Brain Region- and Age-Dependent 5-Hydroxymethylcytosine Activity in the Non-Human Primate

Abstract

Because of the difficulty in collecting fresh brains of humans at different ages, it remains unknown how epigenetic regulation occurs in the primate brains during aging. In the present study, we examined the genomic distribution of 5hmC, an indicator of DNA methylation, in the brain regions of non-human primates (rhesus monkey) at the ages of 2 (juvenile), 8 (young adult), and 17 (old) years. We found that genomic 5hmC distribution was accumulated in the monkey brain as age increased and displayed unique patterns in the cerebellum and striatum in an age-dependent manner. We also observed a correlation between differentially hydroxymethylated regions (DhMRs) and genes that contribute to brain region-related functions and diseases. Our studies revealed, for the first time, the brain-region and age-dependent 5hmC modifications in the non-human primate and the association of these 5hmC modifications with brain region-specific function and potentially aging-related brain diseases.

Keywords: aging; brain region; epigenetics; neurodegenerative diseases; non-human primates.

Figure 1

Quantification and genomic mapping of 5hmC in four brain regions in rhesus monkeys. (A) 5hmC-specific dot blot revealed accumulation of 5hmC in cortex, cerebellum, hippocampus, and striatum of rhesus monkeys at 2, 8, and 17 years. Quantification of 5hmC-specific immunoblot by imageJ software in cortex, cerebellum, hippocampus, and striatum of rhesus monkeys. (n = 2, *p < 0.05; unpaired t-test, error bar = mean±SEM). (B) 5hmC distribution at different genomic components in 2-, 8-, and 17-year-old groups. (C) The ratio of genes with DhMRs to aging-associated genes (734) in different brain regions and the combined regions (totally). *p < 0.05, **p < 0.01, ***p < 0.001; unpaired t-test.

Figure 2

Unique 5hmC pattern in the cerebellum. (A) PCA result of all samples. (B) PCA results for 2-, 8-, and 17-year-old monkeys. (C) Identification and characterization of cerebellum-specific dynamic DhMRs. A total of 901 DhMRs identified as juvenile specific relative to young and old groups, 298 DhMRs identified as young specific relative to juvenile and old groups, and 2,836 DhMRs identified as old specific relative to juvenile and young groups. (D) Identification and characterization of cerebellum-specific stable DhMRs. A total of 364 DhMRs were identified as juvenile and young stable relative to the old group and 1,082 DhMRs were identified as young and old stable relative to the juvenile group. The total numbers of DhMRs are indicated in the upper right-hand corner of each figure.

Figure 3

Identification and characterization of striatum-specific DhMRs. Principal component analysis (PCA) results in the cortex, cerebellum, hippocampus (A), and striatum (B). (C) Dynamic 5hmC DhMRs in the striatum. A total of 4415 DhMRs were identified as 2 years-specific relatives to 8 years and 17 years groups, 3,988 DhMRs were identified as 8 years-specific relatives to 2 years and 17 years groups, and 24,619 DhMRs were identified as 17 years-specific relatives to 2 years and 8 years groups. (D) Stable 5hmC DhMRs in the striatum. A total of 9,910 DhMRs were identified as 2 years and 8 years stable relative to the 17 years group, and 446 DhMRs were identified as 8 years and 17 years stable relative to the 2 years group. The total numbers of DhMRs are indicated in the upper right-hand corner of each figure. (E) GO terms of 17-years-specific DhMRs enriched in cognition (left) and locomotory behavior (right) pathways. The y-axis lists the GO terms, and the x-axis indicates the significance p-value of enrichment.

Figure 4

DhMRs clustering. Five clusters were grouped according to significant differences, which consist of 1,658 DhMRs in Cluster 1, 1,558 DhMRs in Cluster 2, 817 DhMRs in Cluster 3, 433 DhMRs in Cluster 4, and 710 DhMRs in Cluster 5.

Figure 5

Kyoto Encyclopedia of Genes and Genomes (KEGG) terms of DhMRs from 5 clusters. The y-axis lists the KEGG terms, and the x-axis indicated the significance q-value of enrichment.

Figure 6

Sequential features of 5hmC at cerebellum (A), striatum (B), hippocampus (C), and cortex (D) during aging. Time-series analysis revealed 4 clusters in the cerebellum (A), 2 clusters in the striatum (B), 5 clusters in the hippocampus (C), and 4 clusters in the cortex (D) from young to old.

Figure 7

Cluster dendrograms of 5hmC among rhesus, human, and mouse. The 5hmC pattern of the rhesus monkey was closer to that of humans than mice.