TRAUMATIC STRESS RESPONSES IN RATS REVEAL FUNDAMENTAL SEX DIFFERENCES THAT
MIRROR PTSD IN MEN AND WOMEN
By
Apryl Elizabeth Pooley

A DISSERTATION
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Neuroscience—Doctor of Philosophy
2017

ABSTRACT
TRAUMATIC STRESS RESPONSES IN RATS REVEAL FUNDAMENTAL SEX DIFFERENCES THAT
MIRROR PTSD IN MEN AND WOMEN
By
Apryl Elizabeth Pooley
Post-traumatic stress disorder (PTSD) develops after exposure to trauma and is associated with
dysfunction in the normal stress response. Women are twice as likely as men to develop PTSD and tend
to experience different symptoms and comorbidities than men, but the neurobiological basis for these
pervasive sex differences is poorly understood due to the overwhelming male bias in the preclinical
research. My dissertation work tested the novel hypothesis that the neurobiological mechanisms
underlying the traumatic stress response in male and female rats are fundamentally different and may be
related to normal sex differences in circulating levels of adult gonadal hormones. These experiments are
the first to compare adult male and female rats across two rodent models of PTSD, single prolonged
stress and predator exposure. I report a highly sex-specific traumatic stress response that recapitulates
fundamental differences of PTSD in men and women. Surprisingly, these sex differences were largely
independent of adult circulating gonadal hormones, housing conditions, and types of stress. Two
standard measures, the acoustic startle response and dexamethasone suppression test to measure the
negative feedback control of the stress hormone response, suggest that female rats, unlike male rats, are
resilient to the effects of traumatic stress. However, other measures like sucrose preference and social
interaction make it clear that females are not resilient, but simply respond differently to trauma than
males. Dramatic sex differences in how trauma affects cFos activation and glucocorticoid receptor
expression in the brain lend further support to the idea that the trauma response of males and females is
fundamentally different, and likely determined prior to adulthood. Factors that mediate differences in how
individuals adjust after trauma are attractive targets for the prevention and treatment of PTSD, and
identifying such factors of resilience depends on understanding the various ways the traumatic stress
response manifests in different individuals. I propose that sex differences offer a promising inroad for
addressing this issue.

For all those who survived the unthinkable, and especially for those who didn’t.

iii

ACKNOWLEDGEMENTS

First and foremost, I must acknowledge the rats that sacrificed their lives for this work. Their lives
are truly sacred, and I am forever indebted to them for allowing this research to be done. Thank you for
your service.
To my advisor, confidant, and partner-in-crime, Dr. Cindy Jordan, for giving me the space,
freedom, and support to pursue this work with my whole being. You always believed in me and trusted in
my judgement even when I didn’t, and your dedication to science and to mentoring will continue to be a
template for the scientist and mentor I want to be.
To my committee, Dr. Marc Breedlove, Dr. Michelle Mazei-Robison, and Dr. AJ Robison for your
guidance and feedback on my experimental design, abstracts, posters, and manuscripts. To Dr. Shane
Perrine and Dr. Andrew Eagle for your expertise in setting up the SPS model that was crucial for the
completion of this work.
To everyone in the Breedlove-Jordan lab, especially Diane Redenius, Kate Mills, and Beth
Kenyon for your technical assistance. To Dr. Chieh Chen for your help with IHC and for continuing to
check in to see how things were going along the way, I’m so glad our paths crossed in the lab. To the
undergraduate researchers who worked with me, Grace Kamau and Alla Kedzierski for your dedication
and enthusiasm to learn, Susheela Sreedhar for jumping into research head first and making the female
housing study possible, and Cassie Benjamin for three years of commitment to this work and making the
predator exposure study possible (and for making those long days of mounting tissue unbelievably
enjoyable, and especially for your friendship and support outside the lab—you’re going to make an
incredible doctor).
And finally, to all the people who were invested in my health and well-being in times of both
personal struggle and triumph. Kintla Striker, Dr. Ann Ryan, Becky Allen, Lisa Laughman, Jerilyn
Robison, and Gretchen Morse for being the best trauma-informed healthcare allies. To Jessie Shuemaker
for your unwavering friendship and support. To Mandy for being my life-partner-in-crime and co-parent to
two dogs. And to Dr. Christina Ragan for being a friend, colleague, fellow neurd, confidant, pop culture
companion, and conference buddy.

iv

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................................................ vii
LIST OF FIGURES ..................................................................................................................................... viii
KEY TO ABBREVIATIONS ..........................................................................................................................ix
CHAPTER 1: INTRODUCTION .................................................................................................................... 1
Post-traumatic stress disorder in humans ............................................................................................... 1
Traumatic stress is a unique form of stress ..................................................................................... 1
The effects of traumatic stress have been observed throughout recorded history ......................... 3
Risk factors for PTSD....................................................................................................................... 6
Symptoms and subtypes of PTSD ................................................................................................. 10
Dissociative subtype of PTSD ........................................................................................... 11
Child subtypes of PTSD .................................................................................................... 12
Delayed-onset PTSD ........................................................................................................ 12
Internalizing and externalizing subtypes of PTSD ........................................................... 13
Complex PTSD ................................................................................................................. 13
Pathophysiology ............................................................................................................................. 16
Sympathetic nervous system ............................................................................................ 16
Hypothalamic-pituitary-adrenal axis .................................................................................. 17
Limbic system ................................................................................................................... 17
Sex differences in PTSD ................................................................................................................ 19
Sex differences in rodent brain and behavior ........................................................................................ 22
Gonadal hormones, sex chromosomes, and the development of sex differences ....................... 22
Sex differences in the rat HPA axis .............................................................................................. 22
Rat estrous cycle ............................................................................................................................ 23
Merging the fields of PTSD and sex differences research with animal models ................................ 25
Studying sex differences in rodents ............................................................................................... 25
Animal models of PTSD ................................................................................................................. 25
Single prolonged stress .................................................................................................... 27
Predator exposure............................................................................................................. 28
Measuring responses to traumatic stress ...................................................................................... 28
Dexamethasone suppression test ..................................................................................... 28
Acoustic startle response .................................................................................................. 29
Overview of chapters ................................................................................................................................ 30
CHAPTER 2: CHARACTERIZING SEX DIFFERENCES IN RESPONSE TO SPS .................................. 31
Introduction ............................................................................................................................................... 31
Methods and Materials ............................................................................................................................. 32
Experiment 1: Sex differences in the SPS model .......................................................................... 32
Animals ............................................................................................................................. 32
Acoustic startle response .................................................................................................. 32
SPS paradigm ................................................................................................................... 32
Dexamethasone suppression test ..................................................................................... 33
cFos and GR immunohistochemistry ................................................................................ 33
Statistical analysis ............................................................................................................. 35
Experiment 2: Effect of social housing on female SPS response ................................................. 35
Animals ............................................................................................................................. 35
ASR, DST, IHC ................................................................................................................. 35
Sucrose preference test .................................................................................................... 35
Social interaction test ........................................................................................................ 36

vi

Statistical analysis ............................................................................................................. 36
Experiment 3: Conditioned fear in SPS ........................................................................................ 36
Results ....................................................................................................................................................... 37
Males and females respond differently to SPS ............................................................................. 37
Neither housing nor bright lights affect the female response to SPS ........................................... 38
Traumatized rats are sensitive to contextual reminders of SPS .................................................... 39
Discussion ................................................................................................................................................. 40
CHAPTER 3: CHARACTERIZING SEX DIFFERENCES IN RESPONSE TO PREDX ............................. 47
Introduction .............................................................................................................................................. 47
Methods and Materials ............................................................................................................................ 47
Experiment 4: Sex differences in the PredX model ...................................................................... 47
Animals ............................................................................................................................ 47
Predator exposure............................................................................................................. 48
Outcome measures ........................................................................................................... 48
Statistical analysis ............................................................................................................. 48
Results ....................................................................................................................................................... 48
Predator exposure induces comparable sex differences in ASR and DST .................................. 48
Discussion ................................................................................................................................................. 49
CHAPTER 4: THE ROLE OF GONADAL HORMONES IN THE TRAUMATIC STRESS RESPONSE.... 52
Introduction ............................................................................................................................................... 52
Methods and Materials ............................................................................................................................. 52
Experiment 5: Effects of gonadectomy and testosterone replacement in SPS model ................. 52
Animals ............................................................................................................................. 52
Gonadectomy .................................................................................................................... 53
SPS paradigm ................................................................................................................... 53
Outcome measures ........................................................................................................... 53
Statistical analysis ............................................................................................................. 53
Results ....................................................................................................................................................... 53
Sex differences were replicated in the sham surgery group .......................................................... 53
Effects of SPS in males are largely independent of adult testicular hormones ............................ 54
Effects of SPS in females are largely independent of ovarian hormones ..................................... 55
Discussion ................................................................................................................................................. 56
CHAPTER 5: CONCLUSIONS ................................................................................................................... 63
Do sex differences in the traumatic stress response reflect differences in resilience? ................. 63
Future directions: developmental effects of trauma ...................................................................... 64
APPENDIX .................................................................................................................................................. 66
REFERENCES ............................................................................................................................................ 94

vi

LIST OF TABLES

Table 1.

PTSD diagnostic criteria in DSM-5.

15

Table 2.

Statistical results for data shown in Figures 2 and 11.

68

Table 3.

Statistical results for data shown in Figures 3 and 12.

72

Table 4.

Statistical results for data shown in Figures 6 and 14.

75

Table 5.

Statistical results for data shown in Figure 8.

78

Table 6.

Statistical results for data shown in Figures 9 and 16-17.

83

Table 7.

Statistical results for data shown in Figures 10 and 18-19.

89

vii

LIST OF FIGURES

Figure 1.

Timeline of procedures for experiments 1 and 2.

43

Figure 2.

Single prolonged stress (SPS) affects males and females differently.

44

Figure 3.

Housing conditions change how SPS affects social interaction in females.

45

Figure 4.

Contextual reminders to SPS sensitizes anxiety behavior over time.

46

Figure 5.

Timeline of procedures for experiment 4.

50

Figure 6.

Predator exposure (PredX) leads to sex differences in the acoustic startle
response (ASR) and HPA negative feedback comparable to SPS.

51

Figure 7.

Timeline of procedures for experiment 5.

59

Figure 8.

SPS elicits comparable sex differences in sham gonadectomized rats.

60

Figure 9.

Effects of SPS in males are largely independent of adult testicular
hormones.

61

Figure 10.

Effects of SPS in females are largely independent of ovarian hormones.

62

Figure 11.

SPS did not affect body weight.

67

Figure 12.

Neither SPS nor housing affected female body weight.

70

Figure 13.

Females spent more total time in the interaction zone when there was a
target rat present compared to an empty chamber, regardless of SPS
exposure or housing.

71

Figure 14.

As with SPS, predator exposure (PredX) did not affect body weight, but
females weighed less than males.

74

Figure 15.

Rats spent more total time in the interaction zone when there was a target
rat present compared to an empty chamber, regardless of SPS exposure or
sex.

77

Figure 16.

Neither SPS nor castration (GDX) affected body weight of males.

81

Figure 17.

Neither SPS nor GDX affected adrenal weight of males.

82

Figure 18.

SPS did not affect the body weight of females, but as expected, GDX did by
increasing the body weight of ovariectomized females.

87

Figure 19.

Neither SPS nor GDX affected adrenal weight of females.

88

Figure 20.

Individuals are not uniformly susceptible or resilient across measures.

93

viii

KEY TO ABBREVIATIONS

ACC

anterior cingulate cortex

ACTH

adrenal corticotropic hormone

APA

American Psychiatric Association

AR

androgen receptor

ASR

acoustic startle response

BDNF

brain-derived neurotrophic factor

BLA

basolateral amygdala

CBG

corticosterone binding globulin

CORT

corticosterone

CRH

corticotropin-releasing hormone

DESNOS

disorders of extreme stress, not otherwise specified

DEX

dexamethasone

DSM-III/IV/5

Diagnostic and Statistical Manual of Mental Disorders-third, fourth, or fifth edition

DST

dexamethasone suppression test

fMRI

functional magnetic resonance imaging

GDX

gonadectomy/gonadectomized

GR

glucocorticoid receptor

HPA

hypothalamic-pituitary-adrenal

ICD-10

International Statistical Classification of Diseases and Related Health Problems-10th ed.

IHC

immunohistochemistry

IL

infralimbic region of the medial prefrontal cortex

mCPP

meta-Chlorophenylpiperazine

MDD

major depressive disorder

mPFC

medial prefrontal cortex

PET

positron emission tomography

PredX

predator exposure

ix

PrL

prelimbic region of the medial prefrontal cortex

PTSD

post-traumatic stress disorder

PVN

paraventricular nucleus of the hypothalamus

SNS

sympathetic nervous system

SPS

single prolonged stress

T

testosterone

x

CHAPTER 1: INTRODUCTION

Post-traumatic stress disorder in humans
Post-traumatic stress disorder (PTSD) is a disorder that can develop after exposure to a
traumatic experience and is associated with dysfunction in the normal stress response. As many as 10%
of civilians and 30% of military veterans in the United States develop PTSD after trauma (1–4), and
affected individuals can remain chronically symptomatic for decades with debilitating symptoms that
interfere with daily functioning (5). Clinical manifestations of PTSD include a wide range of symptoms that
often overlap or are co-morbid with other psychiatric conditions. At least one other co-morbid psychiatric
condition occurs in 88% of people with PTSD, with major depressive disorder (MDD) diagnosed in half of
all PTSD patients (1), indicating significant heterogeneity in PTSD. A history of experiencing trauma is not
sufficient alone to diagnose PTSD, as over 75% of people in the United States experience a traumatic
event in their lifetime and do not develop PTSD (6), begging the question why do some people who
experience trauma develop PTSD while others do not? To begin to answer this question, we must first
understand what makes an experience traumatic.

Traumatic stress is a unique form of stress. In 1936, Hans Selye introduced the first report of a
biological stress syndrome, in which he described an “alarm reaction” and suggested that the symptoms
are “independent of the nature of the damaging agent” (7). While several early hypotheses regarded
stress as a non-specific response that was essentially the same for all stressors, much evidence exists
today to indicate the contrary—various types of stressors impart distinct “neurochemical signatures” and
activational patterns in the nervous system (8, 9). General classifications of stressor type are based on
four broad categories that include physical stressors, psychological stressors, social stressors, and
cardiovascular/metabolic stressors in acute or chronic duration (8). But traumatic stressors can fall into
any of these categories, and a stressor that is traumatic in nature has its own unique signature, so a more
clinically relevant classification scheme may be one that first identifies a stressor as traumatic or not.
The key to a stressor qualifying as traumatic is that it poses a real or perceived threat to life, physical
integrity, or social affiliation (of oneself or one’s kin) and is uncontrollable in some aspect (10).

1

Uncontrollable stressors are those to which an individual is unable to act upon or to which the available
stress responses (e.g. sympathetic fight-or-flight, neuroendocrine responses, etc.) are not applicable.
Controllability is formally defined as “the probability that a given response will prevent or terminate” the
stressor (11). In response to controllable threats, most animals exhibit physiological and behavioral
changes that function as adaptive, life-preserving responses (12–14). Exposure to mild, controllable
stressors early in life can be beneficial to the developing nervous system to (e.g. the “stress-inoculation
theory”) (15). All animals experience mild, predictable stressors from birth and, in healthy environments,
learn repeatedly which behaviors will terminate or aid in adaptation to cold, hunger, loneliness, fatigue,
and discomfort (16). These associative learning experiences can impart healthy and successful coping
with other stressors throughout life.
On the other hand, little can be learned from an uncontrollable stressor, and maladaptive behaviors
may result from the activation of sympathetic and neuroendocrine stress response systems when the
individual is unable or unequipped to act, or when the responses elicited do not result in consistent
outcomes. Stressors that are more recently-established on the evolutionary time scale may be more likely
to elicit PTSD (e.g. a “neuroevolutionary time-depth principle”) (17), evidenced by the fact that adversities
such as fires are less likely to result in PTSD than adversities for which the human genome has had much
less time to evolve such as motor vehicle accidents (1) or modern surgical procedures and treatments,
after which PTSD rates range from 14-59% (18). After a train accident, for example, where a stress
response was elicited but there was no opportunity to make any behavioral action, the brain subsequently
searches for ways in which the individual was responsible for their own survival (or for their own harm),
and maladaptive thought and behavior patterns can develop. A person might decide that they survived
the train crash because they were sitting at the back of the train, so in the future they only sit at the backs
of trains. Or they might decide that the reason they got in the accident in the first place was because they
took an earlier train than they normally take, so in the future they will never ride trains in the morning.
These types of avoidant behaviors are a key diagnostic criterion for PTSD and primarily occur after
experiencing an uncontrollable stressor. Experimental studies indicate that uncontrollable stress is
necessary for the onset of PTSD-like symptoms and leads to physiological and behavioral consequences

2

different from those associated with controllable stress (11, 16, 19, 20). Those consequences further
diverge if the stress includes a psychosocial aspect.
While physical trauma can have psychological manifestations and psychological trauma can have
physical manifestations (21), physical trauma nonetheless differs from psychosocial trauma in that
physical trauma is caused by objective “changes of the world outside the brain” rather than being
dependent on subjective perception, beliefs, assumptions, expectations, and previous experience within
the brain, as happens in psychosocial conflict, which is the predominant form of stress in humans and
many other social mammals (16). While the ramifications of a gunshot wound, for example, may be
objectively similar at the bullet entry point between two victims, their perceptions of the psychosocial
aspect of the shooting may be completely different—informing unique neurocircuitry effects—depending
on their individual worldviews and experiences. This makes the traumatic stress response, especially
when any psychosocial aspect is involved, extremely heterogenous (22). Considering the various ways
the traumatic stress response has manifested in different individuals throughout history—and not limited
to the ever-changing diagnostic definitions created by professional and political organizations—is a
crucial step to fully understand how to prevent, diagnose, and treat PTSD.

The effects of traumatic stress have been observed throughout recorded history. From
depictions in the Epic of Gilgamesh (300 B.C.) to Homer’s Iliad (800 B.C.) to much of Shakespeare (23,
24), historical descriptions of the effects of war, sexual assault, physical and emotional abuse, natural
disasters, and accidents indicate that PTSD as we know it is not a modern phenomenon, as some have
suggested (25). That life-threatening events can leave a lasting impression upon the body and psyche
has been unwavering in these remarkably parallel historical descriptions of trauma, but the field of
psychiatry has not been so consistent in acknowledging that external factors like trauma can impact an
individual so profoundly and has been peppered with assertions that PTSD merely represents a
vulnerability of the individual that was present with or without trauma (26, 27).
The earliest clinical study on the effects of trauma was published in 1866 when sleep
disturbances, nightmares, chronic pain, depression, memory loss, and avoidance of railway travel
manifested in railway employees and passengers who had experienced train crashes and whiplash from

3

the violent rocking of the train cars (28). Termed railway spine, these symptoms closely mirror current
diagnostic criteria for PTSD (22), and when Dr. Erichsen first studied railway spine, he wrote, “Do not for
a moment suppose that these injuries are peculiar to and are solely occasioned by accidents that may
occur on railways” (28). However, while Erichsen thought the psychological effects of trauma must be a
result of physical injury to the spine, surgeon Herbert Page claimed that both physical and psychological
manifestations of trauma could arise without any detectable lesions to the central nervous system. Page
compared railway spine to hysteria—characterized by almost identical symptoms—suggesting that both
conditions arise from the same psychological processes and that “the change [after trauma] may be a
chemical one” (29). Significant hypotheses regarding the effects of psychological trauma stemmed from
studies of hysteria in the nineteenth century, in which it was suggested that childhood sexual abuse could
cause symptoms of hysteria later in life in both men and women (30), a supposition we know to be true
today. Other responses to trauma were associated with war, including American Civil War and World War
I soldiers who had what was then described as “soldiers’ heart” and “shell shock,” respectively—the
symptoms of which mirror those of currently defined PTSD (31, 32). One landmark study found similar
symptoms in World War II concentration camp survivors (“Concentration Camp Syndrome”) that persisted
15 years after the war had ended (33). The term “rape trauma syndrome” was coined in 1974 to describe
similar effects in some rape survivors (34). After Vietnam, nearly 25% of veterans returned from combat
with psychological problems (“Post-Vietnam Syndrome”) that resembled those of soldiers in previous
wars, leading to the conceptualization of PTSD as a combat disorder (22). The first official recognition of
PTSD, which was issued by the American Psychiatric Association (APA) in the 1980 edition of the
Diagnostic and Statistical Manual of Mental Disorders (DSM-III), was fueled by advocates calling
specifically for the treatment of the psychological problems in returning Vietnam veterans. Nonetheless, in
addition to military combat, the DSM-III explicitly included rape, physical assault, and motor vehicle
accidents as stressors that could lead to PTSD. This recognition was monumental in advancing research
on the effects of different types of traumatic stress in various populations, and it is now widely accepted
that both civilians and veterans who experience any type of trauma are at risk for PTSD. The other major
medical manual, the World Health Organization’s International Statistical Classification of Diseases and
Related Health Problems (ICD) first included PTSD in the 1992 ICD-10, which is still in its current edition.

4

When the current edition of the DSM (DSM-5) was issued in 2013, PTSD was no longer listed as
an anxiety disorder but instead considered part of the newly recognized “trauma and stressor-related
disorders,” reflecting the unique nature of disorders caused by trauma/stress that is not captured in a
simple anxiety phenotype (also included were reactive attachment disorder, acute stress disorder,
adjustment disorders, and disinhibited social engagement disorder). The DSM-5 diagnostic criteria for
PTSD require exposure to “actual or threatened death, serious injury, or sexual violation.” The exposure
must result from one or more of the following scenarios, in which the individual: (1) directly experiences
the traumatic event, (2) witnesses the traumatic event in person, (3) learns that the traumatic event
occurred to a close family member or friend, and (4) experiences first-hand repeated or extreme exposure
to aversive details of the traumatic event (e.g. emergency responders, police officers, corrections officers,
etc.). This DSM revision aligns closely with the ICD-10 classification of PTSD. Notably, the requisite
diagnostic criterion that an individual had responded to the trauma with “fear, helplessness, or horror” was
removed in the DSM-5, citing that it had not been proven that those subjective responses could predict
the onset of PTSD (35). This position was coupled with an impetus to focus research and treatment on
the measurable objective effects of trauma, as the subjective psychosocial factors proved highly variable
and difficult to control or study, as learned from early studies of PTSD.
The initial understanding of PTSD was limited almost exclusively to psychological symptoms
based on self-report, which propagated skepticism about the validity of PTSD as a disorder. But as
progress in biological research confirmed that the disorder was caused by an external agent (i.e. a
traumatic event), and biological readouts of the effects of trauma were identified (36), the medical
community and public began to accept PTSD as a valid disorder worth studying and treating. It is now
widely accepted that a single traumatic event is necessary to induce the onset of PTSD. Nonetheless, the
traumatic event alone is not sufficient to lead to the development of PTSD, as the same traumatic event
triggers PTSD in some individuals but not in others (37). While the characteristics of the traumatic event
itself are invariably important in the subsequent development of PTSD, the subjective perception of
trauma and individual susceptibilities still cannot be ruled out as contributing risk factors for the
development of PTSD.

5

Risk factors for PTSD. Epidemiological studies indicate that 1-in-14 civilians and 1-in-3
veterans develop PTSD, but some people are more susceptible than others (1). Women are twice as
likely as men to develop PTSD, with 10% of women and 5% of men meeting the diagnostic criteria for
PTSD (see section: Sex differences in PTSD) and a history of multiple traumatic experiences increases
the risk for PTSD (38). Additionally, some types of stressors are more likely to lead to PTSD than others,
and the resources available to an individual in the aftermath of trauma have a large impact on whether
PTSD develops (39).
Some hypothesize that “stress exposure only affects individuals with a susceptible genetic
background” (40), but even the most liberal estimates find that only about 30% of the variance in
responses to trauma can be attributed to heredity (41, 42). Genetic backgrounds with certain
polymorphisms in genes encoding the serotonin transporter (43, 44), brain-derived neurotrophic factor
(BDNF) (45, 46), the pituitary adenylate cyclase-activating polypeptide receptor in females (47, 48), 5Îąreductase in males (49), and others (50, 51), confer greater risk for developing PTSD after trauma, but
the effects of genetics alone on PTSD is limited and likely strongly dependent on gene by environment
interactions (52) including epigenetic modifications.
Experience-dependent epigenetic changes, such as changes in gene methylation that correspond
to gene expression (53), occur with PTSD and increase with each traumatic event experienced (54).
Specific types of genes particularly vulnerable to trauma-induced epigenetic changes include the
glucocorticoid receptor (GR) gene, various immune function genes, and sensory perception genes (54–
56). Differential epigenetic states in the genes that regulate the epigenetic process itself may represent
pre-existing risk factors for how an individual responds to trauma (57). Some epigenetic changes remain
even after PTSD symptoms have remitted (58). Increased PTSD risks exist for people whose mother had
PTSD, which are unique from the risks for people whose father had PTSD (59, 60). While trauma-related
epigenetic effects influence parenting behavior (61), these epigenetic imprints themselves are also highly
heritable and passed down through generations (62, 63).
Other physiological risk factors for PTSD include low cortisol level at the time of trauma (64, 65)
and small hippocampal volume (66). While evidence indicates these are pre-trauma risk factors for PTSD,
traumatic stress can also impact cortisol levels and hippocampal volume via changes to hypothalamic-

6

pituitary-adrenal (HPA) axis negative feedback of cortisol and inflammatory responses that decrease
neurogenesis in the hippocampus (67). It is likely that individuals with cortisol and hippocampi on the low
end of the average are simply more affected by the trauma-induced changes to those systems,
representing both a risk factor and consequence of trauma (68).
Nonetheless, to say that only those with a certain genetic background or physiological
predisposition are susceptible to the negative effects of stress likely underestimates the damage that
some severe or repeated stressors can cause. Trauma repeated in occurrence or high in severity may
simply overwhelm even the least vulnerable nervous system. PTSD becomes more likely as individuals
experience more cumulative traumatic events, and experiencing four or more traumatic events is
associated with significantly greater impairment of functioning than experiencing any fewer number of
traumatic events (69, 70). This dose-response of trauma may impart damage by increasing allostatic load
and overwhelming the systems that adapt to environmental challenge and maintain homeostasis (71).
Additionally, some types of traumatic events may constitute a larger “dose” of trauma and are themselves
associated with increased risk for PTSD.
Interpersonal trauma (e.g. physical and sexual assault, emotional abuse, domestic violence) is
unique from other trauma and is the class of events most likely to result in PTSD (72–74). In contrast,
non-interpersonal trauma like natural disasters, accidents, and witnessing trauma all have low conditional
risks for PTSD (4-9%) (1). Sexual trauma is the most common cause of PTSD in both men and women
and accounts for the largest proportion of people with PTSD (1, 70). Moreover, sexual trauma perpetrated
by someone known to the victim is more likely to result in PTSD than sexual trauma perpetrated by a
stranger (75), a phenomenon described by betrayal trauma theory.
Trauma involving social betrayal “violates a fundamental ethic of human relationships” (76) and
presents an especially high risk for the dissociative subtype of PTSD (see section: Dissociative subtype of
PTSD), shame, and more pervasive mental and physical health difficulties than other forms of trauma
(77). Because all interpersonal traumas have some degree of betrayal (i.e. betrayal by fellow humankind),
betrayal trauma theory may explain why interpersonal traumas have such a high risk for PTSD. That
trauma especially high in betrayal (e.g. perpetrated by a partner or family member) represents a risk even
beyond general interpersonal trauma undoubtedly points to the involvement of an individual’s perception

7

of the traumatic event in the trauma sequelae. Indeed, in addition to perceived betrayal, the traumatic
stress response is influenced by perceived level of threat, perceived violations of assumptions, and
perceived level of social support.
Individuals with a high perceived level of danger (those who thought their own lives or the lives of
loved ones were in danger) exhibit a significantly increased risk for PTSD compared to those who did not
perceive a high level of danger (70, 73, 78, 79). Additionally, there is an increased risk for PTSD when the
trauma violates a strongly held worldview (e.g. “My hometown is a safe place.”), and a greater stress
response is elicited when trauma is experienced in a context that previously signaled safety (11).
Subjective perceptions of trauma and the subsequent psychological reactions to trauma are also heavily
influenced by sociocultural norms and widespread ethnocultural variations of PTSD exist (80–82).
While PTSD disproportionally affects socioeconomically disadvantaged populations and people in
conflict zones, (39, 83–86), high-income countries have a significantly greater prevalence of PTSD than
low- or middle- income countries (69). The United States has the second highest worldwide prevalence of
PTSD (behind Northern Ireland and before New Zealand), which may reflect that the PTSD diagnosis is a
Western construct that doesn’t capture the trauma experience of other cultures (i.e. a “culture-bound
disorder”) (81). Alternatively, these nations may have historical and cultural antecedents that predicate a
high risk for PTSD. For example, societies of predominately Western-European heritage de-emphasize
communal living, shared grieving, and other systems of social support in favor of individual responsibility,
which may lead to social isolation and over-reliance on maladaptive coping mechanisms. That people in
Western cultures exhibit a prodromal form of PTSD (see section: Delayed-onset PTSD) more than people
from non-Western cultures might be a reflection of cultural differences in help-seeking behavior or posttrauma resources available (87). Part of the PTSD diagnosis involves respondents reporting distress and
severe role impairment in the domain of work, and disruptions in this domain can be especially distressing
for the working class in capitalist societies where productivity is considered a measure of individual value.
Additionally, many of the nations with the highest PTSD prevalence are also driven by systems that
oppress certain groups of people (i.e. “pathogenic societies”) who are consequently more likely to be
exposed to trauma and to develop PTSD (81). Indeed, in primate social systems with established
dominance hierarchies, individuals of lower social status are faced with persistent uncontrollable social

8

stress such that their adrenal gland weight and circulating glucocorticoid levels are significantly elevated
above individuals of higher social status (88–90), and these biological consequences of social structures
may influence subsequent responses to trauma. While research focuses primarily on those factors that
moderate increased risks for the dysfunctional long-term effects of stress, some of these changes may
actually be adaptive in coping with traumatic stress (91, 92). Whether biological mechanisms confer an
adaptive advantage or an increased risk for psychiatric disorders may be largely dependent on context—
what may be adaptive in one culture or life stage may lead to significant impairment in another (93). And
indeed, life stage is an important mediator of the traumatic stress response (94).
The risk for developing PTSD after interpersonal trauma is even higher if the trauma was
experienced in childhood (75). Early-life interpersonal trauma (age 14 or under) leads to greater
emotional dysregulation and self-destructive behavior compared to early-life non-interpersonal trauma
and any adult trauma (95). While exposure to mild, predictable stressors early in life can be beneficial
(e.g. the “stress-inoculation theory”), chronic adversity or any type of trauma can impart difficulties later in
life including mood and anxiety disorders, susceptibility to drugs of abuse, and learning and memory
problems. (40). While some children fully recover after trauma (37), children exposed to trauma before
age 16 have almost double the rate of any psychiatric disorder compared to those not exposed to trauma
(38), and the effects of childhood abuse can last into adulthood presenting as problems with emotion
regulation, interpersonal skills, sleep difficulties, anxiety, depression, and suicide (96–99). But even if a
child recovers after early trauma, that experience could impart an increased risk for developing PTSD
after subsequent trauma later in life. A history of childhood maltreatment is one of the most robust
predictors of developing PTSD after experiencing trauma in adulthood (73, 100). Similar to the cumulative
effect of trauma on increasing PTSD risk in adults, adverse childhood experiences show a dose-response
in children for increasing risk of a variety of detrimental health outcomes, including PTSD (101, 102).
Some evidence indicates that adult trauma only has a cumulative effect if childhood trauma is also
present (103). As with adult trauma, the mechanisms by which early-life stress can have lasting effects
involves genetic polymorphisms (44, 48, 50) and modification of the epigenome (55, 104, 105). Early-life
stress may also alter the development of brain regions that normally go through postnatal development,
such as the hippocampus and prefrontal cortex (106). While early-life trauma is often thought to act as a

9

kind of toxic agent that damages healthy brain development (e.g. the “glucocorticoid cascade
hypothesis”) (107), these changes may instead promote an “alternative neurodevelopmental pathway”
(e.g. an organizing effect of glucocorticoids) that is adaptive to surviving a high-stress environment (97).
The idea of adaptive responses to trauma can be examined in people who do not develop PTSD after
trauma.
After trauma, most people display symptoms common to PTSD patients, but these effects usually
seem to resolve within a few weeks (108–110), suggesting that at least part of the susceptibility to the
lasting effects of trauma may come not only from vulnerability to the initial impact of trauma but from
factors during the aftermath of recovering from the trauma such as social support. Among 14 separate
risk factors for PTSD, social support is the single greatest factor (100). More specifically, lack of social
support in the form of negative responses from others after trauma significantly predicts PTSD, as
opposed to positive social support, which doesn’t itself protect from PTSD per se but may serve more of a
buffering function (111). Effective social support after trauma may be a mechanism by which an individual
gains a sense of control over an uncontrollable stressor (11). Many individual coping mechanisms are
maladaptive, including thought suppression, avoidance, distraction, and denial, all of which increase
PTSD symptom severity and are often relied upon less in individuals with high levels of social support
(112). Social support on a neighborhood or community-wide level is also important in the PTSD trajectory
(39). While a variety of risk factors influence whether an individual will develop PTSD, even among those
who develop PTSD, symptom severity and presentation can vary tremendously.

