Major Depressive Disorder (MDD) is a widespread mood disorder that affects more than 300 million people worldwide and yet, high relapse rates persist. This current study aimed to use an animal model for depression, unpredictable intermittent restraint (UIR), to investigate changes in a subset of neurons within the hippocampus, a region of high susceptibility in MDD. Adult male and female Sprague-Dawley rats were randomly assigned to four treatment groups based on sex (n = 48, n = 12/group). Half of the rats underwent UIR that involved restraint with orbital shaking (30 min or 1 h) for 2-6 consecutive days, followed by one or two days of no stressors; the other half of the rats were undisturbed (CON). UIR rats were stressed for 28 days (21 days of actual stressors) before behavioral testing began with UIR continuing between testing days for nearly 70 days. Rats were then euthanized between 9 and 11 days after the last UIR session. Brains were processed for Golgi stain and long-shaft (LS) neurons within the hippocampal CA3a and CA3b regions were quantified for dendritic complexity using a Camera Lucida attachment. Our findings failed to support our hypothesis that UIR would produce apical dendritic retraction in CA3 hippocampal LS neurons in both males and females. Given that UIR failed to produce CA3 apical dendritic retraction in males, which is commonly observed in the literature, we discuss several reasons for these findings including, time from the end of UIR to when brains were sampled, and the effects of repeated cognitive testing. Given our published findings that UIR impaired spatial ability in males, but not females, we believe that UIR holds validity as a chronic stress paradigm, as UIR attenuated body weight gain in both males and females and produced reductions in thymus gland weight in UIR males. These findings corroborate UIR as an effective stressor in males and warrant further research into the timing of UIR-induced changes in hippocampal CA3 apical dendritic morphology.

Post-Traumatic Stress Disorder (PTSD) is characterized by intrusive memories from a traumatic event. Current therapies rarely lead to complete remission. PTSD can be modeled in rodents using chronic stress (creating vulnerable phenotype) combined with fear conditioning (modeling a traumatic experience), resulting in attenuated extinction learning and impaired recall of extinction. Studies typically investigate cognition soon after chronic stress ends; however, as days and weeks pass (“rest” period) some cognitive functions may improve compared to soon after stress. Whether a rest period between chronic stress and fear conditioning/extinction would lead to improvements is unclear. In Chapter 2, male rats were chronically stressed by restraint (6hr/d/21d), a reliable method to produce cognitive changes, or assigned to a non-stressed control group (CON). After chronic stress ended, fear conditioning occurred within a day (STR-IMM), or after three (STR-R3) or six weeks (STR-R6). During the first three extinction trials, differences emerged in fear to the non-shock context: STR-R3/R6 showed significantly less fear to the context than did STR-IMM or CON. Differences were unlikely attributable to generalization or to second-order conditioning. Therefore, a rest period following chronic stress may lead to improved fear extinction and discrimination between the conditioned stimulus and environment. In Chapter 3, the infralimbic cortex (IL) was investigated due to the IL’s importance in fear extinction. Rats were infused with chemogenetics to target IL glutamatergic neurons and then assigned to CON, STR-IMM or STR-R3. During the rest period of STR-R3 and the restraint for STR-IMM, the IL was inhibited using CNO (1mg/kg BW, i.p., daily), which ended before behavioral testing. STR-R3 with IL inhibition failed to demonstrate a tone-shock association as spontaneous recovery was not observed. CON with IL inhibition behaved somewhat like STR-IMM; freezing to the extinction context was enhanced. Consequently, inhibiting IL function during the rest period following chronic stress was particularly disruptive for learning in STR-R3, impaired freezing to a safe context for CON, and had no effect in STR-IMM. These studies show that time since the end of chronic stress (recently ended or with a delay) can interact with IL functioning to modify fear learning and response.

