The molecular etiology of Alzheimer’s disease

Abstract

Alzheimer’s disease (AD) is a growing global healthcare epidemic. Owing to advances in technology, genome‐scale studies of various layers of molecular information have been undertaken in recent years and robust variation in key loci have now been published and reproduced by others. This mini‐symposium highlights four key areas of current research in the field of molecular biology in AD, including articles focused on large‐scale genomic profiling, epigenetic research, integrative multi‐omic approaches and how these can be appropriately modeled to address reverse causality. This mini‐symposium provides a timely update on research focused on elucidating the molecular etiology of AD to date and highlights new methodological advances that could enable neuroscientists to identify novel therapeutic targets.

Dementia currently affects 50 million people globally and has been reported to be the fifth leading cause of death by the world health organization (WHO) (1). With an increasingly aging population, the number of dementia cases are predicted to surpass 130 million by 2050, which will have profound healthcare consequences. Alzheimer’s disease (AD) accounts for ~60% of dementia cases and is characterized by the accumulation of two protein hallmarks: extracellular amyloid beta (Aβ) plaques and intracellular neurofibrillary tangles (NFTs) of hyperphosphorylated tau. This is accompanied by neuronal cell loss, inflammation and ultimately death. Recent studies suggest that some of the neuropathological features of the disease start decades before the onset of clinical symptoms, such as memory loss, changes in mood, personality and behavior. Although the cellular pathology of AD has been extensively described, there are currently still no disease‐modifying drugs available for the disorder and available medications only partially alleviate the symptoms of the disorder.

Given the predicted rise in AD incidence, a key priority in the field has been the identification of molecular mechanisms underlying AD etiology, which could represent novel therapeutic targets. Owing to advances in genomic technology, research has primarily focused on identifying genetic variation that could lead to disease using genome wide association studies (GWAS) and a comprehensive review of the AD genetics field features in this mini‐symposium series (2). Over the last decade these studies have grown considerably in size, with the most recently published meta‐analyses incorporating data from nearly 500, 000 individuals (3, 4). Alongside these studies of common variants, sequencing efforts have identified rare variants with larger effect sizes, for example TREM2 (5). There has also been a move in recent years to generate polygenic risk scores (PRS) for AD (6), summing together the effects across multiple susceptibility loci into an aggregate score. Despite these advances in identifying variants robustly associated with disease, these single nucleotide polymorphism (SNP) variants alone do not explain all of sporadic AD incidence and their mode of action is not clear. Most variants associated with AD do not cause changes to protein structure and are, therefore, hypothesized to regulate gene expression. Other epidemiological features of AD, such as monozygotic twin discordance for disease (7), also suggest that mechanisms that regulate gene expression, such as epigenetic processes, are likely to be important in disease.

Because epigenetic modifications can be dynamic and manipulated by the environment they have recently drawn a considerable amount of interest in the field. Indeed, over the last six years a growing number of epigenome‐wide association studies (EWAS) have been undertaken in AD brain samples. These have primarily focused on profiling DNA methylation, and a number of consistent and reproducible loci have been identified across different studies and brain regions. Despite the convincing evidence of epigenetic involvement in AD, the field is currently constrained by several experimental limitations, which are discussed and evaluated in detail in the review provided by Van den Hove, Riemens, Koulousakis and Pishva (8) as part of this mini‐symposium series. This report also discusses how new methodological advances in epigenetics and related data science disciplines could aid the development of novel therapeutic agents and biomarker assays. One important caveat to consider in the context of epigenetic target identification in AD is whether the differences identified represent a cause or a consequence of disease. This question will need to be answered before further progress can be made in the search for epigenetic therapeutics for AD.

Although genetic and epigenomic studies in AD have been invaluable in the identification of novel dysfunctional genes and pathways in disease, an integrated systems biology approach integrating different data modalities will ultimately be required to identify novel therapeutic targets. Genetic, epigenetic, transcriptomic, proteomic and metabolomic variation do not operate in isolation and multi‐omic data integration is imperative. The article in this mini‐symposium series from Ma, Klein and De Jager (9) reviews the requirements and approaches important for integrative omic analyses in AD. However, these approaches are currently hampered by a lack of methods for data integration and a consensus in the genomic field as to the appropriate parameters to use. Ultimately, the ability to integrate complex omics datasets is reliant on data collection to be harmonious across sample cohorts enabling assembly and replication of observations. It is also important to capture all levels of information (genetic, epigenetic, transcriptomic, proteomic and metabolomic) in the same well characterized cohort of samples.

Although molecular studies in AD have highlighted dysfunctional genes and pathways involved in pathogenesis, it is particularly important to address the issue of reverse causality. In the context of genetic variation, this is relatively straightforward as we know that non‐somatic polymorphisms are present from conception and before disease onset. However, for other data modalities related to transcriptional mechanisms it has been difficult to establish causal mechanisms. Recent ground‐breaking developments in gene editing technologies now allow the manipulation of the genome or epigenome to model disease‐associated variation in vitro. In the context of genetic changes, this can allow the identification of downstream functional consequences of specific genetic variants, particularly when compared to isogenic controls. Furthermore, the recent adaptation of these methodologies to specifically alter epigenetic processes such as DNA methylation at specific loci will allow the establishment of whether disease‐associated epigenetic variation are likely causal in disease. The review by Schrauben, Dempster, Lunnon (10) within this mini‐symposium examines the different iterations of gene editing systems and future applications of these technologies for exploring the complex interactions and disruptions in gene regulatory circuits that contribute to AD.

The AD research field has progressed rapidly in recent years, particularly in the context of molecular biology. Key molecular targets have now been identified, with the next step requiring new methods of data analysis, more comprehensive studies and improved modeling of disease risk factors in model systems. Ultimately, this will lead to a better understanding of the molecular mechanisms initiating and driving AD pathology and will allow the identification of novel therapeutic targets, which are urgently required for this disease. This mini‐symposium provides a timely update on the research into the molecular etiology of AD to date, and highlights some of the exciting new methodological advances enabling scientists to take this research forward to ultimately allow the development of disease‐modifying agents.