Identification of risk genes for Alzheimer's disease by gene embedding

Alzheimer's Disease Sequencing Project release 4 whole genome sequencing dataset

INTRODUCTION

The Alzheimer's Disease Sequencing Project (ADSP) is a national initiative to understand the genetic architecture of Alzheimer's disease and related dementias (ADRD) by integrating whole genome sequencing (WGS) with other genetic, phenotypic, and harmonized datasets from diverse populations.

METHODS

The Genome Center for Alzheimer's Disease (GCAD) uniformly processed WGS from 36,361 ADSP samples, including 35,014 genetically unique participants of which 45% are from non-European ancestry, across 17 cohorts in 14 countries in this fourth release (R4).

RESULTS

This sequencing effort identified 387 million bi-allelic variants, 42 million short insertions/deletions, and 6.8 million structural variants. Annotations and quality control data are available for all variants and samples. Additionally, detailed phenotypes from 15,927 participants across 10 domains are also provided. A linkage disequilibrium panel was created using unrelated AD cases and controls.

DISCUSSION

Researchers can access and analyze the genetic data via the National Institute on Aging Genetics of Alzheimer's Disease Data Storage Site (NIAGADS) Data Sharing Service, the VariXam, or NIAGADS GenomicsDB.

HIGHLIGHTS

We detailed the genetic architecture and quality of the Alzheimer's Disease Sequencing Project release 4 whole genome sequences. We identified 435 million single nucleotide polymorphisms, insertions and deletions, and structural variants from diverse genomes. We harmonized extensive phenotypes, linkage disequilibrium reference panel on subset of samples. Data is publicly available at NIAGADS Data Storage Site, variants and annotations are browsable on two different websites.

Alzheimer's Disease Sequencing Project Release 4 Whole Genome Sequencing Dataset

The Alzheimer's Disease Sequencing Project (ADSP) is a national initiative to understand the genetic architecture of Alzheimer's Disease and Related Dementias (AD/ADRD) by sequencing whole genomes of affected participants and age-matched cognitive controls from diverse populations. The Genome Center for Alzheimer's Disease (GCAD) processed whole-genome sequencing data from 36,361 ADSP participants, including 35,014 genetically unique participants of which 45% are from non-European ancestry, across 17 cohorts in 14 countries in this fourth release (R4). This sequencing effort identified 387 million bi-allelic variants, 42 million short insertions/deletions, and 2.2 million structural variants. Annotations and quality control data are available for all variants and samples. Additionally, detailed phenotypes from 15,927 participants across 10 domains are also provided. A linkage disequilibrium panel was created using unrelated AD cases and controls. Researchers can access and analyze the genetic data via NIAGADS Data Sharing Service, the VariXam tool, or NIAGADS GenomicsDB.

Multi-trait association analysis reveals shared genetic loci between Alzheimer's disease and cardiovascular traits

Several cardiovascular traits and diseases co-occur with Alzheimer's disease. We mapped their shared genetic architecture using multi-trait genome-wide association studies. Subsequent fine-mapping and colocalisation highlighted 16 genetic loci associated with both Alzheimer's and cardiovascular diseases. We prioritised rs11786896, which colocalised with Alzheimer's disease, atrial fibrillation and expression of PLEC in the heart left ventricle, and rs7529220, which colocalised with Alzheimer's disease, atrial fibrillation and expression of C1Q family genes. Single-cell RNA-sequencing data, co-expression network and protein-protein interaction analyses provided evidence for different mechanisms of PLEC, which is upregulated in left ventricular endothelium and cardiomyocytes with heart failure and in brain astrocytes with Alzheimer's disease. Similar common mechanisms are implicated for C1Q in heart macrophages with heart failure and in brain microglia with Alzheimer's disease. These findings highlight inflammatory and pleomorphic risk determinants for the co-occurrence of Alzheimer's and cardiovascular diseases and suggest PLEC, C1Q and their interacting proteins as potential therapeutic targets.

Single-Cell Patch-Clamp/Proteomics of Human Alzheimer's Disease iPSC-Derived Excitatory Neurons Versus Isogenic Wild-Type Controls Suggests Novel Causation and Therapeutic Targets

Standard single-cell (sc) proteomics of disease states inferred from multicellular organs or organoids cannot currently be related to single-cell physiology. Here, a scPatch-Clamp/Proteomics platform is developed on single neurons generated from hiPSCs bearing an Alzheimer's disease (AD) genetic mutation and compares them to isogenic wild-type controls. This approach provides both current and voltage electrophysiological data plus detailed proteomics information on single-cells. With this new method, the authors are able to observe hyperelectrical activity in the AD hiPSC-neurons, similar to that observed in the human AD brain, and correlate it to ≈1400 proteins detected at the single neuron level. Using linear regression and mediation analyses to explore the relationship between the abundance of individual proteins and the neuron's mutational and electrophysiological status, this approach yields new information on therapeutic targets in excitatory neurons not attainable by traditional methods. This combined patch-proteomics technique creates a new proteogenetic-therapeutic strategy to correlate genotypic alterations to physiology with protein expression in single-cells.

