Genome-wide association study of brain amyloid deposition as measured by Pittsburgh Compound-B (PiB)-PET imaging

Abstract

Deposition of amyloid plaques in the brain is one of the two main pathological hallmarks of Alzheimer's disease (AD). Amyloid positron emission tomography (PET) is a neuroimaging tool that selectively detects in vivo amyloid deposition in the brain and is a reliable endophenotype for AD that complements cerebrospinal fluid biomarkers with regional information. We measured in vivo amyloid deposition in the brains of ~1000 subjects from three collaborative AD centers and ADNI using ¹¹C-labeled Pittsburgh Compound-B (PiB)-PET imaging followed by meta-analysis of genome-wide association studies, first to our knowledge for PiB-PET, to identify novel genetic loci for this endophenotype. The APOE region showed the most significant association where several SNPs surpassed the genome-wide significant threshold, with APOE*4 being most significant (P-meta = 9.09E-30; β = 0.18). Interestingly, after conditioning on APOE*4, 14 SNPs remained significant at P < 0.05 in the APOE region that were not in linkage disequilibrium with APOE*4. Outside the APOE region, the meta-analysis revealed 15 non-APOE loci with P < 1E-05 on nine chromosomes, with two most significant SNPs on chromosomes 8 (P-meta = 4.87E-07) and 3 (P-meta = 9.69E-07). Functional analyses of these SNPs indicate their potential relevance with AD pathogenesis. Top 15 non-APOE SNPs along with APOE*4 explained 25-35% of the amyloid variance in different datasets, of which 14-17% was explained by APOE*4 alone. In conclusion, we have identified novel signals in APOE and non-APOE regions that affect amyloid deposition in the brain. Our data also highlights the presence of yet to be discovered variants that may be responsible for the unexplained genetic variance of amyloid deposition.

Fig. 1

Distribution of PiB retention in the University of Pittsburgh (PITT) (a), Washington University (WU) (b), and the Alzheimer’s disease Neuroimaging Initiative (ADNI) and the Indiana Memory and Aging Study (ADNI/IU) (c) samples. SUVR standardized uptake volume ratio, BP binding potential

Fig. 2

a Quantile–quantile plot for the individual GWAS results in the University of Pittsburgh (PITT), Washington University (WU), and the Alzheimer’s disease Neuroimaging Initiative (ADNI) and the Indiana Memory and Aging Study (ADNI/IU) datasets and in the meta-analysis. λ is the genomic control value. b Manhattan plot showing the P-values in the meta-analysis. The blue line represents the suggestive significance line (P < E-05). The red line represents the significance threshold (P < 5E-08)

Fig. 3

Regional plot of the APOE region on chromosome 19 in the meta-analysis. The relative location of genes and the direction of transcription are shown in the lower portion of the figure, and the chromosomal position is shown on the x -axis. The light blue line shows the recombination rate across the region (right y -axis) and the left y-axis shows the significance of the associations. The purple diamond shows the P-value for rs429358 that is the most significant SNP in the meta-analysis. The circles show the P-values for all other SNPs and are color coded according to the level of LD with rs429358 in the 1000 Genome Project EUR population

Fig. 4

The functional analysis results for target genes. Out of 257 target genes, only genes meeting at least three functional criteria are listed. The criteria include: (1) differential expression in at least two Alzheimer disease studies that up- or downregulated consistently in different studies; (2) expression in the brain cells (Barres website); (3) having cis-eQTL effects in any brain tissues using GTEx database (P < 0.05); 4) mediating genetic effects on PiB (SMR analysis with P < 0.05) in any brain tissues; (5) having cis-eQTL effects in whole blood (P < 0.05); (6) mediating genetic effects on PiB (SMR analysis with P < 0.05) in whole blood; and (7) included in nominally significant pathways. The detailed results are summarized in Supplementary Table S6

Fig. 4

The functional analysis results for target genes. Out of 257 target genes, only genes meeting at least three functional criteria are listed. The criteria include: (1) differential expression in at least two Alzheimer disease studies that up- or downregulated consistently in different studies; (2) expression in the brain cells (Barres website); (3) having cis-eQTL effects in any brain tissues using GTEx database (P < 0.05); 4) mediating genetic effects on PiB (SMR analysis with P < 0.05) in any brain tissues; (5) having cis-eQTL effects in whole blood (P < 0.05); (6) mediating genetic effects on PiB (SMR analysis with P < 0.05) in whole blood; and (7) included in nominally significant pathways. The detailed results are summarized in Supplementary Table S6