In vitro reporter gene assays for assessment of PPAR- and Nrf2-mediated health effects of tomato and its bioactive constituents

The consumption of food products with health-promoting properties, such as for example margarines with plant sterols, fruit juice enriched with calcium and cereals with (soluble) fibre, has increased rapidly during the last years. The present thesis provides proof-of-principle that reporter gene assays are effective tools to investigate the effects of functional foods and food compounds on gene expression pathways. In order to test fruits and vegetables for selected functions in reporter gene assays, extraction methods for vegetables and fruits are needed which result in extracts suitable for testing in cell-based assays. In this thesis, methods are described to prepare three different tomato extracts, containing 1) apolar compounds such as isoprenoids, 2) semi-polar compounds, such as flavonoid glycosides and 3) deglycosylated semi-polar compounds such as flavonoid aglycones. All three tomato extracts were compatible with cell-physiological and cell culture conditions. Next, the three tomato extracts and individual tomato compounds were tested for their capacity to induce EpRE-mediated, PPARγ-mediated and PPARα-mediated gene expression using three reporter gene assays; the previously described EpRE-LUX assay and the newly developed PPARγ CALUX and PPARα CALUX assays. Both flavonoid glycosides and the tomato extract containing semi-polar compounds were unable to induce EpRE-mediated, PPARγ-mediated and PPARα-mediated gene expression, while the tomato extract containing deglycosylated semi-polar compounds, as well as individual flavonoid aglycones were able to induce EpRE-mediated, PPARγ-mediated and PPARα-mediated gene expression. Some individual isoprenoids and the tomato extract containing apolar compounds were also inducing PPARγ-mediated and PPARα-mediated gene expression, but not EpRE-mediated gene expression. To assess differences between tomato varieties, extracts containing deglycosylated semi-polar compounds were generated from fruits of different tomato varieties. These were then tested for their capacity to induce EpRE-mediated (97 varieties), PPARγ-mediated (7 varieties) and PPARα-mediated (7 varieties) gene expression. The extracts of these varieties were all able to induce both EpRE-mediated, PPARγ-mediated and PPARα-mediated gene expression. For EpRE-mediated gene expression, a 3-fold difference was found between the least potent and the most potent variety; for PPARγ-mediated gene expression, this difference was 1.6-fold, and for PPARα-mediated gene expression this was 1.7-fold. Metabolomics profiles of the 97 extracts were generated in order to identify phytochemicals responsible for the differences in potency of the 97 tomato varieties to induce EpRE-mediated gene expression. The flavonoid aglycone quercetin was identified as one of the main compounds responsible for the ability of tomato extracts to induce EpRE-mediated gene expression using multivariate analysis that combined the reporter gene assay data with metabolite profiles of the same tomato extracts. In addition, yet unidentified compounds correlated with the response in the EpRE-LUX assay and these compounds may also contribute to the observed induction of EpRE-mediated gene expression. In conclusion, the work presented in this thesis provides proof-of-principle that reporter gene assays can be implemented for screening bioactive food compounds as well as whole fruit and vegetable extracts for their capacity to induce gene expression through specific health-related transcription factor.It was shown that reporter gene assays are able to pick up differences in activation of these transcription factors induced by different varieties of tomato. Furthermore, this thesis provides proof-of-principle that active ingredients contributing to the activity of the whole tomato extracts could be identified by combining reporter gene assays and metabolite profiling. Our results indicate that combining multiple reporter gene assays and metabolomics profiling, is useful in fast screening of larger numbers of food items and food compounds. Therefore, the use of reporter gene assays as a first tier in a tiered approach aiming at supporting health claims will limit the number of animal studies and human studies needed, by enabling the ranking and selection of highly promising food items or constituents for further <em>in vivo</em> testing.