the Park in 2006 and 2007.

PAGE 23

23 Figure 2 2. Percent hematocrit (red blood cells/whole blood) of males f rom the Reserve (n = 41) and Park (n = 30).

PAGE 24

24 Figure 2 3. Mean SEM maximum vegetation height at the Park and Reserve (n = 8 transects)

PAGE 25

25 Table 2 1. Mean morphological traits of adult males from the Reserve and the Park in 2006 and and 2007.

PAGE 26

26 CHAPTER 3 BIOCHEMICAL AND PHYS IOLOGICAL VALIDATION OF CORTICOSTEROID RADIOIMMUNOASSAY FOR PLASMA AND FECAL SAM PLES IN THE SHORT TAILED SINGING MOUSE ( Scotinomys T eguina ) Introduction In response to an environmental stressor, animals experience a suite of behavioral and physiological responses that promote a context appropriate stress response. One physiological component of the stress response involves activation of the hypothalamic pituitary adrenal (HPA) axis. The HPA axis is initiated when a stressor stimulates the release of corticotrophin releasing hormone (CRH) from the hypothalamus. CRH stimulates the release of adrenocorticotropin hormone (ACTH) from the anterior pituitary which causes the release of glucocorticoids (the main steroid hormones involved in the stress response) from the adrenals ( Sapolosky et al. 2000 ). Glucocorticoids regulate many facets of the physiological stress response by increasing a vailable glucose, altering cardiovascular activity, and inhibiting immediate non essential physiological processes such as reproduction, growth, and immune system function (review in Nelson 2005). Individuals vary greatly in the magnitude of HPA response t o stressors (De Boer et al. 2003, Ruis et al. 2001, Marchetti and Drent 2000, Dingemanse et al. 2007; Ward et al. 2004). glucorticoid titers or glucorticoid metabolites in b lood and feces. Measuring glucocorticoids in the blood is thought to give an accurate measure of stress response because it evaluates the concentration of glucocorticoids directly available to tissues (Harper and Austad 2000). However, blood hormone assa ys provide hormone concentration at a single point in time and, therefore, do not capture long term hormone profiles (Harper and Austad 2000). Additionally, glucocorticoids increase in as little as two minutes in response to handling stress and, thus, the

PAGE 27

27 procedures involved in blood collection are sufficient to illicit a stress response and increase glucocorticoid titers (Romero and Reed 2005). The use of fecal hormone metabolite measurements is becoming an increasingly common way to assess hormone prof iles. Fecal hormone metabolite assays integrate hormonal fluctuations over the period of time that the feces are formed and stored, providing a broad and perhaps more informative hormonal profile than single point measurements provided by blood samples (H arper and Austad 2000, Touma et al. 2004). Additionally, because of the slow metabolism of glucocorticoids, acute stress associated with handling will have little to no influence on metabolite concentration (Harper and Austad 2000). Blood glucocorticoid s titers are traditionally measured using radioimmunoassay (RIAs) or enzyme immunoassay (EIA). There are pronounced interspecific differences in the metabolic rate and secretion of fecal glucocorticoids that may preclude the use of immunoassays designed f or laboratory mice in exotic animals (Palme 2005, Touma et al. 2005). For this reason, it is necessary to biochemically and physiologically validate immunoassays to ensure that they can detect biologically meaningful hormone measurements. Here, I validat e the corticosterone (CORT) 125I radioimmunoassay kit (MP Biomedicals, Solon, OH; catalog no. 071200103) for plasma CORT and fecal CORT metabolites in the short tailed singing mouse (Scotinomys teguina) using a combination of validation methods described b y Good et al. (2003) and Touma et al. (2004). Methods Physiological Assay Validation of Plasma Corticosterone ( CORT ) To assess the short term effects of handling stress on singing mice, I exposed animals to five treatments of restraint stress (0, 3, 5, 10, and 20 minutes). I restrained animals by hand for the designated restraint time and then immediately collected 100

PAGE 28

28 retro orbital sinus. For control animals receiving zero minutes of restraint stress, I collected blood in less than two minutes to avoid initiating a stress response. I centrifuged blood at 3000 rpm for four minutes and stored the plasma at Biochemical Validation To ascertain assay parallelism, I compared the slope of the standard curve to the slope of a curve generated using serially diluted fecal samples (1:7, 1:15, 1:31, 1:63). I logit transformed percent bound and log transformed CORT titers (ng/ml). I assessed assay precision by calculating inter and intra assay coefficients of variation of the perc ent bound of the internal controls. I used variation between duplicate samples within an assay to calculate intra assay precision and variation between average samples between assays to calculate inter assay precision Radioimmunoassay (RIA) I diluted fec al samples to 1:63 and plasma samples to 1:127 in assay diluent prior to radioimmunoassay. I ran each sample in duplicate, quantified gamma emissions on a Hewlett Packard Cobra II Auto Gamma Counter, and calculated CORT values from the standard curve. R esults Physiological Assay Validation o f Plasma CORT Plasma CORT titers significantly increased between 10 and 20 minutes of restraint stress (p = < 0.01, F = 13.36; Figure 2 1). There were no significant differences in plasma CORT titers between any of t he restraint treatments between zero and 10 minutes (0.88 < p < 0.23 for all ). Biochemical Validation I found log logit transformed curves of serially diluted fecal CORT extracts to be parallel to log logit transformed standard curves (Figure 2 2). I ntra assay variance was 3.67% for the

