CoChat

A genetic link between risk for Alzheimer's disease and severe COVID-19 outcomes via the OAS1 gene

Open access
Magusali, Naciye, Graham, Andrew C, Piers, Thomas M, Panichnantakul, Pantila, Yaman, Umran, Shoai, Maryam, Reynolds, Regina H, Botia, Juan A, Brookes, Keeley J, Guetta-Baranes, Tamar, Bellou, Eftychia, Bayram, Sevinc, Sokolova, Dimitra, Ryten, Mina, Sala Frigerio, Carlo, Escott-Price, Valentina, Morgan, Kevin, Pocock, Jennifer M, Hardy, John, Salih, Dervis A · 2021 · Brain

The DOI matches its Crossref record; 124 of the 124 references it cites were found in a registry or an index.

Abstract

Recently, we reported oligoadenylate synthetase 1 (OAS1) contributed to the risk of Alzheimer’s disease, by its enrichment in transcriptional networks expressed by microglia. However, the function of OAS1 within microglia was not known. Using genotyping from 1313 individuals with sporadic Alzheimer’s disease and 1234 control individuals, we confirm the OAS1 variant, rs1131454, is associated with increased risk for Alzheimer’s disease. The same OAS1 locus has been recently associated with severe coronavirus disease 2019 (COVID-19) outcomes, linking risk for both diseases. The single nucleotide polymorphisms rs1131454(A) and rs4766676(T) are associated with Alzheimer’s disease, and rs10735079(A) and rs6489867(T) are associated with severe COVID-19, where the risk alleles are linked with decreased OAS1 expression. Analysing single-cell RNA-sequencing data of myeloid cells from Alzheimer’s disease and COVID-19 patients, we identify co-expression networks containing interferon (IFN)-responsive genes, including OAS1, which are significantly upregulated with age and both diseases. In human induced pluripotent stem cell-derived microglia with lowered OAS1 expression, we show exaggerated production of TNF-α with IFN-γ stimulation, indicating OAS1 is required to limit the pro-inflammatory response of myeloid cells. Collectively, our data support a link between genetic risk for Alzheimer’s disease and susceptibility to critical illness with COVID-19 centred on OAS1, a finding with potential implications for future treatments of Alzheimer’s disease and COVID-19, and development of biomarkers to track disease progression.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1093/brain/awab337

