This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
The Pennsylvania State University
The Graduate School
Intercollege Graduate Degree Program in Ecology
FACTORS AFFECTING THE PHYSIOLOGICAL CONSEQUENCES OF STRESS IN
The Vertebrate Stress Response and its Outcomes .......................................................... 1Selective Environment .............................................................................................. 3Stressor Characteristics ............................................................................................. 3History of Stress Exposure ....................................................................................... 5
Research Objectives and Study System ............................................................................ 6References ........................................................................................................................ 10
Chapter 2 Immune responses of eastern fence lizards (Sceloporus undulatus) to repeated acute elevations of corticosterone ..................................................................................... 20
Chapter 3 How do duration, frequency, and intensity of exogenous CORT elevation affect immune outcomes of stress? ................................................................................... 45
Study System ............................................................................................................ 49Treatments ................................................................................................................ 49Blood Collection ....................................................................................................... 51
Chapter 4 Ancestry trumps experience: Cross-generational but not early life stress affects the adult physiological stress response ............................................................................. 68
Study System and Animal Collection ....................................................................... 72Animal Husbandry .................................................................................................... 72Treatments ................................................................................................................ 73Blood Collection and Stress Assays ......................................................................... 74ACTH Challenge ...................................................................................................... 75Hormone Analysis .................................................................................................... 76Data Analysis ............................................................................................................ 76
Results .............................................................................................................................. 78Baseline Corticosterone ............................................................................................ 78Corticosterone Reactivity to Restraint ...................................................................... 78Corticosterone Response to Fire Ants ...................................................................... 78Corticosterone Response to ACTH Challenge ......................................................... 79
Discussion ......................................................................................................................... 79Early Life Stress ........................................................................................................ 80Cross-Generational Exposure to Stress ..................................................................... 81Baseline CORT ......................................................................................................... 83Conclusions ............................................................................................................... 83
Chapter 5 Reaping the rewards: High-stressed populations up-regulate immune function in the face of stress ........................................................................................................... 95
Figure 2-1: Plasma CORT concentrations of lizards from high-stress (fire ant invaded) and low-stress (uninvaded) populations after 23 days of treatment with CORT (shaded bars) or control (oil vehicle only; open bars). CORT-treated lizards had significantly higher plasma CORT concentrations than control lizards. This relationship was consistent across high- and low-stress populations (i.e. no significant effect of invasion status). Bars represent means ± one standard error. Sample size for each group is given within each bar. ....................................................... 35
Figure 2-2: A) Hemagglutination scores and B) percent bacterial killing by plasma of lizards from high-stress (fire ant invaded) and low-stress (uninvaded) populations after 23 days of treatment with CORT (shaded bars) or control (oil vehicle only; open bars). CORT-treated lizards had significantly higher hemagglutination scores than but similar bacterial killing ability to control lizards. These relationships were consistent across high- and low-stress populations (i.e. no significant effect of invasion status). Bars represent means ± one standard error. Sample size for each group is given within each bar. ......................................................................................... 36
Figure 3-1: A) The frequency, intensity, and duration of CORT application in each of the treatments used in this study, and the total amount of CORT received in each 3-day period. Text in parentheses indicates: for Frequency, how frequently a CORT-oil solution was applied (oil-vehicle only was applied on remaining days); for Intensity, the amount of CORT applied during each application; and for Duration, whether the period of CORT elevation was short or long. Italicized pairs in each column represents treatments that differ in only the parameter shown in that column. B) A graphical representation of the amount of CORT applied for each of the treatments used in this study (Control (Ctl) had no CORT applied). This is provided for illustrative purposes, to convey the expected duration of CORT release following application. ....................................................................................................................... 61
Figure 3-2: A) Hemagglutination scores and B) Percent bacterial killing by lizard plasma after 9 days of treatment (see Fig. 3-1). Lizards in the Low Acute (LA), Repeated Acute (RA), and Control (Ctl) treatments had significantly higher hemagglutination scores than did those in the High Acute (HA) treatment, and those in the Chronic (Ch) treatment had hemagglutination scores that were intermediate to these groups. Lines above the columns connect treatments that do not significantly differ from one another. Bacterial killing ability did not significantly differ across treatments. Error bars represent means ± one standard error. The sample size for each group is given within each bar. ................................................................................................................. 62
Figure 4-1: CORT reactivity to restraint is greater in offspring of lizards from fire ant-invaded sites. Adult concentrations of CORT at baseline (shaded bars) and following restraint in a bag (white bars) of lizards exposed weekly to fire ants (FA),
ix
exogenous CORT, or control treatment during early life. CORT reactivity (post-restraint stressor minus baseline) was assessed in the statistical model but stress-induced concentrations are plotted here for ease of comparisons between graphs. Bars represent means ± one standard error. The sample size for each group is given above each bar. ................................................................................................................. 85
Figure 4-2: CORT concentrations following adult fire ant exposure are greater in offspring of lizards from fire ant-invaded sites. CORT concentrations following exposure in adulthood to an empty arena (FA control; shaded bars) or attack by fire ants (FA; white bars) of lizards exposed weekly to fire ants (FA), exogenous CORT, or control treatment during early life. Bars represent means ± one standard error. The sample size for each group is given above each bar. ................................................. 86
Figure 4-3: ACTH-induced CORT concentrations are greater in offspring of lizards from fire ant-invaded sites. Adult CORT concentrations following injection with saline solution (shaded bars) or ACTH (white bars) of lizards exposed weekly to fire ants (FA), exogenous CORT, or control treatment during early life. Bars represent means ± one standard error. The sample size for each group is given above each bar. ... 87
Figure 5-1: Early life and cross-generational history of stress exposure interact to affect adult baseline, but not post-inoculation, hemagglutination scores. a) In offspring of lizards from fire ant-uninvaded populations, CORT exposure during early life suppressed adult baseline plasma hemagglutination compared to controls. The opposite effect was seen in offspring of lizards from fire ant-invaded populations: early life CORT exposure enhanced adult baseline hemagglutination compared to controls. b) Post-inoculation hemagglutination scores did not differ across early life stress treatment or fire ant-invasion status. Bars represent means ± one standard error and sample size for each group is shown above each set of bars. ...... 112
Figure 5-2: Bacterial killing by plasma of adult lizards is not related to early life or cross-generational history with stress. Offspring of lizards from fire ant invaded and uninvaded populations exposed weekly to fire ants (FA), CORT, or control treatment from hatching until maturity had similar percent bacterial killing ability of plasma as adults. Bars represent means ± one standard error and sample size for each group is shown above each bar. ........................................................................................ 113
x
LIST OF TABLES
Table 3-1: This table shows the different immune outcomes (enhancement, no change, or suppression) to repeated exposure to stress (exogenous application of CORT, handling stress, or an ecologically-relevant stressor), indicating the study organism and immune component measured. Exogenous CORT was elevated using topical application or feeding. Handling includes handling or chasing, placement in a bag, or air exposure (fish). Ecologically relevant stressors include food deprivation, social isolation, social defeat, or exposure to predator scents. *Animals were simultaneously restrained. Abbreviations as follows: Hemag. = hemagglutination; BKA = bacterial killing ability; DTH = delayed-type hyper sensitivity; Antibody Resp. = antibody response. ............................................................................................... 60
Table A-1: Representative studies from the literature showing the full range of both duration of stress applications (rows) and consequences (columns) of stress, using definitions of “acute” and “chronic” stress from each source paper. Shaded boxes represent combinations expected according to existing theory. ....................................... 131
xi
ACKNOWLEDGEMENTS
Firstly, I thank my advisor, Tracy Langkilde, for her guidance and enthusiasm over the
last five years. Her encouragement, patience, and many many track changes greatly contributed to
my growth as a scientist, and her friendship made my journey through graduate school not only
possible, but also pleasurable. I am also grateful to my committee members for their advice and
thoughtful feedback and to my undergraduate mentors, Phil Myers and Catherine Badgey,
without whom I would not have considered attending grad school.
I am forever indebted to the past and present members of the Langkilde Lab. For their
friendship, manuscript edits, assistance bleeding lizards, movie nights, and all of the Cheez-Its
flavors ever, I thank: Renee Rosier, Lindsey Swierk, Brad Carlson, Jenny Tennessen, Chris
Thawley, Chris Howey, Sean Graham, Travis Robbins, Nicole Freidenfelds, Kelly Brossman,
Caty Tylan, Braulio Assis, Dustin Owen, and Cam Venable. I am also grateful to the horde of
undergraduates who made this work possible, especially Courtney Norjen, Melissa O’Brien,
Tommy Cerri, Mark Herr, and Mark Goldy-Brown for their patience and dedication.
My deepest thanks to my friends and family. Thank you to the State College theatre
community for making State College home; to Katie, Sean, Lia, Nitesh, Heather, and Katelyn for
their never-ending supply of support, understanding, and whimsy; to TL for sharing DPTL, which
translates to all facets of life; and to EJ for encouraging simultaneous pursuit of disparate
passions. I am grateful to my parents and brother for their patience and support of all my
endeavors. And finally, thank you to my boyfriend, Rich, for his unwavering support, constant
presence, and for smashing lizard poop on a Friday night. Thanks for saving the day on multiple
occasions and for the adventures, dinosaurs, and love.
xii
***
I am grateful to the National Science Foundation, the American Society for Ichthyologists and
Herpetologists, the Society for Integrative and Comparative Biology, the Ecological Society of
America Physiology Section, the Huck Institute of the Life Sciences, the Penn State Department
of Biology, and the Penn State Intercollege Graduate Degree Program in Ecology for providing
conference funding as well as financial and logistical support. The research presented here
adheres to the Guidelines for the Use of Animals in Research and the Institutional Guidelines of
Penn State University (IACUC #35780) and animal collection was permitted by the respective
states.
1
Chapter 1
Introduction
The global environment is changing with increasing speed (Vitousek, 1997), exposing
populations to novel stressors (Tylianakis et al., 2008; Vitousek, 1997). Environmental changes
including habitat loss (Homan et al., 2003; Suorsa et al., 2004), pollution (Norris et al., 1999;
Tomei et al., 2003; Wikelski et al., 2001), urbanization (French et al., 2008), and the introduction
of novel predators (Berger et al., 2007; Graham et al., 2012), can elicit a physiological stress
response in vertebrates. Although the physiological stress response is critical in the short-term,
long-term or frequent activation can reduce fitness by altering behavior, reproduction, stress
physiology, and immune function (McEwen, 1998a). The magnitude and direction of these
outcomes may vary with the type of stressor (e.g. social defeat, metabolic stress, immune stress,
trauma; Segerstrom and Miller 2004; Koolhaas et al. 2011; Ariza Traslaviña et al. 2014), stressor
characteristics (e.g. duration, frequency, intensity; Busch et al., 2008; Dhabhar and McEwen,
1997; Martin, 2009; McEwen et al., 1997), or an individual’s or population’s previous exposure
to stress (Carpenter et al., 2007; Franklin et al., 2010; Spencer et al., 2009; Yehuda et al., 2000).
Understanding how such factors may influence the outcomes of stress (e.g. immune costs) will
enable predictions of how organisms respond to environmental change.
The Vertebrate Stress Response and its Outcomes
Stress can be defined as an external challenge to an organism’s normal functioning
(homeostasis; Burchfield, 1979; Dhabhar, 2007; Levine, 2005). A great number of stimuli
(stressors) can present such a challenge, and stress is thus unavoidable (Selye, 1978). When a
2
stressor is encountered, the body undergoes a series of changes in order to survive and recover
from the threat. This physiological response to stress is highly conserved across vertebrates and
includes activation of both the sympathetic nervous system and the hypothalamic–pituitary–
adrenal (HPA) axis (Seaward, 2006). The sympathetic nervous system induces the fight-or-flight
response by rapid production of the catecholamines epinephrine and norepinephrine, whose
effects last only seconds (Seaward, 2006). Longer term effects (minutes to hours, or even weeks)
are mediated by the HPA axis (Seaward, 2006; Stratakis and Chrousos, 1995). Briefly, the
hypothalamus produces corticotropin-releasing factor (CRF), which acts on the pituitary. The
pituitary then secretes adrenocorticotropic hormone (ACTH), which stimulates the adrenal glands
to produce and release glucocorticoid hormones such as cortisol or corticosterone (CORT). The
effects of glucocorticoids are numerous and long-lasting (Sapolsky et al., 2000), and CORT is
frequently measured as a proxy for stress.