PAGE 29

29 high control and 1.71% and for the low control. Inter assay variance for the pooled plasma sample was 8.3% and 6.5% for the pooled fecal extract Conclusion My results demonstrate a biochemical and physiological valid ation of the use of radioimmunography to detect biologically relevant levels of CORT in short tailed singing mice. Restraint stress produced a significant increase in serum CORT titers between 10 and 20 minutes of restraint that was detected by RIA. Log logit transformed curves of serially diluted fecal samples were parallel to a similarly transformed standard curve. This demonstrates the validity of using RIA to measure fecal CORT metabolites at a dilution of 1:63 fecal extract to buffer. In summary, these findings allow us to use RIA to determine serum CORT titers and fecal CORT metabolites in a wild animal, the short tailed singing mouse.

PAGE 30

30 Figure 3 1. Mean serum CORT titer SEM following zero, 3, 5, 10, and 20 minutes of restraint stress (n = 24, 8, 9, 22 respectively). Serum CORT significantly increased between 10 and 20 minutes of restraint stress (*p < 0.01, F = 13.36).

PAGE 31

31 Figure 3 2. Parallelism of log transformed fecal CORT serial dilutions and logit transformed percent bound with transforme d standard curves. Standard curve (triangles): y = 4.34 1.91x, r 2 = 0.87. Fecal extract curve (circle): y = 4.97 1.43x, r 2 = 0.70.

PAGE 32

32 CHAPTER 4 DIVERGENT STRESS COP ING STY LES IN MALE SHORT TAILED SINGING MICE IN TWO DISTINCT HABITAT S Introductio n Physiological and behavioral responses to stress vary greatly between individuals in many taxa. These variable responses have been shown to form distinct complexes of correlated traits, known as stress coping styles which form consistent behavioral pat terns over time (Koolhaas et al. 1999). Stress coping styles are commonly described in a proactive/reactive dichotomy (Koolhaas 1999). Animals that exhibit the proactive stress coping style display low behavioral wariness in response to novelty, actively avoid or manipulate stressful stimuli, and have rigid, learned routines (Koolhaas et al. 1999, Koolhaas et al. 2007). Conversely, reactive animals are behaviorally wary in response to novelty, passively respond to stressors, and display flexible learning patterns (Koolhaas et al.1999, Koolhaas et al. 2007). In addition to behavioral characteristics, proactive/reactive stress coping styles are characterized by a suite of physiological responses to stress. Proactive animals are characterized by high sympa thetic activity and low hypothalamic pituitary adrenal (HPA) activity and reactivity (verli et al. 2007). Because of their low HPA activity/reactivity, proactive animals have low basal levels of glucocorticoids (the main hormones involved in the stress r esponse) and achieve low levels of stress induced glucocorticoids (verli et al. 2007). Compared to proactive animals, reactive animals show higher parasympathetic reactivity in response to stressors (verli et al. 2007). Additionally, reactive animals h ave higher HPA activity and reactivity, resulting in higher titers of basal and stress induced glucocorticoids (verli et al. 2007). Divergent stress coping styles are thought to give animals a survival advantage in specific environmental contexts (Marchet ti and Drent 2000, de Boer et al. 2003). For example, because of their behavioral wariness, reactive animals may have higher survival in environments that are

PAGE 33

33 densely populated by predators. Conversely, in predator depauperate environments, reactive anim als may have lower survival compared to proactive animals because of the substantial morbidity costs associated high levels of glucocorticoids such as reduced immunocompetence and longevity (Cavigelli and McClintock 2003). Additionally, divergent stress c oping styles may influence reproduction by creating tradeoffs between the expression of sexually selected traits, such as display vocalizations, and predator avoidance. Therefore, the frequency and fitness of divergent stress coping styles in populations is likely to be highly dependent on specific environmental factors (Sih et al. 2004). Despite being described in a number of species including laboratory rodents (De Boer et al. 2003), pigs (Ruis et al. 2001), birds (Marchetti and Drent 2000), fish (Dingem anse et al. 2007; Ward et al. 2004), and humans (Tyrka et al. 2006), few studies have examined the prevalence of divergent stress coping styles in wild animals. I examined the prevalence of divergent stress coping styles in short tailed singing mice from two sites in Boquete, Panama (Chapter 2). I measured behavioral response to novelty (open field test) and HPA activity and reactivity. Additionally, to examine the interaction between divergent stress coping styles and reproductive behaviors, I measured the call rate of mice and compared this to behavioral and physiological stress responses. Singing is presumed to be important for mate attraction in singing mice (Chapter 2). Examining the interaction between stress responses and singing behavior may pro vide insight into trade offs between reproduction and survival generated by divergent stress coping styles.