Drug target prediction through deep learning functional representation of gene signatures

Many machine learning applications in bioinformatics currently rely on matching gene identities when analyzing input gene signatures and fail to take advantage of preexisting knowledge about gene functions. To further enable comparative analysis of OMICS datasets, including target deconvolution and mechanism of action studies, we develop an approach that represents gene signatures projected onto their biological functions, instead of their identities, similar to how the word2vec technique works in natural language processing. We develop the Functional Representation of Gene Signatures (FRoGS) approach by training a deep learning model and demonstrate that its application to the Broad Institute's L1000 datasets results in more effective compound-target predictions than models based on gene identities alone. By integrating additional pharmacological activity data sources, FRoGS significantly increases the number of high-quality compound-target predictions relative to existing approaches, many of which are supported by in silico and/or experimental evidence. These results underscore the general utility of FRoGS in machine learning-based bioinformatics applications. Prediction networks pre-equipped with the knowledge of gene functions may help uncover new relationships among gene signatures acquired by large-scale OMICs studies on compounds, cell types, disease models, and patient cohorts.

A review and analysis of key biomarkers in Alzheimer's disease

Alzheimer's disease (AD) is a progressive neurodegenerative disorder that affects over 50 million elderly individuals worldwide. Although the pathogenesis of AD is not fully understood, based on current research, researchers are able to identify potential biomarker genes and proteins that may serve as effective targets against AD. This article aims to present a comprehensive overview of recent advances in AD biomarker identification, with highlights on the use of various algorithms, the exploration of relevant biological processes, and the investigation of shared biomarkers with co-occurring diseases. Additionally, this article includes a statistical analysis of key genes reported in the research literature, and identifies the intersection with AD-related gene sets from databases such as AlzGen, GeneCard, and DisGeNet. For these gene sets, besides enrichment analysis, protein-protein interaction (PPI) networks utilized to identify central genes among the overlapping genes. Enrichment analysis, protein interaction network analysis, and tissue-specific connectedness analysis based on GTEx database performed on multiple groups of overlapping genes. Our work has laid the foundation for a better understanding of the molecular mechanisms of AD and more accurate identification of key AD markers.

Cytokine enrichment in deep cerebellar nuclei is contributed by multiple glial populations and linked to reduced amyloid plaque pathology

Alzheimer's disease (AD) pathology and amyloid-beta (Aβ) plaque deposition progress slowly in the cerebellum compared to other brain regions, while the entorhinal cortex (EC) is one of the most vulnerable regions. Using a knock-in AD mouse model (App KI), we show that within the cerebellum, the deep cerebellar nuclei (DCN) has particularly low accumulation of Aβ plaques. To identify factors that might underlie differences in the progression of AD-associated neuropathology across regions, we profiled gene expression in single nuclei (snRNAseq) across all cell types in the DCN and EC of wild-type (WT) and App KI male mice at age 7 months. We found differences in expression of genes associated with inflammatory activation, PI3K-AKT signalling, and neuron support functions between both regions and genotypes. In WT mice, the expression of interferon-response genes in microglia is higher in the DCN than the EC and this enrichment is confirmed by RNA in situ hybridisation, and measurement of inflammatory cytokines by protein array. Our analyses also revealed that multiple glial populations are responsible for establishing this cytokine-enriched niche. Furthermore, homogenates derived from the DCN induced inflammatory gene expression in BV2 microglia. We also assessed the relationship between the DCN microenvironment and Aβ pathology by depleting microglia using a CSF1R inhibitor PLX5622 and saw that, surprisingly, the expression of a subset of inflammatory cytokines was increased while plaque abundance in the DCN was further reduced. Overall, our study revealed the presence of a cytokine-enriched microenvironment unique to the DCN that when modulated, can alter plaque deposition.

Neuronal γ-secretase regulates lipid metabolism, linking cholesterol to synaptic dysfunction in Alzheimer's disease

Presenilin mutations that alter γ-secretase activity cause familial Alzheimer's disease (AD), whereas ApoE4, an apolipoprotein for cholesterol transport, predisposes to sporadic AD. Both sporadic and familial AD feature synaptic dysfunction. Whether γ-secretase is involved in cholesterol metabolism and whether such involvement impacts synaptic function remains unknown. Here, we show that in human neurons, chronic pharmacological or genetic suppression of γ-secretase increases synapse numbers but decreases synaptic transmission by lowering the presynaptic release probability without altering dendritic or axonal arborizations. In search of a mechanism underlying these synaptic impairments, we discovered that chronic γ-secretase suppression robustly decreases cholesterol levels in neurons but not in glia, which in turn stimulates neuron-specific cholesterol-synthesis gene expression. Suppression of cholesterol levels by HMG-CoA reductase inhibitors (statins) impaired synaptic function similar to γ-secretase inhibition. Thus, γ-secretase enables synaptic function by maintaining cholesterol levels, whereas the chronic suppression of γ-secretase impairs synapses by lowering cholesterol levels.

