Multiomics of Bohring-Opitz syndrome truncating ASXL1 mutations identify canonical and noncanonical Wnt signaling dysregulation

Abstract

ASXL1 (additional sex combs-like 1) plays key roles in epigenetic regulation of early developmental gene expression. De novo protein-truncating mutations in ASXL1 cause Bohring-Opitz syndrome (BOS; OMIM #605039), a rare neurodevelopmental condition characterized by severe intellectual disabilities, distinctive facial features, hypertrichosis, increased risk of Wilms tumor, and variable congenital anomalies, including heart defects and severe skeletal defects giving rise to a typical BOS posture. These BOS-causing ASXL1 variants are also high-prevalence somatic driver mutations in acute myeloid leukemia. We used primary cells from individuals with BOS (n = 18) and controls (n = 49) to dissect gene regulatory changes caused by ASXL1 mutations using comprehensive multiomics assays for chromatin accessibility (ATAC-seq), DNA methylation, histone methylation binding, and transcriptome in peripheral blood and skin fibroblasts. Our data show that regardless of cell type, ASXL1 mutations drive strong cross-tissue effects that disrupt multiple layers of the epigenome. The data showed a broad activation of canonical Wnt signaling at the transcriptional and protein levels and upregulation of VANGL2, which encodes a planar cell polarity pathway protein that acts through noncanonical Wnt signaling to direct tissue patterning and cell migration. This multiomics approach identifies the core impact of ASXL1 mutations and therapeutic targets for BOS and myeloid leukemias.

Keywords: Development; Epigenetics; Genetic diseases; Genetics; Leukemias.

Figure 1. Multiomics study design for Bohring-Opitz syndrome (BOS) caused by pathogenic mutations in ASXL1.

(A) Schematic representation of the ASXL1 transcript (ENST00000375687.10) and protein (GenBank: NM_015338.6; GRCh37), its functional domains, and mutations causing BOS. Mutations listed correspond to patients in this study and are tagged with a Pt identifier. (B) Peripheral blood and dermal fibroblasts were collected and underwent epigenomic assays for ATAC-seq, CUT&RUN, and DNA methylation, and global transcriptome analysis using RNA-seq. (C) Across the multiomics assays and 2 specimen types, we had 8 of 18 BOS samples with fibroblast assays and 14 of 18 with blood assays. Four of 18 BOS patients had data from assays across both specimen types. (D) Promoter hypermethylation and closed chromatin, which can be examined with DNA methylation and ATAC-seq analysis, respectively, are associated with decreased transcription, while activating histone methylation such as H3K4Me3 at promoters and open chromatin are associated with increased transcription. (E) Western blot for representative BOS (n = 5) and representative control (n = 5) fibroblast whole-cell lysate extracts shows no significant difference in total ASXL1 protein. This was repeated 3 times. (F) Western blot for representative BOS (n = 5) and representative control (n = 5). Fibroblast histone extracts showed no significant difference in H2AK119ub, H3K4me3, and H3K27me3. This was repeated 3 times.

Figure 2. Pathogenic mutations in ASXL1 cause tissue-specific and tissue-independent effects on gene expression.

(A) RNA-seq raw read counts for the ASXL1 reference allele (black) and pathogenic allele (white, with black outline) at each BOS patient’s respective mutation in blood (n = 8) and (B) fibroblast (n = 7) samples. (C) RNA-seq heatmap of all significant DEGs with P_adj < 0.05 and abs(log₂FC) ≥ 0.58 between BOS (blue, n = 8) and control (pink, n = 11) blood found 1097 DEGs, with 590 of 1097 (53.8%) being more upregulated and (D) 155 DEGs between BOS (n = 7) and control (n = 7) fibroblasts, with 125 of 155 DEGs (80.6%) being more upregulated in BOS patients. (E) Blood RNA-seq gene ontology highlighted enrichment of genes related to nervous system development and canonical Wnt signaling pathway. (F) Fibroblast RNA-seq gene ontology highlighted enrichment of genes related to nervous system development. (G) Volcano plots for BOS compared to control RNA-seq in blood (x axis) and fibroblast (y axis) identified a core subset of 25 shared dysregulated transcripts, with 21 of 25 DEGs (84%) dysregulated in the same direction.

Figure 3. Epigenetic dysregulation in BOS patient–derived fibroblasts drives transcriptomic dysregulation.

(A) ATAC-seq heatmap of all significant DEGs with P_adj < 0.05 and abs(log₂FC) ≥ 0.58 between BOS (blue, n = 7) and control (pink, n = 7) fibroblasts found 4336 DEGs in fibroblasts, with 3036 of 4336 (70.0%) being more upregulated. (B) Gene set enrichment showed key dysregulated pathways, including nervous system development. (C) Integration of chromatin accessibility (ATAC-seq, x axis) and gene expression (RNA-seq, y axis) in BOS patient (n = 7) compared with control fibroblasts (n = 7) identified a positive correlation (R² = 0.405) between chromatin accessibility and gene expression. We identified a set of 37 common dysregulated transcripts (right). DEGs were considered significant (red) for abs(log₂FC) ≥ 0.58 and P_adj > 0.05.

