Alzheimer’s Disease Linked to Loss of Epigenetic Protection

An international team of researchers has linked a gene known as peptidase M20-domain-containing protein 1 (PM20D1) with Alzheimer’s disease (AD) in humans. An analysis of genetic and epigenetic study data, combined with data from human tissue samples and subsequent experiments in mouse models, indicated that increased levels of PM20D1 in the brain have a protective effect against AD. Genetically boosting PM20D1 expression in AD mice reduced AD-related pathologies, whereas genetically depleting PM20D1 promoted disease progression.

The collaborating teams in Switzerland and Spain claim their findings suggest that in a particular genetic background PM20D1 contributes to neuroprotection against AD. “This variation is associated with the loss of activity of a neuroprotective gene called PM20D1; whoever possesses the variation has a greater probability of suffering from AD, so people carrying these variants could be excellent candidates for clinical prevention trials of the disease in the future,” comments co-researcher Manel Esteller, M.D., Ph.D., director of the Epigenetics and Cancer Biology Program of the Bellvitge Biomedical Research Institute (IDIBELL), ICREA researcher, and professor of genetics of the University of Barcelona.

Dr. Esteller’s team and collaborators, including researchers working with Johannes Gräff, Ph.D., at the Brain Mind Institute, Faculty of Life Sciences, École Polytechnique Fédérale de Lausanne, in Switzerland, describe their studies in a paper in Nature Medicine, entitled “PM20D1 Is a Quantitative Trait Locus Associated with Alzheimer’s Disease.”

Both genetic and nongenetic risk factors impact on the likelihood of developing AD, “the latter likely being mediated by epigenetic mechanisms,” the researchers write. “Over the last decade, the number of epigenomic studies for AD has rapidly increased, which has been paralleled by the discovery of an increasing number of epigenetically dysregulated genes.” However, the researchers point out, only few genes, including ankyrin 1 (ANK1), sorbin and SH3-domain-containing 3 (SORBS3), and histone deacetylase 2 (HDAC2), have been reported by independent studies in humans.

More recently, a combination of epigenetic and genetic approaches has made it possible to identify and link single-nucleotide polymorphisms (SNPs) with changes in DNA methylation levels, “so-called methylation quantitative trait loci (mQTLs) – the importance of which has just started to be recognized for complex diseases,” the authors continue. But while mQTLs have been described for neurological disorders, including schizophrenia and bipolar disorders, until now they haven’t been reported for neurodegenerative disorders such as AD.

Through an analysis of two recent human epigenome-wide association studies (EWAS) carried out by the Swiss team, the researchers initially highlighted an association between variations in the gene PM20D1 and AD. “When we reanalyzed these biologically and technically independent datasets by focusing exclusively on the comparison between samples from healthy controls and patients with advanced-stage AD, we noted that only one gene, PM20D1, which was previously described as a mQTL, consistently displayed promoter hypermethylation in patients with AD (hereafter referred to as AD samples) than in healthy controls in both studies,” they note.

“Over the last seven years, we have created a detailed map of the epigenetic alterations that occur in the brain of people affected by AD and other dementias, such as those associated with the so-called Lewy bodies or Parkinson’s disease,” explains Dr. Esteller.

“That allowed us to collaborate with Dr. Johannes Gräff’s group in Lausanne, who noticed how one of the molecular lesions we had discovered was caused by inheriting a variation in the DNA sequence.”

Further analysis of publicly available DNA methylation data also identified PM20D1 hypermethylation in AD patients. Subsequent SNP studies indicated that PM20D1 hypermethylation is linked with AD in individuals who carry two specific SNPs, rs708727 and rs960603, which are both found in nearly 85% of AD cases. Genetic analysis of human brain cortex samples showed an allele-dependent correlation between the double SNP haplotype and PM20D1 DNA methylation.

Additional in vitro studies then demonstrated that PM20D1 expression increased in neuroblastoma cells treated with reactive oxygen species (ROS) and amyloid-β (Aβ), while an analysis of brain tissue in a mouse model of AD showed that PM20D1 expression levels were raised in the frontal cortex at symptomatic stages of the disease. The link between PM20D1 and AD was also confirmed through human postmortem studies. “In human frontal cortex brain samples, we analyzed PM20D1 expression after correcting for the rs708727–rs960603 haplotype and found that PM20D1 expression was increased in non-risk haplotype carriers with AD compared to non-risk haplotype carriers without AD,” the researchers report.

Genetically increasing or decreasing levels of PM20D1 in cultures of brain cells derived from AD mice similarly demonstrated that PM20D1 overexpression decreased ROS-induced cell death, while in primary cultures PM20D1 overexpression reduced Aβ levels. Inducing overexpression of PM20D1 in the mouse AD model demonstrated what the team comments as a “previously unknown protective role for PM20D1 in the progression of AD-related pathologies.”

They point out that PM20D1 has recently been linked with obesity and diabetes, which are also risk factors for AD. “Furthermore, PM20D1 has been reported to be differentially methylated in individuals with obesity or multiple sclerosis, and it lies within the PD 16 (susceptibility) (PARK16) locus on chromosome 1, which has previously been associated with Parkinson’s disease.”

The team says further studies are now needed to evaluate the mechanisms by which PM20D1 impacts on AD. “The results obtained demonstrate the need for international scientific collaboration, mixing the different areas of experience in epigenetics, genetics, bioinformatics, and neurosciences of each group,” states Dr. Esteller. “We are looking at an example of the usefulness of multidisciplinary research to tackle diseases as complex and devastating as dementia.”