OUP user menu

DNA Methylation of Alzheimer Disease and Tauopathy-Related Genes in Postmortem Brain

Marta Barrachina PhD, Isidre Ferrer MD, PhD
DOI: http://dx.doi.org/10.1097/NEN.0b013e3181af2e46 880-891 First published online: 1 August 2009

Abstract

DNA methylation occurs predominantly at cytosines that precede guanines in dinucleotide CpG sites; it is one of the most important mechanisms for epigenetic DNA regulation during normal development and for aberrant DNA in cancer. To determine the feasibility of DNA methylation studies in the postmortem human brain, we evaluated brain samples with variable postmortem artificially increased delays up to 48 hours. DNA methylation was analyzed in selected regions of MAPT, APP, and PSEN1 in the frontal cortex and hippocampus of controls (n = 26) and those with Alzheimer disease at Stages I to II (n = 17); Alzheimer disease at Stages III to IV (n = 15); Alzheimer disease at Stages V to VI (n = 12); argyrophilic grain disease (n = 10); frontotemporal lobar degeneration linked to tau mutations (n = 6); frontotemporal lobar degeneration with ubiquitin-immunoreactive inclusions (n = 4); frontotemporal lobar degeneration with motor neuron disease (n = 3); Pick disease (n = 3); Parkinson disease (n = 8); dementia with Lewy bodies, pure form (n= 5); and dementia with Lewy bodies, common form (n = 15). UCHL1 (ubiquitin carboxyl-terminal hydrolase 1 gene) was analyzed in the frontal cortex of controls and those with Parkinson disease and related synucleinopathies. DNA methylation sites were very reproducible in every case. No differences in the percentage of CpG methylation were found between control and disease samples or among the different pathological entities in any region analyzed. Because small changes in methylation of DNA promoters in vulnerable cells might have not been detected in total homogenates, however, these results should be interpreted with caution, particularly as they relate to chronic degenerative diseases in which small modifications may be sufficient to modulate disease progression.

Key Words
  • ADORA2A
  • APP
  • DNA methylation
  • MAPT
  • PSEN1
  • RAGE
  • UCHL1

Introduction

DNA methylation is one of the most important mechanisms for epigenetic silencing in mammals; it occurs predominantly at cytosines that precede guanines in dinucleotide CpG sites (1). Approximately half of the human gene promoters contain CpG-rich regions with lengths of 0.5 kb to several kb known as CpG islands. Regions that are actively transcribed generally have promoter regions with predominantly unmethylated CpG islands, whereas transcriptionally silent regions have abundant methylated CpG sites (2). Most CpG islands are located in the 5′ UTR regions and the first exon (3). DNA methylation is a normal process that occurs in mammalian embryonic development, X-chromosome inactivation, and repression of proviral genes and endogenous transposons (1). Methylation of CpG islands is usually accompanied by posttranslational histone modifications that modulate gene expression (4, 5). For example, MeCP2 is a methyl CpG-binding protein that recruits members of chromatin remodeling complexes to repress gene transcription (6). Several mutations in MeCP2 genes have been reported in Rett syndrome, one of the most common genetic causes of mental retardation in females (7).

CpG islands are usually unmethylated in normal cells (8), whereas hypermethylation of CpG islands is a major event in many cancers (9). The role of DNA methylation in the brain is an emerging field of scientific analysis. It was recently reported that DNA methylation signatures exist for every cerebral region (10) and that neuronal DNA methylation is modified with life span (11).