Elevation of plasma CORT can help an organism appropriately respond to a stressor
(Munck et al., 1984; Sapolsky et al., 2000; Stratakis and Chrousos, 1995). For example, elevation
of CORT can trigger important behavior (including anti-predatory behavior, Remage-Healey and
Romero, 2001; Thaker et al., 2009), mobilize stored energy (Sapolsky et al., 2000), enhance
immune function in preparation for wounding or subsequent risk of infection (Dhabhar, 2009;
Martin, 2009), and alter metabolism to help an individual maintain homeostasis (Stratakis and
Chrousos, 1995). For these reasons, the stress response is generally considered adaptive in the
short term (McEwen, 2008). By contrast, persistent or long-duration elevation of CORT, such as
that experienced when an organism is continuously or repeatedly exposed to a threat, may
suppress reproduction (Moore and Jessop, 2003; Salvante and Williams, 2003), growth (Barton et
al., 1986; Bourgeon and Raclot, 2006; Morici et al., 1997), and immune function (Dhabhar, 2009;
Martin, 2009). This may result from the long-term diversion of energy away from these energy-
sensitive functions (McEwen and Wingfield, 2003; McEwen, 1998b; Romero et al., 2009).
3
Despite these general trends, documented outcomes of both short- and long-duration stress vary
greatly (Table A-1). The factors underlying this variation are likely numerous and may include
the selective environment (French et al., 2008; Martin II et al., 2005), characteristics of the
stressor (Busch et al., 2008; Martin, 2009; McEwen et al., 1997), and an organism’s or
population’s history with stress (Harris and Seckl, 2011; McCormick and Green, 2013;
Veenema, 2009; Yehuda et al., 2000). The potential effects of these factors on the outcomes of
stress are addressed below.
Selective Environment
The nature of the costs of the stress response may vary depending on context. For
example, in populations where immune function is critical to survival, such as where the
prevalent stressor is likely to injure an individual or induce behavior that increases risk of
infection (Cox and John-Alder, 2007; Ezenwa et al., 2012), immune function may be maintained
by reallocating energy from growth or reproduction. Over time, one might expect to see selection
against immune suppression (when it exists), and maybe even selection for immune enhancement
within populations that are frequently exposed to immune-activating stressors (French et al.,
2008; Martin II et al., 2005). Understanding how populations vary in their response to stress may
have important management implications, as management strategies may need to be altered at a
population level according to the local selective environment.
Stressor Characteristics
Stress is typically characterized by the duration of the stressor: short-duration exposure to
a stressor lasting from minutes to a few days (Harbuz and Lightman, 1992; Martin, 2009) is
4
typically termed “acute” (Burchfield, 1979; Romero, 2004), while long-duration continuous
exposure to a stressor over days or months (Dhabhar, 2009; Martin, 2009) is typically termed
“chronic” (Burchfield, 1979; Romero, 2004; Sapolsky et al., 2000). Repeated exposure to short-
duration stressors, such as frequent disturbances by humans or sub-lethal predator attacks, are
sometimes referred to as “repeated acute” (Burchfield, 1979; Busch et al., 2008) but usually
classified as “chronic” (Harbuz and Lightman, 1992; Romero, 2004). In spite of their frequent use
in the literature and in medical practices, the terms “acute” and “chronic” are inconsistently
applied (Appendix). It is also unclear when “acute” stress becomes “chronic” and where repeated
acute stressors fall on the acute-chronic spectrum. This distinct terminology can be problematic,
as beneficial or harmful physiological outcomes are typically associated with acute and chronic
stress, respectively (Martin, 2009; McEwen, 2008), but observed outcomes commonly vary from
these associations (Table A-1; e.g., Chester et al., 2010; Harris et al., 2002; Merrill et al., 2012;
Ottenweller et al., 1992).
Although prediction of stress outcomes is almost exclusively associated with stressor
duration, other aspects such as stressor intensity and frequency may play important roles. For
example, exposure to conspecifics and predators are stressors that may differ in intensity and may
be encountered at different rates. However, stressor intensity and frequency are largely ignored in
the literature (but see Busch et al., 2008; McCormick et al., 1998; McEwen et al., 1997;
Ottenweller et al., 1989). Predicting the independent and synergistic effects of stressor
characteristics is critical if we are to gain a complete understanding of how organisms respond to
stress.
5
History of Stress Exposure
Stress experienced by an organism within its lifetime may affect its physiological
response to stress (Ladd et al., 1996; Spencer et al., 2009) and may affect HPA activity in future
generations (Jenkins et al., 2014; Yehuda et al., 2000). When persistently exposed to stress either
within a lifetime or across generations, an organism’s physiological stress response may change
to balance associated costs and benefits (Matthews, 2002; Meaney et al., 1994; Oitzl et al., 2010).
Down-regulation of HPA activity may reduce the costs associated with this stress response
(Martin, 2009; Romero, 2004; Romero et al., 2009). Alternatively, if the benefits of the stress
response outweigh the costs, organisms frequently exposed to stress may up-regulate HPA
activity to take advantage of these benefits (Romero, 2004; Romero et al., 2009; Sapolsky et al.,
2000). Altering the stress response to balance these costs and benefits may alter energy usage,
which may in turn affect other traits, such as growth or immune function.
Within an individual’s lifetime, stress exposure can have effects on behavior and
physiology that persist into adulthood. For example, early life stress can increase risk of
aggression, anxiety, and depressive behaviors in adult rodents and primates (reviewed in
McCormick and Green, 2013; Veenema, 2009), can affect adult HPA activity in a variety of
species (Carpenter et al., 2007; Ladd et al., 1996; Spencer et al., 2009), and has been linked to
increased risk of inflammatory, liver, lung, and ischemic heart disease in adult humans (Danese et
al., 2007; Dong et al., 2004; Felitti MD et al., 1998). The effects of early life stress are relatively
well studied in humans and rodents (McCormick and Green, 2013; Veenema, 2009). Expanding
this understanding to other taxa will inform the evolutionary pressures leading to the
consequences of early life stress.
A population’s experience with stress can affect stress physiology in the next generation
(Franklin et al., 2010; Harris and Seckl, 2011; Storm and Lima, 2010; Yehuda et al., 2000). For
6
example, adult human offspring of Holocaust survivors have lower baseline CORT
concentrations compared to their peers (Yehuda et al., 2000). These cross-generational changes
could operate via a number of mechanisms, including selection favoring particular CORT profiles
(Jenkins et al., 2014; Lightman, 2008; Oitzl et al., 2010), maternal effects (e.g. mothers
differentially allocating hormones or nutrients to young or altering maternal behavior;
Champagne and Meaney, 2001; Liu et al., 1997; Love et al., 2013), or epigenetic processes (e.g.
altering gene expression within a lifetime at the level of the HPA axis that may be passed to
offspring; Anacker et al., 2014; Harper, 2005; Jablonka and Raz, 2009; Weaver et al., 2004).
Cross-generational exposure to perturbations following the introduction of novel threats, such as
predatory or competitive invasive species, may also drive changes in other traits that enhance
fitness, such as behavior (e.g. Griffiths et al., 1998; Langkilde, 2009) and morphology (e.g.
Langkilde, 2009; Phillips and Shine, 2004). However, the role of stress in driving these changes
is unknown.
Research Objectives and Study System
My dissertation broadly investigates how characteristics of a stressor (duration,
frequency, intensity) and those of an organism (history of stress exposure) affect the
physiological consequences of stress, including HPA and immune function. I utilized eastern
fence lizards (Sceloporus undulatus) as a study system, as populations of this species exhibit
documented variation in exposure to a known ecological stressor: red imported fire ants
(Solenopsis invicta). Fire ants are invasive across parts of the natural range of fence lizards
(Callcott and Collins, 1996), and these species occupy similar habitat and frequently interact
where their ranges overlap (Freidenfelds et al., 2012; Langkilde, 2009b). Fire ants prey upon
fence lizards, and attacks involve bites and stings that can break the skin of lizards, leaving
7
lizards vulnerable to infection (Elkan and Cooper, 1980; Murphy, 2001). Sub-lethal encounters
elevate lizard plasma CORT concentrations (Langkilde, unpubl. data), and my early research
revealed that lizards from fire ant-invaded sites have higher baseline CORT concentrations and
post-stress CORT reactivity than lizards from uninvaded sites (Graham et al., 2012). This system
thus allows us to compare populations that differ in lifetime and cross-generational histories of
high versus low stress (sites with different histories of fire ant invasion).
I take advantage of elements of both field and lab studies by bringing field-caught
animals from populations with different histories of stress exposure into the lab. Field studies are
more realistic and ecologically relevant than laboratory manipulations, but stress exposure can be
challenging to manipulate in the field if not using long-term implants (which are not always
physiologically relevant). In contrast, short and long duration stress exposure can be easily
manipulated in the lab. By utilizing wild-caught lizards from high- and low-stress sites, I was able
to assess the effects of cross-generational history with stress and the effects of lifetime exposure
to stress by carefully manipulating animals reared in the lab from hatching.
In fence lizards, one can easily manipulate exposure to a variety of stress regimes,
including ecologically relevant fire ant stressors, restraint and handling (a common experimental
stressor), and topical application of exogenous CORT. In some cases, I chose to manipulate
CORT to replicate the CORT elevation that would occur in response to a stressor. Manipulation
of exogenous CORT allows one to target the role of this stress-relevant hormone without
potentially confounding effects of the stressor itself (e.g. wounding, venom). In some cases, I
used fire ant attack as a direct stressor to complement exogenous CORT treatments as it is more
realistic. Utilizing both CORT and fire ant exposure allows us to determine if any effects of fire
ants are due to CORT elevation or to something else, such as higher order HPA activity,
including sensitivity to other stress-relevant hormones (CRF, ACTH), or to other elements of the
attack (e.g. wounding).
8
In my dissertation research, I measured HPA and immune function, both of which are
fitness-relevant and can be rapidly assessed. Many components of the HPA axis could be
measured to assess HPA activity, including concentrations of receptor numbers for CRF, ACTH,
and CORT. Identifying receptor density and location can increase our understanding of the
function of glucocorticoid production, but this can be challenging, particularly in non-model
organisms whose brain structures are not well mapped, and collection of the brain or tissue is
necessary, precluding serial sampling. By contrast, hormone concentrations or metabolites can be
determined using small amounts of blood, saliva, hair, feathers, urine, or feces, permitting non-
lethal and repeated sampling from the same individual. Hair and feathers provide a long-term
record of glucocorticoids due to their slow growth and replacement (Sheriff et al., 2011). Urine
and fecal metabolites reflect a shorter record of glucocorticoids (though metabolism is
complicated and species-specific; Sheriff et al., 2011). Saliva and plasma provide an immediate
snapshot of the circulating glucocorticoid concentrations (Sheriff et al., 2011). Following
extraction, glucocorticoid concentrations or metabolites can be easily measured using
commercially available kits and are repeatable within species (Tarlow and Blumstein, 2007). In
my research, I chose to measure circulating plasma concentrations of CORT due to its rapid
production, repeatability, and known methodological validity for this species (Trompeter and
Langkilde, 2011).
The immune system plays a critical role in maintaining health by protecting an organism
from pathogens and parasites, repairing damage, and responding to infection (Møller and Saino,
2004; Murphy, 2001). Fence lizards have high rates of wounding across their range (McCormick
and Langkilde, 2014), and immune function may be particularly important at sites where risk of
fire ant attack and associated wounding is high. Immune function can be assessed in a variety of
ways. For example, measuring rates of wound healing provides an integrative assessment of
immune function, but is challenging to measure in the field and does not indicate the specific
9
immune mechanism responsible for any changes (Detillion et al., 2004; French et al., 2006;
Martin and Martin, 2014). By contrast, quantifying specific components of the immune system
(e.g. activity of natural antibodies, complement) can be carried out using relatively simple assays
with small amounts of blood, and can be used in the field (Matson et al., 2006, 2005). Lizards
rely primarily on innate rather than adaptive immune function (Zimmerman et al., 2010), and I
therefore utilized two measures of innate immune function that lizards may employ in response to
infection: specifically, the ability of antibodies to agglutinate (clump) foreign cells, providing
fewer targets to be engulfed by phagocytes (Matson et al., 2005; Millet et al., 2007; Sharon,
1998), and of complement to kill bacteria (bacterial killing; Graham et al., 2012; Matson et al.,
2006).
For my dissertation research, I used the above approaches to determine the effects of
stressor characteristics and history of stress exposure on stress physiology and immune function.
In Chapter 2, I investigate whether repeated acute elevations of CORT affect immune function in
lizards, and if these immune outcomes differ in lizards from high- and low-stress sites. In Chapter
3, I experimentally manipulate CORT in fence lizards to examine how stressor duration,
frequency, and intensity affect immune outcomes. In Chapters 4 and 5, I investigate the
consequences of early life stress on adult HPA activity and immune function, respectively, using
lizards from high-and low-stress sites raised under different stress regimes (fire ant exposure and
CORT elevation). Chapters 2, 3 and 5 describe how stressor characteristics and an organism’s
experience with stress affect immune outcomes of stress. Chapters 2 and 4 contribute to our
understanding of naturally occurring patterns of stress in this system and, taken together, allow us
to identify the cross-generational (genetic and maternal effects) and within-lifetime (plasticity)
mechanisms driving the ecology of this system. As a whole, my dissertation sheds light on factors
that influence the physiological outcomes of stress, guiding future research in this field.