Hysterectomy is the second most common gynecological surgery performed in women. Half of these surgeries involve removal of the uterus alone, and half involve concomitant removal of the ovaries. While the field has retained the notion that the nonpregnant uterus is dormant, more recent findings suggest that hysterectomy is associated with cognitive detriment. Of note, the clinical literature suggests that an earlier age at hysterectomy, with or without concomitant ovarian removal, increases dementia risk, implicating age at surgery as a variable of interest. While preclinical work in a rodent model of hysterectomy has demonstrated spatial working memory impairments, the role of age at surgery has yet to be addressed. The current experiment utilized a rodent model of hysterectomy to investigate the importance of age at surgery in post- surgical cognitive outcomes and to evaluate relative protein expression related to brain activity, FosB and ∆FosB, in regions critical to spatial learning processes. Young adult and middle-aged female rats underwent sham surgery, hysterectomy, or hysterectomy with ovariectomy, and were tested on a behavioral battery that evaluated spatial working and reference memory. Following the behavioral battery, animals were sacrificed and brain tissues from the Dorsal Hippocampus and Entorhinal Cortex were processed via Western Blot for relative FosB and ∆FosB expression. Behavioral analyses demonstrated that animals receiving hysterectomy, regardless of age or ovarian status, were generally impaired in learning a complex spatial working memory task. However, rats that received hysterectomy in middle-age uniquely demonstrated persistent working memory impairment, particularly with a high working memory demand. Subsequent neurobiological analyses revealed young rats that underwent hysterectomy had reduced relative FosB expression in the Entorhinal Cortex compared to sham controls, where no significant effects were observed for rats that received surgery in middle-age. Finally, unique relationships between neurobiological and behavioral outcomes were observed largely for sham rats, suggesting that such surgical manipulations might modulate these relationships. Taken together, these findings suggest that age at surgery plays an important role in learning and memory outcomes following hysterectomy, and demonstrate the need for further research into the role of the uterus in communications between the reproductive tract and brain.

Reproductive hormones are recognized for their diverse functions beyond reproduction itself, including a vital role in brain organization, structure, and function throughout the lifespan. From puberty to reproductive senescence, the female is characterized by inherent responsiveness to hormonal cyclicity. For most women, a natural transition to menopause occurs in midlife, wherein the endogenous hormonal milieu undergoes significant changes and marks the end of the reproductive life stage. Although most women experience natural menopause, many women will undergo gynecological surgery during their lifetime, which can lead to an abrupt surgical menopause. It is of critical importance to better understand how endogenous and exogenous reproductive hormone exposures across the lifespan influence cognitive and brain aging, as women are at a greater risk for developing a variety of diseases after menopause, including dementia. Using rodent models, this dissertation explores how the etiology of reproductive senescence, that is, whether it is transitional or surgical, influences the female phenotype to result in divergent cognitive outcomes dependent upon a variety of factors, with an emphasis on age at the time of intervention playing a key role in brain outcomes. Furthermore, the impact of exogenous hormone therapy on cognition is evaluated in the context of surgical menopause. A novel rat model of hysterectomy is also presented, with results demonstrating for the first time that the nonpregnant uterus, which is typically considered to be a quiescent organ, may play a unique, direct role in modulating cognitive outcomes. Neurobiological mechanisms associated with reproductive hormones and aging are assessed to better recognize neural correlates underlying the observed behavior changes. The overarching goal of this dissertation was to elucidate novel factors contributing to cognitive aging outcomes in females. Collectively, the data presented herein indicate that the age at the onset of reproductive senescence has significant implications for learning and memory outcomes, and that variations in gynecological surgery can have unique, long-lasting effects on the brain and cognition. Translationally, this series of experiments moves the field forward toward the goal of improving the health and quality of life for women throughout the lifespan.

The present series of studies examined whether a novel implementation of an

intermittent restraint (IR) chronic stress paradigm could be used to investigate hippocampal-dependent spatial ability in both sexes. In experiments 1 and 2, Sprague- Dawley male rats were used to identify the optimal IR parameters to assess spatial ability. For IR, rats were restrained for 2 or 6hrs/day (IR2, IR6, respectively) for five days and then given two days off, a process that was repeated for three weeks and compared to rats restrained for 6hrs/d for each day (DR6) and non-stressed controls (CON). Spatial memory was tested on the radial arm water maze (RAWM), object placement (OP), novel object recognition (NOR) and Y-maze. The results for the first two experiments revealed that IR6, but not IR2, was effective in impairing spatial memory in male rats and that task order impacted performance. In experiment 3, an extended IR paradigm for six weeks was implemented before spatial memory testing commenced in male and female rats (IR- M, IR-F). Unexpectedly, an extended IR paradigm failed to impair spatial memory in either males or females, suggesting that when extended, the IR paradigm may have become predictable. In experiment 4, an unpredictable IR (UIR) paradigm was implemented, in which restraint duration (30 or 60-min) combined with orbital shaking, time of day, and the days off from UIR were varied. UIR impaired spatial memory in males, but not females. Together with other reports, these findings support the interpretation that chronic stress negatively impairs hippocampal-dependent function in males, but not females, and that females appear to be resilient to spatial memory deficits in the face of chronic stress.