The aims of this study were to 1) determine whether postmortem delay between death and tissue processing may affect CpG methylation, thus hampering further studies in the human postmortem brain; and 2) study DNA methylation in 5′ UTR and/or intronic regions of genes related to Alzheimer disease (AD) and other tauopathies, including Pick disease (PiD), argyrophilic grain disease (AGD), and frontotemporal lobar degeneration (FTLD) linked to mutations in tau (FTLD-tau). For the first purpose (and in line with previous studies focused on protein, RNA, and DNA preservation in postmortem carried out within the context of the European Brain Bank network [BrainNet Europe II] 12-14), we analyzed RAGE (advanced glycation end product receptor), ADORA2A (adenosine A2A receptor), and MAPT (tau) genes in brain samples with short and artificially prolonged postmortem delays. For the second purpose, CpG methylation in selected regions of MAPT, PSEN1 (presenilin 1 gene), and APP (β-amyloid precursor protein gene) were analyzed in the frontal cortex and hippocampus in AD cases at different stages of disease progression, AGD, and PiD. This was carried out in parallel with the study of age-matched controls and cases with other neurodegenerative diseases, including Parkinson disease (PD); dementia with Lewy bodies, pure and common forms (DLBp and DLBc); FTLD with ubiquitin- and TDP-43-immunoreactive inclusions not associated with motor neuron disease (FTLD-U-TDP-43); and FTLD associated with motor neuron disease (FTLD-MND). UCHL1 (ubiquitin carboxyl-terminal hydrolase 1 gene) was analyzed in the frontal cortex in PD and related synucleinopathies.

Materials and Methods

Case Material

To document variations in DNA methylation profiles with postmortem delay, we extracted genomic DNA from the frontal cortex of 5 different subjects obtained after a short postmortem delay and immediately frozen or stored at 4°C for 3, 6, 15.5, 24, or 48 hours and then frozen to mimic variable postmortem delay in tissue processing (Table 1).

View this table:
TABLE 1.

Cases studied for CpG methylation in MAPT, PSEN1, APP, and UCHL1 were as follows: controls (n = 26), PD Stages 3 and 4 (n = 8), DLBp (n = 5), DLBc (n = 15), AD Stages I to II (n = 17), AD Stages III to IV (n = 15), AD Stages V to VI (n = 12), AGD (n = 10), FTLD-tau (n = 6), FTLD-U (n = 4), FTLD-MND (n = 3), and PiD (n = 3). The total number of cases was 124, with 75 males and 49 females. Postmortem delay was 2 to 20 hours. The neuropathologic diagnoses were made according to well-established criteria for AD (15, 16), AGD (17), FTLD-tau (18), FTLD-U and FTLD-MND (19, 20), PD (21), DLB (20-23), and PiD (24, 25). Clinical and neuropathologic data are summarized in Table 2.

View this table:
TABLE 2.

Cell Culture

HeLa and SH-SY5Y cells were maintained in Dulbecco minimal essential medium (Invitrogen, El Prat de Llobregat, Spain) supplemented with 10% fetal bovine serum and 2 mmol/L L-glutamine. Both cell lines were grown at 37°C in a humidified atmosphere of 5% carbon dioxide.

DNA Purification and Bisulfite Treatment

Genomic DNA from human postmortem frozen brain samples was purified using DNeasy Tissue kit (Qiagen, Las Matas, Madrid, Spain) following the indications of the supplier. Bisulfite DNA treatment was performed using EZ DNA Methylation kit (Zymo Research, Ecogen, Barcelona, Spain). One microgram of genomic DNA for every sample was mixed with 5 μL of M-dilution buffer in a final volume of 50 μL and incubated at 37°C for 15 minutes. After incubation, 100 μL of prepared CT conversion reagent was added to each sample. The tubes were then incubated in the dark with a thermal cycler (GeneAmp PCR System 9700, Applied Biosystems, Madrid, Spain) as follows: 20 cycles at 95°C for 30 seconds and 50°C for 15 minutes followed by a last hold at 4°C for 10 minutes (the total duration is approximately 5.5 hours). The samples were then mixed with 400 μL of M-binding buffer and loaded into a Zymo-Spin I column and centrifuged (≥10,000 × g) for 30 seconds. The columns were washed with 200 μL of M-wash buffer and centrifuged at full speed for 30 seconds. After this, 200 μL of M-desulphonation buffer was added at room temperature for 15 minutes. After incubation, the columns were centrifuged at full speed for 30 seconds and then washed twice with 200 μL of M-wash buffer at full speed for 30 seconds and 1 minute. A third centrifugation without buffer was performed to remove wash buffer residues. Finally, 100 μL of water was added to the column, and DNA was eluted in a new tube after a centrifugation at 3,000 × g for 30 seconds.

