Gene expression profiling of early-activated targets of antidepressants in glia cells

The identification of cell-type-specific and early-activated pharmacological targets of antidepressants might help to implement the search for alternative targets to develop drugs with safer side-effect profiles and faster onset of actions. Glia cells are active partners of neurons and blood vessels in supporting a variety of functions such as synaptic communication and blood-brain barrier permeability. Impaired glia cells (astrocytes and microglia cells) characterize postmortem brains of major depressive disorder patients, suggesting that they might be specifically modulated by pharmacological treatments to reverse disease phenotypes. Here, we have treated rat C6 glioma cells, used as a model of astrocytes, with two different classes of antidepressants (desipramine and fluoxetine) and two drugs without antidepressant properties (haloperidol and diazepam) to compare transcriptomes of glia cells after treatment with antidepressants versus drugs without antidepressant properties. To this aim, we used microarrays to detail the global early-activated cell-type-specific gene expression program and identified multiple classes of genes selectively modulated by antidepressants in astrocytes. Rat C6 glioma cells were chosen as they are considered a model of astrocytes. They were treated with antidepressants from two different classes (desipramine, a norepinephrine and serotonine reuptake inhibitor, and fluoxetine, a selective serotonin reuptake inhibitor) and drugs without antidepressant properties (haloperidol, a typical antipsychotic, and diazepam, a benzodiazepine). RNA extraction and hybridization on Illumina microarrays followed after 2 hours of drug treatment. Six replicates each treatment. Two diazepam-treated samples did not meet quality control, and were not included in this submission. We used C6 glioma cells instead of primary astrocytes to obtain a homogeneous populations of cells in order to increase the consistency and reproducibility of data analysis after expression profiling.