Nicotine self-administration is associated with decreased expression of the glial glutamate transporter 1 (GLT-1) and the cystine-glutamate exchange protein xCT in the nucleus accumbens core (NAcore). N-acetylcysteine (NAC), which is an antioxidant, anti-inflammatory, and glutamatergic agent, restores these proteins associated with increased relapse vulnerability. However, the specific molecular mechanisms driving NAC inhibitory effects on cue-induced nicotine reinstatement are unknown. Thus, the present study assessed NAC’s effects on cue-induced nicotine reinstatement are dependent on NAcore GLT-1 expression. Here, rats were treated with NAC in combination with intra-NAcore vivo-morpholinos to examine the role of GLT-1 in NAC-mediated inhibition of cue-induced nicotine seeking. Subchronic NAC treatment attenuated cue-induced nicotine seeking in male rats and an antisense vivo-morpholino (AS) designed to selectively suppress GLT-1 expression in the NAcore blocked this effect. NAC treatment was also associated with an inhibition of pro-inflammatory tumor necrosis factor alpha (TNFα) expression in the NAcore. As well, GLT-1 AS markedly increased expression of CD40, a known marker of pro-inflammatory M1 activation of microglia and macrophages. To further examine whether NAC-induced decreases in nicotine seeking involve suppression of TNFα, we manipulated a downstream mediator of this pathway, nuclear factor kappa B (NF-kB). Considering the putative role of NF-κB in learning, memory, and synaptic plasticity, separate experiments were performed where rats were treated with herpes simplex virus (HSV) vectors designed to increase (HSV-IKKca) or decrease (HSV-IKKdn) NF-κB signaling through interactions with IκB Kinase (IKK). The goal was to examine the role of NF-κB signaling in mediating nicotine seeking behavior and if NF-κB signaling regulates GLT-1 expression. HSV-IKKdn alone and in combination with NAC inhibited cue-induced nicotine reinstatement, while HSV-IKKca blocked the attenuating effect of NAC on reinstatement. Interestingly, both HSV-IKKdn and HSV-IKKca, regardless of NAC treatment, inhibited GLT-1 expression. Taken together, these results suggest that while GLT-1 may be a conserved neurobiological substrate underlying relapse vulnerability across drugs of abuse, immunomodulatory mechanisms may regulate drug-induced alterations in glutamatergic plasticity that mediate cue-induced drug-seeking behavior through GLT-1-independent mechanisms.

The goal of the present study was to investigate whether a rest period following the end of chronic stress would impact fear extinction. Past research has indicated that chronic stress leads to impairments in the learning and recall of fear conditioning extinction. Moreover, the effects of chronic stress can return to levels similar to controls when a post-stress “rest” period (i.e., undisturbed except for normal husbandry) is given prior to testing. Male rats underwent chronic restraint stress for 6hr/day/21days (STR-IMM). Some rats, underwent a post-stress rest period for 6- or 3-weeks after the end of stress (STR-R6, STR-R3). Control (CON) rats were unrestrained for the duration of the experiment. In Experiment 1, following the stress or rest manipulation, all rats were acclimated to conditioning and extinction contexts, fear conditioned with 3 tone-foot shock pairings, and then had two days of extinction training. All groups froze similarly to the tone across all training sessions. However, STR-R6/R3 froze less in the non-shock context than did STR-IMM or CON. During extinction training, STR-IMM showed high levels of freezing to the non-shock context, leading to a concern they may be generalizing across contexts. Consequently, a follow-up experiment tested for context generalization. In Experiment 2, STR-IMM rats underwent a generalization test in an environment that was either different or the same as the conditioning environment, using STR-R6 as a comparison. STR-IMM and STR-R6 showed similar relative levels of freezing to tone and context, regardless of their conditioning environment to reveal that STR-IMM did not generalize and instead, maybe expressing hypervigilance. Thus, the present study demonstrated the novel finding that a rest period from chronic stress can lead to reduced fear responsiveness in a non-shock environment.

Chronic stress results in functional and structural changes to the hippocampus. Decades of research has led to insights into the mechanisms underlying the chronic stress-induced deficits in hippocampal-mediated cognition and reduction of dendritic complexity of hippocampal neurons. Recently, a considerable focus of chronic stress research has investigated the mechanisms behind the improvements in hippocampal mediated cognition when chronic stress ends and a post-stress rest period is given. Consequently, the goal of this dissertation is to uncover the mechanisms that allow for spatial ability to improve in the aftermath of chronic stress. In chapter 2, the protein brain derived neurotrophic factor (BDNF) was investigated as a mechanism that allows for spatial ability to show improvements following the end of chronic stress. It was found that decreasing the expression of BDNF in the hippocampus prevented spatial memory improvements following a post-stress rest period. Chapter 3 was performed to determine whether hippocampal CA3 apical dendritic complexity requires BDNF to show improvements following a post-stress rest period, and whether a receptor for BDNF, TrkB, mediates the improvements of spatial ability and dendritic complexity in a temporal manner, i.e. during the rest period only. These experiments showed that decreased hippocampal BDNF expression prevented improvements in dendritic complexity, and administration of a TrkB antagonist during the rest period also prevented the improvements in spatial ability and dendritic complexity. In chapter 4, the role of the GABAergic system on spatial ability following chronic stress and a post-stress rest period was investigated. Following chronic stress, it was found that male rats showed impairments on the acquisition phase of the RAWM and this correlated with limbic glutamic acid decarboxylase, a marker for GABA. In chapter 5, a transgenic mouse that expresses a permanent marker on all GABAergic interneurons was used to assess the effects of chronic stress and a post-stress rest period on hippocampal GABAergic neurons. While no changes were found on the total number of GABAergic interneurons, specific subtypes of GABAergic interneurons were affected by stressor manipulations. Collectively, these studies reveal some mechanisms behind the plasticity seen in the hippocampus in response to a post-stress rest period.

