Classical fear conditioning – including acquisition and extinction – is a model for fear learning and memory in health and disease. Moreover, trauma-related disorders can be viewed as comprising fear acquisition under severe stress. Yet, in humans, we know comparatively little about how acute stress affects fear conditioning. Therefore, the first aim of this thesis was to investigate the effect of stress on fear acquisition or extinction. Stress induces multiple hormonal and neurotransmitter changes dynamically developing over time, including a fast first-wave and a slower second-wave stress response. Models derived from avoidance learning and declarative memory studies suggest that stress effects on memory depend on the temporal proximity between learning and stressor: encoding close to the stressor will be enhanced, but encoding and recall later in time (during the second-wave) will be suppressed (e.g., Schwabe, Joëls, Roozendaal, Wolf, & Oitzl, 2012). So far, these predictions were not related to fear conditioning. Therefore, we investigated if the model-based predictions are also valid in human fear conditioning. We used two stressors to investigate first-wave and second-wave stress effects: the cold pressor test (CPT) inducing a strong first-wave but little second-wave activation and a psychosocial stressor, reliably inducing both, first- and second-wave stress responses. Conditioning was measured via skin conductance responses (SCRs). Investigating the first-wave (Experiment 2), we placed fear acquisition and immediate extinction directly after the CPT (n = 20) or after the control treatment (n = 20). We found no group difference in acquisition performance, but significantly increased extinction resistance in the stressed CPT group. In Experiment 3, CPT (n = 20) or control (n = 20) was placed after acquisition but directly prior to extinction training. Here, we found improved extinction and 24h-delayed extinction recall after CPT. Investigating the second-wave (Experiment 1), we placed fear acquisition and immediate extinction 45 min after the psychosocial stressor (i.e., at the peak of salivary cortisol, n = 12) or after control (n = 12). Here, we found no significant stress effects. Sex and female sex hormones also influence fear conditioning: Women are at a higher risk to develop anxiety and stressor-related disorders than men. Interestingly, patients with these disorders show impaired fear extinction and extinction recall, and low levels of the sex hormone 17β-estradiol (E2) are linked to impaired extinction in both, healthy and patient female samples. So far, there is little data on how acute stress and circulating E2-levels might interact in fear acquisition and especially in fear extinction. Therefore, the second aim of this thesis was to explore this possible interaction in healthy women in different cycle phases compared to men. Thus, in Experiment 4, we included hormone status as a quasi-experimental variable and compared free cycling women in the midcycle phase (high E2, low progesterone, n = 24), women in the early follicular phase of the menstrual cycle (low E2, low progesterone, n = 24), and men (n = 24). We placed fear acquisition and extinction 45 min after the psychosocial stressor (n = 36) or control (n = 36), and tested extinction recall after 24 h. In line with Experiment 1, the second-wave stressor did not affect fear acquisition and immediate extinction. However, we found a stress by hormone status interaction within women at the 24h-delayed extinction recall test: in the stressed group, early follicular women showed impaired extinction recall and a higher return of fear compared to midcycle women, whereas there was no difference between early follicular and midcycle women after control treatment. Collectively our results support a different role for the first- and second-wave stress response in human fear conditioning. Fear acquisition near the first-wave stress response results in enhanced fear memory, which is resistant to extinction. Extinction training near the first-wave enhances extinction learning. In contrast, fear conditioning at the peak of the peripheral second-wave cortisol response had no effect on acquisition or extinction performance. However, second-wave stress interacted with the hormone status of women, where only women in a low E2 state were vulnerable to negative stress effects in extinction recall. The last result will encourage further investigation of the interplay between E2 and stress in fear extinction. Enhancement of extinction by the CPT could – if replicated – be translated into strategies for optimizing exposure therapy.

