key: cord-1007159-x1f12ang authors: Liu, D.; Zhang, X.; Cao, W.; Zhang, J.; Song, M.; Xing, W.; Wang, W.; Meng, Q.; Wang, Y. title: Association between Alzheimer's disease and COVID-19: A bidirectional Mendelian randomization date: 2020-07-30 journal: nan DOI: 10.1101/2020.07.27.20163212 sha: 95e80dfdb41c1eb09b0a473c4d2b29a3103b74c6 doc_id: 1007159 cord_uid: x1f12ang Background In observational studies, Alzheimer's disease (AD) has been associated with an increased risk of Coronavirus disease 2019 (COVID-19), and the prognosis of COVID-19 can affect nervous systems. However, the causality between these conditions remains to be determined. Methods This study sought to investigate the bidirectional causal relations of AD with COVID-19 using two-sample Mendelian randomization (MR) analysis. Results We found that genetically predicted AD was significantly associated with higher risk of severe COVID-19 (odds ratio [OR], 3.329; 95% confidence interval [CI], 1.139-9.725; P=0.028). It's interesting that genetically predicted severe COVID-19 was also significantly associated with higher risk of AD (OR, 1.004; 95% CI, 1.001-1.007; P=0.018). In addition, the two strong genetic variants associated with severe COVID-19 was associated with higher AD risk (OR, 1.018; 95% CI, 1.003-1.034; P=0.018). There is no evidence to support that genetically predicted AD was significantly associated with COVID-19 susceptibility, and vice versa. No obvious pleiotropy bias and heterogeneity were observed. Conclusion Overall, AD may causally affect severe COVID-19, and vice versa, performing bidirectional regulation through independent biological pathways. Coronavirus disease , which is caused by the severe acute respiratory syndrome coronavirus 2 (SARS-COV2) and represents the causative agent of a potentially fatal disease, rapidly developing into a global pandemic 1 .The prognosis of COVID-19 can cause impaired vision, loss of smell, and affect reproductive and nervous systems 2 3 , that is of great global public health concern. A wide range of studies focused on susceptible populations for COVID-19. In the susceptible population of the elderly, specific risk factors such as dementia have been found to be a risk factor (odds ratio [OR] = 3.50, 95% CI: 1.93 to 6.34) for COVID-19 severity in the UK Biobank (UKB) 4 . However, findings from epidemiological studies are prone to unmeasured confounding, and reverse causation. Mendelian randomization (MR)has becoming a popular method by which genetic variants robustly associated with exposures can be used as instrumental variables to infer causality from routinely conducted observational studies 5 Mendelian randomization (SMR) is an excellent strategy to evaluate the causality using summary statistics from GWAS data 6 . Therefore, we performed bidirectional MR analyses for causal inference between Alzheimer's disease (AD), the main type of dementia, and COVID-19 using summary statistics GWAS results of AD and COVID-19. Understanding the bidirectional relations between AD and COVID-19 is of significant public health importance about disease prevention and management of complications. Genetic association datasets for COVID -19 The summarized data was obtained from the most recent version of GWAS analyses from the COVID-19 host genetics initiative from UK Biobank individuals, which released on July 1, 2020 (https://www.covid19hg.org/results/) 7 Summarized data on AD were obtained from the International Genomics of Alzheimer Project with 17008 Alzheimer's disease cases and 37154 controls (https://www.pasteur-lille.fr) 9 . All GWAS summary statistic data were based on European ancestry populations. For the bidirectional MR analysis, we first applied forward MR analysis to assess the effect of predisposition to AD on the phenotypes of COVID-19 or severe COVID-19 by using genetic variants robustly associated with AD. Then, we performed reverse MR using genetic variants robustly associated with the phenotypes of COVID-19 or severe COVID-19 as genetic instruments to investigate their effect on AD. Single nucleotide polymorphisms (SNPs) that achieved significance (P<5×10 -8 ) for the AD were selected as instrumental variables (IVs). SNPs were chosen as IVs for COVID-19 or severe COVID-19 at a significance threshold (P<1×10 -5 ) from the COVID-19 host genetics initiative, because we didn't filter out SNPs as IVs for COVID-19 or severe COVID-19 at a significance threshold (P<5×10 -8 ). In addition, We only retained independent variants from each other (Linkage disequilibrium [LD], r 2 < 0.001). When we encountered genetic variants in LD, the SNP with the lowest p-value was selected. The LD proxies were defined using 1000 genomes European samples. The same method was applied to filter instrumental variables in reverse MR analysis. In the main analyses, inverse-variance weighted (IVW) method was used to estimate the overall causal association of AD on COVID-19 and severe COVID-19 susceptibility. We additionally conducted sensitivity analyses using the weighted median, penalised weighted median, and MR-Egger regression to account