Transcriptomics of the fetal hypothalamic response to brachiocephalic occlusion and estradiol treatment

Abstract

Estradiol (E2) is a well-known modulator of fetal neuroendocrine activity and has been proposed as a critical endocrine signal readying the fetus for birth and postnatal life. To investigate the modulatory role of E2 on fetal stress responsiveness and the response of the fetal brain to asphyxic stress, we subjected chronically catheterized fetal sheep to a transient (10 min) brachiocephalic artery occlusion (BCO) or sham occlusion. Half of the fetuses received subcutaneous pellets that increased plasma E2 concentrations within the physiological range. Hypothalamic mRNA was analyzed using the Agilent 8x15k ovine array (019921), processed and annotated as previously reported by our laboratory. Analysis of the data by ANOVA revealed that E2 differentially regulated (DR) 561 genes, and BCO DR 894 genes compared with control and E2+BCO DR 1,153 genes compared with BCO alone (all P < 0.05). E2 upregulated epigenetic pathways and downregulated local steroid biosynthesis but did not significantly involve genes known to directly respond to the estrogen receptor. Brachiocephalic occlusion upregulated kinase pathways as well as genes associated with lymphocyte infiltration into the brain and downregulated neuropeptide synthesis. E2 upregulated immune- and apoptosis-related pathways after BCO and reduced kinase and epigenetic pathway responses to the BCO. Responses to BCO are different from responses to hypoxic hypoxia suggesting that mechanisms of responses to these two forms of brain hypoxia are distinct. We conclude that cerebral ischemia caused by BCO might stimulate lymphocyte infiltration into the brain and that this response appears to be modified by estradiol.

Keywords: cortisol; fetal heart; late gestation; metabolism; mitochondria.

Fig. 1.

A: expression [plotted as log2(ΔIntensity)] of AGRP as measured in the 4 groups using qPCR (open bars) and microarray (filled bars). Data are represented as means ± SE. BCO, brachiocephalic artery occlusion; E, 17β-estradiol; CON, control; qPCR, quantitative PCR. B: correlation of expression of AGRP, CRHBP, IL1B, POMC, TNF, and NPY as measured by qPCR and microarray methodology.

Fig. 2.

Volcano plot (bottom) representing the relationship between calculated P value [shown as −log₁₀(P)] and difference between control and BCO groups for each gene [represented as log₂(BCO − Control)]. Statistically significant increases in gene expression are shown in red, and statistically significant decreases in gene expression are shown in green. Above the volcano plot are inferred networks of the upregulated (red) and downregulated (green) genes, plotted as force-directed layouts of the networks. These networks are accompanied by gene ontology terms significantly associated with the up- and downregulated genes. Note that the network of downregulated genes is less compact, indicating fewer known associations among these genes.

Fig. 3.

Changes in mRNA abundance, as measured by microarray, for 8 neuropeptides (GHRH, AGRP, NPY, PRL, GALP, OXT, POMC, and CRHBP) in fetuses treated with E2 (represented as group E), BCO (represented as group BCO), and E2 plus BCO (represented as group EBCO). Changes were calculated as differences from expression values in control fetuses (represented as group C). Data are represented as group mean values.

Fig. 4.

Volcano plot (bottom) representing the relationship between calculated P value [shown as −log₁₀(P)] and difference between control and estradiol (E2) groups for each gene [represented as log₂(E2 − Control)]. Statistically significant increases in gene expression are shown in red, and statistically significant decreases in gene expression are shown in green. Above the volcano plot are inferred networks of the upregulated (red) and downregulated (green) genes, plotted as force-directed layouts of the networks. These networks are accompanied by gene ontology terms significantly associated with the up- and downregulated genes.

Fig. 5.

Venn diagram (left) and network overlap representation (right) of genes up- and downregulated by estradiol with genes known to be directly responsive to estradiol via binding of the estrogen response (ER)α with the estrogen response element (ERE) (13).

Fig. 6.

Volcano plot (bottom) representing the relationship between calculated P value [shown as −log₁₀(P)] and difference between BCO and combined estradiol and brachiocephalic occlusion (EBCO) groups for each gene [represented as log₂(BCO − Control)]. Statistically significant increases in gene expression are shown in red, and statistically significant decreases in gene expression are shown in green. Above the volcano plot are inferred networks of the upregulated (red) and downregulated (green) genes, plotted as force-directed layouts of the networks. These networks are accompanied by gene ontology terms significantly associated with the up- and downregulated genes. Note that the network of downregulated genes is less compact, indicating fewer known associations among these genes.

Fig. 7.

Networks of genes differentially regulated by hypoxia (red) or by BCO (purple). Genes differentially regulated by both stimuli are shown in yellow. Genes responsive to hypoxia are as previously reported (41).