Symptoms and subtypes of PTSD. Defining features of PTSD in the DSM-5 include 20
symptoms in 4 symptom clusters: (1) persistent re-experiencing of the traumatic event, (2) avoidance of
stimuli associated with the trauma, (3) negative cognitions and mood, and, (4) increased physiological
arousal (Table 1). Each diagnostic cluster consists of diverse symptoms that have unique presentations in
each individual, and no individual will meet all 20 symptoms at any one time. Indeed, some of these
diagnostic measures seem to oppose one another (e.g. recurrent intrusive memories of the trauma vs.
amnesia of the trauma). There are 636,120 unique combinations of DSM-5 symptoms that could lead to a
PTSD diagnosis (113). To explain this symptom heterogeneity, subtypes of PTSD have been identified.

10

Dissociative subtype of PTSD. The dissociative subtype of PTSD includes the standard DSM-5
criteria plus prominent depersonalization and derealization symptoms, which presents as disruptions in
memory (including amnesia for all or part of the trauma), identity, body awareness, and self-perception.
Depersonalization is an “out-of-body” experience during which individuals may describe observing their
own body from above and is typically accompanied by a cognitive impression that “this is not happening
to me.” Derealization creates the perception that an experience itself is not real or may be a dream, and
both derealization and depersonalization are associated with an attenuation of the intensity of emotional
experience (84).
Over a hundred years ago, Pierre Janet coined the term dissociation as we know it and
developed the original theories of two distinct types of responses to traumatic stress: hysteria,
characterized by the dissociation of feelings or memories related to traumatic experiences, and
psychasthenia, characterized by ruminations, phobias, and anxiety related to traumatic experiences
(114). Janet hypothesized that some traumatized individuals cannot integrate the memories of painful
events and the intense emotions associated with them into their narrative memory—so both the memory
and the accompanying emotions remain dissociated from consciousness (115). Independently from
Janet, Freud drew similar conclusions about trauma and memory when he described patients recalling
memories of childhood abuse who “are recalling these infantile experiences to consciousness…suffer the
most violent sensations of which they are ashamed and try to conceal, and even after having gone
through them once more in such a convincing manner…they have no feeling of remembering the
scenes…and assure so emphatically of their unbelief” (30).
As many as 30% of PTSD patients report with symptoms of depersonalization and derealization,
and individuals with the dissociative subtype of PTSD are more likely to have experienced early-life
trauma (prior to age 14), sexual trauma, and have comorbid MDD (116–120). Additionally, dissociative
PTSD is accompanied by a distinct neurobiological, physiological, and stress-hormone profile (121–125).
Because individuals with dissociative PTSD show blunted, rather than increased, heart rate in response
to stress (126), the ICD-10 classification of PTSD does not include a dissociative subtype but allows for
the diagnosis of PTSD in the absence of hyperarousal symptoms if dissociation or amnesia is present.

11

Peritraumatic dissociation is one of the strongest predictors of subsequent development of PTSD
(73, 79). Dissociation may be a survival mechanism by which consciousness is altered in the face of
overwhelming experience when actual escape is not possible. This may reflect the strong link between
child abuse and dissociation where children are most often abused by caregivers or other power figures,
and because children rely on those adults for survival, traumatic amnesia may not simply reduce suffering
but promote survival (76). Dissociative PTSD most often follows trauma high in social or institutional
betrayal (e.g. abuse that occurs with complicity from a church or educational institution) as an adaptive
response when an individual must “preserve a necessary relationship in the face of mistreatment” and
this most often occurs when the betrayal trauma occurred in childhood (77, 127, 128).
Child subtypes of PTSD. The DSM-IV PTSD diagnostic criteria were developed from adult
literature and did not identify the children most adversely affected by trauma or reflect how trauma
manifests in children (38, 97). The same trauma that leads to hypervigilance in adults leads to
dissociation in children (127), indicating fundamental differences in how trauma is processed across
development. Indeed, people with PTSD who experienced childhood trauma have a completely different
(98% non-overlapping) epigenetic profile than people with PTSD whose trauma occurred only in
adulthood (129). A new diagnosis called Developmental Trauma Disorder was proposed for inclusion in
the DSM-5, citing that children who experience trauma are diagnosed with 3-8 co-morbid disorders that
can all be explained by one etiological factor: trauma (130, 131). Developmental Trauma Disorder was
not accepted as a new diagnosis, but the DSM-5 included a child subtype (or “preschool subtype”)
specification that simply requires fewer symptoms than necessary for an adult diagnosis, but does not
take into account the evidence indicating that PTSD in children is not simply a “miniature” form of adult
PTSD (132, 133). Interestingly, some children who experience trauma do not present with detectable
symptoms until later in life, which may be related to delayed-onset PTSD.
Delayed-onset PTSD. While it isn’t a subtype per se, the delayed-onset specification of PTSD has
been included since the first release of the DSM-IV based on the indication that some individuals do not
meet full PTSD criteria until more than six months (sometimes even years) after the traumatic event (35).
About 25% of people who develop PTSD are considered to have delayed PTSD (87). Individuals with
delayed-onset PTSD usually still experience several PTSD symptoms immediately after the trauma (134,

12

135), but at a clinical subthreshold level that does not progress to full PTSD until some time later, often
after exposure to new traumatic events or reminder triggers (136). The ICD-10 has a separate diagnosis
for “the late chronic sequelae of devastating stress, i.e. those manifest decades after the stressful
experience,” which is classified under “enduring personality changes, not attributable to brain damage
and disease.” Roger Pitman, building off of Eysenck and Kelly’s 1987 hypotheses on the “incubation of
neurotic disorders,” proposed a model for delayed onset of PTSD as a stress-hormone facilitated “positive
feedback loop in which subclinical PTSD may escalate into clinical PTSD” (36). Another possibility is that
what was speculated to be “subclinical” PTSD was really an unrecognized phenotype of PTSD, such as
the internalizing subtype of PTSD.
Internalizing and externalizing subtypes of PTSD. While not yet a formal diagnostic distinction,
subgroups of individuals with either dominant internalizing symptoms or dominant externalizing symptoms
have been revealed as distinct from “simple PTSD” by their presentation of prominent personality
alterations (117, 137). People with simple PTSD have low comorbidity with other disorders and normal
personality functioning (138). The internalizing phenotype is characterized by low-positive emotion,
avoidant personality, and high comorbid depression and anxiety, and the externalizing phenotype is
characterized by low levels of constraint, aggression, antisocial or narcissistic personality, and comorbid
substance abuse (139, 140). The idea that trauma can lead to pronounced personality alterations in some
individuals was addressed in a previously proposed phenotype: complex PTSD. Internalizing and
externalizing phenotypes of PTSD have been found to be an analogous construct to complex PTSD and
may represent subtypes of complex PTSD (138).
Complex PTSD. The complex PTSD construct was proposed by Judith Herman in 1992 for
inclusion in the DSM-IV under “disorders of extreme stress, not otherwise specified (DESNOS)”. Herman
argued that the DSM-III PTSD criteria did not capture the effects of prolonged, repeated trauma such as
those observed in survivors of captivity (prisons, concentration camps, sex trafficking) and domestic
abuse (intimate partner violence, child abuse) who exhibit pronounced changes in personality and identity
and diffuse symptoms in multiple domains (somatic, cognitive, affective, behavioral, relational) (141).
DESNOS was ultimately not included in the manual because most people who met the criteria for
DESNOS also met the criteria for PTSD, so it was thought of as just a more severe form of PTSD (142).

13

Calls for a complex PTSD diagnosis continued during the preparation of DSM-5, as some thought the
practice of diagnosing PTSD along with several comorbidities was less clinically useful than a single
complex PTSD diagnosis (103, 143), and the complex PTSD symptoms of dissociation, affect
dysregulation, and somatization were not captured with the standard PTSD diagnosis. Nonetheless,
complex PTSD was again not included in the DSM-5, but the new PTSD criteria moved closer toward
accounting for complex PTSD symptoms (e.g. the addition of the “negative alterations in cognitions and
mood” symptom category and the dissociative subtype) (35). What still remains to be recognized by any
official diagnosis is the well-documented somatic symptomatology related to trauma: headache,
gastrointestinal disturbance, autoimmune disorder, non-epileptic seizure, and chronic pain (141). The link
between trauma and the various symptoms and phenotypes described here can only be fully elucidated
by research on the pathophysiology of the traumatic stress response.

14

Table 1. PTSD diagnostic criteria in DSM-5. Items highlighted in red are new additions to the DSM-5. Adapted from
Friedman, 2013.
# symptoms
Criterion
Symptom category
Specific symptoms
required

A

Exposure to a traumatic
event

N/A

1.
2.
3.
4.

B

Intrusion symptoms

1

C

Avoidance symptoms

1

D

Negative alterations in
cognitions and mood

2

E

Alterations in arousal and
reactivity

2

F
G

1.
2.
3.
4.
5.
1.
2.
1.
2.
3.
4.
5.
6.
7.
1.
2.
3.
4.
5.
6.

Duration of symptoms
>1 month
Symptoms cause
significant distress or
functional impairment
1.

H

Directly experiencing the event(s)
Witnessing the event(s)
Learning that the event(s) occurred to a close relative or close
friend
Experiencing repeated or extreme exposure to aversive details of
the event(s)
Intrusive distressing memories of the event(s)
Recurrent distressing trauma-related dreams
Dissociative reactions (e.g. flashbacks)
Intense psychological distress when exposed to traumatic
reminders
Marked physiological reactions to traumatic reminders
Persistent avoidance of thoughts and memories
Persistent avoidance of external reminders
Dissociative amnesia of the event(s)
Persistent negative expectations
Persistent distorted blame of self or others about the event(s)
Persistent negative emotional state
Diminished interest or participation in significant activities
Feeling of detachment or estrangement from others
Persistent inability to experience positive emotions
Irritable behavior or angry outbursts
Reckless or self-destructive behavior
Hypervigilance
Exaggerated startle response
Problems with concentration
Sleep disturbance

Symptoms not due to
alcohol, drugs, or
medication

2.
3.

Specify if: dissociative subtype (full PTSD + derealization or
depersonalization
Specify if: preschool subtype (1 B and 2 E, but only 1 C or D
symptoms needed)
Specify if: with delayed expression of symptoms

15

Pathophysiology. While identifying PTSD subtypes has been important in recognizing symptom
heterogeneity and providing more accurate diagnoses and treatment options for patients, PTSD cannot
fully be prevented, diagnosed, or treated without a clear fundamental understanding of the underlying
neurobiology of the traumatic stress response. Studying PTSD pathophysiology in humans is mostly
limited to brain functional magnetic resonance imaging (fMRI) and positron emission tomography (PET)
scans and biomarkers in blood, urine, and saliva. The effects of PTSD are most reliably measured during
“challenge” tests (e.g. traumatic script reminders, glucocorticoid injection), as baseline measures have not
consistently detected PTSD (144). A postmortem PTSD brain collection is currently being built by the
Harvard Brain Bank and Cohen Veterans Bioscience, but to date, no studies from this initiative have been
published. Like other types of stress, responses to traumatic stress involve neuroendocrine activation of
the HPA axis, neuronally mediated activation of the sympathetic nervous system (SNS), and limbic
system responses (110), but the neurobiological underpinnings of these trauma-induced alterations are
very different from those seen in non-traumatic stress responses (145). These systems likely interact with
one another in synergistic and inhibitory ways to modulate the traumatic stress response.
Sympathetic nervous system. Within seconds of stress exposure, catecholamines (e.g.
epinephrine/adrenaline and norepinephrine/noradrenaline) are released from the adrenal medulla through
sympathetic neuron activity (146). In PTSD, the SNS response is exaggerated, exhibited by elevated
resting heart rate and blood pressure, exaggerated acoustic startle responses, and increased skin
conductance, which do not seem to be risk factors for PTSD but indicators of PTSD (110, 112).
Catecholamines enhance memory storage for events associated with emotional arousal (147) and
norepinephrine is specifically associated with the hyperarousal and re-experiencing symptom clusters of
PTSD (148). Flashbacks and panic attacks can be elicited by acute yohimbine injection in PTSD patients
but not healthy controls, indicating an increased noradrenergic sensitivity in PTSD; however, a subset of
PTSD patients do not experience yohimbine-induced reactivity but rather exhibit metaChlorophenylpiperazine (mCPP)-induced flashbacks and panic, indicating enhanced serotonergic
sensitivity (149). This evidence suggests that flashbacks and hyperarousal are not specific to SNS
hyperactivity and distinct biological subtypes of PTSD exist. These SNS responses both influence and are
influenced by HPA responses to stress (110).

16

Hypothalamic-pituitary-adrenal axis. Within minutes after stress exposure, glucocorticoids (e.g.
cortisol) are released from the adrenal cortex via activation of the HPA axis. In response to stress, the
paraventricular nucleus of the hypothalamus (PVN) releases corticotropin-releasing hormone (CRH),
stimulating the release of adrenal corticotropic hormone (ACTH) from the pituitary, which then stimulates
the release of adrenal glucocorticoids. Cortisol in humans is the major adrenal HPA stress response
hormone, and its release is regulated by a negative feedback system modulated by GR in the PVN and
pituitary immediately after a cortisol increase, and again hours later, in distinct processes of fast feedback
and delayed feedback (150). GR in the hippocampus is a primary indirect mediator of HPA negative
feedback (151). This system is specifically disrupted in PTSD, characterized by decreased levels of
ACTH and cortisol due to exaggerated or sensitized negative feedback control over cortisol release,
resulting in persistent low levels of cortisol (152–154), the opposite of what is seen in MDD (155, 156)
and the dissociative subtype of PTSD (125). These differences may reflect that some patients with PTSD
have an increased number of GR and patients with MDD have a decreased number of GR on blood
lymphocytes (157). However, MDD co-occurs in 48% of PTSD patients (1), which may explain some
inconsistencies in the literature regarding the HPA axis and PTSD. Nonetheless, low cortisol is
associated with the avoidance symptom cluster in PTSD (158). Since cortisol potentiates memory
consolidation, low cortisol in response to a traumatic event could contribute to aberrant memory formation
in PTSD (159). Alternatively, low cortisol in response to trauma may permit catecholamines release to
remain unchecked, leading to PTSD symptoms (160), as glucocorticoids, in addition to their function in
the HPA axis, suppress stress-induced increases in catecholamines (161). Indeed, high-dose cortisol
administration in the hours before or after trauma exposure can attenuate or prevent the development of
PTSD symptoms (162, 163). Together, these results suggest that both adrenal glucocorticoid and
catecholamine response at the time of trauma exposure may influence the subsequent trauma sequelae.
While these biomarker tests have provided insight to the biological underpinnings of PTSD, structural and
functional neuroimaging studies of the brain have provided a broader lens of the effects of trauma and
PTSD on psychological functioning.
Limbic system. Limbic system structures including the amygdala, hippocampus, medial prefrontal
cortex (mPFC), and anterior cingulate cortex (ACC) are involved in the emotional and memory-related

17

processes of PTSD and show functional and structural changes with PTSD (9, 164–166). The amygdala
detects threats in the environment, the ACC and mPFC are activated during emotional states and
decision making, and the hippocampus is involved in explicit memory processes and fear memory (144).
The widely-reported smaller hippocampal volume in PTSD patients (see section: Risk factors for
PTSD) was initially thought to be caused by damage from excess glucocorticoids, but as evidence of
chronic low cortisol in PTSD patients mounted, it became clear that this was not the case and lower
hippocampal volume is a pre-trauma risk factor for PTSD (66, 145). Results of hippocampal functional
tests have been inconsistent, with some showing decreased hippocampal activation and gray matter
volume in PTSD and some showing increased hippocampal activation and gray matter volume, possibly
depending on differences in tasks and analyses used (9, 165, 166).
PTSD is traditionally associated with decreased activity in the prefrontal cortex and increased
activity in the amygdala (167–169). While overactive amygdala activity is also observed in people with
social anxiety disorder, phobic disorder, and fear conditioning of healthy subjects, underactive mPFC
activity is unique to PTSD (169). The reduction of mPFC activity and gray matter in PTSD patients is
coupled with reduced ACC activity and gray matter, and some hypotheses posit that this underactivation
of frontal circuits results in amygdala hyperactivity, inhibiting fear extinction (166, 170–172), but others
show no evidence that amygdala hyperactivity is due to a failure of frontal lobe inhibition (173), while still
others show no amygdala hyperactivity at all (9). The pattern of reduced frontal cortex and enhanced
amygdala activity has been associated with an inability of people with PTSD to use contextual cues to
assess safety, resulting in feelings of endangerment in safe contexts, and “low fear” in danger contexts
(174). The prefrontal cortex detects whether a stressor is under a person’s control, and if it is, it inhibits
brainstem stress responses; thus, the presence of control blocks the “default” behavioral sequelae of
uncontrollable stress (19). This evidence argues that “from an evolutionary perspective, it may be
sensible that activation of ‘lower’ centers by strongly aversive events came first, and that as species
developed the ability to cope with such events by behavioral means, inhibition from ‘higher’ centers under
conditions of behavioral coping then developed;” so, if trauma reduces activation of the mPFC, then a
loss in the “perception of control” may follow (19).

18

Recently, however, people with dissociative PTSD have been observed with increased prefrontal
cortex activity and decreased amygdala activity (123, 124), the opposite of the presumed typical PTSD
pattern. What this means for the perception of control is yet to be examined, but this natural dichotomy
will be a useful tool to examine the role of stressor controllability and mPFC activity in PTSD and certainly
strengthens the argument that distinct subtypes of PTSD exist.
While fMRI scans can aid in the diagnosis of PTSD (175), discrepancies in the literature as to
what a “PTSD brain” looks like preclude robust imaging diagnostics, and similar challenges exist for
physiological biomarker tests. One contributor to these inconsistencies is related to the fact that sex
differences in PTSD, while well-documented in epidemiological reports, have received little attention in
PTSD physiological research. To the extent that sex differences have been reported, women with PTSD
almost always show very different responses than men with PTSD.

Sex differences in PTSD. Sex differences pervade nearly every aspect of the traumatic stress
response, from PTSD prevalence to epigenetic changes at the molecular level (176, 177). One of the
most commonly reported epidemiological findings is that PTSD is twice as prevalent in women than it is in
men, even though women are less likely to experience a traumatic event (1, 3, 70, 108, 178, 179). This
increased prevalence in PTSD in women is a reflection of a higher overall conditional risk of developing
PTSD after any kind of trauma in women compared to men, a sex difference that is partially mediated by
women’s higher risk for revictimization (i.e. cumulative trauma effect) (1, 180) and for developing PTSD
after physical attack or threat with a weapon (1). Interestingly, the latter two traumas confer the lowest risk
for developing PTSD among men, which is perhaps a reflection that women experience physical violence
as more threatening than men do (70). Women have been reported to perceive any type of traumatic
event as more distressing than do men (70), and perception of danger is a significant risk factor for PTSD
(78). Indeed, women respond to negative emotional stimuli with greater amygdala activation than men
and men respond to positive emotional stimuli with greater amygdala activation than women (181).
Additionally, women face a more negative social environment (e.g. victim blaming) than men following
trauma (111). A meta-analysis confirmed that decreased hippocampal volume is correlated with PTSD but
revealed that this effect is independent of gender, suggesting that the higher prevalence of PTSD in

19

women is not attributed to decreased hippocampal volume (182), further implicating the amygdala as an
important neurobiological substrate for women’s increased response to traumatic stress. While women’s
increased risk for PTSD certainly calls for more research on the effects of trauma in women, what is
perhaps more intriguing is that, of the women who do develop PTSD, their phenotype is characteristically
and biologically different from men with PTSD.
Since its first inception in 1980, the DSM PTSD diagnosis has reflected the traumatic stress
sequelae in primarily men, with the assumption that women with PTSD would show the same responses.
Much evidence exists to counter this assumption. Indeed, while exaggerated HPA negative feedback is
considered a hallmark symptom of PTSD (145), many women with PTSD do not show this response (125,
183–185). Similarly, an enhanced startle response to an acoustic stimulus is also a presumed core
symptom of PTSD, but women with PTSD have been found to show a diminished startle (186).
Additionally, the symptoms of PTSD may be more severe (116) and persist up to four times longer in
women than in men (187).
Sex differences in the traumatic stress response also interact with PTSD subtypes. Men tend to
exhibit the externalizing PTSD phenotype and women tend to exhibit the internalizing PTSD phenotype
(188). While both men and women can exhibit any of the PTSD phenotypes, what factors contribute to
the observed sex bias in not yet known. Furthermore, the dissociative subtype of PTSD was initially found
in a male sample to be associated with increased severity of flashbacks and exaggerated startle (118),
but when female samples were examined by the same group, the dissociative subtype was not
associated with increased flashbacks or startle (120). Additionally, women are more likely to exhibit
dissociative symptoms, with 30% of women with PTSD fitting the dissociative subtype of PTSD compared
to 15% of men with PTSD, but this could be due to the increased exposure of female sexual trauma in
this sample, which has been specifically linked to dissociation (116, 120). In contrast, a larger worldwide
survey found males were at higher risk for dissociative symptoms (119). Additionally, women but not men
with dissociative PTSD have a higher rate of comorbid personality disorders compared to same-sex
groups without dissociative PTSD (120), further indicating that regardless of differences in prevalence,
women seem to respond to traumatic stress with a different phenotype than men. This is reflected in

20

studies indicating that, in general, men and women with PTSD exhibit different physiological responses to
stressful stimuli (189–192).
Women with PTSD show greater psychophysiological (skin conductance and electromyography)
responses to recollection of personal trauma than men with PTSD, but only men with PTSD show greater
amygdala activation to personal traumatic reminders (193). However, another study showed that women
with PTSD did have increased amygdala activation in response to fearful faces compared to happy faces
(194). This may reflect sex differences in how stressful stimuli are processed, with male amygdalae more
reactive to personal trauma reminders and female amygdalae more reactive to general threats,
highlighting further nuances in the sex-specific amygdala activation mentioned earlier in this section.
Current diagnostic criteria and clinical practices have not been revised to reflect these sex differences in
the trauma response, largely due to a lack of basic research on the sex differences in the traumatic stress
response. The vast extant literature on sex differences in the brain and behavior of rodents provides an
invaluable guidepost for approaching this gap in the traumatic stress literature.

21

Sex differences in rodent brain and behavior

Gonadal hormones, sex chromosomes, and the development of sex differences. The
organizational-activational hypothesis of sexual differentiation posits that some effects of gonadal
hormones that occur early in development permanently and irreversibly influence brain structure and
circuitry (organizational effects), while other effects of gonadal hormones change as circulating steroid
hormone levels fluctuate throughout life (activational effects) (195). In rats, males undergo two
testosterone (T) surges, one during prenatal day 17-18 and another 1-3 hours after birth that, via both
androgen receptors (AR) and estrogen receptors (e.g. the “aromatization hypothesis”), are necessary for
the permanent masculinization of the male rat (196). While feminization was once thought to be a
“default” state that resulted from the absence of testicular secretions (197), feminization is likely an active
process that is independent of masculinization (198). Further evidence indicates that sex differences
resulting from gonadal secretions are downstream from the primary driver: the sex chromosomes.
Sex differences in the mammalian brain and behavior all stem from the “inherent sexual inequality
of the sex chromosomes,” which lead to a host of sexually-biased effects including sex-specific uses of
epigenetic modifications (199), downstream effects of gonadal differentiation and gonadal hormone
secretion, sex differences in autosomal gene expression, and cell-autonomous use of the sex
chromosome complement (199, 200). The postnatal environment is also a source of sex-specific
information, especially in humans where males and females are socially treated differently (200).
Together, these sex differences may create or reduce sex differences in physiology and behavior, and the
well-documented sex differences in the HPA axis are especially relevant to PTSD.

Sex differences in the rat HPA axis. Compared to males, female rats have a higher
concentration of total plasma corticosterone (CORT), which is analogous to human cortisol (201, 202),
and possibly to compensate for this sex difference, females also have higher CORT binding globulin
(CBG) that brings their free CORT levels down to a level similar to that of males (203, 204). However,
elevated CBG doesn’t account for the sex differences in CORT secretion in which females release more
CORT in response to ACTH stimulation (196), which may be a result of hypertrophied adrenal glands in

22

females due to low T (202). In rodents, activational T inhibits the HPA response to stress (202, 205–207)
while activational estradiol enhances the HPA response to stress (206, 208). This gonadal hormonedependent aspect of the HPA response is regulated by AR in males and by GR in females (203, 206,
209–212). It is likely that both AR, which is expressed in the brain of female rodents and humans (213,
214), and ER may have a role in both the male and female stress response (196, 215, 216).
Higher CORT in females may also be due to diminished negative feedback of the HPA axis in
females (203, 210). Indeed, female rats are less sensitive to the negative feedback effects of
glucocorticoids (204, 215, 216). Sex differences in rat CORT binding capacity and GR binding site affinity
in the hippocampus and hypothalamus are regulated by ovarian hormones and may contribute to sex
differences in HPA negative feedback or may be a compensatory mechanism for differences in binding
capacity to prevent sex differences in downstream effects (217). The fast feedback inhibition of the HPA
axis is sensitive to the rate at which plasma glucocorticoids increases, such that inhibition occurs when
the glucocorticoid rate of increase rises above a certain threshold—and that threshold is higher in female
rats than in male rats (150). This sex difference could be a mechanism to compensate for females’ faster
rate CORT increase (203). These sex differences are only partially reversed by ovariectomy, indicating
that ovarian hormones play some role in modulating the HPA axis, but organizational effects are also
likely involved (204). Hormonal fluctuations in the female reproductive cycle have historically been
thought to be a primary source of activational sex differences.

Rat estrous cycle. While the influence of cycling estrogens have been presumed to play a large
role in stress-related sex differences, the overall increased HPA activity in females is apparent in
populations of normally cycling females at random stages of the estrous cycle, which is analogous to the
human menstrual cycle (211, 218). Additionally, most neurobiological measures that show sex differences
are not driven by changes in the female hormone cycle (219), including footshock-induced CORT release
(220), behavioral responses to predator exposure (221), and the higher ACTH and CORT response to
restraint stress in females compared to males (222). However, even though females have a greater
stress-induced activation of the HPA axis compared to males, females exhibit reduced fear-related
behavior and estradiol shows a protective effect in fear conditioning (206, 223, 224) possibly via estradiol-

23

mediated increases mPFC dendritic spine development and axonal branching in response to stress in
females (225). This indicates that while the female reproductive cycle may not play as large of a role in
activational sex differences as previously thought, ovarian hormones such as estradiol clearly do mediate
some aspects of the stress response. The relationship between gonadal hormones and the stress
response is well-documented, but the neurobiological mechanisms underlying these sex differences
needs further basic research, particularly with respect to traumatic stress.