PAGE 34

34 Methods Animal Trapping and Site Descriptions I trapped male singing mice from May July 2006 and May June 2007 using the protocol described in Chapter 1. I housed mice in the laboratory for seven days using a standard protocol (Chapter 1). Behavioral Stress Response: Open Field Assay To measure behavioral stress response I conducted open field trials the morning following capture (Day 2) betwe en seven AM and noon. The open field test assays behavioral stress by taking advantage of the natural tendency of rodents to avoid open spaces. The open field arena consisted of a circular plastic container 50 cm in diameter and 22 cm in depth with an in ner circle 35 cm in diameter marked with push pins. I lined the arena with corn cob bedding. I changed the bedding between trials and cleaned the arena with ethanol. Prior to the initiation of the ten minute open field trial, I placed mice in an opaque c overed container in the center of the arena for five minutes. I videotaped trials using a Sony CCD RTV118 Handycam and used JWatcher (Blumstein et al. 2000) to quantify the following behaviors: immobility (sec), singing, grooming (sec), jumping, posting, defecating, and crossing between the inner and outer circle (described in Table 4 1). I collected the feces produced by the mice during the ten minute period to quantify fecal CORT titers. Hypothalamic Pituitary Adrenal (HPA) Activity and Reactivity: Feca l and Serum Corticosterone (CORT) I measured HPA reactivity by exposing mice to a zero (control) or 20 minute restraint stress treatments. For the 20 minute restraint stress treatment, I held animals by the scruff and restrained them on their back. Afte r restraint I immediately collected 100 the retro orbital sinus. For animals in the control treatment, I collected a blood sample from the

PAGE 35

35 retro oribital sinus in less than two minutes (including capture time) to obtain a measure of b asal CORT titers (Romero and Reed 2005). I centrifuged whole blood at 3000 rpm for four minutes and then isolated and stored plasma at Using a standard protocol, I extracted CORT from fecal samples (Mateo and Cavigelli 2005). I weighed feces and placed them in individual test tubes with 1 ml of 90% methanol. Using a spatula, I homogenized the samples and agitated them for 24 hours. I centrifuged the samples at 2000 rpm for five minutes, decanted the extract, and stored it glass test tubes at 2 until assayed. Radioimmunoassays (RIA) I measured serum and fecal CORT using 125 I radioimmunoassay kit (MP Biomedicals, Solon, OH; catalog no. 071200103). Using a combination of validation methods described by Good et al. (2003) and Touma et al. (20 04; Chapter 2), I physiologically validated the blood and fecal assay to ensure that it was sensitive enough to detect biological meaningful alternations in hormone alterations I diluted fecal samples to 1:63 and plasma samples to 1:255 in assay diluent prior to radioimmunoassay. I ran each sample in duplicate, quantified gamma emissions on a Hewlett Packard Cobra II Auto Gamma Counter, and calculated CORT values from a standard curve. Adrenocorticotropin Releasing Hormone (ACTH) Supplementation and Singing I measured spontaneous call rate from seven AM until noon the day following capture (Day 2). I placed mice in opaque acoustically insulated chambers (40cm x 40cm x 40cm) and allowed them to acclimate for 30 minutes prior to determining call rate. To measure call rate, I counted the number of songs in a two hour period. To examine the direct role that the HPA axis plays in singing behavior, I experimentally manipulated CORT titers and measured the resulting change in singing behavior. In a pair ed

PAGE 36

36 design, I exposed animals to an ACTH and a saline treatment, alternating the order of treatments to account for treatment order effects. I injected animals with 0.03 IU/ml of ACTH suspended in gelatin or 0.03ml of saline. I allowed animals to acclimat e to acoustically insulated chambers for 30 minutes and then counted the number of songs produced in one hour. Results Behavioral Stress Assay: Open Field Behavior There were significant site by year interactions for all open field behaviors. For this reason, I analyzed open field behavior separately for each year. In 2006, compared to males from the Park, males from the Reserve spent significantly less time immobile after trial initiation (p = 0.03, F = 5.03), jumped significantly more (p = 0.02, F = 5.69), crossed from the inner to outer circle significantly more times (p = 0.03, F = 5.69, and had a non significant trend to post with paws on the side of the arena more (p = 0.08, F = 3.19; Figure 4 1). There were no significant differences betwee n the Park and Reserve in number of defecations (p = 0.14, F = 2.13) or time spent grooming (sec; p = 0.42, F = 0.66; n = 29 for Reserve and 14 for Park). Finally, there were significantly more males from the Reserve who sang in the open field (n = 14) co mpared 2 = 10.02). In 2007, there were no significant differences between males from the Park (n =22) and the Reserve (n = 17) in immobility behavior (p = 0.37, F = 0.83) jumps (p = 0.42, F = 0.67), number of crosses between the inner and the outer circle (p = 0.32, F = 1.02), number of posts with paws on arena wall (p = 0.22, F = 1.57), number of defecations (p = 0.92, F = 0.10), or time spent grooming (sec; p = 0.98, F = 0.001). Though there were no signific ant differences between sites in any open field behaviors in 2007, it is interesting to note that many of the trends in the data or the opposite of the trends from 2006 (Figure 4 2).

