Transcriptomics Modeling of the Late-Gestation Fetal Pituitary Response to Transient Hypoxia

Abstract

Background: The late-gestation fetal sheep responds to hypoxia with physiological, neuroendocrine, and cellular responses that aid in fetal survival. The response of the fetus to hypoxia represents a coordinated effort to maximize oxygen transfer from the mother and minimize wasteful oxygen consumption by the fetus. While there have been many studies aimed at investigating the coordinated physiological and endocrine responses to hypoxia, and while immunohistochemical or in situ hybridization studies have revealed pathways supporting the endocrine function of the pituitary, there is little known about the coordinated cellular response of the pituitary to the hypoxia.

Results: Thirty min hypoxia (from 17.0±1.7 to 8.0±0.8 mm Hg, followed by 30 min normoxia) upregulated 595 and downregulated 790 genes in fetal pituitary (123-132 days' gestation; term = 147 days). Network inference of up- and down- regulated genes revealed a high degree of functional relatedness amongst the gene sets. Gene ontology analysis revealed upregulation of cellular metabolic processes (e.g., RNA synthesis, response to estrogens) and downregulation of protein phosphorylation, protein metabolism, and mitosis. Genes found to be at the center of the network of upregulated genes included genes important for purine binding and signaling. At the center of the downregulated network were genes involved in mRNA processing, DNA repair, sumoylation, and vesicular trafficking. Transcription factor analysis revealed that both up- and down-regulated gene sets are enriched for control by several transcription factors (e.g., SP1, MAZ, LEF1, NRF1, ELK1, NFAT, E12, PAX4) but not for HIF-1, which is known to be an important controller of genomic responses to hypoxia.

Conclusions: The multiple analytical approaches used in this study suggests that the acute response to 30 min of transient hypoxia in the late-gestation fetus results in reduced cellular metabolism and a pattern of gene expression that is consistent with cellular oxygen and ATP starvation. In this early time point, we see a vigorous gene response. But, like the hypothalamus, the transcriptomic response is not consistent with mediation by HIF-1. If HIF-1 is a significant controller of gene expression in the fetal pituitary after hypoxia, it must be at a later time.

Fig 1. Volcano plot illustrating increases and decreases in pituitary gene expression in hypoxic versus normoxic fetuses.

Differences between groups are represented on the x-axis as differences in expressed in values of logarithm of 2. Probability values are represented on the y-axis as values of logarithm of 10. Points drawn above the dashed line are statistically significant (P<0.05). 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.

Fig 2. Hierarchical clustering.

Top panel: Hierarchical clustering of gene expression in the normoxic (N1, N2, N3, and N4) and hypoxic (H1, H2, H3, H4) fetal pituitaries. Bottom panel: Principal component analysis of gene expression in normoxic (red symbols) and hypoxic (blue symbols) fetal pituitaries.

Fig 3. Network of gene ontology terms significantly (P<0.05) associated with the genes that were upregulated after hypoxia.

Note that the highest order ontology terms (e.g., glucocorticoid signaling, or prostaglandin biosynthesis, etc) are in the outermost region of the network, while more basic ontology terms (e.g., “biological process”, which is not clearly shown in this figure because of reduced size of the node labels) are in the center of the network. The intensity of the color of each node denotes statistical significance (i.e., orange greater statistical significance than yellow, and uncolored nodes are not statistically significant). All nodes labeled with arrows are statistically significant.

Fig 4. Network of gene ontology terms significantly (P<0.05) associated with the genes that were downregulated after hypoxia.

The intensity of the color of each node denotes statistical significance (i.e., orange greater statistical significance than yellow, and uncolored nodes are not statistically significant). All nodes labeled with arrows are statistically significant.

Fig 5. Overlap of inferred networks of differentially-regulated genes and genes that are potentially controlled by HIF-1.

Genes that were significantly upregulated in the pituitary after hypoxia are represented as a network with red nodes (as in Fig 1). Genes that were significantly downregulated in the pituitary after hypoxia are represented as a network with green nodes (as in Fig 1). Genes with HIF-1 consensus binding sites in their putative promoter regions (-2 to +2 kB) are represented as yellow nodes. Genes that were found to be in both the red- and yellow networks, and genes that were found to be in both the green- and yellow- networks after network merge are colored blue. The merged network is represented in a force-directed layout.