Dalmasso et al. Translational Psychiatry (2019)9:55 https://doi.org/10.1038/s41398-019-0394-9 Translational Psychiatry ARTICLE Open Access Transethnic meta-analysis of rare coding variants in PLCG2, ABI3, and TREM2 supports their general contribution to Alzheimer’s disease Maria Carolina Dalmasso1, Luis Ignacio Brusco2,3,4, Natividad Olivar2,3, Carolina Muchnik5, Claudia Hanses6, Esther Milz7, Julian Becker6, Stefanie Heilmann-Heimbach8,9, Per Hoffmann 8,9,10, Federico A. Prestia1, Pablo Galeano 1, Mariana Soledad Sanchez Avalos11, Luis Eduardo Martinez4, Mariana Estela Carulla4, Pablo Javier Azurmendi 5, Cynthia Liberczuk2, Cristina Fezza5, Marcelo Sampaño2, Maria Fierens2, Guillermo Jemar2, Patricia Solis12, Nancy Medel12, Julieta Lisso12, Zulma Sevillano12, Paolo Bosco13, Paola Bossù14, Gianfranco Spalletta15, Daniela Galimberti16, Michelangelo Mancuso17, Benedetta Nacmias18, Sandro Sorbi18,19, Patrizia Mecocci20, Alberto Pilotto21, Paolo Caffarra22,23, Francesco Panza 24, Maria Bullido25,26,27, Jordi Clarimon26,28, Pascual Sánchez-Juan29, Eliecer Coto30, Florentino Sanchez-Garcia31, Caroline Graff32,33, Martin Ingelsson34, Céline Bellenguez 35,36,37, Eduardo Miguel Castaño1, Claudia Kairiyama4, Daniel Gustavo Politis4, Silvia Kochen12, Horacio Scaro11, Wolfgang Maier38,39, Frank Jessen38,40, Carlos Alberto Mangone2, Jean-Charles Lambert 35,36,37, Laura Morelli1 and Alfredo Ramirez 7,39 1234567890():,; 1234567890():,; 1234567890():,; 1234567890():,; Abstract Rare coding variants in TREM2, PLCG2, and ABI3 were recently associated with the susceptibility to Alzheimer’s disease (AD) in Caucasians. Frequencies and AD-associated effects of variants differ across ethnicities. To start filling the gap on AD genetics in South America and assess the impact of these variants across ethnicity, we studied these variants in Argentinian population in association with ancestry. TREM2 (rs143332484 and rs75932628), PLCG2 (rs72824905), and ABI3 (rs616338) were genotyped in 419 AD cases and 486 controls. Meta-analysis with European population was performed. Ancestry was estimated from genome-wide genotyping results. All variants show similar frequencies and odds ratios to those previously reported. Their association with AD reach statistical significance by meta-analysis. Although the Argentinian population is an admixture, variant carriers presented mainly Caucasian ancestry. Rare coding variants in TREM2, PLCG2, and ABI3 also modulate susceptibility to AD in populations from Argentina, and they may have a European heritage. Correspondence: Laura Morelli (lmorelli@leloir.org.ar) or Alfredo Ramirez Introduction (alfredo.ramirez@uk-koeln.de) Alzheimer’s disease (AD) is the most common form of 1 Laboratory of Amyloidosis and Neurodegeneration, Fundación Instituto dementia, and has an estimated genetic component of Leloir-IIBBA-CONICET, Ciudad Autónoma de Buenos Aires (C.A.B.A.), Buenos 60–80%1. Over the last decade, more than 20 loci con- Aires, Argentina 2 Centro de Neuropsiquiatría y Neurología de la Conducta (CENECON), Facultad taining common genetic variants (minor allele frequency de Medicina, Universidad de Buenos Aires (UBA), C.A.B.A, Buenos Aires, (MAF) >5%) have been associated with AD2. The advent Argentina of new genetic sequencing technologies has enabled the Full list of author information is available at the end of the article. These authors contributed equally: Maria Carolina Dalmasso, Luis Ignacio identification of several rare variants (MAF <1%) with Brusco, Laura Morelli, Alfredo Ramirez © The Author(s) 2019 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Dalmasso et al. Translational Psychiatry (2019)9:55 Page 2 of 6 moderate effects on AD susceptibility3. In 2017, the function, of which one or more should be altered; a International Genomics of Alzheimer’s Project (IGAP) computerized tomography and/or magnetic resonance