Acta Neuropathologica (2019) 138:237–250 https://doi.org/10.1007/s00401-019-02026-8 ORIGINAL PAPER A nonsynonymous mutation in PLCG2 reduces the risk of Alzheimer’s disease, dementia with Lewy bodies and frontotemporal dementia, and increases the likelihood of longevity Sven J. van der Lee1,2 · Olivia J. Conway3 · Iris Jansen1,4 · Minerva M. Carrasquillo3 · Luca Kleineidam5,6,7 · Erik van den Akker8,9 · Isabel Hernández10,11 · Kristel R. van Eijk12 · Najada Stringa13 · Jason A. Chen14 · Anna Zettergren15 · Till F. M. Andlauer16,17,18 · Monica Diez‑Fairen19,20 · Javier Simon‑Sanchez21,22 · Alberto Lleó11,23 · Henrik Zetterberg24,25,26 · Marianne Nygaard27 · Cornelis Blauwendraat28 · Jeanne E. Savage4 · Jonas Mengel‑From29 · Sonia Moreno‑Grau10 · Michael Wagner5,6 · Juan Fortea11,23 · Michael J. Keogh30,31 · Kaj Blennow24,25 · Ingmar Skoog15 · Manuel A. Friese18,32 · Olga Pletnikova33 · Miren Zulaica11,34 · Carmen Lage11,35,36 · Itziar de Rojas10,11 · Steffi Riedel‑Heller37 · Ignacio Illán‑Gala11,23 · Wei Wei31 · Bernard Jeune29 · Adelina Orellana10,11 · Florian Then Bergh18,38 · Xue Wang3 · Marc Hulsman1,2,9 · Nina Beker1 · Niccolo Tesi1,2,9 · Christopher M. Morris39 · Begoña Indakoetxea11,34,40 · Lyduine E. Collij41 · Martin Scherer42 · Estrella Morenas‑Rodríguez11,23 · James W. Ironside43 · Bart N. M. van Berckel41 · Daniel Alcolea11,23 · Heinz Wiendl18,44 · Samantha L. Strickland3 · Pau Pastor19,20 · Eloy Rodríguez Rodríguez11,35,36 · DESGESCO (Dementia Genetics Spanish Consortium), EADB (Alzheimer Disease European DNA biobank) · EADB (Alzheimer Disease European DNA biobank) · IFGC (International FTD-Genomics Consortium), IPDGC (The International Parkinson Disease Genomics Consortium) · IPDGC (The International Parkinson Disease Genomics Consortium) · RiMod-FTD (Risk and Modifying factors in Fronto-Temporal Dementia) · Netherlands Brain Bank (NBB) · Bradley F. Boeve45 · Ronald C. Petersen45 · Tanis J. Ferman46 · Jay A. van Gerpen47 · Marcel J. T. Reinders64 · Ryan J. Uitti47 · Lluís Tárraga10,11 · Wolfgang Maier5,6 · Oriol Dols‑Icardo11,23 · Amit Kawalia7 · Maria Carolina Dalmasso7,48 · Mercè Boada10,11 · Uwe K. Zettl18,49 · Natasja M. van Schoor13 · Marian Beekman8 · Mariet Allen3 · Eliezer Masliah50 · Adolfo López de Munain11,34,51 · Alexander Pantelyat52 · Zbigniew K. Wszolek47 · Owen A. Ross3 · Dennis W. Dickson3 · Neill R. Graff‑Radford47 · David Knopman45 · Rosa Rademakers3 · Afina W. Lemstra1 · Yolande A. L. Pijnenburg1 · Philip Scheltens1 · Thomas Gasser53 · Patrick F Chinnery31,54 · Bernhard Hemmer17,18,55 · Martijn A. Huisman13,56 · Juan Troncoso33 · Fermin Moreno11,34,40 · Ellen A. Nohr57 · Thorkild I. A. Sørensen58,59,60 · Peter Heutink21,22 · Pascual Sánchez‑Juan11,35,36 · Danielle Posthuma2,4 · The GIFT (Genetic Investigation in Frontotemporal Dementia and Alzheimer’s Disease) Study Group · Jordi Clarimón11,23 · Kaare Christensen27,61,62 · Nilüfer Ertekin‑Taner3,47 · Sonja W. Scholz28,52 · Alfredo Ramirez5,7 · Agustín Ruiz10,11 · Eline Slagboom8,63 · Wiesje M. van der Flier1 · Henne Holstege1,2 Received: 23 March 2019 / Revised: 3 May 2019 / Accepted: 5 May 2019 / Published online: 26 May 2019 © The Author(s) 2019 Abstract The genetic variant rs72824905-G (minor allele) in the PLCG2 gene was previously associated with a reduced Alzheimer’s disease risk (AD). The role of PLCG2 in immune system signaling suggests it may also protect against other neurodegenera- tive diseases and possibly associates with longevity. We studied the effect of the rs72824905-G on seven neurodegenerative diseases and longevity, using 53,627 patients, 3,516 long-lived individuals and 149,290 study-matched controls. We rep- licated the association of rs72824905-G with reduced AD risk and we found an association with reduced risk of dementia with Lewy bodies (DLB) and frontotemporal dementia (FTD). We did not find evidence for an effect on Parkinson’s disease Electronic supplementary material The online version of this article (https​://doi.org/10.1007/s0040​1-019-02026​-8) contains supplementary material, which is available to authorized users. Extended author information available on the last page of the article 13 Vol.:(0123456789) 238 Acta Neuropathologica (2019) 138:237–250 (PD), amyotrophic lateral sclerosis (ALS) and multiple sclerosis (MS) risks, despite adequate sample sizes. Conversely, the rs72824905-G allele was associated with increased likelihood of longevity. By-proxy analyses in the UK Biobank sup- ported the associations with both dementia and longevity. Concluding, rs72824905-G has a protective effect against multiple neurodegenerative diseases indicating shared aspects of disease etiology. Our findings merit studying the PLCγ2 pathway as drug-target. Keywords Alzheimer’s disease · Frontotemporal dementia · Dementia with Lewy bodies · Progressive supranuclear palsy · Parkinson’s disease · Amyotrophic lateral sclerosis · Multiple sclerosis · Neurodegenerative disease · Longevity · PLCG2 · Phospholipase C Gamma 2 Introduction could lead to an increased likelihood to survive to old age, we tested whether rs72824905-G associated with longevity. The protein product of the phospholipase Cγ2 (PLCG2) gene is involved in the transmembrane transduction of immune signals [30, 42, 45] that determine the fate and function of Materials and methods various immune cell types, both in the periphery and the brain [42, 45]. It is known that gain-of-function mutations Study populations and genotyping in the PLCG2 gene cause autoimmune disorders [40, 46, 58, 59] and resistance to treatment of chronic lymphocytic We present a short description of 16 cohorts, often includ- leukemia [56]. ing multiple sites or studies, which contributed to this In 2017, a genome-wide association (GWA) study of manuscript in Suppl. Table 1, Online Resource. Studies Alzheimer’s disease (AD) showed that the rare nonsyn- were approved by corresponding ethics committees and onymous variant in the PLCG2 gene (rs72824905-G; informed consent was obtained for all participants (Suppl. p.Pro522Arg; NC_000016.9:g.81942028C > G) reduced Table 1, Online Resource). Study characteristics (age, per- AD risk (OR = 0.68, p = 5.4 × 10−10) [47]. In both mouse centage female, apolipoprotein E (APOE) status and age) and human brain tissues, PLCG2 has been shown to be over- are described in Suppl. Table 2, Online Resource. In most expressed > 6-log fold in microglia compared to other brain cohorts, the average age of the controls was lower than that cells [12]. Further, PLCG2 has higher expression levels in of cases (Suppl. Fig 3, Online Resource). We determined pathologically affected brain regions of AD patients, which rs72824905-G genotypes (NC_000016.9:g.81942028C > G, seems to be driven by microgliosis [7]. Since microglia are p.Pro522Arg) using direct genotyping with a variety of gen- the brain’s immune cells, these findings suggest an impor- otyping arrays or TaqMan genotyping. If direct genotyping tant role for PLCG2 in the neural immune response. Next to was not available, we used imputation to 1000 Genomes PLCG2, GWA studies of AD identified additional immune- phase I version 3 [15] or the Haplotype Reference Consor- and microglia-related genes that associate with AD, e.g. the tium (HRC) reference panels [37]. Details on genotyping or triggering receptor expressed on myeloid cells 2 (TREM2) imputation by study can be found in Suppl. Table 3, Online gene and pathway analysis based on these same GWA stud- Resource. We studied participants from European descent. ies indicate that the immune system plays a key role in the development of AD [47]. Likewise, human genetic studies Study populations of AD, FTD, DLB and PSP patients imply the immune system plays a role in other neurodegen- erative diseases such as frontotemporal dementia (FTD) [3], We compared rs72824905-G genotypes in a total of 4,985 Parkinson’s disease (PD) [13], and multiple sclerosis (MS) AD patients and 9,238 controls from eight cohorts. All sam- [17, 20, 41]. We reasoned that next to AD, PLCG2-related ples were independent from Sims et al. [47], but include the immune signaling may be involved in the etiology of these samples from Conway et al. [7]. We compared in total 2,437 other neurodegenerative diseases. This led us to question FTD patients with 10,647 controls from four studies and two whether the rs72824905-G variant in PLCG2 is also associ- consortia. Further, we studied 1446 DLB patients with 5509 ated with a reduced risk of other neurodegenerative diseases. controls from five cohorts and 882 PSP patients with 3187 Here, we tested whether rs72824905-G protects against controls from five cohorts. Details on sample size by cohort other neurodegenerative diseases. We first tested whether and which cohort contributed to which analysis can be found rs72824905-G associates with reduced risk of AD, FTD, in Suppl. Table 2, Online Resource. dementia with Lewy bodies (DLB), progressive supranu- clear palsy (PSP), PD, amyotrophic lateral sclerosis (ALS) and MS. Since a reduced risk of neurodegenerative diseases 13 Acta Neuropathologica (2019) 138:237–250 239 Study populations of ALS, PD and MS patients For the