10
References
Anacker, C., O’Donnell, K.J., Meaney, M.J., 2014. Early life adversity and the epigenetic
programming of hypothalamic-pituitary-adrenal function. Dialogues Clin Neurosci 16, 321–
33.
Ariza Traslaviña, G. a, de Oliveira, F.L., Franci, C.R., 2014. Early adolescent stress alters
behavior and the HPA axis response in male and female adult rats: the relevance of the nature
and duration of the stressor. Physiol. Behav. 133, 178–89.
Barton, B.A., Schreck, C.B., Barton, L.D., 1986. Effects of chronic Cortisol administration and
daily acute stress on growth, physiological conditions, and stress responses in juvenile
rainbow trout. Dis. Aquat. Organ. 2, 173–185.
Berger, S., Wikelski, M., Romero, L.M., Kalko, E.K. V, Rödl, T., 2007. Behavioral and
physiological adjustments to new predators in an endemic island species, the Galápagos
showed higher hemagglutination of novel proteins by their plasma (a test of constitutive humoral
immunity) than control lizards, a pattern that was consistent across sites. There was no significant
effect of CORT treatment on bacterial killing ability of plasma. These results suggest that
repeated elevations of CORT, which are common in nature, produce immune effects more typical
of those expected at the acute end of the acute-chronic spectrum and provide no evidence of
modulated consequences of elevated CORT in animals from high-stress sites.
Introduction
An organism’s physiological response to stress is generally adaptive, promoting
important behavioral and physiological changes to deal with the stressor (Sapolsky et al., 2000;
Stratakis and Chrousosa, 1995). The vertebrate stress response includes the release of
glucocorticoids, such as cortisol or corticosterone (CORT), by the hypothalamic–pituitary–
22
adrenal (HPA) axis. The production of CORT in response to short-duration (acute) stress, such as
that imposed by a predator encounter (Creel, 2001), may increase survival by triggering important
behavior (including anti-predatory behavior, Remage-Healey et al., 2006; Thaker et al., 2009),
enhancing immune responsiveness in preparation for wounding or subsequent infection that
might result (Deak et al., 1999; Dhabhar, 2009; Martin, 2009), and mobilizing stored energy
(Remage-Healey and Romero, 2001; Sapolsky et al., 2000).
While reallocation of resources to permit these changes is integral to dealing with acute
stressors, long-term reallocation of resources due to chronic stress may reduce fitness, by
suppressing reproduction (Moore and Miller, 1984; Moore and Jessop, 2003; Salvante and
Williams, 2003), growth (Barton et al., 1986; Bourgeon and Raclot, 2006; Davis et al., 1985;
Morici et al., 1997), and immune function (Dhabhar, 2009; Martin, 2009; Sapolsky et al., 2000;
Stratakis and Chrousosa, 1995). Immune effects of long-duration (chronic) stress can include a
reduced ability to heal wounds (Bourgeon and Raclot, 2006; French et al., 2010), suppressed
inflammatory (El-Lethey et al., 2003; Martin II et al., 2005) and antibody response to novel
foreign bodies (El-Lethey et al., 2003; Stier et al., 2009), reduced ability of plasma to kill bacteria
(French et al., 2010), and lymphocyte apoptosis (Sapolsky et al., 2000). There are examples,
however, in which long-term elevations of CORT do not affect immune function (Dabbert et al.,
1997; Klein et al., 1992; Martin II et al., 2005; Vegas et al., 2012). These conflicting findings
could result from differences in energy limitation during the study, as some species exhibit an
immune-suppressive effect of CORT only when resources are limited (food is limited or during
reproduction: French et al., 2010, 2007). Alternatively, it may be maladaptive to suppress
immune function in response to chronic or persistent stress in situations where the immune
system is particularly important for dealing with this stress, such as in response to agonistic social
interactions or predator attacks that can damage an organism or induce behavior that increases
risk of infection (Cox and John-Alder, 2007; Ezenwa et al., 2012). We may expect to see
23
selection against immune compromise (when it exists) and maybe even selection for immune
enhancement within populations chronically exposed to these types of immune-activating
stressors (French et al., 2008; Martin II et al., 2004).
Chronic stress, such as that induced by a storm or drought (Fitze et al., 2009; Wilson and
Wingfield, 1994; Wingfield and Kitaysky, 2002), is often studied by exposing animals to constant
levels of stress (e.g. implanting animals with silastic implants, gel, or pellets that provide a
continuous release of glucocorticoids for days to months; French et al., 2007; Martin II et al.,
2005; Morici et al., 1997). Acute stressors, such as predator attacks or disturbances by humans,
are much shorter in duration, and CORT concentrations following exposure to these stressors
typically return to baseline within a few hours (Malisch et al., 2010), which can be mimicked by
topical application or injection of CORT. However, acute stressors may occur at high frequency.
Such repeated exposure to an acute stressor is sometimes termed “chronic” (Burchfield, 1979;
Harbuz and Lightman, 1992; Romero, 2004), and studies investigating the consequences of
chronic stress often do so by exposing animals to repeated acute stress (Barton et al., 1986; Davis
et al., 1985; Fitze et al., 2009; Retana-Marquez et al., 1998). Repeated acute stress has been
shown to have consequences typical of chronic stress, including suppressed reproduction (Robert
et al., 2009) and growth (Busch et al., 2008a; McCormick et al., 1998; McGraw et al., 2011).
There are mixed outcomes of repeated acute stress on immune function: reduced wound healing
in tree lizards (Urosaurus ornatus; French et al., 2006); suppressed hemagglutination and
inflammatory response to phytohemagglutinin (PHA) in chickens (El-Lethey et al., 2003); and no
change in wound healing in mice (Vegas et al., 2012). Further research examining the immune
consequences of repeated acute stress is necessary to better understand how these ecologically-
relevant repeated acute stressors fit into the existing acute-chronic gradient of stress.
We examined the immune consequences of repeated acute elevations of CORT in eastern
fence lizards, Sceloporus undulatus. Some populations of this species have had their habitat
24
invaded by red imported fire ants, Solenopsis invicta, which prey upon fence lizards (along with
several native predators including birds and snakes; Crowley, 1985). Eastern fence lizards and red
imported fire ants frequently encounter one another in nature (Langkilde, 2009a); lizards exhibit
elevated plasma CORT concentrations following fire ant attack (Langkilde, unpublished data),
and lizards from fire ant invaded sites have higher baseline plasma CORT concentrations than
those from uninvaded sites (Graham et al., 2012a). This provides an excellent opportunity for
understanding whether the immune costs of repeated acute exposure to CORT vary between
animals from high- and low-stress populations.
Methods
Collection and Housing
During April and May 2012, we captured a total of 46 adult male fence lizards (S.
undulatus) from six sites across the southern United States using a hand-held noose. These sites
are similar in habitat (Langkilde, 2009a; Langkilde, unpublished data), but differ in fire ant
invasion history (Callcott and Collins, 1996). Three of these sites have no previous history of fire
ant invasion and lizards at these sites have relatively low baseline CORT concentrations (referred
to as “low-stress” populations): (1) St Francis National Forest, Lee County, Arkansas; (2) Edgar
Evins State Park, DeKalb County, Tennessee; and (3) Standing Stone State Park, Overton
County, Tennessee. The remaining sites were first invaded by fire ants 55-70 years ago and
lizards at these sites have relatively high baseline CORT concentrations (referred to “high-stress”
populations): (4) Blackwater River State Forest, Santa Rosa County, Florida; (5) Geneva State
Forest, Geneva County, Alabama; and (6) Conecuh National Forest, Covington County, Alabama.
25
We measured all lizards for snout-vent length (SVL) upon capture and transported them to our
lab in Pennsylvania, in individual cloth bags inside coolers, for this study.
Lizards were housed individually in plastic enclosures (42 x 28 x 27 cm L x W x H).
Outer walls of the enclosures were wrapped with dark paper to prevent lizards from seeing each
other. The floor of each enclosure was lined with paper towel and furnished with a water dish and
plastic shelter for refuge and basking. The shelter was placed underneath a 60-W incandescent
light bulb that provided heat for 2 h each day. Overhead florescent lights provided additional heat
and were set to a 12:12 light:dark schedule. Each lizard was fed 3 crickets (Acheta domestica)
every other day. All cages were cleaned on the same days to standardize stress (once during the
experimental phase of the study).
Hormone Manipulation and Blood Collection
Lizards were allowed to acclimate for at least 1.5 months (47 - 77 days depending on date
of capture; lizards from fire ant invaded sites were collected approximately 2 weeks earlier than
those from invaded sites) prior to starting the study. This provided adequate time for lizards to
acclimate to laboratory conditions and CORT concentrations to return to baseline (CORT levels
are reduced and remain stable after 2 weeks in captivity; Trompeter and Langkilde, 2011).
Lizards from each site were randomly assigned to a treatment or control group. Those in the
treatment group (“+CORT”) received daily application of CORT (≥92%, Sigma C2505) dissolved
in commercial sesame oil (1 µg CORT / 1 µL sesame oil) and those in the control group received
the sesame oil vehicle only. Every day for 23 days, we applied 6 µL of either CORT solution or
sesame oil only (control) to the backs of lizards using a repeat pipette. Due to the lipophilic nature
of lizard skin, both the oil and hormone/oil mixture were quickly absorbed (Belliure and Clobert,
2004; Meylan et al., 2003; Trompeter and Langkilde, 2011). This CORT application elevates
26
plasma CORT to ecologically relevant concentrations approximate 30 min after application, and
these return to baseline within 4 h after application (Knapp and Moore, 1997; Trompeter and
Langkilde, 2011). We treated lizards in the evening (between 1545 and 1750) because animals
were less active at these hours, negating the need to handle lizards during hormone application
and thus minimizing elevations of CORT caused by this methodology.
Eighteen to 21 h after the final treatment was applied, we collected blood samples from
the post-orbital sinus using 70 µl heparinized microhematocrit tubes (VWR, San Francisco, CA).
This timing allowed us to measure baseline CORT concentrations, as CORT elevations resulting
from CORT application would no longer be evident. All samples for hormone assays were
collected within 2 min of capture (mean 95.7 ± 2.97 SE sec) to prevent handling stress from
influencing plasma corticosterone concentrations in our samples. For a subset of animals (n=28),
additional time was needed to collect blood for immune assays, and this was collected within 3.5
min of capture (mean 194.4 ± 9.0 SE seconds). Blood samples were centrifuged, and plasma was
drawn off and assigned to 3 tubes, one for each assay, and immediately frozen (-20°C) until
assays were performed. Sufficient blood samples were not always available to perform all assays
for each lizard, so sample sizes were reduced in some cases (provided for each assay, below).
Hormone Assays
For 39 individuals, we measured CORT by enzyme immunoassay (Corticosterone High
Sensitivity EIA Kits, Immunodiagnostic Systems Ltd., Fountain Hills, AZ, USA) following
directions provided in the kit. These kits have been validated for S. undulatus (Trompeter and
Langkilde, 2011). Plasma was stored for 49 or 64 days before the assay was performed. We
diluted plasma 1:9 with buffer (5 µl plasma : 45 µl buffer) to ensure that samples fell within the
range of detection of this assay’s standard curve. We ran all samples in duplicate. The mean
27
intraassay coefficient of variation within the two kits was 7.73%, and mean interassay coefficient
of variation between the two kits was 2.79%.
Hemagglutination Assay
For 40 individuals, we measured the ability of plasma to hold sheep red blood cells
(SRBC) in suspension (hemagglutination) in vitro. This assay has previously been used as a
measure of innate immunity in eastern fence lizards (Graham et al., 2012a) as well as birds
(Matson et al., 2005), toads (Graham et al., 2012b), and snakes (Sparkman and Palacios, 2009).
SRBC (Innovative Research, Novi, MI) was washed with phosphate-buffered saline (PBS) up to
three times by diluting 2mL SRBC with 4mL PBS, then gently vortexing and centrifuging for 5
min. We drew off excess PBS and lysed SRBC and repeated the process until the supernatant
became clear. We then brought the washed SRBC to a 2% solution with PBS. After 16 hours in
the freezer, plasma (25 µL) was thawed and diluted 1:1 with PBS (25 µL) and then serially
diluted to 1:64 in a 96-well plate using a multichannel pipette. Control wells contained PBS only
(25 µL). The 2% SRBC solution (25 µL) was then added to each well. Plates were gently mixed
by tapping and incubated at room temperature for 1 h, after which plates were scored for
agglutination (action of antibodies in plasma to hold SRBC in suspension). Scores were
calculated as the negative log2 of the highest dilution at which agglutination was attained – higher
scores indicate a higher concentration of SRBC-specific antibodies in the plasma (Matson et al.,
2005). Half scores were recorded when SRBC precipitated partially but not to the extent of
control wells. Lysis of erythrocytes was not scored since this process has been shown to be
subjective (Matson et al., 2005).