reported four rare coding variants significantly associated imaging to assess cortical–hippocampal atrophy and with AD4, all of them being involved in microglial- vascular events, and a blood test analysis to exclude mediated innate immunity. Two of them are novel non- metabolic or infectious causes of dementia. Individuals synonymous variants: a protective one in PLCG2 were included as controls if neurocognitive and clinical (rs72824905) and a risk one in ABI3 (rs616338). The other assessments were normal. two were previously reported in the susceptibility gene TREM2 (rs143332484 and rs75932628), and are respon- Genotyping and statistics sible for p.R62H and p.R47H substitutions, respectively. Genomic DNA was isolated using standard procedures The findings in TREM2 have been consistently replicated from whole blood or saliva samples. TREM2 (rs143332484 in Caucasian4–6 and African-American populations7. and rs75932628), PLCG2 (rs72824905), and ABI3 However, the association of TREM2 p.R47H with AD (rs616338) variants were genotyped using custom- could not be found in East Asian population, because its designed TaqMan assays (Thermo Fisher). Assay accu- frequency is extremely low8,9. This latter observation racy was checked by including positive and negative suggests ethnic variability fostering investigation of these controls in each experiment. APOE alleles were deter- variants in different ethnic groups. Given the increasing mined by genotyping rs429358 and rs7412. Association population diversity observed in countries all over the with AD was calculated using Fisher’s exact test with world, understanding population-shared and -specific risk statistical significance of p < 0.05. All variants were in factors of AD will translate into improved and specific Hardy–Weinberg equilibrium (HWE, p > 0.05). Power prevention and/or treatment for people. calculations were performed using Genetic Power Cal- Latin America is a vast territory, with a wide diverse culator for discrete traits (http://zzz.bwh.harvard.edu/ admixture of European, Native American, and African gpc/cc2.html). Meta-analysis for the effect of rare var- ancestral populations. The genetic architecture of spora- iants in association with AD was conducted using beta dic AD has not been studied in this population beyond and standard error with Metafor R-package11. Populations APOE-ε4. In this report, we provide the first evidence for from France (n = 8514), Italy (n = 2306), Spain (n = an association between TREM2, PLCG2, and ABI3 rare 3966), and Sweden (n = 2286) from the European Alz- variants and AD in the Argentinian population. heimer's Disease Initiative (EADI)4 were included in the meta-analysis. Methods Subjects Ancestry of the population Individuals with AD and without cognitive impairment European (CEU, n = 85), Yorubas African (AFR, n = older than 60 years were recruited from outpatient Neu- 88), and Native American (NAM, n = 46) ancestral rology Departments of the following hospitals: Instituto populations were obtained from 1000 Genomes (http:// de Investigaciones Médicas “Alfredo Lanari” and Hospital www.internationalgenome.org/). Argentinian samples de Clínicas (Buenos Aires City), Hospital Interzonal (ARG) were subjected to genome-wide genotyping using General de Agudos “Eva Perón” (General San Martín the Infinium Global Screening Array (GSA) v.1.0+GSA county), Hospital El Cruce “Dr. Néstor Kirchner” (Flor- shared custom content (Illumina). Quality controls (QC) encio Varela county), and several assistance centers were performed as described before12, using PLINK v1.913 located across Jujuy province. Only samples from people and R v3.4.414. After QC, remaining samples (n = 834) born in Argentina or in South America were included in have <5% of missing genotypes and passed sex-check and the analysis. This study was approved by the ethical identity-by-state