for potential violations of valid instrumental variable assumptions. The MR-Egger analysis was performed to valuate pleiotropy based on the intercept. We conducted heterogeneity test in MR analyses using IVWQ test and a leave one out analysis in which we omitted one genetic variant to investigate the influence of pleiotropic genetic variants. Results are presented as odds ratios with their 95% confidence intervals (95% CIs) of outcomes per genetically predicted increase in each exposure factor. All data analyses were performed by "twosampleMR" package in R (R version 3.6.3, R Core Team, 2017). In terms of various estimates for different measures, we chose the result of main MR method as following rules: (1) If no directional pleiotropy in MR estimates (Q statistic: P value > 0.05, MR-Egger intercept: P value > 0.05), we reported the results of IVW method. (2) If directional pleiotropy was detected (MR-Egger intercept: P value < 0.05) and P value > 0.05 for the Q test, we reported the results of MR-Egger method. (3) If directional pleiotropy was detected (MR-Egger intercept: P value < 0.05) and P value < 0.05 for the Q test, we reported the results of weighted median method. The causal relationships between AD and COVID-19 were delineated into four potential categories, which were observed by bidirectional MR. Figure 1 provided an overview of the bidirectional MR study to investigate these explanations. If the P value was less than 0.05 only in forward MR, genetic variants determined AD was associated with higher COVID-19 risk (Explanation 1). To robustly test Explanation 2, we performed the reverse MR analysis, evaluating whether COVID-19 affected AD. A P value greater than 0.05 indicated that Explanation 2 was unlikely. There are bidirectionally causal associations between AD and COVID-19, if both P values were less than 0.05 in forward and reverse MR. We identified the results to evaluate Explanation 3, that is, the finding is based on that the selected SNPs as IVs for forward and reverse MR analysis were not overlapped between AD and COVID-19. If the P value was greater than 0.05 in forward and reverse MR, it was interpreted as Explanation 4. The annotations of SNPs residing in ± 100 kb of the corresponding genes were obtained from annotation files of the Grch37/hg19. To functionally annotate putative SNPs, we conducted functional enrichment analysis for the identified genes using the functional annotation tool "Metascape" 10 and the R package "clusterProfiler". Gene symbols corresponding to putative genes (P < 0.05) were used as the input of the gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes enrichment analysis. In the AD→COVID-19 MR analysis, we used 20 independent SNPs as the instrument for AD. As shown in Table 1 and heterogeneity (P=0.335). The association of AD with COVID-19 and severe COVID-19 remained robust in the weighted median and penalised weighted median methods, and leave one out analyses ( Figure S1 and Figure 2 ). As shown in Table 2 (Figure 3 ). In addition, genetically predicted severe COVID-19 was significantly associated with higher risk of AD using the two strong genetic variants of rs11385942 (locates at 3p21.31) and rs657152 (9q34.2, ABO blood-group) as IVs for severe COVID-19 (OR, 1.018; 95% CI, 1.003-1.034; P=0.018). The SNPs as IVs for AD were annotated to 126 unique genes (Table S1) . GO enrichment analysis of biological process, molecular function, and cellular component pathways showed that the identified genes were involved in 20 GO terms (Figure 4) To our knowledge, this is the first study to investigate the causal relationship between 1 0 AD and severeCOVID-19 using a bidirectional MR. Our study showed that AD may causally affect severe COVID-19, and vice versa, performing bidirectional regulation through independent biological pathways. Our study found that AD was causally associated with the susceptibility of severe COVID-19, but not COVID-19, might indicating that AD patients infected with severe COVID-19 were more likely to be severe phenotype. The finding was consistent with the observational epidemiology study, which found that pre-existing dementia was associated with the risk of COVID-19 hospitalization and death increase 4 . In addition, the study showed that ApoEε4ε4 homozygotes were more likely to be COVID-19 positives (OR = 2.31, 95% CI: 1.65 to 3.24, P = 1.19E-6) compared to ε 3ε3 homozygotes in the UKBiobank Community Cohort. The ApoEε4ε4 genotype was associated with a 14-fold increase in risk of AD compared to the common ε 3ε3 genotype in populations with European ancestries 11 . Therefore, AD patients with the ApoE ε 4ε4 allele might increase risks of severe COVID-19 infection 12 . In addition, the genes annotated by genetic variants strongly associated with AD were involved in the pathway of immune response. That could explain the death of most severe COVID-19 patients due to the inflammatory storm 13 14 . This is the first study to identify the prognosis of COVID-19. We found that severe COVID-19 might cause AD. Notably, the COVID-19 cases endanger the elderly, especially