A clustering of heterozygous missense variants in the crucial chromatin modifier WDR5 defines a new neurodevelopmental disorder

Abstract

WDR5 is a broadly studied, highly conserved key protein involved in a wide array of biological functions. Among these functions, WDR5 is a part of several protein complexes that affect gene regulation via post-translational modification of histones. We collected data from 11 unrelated individuals with six different rare de novo germline missense variants in WDR5; one identical variant was found in five individuals and another variant in two individuals. All individuals had neurodevelopmental disorders including speech/language delays (n = 11), intellectual disability (n = 9), epilepsy (n = 7), and autism spectrum disorder (n = 4). Additional phenotypic features included abnormal growth parameters (n = 7), heart anomalies (n = 2), and hearing loss (n = 2). Three-dimensional protein structures indicate that all the residues affected by these variants are located at the surface of one side of the WDR5 protein. It is predicted that five out of the six amino acid substitutions disrupt interactions of WDR5 with RbBP5 and/or KMT2A/C, as part of the COMPASS (complex proteins associated with Set1) family complexes. Our experimental approaches in Drosophila melanogaster and human cell lines show normal protein expression, localization, and protein-protein interactions for all tested variants. These results, together with the clustering of variants in a specific region of WDR5 and the absence of truncating variants so far, suggest that dominant-negative or gain-of-function mechanisms might be at play. All in all, we define a neurodevelopmental disorder associated with missense variants in WDR5 and a broad range of features. This finding highlights the important role of genes encoding COMPASS family proteins in neurodevelopmental disorders.

Keywords: COMPASS; Mendelian disorders; WDR5; de novo variants; intellectual disability; missense variants; multiple congenital abnormalities; neurodevelopmental disorders; next generation sequencing.

Figure 1

Clinical features reported in individuals with WDR5 variants Graphical overview of clinical features reported in 11 individuals with WDR5 missense variants. Growth parameters are shown as standard deviations to the mean for a certain age. All graphs include data for 11 individuals (N = 11). EEG abn., EEG abnormalities. A more detailed overview of clinical features can be found in Table S1.

Figure 2

Facial features in individuals with six different WDR5 variants Facial images of eight individuals with a heterozygous WDR5 variant. Several overlapping facial features are seen, such as a bulbous nasal tip (individuals 2, 4, and 10), low-set, posteriorly rotated, and/or dysplastic ears (individuals 2, 4, 7, and 10), ptosis (individual 11), and thin upper lip vermilion (individuals 4, 10, and 11). In addition, individuals 4 and 11 have severe micrognathia, a small mouth, and down-slanting palpebral fissures.

Figure 3

In silico, in vitro, and in vivo studies of the effect of WDR5 missense variants (A) Linear structure of WDR5 protein (334 amino acids) with the seven different WD40 domains and all identified missense variants shown; in total, six different missense variants were found in 11 individuals: one variant (p.(Thr208Met)) was found in five unrelated individuals and another variant (p.(Arg196Cys)) in two individuals. (B) Three-dimensional visualization of WDR5 (PDB: 2GNQ); locations of the amino acids involved in missense variants are shown with magenta balls. Colors of the different domains match with the colors used in (A). (C) Courtship memory was assessed in mushroom-body-specific wds RNAi knockdown flies (wds^RNAi) compared with controls expressing an RNAi against mCherry (mCherry^RNAi). Boxplots show the distribution of courtship indices (CIs) for naive (N) and trained (T) flies (top panel). Memory was observed when a significant reduction in CI occurred between naive and trained conditions (Mann-Whitney test). + indicates the mean. N is indicated along the x axis. Bar graphs show the memory indices (MIs), which are single values derived from the above CIs (indicated by arrows) according to the formula: MI = ($\overline{\text{X}}$ CI_naive − $\overline{\text{X}}$ CI_trained)/ $\overline{\text{X}}$ CI_naive. MIs were consistent between controls and wds^RNAi lines (randomization test, 10,000 bootstrap replicates). (D) Detection of WDR5 reference and variant proteins in adult flies by western blot. Representative bands for HA-tagged WDR5 reference and variant proteins at 36.6 kDa (top) along with the actin loading control 41 kDa (bottom). UAS-WDR5::HA reference and variant transgenes were expressed ubiquitously using Actin-Gal4. (E) Direct fluorescence micrographs of nuclei of HEK293T/17 cells expressing YFP-WDR5 fusion proteins (green). Nuclei were stained with Hoechst 33342 (blue). Scale bar: 10 μm. (F) WDR5 (green) is shown as part of the core COMPASS complex, with RbBP5 (yellow), ASH2L (blue), DPY30 (purple), and KMT2A (cyan) (PDB: 6KIV). The nucleosome is shown in gray. The locations of affected amino acids in individuals with missense variants are shown with magenta balls. (G) WDR5 (green; p.33–332) is shown together with RbBP5 (yellow; p.1–380) and KMT2A (cyan; p.3764–3969) as part of the core COMPASS complex (PDB: 6KIV). The locations of affected amino acids in individuals with missense variants are shown with magenta balls from three different angles facing the WIN site (I), the WBM site (II), and a side between WIN and WBM (III). (H) BRET assays for WDR5-RbBP5 interaction in live cells. Left, mean BRET saturation curves ±95% confidence interval fitted using a nonlinear regression equation assuming a single binding site (n = 4; y = BRETmax ∗ x/(BRET50/x); GraphPad) showing a strong BRET signal for Rluc-RbBP5 with YFP-WDR5. Right, corrected BRET values measured with an acceptor/donor ratio of 1:1 (n = 3, one-way ANOVA and post-hoc Bonferroni test). NLS, YFP fused to a C-terminal nuclear localization signal as control protein. p values show significance for the comparison of WDR5 variant with the WDR5 wild-type.