filters. Remaining single nucleotide committee “Comité de Bioética Fundación Insituto Leloir polymorphisms (SNPs) have >95% call rate, MAF >1%, are (HHS IRB #00007572, IORG # 006295, FWA00020769)” in HWE (p > 10−6), and without differences in call rate by the approval of protocol CBFIL #22. All participants between cases and controls (p < 1 × 10−5). and/or family members gave their informed consent. For the ancestry analyses, overall population ancestry Diagnosis of AD followed diagnostic criteria from the was first evaluated for ARG by extracting 446 ancestry National Institute of Neurological and Communicative informative markers (AIMs), which were specifically Disorders and Stroke and the Alzheimer’s disease and selected to estimate ancestry in Latin America15, from the Related Disorders Association (NINCDS-ADRDA)10. The genotyped data. Second, ancestry of chromosomes con- diagnosis of AD includes a clinical examination to eval- taining rare variants was evaluated by extracting from the uate functionality and activities of daily living that should 446 AIMS, the AIMs in chromosome 6 from people be compromised; a complete panel of neurocognitive tests carrying TREM2 p.R47H and TREM2 p.R62H, those to evaluate memory, attention, language, and executive AIMs in chromosome 16 from people carrying PLCG2 Dalmasso et al. Translational Psychiatry (2019)9:55 Page 3 of 6 Table 1 Argentinian sample demographics No. of subjects Age (years) AAO mean MMSE mean CDR mean APOE freq (%) APOE-ε4 (female %) (SD) (SD) (SD) carriers (%) Mean (SD) Range ε2 ε3 ε4 Cases 419 (64.4) 77.2 (6.3) 62–96 72.5 (6.5) 18.3 (5.8) 1.4 (0.75) 3.5 69.5 27.0 45.6 Controls 486 (65.6) 74.6 (7.5) 59–105 28.5 (1.2) 0.3 (0.3) 5.5 84.2 10.4 19.6 SD standard deviation, AAO age at onset, MMSE Mini-Mental State Examination, CDR Clinical Dementia Rating scale, APOE freq apolipoprotein E allele frequency Table 2 Genotyping results for TREM2, PLCG2, and ABI3 Gene Protein variation MAF cases MAF controls Allele cases Alleles controls OR 95% CI P value ORIGAP TREM2 p.R47H 0.005 0.001 4|816 1|948 4.68 0.46–230.84 0.19 2.46 TREM2 p.R62H 0.012 0.009 10|816 9|956 1.31 0.47–3.68 0.64 1.67 PLCG2 p.P522R 0.004 0.006 3|810 6|944 0.58 0.09–2.73 0.52 0.68 ABI3 p.S209F 0.012 0.004 9|772 4|912 2.70 0.75–12.07 0.10 1.43 MAF minor allele frequency, OR odds ratio, CI confidence interval, IGAP International Genomics of Alzheimer’s Project p.P522R, and finally AIMs in chromosome 17 from people Table 3 Contribution of Argentinian samples to meta- carrying ABI3 p.S209F. For each ancestry estimation, the analysis same AIMs were extracted from CEU, AFR, and NAM. Gene Protein Populations OR 95% CI P value I2 Ancestry was predicted using ADMIXTURE v1.3.016. variation Results TREM2 p.R47H EADI 2.10 1.04–4.27 0.04 0.00 To evaluate the association of the four rare variants EADI+ARG 2.29 1.17–4.47 0.02 0.00 recently reported by IGAP4 with AD in an Argentinian population sample (ARG), 905 participants were recruited PLCG2 p.P522R EADI 0.60 0.35–1.03 0.06 4.17 from different regions of the country. Demographic and EADI+ARG 0.60 0.36–0.99 0.05 0.00 clinical information of the 419 AD cases and 486 controls ABI3 p.S209F EADI 1.49 0.90–2.48 0.12 0.00 is summarized in Table 1. We first explored the risk effect EADI+ARG 1.58 0.98–2.57 0.06 0.00 of APOE-ε4 allele on AD susceptibility confirming thereby previous reports (odds ratio (OR) = 3.14, p < 0.0001)17. 