Aging and the menopause transition are both intricately linked to cognitive changes

during mid-life and beyond. Clinical literature suggests the age at menopause onset can differentially impact cognitive status later in life. Yet, little is known about the relationship between behavioral and brain changes that occur during the transitional stage into the post-menopausal state. Much of the pre-clinical work evaluating an animal model of menopause involves ovariectomy in rodents; however, ovariectomy results in an abrupt loss of circulating hormones and ovarian tissue, limiting the ability to evaluate gradual follicular depletion. The 4-vinylcyclohexene diepoxide (VCD) model simulates transitional menopause in rodents by selectively depleting the immature ovarian follicle reserve and allowing animals to retain their follicle-deplete ovarian tissue, resulting in a profile similar to the majority of menopausal women. Here, Vehicle or VCD treatment was administered to ovary-intact adult and middle-aged Fischer-344 rats to assess the cognitive effects of transitional menopause via VCD-induced follicular depletion over time, as well as to understand potential interactions with age, with VCD treatment beginning at either six or twelve months of age. Results indicated that subjects that experience menopause onset at a younger age had impaired spatial working memory early in the transition to a follicle-deplete state. Moreover, in the mid- and post- menopause time points, VCD-induced follicular depletion amplified an age effect, whereby Middle-Aged VCD-treated animals had poorer spatial working and reference memory performance than Young VCD-treated animals. Correlations suggested that in middle age, animals with higher circulating estrogen levels tended to perform better on spatial memory tasks. Overall, these findings suggest that the age at menopause onset is a critical parameter to consider when evaluating learning and memory across the transition to reproductive senescence. From a translational perspective, this study informs the field with respect to how the age at menopause onset might impact cognition in menopausal women, as well as provides insight into time points to explore for the window of opportunity for hormone therapy during the menopause transition to attenuate age- and menopause- related cognitive decline, and produce healthy brain aging profiles in women who retain their ovaries throughout the lifespan.

Social influences are important determinants of drug initiation in humans, particularly during adolescence and early adulthood. My dissertation tested three hypotheses: 1) conditioned and unconditioned nicotine and social rewards elicit unique patterns of neural signaling in the corticolimbic neurocircuitry when presented in combination versus individually; 2) play behavior is not necessary for expression of social reward; and 3) social context enhances nicotine self-administration. To test the first hypothesis, Fos protein was measured in response to social and nicotine reward stimuli given alone or in combination and in response to environmental cues associated with the rewards in a conditioned place preference (CPP) test. Social-conditioned environmental stimuli attenuated Fos expression in the nucleus accumbens core. A social partner elevated Fos expression in the caudate-putamen, medial and central amygdala, and both nucleus accumbens subregions. Nicotine decreased Fos expression in the cingulate cortex, caudate-putamen, and the nucleus accumbens core. Both stimuli combined elevated Fos expression in the basolateral amygdala and ventral tegmental area, suggesting possible overlap in processing both rewards in these regions. I tested the second hypothesis with an apparatus containing compartments separated by a wire mesh barrier that allowed limited physical contact with a rat or object. While 2 pairings with a partner rat (full physical contact) produced robust CPP, additional pairings were needed for CPP with a partner behind a barrier or physical contact with an object (i.e., tennis ball). The results demonstrate that physical contact with a partner rat is not necessary to establish social-reward CPP. I tested the third hypothesis with duplex operant conditioning chambers separated either by a solid or a wire mesh barrier to allow for social interaction during self-administration sessions. Nicotine (0.015 and 0.03 mg/kg, IV) and saline self-administration were assessed in male and female young-adult rats either in the social context or isolation. Initially, a social context facilitated nicotine intake at the low dose in male rats, but suppressed intake in later sessions more strongly in female rats, suggesting that social factors exert strong sex-dependent influences on self-administration. These novel findings highlight the importance of social influences on several nicotine-related behavioral paradigms and associated neurocircuitry.