Genomic, transcriptomic and proteomic analysis of the nuclear receptor PPARa

Accumulating wealth in the western world has led to an increase in chronic health problems such as obesity. Obesity is often associated with abnormal metabolic function, resulting in increased glucose, cholesterol and triglyceride levels in the blood. Development of cardiovascular diseases and type 2 diabetes is often observed as a result of obesity. Disturbed lipid metabolism is one of the major processes altered with obesity. Investigation of the fundamental molecular mechanisms involved in lipid metabolism may be beneficial to the prevention and treatment of this disease. One of the proteins involved in the regulation of the lipid metabolism is the nuclear receptor named Peroxisome Proliferator-Activated Receptor alpha (PPARα). PPARα resides in the nucleus and functions as an inducible transcription factor, which upon binding of fatty acid binds to specific regions within the DNA. As a consequence, the transcription of several genes that encode for enzymes and other proteins that are responsible for the catabolism of fatty acids and energy homeostasis is increased.
In this thesis, several aspects related to the mechanism of PPARα function are described. First it was investigated what proteins physically interact with PPARα and may thus be necessary for proper transcription of target genes. Secondly it was investigated what genomic locations PPARα is bound to and if the binding results in enhanced expression of genes located near these binding locations. Thirdly, the predictability of PPARα DNA binding based on specific DNA sequences was studied. Finally, a comparison was made between PPARα regulated genes in primary human hepatocytes and the human liver cell line HepG2.
The presence of PPARα alone is not sufficient to induce the expression of a target gene. Additional proteins, coactivators, have to interact with PPARα to form a larger protein complex that stimulates gene expression. Currently, it is not completely known which coregulator proteins are involved in PPARα-mediated gene regulation. With the use of immunoprecipitation and mass spectrometry, it is possible to isolate protein complexes and identify protein interactions with a tagged protein. Those techniques were used to identify proteins that interact with PPARα, resulting in the identification of several proteins that may play a role in the regulation of gene expression by PPARα.
To investigate which genomic locations are bound by PPARα after ligand activation, the ChIP-Chip technology was used. This technique combines chromatin immunprecipitation with specialized microarrays called DNA tiling arrays . With the use of ChIP it is possible to isolate DNA fragments of approximately 1000 base pairs that are bound by PPARα. By hybridizing the isolated DNA fragments to a promoter tiling microarray, it is possible to create a profile of PPARα binding across the whole genome. Over 4000 DNA regions were identified using this approach, including binding regions near genes that were previously described to be regulated by PPARα The identified genes located near PPARα binding regions were compared with gene expression data. PPARα binding regions near genes that were transcriptionally downregulated were found to be enriched for a binding motif for the STAT transcription factors. Binding of STAT3 and STAT1 to these regions was shown to be reduced upon PPARα activation. Interestingly, STATs are involved in inflammatory processes that are known to be inhibited by PPARα activation and were previously shown to be inhibited by PPARγ. Accordingly, it can be hypothesized that down-regulation of gene expression by PPARα ligands is partially mediated by interfering with binding of STAT3 and STAT1 to the DNA. Similarly, PPARα binding regions near genes that were transcriptionally upregulated were found to be enriched for a binding motif for TBP and C/EBPα. In addition, important cross-talk between PPARα and SREBP was found with respect to upregulation of gene expression. Possibly, these combinations of transcription factors play an important role in PPARα-mediated gene regulation.
To examine the relation between DNA binding by PPARα, changes in gene expression, and predicted PPARα DNA binding, we compared various data sets via a systems biology type of analysis. This comparison clearly showed that the number of locations where PPARα is able to bind greatly exceeds the number of genes that are transcriptionally regulated. This indicates that in many cases DNA binding by PPARα does not result in changes in expression of genes located nearest to the binding location. Possibly, PPARα binding has another unknown function at those locations. It appeared also from this comparison that a prediction of PPARα binding on the DNA on the basis of sequence alone did not give an enrichment of PPARα binding sites in the identified ChIP-Chip regions compared to control promotor regions. However, one of the prediction algorithms did give an enrichment on basis of sequence when only considering PPARα binding locations associated with upregulated genes. The latter method may be useful for identification of new genes that are regulated by PPARα.
Finally, a comparison was made between the gene expression patterns after PPARα activation in the human liver cell line HepG2 and human primary hepatocytes. It appears that the HepG2 cell line poorly reflects the known PPARα function in comparison to primary hepatocytes. Several known PPARα target genes are induced by PPARα agonist in primary hepatocytes but are unresponsive in the HepG2 cell line. Therefore, it is recommended to exercise caution in the interpretation of results using HepG2 cells and if possible select primary hepatocytes. However, since an alternative human liver cell line is not widely available, HepG2 still remains the best model for several applications.