those with cardiovascular illness including diabetes mellitus, hypertension, and coronary heart disease 15 16 . The increasing MR studies have found that coronary artery disease, higher body mass index, lifetime smoking and dyslipidemia were causally associated with the susceptibility of COVID-19 17-20 , which were also risk factors for AD 21 22 . The reasons underlying AD also may relate to the proinflammatory effect associated with severe COVID-19. An exaggerated inflammatory response appears a major driver of immunopathology in SARS-CoV-2 infected patients 23 , with benefits of targeting the acute inflammatory response in severe COVID-19 cases 24 . It has been reported that bi-directional communication existed between the central nervous system and the peripheral immune system 25 . Activated peripheral immune cells that enter the brain through the blood brain barrier (BBB) may modulate the pathogenesis of AD 26 27 . The genes DRD1 and MCTP1 annotated by genetic variants strongly associated with severe COVID-19 were involved in the pathway of regulation of neurotransmitter secretion and transport. SARS-COV2 might be a causative factor involved in the infectious etiology of AD. In addition, ABO blood type might play an important role in the development of COVID-19 and AD 8 28 . Taken together, the findings indicated that subjects infected with severe COVID-19 with respiratory failure might be of higher risk of AD after survival of COVID-19. No overlapped genetic variants as IVs for AD and severe COVID-19 was overserved, indicating that AD and severe COVID-19 might be bidirectionally regulated by different pathogenic pathways. The shared pathway of the regulation of GTPase activity might underly the association between AD and severe COVID-19 [29] [30] [31] , which needed further investigation to get more data support and further experimental study to verify. In conclusion, our study showed that AD may causally affect severe COVID-19, 1 3 of interest were disclosed by the authors. The study was supported by grants from the National Key R&D Program of China The funding organization had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication. on AD using all instruments expect for the SNP listed on the y-axis. The point represents the odds ratio with that SNP removed and the line represents 95% confidence interval. An interactive web-based dashboard to track COVID-19 in real time The potential long-term impact of the COVID-19 outbreak on patients with non-communicable diseases in Europe: consequences for healthy ageing Neurobiology of COVID-19 Preexisting Comorbidities Predicting COVID-19 and Mortality in the UK Biobank Community Cohort. The journals of gerontology Series A, Biological sciences and medical sciences Mendelian Randomization as an Approach to Assess Causality Using Observational Data A Primer in Mendelian Randomization Methodology with a Focus on Utilizing Published Summary Association Data The COVID-19 Host Genetics Initiative, a global initiative to elucidate the role of host genetic factors in susceptibility and severity of the SARS-CoV-2 virus pandemic Genomewide Association Study of Severe Covid-19 with Respiratory Failure Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer's disease Metascape provides a biologist-oriented resource for the analysis of systems-level datasets Effects of age, sex, and ethnicity on the association between apolipoprotein E genotype and Alzheimer disease. A meta-analysis. APOE and Alzheimer Disease Meta Analysis Consortium ApoE e2 and aging-related outcomes in 379,000 UK Biobank participants The cytokine storm and COVID-19 Cytokine release syndrome in severe COVID-19 Clinical Characteristics of Coronavirus Disease 2019 in China Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study Causally Associations of Blood Lipids Levels with COVID-19 Risk: Mendelian Randomization Study Cardiometabolic traits, sepsis and severe covid-19 with respiratory failure: a Mendelian randomization investigation Obesity has an impact on COVID-19 susceptibility and severity: a two-sample Mendelian randomization study Mendelian randomization analysis to characterize causal association between coronary artery disease and Dissecting the genetic relationship between cardiovascular risk factors and Alzheimer's disease Dyslipidemia and the risk of Alzheimer's disease Pathological findings of COVID-19 associated with acute respiratory distress syndrome Inhibition of Bruton tyrosine kinase in patients with severe COVID-19 Systemic inflammation and Alzheimer's disease Brain-immune communication pathways Systemic inflammation, blood-brain barrier vulnerability and cognitive/non-cognitive symptoms in Alzheimer disease: relevance to pathogenesis and therapy O' blood type is associated with larger grey-matter volumes in the cerebellum Rho GTPases as therapeutic targets in Alzheimer's disease SARS-CoV-2 receptor ACE2 and TMPRSS2 are primarily expressed in bronchial transient secretory cells Boning up: amino-bisphophonates as immunostimulants and endosomal disruptors of dendritic cell in SARS-CoV-2 infection