24

Merging the fields of PTSD and sex differences research with animal models

Studying sex differences in rodents. There exists some discord among neuroendocrinologists
about what constitutes “true” sex differences, what they mean, and how to study them. Over the past ten
years, researchers have shifted from suggesting that only those differences initiated by sex chromosomeencoded genes “count” as sex differences, while differences that can be attributed to gonadal hormonal
modulation do not (226) to later differentiating types of sex differences as sexual dimorphisms, sex
differences, and sex convergences/divergences (227) to more recently, simply calling for the use of “sex
as a biological variable,” and acknowledging that contributions of sex chromosomes and gonadal
hormones likely shape a spectrum of differences between sexes (228) while warning that not all sex
differences are “direct, context-independent and persistent” (229). Regardless of the semantics of
defining sex differences, designing a study to detect sex differences can be done in several ways.
Strategies to parse out which sex differences are related to gonadal sex and which are related to
chromosomal sex include using the “four core genotypes” that result from mating a rat with the Sry gene
knocked out (XY- female) with a rat in which the Sry gene is moved to an autosome (XY-Sry male), using
a “classic two-step endocrine experiment” in which the gonads are removed via gonadectomy and
specific hormones replaced (227, 230), and using rodents with a spontaneous Testicular Feminization
Mutation which renders ARs nonfunctional (231). While a handful of studies have used both sexes in
basic PTSD research, to date, no formal sex differences studies have been applied to animal models of
PTSD.

Animal models of PTSD. Because trauma exposure is generally an unpredictable event, most
clinical PTSD studies are retrospective; thus, it is not known whether many of the functional, structural,
and chemical brain abnormalities associated with PTSD are a consequence of the disorder or constitute a
pre-existing condition that predisposes a person to PTSD. For this reason, animal models of PTSD are
invaluable in examining the pathophysiology of PTSD. Moreover, because responses to life threatening
challenges are highly conserved across mammalian species, PTSD may need not necessarily be
“modeled” in animals, as PTSD-like responses are not unique to humans—other animals that survive life-

25

threatening experiences also show lasting biological effects on behavior, reproduction, and other domains
of life (12). Still, it is important to develop criteria to ensure we are studying what we intend to study. In
1993, Rachel Yehuda and Seymour Antleman defined five criteria for evaluating the relevance of animal
stress paradigms to PTSD (232):

(1) Very brief exposure to the stressor should be able to induce the biological and
behavioral symptoms of PTSD. In humans, the same magnitude of PTSD symptoms can
result from traumatic experiences that range in duration from seconds (e.g. traffic accidents)
to months (e.g. captivity), so the extent to which a stressor is traumatizing should predict the
development of PTSD rather than the duration of the event (i.e. acute or chronic).

(2) The stressor should produce PTSD symptoms in a dose-dependent manner, in which
the measured endpoints of an animal model respond differentially to different levels of
the same stressor. In humans, PTSD only occurs after experiencing a threshold dose of
stress and there is evidence for a relationship between stressor intensity and severity of
PTSD symptoms.

(3) The stressor should induce PTSD symptoms that persist or increase in severity over
time. Any biological changes that occur immediately following stressor exposure and return
to baseline shortly thereafter can be considered a normal acute stress response. In humans,
the onset of PTSD could occur immediately following trauma exposure or could be delayed
for months or years, but the symptoms can persist for many years; thus, the time since
stressor exposure is correlated with development of PTSD symptoms.

(4) The stressor should be capable of inducing both increased and decreased
environmental responsiveness to stimuli that recall the stressor. In humans, both
increased (i.e. intrusive re-experiencing, hyperarousal) and decreased (i.e. avoidance,
numbing) responsiveness to stimuli can exist concurrently but usually alternate between
dominant re-experiencing and dominant avoidance.

26

(5) Individual responses to a particular stressor should vary based on previous
experience or genetics. A traumatic experience can induce PTSD in some people, while
other people experiencing the exact same event do not develop PTSD or develop unique
individual phenotypes, suggesting that factors other than the trauma contribute to the
induction of PTSD.

In addition to these criteria, Siegmund and Wotjak suggested that a PTSD animal model should
exhibit both the associative (re-experiencing, avoidance of and exaggerated response to trauma
reminders) and non-associative (hyperarousal, irritability, hypervigilance, increased startle, emotional
numbing, social withdrawal) components after trauma exposure (233). Animal models used in PTSD
research include predator exposure, footshock, single prolonged stress, and fear conditioning. Because
the fear circuit is conserved across most vertebrate species, fear conditioning in animal models has been
suggested as clinically applicable way to explore mechanisms of the fear circuit in PTSD (167), but
because fear conditioning involves repeated exposure to an unconditioned stimulus in order to produce
the fear response, this may not be an appropriate model for PTSD unless a single exposure to the
unconditioned stimulus can induce PTSD symptoms in a manner defined by the previously described
criteria. Fear conditioning is likely more useful in examining the phenomenon of “triggers” that induce
PTSD symptoms rather than being used as a PTSD model itself.
Single prolonged stress. Single prolonged stress (SPS) is one of the most replicated and wellvalidated models for PTSD, and experimental evidence supports the face, construct, and predictive
validity of the model (234). The effects of SPS recapitulate the PTSD symptoms in humans, including
increased anxiety behavior (235–238), enhanced arousal (239, 240), and increased negative feedback of
the HPA axis (241, 242), with the caveat that these responses can only be assumed to relevant to males.
In humans, at least one month post-trauma exposure is required for a PTSD diagnosis (DSM-5).
Similarly, in rodents, a minimum of 7 days post-SPS (equivalent to one month in humans) is required to
produce the full PTSD-like response in regard to HPA axis dysregulation (242) and fear extinction deficits
(243). Selective serotonin reuptake inhibitors and atypical antipsychotics are the standard

27

pharmaceutical treatment for PTSD (159, 244, 245), and these same drugs reduce the SPS-induced
enhanced fear memory and anxiety behavior in rats (246–248).
The SPS model meets all five criteria proposed by Yehuda and Antleman. The SPS paradigm
employs psychological (restraint), physiological (forced swim), and chemical/endocrinological (ether)
stress, which are often all simultaneously part of the human traumatic experience. The combination of
the three SPS stressors is required to produce the full PTSD phenotype, as any combination of two of the
three stressors does not produce all of the effects observed with the full SPS paradigm (243). Using three
different stressors, as with SPS, can also reduce the risk for habituation that sometimes occurs in other
PTSD models that use repeated footshock or predator exposures (234).
Predator exposure. The predator exposure (PredX) model is another valid animal model for
PTSD (249). A single, short (10-60 min) exposure to a predator or predator scent in a closed (i.e.
inescapable) environment results in lasting changes in HPA axis activity, startle response, anxiety
behaviors, social interaction, conditioned fear responses, prefrontal cortex activation, and hippocampal
cell morphology (250–255). These responses to predator stress exposures can be prevented with earlyintervention selective serotonin reuptake inhibitor treatment (256), but as with SPS, can only be assumed
to be relevant to males.

Measuring responses to traumatic stress. While PTSD shares many symptoms with other
psychiatric disorders, increases in acoustic startle response (ASR) and HPA negative feedback are two
that are unique to PTSD and readily quantifiable in both clinical and preclinical research settings.
Dexamethasone suppression test. One of the most robust measures used in the SPS paradigm is
the dexamethasone suppression test (DST) (235, 239, 240), which surprisingly has not been tested in
any PredX paradigm. The DST is a tool used in clinical and experimental studies to detect disruption of
the HPA axis (257). Dexamethasone (DEX) is a pituitary GR agonist (258) that diminishes any
subsequent CORT response via negative feedback. In humans, high-dose DEX administration at night
normally results in very low baseline CORT levels the next day; thus, shifting down the post-stress CORT
levels (259). But under conditions of acute stress such as a CRH injection (260), patients with MDD show
no CORT suppression with DEX (DEX non-suppression) (257, 261–263), whereas patients with PTSD
typically show exaggerated suppression of CORT with DEX (155, 156, 160). While a high dose of DEX

28

(1.0mg in humans) is more sensitive in detecting DEX non-suppression, a low dose of DEX (0.5mg in
humans) is more sensitive in detecting exaggerated DEX suppression (156). Interestingly, as many as
10% of healthy individuals are reported to be DEX non-suppressors (257), and women are more likely to
be DEX non-suppressors than men (261). Women with PTSD do not show any different response to the
DST compared to women without PTSD (183). Post-menopausal women are more likely to have DEX
non-suppression but age does not appear to affect males (264).
Acoustic startle response. The startle response is a brainstem reflex to sudden stimuli
manifesting as a whole-body or an eyeblink reflex, and the exaggeration of this response is a hallmark
symptom of PTSD that indicates trauma-induced increased arousal and fear (265–268). In humans,
eyeblink reflex is the most common method for measuring startle response, and in rodents, whole-body
movement is the most common method for measuring startle response. However, women with PTSD are
less likely to show enhanced ASR, and may in fact show a diminished startle (186). The ASR is a reliable
measure for enhanced arousal in both the SPS and PredX models, and the sex differences in both the
ASR and DST warrant more investigation as a potential mechanism by which to measure sex differences
in the traumatic stress response.

29

Overview of chapters

While much attention has been given to how sex differences in gonadal hormones may influence
sex differences in the prevalence of depression (269), the same consideration has not been given to the
influence of gonadal hormones on the traumatic stress response. Indeed, gonadal hormones modulate
negative feedback of the HPA axis (264, 270, 271), and to the extent that gonadal hormones have been
studied relevant to PTSD, the focus is limited almost entirely on the female menstrual cycle with no direct
sex comparisons to men (272–275). The few studies that have examined the role of T in PTSD are limited
only to men, with no comparisons to women (276). While some evidence indicates that T in healthy males
correlates with increased fear reactivity (276), treating healthy women with T reduces fear and startle
(277, 278), indicating sex differences in how T mediates stress responses. The following experiments will
begin to address this gap in the understanding of the traumatic stress response.
Chapter 2 begins by characterizing sex differences in response to SPS in gonadally intact, adult
male and female Sprague Dawley rats with the DST, ASR, and relevant measures in the brain (cFos
activation and GR expression). The potential sex-specific effects of social support are tested, and
whether measures of depression (sucrose preference and social interaction) capture the traumatic stress
phenotype in females is examined. Finally, whether SPS induces conditioned anxiety behavior in male
and females is tested.
Chapter 3 characterizes sex differences in a PredX model of PTSD to determine whether the sex
differences found in SPS are generalizable to another type of traumatic stress. The DST, ASR, and GR
expression in the PVN are examined.
Chapter 4 examines the role of gonadal hormones in mediating the traumatic stress response
using a classic-two step endocrine experiment in which adult male and female rats are gonadectomized
and later exposed to SPS. A subset of rats receive T replacement to determine the role of activational T
in the traumatic stress response.
Chapter 5 presents concluding remarks on the interpretation of the data presented and future
directions.

30

CHAPTER 2: CHARACTERIZING SEX DIFFERENCES IN RESPONSE TO SPS

Introduction
The neurobiological basis of marked sex differences in PTSD is not understood (178). While
women are twice as likely as men to develop PTSD following a traumatic experience and tend to
experience different symptoms and comorbidities than men (1, 139, 140, 279), current understanding of
PTSD is largely based on studies of males. Recent reviews emphasize the need to examine the
neurobiology behind sex differences in PTSD (280), but this appeal has been largely unfulfilled. While the
animal literature is replete with reports showing various forms of stress affect males and females
differently, with often opposing outcomes on behavior, physiology and the brain (281–283), focus has
been largely on acute or chronic stress as models of anxiety and depression, with negligible attention
given to traumatic stress and how it might also affect males and females differently. One of the most wellvalidated and commonly used rodent models of PTSD is SPS. To date, over 140 published SPS studies
have been published, but only six have used females (223, 284–287), with only one directly comparing
males and females (288). It is clear that the sex-specific effects of traumatic stress merit further
investigation.
Measures of hyper-responsiveness to stressful stimuli, including enhanced ASR and exaggerated
negative feedback control of the HPA axis, are presumed core attributes of PTSD and readily observed in
men with PTSD (159) and trauma-exposed male rodents (239, 256). However, whether these same
readouts reflect the effects of trauma in females is unclear (289). Indeed, more than half of women with
PTSD do not show the male-typical increase in negative feedback of the HPA axis (183, 290). Similarly,
women with PTSD are less likely to show enhanced ASR, and may in fact show a diminished startle
(186). Current diagnostic criteria and clinical practices have not been revised to reflect these sex
differences in the trauma response, largely due to a lack of basic research on females.
To address this issue, we examined the response of male and female rats to the same traumatic
stress, SPS. We began by directly comparing intact males to intact, normally cycling females, since sex
differences in fear-conditioning are present regardless of estrogen levels in women with PTSD (190). We

31

now report robust sex differences in the traumatic stress response at every level of analysis, from
behavior to the stress hormone response to cellular measures in the brain.

Methods and Materials
Experiment 1: Sex differences in the SPS model. Refer to Fig 1 for experimental timeline.
Animals. 8wk old adult Sprague-Dawley male and female rats were housed in same-sex pairs on
the day of arrival and handled 3min daily for one week before any testing or stress exposure. Rats were
purchased from Charles River and housed with 12h reversed light-dark cycle, ad lib food and water. Cage
bedding was changed weekly and no testing was conducted on days of cage changes. All behavior tests
were conducted in the dark-phase > 2h after beginning of dark phase. Female rats were freely-cycling
and assigned to treatment groups without regard to estrous cycle stage. All animal procedures and care
met or exceeded the NIH guidelines, were approved by Michigan State University Institutional Animal
Care and Use Committee.
Acoustic startle response. ASR was tested as previously described (239). Rats were placed in a
Plexiglas tube attached to an accelerometer inside a dark, soundproof chamber (SR-Lab, San Diego
Instruments) and allowed to acclimate for 5 min (68 dB background noise) before delivery of a startle
stimulus (50 ms burst of 110 dB white noise every 30 sec for 15 min). The chamber and Plexiglas tube
were cleaned with 70% ethanol between each test. Peak whole-body startle response was recorded
every 1 ms for 100 msec, beginning with each startle stimulus. The average peak value for each rat was
normalized to body weight. Baseline measures were taken on the day before stress exposure, with rats
randomly selected and counterbalanced by group and cagemates tested simultaneously in two separate
chambers. Rats were then assigned to control or stress groups so that each group had equal average
ASR (291). ASR testing was repeated 11 days after stress exposure in the same order as baseline ASR.
SPS paradigm. All rats were singly housed immediately before exposure to SPS or control
conditions to eliminate any sex-specific effects of social support on the stress response. SPS was
performed as previously described (241). SPS consists of a psychological stressor (2 hr tube-restraint
stress), a physiological stressor (20 min group forced swim in 24°C water, n=6 same sex rats per 75-liter
tub with 28cm water depth), and a chemical stressor (brief exposure to diethyl ether until immobile and

32

lacking toe-pinch response). Rats were given a 15min rest period in their home cages after the forced
swim before ether exposure. Ether was used as a chemical/endocrinological stressor to recapitulate an
important phenotypic aspect of PTSD and not as an anesthetic. Control rats were also similarly removed
from the vivarium for 2.75h. Rats were left undisturbed for one week after SPS, a requisite delay for the
long-lasting PTSD responses to develop (239, 243).
Dexamethasone suppression test. To assess the strength of negative feedback control of the
HPA axis, the DST was done two wks post-SPS, as previously described (239). DEX (Sigma-Aldrich) was
dissolved with ethanol and diluted to 5% in sterile saline. Low-dose DEX (0.05 mg/kg, i.p.) or vehicle was
administered 2h prior to 30min tube-restraint. Tail-nick blood samples were collected at 0 and 30min of
restraint. Blood samples were collected in the rats’ dark phase, matching the time of day across
experimental groups. Plasma CORT levels were determined using an enzyme immunoassay kit. Rats
were sacrificed 30min after the restraint ended by pentobarbital (i.p.) overdose 30min after the end of
restraint (1h after restraint onset) then intracardially perfused with saline and 4% buffered
paraformaldehyde, with brains harvested for IHC staining.
cFos and GR immunohistochemistry. Only rats that received vehicle injections were used to map
specific neuronal populations activated by restraint stress and expressing GR. Brains were sectioned and
labeled for cFos or GR, as previously described (292). A rabbit IgG polyclonal cFos antiserum (1:10,000;
Santa Cruz Biotech, cat# sc-52) and an immunohistochemistry (IHC) approach was used to visualize
cFos expression with biotinylated goat-anti-rabbit antibody (1:500; Vector Labs; cat# BA-1000) and
avidin-biotin complex Vectastain Elite ABC kit (Vector Labs, cat# PK-6200). GR IHC used the same basic
protocol, with a GR primary antiserum (1:2500; rabbit polyclonal IgG; Santa Cruz Biotech, cat# M-20) on
alternate sections from the same brains. Specificity of cFos staining was confirmed by observing a loss of
nuclear staining in the suprachiasmatic nucleus when the cFos antiserum was preadsorbed with the
immunizing peptide. Specificity of GR staining was confirmed by observing a loss of nuclear staining
when the GR antiserum was preadsorbed with the immunizing peptide and observing the expected
regional staining (e.g. in the dentate gyrus and CA1 but not CA3).
Using a stereotaxic rat brain atlas (Paxinos and Watson 6th ed., 2007) regions of interest were
defined as follows: the mPFC included the prelimbic (PrL) and infralimbic (IL) areas before the corpus

33

callosum crossed midline (5.16 - 2.52 mm bregma), the dorsal hippocampus included the CA1, CA2, and
dentate gyrus regions before the lateral ventricle appeared next to the optic tract (-1.72 – 3.80 mm
bregma), and the amygdala included the central, basal, and medial amygdalar nuclei in the same tissue
sections as the hippocampal regions of interest. These distinct landmarks allowed for the collection of
comparable sections from each animal. Four serial sections (thickness 30 Âľm) from a 1 in 6 series were
analyzed for each region of interest in each animal. A Zeiss Axioplan II microscope equipped with a
motorized stage and an MBF CX9000 digital video camera was used to quantify immuno-labeled cells
within each region of interest. Stereo Investigator software (SI v. 10.55, MBF Biosciences, Williston, VT)
was used to guide the microscope with 1Âľm precision to trace the perimeter of each region of interest in
serial sections at low magnification (2.5x). After tracing regions of interest, an optical fractionator probe
(SI v. 10.55) inserted a counting frame at a random location within the outlined region, and the number of
labelled cells within the counting frame were identified by the observer at 20x magnification and recorded
by the software. Criteria for identifying a labelled cell included distinct black nuclear staining within the
plane of focus of the counting frame. Counting frame dimensions were 130 Âľm x 130 Âľm with a height of
10 Âľm, which allows for the inclusion of ~3 cells per counting frame, and the coefficient of error
(Gundersen m=1) for each hemisphere was at or below 0.10. A 2Âľm no-counting guard volume was
employed at the top and bottom of the 3D counting probe to avoid edge effects that arise from distortion
of objects passing through the leading edge of the knife during sectioning. The systematic and random
counting frame placement proceeded at fixed intervals for a total of 25-30 frames to generate an
unbiased estimate of labelled cells within each region of interest. Sampling parameters were optimized to
minimize systematic error from oversampling while still allowing for reproducible estimates of labelled-cell
density, as determined by the Gundersen coefficient of error. The observer was blind to the experimental
condition of the tissue. In some brains, the entire region of interest was not available for analysis (due to
loss during processing); for those brains, only the regions in which tissue was fully intact were analyzed.
This method allows for a precise estimate of the number of labelled cells within a known volume. Total
immunoreactivity for each hemisphere of each region was quantified by stereological analysis as the total
number of labelled cells counted in each region divided by the total volume of each counting frame

34

analyzed (added for each of the 4 serial sections) to obtain an estimate of the density of labelled-cells in
each region of interest for each animal:

Density of labelled cells =

𝑵𝒖𝒎𝒃𝒆𝒓 𝒍𝒂𝒃𝒆𝒍𝒍𝒆𝒅 𝒄𝒆𝒍𝒍𝒔
𝑽𝒐𝒍𝒖𝒎𝒆 𝒔𝒂𝒎𝒑𝒍𝒆𝒅

Statistical analysis. Two-, three-, or four-way ANOVAs were run for comparisons in intact rats.
See Table 2 for all statistical tests performed. For brain measures, if no main effect or interactions of
hemisphere was present, data were collapsed across hemisphere. The conservative Bonferroni test was
used to correct for multiple tests to hold alpha at 0.05.
Experiment 2: Effect of social housing on female SPS response. The same experimental
timeline from experiment 1 was followed in experiment 2 (Fig 1), with the addition of two measures.
Because female rats exposed to SPS in experiment 1 showed a stress hormone response reminiscent of
depression, we also enlisted new measures (sucrose preference and social interaction) not typically used
to assess the effects of trauma.
Animals. 8wk old adult Sprague-Dawley female rats were housed in pairs on the day of arrival
and handled 3min daily for one week before any testing or stress exposure. Rats were purchased from
Charles River and housed with 12h reversed light-dark cycle, ad lib food and water. Cage bedding was
changed weekly and no testing was conducted on days of cage changes. All behavior tests were
conducted in the dark-phase > 2h of dark. Female rats were freely-cycling and assigned to treatment
groups without regard to estrous cycle stage. All animal procedures and care met or exceeded the NIH
guidelines, were approved by Michigan State University Institutional Animal Care and Use Committee.
ASR, DST, and IHC. The same outcome measures used in experiment 1 were assessed in
experiment 2: ASR, DEX suppression test, and cFos and GR immunohistochemistry.
Sucrose preference test. To measure anhedonia, rats were given access to two bottles containing
either 0.8% sucrose solution or tap water for 24h. To account for any circadian influences and sidepreferences, the two water bottles were presented halfway through the rats’ dark cycle and the bottle
position switched after 12h, as previously described (293). The bottles were initially weighed and a final
weight taken at 24 hrs. Sucrose preference was calculated as % total intake. To minimize the influence of
metabolic factors and reduce the effect of acute stress, food and water were available ad lib.

35

Social interaction test. Each test rat was placed in a clear Plexiglas box at the end opposite to an
empty wire enclosure. The test arena was a 60cm 3 clear Plexiglas box with an interaction zone measuring
8cm out from the edge of a circular wire enclosure. After 2.5 min, a probe rat matched for sex and age
was placed inside the wire enclosure for an additional 2.5 min, as previously described (294). Videos
were analyzed for measures of social interaction. Social interaction behaviors analyzed included latency
to first enter the interaction zone, # of social interactions, and total time spent in the interaction zone.
While total social interaction time is a general measure of anxiety in rats (295), latency to first contact is
another valuable measure of social interaction that captures a different aspect of the stress response
(296). As rats generally approach more quickly and spend more time interacting with a social target
compared to a non-social target (in this case, the empty wire enclosure), the ratio of time spent in the
interaction zone in the presence vs the absence of a social probe rat gives a readout of social interaction
while accounting for individual variability in level of activity (297–299).
Statistical analysis. The same statistical analyses performed in experiment 1 were performed in
experiment 2, except instead of using sex as a factor, housing (single or pair) was the independent
variable. See Table 3 for all statistical tests performed.
Experiment 3: Conditioned fear in SPS. In a separate cohort of 48 rats (24 of each sex), the
conditioned behavioral response to SPS was recorded by pairing the SPS procedure with a neutral tone.
A neutral tone of 2kHz, 80dB 5-pulse beeps (500ms silence in between 750ms tone duration, NCH Tone
Generator software) was played for the duration of the SPS paradigm. Control rats were exposed to the
neutral tone for the same amount of time the SPS-exposed rats were exposed to the tone (2 hours 45
minutes). The only outcome measure of this experiment was the open field test, which was conducted
starting on day 8 post-SPS.
The test arena was an empty white plastic box (122cm x 122cm) illuminated by a dim red light
above and recorded by an overhead video camera. The floor of the arena was marked by a grid
(individual grids were 20.3cm x 20.3cm) to demarcate entries into the center area. The center zone
(approximately 20% the size of the open field chamber) was used to assess anxiety-related behavior.
Each rat was placed into the corner of the empty arena and allowed to explore. After 5 minutes, the rat
was returned to its home cage and the arena cleaned with 70% ethanol. After three minutes, the rat was

36

returned to the same corner of the arena, which now contained a novel object in the center of the
chamber to assess behavioral response to a mild stressor, as previously described (292). An
unconditioned tendency for “wall hugging” in an open field is a measure of anxiety-related behavior in rats
(300). The duration of time spent in the center of the testing chamber, which is a measure of exploratory
behavior and anxiety (with time spent in the center inversely related to anxiety behaviors), was recorded
and analyzed from video recordings with Noldus software. The ratio of time spent in the center of the
arena in the presence vs the absence of a novel object rat gives a readout of anxiety and exploratory
behavior while accounting for individual variability in level of activity. Half of the rats were tested while the
neutral tone from the SPS procedure was played while the other half were tested without the neutral tone
playing. To account for any effect of noise alone, white noise background was continuously played
throughout all testing sessions. This procedure was conducted for three consecutive days to asses if any
contextual reminder-induced anxiety would be extinguished by repeated exposure.
Separate repeated measures three-way ANOVAs (SPS x day x sex) were run for comparisons in
rats exploring the open field with no contextual reminder tone and for rats exploring the open field with the
contextual reminder tone. Only day 1 and day 3 behavior were analyzed. The conservative Bonferroni
test was used to correct for multiple tests to hold alpha at 0.05.

Results
Males and females respond differently to SPS. As expected, males exposed to SPS showed
an enhanced ASR (Fig 2A), replicating a well-established effect of SPS (239). In females, on the other
hand, SPS had no effect on ASR (Fig 2A). The DST also revealed the expected enhanced sensitivity to
DEX in SPS-exposed males, blocking the restraint-induced increase in circulating CORT levels typical of
control males (Fig 2B). In contrast, females exposed to SPS showed a reduced sensitivity—DEX
pretreatment, which competes with endogenous CORT for GR binding, had no effect on stress-related
CORT levels in SPS-exposed females (Fig 2B). In short, these data show the expected exaggerated DEX
suppression of CORT in SPS-exposed males but not females. This sex difference in DEX sensitivity is
due to trauma exposure per se and not because females are resistant to DEX, since DEX lowered postrestraint CORT levels in control females and drove baseline CORT levels down in both males and

37

females, SPS-exposed or not (Fig 2B). SPS did not affect baseline CORT levels in either sex, as
previously shown for males (241) and control females showed significantly higher baseline and poststress CORT levels than males (Fig 1B), as previously reported (211, 216). Additionally, body weight was
not a factor (Fig 11).
These striking sex differences in response to SPS suggested that SPS might also affect cellular
measures in the brain differently in males and females. This is indeed what we found when we examined
GR and cFos expression in brain regions implicated in the traumatic stress response (159, 301). SPS
moderately increased the number of GR-expressing neurons in the PVN of males (P=0.06), but
significantly decreased their number in females (Fig 2C). GR expression in the PVN of control
(unstressed) females was also higher than in control males (Fig 2C). SPS had surprisingly little effect on
the cFos response to acute restraint stress of males, including in the PrL or IL sugbregions of the mPFC
(Fig 2D), the basolateral amygdala (BLA) and medial amygdala (MeA; Fig 2E, F). On the other hand, SPS
extensively affected cFos expression in females, increasing the number of cFos+ neurons more than a
week after trauma in both the PrL and IL (Fig 2D), and in the right BLA (Fig 2E) compared to control
females. The number of cFos+ neurons induced by acute stress was also sexually differentiated in the
mPFC of control rats, with control females having fewer such neurons than control males (Fig 2D), as
previously reported (220, 302). GR expression in the CA1/2 region of the dorsal hippocampus was
differentially affected in males and females with SPS decreasing the number of GR+ neurons in males but
increasing their number in females (Fig 2G), as previously reported (288). Taken together, we found sex
differences in the response to SPS in every outcome measure examined.

Neither housing nor bright lights affect the female response to SPS. Females may have
responded differently to SPS than males because of being singly housed, so we next compared the
effects of SPS in females that were single- or pair-housed. Because our results based on single-housed
males replicated those of group-housed males (240), we did not include a cohort of males in this next
study. We also tested the effect of bright ambient light which can enhance acoustic startle for females
(303), thinking that this might reveal an effect of SPS on the ASR in females that was masked in the dark.
All other conditions were the same as in the first study. ASR in females, even under bright lights, was not

38

affected by SPS or housing condition (Fig 3A). Likewise, DEX failed to block a stress-induced CORT
response in SPS-exposed females (Fig 3B), as seen previously (Fig 2B), although CORT levels overall
were lower in this study. DEX again lowered baseline CORT levels to near zero in all groups, and neither
housing condition nor SPS affected baseline CORT levels. Thus, female rats show a unique response to
SPS that is distinct from the male phenotype; neither acoustic startle nor HPA negative feedback
captured the effect of SPS in female rats.
Because SPS reduced the sensitivity to DEX for females in the first study, suggestive of DEX
non-suppression, a marker of depression in humans (261) and rodents (304), we added two measures
routinely used to assess a depressive phenotype in rodents, the social interaction and sucrose preference
tests (295, 305). Contrary to expectation, SPS had no effect on sucrose preference in either housing
condition (Fig 3C). On the other hand, social interaction was affected by SPS in females, although the
direction of effect depended on housing. For pair-housed females, SPS increased the latency to approach
a novel rat (social target), suggestive of an anxious phenotype, but for singly housed females, SPS
decreased the latency to approach a novel rat, perhaps reflecting an inclination to seek social support
(Fig 3D). Because neither housing nor SPS affected the latency to approach the empty enclosure (lacking
a social target) (Fig 3E), the effect of SPS is on social interaction per se. These data indicate that social
interaction is a sensitive readout of the traumatic stress response for female rats exposed to SPS. Body
weight was not affected (Fig 12). Total time spent with a social target was not affected by housing or SPS
(Fig 13). For experimental consistency, hereafter we continued to use single housing.

Traumatized rats are sensitive to contextual reminders of SPS. With no contextual reminder
of the SPS, both unstressed controls and SPS+tone-exposed rats spent more time in the center of an
open field when there was a novel object to explore relative to when the arena was empty, regardless of
sex, and this behavior was consistent through all three days of testing (Fig 4A). When the contextual
reminder tone was played during the OFT, unstressed control rats behaved similarly to when there was
no tone; spending more time in the center of the open field when there was an object consistently through
all three days of testing, regardless of sex (Fig 4B). On the other hand, both males and females that had
been exposed to SPS+tone a week prior, spent significantly less time with the novel object on the third

39

day of testing compared to the first day of testing, suggesting that with repeated contextual reminders of
SPS, they showed sensitized anxiety behavior in an exploratory task (Fig 4B).