analysis of longevity-by-proxy, we chose the age of 90 years as a cut-off for the minimum age reached by To study the association of rs72824905-G with ALS, PD and the parents. By principle, phenotype by-proxy analyses suf- MS, we obtained summary statistics from existing GWAS fer from dilution effect [34, 36]; therefore, a more extreme meta-analyses, see Suppl. Table 1, 2, 3, Online Resource, parental age cut-off of 95 years was also studied. In this for study descriptions. We present results of a combined analysis, we compared 35,256 UK Biobank participants who total of 28,448 PD patients that were compared with 108,438 had a mother who reached at least 90 years (7790 moth- controls: data from 27,595 PD patients and 106,951 con- ers reached the age of 95 years) with 342,810 participants trols from the International Parkinson Disease Genomics whose mother did not reach 90 years of age. Likewise, we Consortium (IPDGC) [39] were combined with data from compared 17,558 UK Biobank participants with a father 853 PD patients and 1,487 controls from the Mayo Clinic. who reached at least 90 years (3,043 fathers reached the age Furthermore, we studied 10,953 ALS patients and 20,673 of 95 years) with 353,100 participants whose father did not controls, which represents the subset of the data presented reach 90 years of age. by van Rheenen et al. [54], for which rs72824905-G was imputed with sufficient quality (imputation quality > 0.3). Statistical analysis Last, we studied 4476 MS patients and 5714 controls which were previously described by Dankowski et al. [8]. R (version 3.5.1) was used for all analysis [50]. Logistic regression models were fitted within studies to assess the Study populations of longevity association of rs72824905-G with AD, FTD, DLB, and PSP patients, and long-lived individuals, compared to controls. We investigated the association of rs72824905-G with For each study, we calculated the odds ratio’s (OR) and longevity in five different cohorts; in total, we compared 95% confidence intervals (CI). We accounted for popula- 3516 individuals who reached at least 90 years with 9677 tion substructure by adjusting for principal components or control individuals who died before age 90 years or were by comparing cases and controls from the same study or last screened before 90 years (Suppl. Table 1–3, Online country of origin. We meta-analyzed the effect estimates Resource). The data from Tesi et al. [51] were included in (log(OR)) from the studies using inverse-variance fixed- this study. A subset of 1136 Dutch long-lived individuals effect meta-analyses (R-package ‘rmeta’ v3.0). The fraction for whom follow-up data until death were available [22] was of variance that is due to heterogeneity was estimated by included. In this subset, we compared the survival of carriers the ­I2 statistic [21]. We visualized survival of rs72824905- of rs72824905-G with non-carriers. G carriers compared to non-carriers using Kaplan–Meier curves. Differences in survival were tested using a Cox pro- Studies of dementia and longevity by‑proxy portional hazards model correcting for (age at inclusion, sex in the UK Biobank and relatedness). For MS, the results originate from a single study, which The UK Biobank is a study of genetic and health of a half used ancestry principal components (PCs) to adjust for million people from the United Kingdom [49]. Information population stratification [8]. The statistical methods of from parents or first-degree relatives can be used as a proxy- the GWAS meta-analyses of ALS and PD were previously phenotype for the participants [34]. In this study, we used described [39, 54]. In short, individual cohorts calculated maternal and paternal history of Alzheimer’s/dementia as logistic regression models and then summary statistics of proxy for dementia [34, 36] and the reported age of the par- cohorts were combined using inverse-variance fixed-effect ents (at completing the survey or death) as proxy phenotype meta-analyses. PCs were used to adjust for population strati- for longevity [44]. In the UK Biobank, the rs72824905-G fication. Analysis in the UK Biobank were performed using variant was imputed using the available genotyping arrays logistic regression models adjusted for genotyping array and and the HRC-reference panel as previously described [25]. the first 12 PCs. Effect estimates of the paternal and mater- The maternal and paternal by-proxy phenotypes were ana- nal analysis were combined using inverse-variance fixed- lyzed separate using the genotypes of the participants and effect meta-analysis (R-package ‘rmeta’ v3.0). We reported the results were meta-analyzed. two-sided p values and considered p values < 0.05 as signifi- We compared rs72824905-G genotypes of 32,262 partici- cant; p values are not corrected for multiple testing. pants whose mother was reported to have dementia with the genotypes of 346,999 participants whose mothers did not Power analysis have dementia. Likewise, we compared 16,968 participants whose father had dementia with 358,468 whose fathers did For all diseases studied, we performed power analysis using not have dementia. the online tool Genetic Association Study (GAS) power 13 240 Acta Neuropathologica (2019) 138:237–250 Calculator implementing the methods described in Skol (IQR = 1.9–7.4) while 1108 non-carriers survived a median et al. [48]. We calculated power of our analysis to attain a p of 3.3 years (IQR = 1.4–5.8) (Suppl. Fig 9, Online Resource). value of 0.05 and used the total number of cases and controls However, the difference was not significant (HR 0.75, 95% from our analysis. We assumed an additive model, a minor CI 0.51–1.09, p = 0.078), likely due to the low number of allele frequency of 0.009 and a disease frequency of 0.01 rs72824905-G carriers in the analysis, as a consequence of for all diseases (higher disease frequency assumption would variant rareness (MAF ~ 1%). lead to higher power estimates). We report the power for an OR between 1 and 2. This corresponds to protective OR (the Association with by‑proxy dementia and longevity inverse OR = 1/OR) between 0.50 and 1. In line with the protection against AD, the by-proxy analy- sis showed that PLCG2 variant carriers had a reduced risk Results of having a parent with dementia, OR = 0.88 (0.81–0.95, p = 1.9 × 10−3) (Fig. 2). Next, we tested the association of An overview of study sample, contributing studies, cor- rs72824905-G with longevity-by-proxy. Carriers of the rections applied by study and counts of carriers split by rs72824905-G variant did not have an increased likelihood case–control status is shown in Table 1. of having a parent who reached 90 years of age compared to non-carriers (OR = 1.05, p = 0.24). However, carriers Association with brain diseases did have an increased likelihood of having a parent who reached 95 years (OR = 1.19, p = 2.1 × 10−2). The threshold We replicated the association of rs72824905-G in PLCG2 of 95 years was chosen as too few parents reached the age with a reduced AD risk (OR = 0.57, p = 6.0 × 10−4, I2 = 0%). of 100 years. In addition, we found that rs72824905-G associated with a reduced risk of both DLB (OR = 0.54, p = 0.045, I2 = 0%) Power analysis and FTD (OR = 0.61, p = 0.011, I2 = 0%). In contrast, we found no evidence that rs72824905-G is associated with Power analysis (Suppl. Fig 10, Online Resource) showed PSP (OR = 1.46, p = 0.19, I 2 = 0%), ALS (OR = 1.07, that the PD, MS and ALS analysis had adequate statisti- p = 0.52, I2 = 0%), PD (OR = 1.18, p = 0.10, I2 = 0) and MS cal power (power > 0.8) to detect a protective association (OR = 0.99, p = 0.95). The association of rs72824905-G (p = 0.05) with an OR ~ 0.68 (the OR for AD reported in with these seven brain diseases is shown in Fig. 1. In Suppl. Sims et al. [47].). The PSP analysis had the lowest statistical Figs. 2–7, Online Resource, we show the association esti- power (0.32 at the expected OR = 0.67). mates for each study separately in the meta-analyses for AD, DLB, FTD, PSP, ALS and PD (the MS study consisted of a single study). Discussion Association with longevity A recent study showed the protective effect against AD risk of the p.Pro552Arg nonsynonymous amino acid change in In line with a reduced risk of neurodegenerative diseases, PLCG2 (rs72824905-G) [47]. We replicated this protective we found that rs72824905-G associated with a 1.49 (95% effect in independent AD patients and controls. Addition- CI 1.12–1.98) increased likelihood (p = 6.3 × 10−3, I2 = 0%) ally, we found that the variant also protected against FTD to reach the age of 90 years. Although no heterogene- and DLB, but not against ALS, PD and MS. The analysis ity was observed between studies, it is of interest that a of PSP was inconclusive because of insufficient power. We cohort of centenarians who were selected based on being also found that rs72824905-G associated with increased 100 years old and cognitively healthy (description of ‘100- likelihood of longevity, which is according to expectations, plus Study’ in Suppl. Table 1, Online Resource) was most since overall dementia is the leading cause of death at older enriched with rs72824905-G (OR = 2.36, 95% CI 1.34–4.15, age [1]. Indeed, the strongest effect of PLCG2 variant was p = 2.8 × 10−3) (Suppl. Fig 8, Online Resource). Next, we observed in cognitively healthy centenarians, individuals tested whether carrying the rs72824905-G variant was where an absence of dementia and extreme longevity is com- associated with longer survival after the age of 90 years in bined. Our findings were supported by analyses of by-proxy 1,136 Dutch long-lived individuals of which 96.3% were phenotypes for dementia and longevity in the UK Biobank. followed until death [median age at inclusion 93.2, IQR Taken together, the association of the rs72824905-G variant 91.6–95.0 years, 63% female; mean survival after inclusion with a decreased risk of multiple dementia types and the was 3.3 years; inter quartile range (IQR) 1.4–5.8 years]. increased risk of longevity warrants thorough investigation We found that 28 carriers survived a median of 4.7 years 13 Acta Neuropathologica (2019) 138:237–250 241 Table 1  Study sample description Trait Consortium or combined Studies/sites included Corrections Cases Controls cohort name N N-carriers MAF N N-carriers MAF AD Amsterdam UMC ADC, NBB, LASA PC1-3 1893 24 0.63 2571 64 1.24 Brain compendium Keogh et al. [29] None 277 0 0 362 6 0.83 Mayo Clinic Conway et al. [7] None 1477 19 0.64 1487 29 0.98 NDRU cohort NDRU cohort None 527 7 0.66 343 8 1.17 Spanish cohorts Valdecilla Cohort, Fundació ACE, None 23 0 0 746 10 0.67 Oviedo, Sant Pau (SPIN cohort), San Sebastian Swedish studies GBC Studies, Clinical AD cohort None 564 6 0.53 3480 61 0.88 sweden. UCLA/UCSF GIFT Chen et al. [6] None 224 0 0 249 10 2.01 Combined AD 4985 56 0.56 9238 188 1.02 DLB Amsterdam UMC ADC, NBB, LASA PC1-3 189 2 0.53 2571 64 1.24 Brain compendium Keogh et al. [29] None 97 1 0.52 362 6 0.83 Mayo Clinic Conway et al. [7] None 306 2 0.33 1487 29 0.98 NDRU cohort NDRU cohort None 622 8 0.64 343 8 1.17 Spanish cohorts Valdecilla Cohort, Fundació ACE, None 232 3 0.65 746 10 0.67 Oviedo, Sant Pau (SPIN cohort), San Sebastian Combined DLB 1446 16 0.55 5509 117 1.06 FTD Amsterdam UMC ADC, NBB, LASA PC1-3 231 1 0.22 2571 64 1.24 Brain compendium Keogh et al. [29] None 93 2 1.08 362 6 0.83 IFGC Ferrari et al. [11] None 1360 22 0.81 5059 118 1.17 RiMoD-FTD (Consortium) None 255 3 0.59 1660 38 1.17 Spanish cohorts Valdecilla Cohort, Fundació ACE, None 366 1 0.14 746 10 0.67 Oviedo, Sant Pau (SPIN cohort), San Sebastian UCLA/UCSF GIFT Chen et al. [6] None 132 2 0.76 249 10 2.01 Combined FTD 2437 31 0.64 10,647 246 1.19 PSP Brain compendium Keogh et al. [29] None 17 1 2.94 362 6 0.83 Mayo Clinic Conway et al. [7] None 231 9 1.95 1487 29 0.98 NDRU cohort NDRU cohort None 613 11 0.9 343 8 1.17 UCLA/UCSF GIFT Chen et al. [6] None 12 0 0 249 10 2.01 Combined PSP 873 21 1.20 2441 53 1.09 PD IPDGC Nalls et al. [39] ≥3PCs 27,595 340a 0.81a 106,951 391a 0.81a Mayo Clinic Conway et al. [7] None 853 18 1.06 1487 29 0.98 MS KKNMS Dankowski et al. [3] PC1 and 2 4476 82 0.92 5714 107 0.94 ALS Project MinE Van Rheenen et al. [8] PC1-4 10,953 214 0.98 20,673 385 0.93 longevity AgeCoDe AgeCoDe None 462 14 1.52 861 19 1.12 Amsterdam UMC 100-Plus Study, LASA, NBB PC1-3 293 16 2.73 2571 64 1.24 Danish studies Multiple Danish studies None 853 10 0.59 2793 33 0.59 Leiden Longevity Study LLS, GEHA-NL None 1138 28 1.23 743 11 0.74 GBC Studies GBC Studies None 770 16 1.04 2709 45 0.83 Combined longevity 3516 84 1.19 9677 172 0.89 Consortium or combined cohort name corresponds to the name used in the figures of this manuscript. Studies/sites included or reference to cohort shows the studies combined to form one site (if more then one). Additional information on studies included can be found in supplemen- tary Table 2. If studies/sites include a reference, the exact methods described in the reference were used to obtain the genotypes and association results AD Alzheimer’s disease, FTD frontotemporal dementia, DLB dementia with Lewy bodies, PSP progressive supranuclear palsy, PD Parkinson’s Disease, ALS Amyotrophic Lateral Sclerosis, MS multiple sclerosis, MAF Minor allele frequency, ADC Amsterdam Dementia Cohort, NBB Netherlands Brain Bank, LASA Longitudinal Aging Study Amsterdam, GEHA Genetics of Healthy Ageing Study, NL, AgeCoDe German Study on Ageing, Cognition and Dementia in Primary Care Patients, GBC Gothenburg Birth Cohort Studies, IFGC International FTD-Genomics Con- sortium, IPDGC The International Parkinson Disease Genomics Consortium, KKNMS German Competence Network Multiple Sclerosis, LLS Leiden Longevity study, RiMoD-FTD Risk and modifying factors in Frontotemporal Dementia, UCLA/UCSF Genetic Investigation in Fronto- temporal Dementia and Alzheimer’s Disease (GIFT) Study a The number of carriers and minor allele frequency were calculated in a subset of 21092 cases and 23896 controls. No combined estimate of 13 242 Acta Neuropathologica (2019) 138:237–250 Table 1  (continued) MAF can be given Fig. 1  Association results of rs72824905-G with seven brain diseases and longevity. *P values < 0.05. Numbers (N) of cases (patients or long- lived individuals) and controls studied. The figure shows the odds-ratio (box) of the rs72824905-G with the 95% confidence intervals (whiskers) Fig. 2  Association results of rs72824905-G with dementia by-proxy and longevity by-proxy analysis in the UK Biobank. *P values < 0.05. The figure shows the odds-ratio (box) of the rs72824905-G with the 95% confidence intervals (whiskers) of the molecular mechanisms underlying this protective cannot exclude that rs72824905-G has an independent effect effect. of rs72824905-G on the risk of longevity and/or the risk Thus far, the common APOE ɛ4 allele and the rare of maintaining cognitive health. In line with this observa- TREM2.R47H variant are strong genetic risk factors shared tion, we anecdote one cognitively healthy centenarian who across AD, FTD and DLB (not TREM2 [55]). [2, 11, 18, 27, is homozygous for the APOE ɛ4 risk allele, but also car- 47] The HLA-locus and the microtubule-associated protein ried the rs72824905-G allele. On MRI scan and amyloid tau (MAPT) loci (not individual variants) also have (sug- scan (PiB-PET), this person has some global atrophy and gestive) effects on the risks of AD, FTD and DLB [3, 11, only amyloid-β positivity in the precuneus and in the frontal 18, 43]. The APOE gene has been implied in a multitude of lobes (Fig. 3). At the age of 90 years, the dementia risk for pathways [52], TREM2 and HLA are involved in immunity homozygous carriers of the APOE ɛ4 genotype is approxi- [2], and MAPT encodes the tau protein. These shared genetic mately 80% [53] and virtually all are amyloid positive by age risk factors indicate a partial overlap in AD, DLB and FTD 90 [26]. The literature reports only a handful of centenarians etiology. It is of interest that, like the PLCG2 variant, APOE who are homozygous for the APOE ɛ4 allele [14, 16, 23]. and the HLA-DR locus were also associated with longev- It is unknown if these individuals were cognitively healthy. ity [4, 10, 28, 38]. A possible explanation is that APOE, This case shows that cognitively healthy aging in the pres- PLCG2 and HLA are involved in the processing of accumu- ence of the APOE ɛ4ɛ4 genotype is possible, likely due to lated aging-associated proteins [9]. In line with this hypoth- the protective effect of other genetic variants, such as the esis, rs72824905-G is associated with reduced ­pTau181 in the rs72824905-G variant in PLCG2 [26, 53]. CSF of memory clinic patients with pathologic Aβ1-42 CSF The mechanism that explains the protective effect of levels (L. Kleineidam et al. submitted). It is well known that rs72824905-G variant in the PLCG2 gene is currently having a dementia-associated neurodegenerative disease is unclear. We find that the associations of the rs72824905-G associated with shorter life-span [57]. Conversely, resilience variant with disease risk differ between diseases that have to diseases is associated with a longer life-span [24]. It is overlapping pathological features. For example, we observe likely that the association of rs72824905-G with longevity that carrying the rs72824905-G variant is protective against is due to the protection against dementia-associated neuro- DLB, but not against PD, while a common characteristic for degenerative diseases. However, with the available data we both diseases is the presence of α-synuclein-positive Lewy 13 Acta Neuropathologica (2019) 138:237–250 243 Fig. 3  MRI scan and PiB-PET scan, of a 102-year-old centenar- administration): Abnormal retention in the posterior cingulate/pre- ian carrying the homozygote APOE ɛ4 genotype as well as the cuneus and frontal lobes. Neuropsychological testing around time of rs72824905-G allele in PLCG2. MRI scan (Titan 3T MR scanner) scanning showed average performance on global cognitive function- shows some hippocampal atrophy (MTA grade 2), some global corti- ing/MMSE, memory, attention, working memory, fluency and visuo- cal atrophy (GCA-scale grade 1), but pronounced posterior cortical spatial tests compared to the cohort of cognitively healthy centenar- atrophy (grade 2), moderate white matter