Exome sequencing identifies rare damaging variants in ATP8B4 and ABCA1 as risk factors for Alzheimer's disease

Abstract

Alzheimer's disease (AD), the leading cause of dementia, has an estimated heritability of approximately 70%¹. The genetic component of AD has been mainly assessed using genome-wide association studies, which do not capture the risk contributed by rare variants². Here, we compared the gene-based burden of rare damaging variants in exome sequencing data from 32,558 individuals-16,036 AD cases and 16,522 controls. Next to variants in TREM2, SORL1 and ABCA7, we observed a significant association of rare, predicted damaging variants in ATP8B4 and ABCA1 with AD risk, and a suggestive signal in ADAM10. Additionally, the rare-variant burden in RIN3, CLU, ZCWPW1 and ACE highlighted these genes as potential drivers of respective AD-genome-wide association study loci. Variants associated with the strongest effect on AD risk, in particular loss-of-function variants, are enriched in early-onset AD cases. Our results provide additional evidence for a major role for amyloid-β precursor protein processing, amyloid-β aggregation, lipid metabolism and microglial function in AD.

Fig. 1. Study setup and power.

a, Schematic of the study setup. The AD association of genes identified in stage 1 was confirmed in stage 2 and significance was determined by meta-analysis. Variant characteristics were investigated in a merged mega-sample rather than the meta-sample, allowing more accurate variant effect size estimates for variant categories/age-at-onset bins. The mega-sample (without exome extracts) was also used for the GWAS gene burden analysis. MTC, multiple testing correction. b, Top, number of genes (y axis) with at least a certain cumulative carrier frequency of prioritized variants (x axis), prioritized according to different deleteriousness thresholds. White box, genes with a cMAC ≥ 10 (cumulative minor allele count of ≥10 prioritized alleles identified across the 12,652 cases and 8,693 controls in the stage 1 sample) were considered to have sufficient carrier frequency to allow burden analysis. The SORL1, TREM2 and ABCA7 genes are indicated, revealing that carriers of rare damaging variants in these genes are relatively common, allowing identification in smaller sample sizes^–. Bottom, power analysis for stage 1, to attain a P < 1 × 10⁻⁶, at the same scale as the top figure. For comparison, we indicate 80% power thresholds for sample sizes of 1,000 and 5,000 individuals (subsampled from stage 1). Cumulative carrier frequency and estimated effect size ranges are indicated for common variants identified to associate with AD by GWAS (green), rare-variant burdens in SORL1, TREM2 and ABCA7 identified using sequencing studies^– (grey/blue), and for rare variants observed in autosomal dominant AD (magenta). Common variants with high effect sizes (red) are not expected to exist. Genes with cMAC < 10 were not analyzed (pink). Power calculations show that aggregating more cases and controls might allow for the identification of rare-variants that have a large effect on AD but for which only few carriers are observed, or for variants that have a modest/average effect on AD, for which many carriers are observed (power calculations shown in Supplementary Table 6). c, Quantile–quantile plot of P values determined in the stage 1 discovery analysis based on an ordinal logistic burden test. For each of 13,222 genes, we tested the burden of variants adhering to four variant deleteriousness thresholds, conditional on having a cMAC ≥ 10 (n = 31,204 tests). Threshold for multiple testing correction: FDR < 0.1, P value inflation, 1.046. Gene names in black indicate the deleteriousness threshold of the most significant burden test in that gene.

Fig. 2. Characterization of gene-specific variant features based on the mega-sample.

For all variant features, we considered the deleteriousness threshold that provides the most evidence for AD association in the meta-analysis. Variant features were investigated in a merged mega-sample (n = 31,905) instead of the meta-sample because this allows for increased accuracy for estimations of variant effect sizes for each variant category/age-at-onset bin (Table 3, refined burden). a, Carrier frequency according to age at onset. A carrier carries at least one damaging variant in the considered gene. b, ORs according to age at onset. The effect size significantly decreased with age at onset for SORL1, TREM2, ABCA7, ABCA1 and ADAM10 (after multiple testing correction; Supplementary Table 9). c, ORs according to variant frequency. The rareness of variants in SORL1 was significantly associated with the effect size (Supplementary Table 11). d, cMAC by variant frequency: the stacked total number of cases (dark) and controls (light) that carry gene variants with allele frequencies as observed in the mega-sample. The numbers above the bars indicate the number of contributing variants. Whiskers: 95% CI. Genes in black: genes identified to significantly associate with AD in the meta-analysis; gray: genes not significantly associated with AD in the meta-analysis; blue: genes identified by the targeted GWAS analysis, these were not significantly associated with AD in the meta-analysis. Source data

Fig. 3. ORs according to age at onset and variant pathogenicity.