Discussion
Sex differences in the traumatic stress response are among the most widely reported phenomena
in epidemiological and clinical studies, but the neurobiological basis for these differences is unknown,
largely due to an overwhelming male bias in the research (289). We find robust and comparable sex
differences in the traumatic stress response to SPS that recapitulate sex differences in PTSD
symptomology. Male rats showed a hyper-responsive phenotype (ASR and negative feedback control of
CORT levels were both enhanced) typical of PTSD in men (160, 239) but female rats showed a unique
phenotype with more affective behavioral changes (306). Trauma had no effect on the ASR or negative
feedback control of CORT in female rats, aligning well with the increased likelihood for women with PTSD
to show more “internalizing” characteristics. The only measure that did not show sex differences was the
open field test, which interestingly indicated that both males and female anxiety behavior can become
sensitized over time to otherwise innocuous situations when a neutral contextual reminder of their
previous trauma exposure is repeatedly present. This evidence has implications for the use of prolonged
exposure therapy, which has been shown useful in some studies but not in others.
These sex differences in behavior and physiology were accompanied by sex differences in how
the brain responded to trauma, including increased activity in the mPFC and in the right amygdala only in
females. The right amygdala is uniquely implicated in contextual fear-conditioning in rats (307) and
humans with PTSD (308). Notably, a recent study also shows no apparent effect of trauma on cFos
activation in these same brain regions of males, but a minority (15%) of trauma-exposed males show an
anhedonic phenotype with increased cFos in both IL mPFC and BLA (309), similar to what we see after
SPS (Figure 2D). Additionally, males and females recruit different aspects of the mPFC-amygdala during
fear conditioning and extinction (283). These data are consistent with the idea that anhedonia associated
with mPFC and amygdala activation may be relevant markers of trauma pathology for females but not
males.

40

We also find divergent effects of trauma on GR expression in the PVN and in the CA1/2 region of
the dorsal hippocampus, reflecting yet another sex-specific response that likely alters the sensitivity of the
PVN to stress hormones, and may explain why trauma-exposed males show enhanced negative
feedback control of the HPA axis while trauma-exposed females do not. Notably, PTSD patients have an
increased number of GR and increased sensitivity to glucocorticoids, while depressed patients have a
lower number of GR+ blood lymphocytes (153, 157), but the effects of stress on GR are likely regionspecific in the brain, as our data indicate. Our discovery of such sex differences in the trauma response of
rats aligns well with the female-biased internalizing and male-biased externalizing phenotypes identified
in people with PTSD (188) and in the general population (310). Because non-stressed females have
higher baseline and post-stress CORT levels than males, higher GR expression in the PVN, and lower
cFos expression in the mPFC after acute restraint than males, complimenting previous findings (216, 220,
222, 311), such sex differences may well predispose males and females to respond differently to
traumatic stress. These results align with other studies showing opposite effects of acute stress on males
and females (312, 313).
Because major depression is highly comorbid with PTSD, which is female biased like PTSD (1,
183, 314, 315), we included standard measures of depression not typically used to assess the effects of
traumatic stress on rodents. SPS affected social interaction in females; however, this effect depended on
housing conditions, with SPS decreasing their latency to socially interact when single-housed but
increasing latency when pair-housed, which is a measure of time spent evaluating safety cues (316).
While the increased latency to approach a social target shown in SPS-exposed pair-housed females
could be indicative of anxiety-like behavior or the social withdrawal symptom of PTSD, the decreased
latency to approach a social target shown in SPS-exposed single-housed females may seem more
perplexing, as social support in humans is a robust protective factor from developing PTSD. However,
while perceived or received social support may impart resilience to PTSD, social support-seeking
behavior as an early coping mechanism after trauma is actually a risk factor for PTSD, as the need for
social support may be greater in individuals more adversely affected by trauma and they may or may not
receive meaningful support (317, 318). So, social support seeking behavior in itself is not protective, and
whether an individual seeks social support or withdraws from social support may depend on the social

41

context of their lives before trauma. Contrary to expectation, SPS had no effect on sucrose preference,
suggesting that the female response to SPS may not be a depressive, anhedonic-like response.
The diametrically opposed responses to trauma exhibited by male and female rats conforms with
the well-known opposing responses to acute and chronic stress of males and females (319, 320). While
the same neural substrates may be enlisted to manage stress, the specific mechanisms or outcomes
within these substrates seems fundamentally different in males and females (283, 321). This conclusion
has wide reaching implications for therapeutics to treat PTSD in men and women.

42

Veh Blood
or DEX sample

Quiescent period

Day to
SPS

Blood
sample Sacrifice

Restraint

Hour

-2

0

9

11

0.5

1

Daily handling

-7

-1

0

Baseline SPS
ASR

1 2 3 4 5 6 7 8

Social ASR
Sucrose
preference interaction

13

DST

Figure 1. Timeline of procedures for experiments 1 and 2. Experimental timeline begins with daily handling one week before (-7) single prolonged stress (SPS)
and baseline acoustic startle response (ASR) testing the day before (-1) SPS. In both experiments 1 and 2, post-stress ASR is conducted 11 days later and
dexamethasone suppression test (DST) 13 days later. In experiment 2, sucrose preference is tested 8 days after SPS and social interaction 9 days after SPS.

43

B

*

6000
4000
2000
0
Ctrl

SPS

Ctrl

Male

*

n=5-8/gp

6000

*

0 min
30 min

5000

*

*

2000

*

*

1000
0
Veh

DEX

Veh

Ctrl

DEX

Veh

SPS

DEX

*

*
*

Male

Female

Male

F

n=5-6/gp

15

*
*

Female

*
*

10

5

0
Left

Right

Male

Left

Veh

Ctrl

cFos-IR (cells/mm3) x103

*

*

80
60

*

40
20

DEX

Male

Female

SPS
Female

BLA

E
cFos-IR (cells/mm3) x103

cFos-IR (cells/mm3) x103

40
35
30
25
20
15
10
5
0

IL

Control
SPS

n=5-6/gp

0

SPS

Female

PrL

*

*

*

4000
3000

PVN
100

Male

D n=6-7/gp

C
GR-IR (cells/mm3) x103

Baseline
Post-test

MeA

G

n=5-6/gp

25

CA1/2
140

GR-IR (cells/mm3) x103

n=12/gp

8000

Plasma corticosterone (nmol/L)

Average max ASR
block 1 (A.U./kg b.w.)

A

n=4-5/gp

120

20

100

15
10
5
0

60
40
20
0

Male

Right

80

Female

Male

Female

Female

Figure 2. Single prolonged stress (SPS) affects males and females differently. (A) Males exposed to SPS one week earlier showed, as expected, a significant
increase in the acoustic startle response (ASR), but females exposed to SPS showed no such increase. (B) Negative feedback control of the HPA axis was
assessed by measuring corticosterone (CORT) response to acute stress in the presence or absence of a low dose dexamethasone (DEX) pretreatment. SPS
males who received DEX had significantly lower CORT levels after 30 min of stress compared to SPS males who received only vehicle. Note that controls males
were not as sensitive to the DEX treatment, failing to cause a significant reduction in CORT level, demonstrating the well-established enhancement DEX
suppression induced by prior exposure to SPS for males. In contrast, SPS females showed comparable levels of CORT 30 min post-stress regardless of DEX
treatment, suggesting DEX-nonsuppression that is more typical of depression. All females had significantly higher levels of both baseline and stress-induced
CORT levels than males, as expected. DEX drove down baseline CORT levels (0 min) in all groups, demonstrating its effectiveness in both sexes. (C) SPS also
had opposing actions on glucocorticoid receptor (GR) expression in the paraventricular nucleus of the hypothalamus (PVN), with a moderate increase (P=.060) in
males but a significant decrease in females. Control females also had significantly higher GR expression in the PVN than control males, indicating that GR
expression in the PVN is normally sexually differentiated. (D - F) Prior exposure to SPS had surprisingly little effect on the cFos response to restraint stress in
males, with no effect of SPS in the prelimbic (PrL) or infralimbic (IL) subregions of the medial prefrontal cortex (mPFC), or in the basal lateral amygdala (BLA) or
medial amygdala (MeA). On the other hand, SPS females showed significant increases in cFos in both the PrL and IL. SPS also increased the cFos response in
the right BLA of females. Control females showed a lower cFos response to acute restraint stress in the mPFC than control males. (G) GR expression in the CA1/2
region of the dorsal hippocampus was also affected differently in males and females (sex*SPS interaction P=.050). Data are presented as meanÂąSEM.
Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons (Bonferroni). Refer to Table 2 for full statistical results.

44

B

Average max ASR
block 1 (A.U./kg b.w.)

n=16-17/gp

Baseline
Post-test

8000
6000
4000
2000
0
Ctrl

SPS

Single-housed

Ctrl

SPS

n=10-12/gp

Plasma corticosterone (nmol/L)

A

6000

0 min
30 min

5000

*

4000

*

3000

*

*

*

*

2000

*

0
Veh

DEX

Veh

Control

Pair-housed

DEX
SPS

D

*

n=21-24gp

180

E

*

160

Latency to interaction zone
with empty chamber (sec)

Sucrose preference
(% total intake)

100

Control
SPS

Relative latency to social
interaction

n=20-22/gp

140

80

120
100

60
40
20

80
60

*

40
20

0

0
Single-housed Pair-housed

*

1000

Veh

DEX

DEX
SPS

Pair-housed
n=21-24gp

14
12
10
8
6
4
2
0
Single housed

Single-housed Pair-housed

Veh

Control

Single-housed

C

*

*

Pair housed

Figure 3. Housing conditions change how SPS affects social interaction in females. (A) The null effects of SPS on the ASR persisted regardless of housing
condition (single v. paired) and under conditions of bright light. (B) Females exposed to SPS again did not show the enhancement of DEX suppression of CORT
levels that is typical of SPS exposed males (Fig 1B). Note however, that pair-housing increased the sensitivity to DEX, and did so for both SPS and control
females. Thus, this effect is due to housing and not SPS. (C) SPS did not affect sucrose preference in females, regardless of housing condition, inconsistent with a
depressive-like phenotype. (D) SPS affected social interaction (based on latency to approach a novel female) but the direction of effect depended on the housing
condition. SPS decreased the latency to approach when females were single-housed but increased the latency to approach when females were pair-housed. (E)
Neither housing nor SPS affected the latency of females to approach the empty rat enclosure, indicating that the effect of SPS is on social interaction per se and
not on general activity or exploration of the chamber. These data are consistent with the idea that the traumatic stress phenotype for females is distinctly different
from that of males, and may share some traits of depression (given the consistent failure of SPS to enhance ASR or the effect of DEX on CORT) unlike the
traumatic stress phenotype for males. Data are presented as meanÂąSEM. Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons
(Bonferroni). Refer to Table 3 for full statistical results.

45

A

B
n=4-6/gp

Ratio of total time in arena center

2000

Ctrl+tone
SPS+tone No contextual reminder
(white noise)

2000

1500

Contextual reminder
(tone)

1500

*

P=0.075

1000

1000

500

500

0

d1

d3
Male

d1

d3
Female

0

d1

d3
Male

d1

d3
Female

Figure 4. Contextual reminders to SPS sensitizes anxiety behavior over time. With no contextual SPS reminder (white noise), all rats spend more time in the
center when it contains an object, regardless of sex or SPS or testing day (d1 or d3). After 3 days of testing with the contextual SPS reminder (white noise+neutral
tone), both SPS males and SPS females spent less time in the center when it had an object, indicating that repeated exposure to contextual trauma reminders
sensitized anxiety behavior over time. Data are presented as meanÂąSEM. Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons
(Bonferroni).

46

CHAPTER 3: CHARACTERIZING SEX DIFFERENCES IN RESPONSE TO PREDX

Introduction
While we clearly identified sex differences in response to SPS, how generalizable these
responses are to other types of traumatic stress is not yet known. Examining sex differences across
different traumatic stress models is useful to determine whether males and females have fundamental
differences in their underlying neurobiology that predisposes them to respond differently to traumatic
stress, and to determine if the sex differences we observed in SPS were just an artifact of that particular
paradigm. To our knowledge, no studies have used two traumatic stress models in the same lab, an
approach that limits confounding variables to accurately compare the effects of trauma across models.
We carefully chose a second model, PredX, that was different enough from SPS in stressor type to allow
for inferences on the generalizability of effects but would also not introduce possible confounding effects
of some types of stressors that include pain (i.e. footshock, immobilization) that may not reflect PTSD per
se. There is a spectrum of PredX models ranging in stressor severity from predator odor exposure (low
severity) to full-contact live predator exposure (high severity) (252). We chose a paradigm in the middle of
the spectrum that exposes rats to a live predator (cat) but does not allow physical contact between the
animals, which eliminates the chance of introducing pain if the cat should attack the rats. PredX is a wellvalidated and commonly used rodent model of PTSD, but as with SPS, only a handful of the >100
published PredX studies have examined sex differences (221, 322–327), and no studies have looked
across models.

Methods and Materials
Experiment 4: Sex differences in the PredX model. The same experimental timeline from
experiment 2 was followed in experiment 4 (Fig 5).
Animals. 8wk old adult Sprague-Dawley male and female rats were housed in same-sex pairs on
the day of arrival and handled 3min daily for one week before any testing or stress exposure. Rats were
purchased from Charles River and housed with 12h reversed light-dark cycle, ad lib food and water. Cage
bedding was changed weekly and no testing was conducted on days of cage changes. All behavior tests
were conducted in the dark-phase > 2h of dark. Female rats were freely-cycling and assigned to

47

treatment groups without regard to estrous cycle stage. All animal procedures and care met or exceeded
the NIH guidelines, were approved by Michigan State University Institutional Animal Care and Use
Committee.
Predator exposure. PredX rats were placed in individual wedge-shaped enclosures of a circular
Plexiglas “pie restrainer” (Braintree Scientific) on which cat food was smeared and placed inside a
Plexiglas arena (60cm3). This paradigm allows rats to be exposed to the cat without physical contact.
Rats were exposed to a live female cat for 1h, as previously described (323). All rats were singly housed
immediately before the PredX and control procedure. Control rats were removed from the vivarium for 1h
but not put in the restrainer nor exposed to the cat, and housed in a separate room to prevent exposure to
possible residual predator scent on PredX rats. Rats were left undisturbed for one week after PredX to
allow the acute stress responses to resolve and long-term PTSD responses to develop.
Outcome measures. The same outcome measures used in experiments 1 and 2 were used in
experiment 4: acoustic startle response, DEX suppression test, sucrose preference, social interaction,
and cFos and GR IHC.
Statistical analysis. The same statistical analyses used in experiment 1 were used in experiment
4. See Table 4 for all statistical tests performed.

Results
Predator exposure induces comparable sex differences in ASR and DST. To test how
general the sex difference is, rats were exposed to a live cat for 1h. Like SPS, PredX enhanced both ASR
(Fig 6A) and DEX suppression of CORT for males but not females (Fig 6B). Indeed, the pattern of sex
differences was remarkably similar following SPS and PredX, and occurred independent of effects on
body weight (Fig 14). Specifically, DEX blocked an increase in CORT levels for PredX-exposed males,
leading to a significant difference in CORT level between DEX and vehicle treated PredEx males not
shown by control males, but failed to do so for PredX-exposed females, again showing that prior
exposure to a traumatic event enhances sensitivity to DEX in males but not in females. Females again
had higher baseline and post-restraint CORT levels compared to males (Fig 6B) and DEX drove baseline
CORT levels down in all four groups (Fig 6B), confirming the effectiveness of the DEX treatment. Females

48

again had higher GR expression in the PVN than males, but unlike SPS, PredX did not affect GR
expression in the PVN of either sex (Fig 6C).

Discussion
Regarding the ASR and DST tests, sex differences in PredX mirror sex differences in SPS and
recapitulate sex differences in PTSD symptomology. While we replicated the sex difference that females
normally have higher GR+ neurons in the PVN compared to males, PredX did not affect PVN GR as SPS
did. The effects of SPS on GR expression may reflect a stressor-specific response to traumatic exposure,
likely the ether component of SPS, since the replacement of ether with isoflurane abolishes the effect of
SPS on hippocampal GR (328). Such dissociations in the stress response across different stressors help
to identify core traits of the sex-specific phenotype that are independent of stressor type. Taken together,
these data establish sex-specific responses to traumatic stress that are independent of the type of
stressor and housing condition (ASR and DST), indicating fundamental sex differences in the
neurobiology underlying the traumatic stress response. We are the first to recapitulate in two different
animal models distinct subtypes of PTSD recognized in clinical studies (116, 120, 139, 140) that are
linked to a single biological factor: sex. The next logical step is to examine what mediates these sex
differences.

49

Blood
sample Sacrifice

Veh Blood
or DEX sample

Quiescent period

Day to
SPS

Restraint

Hour

-2

0

0.5

1

Daily handling

-7

-1

0

1 2 3 4 5 6 7

Baseline PredX
ASR

11

13

ASR

DST

Figure 5. Timeline of procedures for experiment 4. Experimental timeline begins with daily handling one week before (-7) predator exposure (PredX) and
baseline acoustic startle response (ASR) testing the day before (-1) PredX. Rats are left undisturbed for one week after PredX, and post-stress ASR is assessed
11 days later, and dexamethasone suppression test (DST) 13 days later.

50

A
Average max ASR
block 1 (A.U./kg b.w.)

8000

n=11-12/gp

Baseline
Post-test

6000

*

4000

2000

0
Ctrl

Ctrl

Male

Plasma corticosterone (nmol/L)

B

PredX

6000

Female

n=5-8/gp

0 min
30 min

5000
4000

*

*

*

*

*

1000
0
Veh

DEX

Veh

Ctrl

DEX

PredX

GR-IR (cells/mm3)

x103

Male

C

*

*

3000
2000

PredX

100

n=5-6/gp

*

80

Veh

DEX

Veh

Ctrl

DEX

PredX
Female

Control
PredX

60
40
20
0
Male

Female

Figure 6. Predator exposure (PredX) leads to sex differences in the acoustic startle response (ASR) and HPA
negative feedback comparable to SPS. (A) Only males and not females show an enhanced ASR after PredX
exposure, replicating the sex difference found with SPS. (B) Likewise, PredX also enhanced HPA negative feedback
in males but not in females, as indicated by the DEX suppression test. Again, DEX blocked an increase in acute
stress-induced CORT levels in PredX males but not in PredX females, leading to a significant deficit in CORT levels
of PredEx males treated with DEX compared to vehicle. This same difference is not evident for PredEx females, as in
the SPS model (Fig 1B). CORT levels after 30 min of restraint stress were significantly higher than baseline for all
female groups, independent of DEX or PredX. This pattern was not seen in males treated with DEX. CORT levels
were significantly higher in females than males, replicating previous reports in rats. DEX lowered baseline CORT
levels (0 min) to near zero in all groups, demonstrating that DEX was effective in both sexes. (C) Unlike SPS, PredX
did not affect glucocorticoid receptor (GR) expression in the PVN of either sex, but we did replicate the sex difference
in GR expression seen previously (see Figure 1C), with females having more GR+ neurons in the PVN than males.
Because the sex difference in GR expression in the PVN was replicated across studies, it is likely some unique
aspect of SPS that affected GR expression in the PVN and this response may be stressor-specific. Data are
presented as meanÂąSEM. Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons
(Bonferroni). Refer to Table 4 for full statistical results.

51

CHAPTER 4: THE ROLE OF GONADAL HORMONES IN THE TRAUMATIC STRESS RESPONSE

Introduction
Given the novel sex differences we discovered in the traumatic stress response to SPS and
PredX, we next tackled the question of whether such sex differences are related to sex differences in
circulating levels of adult gonadal hormones. While some studies have found sex differences in these
traumatic stress models, none have examined the possible role of gonadal hormones, despite the wellestablished role they serve in determining sex differences in brain morphology and function (329)
We find that the traumatic stress response in rats is highly sex-specific and recapitulates
fundamental differences of PTSD in humans, with males exhibiting externalizing symptoms (e.g.
hyperarousal, aggression, and risk-taking behaviors), while females exhibit internalizing symptoms (e.g.
sadness, loss of pleasure, and social withdrawal) (117, 137–140). Because other sex-biased psychiatric
disorders share this same divide between men and women (310), our studies of the underlying
neurobiology of these sex differences in the traumatic stress response may lend insight into the
neurobiological underpinnings of other psychiatric disorders.

Methods and Materials
Experiment 5: Effects of gonadectomy and testosterone replacement in SPS model. The
same experimental timeline used in experiments 2 and 3 were used in experiment 5, with the addition of
gonadectomy (GDX) surgery two weeks prior to any testing or SPS exposure (Fig 7).
Animals. 8wk old adult Sprague-Dawley male and female rats were housed in same-sex pairs on
the day of arrival and handled 3min daily for one week before any testing or stress exposure. Rats were
purchased from Charles River and housed with 12h reversed light-dark cycle, ad lib food and water. Cage
bedding was changed weekly and no testing was conducted on days of cage changes. All behavior tests
were conducted in the dark-phase > 2h of dark. Female rats were freely-cycling and assigned to
treatment groups without regard to estrous cycle stage. All animal procedures and care met or exceeded
the NIH guidelines, were approved by Michigan State University Institutional Animal Care and Use
Committee.

52

Gonadectomy. The day after arrival, rats were randomly assigned to a treatment group
(sham+blank, GDX+blank, GDX+T). Cagemates were placed in the same surgery treatment group. GDX
was performed using sterile technique under isofluorane anesthesia. Ketoprofen analgesia was given at
the onset of the procedure and the following day. For GDX, the outer layer of the scrotum was opened
(for males) or the abdominal cavity opened (for females) and then closed with wound clips. Under the
same about of anesthesia, Silastic capsules (3.2mm diameter, 40mm length) filled with T or blank were
implanted s.c. at the back of the neck. Such implants provide a time-release of the hormone of interest for
up to 4 wks without the stress of daily injections. The rationale for treating females, as well as males, with
T was to see whether T might induce in females a male-like response to SPS. Rats recovered for 2
weeks, and were handled daily during the second week.
SPS paradigm. The SPS paradigm was followed as described in experiment 1, two weeks after
GDX.
Outcome measures. The same outcome measures used in experiments 1 and 2 were used in
experiment 5: ASR, DEX suppression test, sucrose preference, social interaction, and cFos and GR IHC.
Statistical analysis. Two-, three-, or four-way ANOVAs were run for comparisons in intact rats.
See Tables 5-7 for all statistical tests performed. For brain measures, if no main effect or interactions of
hemisphere was present, data were collapsed across hemisphere. In the GDX experiment, analysis was
first conducted on only sham+blank rats to assess whether previously identified sex differences were
replicated. Then, the effects of GDX were assessed within each sex, comparing GDX+blank to
sham+blank. If there was a main effect or interaction with GDX, the effect of T treatment was assessed by
comparing GDX+blank to GDX+T. The conservative Bonferroni test was used to correct for multiple tests
to hold alpha at 0.05.

Results
Sex differences were replicated in the sham surgery group. We first compared sham
operated males and females and found the same sex differences (Fig 8A,B,D), indicating that sham
surgery itself did not alter the pattern of differences. SPS in this study replicated effects on social
interaction latency in females, and notably, revealed a decrease in sucrose preference in females (Fig 8F,

53

G, Fig 15). These data indicate that both sucrose preference and social interaction are useful measures
for capturing the effects of SPS on behavior in females. Interestingly, males appear resilient based on
these particular measures.

Effects of SPS in males are largely independent of adult testicular hormones. Overall, we
find that male rats castrated as adults respond to SPS much like gonadally intact male rats. To the extent
that adult testicular hormones influenced our measures, they were independent of SPS. Specifically,
gonadal status did not alter the effect of SPS on the ASR in males (Fig 9A). Likewise, the DST revealed
the same pattern of differences in castrated (GDX) males as in gonadally intact (sham) males (Fig 9B).
Both groups showed an exaggerated DEX suppression after exposure to SPS. Removing the testes did
however significantly increase both baseline and restraint-induced CORT levels in control males, which
was reversed by T treatment of castrated males (GDX+T), as previously reported (215). However, T
treatment also seemed to mask an effect of SPS on HPA negative feedback, since both control and SPS
males had comparably low, near baseline, CORT levels after restraint stress in the presence of DEX. This
outcome may reflect the fact that exogenous T does not faithfully recapitulate normal levels of
endogenous T, or that the low number of individuals in the GDX+T group (n=2-6) was insufficient to
detect an effect of SPS. Regardless, these data indicate that the effect of SPS in males on the ASR and
negative feedback control of stress-induced CORT levels does not depend on endogenous testicular
hormones in adulthood.
SPS and castration each significantly increased the number of GR+ neurons in the PVN (Fig 9C).
SPS increased GR expression in gonadally intact but not castrated males while castration increased GR
expression in control but not SPS-exposed males, appearing to eliminate the effect of SPS in castrated
males. These results indicate that testicular hormones normally regulate GR expression in the PVN, and
depending on gonadal status, an effect of SPS on GR expression in the PVN may or may not be
detected. However, T treatment did not reverse this effect of castration. Since other measures (e.g., Fig
9E-F, described below) confirm that the T capsules were effective, these data raise the question of
whether other testicular factors regulate GR expression in the PVN.

54

While adult gonadal hormones had no effect on sucrose preference in males (Fig 9D), they did
affect social interaction, in turn revealing a significant effect of SPS in GDX males. Castration of control
males increased the latency to approach a novel male (Fig 9E) while decreasing the latency to approach
the empty interaction chamber (Fig 9F), pushing both measures toward a more feminine phenotype in
castrated males (Fig 9E, F). Consequently, an effect of SPS is detected in castrated males that is not
apparent in gonadally intact males or castrates given T, suggesting that androgens in males may
normally override an effect of SPS on social interaction. Body and adrenal weight in males were not
affected by any treatment (Fig 16, 17).

Effects of SPS in females are largely independent of ovarian hormones. Regardless of
ovarian hormone status, ASR is not influenced by SPS in females (Fig 10A). Adult T treatment also had
no effect on the ASR in females, consistent with the lack of an effect of adult T on this measure in males
(Fig 9A). The DEX suppression test revealed the same pattern of differences in CORT levels irrespective
of gonadal status in females (Fig 10B). As expected, 30 min of acute restraint stress significantly
increased the level of CORT in all vehicle-treated groups, although the level was significantly reduced in
both groups of GDX control females (blank and T-treated) compared to sham control females. That T did
not reverse the effect of ovariectomy suggests that ovarian hormones are involved in the blunted CORT
response to restraint stress in GDX females. Importantly, the same low dose of DEX that blocked a
significant increase in CORT in SPS-exposed males after 30 min of restraint stress did not block this
increase in CORT in SPS-exposed females, whether GDX or treated with T. Nonetheless, DEX did
significantly reduce baseline (0 min) CORT and the level of CORT induced by restraint stress in SPSexposed females under all three hormonal conditions, indicating that DEX was working. As expected, T
treatment significantly decreased baseline CORT levels in females (330). These data indicate that the
enhanced CORT suppression characteristic of traumatized males is not a characteristic feature of the
trauma response of females. This distinct sex difference in how trauma influences regulation of the
negative feedback control of HPA axis of males and females is apparently independent of adult gonadal
hormones in both sexes.

55

Interestingly, the effect of SPS on the number of GR+ neurons in the PVN depended on ovarian
status. Without ovarian hormones, SPS had no effect on GR expression in the PVN as opposed to
decreasing their number in gonadally-intact females (Fig 10C). Because ovariectomy per se does not
affect the number of GR+ neurons in the PVN, ovarian hormones likely regulate how sensitive PVN
neurons are to stress, with stress influencing GR expression in this brain region when ovarian hormones
are present. SPS significantly and selectively reduced sucrose preference in sham operated females (Fig
10D), an effect that was reversed by ovariectomy. Because T did not reverse the effect of adult
ovariectomy, estrogens and/or other ovarian factors likely mediate the effect of SPS on sucrose
preference. On the other hand, the effect of SPS on social interaction did not depend on ovarian
hormones, since SPS shortened the latency to approach a novel female, regardless of gonadal status
(Fig 10E). T treatment, on the other hand, clearly reduced social interaction latency in general, eliminating
the effect of SPS on this measure in GDX females. Latency to enter the interaction zone when it had an
empty chamber was unaffected by SPS in females except in the presence of T, when approach time was
increased by SPS compared to SPS-exposed females in the other two hormone groups (Fig 10F). Again,
SPS did not affect body weight in females, but GDX led to significant increases in body weight (Fig 18),
as expected (331). GDX had no effect on adrenal weight in females (Fig 19).