Quantitative DNA Methylation Analysis

DNA regions surrounding transcriptional start sites of RAGE, ADORA2A, UCHL1, MAPT, PSEN1, and APP (26-33) were analyzed. Primers for each region were designed using MethPrimer (http://www.urogene.org/methprimer/). Each reverse primer presented a T7 promoter tagged to obtain an appropriate product for in vitro transcription and an 8-bp insert to prevent abortive cycling. The forward primers contain a 10-mer tag to balance the polymerase chain reaction (PCR) primer length. The sequence of each primer used for amplification of bisulfite-treated DNA (tags incorporated are labeled in lower case and underlined) is indicated as follows:

RAGE: forward, 5′-aggaagagagAGAGTGGGGAATTTTTTTTATTAAAG-3′; reverse, 5′-cagtaatacgactcactatagggagaaggctCACCCCTAAATACTACCAACCTCTA-3′;

ADORA2A: forward, 5′-aggaagagagTTAGTTAGGTAGAGGAGTAGGTGGG-3′; reverse, 5′-cagtaatacgactcactatagggagaaggctCCACTCTAAACTCAAAACCAAAAAT-3′;

MAPT-640/-294: forward, 5′-aggaagagagTGTAATTGAGTTAGTTTGTTTTAAGT-3′; reverse, 5′-cagtaatacgactcactatagggagaaggctCCTCCTATAATTAAAATCTTTATATC-3′;

MAPT+1411/+1978: forward, 5′-aggaagagagTTTTTTGTTTTGTTTGTAGAGGTTA-3′; reverse, 5′-cagtaatacgactcactatagggagaaggctCACCTATAATTTCCATAACAATCCC-3′;

MAPT+54336/+54905: forward, 5′-aggaagagagttTTtggtggtgTagaaTaggagaT-3′; reverse, 5′-cagtaatacgactcactatagggagaaggctAAATCCAAAAACTCAAAAAAAACTC-3′;

PSEN1: forward, 5′-aggaagagagTGGGTTTAATTTATATAGGGGTTTT-3′; reverse, 5′-cagtaatacgactcactatagggagaaggctTAACTCAAATTCCTTCCAAACCA-3′;

APP-526/-234: forward, 5′-aggaagagagTTGTTGTTTTAATAAGTAAAGAAAATTTTA-3′; reverse, 5′-cagtaatacgactcactatagggagaaggctAAAAAAAATCTAAAACCAAAAAAAA-3′.

APP-2572/-2108: forward, 5′-aggaagagagTTTGATTAGGGAATGTGTTAGTGTT-3′; reverse, 5′-cagtaatacgactcactatagggagaaggctCTCCAAAATTACATACCCATAAAAC-3′.

UCHL1: forward, 5′-aggaagagagGTTTAAAATTAAAGATTTTATTAAAAGGAT-3′; reverse, 5′-cagtaatacgactcactatagggagaaggctATCTAAAAAACAAATACAAAAAAAA-3′.

For PCR amplification, 2 μL of bisulfite-treated DNA was used as a template, along with 1 μL of each primer 1 μmol/L, 0.04 μL dNTPs 25 mmol/L, 0.42 μL water, 0.5 μL 10× Hot Star buffer and 0.04 μL of Hot Star Taq Flexi polymerase (Qiagen). The reaction was carried out using the following parameters: 94°C for 15 minutes and 45 cycles of 94°C for 20 seconds, annealing temperature of every set of primers for 30 seconds (Table 3), 72°C for 1 minute, and a last hold at 72°C for 3 minutes. Then, 0.5 μL of every PCR product was checked in 1.5% agarose gel to confirm successful PCR amplification. The rest of the PCR product of each sample was sent to SEQUENOM (Hamburg, Germany) to be analyzed using the MassArray System platform (34, 35). This technology consists of the analysis of DNA methylation by gene-specific amplification of bisulfite-treated DNA followed by in vitro transcription, base-specific cleavage, and matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF) analysis without cloning of the PCR products. Because CpG sites that are very close to each other cannot be differentiated in the MALDI-TOF analysis, the percentage of DNA methylation is considered to be the same for each.

View this table:
TABLE 3.

