Modulation of gene expression by chronic stress in astroglia in hippocampus and prefrontal cortex of the rat

Abstract

It has been suggested that there are causal relationships between alterations in brain glia and major depression. In order to investigate whether a depressive-like state induces changes in brain astrocytes, we analyzed the modulation of astroglia-specific gene expression in the hippocampus of male rats using two models of chronic stress: chronic social stress (5 weeks, daily confrontations with a dominant male) and chronic restraint stress (3 weeks, immobilization 6h/day). Furthermore, the effects of chronic citalopram administration on hippocampal as well as prefrontal cortex (PFC) gene expression were assessed in the chronic social stress experiment. The expression of six astroglial genes was determined: N-myc-downregulated gene 2 (Ndrg2), glial fibrillary acidic protein (GFAP), glutamate transporter 1 (GLT-1), glutamate and aspartate transporter (GLAST), aquaporin 4 (AQP4) and the inward rectifying potassium channel 4.1 (Kir4.1). Furthermore, protein analyses in the hippocampus were performed by means of quantitative Western blots. Since so far, there was no reference gene validated for the PFC in experiments where animals were subjected to chronic social stress, the first step was to test several commonly used reference genes for expression stability in this part of the brain. In order to study hemispheric effects of stress in the PFC, the left and right hemispheres were analyzed separately. The results of astroglial gene expression after chronic stress suggest differential regulation depending of the experimental stress paradigm. Chronic restraint stress altered expression of astroglial genes which have a direct effect on neuronal activity: GLT-1 plays an essential role in glutamate clearance from the synaptic cleft, and Kir4.1 is fundamental in keeping low K+ concentrations in the interstitial space. The upregulation of GLT-1 at the mRNA level and the downregulation of Kir4.1 at the protein level, suggest altered glutamate and potassium ion homeostasis after chronic restraint stress. Also chronic social stress induced profound changes in astroglia. It downregulated GFAP, which might indicate the presence of resting astrocytes. Whether this represents a pathological process or is an adaptive mechanism that protects the system from overshooting remains to be elucidated. Chronic social stress also upregulated NDRG2 expression which was not due to alterations in the number of astrocytes, but to changes in the amount of NDRG2 expressed per cell. Altered NDRG2 expression might have an impact on cell proliferation. The observed changes in gene expression in the hippocampus after chronic social stress were not reversed by a 4-weeks treatment with citalopram, in contrast to previous findings in the dorsal raphe nucleus. However, in the hippocampal formation, citalopram reversed the stress-induced changes in two neuronal genes involved in synaptic transmission, the synaptosomal-associated protein 25 (SNAP-25) and syntaxin-1A. Taken together, these results might indicate that citalopram's therapeutic effects depend on the brain region with its specific neurochemical environment as well as features of the target cells. Also, it appears that restoration of normal astroglial gene expression in the hippocampus is not a prerequisite for the therapeutic effects of citalopram. These findings are in concordance with the hypothesis that chronic antidepressant treatments stimulate plasticity of brain cells; however, glial changes may show a different time course in comparison to neuronal alterations. In regard to the quantification of gene expression in the PFC, it is not possible to draw clear conclusions because expression of several presumptive reference genes was also affected by the chronic stress, at least in the left PFC. Nonetheless, analysis of reference gene stability revealed that cyclophilin was stably expressed in the right PFC. Furthermore, in an attempt to gain insight into the potential role of Ndrg2, astrocyte cultures were generated. Since NDRG2 has been previously related to processes of cell proliferation and stabilization of cell morphology, the astrocyte cultures were transduced with AAV vectors expressing two isoforms of Ndrg2 (Ndrg2S and Ndrg2L). EGFP-transduced cultures as well as buffer treated ones were used as controls. Subsequently, morphological measurements, proliferation studies and analysis of gene expression were performed on the astrocytes. The results suggest that the EGFP-transduced cultures were not an appropriate control for transduction, as these cultures showed differences compared to the buffer control in terms of morphological parameters and proliferation. Therefore, it was not possible to conclude if the changes in cell proliferation observed after NDRG2 transduction were related to NDRG2 expression or to the transduction procedure per se. No significant changes were observed in the morphological parameters measured, and GFAP quantification did not show significant alterations after NDRG2 transduction while there was a high variability in the results from the different experiments