hyperintensities (Fazekas ians. The result of the clock drawing test is shown. The patient was grade 2), no lacunar infarcts or microbleeds. PET-PiB (scan after asked to draw a clock and put the time at 10 before 11 admission of 396 MBq C-11 PIB, 20-min image starting 90 min after bodies. The same holds for pathologies associated with the PLAID) [40, 46, 59] while somatic variants in PLCɣ2 are FTD-ALS and PSP spectrum of diseases (e.g. TDP-43, FUS associated with resistance to treatment of leukemia [56] inclusions as well as aggregations of tau). The observation (reviewed in Koss et al. [32].). The mutations that cause that our results do not point to a single pathological condi- PLAID and APLAID contribute to a strong hyperactivation tion does not preclude that PLCG2 is involved in a single of PLCɣ2 upon activation. In the case of APLAID (caused biological process. In fact, determining the involvement of by a p.Ser707Tyr substitution), the auto-inflammation has the PLCG2-related pathway might be an asset in pathologi- been suggested to be partially driven by PLCɣ2-dependent cal classifications of diseases, e.g. differentiating between activation of the pyrin (PYD)-domain-containing protein DLB and PD. Thus far, only one publication investigated 3 (NLRP3) inflammasome [5]. The potential of PLCG2 to the functional effect of the rs72824905-G variant in in vitro activate the inflammasome is further supported by in vitro experiments [35]. The authors suggest that in the mouse experiments [31]. The NLRP3 inflammasome is a crucial and human brain, PLCɣ2 is expressed in microglia [12]. signaling node in microglia that ultimately controls the mat- They show that PLCɣ2 mRNA co-localized with microglia- uration of pro-inflammatory interleukin (IL)-1β and IL-18 specific markers in healthy brain tissue and is expressed in [19] and has been linked to a multitude of neurodegenerative microglia near amyloid-β plaques in an APP mouse model diseases [60]. Although functional studies will need to eluci- [35]. Furthermore, functional characterization of PLCɣ2 date the effects of the rs72824905-G on PLCɣ2 function, we with the p.Pro552Arg amino acid substitution suggested only speculate that subtle changes in the NLRP3 inflammasome a slight increase in activity compared to wild-type PLCɣ2 activation may explain its protective effect. [35]. While additional functional experiments will be needed to confirm these findings, these experiments suggest that Strengths and weaknesses the functional changes induced by the PLCɣ2 p.Pro552Arg genetic variant may be subtle and, therefore, difficult to pin- The most important strength of our study is that we investi- point. This is according to expectations, as major changes gated the effect of the rs72824905-G variant in seven neu- to the immune system will most likely be harmful. Indeed, rological diseases in more than 53,000 patients and almost known germline mutations in PLCG2 cause the immune 150,000 controls. The AD cases and controls studied here disorders PLAID (PLCG2-associated antibody deficiency were all independent from the AD patients and controls and immune dysregulation) and APLAID (autoinflammatory in which the protective effect of rs72824905-G was first 13 244 Acta Neuropathologica (2019) 138:237–250 identified [47], but includes the samples used in Conway of the Alzheimer center Amsterdam is part of the neurodegeneration et al. [7] and Tesi et al. [51]. This report offers a robust research program of Amsterdam Neuroscience. The Alzheimer Center Amsterdam is supported by Stichting Alzheimer Nederland and Sticht- replication of the protection against AD. Some may argue ing VUmc fonds. The clinical database structure was developed with that the protective effect observed in FTD and DLB cases is funding from Stichting Dioraphte. Genotyping of the Dutch case-con- driven by misclassified AD cases. However, the effect size of trol samples was performed in the context of EADB (European Alz- rs72824905-G in these cases is very similar to the protective heimer DNA biobank) funded by the JPco-fuND FP-829-029 (ZonMW projectnumber 733051061). 100-Plus study: We are grateful for the effect in AD, which makes it unlikely that the effect can be collaborative efforts of all participating centenarians and their family ascribed purely to misclassified AD. Moreover, the age of members and/or relations. This work was supported by Stichting Alz- the controls was mostly younger than that of cases, making heimer Nederland (WE09.2014-03), Stichting Diorapthe, horstingstuit the protection from dementia not a longevity effect. foundation, Memorabel (ZonMW projectnumber 733050814) and Stichting VUmc Fonds. Genotyping of the 100-Plus Study was per- The large numbers under study were necessary because formed in the context of EADB (European Alzheimer DNA biobank) rs72824905-G has a minor allele frequency (MAF) ~ 1% in funded by the JPco-fuND FP-829-029 (ZonMW projectnumber European ancestry populations, which makes it a relatively 733051061). German Study on Ageing, Cognition and Dementia in rare genetic variant. Therefore, we ensured that our samples Primary Care Patients (AgeCoDe): This study/publication is part of the German Research Network on Dementia (KND), the German Research provided adequate statistical power to observe a similar pro- Network on Degenerative Dementia (KNDD; German Study on Age- tective effect of rs72824905-G against other neurodegen- ing, Cognition and Dementia in Primary Care Patients; AgeCoDe), and erative diseases. Despite the large sample sizes, we found the Health Service Research Initiative (Study on Needs, health service no evidence for this effect in our sample of PD, ALS and use, costs and health-related quality of life in a large sample of oldest- old primary care patients (85+; AgeQualiDe)) and was funded by the MS, which makes it unlikely that larger meta-analyses will German Federal Ministry of Education and Research (grants KND: observe an association between rs72824905-G and these 01GI0102, 01GI0420, 01GI0422, 01GI0423, 01GI0429, 01GI0431, three diseases. An association of rs72824905-G with an 01GI0433, 01GI0434; grants KNDD: 01GI0710, 01GI0711, 01GI0712, increased risk of PSP has been reported previously [7]. In 01GI0713, 01GI0714, 01GI0715, 01GI0716; grants Health Service Research Initiative: 01GY1322A, 01GY1322B, 01GY1322C, our analysis, which includes additional PSP cases, we were 01GY1322D, 01GY1322E, 01GY1322F, 01GY1322G). Alfredo Ram- not able to replicate this finding. Larger studies are needed irez was partly supported by the ADAPTED consortium: Alzheimer’s to determine the association with PSP. Including as large as disease Apolipoprotein Pathology for Treatment Elucidation and possible samples came with the consequence that we were Development, which has received funding from the Innovative Medi- cines Initiative 2 Joint Undertaking under grant agreement No 115975. not able to correct for population stratification using ances- Brain compendium: This work was funded by the UK Medical try principal components in all studies. Therefore, we used Research Council (13044). P.F.C. is a Wellcome Trust principal Fellow PCs were possible and further matching cases and controls (212219/Z/18/Z) and a UK NIHR Senior Investigator, who receives by study or country of origin. Finally, we indicate that the support from the Medical Research Council Mitochondrial Biology Unit (MC_UU_00015/9), and the National Institute for Health Research identified effects need to be replicated in other ethnicities in (NIHR) Biomedical Research Centre based at Cambridge University which rs72824905-G occurs. We note that in some ethnicities Hospitals NHS Foundation Trust and the University of Cambridge. The rs72824905-G plays no role as the frequency is very low in views expressed are those of the author(s) and not necessarily those of African (MAF = 0.0012) and African-American populations the NHS, the NIHR, or the Department of Health.Clinical AD, Sweden: We would like to thank UCL Genomics for performing the genotyping (MAF = 0.0004), and is not observed in East Asian [7, 33]. analyses. Danish data: The studies behind the Danish long-lived cases received funding from The National Program for Research Infrastruc- ture 2007 (grant no. 09-063256), the Danish Agency for Science Tech- Conclusions nology and Innovation, the Velux Foundation, the US National Institute of Health (P01 AG08761), the Danish Agency for Science, Technology and Innovation/The Danish Council for Independent Research (grant Our study shows that the rs72824905-G allele in PLCG2 no. 11-107308), the European Union’s Seventh Framework Programme associates with a decreased risk for AD, FTD, DLB and (FP7/2007-2011) under grant agreement no. 259679, the INTERREG concurrently with an increased chance of longevity. The pro- 4 A programme Syddanmark-Schleswig-K.E.R.N. (by EU funds from the European Regional Development Fund), the CERA Foundation tective effect of the rs72824905-G allele was not observed (Lyon), the AXA Research Fund, Paris, and The Health Foundation in ALS, PD and MS cases, which suggests that PLCG2- (Helsefonden), Copenhagen, Denmark. The