Cites

124 of 124 references verified0 citation errors · 0 not checkable

Statistics

Figures

Figure 1 The genomic position of four SNPs on 12q24: rs1131454 and rs4766676 (associated with Alzheimer’s disease) and rs10735079 and rs6489867 (associated with critical illness with COVID-19) relative to OAS1. We have confirmed rs1131454 is associated with Alzheimer’s disease (Table 1) following our analysis of Alzheimer’s disease GWAS data (our data is given in Salih et al.,60 which was derived from further analysis of the GWAS data in Lambert et al.83). We identified the association of rs4766676 with Alzheimer’s disease (our data is given in Salih et al.,12 which was derived from further analysis of the GWAS data in Kunkle et al.84). SNPs rs10735079 and rs6489867 are associated with severe illness with COVID-19.32 The R2 scale indicates LD between rs1131454 (the reference SNP, purple), compared with the other three SNPs. Drawn with LocusZoom.61
Figure 1 The genomic position of four SNPs on 12q24: rs1131454 and rs4766676 (associated with Alzheimer’s disease) and rs10735079 and rs6489867 (associated with critical illness with COVID-19) relative to OAS1. We have confirmed rs1131454 is associated with Alzheimer’s disease (Table 1) following our analysis of Alzheimer’s disease GWAS data (our data is given in Salih et al.,60 which was derived from further analysis of the GWAS data in Lambert et al.83). We identified the association of rs4766676 with Alzheimer’s disease (our data is given in Salih et al.,12 which was derived from further analysis of the GWAS data in Kunkle et al.84). SNPs rs10735079 and rs6489867 are associated with severe illness with COVID-19.32 The R2 scale indicates LD between rs1131454 (the reference SNP, purple), compared with the other three SNPs. Drawn with LocusZoom.61CC BY-NC · PubMed Central
Figure 2 An interferon response-associated gene module is present along a distinct microglial activation trajectory upregulated in aged mice and mice with amyloid pathology. (A) The genetic network containing Oas1a from microglial cells isolated from wild-type and APPNL-G-F knock-in mice at 3, 6, 12 and 21 months of age analysed by scRNA-seq.9 The 50 genes showing the highest connectivity are plotted, and Oas1a is highlighted with a black oval. Green nodes represent genes, edge lines represent co-expression connections, and the central large red nodes are the hub genes (the full network given in Supplementary Table 4). (B) Semi-supervised pseudotime ordering of microglial cells isolated from wild-type and APPNL-G-F knock-in mice based on expression,9 with Monocle 2, shows homeostatic cells as the root state, and ARM and IRM as the end points of distinct activation trajectories. (C) The gene module containing Oas1a is upregulated (yellow data-points) along the IRM-associated activation trajectory. The expression of this module is relatively absent from both the root homeostatic state and the ARM trajectory (purple data-points). (D) Mean normalized expression of the 60 most central genes in the interferon-response module is greater in microglia isolated from aged wild-type relative to young adult mice (6–8 weeks versus 16–18 months of age; n = 6 mice per group). Data are shown as mean ± SEM. Student’s t-test; **P < 0.01. Further analysis of data from O’Neil et al.69 (E) Mean normalized expression of the 60 most central genes in the interferon-response module is greater in microglia isolated from PSEN2/APP relative to wild-type mice at 14–15 months of age (n = 6–9 mice per group). Data are shown as mean ± SEM. Student’s t-test; ****P < 0.0001. Further analysis of data from Friedman et al.13
Figure 2 An interferon response-associated gene module is present along a distinct microglial activation trajectory upregulated in aged mice and mice with amyloid pathology. (A) The genetic network containing Oas1a from microglial cells isolated from wild-type and APPNL-G-F knock-in mice at 3, 6, 12 and 21 months of age analysed by scRNA-seq.9 The 50 genes showing the highest connectivity are plotted, and Oas1a is highlighted with a black oval. Green nodes represent genes, edge lines represent co-expression connections, and the central large red nodes are the hub genes (the full network given in Supplementary Table 4). (B) Semi-supervised pseudotime ordering of microglial cells isolated from wild-type and APPNL-G-F knock-in mice based on expression,9 with Monocle 2, shows homeostatic cells as the root state, and ARM and IRM as the end points of distinct activation trajectories. (C) The gene module containing Oas1a is upregulated (yellow data-points) along the IRM-associated activation trajectory. The expression of this module is relatively absent from both the root homeostatic state and the ARM trajectory (purple data-points). (D) Mean normalized expression of the 60 most central genes in the interferon-response module is greater in microglia isolated from aged wild-type relative to young adult mice (6–8 weeks versus 16–18 months of age; n = 6 mice per group). Data are shown as mean ± SEM. Student’s t-test; **P < 0.01. Further analysis of data from O’Neil et al.69 (E) Mean normalized expression of the 60 most central genes in the interferon-response module is greater in microglia isolated from PSEN2/APP relative to wild-type mice at 14–15 months of age (n = 6–9 mice per group). Data are shown as mean ± SEM. Student’s t-test; ****P < 0.0001. Further analysis of data from Friedman et al.13CC BY-NC · PubMed Central