Figure 4. DNA methylation drives transcriptomic dysregulation in BOS samples and identifies common dysregulated transcripts enriched in Wnt signaling genes.

(A) Integration of patient blood samples across DNAm (BOS n = 13, control n = 26) and RNA transcriptomic (BOS n = 8, control n = 11) dysregulation identified 672 differentially methylated CpG sites (P_adj < 0.05) that correlated to 341 RNA-seq DEGs (P_adj < 0.05). These significant DEGs were further filtered for RNA-seq abs(log₂FC) ≥ 0.58, and DNAm abs(Δβ) ≥ 0.05, shown by the dotted red lines. After filtering, we retained 50 of 672 CpG sites (7.44%) and 24 of 341 unique genes (7.04%). (B) Analysis of enriched biological processes identified canonical Wnt signaling, anterior-posterior body patterning, regulation of neuron projection development, and other biologically relevant pathways. (C) In BOS patients, PSMA8, which encodes a key component of the β-catenin destruction complex, is hypermethylated in blood DNAm across 8 CpG sites (Δβ 6.1% to 18.9%) and (D) shows strong downregulation in blood RNA-seq (log₂FC = –2.92). Control sample–normalized counts ranged from 0 to 17.7 (whiskers), with a mean (horizontal line) of 8.1, and quartile bounds (box limits) of 4.6 and 10.8. BOS normalized sample counts ranged from 0 to 2.1, with a mean of 0.8, and quartile bounds of 0 and 1.5.

Figure 5. Truncated ASXL1 dysregulates the canonical and noncanonical Wnt signaling pathways.

(A) The canonical Wnt signaling pathway (left) is activated when Wnt ligand stimulates its receptors. This inactivates the β-catenin destruction complex, allowing nuclear translocation of β-catenin and activation of target genes. Van Gogh–like 2 (VANGL2) intersects with the canonical pathway through activation of Dishevelled (DVL) to activate noncanonical pathways (right) and cell migration. (B) Whole-cell lysate (15 μg) of representative BOS- (n = 5) and control-derived (n = 5) fibroblasts show downstream Wnt pathway activation at the protein level through staining for VANGL2, β-catenin, axis-inhibition protein 1 (AXIN1), AXIN2, DVL2, and DVL3. (C) ImageJ (NIH) quantification identified an increase of 1.5-fold for AXIN1, 2.8-fold for AXIN2, 3.5-fold for DVL2, and 1.5-fold for DVL3 averaged across BOS patient samples compared to controls. This was repeated 2 times. In BOS patient samples, the Wnt pathway coreceptor LRP5 (green) transcriptional upregulation in (D) blood (log₂FC = 1.64, P_adj = 3.58 × 10^–9) and (E) fibroblast RNA-seq and (F) DNA hypomethylation in blood (Δβ –3.5% to –8.0%, FDR < 0.05) at multiple CpG sites. Similarly, LRP6 (green) shows that transcriptional upregulation in (G) blood RNA-seq (log₂FC = 1.63, P_adj = 1.17 × 10^–12), (H) fibroblast RNA-seq, and (I) DNA hypomethylation in blood (Δβ –2.7% to –4.0%, FDR < 0.05) BOS samples exhibit strong dysregulation of VANGL2 across tissue and assay types (pink). For BOS samples, VANGL2 is (J) hypomethylated (Δβ –7.6%) at CpG site cg17024258, and shows (K) increased chromatin accessibility at the 5′ UTR (log₂FC = 1.20). VANGL2 has significant transcriptional upregulation in (L) blood RNA-seq (log₂FC = 3.80) and (M) fibroblast RNA-seq (log₂FC = 2.55). In representative BOS patient samples, the VANGL2 promoter shows (N) increased chromatin accessibility and (O) increased H3K4me3 marks compared with control. The box-and-whisker plots in D–I, L, and M show the mean (horizontal line), range (whiskers), and IQR [upper and lower box boundaries]. LRP5 RNA-seq blood: control, 25.6 (18.3–39.1) [20.7–30.4]; BOS, 84.2 (24.1–134.6) [72.8–106.8]. LRP5 RNA-seq fibroblast: control, 3829 (2996–5177) [3187–4351]; BOS, 4392 (2824–5390) [4268–4719]. LRP6 RNA-seq blood: control, 75.0 (32.2–110.1) [67.8–86.6]; BOS, 234.5 (123.1–346.3) [199.9–266.9]. LRP6 RNA-seq fibroblast: control, 2804 (2286–3262) [2602–3094]; BOS, 2904 (2381–3596) [2725–3006]. VANGL2 RNA-seq blood: control: 2.3 (0–5.7) [0–4.1]; BOS, 35.3 (10.1–51.8) [27.2–45.2]. VANGL2 RNA-seq fibroblast: control, 134.8 (39.4–282.8) [90.8–163.4]; BOS, 1002.7 (464–1235.6) [894.7–1210.1].