Post-Traumatic Stress Disorder (PTSD) is a complex, multi-faceted disease that affects a subset of individuals who undergo a traumatic experience. To understand the variety of psychological symptoms and their underpinning neural mechanisms, our laboratory has developed a rodent model of the disease in which animals experience an acute traumatic experience (15 unsignaled footshocks) that leads to a variety of changes in fear, anxiety, and depression (PTSD-like phenotypes). Through using a behavioral model of the disease, we can use pharmacological, behavioral, or genetic manipulations to dissect the neural mechanisms underlying the variety of changes in behavior seen following the acute stressor. In this dissertation, I seek to understand how these different symptoms (exaggerated subsequent fear conditioning, anxiety, and depression) manifest selectively and are driven by distinct neural mechanisms, each of which are sensitive to the effects of stress. As we begin to disentangle the variety of changes that occur in the brain following stress and how these changes in the brain manifest behaviorally, we can ultimately hope to better diagnose and treat individuals who are suffering from the debilitating symptoms associated with PTSD.

In Chapter 2, I test for the effects of pair housing animals prior to, during, and following the acute stressor. Rodents are extremely social animals, and isolation housing has been shown to magnify or sufficiently cause effects of stress. Therefore, it was critical to understand whether isolation housing is necessary for developing the PTSD-like symptoms observed after an acute stressor. Interestingly, I found that housing condition (isolation versus pair) had no effect on subsequent fear and anxiety phenotypes in adult, male rats undergoing an acute stressor.

In Chapter 3, I test for the role of kappa opioid receptors (KORs) in the subsequent expression of PTSD-like phenotypes following an acute stressor. The KOR antagonist JDTic was administered immediately after trauma, then animals were tested for exaggerated fear conditioning, and anxiety and depression assays. JDTic administration did not perturb the enhanced fear conditioning phenotype observed following stress, but it did mitigate anxiety behavior on the elevated plus maze (EPM). JDTic administered to stressed animals caused an increase in time spent in the open arms of the EPM across an 8 minute session. Shockingly, JDTic administered to unstressed animals caused an anxiogenic phenotype as seen by a failure to habituate to the EPM across an 8 minute session. Animals were also tested in the open field test and forced swim test, but there was no effect of JDTic on these measures. Together, these data indicate that KORs have a selective role in the anxiety-like phenotypes seen in rats following an acute stressor. These data are one example of how different neural circuits could contribute differentially to the array of phenotypes observed following acute stress.

In Chapter 4, I use a variety of techniques to assess the neural mechanisms underlying the enhanced fear phenotype observed in our PTSD model. Specifically, I test for changes in excitatory receptors expressed in brain regions associated with fear and anxiety through Western blotting, RT-PCR, and a genetic manipulation to test the contribution of NMDA-R’s to the sensitized fear response. Taken together, these studies begin to describe the neural and molecular changes that lead to the robust enhancement in subsequent fear conditioning observed in stressed animals.

Collectively, these studies serve as a beginning of exploring the many behavioral and neural mechanisms underlying our rodent model of PTSD. Future and subsequent studies may build off of this work to further characterize the PTSD-like phenotypes of our model, and relate these behavioral changes to the specific neural changes that guide them.

