Limits...
Glucocorticoid-dependent hippocampal transcriptome in male rats: pathway-specific alterations with aging.

Chen KC, Blalock EM, Curran-Rauhut MA, Kadish I, Blalock SJ, Brewer L, Porter NM, Landfield PW - Endocrinology (2013)

Bottom Line: Short-term CORT (4 days) did not recapitulate this transcriptome.We then compared the GC transcriptome with a previously defined hippocampal aging transcriptome, revealing a high proportion of common genes.These results contradict the hypothesis that GCs simply promote brain aging and also suggest that the opposite direction shifts during aging reflect resistance to CORT regulation.

View Article: PubMed Central - PubMed

Affiliation: Department of Molecular and Biomedical Pharmacology, University of Kentucky College of Medicine, Lexington, KY 40536, USA.

ABSTRACT
Although glucocorticoids (GCs) are known to exert numerous effects in the hippocampus, their chronic regulatory functions remain poorly understood. Moreover, evidence is inconsistent regarding the long-standing hypothesis that chronic GC exposure promotes brain aging/Alzheimer disease. Here, we adrenalectomized male F344 rats at 15 months of age, maintained them for 3 months with implanted corticosterone (CORT) pellets producing low or intermediate (glucocorticoid receptor-activating) blood levels of CORT, and performed microarray/pathway analyses in hippocampal CA1. We defined the chronic GC-dependent transcriptome as 393 genes that exhibited differential expression between intermediate and low CORT groups. Short-term CORT (4 days) did not recapitulate this transcriptome. Functional processes/pathways overrepresented by chronic CORT-up-regulated genes included learning/plasticity, differentiation, glucose metabolism, and cholesterol biosynthesis, whereas processes overrepresented by CORT-down-regulated genes included inflammatory/immune/glial responses and extracellular structure. These profiles indicate that GCs chronically activate neuronal/metabolic processes while coordinately repressing a glial axis of reactivity/inflammation. We then compared the GC transcriptome with a previously defined hippocampal aging transcriptome, revealing a high proportion of common genes. Although CORT and aging moved expression of some common genes in the same direction, the majority were shifted in opposite directions by CORT and aging (eg, glial inflammatory genes down-regulated by CORT are up-regulated with aging). These results contradict the hypothesis that GCs simply promote brain aging and also suggest that the opposite direction shifts during aging reflect resistance to CORT regulation. Therefore, we propose a new model in which aging-related GC resistance develops in some target pathways, whereas GC overstimulation develops in others, together generating much of the brain aging phenotype.

Show MeSH

Related in: MedlinePlus

Microarray and pathway analysis. A, Filtering and statistical testing algorithm. Total probe sets were filtered to retain probe sets with at least 5 presence calls, and an A grade unique gene symbol level annotation. The retained 52% of probe sets were each tested by 1-way ANOVA (P ≤ .05) across the 4 treatment groups. The FDR for statistically significant genes is shown. B, Post hoc pairwise analysis (Fisher protected least significant difference, P ≤ .05) identified 393 genes that differed between low- and intermediate-CORT conditions (separated into up- and down-regulated categories), defining the GC-dependent transcriptome. C, Functional process analysis (DAVID; see Materials and Methods) revealed processes/pathways overrepresented by GC-dependent genes. For each overrepresented GO process, the number of GC-dependent genes identified for that process (#), and the overrepresentation P value (Overrep p-value) are shown. Underlined processes were added by the authors.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3713214&req=5

Figure 2: Microarray and pathway analysis. A, Filtering and statistical testing algorithm. Total probe sets were filtered to retain probe sets with at least 5 presence calls, and an A grade unique gene symbol level annotation. The retained 52% of probe sets were each tested by 1-way ANOVA (P ≤ .05) across the 4 treatment groups. The FDR for statistically significant genes is shown. B, Post hoc pairwise analysis (Fisher protected least significant difference, P ≤ .05) identified 393 genes that differed between low- and intermediate-CORT conditions (separated into up- and down-regulated categories), defining the GC-dependent transcriptome. C, Functional process analysis (DAVID; see Materials and Methods) revealed processes/pathways overrepresented by GC-dependent genes. For each overrepresented GO process, the number of GC-dependent genes identified for that process (#), and the overrepresentation P value (Overrep p-value) are shown. Underlined processes were added by the authors.

Mentions: Filtered genes were analyzed by 1-way ANOVA (Figure 2), and significant results were evaluated with post hoc all-pairwise Fisher protected least significant difference testing. Multiple testing error (false-positive results) was estimated by the false discovery rate (FDR = expected/observed) (71) procedure, and median FDRs are reported.