2 OR odds ratio, CI confidence interval, I heterogeneity statistic For APOE-ε2, we observed the expected protective effect, although it did not reach statistical significance (OR = gain in power is enough to help in reaching statistical 0.77, p < 0.33). Next, we genotyped the recently described significance for the variants analyzed, in particular the rare variants4, i.e., TREM2 p.R47H (rs75932628) and p. TREM2 p.R47H variant (Table 3, Figure S1). Unfortu- R62H (rs143332484), PLCG2 p.P522R (rs72824905) and nately, information for TREM2 p.R62H was not available. ABI3 p.S209F (rs616338). All of them were detected in This can be explained by the observation that the variant ARG with MAFs similar to those reported by IGAP (Table effects in ARG are similar to those detected in the other 2)4. They also showed similar magnitude of association, European populations analyzed (as indicated by hetero- even though neither of these variants reached statistical geneity statistic (I2), see Table 3). All these results toge- significance (Table 2)4. This observation was expected, ther support the hypothesis that these rare variants are since our sample had a power of 60% to detect OR = 3 of also associated to AD in ARG at a similar level than the a variant with MAF = 0.01. Notwithstanding, the fact that one observed in Europe. all variants showed similar effect sizes and effect direc- Although it is generally accepted that the population tions as in the report of IGAP prompted us to perform a from Argentina is mostly originated from Europe, several meta-analysis using samples from France, Italy, Spain, and studies have shown that this population is an admixture of Sweden (EADI)4. Despite larger sample size, statistical predominantly European and Native American ancestry18. power associated with EADI does not allow to reach To estimate the ancestry of ARG, we used a panel of 446 nominal significant association with AD risk. However, SNPs, reported to be precisely balanced to study Latin when meta-analyzing both EADI and ARG samples, this American populations15. ARG showed to be an admixture Dalmasso et al. Translational Psychiatry (2019)9:55 Page 4 of 6 Fig. 1 Ancestry analysis of DNA samples that passed quality controls. a Distribution of genetic ancestry in Alzheimer’s disease (AD) cases and controls. Bar-plots represent each participant on the x-axis, and his percent of European (CEU), African (AFR), and Native American (NAM) ancestry on the y-axis. b Ancestry of rare variant carriers. Box-plots show ancestry composition in percent of people carrying TREM2 p.R47H (n = 3), TREM2 p.R62H (n = 16), PLCG2 p.P522R (n = 6), and ABI3 p.S209F (n = 11) mutations of mainly CEU and NAM, and to a lesser extent of AFR the susceptibility to AD in this population. In addition, we (Fig. S2). Proportions of ancestries are equally distributed confirmed, as previously reported18, that ARG is an among cases and controls (Fig. 1a), indicating that asso- admixture of mainly NAM and CEU. NAM ancestry ciation analysis may not be biased by population stratifi- stemmed from the first settlers of the Americas, who cation. Furthermore, we looked at the ancestry of people originated from an East Asian population that migrated carrying the rare variant mutations in ARG (Fig. S3) and from Siberia19. On the other hand, CEU ancestry is a detected that CEU component was predominant in all the consequence of the pro-immigration legislation to popu- carriers (Fig. 1b). Although ancestry estimation at the late Argentina during nineteenth–twentieth century20. In specific locus was not possible due to the lack of AIMs in this context, we observed that while APOE-ε4 OR was close proximity to the rare variants, the ancestry of the similar to that reported for Caucasians (ORARG = 3.14 vs chromosomes containing the rare variants showed to be ORCEU = 3.6, http://www.alzgene.org/), its frequency in CEU (Fig. 2), suggesting a European heritage for the AD cases was lower (26.9% in ARG vs. 38% in CEU, studied variants. http://www.alzgene.org/). This lower frequency is in agreement with data previously reported in other Latin Discussion American countries21. Interestingly, it is among the lowest Here we show information from a case–control study worldwide together with that of Mediterranean basin and performed in Argentina, being the first study on