Gene expression profiling in molecular studies of hormone actions

Abstract

Although the link between hormones and their physiological effects have been known for a long time, large pieces are still missing in our understanding of how these extracellular signals induce their effects at the cellular level. Signal transduction studies have gathered plenty of information about hormone signaling, but the complex network of interactions between different hormones, signaling pathways and cell types is still not completely understood. Advances in genome research and the development of high throughput techniques have made it possible to approach these questions from a new angle. Gene expression profiling using microarrays enables us to study how the expression of thousands of genes is hormonally regulated in a single experiment. Efforts have been made within the last years to establish standards for microarray data handling and to explore the possibilities of downstream analysis of these large sets of data. Microarray analysis is becoming a well established technique, but its use in endocrine research has only just begun.

The aim of this thesis was to evaluate the use of gene expression profiling in studies of hormone actions. Therefore, microarrays were employed to study the effects of growth hormone (GH) on gene expression patterns in different tissues and endocrine situations in the rat. The main focus of these studies was placed on the liver, as it is one of the major targets of GH. The effects of short- and longterm treatment with thyroid hormone on liver gene expression were also investigated, as well as the dependence of those effects on thyroid hormone receptor beta. An in silico analysis of the promoter regions of some coregulated genes was performed to investigate the mechanisms behind their hormonal regulation. Finally, the effects of several hormones and gender on hepatic gene expression patterns were compared to identify hormone-specific and overlapping effects. All collected gene expression profiles have been made publicly available through the creation of EndoGED, the Endocrinology Gene Expression Database.

The analysis of a large number of gene expression profiles of endocrine relevance led to the following results: (i) It was shown that changes in hepatic gene expression during aging are partially related to changes in GH levels, and that GH is responsible for a significant portion of gender differences in hepatic gene expression. (ii) Our finding that FAT/CD36 expression is female-predominant in the liver, which may contribute to gender differences in serum lipids, can serve as an example of how expression profiling may generate new hypotheses regarding the mechanisms behind physiological effects previously lacking molecular explanation. (iii) It was shown that microarrays can be used in a novel way to classify hormonally regulated genes, and the procedure to identify putative response elements in the promoters of hormone responding genes was demonstrated. (iv) Since the large scale data output from gene expression profiling requires efficient data management to enable data mining, the database EndoGED was constructed.

In conclusion, gene expression profiling using microarrays is a valuable technique with great possibilities to resolve endocrine research questions. In combinations with other high throughput techniques, it has the potential to reveal the complex regulatory networks that mediate the path from hormonal signal to physiological effects.