Studies of Central Nervous System Functional Anatomy

National Institute Of Mental Health

Investigators

Linked publications & trials

Abstract

Chronic psychosocial stress has been implicated in the etiology and progression of psychiatric disorders such as major depression and post-traumatic stress disorder (PTSD). We study the effects of psychosocial stress in a chronic social defeat (CSD) paradigm in mice that generates behavioral alterations manifested as social avoidance, anhedonia, anxiety, and depressive-like states. In CSD, two male mice are placed in a continuous dyadic living relationship in which the subordinate experimental mouse is chronically exposed to and periodically briefly defeated by a dominant mouse of a different strain. Over the course of weeks in this living situation, the experimental mouse develops asocial, anhedonic, and anxious behaviors that can be experimentally linked with neurochemical and physical alterations in identified stress-related, limbic brain circuits. For example, our past research showed that animals undergoing CSD have reduced hippocampal new-cell proliferation and reduced medial prefrontal cortical myelination. These kinds of changes have been associated with anxiety-like and depressive-like behaviors in animal and human studies. Thus, the paradigm allows us to examine mechanistic bases for mental illness in humans. Stressful events engage the periphery via well-characterized activation of the hypothalamic-pituitary-adrenal (HPA) axis and by activity within the sympathetic nervous system (SNS). Stress-induced hormonal and neural activity affects all peripheral organs, and these organs, in turn, generate neuronal, cellular, and chemical signals that can reach the brain to adjust the homeostatic balance in response to challenges. Many of these returning signals must first encounter the blood-brain barrier (BBB), which largely excludes blood-borne signals generated by the peripheral organs. However, numerous bypass mechanisms have evolved to permit information transfer into the brain. Furthermore, the BBB at a cellular level is highly dynamic, and transient breeches can occur under stressful conditions. We were surprised to discover that animals susceptible to CSD (called CSD-S mice), i.e., mice that showed asocial and anxiety-like behaviors, had microglial gene expression changes that suggested breakdown of the extracellular matrix, phagocytic activity, and BBB extravasation. To investigate this further, we used histochemical methods to discover that indeed, CSD-S animals showed scattered, rare microhemorrhages throughout the brain. These small BBB breaches allowed inflammatory blood products such as fibrinogen to gain access to perivascular spaces and the brain parenchyma. At the level of the blood vessel, the main constituent of the BBB is the vascular brain endothelial cell, so we endeavored to characterize the response of these cells to CSD. Microarray transcriptomics analysis revealed that CSD induced in endothelial cells a dynamic temporal pattern of gene activation reflecting the biological process of wounding and repair. When we examined the recovery period following the cessation of CSD, we found that peripheral monocytes, which are not involved in the CSD and therefore do not accumulate in the 14-day period of CSD, do accumulate at perivascular areas during a one-week recovery period. Here, they phagocytose fibrinogen, an inflammatory blood product. They also appear to participate in vascular repair and behavioral recovery in this period. Elimination of the peripheral monocyte population hindered fibrinogen clearance and recovery from CSD-induced behavioral deficits in the urine scent marking (USM) task. Peripheral interactions with the brain occur at the blood-brain barrier (BBB). We showed that CSD compromised the integrity of the BBB. To get a detailed picture of the role of NVU cells in reacting to and contributing to the interaction between the brain and the periphery during and after psychosocial stress, we developed methods to isolate NVU cells and subject them to gene expression analysis by the technique of single cell RNA sequencing (scRNA-Seq). Initially we used gene-chip microarray analysis on cells isolated by CD-31-tagged magnetic bead separation of NVU elements. We examined cellular responses at 1, 7, and 14 days of CSD, and then at 7 days following termination of defeat. We published work showing that inflammatory processes characterize initiation of stress, and anti-inflammatory responses dominate in the recovery period. We validated these changes using histochemical, cellular, and molecular techniques. Seeking to extend these findings, we have optimized further the method of isolation of NVU cells and subjected the captured cells to scRNA-Seq analysis on 10x Genomic arrays. In a parallel endeavor that seeks to examine whether similar processes might occur in the brains of humans with major depression, we developed isolation methods to perform RNA-Seq on fresh-frozen tissue samples of prefrontal cortex (Brodmanns Area 10) of postmortem brains with clinical histories of depression, versus normal control cases. In order to maximize yields of the desired cell types, a negative selection strategy was developed, by which neurons, astrocytes, and oligodendrocytes were removed using antibodies that selectively mark these cell types. Following tissue homogenization, centrifugation, antibody application, and fluorescent activated cell sorting (FACS) in a flow cytometer, isolated nuclei were applied to the 10xGenomics chip for single-nuclei (sn) RNA-Seq analysis. Discrete cell populations were identified by signature gene expression, and differences among the individual populations were identified between MDD (n = 11) and control (n = 12 ) cases. The differentially expressed genes follow patterns similar to those seen in the samples from CSD mouse brain, supporting our hypothesis that psychosocial stress and is actions of the blood-brain barrier may have relevance for the etiology of human depression.