Figure 3 An interferon-response-associated gene module in the microglia of humans with Alzheimer’s disease shows significant overlap with an interferon-response gene module in lung macrophages of patients with severe COVID-19. (A) Genetic network plot of an interferon-response associated module detected in microglial cells isolated from human Alzheimer’s disease patients and individuals with MCI analysed by scRNA-seq.19 This module shows a significant overlap with the interferon-response module detected in mice (Fig. 2A) (the full network is given in Supplementary Table 5). A hub gene of this module is CD163, and a number of other macrophage marker genes are prominent within this module (TGFBI, F13A1, LY6E and LYZ), indicating that expression of OAS1 and other interferon-response genes in the human Alzheimer’s disease patient brains are either associated with a population of CNS border-associated or invading macrophages, or potentially a microglial subpopulation that upregulates anti-inflammatory genes. OAS1 is highlighted with a black oval. (B) Linear regression of the mean normalized expression of the 60 most connected genes from the human interferon-response network in microglia isolated from the cortex of humans of different ages (range: 34–102 years of age) and profiled by RNA-seq by Galatro et al.70 Linear regression formula: Log2(mean normalized expression + 1) = age + gender + site of origin + post-mortem delay. Data fulfils the assumptions of linearity, normality of residuals, and homoscedasticity, and shows no significant outliers. Red line depicts the line that minimizes distances to the observed values. Grey area depicts 95% confidence intervals. (C) Genetic network plot of an interferon-response-associated module detected in macrophages isolated from the bronchoalveolar fluid of patients with severe/critical and moderate COVID-19, alongside healthy individuals analysed by scRNA-seq.66 This module shows a significant overlap with the interferon-response module detected in human Alzheimer’s disease patients and individuals with MCI (Fig. 3A) (the full network given in Supplementary Table 6). A hub gene of this module is ISG15, and a number of other macrophage marker genes are prominent within this module (TGFB1, LY6E and LYZ). OAS1 is highlighted with a black oval. (D) Expression of the OAS1-containing module in macrophages isolated from individuals with severe and modest COVID-19, alongside healthy controls. The mean module eigengene score was calculated as the first principal component of module expression, giving a reliable indicator of module expression for macrophages from each patient (n = 6 severe COVID-19 individuals, n = 3 modest COVID-19, and n = 3 healthy controls). Data are shown as a box-and-whisker plot, where the median is the box middle, the box is the interquartile range, and the whiskers are the minimum and maximum data-points. Kruskal-Wallis test showed a significant effect of disease status (P < 0.01). Dunn’s multiple comparisons tests were then performed to test for pairwise significance between the healthy control and each COVID-19 group; *P < 0.05. Further analysis of data from Liao et al.66
Figure 3 An interferon-response-associated gene module in the microglia of humans with Alzheimer’s disease shows significant overlap with an interferon-response gene module in lung macrophages of patients with severe COVID-19. (A) Genetic network plot of an interferon-response associated module detected in microglial cells isolated from human Alzheimer’s disease patients and individuals with MCI analysed by scRNA-seq.19 This module shows a significant overlap with the interferon-response module detected in mice (Fig. 2A) (the full network is given in Supplementary Table 5). A hub gene of this module is CD163, and a number of other macrophage marker genes are prominent within this module (TGFBI, F13A1, LY6E and LYZ), indicating that expression of OAS1 and other interferon-response genes in the human Alzheimer’s disease patient brains are either associated with a population of CNS border-associated or invading macrophages, or potentially a microglial subpopulation that upregulates anti-inflammatory genes. OAS1 is highlighted with a black oval. (B) Linear regression of the mean normalized expression of the 60 most connected genes from the human interferon-response network in microglia isolated from the cortex of humans of different ages (range: 34–102 years of age) and profiled by RNA-seq by Galatro et al.70 Linear regression formula: Log2(mean normalized expression + 1) = age + gender + site of origin + post-mortem delay. Data fulfils the assumptions of linearity, normality of residuals, and homoscedasticity, and shows no significant outliers. Red line depicts the line that minimizes distances to the observed values. Grey area depicts 95% confidence intervals. (C) Genetic network plot of an interferon-response-associated module detected in macrophages isolated from the bronchoalveolar fluid of patients with severe/critical and moderate COVID-19, alongside healthy individuals analysed by scRNA-seq.66 This module shows a significant overlap with the interferon-response module detected in human Alzheimer’s disease patients and individuals with MCI (Fig. 3A) (the full network given in Supplementary Table 6). A hub gene of