Fear is an adaptive response that allows animals to respond appropriately to danger, and Pavlovian fear learning provides a mechanism that allows animals to learn about, predict, and effectively respond to dangerous situations. In the laboratory, fear conditioning is a frequently used procedure used to study the learning, expression, and underlying neurobiology of fear and emotional behavior. In such an experiment, animals learn to associate some cue/conditional stimulus (CS) or contextual location with an aversive outcome/unconditional stimulus (US), such as a footshock. As a result of this learned association, the animal will now respond to the initially neutral CS with defensive behavior, a conditional response (CR) as a result of learning. This dissertation contains two chapters which investigate different aspects of cued and contextual fear learning in male and female rodents, first with a focus on the selection and associative nature of defensive responding, then with a focus on potential sex differences and any potential underlying mechanisms.Chapter #1 of this dissertation presents behavioral experiments in mice that assess defensive behavior after various auditory fear conditioning procedures. Particularly, we employed a serial compound conditional stimulus (SCS), consisting of a tone followed by a white noise, training procedure which allowed us to asses differential freezing and activity burst behaviors to each component of the SCS. A number of experiments were run with various control groups to assess the associative or non-associative nature of defensive responding following such a procedure. Contrary to some conclusions in the recent literature, while associative processes had some impact on the topography of flight behavior, cue-elicited activity bursts were primarily due to nonassociative learning processes. Such activity burst behavior was viewed as being similar to a fear-potentiated startle response. The results of the experiments led to the production of a rule for defensive behavior selection: “When afraid, freeze until there is a sudden novel change in stimulation, then burst into vigorous flight attempts,” (Trott et al., 2022, p. 2). Chapter #2 of this dissertation presents behavioral contextual fear learning experiments in rats and computational simulations of the same experiments using BACON, a conceptual and computational model of hippocampal function. Behaviorally, animals underwent a 3-day contextual fear conditioning procedure in which they received pre-exposure to a to-be-shocked context on Day 1, a shock after some placement-to-shock interval (PSI) on Day 2, and a fear test on Day 3. Experiments 1A and 1B results revealed that after shorter pre-exposure periods, male rats showed higher levels of contextual fear; however, after longer pre-exposure periods, female rats showed greater fear. In Experiment 2, when testing recently vs remotely acquired fear memories, males again showed more contextual fear. Additionally, while pre-exposure and PSI are both periods of time prior to shock, Experiment 3 revealed that they were not equivalent to each other such that there were overall sex differences and differences in the timing of fear expression based on whether animals received the majority of pre-shock context exposure as pre-exposure or PSI. Through the use of BACON, a computational model of hippocampal function, we simulated many of our key findings and suggest a potential mechanism for such findings, that of an increased contextual feature sampling rate in males. The experiments presented here asses both cued auditory and contextual fear learning in both rats and mice and add to a rich literature on fear conditioning and defensive behavior. The experiments and discussion within Chapter #1 introduce important controls and considerations for the conduction and interpretation of fear conditioning experiments, particularly for a set of increasingly used fear conditioning procedures which result in different freezing and flight behavior to different cues paired with a shock US. Results from Chapter #2 provide insight into sex differences in contextual learning and suggest a hippocampal mechanism for such differences, adding to a growing literature on sex differences in learning and hippocampal function.

Fear is critical for survival but can become harmful when it is disproportionate to the level of threat present. Post-traumatic stress disorder (PTSD) is characterized by such dysregulation of fear and is frequently comorbid with substance use disorders (SUDs), particularly alcohol abuse. However, the factors that mediate susceptibility to these conditions and the biological mechanisms that support these conditions are unclear. Stress-enhanced fear learning (SEFL) is a rodent model of stress exposure that captures the sensitization of fear learning seen in PTSD patients. In this dissertation I used the SEFL model to investigate the factors that govern stress susceptibility as well as the potential role of increased levels of Ca2+-permeable α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors in supporting enhanced fear learning following stress.PTSD is twice as common in women compared to men, yet the vast majority of rodent studies investigating this condition have been conducted exclusively in males. In Chapter 1 I investigated sex differences in responses to an intense, acute stressor (15 footshocks over 90 minutes). I found that male and female rats showed similar changes in fear learning following stress, while some aspects of anxiety-related behavior were differentially impacted by stress. While levels of the GluA1 subunit of the AMPA receptor were altered within the basolateral amygdala (BLA) of male rats, no changes in GluA1 levels were observed in females. In Chapter 2 I demonstrated that a reduced severity stressor (4 footshocks over 90 minutes) revealed distinct susceptible and resilient groups in both male and female rats. Susceptible male and female rats showed increased fear and anxiety compared to their resilient counterparts. Some aspects of stress susceptibility manifested differently in males compared to females, with susceptibility associated with increased alcohol intake in males and increased baseline anxiety in females. In Chapter 3 I tested the role of Ca2+-permeable AMPA receptors in supporting changes in fear learning following stress. I found that neither overexpression of GluA1 nor genetic deletion of GluA2 was sufficient to enhance fear learning in the absence of stress. However, GluA2 deletion paired with prior footshock exposure enhanced subsequent fear learning in male mice but not females. Collectively, these findings indicate that while males and females show similar changes in fear learning following stress, these behavioral changes may be supported by different mechanisms. These results provide insight into the differential impacts of stress on behaviors relevant to PTSD symptomology across sexes.