sporadic Native Americans22,23, which are the main contributors to AD genetics in South America, beyond APOE. Our results Argentinian admixture. strongly suggest that rare coding variants described by It is of note that the rare variants studied here showed IGAP in TREM2, PLCG2, and ABI3 might also modulate MAFs similar to those reported by IGAP in Caucasians4. Dalmasso et al. Translational Psychiatry (2019)9:55 Page 5 of 6 Fig. 2 Ancestry of chromosomes containing the rare variants. Principal component analysis (PCA) of ancestry results for a chromosome 6, containing TREM2 p.R47H (black) and TREM2 p.R62H (gray); b chromosome 16, containing PLCG2 p.P522R (black); and c chromosome 17, containing ABI3 p.S209F. Ancestral populations are European (blue), African (red), and Native American (green). Percent of distribution explained by each principal component (PC) it is shown in parenthesis Acknowledgements Unfortunately, since there are no reports in Amer- This work was supported by grants from the International Society for indians, we could not compare their MAFs. However, Neurochemistry (ISN) and Alexander von Humboldt Foundation (to M.C.D.); TREM2 p.R47H is almost absent in East Asians8,9, sug- Agencia Nacional de Promoción Científica y Tecnológica (PBIT/09 2013, PICT- 2015-0285 and PICT-2016-4647 to L.M.; PICT-2014-1537 to M.C.D.); GENMED gesting that it might also be extremely rare in Native Labex and JPND PERADES grant; and JPND EADB grant (German Federal Americans. For the rest of the variants, we performed a Ministry of Education and Research, BMBF: 01ED1619A). search in ExaC database (exac.broadinstitute.org), and found that PLCG2 p.P522R, and TREM2 p.R62H were Author details 1 not detected in approximately 4300 East Asian indivi- Laboratory of Amyloidosis and Neurodegeneration, Fundación Instituto Leloir-IIBBA-CONICET, Ciudad Autónoma de Buenos Aires (C.A.B.A.), Buenos duals. Unfortunately, results for ABI3 p.S209F are not Aires, Argentina. 2Centro de Neuropsiquiatría y Neurología de la Conducta reliable due to low quality. Notwithstanding, these (CENECON), Facultad de Medicina, Universidad de Buenos Aires (UBA), C.A.B.A, observations, together with our results showing that the Buenos Aires, Argentina. 3Departamento Ciencias Fisiológicas UAII, Facultad de Medicina, UBA, C.A.B.A, Buenos Aires, Argentina. 4Hospital Interzonal General chromosomes containing these rare variants have de Agudos Eva Perón, San Martín, Buenos Aires, Argentina. 5Laboratorio de mainly Caucasian ancestry in the identified carriers, Bioquímica Molecular, Facultad de Medicina, Instituto de Investigaciones strongly suggest a European origin for these variants. Médicas A. Lanari, UBA, C.A.B.A, Buenos Aires, Argentina. 6Department of Psychiatry and Psychotherapy, University of Bonn, 53127 Bonn, Germany. However, the possibility still remains open that the loci 7 Division of Neurogenetics and Molecular Psychiatry, Department of Psychiatry containing the studied rare variants might have an and Psychotherapy, University of Cologne, 50937 Cologne, Germany. 8Institute ancestry other than Caucasian. To answer this question, of Human Genetics, University of Bonn, School of Medicine & University Hospital, 53127 Bonn, Germany. 9Department of Genomics, Life & Brain Center, additional studies comparing SNPs among ancestral University of Bonn, 53127 Bonn, Germany. 10Division of Medical Genetics, populations are needed to identify AIMs that better University Hospital and Department of Biomedicine, University of Basel, CH- explain the ancestry of these loci. 4058 Basel, Switzerland. 11Ministerio de Salud de la Provincia de Jujuy, Programa del Adulto Mayor, San Salvador de Jujuy, Jujuy, Argentina. In conclusion, we report the first genetic data from 12 Neurosciences and Complex Systems Unit (EnyS), CONICET, Hospital El Cruce Argentinian population, which support the contribution “Dr. Néstor Kirchner”, Univ Arturo Jauretche, F. Varela, Buenos Aires, Argentina. 