Rare deletions at 16p13.11 predispose to a diverse spectrum of sporadic epilepsy syndromes

Abstract

Deletions at 16p13.11 are associated with schizophrenia, mental retardation, and most recently idiopathic generalized epilepsy. To evaluate the role of 16p13.11 deletions, as well as other structural variation, in epilepsy disorders, we used genome-wide screens to identify copy number variation in 3812 patients with a diverse spectrum of epilepsy syndromes and in 1299 neurologically-normal controls. Large deletions (> 100 kb) at 16p13.11 were observed in 23 patients, whereas no control had a deletion greater than 16 kb. Patients, even those with identically sized 16p13.11 deletions, presented with highly variable epilepsy phenotypes. For a subset of patients with a 16p13.11 deletion, we show a consistent reduction of expression for included genes, suggesting that haploinsufficiency might contribute to pathogenicity. We also investigated another possible mechanism of pathogenicity by using hybridization-based capture and next-generation sequencing of the homologous chromosome for ten 16p13.11-deletion patients to look for unmasked recessive mutations. Follow-up genotyping of suggestive polymorphisms failed to identify any convincing recessive-acting mutations in the homologous interval corresponding to the deletion. The observation that two of the 16p13.11 deletions were larger than 2 Mb in size led us to screen for other large deletions. We found 12 additional genomic regions harboring deletions > 2 Mb in epilepsy patients, and none in controls. Additional evaluation is needed to characterize the role of these exceedingly large, non-locus-specific deletions in epilepsy. Collectively, these data implicate 16p13.11 and possibly other large deletions as risk factors for a wide range of epilepsy disorders, and they appear to point toward haploinsufficiency as a contributor to the pathogenicity of deletions.

Copyright (c) 2010 The American Society of Human Genetics. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1

16p13.11 Deletions Observed Exclusively in Epilepsy Patients A total of 23 deletions were observed in the region (indicated by blue bars marking the locations of patient-specific deletions), 22 of them sharing a common segment including or disrupting the NDE1 gene. Deletions in the same patient that were called in tandem and separated only by SNPs that failed genotyping quality control were assumed to be continuous and merged together (shown as a single bar in this display). Using the exact same criteria in controls, we observed no deletions exceeding 16 kb. Segmental duplications flanking the boundaries of these deletions are shown. Figure produced in part with the use of the UCSC Genome Browser.

Figure 2

Confirmation of 16p13.11 Deletions (A) CGH experiment confirming the 16p13.11 deletion. Intensity signals in a single patient with a 16p13.11 deletion detected via PennCNV on Illumina-based genotyping technology (top panel) compared to data collected on the same deletion patient via CGH (Roche NimbleGen, bottom panel). Deletion regions called by the two technologies are shown in red, and the coordinates defining the start and end points of the deletions are provided above each panel. (B) A set of 14 individuals were regenotyped on the Illumina HumanHap 610 genotyping chips for replication of initial calls. All 14 deletions replicated, and the sizes of deletions called were highly correlated. (C) Confirmation of the 16p13.11 deletion with the use of Taqman-based real-time genomic amplification.

Figure 3

Expression Effects of a Subset of 16p13.11 Deletions in Leukocytes (A) Shown is the number of SDs away from the group mean for highly expressed transcripts located within the deletion (red) and a 1 Mb surrounding region (blue). Data are summarized for seven patients with the 16p13.11 deletion (mean ± standard error of the mean) compared to eight controls. (B) Shown is the number of SDs away from the mean for each individual transcript evaluated in individual subjects. The deleted region in each patient is highlighted in yellow. Position of the transcript along the x axis in both panels is equal to the position of the midpoint of the probe measuring the transcript.

Figure 4

Screen for Recessive Variants on the Intact Chromosome of 16p13.11 Individuals with deletions (indicated by the blue bars marking the locations of patient-specific deletions) were selected for a hybridization-based capture and next-generation sequencing experiment in which the intact homologous stretch of chromosome corresponding the deleted segment was sequenced in order to screen for pathogenic recessive mutations that may contribute to the risk associated with the deletion. The average read depth averaged across all ten subjects at each base is shown, and purple bars below mark the regions targeted in the hybridization-based capture. Figure produced in part with the use of the UCSC Genome Browser.