PAGE 37

37 To determine whether individual differences in open field behavior rep resented a cohesive set of behavioral traits, I examined correlations between measures of immobility (latency to move, latency to cross pegs) and measures that could be interpreted to reflect active responses to the stressful environment (jumps, posts, cro ssing from inner to outer circle), using animals from both sites and years. All pairwise correlations among these variable were significant (Pearson's coefficient, 0.50
PAGE 38

38 0.21, F = 1.68, n = 21). There was a significant negative correlation between fecal CORT titers and the number of feces produced during the open field trial. No other open field behaviors were significantly relat ed to fecal corticosterone titers. Examining acute stress reactivity, I found no significant difference between Reserve and Park males in basal (control) serum CORT titers (p = 0.72, F = 0.13, n = 7 and 17 respectively). Conversely, males from the Par k achieved significantly higher serum CORT titers following 20 minutes of restraint stress (ANCOVA, p = 0.02, F = 6.68, n = 12 and 10 respectively, Figure 4 5). Singing Behavior and Stress Responses Combining data from 2006 and 2007, significantly more males from the Reserve (n = 46) called than males from the Park site (n = 35; p 2 = 17.01; Figure 4 6) in the two hour sampling period. 52% of males from the Reserve (n = 46) sang while only 8% of males from the Park site sang (n = 35). However, calling males from the Reserve (n = 24) did not call significantly more than c alling males from the Park (n = 3; MannWhitney U; p = 0.17; U = 18.0). Interestingly, spontaneous call rate is significantly positively correlated with anogenital distance (AGD) of calling males in both populations (n = 24; p = 0.04; Figure 4 7). In 2006, after seven days of captivity, there was no significant difference between the number of callers and noncallers from the Reserve 2 = 0.70, n = 21 and 11 respectively). Calling males from the Reserve sang significantly more on the seventh day in captivity compared to the second day (p = 0.1, t = 3.41, n =13). Although not significant, there was a trend for Park males to increase singing rate on the seventh day (p = 0.08, t = 2.24, n = 6). Among seventh day callers from the Park and the Reserve, there was a non significant positive correlation between fecal CORT titers and calling rate (Spearman correlatio n, p = 0.07, z = 1.83, n = 11).

PAGE 39

39 Among calling males from both the Park and the Reserve in 2006 and 2007 (n = 24), there was non significant negative correlation between fecal CORT titers and spontaneous singing rate (p = 0.21, F = 1.68). Additionally, th ere was no significant relationship between basal serum CORT titers (as measured from the restraint stress experiment) and calling rate in calling males from both sites and both years (p = 0.47, F = 0.55). Looking at differences in open field data between callers and non callers, calling males spent significantly less time in the inner circle after trial initiation than non calling males (p = 0.03, F= 5.10) and posted with their front legs against the wall significantly more than non calling males (p = 0.0 01, F = 11.59; Figure 4 8). However, there were no significant difference between calling and non calling males in other open field behaviors including the number of jumps against the arena wall (p = 0.56, F = 0.34) or the number of times an animal crosse d between the inner and outer circle ( p = 0.25, F = 1.36; Figure 4 8). Effect of ACTH on Singing Behavior Experimentally manipulating ACTH titers (and thus increasing CORT) did not alter singing rate compared to control treatments (Figure 4 9, Wilcoxon signed rank, p = 0.21). Discussion The results from this study indicate the existence of differences in the behavioral and physiological stress responses between short tailed singing mice from two distinct habitats. In 2006, male singing mice from the R eserve display a proactive like stress phenotype while males from the Park fit the reactive coping style. In the open field test they spent less time immobile and were more active (jumped and moved between circles more) compared to males from the Park. P roactive animals will often attempt to repeatedly solve a problem in the same way even when it proves ineffective; reactive animals will either attempt a new solution or abstain from further problem solving attempts (Korte et al. 1996). The males from the Reserve jumped

PAGE 40

40 significantly more in the open field than mice from the Park. If this behavior constituted an attempt to escape the open field apparatus, this routine formation lends further support to the proactive behavioral nature of Reserve males. I nterestingly, the open field behavioral trends I observed in 2006 were absent in 2007. In fact, though not significant, many of the trends were the opposite of what I observed in 2006. that some extrinsic ecological factor influenced open field behavior between years. Wild type rats ( Rattus norvegicus ) show no difference in some behavioral response tests such as the 8 arm radial maze and the Morris water maze on first exposure, but reac tive animals will behave more anxiously during the second exposure (File et al. 1996, Veenema et al. 2003). This suggests that prior constantly and unpredic habitat changes at this site between 2006 and 2007 influenced previous stress exposure and that these alterations were responsible for the shifts in open field behavioral trends Although many of the open field measures varied by site and year, there were strong overall correlations among open field behaviors across individuals. This supports the interpretation that these reflect a common suite of behaviors that can be interpret ed as a coping style. Unlike the behavioral data, the assays of physiological stress response yielded consistent results across years. Physiologically, the proactive/reactive dichotomy predicts that males from the Park would have higher basal CORT titers than males from the Reserve. Contrary to this prediction, I observed no significant difference between basal serum CORT between Reserve males and Park males. However, Park males achieved higher serum CORT titers following a 20

