Quantification of cytosine modifications in the aged mouse brain

Abstract Quantifying cytosine modifications in various brain regions provides important insights into the gene expression regulation and pathophysiology of neuropsychiatric disorders. In this study, we quantified 5‐methylcytosine (5‐mC), 5‐hydroxymethylation (5‐hmC), and 5‐formylcytosine (5‐fC) levels in five brain regions (the frontal lobe, cerebral cortical region without frontal lobe, hippocampus, basal ganglia, and the cerebellum) and the heart at three developmental periods (12, 48, and 101 weeks). We observed significant regional variations in cytosine modification. Notably, regional variations were generally maintained throughout development, suggesting that epigenetic regulation is unique to each brain region and remains relatively stable with age. The 5‐mC and 5‐hmC levels were positively correlated, although the extent of the correlations seemed to differ in different brain regions. On the contrary, 5‐fC levels did not correlate with 5‐mC or 5‐hmC levels. Additionally, we observed an age‐dependent decrease in 5‐fC levels in the basal ganglia, suggesting a unique epigenetic regulation mechanism. Further high‐resolution studies using animal models of neuropsychiatric disorders as well as postmortem brain evaluation are warranted.


| INTRODUC TI ON
Epigenetics is the study of the genetic regulation of gene expression in which DNA sequences are not altered.DNA methylation is a major form of DNA modification and contributes to long-lasting alterations in gene expression. 1 This process can be affected by environmental factors and has been considered to reflect gene-environment interactions. 2Numerous studies have suggested that DNA methylation plays an important role in brain development and the pathophysiology of neuropsychiatric disorders. 3,4though DNA methylation at the 5′ position of cytosine (5-methylcytosine: 5-mC) has been studied for decades, other modifications such as 5-hydroxymethylation (5-hmC) have recently attracted significant attention. 5,6In addition, researchers have identified multiple oxidized 5-mC variants, which are the products of sequential oxidation of 5-mC to 5-hmC, 5-hmC to 5-formylcytosine (5-fC), and 5-fC to 5-carboxylcytosine by Ten-eleven translocation (TET) enzymes. 7,8The functional significance of cytosine modifications has not yet been fully elucidated; however, it has been suggested that they serve a unique regulatory function in active DNA demethylation in the brain. 9,10 mammals, global 5-hmC content varies between tissues in the highest levels of the brain. 11,12In mice, 5-mC and 5-hmC content in brain tissue is increased during early postnatal development and stabilizes after approximately 10 weeks, [13][14][15][16] although timing differs between 5-mC and 5-hmC and between the cerebellum and cortex. 17 contrast, 5-fC levels drastically decrease after birth until postnatal Day 14, after which its level seems to stabilize in the mouse cerebral cortex. 18[21][22] Few studies have conducted systematic analyses of cytosine modifications of multiple brain regions at different developmental stages.
In this study, we quantified 5-mC, 5-hmC, and 5-fC levels in various brain regions of adult mice and focused on aging-related changes to these levels throughout development.

| Animals
Two pairs of male and female C57BL/6 mice, aged 12, 48, and 101 weeks, were used for the experiment.Genomic DNA was isolated from five brain regions (frontal lobes, Fl; cerebral cortical regions without the frontal lobe, Cr; hippocampus, Hp; basal ganglia, Bg; and cerebellum, Ce), and the heart.All experimental procedures involving animals were approved by the Animal Experiment Committees of RIKEN (Wako, Saitama, Japan) and were conducted in accordance with the Guidelines for Proper Conduct of Animal Experiments by the Science Council of Japan.

| Measurement of global 5-mC level
To quantify the global 5-mC level, the luminometric methylation assay (LUMA) 23 was used with slight modifications. 24

| Measurement of 5-hmC level
A dot blot assay was performed to calculate 5-hmC signal intensity.

| Measurement of global 5-fC level
The assay was performed using a MethylFlash 5-Formylcytosine DNA Quantification Kit (Epigentek, Farmingdale, NY, USA).In this assay, Fl, Cr, Bg, and Ce from 12-and 101-week aged mice have suggested distinct epigenetic regulation in these brain regions, such as altered TET expression in the NAc in response to rewards. 26,27ese results imply Bg-specific epigenetic regulation of 5-hmC to 5-fC conversion, and gene expression analysis including TET would provide new insights.
This study was constrained by limitations.First, the LUMA measured both 5-mC and 5-hmC levels, thus complicating the interpretation of the results.Second, we were unable to compare the levels of the three cytosine modifications in each brain region because we employed independent assays with different principles.Third, because we used large amounts of frozen brain tissue, the resolution of the analysis was low.Therefore, we were unable to assess the cytosine modification status in Cr, Hp, and Bg subregions, and cell-type differences were not considered.The latter is important because 5-hmC accumulates in neurons, 17 whereas 5-fC is more enriched in non-neuronal cells. 18Therefore, the identified age-related changes in 5-fC levels may be associated with altered composition of brain cell types.Fourth, with regard to sex, we compared the average 5-mC, 5-hmC, and 5-fC levels between males and females.We did not observe large differences in 5-mC or 5-fC levels.In Fl, we observed higher 5-hmC levels in females than in males.However, considering the limited number of samples with high interindividual variations, especially in 5-hmC, more studies using enough numbers of animals will be required.Finally, although our study focused on later developmental stage of the brain, it is also important to investigate at an early developmental stage which is a critical period for psychiatric disorders such as schizophrenia and autism.Further high-resolution studies using animal models of neuropsychiatric disorders or examinations of postmortem brains of patients are warranted.
Genomic DNA was digested with EcoRI (New England Biolabs, Beverly, MA, USA) and either HpaII (New England Biolabs) or MspI (New England Biolabs).The enzymes HpaII and MspI are isoschizomers that recognize the CCGG sequence and cleave CG.MspI is a CpG methylation-insensitive restriction endonuclease whereas HpaII is sensitive to CpG methylation.After digestion, the 5′-CG overhang was quantified by a luminometric polymerase extension assay using a PSQ 96MA instrument (QIAGEN, Hilden, Germany) according to the manufacturer's instructions.The relative amount of DNA methylation is expressed as the HpaII/MspI ratio, which indicates the amount of CpG methylation (5-mC and 5-hmC, not including 5-fC) in the context of CCGG.The amount of DNA used for each restriction cleavage reaction was 500 ng, and each assay was replicated once.In each assay, methylation level was determined based on the standard curves using the samples containing various amounts of 100% methylated and unmethylated genomic DNA using 5-Methylcytosine & 5-Hydroxymethylcytosine DNA Standard Set (Zymo Research, Irvine, CA, USA).

F I G U R E 1
Cytosine modification levels of adult mouse brain.(A) 5-mC, (B) 5-hmC, and (C) 5-fC levels.Left panels show comparisons of aging effect in each brain region.Right panels show comparisons across brain subregions and heart in each aging stage.In each group, four animals were used (two males and two females).Note that 5-mC level includes both 5-mC and 5-hmC.Mean ± SD is shown.ANOVA followed by Tukey test (p < 0.05) is indicated.*, **, and *** indicate that the heart showed significance in all brain subregions except for Bg, both Bg and Ce, and Ce, respectively.Bg, basal ganglia; Ce, cerebellum; Cr, cerebral cortical regions without frontal lobe; Fl, frontal lobes; Hp, hippocampus.