Glucocorticoid-dependent hippocampal transcriptome in male rats: pathway-specific alterations with aging.

Chen KC, Blalock EM, Curran-Rauhut MA, Kadish I, Blalock SJ, Brewer L, Porter NM, Landfield PW - Endocrinology (2013)

Microarray and pathway analysis. A, Filtering and statistical testing algorithm. Total probe sets were filtered to retain probe sets with at least 5 presence calls, and an A grade unique gene symbol level annotation. The retained 52% of probe sets were each tested by 1-way ANOVA (P ≤ .05) across the 4 treatment groups. The FDR for statistically significant genes is shown. B, Post hoc pairwise analysis (Fisher protected least significant difference, P ≤ .05) identified 393 genes that differed between low- and intermediate-CORT conditions (separated into up- and down-regulated categories), defining the GC-dependent transcriptome. C, Functional process analysis (DAVID; see Materials and Methods) revealed processes/pathways overrepresented by GC-dependent genes. For each overrepresented GO process, the number of GC-dependent genes identified for that process (#), and the overrepresentation P value (Overrep p-value) are shown. Underlined processes were added by the authors.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3713214&req=5

Figure 2: Microarray and pathway analysis. A, Filtering and statistical testing algorithm. Total probe sets were filtered to retain probe sets with at least 5 presence calls, and an A grade unique gene symbol level annotation. The retained 52% of probe sets were each tested by 1-way ANOVA (P ≤ .05) across the 4 treatment groups. The FDR for statistically significant genes is shown. B, Post hoc pairwise analysis (Fisher protected least significant difference, P ≤ .05) identified 393 genes that differed between low- and intermediate-CORT conditions (separated into up- and down-regulated categories), defining the GC-dependent transcriptome. C, Functional process analysis (DAVID; see Materials and Methods) revealed processes/pathways overrepresented by GC-dependent genes. For each overrepresented GO process, the number of GC-dependent genes identified for that process (#), and the overrepresentation P value (Overrep p-value) are shown. Underlined processes were added by the authors.
Mentions: Filtered genes were analyzed by 1-way ANOVA (Figure 2), and significant results were evaluated with post hoc all-pairwise Fisher protected least significant difference testing. Multiple testing error (false-positive results) was estimated by the false discovery rate (FDR = expected/observed) (71) procedure, and median FDRs are reported.

Bottom Line: Short-term CORT (4 days) did not recapitulate this transcriptome.We then compared the GC transcriptome with a previously defined hippocampal aging transcriptome, revealing a high proportion of common genes.These results contradict the hypothesis that GCs simply promote brain aging and also suggest that the opposite direction shifts during aging reflect resistance to CORT regulation.

View Article: PubMed Central - PubMed

Affiliation: Department of Molecular and Biomedical Pharmacology, University of Kentucky College of Medicine, Lexington, KY 40536, USA.

ABSTRACT
Although glucocorticoids (GCs) are known to exert numerous effects in the hippocampus, their chronic regulatory functions remain poorly understood. Moreover, evidence is inconsistent regarding the long-standing hypothesis that chronic GC exposure promotes brain aging/Alzheimer disease. Here, we adrenalectomized male F344 rats at 15 months of age, maintained them for 3 months with implanted corticosterone (CORT) pellets producing low or intermediate (glucocorticoid receptor-activating) blood levels of CORT, and performed microarray/pathway analyses in hippocampal CA1. We defined the chronic GC-dependent transcriptome as 393 genes that exhibited differential expression between intermediate and low CORT groups. Short-term CORT (4 days) did not recapitulate this transcriptome. Functional processes/pathways overrepresented by chronic CORT-up-regulated genes included learning/plasticity, differentiation, glucose metabolism, and cholesterol biosynthesis, whereas processes overrepresented by CORT-down-regulated genes included inflammatory/immune/glial responses and extracellular structure. These profiles indicate that GCs chronically activate neuronal/metabolic processes while coordinately repressing a glial axis of reactivity/inflammation. We then compared the GC transcriptome with a previously defined hippocampal aging transcriptome, revealing a high proportion of common genes. Although CORT and aging moved expression of some common genes in the same direction, the majority were shifted in opposite directions by CORT and aging (eg, glial inflammatory genes down-regulated by CORT are up-regulated with aging). These results contradict the hypothesis that GCs simply promote brain aging and also suggest that the opposite direction shifts during aging reflect resistance to CORT regulation. Therefore, we propose a new model in which aging-related GC resistance develops in some target pathways, whereas GC overstimulation develops in others, together generating much of the brain aging phenotype.

Show MeSH
Related in: MedlinePlus