PAGE 41

41 minute period of restraint stress than Reserve males, indicating a more reactive HPA response as I would predict from a reactive phenotype. Interestingly, Reserve males had significantly higher fecal CORT titers than males from the Park when assayed on the second day of captivity. Fecal samples integrate circulating hormone titers over time including basal and acute episodic hormone fluctuations. For this reason, measuring hormones from fecal samples is thought to more accurately represent an es not capture a single point in time like blood samples (Touma et al. 2004). If this is true, it is interesting and unexpected that males from the Reserve had higher fecal CORT titers than males from the Park. One possible explanation is that, though Pa rk males achieve a higher stress induced CORT response, they are able to attenuate HPA activity more rapidly or metabolize CORT more rapidly than males from the Reserve. In an environmental context where animals are exposed to chronic stress, the ability to rapidly attenuate HPA activity or metabolize endogenous CORT can reduce morbidity costs associated with elevated levels of glucocorticoids such as suppression of reproduction, muscle wasting, growth suppression, neuronal death, and immune system suppres sion (Sapolsky et al. 2000, Wingfield et al. 1997). Attenuation of HPA activity in response to chronic stress has been demonstrated in European starlings (Rich and Romero 2005, Cyr and Romero 2007), pigs (Ruis et al. 2001), and rats (Swiergiel et al. 2007 ). Mice from the Park may chronically be exposed to stress as their environment is unpredictably manipulated for agricultural purposes, which could influence resource availability, conspecific population density, or predator density. If this is true, it would explain why mice from the Park have a higher HPA response to an acute stressor, but lower levels of fecal CORT.

PAGE 42

42 I found a greater prevalence of mice that sang at the Reserve than at the Park. Looking across years and sites, animals that vocalized (c allers) were more likely to display some open field behaviors associated with the proactive stress coping style (jumping and shorter periods of immobility). While singing behavior is associated with behavioral stress coping style, it does not appear to be related to the physiological component of stress coping styles. Singing behavior was not predicted by second day fecal CORT titers or basal serum CORT titers and ACTH supplementation did not decrease singing behavior as would be predicted if propensity t o sing was modulated by a CORT mechanism. Although I observed an increase in singing behavior and a concordant decrease in fecal CORT titers from the second to the seventh day in captivity, this could represent an overall attenuation of the stress respons e to captivity and not a causative mechanism. In addition to differential HPA activity and stress response behaviors, proactive/reactive animals are also characterized by differential levels of conspecific aggression. Serotonin is an important neuromodu lator in the regulation of social behavior and aggression and has been shown to modulate display vocalizations associated with dominance in weakly electric fish ( Apteronotus leptorhynchus si nging mice is controlled by a similar mechanism and would explain why ACTH supplementation had no effect on singing behavior. In conclusion, I describe behavioral and physiological differences in stress responses in 2006 that form phenotypic assemblage s that closely mirror the characteristics of proactive and reactive phenotypes described by Koolhaas et al. (1999). However, in 2007, behavioral stress responses in the open field assay do not conform to the proactive/reactive dichotomy, nor were they cons istent with the behavioral trends in 2006. Conversely, differences in physiological stress response between sites are consistent between years and, in part, conform to the

PAGE 43

43 proactive/reactive dichotomy. The anomalous behavioral trends between years and th e lack of correlation between behavioral and physiological responses that typify the proactive/reactive stress coping style dichotomy suggest that the proactive/reactive stress coping style dichotomy does not appropriately or consistently describe differen ces in stress responses between Park and Reserve animals. However, the consistency in the relationship between stress coping style and singing across years and between sites suggests that vocalization behavior can be mapped onto the proactive/reactive dic hotomy. Specifically, I suggest that behaviorally proactive animals are more likely to engage in display vocalizations that reactive animals. This suggests that animals may face trade offs between reproductive success and stress coping style. The unique vocalization behavior of singing mice provides the opportunity to examine these tradeoffs in wild animals, making them an ideal system in which to study the ecological and evolutionary relevance of divergent stress coping styles.

PAGE 44

44 Figure 4 1. Me an SEM measures of open field behavior in 2006 with n = 29 for Reserve and n = 14 for Park. M ales from the Reserve (A) had significantly shorter periods of immobility (p = 0.03, F = 5.03), (B) had a non significant trend to post with paws on the wall mo re than Park males (p = 0.08, F = 3.19), (C) jumped significantly more against the arena wall (p = 0.02, F = 5.69), and (D) crossed from the inner to the outer circle significantly more than Park males (p = 0.03, F = 5.69). indicates p < 0.05.