UCHL1 methylation levels were quantified in non-UCHL1-expressed HeLa cells (28) versus UCHL1-expressed SH-SY5Y cells as positive and negative controls of the reaction (see Supplemental Figure 1, Supplemental Digital Content 1, http://www.jnen.oxfordjournals.org/lookup/suppl/doi:10.1097/NEN.0b013e3181af2e46/-/DC1).

Statistical Analysis

Statgraphics Plus v5 software was used for statistical analysis.

Results

DNA Methylation Is Preserved in Human Postmortem Brain Samples

The 4 CpG sites quantified in RAGE were highly methylated (especially CpG Site Numbers 4 and 5); this methylation profile was not modified with artificially prolonged postmortem delay up to 48 hours (Fig. 1A). Similar results were obtained for all 13 CpG sites in ADORA2A (Fig. 1B). The region analyzed, 5′ upstream region of Exon 0 in MAPT, which was predicted as a CpG island by MethPrimer software, was lightly methylated, and no increase in methylcytosine content was observed with a progressive postmortem delay (Fig. 1C).

FIGURE 1.

DNA methylation is preserved in human postmortem brain samples. DNA sequences in gene promoters of RAGE (A), ADORA2A (B), and MAPT (C) show corresponding CpG sites in gray boxes. The graphs represent the percentages of DNA methylation (mean ± SD) of each CpG site amplified by polymerase chain reaction of bisulfite-treated DNA samples. Samples are those shown in Table 1 that had been subjected to artificially increased postmortem delay for 6, 15.5, 24, or 48 hours after autopsy performed between 2 and 5 hours after death. No significant differences are seen in the methylation levels of each specific site at different time points of postmortem delay up to 48 hours.

Analysis of DNA Methylation in Noncoding Regions of Genes Related With AD and Other Tauopathies

MAPT

The percentages of DNA methylation were analyzed in 3 MAPT regions, a locus surrounding Exon 0 (Fig. 2) and 2 intronic loci (Figs. 3, 4). The analysis revealed low levels of DNA methylation in the 5′ upstream flanking region to Exon 0, only detecting around 20% to 40% of DNA methylation for CpG Sites 5, 6, and 13. No differences in the frontal cortex were detected between AD and tauopathies and age-matched control samples. Moreover, no differences were seen between these groups and other degenerative diseases, including FTLD-U, FTLD-MNS, PD, and DLB. Finally, no differences were observed in the hippocampus among different stages of AD, AGD, and age-matched controls (Fig. 2). The analysis of the other 2 MAPT regions in the intronic region close to and distant from Exon 0 was restricted to AD, tauopathies, and FTLD, and to AD and tauopathies, respectively. The analysis of the 3′ downstream flanking region to Exon 0 (positions at +1411/+1978 and at +54336/+54905) presented a technical problem because base-specific cleavage of the in vitro transcription product was not discriminated by MALDI-TOF, obtaining the same percentage of DNA methylation for CpG Site Numbers 3 and 7, CpG site Numbers 15 to 17 and 33 (Fig. 3), and CpG Site Numbers 1 and 19 (Fig. 4). The global DNA methylation profile for these 2 loci is similar to that observed in the 5′ upstream flanking region to Exon 0, including low levels of DNA methylation and no differences between control and diseased cases in the frontal cortex and hippocampus. The only difference occurred at Positions 1 and 19 in the +54336/+54905 region in PiD, where increased methylation was observed in comparison with controls (p < 0.01, analysis of variance with post hoc Scheffé test) (Fig. 4).

FIGURE 2.

DNA methylation analysis of MAPT region −640/−294. (A) Schematic representation of the location of the bisulfite-sequencing region analyzed. The bp annotations are relative to the transcriptional start site. (B) Graphs represent the percentage of DNA methylation (mean ± SD) of each CpG site located in the locus amplified by polymerase chain reaction of bisulfite-treated DNA samples. The analysis was carried out in the frontal cortex of controls (C) and those with Parkinson disease (PD), dementia with Lewy bodies pure (DLBp) and common forms (DLBc), Alzheimer disease at different Braak stages (ADI-II, ADIII-IV, ADV-VI), argyrophilic grain disease (AGD), frontotemporal lobar degeneration associated with mutations in the tau gene (FTLD-tau), and FTLD associated with motor neuron disease (FTLD-MND); and in the hippocampus in AD and AGD cases. No differences are seen among control and diseased cases.

