Predicting targets and signaling pathways of steroid hormones using the Allen Brain Atlas

Abstract

Event Abstract Back to Event Predicting targets and signaling pathways of steroid hormones using the Allen Brain Atlas Ahmed Mahfouz1, 2*, Onno C. Meijer3, Boudewijn P. Lelieveldt2, 4 and Marcel J. Reinders1 1 Delft University of Technology, Department of Intelligent Systems, Netherlands 2 Leiden University Medical Center, Department of Radiology, Netherlands 3 Leiden University Medical Center, Department of Endocrinology, Netherlands 4 Delft University of Technology, Department of Intelligent Systems, Netherlands Steroid hormones have strong modulatory effects on cognitive function, mood and physiological regulation in the brain. Their effects are highly cell type dependent, and mediated via steroid receptors which are part of the superfamily of nuclear receptors (NRs) and act mainly as transcription factors. These effects are mediated by specific target genes and depend on cellular signaling partners, such as cell/tissue type-specific co-regulators [1, 2]. Even if these co-regulators are known for a number of well-studied brain regions [3], targets, mediators and in fact responsive brain areas remain unknown for most steroid effects on the brain. The Allen Mouse Brain Atlas (ABA) (http://mouse.brain-map.org/) provides genome-wide cellular-resolution in situ hybridization (ISH) gene expression map of the adult mouse brain [4]. This high resolution data allows the study of the effects of co-regulators on steroid receptors activity, by analyzing the co-localization of the receptors and co-regulators in different brain nuclei. We used a data-driven approach to analyze the spatial co-expression patterns between each gene in the mouse genome and estrogen receptor alfa (Esr1, NR3a1). We find that a set of 15 genes which were shown to have sexually dimorphic gene expression patterns in the mouse brain [5] is significantly co-expressed with Esr1 in sexually important brain regions, such as the hypothalamus. We have also found that the top Esr1-co-expressed genes are enriched for known estrogen response element sequences. The proposed method provides an immediately testable a set of candidate Esr1 targets and signaling partners with a role in sexual behavior. After validation, the candidate set of targets and signaling partners could allow selective targeting of brain regions. Knowledge on brain region specific target genes will greatly facilitate our understanding of the way in which steroid hormones affect so many aspects of the central nervous system (CNS) function, and will in fact lead to the prediction of unrecognized effects of steroids on the nervous system. While the approach is applicable to all nuclear receptors in the brain Esr1 based on available sources and tools to validate our predictions.