Discussion
Surprisingly, gonadal hormones had little role in these sex differences. This is particularly
unexpected, given the current view that cycling estrogens robustly regulate stress susceptibility in
females (332, 333). It seems these sex differences are determined earlier in life, predisposing individuals
to certain endophenotypes of PTSD as adults. Indeed, adult gonadal hormones only partially modulate
certain aspects of the stress response in rats, with neonatal organizational effects also at play (196).
Individual differences in the stress response are apparent in rats at weaning and persist throughout life
(334). We propose that such sex differences reflect differences in the underlying neurobiology,
orchestrated earlier in life by sex hormones and/or genes.
Sex differences in the traumatic stress response were apparent at every level of analysis:
traumatized males are hyper-responsive while traumatized females seem depressed. Traumatic stress

56

causes male rats to show an enhanced ASR while not affecting this same measure in females. Moreover,
HPA negative feedback, which is enhanced in trauma-exposed males, is unaffected or even decreased
by trauma in females, suggesting that the female trauma response may share common attributes with
depression, characterized by a blunted ASR (335) and reduced HPA negative feedback (160, 261). We
again included two common measures of depression, social interaction and sucrose preference. Each
revealed effects of trauma in females but not males, further supporting the view that the female response
to traumatic stress is distinct from that of males and is reminiscent of depression. While the overall pattern
of effects of SPS in females points toward a depressive phenotype, the effects on sucrose preference
was not consistent, indicating that this measure is only marginally sensitive to the effect of trauma in
female rats and that the traumatic stress response in females is not equivalent to depression but is more
affective in nature than in males.
Sex differences in the traumatic stress response were largely independent of adult circulating
gonadal hormones. We find that increased ASR and exaggerated DEX suppression of CORT,
characteristic of trauma-exposed males, do not depend on adult circulating gonadal hormones, as shown
previously for ASR in males (336). These measures were also unaffected by gonadal hormones in
females, further implicating the role for early life factors in laying the groundwork for sex differences in the
trauma response. While sucrose preference was not affected by SPS or gonadal hormones in males,
sucrose preference was affected by both in females. Because GDX did not alter sucrose preference in
control rats, the effect of trauma on sucrose preference in females appears to be mediated by adult
ovarian hormones. This could mean that fluctuating estrogen levels might influence the emergence and/or
severity of depressive symptoms following exposure to trauma. However, whether ovarian hormones
matter during and/or after experiencing trauma is entirely unanswered.
SPS affected social interaction only in females. Because stress effects on the latency to socially
interact are sex-specific (337, 338), we chose this measure rather than the more typical measure of total
social interaction time. Additionally, we normalized social interaction latency relative to latency to
approach a non-social target, which serves the important function of controlling for individual differences
in general inclination to move and explore, another sexually differentiated attribute (339), making this
measure the more valid measure in assessing whether SPS affects social interaction in a sexually

57

differentiated manner. We are the first to test this measure in a traumatic stress paradigm and
demonstrate that stress effects on this measure extend to traumatic stress, with effects only in females.
This effect was also seen in GDX females, indicating that the effect of SPS on this measure is
independent of adult ovarian hormones.
While SPS had no apparent effect on social interaction in gonadally intact males, castration did
increase social interaction latency in control males, introducing an apparent effect of SPS to reduce
latency in castrated males (Fig 9E). Whether the decreased latency to approach a novel rat reflects a
genuine effect of SPS in castrates or an artifact of the effect of castration in control males is unclear.
Likewise, SPS increased GR expression in the PVN of intact males, an effect lost in castrated males
because castration increased GR expression in the PVN only of control males (Fig 9C). Thus, castration
may have protected GR expression from trauma in the PVN of males, or may have simply masked the
effect of trauma due to a ceiling effect. On the other hand, the effect of SPS to decrease GR expression
in the PVN of females was not apparent after ovariectomy, but ovariectomy alone did not change the
number of GR+ neurons in the PVN of non-traumatized females, indicating that the effect of SPS on this
measure, like sucrose preference, depends on ovarian hormones in females. These data raise questions
about the potential importance of gonadal hormones at the time of trauma versus at the time of
assessment, an entirely unexplored problem in trauma research. It is clear that differences in gonadal
hormones change the response to acute stress. Hence, such changes could also potentially reveal or
mask an effect of prior trauma. While the effect of trauma on the ASR and DEX suppression of CORT
appear independent of adult gonadal hormones, other measures like social interaction and sucrose
preference may well depend on hormonal status, and could potentially lead to different conclusions about
whether trauma exposure led to pathology.

58

Veh Blood
or DEX sample

Surgery recovery

Day to
SPS -15 -14

GDX and
sham

Quiescent period

Blood
sample Sacrifice

Restraint

Hour

-2

0

9

11

0.5

1

Daily handling

-7

-1

0

1 2 3 4 5 6 7 8

Social ASR
Sucrose
preference interaction

Baseline SPS
ASR

13

DST

Figure 7. Timeline of procedures for experiment 5. Experimental timeline begins with GDX or sham surgery performed fifteen days before single prolonged
stress (SPS), with daily handling one week before (-7) SPS and baseline acoustic startle response (ASR) testing the day before (-1) SPS. Rats are left undisturbed
for one week after SPS, and sucrose preference is assessed 8 days later, social interaction 9 days later, post-stress ASR 11 days later, and dexamethasone
suppression test (DST) 13 days later.

59

Baseline
Post-test

7000

Average max ASR
block 1 (AU/kg b.w.)

B

n=14-16/gp

8000

Plasma corticosterone (nmol/L)

A

6000
5000

*

*

4000

*

*

3000
2000
1000
0

Ctrl

SPS

Ctrl

Male

SPS

6000

n=6-10/gp

0 min
30 min

5000

*
4000

*

3000

*

2000

*
*

*
*

1000
0
Veh

DEX

Veh

Ctrl

DEX

Veh

SPS

DEX

Female

Sucrose preference
(%total intake)

*

80
60
40
20

120
90
60

*

30
0
Male

Female

*

60
40
20

Female

Male

Female

H
n=14-15/gp

*

30

n=16/gp

Body weight (g)

*

150

*

80

0
Male

G

180

n=11-14/gp

100

100

Female

n=14-15/gp

Relative latency to social
interaction

*

140
120

DEX

SPS

E

0
Male

F

GR-IR (cells/mm3) x103

*

*

Latency to interaction zone
with empty chamber (sec)

Adrenal gland weight (wet
wt (mg)/body wt (g)

n=6/gp

0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0

D 160 n=4-6/gp

Veh

Ctrl

Male

Control
SPS

*

*

*

Female

C

*

25
20
15
10
5

400
350
300
250
200
150
100
50
0

*
*

*

Ctrl

0
Male

Female

Baseline
Post-test

SPS
Male

*

*

Ctrl

SPS

Female

Figure 8. SPS elicits comparable sex differences in sham gonadectomized rats. (A) Despite prior exposure to
anesthesia and sham gonadectomy, SPS enhanced the acoustic startle response (ASR) only in males, not females.
While females showed a significant increase in the ASR in the post-test compared to baseline, this effect was not
specific to SPS exposure as it is for males. Control males showed a habituation to the acoustic stimulus in the posttest compared to baseline, which was not seen in the previous study (Figure 1A). (B) As before, SPS enhanced the
negative feedback control of the HPA axis in males, but not in females. Note that while DEX completely blocked the
increase in CORT levels induced by acute restraint stress in SPS-exposed males, it did not block this increase in
SPS-exposed females, nor did DEX block this increase in control males and females. SPS did increase baseline
CORT levels in females but had no effect on baseline CORT in males. Again, our dose of DEX drove baseline CORT
levels to near zero in all four groups, confirming its overall efficacy in both sexes. (C) The significantly higher CORT
level in females than males was paralleled by a sex difference in adrenal weight, as females had significantly heavier
adrenal glands than males. Note that SPS had no effect on adrenal weight in either sex. (D) The opposing effects of
SPS on glucocorticoid receptor (GR) expression in the paraventricular nucleus of the hypothalamus (PVN) in males
versus females was comparable to that seen in the previous study (Fig 1C), but this time, the effect of increasing GR
expression in males was significant while the effect of decreasing GR expression in females fell short of significance
(P=.056). As in the first study, control females had significantly more GR+ PVN neurons than control males. (E - G)
SPS affected sucrose preference and social interaction only in females, significantly decreasing their preference for
sucrose and latency to approach a novel conspecific, while having no effect on these measures in males. Control
females also took longer than control males to approach a novel rat and this sex difference flipped for latency to
approach the interaction zone. (H) SPS did not affect body weight, but females weighed less than males. Data are
presented as meanÂąSEM. Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons
(Bonferroni). Refer to Table 5 for full statistical results.

60

B
Baseline
Post-test

8000
6000

*
*

*

4000
2000

n=6-10/gp

2500
2000
1500

n=6-10/gp

0 min
30 min

*

*

1000

*

*

*

*
*

*
*

*

*

0
Ctrl

SPS

Sham

Ctrl

SPS

Ctrl

GDX

Veh

SPS

DEX

Veh

Ctrl

GDX+T

DEX

Veh

SPS

DEX

Control
SPS

n=5-6/gp

150

*

100

*

50

E

n=8-13/gp

Relative latency to
social interaction

D

PVN

Sucrose preference
(%total intake)

C

80
60
40
20
0

0
Sham

GDX

GDX+T

GDX

GDX+T

Veh

DEX

F

*

100
80
60

*

40
20
Sham

GDX

Veh

Ctrl

DEX

SPS
GDX+T

0
Sham

DEX

SPS
GDX

n=11-15/gp

120

Veh

Ctrl

Sham

GR-IR
(cells/mm3) x103

*

*

500

0

100

n=2-6/gp

*

GDX+T

Latency to interaction zone
with empty chamber(sec)

Average max ASR
block 1 (AU/kg b.w.)

n=12-16/gp

Plasma corticosterone
(nmol/L)

A

n=11-15/gp

35
30

*

*

25
20
15
10
5
0
Sham

GDX

GDX+T

Figure 9. Effects of SPS in males are largely independent of adult testicular hormones. (A, B) The ability of SPS to enhance the ASR and negative feedback
control of CORT levels was unaffected by gonadal hormone status in males. In each hormonal condition, DEX inhibited the rise in CORT levels after 30 min of
restraint stress only for SPS males, with the single exception of control males that were gonadectomized and treated with testosterone (GDX+T), which may reflect
the low number of individuals in the GDX+T group (n=2-6). Importantly, removal of the testes (GDX) did not change the pattern of differences seen in gonadally
intact males (shams). (C) While SPS increased GR expression in the PVN of males, this effect was not apparent in GDX males because the number of GR+ PVN
neurons increased by GDX in sham control males, presumably masking the effect of SPS. This effect was not reversed by T treatment, suggesting other testicular
hormones may normally inhibit GR expression in the PVN. (D) Neither SPS nor gonadal hormone status affected sucrose preference in males. (E) Castration
(GDX) of control males increased their latency to approach a novel rat, without similar influences on SPS males. Thus, SPS may have no genuine effect on this
measure in males. These data highlight the difficult issue of gonadal hormones possibly masking real effects of trauma or spuriously introducing apparent effects
of trauma, simply because baseline measures are sensitive to gonadal hormone levels independent of the trauma experience itself. The effect of GDX on this
measure was reversed by T treatment. (F) SPS had no effect on latency to enter the interaction zone when it had an empty chamber, regardless of gonadal status,
but GDX reduced this latency for both control and SPS-exposed males, an effect reversed with T replacement. These results suggest that T normally affects this
measure independent of SPS exposure. Data are presented as meanÂąSEM. Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons
(Bonferroni). Refer to Table 6 for full statistical results.

61

B
Baseline
Post-test

Average max ASR
block 1 (AU/kg b.w.)

6000
4000
2000
0
Ctrl

SPS

Ctrl

Sham

SPS

GDX

Ctrl

SPS

5000
4000

n=6-8/gp

*

*
*

*

*

*

3000

*

2000

*

*

Veh

DEX

Veh

Ctrl

D 100

*

80

DEX

SPS

E

Veh

GDX

GDX+T

180

DEX

Veh

Ctrl

*

40
20

*

*

*

30

*

0
GDX

GDX+T

Veh

DEX

Sham

GDX GDX+T

Veh

Ctrl

DEX

SPS
GDX+T

F

n=10-15/gp

90
60

DEX

SPS

120

60

Sham

*

GDX

150

0
Sham

*

0

Relative latency to
social interaction

*

*

*

1000

GDX+T

Sucrose preference
(%total intake)

GR-IR (cells/mm3) x103

160
140
120
100
80
60
40
20
0

Control
SPS

PVN
n=4-6/gp

*

*

*

Sham

C

n=4-6/gp

0 min
30 min

6000

Latency to interaction zone
with empty chamber (sec)

n=11-16/gp

8000

n=6-8/gp

Plasma corticosterone (nmol/L)

A

30

n=10-15/gp

*

25
20
15
10
5
0
Sham

GDX

GDX+T

Figure 10. Effects of SPS in females are largely independent of ovarian hormones. (A, B) SPS again had no effect on ASR or negative feedback control of
CORT levels in females, and this was true regardless of gonadal hormone status, neither enhancing ASR nor negative feedback of CORT levels, in contrast to the
effects of SPS on males (Fig 4A,B). GDX did significantly decrease restraint-induced CORT levels in vehicle-treated control females, but T treatment did not
reverse this effect, implicating ovarian hormones other than androgens. (C) SPS decreased GR expression in the PVN of gonadally intact (sham) females, an
effect that was abolished by GDX. Given that this effect was specific to SPS exposure, ovarian hormones may mediate this response to trauma in the brain.
However, what this means for the functional response to stress is not clear since corresponding changes were not seen in the DEX suppression test. T treatment
lowered the number of GR+ neurons overall in the PVN, but did not restore the effect of SPS on GR expression in the PVN, suggesting that the effect of ovarian
hormones on this measure is via estrogen or progesterone receptors and not androgen receptors. (D) While decreased sucrose preference was decreased by
SPS in sham females, the effect of SPS was abolished by GDX implicating ovarian hormones in this female-specific effect of SPS. That T did not reverse the effect
of GDX on sucrose preference suggests that ovarian hormones may act via estrogen and/or progesterone receptors to modulate the effect of SPS on sucrose
preference. (E) SPS decreased the latency to approach a novel rat in both sham and GDX females, indicating that the effect of SPS on social interaction is
independent of adult ovarian hormones. However, T treatment lowered the latency of control females to approach a social target, eliminating an apparent effect of
SPS on this measure. Perhaps T acted as an anxiolytic to have this effect. (F) Latency to enter the interaction zone when it had an empty chamber was affected
only by the combination of SPS and T treatment to significantly increase latency. Data are presented as meanÂąSEM. Significance set at P<.05 (indicated by
asterisk) for planned pairwise comparisons (Bonferroni). Refer to Table 7 for full statistical results.

62

CHAPTER 5: CONCLUSIONS

Factors that mediate differences in how individuals adjust after traumatic stress are attractive
targets for the prevention and treatment of PTSD. At least 75% of people in the United States experience
a traumatic event in their lifetime and do not develop PTSD (6), begging the question of why some people
who experience trauma develop PTSD while others do not? The significant sex differences in the
prevalence of stress-related disorders in humans is a call for inquiries of the factors behind such
differences, which undoubtedly offer insight into the brain regions and mechanisms that underlie
individual differences in the response to trauma stress. While psychological theories of PTSD give great
insight into how trauma is experienced by people and how to address the effects of traumatic stress with
psychotherapeutic approaches (340), underlying all psychological experiences are neurobiological
substrates. The experiments described here indicate that neurobiological factors mediating the traumatic
stress response are fundamentally different in males and females, challenging the assumption that what
is known about PTSD in males can be readily applied to females.
This is not to suggest that there is a dichotomous “male PTSD” and “female PTSD,” but rather
that sex accounts for some of the variability we see in the traumatic stress response with males more
likely to exhibit certain phenotypes and females more likely to exhibit other phenotypes. The extent that
adult circulating gonadal hormones mediate these sex differences is likely limited to specific aspects of
the traumatic stress response such as social or affective symptoms, while the greatest effects are likely
organized prior to adulthood. Activational effects of steroid hormones do not themselves regulate
physiology or behavior, but rather they have “permissive, preparative, stimulating, and suppressive
actions” that make certain outcomes more likely in certain contexts (e.g. “Testosterone does not cause
aggression, it simply exaggerates the pre-existing pattern and response to environmental triggers of
aggression.”) (341). That neither sex differences in PTSD nor any traumatic stress responses are
dichotomously determined has implications for the identification of resilience vs. susceptibility factors.

Do sex differences in the traumatic stress response reflect differences in resilience?
Based on the ASR and DST, females might appear more resilient to the effects of traumatic stress than

63

males, but as we looked further, it became clear that females were not more resilient but rather,
responded differently to trauma. Similarly, in response to uncontrollable footshocks, females appear more
resilient than males to the negative effects of stress on learning that involves operant conditioning (342),
but if classical conditioning is involved, females are less resilient than males (20), suggesting that sex per
se may not be a factor that confers susceptibility or resilience to stress but simply leads to different
responses, so that the choice of outcome measures may profoundly shape the conclusions drawn. Our
data suggest that the line between resiliency and susceptibility may not be clear enough to separate
individuals into these two groups as other stress studies have done (343), because perceived
susceptibility can change depending on the measures used and the gonadal hormonal status of the
individual (Fig 20). Resilience likely exists on a continuum that also depends on other factors, such that
an individual may be resilient in one domain of functioning but not in another, or at one time during the
lifespan but not another (344). While our data seem to question the heuristic value of the terms
“resilience” and “susceptibility,” these terms remain useful and valid. Not only does the epidemiological
data tell us that only a small fraction of people exposed to trauma develop PTSD, indicating that
differences in susceptibility are real, but also a growing list of gene polymorphisms and epigenetic states
have been identified that appear to bias the nervous system toward more or less resilience to the
negative effects of stress.

Future directions: developmental effects of trauma. The evidence presented here indicates
that robust sex differences in rodent models of PTSD are largely independent of adult circulating gonadal
hormones, which undoubtedly points to organizational effects earlier in development that predisposed
males and females to respond differently to trauma. Indeed, most adult women with PTSD do not show
exaggerated HPA negative feedback unless trauma occurred before puberty (290, 345, 346), and the
increased risk for PTSD associated with low cortisol at the time of trauma is reversed in children
(increased cortisol imparts increased risk in children) but only in boys, not girls (347). That the DSM-5
recognized the child subtype of PTSD is a start to acknowledging the heterogeneity in the traumatic
stress response across development and across sexes. Converging at the intersection of sex and
nervous system development will undoubtedly lie valuable answers to questions regarding the

64

heterogeneity of the traumatic stress response that will advance efforts in the prevention, diagnosis, and
treatment of PTSD.

65

APPENDIX

66

n=12/gp

Body weight (g)

400
350

Baseline
Post-test

*

*

*

300
250

*

200

*

150
100
50
0
Ctrl

SPS
Male

Ctrl

SPS
Female

Figure 11. SPS did not affect body weight. Rats in all groups showed a significant gain in body weight between the
two time points regardless of traumatic stress exposure. As expected, females weighed less than males. Data are
presented as meanÂąSEM. Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons
(Bonferroni). Refer to Table 2 for full statistical results.

67

Table 2. Statistical results for data shown in Figures 2 and 11. All pairwise comparisons use
Bonferroni adjustment for multiple comparisons. RM denotes repeated measure; otherwise assume
between group measures. Only statistically significant results are shown.
Outcome
Power
Statistical test
Significant effects
p value
measure
(Îą=0.05)
ASR
(Fig. 2A)

DST
(Fig. 2B)

RM 3-way ANOVA
(stress*sex*time)

RM 4-way ANOVA
(stress*sex*time*DEX)

Interaction: stress*time

0.043

0.529

Pairwise: SPS male baseline v. posttest

0.017

0.683

Main effect: time

<0.0001

1.000

Main effect: DEX

<0.0001

1.000

Main effect: sex

<0.0001

1.000

Interaction: sex*time

<0.0001

1.000

Interaction: sex*time*DEX

0.004

0.835

Interaction: sex*time*DEX*stress

0.037

0.557

Pairwise: SPS male time 2 DEX v. veh

0.022

0.639

<0.0001

0.999

0.05

0.506

<0.0001

0.995

0.019

0.663

<0.0001

1.000

0.001

0.953

<0.0001

0.974

<0.0001

0.977

0.006

0.802

<0.0001

1.000

Pairwise: SPS DEX female time 1 v. 2

<0.0001

1.000

Pairwise: SPS vehicle male time 1 v. 2

0.002

0.886

Pairwise: SPS vehicle female time 1 v.
2

<0.0001

1.000

Pairwise: control DEX male time 1 v. 2

0.019

0.665

<0.0001

1.000

0.001

0.943

<0.0001

1.000

Pairwise: SPS female time 1 DEX v.
veh
Pairwise: control male time 1 DEX v.
veh
Pairwise: control female time 1 DEX v.
veh
Pairwise: control female time 2 DEX v.
veh
Pairwise: SPS DEX time 2 male v.
female
Pairwise: SPS vehicle time 1 male v.
female
Pairwise: SPS vehicle time 2 male v.
female
Pairwise: control DEX time 2 male v.
female
Pairwise: control vehicle time 1 male v.
female
Pairwise: control vehicle time 2 male v.
female

Pairwise: control DEX female time 1 v.
2
Pairwise: control vehicle male time 1 v.
2
Pairwise: control vehicle female time 1
v. 2

68

Table 2 (cont’d).
PVN GR
(Fig. 2C)

PrL cFos
(Fig. 2D)

IL cFos
(Fig. 2D)

BLA cFos
(Fig. 2E)

2-way ANOVA
(stress*sex)

2-way ANOVA
(stress*sex)

2-way ANOVA
(stress*sex)

3-way ANOVA
(stress*sex*side)

Main effect: stress

0.040

0.553

Main effect: sex

<0.0001

1.000

Interaction: stress*sex

<0.0001

0.998

Pairwise: control male v. female

<0.0001

1.000

Pairwise: female SPS v. control

<0.0001

0.999

Pairwise: male SPS v. control

0.060

0.477

Interaction: stress*sex

0.002

0.916

Pairwise: SPS male v. female

0.040

0.552

Pairwise: control male v. female

0.011

0.754

Pairwise: female SPS v. control

0.001

0.960

Main effect: stress

0.030

0.603

Interaction: stress*sex

0.003

0.891

Pairwise: SPS male v. female

0.018

0.688

Pairwise: control male v. female

0.042

0.542

Pairwise: female SPS v. control

0.001

0.974

Main effect: stress

0.015

0.697

Pairwise: female SPS left v. right

0.029

0.602

Pairwise: female right SPS v. control

0.021

0.650

MeA cFos
(Fig. 2F)

2-way ANOVA
(stress*sex)

None

CA1/2 GR
(Fig. 2G)

2-way ANOVA
(stress*sex)

Interaction: sex*group

0.050

0.517

Body wt.
(Fig. 10)

RM 3-way ANOVA
(stress*sex*time)

Main effect: sex

<0.0001

1.000

Main effect: time

<0.0001

1.000

Interaction: time*sex

<0.0001

1.000

Pairwise: SPS time 1 male v. female

<0.0001

1.000

Pairwise: SPS time 2 male v. female

<0.0001

1.000

Pairwise: Control time 1 male v. female

<0.0001

1.000

Pairwise: Control time 2 male v. female

<0.0001

1.000

Pairwise: SPS male time 1 v. 2

<0.0001

1.000

Pairwise: Control male time 1 v. 2

<0.0001

1.000

Pairwise: SPS female time 1 v. 2

<0.0001

1.000

Pairwise: Control female time 1 v. 2

<0.0001

1.000

69

Baseline
Post-test

Body weight (g)

n=24/gp

200
180
160
140
120
100
80
60
40
20
0

*

*

Ctrl

SPS

Single-housed

*

*

Ctrl

SPS

Pair-housed

Figure 12. Neither SPS nor housing affected female body weight. Rats in all groups showed a significant gain in
body weight between the two time points regardless of traumatic stress exposure. Data are presented as
meanÂąSEM. Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons (Bonferroni). Refer to
Table 3 for full statistical results.

70

Time in interaction zone (sec)

Control Target absent
Control Target present
SPS Target absent
SPS Target present
140

*

*

*

*

120
100
80
60
40
20
0
Single housed

Pair housed

Figure 13. Females spent more total time in the interaction zone when there was a target rat present
compared to an empty chamber, regardless of SPS exposure or housing. Main effect of target (p<0.0001). Data
are presented as meanÂąSEM. Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons
(Bonferroni).

71

Table 3. Statistical results for data shown in Figures 3 and 12. All pairwise comparisons use
Bonferroni adjustment for multiple comparisons. RM denotes repeated measure; otherwise assume
between group measures. Only statistically significant results are shown.
Outcome
Power
Statistical test
Significant effects
p value
measure
(Îą=0.05)
RM 3-way ANOVA
ASR (Fig. 3A)
None
(stress*housing*time)
RM 4-way ANOVA
DST (Fig. 3B)
Main effect: time
<0.0001
1.000
(stress*housing*time*DEX)
Main effect: stress
0.013
0.712
Main effect: DEX

<0.0001

1.000

Interaction: time*stress
0.024
0.623
Because there was no effect of housing on DST, a separate 3-way ANOVA was run for single- and
pair-housed females
DST, singleRM 3-way ANOVA
housed (Fig.
Main effect: time
<0.0001
1.000
(stress*time*DEX)
3B)
Main effect: DEX
<0.0001
0.999
Pairwise: SPS time 1 DEX v.
<0.0001
0.999
veh
Pairwise: control time 1 DEX v.
<0.0001
1.000
veh
Pairwise: SPS DEX time 1 v. 2 <0.0001
0.984
Pairwise: SPS vehicle time 1 v.
<0.0001
0.980
2
Pairwise: control DEX time 1 v.
<0.0001
1.000
2
Pairwise: control vehicle time 1
<0.0001
0.995
v. 2
DST, pairRM 3-way ANOVA
housed (Fig.
Main effect: time
<0.0001
1.000
(stress*time*DEX)
3B)
Main effect: DEX
<0.0001
1.000
Pairwise: DEX time 2 SPS v.
0.020
0.654
control
Pairwise: SPS time 1 DEX v.
<0.0001
1.000
veh
Pairwise: SPS time 2 DEX v.
<0.0001
0.984
veh
Pairwise: control time 1 DEX v.
<0.0001
1.000
veh
Pairwise: control time 2 DEX v.
0.032
0.585
veh
Pairwise: SPS DEX time 1 v. 2 0.007
0.788
Pairwise: SPS vehicle time 1 v.
<0.0001
0.999
2
Pairwise: control DEX time 1 v.
<0.0001
1.000
2
Pairwise: control vehicle time 1
<0.0001
1.000
v. 2

72

Table 3 (cont’d).
Sucrose pref.
(Fig. 3C)
Social
interaction
(Fig. 3D)

2-way ANOVA
(stress*housing)

None

2-way ANOVA
(stress*housing)

Interaction: stress*housing
Pairwise: pair-housed SPS v.
control
Pairwise: single-housed SPS
v. control
Pairwise: SPS pair-housed v.
single-housed

Latency/empty
zone (Fig. 3E)
Body wt.
(Fig. 11)

2-way ANOVA
(stress*housing)
RM 3-way ANOVA
(stress*housing*time)

0.001

0.902

0.009

0.751

0.049

0.505

0.009

0.758

<0.0001

1.000

<0.0001

1.000

<0.0001

1.000

<0.0001

1.000

<0.0001

1.000

None
Main effect: time
Pairwise: SPS single-housed
time 1 v. 2
Pairwise: Control singlehoused time 1 v. 2
Pairwise: SPS pair-housed
time 1 v. 2
Pairwise: Control pair-housed
time 1 v. 2

73

*

n=16/gp

Weight (g)

350
300

Baseline
Post-test

*

*

250

*

200

*

150
100
50
0
Ctrl

PredX
Male

Ctrl

PredX
Female

Figure 14. As with SPS, predator exposure (PredX) did not affect body weight, but females weighed less than
males. Likewise, rats in all groups showed a significant gain in body weight regardless of traumatic stress exposure
or sex. Data are presented as meanÂąSEM. Significance set at P<.05 (indicated by asterisk) for planned pairwise
comparisons (Bonferroni). Refer to Table 4 for full statistical results.

74

Table 4. Statistical results for data shown in Figures 6 and 14. All pairwise comparisons use
Bonferroni adjustment for multiple comparisons. RM denotes repeated measure; otherwise assume
between group measures. Only statistically significant results are shown.
Outcome
Statistical test
Significant effects
p value
Power
measure
(Îą=0.05)
ASR (Fig.
RM 3-way ANOVA
Interaction: stress*sex*time
0.034
0.570
6A)
(stress*sex*time)
Pairwise: SPS male time 1 v. 2
0.016
0.688
DST (Fig.
6B)

RM 4-way ANOVA
(stress*sex*time*DEX
)

Main effect: time

<0.000
1

1.000

Main effect: sex

<0.000
1
<0.000
1
<0.000
1
<0.000
1
<0.000
1
<0.000
1
0.007

1.000

0.012

0.730

0.001

0.918

0.042

0.537

Pairwise: Male PredX time 2 DEX v. veh

0.010

0.752

Pairwise: Male control time 1 DEX v. veh

0.010

0.755

Pairwise: female PredX time 1 DEX v.
veh
Pairwise: female control time 1 DEX v.
veh
Pairwise: Male PredX veh time 1 v. 2

<0.000
1
<0.000
1
0.006

1.000

Pairwise: Male control vehicle time 1 v. 2

0.043

0.532

Pairwise: female PredX DEX time 1 v. 2

1.000

Pairwise: female control vehicle time 1 v.
2
Main effect: sex

<0.000
1
<0.000
1
<0.000
1
<0.000
1
0.012

Pairwise: Control male v. female

0.028

0.621

Main effect: sex

<0.000
1

1.000

Main effect: DEX
Interaction: sex*time
Pairwise: PredX DEX time 2 male v.
female
Pairwise: PredX vehicle time 1 male v.
female
Pairwise: PredX vehicle time 2 male v.
female
Pairwise: control DEX time 2 male v.
female
Pairwise: control vehicle time 1 male v.
female
Pairwise: control vehicle time 2 male v.
female
Pairwise: Male PredX time 1 DEX v. veh

Pairwise: female PredX vehicle time 1 v.
2
Pairwise: female control DEX time 1 v. 2

PVN GR
(Fig. 6C)
Body wt.
(Fig. 13)

2-way ANOVA
(stress*sex)
RM 3-way ANOVA
(stress*sex*time)

75

1.000
0.998
0.995
0.992
0.984
0.784

0.994
0.801

0.999
0.999
0.979
0.756

Table 4 (cont’d).
Main effect: time
Interaction: time*sex
Pairwise: PredX time 1 male v. female
Pairwise: PredX time 2 male v. female
Pairwise: Control time 1 male v. female
Pairwise: Control time 2 male v. female
Pairwise: PredX male time 1 v. 2
Pairwise: Control male time 1 v. 2
Pairwise: PredX female time 1 v. 2
Pairwise: Control female time 1 v. 2

76

<0.000
1
<0.000
1
<0.000
1
<0.000
1
<0.000
1
<0.000
1
<0.000
1
<0.000
1
<0.000
1
<0.000
1

1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000
1.000

Time in interaction zone (sec)

Control Target absent
Control Target present
SPS Target absent
SPS Target present
140

*

120

*

*

*

100
80
60
40
20
0
Male

Female

Figure 15. Rats spent more total time in the interaction zone when there was a target rat present compared to
an empty chamber, regardless of SPS exposure or sex. Main effect of target (p<0.0001). Data are presented as
meanÂąSEM. Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons (Bonferroni).