FIGURE 3.

DNA methylation analysis of the intronic region +1411/+1978 of MAPT. (A) Schematic representation of the location of the bisulfite-sequencing region analyzed. The bp annotations are relative to the transcriptional start site. (B) The graphs represent the percentage of DNA methylation (mean ± SD) of each CpG site located in the locus amplified by polymerase chain reaction of bisulfite-treated DNA samples. The analysis was carried out in the frontal cortex of controls and those with Alzheimer disease at different Braak stages (AD) I to II, AD III to IV, AD V to VI, argyrophilic grain disease (AGD), Pick disease (PiD), frontotemporal lobar degeneration associated with mutations in the tau gene (FTLD-tau), FTLD with ubiquitin-immunoreactive inclusions (FTLD-U), and FTLD associated with motor neuron disease (FTLD-MND); and in the hippocampus in AD and AGD cases. No differences are seen among control and diseased cases.

FIGURE 4.

DNA methylation analysis of the intronic region +54336/+54905 of MAPT. (A) Schematic representation of the location of the bisulfite sequencing region analyzed. The bp annotations are relative to the transcriptional start site. (B) The graphs represent the percentage of DNA methylation (mean ± SD) of each CpG site located in the locus amplified by polymerase chain reaction of bisulfite-treated DNA samples. The analysis was performed in the frontal cortex of controls (C) and those with Alzheimer disease at different Braak stages (AD) I to II, AD III to IV, AD V to VI, argyrophilic grain disease (AGD), and Pick disease (PiD), and hippocampus in C, AD I to II, AD III to IV, and AD V to VI cases. Significant differences are only seen in Sites 1 and 19 in PiD.

PSEN1

DNA methylation levels in the regions of PSEN1 analyzed were very low, and no differences were seen among controls and diseased cases in frontal cortex and hippocampus. Cases analyzed included AD, tauopathies, FTLD-U, FTLD-MNS, PD, DLBp, DLBc, and controls (Fig. 5).

FIGURE 5.

DNA methylation analysis of the PSEN1 region at the transcriptional start site. (A) Schematic representation of the location of the bisulfite sequencing region analyzed covering the region −228/+101 of PSEN1. The bp annotations are relative to the transcriptional start site. (B) Graphs represent the percentages of DNA methylation (mean ± SD) of each CpG site located in the locus amplified by polymerase chain reaction of bisulfite-treated DNA samples. The analysis was performed in the frontal cortex of controls (C) and those with Parkinson disease (PD), dementia with Lewy bodies pure and common forms (DLBp and DLBc), Alzheimer disease at different Braak stages (AD) I to II, AD III to IV, AD V to VI, argyrophilic grain disease (AGD), Pick disease (PiD), frontotemporal lobar degeneration associated with mutations in the tau gene (FTLD-tau), FTLD with ubiquitin-immunoreactive inclusions (FTLD-U), and FTLD associated with motor neuron disease (FTLD-MND), and in the hippocampus in AD I to II, AD III to IV, AD V to VI, and AGD. No differences are seen among control and diseased cases.

APP

Low CpG methylation in the APP gene promoter region close to the transcriptional start site occurred in control and diseased cases. No significant differences were seen among AD cases and cases with Lewy body-associated pathology (the only ones examined for APP); these did not differ from percentages of methylation encountered in controls (Fig. 6). In contrast, a high percentage of CpG methylation was present in a distal 5′ region of APP promoter. Again, no differences were seen among AD at different stages and Lewy body diseases (PD, DLBp, and DLBc) or between diseased cases and controls (Fig. 6).

FIGURE 6.