ORs for LOF (red) and missense (yellow) variants as observed in the mega-sample (n = 31,905). Case/control OR (square, 95% CI), EOAD OR (triangle pointing upward), LOAD OR (triangle pointing downward). Missense variants in the considered gene appertained to the variant deleteriousness threshold that provides the most evidence for its AD association (Table 3, refined). The LOF burden effect size was significantly larger than the missense burden effect size in the SORL1 and we observed similar trends in ABCA7 and ABCA1 (Supplementary Table11). Of note, for ZCWPW1 only the burden of the LOF variants was significantly associated with AD; missense variants are shown for reference purposes (REVEL > 25).Grey: gene was not significantly associated with AD in the meta-analysis. Source data

Extended Data Fig. 1. Age, gender, APOE genotype distribution.

Age, gender and APOE genotype distribution of all samples, stratified by case/control status.

Extended Data Fig. 2. PCA: Sample population compared to 1,000 G population samples.

Sample population compared to 1,000 G population samples. First two PCA components of the study samples used for the Stage 1 and Stage 2 analysis, shown in context of the 1000 Genomes samples for reference (see Supplementary Note section 1.3.4). Samples in red are considered population outliers. Samples with only exome-extracts were not included in this analysis.

Extended Data Fig. 3. P value inflation in Stage-2 analysis.

P value inflation in Stage-2 analysis: Quantile-quantile plot for Stage-2 (without exome-extract samples), based on a ordinal logistic burden test (see Methods). Results are shown for all burden tests (n = 20,681) for which at least 10 damaging alleles were present in this dataset (based on 4 different variant deleteriousness thresholds per gene). While not used in this analysis, the threshold for multiple testing correction based on FDR < 0.1 is shown for reference. The genomic p-value inflation was 1.016. Note that causative mutations were not separately removed in Stage-2, as we focused on a specific set of genes.

Extended Data Fig. 4. P value inflation in the mega-analysis dataset.

P value inflation in the mega-analysis dataset: Quantile-quantile plot for the mega-analysis dataset (without exome-extract samples) based on a ordinal logistic burden test (see Methods). Results are shown for all burden tests (n = 37,710) for which at least 10 damaging alleles were present in this dataset (based on 4 different variant deleteriousness thresholds per gene). For reference, the threshold for multiple testing correction based on a false discovery rate threshold of 0.1 is shown. P values for the mega-analysis are shown in Supplementary Table 15. The genomic p-value inflation was 1.025.

Extended Data Fig. 5. Variant carrier frequency in controls by age last seen.

Variant carrier frequency in controls by age last seen: Carrier frequency in controls by age last seen for the variant selection threshold with the strongest association, as observed in the mega-analysis (n = 31,905 unique individuals); RIN3, CLU, ZCWPW1, ACE (n = 29,727 unique individuals; that is without exome-extracts) (Table 3, refined). Black: genes significant in the meta-analysis. Grey: genes not significant in meta-analysis. Blue: genes detected in the GWAS targeted analysis.

Extended Data Fig. 6. Age-at-onset by variant deleteriousness category.

Age-at-onset by variant deleteriousness category: Age-at-onset (median and IQR) in the mega-analysis (n = 31,905 unique individuals); RIN3, CLU, ZCWPW1, ACE (n = 29,727 unique individuals; that is without exome-extracts). Samples in variant deleteriousness categories with <10 samples are shown individually. The median age at onset and IQR for the complete mega-analysis dataset is shown on the right. Black: genes significant in the meta-analysis. Grey: genes not significant in meta-analysis. Blue: genes detected in the GWAS targeted analysis.

Extended Data Fig. 7. Attributable fraction per gene and age-at-onset category.

Attributable fraction per gene and age-at-onset category: Attributable fractions as derived based on the mega-analysis in the mega-analysis (n = 31,905 unique individuals); RIN3, CLU, ZCWPW1, ACE (n = 29,727 unique individuals; that is without exome-extracts). The attributable fraction of a gene is an estimate of the fraction of AD cases in a specific age group that have become part of this dataset due to carrying a rare damaging variant in the respective gene (Methods). This estimate accounts only for variants in the burden selection. Black: genes significant in the meta-analysis. Grey: genes not significant in meta-analysis. Blue: genes detected in the GWAS targeted analysis.

Extended Data Fig. 8. Sensitivity Analysis: AD vs Age association.

AD vs Age association: Sensitivity analysis of the gene burden tests (for the most significant deleteriousness thresholds, Table 2) for the mega-analysis dataset (RIN3, CLU, ZCWPW1, ACE: without exome-extracts) (respectively n = 31,905 and n = 29,727 unique individuals). Comparison of the case/control odds ratio of an age-matched and a non-age-matched analysis. Age-matching was performed as described in the methods. Based on the confidence intervals, we cannot exclude that the signals in ACE, ADAM10 and ZCWPW1 are affected by other age-related conditions. Note however, that the signals in ADAM10 and ZCWPW1 are based on very few variants, such that confidence intervals are expected to be wide.