GOYA study was con- associated processes overlap in the etiology of AD, FTD ducted as part of the activities of the Danish Obesity Research Centre and DLB, but not in the etiologies of ALS, PD and MS (DanORC, www.danor​c.dk) and The MRC centre for Causal Analyses in Translational Epidemiology (MRC CAiTE). The genotyping for (PSP too little power). Explaining the protective effect of the GOYA was funded by the Wellcome Trust (WT 084762). GOYA is a PLCγ2 protein on brain immune function may contribute to nested study within The Danish National Birth Cohort which was the design of successful therapeutic intervention strategies established with major funding from the Danish National Research applicable to those at risk for neurodegenerative diseases. Foundation. Additional support for this cohort has been obtained from the Pharmacy Foundation, the Egmont Foundation, The March of Dimes Birth Defects Foundation, the Augustinus Foundation, and the Acknowledgements The following studies and consortia have contrib- Health Foundation. Fundació ACE (FACE): We would like to thank uted to this manuscript. Amsterdam dementia Cohort (ADC): Research patients and controls who participated in this project. We are indebted 13 Acta Neuropathologica (2019) 138:237–250 245 to Trinitat Port-Carbó and her family for their support of Fundació for Regenerative Medicine, Mayo Clinic Center for Individualized ACE research programs. Fundació ACE collaborates with the Centro Medicine, Mayo Clinic Neuroscience Focused Research Team (Cecilia de Investigación Biomédica en Red sobre Enfermedades Neurodegen- and Dan Carmichael Family Foundation, and the James C. and Sarah erativas (CIBERNED, Spain) and is one of the participating centers of K. Kennedy Fund for Neurodegenerative Disease Research at Mayo the Dementia Genetics Spanish Consortium (DEGESCO). Agustín Clinic in Florida), the gift from Carl Edward Bolch, Jr., and Susan Bass Ruiz has received support from the EU/EFPIA Innovative Medicines Bolch, and The Sol Goldman Charitable Trust. Samples included in Initiative Joint Undertaking ADAPTED Grant No. 115975 and by this study are from the brain bank at Mayo Clinic in Jacksonville which grants PI13/02434 and PI16/01861. Acción Estratégica en Salud, inte- is supported by CurePSP|Society for Progressive Supranuclear Palsy grated in the Spanish National R + D + I Plan and financed by ISCIII and the Tau Consortium. NDRU cohort: We would like to thank the (Instituto de Salud Carlos III)-Subdirección General de Evaluación and NIH Neuro Brain Bank for contributing tissue samples; this study was the Fondo Europeo de Desarrollo Regional (FEDER- “Una manera de supported in part by grants from the National Institutes of Health: Hacer Europa”), by Fundación bancaria “La Caixa” and Grifols SA U19-AG03365, P50 NS38377, and P50-AG005146. Tissue samples for (GR@ACE project). Genetics of Healthy Ageing Study (GEHA – NL): genotyping were provided by the Johns Hopkins Morris K. Udall The work described in this paper was funded mainly by the EU GEHA Center of Excellence for Parkinson’s Disease Research (NIH P50 Project contract no. LSHM-CT-2004-503-270. Gothenburg Birth NS38377) and the Johns Hopkins Alzheimer’s Disease Research Cohort (GBC) Studies: We would like to thank UCL Genomics for Center. We aregrateful for the support of the entire BIOCARD study performing the genotyping analyses. The studies were supported by team at Johns Hopkins University. Additionally, we acknowledge the The Stena Foundation, The Swedish Research Council (2015-02830, contributions of the Geriatric Psychiatry Branch (GPB) in the intramu- 2013-8717), The Swedish Research Council for Health, Working Life ral program of NIMH who initiated the BIOCARD study. We would and Wellfare (2013-1202, 2005-0762, 2008-1210, 2013-2300, 2013- like to thank the NIA Baltimore Longitudinal Study of Aging for con- 2496, 2013-0475), The Brain Foundation, Sahlgrenska University tributing tissue samples to the Johns Hopkins Alzheimer’s Disease Hospital (ALF), The Alzheimer’s Association (IIRG-03-6168), The Research Center. DNA panels from the NINDS Human Genetics Alzheimer’s Association Zenith Award (ZEN-01-3151), Eivind och Resource Center DNA and Cell Line Repository (http://ccr.corie​ll.org/ Elsa K:son Sylvans Stiftelse, The Swedish Alzheimer Foundation. ninds​) were used in this study, as well as clinical data. We thank the International FTD-Genomics Consortium (IFGC): International FTD- following brain banks for providing brain tissues: Banner Sun Health Genomics Consortium (IFGC): The authors thank the IFGC for provid- Research Institute, New York Brain Bank, Newcastle Brain Tissue ing relevant data to support the analyses presented in this manuscript. Resource, Human Brain and Spinal Fluid Resource Center, Netherlands Further acknowledgments for IFGC (https​://ifgcs​ite.wordp​ress.com/), Brain Bank Amsterdam, Mount Sinai Brain Bank, Harvard Brain Bank, e.g. full members list and affiliations, are found in the online supple- Duke University Brain Bank, Virginia Commonwealth University Brain mentary files. IPDGC (​The International Parkinson Disease Genomics Bank, and the Georgetown University Brain Bank. We also thank the Consortium): We also would like to thank all members of the Interna- following research centers for providing patient samples: University of tional Parkinson Disease Genomics Consortium (IPDGC). See for a Michigan, University College London Institute of Neurology, Univer- complete overview of members, acknowledgements and funding http:// sity of Maryland, University of California – San Diego, and University pdgene​ tics.​ org/partne​ rs. Kompetenznetz Multiple Sklerose (KKNMS): of Miami. We are grateful to members of the North American Brain This work was supported by the German Ministry for Education and Expression Consortium for contributing DNA samples. The authors Research (BMBF) as part of the “German Competence Network Mul- thank the patients and families who have donated DNA samples and tiple Sclerosis” (KKNMS) (grant nos. 01GI0916 and 01GI0917) and brain tissue for scientific research. The research was supported in part the Munich Cluster for Systems Neurology (SyNergy). TA was sup- by the Intramural Research Program of the NIH National Institute of ported by the BMBF through the Integrated Network IntegraMent, Neurological Disorders and Stroke and the National Institute on Aging under the auspices of the e:Med Programme (01ZX1614J). BH was (project numbers: ZIA-NS003154, Z01-AG000949).” Oviedo: This supported by the EU Horizon 2020 project MultipleMS.Longitudinal work was partly supported by Grant from Fondo de Investigaciones Aging Study Amsterdam (LASA) is largely supported by a grant from Sanitarias-Fondos FEDER EuropeanUnion to Victoria Alvarez the Netherlands Ministry of Health, Welfare and Sports, Directorate of PI15/00878. Pascual Sánchez-Juan: Pascual Sánchez-Juan is supported Long-Term Care. The authors are grateful to all LASA participants, the by CIBERNED and Carlos III Institute of Health, Spain (PI08/0139, fieldwork team and all researchers for their ongoing commitment to the PI12/02288, and PI16/01652), jointly funded by Fondo Europeo de study. Leiden Longevity Study: This study was supported by a grant Desarrollo Regional (FEDER), Unión Europea, “Una manera de hacer from the Innovation-Oriented Research Program on Genomics (Sen- Europa”. Project MinE: The ProjectMinE study was supported by the terNovem IGE05007), the Centre for Medical Systems Biology, and ALS Foundation Netherlands and the MND association (UK) (Project the Netherlands Consortium for Healthy Ageing (Grant 050-060-810), MinE, www.proje​ctmin​e.com). Risk and modifying factors in Fronto all in the framework of the Netherlands Genomics Initiative/Nether- Temporal Dementia (RiMoD-FTD): follows: The SPIN cohort: We are lands Organization for Scientific Research (NWO) and by Unilever indebted to patients and their families for their participation in the Colworth.Maria Carolina Dalmasso: Georg Forster Research Award “Sant Pau Initiative on Neurodegeneration cohort”, at the Sant Pau (Alexander von Humboldt Foundation). Mayo Clinic AD, DLB, PD, Hospital (Barcelona). This is a multimodal research cohort for bio- PSP: We thank the patients and their families for their participation, marker discovery and validation that is partially funded by Generalitat without whom these studies would not have been possible. Funding for de Catalunya (2017 SGR 547 to JC), as well as from the Institute of this work was supported by National Institute on Aging [RF AG051504 Health Carlos III-Subdirección General de Evaluación and the Fondo to NET.; U01 AG046139 to NET]; and National Institute of Neurologi- Europeo de Desarrollo Regional (FEDER- “Una manera de Hacer cal Disorders and Stroke [R01 NS080820 to NET; P50 NS072187]. Europa”) (grants PI11/02526, PI14/01126, and PI17/01019 to JF; The Mayo Clinic is a Lewy Body Dementia Association (LBDA) PI17/01895 to AL), and the Centro de Investigacion Biomedica en Red Research Center of Excellence, American Parkinson Disease Associa- Enfermedades Neurodegenerativas programme (Program 1, Alzheimer tion (APDA) Information and Referral Center and Center for Advanced Disease to AL). We would also like to thank the Fundació Bancària Research, NINDS Tau Center without Walls (U54-NS100693) and is Obra Social La Caixa (DABNI project) to JF and AL; and Fundacion supported by Mayo Clinic AD and related dementias genetics program, BBVA (to AL), for their support in funding this follow-up study. San The Little Family Foundation, the Mangurian Foundation for Lewy Sebastian: We would like to thank patients and control volunteers who body research and NINDS R01 NS078086 (to OAR). The PD program participated in this study. We thank Ana Gorostidi for his work at the at the Mayo Clinic Florida is also supported by the Mayo Clinic Center Biodonostia HRI Genomics Platform. This work was partly supported 13 246 Acta Neuropathologica (2019) 138:237–250 by CIBERNED. Adolfo López de Munain is supported by Fundación 4. Broer L, Buchman AS, Deelen J, Evans DS, Faul JD, Lunetta KL Salud 2000 (PI2013156) and Diputación Foral de Gipuzkoa (Exp. et al (2015) GWAS of longevity in CHARGE consortium confirms 114/17). UK Biobank analysis: This work was funded by The Nether- APOE and FOXO3 candidacy. J Gerontol A Biol Sci Med Sci lands Organization for Scientific Research (NWO VICI 453-14-005). 