28
Bacterial Killing Ability
For 32 individuals, we measured the ability of plasma to lyse Escherichia coli bacteria.
Heat inactivation experiments completely inhibit the ability of S. undulatus plasma to lyse
bacteria, indicating that complement (e.g., a non-specific protein cascade involved in innate and
adaptive immune lysis) plays a major role in lysis of E. coli in this assay (Graham et al., 2012a).
We use this assay as an additional measure of innate immunity. Detailed methods of this assay
are provided in (Graham et al., 2012a). Briefly, after 4 days in the freezer, thawed lizard plasma
(14 µl) was combined with E. coli (10 µl of 200 CFU bacteria dilution) and allowed to react. This
solution was combined with a growth medium solution (126 µl of a CO2 L-glutamine solution,
containing 400 µl L-glutamine and 19.6 mL CO2 medium). 50µl of each sample was spread in
duplicate on agar plates and incubated at 37°C for 16 h. Colonies on each plate were counted,
averaged (across duplicates) and compared to mean colony counts of two replicated control plates
that contained no plasma (140 µl CO2 L-glutamine medium + 10 µl E. coli). Percent bacterial
killing was calculated as 100 - (mean plasma treatment colony count/mean control colony count)
x 100. Plates that ranged from 0 to negative 10% killing (n=4) were corrected to 0%. Plates with
less than negative 10% killing (n=4) were discarded.
Data Analysis
Plasma CORT concentrations, SVL, and time to obtain blood were log transformed, and
percent lysis data were angular transformed, prior to analysis to meet assumptions of parametric
tests. We compared mean CORT concentrations using ANCOVA with treatment, fire ant invasion
status, and source population (site) nested within invasion status as factors and SVL and time to
obtain blood sample as covariates. We analyzed hemagglutination and bacterial killing ability of
29
plasma separately using ANCOVA with treatment, invasion status, and source population nested
within invasion status as factors and SVL as a covariate. Statistical analyses were performed
using JMP (version 7.0, SAS Institute Inc., Cary NC) with α = 0.05.
Results
Plasma Corticosterone
After treatment, all lizards exhibited plasma CORT concentrations within the
physiological limits of this species (Trompeter and Langkilde, 2011). Lizards repeatedly treated
with CORT had significantly higher plasma CORT concentrations at the end of this study than
control-treated lizards (Fig. 2-1; F1,29=4.44, P=0.04). This relationship was consistent across high-
and low-stress populations (treatment*invasion status: F1,29=0.20, P=0.66), and concentration of
CORT did not significantly differ between lizards from high- and low-stress populations
(invasion status: F1,29=0.15, P=0.70; site within invasion status: F4,29=0.65, P=0.63; SVL:
F1,29=0.52, P=0.48; time to bleed: F1,29=0.28, P=0.60).
Hemagglutination
Hemagglutination scores of lizards repeatedly treated with CORT were significantly
greater than those of control-treated lizards (Figh. 2-2a; F1,31=5.07, P=0.03). This relationship was
consistent across lizards from high- and low-stress populations (treatment*invasion status:
F1,31=0.23, P=0.64), and hemagglutination scores did not significantly differ between lizards from
high- and low-stress populations (invasion status: F1,31=2.32, P=0.14; site within invasion status:
F4,31=0.93, P=0.46; SVL: F1,31=0.01, P=0.92).
30
Bacterial Killing Ability
Lizards repeatedly treated with CORT had similar bacterial killing ability to control-
treated lizards (Fig. 2-2b; F1,19 = 1.56, P = 0.23). This was consistent across lizards from high-
and low-stress populations (treatment*invasion status: F1,19 = 0.00, P = 0.97). Bacterial killing
also did not differ between high- and low-stress populations (invasion status: F1,19 = 0.43, P =
0.52; site within invasion status: F4,19 = 0.38, P = 0.82).
Discussion
Repeated exposure to acute stress is typically considered chronic (Burchfield, 1979;
Harbuz and Lightman, 1992; Romero, 2004) and is thus expected to have “chronic” consequences
including immune suppression. Although each application of CORT should have resulted in only
a short-duration (4 h) elevation in CORT (Trompeter and Langkilde, 2011), baseline CORT
levels of CORT-treated individuals became elevated by the end of the study. However, we found
no evidence that daily application of CORT suppressed immune function of eastern fence lizards.
In fact, our results reveal that repeated acute exposure to CORT produces immune effects more
typical of those expected from stress categorized at the acute end of the spectrum (enhanced;
Dhabhar, 2009; Romero, 2004; Sapolsky et al., 2000): lizards repeatedly treated with acute levels
of CORT exhibited enhanced immune function (hemagglutination) compared to control-treated
lizards. There may be a number of possible reasons for this:
1) We treated lizards at the same time each evening, resulting in elevated plasma CORT
concentrations each evening. Although CORT levels in eastern fence lizards do not vary
considerably during the activity period of this species (1000 - 1600; Trompeter and Langkilde,
2011), quantifying CORT levels outside of this period would be informative for understanding
31
how timing of CORT application may affect our results. If CORT is typically low in the evening,
for example, addition of exogenous CORT at this time may not increase CORT to concentrations
necessary to incite changes typical of the stress response (i.e. concentrations to induce allostatic
overload, McEwen and Wingfield, 2003; Romero, 2004). Although lizards may be more likely to
experience stress during the day, evening elevations of CORT are ecologically relevant, as
eastern fence lizards likely experience stressful fire ant attacks in the evening (Calcaterra et al.,
2008; Porter and Tschinkel, 1987). Studies examining how timing affects immune outcomes
would be valuable.
2) Lizards may have acclimated to these daily elevations in plasma CORT due to the
consistency in timing of application (Harbuz and Lightman, 1992), and some studies randomize
treatment time to prevent this (Boyd, 2007; Busch et al., 2008a, 2008b). Again, however, our
treatment is likely ecologically relevant, as animals commonly experience stressors at predictable
times.
3) Chronic stress is often defined by its long duration (Burchfield, 1979; Romero, 2004;
Wingfield and Kitaysky, 2002). It is possible that we would have detected an immune-
suppressive effect if we continued our CORT treatment for longer than 23 days; however,
physiological effects have been detected following similar durations of CORT application in
common lizards (Lacerta vivipara; Cote et al., 2006) and Gambel’s white-crowned sparrows
(Zonotrichia leucophrys gambelii; Boyd, 2007; Busch et al., 2008a). Further research is necessary
to better understand the frequency or duration necessary for repeated acute stressors to be
interpreted as chronic by an organism.
4) It is also possible that we would have seen immune-suppressive effects of our CORT
treatment under more food-limited conditions (French et al., 2010, 2007). Lizards in our
experiment were well fed and likely not energy limited. Because resources are likely more limited
in nature (body condition (mass/SVL) of field-caught adults = 1.43 +/- 0.03, lab reared adults =
32
1.50 +/- 0.04; P<0.01; TL unpublished data), we might expect to see immune-suppression from
chronic stress in the wild even when we do not see it in the lab. In some cases, however, immune
suppression has been observed even when resources were not limited (repeated acute stress:
Barton et al., 1986; Boyd, 2007; El-Lethey et al., 2003; chronic stress: Morici et al., 1997) and so
this may not have been a constraining factor in this study.
5) Some components of the immune system may trade-off differentially with the stress
response based on the costs and benefits of suppressing that component (Kurtz et al., 2000; Stier
et al., 2009). This may explain why differential effects of CORT on various immune components
have been observed within the same study (this study; Bourgeon and Raclot, 2006; Graham et al.,
2012b; Ilmonen et al., 2003; Stier et al., 2009) and could contribute to the varied effects of CORT
on immune function described in literature. We would benefit from a better understanding of
which components of the immune system trade off with other systems under stress.
Exposure of fence lizard populations to fire ants has driven evolutionary changes in
behavior and morphology within seven decades (35 lizard generations; Langkilde, 2009b;
Robbins, unpubl. data). Plasma CORT concentrations at the end of our study did not, however,
differ between lizards from high- and low-stress populations, suggesting lifetime and/or
evolutionary exposure to stress may not affect an animal’s response to exogenous CORT. Lizards
from high-stress (fire ant invaded) and low-stress (uninvaded) populations return to similar
baseline CORT concentrations in the lab (this study; Trompeter and Langkilde, 2011), despite
lizards from high-stress populations having higher baseline CORT concentrations than lizards
from low-stress populations in the field (Graham et al., 2012a). This supports the notion that
differences in CORT concentrations between invaded and uninvaded populations in the field are a
direct result of exposure to environmental stressors (likely attack by fire ants) rather than a result
of genetic or acquired sensitivity to CORT.
33
Similarly, immune function did not differ between high- and low-stress populations in
this study. The lack of immune-suppressive effects of CORT in high-stress fire ant invaded sites
may not be surprising given that the immune system may be important for dealing with these
predatory ants. Immune function may be important for these lizards in general - they exhibit high
rates of tail autotomy and other wounding (30% (100/337) of individuals surveyed; T.L. unpubl.
data). To maintain a high-functioning immune system, it is possible that less immediately
important systems, such as growth or reproduction, may be traded-off against CORT in this and
possibly other wild species. Additionally, the risk of injury from encounters with fire ants can be
high in this species—fire ants puncture the skin when they sting lizards—and suppressing
immune function when stressed would likely be maladaptive for this species. We may, in fact,
expect lizards from fire ant invaded populations to up-regulate immune function, but we have
found no evidence of this in the field (Graham et al., 2012a). Further research on a range of
organisms that have evolved under different stressors is necessary to assess the contexts under
which costs of CORT are incurred and the nature of these costs.
Acknowledgements
We thank T. Robbins, S. Graham, and J. Newman for help with lizard collection, S.
Graham, G. DeWitt, and S. McGinley for assistance with immune assays, the Cavener lab for use
of their plate reader, and B. Chitterling for valuable comments on this manuscript. We thank the
Landsdale family for access to their land and lizards and personnel at St Francis National Forest,
Edgar Evins State Park, Standing Stone State Park, Blackwater River State Forest, Geneva State
Forest, Conecuh National Forest, and especially the Solon Dixon Forestry Education Center for
logistical support. The research presented in this article adheres to the Guidelines for the Use of
Animals in Research and the Institutional Guidelines of Penn State University, and animal
34
collection was permitted by the respective states. Funding was provided in part by the American
Society of Ichthyologists and Herpetologists (Gaige Award to GLM) and the National Science
Foundation (DGE1255832 to GLM and IOS1051367 to TL).
35
Figures
Figure 2-1: Plasma CORT concentrations of lizards from high-stress (fire ant invaded) and low-stress (uninvaded) populations after 23 days of treatment with CORT (shaded bars) or control (oil vehicle only; open bars). CORT-treated lizards had significantly higher plasma CORT concentrations than control lizards. This relationship was consistent across high- and low-stress populations (i.e. no significant effect of invasion status). Bars represent means ± one standard error. Sample size for each group is given within each bar.
36
Figure 2-2: A) Hemagglutination scores and B) percent bacterial killing by plasma of lizards from high-stress (fire ant invaded) and low-stress (uninvaded) populations after 23 days of treatment with CORT (shaded bars) or control (oil vehicle only; open bars). CORT-treated lizards had significantly higher hemagglutination scores than but similar bacterial killing ability to control lizards. These relationships were consistent across high- and low-stress populations (i.e. no significant effect of invasion status). Bars represent means ± one standard error. Sample size for each group is given within each bar.
37
References
Barton, B.A., Schreck, C.B., Barton, L.D., 1986. Effects of chronic Cortisol administration and
daily acute stress on growth, physiological conditions, and stress responses in juvenile
study may have constituted moderate (LA, immune enhancement) and intense (HA, immune
suppression) levels of stress. Only a study that systematically examines outcomes for all possible
combinations of frequency, intensity and duration of stress can comprehensively delineate the
complete stress response surface (Miller, Roxburgh, & Shea, 2011). And these stress aspects will,
of course, need to be scaled to the relevant traits of the organism of interest (e.g. life cycle, HPA
function) in order to make them generally applicable (Roxburgh et al., 2004).
59
Conclusions
Our results reveal that average, total, or other single measures of stress (CORT) do not
satisfactorily encompass either how stress is experienced (aspects such as frequency, intensity,
and duration) or the physiological outcomes (immune function). Additionally, categorizing stress
as acute or chronic – by duration alone – and the lack of consistency in use of these terms may be
hindering progress in this field. Our results suggest that additional underlying aspects of stress,
such as stressor intensity and frequency, can affect the outcomes of stress and should also be
considered and reported if we are to adequately describe and assess the ecological and human
health consequences of stress.