DNA methylation analysis of APP regions proximal and distal to the transcriptional start site. (A) Proximal and (B) distal regions of the transcriptional start site were examined. Schematic representations of the location of the bisulfite sequencing region analyzed are illustrated with the bp annotations. The graphs represent the percentage of DNA methylation (mean ± SD) of each CpG site located in the locus amplified by polymerase chain reaction from bisulfite-treated DNA samples. The analysis was performed in the frontal cortex of controls (C) and those with Parkinson disease (PD), dementia with Lewy bodies pure and common forms (DLBp and DLBc), Alzheimer disease at different Braak stages (AD) I to II, AD III to IV, and AD V to VI. These sites have very different levels of CpG methylation. CpG sites in the region −526/−234 show low levels of methylation, whereas very high levels of methylation occur in the −2572/−2108 region. Yet no differences between controls and disease cases and among different pathological phenotypes are observed in the 2 regions.

UCHL1

A low percentage of DNA methylation was observed in controls. This pattern was not modified in PD, DLBp, and DLBc, although there were individual variations in Sites 6 to 8 and 16 to 18 (Fig. 7).

FIGURE 7.

DNA methylation analysis of UCHL1 gene in α-synucleinopathies. DNA sequence amplified (−251/+218) in UCHL1 showing CpG sites. (A) Each CpG site is indicated with gray boxes. (B) The graph represents the percentage of DNA methylation of each CpG site. DNA methylation analysis (mean ± SD) of UCHL1 was carried out in the frontal cortex in controls (C) and those with Parkinson disease (PD) and dementia with Lewy bodies pure (DLBp) and common forms (DLBc). No differences between controls and disease cases are observed.

Discussion

MassArray platform has been used to determine the percentage of DNA methylation in selected loci of gene promoters related to AD and other tauopathies. This technology consists of the analysis of DNA methylation by gene-specific amplification of bisulfite-treated DNA followed by in vitro transcription, base-specific cleavage, and MALDI-TOF without cloning of the PCR products. Although this is a robust method, it is worth stressing that CpG sites very close to each other cannot be differentiated in the MALDI-TOF analysis, and the percentage of DNA methylation is then considered to be the same for both. The MassArray technique does not evaluate non-CpG methylation (34, 35).

The genes for the study of possible effects of postmortem delay on CpG methylation were selected for several reasons. On 1 hand, DNA methylation in the RAGE promoter seems to be reduced with age (36), whereas the region analyzed in ADORA2A is heavily methylated. Finally, the region analyzed in MAPT is poorly methylated, and the selection of this gene was within the context of the study of genes-the encoded proteins of which are associated with AD. No modifications in CpG methylation with artificially increased postmortem delay up to 48 hours were observed in the regions examined in RAGE, ADORA2A, and MAPT. Recent studies have also shown that postmortem delay does not affect methylation of histone tails (37).

MAPT contains 16 exons and is devoid of TATA or CAAT boxes. A CpG island encompasses Exon 0 and spans more than 3 kb (29); 11 CpG islands are present in the adjacent Intron 0 (30). Previous studies have shown a decrease with age in the total number of methylcytosines in the 5′ flanking region close to Exon 0 in the human parietal cortex (38). Because 3 CpG islands analyzed here (one located upstream from Exon 0 and the other 2 in Intron 0) show a low percentage of DNA methylation in control and disease cases, this is difficult to assess. Interestingly, the same patterns were observed in control and disease cases, and similar methylation sites occurred in the frontal cortex and hippocampus. Moreover, no differences were seen among the different stages in AD or among AD, AGD, FTLD-tau, FTLD-U, FTL-MND, PD, DLBp, and DLBc. The only exception was PiD, in which increased DNA methylation was noted in CpG Sites 1 and 19. These results must be critically examined, however, first because the number of PiD cases was small, and second because in silico analysis did not reveal any putative binding for a transcription factor candidate in these particular sites as revealed by bioinformatic analysis with MatInspector software (see Supplemental Figure 2, Supplemental Digital Content 2, http://www.jnen.oxfordjournals.org/lookup/suppl/doi:10.1097/NEN.0b013e3181af2e46/-/DC1).

Recent studies have shown that PSEN1 is hypomethylated (11), and that no variations in DNA methylation are found in AD brains (39). We corroborate these findings at the different stages in AD, and we point out that modifications are also not observed in the other tauopathies, that is, FTLD-U and FTLD-MND, or in PD, DLBp, and DLBc.