77

Table 5. Statistical results for data shown in Figure 8. All pairwise comparisons use Bonferroni
adjustment for multiple comparisons. RM denotes repeated measure; otherwise assume between
group measures. Only statistically significant results are shown.
Outcome
Power
Statistical test
Significant effects
p value
measure
(Îą=0.05)
ASR
RM 3-way ANOVA
Main effect: time
0.020
0.655
(Fig. 8A)
(stress*sex*time)
Interaction: stress*time
0.008
0.767

DST
(Fig. 8B)

RM 4-way ANOVA
(stress*sex*time*DEX)

Interaction: sex*time

0.032

0.582

Interaction: stress*sex*time

0.022

0.639

Pairwise: SPS male time 1 v. 2

0.008

0.767

Pairwise: SPS female time 1 v. 2

0.019

0.660

Pairwise: control male time 1 v. 2

0.021

0.644

Pairwise: control female time 1 v. 2

0.033

0.575

Main effect: time

<0.0001

1.000

Main effect: DEX

<0.0001

1.000

Main effect: sex

<0.0001

1.000

Interaction: stress*time

0.005

0.814

Interaction: sex*time

<0.0001

0.999

Interaction: sex*DEX
Pairwise: veh female time 1 SPS v.
control
Pairwise: SPS male time 1 DEX v.
veh
Pairwise: SPS male time 2 DEX v.
veh
Pairwise: SPS female time 1 DEX v.
veh
Pairwise: SPS female time 2 DEX v.
veh
Pairwise: control male time 1 DEX v.
veh
Pairwise: control male time 2 DEX v.
veh
Pairwise: control female time 1 DEX
v. veh
Pairwise: control female time 2 DEX
v. veh
Pairwise: SPS DEX time 2 male v.
female
Pairwise: SPS vehicle time 1 male v.
female

0.004

0.840

<0.0001

0.969

0.002

0.887

0.001

0.914

<0.0001

1.000

0.001

0.950

0.048

0.512

0.010

0.746

<0.0001

1.000

0.001

0.951

0.003

0.875

<0.0001

1.000

78

Table 5 (cont’d).
Pairwise: SPS vehicle time 2 male v.
female
Pairwise: control DEX time 2 male v.
female
Pairwise: control vehicle time 1 male
v. female
Pairwise: control vehicle time 2 male
v. female
Pairwise: SPS DEX female time 1 v.
2
Pairwise: SPS vehicle male time 1 v.
2
Pairwise: SPS vehicle female time 1
v. 2
Pairwise: control DEX male time 1 v.
2
Pairwise: control DEX female time 1
v. 2
Pairwise: control vehicle male time 1
v. 2
Pairwise: control vehicle female time
1 v. 2
Adrenal wt
(Fig. 8C)

PVN GR
(Fig. 8D)

Sucrose pref.
(Fig. 8E)

Social
interaction
(Fig. 8F)

Latency/empt
y zone
(Fig. 8G)
Body weight
(Fig. 8H)

2-way ANOVA
(stress*sex)

2-way ANOVA
(stress*sex)

2-way ANOVA
(stress*sex)

2-way ANOVA
(stress*sex)

2-way ANOVA
(stress*sex)

RM 3-way ANOVA
(stress*sex*time)

<0.0001

0.988

<0.0001

0.960

<0.0001

0.996

<0.0001

0.999

<0.0001

0.993

0.001

0.935

<0.0001

0.978

0.034

0.571

<0.0001

1.000

<0.0001

0.993

<0.0001

1.000

Main effect: sex

<0.0001

1.000

Pairwise: SPS male v. female

<0.0001

1.000

Pairwise: control male v. female

<0.0001

0.998

Interaction: stress*sex

0.005

0.869

Pairwise: control male v. female

0.004

0.876

Pairwise: male SPS v. control

0.020

0.490

Pairwise: female SPS v. control

0.056

0.682

Interaction: stress*sex

0.015

0.699

Pairwise: female SPS v. control

0.034

0.570

Pairwise: SPS male v. female

0.017

0.675

Interaction: stress*sex

0.023

0.632

Pairwise: female SPS v. control

0.023

0.632

Pairwise: control female v. male

0.004

0.849

Main effect: sex

0.011

0.736

Pairwise: control male v. female

0.022

0.639

Main effect: sex

<0.0001

1.000

Main effect: time

<0.0001

1.000

Interaction: time*sex

<0.0001

1.000

79

Table 5 (cont’d).
Pairwise: SPS time 1 male v. female

<0.0001

1.000

Pairwise: SPS time 2 male v. female
Pairwise: Control time 1 male v.
female
Pairwise: Control time 2 male v.
female
Pairwise: SPS male time 1 v. 2

<0.0001

1.000

<0.0001

1.000

<0.0001

1.000

<0.0001

1.000

Pairwise: Control male time 1 v. 2

<0.0001

1.000

Pairwise: SPS female time 1 v. 2

<0.0001

0.911

Pairwise: Control female time 1 v. 2

<0.0001

0.994

80

Weight (g)

n=12-16/gp

400
350
300
250
200
150
100
50
0

Baseline
Post-test

*

*

*

Ctrl

SPS

Sham

*

*

*

*

Ctrl

SPS

Ctrl

SPS

GDX+blank

GDX+T

Figure 16. Neither SPS nor castration (GDX) affected body weight of males. Behind a main effect of GDX was a
significant lower body weight in SPS castrated males compared to SPS sham males, but this effect was present at
baseline before any testing, indicating that particular group had smaller males independently of surgery or SPS. Data
are presented as meanÂąSEM. Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons
(Bonferroni). Refer to Table 6 for full statistical results.

81

Adrenal gland weight (wet
wt (mg)/body wt (g)

n=5-6gp

Control
SPS

0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
Sham

GDX

GDX+T

Figure 17. Neither SPS nor GDX affected adrenal weight of males. Data are presented as meanÂąSEM.
Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons (Bonferroni). Refer to Table 6 for
full statistical results.

82

Table 6. Statistical results for data shown in Figures 9 and 16-17. All pairwise comparisons use
Bonferroni adjustment for multiple comparisons. RM denotes repeated measure; otherwise assume
between group measures. Only statistically significant results are shown.
Outcome
Power
Statistical test
Significant effects
p value
measure
(Îą=0.05)
ASR (Fig. 9A)

RM 3-way ANOVA
(stress*GDX*time)
sham v. GDX

Main effect: time

0.016

0.684

Interaction: stress*time

0.003

0.874

Interaction: GDX*time

0.024

0.627

0.027

0.607

0.001

0.910

Main effect: time

0.001

0.93

Interaction: stress*time

0.046

0.52

0.005

0.817

0.011

0.741

Main effect: time

<0.0001

1.000

Main effect: stress

0.055

0.485

Main effect: DEX

<0.0001

1.000

Main effect: GDX

0.006

0.795

Interaction: stress*time

0.007

0.783

Interaction: DEX*time

<0.0001

1.000

<0.0001

0.974

<0.0001

0.999

<0.0001

0.982

<0.0001

1.000

0.015

0.689

<0.0001

0.980

<0.0001

1.000

<0.0001

1.000

0.004

0.842

Pairwise: SPS sham time 1 v.
time 2
Pairwise: SPS GDX time 1 v. time
2
RM 3-way ANOVA
(stress*GDX*time)
GDX v. GDX+T

Pairwise: SPS GDX time 1 v. time
2
Pairwise: SPS GDX+T time 1 v.
time 2
DST (Fig. 9B)

RM 4-way ANOVA
stress*GDX*time*DEX
sham v. GDX

Pairwise: SPS sham time 1 DEX
v. veh
Pairwise: SPS sham time 2 DEX
v. veh
Pairwise: SPS GDX time 1 DEX
v. veh
Pairwise: SPS GDX time 2 DEX
v. veh
Pairwise: control sham time 1
DEX v. veh
Pairwise: control sham time 2
DEX v. veh
Pairwise: control GDX time 1
DEX v. veh
Pairwise: control GDX time 2
DEX v. veh
Pairwise: control vehicle time 1
sham v. GDX

83

Table 6 (cont’d).
Pairwise: control vehicle time 2
sham v. GDX
Pairwise: SPS vehicle sham time
1 v. time 2
Pairwise: SPS vehicle GDX time
1 v. time 2
Pairwise: control DEX sham time
1 v. time 2
Pairwise: control DEX GDX time
1 v. time 2
Pairwise: control vehicle sham
time 1 v. time 2
Pairwise: control vehicle GDX
time 1 v. time 2
RM 4-way ANOVA
stress*GDX*time*DEX
GDX v. GDX+T

0.011

0.733

<0.0001

1.000

<0.0001

1.000

0.001

0.909

<0.0001

0.993

<0.0001

1.000

<0.0001

1.000

Main effect: time

<0.0001

1.000

Main effect: DEX

<0.0001

1.000

Main effect: GDX

<0.0001

0.979

Interaction: DEX*time

<0.0001

0.989

Interaction: GDX*time

0.003

0.867

Interaction: DEX*GDX

0.033

0.574

<0.0001

0.970

<0.0001

1.000

0.004

0.835

<0.0001

1.000

<0.0001

1.000

0.025

0.621

0.038

0.553

0.020

0.657

<0.0001

1.000

<0.0001

1.000

0.015

0.697

<0.0001

0.996

<0.0001

1.000

Pairwise: SPS GDX time 1 DEX
v. veh
Pairwise: SPS GDX time 2 DEX
v. veh
Pairwise: SPS GDX+T time 2
Dex v. veh
Pairwise: control GDX time 1
DEX v. veh
Pairwise: control GDX time 2
DEX v. veh
Pairwise: control GDX+T time 2
DEX v. veh
Pairwise: control DEX time 2
GDX v. GDX+T
Pairwise: control veh time 1 GDX
v. GDX+T
Pairwise: control veh time 2 GDX
v. GDX+T
Pairwise: SPS veh GDX time 1 v.
time 2
Pairwise: SPS veh GDX+T time 1
v. time 2
Pairwise: control DEX GDX time
1 v. time 2
Pairwise: control veh GDX time 1
v. time 2

84

Table 6 (cont’d).

PVN (Fig. 9C)

2-way ANOVA
(stress*GDX)
sham v. GDX

2-way ANOVA
(stress*GDX)
GDX v. GDX+T
Sucrose (Fig.
9D)

Social int. (Fig.
9E)

2-way ANOVA
(stress*GDX)
sham v. GDX
2-way ANOVA
(stress*GDX)
GDX v. GDX+T
2-way ANOVA
(stress*GDX)
sham v. GDX

2-way ANOVA
(stress*GDX)
GDX v. GDX+T

Latency/empty
zone (Fig. 9F)

2-way ANOVA
(stress*GDX)
sham v. GDX
2-way ANOVA
(stress*GDX)
GDX v. GDX+T

Body weight
(Fig. 15)

RM 3-way ANOVA
(stress*GDX*time)
sham v. GDX

Pairwise: control veh GDX+T
time 1 v. time 2

0.020

0.656

Main effect: stress

0.021

0.670

Pairwise: control GDX v. sham

0.031

0.601

Pairwise: sham SPS v. control

0.010

0.780

Main effect: GDX

0.056

0.485

Interaction: stress*GDX

0.025

0.623

Pairwise: GDX SPS v. control

0.049

0.507

Pairwise: control sham v. GDX

0.004

0.842

Main effect: stress

0.034

0.573

Main effect: GDX

0.066

0.454

Pairwise: GDX SPS v. control

0.035

0.564

Main effect: GDX

0.001

0.934

Pairwise: control sham v. GDX

0.050

0.503

Main effect: GDX

<0.0001

0.992

Pairwise: SPS GDX v. GDX+T

0.003

0.861

Pairwise: control GDX v. GDX+T

0.002

0.883

Main effect: GDX

0.002

0.887

Main effect: time

<0.0001

1.000

0.005

0.815

0.002

0.887

<0.0001

1.000

None

None

None

Pairwise: SPS time 1 sham v.
GDX
Pairwise: SPS time 2 sham v.
GDX
Pairwise: SPS sham time 1 v. 2

85

Table 6 (cont’d).

RM 3-way ANOVA
(stress*GDX*time)
GDX v. GDX+T

Adrenal weight
(Fig 17)

RM 3-way ANOVA
(stress*GDX*time)
sham v. GDX
RM 3-way ANOVA
(stress*GDX*time)
GDX v. GDX+T

Pairwise: SPS GDX time 1 v. 2

<0.0001

1.000

Pairwise: control sham time 1 v. 2

<0.0001

1.000

Pairwise: control GDX time 1 v. 2

<0.0001

1.000

Main effect: time

<0.0001

1.000

Pairwise: SPS GDX time 1 v. 2

<0.0001

1.000

Pairwise: SPS GDX+T time 1 v. 2

<0.0001

1.000

Pairwise: control GDX time 1 v. 2

<0.0001

1.000

Pairwise: control GDX+T time 1
v. 2

<0.0001

1.000

None

None

86

300

Weight (g)

250
200

n=12/gp

*

*

Ctrl

SPS

$

*

$
$

*$

Baseline
Post-test
$
$
*
$
$*

150
100
50
0
Sham+blank

Ctrl

SPS
GDX

Ctrl

SPS

GDX+T

Figure 18. SPS did not affect the body weight of females, but as expected, GDX did by increasing the body
weight of ovariectomized females. Interestingly, the effect of GDX was not reversed by testosterone (T) treatment.
Data are presented as meanÂąSEM. Significance set at P<.05 (indicated by * or $) for planned pairwise comparisons
(Bonferroni). *, vs baseline same group; $, vs sham same group. Refer to Table 7 for full statistical results.

87

Control
SPS

Adrenal gland weight (wet
wt (mg)/body wt (g)

n=5-6gp

0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
Sham

GDX

GDX+T

Figure 19. Neither SPS nor GDX affected adrenal weight of females. Data are presented as meanÂąSEM.
Significance set at P<.05 (indicated by asterisk) for planned pairwise comparisons (Bonferroni). Refer to Table 7 for
full statistical results.

88

Table 7. Statistical results for data shown in Figures 10 and 18-19. All pairwise comparisons use
Bonferroni adjustment for multiple comparisons. RM denotes repeated measure; otherwise assume
between group measures. Only statistically significant results are shown.
Outcome
Power
Statistical test
Significant effects
p value
measure
(Îą=0.05)
ASR (Fig. 10A)

RM 3-way ANOVA
(stress*GDX*time)
sham vs GDX

RM 3-way ANOVA
(stress*GDX*time)
GDX v. GDX+T

DST (Fig. 10B)

RM 4-way ANOVA
(stress*GDX*time*DEX)
sham v. GDX

Main effect: time

<0.0001

0.972

Pairwise: SPS sham time 1 v. 2

0.024

0.625

Pairwise: Control sham time 1 v. 2

0.040

0.541

Pairwise: Control GDX time 1 v. 2

0.013

0.715

Main effect: time

<0.0001

0.705

Pairwise: Control GDX time 1 v. 2

0.037

0.555

Pairwise: Control GDX+T time 1
v. 2

0.045

0.524

Main effect: time

<0.0001

1.000

Main effect: DEX

<0.0001

1.000

Interaction: stress*time*GDX

0.044

0.526

<0.0001

0.962

<0.0001

1.000

0.002

0.897

<0.0001

1.000

0.027

0.608

<0.0001

1.000

0.002

0.899

<0.0001

1.000

0.038

0.550

0.012

0.730

<0.0001

0.979

<0.0001

0.997

0.001

0.946

Pairwise: vehicle sham time 1
SPS v. control
Pairwise: SPS sham time 1 DEX
v. veh
Pairwise: SPS sham time 2 DEX
v. veh
Pairwise: SPS GDX time 1 DEX v.
veh
Pairwise: SPS GDX time 2 DEX v.
veh
Pairwise: control sham time 1
DEX v. veh
Pairwise: control sham time 2
DEX v. veh
Pairwise: control GDX time 1 DEX
v. veh
Pairwise: SPS vehicle time 1
sham v. GDX
Pairwise: control vehicle time 2
sham v. GDX
Pairwise: SPS DEX sham time 1
v. time 2
Pairwise: SPS DEX GDX time 1 v.
time 2
Pairwise: SPS vehicle sham time
1 v. time 2

89

Table 7 (cont’d).
Pairwise: SPS vehicle GDX time 1
v. time 2
Pairwise: control DEX sham time
1 v. time 2
Pairwise: control DEX GDX time 1
v. time 2
Pairwise: control vehicle sham
time 1 v. time 2
Pairwise: control vehicle GDX
time 1 v. time 2
RM 4-way ANOVA
stress*GDX*time*DEX
GDX vs GDX+T

<0.0001

0.989

<0.0001

1.000

<0.0001

1.000

<0.0001

1.000

0.005

0.821

Main effect: time

<0.0001

1.000

Main effect: DEX

<0.0001

1.000

Main effect: GDX

0.001

0.926

Interaction: time*DEX*GDX

0.041

0.539

<0.0001

1.000

0.015

0.697

0.004

0.850

0.008

0.775

<0.0001

1.000

0.051

0.502

<0.0001

0.999

<0.0001

0.974

<0.0001

1.000

0.009

0.764

<0.0001

1.000

<0.0001

0.999

<0.0001

1.000

<0.0001

0.997

0.001

0.953

0.001

0.957

Pairwise: SPS GDX time 1 DEX v.
veh
Pairwise: SPS GDX time 2 DEX v.
veh
Pairwise: SPS GDX+T time 1 Dex
v. veh
Pairwise: SPS GDX+T time 2 Dex
v. veh
Pairwise: control GDX time 1 DEX
v. veh
Pairwise: control GDX+T time 1
Dex v. veh
Pairwise: SPS vehicle time 1 GDX
v. GDX+1
Pairwise: control veh time 1 GDX
v. GDX+T
Pairwise: SPS DEX GDX time 1 v.
time 2
Pairwise: SPS DEX GDX+T time
1 v. time 2
Pairwise: SPS veh GDX time 1 v.
time 2
Pairwise: SPS veh GDX+T time 1
v. time 2
Pairwise: control DEX GDX time 1
v. time 2
Pairwise: control DEX GDX+T
time 1 v. time 2
Pairwise: control veh GDX time 1
v. time 2
Pairwise: control veh GDX+T time
1 v. time 2

90

Table 7 (cont’d).
PVN (Fig. 10C)

2-way ANOVA
(stress*GDX)
sham v. GDX

2-way ANOVA
(stress*GDX)
GDX v. GDX+T

Sucrose (Fig.
10D)

Social int.
(Fig. 10E)

2-way ANOVA
(stress*GDX)
sham v. GDX

Body weight
(Fig. 17)

0.010

0.780

Pairwise: SPS GDX v. sham

0.006

0.848

Pairwise: sham SPS v. control

0.023

0.654

Main effect: GDX

<0.0001

0.999

Main effect: stress

0.007

0.834

Pairwise: SPS GDX v. GDX+T

0.001

0.977

Pairwise: control GDX v. GDX+T

0.002

0.934

Pairwise: GDX SPS v. control

0.017

0.701

Main effect: GDX

0.046

0.518

Main effect: stress

0.057

0.480

Pairwise: sham SPS v. control

0.029

0.596

Pairwise: SPS sham v. GDX

0.027

0.608

2-way ANOVA
(stress*GDX)
GDX v. GDX+T

None

2-way ANOVA
(stress*GDX)
sham v. GDX

Main effect: stress

0.014

0.702

Pairwise: sham SPS v. control

0.039

0.549

Main effect: stress

0.044

0.528

Pairwise: GDX SPS v. control

0.031

0.585

Main effect: stress

0.016

0.690

Pairwise: GDX+T SPS v. control

0.008

0.771

Pairwise: SPS GDX v. GDX+T

0.018

0.670

Main effect: GDX

0.002

0.887

Main effect: time

<0.0001

1.000

0.005

0.815

0.002

0.887

2-way ANOVA
(stress*GDX)
GDX v. GDX+T

Latency/empty
zone (Fig. 10F)

Interaction: stress*GDX

2-way ANOVA
(stress*GDX)
sham v. GDX
2-way ANOVA
(stress*GDX)
GDX v. GDX+T

RM 3-way ANOVA
(stress*GDX*time)
sham v. GDX

None

Pairwise: SPS time 1 sham v.
GDX
Pairwise: SPS time 2 sham v.
GDX

91

Table 7 (cont’d).

RM 3-way ANOVA
(stress*GDX*time)
GDX v. GDX+T

Adrenal
weight
(Fig. 18)

RM 3-way ANOVA
(stress*GDX*time)
Sham v. GDX
RM 3-way ANOVA
(stress*GDX*time)
GDX v. GDX+T

Pairwise: SPS sham time 1 v. 2

<0.0001

1.000

Pairwise: SPS GDX time 1 v. 2

<0.0001

1.000

Pairwise: control sham time 1 v. 2

<0.0001

1.000

Pairwise: control GDX time 1 v. 2

<0.0001

1.000

Main effect: time

<0.0001

1.000

Pairwise: SPS GDX time 1 v. 2

<0.0001

1.000

Pairwise: SPS GDX+T time 1 v. 2

<0.0001

1.000

Pairwise: control GDX time 1 v. 2

<0.0001

1.000

Pairwise: control GDX+T time 1 v.
2

<0.0001

1.000

None

None

92

DST

ASR
2500

6000
4000
2000
0
-2000
-4000

60

50

2000

GR-IR (cells/mm3) x103

Plasma CORT release (nmol/L) after 30min
restraint (from baseline at 0min)

ASR change from baseline (A.U./kg b.w.)

8000

-6000

PVN GR

1500

1000

500

SPS

30

20

10

0
Control

40

Vehicle

0

DEX

Control

SPS

Figure 20. Individuals are not uniformly susceptible or resilient across measures. SPS affects one individual male (red triangle) differently than another (blue
circle) across measures of acoustic startle response (ASR), dexamethasone (DEX) suppression test (DST), and glucocorticoid receptor (GR) expression in the
paraventricular nucleus of the hypothalamus (PVN). All individuals in the control groups are shown with the mean (black circle) and standard deviation (black
bars). Responses that lie outside the standard deviation are considered “affected” by SPS and are denoted with a black box, an approach used to infer
susceptibility in chronic unpredictable and acute stress paradigms in rodents (343). While one individual is affected by SPS in terms of ASR, that rat is unaffected
in terms of DST—and vice versa for another individual. Both individuals, however, are equally affected by SPS in terms of GR expression in the PVN, which
challenges the notion that an individual is either resilient or susceptible to trauma.

93

REFERENCES

94

REFERENCES

1.

Kessler RC, Sonnega A, Bromet E, Hughes M, Nelson C (1995) Posstraumatic Stress Disorder in
the National Comorbidity Survey. Arch Gen Psychiatry 52:1048–1060.

2.

Kessler RC, et al. (2005) Lifetime Prevalence and Age-of-Onset Distributions of DSM-IV Disorders
in the National Comorbidity Survey Replication. Arch Gen Psychiatry 62(June 2005):593–602.

3.

Breslau N, Davis GC, Andreski P, Peterson E (1991) Traumatic Events and Posttraumatic Stress
Disorder in an Urban Population of Young Adults. Arch Gen Psychiatry 48:216–222.

4.

Cifu DX, et al. (2014) Traumatic brain injury, posttraumatic stress disorder, and pain diagnoses in
OIF/OEF/OND Veterans. J Rehabil Res Dev 50(9):1169–76.

5.

Orr SP, Pitman RK, Lasko NB, Herz LR (1993) Psychophysiological assessment of posttraumatic
stress disorder imagery in World War II and Korean combat veterans. J Abnorm Psychol
102(1):152–9.

6.

Sledjeski EM, Speisman B, Dierker LC (2008) Does number of lifetime traumas explain the
relationship between PTSD and chronic medical conditions? Answers from the National
Comorbidity Survey-Replication (NCS-R). J Behav Med 31(4):341–9.

7.

Seyle H (1936) A Syndrome Produced by Diverse Nocuous Agents. Nature 138:32.

8.

PacĂĄk K, Palkovits M (2001) Stressor specificity of central neuroendocrine responses: implications
for stress-related disorders. Endocr Rev 22(4):502–48.

9.

Boccia M, et al. (2016) Different neural modifications underpin PTSD after different traumatic
events: an fMRI meta-analytic study. Brain Imaging Behav 10:226–237.

10.

Freed S, D’Andrea W (2015) Autonomic Arousal and Emotion in Victims of Interpersonal
Violence : Shame Proneness But Not Anxiety Predicts Vagal Tone Autonomic Arousal and
Emotion in Victims of Interpersonal Violence : Shame Proneness. J trauma dissociation
16(September):367–383.

11.

Foa EB, Zinbarg R, Rothbaum BO (1992) Uncontrollability and unpredictability in post-traumatic
stress disorder: an animal model. Psychol Bull 112(2):218–38.

12.

Clinchy M, et al. (2010) The Neurological Ecology of Fear: Insights Neuroscientists and Ecologists
Have to Offer one Another. Front Behav Neurosci 4(April):21.

13.

Hawlena D, Schmitz OJ (2010) Physiological stress as a fundamental mechanism linking
predation to ecosystem functioning. Am Nat 176(5):537–56.

14.

Creel S, Christianson D (2008) Relationships between direct predation and risk effects. Trends
Ecol Evol 23(4):194–201.

15.

Parker KJ, Maestripieri D (2011) Identifying key features of early stressful experiences that
produce stress vulnerability and resilience in primates. Neurosci Biobehav Rev 35(7):1466–1483.

16.

Huether G (1996) The central adaptation syndrome: psychosocial stress as a trigger for adaptive
modifications of brain structure and brain function. Prog Neurobiol 48(6):569–612.

95

17.

Bracha HS (2006) Human brain evolution and the “Neuroevolutionary Time-depth Principle:”
Implications for the Reclassification of fear-circuitry-related traits in DSM-V and for studying
resilience to warzone-related posttraumatic stress disorder. Prog Neuropsychopharmacol Biol
Psychiatry 30(5):827–53.

18.

Tedstone JE, Tarrier N (2003) Posttraumatic stress disorder following medical illness and
treatment. Clin Psychol Rev 23(3):409–448.

19.

Amat J, et al. (2005) Medial prefrontal cortex determines how stressor controllability affects
behavior and dorsal raphe nucleus. Nat Neurosci 8(3):365–71.

20.

Leuner B, Mendolia-Loffredo S, Shors TJ (2004) Males and females respond differently to
controllability and antidepressant treatment. Biol Psychiatry 56(12):964–970.

21.

Turnbull GJ (1998) A review of post-traumatic stress disorder. Part I: Historical development and
classification. Injury 29(2):87–91.

22.

Dimauro J, Carter S, Folk JB, Kashdan TB (2014) A historical review of trauma-related diagnoses
to reconsider the heterogeneity of PTSD. J Anxiety Disord 28(8):774–786.

23.

Birmes P, Hatton L, Brunet A, Schmitt L (2003) Early historical literature for post-traumatic
symptomatology. Stress Heal 19(1):17–26.

24.

Shay BJ (1991) Learning About Combat Stress from Homer ‘ s Iliad. J Trauma Stress 4(4):561–
579.

25.

Jones E, et al. (2013) Flashbacks and post-traumatic stress disorder : the genesis of a 20thcentury diagnosis Flashbacks and post-traumatic stress disorder : the genesis of a 20th-century
diagnosis. Br J Psychiatry 182:158–163.

26.

Lasiuk GC, Hegadoren KM (2006) Posttraumatic stress disorder part I: historical development of
the concept. Perspect Psychiatr Care 42(1):13–20.

27.

van der Kolk BA, Herron N, Hostetler A (1994) The History of Trauma of Psychiatry. Pyschiatric
Clin North Am 17(3):583–600.

28.

Erichsen JE (1866) THE CLASSIC: Injuries of the Nervous System. Clin Orthop Relat Res
458:47–51.

29.

Page HW (1891) Railway Injuries: In their Medico-Legal and Clinical Aspects.

30.

Freud S (1896) The Aetiology of Hysteria. 170–189.

31.

Myers CS (1915) A Contribution to the Study of Shell Shock. Lancet 188:316–320.

32.

Hartshorne H (1864) Summary of the transactions of the College of Physicians of Philadelphia.
1863. June 3. On heart disease in the army. Am J Med Sci 48(7):59–63.

33.

Chodoff P (1963) Late Effects of the Concentration Camp Syndrome. Arch Gen Psychiatry
8(4):323–333.

34.

Burgess AW, Holmstrom LL (1974) Rape Trauma Syndrome. Am J Psychiatry 131(9):981–986.

35.

Friedman MJ (2013) Finalizing PTSD in DSM-5 : Getting Here From There and Where to Go Next.
J Trauma Stress (26):548–556.

96

36.

Pitman RK (1989) Post-Traumatic Stress Disorder, Hormones, and Memory. Biol Psychiatry
26:221–223.

37.

Rausch S, et al. (2016) Women with exposure to childhood interpersonal violence without
psychiatric diagnoses show no signs of impairment in general functioning, quality of life and
sexuality. Borderline Personal Disord Emot Dysregulation 3(1):13.

38.

Copeland WE, Keeler G, Angold A, Costello EJ (2007) Traumatic Events and Posttraumatic Stress
in Childhood. Arch Gen Psychiatry 64(5):577.

39.

Lowe SR, Galea S, Uddin M, Koenen KC (2014) Trajectories of Posttraumatic Stress Among
Urban Residents. Am J Community Psychol 53(0):159–172.

40.

Daskalakis NP, Bagot RC, Parker KJ, Vinkers CH, de Kloet ER (2013) The three-hit concept of
vulnerability and resilience: Toward understanding adaptation to early-life adversity outcome.
Psychoneuroendocrinology 38(9):1858–1873.

41.

True WR, et al. (1993) A twin study of genetic and environmental contributions to liability for
posttraumatic stress symptoms. Arch Gen Psychiatry 50:257–264.

42.

Stein MB, Jang K, Taylor S, Vernon P, Livesley WJ (2002) Genetic and environmental influences
on trauma exposure and ... Am J Psychiatry 159(10):1675–1681.

43.

Xie P, et al. (2009) Interactive Effect of Stressful Life Events and the Serotonin Transporter 5HTTLPR Genotype on Posttraumatic Stress Disorder Diagnosis in 2 Independent Populations.
Arch Gen Psychiatry 66(11):1201.

44.

Walsh K, Uddin M, Soliven R, Wildman DE, Bradley B (2014) Associations between the SS variant
of 5-HTTLPR and PTSD among adults with histories of childhood emotional abuse: Results from
two African American independent samples. J Affect Disord 161:91–96.