Acknowledgements
We thank S. Graham, C. Thawley, M. Goldy-Brown, and M. Herr for help with lizard
collection, C. Thawley for assistance with blood collection, C. Norjen and D. McGregor for
assistance with immune assays, C. Thawley, C. Norjen, M. Goldy-Brown, and S. McGinley for
lizard care, the Cavener lab for use of their plate reader, and B. Chitterlings for valuable
comments on this manuscript. We thank personnel at Standing Stone State Park, Fall Creek Falls
State Park, Holly Spring National Forest, Blakeley State Park, Conecuh National Forest, and
especially the Solon Dixon Forestry Education Center for logistical support. The research
presented in this article adheres to the Guidelines for the Use of Animals in Research and the
Institutional Guidelines of Penn State University (IACUC #35780), and animal collection was
permitted by the respective states. Funding was provided in part by the National Science
Foundation (DGE1255832 to GLM, DEB0815373 to KS, and IOS1051367 to TL).
60
Tables
Table 3-1: This table shows the different immune outcomes (enhancement, no change, or suppression) to repeated exposure to stress
(exogenous application of CORT, handling stress, or an ecologically-relevant stressor), indicating the study organism and immune component measured. Exogenous CORT was elevated using topical application or feeding. Handling includes handling or chasing, placement in a bag, or air exposure (fish). Ecologically relevant stressors include food deprivation, social isolation, social defeat, or exposure to predator scents. *Animals were simultaneously restrained. Abbreviations as follows: Hemag. = hemagglutination; BKA = bacterial killing ability; DTH = delayed-type hyper sensitivity; Antibody Resp. = antibody response.
61
Figures
Figure 3-1: A) The frequency, intensity, and duration of CORT application in each of the treatments used in this study, and the total amount of CORT received in each 3-day period. Text in parentheses indicates: for Frequency, how frequently a CORT-oil solution was applied (oil-vehicle only was applied on remaining days); for Intensity, the amount of CORT applied during each application; and for Duration, whether the period of CORT elevation was short or long. Italicized pairs in each column represents treatments that differ in only the parameter shown in that column. B) A graphical representation of the amount of CORT applied for each of the treatments used in this study (Control (Ctl) had no CORT applied). This is provided for illustrative purposes, to convey the expected duration of CORT release following application.
62
Figure 3-2: A) Hemagglutination scores and B) Percent bacterial killing by lizard plasma after 9 days of treatment (see Fig. 3-1). Lizards in the Low Acute (LA), Repeated Acute (RA), and Control (Ctl) treatments had significantly higher hemagglutination scores than did those in the High Acute (HA) treatment, and those in the Chronic (Ch) treatment had hemagglutination scores that were intermediate to these groups. Lines above the columns connect treatments that do not significantly differ from one another. Bacterial killing ability did not significantly differ across treatments. Error bars represent means ± one standard error. The sample size for each group is given within each bar.
63
References
Altemus, M., Dhabhar, F.S., Yang, R., 2006. Immune function in PTSD. Ann. N. Y. Acad. Sci.
1071, 167–83.
Barton, B.A., Schreck, C.B., Barton, L.D., 1986. Effects of chronic Cortisol administration and
daily acute stress on growth, physiological conditions, and stress responses in juvenile
rainbow trout. Dis. Aquat. Organ. 2, 173–185.
Belliure, J., Clobert, J., 2004. Behavioral sensitivity to corticosterone in juveniles of the wall
CRH) (e.g. Carpenter et al., 2007; Heim et al., 2000; Ladd et al., 1996; Plotsky and Meaney,
1993). Suppression of the stress response could reflect habituation to a specific stressor (Grissom
and Bhatnagar, 2009; Romero and Reed, 2005; Romero et al., 2009) or general down-regulation
of the response (Romero, 2004). This suppression may protect the organism from stress-related
costs (e.g. diversion of energy away from non-critical functions such as growth, reproduction, and
immune function; Chrousos and Gold, 1992; Greenberg and Wingfield, 1987; Martin, 2009) but
could reduce benefits associated with heightened stress reactivity (e.g. supporting behavioral and
71
metabolic responses to stressors; Sapolsky, 2000; Thaker et al., 2009). Alternatively, up-
regulation of the stress response may allow an organism to more effectively mount responses to
current threats (Romero, 2004) while potentially incurring longer-term costs of elevated
corticosterone (Korte et al., 2005; Romero et al., 2009; Sapolsky et al., 2000). Both up- and
down-regulation of the adult stress response has been observed in response to early life stress
(Ariza Traslaviña et al., 2014; Carpenter et al., 2007; Caruso et al., 2014; Spencer et al., 2009)
and in some cases adult baseline CORT is not altered (e.g. Mirescu et al., 2004; Plotsky and
Meaney, 1993). The effects of early life stress are relatively well studied in humans and rodents
(McCormick and Green, 2013; Veenema, 2009); expanding this to other organisms with
documented differences in ancestral stress exposure will inform the evolutionary pressures
leading to the consequences of both early life and cross-generational stress.
Physiological stress has been documented in response to a variety of environmental
perturbations including habitat loss (Homan et al., 2003; Suorsa et al., 2004), urbanization
(French et al., 2008), and the introduction of invasive species (Berger et al., 2007; Graham et al.,
2012). We took advantage of naturally occurring high- and low-stress populations of native
lizards associated with cross-generational presence or absence, respectively, of predatory invasive
fire ants (Graham et al., 2012). We manipulated early life stress of offspring from these
populations and measured aspects of their stress physiology in adulthood. This allowed us to
experimentally investigate whether adult HPA activity is affected by early life and cross-
generational exposure to stress and whether these stress histories interact.
72
Methods
Study System and Animal Collection
Red imported fire ants (Solenopsis invicta) are invasive predators of eastern fence lizards
(Sceloporus undulatus) (Langkilde, 2009a), and these species utilize similar habitat where their
ranges overlap (Langkilde, 2009b). Fire ants frequently attack fence lizards in nature
(Freidenfelds et al., 2012). These encounters induce anti-predator behavior (Langkilde, 2009a)
and trigger the release of CORT in lizards (Trompeter and Langkilde, 2011). Lizards in areas
invaded by fire ants have higher baseline concentrations of CORT than do those from uninvaded
sites (Graham et al., 2012).
During April and May 2012, we captured gravid female eastern fence lizards (n = 86)
from 6 sites across the southeastern United States: (1) Blackwater River State Forest, Santa Rosa
County, Florida; (2) Geneva State Forest, Geneva County, Alabama; (3) Conecuh National
Forest, Covington County, Alabama; (4) St Francis National Forest, Lee County, Arkansas; (5)
Edgar Evins State Park, DeKalb County, Tennessee; and (6) Standing Stone State Park, Overton
County, Tennessee. All sites have similar habitat (Langkilde, 2009, unpubl. data) but differ in fire
ant invasion history (Callcott and Collins, 1996): Sites 1, 2, and 3 were first invaded by fire ants
57 to 75 years ago (“invaded”), while sites 4, 5, and 6 have no previous history of fire ant
invasion (“uninvaded”).
Animal Husbandry
Gravid lizards were housed in pairs in plastic enclosures (56 x 40 x 30 cm, L x W x H)
furnished with a shelter for refuge and basking, a water bowl, and moist sand for nesting.
Overhead lights were set to a 12:12 hour light:dark schedule (light: 0800 – 2000 hours), and a 60-
73
W incandescent light bulb was placed at one end of the enclosure to provide heat for 6 hours each
day to allow lizards to thermoregulate. Lizards were fed crickets (Acheta domestica) dusted with
calcium and vitamin supplements every second day, and water was available ad libitum.
We checked enclosures at least twice daily for eggs and immediately placed clutches in
plastic containers (11 x 11 x 7.5 cm) filled with moist vermiculite (-200 kpa), covered with plastic
wrap, and sealed with a rubber band (Langkilde and Freidenfelds, 2010). We placed containers in
an incubator (29o C ± 1oC) until eggs hatched (approximately 45 days), rotating the containers
every other day to avoid any within-incubator effects of position. We checked incubators twice
daily for hatchlings.
We toe-clipped hatchlings for unique identification and housed them in groups of six
based on age. Each enclosure contained two lizards from each of the three treatments (described
under Treatments) and no more than two lizards from each clutch, each from different treatments.
Lizards from fire ant-invaded sites never shared enclosures with those from uninvaded sites.
Hatchlings were housed under similar conditions as gravid females but without sand; the floor of
each enclosure was instead lined with paper towel. At 42 weeks of age, we measured all lizards
for mass and snout-vent length (SVL).
Treatments
To determine the effects of exposure to CORT or an ecologically-relevant stressor during
early life, hatchlings were assigned to one of 3 treatments using a split-clutch design. Starting at 2
weeks of age, lizards were exposed to topical application of CORT (CORT), fire ants (FA), or a
handling and oil-vehicle control (Ctl) once a week for 42 weeks, at which time lizards had
reached maturity. This regimen was selected to be ecologically-relevant, while avoiding
potentially lethal effects of frequent exposure to fire ant venom (Freidenfelds et al., 2012). To
74
ensure all lizards received the same handling and topical application, each week all lizards were
individually placed in a sand-lined arena (with or without fire ants; 9 x 22 cm R x H) for 30
seconds, after which 3 µl sesame oil (with or without CORT) was applied to their backs with a
pipette. Lizards were returned to their home enclosure after treatment.
Lizards in the fire ant treatment were placed inside the testing arena with 15 - 20 fire ants.
Ants were allowed to encounter and sting the lizard, as they do in nature. A trial ended 30
seconds after the first ant contacted the lizard, and any attacking fire ants were removed from the
lizard. This provided a non-lethal exposure that induces CORT elevation (Trompeter and
Langkilde, 2011). They then had sesame oil applied to their backs.
Lizards in the CORT treatment topically received CORT (≥92%, Sigma C2505, Saint
Louis MO) dissolved in commercial sesame oil, after being removed from the sand-lined testing
arena. The oil and hormone were quickly absorbed due to the lipophilic nature of lizard skin
(Belliure and Clobert, 2004) and resulted in physiologically-relevant increases in plasma CORT
concentrations that simulated CORT responses to fire ant exposure (Knapp and Moore, 1997;
Trompeter and Langkilde, 2011). CORT doses were calculated based on the average growth of
this species in the laboratory to avoid stress associated with measuring size each week
(Freidenfelds et al., 2012) and ranged from 0.6 to 1 µg CORT/g body mass.
Lizards in the control treatment were placed in a sand-lined testing arena for 30 seconds
and then had sesame oil applied to their backs.
Blood Collection and Stress Assays
We measured baseline and stress reactive concentrations of CORT following restraint for
a subset of lizards (n=32) on a single day eight to 14 weeks after treatments ended. Baseline
blood samples were obtained from the post-orbital sinus using 70 µl heparinized microhematocrit
75
tubes (VWR, San Francisco CA) within 3 minutes of capture from their home enclosures
(Romero and Reed, 2005). We then individually placed lizards in cloth bags for 30 minutes, after
which we collected a second blood sample to assess CORT reactivity to this standardized restraint
stressor (Romero and Reed, 2005).
We measured the CORT response to an ecologically-relevant stressor, attack by fire ants,
for a separate subset of lizards (n=31) on a single day eight to 16 weeks after treatments ended.
Lizards were placed inside a sand-lined arena containing 15-20 fire ants for 60 seconds after the
first ant attacked. The number of ants attacking lizards during these trials was similar across early
life treatments (ANOVA: F1,30=1.068, p=0.358) and invasion statuses of the source population
(F1,30=0.041, p=0.841). A separate subset of lizards (n=30) was placed in a sand-lined arena with
no ants for 70 seconds to serve as a control. The difference in duration of these fire ant versus
control trials (60 vs. 70 seconds) reflects the average time it took for fire ants to attack after
placing a lizard in the arena (Robbins, unpubl. data). Lizards in both the fire ant-exposure and
control groups were immediately returned to their home enclosures for 60 minutes, after which
we obtained a blood sample as described above. All blood samples were maintained on ice during
collection, then centrifuged, and plasma drawn off and immediately frozen (-20°C) until assays
were performed.
ACTH Challenge
To determine the ability of lizards' adrenal glands to mount a CORT response, we
conducted an ACTH challenge on a separate subset of lizards (n=65) over two days 11 to 18
weeks after treatments ended. Lizards were injected intraperitoneally with 70 or 100 µl of either
adrenocorticotrophic hormone (ACTH), a pituitary hormone that stimulates the adrenal glands to
excrete CORT in lizards, (n=34) or saline solution as a control (n=31). ACTH (Sigma A6303,
76
Saint Louis MO) was dissolved in saline prior to injection with doses ranging from 0.56-0.80 IU
(Klukowski 2011). Dosages did not significantly affect CORT concentrations (see Data
Analysis). After injection, all lizards were placed in individual cloth bags for 60 minutes, after
which we obtained a blood sample. Blood samples were processed using the previously described
methods.