Le stress peut induire des altérations durables de la mémoire, particulièrement via une accentuation de la mémoire émotionnelle (Roozendaal B,McEwen BS and Chattarji S, 2009). Les personnes ayant vécu une adversité extrême au cours leur enfance (stress précoce) présentent une incidence plus élevée de troubles psychiatriques à l’âge adulte, incluant les troubles de stress post-traumatique (Malter Cohen M et al., 2013). De nombreuses dysfonctions comportementales associées au stress précoce se reflètent fidèlement chez les modèles murins, notamment l’exacerbation des souvenirs de peur (Guadagno A et al., 2021). Dans la réponse au stress, les glucocorticoïdes jouent un rôle central. Alors, les astrocytes ont une expression de récepteurs aux glucocorticoïdes beaucoup plus importante par rapport aux neurones (Zhang Y et al., 2014). Malgré cela, le rôle des astrocytes dans les réponses de stress demeure à définir. Ce projet a pour but de déterminer si les déficits comportementaux induits par le stress précoce peuvent expliquer la variabilité des thérapies d’expositions ainsi que d’élucider le rôle de la signalisation glucocorticoïde astrocytaire dans ces processus. La thérapie d’exposition, type le plus employé dans la gestion des troubles de stress, démontre en effet un taux d’abandon élevé, souvent justifié par la régression ou l’immuabilité des symptômes (Lewis C et al., 2020; Wells SY et al., 2023). Pour ce faire, des souris naïves ou avec stress précoce furent soumises à un protocole court d’extinction de peur. Au contraire des naïves, les souris avec stress précoce ne furent pas en mesure d’atteindre le seuil d’extinction efficace défini. Afin de modéliser une complétion optimale de thérapie, les souris furent soumises à un protocole long d’extinction. Les souris avec stress précoce furent en mesure de rattraper le déficit d’extinction et d’atteindre des niveaux semblables aux autres groupes, néanmoins en demandant davantage de sessions. Aussi, une délétion des récepteurs astrocytaires aux glucocorticoïdes dans l’amygdale fut en mesure de renverser ce déficit tant dans le protocole court que dans le protocole long. Ces données suggèrent la signalisation glucocorticoïde astrocytaire comme déterminant clé des déficits d’extinction induits en réponse au stress précoce et permettent de suggérer un mécanisme par lequel les astrocytes médient certaines susceptibilités psycho-pathologiques induites par le stress précoce. / Stress can induce robust memory alterations, particularly through the magnification of emotionally salient memories (Roozendaal B,McEwen BS and Chattarji S, 2009). Individuals that have previously experienced extreme adversity in their childhood (ELS: early-life stress) present a higher incidence of psychiatric disorders in adulthood, including post-traumatic stress disorder (Malter Cohen M et al., 2013). Many behavioural dysfunctions associated with early-life stress are faithfully reflected in rodent models, notably the exacerbation of fear memories (Guadagno A et al., 2021). Glucocorticoids play a central role in stress responses. In fact, astrocytes have been shown to express a higher level of glucocorticoid receptors in comparison to neurons (Zhang Y et al., 2014). Despite this, very little is known regarding the implication of astrocytes in stress responses. This project aimed to determine whether behavioural impairments induced by early-life stress could explain the variability observed in exposure therapies, and to define whether astrocyte glucocorticoid signaling was involved in this process. Exposure therapy, the main paradigm used in the treatment of stress disorders, presents an important drop-out rate, often justified by a regression or a lack of improvement of the symptoms (Lewis C et al., 2020; Wells SY et al., 2023). In this optic, naive and ELS mice were subjected to a short fear extinction protocol. In contrast to naives, ELS mice were unable to reach the defined extinction efficacy threshold. With the goal of modeling an optimal treatment completion, mice underwent a prolonged extinction protocol. ELS mice were able to catch up to naive levels of extinction, although requiring more sessions. Additionally, an astrocytic glucocorticoid receptor knock-out (GRKO) was sufficient to reverse this deficit in both the short and long extinction protocol, implying a role for astrocyte glucocorticoid signaling in the long-term adverse effects of early-life stress. These data suggest astrocyte glucocorticoid signaling as a key player in extinction deficits occurring in response to early-life stress and suggest a mechanism through which astrocytes mediate some ELS-induced psycho-pathological susceptibilities.