NEW YORK (GenomeWeb) – A pair of studies appearing in Nature Neuroscience yesterday uncovered epigenetic changes associated with the development of Alzheimer's disease.
For the studies, the two research teams conducted epigenome-wide association studies to search for DNA regions that were differentially methylated in people who had Alzheimer's disease as compared to those without the disease. While the researchers uncovered nearly a dozen differentially methylated regions, both groups homed in on a hypermethylated region at the ANK1 gene.
"This is the strongest evidence yet to suggest that epigenetic changes in the brain occur in Alzheimer's disease, and offers potential hope for understanding the mechanisms involved in the onset of dementia," Jonathan Mill from the University of Exeter Medical School and the senior author of one of the studies said in a statement. "We don't yet know why these changes occur — it's possible that they are involved in disease onset, but they may also reflect changes induced by the disease itself."
Alzheimer's disease affects some 26 million people worldwide, a figure that is likely to grow as the population ages, but much about how the condition develops and why it affects some parts of the brain and not others is unclear.
In one study, Rush University Medical Center's David Bennett and colleagues examined 708 prospectively collected autopsied brains from the Religious Order Study and the Memory and Aging Project to examine the methylation state of DNA in Alzheimer's brains.
Using the Illumina HumanMethylation 450 BeadChip, they examined the methylation status at more than 416,000 CpG sites in tissue samples taken from the dorsolateral prefrontal cortex of each individual's brain.
From this, the researchers identified 137 CpGs with methylation levels that correlated with the level of neurolytic amyloid plaques, a measure of Alzheimer's disease pathology. Seventy-one of those CpGs, located in 60 discrete differentially methylated regions, were replicated in an independent set of samples.
The researchers also evaluated whether methylation differences at these 71 CpG sites were associated with another measure of AD pathology called Braak staging in another cohort of 117 subjects.
Based on this, the researchers validated 12 CpG sites located in 11 differentially methylated regions.
Functional analysis of those regions revealed that seven genes in the vicinity of those CpGs had expression level changes associated with Alzheimer's disease onset: ANK1, CDH23, DIP2A, RHBDF2, RPL13, SERPINF1, and SERPINF2. Six of these genes are connected, the researchers added, to a known Alzheimer's disease susceptibility network.
"[We] found several replicated, functionally validated associations between altered DNA methylation and the presymptomatic accumulation of AD pathology," Bennett and his colleagues wrote in their paper.
Meanwhile, in the second study Exeter's Mill and his colleagues examined the methylation state in a number of different brain tissues from people with and without Alzheimer's disease.
Drawing on tissue samples from the parts of the brain known to be affected by Alzheimer's disease, samples from regions that are not affected, and premortem blood samples from some of the participants, Mill and his colleagues searched for methylation changes in the genome related to disease onset.
Using the Illumina 450K HumanMethylation assay, they searched for differentially methylated regions that were associated with Braak staging of the disease. Two of the top-ranked regions in the entorhinal cortex — which is affected by the disease — were located less than 100 base pairs from each other in the ANK1 gene, they noted.
ANK1, which the Bennett team also found, encodes a protein that's expressed in the brain and is involved in the compartmentalization of the neuronal plasma membrane.
That differentially methylated stretch of the genome, the researchers found, was also differentially methylated in two other cortical regions — the superior temporal gyrus and the prefrontal cortex — though not in the cerebellum, which isn't affected by Alzheimer's disease, and not in blood samples taken from some of the patients before they died.
The researchers confirmed these findings in a separate cohort using bisulfite sequencing.
"We identified evidence for cortex-specific hypermethylation at CpG sites in the ANK1 gene associated with AD neuropathology," Mill and his colleagues said in their paper.
"[O]ur observation of highly consistent changes across multiple regions of the cortex in several independent sample cohorts suggests that the identified loci are directly relevant to the pathogenesis of AD," they added.
Additionally, Mill and colleagues noted that some methylations changes — though not in ANK1 — were apparently in the premortem blood samples, and Bennett and his colleagues similarly reported that some of the methylation changes they spotted appeared to begin early in the development of Alzheimer's disease-related cognitive decline.
"Going forward in the aging brain, we clearly need to more precisely map the alterations of chromatin structure that contribute to AD pathophysiology and to assess, using model systems, whether remodeling the epigenome is a fruitful goal for the development of AD therapies," Bennett and his colleagues said.