Hormone Analysis
We measured plasma CORT concentrations using Corticosterone High Sensitivity EIA
Kits (Immunodiagnostic Systems Ltd., Fountain Hills, AZ, USA) following directions provided
in the kit. These kits have been validated for Eastern fence lizards (Trompeter and Langkilde,
2011). We diluted plasma 1:9 with buffer (5 µl plasma : 45 µl buffer) to ensure that samples fell
within the range of detection of the assay’s standard curve. We ran all samples in duplicate. The
mean intrassay coefficient of variation within the six kits was 2.35% (1.53% to 2.91%), and the
mean interassay coefficient of variation was 5.11%.
Data Analysis
CORT concentrations at baseline, following ACTH- and saline-injection, and following
exposure to fire ants or the associated control were log transformed prior to analysis to meet
assumptions of parametric tests. One data point was omitted from analysis of baseline CORT and
one from the analysis of CORT reactivity to restraint, as their values were greater than 2 standard
deviations from the mean.
For restraint stress, we calculated CORT reactivity as the baseline CORT value
subtracted from the CORT value after the 30-minute stressor. Because baseline CORT
77
concentrations were not taken for lizards before exposure to fire ants or the control arena or
before ACTH- or saline-injection, we analyzed the CORT response to these stressors (post-
stressor concentrations). We analyzed baseline CORT, CORT reactivity to restraint, and CORT
response to fire ant exposure, ACTH-, or saline-injection separately using mixed-model
ANCOVA with early life treatment, fire ant invasion status, source population (nested within
invasion status), and sex as factors, maternal ID as a random effect, and snout-vent length and age
as covariates. Time to bleed was included as a covariate in the model for baseline CORT, baseline
CORT was included as a covariate in the model for CORT reactivity to restraint, and fire ant
exposure assay (FA vs. control exposure) was included as a factor in the model of CORT
response to fire ant exposure.
Site, sex, and age did not significantly explain variation in any of the CORT
concentration data (p>0.100); time to bleed and SVL did not significantly explain variation in
baseline CORT concentrations (p>0.146); SVL did not significantly explain variation in CORT
reactivity to restraint (p=0.503); the amount of ACTH or saline injected did not significantly
explain variation in the CORT response (p>0.831); and age and SVL did not significantly explain
variation in CORT responses to injection (p>0.203). These variables were thus omitted from the
respective final models. In cases where interactions were non-significant, they were removed
from the final model. All statistical analyses were performed using JMP (version 12.1, SAS
Institute Inc., Cary NC) with α = 0.05.
78
Results
Baseline Corticosterone
Neither early life treatment (Fig. 4-1; F2,20=1.535, p=0.240) nor invasion status of the
source population affected baseline CORT concentrations of adult lizards (F1,6=2.094, p=0.196;
early life treatment x invasion status F2,14=0.713, p=0.508).
Corticosterone Reactivity to Restraint
The change in CORT following restraint (CORT reactivity) was not significantly affected
by early life treatment (Fig. 4-1; F2,10=0.293, p=0.752). However, CORT reactivity was
significantly greater in offspring of lizards from high-stress fire ant-invaded populations than in
those from low-stress uninvaded populations (F1,9=5.239, p=0.048; baseline F1,15=2.454,
p=0.140). The effect of early life treatment on CORT reactivity did not differ with invasion status
(early life treatment x invasion status F2,10=0.031, p=0.970).
Corticosterone Response to Fire Ants
CORT concentrations were significantly greater in lizards following exposure to fire ants
in adulthood compared to those that were control-handled (Fig 4-2; F1,33=12.954, p=0.010). This
result was not affected by the invasion status of the source population or early life treatment
(invasion status x fire ant exposure assay F1,25=0.078, p=0.782; early life treatment x fire ant
exposure assay F2,34=1.831, p=0.176; early life treatment x invasion status x fire ant exposure
assay F2,34=0.546, p=0.584). Among lizards exposed to fire ants in adulthood, offspring of lizards
from high-stress fire ant-invaded populations had greater CORT concentrations than did those
79
from low-stress uninvaded populations (F1,22=4.825, p=0.039; early life treatment F2,51=0.937,
p=0.398; SVL F1,43=12.222, p=0.001). This may have been driven by the fact that CORT-treated
offspring of lizards from high-stress populations had elevated CORT concentrations following
fire ant exposure relative to controls, whereas this was not observed in offspring from uninvaded
source populations; a trend which approached significance (early life treatment x invasion status
F2,50=2.788, p=0.070).
Corticosterone Response to ACTH Challenge
Injection of saline had a similar effect on CORT concentrations of lizards regardless of
their early life treatment (Fig. 4-3; F2,13=0.246 p=0.786) or the invasion status of the source
population (F1,8=0.814, p=0.394). These factors did not interact to explain the CORT
concentrations following injection with saline solution (early life treatment x invasion status
F2,9=0.182, p=0.837).
CORT concentrations following ACTH-injection were not affected by early life treatment
(Fig. 4-3; F2,22=2.389, p=0.115) but were significantly greater in offspring from high-stress fire
ant-invaded source populations than in those from low-stress uninvaded populations (F1,17=7.604,
p=0.013). This result was not affected by early life treatment (invasion status x early life
treatment F2,17=1.405, p=0.272).
Discussion
We investigated the effects of early life and cross-generational exposure to stress, and the
interaction of these exposure histories, on adult HPA activity. We found no effect of early life
stress (weekly exposure to CORT or fire ant attack) on adult baseline concentrations of CORT,
80
CORT reactivity to restraint, or the CORT response to fire ant exposure or ACTH injection.
Offspring of lizards from high-stress fire ant-invaded and low-stress uninvaded populations had
similar baseline concentrations of CORT. However, cross-generational exposure to stress did
influence our measures of HPA reactivity: offspring of lizards from high-stress invaded
populations had greater CORT responses to restraint, fire ant exposure, and ACTH injection than
did offspring of lizards from low-stress uninvaded populations. These results suggest cross-
generational history with stress has important effects on adult HPA activity, while early life stress
may play a lesser role in this system.
Early Life Stress
We did not observe an effect of early life exposure to CORT or fire ants on HPA activity.
Early life stress does not affect adult baseline concentrations of CORT in rats (Mirescu et al.,
2004; Plotsky and Meaney, 1993); however, adult CORT reactivity is altered by early life stress
in rats and humans (Carpenter et al., 2007; Gunnar et al., 2009; Liu et al., 1997; Matthews, 2002)
and there is some evidence of this in birds (Spencer et al., 2009). We also did not observe any
effect of early life stress on CORT following ACTH injection, suggesting that this stress exposure
had no effect on adrenal function. It is important to note that we measured HPA activity several
months after treatments had ended. It is thus possible that early life stress exposure affected HPA
activity during or immediately following the treatment period, but any potential effects did not
persist two months beyond the completion of treatments. This is counter to the literature
documenting persistent effects of early life stress on adult behavior and physiology (McCormick
and Green, 2013; Meaney et al., 1994; Veenema, 2009). Lizards may have habituated to the
regular exposure to fire ants in the laboratory environment (Cyr and Romero, 2009; Romero,
2004; Romero et al., 2009) or the fire ant colony may have become less venomous in captivity
81
(Tschinkel, 2006; Xian-Fu et al., 2015), although this does not explain why no effects of early life
CORT treatment were observed. Lizards exposed to fire ants in adulthood had greater CORT
concentrations compare to those exposed to a sand-lined enclosure, suggesting that our fire ants
still induced a CORT response. Future research on whether more frequent, intense, or varied
duration of early life stress treatments produce lasting effects into adulthood would be
informative (Busch et al., 2008; McCormick et al., 1998; McCormick et al., 2015; McEwen et al.,
1997). The absence of maternal care in this species may also play a role, as maternal care may
exacerbate the effects of early life stress in mammals (Champagne and Meaney, 2001; Meaney,
2001).
Cross-Generational Exposure to Stress
Cross-generational exposure to stress predicted adult HPA activity in the current study.
Offspring of lizards from high-stress fire ant-invaded populations had greater CORT reactivity to
restraint and higher CORT concentrations following fire ant exposure than did offspring from
low-stress uninvaded populations, irrespective of early life stress treatment. These results mirror
the differences in adult HPA activity of these populations in the wild (Graham et al., 2012).
Because these patterns were observed even in lizards from the control treatment, the current study
indicates that field patterns of elevated CORT responsiveness in fire ant-invaded populations
(Graham et al., 2012) are not driven by within-lifetime stress, but rather by cross-generational
mechanisms. This could take two forms:
a) Mothers from high-stress invaded populations may differentially provision their eggs
(with nutrients, CORT, etc.; Hayward and Wingfield, 2004; Seckl and Meaney, 2004) or
alter their behavior (e.g. feeding, thermoregulation, maternal care; Champagne and
Meaney, 2001; Shine and Harlow, 1993). These changes may affect the perinatal
82
environment and lead to epigenetic changes (Fish et al., 2006; Weaver et al., 2004),
which can affect the stress responsiveness of offspring (maternal effects; Champagne and
Meaney, 2001; Liu et al., 1997; Love et al., 2013, 2008). These changes may prepare
offspring for a stressful environment (maternal matching: Sheriff and Love, 2013). Prior
research in this species argues against a role for increased yolk CORT in explaining
elevated CORT reactivity, as 3 month-old hatchlings from eggs treated with CORT had
lower, not higher, CORT reactivity to restraint than those from eggs treated with oil
vehicle only (Norjen, unpubl. data). Future research should investigate how maternal
mechanisms may affect adult HPA activity.
b) At high-stress fire ant-invaded sites, selection may favor heightened CORT
responsiveness, as elevated concentrations of CORT trigger important survival behaviors
that allow these lizards to escape attack from fire ants (Trompeter and Langkilde, 2011;
Langkilde, unpubl. data). The adult CORT response to fire ant exposure appeared to be
heightened by early life exposure to CORT in offspring from fire ant-invaded sites (a
trend which approached statistical significance; Fig. 2). Cross-generational history with
stress may thus select for greater sensitivity to early life stressors. This combination of
genetic and early life environmental factors affecting adult stress reactivity has been
documented in other systems (Gariépy et al., 2002; Jenkins et al., 2014). Further research
on how these stress histories interact in natural populations will increase our
understanding of the ecological and evolutionary significance of these patterns.
The greater CORT reactivity and CORT response to fire ant exposure in offspring of
lizards from high-stress fire ant-invaded populations is mirrored in our ACTH results: offspring
of lizards from high-stress sites had greater CORT responses to ACTH compared to those from
low-stress uninvaded sites. This indicates a general up-regulation of CORT reactivity in offspring
83
from high-stress populations, with differences occurring at the adrenal glands rather than at the
level of the hypothalamus or the pituitary.
Baseline CORT
Neither early life stress nor cross-generational history with stress affected baseline
CORT. Baseline CORT varies greatly within individuals as seasons and metabolic demands
change (Bonier et al., 2009; Landys et al., 2006). Because this flexibility occurs on short time
scales, variation in baseline CORT concentrations may be best explained by environmental rather
than genetic factors (Jenkins et al., 2014). The results of this study suggest that field patterns of
lizards having higher baseline CORT at fire ant-invaded than at uninvaded sites (Graham et al.,
2012) are likely not due to early life stress or cross-generational history with stress. Instead, these
field patterns may reflect CORT responses to recent and frequent (on the order of minutes) fire
ant attack (Freidenfelds et al., 2012; Langkilde unpubl. data) rather than true baseline CORT.
This is supported by the fact that lizards from both fire ant-invaded and uninvaded sites return to
similar baseline CORT concentrations within one week in captivity (Langkilde, unpubl. data).
Alternatively, these field differences may reflect increases in true baseline CORT concentrations
as a result of persistent increases in CORT due to frequent attack by fire ants over long time
periods (McCormick and Langkilde, 2014).
Conclusions
This study demonstrates that measuring early life stress alone may not adequately capture
the population level drivers of changes in adult physiology, which may be affected by both cross-
generational and early life exposure to stress. The exposure of an individual’s ancestors to stress
84
may be more important in regulating physiological stress responses than the individual’s own
lifetime stress experiences. This has important consequences for predicting and managing the
effects of stress and for establishing whether these effects can be reversed within a lifetime.
Determining the mechanisms that drive changes in the physiological stress response would aid
predictions of how species will respond to environmental perturbations.
Acknowledgements
We thank S. Graham and C. Thawley for assistance with planning, S. Graham, C.
Thawley, and J. Newman for help with lizard collection, C. Thawley, S. McGinley and E. Baron
for assistance with adult lizard trials, C. Thawley for assistance with blood collection, C.