70:110–118. https​://doi.org/10.1093/geron​a/glu16​6 The analyses were carried out on the Genetic Cluster Computer, which 5. Chae JJ, Park YH, Park C, Hwang IY, Hoffmann P, Kehrl JH is financed by the Netherlands Scientific Organization (NWO: 480-05- et al (2015) Connecting two pathways through ­Ca2+ signaling: 003), by the VU University, Amsterdam, The Netherlands, and by the NLRP3 inflammasome activation induced by a hypermorphic Dutch Brain Foundation, and is hosted by the Dutch National Comput- PLCG2 mutation. Arthritis Rheumatol 67:563–567. https​://doi. ing and Networking Services SurfSARA. This research has been con- org/10.1002/art.38961​ ducted using the UK Biobank resource under application number 6. Chen JA, Wang Q, Davis-Turak J, Li Y, Karydas AM, Hsu SC 16406. We are grateful to the numerous participants, researchers, and et al (2015) A multiancestral genome-wide exome array study staff who collected and contributed to the data. of Alzheimer disease, frontotemporal dementia, and progres- sive supranuclear palsy. JAMA Neurol 72:414–422. https​://doi. Data availability statement The authors declare that the data support- org/10.1001/jaman​eurol​.2014.4040 ing the findings of this study are available within the paper and its 7. Conway OJ, Carrasquillo MM, Wang X, Bredenberg JM, Reddy online Resource. JS, Strickland SL et al (2018) ABI3 and PLCG2 missense vari- ants as risk factors for neurodegenerative diseases in Caucasians and African Americans. Mol Neurodegener 13:53. https​://doi. Author contributions SvdL and HH had full access to the meta-anal- org/10.1186/s1302​4-018-0289-x ysis results and wrote the draft of the manuscript. SvdL, OJC, MMC, 8. Dankowski T, Buck D, Andlauer TF, Antony G, Bayas A, Bech- LK, EBvdA, NS, JAC, AZ, TFMA, MD-F, JS-S, HZ, MN, CB, JES, mann L et al (2015) Successful replication of GWAS hits for mul- JM-F, SM-G, MJK, KB, CL, IdR, WW, BJ, XW, NB, NT, JEC, SLS, tiple sclerosis in 10,000 germans using the exome array. Genet AK, MCD, MA, NE-T, SWS and Aru analyzed parts of the data. MJTR Epidemiol 39:601–608. https​://doi.org/10.1002/gepi.21933​ provided compute for this study. MCD compiled demographic informa- 9. Deane R, Sagare A, Hamm K, Parisi M, Lane S, Finn MB et al tion of all Spanish cohorts. SvdL, EBvdA, NS, MD-F, JS-S, SM-G, (2008) apoE isoform-specific disruption of amyloid beta peptide CL, IdR, WW, BJ, NB, NT, NE-T, SWS, Aru, IJ, IH, AL, MW, JF, clearance from mouse brain. J Clin Invest 118:4002–4013. https​ IS, MAF, OP, MZ, SR-H, II-G, AO, FTB, MH, CM, BI, MS, EM-R, ://doi.org/10.1172/JCI36​663 JWI, MNMvB, DA, HW, ERR, BFB, RCP, TJF, JAvG, MJU, LT, WM, 10. Deelen J, Beekman M, Uh HW, Broer L, Ayers KL, Tan Q et al OD-I, MBo, MB, EM, ALdM, AP, ZKW, OAR, DWD, NRG-R, DK, (2014) Genome-wide association meta-analysis of human longev- RR, AWL, YALP, PS, TG, PFC, BH, JT, FM, EAN, TIAS, PS-J, DP, ity identifies a novel locus conferring survival beyond 90 years of GC, JC, KC, AR, WMvdF and HH contributed to data collection. JT age. Hum Mol Genet 23:4420–4432. https:​ //doi.org/10.1093/hmg/ and OP contributed to neuropathological diagnoses. MW, IS, SR-H, ddu13​9 II-G, MS, HW, LT, WM, MBo, MB, ZKW, AWL, PS, TG, PFC, BH, 11. Ferrari R, Wang Y, Vandrovcova J, Guelfi S, Witeolar A, Karch MAH, TIAS, PH, JC, KC, NE-T, SWS, AR, Aru, ES, WMvdF and HH CM et al (2017) Genetic architecture of sporadic frontotempo- secured funding for their respective studies. MMC, MW, SR-H, MH, ral dementia and overlap with Alzheimer’s and Parkinson’s dis- JEC, MS, MNMvB, HW, WM, NMvS, PS, TG, PFC, BH, MAH, EAN, eases. J Neurol Neurosurg Psychiatry 88:152–164. https​://doi. PH, NE-T, SWS, AR, ES, WMvdF and HH supervised their respec- org/10.1136/jnnp-2016-31441​1 tive studies. HH was responsible for the overall study supervision. All 12. Friedman BA, Srinivasan K, Ayalon G, Meilandt WJ, Lin H, authors critically revised the manuscript for important intellectual con- Huntley MA et al (2018) Diverse brain myeloid expression tent and approved the final manuscript. profiles reveal distinct microglial activation states and aspects of Alzheimer’s disease not evident in mouse models. Cell Rep Funding statement No funding was received specifically for this study. 22:832–847. https​://doi.org/10.1016/j.celre​p.2017.12.066 The studies and consortia all have their respective funding sources 13. Gagliano SA, Pouget JG, Hardy J, Knight J, Barnes MR, Ryten M described in the acknowledgement section. et al (2016) Genomics implicates adaptive and innate immunity in Alzheimer’s and Parkinson’s diseases. Ann Clin Transl Neur Open Access This article is distributed under the terms of the Crea- 3:924–933. https​://doi.org/10.1002/acn3.369 tive Commons Attribution 4.0 International License (http://creat​iveco​ 14. Garatachea N, Marin PJ, Santos-Lozano A, Sanchis-Gomar F, mmons​.org/licen​ses/by/4.0/), which permits unrestricted use, distribu- Emanuele E, Lucia A (2015) The ApoE gene is related with tion, and reproduction in any medium, provided you give appropriate exceptional longevity: a systematic review and meta-analysis. credit to the original author(s) and the source, provide a link to the Rejuvenation Res 18:3–13. https​://doi.org/10.1089/rej.2014.1605 Creative Commons license, and indicate if changes were made. 15. Genomes Project C, Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, et al (2015) A global reference for human genetic variation. Nature 526:68–74. https​://doi.org/10.1038/natur​e1539​ 3 References 16. Gerdes LU, Jeune B, Ranberg KA, Nybo H, Vaupel JW (2000) Estimation of apolipoprotein E genotype-specific relative mor- 1. Alzheimer’s A (2016) 2016 Alzheimer’s disease facts and figures. tality risks from the distribution of genotypes in centenarians Alzheimers Dement 12:459–509 and middle-aged men: apolipoprotein E gene is a “frailty gene,” 2. Borroni B, Ferrari F, Galimberti D, Nacmias B, Barone C, Bag- not a “longevity gene”. Genet Epidemiol 19:202–210. https​:// noli S et al (2014) Heterozygous TREM2 mutations in frontotem- doi.org/10.1002/1098-2272(20001​0)19:3%3c202​:AID-GEPI2​ poral dementia. Neurobiol Aging 35(934):e937-910. https​://doi. %3e3.0.CO;2-Q org/10.1016/j.neuro​biola​ging.2013.09.017 17. Glass CK, Saijo K, Winner B, Marchetto MC, Gage FH (2010) 3. Broce I, Karch CM, Wen N, Fan CC, Wang YP, Tan CH et al Mechanisms underlying inflammation in neurodegeneration. Cell (2018) Immune-related genetic enrichment in frontotemporal 140:918–934. https​://doi.org/10.1016/j.cell.2010.02.016 dementia: an analysis of genome-wide association studies. Plos 18. Guerreiro R, Ross OA, Kun-Rodrigues C, Hernandez DG, Orme Med. https​://doi.org/10.1371/journ​al.pmed.10024​87 T, Eicher JD et al (2018) Investigating the genetic architecture 13 Acta Neuropathologica (2019) 138:237–250 247 of dementia with Lewy bodies: a two-stage genome-wide asso- 35. Magno L, Lessard CB, Martins M, Lang V, Cruz P, Asi Y et al ciation study. Lancet Neurol 17:64–74. https​://doi.org/10.1016/ (2019) Alzheimer’s disease phospholipase C-gamma-2 (PLCG2) S1474​-4422(17)30400​-3 protective variant is a functional hypermorph. Alzheimers Res 19. He Y, Hara H, Nunez G (2016) Mechanism and regulation of Ther 11:16. https​://doi.org/10.1186/s1319​5-019-0469-0 NLRP3 inflammasome activation. Trends Biochem Sci 41:1012– 36. Marioni RE, Harris SE, Zhang Q, McRae AF, Hagenaars SP, 1021. https​://doi.org/10.1016/j.tibs.2016.09.002 Hill WD et al (2018) GWAS on family history of Alzheimer’s 20. Hemmer B, Kerschensteiner M, Korn T (2015) Role of the innate disease. Transl Psychiatry 8:99. https​://doi.org/10.1038/s4139​ and adaptive immune responses in the course of multiple scle- 8-018-0150-6 rosis. Lancet Neurol 14:406–419. https​://doi.org/10.1016/S1474​ 37. McCarthy S, Das S, Kretzschmar W, Delaneau O, Wood AR, -4422(14)70305​-9 Teumer A et al (2016) A reference panel of 64,976 haplotypes 21. Higgins JP, Thompson SG (2002) Quantifying heterogeneity in a for genotype imputation. Nat Genet 48:1279–1283. https​://doi. meta-analysis. Stat Med 21:1539–1558. https​://doi.org/10.1002/ org/10.1038/ng.3643 sim.1186 38. McDaid AF, Joshi PK, Porcu E, Komljenovic A, Li H, Sorrentino 22. Holstege H, Beker N, Dijkstra T, Pieterse K, Wemmenhove E, V et al (2017) Bayesian association scan reveals loci associated Schouten K et al (2018) The 100-plus Study of cognitively healthy with human lifespan and linked biomarkers. Nat Commun. https​ centenarians: rationale, design and cohort description. Eur J Epi- ://doi.org/10.1038/ncomm​s1584​2 demiol. https​://doi.org/10.1007/s1065​4-018-0451-3 39. Nalls MA (2018) Parkinson’s disease genetics: identifying novel 23. Ishioka YL, Gondo Y, Fuku N, Inagaki H, Masui Y, Takayama M risk loci, providing causal insights and improving estimates of (2016) Effects of the APOE epsilon4 allele and education on cog- heritable risk. BioRxiv. https​://doi.org/10.1101/38816​5 nitive function in Japanese centenarians. Age (Dordr) 38:495–503. 