45.

Zhang L, et al. (2013) PTSD risk is associated with BDNF Val66Met and BDNF overexpression.
Mol Psychiatry:1–2.

46.

Frielingsdorf H, et al. (2010) Variant brain-derived neurotrophic factor Val66Met endophenotypes:
implications for posttraumatic stress disorder. Ann N Y Acad Sci 1208:150–7.

47.

Ressler KJ, et al. (2011) Post-traumatic stress disorder is associated with PACAP and the PAC1
receptor. Nature 470(7335):492–7.

48.

Uddin M, et al. (2013) ADCYAP1R1 GENOTYPE, POSTTRAUMATIC STRESS DISORDER, AND
DEPRESSION AMONG WOMEN EXPOSED TO CHILDHOOD MALTREATMENT. Depress
Anxiety 30(3):251–258.

49.

Gillespie CF, et al. (2013) Sex dependent influence of a functional polymorphism in steroid 5-Îąreductase type 2 (SRD5A2) on post-traumatic stress symptoms. Am J Med Genet Part B,
Neuropsychiatr Genet 162B(3):283–92.

50.

Lowe SR, et al. (2015) RORA and posttraumatic stress trajectories: Main effects and interactions
with childhood physical abuse history. Brain Behav 5(4):1–11.

51.

Wingo AP, et al. (2015) DICER1 and microRNA regulation in post-traumatic stress disorder with
comorbid depression. Nat Commun 6:10106.

97

52.

Norrholm SD, Ressler KJ (2009) Genetics of anxiety and trauma-related disorders. Neuroscience
164(1):272–87.

53.

Bird A (2002) DNA methylation patterns and epigenetic memory DNA methylation patterns and
epigenetic memory. Genes Dev:6–21.

54.

Uddin M, et al. (2010) Epigenetic and immune function profiles associated with posttraumatic
stress disorder. Proc Natl Acad Sci 107(20).

55.

Perroud N, et al. (2011) Increased methylation of glucocorticoid receptor gene (NR3C1) in adults
with a history of childhood maltreatment: a link with the severity and type of trauma. Transl
Psychiatry 1(12):e59.

56.

Smith AK, et al. (2011) Differential immune system DNA methylation and cytokine regulation in
post-traumatic stress disorder. Am J Med Genet Part B, Neuropsychiatr Genet 156B(6):700–8.

57.

Sipahi L, et al. (2014) Longitudinal epigenetic variation of DNA methyltransferase genes is
associated with vulnerability to post-traumatic stress disorder. Psychol Med (2014):1–15.

58.

LabontĂŠ B, Azoulay N, Yerko V, Turecki G, Brunet a (2014) Epigenetic modulation of
glucocorticoid receptors in posttraumatic stress disorder. Transl Psychiatry 4(November
2013):e368.

59.

Yehuda R, et al. (2014) Influences of Maternal and Paternal PTSD on Epigenetic Regulation of the
Glucocorticoid Receptor Gene in Holocaust Survivor Offspring. Am J Psychiatry 171:872–880.

60.

Yehuda R, Bell A, Bierer LM, Schmeidler J (2008) Maternal, not paternal, PTSD is related to
increased risk for PTSD in offspring of Holocaust survivors. J Psychiatr Res 42(13):1104–11.

61.

Schechter DS, et al. (2015) Methylation of NR3C1 is related to maternal PTSD, parenting stress
and maternal medial prefrontal cortical activity in response to child separation among mothers with
histories of violence exposure. Front Psychol 6(May):1–12.

62.

Rodgers AB, Bale TL (2015) Germ Cell Origins of Posttraumatic Stress Disorder Risk: The
Transgenerational Impact of Parental Stress Experience. Biol Psychiatry 78(5):307–314.

63.

Bader HN, et al. (2014) Maternal Age at Holocaust Exposure and Maternal PTSD Independently
Influence Urinary Cortisol Levels in Adult Offspring. Front Endocrinol (Lausanne) 5(July):103.

64.

McFarlane AC, Atchison M, Yehuda R (1997) The acute stress response following motor vehicle
accidents and its relation to PTSD. Psychobiol Posttraumatic Stress Disord 821:437–441.

65.

Delahanty DL, Raimonde AJ, Spoonster E (2000) Initial posttraumatic urinary cortisol levels
predict subsequent PTSD symptoms in motor vehicle accident victims. Biol Psychiatry 48(9):940–
947.

66.

Gilbertson MW, et al. (2002) Smaller hippocampal volume predicts pathologic vulnerability to
psychological trauma. Nat Neurosci 5(11):1242–1247.

67.

Andrews JA, Neises KD (2012) Cells, biomarkers, and post-traumatic stress disorder: evidence for
peripheral involvement in a central disease. J Neurochem 120:26–36.

68.

Yehuda R, et al. (2000) Low cortisol and risk for PTSD in adult offspring of holocaust survivors.
Am J Psychiatry 157(8):1252–9.

98

69.

Karam EG, et al. (2014) Cumulative Traumas and Risk Thresholds: 12-Month PTSD in the World
Mental Health (WMH) Surveys. Depress Anxiety 31(2):130–142.

70.

Frans O, RimmĂś P, Aberg L, Fredrikson M (2005) Trauma exposure and post-traumatic stress
disorder in the general population. Acta Psychiatr Scand 111(4):291–9.

71.

McEwen BS (1998) Stress, Adaptation, and Disease: Allostasis and Allostatic Load. Ann N Y Acad
Sci 840(1):33–44.

72.

Kilpatrick DG, et al. (2013) National Estimates of Exposure to Traumatic Events and PTSD
Prevalence Using DSM-IV and DSM-5 Criteria. J Trauma Stress 26:537–547.

73.

Ozer EJ, Best SR, Lipsey TL, Weiss DS (2003) Predictors of posttraumatic stress disorder and
symptoms in adults: A meta-analysis. Psychol Bull 129(1):52–73.

74.

Ullman SE, Relyea M, Peter-Hagene L, Vasquez AL (2013) Trauma histories, substance use
coping, PTSD, and problem substance use among sexual assault victims. Addict Behav
38(6):2219–2223.

75.

Molnar BE, Buka SL, Kessler RC (2001) Child sexual abuse and subsequent psychopathology:
results from the National Comorbidity Survey. Am J Public Health 91(5):753–60.

76.

Freyd JJ (1994) Betrayal Trauma: Traumatic Amnesia as an Adaptive Response to Childhood
Abuse. Ethics Behav 4(4):307–329.

77.

Goldsmith RE, Freyd JJ, DePrince a. P (2012) Betrayal Trauma: Associations With Psychological
and Physical Symptoms in Young Adults. J Interpers Violence 27:547–567.

78.

Spindler H, Elklit A, Christiansen D (2010) Risk factors for posttraumatic stress disorder following
an industrial disaster in a residential area: A note on the origin of observed gender differences.
Gend Med 7(2):156–65.

79.

Ehlers A, Mayou R, Bryant B (1998) Psychological predictors of chronic posttraumatic stress
disorder after motor vehicle accidents. J Abnorm Psychol 107(3):508–519.

80.

Lee CS, et al. (2009) Acculturation, psychiatric comorbidity and posttraumatic stress disorder in a
Taiwanese aboriginal population. Soc Psychiatry Psychiatr Epidemiol 44(1):55–62.

81.

Marsella AJ (2010) Ethnocultural Aspects of PTSD: An Overview of Concepts, Issues, and
Treatments. Traumatology (Tallahass Fla) 16(4):17–26.

82.

Hinton DE, Lewis-FernĂĄndez R (2011) The cross-cultural validity of posttraumatic stress disorder:
implications for DSM-5. Depress Anxiety 28(9):783–801.

83.

Algeria M, et al. (2013) Prevalence, Risk, and Correlates of Posttraumatic Stress Disorder across
Ethnic and Racial Minority Groups in the U.S. Med Care 51(12):1114–1123.

84.

Marshall GN, Orlando M (2002) Acculturation and peritraumatic dissociation in young adult Latino
survivors of community violence. J Abnorm Psychol 111(1):166–174.

85.

Fortuna LR, Porche M V., Alegria M (2008) Political violence, psychosocial trauma, and the
context of mental health services use among immigrant Latinos in the United States. Ethn Health
13(5):435–463.

99

86.

Hobfoll SE (2012) Conservation of resources and disaster in cultural context: the caravans and
passageways for resources: commentary on "a conceptual framework for understanding the
mental health impacts of oil spills: lessons from the Exxon Valdez oil spill. Psychiatry 75(3):227–
232.

87.

Smid GE, Mooren TTM, Van Der Mast RC, Gersons BPR, Kleber RJ (2009) Delayed
posttraumatic stress disorder: Systematic review, meta-analysis, and meta-regression analysis of
prospective studies. J Clin Psychiatry 70(11):1572–1582.

88.

Shively C, Kaplan J (1984) Effects of social factors on adrenal weight and related physiology of
Macaca fascicularis. Physiol Behav 33(5):777–782.

89.

Abbott DH, et al. (2003) Are subordinates always stressed? A comparative analysis of rank
differences in cortisol levels among primates. Horm Behav 43(1):67–82.

90.

Golub MS, Sassenrath EN, Goo GP (1979) Plasma cortisol levels and dominance in peer groups
of rhesus monkey weanlings. Horm Behav 12(1):50–59.

91.

Russo SJ, Murrough JW, Han M-H, Charney DS, Nestler EJ (2012) Neurobiology of resilience.
Nat Neurosci 15(11):1475–84.

92.

Caspi A, Moffitt TE (2006) Gene – environment interactions in psychiatry : joining forces with
neuroscience. 7(July):583–590.

93.

Belsky J, Beaver KM (2011) Cumulative-genetic plasticity, parenting and adolescent selfregulation. J Child Psychol Psychiatry 52(5):619–626.

94.

Ogle CM, Rubin DC, Siegler IC (2013) The Impact of the Developmental Timing of Trauma
Exposure on PTSD Symptoms and Psychosocial Functioning Among Older Adults. Dev Psychol
49(11):1–18.

95.

Ehring T, Quack D (2010) Emotion Regulation Difficulties in Trauma Survivors: The Role of
Trauma Type and PTSD Symptom Severity. Behav Ther 41(4):587–598.

96.

Stevens NR, et al. (2013) Emotion Regulation Difficulties, Low Social Support, and Interpersonal
Violence Mediate the Link Between Childhood Abuse and Posttraumatic Stress Symptoms. Behav
Ther 44(1):152–161.

97.

Teicher MH, Andersen SL, Polcari A, Anderson CM, Navalta CP (2002) Developmental
neurobiology of childhood stress and trauma. Psychiatr Clin North Am 25(2):397–426.

98.

Widom CS, DuMont K, Czaja SJ (2007) A Prospective Investigation of Major Depressive Disorder
and Comorbidity in Abused and Neglected Children Grown Up. Arch Gen Psychiatry 64:49–56.

99.

Molnar BE, Berkman LF, Buka SL (2001) Psychopathology, childhood sexual abuse and other
childhood adversities: relative links to subsequent suicidal behaviour in the US. Psychol Med
31(6):965–977.

100.

Brewin CR, Andrews B, Valentine JD (2000) Meta-analysis of risk factors for posttraumatic stress
disorder in trauma-exposed adults. J Consult Clin Psychol 68(5):748–766.

101.

Anda RF, et al. (2006) The enduring effects of abuse and related adverse experiences in
childhood: A convergence of evidence from neurobiology and epidemiology. Eur Arch Psychiatry
Clin Neurosci 256(3):174–186.

100

102.

Felitti VJ, et al. (1998) Relationship of childhood abuse and household dysfunction to many of the
leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study.[see
comment]. Am J Prev Med 14(4):245–258.

103.

Cloitre M, et al. (2009) A Developmental Approach to Complex PTSD: Childhood and Adult
Cumulative Trauma as Predictors of Symptom Complexity. J Trauma Stress 22(5):399–408.

104.

Bustamante AC, et al. (2016) Glucocorticoid receptor DNA methylation, childhood maltreatment
and major depression. J Affect Disord 206:181–188.

105.

Nugent NR, Goldberg A, Uddin M (2015) Topical Review: The Emerging Field of Epigenetics:
Informing Models of Pediatric Trauma and Physical Health. J Pediatr Psychol 41(March 2015):1–
8.

106.

Teicher MH, et al. (2003) The neurobiological consequences of early stress and childhood
maltreatment. Neurosci Biobehav Rev 27(1–2):33–44.

107.

Sapolsky RM, Krey LC, McEwen BS (1986) The Neuroendocrinology of Stress and Aging: The
Glucorticoid Cascade Hypothesis. Endocr Rev 7(3):284–301.

108.

Zoladz PR, Diamond DM (2013) Current status on behavioral and biological markers of PTSD: A
search for clarity in a conflicting literature. Neurosci Biobehav Rev 37(5):860–95.

109.

Mcfarlane AC (2000) Posttraumatic Stress Disorder : A Model of the Longitudinal Course and the
Role of Risk Factors r py Co ig a du es r Co ht ig In. J Clin Psychiatry 61(suppl 5):15–20.

110.

Yehuda R, McFarlane AC, Shalev AY (1998) Predicting the development of posttraumatic stress
disorder from the acute response to a traumatic event. Biol Psychiatry 44(12):1305–1313.

111.

Andrews B, Brewin CR, Rose S (2003) Gender, social support, and PTSD in victims of violent
crime. J Trauma Stress 16(4):421–427.

112.

Morris MC, Rao U (2013) Psychobiology of PTSD in the acute aftermath of trauma: Integrating
research on coping, HPA function and sympathetic nervous system activity. Asian J Psychiatr
6(1):3–21.

113.

Galatzer-Levy IR, Bryant RA (2013) 636,120 Ways To Have Posttraumatic Stress Disorder.
Perspect Psychol Sci 8(6):651–662.

114.

van der Kolk BA, Van der Hart O (1989) Pierre Janet and the breakdown of adaption in
psychological trauma. Am J Psychiatry 146(12):1530–1540.

115.

Janet P (1889) L’automatisme Psychologique ed Alcan F (Ancienne Librairie Germer Bailliere et
Cie, Paris) doi:10.1176/ajp.148.12.1730.

116.

Steuwe C, Lanius RA, Frewen PA (2012) Evidence for a dissociative subtype of PTSD by latent
profile and confirmatory factor analyses in a civilian sample. Depress Anxiety 29(8):689–700.

117.

Wolf EJ, Miller MW, Harrington KM, Reardon A (2012) Personality-based latent classes of
posttraumatic psychopathology: personality disorders and the internalizing/externalizing model. J
Abnorm Psychol 121(1):256–62.

118.

Wolf EJ, et al. (2012) A Latent Class Analysis of Dissociation and PTSD: Evidence for a
Dissociative Subtype. Arch Gen Psychiatry 69(7):698–705.

101

119.

Stein DJ, et al. (2013) Dissociation in Posttraumatic Stress Stress Disorder: Evidence from the
World Mental Health Surveys. Biol Psychiatry 73(4):302–312.

120.

Wolf EJ, et al. (2012) The dissociative subtype of PTSD: A replication and extension. Depress
Anxiety 29(8):679–688.

121.

Lanius RA, et al. (2005) Functional connectivity of dissociative responses in posttraumatic stress
disorder: A functional magnetic resonance imaging investigation. Biol Psychiatry 57(8):873–884.

122.

Lanius RA, et al. (2010) Emotion Modulation in PTSD: Clinical Neurobiological Evidence for a
Dissociative Subtype. Am J Psychiatry 167(6):640–647.

123.

Lanius RA, Brand B, Vermetten E, Frewen PA, Spiegel D (2012) The dissociative subtype of
posttraumatic stress disorder: rationale, clinical and neurobiological evidence, and implications.
Depress Anxiety 29(8):701–8.

124.

Hopper JW, Frewen PA, Lanius RA (2007) Neural Correlates of Reexperiencing , Avoidance , and
Dissociation in PTSD : Symptom Dimensions and Emotion Dysregulation in Responses to ScriptDriven Trauma Imagery. J Trauma Stress 20(5):713–725.

125.

Simeon D, et al. (2007) Hypothalamic-Pituitary-Adrenal Axis Function in Dissociative Disorders,
Post-Traumatic Stress Disorder, and Healthy Volunteers. Biol Psychiatry 61(8):966–973.

126.

Griffin MG, Resick PA, Mechanic MB (1997) Objective Assessment of Peritraumatic Dissociation:
Psychophysiological Indicators. Am J Psychiatry 154(8):1081–1088.

127.

Bernstein RE, Delker BC, Knight JA, Freyd JJ (2015) Hypervigilance in college students :
Associations with betrayal and dissociation and psychometric properties in a brief hypervigilance
scale. Psychol Trauma 7(5):448–455.

128.

Smith CP, Freyd JJ (2013) Dangerous Safe Havens: Institutional Betrayal Exacerbates Sexual
Trauma. J Trauma Stress 26(February):119–127.

129.

Mehta D, et al. (2013) Childhood maltreatment is associated with distinct genomic and epigenetic
profiles in posttraumatic stress disorder. Proc Natl Acad Sci 110(20):8302–8307.

130.

Schmid M, Petermann F, Fegert JM (2013) Developmental trauma disorder: pros and cons of
including formal criteria in the psychiatric diagnostic systems. BMC Psychiatry 13(1):3.

131.

Rahim M (2014) Developmental trauma disorder: An attachment-based perspective. Clin Child
Psychol Psychiatry 19(4):548–560.

132.

Ford JD, et al. (2013) Clinical significance of a proposed developmental trauma disorder
diagnosis: Results of an international survey of clinicians. J Clin Psychiatry 74(8):841–849.

133.

Stolbach BC, et al. (2013) Complex Trauma Exposure and Symptoms in Urban Traumatized
Children: A Preliminary Test of Proposed Criteria for Developmental Trauma Disorder. J Trauma
Stress 26:483–491.

134.

Andrews B, Brewin CR, Philpott R, Stewart L (2007) Delayed-onset posttraumatic stress disorder:
a systematic review of the evidence. Am J Psychiatry 164(9):1319–26.

135.

Utzon-Frank N, et al. (2014) Occurrence of delayed-onset post-traumatic stress disorder: A
systematic review and meta-analysis of prospective studies. Scand J Work Environ Health
40(3):215–229.

102

136.

Friedman MJ, Resick PA, Bryant RA, Brewin CR (2011) Considering PTSD for DSM-5. Depress
Anxiety 28(9):750–769.

137.

Forbes D, Elhai JD, Miller MW, Creamer M (2010) Internalizing and Externalizing Classes in
Posttraumatic Stress Disorder : A Latent Class Analysis. J Trauma Stress 23(3):340–349.

138.

Miller MW, Resick PA (2007) Internalizing and externalizing subtypes in female sexual assault
survivors: implications for the understanding of complex PTSD. Behav Ther 38(1):58–71.

139.

Castillo DT, et al. (2014) Externalizing and Internalizing Subtypes of Posttraumatic
Psychopathology and Anger Expression. J Trauma Stress 27:108–111.

140.

Rielage JK, Hoyt T, Renshaw K (2010) Internalizing and Externalizing Personality Styles and
Psychopathology in OEF – OIF Veterans. J Trauma Stress 23(3):350–357.

141.

Herman JL (1992) Complex PTSD: A syndrome in survivors of prolonged and repeted trauma. J
Trauma Stress 5(3):377–392.

142.

Resick PA, et al. (2012) A Critical Evaluation of the Complex PTSD Literature: Implications for
DSM-5. J Trauma Stress 25:241–251.

143.

Herman JL (2012) CPTSD is a Distinct Entity: Comment on Resick et al. (2012). J Trauma Stress
25:256–257.

144.

Shin LM, et al. (2009) Resting Metabolic Activity in the Cingulate Cortex and Vulnerability to
Posttraumatic Stress Disorder. Arch Gen Psychiatry 66(10):1099–1107.

145.

Yehuda R (1997) Sensitization of the hypothalamic-pituitary-adrenal axis in posttraumatic stress
disorder. Ann N Y Acad Sci 821(7 18):57–75.

146.

Rodrigues SM, Ledoux JE, Sapolsky RM (2009) The Influence of Stress Hormones on Fear
Circuitry. Annu Rev neurosience 32:289–313.

147.

Cahill L, Prins B, Weber M, McGaugh JL (1994) β-Adrenergic activation and memory for emotional
events. Nature 371(6499):702–704.

148.

O’Donnell T, Hegadoren KM, Coupland NC (2004) Noradrenergic mechanisms in the
pathophysiology of post-traumatic stress disorder. Neuropsychobiology 50(4):273–283.

149.

Southwick SM, et al. (1997) Noradrenergic and Serotonergic Function in Posttraumatic Stress
Disorder. Arch Gen Psychiatry 54:749–758.

150.

Keller-Wood ME, Dallman MF (1984) Corticosteroid inhibition of ACTH secretion. Endocr Rev
5(1):1–24.

151.

Jacobson L, Sapolsky RM (1991) The role of the hippocampus in feedback regulation of the
hypothalamic-pituitary-adrenocortical axis. Endocr Rev 12(2):118–34.

152.

Yehuda R, et al. (1993) Enhanced suppression of cortisol following dexamethasone administration
in posttraumatic stress disorder. Am J Psychiatry 150(1):83–6.

153.

Rohleder N, Joksimovic L, Wolf JM, Kirschbaum C (2004) Hypocortisolism and increased
glucocorticoid sensitivity of pro-Inflammatory cytokine production in Bosnian war refugees with
posttraumatic stress disorder. Biol Psychiatry 55(7):745–51.

103

154.

Yehuda R, Bierer LM (2008) Transgenerational transmission of cortisol and PTSD risk. Prog Brain
Res 167(7):121–35.

155.

Yehuda R, Halligan SL, Golier JA, Grossman R, Bierer LM (2004) Effects of trauma exposure on
the cortisol response to dexamethasone administration in PTSD and major depressive disorder.
Psychoneuroendocrinology 29(3):389–404.

156.

Grossman R, et al. (2003) Dexamethasone Suppression Test Findings in Subjects With
Personality Disorders : Associations With Posttraumatic Stress Disorder and Major Depression.
Am J Psychiatry 160(July):1291–1297.

157.

Yehuda R, Boisoneau D, Mason JW, Giller EL (1993) Glucocorticoid receptor number and cortisol
excretion in mood, anxiety, and psychotic disorders. Biol Psychiatry 34(1–2):18–25.

158.

Wahbeh H, Oken BS (2013) Salivary Cortisol Lower in Posttraumatic Stress Disorder. J Trauma
Stress 26:1–8.

159.

Pitman RK, et al. (2012) Biological studies of post-traumatic stress disorder. Nat Rev Neurosci
13(11):769–87.

160.

Yehuda R (2002) Post-Traumatic Stress Disorder. N Engl J Med 346(2):108–114.

161.

Pacak K, Palkovits M, Kopin IJ, Goldstein DS (1995) Stress-induced norepinephrine release in the
hypothalamic paraventricular nucleus and pituitary-adrenocortical and sympathoadrenal activity: in
vivo microdialysis studies. Front Neuroendocr 16(2):89–150.

162.

Schelling G, et al. (2004) Stress doses of hydrocortisone, traumatic memories, and symptoms of
posttraumatic stress disorder in patients after cardiac surgery: a randomized study. Biol Psychiatry
55(6):627–33.

163.

Zohar J, et al. (2011) High dose hydrocortisone immediately after trauma may alter the trajectory
of PTSD: interplay between clinical and animal studies. Eur Neuropsychopharmacol 21(11):796–
809.

164.

Shin LM, Rauch SL, Pitman RK (2006) Amygdala, medial prefrontal cortex, and hippocampal
function in PTSD. Ann N Y Acad Sci 1071:67–79.

165.

Hughes KC, Shin LM (2011) Functional neuroimaging studies of post-traumatic stress disorder.
Expert Rev Neurother 11(2):275–285.

166.

Weber M, et al. (2013) Voxel-Based Morphometric Gray Matter Correlates of Posttraumatic Stress
Disorder. J Anxiety Disord 27(4):413–419.

167.

Mahan A, Ressler KJ (2012) Fear conditioning, synaptic plasticity, and the amygdala: implications
for posttraumatic stress disorder. Trends Neurosci 35(1):24–35.

168.

Hayes JP, Hayes SM, Mikedis AM (2012) Quantitative meta-analysis of neural activity in
posttraumatic stress disorder. Biol Mood Anxiety Disord 2(1):9.

169.

Etkin A, Wager T (2007) Functional Neuroimaging of Anxiety: A Meta-Analysis of Emotional
Processing in PTSD, Social Anxiety Disorder, and Specific Phobia. Am J Psychiatry
164(10):1476–1488.

170.

Kasai K, et al. (2008) Evidence for acquired pregenual anterior cingulate gray matter loss from a
twin study of combat-related posttraumatic stress disorder. Biol Psychiatry 63(6):550–6.

104

171.

Rauch SL, Shin LM, Phelps EA (2006) Neurocircuitry Models of Posttraumatic Stress Disorder and
Extinction: Human Neuroimaging Research-Past, Present, and Future. Biol Psychiatry 60(4):376–
382.

172.

KĂźhn S, Gallinat J (2013) Gray matter correlates of posttraumatic stress disorder: A quantitative
meta-analysis. Biol Psychiatry 73(1):70–74.

173.

Gilboa A, et al. (2004) Functional connectivity of the prefrontal cortex and the amygdala in
posttraumatic stress disorder. Biol Psychiatry 55(3):263–72.

174.

Garfinkel SN, et al. (2014) Impaired Contextual Modulation of Memories in PTSD: An fMRI and
Psychophysiological Study of Extinction Retention and Fear Renewal. J Neurosci 34(40):13435–
13443.

175.

Christova P, James LM, Engdahl BE, Lewis SM, Georgopoulos AP (2015) Diagnosis of
posttraumatic stress disorder (PTSD) based on correlations of prewhitened fMRI data: outcomes
and areas involved. Exp Brain Res 233(9):2695–2705.

176.

Uddin M, Sipahi L, Li J, Koenen KC (2013) Sex differences in DNA methylation may contribute to
risk of PTSD and depression: A review of existing evidence. Depress Anxiety 30(12):1151–1160.

177.

Vukojevic V, et al. (2014) Epigenetic Modification of the Glucocorticoid Receptor Gene Is Linked to
Traumatic Memory and Post-Traumatic Stress Disorder Risk in Genocide Survivors. J Neurosci
34(31):10274–10284.

178.

Tolin DF, Foa EB (2006) Sex differences in trauma and posttraumatic stress disorder: a
quantitative review of 25 years of research. Psychol Bull 132(6):959–92.

179.

Zlotnick C, Zimmerman M, Wolfsdorf B a, Mattia JI (2001) Gender differences in patients with
posttraumatic stress disorder in a general psychiatric practice. Am J Psychiatry 158(11):1923–5.

180.

Yehuda R, et al. (2015) Post-traumatic stress disorder. Nat Rev Dis Prim 1(October):1–22.

181.

Stevens JS, Hamann S (2012) Sex differences in brain activation to emotional stimuli: a metaanalysis of neuroimaging studies. Neuropsychologia 50(7):1578–93.

182.

Woon F, Hedges DW (2011) Gender does not moderate hippocampal volume deficits in adults
with posttraumatic stress disorder: a meta-analysis. Hippocampus 21(3):243–52.

183.

Metzger LJ, et al. (2008) Basal and suppressed salivary cortisol in female vietnam nurse veterans
with and without PTSD. Psychiatry Interpers Biol Process 161(3):330–335.

184.

Rasmusson a M, et al. (2001) Increased pituitary and adrenal reactivity in premenopausal women
with posttraumatic stress disorder. Biol Psychiatry 50(12):965–77.

185.

Resnick HS, Yehuda R, Pitman RK, Foy DW (1995) Effect of previous trauma on acute plasma
cortisol level following rape. Am J Psychiatry 152(11):1675–1677.

186.

Medina AM, Mejia VY, Schell a M, Dawson ME, Margolin G (2001) Startle reactivity and PTSD
symptoms in a community sample of women. Psychiatry Res 101(2):157–69.

187.

Breslau N (2002) Gender Differences in Trauma and Posttraumatic Stress Disorder. J Gend
Specif Med 5(1):34–40.

105

188.

Kucharska J (2016) Sex Differences in the Appraisal of Traumatic Events and Psychopathology.
Psychol Trauma Theory , Res , Pract , Policy. doi:http://dx.doi.org/10.1037/tra0000244.

189.

Kamkwalala A, et al. (2012) Dark-enhanced startle responses and heart rate variability in a
traumatized civilian sample: putative sex-specific correlates of posttraumatic stress disorder.
Psychosom Med 74(2):153–9.

190.

Shvil E, et al. (2014) Sex Differences in Extinction Recall in Posttraumatic Stress Disorder: A Pilot
fMRI Study. Neurobiol Learn Mem 113:101–108.

191.

Inslicht SS, et al. (2013) Sex differences in fear conditioning in posttraumatic stress disorder. J
Psychiatr Res 47(1):64–71.

192.

Hawk LW, Dougall a L, Ursano RJ, Baum a (2000) Urinary catecholamines and cortisol in recentonset posttraumatic stress disorder after motor vehicle accidents. Psychosom Med 62(3):423–434.

193.

Shin LM, et al. (2004) Regional cerebral blood flow in the amygdala and medial prefrontal cortex
during traumatic imagery in male and female Vietnam veterans with PTSD. Arch Gen Psychiatry
61:168–176.

194.

Stevens JS, et al. (2013) Disrupted amygdala-prefrontal functional connectivity in civilian women
with posttraumatic stress disorder. J Psychiatr Res 47(10):1469–1478.

195.

Arnold AP (2009) The organizational-activational hypothesis as the foundation for a unified theory
of sexual differentiation of all mammalian tissues. Horm Behav 55(5):570–578.

196.

Seale J V, Wood S a, Atkinson HC, Lightman SL, Harbuz MS (2005) Organizational role for
testosterone and estrogen on adult hypothalamic-pituitary-adrenal axis activity in the male rat.
Endocrinology 146(4):1973–82.

197.

Mccarthy MM (2016) Multifaceted origins of sex differences in the brain Multifaceted origins of sex
differences in the brain. Philos Trans R Soc London B (FEBRUARY). doi:10.1098/rstb.2015.0106.

198.

