RESUMO
The potent neurotoxicity of benzo[a]pyrene (B[a]P) has been suggested to be a susceptibility factor accelerating the onset of brain tumours and the emergence of neurobehavioural disturbances. B[a]P has been shown to be neurotoxic, acting directly on both the central and peripheral nervous systems, as well as indirectly via peripheral organs like liver and gut. By using a realistic B[a]P exposure scenario (0.02-200 mg/kg/day, 10 days) in mice, we elucidated brain-specific B[a]P metabolism and at identified hydroxylated B[a]P metabolites in serum which could be used as markers of cognitive impairment. Repeated oral administration of B[a]P led to, at the doses of 20 and 200 mg/kg/day, significant overexpression of Cyp1a1/Cyp1b1 in 2 out of the 3 brain regions considered, thereby suggesting the ability of the brain to metabolize B[a]P itself. At the same doses, mice exhibited a reduction in anxiety in both the elevated plus maze and the hole board apparatus. Concomitantly, B[a]P triggered dose-dependent changes in Nmda subunit expression (Nr1 and Nr2a/Nr2b) in areas involved in cognition. We detected 9-OH-B[a]P and 7,8-diol-B[a]P in serum at the level for which cognitive impairment was observed. We suggest that these metabolites may, in the future be exploited as potent biomarkers of B[a]P-induced cognitive impairments.
RESUMO
Glucocorticoids and the glucocorticoid (GR) and mineralocorticoid (MR) receptors have been implicated in many processes, particularly in negative feedback regulation of the hypothalamic-pituitary-adrenal axis. Epigenetically programmed GR alternative promoter usage underlies transcriptional control of GR levels, generation of GR 3' splice variants, and the overall GC response in the brain. No detailed analysis of GR first exons or GR transcript variants throughout the human brain has been reported. Therefore we investigated post mortem tissues from 28 brain regions of 5 individuals. GR first exons were expressed throughout the healthy human brain with no region-specific usage patterns. First exon levels were highly inter-correlated suggesting that they are co-regulated. GR 3' splice variants (GRα and GR-P) were equally distributed in all regions, and GRß expression was always low. GR/MR ratios showed significant differences between the 28 tissues with the highest ratio in the pituitary gland. Modification levels of individual CpG dinucleotides, including 5-mC and 5-hmC, in promoters 1D, 1E, 1F, and 1H were low, and diffusely clustered; despite significant heterogeneity between the donors. In agreement with this clustering, sum modification levels rather than individual CpG modifications correlated with GR expression. Two-way ANOVA showed that this sum modification was both promoter and brain region specific, but that there was however no promoter*tissue interaction. The heterogeneity between donors may however hide such an interaction. In both promoters 1F and 1H modification levels correlated with GRα expression suggesting that 5-mC and 5-hmC play an important role in fine tuning GR expression levels throughout the brain.