40. Ombrello MJ, Remmers EF, Sun G, Freeman AF, Datta S, Torabi- https​://doi.org/10.1007/s1135​7-016-9944-8 Parizi P et al (2012) Cold urticaria, immunodeficiency, and auto- 24. Ismail K, Nussbaum L, Sebastiani P, Andersen S, Perls T, Bar- immunity related to PLCG2 deletions. N Engl J Med 366:330– zilai N et al (2016) Compression of morbidity is observed across 338. https​://doi.org/10.1056/NEJMo​a1102​140 cohorts with exceptional longevity. J Am Geriatr Soc 64:1583– 41. Patsopoulos NA (2018) Genetics of multiple sclerosis: an over- 1591. https​://doi.org/10.1111/jgs.14222​ view and new directions. Cold Spring Harb Perspect Med 8:45. 25. Jansen IE (2017) Genetic meta-analysis identifies 9 novel loci https​://doi.org/10.1101/cshpe​rspec​t.a0289​51 and functional pathways for Alzheimers disease risk. Biorxiv 42. Patterson RL, van Rossum DB, Ford DL, Hurt KJ, Bae SS, Suh 1:258533 PG et al (2002) Phospholipase C-gamma is required for agonist- 26. Jansen WJ, Ossenkoppele R, Knol DL, Tijms BM, Scheltens P, induced ­Ca2+ entry. Cell 111:529–541 Verhey FRJ et al (2015) Prevalence of cerebral amyloid pathology 43. Peuralinna T, Myllykangas L, Oinas M, Nalls MA, Keage HA, Iso- in persons without dementia a meta-analysis. J Am Med Assoc viita VM et al (2015) Genome-wide association study of neocorti- 313:1924–1938. https​://doi.org/10.1001/jama.2015.4668 cal Lewy-related pathology. Ann Clin Transl Neurol 2:920–931. 27. Jonsson T, Stefansson H, Steinberg S, Jonsdottir I, Jonsson PV, https​://doi.org/10.1002/acn3.231 Snaedal J et al (2013) Variant of TREM2 associated with the risk 44. Pilling LC, Kuo CL, Sicinski K, Tamosauskaite J, Kuchel GA, of Alzheimer’s disease. N Engl J Med 368:107–116. https​://doi. Harries LW et al (2017) Human longevity: 25 genetic loci associ- org/10.1056/NEJMo​a1211​103 ated in 389,166 UK biobank participants. Aging (Albany NY) 28. Joshi PK, Pirastu N, Kentistou KA, Fischer K, Hofer E, Schraut 9:2504–2520. https​://doi.org/10.18632​/aging​.10133​4 KE et al (2017) Genome-wide meta-analysis associates HLA- 45. Rhee SG (2001) Regulation of phosphoinositide-specific phospho- DQA1/DRB1 and LPA and lifestyle factors with human longevity. lipase C. Annu Rev Biochem 70:281–312. https:​ //doi.org/10.1146/ Nat Commun 8:910. https:​ //doi.org/10.1038/s41467​ -017-00934-​ 5 annur​ev.bioch​em.70.1.281 29. Keogh MJ, Wei W, Wilson I, Coxhead J, Ryan S, Rollinson S 46. Schade A, Walliser C, Wist M, Haas J, Vatter P, Kraus JM et al et al (2017) Genetic compendium of 1511 human brains available (2016) Cool-temperature-mediated activation of phospholipase through the UK Medical Research Council Brain Banks Network C-gamma2 in the human hereditary disease PLAID. Cell Signal Resource. Genome Res 27:165–173. https​://doi.org/10.1101/ 28:1237–1251. https​://doi.org/10.1016/j.cells​ig.2016.05.010 gr.21060​9.116 47. Sims R, van der Lee SJ, Naj AC, Bellenguez C, Badarinarayan 30. Kim D, Jun KS, Lee SB, Kang NG, Min DS, Kim YH et al N, Jakobsdottir J et al (2017) Rare coding variants in PLCG2, (1997) Phospholipase C isozymes selectively couple to specific ABI3, and TREM2 implicate microglial-mediated innate immu- neurotransmitter receptors. Nature 389:290–293. https​://doi. nity in Alzheimer’s disease. Nat Genet 49:1373–1384. https:​ //doi. org/10.1038/38508​ org/10.1038/ng.3916 31. Kleineidam L, Chouraki V, Próchnicki T, van der Lee S, Madrid- 48. Skol AD, Scott LJ, Abecasis GR, Boehnke M (2006) Joint analy- Márquez L, Wagner-Thelen H et al (2018) PLCG2 protective vari- sis is more efficient than replication-based analysis for two-stage ant p.P522R modulates Tau pathology and disease progression in genome-wide association studies. Nat Genet 38:209–213. https​:// patients with mild cognitive impairment. Lancet. https:​ //ssrn.com/ doi.org/10.1038/ng170​6 abstr​act=33076​49 49. Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J 32. Koss H, Bunney TD, Behjati S, Katan M (2014) Dysfunction et al (2015) UK biobank: an open access resource for identifying of phospholipase Cgamma in immune disorders and cancer. the causes of a wide range of complex diseases of middle and Trends Biochem Sci 39:603–611. https​: //doi.org/10.1016/j. old age. Plos Med 12:e1001779. https​://doi.org/10.1371/journ​ tibs.2014.09.004 al.pmed.10017​79 33. Lek M, Karczewski KJ, Minikel EV, Samocha KE, Banks E, Fen- 50. Team RC (2018) R: a language and environment for statistical nell T et al (2016) Analysis of protein-coding genetic variation computing. R Foundation for Statistical Computing, Vienna in 60,706 humans. Nature 536:285–291. https​://doi.org/10.1038/ 51. Tesi N, van der Lee SJ, Hulsman M, Jansen IE, Stringa N, van natur​e1905​7 Schoor N et al (2018) Centenarian controls increase variant effect 34. Liu JZ, Erlich Y, Pickrell JK (2017) Case-control association sizes by an average twofold in an extreme case-extreme control mapping by proxy using family history of disease. Nat Genet analysis of Alzheimer’s disease. Eur J Hum Genet. https​://doi. 49:325–331. https​://doi.org/10.1038/ng.3766 org/10.1038/s4143​1-018-0273-5 13 248 Acta Neuropathologica (2019) 138:237–250 52. Tudorache IF, Trusca VG, Gafencu AV (2017) Apolipoprotein 57. Xie J, Brayne C, Matthews FE, Medical Research Council Cog- E—a multifunctional protein with implications in various patholo- nitive F, Ageing Study c (2008) Survival times in people with gies as a result of its structural features. Comput Struct Biotechnol dementia: analysis from population based cohort study with J 15:359–365. https​://doi.org/10.1016/j.csbj.2017.05.003 14 year follow-up. BMJ 336:258–262. https​://doi.org/10.1136/ 53. van der Lee SJ, Wolters FJ, Ikram MK, Hofman A, Ikram MA, bmj.39433​.61667​8.25 Amin N et al (2018) The effect of APOE and other common 58. Yu P, Constien R, Dear N, Katan M, Hanke P, Bunney TD genetic variants on the onset of Alzheimer’s disease and demen- et al (2005) Autoimmunity and inflammation due to a gain-of- tia: a community-based cohort study. Lancet Neurol 17:434–444. function mutation in phospholipase C gamma 2 that specifically https​://doi.org/10.1016/S1474​-4422(18)30053​-X ­ a2+ entry. Immunity 22:451–465. https​://doi. increases external C 54. van Rheenen W, Shatunov A, Dekker AM, McLaughlin RL, org/10.1016/j.immun​i.2005.01.016 Diekstra FP, Pulit SL et al (2016) Genome-wide association 59. Zhou Q, Lee GS, Brady J, Datta S, Katan M, Sheikh A et al analyses identify new risk variants and the genetic architecture (2012) A hypermorphic missense mutation in PLCG2, encoding of amyotrophic lateral sclerosis. Nat Genet 48:1043. https​://doi. phospholipase Cgamma2, causes a dominantly inherited autoin- org/10.1038/ng.3622 flammatory disease with immunodeficiency. Am J Hum Genet 55. Walton RL, Soto-Ortolaza AI, Murray ME, Lorenzo-Betancor O, 91:713–720. https​://doi.org/10.1016/j.ajhg.2012.08.006 Ogaki K, Heckman MG et al (2016) TREM2 p.R47H substitution 60. Zhou K, Shi L, Wang Y, Chen S, Zhang J (2016) Recent is not associated with dementia with Lewy bodies. Neurol Genet Advances of the NLRP3 Inflammasome in Central Nervous 2:85. https​://doi.org/10.1212/nxg.00000​00000​00008​5 System Disorders. J Immunol Res 2016:9238290. https​://doi. 56. Woyach JA, Furman RR, Liu TM, Ozer HG, Zapatka M, Ruppert org/10.1155/2016/92382​90 AS et al (2014) Resistance mechanisms for the Bruton’s tyrosine kinase inhibitor ibrutinib. N Engl J Med 370:2286–2294. https​:// Publisher’s Note Springer Nature remains neutral with regard to doi.org/10.1056/NEJMo​a1400​029 jurisdictional claims in published maps and institutional affiliations. Affiliations Sven J. van der Lee1,2 · Olivia J. Conway3 · Iris Jansen1,4 · Minerva M. Carrasquillo3 · Luca Kleineidam5,6,7 · Erik van den Akker8,9 · Isabel Hernández10,11 · Kristel R. van Eijk12 · Najada Stringa13 · Jason A. Chen14 · Anna Zettergren15 · Till F. M. Andlauer16,17,18 · Monica Diez‑Fairen19,20 · Javier Simon‑Sanchez21,22 · Alberto Lleó11,23 · Henrik Zetterberg24,25,26 · Marianne Nygaard27 · Cornelis Blauwendraat28 · Jeanne E. Savage4 · Jonas Mengel‑From29 · Sonia Moreno‑Grau10 · Michael Wagner5,6 · Juan Fortea11,23 · Michael J. Keogh30,31 · Kaj Blennow24,25 · Ingmar Skoog15 · Manuel A. Friese18,32 · Olga Pletnikova33 · Miren Zulaica11,34 · Carmen Lage11,35,36 · Itziar de Rojas10,11 · Steffi Riedel‑Heller37 · Ignacio Illán‑Gala11,23 · Wei Wei31 · Bernard Jeune29 · Adelina Orellana10,11 · Florian Then Bergh18,38 · Xue Wang3 · Marc Hulsman1,2,9 · Nina Beker1 · Niccolo Tesi1,2,9 · Christopher M. Morris39 · Begoña Indakoetxea11,34,40 · Lyduine E. Collij41 · Martin Scherer42 · Estrella Morenas‑Rodríguez11,23 · James W. Ironside43 · Bart N. M. van Berckel41 · Daniel Alcolea11,23 · Heinz Wiendl18,44 · Samantha L. Strickland3 · Pau Pastor19,20 · Eloy Rodríguez Rodríguez11,35,36 · DESGESCO (Dementia Genetics Spanish Consortium), EADB (Alzheimer Disease European DNA biobank) · EADB (Alzheimer Disease European DNA biobank) · IFGC (International FTD-Genomics Consortium), IPDGC (The International Parkinson Disease Genomics Consortium) · IPDGC (The International Parkinson Disease Genomics Consortium) · RiMod-FTD (Risk and Modifying factors in Fronto-Temporal Dementia) · Netherlands Brain Bank (NBB) · Bradley F. Boeve45 · Ronald C. Petersen45 · Tanis J. Ferman46 · Jay A. van Gerpen47 · Marcel J. T. Reinders64 · Ryan J. Uitti47 · Lluís Tárraga10,11 · Wolfgang Maier5,6 · Oriol Dols‑Icardo11,23 · Amit Kawalia7 · Maria Carolina Dalmasso7,48 · Mercè Boada10,11 · Uwe K. Zettl18,49 · Natasja M. van Schoor13 · Marian Beekman8 · Mariet Allen3 · Eliezer Masliah50 · Adolfo López de Munain11,34,51 · Alexander Pantelyat52 · Zbigniew K. Wszolek47 · Owen A. Ross3 · Dennis W. Dickson3 · Neill R. Graff‑Radford47 · David Knopman45 · Rosa Rademakers3 · Afina W. Lemstra1 · Yolande A. L. Pijnenburg1 · Philip Scheltens1 · Thomas Gasser53 · Patrick F Chinnery31,54 · Bernhard Hemmer17,18,55 · Martijn A. Huisman13,56 · Juan Troncoso33 · Fermin Moreno11,34,40 · Ellen A. Nohr57 · Thorkild I. A. Sørensen58,59,60 · Peter Heutink21,22 · Pascual Sánchez‑Juan11,35,36 · Danielle Posthuma2,4 · The GIFT (Genetic Investigation in Frontotemporal Dementia and Alzheimer’s Disease) Study Group · Jordi Clarimón11,23 · Kaare Christensen27,61,62 · Nilüfer Ertekin‑Taner3,47 · Sonja W. Scholz28,52 · Alfredo Ramirez5,7 · Agustín Ruiz10,11 · Eline Slagboom8,63 · Wiesje M. van der Flier1 · Henne Holstege1,2 1 * Sven J. van der Lee Alzheimer Center Amsterdam, Department of Neurology, s.j.vanderlee@amsterdamumc.nl Amsterdam Neuroscience, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, The Netherlands * Henne Holstege 2 h.holstege@amsterdamumc.nl Department of Clinical Genetics, Vrije Universiteit Amsterdam, Amsterdam UMC, Amsterdam, The Netherlands 13 Acta Neuropathologica (2019) 138:237–250 249 3 25 Department of Neuroscience, Mayo Clinic Florida, Department of Psychiatry and Neurochemistry, Institute Jacksonville, FL 32224, USA of Neuroscience and Physiology, Sahlgrenska Academy 4 at the University of Gothenburg, Gothenburg, Sweden Department of Complex Trait Genetics, Center 26 for Neurogenomics and Cognitive Research, Amsterdam Department of Molecular Neuroscience, UCL Institute Neuroscience, Vrije Universiteit Amsterdam, Amsterdam of Neurology, Queen Square, London, UK UMC, Amsterdam, The Netherlands 27 The Danish Aging Research Center, Epidemiology, 5 Department for Neurodegenerative Diseases and Geriatric Biostatistics and Biodemography, Department of Public Psychiatry, University of Bonn, Bonn, Germany Health, University of Southern Denmark, Odense, Denmark 6 28 DZNE, German Center for Neurodegenerative Diseases, Neurodegenerative Diseases Research Unit, National Bonn, Germany Institute of Neurological Disorders and Stroke, Bethesda, 7 MD 20892‑3707, USA Division of Neurogenetics and Molecular Psychiatry, 29 Department of Psychiatry and Psychotherapy, Faculty Epidemiology, Biostatistics and Biodemography, Department of Medicine, University Hospital Cologne, Cologne, of Public Health, University of Southern Denmark, Odense, Germany Denmark 8 30 Molecular Epidemiology, Leiden University Medical Center, Institute of Genetic Medicine, Newcastle University, Leiden, The Netherlands Newcastle upon Tyne NE1 3BZ, UK 9 31 Pattern Recognition and Bioinformatics, Delft University Department of Clinical Neurosciences, University of Technology, Delft, The Netherlands of Cambridge, Cambridge CB2 0QQ, UK 10 32 Research Center and Memory Clinic, Fundació ACE, Institut Institut für Neuroimmunologie und Multiple Sklerose Català de Neurociències Aplicades, Universitat Internacional (INIMS), Universitätsklinikum Hamburg‐Eppendorf, de Catalunya, Barcelona, Spain Hamburg, Germany 11 33 Centro de Investigacion Biomedica en Red en Enfermedades Department of Pathology (Neuropathology), Johns Hopkins Neurodegenerativas (CIBERNED), Madrid, Spain University Medical Center, Baltimore, MD, USA 12 34 Department of Neurology, Brain Center Rudolf Magnus, Instituto Biodonostia, San Sebastian, Spain University Medical Center Utrecht, Utrecht, The Netherlands 35 University Hospital “Marques de Valdecilla”, Santander, 13 Amsterdam UMC‑Vrije Universiteit Amsterdam, Department Spain of Epidemiology and Biostatistics, Amsterdam Public Health 36 IDIVAL, Santander, Spain Research Institute, Amsterdam, The Netherlands 37 14 Institute of Social Medicine, Occupational Health and Public Interdepartmental Program in Bioinformatics, University Health (ISAP), University of Leipzig, Leipzig, Germany of California, Los Angeles, USA 38 15 Department of Neurology, University of Leipzig, Leipzig, Neuropsychiatric Epidemiology Unit, Department Germany of Psychiatry and Neurochemistry, Institute of Neuroscience 39 and Physiology, Sahlgrenska Academy, Centre for Ageing Newcastle Brain Tissue Resource, Edwardson Building, and Health (AgeCap) at the University of Gothenburg, Institute of Neuroscience, Newcastle University, Gothenburg, Sweden Newcastle upon Tyne NE4 5PL, UK 16 40 Max Planck Institute of Psychiatry, Munich, Germany Cognitive Disorders Unit, Department of Neurology, 17 Hospital Universitario San Sebastian, San Sebastian, Spain Department of Neurology, Klinikum rechts der Isar, 41 Technical University of Munich, Munich, Germany Department of Radiology and Nuclear Medicine, Amsterdam 18 Neuroscience, Vrije Universiteit Amsterdam, Amsterdam German Competence Network Multiple Sclerosis (KKNMS), UMC, Amsterdam, The Netherlands Munich, Germany 42 19 Department of Primary Medical Care, Center Movement Disorders and Memory Unit, Department for Psychosocial Medicine, University Medical Center, of Neurology, University Hospital Mutua de Terrassa, Hamburg‑Eppendorf, Germany Barcelona, Spain 43 20 Centre for Clinical Brain Sciences, University of Edinburgh, Fundacio per la Recerca Biomedica I Social Mutua Terrassa, Edinburgh EH4 2XU, UK Terrassa, Barcelona, Spain 44 21 Department of Neurology, Klinik für Neurologie mit Institut German Center for Neurodegenerative Diseases für Translationale Neurologie, University of Münster, (DZNE)-Tübingen, Tübingen, Germany Münster, Germany 22 Hertie Institute for Clinical Brain Research, University 45 Department of Neurology, Mayo Clinic Minnesota, of Tübingen, Tübingen, Germany Rochester, MN 55905, USA 23 Memory Unit, Department of Neurology, IIB Sant Pau, 46 Department of Psychiatry and Psychology, Mayo Clinic Hospital de la Santa Creu i Sant Pau, Universitat Autonoma Florida, Jacksonville, FL 32224, USA de Barcelona, Barcelona, Spain 47 24 Department of Neurology, Mayo Clinic Florida, Jacksonville, Clinical Neurochemistry Laboratory, Sahlgrenska University FL 32224, USA Hospital, Mölndal, Sweden 13 250 Acta Neuropathologica (2019) 138:237–250 48 57 Fundación Instituto Leloir-IIBBA-CONICET, Buenos Aires, Research Unit of Gynecology and Obstetrics, Department Argentina of Clinical Research, University of Southern Denmark, 49 Odense, Denmark Department of Neurology, University of Rostock, Rostock, 58 Germany Novo Nordisk Foundation Center for Basic Metabolic 50 Research, Section of Metabolic Genetics, Copenhagen, Laboratory of Neurogenetics, National Institute on Aging, Denmark National Institutes of Health, Bethesda, MD, USA 59 51 Department of Public Health, Section of Epidemiology, Department of Neurology, Hospital Universitario San Faculty of Health and Medical Sciences, University Sebastian, San Sebastian, Spain of Copenhagen, Copenhagen, Denmark 52 Department of Neurology, Johns Hopkins University Medical 60 MRC Integrative Epidemiology Unit, Bristol University, Center, Baltimore, MD 21287, USA Bristol, UK 53 Center of Neurology, Department of Neurodegenerative 61 Clinical Biochemistry and Pharmacology, Odense University diseases, Hertie‑Institute for Clinical Brain Research, Hospital, Odense, Denmark University of Tuebingen, Tuebingen, Germany 62 54 Department of Clinical Genetics, Odense University MRC Mitochondrial Biology Unit, University of Cambridge, Hospital, Odense, Denmark Cambridge CB2 0QQ, UK 63 55 Dutch Society for Research on Ageing, Leiden, Munich Cluster for Systems Neurology (SyNergy), Munich, The Netherlands Germany 64 56 Delft Bioinformatics Lab, Delft University of Technology, Department of Sociology, VU University, Amsterdam, Delft, The Netherlands The Netherlands 13