this module is ISG15, and a number of other macrophage marker genes are prominent within this module (TGFB1, LY6E and LYZ). OAS1 is highlighted with a black oval. (D) Expression of the OAS1-containing module in macrophages isolated from individuals with severe and modest COVID-19, alongside healthy controls. The mean module eigengene score was calculated as the first principal component of module expression, giving a reliable indicator of module expression for macrophages from each patient (n = 6 severe COVID-19 individuals, n = 3 modest COVID-19, and n = 3 healthy controls). Data are shown as a box-and-whisker plot, where the median is the box middle, the box is the interquartile range, and the whiskers are the minimum and maximum data-points. Kruskal-Wallis test showed a significant effect of disease status (P < 0.01). Dunn’s multiple comparisons tests were then performed to test for pairwise significance between the healthy control and each COVID-19 group; *P < 0.05. Further analysis of data from Liao et al.66CC BY-NC · PubMed Central
Figure 4 OAS1 knockdown in the presence of IFN-γ resulted in exaggerated expression of TNF in hman iPSCs differentiated to microglia. (A) OAS1 expression was knocked-down using siRNA under basal conditions and in response to IFN-γ. Two-way ANOVA with significant main effects of IFN-γ treatment (P < 0.01), and siRNA treatment (P < 0.001) indicated as horizontal and vertical red lines, respectively, with a significant interaction (P < 0.001). (B) TNF expression, as a marker of the pro-inflammatory response, showed significant upregulation in response to IFN-γ application, and accentuated expression with OAS1 knockdown. Two-way ANOVA with significant main effects of IFN-γ treatment (P < 0.001), and siRNA treatment (P = 0.015) indicated as horizontal and vertical red lines, respectively, with a significant interaction (P = 0.023). (C) ELISA of TNF-α levels in the conditioned human iPSC-derived microglia media. TNF-α levels were normalized to the geometric mean of GAPDH, HPRT1 and RPS18 assessed by RT-qPCR. Note that TNF-α protein was not detectable in the media of human iPSC-derived microglia not treated with IFN-γ. n = 5 independent plates. Data are shown as mean ± SEM. Paired Student’s t-test (paired for wells on same plate); *P < 0.05. (D) C1QA expression was increased with OAS1 knockdown. Two-way ANOVA with significant main effect of only siRNA treatment (P = 0.029), not IFN-γ treatment (P > 0.05), indicated as a vertical red line, with no significant interaction (P > 0.05). Gene expression levels were normalized to the geometric mean of GAPDH, HPRT1 and RPS18, then calculated as fold change relative to the non-targeting siRNA control (vehicle control) without IFN-γ treatment in each individual culture preparation. n = 5 independent plates. Data are shown as mean ± SEM. Two-way ANOVA; significant main effects of IFN-γ treatment and OAS1-knockdown indicated by red lines. When a significant interaction was seen between IFN-γ treatment and siRNA treatment, Tukey’s multiple comparisons tests were then performed to test for pairwise significance of all groups; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.
Figure 4 OAS1 knockdown in the presence of IFN-γ resulted in exaggerated expression of TNF in hman iPSCs differentiated to microglia. (A) OAS1 expression was knocked-down using siRNA under basal conditions and in response to IFN-γ. Two-way ANOVA with significant main effects of IFN-γ treatment (P < 0.01), and siRNA treatment (P < 0.001) indicated as horizontal and vertical red lines, respectively, with a significant interaction (P < 0.001). (B) TNF expression, as a marker of the pro-inflammatory response, showed significant upregulation in response to IFN-γ application, and accentuated expression with OAS1 knockdown. Two-way ANOVA with significant main effects of IFN-γ treatment (P < 0.001), and siRNA treatment (P = 0.015) indicated as horizontal and vertical red lines, respectively, with a significant interaction (P = 0.023). (C) ELISA of TNF-α levels in the conditioned human iPSC-derived microglia media. TNF-α levels were normalized to the geometric mean of GAPDH, HPRT1 and RPS18 assessed by RT-qPCR. Note that TNF-α protein was not detectable in the media of human iPSC-derived microglia not treated with IFN-γ. n = 5 independent plates. Data are shown as mean ± SEM. Paired Student’s t-test (paired for wells on same plate); *P < 0.05. (D) C1QA expression was increased with OAS1 knockdown. Two-way ANOVA with significant main effect of only siRNA treatment (P = 0.029), not IFN-γ treatment (P > 0.05), indicated as a vertical red line, with no significant interaction (P > 0.05). Gene expression levels were normalized to the geometric mean of GAPDH, HPRT1 and RPS18, then calculated as fold change relative to the non-targeting siRNA control (vehicle control) without IFN-γ treatment in each individual culture preparation. n = 5 independent plates. Data are shown as mean ± SEM. Two-way ANOVA; significant main effects of IFN-γ treatment and OAS1-knockdown indicated by red lines. When a significant interaction was seen between IFN-γ treatment and siRNA treatment, Tukey’s multiple comparisons tests were then performed to test for pairwise significance of all groups; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.CC BY-NC · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Cited by