PAGE 45

45 Figu re 4 2. Mean SEM measures of open field behavior in 2007 with n = 17 for Reserve and n = 22 for Park. There were no significant differences between (A) period of immobility following trail initiation (p = 0.37, F = 0.83), (B) number of posts with paws on the arena wall (p = 0.22, F = 1.57), (C) number of jumps against the arena wall (p = 0.42, F = 0.67), and (D) number of times individuals crossed from the inner to the outer circle (p = 0.32, F = 1.02).

PAGE 46

46 Figure 4 3. Mean fecal CORT SEM for 2006 and 2007 combined. Reserve males had significantly higher titers of fecal CORT than Park males (*p = 0.03, F = 4.94, n = 38 and 27 respectively.

PAGE 47

47 Figure 4 4. Mean fecal CORT (ng/g) SEM on the second and seventh day in captivity. Mice from the Reserve (A) had significantly lower fecal CORT titers on the seventh day in captivity compared to the second day (*p = 0.01, t = 4.81, n = 6). Likewise, mice from the Park showed a significant decrease in fecal CORT titers from the second to seventh day in captivity (**p = 0.03, t = 2.69, n = 8.

PAGE 48

48 Figure 4 5. Mean SEM CORT secretion in basal and stress (20 minute restraint) conditions. Animals from the Park showed higher post stress levels of CORT compared to animals from the Reserve (n = 12 a nd 10 respectively, *p = 0.02, F =6.68 ANCOVA). No site differences were observed for the control (n = 7 and 17, p = 0.72, F = 0.13,

PAGE 49

49 Figure 4 6. Percentage of males that sang at least once in a two hour period ( c aller s) and males that did not sing (n oncaller s ). The Reserve had significa ntly more c Park (P < 0.00, 2 = 17.0). 0% 10% 20% 30% 40% 5 0% 60% 70% 80% 90% 100% Reserve Park Noncaller Caller n = 22 n = 24 n = 32 n = 3 Percentage of individuals

PAGE 50

50 Figure 4 7. Singing rate in a two hour period in relation to anogenital distance. Linear regression with 95% confidence bands for mean (r 2 = 0.14, P = 0.04, n = 25). 0 2 4 6 8 10 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 Anogential distance (cm) Number of songs in a two hour period

PAGE 51

51 Figure 4 8. Mean SEM measures of ope n field behavior in 2006 and 2007 with n = 20 for Callers and n = 62 for Noncallers. Compared to Noncallers, Callers (A) spent significantly less time in the inner circle after trial initiation (p = 0.03, F = 5.10) and (B) posted with their front legs aga inst the arena significantly more (p = 0.001, F = 11.59. There were no significant differences between (C) number of jumps against the arena wall (p = 0.56, F = 0.34) or the number of times an animal traversed between the inner and outer circle ( p = 0.25 F = 1.36). indicates p < 0.05.

PAGE 52

52 Figure 4 9. ACTH treatment did not significantly change singing behavior compared to control treatments (Wilcoxon signed rank, p = 0.21, n = 7 per treatment). 1 2 3 4 5 6 Mean songs in 2 hours SEM First Treatment Second Treatment Saline ACTH

PAGE 53

53 Table 4 1. De scription of behaviors and activities measured in the open field behavior assay.

PAGE 54

54 LIST OF REFERENCES Ashby J Tinwell H Lefevre P A Odum D Paton S W Millward S Tittensor S Brooks A N. (1997) Normal sexual development of rats exposed to butyl benzyl phthalate from conception to weaning Regulatory Toxicology and Pharmacology 26: 102 118. Arnqvist, G., Henriksson, S. (1997) Sexual cannibalism in the fi shing spider and a model for the evolution of sexual cannibalism based on genetic constraints Evolutionary Ecology 11: 255 273. Bell, A.M. (2005) Behavioral differences between individuals in two populations of stickleback ( Gasterosteus aculeatus ). Journal of Evolutionary Biology 18: 261 276. Boinski, S., Gross, T.S., Davis, J.K. (1999) Terrestrial Predator alarm vocalizations are a valid monitor of stress in captive brown capuchins ( Cebus paella ). Zoo Biology 18: 295 312. Macquarie Univer sity. Blumstein, D.T., Patton, M.L., Saltzman, W. (2006) Faecal glucocorticoid metabolites and alarm calling in free living yellow bellied marmots. Biology Letters 2: 29 32. Cavigelli, S.A., McClintock (2003) Fear of novelty in infant rats predicts a dult corticosterone dynamics and an early death. Proceedings of the National Academy of Sciences USA, 100: 16131 16136. Cyr, N.E., Romero, L.M. (2007) Chronic stress in free living European starlings reduces corticosterone concentrations and reproductive success. General and Comparative Endocrinology 151: 82 89. De Boer, S.F., Van Der Vegt, B.J., Koolhaas, J.M. (2003) Individual variation in aggression of feral rodent strains: A standard for the genetics of aggression and violence? Behavior Genetics 3 3(5): 485 501. Dingemanse N.J., Bot, C., Drent, P.J., Tinbergen, J.M. (2004) Fitness consequences of avian personality in a fluctuating environment. Proceedings of Biological Science 271: 847 852. Dingemanse, N.J., Wright, J., Kazem, A.J.N., Thomas, D.K., Hickling, R., Dawnay, N. (2007) Behavioral syndromes differ predictably between 12 populations of three spined stickleback. Journal of Animal Ecology 76: 1128 1138. Faure, P.A., Hoy, R.R. (2000 ) The sounds of silence: cessation of singing and song pausing are ultrasound induced acoustic startle behaviors in the katydid Neoconcocephalus ensiger (Orthoptera: Tettigoniidae).