Joel D, et al. (2015) Sex beyond the genitalia : The human brain mosaic. Proc Natl Acad Sci:1–6.

199.

Forger NG (2016) Epigenetic mechanisms in sexual differentiation of the brain and behaviour.
Philos Trans R Soc London B Biol Sci 371(1688). doi:10.1098/rstb.2015.0114.

200.

Arnold AP (2012) The end of gonad-centric sex determination in mammals. Trends Genet
28(2):55–61.

201.

Sterrenburg L, et al. (2012) Sex-dependent and differential responses to acute restraint stress of
corticotropin-releasing factor-producing neurons in the rat paraventricular nucleus, central
amygdala, and bed nucleus of the stria terminalis. J Neurosci Res 90(1):179–92.

202.

Goel N, Bale TL (2010) Sex differences in the serotonergic influence on the hypothalamicpituitary-adrenal stress axis. Endocrinology 151(4):1784–94.

203.

Young EA, Altemus M, Parkison V, Shastry S (2001) Effects of estrogen antagonists and agonists
on the ACTH response to restraint stress in female rats. Neuropsychopharmacology 25(6):881–
91.

204.

Young EA (1996) Sex differences in response to exogenous corticosterone: a rat model
hypercortisolemia. Mol Psychiatry 1:313–319.

106

205.

Williamson M, Bingham B, Gray M, Innala L, Viau V (2010) The medial preoptic nucleus integrates
the central influences of testosterone on the paraventricular nucleus of the hypothalamus and its
extended circuitries. J Neurosci 30(35):11762–70.

206.

Lund TD, Munson DJ, Haldy ME, Handa RJ (2004) Androgen inhibits, while oestrogen enhances,
restraint-induced activation of neuropeptide neurones in the paraventricular nucleus of the
hypothalamus. J Neuroendocrinol 16(3):272–8.

207.

Handa RJ, Kudwa AE, Donner NC, McGivern RF, Brown R (2013) Central 5-alpha reduction of
testosterone is required for testosterone’s inhibition of the hypothalamo-pituitary-adrenal axis
response to restraint stress in adult male rats. Brain Res 1529:74–82.

208.

Larkin JW, Binks SL, Li Y, Selvage D (2010) The role of oestradiol in sexually dimorphic
hypothalamic-pituitary-adrena axis responses to intracerebroventricular ethanol administration in
the rat. J Neuroendocrinol 22(1):24–32.

209.

Handa RJ, Weiser MJ (2013) Gonadal steroid hormones and the hypothalamo-pituitary-adrenal
axis. Front Neuroendocrinol. doi:10.1016/j.yfrne.2013.11.001.

210.

Burgess L, Handa RJ (1992) Chronic Estrogen-Induced Alterations in Adrenocorticotropin and
Corticosterone Secretion, and Glucocorticoid Receptor-Mediated Functions in Female Rats.
Endocrinology 131(3):1261–1269.

211.

Handa RJ, Burgess L, Kerr J, O’Keefe J (1994) Gonadal Steroid Hormone Receptors and Sex
Differences in the Hypothalamic-Pituitary-Adrenal Axis. Horm Behav 28:464–476.

212.

Handa RJ, et al. (1994) Androgen regulation of adrenocorticotropin and corticosterone secretion in
the male rat following novelty and foot shock stressors. Physiol Behav 55(1):117–24.

213.

Dart DA, Waxman J, Aboagye EO, Bevan CL (2013) Visualising androgen receptor activity in male
and female mice. PLoS One 8(8):e71694.

214.

FernĂĄndez-Guasti A, Kruijver FP, Fodor M, Swaab DF (2000) Sex differences in the distribution of
androgen receptors in the human hypothalamus. J Comp Neurol 425(3):422–35.

215.

Seale J V, Wood S a, Atkinson HC, Harbuz MS, Lightman SL (2004) Gonadal steroid replacement
reverses gonadectomy-induced changes in the corticosterone pulse profile and stress-induced
hypothalamic-pituitary-adrenal axis activity of male and female rats. J Neuroendocrinol
16(12):989–98.

216.

Seale J V, et al. (2004) Gonadectomy reverses the sexually diergic patterns of circadian and
stress-induced hypothalamic-pituitary-adrenal axis activity in male and female rats. J
Neuroendocrinol 16(6):516–24.

217.

Turner BB, Weaver DA (1985) Sexual dimorphism of glucocorticoid binding in rat brain. Brain Res
343:16–23.

218.

Spinedi E, Suescun MO, Hadid R, Gaillard RC (1992) Effects of Gonadectomy and Sex Hormone
Therapy on the Endotoxin-Stimulated Hypothalamo-Pituitary-Adrenal Axis: Evidence for a
Neuroendocrine-Immunological Sexual Dimorphism EDUARDO. Endocrinology 131(5):2430–36.

219.

Bale TL, Epperson CN (2016) Sex As a Biological Variable: Who, What, When, Why and How.
Neuropsychopharmacology:1–11.

107

220.

Bland ST, et al. (2005) Expression of c-fos and BDNF mRNA in subregions of the prefrontal cortex
of male and female rats after acute uncontrollable stress. Brain Res 1051(1–2):90–9.

221.

Mazor A, et al. (2009) Gender-related qualitative differences in baseline and post-stress anxiety
responses are not reflected in the incidence of criterion-based PTSD-like behaviour patterns.
World J Biol Psychiatry 10(4):856–869.

222.

Babb JA, Masini CV, Day HEW, Campeau S (2013) Sex differences in activated corticotropinreleasing factor neurons within stress-related neurocircuitry and hypothalamic-pituitaryadrenocortical axis hormones following restraint in rats. Neuroscience 234:40–52.

223.

Mirshekar M, Abrari K, Goudarzi I, Rashidy-Pour A (2013) Systemic administrations of β-estradiol
alleviate both conditioned and sensitized fear responses in an ovariectomized rat model of posttraumatic stress disorder. Neurobiol Learn Mem 102:12–9.

224.

McDermott CM, Liu D, Schrader L a (2012) Role of gonadal hormones in anxiety and fear memory
formation and inhibition in male mice. Physiol Behav 105(5):1168–74.

225.

Garrett JE, Wellman CL (2009) Chronic stress effects on dendritic morphology in medial prefrontal
cortex: sex differences and estrogen dependence. Neuroscience 162(1):195–207.

226.

McCarthy MM, Konkle ATM (2005) When is a sex difference not a sex difference? Front
Neuroendocrinol 26(2):85–102.

227.

McCarthy MM, Arnold AP, Ball GF, Blaustein JD, De Vries GJ (2012) Sex differences in the brain:
the not so inconvenient truth. J Neurosci 32(7):2241–7.

228.

Bale TL, Epperson CN (2015) Sex differences and stress across the lifespan. Nat Neurosci
18(10):1413–20.

229.

Joel D, McCarthy MM (2016) Incorporating Sex As a Biological Variable in Neuropsychiatric
Research: Where Are We Now and Where Should We Be? Neuropsychopharmacology (May):1–
23.

230.

Becker JB, et al. (2005) Strategies and methods for research on sex differences in brain and
behavior. Endocrinology 146(4):1650–1673.

231.

Zuloaga DG, Puts D a, Jordan CL, Breedlove SM (2008) The role of androgen receptors in the
masculinization of brain and behavior: what we’ve learned from the testicular feminization
mutation. Horm Behav 53(5):613–26.

232.

Yehuda R, Antelman SM (1993) Criteria for rationally evaluating animal models of posttraumatic
stress disorder. Biol Psychiatry 33(7):479–86.

233.

Siegmund A, Wotjak CT (2006) Toward an animal model of posttraumatic stress disorder. Ann N Y
Acad Sci 1071:324–34.

234.

Yamamoto S, et al. (2009) Single prolonged stress: toward an animal model of posttraumatic
stress disorder. Depress Anxiety 26(12):1110–7.

235.

Serova LI, et al. (2013) Single intranasal neuropeptide Y infusion attenuates development of
PTSD-like symptoms to traumatic stress in rats. Neuroscience 236:298–312.

108

236.

Brand L, Groenewald I, Stein DJ, Wegener G, Harvey BH (2008) Stress and re-stress increases
conditioned taste aversion learning in rats: possible frontal cortical and hippocampal muscarinic
receptor involvement. Eur J Pharmacol 586(1–3):205–11.

237.

Wang W, et al. (2008) A modified single-prolonged stress model for post-traumatic stress disorder.
Neurosci Lett 441(2):237–41.

238.

Imanaka A, Morinobu S, Toki S, Yamawaki S (2006) Importance of early environment in the
development of post-traumatic stress disorder-like behaviors. Behav Brain Res 173(1):129–37.

239.

Kohda K, et al. (2007) Glucocorticoid receptor activation is involved in producing abnormal
phenotypes of single-prolonged stress rats: a putative post-traumatic stress disorder model.
Neuroscience 148(1):22–33.

240.

Khan SA, Liberzon I (2004) Topiramate attenuates exaggerated acoustic startle in an animal
model of PTSD. Psychopharmacology (Berl) 172(2):225–9.

241.

Liberzon I, Krstov M, Young EA (1997) Stress-restress: effects on ACTH and fast feedback.
Psychoneuroendocrinology 22(6):443–53.

242.

Liberzon I, LĂłpez JF, Flagel SB, VĂĄzquez DM, Young EA (1999) Differential regulation of
hippocampal glucocorticoid receptors mRNA and fast feedback: relevance to post-traumatic stress
disorder. J Neuroendocrinol 11(1):11–7.

243.

Knox D, et al. (2012) Single prolonged stress disrupts retention of extinguished fear in rats. Learn
Mem 19(2):43–9.

244.

Schneier FR, et al. (2012) Combined Prolonged Exposure Therapy and Paroxetine for
Posttraumatic Stress Disorder Related to the World Trade Center Attacks: A Randomized
Controlled Trial. Am J Psychiatry 169(1):80–88.

245.

Corchs F, Nutt DJ, Hood S, Bernik M (2009) Serotonin and sensitivity to trauma-related exposure
in selective serotonin reuptake inhibitors-recovered posttraumatic stress disorder. Biol Psychiatry
66(1):17–24.

246.

Wang J-W, David DJ, Monckton JE, Battaglia F, Hen R (2008) Chronic fluoxetine stimulates
maturation and synaptic plasticity of adult-born hippocampal granule cells. J Neurosci 28(6):1374–
84.

247.

Takahashi T, Morinobu S, Iwamoto Y, Yamawaki S (2006) Effect of paroxetine on enhanced
contextual fear induced by single prolonged stress in rats. Psychopharmacology (Berl)
189(2):165–73.

248.

Peng Z, et al. (2013) Ziprasidone ameliorates anxiety-like behaviors in a rat model of PTSD and
up-regulates neurogenesis in the hippocampus and hippocampus-derived neural stem cells.
Behav Brain Res 244C:1–8.

249.

Adamec R, Holmes A, Blundell J (2008) Vulnerability to Lasting Anxiogenic Effects of Brief
Exposure to Predator Stimuli: Sex, Serotonin and Other Factors - Relevance to PTSD. Neurosci
Biobehav Rev 32(7):1287–1292.

250.

Campos AC, Ferreira FR, da Silva WA, GuimarĂŁes FS (2013) Predator threat stress promotes
long lasting anxiety-like behaviors and modulates synaptophysin and CB1 receptors expression in
brain areas associated with PTSD symptoms. Neurosci Lett 533:34–8.

109

251.

Cohen H, et al. (2006) Blunted HPA axis response to stress influences susceptibility to
posttraumatic stress response in rats. Biol Psychiatry 59(12):1208–18.

252.

Cohen H, et al. (2004) Setting apart the affected: the use of behavioral criteria in animal models of
post traumatic stress disorder. Neuropsychopharmacology 29(11):1962–70.

253.

Cohen H, Zohar J, Matar MA (2003) The Relevance of Differential Response to Trauma in an
Animal Model of Posttraumatic Stress Disorder. Biol Psychiatry 53:463–473.

254.

Mackenzie L, Nalivaiko E, Beig MI, Day T a, Walker FR (2010) Ability of predator odour exposure
to elicit conditioned versus sensitised post traumatic stress disorder-like behaviours, and forebrain
deltaFosB expression, in rats. Neuroscience 169(2):733–42.

255.

Adamec R, Muir C, Grimes M, Pearcey K (2007) Involvement of noradrenergic and corticoid
receptors in the consolidation of the lasting anxiogenic effects of predator stress. Behav Brain Res
179(2):192–207.

256.

Matar MA, Cohen H, Kaplan Z, Zohar J (2006) The effect of early poststressor intervention with
sertraline on behavioral responses in an animal model of post-traumatic stress disorder.
Neuropsychopharmacology 31(12):2610–8.

257.

Fountoulakis K, et al. (2004) Relationship among Dexamethasone Suppression Test, personality
disorders and stressful life events in clinical subtypes of major depression: An exploratory study.
Ann Gen Hosp Psychiatry 3(1):15.

258.

Cole M a, Kim PJ, Kalman B a, Spencer RL (2000) Dexamethasone suppression of corticosteroid
secretion: evaluation of the site of action by receptor measures and functional studies.
Psychoneuroendocrinology 25(2):151–67.

259.

Masi G, Brovedani P (2011) The hippocampus, neurotrophic factors and depression: possible
implications for the pharmacotherapy of depression. CNS Drugs 25(11):913–31.

260.

Watson S, Gallagher P, Smith MS, Ferrier IN, Young AH (2006) The dex/CRH test--is it better than
the DST? Psychoneuroendocrinology 31(7):889–94.

261.

Jokinen J, NordstrĂśm P (2009) HPA axis hyperactivity and attempted suicide in young adult mood
disorder inpatients. J Affect Disord 116(1–2):117–20.

262.

Jokinen J, Mårtensson B, Nordström A-L, Nordström P (2008) CSF 5-HIAA and DST nonsuppression -independent biomarkers in suicide attempters? J Affect Disord 105(1–3):241–5.

263.

Menke A, et al. (2012) Dexamethasone stimulated gene expression in peripheral blood is a
sensitive marker for glucocorticoid receptor resistance in depressed patients.
Neuropsychopharmacology 37(6):1455–64.

264.

Young EA, et al. (1993) Dissociation Between Pituitary and Adrenal Suppression to
Dexamethasone in Depression. Arch Gen Psychiatry 50:395–403.

265.

Grillon C, Morgan C a, Southwick SM, Davis M, Charney DS (1996) Baseline startle amplitude and
prepulse inhibition in Vietnam veterans with posttraumatic stress disorder. Psychiatry Res
64(3):169–78.

266.

Shalev a Y, Orr SP, Peri T, Schreiber S, Pitman RK (1992) Physiologic responses to loud tones in
Israeli patients with posttraumatic stress disorder. Arch Gen Psychiatry 49(11):870–5.

110

267.

Charney DS, Deutch AY, Krystal JH, Southwick SM, Davis M (1993) Psychobiologic Mechanisms
of Posttraumatic Stress Disorder. Arch Gen Psychiatry 50:294–305.

268.

Pitman RK (1997) Overview of biological themes in PTSD. Ann N Y Acad Sci 821(603):1–9.

269.

Young EA, Korszun A (2010) Sex, trauma, stress hormones and depression. Mol Psychiatry
15(1):23–8.

270.

Altemus M, et al. (1997) Reduced Sensitivity to Glucocorticoid Feedback and Reduced
Glucocorticoid Receptor mRNA Expression in the Luteal Phase of the Menstrual Cycle.
Neuropsychopharmacology 17:100–109.

271.

Kudielka BM, Kirschbaum C (2005) Sex differences in HPA axis responses to stress: a review.
Biol Psychol 69(1):113–32.

272.

Glover EM, et al. (2012) Estrogen levels are associated with extinction deficits in women with
posttraumatic stress disorder. Biol Psychiatry 72(1):19–24.

273.

Glover EM, et al. (2013) Inhibition of fear is differentially associated with cycling estrogen levels in
women. J psychiatry Neurosci 38(5):341–348.

274.

Goldstein JM, et al. (2005) Hormonal cycle modulates arousal circuitry in women using functional
magnetic resonance imaging. J Neurosci 25(40):9309–16.

275.

Zeidan MA, et al. (2011) Estradiol modulates medial prefrontal cortex and amygdala activity during
fear extinction in women and female rats. Biol Psychiatry 70(10):920–7.

276.

Manuck SB, et al. (2010) Salivary testosterone and a trinucleotide (CAG) length polymorphism in
the androgen receptor gene predict amygdala reactivity in men. Psychoneuroendocrinology
35(1):94–104.

277.

Hermans EJ, Putman P, Baas JM, Koppeschaar HP, van Honk J (2006) A Single Administration of
Testosterone Reduces Fear-Potentiated Startle in Humans. Biol Psychiatry 59(9):872–874.

278.

Hermans EJ, et al. (2007) Exogenous testosterone attenuates the integrated central stress
response in healthy young women. Psychoneuroendocrinology 32(8–10):1052–1061.

279.

King MW, Street A, Gradus JL, Vogt DS, Resick PA (2013) Gender Differences in Posttraumatic
Stress Symptoms Among OEF/OIF Veterans: An Item Response Theory Analysis. J Trauma
Stress 26:175–183.

280.

Cohen H, Kozlovsky N, Alona C, Matar MA, Joseph Z (2012) Animal model for PTSD: from clinical
concept to translational research. Neuropharmacology 62(2):715–24.

281.

Farrell MR, Sengelaub DR, Wellman CL (2013) Sex differences and chronic stress effects on the
neural circuitry underlying fear conditioning and extinction. Physiol Behav 122:208–215.

282.

Leuner B, Shors TJ (2013) Stress, anxiety, and dendritic spines: What are the connections?
Neuroscience 251:108–119.

283.

Gruene TM, Roberts E, Thomas V, Ronzio A, Shansky RM (2015) Sex-specific neuroanatomical
correlates of fear expression in prefrontal-amygdala circuits. Biol Psychiatry 78(3):186–193.

284.

Zer-Aviv TM, Akirav I (2016) Sex differences in hippocampal response to endocannabinoids after
exposure to severe stress. Hippocampus 26(7):947–957.

111

285.

He Y-Q, et al. (2013) PKCÎł Receptor Mediates Visceral Nociception and Hyperalgesia following
Exposure to PTSD-Like Stress in the Spinal Cord of Rats. Mol Pain 9(1):1744-8069-9–35.

286.

Yu H, Watt H, Kesavan C, Mohan S (2013) The negative impact of single prolonged stress (SPS)
on bone development in mice. Stress 16(5):564–570.

287.

Arai A, et al. (2016) A single prolonged stress paradigm produces enduring impairments in social
bonding in monogamous prairie voles. Behav Brain Res 315:83–93.

288.

Keller SM, Schreiber WB, Staib JM, Knox D (2015) Sex differences in the single prolonged stress
model. Behav Brain Res 286:29–32.

289.

Shansky RM (2015) Sex differences in PTSD resilience and susceptibility: Challenges for animal
models of fear learning. Neurobiol Stress 1(1):60–65.

290.

Zaba M, et al. (2015) Identification and characterization of HPA-axis reactivity endophenotypes in
a cohort of female PTSD patients. Psychoneuroendocrinology 55:102–115.

291.

Jia M, et al. (2014) Corticosterone mitigates the stress response in an animal model of PTSD. J
Psychiatr Res:1–11.

292.

Zuloaga DG, Poort JE, Jordan CL, Breedlove SM (2011) Male rats with the testicular feminization
mutation of the androgen receptor display elevated anxiety-related behavior and corticosterone
response to mild stress. Horm Behav 60(4):380–8.

293.

Strekalova T, Spanagel R, Bartsch D, Henn FA, Gass P (2004) Stress-induced anhedonia in mice
is associated with deficits in forced swimming and exploration. Neuropsychopharmacology
29(11):2007–2017.

294.

Walsh JJ, et al. (2013) Stress and CRF gate neural activation of BDNF in the mesolimbic reward
pathway. Nat Neurosci 17(1):27–29.

295.

Louvart H, Maccari S, Ducrocq F, Thomas P, DarnaudĂŠry M (2005) Long-term behavioural
alterations in female rats after a single intense footshock followed by situational reminders.
Psychoneuroendocrinology 30(4):316–24.

296.

Rex A, Voigt JP, Gustedt C, Beckett S, Fink H (2004) Anxiolytic-like profile in Wistar, but not
Sprague-Dawley rats in the social interaction test. Psychopharmacology (Berl) 177(1–2):23–34.

297.

Vialou V, et al. (2010) DeltaFosB in brain reward circuits mediates resilience to stress and
antidepressant responses. Nat Neurosci 13(6):745–752.

298.

Krishnan V, et al. (2007) Molecular adaptations underlying susceptibility and resistance to social
defeat in brain reward regions. Cell 131(2):391–404.

299.

Dong X, Li Y (2014) Peritraumatic startle response predicts the vulnerability to develop PTSD-like
behaviors in rats: a model for peritraumatic dissociation. Front Behav Neurosci 8(January):14.

300.

Eagle AL, et al. (2013) Single prolonged stress enhances hippocampal glucocorticoid receptor and
phosphorylated protein kinase B levels. Neurosci Res 75(2):130–7.

301.

Myers B, McKlveen JM, Herman JP (2014) Glucocorticoid actions on synapses, circuits, and
behavior: Implications for the energetics of stress. Front Neuroendocrinol 35(2):180–196.

112

302.

Figueiredo HF, Dolgas CM, Herman JP (2002) Stress activation of cortex and hippocampus is
modulated by sex and stage of estrus. Endocrinology 143(7):2534–40.

303.

Toufexis D, Davis C, Hammond A, Davis M (2005) Sex differences in hormonal modulation of
anxiety measured with light-enhanced startle: possible role for arginine vasopressin in the male. J
Neurosci 25(39):9010–6.

304.

Pisu MG, et al. (2016) Sex differences in the outcome of juvenile social isolation on HPA axis
function in rats. Neuroscience 320:172–182.

305.

Vialou V, et al. (2014) Prefrontal Cortical Circuit for Depression- and Anxiety-Related Behaviors
Mediated by Cholecystokinin: Role of ΔFosB. J Neurosci 34(11):3878–87.

306.

Nestler EJ, et al. (2002) Neurobiology of Depression. Neuron 34:13–25.

307.

Baker KB, Kim JJ (2004) Amygdalar lateralization in fear conditioning: evidence for greater
involvement of the right amygdala. Behav Neurosci 118(1):15–23.

308.

Georgopoulos AP, et al. (2010) The synchronous neural interactions test as a functional
neuromarker for post-traumatic stress disorder (PTSD): a robust classification method based on
the bootstrap. J Neural Eng 7(1):16011.

309.

Ritov G, Boltyansky B, Richter-Levin G (2016) A novel approach to PTSD modeling in rats reveals
alternating patterns of limbic activity in different types of stress reaction. Mol Psychiatry 21:630–
641.

310.

Kramer MD, Krueger RF, Hicks BM (2008) The role of internalizing and externalizing liability
factors in accounting for gender differences in the prevalence of common psychopathological
syndromes. Psychol Med 38(1):51–61.

311.

Zavala JK, Fernandez AA, Gosselink KL (2011) Female responses to acute and repeated restraint
stress differ from those in males. Physiol Behav 104(2):215–21.

312.

Shors TJ, Beylin A V., Wood GE, Gould E (2000) The modulation of Pavlovian memory. Behav
Brain Res 110(1–2):39–52.

313.

Wood GE, Shors TJ (1998) Stress facilitates classical conditioning in males, but impairs classical
conditioning in females through activational effects of ovarian hormones. Proc Natl Acad Sci
95(7):4066–4071.

314.

Kessler RC (2003) Epidemiology of women and depression. J Affect Disord 74:5–13.

315.

Horesh D, Lowe SR, Galea S, Uddin M, Koenen KC (2014) Gender Differences in the Long-Term
Associations Between Posttraumatic Stress Disorder and Depression Symptoms: Findings From
the Detroit Neighborhood Health Study. Depress Anxiety 11:1–11.

316.

van Hasselt FN, et al. (2012) Within-litter variation in maternal care received by individual pups
correlates with adolescent social play behavior in male rats. Physiol Behav 106(5):701–706.

317.

Silver RC, Alison Holman E, McIntosh DN, Poulin M, Gil-Rivas V (2002) Nationwide longitudinal
study of psychological responses to September 11. J Am Med Assoc 288(10):1235–1244.

318.

McFarlane AC (1989) The Aetiology of Post-traumatic Precipitating Morbidity : Factors. Br J
Psychiatry 154:221–228.

113

319.

Maeng LY, Milad MR (2015) Sex differences in anxiety disorders: Interactions between fear,
stress, and gonadal hormones. Horm Behav 76:106–117.

320.

Shors TJ (2016) A trip down memory lane about sex differences in the brain. Philos Trans R Soc B
Biol Sci 371(1688):20150124.

321.

Wood GE, Beylin A V, Shors TJ (2001) The contribution of adrenal and reproductive hormones to
the opposing effects of stress on trace conditioning in males versus females. Behav Neurosci
115(1):175–187.

322.

Koresh O, et al. (2016) Distinctive cardiac autonomic dysfunction following stress exposure in both
sexes in an animal model of PTSD. Behav Brain Res 308:128–142.

323.

Burke HM, et al. (2013) Sex-specific impairment of spatial memory in rats following a reminder of
predator stress. Stress 16(4):469–476.

324.

Cozzoli D, Tanchuck-Nipper M, Kaufman M, Horowitz C, Finn D (2015) Environmental stressors
influence limited-access ethanol consumption by C57BL/6J mice in a sex-dependent manner.
Alcohol 48(8):741–754.

325.

Adamec R, Head D, Blundell J, Burton P, Berton O (2006) Lasting anxiogenic effects of feline
predator stress in mice: sex differences in vulnerability to stress and predicting severity of
anxiogenic response from the stress experience. Physiol Behav 88(1–2):12–29.

326.

Daskalakis NP, Cohen H, Cai G, Buxbaum JD, Yehuda R (2014) Expression profiling associates
blood and brain glucocorticoid receptor signaling with trauma-related individual differences in both
sexes. Proc Natl Acad Sci 111(37):1–6.

327.

Toth M, et al. (2015) Overexpression of Forebrain CRH During Early Life Increases Trauma
Susceptibility in Adulthood. Neuropsychopharmacology 41(6):1–10.

328.

Knox D, Nault T, Henderson C, Liberzon I (2012) Glucocorticoid receptors and extinction retention
deficits in the single prolonged stress model. Neuroscience 223:163–73.

329.

Farrell MR, Gruene TM, Shansky RM (2015) The influence of stress and gonadal hormones on
neuronal structure and function. Horm Behav 76:118–124.

330.

Goel N, Bale TL (2008) Organizational and activational effects of testosterone on masculinization
of female physiological and behavioral stress responses. Endocrinology 149(12):6399–405.

331.

Serova LI, Harris HA, Maharjan S, Sabban EL (2010) Modulation of responses to stress by
estradiol benzoate and selective estrogen receptor agonists. J Endocrinol 205(3):253–262.

332.

Shors TJ, Leuner B (2003) Estrogen-mediated effects on depression and memory formation in
females. J Affect Disord 74(1):85–96.

333.

Shansky RM, et al. (2010) Estrogen promotes stress sensitivity in a prefrontal cortex-amygdala
pathway. Cereb cortex 20(11):2560–7.

334.

Cavigelli SA, McClintock MK (2003) Fear of novelty in infant rats predicts adult corticosterone
dynamics and an early death. Proc Natl Acad Sci U S A 100(26):16131–6.

335.

Mcteague LM, et al. (2010) Aversive Imagery in Posttraumatic Stress Disorder : Trauma
Recurrence , Comorbidity , and Physiological. Biol Psychiatry 67(4):346–356.

114

336.

Fenchel D, et al. (2015) Beyond the HPA-axis: the role of the gonadal steroid hormone receptors
in modulating stress related responses in an animal model of PTSD. Eur J Psychotraumatol
3(0):1–14.

337.

Farrell MR, Holland FH, Shansky RM, Brenhouse HC (2016) Sex-specific effects of early life
stress on social interaction and prefrontal cortex dendritic morphology in young rats. Behav Brain
Res 310:119–125.

338.

Holland FH, Ganguly P, Potter DN, Chartoff EH, Brenhouse HC (2014) Early life stress disrupts
social behavior and prefrontal cortex parvalbumin interneurons at an earlier time-point in females
than in males. Neurosci Lett 566:131–136.

339.

Archer J (1975) Rodent Sex Differences in Emotional and Related Behavior I Ethology and
Neurophysiology Group , School of Biological Sciences , University of Sussex ,. Neurophysiology
479(4298):451–479.

340.

Brewin CR, Holmes E a (2003) Psychological theories of posttraumatic stress disorder. Clin
Psychol Rev 23(3):339–376.

341.

Korte SM, Koolhaas JM, Wingfield JC, McEwen BS (2005) The Darwinian concept of stress:
benefits of allostasis and costs of allostatic load and the trade-offs in health and disease. Neurosci
Biobehav Rev 29(1):3–38.

342.

Shors TJ, et al. (2007) Neurogenesis and helplessness are mediated by controllability in males but
not in females. Biol Psychiatry 62(5):487–95.

343.

Nasca C, Bigio B, Zelli D, Nicoletti F, McEwen BS (2015) Mind the gap: glucocorticoids modulate
hippocampal glutamate tone underlying individual differences in stress susceptibility. Mol
Psychiatry 20(6):755–763.

344.

Southwick SM, Bonanno GA, Masten AS, Panter-brick C, Yehuda R (2014) Resilience definitions,
theory, and challenges: interdisciplinary perspectives. Eur J Psychotraumatol 1:1–14.

345.

Newport DJ, Heim C, Bonsall R, Miller AH, Nemeroff CB (2004) Pituitary-adrenal responses to
standard and low-dose dexamethasone suppression tests in adult survivors of child abuse. Biol
Psychiatry 55(1):10–20.

346.

Stein MB, Yehuda R, Koverola C, Hanna C (1997) Enhanced dexmethasone suppression of
plasma cortisol in adult women traumatized by childhood sexual abuse. Biol Psychiatry
42(96):680–686.

347.

Delahanty DL, Nugent NR, Christopher NC, Walsh M (2005) Initial urinary epinephrine and cortisol
levels predict acute PTSD symptoms in child trauma victims. Psychoneuroendocrinology
30(2):121–128.

115