13 of rare coding variants to AD susceptibility. Although this Instituto di Ricovero e Cura a Carattere Scientifico (IRCCS) Associazione Oasi Maria Santissima Srl, Troina, Italy. 14Department of Clinical and Behavioural population size is not enough to reach statistical sig- Neurology, Experimental Neuropsychobiology Laboratory, Rome, Italy. nificance for the rare variants studied here, it is a relevant 15 Department of Clinical and Behavioural Neurology, Neuropsychiatry opportunity to start filling the gap on AD genetic archi- Laboratory, IRCCS Santa Lucia Foundation, Rome, Italy. 16Neurodegenerative Diseases Center, University of Milan, Centro Dino Ferrari, Fondazione Ca′ tecture in Latin American admixed populations. Our Granda, IRCCS Ospedale Maggiore Policlinico, Milan, Italy. 17Department of analysis fosters further analysis of these rare variants in Experimental and Clinical Medicine, Neurological Institute, University of Pisa, other Latin populations to confirm our initial observation. Pisa, Italy. 18NEUROFARBA (Department of Neuroscience, Psychology, Drug Research and Child Health), University of Florence, Florence, Italy. 19IRCCS ′Don Importantly, expanding research to admixed populations, Carlo Gnocchi′, Florence, Italy. 20Section of Gerontology and Geriatrics, like this one, will help to identify potential population- Department of Medicine, University of Perugia, Perugia, Italy. 21Geriatric Unit specific effects on the genetic structure of AD, in addition and Gerontology-Geriatrics Research Laboratory, Department of Medical Sciences, IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy. to better define conserved relevant pathways involved in 22 Section of Neuroscience, DIMEC, University of Parma, Parma, Italy. the disease. 23 Alzheimer Center, FERB, Gazzaniga, Bergamo, Italy. 24Neurodegenerative Dalmasso et al. Translational Psychiatry (2019)9:55 Page 6 of 6 Disease Unit, Department of Basic Medicine, Neuroscience, and Sense Organs, 3. Del-Aguila, J. L. et al. Alzheimer’s disease: rare variants with large effect sizes. University of Bari Aldo Moro, Bari, Italy. 25Instituto de Investigación Sanitaria Curr. Opin. Genet. Dev. 33, 49–55 (2015). Hospital la Paz (IdiPAZ), Madrid, Spain. 26Centro de Investigación Biomédica en 4. Sims, R. et al. Rare coding variants in PLCG2, ABI3, and TREM2 implicate Red de Enfermedades Neurodegenerativas (CIBERNED), Instituto de Salud microglial-mediated innate immunity in Alzheimer’s disease. Nat. Genet. 49, Carlos III, Madrid, Spain. 27Centro de Biología Molecular Severo Ochoa (CSIC- 1373–1384 (2017). UAM), Madrid, Spain. 28Memory Unit, Neurology Department and Sant Pau 5. Benitez, Ba et al. TREM2 is associated with the risk of Alzheimer’s disease in Biomedical Research Institute, Hospital de la Santa Creu i Sant Pau, Spanish population. Neurobiol. Aging 34, 1711.e15–7 (2013). Autonomous University of Barcelona, Barcelona, Spain. 29Neurology Service 6. Finelli, D. et al. TREM2 analysis and increased risk of Alzheimer’s disease. and CIBERNED, ′Marqués de Valdecilla′ University Hospital (University of Neurobiol. Aging 36, 546.e9–546.e13 (2015). Cantabria and IDIVAL), Santander, Spain. 30Molecular Genetics Laboratory- 7. Jin, S. C. et al. TREM2 is associated with increased risk for Alzheimer’s disease in Hospital, University of Central Asturias, Oviedo, Spain. 31Immunology Service, African Americans. Mol. Neurodegener. 10, 19 (2015). Hospital Universitario de Gran Canaria Doctor Negrín, Las Palmas de Gran 8. Miyashita, A. et al. Lack of genetic association between TREM2 and late-onset Canaria, Spain. 32Genetics Unit, Theme Aging, Karolinska University Hospital, Alzheimer’s disease in a Japanese population. J. Alzheimers Dis. 41, 1031–1038 Solna, Sweden. 