Thawley, G. Dewitt, D. Fricken, M. Goldy-Brown, A. Hollowell, L. Horne, M. Hook, A. Jacobs,
C. Norjen, S. McGinley, and M. O’Brien for lizard care and assistance with early life lizard
treatments, the Cavener lab for use of their plate reader, and B. Chitterlings for valuable
comments on this manuscript. We thank the Landsdale family for access to their land and lizards
and personnel at St. Francis National Forest, Edgar Evins State Park, Standing Stone State Park,
Blakeley State Park, Blackwater River State Forest, Geneva State Forest, Conecuh National
Forest, and especially the Solon Dixon Forestry Education Center for logistical support. All
methods detailed here adhere to the Guidelines for the Use of Animals in Research and the
Institutional Guidelines of Penn State University (IACUC #35780), and animal collection was
permitted by the respective states. Funding was provided in part by the National Science
Foundation (DGE1255832 to GLM and IOS1051367 to TL and SAC).
85
Figures
Figure 4-1: CORT reactivity to restraint is greater in offspring of lizards from fire ant-invaded sites. Adult concentrations of CORT at baseline (shaded bars) and following restraint in a bag (white bars) of lizards exposed weekly to fire ants (FA), exogenous CORT, or control treatment during early life. CORT reactivity (post-restraint stressor minus baseline) was assessed in the statistical model but stress-induced concentrations are plotted here for ease of comparisons between graphs. Bars represent means ± one standard error. The sample size for each group is given above each bar.
86
Figure 4-2: CORT concentrations following adult fire ant exposure are greater in offspring of lizards from fire ant-invaded sites. CORT concentrations following exposure in adulthood to an empty arena (FA control; shaded bars) or attack by fire ants (FA; white bars) of lizards exposed weekly to fire ants (FA), exogenous CORT, or control treatment during early life. Bars represent means ± one standard error. The sample size for each group is given above each bar.
87
Figure 4-3: ACTH-induced CORT concentrations are greater in offspring of lizards from fire ant-invaded sites. Adult CORT concentrations following injection with saline solution (shaded bars) or ACTH (white bars) of lizards exposed weekly to fire ants (FA), exogenous CORT, or control treatment during early life. Bars represent means ± one standard error. The sample size for each group is given above each bar.
88
References
Belliure, J., Clobert, J., 2004. Behavioral sensitivity to corticosterone in juveniles of the wall
Adult percent killing of E. coli by plasma was not related to early life stress exposure
(Fig. 2; F2,19=0.014, p=0.986), invasion status of the source population (F1,15=1.875, p=0.192), or
an interaction of the two (F2,19=2.045, p=0.157).
Discussion
In offspring of lizards from fire ant-uninvaded sites, early life CORT (but not fire ant)
exposure suppressed adult baseline hemagglutination compared to controls, but in offspring from
invaded sites, early life CORT exposure enhanced baseline hemagglutination. Neither early life
nor cross-generational exposure to stress affected adult bacterial killing ability or post-inoculation
hemagglutination scores.
Suppression of adult baseline hemagglutination by early life CORT exposure in lab-
reared offspring of lizards from low-stress fire ant-uninvaded sites is consistent with the
suppressive effects of long-term and early life stress on innate immune function of rodents, birds,
and humans (Michaut et al. 1981; Avitsur et al. 2006; De Coster et al. 2011; Kriengwatana et al.
2013; Schmidt et al. 2015). Because development of the immune system is energetically costly
(Sheldon and Verhulst 1996; Lochmiller and Deerenberg 2000; Norris and Evans 2000; Klasing
2004), immune function development may not be prioritized if energy is limited as a result of
early life stress. The opposite pattern, however, was observed in lab-reared offspring of lizards
from high-stress fire ant-invaded sites; when exposed to CORT during early life, these offspring
had higher baseline hemagglutination scores in adulthood compared to controls. Enhancing
immune function in response to early life stress may be beneficial for lizards from these
106
populations. Fire ants are the predominant stressor facing lizards in these populations (Langkilde
2009a; Graham et al. 2012), and fire ant attacks are frequent (on the order of every few minutes;
Freidenfelds et al., 2012). The bites and stings of fire ants break the skin, increasing risk of
infection (Elkan and Cooper 1980; Murphy 2001), and lizards from fire ant-invaded sites have
higher rates of tail autonomy and other wounding (Thawley unpubl. data), possibly due to
behavioral responses of lizards to fire ants attracting the attention of visual predators
(Freidenfelds et al. 2012). This frequent wounding may lead to selection against stress-induced
innate immune suppression at these sites, and may even favor enhancement of innate immunity
(including hemagglutination). Innate immune function is essential for a rapid response to frequent
low-grade wounding and infection (Murphy 2001), and may be favored over the slower adaptive
immune function in this context (McDade et al. 2001). Other energy-sensitive traits (e.g.
behavior, growth, reproduction) may also be traded-off in order to maintain a high-functioning
innate immune system at these sites (Svensson et al. 1998; Norris and Evans 2000; Van Der Most
et al. 2011; Rauw 2012), and should be further investigated.
The mechanism behind the documented interaction between early-life and cross-
generational stress exposure are unclear, but could be induced by epigenetic processes
(Mostoslavsky and Bergman 1997; Fitzpatrick and Wilson 2003; Teitell and Richardson 2003),
and/or maternal effects, such as the transfer of maternal immunological memory (Hasselquist et
al. 2012; Ismail et al. 2015). These processes could affect the immune system directly or could
affect other systems, such as the HPA axis, that have cascading effects on the immune system.
We have found that lizards from fire ant-invaded populations have greater stress (CORT)
responsiveness as adults in the lab regardless of stress exposure in early life (McCormick, unpubl.
data), but it is unclear if underlying mechanisms affect HPA and immune function in a similar
manner. Additional research on the mechanisms behind these early life and cross-generational
changes would be useful.
107
In contrast to the effects of CORT exposure in early life, exposure to fire ants in early life
did not affect baseline hemagglutination scores. Whereas CORT exposure remained consistent
throughout the treatment period, fire ant exposure may have become less stressful over time if
lizards habituated to this exposure (Romero 2004; Cyr and Romero 2009; Romero et al. 2009) or
if the fire ant colony became less venomous in captivity (Tschinkel 2006; Xian-Fu et al. 2015),
rendering this treatment similar to the control. Alternatively, it may be that the immune enhancing
effect of CORT elevation due to fire ant attack is counteracted by potentially immune suppressive
effects of venom (Yi et al. 2003; Tankersley 2008) or frequent wounding (Plaistow et al. 2003).
Patterns of immune function in the field mirror the lack of immune modulation in
response to early life fire ant exposure in this study. In the field, baseline hemaggglutination
scores are similar for lizards at fire ant-invaded and -uninvaded sites (Graham et al. 2012) despite
the presence of both early life and cross-generational exposure to fire ants at invaded sites. The
field patterns do not, however, reflect the up-regulation of baseline hemagglutination following
lifetime CORT exposure in the present study. Several possibilities could explain these conflicting
findings: 1) CORT doses in the lab may have been greater than those elicited by stress in the
field. However, we selected the laboratory CORT dosage to reflect CORT elevations that occur in
response to natural encounters with fire ants (Trompeter and Langkilde 2011). 2) Adult lizards
used in this study were smaller and likely younger than those surveyed in the field (SVL;
McCormick unpubl. data). Younger lizards may up-regulate immune function in response to early
life CORT, as seen in this study, but older lizards may prioritize other traits, such as reproduction
(reviewed in Forslund and Pärt, 1995). 3) It is possible that intrinsic immune function is up-
regulated in lizards at fire ant-invaded sites in the field, but suppressed by extrinsic environmental
factors (e.g. frequent wounding), resulting in baseline levels that are similar to lizards at
uninvaded sites (Martin et al. 2011; Du et al. 2012). 4) Energy limitations in the field may prevent
lizards at high-stress fire ant-invaded sites from up-regulating immune function. This is supported
108
by the fact that immune effects of early life stress in adult birds are only observed under
energetically favorable conditions, and are not observed when energy demands are high (De
Coster et al. 2011). 5) Our application of CORT does not mimic the full spectrum of the HPA
response to an environmental stressor, as would be experienced in the field, including higher-
order effects (e.g. norepinephrine response) that may offset CORT effects. Further research into
how these factors affect the interaction of within- and across-generation stress exposure are
needed.
In the current study, adult acquired hemagglutination scores 2-weeks after inoculation
with sheep red blood cells (SRBC) were not affected by either early life or cross-generational
history with stress. This mirrors work in birds, in which early life nutritional stress did not affect
post-inoculation hemagglutination scores (Kriengwatana et al. 2013). It may not be surprising that
we observed modulation of innate (baseline hemagglutination) but not adaptive (post-inoculation)
immunity following exposure to stress. In reptiles, the adaptive immune system generally takes
longer to respond than in mammals and birds, in part because the innate immune system produces
a stronger response than in mammals and birds (Zimmerman et al. 2010). It is important to note
that hemagglutination scores did not increase overall in response to SRBC-injection, as would be
expected if an adaptive response had occurred (Ochsenbein and Zinkernagel 2000; Zimmerman et
al. 2010; Graham et al. 2012). The incubation period of fifteen days should have allowed an
appropriate adaptive response of SRBC-specific antibodies in this species, as demonstrated in
wild-caught lizards (Graham et al. 2012). However, in some cases the maximal response of
humoral immunity in reptiles takes six to eight weeks (Zimmerman et al. 2010) and our younger
lizards may have taken longer to respond than the older field lizards (Palacios et al. 2009;
Hopkins and Durant 2011).
In contrast to the effects of stress on baseline hemagglutination, early life stress exposure
(CORT or fire ants) did not affect bacterial killing ability of adult lizard plasma. Previous work
109
on this species has revealed similar short-term effects of stress on baseline hemagglutination but
not bacterial killing of E. coli (McCormick and Langkilde 2014; McCormick et al. 2015). In
birds, the opposite pattern is found—early life stress does not affect baseline hemagglutination
but does affect bacterial killing (De Coster et al. 2011; Kriengwatana et al. 2013; Schmidt et al.
2015). This varied sensitivity to early life stress may be a result of differential resource allocation
to specific immune responses (i.e. immune components, including complement, antibodies,
cellular activity) in ectotherms versus endotherms (Zimmerman et al. 2010), or due to species-
specific effects of early life stress on immune function (Schmidt et al. 2015). However, the
immune consequences of stress are known to vary depending on the immune component
measured (in response to short term stress: Matson et al., 2006; Stier et al., 2009; Brooks and
Mateo, 2013; McCormick and Langkilde, 2014; in response to early life stress: von Hoersten et
al., 1993; De Coster et al., 2011; Kriengwatana et al., 2013; Schmidt et al., 2015). Even within the
same branch of the immune system (e.g. innate), there is modulation of some components, such
as activity of phagocytes, but not others, such as complement (Schmidt et al. 2015) in response to
stress. It has been suggested that up-regulating a specific immune component may allow animals
to avoid an energetically-costly generalized immune response (Wegner et al. 2007). Bacterial
killing of E. coli 8739 (Millet et al. 2007; Graham et al. 2012) and hemagglutination (Matson et
al. 2005; Graham et al. 2012) both involve complement and natural antibody immune responses,
and we are unable to tease apart the effects of early life stress on these components. However,
early life and cross-generational history with stress may have different effects on other
components of the immune system.
110
Conclusions
The results of this study demonstrate that exposure to stress within a lifetime and across
generations interact to affect adult immune function. This is in contrast to our prior findings on
adult lizard HPA axis regulation, where cross-generational, but not early life, stress exposure was
shown to affect adult HPA activity in this species (Chapter 4). Cross-generational history with
stress, but not early life stress, also influenced survival and morphology of this species
(Langkilde, unpubl. data). Together, these results suggest that cross-generational and early life
stress affect different traits in different manners, and caution that cross-generational history with
stress likely plays an important role in determining adult phenotype. Thus, the interaction
between early life and cross-generational stress exposure should be considered.
Many organisms will be exposed to novel stressors as a result of global environmental
change, and it is critical that we understand how immune function and other fitness-relevant traits
trade off to balance the energetic costs of responding to these stressors. Assessing how these
responses may vary across taxa (including endotherms versus ectotherms), traits, and
environmental gradients is necessary to allow generalizations that permit the prediction and
management of the effects of stress on wild populations.
Acknowledgements
We thank S. Graham and C. Thawley for assistance with planning, S. Graham, C.
Thawley, and J. Newman for help with lizard collection; C. Thawley, S. McGinley and E. Baron
for assistance with lizard trials; C. Thawley for assistance with blood collection, D. McGregor for
assistance with immune assays; C. Thawley, G. Dewitt, D. Fricken, M. Goldy-Brown, A.
