Logo Logo
Switch language to English
Plendl, Wolfgang (2010): Extinction learning in mice. Dissertation, LMU München: Fakultät für Biologie



Anxiety disorders are characterized by an exaggerated fear response towards a non-harmful stimulus or situation. Exposure therapy is an effective treatment and consists of repeated non-reinforced exposures to the stimulus. Currently the index of beneficial learning is the decreased level of fear at the end of an exposure session. For patients this therapy is emotionally very demanding and dropout rates are high. In addition, many patients are prone to relapse in a context and time dependent manner. In rodents, the explicit model for exposure therapy is extinction learning. After initial pairing of a previous neutral stimulus (tone) with a negative outcome (shock), animals will show a directed fear reaction (conditioned response) during re-exposure. Multiple re-exposures of the animal to the stimulus without further reinforcement lead to a gradual decay of fear (extinction training). This fear decay is shown to be based on relearning processes. Over time the original memory trace of the tone predicting the shock is opposed and inhibited by the new association that the tone does not predict the shock any longer. The focus of this work was to dissect extinction learning at the behavioural and the molecular level in a mouse model of conditioned fear. We want to critically revise nomenclature and to provide mechanistic insights and possible transfers to clinical application and to increase the persistence of fear extinction. In the first set of experiments, we showed that fear deacy during exposure (within-session) to a permanent stimulus is mediated by the endocannabionid system, involving glutamatergic neurons but not stress hormone signalling. In the second set of experiments, we challenged the view that the fear level at the end of an exposure session (within-session extinction) is the indicator for the strenght of between-session extinction. By comparing different exposure approaches, we were able to show that the decline of fear during exposures did not predict formation of between-session extinction, nor was it a prerequisite. Multiple short and unpredictable exposures to the stimuli were most effective in producing between-session extinction, compared to a permanent exposure. Starting from our results obtained in the first set of experiments we further demonstrated that the endocannabinoid system is essential for within-session extinction but not for between-session extinction. The dissociation of within-session and between-session extinction allowed us to pin down the basolateral amygdala, the cingulate cortex and the dentate gyrus as possible centers of relearning by measuring c-Fos immunoreactivity in various brain areas. Even after successful completion of extinction training subjects are prone to relapse. Reconsolidation, elicited by re-exposure, was recently identified as a phase where the original fear memory becomes vulnerable again to selective modifications. Therefore in the third set of experiments, we tried to interfere with reconsolidation in order to achieve a long lasting decrease in conditioned fear. By inhibition of intracerebral activity of PKC, within- and between- session extinction were facilitated and moreover renewal and spontanous recovery, as a control for relapse, were abolished. We showed for the first time that reconsolidation can be manipulated without having any negative influences on memory consolidation during extinction training. By means of c-Fos immunohistochemistry we were able to identifiy the cin7 gulate cortex as a candidate region to mediate reconsolidation processes. In summary this study demonstrates for the first time that endocannabinoid signalling at glutamatergic neurons in the forebrain is essential for within-session extinction but also independent of CRH signalling and corticosterone secretion. Second, within-session and between- session extinction can be dissociated at the behavioural, anatomical and molecular level. Third, by blocking reconsolidation, the conditioned fear memory trace can be degraded during extinction, thus resulting in a lasting decrease of fear. These findings challenge a number of facts, which have been taken for granted in current psychotherapy practice. We hope that this thesis will contribute to develop new therapeutic approaches for the treatment of phobias or trauma-related anxiety disorders.