The genetic architecture of the human thalamus and its overlap with ten common brain disorders

Abstract

The thalamus is a vital communication hub in the center of the brain and consists of distinct nuclei critical for consciousness and higher-order cortical functions. Structural and functional thalamic alterations are involved in the pathogenesis of common brain disorders, yet the genetic architecture of the thalamus remains largely unknown. Here, using brain scans and genotype data from 30,114 individuals, we identify 55 lead single nucleotide polymorphisms (SNPs) within 42 genetic loci and 391 genes associated with volumes of the thalamus and its nuclei. In an independent validation sample (n = 5173) 53 out of the 55 lead SNPs of the discovery sample show the same effect direction (sign test, P = 8.6e-14). We map the genetic relationship between thalamic nuclei and 180 cerebral cortical areas and find overlapping genetic architectures consistent with thalamocortical connectivity. Pleiotropy analyses between thalamic volumes and ten psychiatric and neurological disorders reveal shared variants for all disorders. Together, these analyses identify genetic loci linked to thalamic nuclei and substantiate the emerging view of the thalamus having central roles in cortical functioning and common brain disorders.

Conflict of interest statement

T.E. is a consultant to BrainWaveBank and received speaker’s honoraria from Lundbeck and Janssen Cilag. O.A.A. is a consultant to BrainWaveBank and HealthLytix, and received speaker’s honoraria from Lundbeck. The remaining authors declare no competing interests.

Figures

Fig. 1. Analysis of the GWAS discovery sample identifies 42 loci associated with thalamic volumes.

a The thalamus was segmented into six nuclei groups—anterior, lateral, ventral, intralaminar, medial, and posterior nuclei—using Bayesian thalamus segmentation, . All data sets were assessed by visually inspecting axial view figures of the segmentations for each participant and we removed sets with segmentation errors, insufficient data quality, and pathologies. The nuclei volumes in the left and right thalamus were summed and these were used in the analyses. b Heritability estimates for the thalamic volumes in the discovery sample of n = 30,114 participants from the UK Biobank. All thalamic volumes showed substantial heritability. Data are presented as mean ± SE. cQ–Q plots for the thalamic volumes of the discovery sample. d Circular Manhattan plots of GWAS for thalamus volumes of the discovery sample. The innermost plot reflects the GWAS of whole thalamus volume, whereas from center to the periphery, the plots indicate the GWAS of the anterior, lateral, ventral, intralaminar, medial, and posterior nuclei, respectively. Black circular dashed lines indicate genome-wide significance (two-sided P < 7e − 9). Horizontal Manhattan plots for the seven volumes are shown in Supplementary Fig. 2. The colors in a, c, and d indicate the same volumes, i.e., red color reflects whole thalamus; orange, yellow, and light green indicate the anterior, lateral, and ventral nuclei, respectively; whereas dark green, blue, and magenta reflect intralaminar, medial, and posterior nuclei volumes, respectively. GWAS, genome-wide association studies; h2, heritability.

Fig. 2. GWGAS identifies 127 unique genes associated with thalamic volumes.

Nineteen genes were associated with whole thalamus; 4, 29, and 17 genes were associated with volumes of the anterior, lateral, and ventral nuclei; and 37, 11, and 21 genes were associated with intralaminar, medial, and posterior nuclei volumes, respectively. Lighter font color and higher position for gene names indicate greater Z-score. Underlined gene names designate genes that were significantly associated with more than one volume, whereas gene names not underlined indicate genes associated with only one volume. Additional results of the GWGAS are presented in Supplementary Data 12. GWGAS, genome-wide gene-based association analysis.

Fig. 3. Thalamocortical genetic relationships.

We found significant genetic correlations between all thalamic nuclei and distinct cortical regions. There were significant positive genetic correlations between anterior nuclei and medial premotor cortex; between lateral nuclei and mainly medial prefrontal, anterior cingulate, parietal, and visual cortices; between medial nuclei and prefrontal and temporal cortices; and between posterior nuclei and visual cortices. We also found significant negative genetic associations between ventral nuclei and visual, prefrontal, and temporal cortical regions, and between intralaminar nuclei and rostral medial prefrontal cortex. Warm and cool colors in cortical regions indicate significant positive and negative genetic correlations, respectively, after adjusting for analyses of 180 cortical regions and 6 volumes (two-sided P < FDR). Corresponding statistics are provided in Supplementary Data 15. FDR, false discovery rate; Rg, genetic correlation.

Fig. 4. LD-score regression-based genetic correlations between thalamic volumes and ten brain disorders.

We assessed genetic correlations between thalamic volumes and ten brain disorders using LD-score regression. Warm and cool colors indicate positive and negative genetic associations, respectively. There were significant positive correlations between the whole thalamus and PD, between posterior nuclei and BD, and between intralaminar nuclei and MS, as indicated by a black frame, after FDR correcting across all 70 analyses (7 volumes x 10 disorders; two-sided P < FDR). AD; Alzheimer’s disease. ADHD; attention-deficit hyperactivity disorder. ASD, autism spectrum disorder; BD, bipolar disorder; FDR, false discovery rate; FEP, focal epilepsy; GEP, generalized epilepsy; MD, major depression; MS, multiple sclerosis; PD, Parkinson’s disease; SCZ, schizophrenia.

Fig. 5. Genetic loci shared between thalamic volumes and ten brain disorders.

Conjunctional FDR analysis detected shared genetic loci across thalamic volumes and the ten disorders. The figure shows results with FDR thresholds of both 0.05 (circles) and 0.01 (triangles). We found the largest number of overlapping loci for SCZ (66), PD (26), and BD (15), when applying a conjunctional FDR threshold of 0.05. For ASD, ADHD, MD, MS, GEP, FEP, and MS, there were 8, 8, 17, 10, 14, 4, and 14 genetic loci jointly associated with thalamic volumes and disorders when using a threshold of 0.05, respectively. When using a conjunctional FDR threshold of 0.01, there were overlapping loci associated with thalamic volumes and SCZ (17), PD (14), BD (5), ASD (2), ADHD (1), MDD (3), MS (2), and AD (2), and no shared locus for GEP or FEP. AD, Alzheimer’s disease; ADHD, attention-deficit hyperactivity disorder; ASD, autism spectrum disorder; BD, bipolar disorder; FDR, false discovery rate; FEP, focal epilepsy; GEP, generalized epilepsy; MD, major depression; MS, multiple sclerosis; PD, Parkinson’s disease; SCZ, schizophrenia.