See all 113

Metadata

Authors
Magusali, N., Graham, A. C., Piers, T. M., Panichnantakul, P., Yaman, U., Shoai, M., Reynolds, R. H., Botia, J. A., Brookes, K. J., Guetta-Baranes, T., Bellou, E., Bayram, S., Sokolova, D., Ryten, M., Sala Frigerio, C., Escott-Price, V., Morgan, K., Pocock, J. M., Hardy, J., & Salih, D. A.
Affiliations
  • UK Dementia Research Institute (GB)
  • UCL Queen Square Institute of Neurology (GB)
  • UK Dementia Research Institute at UCL, Gower Street, London WC1E 6BT, UK
  • UCL Biomedical Research Centre (GB)
  • Department of Neurodegenerative Diseases, Queen Square Institute of Neurology, UCL, London WC1N 1PJ, UK
  • Nottingham Trent University (GB)
  • Genetics, School of Life Sciences, Life Sciences Building, University Park, University of Nottingham, Nottingham NG7 2RD, UK
  • Cardiff University (GB)
  • Hitachi Rail Europe Ltd, New Ludgate, London EC4M 7HX, UK
  • Dementia Research Institute, MRC Centre for Neuropsychiatric Genetics and Genomics, Cardiff University, Cardiff CF24 4HQ, UK
  • University of Nottingham (GB)
Corresponding author
Salih, Dervis A
Venue
Brain
Published
2021-10-07
Volume
144
Issue
12
Pages
3727-3741
ISSN
  • 0006-8950 (ISSN-L)
  • 1460-2156
OpenAlex
W3201962999
Cited by
111
References
123
Citations by year
2026: 6 · 2025: 18 · 2024: 20 · 2023: 25 · 2022: 41 · 2021: 1
FWCI
13.98
Type
journal article
Language
en
Open access
Locations
  • Brain (publishedVersion, cc-by-nc)
  • Europe PMC (PubMed Central) (publishedVersion, other-oa)
  • PubMed Central (publishedVersion, cc-by-nc)
  • PubMed (publishedVersion)
  • ORCA Online Research @Cardiff (Cardiff University) (acceptedVersion)
  • UCL Discovery (University College London) (submittedVersion)
Funders
  • Innovative Medicines Initiative 2 Joint Undertaking (115976)
  • European Union's Horizon 2020
  • European Federation of Pharmaceutical Industries and Associations
  • Science and Technology Agency (00007/COVI/20)
  • The University of Nottingham Group
  • National Institute for Health Research University College London Hospitals Biomedical Research Centre
  • Alzheimer’s Research UK
  • UK Medical Research Council
  • Alzheimer’s Society and ARUK
Topics
  • Life Sciences › Neuroscience › Cellular and Molecular Neuroscience › Alzheimer's disease research and treatments
  • Life Sciences › Neuroscience › Neurology › Neuroinflammation and Neurodegeneration Mechanisms
  • Life Sciences › Immunology and Microbiology › Immunology › interferon and immune responses
Keywords
Alzheimer's disease, COVID-19 pandemic, OAS1, genetic risk, microglia, interferon-stimulated genes, TNF-α, interferon-gamma, myeloid cells, inflammation, single nucleotide polymorphisms, induced pluripotent stem cell-derived microglia
MeSH terms
COVID-19, Adolescent, Aged, Aged, 80 and over, Alzheimer Disease, Cells, Cultured, Female, Humans, Genetic Linkage, Male, Middle Aged, 2',5'-Oligoadenylate Synthetase, Genetic Predisposition to Disease, Polymorphism, Single Nucleotide, Gene Regulatory Networks, Young Adult, Induced Pluripotent Stem Cells, Patient Acuity

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.