PAGE 55

55 Good, T., Khan, M.Z., Lynch, J.W. (2003) Biochemical and physiological validation of a corticosteroid radioimmunoassay for pla sma and fecal samples in oldfield mice ( Peromyscus polionotus ). Physiological and Behavior 80: 405 411. Harper, J.M., Austad, S.M. (2000) Fecal glucocorticoids: a noninvasive method of measuring adrenal activity in wild and captive rodents. Physiologi cal and Biochemical Zoology 73(1): 12 22. Hooper, E.T., Carelton, M.D. (1976) Reproduction, growth and development in two contiguously allopatric rodent species, Genus Scotinomys in Miscellaneous Publications, Museum of Zoology, University of Michiga n, No. 151. Koolhass, J.M., Korte, S.M., De Boer, S.F., Van Der Vegt, B.J., Van Reenen, C.G., Hopster, H., De Jong, I.C., Ruis, M.A.W., Blokhuis, H.J. (1999) Coping styles in animals: current status in behavior and stress physiology. Neuroscience and Bio behavioral Reviews 23: 925 935. Koolhass, J.M., de Boer, S.F., Buwalda, B., van Reenen, K. (2007) Individual variation in coping with stress: a multidimensional approach of ultimate and proximate mechanisms. Brain, Behavior, and Evolution 70: 218 226. Korte, S.M., De Kloet, E.R., Buwalda, B., Bouman, S.D., Bohus, B. (1996) Antisense to the glucocorticoid receptor in hippocampal dentate gyrus reduces immobility in forced swim test. European Journal of Pharmacology 301: 19 25. Korte, S.M., Koolhaas, J.M., Wingfield, J.C., McEwen, B.S. (2005) The Darwinian concept of stress: benefits of allostasis and costs of allostatic load and the trade offs in health and disease. Neuroscience and Biobehavioral Reviews 29: 3 38. Leary, C.J., Garcia, A.M., Knapp, R. (2006) Stress hormone is implicated in satellite caller associations and sexual selection in the Great Plains toad. The American Naturalist 168(4): 431 440. Marchetti, C., Drent, P.J. (2000) Individual differences in the use of social information in f oraging by captive great tits. Animal Behavior 60: 131 140. Magnhagen, C. Staffan, F. (2005) Is boldness affected by group composition in young of the year perch ( Perca fluviatilis )? Behavioral Ecology and Sociobiology 57: 295 303. Mateo, J.M., Cavi gelli, S.A. (2005) A validation of extraction methods for noninvasive sampling of glucocorticoids in free living ground squirrels. Physiological and Biochemical Zoology 78: 1069 1084. Nelson, R. (2005) Stress. In: An Introduction to Behavioral Endocrin ology Sunderland, MA, Sinauer Associates, Inc. Publishers, pages 669 720.