Prioritization of risk genes for Alzheimer's disease: an analysis framework using spatial and temporal gene expression data in the human brain based on support vector machine

Background: Alzheimer's disease (AD) is a complex disorder, and its risk is influenced by multiple genetic and environmental factors. In this study, an AD risk gene prediction framework based on spatial and temporal features of gene expression data (STGE) was proposed. Methods: We proposed an AD risk gene prediction framework based on spatial and temporal features of gene expression data. The gene expression data of providers of different tissues and ages were used as model features. Human genes were classified as AD risk or non-risk sets based on information extracted from relevant databases. Support vector machine (SVM) models were constructed to capture the expression patterns of genes believed to contribute to the risk of AD. Results: The recursive feature elimination (RFE) method was utilized for feature selection. Data for 64 tissue-age features were obtained before feature selection, and this number was reduced to 19 after RFE was performed. The SVM models were built and evaluated using 19 selected and full features. The area under curve (AUC) values for the SVM model based on 19 selected features (0.740 [0.690-0.790]) and full feature sets (0.730 [0.678-0.769]) were very similar. Fifteen genes predicted to be risk genes for AD with a probability greater than 90% were obtained. Conclusion: The newly proposed framework performed comparably to previous prediction methods based on protein-protein interaction (PPI) network properties. A list of 15 candidate genes for AD risk was also generated to provide data support for further studies on the genetic etiology of AD.

Downregulation of Krüppel-like factor 14 accelerated cellular senescence and aging

Aging has been considered as a risk factor in many diseases, thus, comprehensively understanding the cellular and molecular mechanisms of delayed aging is important. Here we investigated whether Krüppel-like factor 14 (KLF14) is a suppressor of cellular senescence and aging. In our research, KLF14 levels significantly decreased not only in the lymphocytes of healthy people but also in the cells and tissues of mice with aging. We performed in vitro and in vivo experiments on cells and mice to reveal the function of KLF14 in aging. KLF14 deficiency facilitates cellular senescence and aging-related pathologies in C57BL/6J mice, whereas KLF14 overexpression attenuates cellular senescence. Mechanistically, KLF14 delays aging by binding to the POLD1 promoter to positively regulate POLD1 expression. Remarkably, cellular senescence mediated by KLF14 downregulation could be alleviated by POLD1 expression. In addition, perhexiline, an agonist of KLF14, could delay cellular senescence and aging-related pathologies in senescence-accelerated P8 mice by inducing POLD1 expression, as perhexiline could enhance the effect of KLF14's transcription activation to POLD1 by elevating the binding level of KLF14 to the POLD1 promoter. Our data indicate that KLF14 might be a critical element in aging by upregulating POLD1 expression, indicating that the activation of KLF14 may delay aging and aging-associated diseases.

Evolutionary Action-Machine Learning Model Identifies Candidate Genes Associated With Early-Onset Coronary Artery Disease

Background Coronary artery disease is a primary cause of death around the world, with both genetic and environmental risk factors. Although genome-wide association studies have linked >100 unique loci to its genetic basis, these only explain a fraction of disease heritability. Methods and Results To find additional gene drivers of coronary artery disease, we applied machine learning to quantitative evolutionary information on the impact of coding variants in whole exomes from the Myocardial Infarction Genetics Consortium. Using ensemble-based supervised learning, the Evolutionary Action-Machine Learning framework ranked each gene's ability to classify case and control samples and identified 79 significant associations. These were connected to known risk loci; enriched in cardiovascular processes like lipid metabolism, blood clotting, and inflammation; and enriched for cardiovascular phenotypes in knockout mouse models. Among them, INPP5F and MST1R are examples of potentially novel coronary artery disease risk genes that modulate immune signaling in response to cardiac stress. Conclusions We concluded that machine learning on the functional impact of coding variants, based on a massive amount of evolutionary information, has the power to suggest novel coronary artery disease risk genes for mechanistic and therapeutic discoveries in cardiovascular biology, and should also apply in other complex polygenic diseases.

Clinical evidence of human pathogens implicated in Alzheimer's disease pathology and the therapeutic efficacy of antimicrobials: an overview

A wealth of pre-clinical reports and data derived from human subjects and brain autopsies suggest that microbial infections are relevant to Alzheimer's disease (AD). This has inspired the hypothesis that microbial infections increase the risk or even trigger the onset of AD. Multiple models have been developed to explain the increase in pathogenic microbes in AD patients. Although this hypothesis is well accepted in the field, it is not yet clear whether microbial neuroinvasion is a cause of AD or a consequence of the pathological changes experienced by the demented brain. Along the same line, the gut microbiome has also been proposed as a modulator of AD. In this review, we focus on human-based evidence demonstrating the elevated abundance of microbes and microbe-derived molecules in AD hosts as well as their interactions with AD hallmarks. Further, the direct-purpose and potential off-target effects underpinning the efficacy of anti-microbial treatments in AD are also addressed.