Hollowell, M. Hook, C. Norjen, S. McGinley, L. Horne, A. Jacobs and M. O’Brien for lizard care
111
and assistance with developmental lizard treatments; the Cavener lab for use of their plate reader;
and B. Chitterlings for valuable comments on this manuscript. We thank the Lansdale family for
access to their land and lizards and personnel at St. Francis National Forest, Edgar Evins State
Park, Standing Stone State Park, Blackwater River State Forest, Geneva State Forest, Conecuh
National Forest, and especially the Solon Dixon Forestry Education Center for logistical support.
All methods detailed here adhere to the Guidelines for the Use of Animals in Research and the
Institutional Guidelines of Penn State University (IACUC #35780), and animal collection was
permitted by the respective states. Funding was provided in part by the National Science
Foundation (DGE1255832 to GLM and IOS1051367 to TL and SAC).
112
Figures
Figure 5-1: Early life and cross-generational history of stress exposure interact to affect adult baseline, but not post-inoculation, hemagglutination scores. a) In offspring of lizards from fire ant-uninvaded populations, CORT exposure during early life suppressed adult baseline plasma hemagglutination compared to controls. The opposite effect was seen in offspring of lizards from fire ant-invaded populations: early life CORT exposure enhanced adult baseline hemagglutination compared to controls. b) Post-inoculation hemagglutination scores did not differ across early life stress treatment or fire ant-invasion status. Bars represent means ± one standard error and sample size for each group is shown above each set of bars.
113
Figure 5-2: Bacterial killing by plasma of adult lizards is not related to early life or cross-generational history with stress. Offspring of lizards from fire ant invaded and uninvaded populations exposed weekly to fire ants (FA), CORT, or control treatment from hatching until maturity had similar percent bacterial killing ability of plasma as adults. Bars represent means ± one standard error and sample size for each group is shown above each bar.
114
References
Avitsur R., J. Hunzeker, and J.F. Sheridan. 2006. Role of early stress in the individual differences
in host response to viral infection. Brain Behav Immun 20:339–348.
Belliure J. and J. Clobert. 2004. Behavioral sensitivity to corticosterone in juveniles of the wall
rapidly increases activity in Gambel’s white-crowned sparrows (Zonotrichia leucophrys
gambelii). Gen Comp Endocrinol 111:386–94.
De Coster G., S. Verhulst, E. Koetsier, L. De Neve, M. Briga, and L. Lens. 2011. Effects of early
developmental conditions on innate immunity are only evident under favourable adult
conditions in zebra finches. Naturwissenschaften 98:1049–56.
De Frenne P., B.J. Graae, F. Rodr??guez-S??nchez, A. Kolb, O. Chabrerie, G. Decocq, H. De
Kort, et al. 2013. Latitudinal gradients as natural laboratories to infer species’ responses to
temperature. J Ecol 101:784–795.
Dhabhar F.S., W.B. Malarkey, E. Neri, and B.S. McEwen. 2012. Stress-induced redistribution of
immune cells--from barracks to boulevards to battlefields: a tale of three hormones--Curt
Richter Award winner. Psychoneuroendocrinology 37:1345–68.
Dhabhar F.S. and B.S. McEwen. 1997. Acute stress enhances while chronic stress suppresses
cell-mediated immunity in vivo: a potential role for leukocyte trafficking. Brain Behav
Immun 11:286–306.
French S.S., H.B. Fokidis, and M.C. Moore. 2008. Variation in stress and innate immunity in the
tree lizard (Urosaurus ornatus) across an urban-rural gradient. J Comp Physiol B 178:997–
1005.
127
Greenberg N. and J.C. Wingfield. 1987. Stress and Reproduction: Reciprocal Relationships. Pp.
461–503 in D.O. Norris and R.E. Jones eds. Horm Reprod Fishes, Amphib Reptil. Springer
US, New York.
Homan R.N., J. V. Regosin, D.M. Rodrigues, J.M. Reed, B.S. Windmiller, and L.M. Romero.
2003. Impacts of varying habitat quality on the physiological stress of spotted salamanders
(Ambystoma maculatum). Anim Conserv 6:11–18.
Ilmonen P., D. Hasselquist, A. Langefors, and J. Wiehn. 2003. Stress, immunocompetence and
leukocyte profiles of pied flycatchers in relation to brood size manipulation. Oecologia
136:148–54.
McEwen B.S. and J.C. Wingfield. 2003. The concept of allostasis in biology and biomedicine.
Horm Behav 43:2–15.
Meylan S. and J. Clobert. 2005. Is corticosterone-mediated phenotype development adaptive?
Maternal corticosterone treatment enhances survival in male lizards. Horm Behav 48:44–52.
Murphy K.M. 2001. Janeway’s Immunobiology (revised.). Garland Science, Taylor & Francis
Group, LLC, New York.
Norris K. and M.R. Evans. 2000. Ecological immunology : life history trade-offs and immune
defense in birds. Behav Ecol 11:19–26.
Schmidt K.L., S.P. Kubli, E. a. MacDougall-Shackleton, and S. a. MacDougall-Shackleton. 2015.
Early-Life Stress Has Sex-Specific Effects on Immune Function in Adult Song Sparrows*.
Physiol Biochem Zool 88:183–194.
Schmidt K.L., E. a Macdougall-Shackleton, and S. a Macdougall-Shackleton. 2012.
Developmental stress has sex-specific effects on nestling growth and adult metabolic rates
but no effect on adult body size or body composition in song sparrows. J Exp Biol 215:3207–
17.
128
Schreck C.B., B.L. Olla, and M.W. Davis. 1997. Behavioral responses to stress. Pp. 145–170 in
G.K. Iwama, A.D. Pickering, J.P. Sumpter, and C.B. Schreck eds. Fish Stress Heal Aquac.
Cambridge University Press, New York.
Stier K.S., B. Almasi, J. Gasparini, R. Piault, A. Roulin, and L. Jenni. 2009. Effects of
corticosterone on innate and humoral immune functions and oxidative stress in barn owl
nestlings. J Exp Biol 212:2085–91.
Tomei F., M.V. Rosati, M. Ciarrocca, T.P. Baccolo, M. Gaballo, T. Caciari, and E. Tomao. 2003.
Plasma cortisol levels and workers exposed to urban pollutants. Ind Health 41:320–6.
Trompeter W.P. and T. Langkilde. 2011. Invader danger: lizards faced with novel predators
exhibit an altered behavioral response to stress. Horm Behav 60:152–8.
Vitousek P.M. 1997. Human Domination of Earth’s Ecosystems. Science (80- ) 277:494–499.
Wegner K.M., M. Kalbe, and T.B.H. Reusch. 2007. Innate versus adaptive immunity in
sticklebacks: Evidence for trade-offs from a selection experiment. Evol Ecol 21:473–483.
Zimmerman L.M., L. a Vogel, and R.M. Bowden. 2010. Understanding the vertebrate immune
system: insights from the reptilian perspective. J Exp Biol 213:661–671.
129
Appendix
What Makes Stress Stressful? Extending the Acute-Chronic Stress Paradigm
The ecological causes and consequences of stress have received increasing attention in
the last decade and have been the focus of many special features and symposia (e.g. Functional
Ecology Special Feature (2013): The Ecology of Stress; Society of Integrative and Comparative
Biology Symposium (2013): Coping with Uncertainty; Journal of Experimental Biology Special
Feature (2014): Stress, Challenging Homeostasis).
Stress is typically characterized by the duration of the stressor. Acute stress is
characterized as “short” in duration (Burchfield, 1979; Romero, 2004), lasting from minutes to a
few days (Boonstra, 2012; Martin, 2009), and is usually not repeated. Chronic stress is
characterized as “long” in duration (Burchfield, 1979; Romero, 2004; Sapolsky et al., 2000),
lasting from days to months (Boonstra, 2012; Dhabhar, 2009; Martin, 2009): a stimulus to which
the organism is continuously or repeatedly exposed. This terminology is frequently used in the
animal stress literature (Table A-1), as well as in human medical practice (A.D.A.M. Inc., 2013;
Merriam-Webster Incorporated, n.d.). These terms are also used widely in popular culture.
Despite the fact that some studies have independently addressed stressor intensity
(McEwen et al., 1997; Ottenweller et al., 1989), or frequency (Busch et al., 2008; McCormick et
al., 1998) we know of only one study that examines more than one aspect simultaneously
(McCormick et al., 2015). Furthermore, while stress is usually characterized by stressor duration,
it is also characterized by the consequence of the stressor (e.g. typically acute stress enhances,
while chronic stress suppresses, immune function).
130
Unfortunately, common usage of these terms for both the cause and the consequence
of stress, as well as the acute-chronic paradigm’s focus on duration, results in confusion.
Inconsistent use of terminology contributes to circular definitions and conflicting results on the
consequences of stress with different durations (Table A-1), and hinders our ability to predict
how an organism will be affected by stress. Relying on “acute” and “chronic” to encompass a
plethora of meanings (different stressor aspects; cause vs. consequence), and limiting our
definitions to duration of stress, may be hampering progress towards understanding how stress
can have different consequences (Romero et al., 2009). A more complete stress framework, that
includes other aspects of stress that are likely critically important in determining the
consequences of stress, will allow more rapid progress in an increasingly important field that is
currently hampered by a limited terminological framework.
131
Tables
Table A-1: Representative studies from the literature showing the full range of both duration of stress applications (rows) and consequences (columns) of stress, using definitions of “acute” and “chronic” stress from each source paper. Shaded boxes represent combinations expected according to existing theory.
behaviors in male tree lizard morphs. Horm. Behav. 56, 51–7.
Tokarz, R.R., 1987. Effects of the antiandrogens cyproterone acetate and flutamide on male
reproductive behavior in a lizard (Anolis sagrei). Horm. Behav. 21, 1–16.
VITA — GAIL LINDSEY MCCORMICK
EDUCATION Ph.D. Ecology Intercollege Graduate Degree Program. Penn State University. May 2016. B.S. with high distinction. (Ecology & Evolutionary Biology) University of Michigan. 2010. B.T.A. with highest honors. (Theatre Arts) University of Michigan. Ann Arbor, MI. 2010.
PUBLICATIONS McCormick, GL, T Robbins, S Cavigelli, T Langkilde. Under review. Ancestry
trumps experience: Cross-generational but not developmental stress affects the physiological stress response. Horm Behav.
McCormick, GL, K Shea, T Langkilde. 2015. How do duration, frequency, and intensity of exogenous CORT elevation affect immune outcomes of stress? Gen Comp Endocrinol 222: 81-87.
McCormick, GL, T Langkilde. 2014. Immune responses of Eastern fence lizards (Sceloporus undulatus) to repeated acute elevation of corticosterone. Gen Comp Endocrinol 204: 135-140.
Stuble, KL, MA Rodriguez-Cabal, GL McCormick, RR Dunn, NJ Sanders. 2013 Tradeoffs, competition, and coexistence in eastern deciduous forest ant communities. Oecologia 171(4): 981-992.
Graham, SP, NA Freidenfelds, GL McCormick, T Langkilde. 2012. The impacts of invaders: Basal and acute stress glucocorticoid profiles and immune function in native lizards threatened by invasive ants. Gen Comp Endocrinol 176(3): 400-408.
GRANTS, AWARDS & FELLOWSHIPS 2016 Alumni Association Dissertation Award. Penn State Alumni Association. $5,000 2015 Ecology Research Assistantship. PSU Ecology Program. $11,800 (declined) ESA Travel Award. Ecological Society of America Physiology Section. $500 2014 Ecology Travel Award. PSU Ecology Intercollege Graduate Degree Program. $200 Biology Travel Grant. PSU Department of Biology. $250 2013 Charlotte Mangum Student Support. Society for Integrative and Comparative Biol. $300 2012 Graduate Research Fellowship. National Science Foundation. $126,000 Gaige Award. American Society of Ichthyologists and Herpetologists. $500 2011 University Graduate Fellowship. Penn State University. $41,730 Arthur J. Schmitt Presidential Fellowship. University of Notre Dame. $284,800 (declined)
TEACHING AND LEADERSHIP 2015 Teaching Assistant. Populations and Communities. BIO 220. Penn State University. 2014 Guest lecturer. BIOL 429. Animal Behavior. Penn State University. 2014 Guest lecturer. WFS 462. Amphibians & Reptiles Penn State 2010 Teaching Assistant. Field Mammalogy. University of Michigan Biological Station 2010 Teaching Assistant. Inside the Dramatic Process. University of Michigan
LEADERSHIP AND OUTREACH EXPERIENCE 2014 Ecology Graduate Student Association. President. Penn State University. 2013–2015 Science Café. Committee Chair. Penn State University Ecology Program. 2014 Science Café. Presenter. Penn State University Ecology Program. 2012–2013 Ecology Graduate Student Association. Webmaster. Penn State University. 2012–2013 Ecology Spring Seminar Series. Co-Coordinator. Penn State University