PAGE 56

56 verli, Srensen, C., Pulman, K.G.T., Pottinger, T.G., Korzan, W., Summers, C.H., Nillson, G.E. (2007) Evolutionary background for stress coping styles: Relationships between physiological, behavioral, and cognitive traits in non mammalian vertebrates. Neuroscience and Biobehavioral Reveiws 31: 396 412. Palme, R. (2005) Measuring fecal steroids guidelines for practical application. Romero, L.M., Reed, J.M. (2005). Collecti ng baseline corticosterone samples in the field: is under 3 min good enough? Comparative Biochemistry and Physiology, Part A, 140: 73 79. Pocock, V.J., Sales, G.D., Wilson, C.A., Milligan, S.R. (2002) Effects of perinatal octylphenol on ultrasound vocali zation, behavior and reproductive physiology in rats Physiology and Behavior, 76: 645 653. Remage Healey, L.R., Nowacek, D.P., Bass, A.H. (2006) Dolphin foraging sounds suppress calling and elevate stress hormone levels in a prey species, the Gulf toad fish. The Journal of Experimental Biology, 209: 4444 4451. Rich, E.L., Romero, L.M. (2005) Exposure to chronic stress downregulates corticosterone responses to acute stressors. American Journal of Regulatory, Integrative, and Comparative Physiology 288: R1628 R1636. Romero, L.M., Reed, J.M. (2005). Collecting baseline corticosterone samples in the field: is under 3 min good enough? Comparative Biochemistry and Physiology, Part A, 140: 73 79. Ruis, M.A.W, te Brake, J.H.A., Engel, B., Buist, W.G., Blo khuis, H.J., Koolhaas, J.M. (2001) Adaptation to social isolation acute and long term stress responses in growing gilts with different coping characteristics. Physiology and Behavior, 73: 541 551. Ryan, M.J. (1985) The Tngara Frog. A Study in Sexual Se lection and Communication Chicago, IL: University of Chicago Press. Sapolosky, R.M., Romero, L.M., Munck, A.U. (2000) How do glucocorticoids influence stress responses? Integrating permissive, suppressive, stimulatory, and preparative actions. Endocrine Reviews 21(1): 55 89. Sih, A., Bell, A., Johnson, J.C. (2004) Behavioral syndromes: an ecological and evolutionary overview. Trends in Ecology and Evolution 19(7): 372 378. Stoker, T. E., Laws, S. C., Guidici, D. L., and Cooper, R. L. (2000) The eff ect of atrazine on puberty in male Wistar rats: An evaluation in the protocol for the assessment of p ubertal development and thyroid function. Toxicological Sciences, 58, 50 59. Swiergiel, A.H., Zhou, Y., Dunn, A.J. (2007) Effects of chronic footshock, re straint and corticotrophin releasing factor on freezing, ultrasonic vocalization and forced swim behavior in rats. Behavioral Brain Research 183: 178 187.

PAGE 57

57 Telgkamp, P., Combs, N., Smith, G.T. (2007) Serotonin in a diencephalic nucleus c ontrolling comm unication in an electric fish: s exual dimorphism and relationship to i ndicators of d ominance Developmental Neurobiology 67(3): 339 354. Touma, C., Sachser, N., M stl, E., Palme R. (2003) Effects of sex and time of day on metabolism and excretion of c orticosterone in urine and feces of mice General and Comparative Endocrinology 130: 267 278. Touma, C., Palme, R., Sachser, N. (2004) Analyzing corticosterone metabolites in fecal samples of mice: a noninvasive technique to monitor stress hormones. Hormones and Behavior 45: 10 22. Tyrka, A.R., Mello, A.F., Mellow, M.F., Gagne, G.G., Grover, K.E., Anderson, G.M., Price, L.H., Carpenter, L.L. (2006) Temperament and hypothalamic pituitary adrenal axis function in healthy adults. Psychoneuroendocrin ology 31: 1036 1045. Ward, A.J.W., Thomas, P., Hart, P.J.B., Krause, J. (2004) Correlates of boldness in three spined sticklebacks ( Gasterosteus aculeatus ). Behavioral Ecology and Sociobiology 55: 561 568.b Wilhelms, K.W., Cutler, S.A., Proudman, J.A., Anderson, L.L., Scanes C.G. (2005) Atrazine and the hypothalamo pituitary gonadal axis in s exually m aturing p recocial b irds: s tudies in m ale Japanese Quail Toxicological Sciences, 86(1): 152 160. Wingfield, J.C., Hunt, K., Breuner, C., Dunlap, K., Fowler, G.S., Freed, L., Lepson, J. (1997) Environmental stress, field endocrinology, and conservation biology. In: Behavioral Approaches to Conservation in the Wild edited by Clemmons, J.R. and Buchholz, R. Campbridge, UK: Cambridge University Press, 1997, p 95 131.

PAGE 58

58 BIOGRAPHICAL SKETCH Andrea Crino was born in Forest Grove, Oregon, in a cabin in the woods. A childhood spent rampaging through the wilderness, eating questionable plants, and poking unsuspecting animals with sticks instilled Ms. Cr ino with a deep respect and sense of passion for the natural world. She pursued this interest by earning a Bachelor of Arts degree in biology at Lewis & Clark College in 2002. As an undergraduate, Ms. Crino spent a semester in Costa Rica studying tropica l ecology and conservation with the Organization of Tropical Studies. This fantastic experience motivated Ms. Crino to pursue independent research through two Research Experience for Undergraduates fellowships with the National Science Foundation at Baylo r University and the Rocky Mountain Biological Station. Ms. Crino was actively involved in community service at her undergraduate institution. She served as a teaching assistant for introductory biology courses, led middle school girls in weekly science activities to increase confidence, and volunteered with the Nature Conservancy. After graduating Ms Crino, worked as an intern at Archbold Biological Station and the Smithsonian Tropical Research Institute and as a research technician at Arizona State U niversity before returning to school to pursue a Master of Science at the University of Florida. She completed her degree in August of 2008. While in graduate school, Ms Crino completed two half marathons, two full marathons, and two Olympic distance tri athlons. She is a devoted rock climber, an amateur surfer, consummate booty shaker, and enjoys a fine Pinot Grigio on warm, summer days in Florida.