Two research lines argue for rapid stress-induced reallocations of neural network activity involving the amygdala. One focuses on the role of norepinephrine (NE) in mediating a shift towards the salience network and improving vigilance processing, whereas the other focuses on the role of cortisol in enhancing automatic, habitual responses. It has been suggested that the mineralocorticoid receptor (MR) is critical in shifting towards habitual responses, which are supported by the dorsal striatum. However, until now it remained unclear whether these two reallocations of neural recourses might be part of the same phenomenon and develop immediately after stress onset. We combined methods used in both approaches and hypothesized specifically that stress would lead to rapidly enhanced involvement of the striatum as assessed by amygala-striatal connectivity. Furthermore, we tested the hypothesis that this shift depends on cortisol interacting with the MR, by using a randomized, placebo-controlled, full-factorial, between-subjects design with the factors stress and MRblockade (spironolactone). We investigated 101 young, healthy men using functional magnetic resonance imaging after stress induction, which led to increased negative mood, heart rate, and cortisol levels. We confirmed our hypothesis by revealing a stress-by-MR-blockade interaction on the functional connectivity between the centromedial amygdala (CMA) and the dorsal striatum. Stress rapidly enhanced CMA-striatal connectivity and this effect was correlated with the stress-induced cortisol response, but required MR availability. This finding might suggest that the stress-induced shift described by distinct research lines might capture different aspects of the same phenomenon, ie, a reallocation of neural resources coordinated by both NE and cortisol. Neuropsychopharmacology (2015) 40, 947-956;

Beta and gamma rhythms have been hypothesized to be involved in global and local coordination of neuronal activity, respectively. Here, we investigated how cells in rodent area S1BF are entrained by rhythmic fluctuations at various frequencies within the local area and in connected areas, and how this depends on behavioral state and cell type. We performed simultaneous extracellular field and unit recordings in four connected areas of the freely moving rat (S1BF, V1M, perirhinal cortex, CA1). S1BF spiking activity was strongly entrained by both beta and gamma S1BF oscillations, which were associated with deactivations and activations, respectively. We identified multiple classes of fast spiking and excitatory cells in S1BF, which showed prominent differences in rhythmic entrainment and in the extent to which phase locking was modulated by behavioral state. Using an additional dataset acquired by whole-cell recordings in head-fixed mice, these cell classes could be compared with identified phenotypes showing gamma rhythmicity in their membrane potential. We next examined how S1BF cells were entrained by rhythmic fluctuations in connected brain areas. Gamma-synchronization was detected in all four areas, however we did not detect significant gamma coherence among these areas. Instead, we only found long-range coherence in the theta-beta range among these areas. In contrast to local S1BF synchronization, we found long-range S1BF-spike to CA1–LFP synchronization to be homogeneous across inhibitory and excitatory cell types. These findings suggest distinct, cell-type contributions of low and high-frequency synchronization to intra- and inter-areal neuronal interactions.

Stress is assumed to cause a shift from flexible 'cognitive' memory to more rigid 'habit' memory. In the spatial memory domain, stress impairs place learning depending on the hippocampus whereas stimulus-response learning based on the striatum appears to be improved. While the neural basis of this shift is still unclear, previous evidence in rodents points towards cortisol interacting with the mineralocorticoid receptor (MR) to affect amygdala functioning. The amygdala is in turn assumed to orchestrate the stress-induced shift in memory processing. However, an integrative study testing these mechanisms in humans is lacking. Therefore, we combined functional neuroimaging of a spatial memory task, stress-induction, and administration of an MR-antagonist in a full-factorial, randomized, placebo-controlled between-subjects design in 101 healthy males. We demonstrate that stress-induced increases in cortisol lead to enhanced stimulusresponse learning, accompanied by increased amygdala activity and connectivity to the striatum. Importantly, this shift was prevented by an acute administration of the MR-antagonist spironolactone. Our findings support a model in which the MR and the amygdala play an important role in the stress-induced shift towards habit memory systems, revealing a fundamental mechanism of adaptively allocating neural resources that may have implications for stress-related mental disorders.

No abstract