APP is also GC rich, contains multiple initiation sites, and lacks a TATA box (33). Previous studies have shown a reduction in methylcytosines in the promoter region of APP with age, suggesting that demethylation may have some role in β-amyloid deposition in the aged brain (36, 40). These findings have not been replicated using a more sensitive technique in a larger number of cases, however (39). The present findings show no differences in the percentages of CpG methylation sites among different pathological conditions and age-matched controls.

Finally, a previous study revealed 35 CpG sites in the UCHL1 promoter, spanning the putative transcription start site and Exons 1 and 2; the gene promoter was fully methylated in non-UCHL1-expressed HeLa cells (28). Our study revealed that the UCHL1 promoter region has very low levels of DNA methylation in control samples. This finding correlates with the abundance of UCHL1 in the brain, constituting up to 2% of total protein (41). Inclusion of the study of UCHL1 in the present context was because previous studies demonstrated a reduced expression of UCHL1 mRNA and protein levels in the cerebral cortex in DLB (42), thus making CpG methylation at the UCHL1 promoter a putative mechanism of reduced UCHL1 mRNA expression and, therefore, a good candidate gene for comparative purposes. No differences in CpG methylation sites were observed in DLB cases when compared with controls.

Wang et al (39) recently reported 12 potential AD low-methylated loci in late-onset AD. Similar sites are observed in other neurodegenerative diseases including AGD, PiD, FTLD-tau, FTLD-U, FTLD-MND, PD, DLBp, and DLBc, indicating that putative sites are not exclusive to AD but rather are common in control and disease cases. Importantly, low methylation sites are susceptible to drug intervention. In this regard, putative binding sites to Sp1 and Ets family factors, such as Elk1, occur in the low-methylated −118/+178 region of PSEN1 (43, 44). Therefore, methylation by specific drugs might reduce the expression of PSEN1 and eventually curve β-amyloid deposition in AD. Indeed, S-adenosylmethionine administration in cell lines downregulates PSEN1 and reduces β-amyloid production and is thus a putative candidate for the treatment of AD (45, 46). Moreover, deprivation of S-adenosylmethionine upregulates PSEN1 and increases β-amyloid deposits in APP transgenic mice (47).

It is worth noting that the MAPT loci analyzed in the present study contain binding sites for Sp1, upstream stimulatory factor (USF1), XBP-1, ZNF219, heat-shock transcription factor 2, FAP-2, MAZ, Elk1, AP1, and NFKβ, and that all of these putative binding sites are hypomethylated. All of these transcription factors regulate tau expression (48-52) or have been implicated in brain function (53-58). ZNF219 is a member of the Kruppel-like zinc finger family that acts as a repressor (59), but its function in the central nervous system is not known.

The present findings in this very large number of cases and regions have shown the following: 1) preservation of CpG methylation of gene promoters with postmortem delay, thus enabling the study of DNA methylation in human postmortem brain; 2) highly reproducible methylation sites for a given gene in control and diseased cases in the human frontal cortex and hippocampus; and 3) a lack of significant modifications in the methylation state of selected loci in disease cases when compared with controls and among various pathological entities.

It is important to stress, however, that all of these studies were carried out in total homogenates of only relatively circumscribed areas that include presumed normal and damaged neurons and different populations of glial and other cells. Therefore, small changes in methylation of DNA promoters might have gone undetected. This caveat is particularly important in neurodegenerative diseases because they are long-lasting disorders in which small methylation changes might be sufficient to modulate their progression.

Acknowledgments

The authors thank Jesús Moreno and Salvador Juvés for excellent technical support in genomic DNA extraction, bisulfite DNA treatment, and analysis of PCR products in agarose gels. The authors also thank Dr S. Boluda for help in tissue sampling and T. Yohannan for editorial assistance.

Footnotes

  • This work was funded by the European Commission under the Sixth Framework Programme (BrainNet Europe II, LSHM-CT-2004-503039) and by FIS grants from the Spanish Ministry of Health, Instituto de Salud Carlos III (PI05/1631 and PI08/0582).

  • Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Web site (http://jnen.oxfordjournals.org/).

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
View Abstract