A risk assessment-driven quantitative comparison of gene expression profiles in PBMCs and white adipose tissue of humans and rats after isoflavone supplementation. (September 2016)

• Record Type: Journal Article
• Authors: van der Velpen, Vera; van 't Veer, Pieter; Islam, M. Ariful; ter Braak, C.J.F.; van Leeuwen, F.X. Rolaf; Afman, Lydia A.; Hollman, Peter C.; Schouten, Evert G.; Geelen, Anouk
• Is Part Of: Food and chemical toxicology. Volume 95(2016)
• Page Start: 203
• Page End: 210
• Publication Date: 2016-09
• Subjects: Risk assessment -- Gene expression -- Species and tissue differences -- Quantitative evaluation -- Isoflavones -- Multivariate model; ER estrogen receptor -- ERGDB estrogen responsive gene database -- GSEA gene set enrichment analysis -- OVX ovariectomized -- PBMC peripheral blood mononuclear cells -- PCA principal component analysis -- RDA redundancy analysis -- RMSE root mean square effect -- SES summary effect size -- WAT white adipose tissue; Toxicology -- Periodicals; Food poisoning -- Periodicals; Food Poisoning -- Periodicals; Toxicology -- Periodicals; Toxicologie -- Périodiques; Intoxications alimentaires -- Périodiques; Food poisoning; Toxicology; Periodicals; Electronic journals; 615.9
• DOI: 10.1016/j.fct.2016.07.013
• Languages: English
• ISSNs: 0278-6915

Abstract

Abstract: Quantitative insight into species differences in risk assessment is expected to reduce uncertainty and variability related to extrapolation from animals to humans. This paper explores quantification and comparison of gene expression data between tissues and species from intervention studies with isoflavones. Gene expression data from peripheral blood mononuclear cells (PBMCs) and white adipose tissue (WAT) after 8wk isoflavone interventions in postmenopausal women and ovariectomized F344 rats were used. A multivariate model was applied to quantify gene expression effects, which showed 3–5-fold larger effect sizes in rats compared to humans. For estrogen responsive genes, a 5-fold greater effect size was found in rats than in humans. For these genes, intertissue correlations (r = 0.23 in humans, r = 0.22 in rats) and interspecies correlation in WAT (r = 0.31) were statistically significant. Effect sizes, intertissue and interspecies correlations for some groups of genes within energy metabolism, inflammation and cell cycle processes were significant, but weak. Quantification of gene expression data reveals differences between rats and women in effect magnitude after isoflavone supplementation. For risk assessment, quantification of gene expression data and subsequent calculation of intertissue and interspecies correlations within biological pathways will further strengthen knowledge on comparability between tissues and species. Graphical abstract: Highlights: Gene expression data from one animal experiment and two human intervention studies were quantitatively compared. A multivariate model was used to estimate effect sizes of gene expression in PBMCs and WAT of both species. Intertissue correlations between PBMCs and WAT were significant for estrogen-related genes in both humans and rats. Interspecies correlations were significant for oxidative phosphorylation in PBMCs and estrogen-related genes in WAT. Estimation of gene expression effect size and correlations between tissues and species can be useful for risk assessment.