33Division of Neurogeriatrics, Department NVS, Karolinska (2014). Institutet, Bioclincum J10:20, Solna, Sweden. 34Department of Public Health/ 9. Huang, M. et al. Lack of genetic association between TREM2 and Alzheimer’s Geriatrics, Uppsala University, Uppsala, Sweden. 35INSERM, U1167, RID-AGE-Risk disease in East Asian population. Am. J. Alzheimer’s Dis. Other Dement. 30, Factors and Molecular Determinants of Aging-Related Diseases, F-59000 Lille, 541–546 (2015). France. 36Institut Pasteur de Lille, F-59000 Lille, France. 37University Lille, U1167- 10. McKhann, G. M. et al. The diagnosis of dementia due to Alzheimer’s disease: Excellence Laboratory LabEx DISTALZ, F-59000 Lille, France. 38German Center recommendations from the National Institute on Aging-Alzheimer’s Associa- for Neurodegenerative Diseases (DZNE), 53127 Bonn, Germany. 39Department tion workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimer’s of Neurodegenerative Diseases and Geriatric Psychiatry, University of Bonn, Dement. 7, 263–269 (2011). 53127 Bonn, Germany. 40Department of Psychiatry and Psychotherapy, 11. Viechtbauer, W. Conducting meta-analyses in R with the metafor package. J. University of Cologne, 50937 Cologne, Germany Stat. Softw. 36, 1–48 (2010). 12. Anderson, C. A. et al. Data quality control in genetic case-control association Conflict of interest studies. Nat. Protoc. 5, 1564–1573 (2010). The authors declare that they have no conflict of interest. 13. Chang, C. C. et al. Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience 4, 7 (2015). Informed consent 14. R Core Team. R: A Language and Environment for Statistical Computing (R All participants had agreed by signed informed consent to participate in Foundation for Statistical Computing, Vienna, 2018). genetic studies approved by our Institutional Review Board. 15. Galanter, J. M. et al. Development of a panel of genome-wide ancestry informative markers to study admixture throughout the americas. PLoS Genet. 8, e1002554 (2012). 16. Alexander, D. H., Novembre, J. & Lange, K. Fast model-based estimation of Publisher’s note ancestry in unrelated individuals. Genome Res. 19, 1655–1664 (2009). Springer Nature remains neutral with regard to jurisdictional claims in 17. Morelli, L., Leoni, J., Castano, E. M., Mangone, C. A. & Lambierto, A. Apolipo- published maps and institutional affiliations. protein E polymorphism and late onset Alzheimer’s disease in Argentina. J. Neurol. Neurosurg. Psychiatry 61, 426–427 (1996). Supplementary information accompanies this paper at (https://doi.org/ 18. Avena, S. et al. Heterogeneity in genetic admixture across different regions of 10.1038/s41398-019-0394-9). argentina. PLoS One 7, e34695 (2012). 19. Bodner, M. et al. Rapid coastal spread of First Americans: novel insights from Received: 2 August 2018 Revised: 26 November 2018 Accepted: 1 January South America’s Southern Cone mitochondrial genomes. Genome Res. 22, 2019 811–820 (2012). 20. Devoto, F. J. Argentine migration policy and movements of the European population (1876-1925). Estud. Migr. Latinoam. 4, 135–158 (1989). 21. Jacquier, M. et al. APOE e4 and Alzheimer’s disease: positive association in a Colombian clinical series and review of the Latin-American studies. Arq. References Neuropsiquiatr. 59, 11–17 (2001). 1. Gatz, M. et al. Role of genes and environments for explaining Alzheimer 22. Ward, A. et al. Prevalence of Apolipoprotein E4 genotype and homozygotes Disease. Arch. Gen. Psychiatry 63, 168 (2006). (APOE e4/4) among patients diagnosed with Alzheimer’s disease: a systematic 2. Naj, A. C. & Schellenberg, G. D. Genomic variants, genes, and pathways of review and meta-analysis. Neuroepidemiology 38, 1–17 (2012). Alzheimer’s disease: an overview. Am. J. Med Genet. B Neuropsychiatr. Genet. 23. Henderson, J. N. et al. Apolipoprotein E4 and tau allele frequencies among 174, 5–26 (2017). Choctaw Indians. Neurosci. Lett. 324, 77–79 (2002).