Sex-biasing influence of autism-associated Ube3a gene overdosage at connectomic, behavioral, and transcriptomic levels

Abstract

Genomic mechanisms enhancing risk in males may contribute to sex bias in autism. The ubiquitin protein ligase E3A gene (Ube3a) affects cellular homeostasis via control of protein turnover and by acting as transcriptional coactivator with steroid hormone receptors. Overdosage of Ube3a via duplication or triplication of chromosomal region 15q11-13 causes 1 to 2% of autistic cases. Here, we test the hypothesis that increased dosage of Ube3a may influence autism-relevant phenotypes in a sex-biased manner. We show that mice with extra copies of Ube3a exhibit sex-biasing effects on brain connectomics and autism-relevant behaviors. These effects are associated with transcriptional dysregulation of autism-associated genes, as well as genes differentially expressed in 15q duplication and in autistic people. Increased Ube3a dosage also affects expression of genes on the X chromosome, genes influenced by sex steroid hormone, and genes sex-differentially regulated by transcription factors. These results suggest that Ube3a overdosage can contribute to sex bias in neurodevelopmental conditions via influence on sex-differential mechanisms.

Fig. 1.. Increased Ube3a dosage affects global fMRI connectivity in a sex-dependent manner.

(A) Contrast maps (left panel) illustrating difference in global fMRI connectivity strength between WT (n = 20) and Ube3a2X (n = 20) animals, irrespective of sex (blue indicates reduced connectivity, t test, t > 2; FWE cluster-corrected). Panel on the right illustrates quantification of global fMRI connectivity strength in representative regions of interest (t test; Hypothalamus, t = 2.44, P = 0.019; Thalamus, t = 3.36, P = 0.002). (B) Contrast maps (left panel) showing areas exhibiting sex*genotype interaction in global fMRI connectivity strength (purple-yellow indicates areas with significant interaction, t > 2; FWE cluster-corrected). Panel on the right illustrates the quantification of sex*genotype interaction in the hypothalamus (ANOVA, sex*genotype interaction, F = 5.85, P = 0.02). (C) Contrast maps (left panel) showing areas exhibiting decreased global fMRI connectivity strength in female Ube3a2X mice (n = 10) compared to female WT (n = 10) littermates (t test, t > 2; FWE cluster-corrected). The plot on the right illustrates quantification of global fMRI connectivity strength in representative regions of interest (t test; Hypothalamus, t = 4.50, P < 0.001; Thalamus, t = 2.44, P = 0.026). (D) Contrast maps (left panel) showing regions exhibiting increased global fMRI connectivity strength in male Ube3a2X mice (n = 10) compared to male WT (n = 10) littermates (red indicates increased connectivity, t test, t > 2; FWE cluster-corrected). Panel on the right illustrates quantification of global fMRI connectivity strength in both groups of males in representative regions of interest (t test; VDB, t = 3.33, P = 0.004; PFC, t = 2.90, P = 0.009). BF, basal forebrain; Hypo, hypothalamus; PFC, prefrontal cortex; Thal, thalamus; VDB, ventral diagonal band; Mid, Midbrain. *P < 0.05, **P < 0.01. Error bars indicate SEM.

Fig. 2.. Divergent fMRI connectivity profiles in male and female Ube3a2X mutants.

Seed-based connectivity mapping of (A) hypothalamus and (B) PFC. Contrast maps show areas exhibiting sex*genotype interaction of connectivity to the seed (purple-yellow coloring, sex*genotype interaction, t > 2; FWE cluster-corrected). The plots on the right illustrate the quantification of sex*genotype interaction of connectivity strength between the seed and the region of interest (ANOVA, Hypo-SS F = 4.42, P = 0.04, PFC-Thal F = 4.62, P = 0.04). Seed-based connectivity mapping of (C) hypothalamus in WT and Ube3a2X female mice, and (D) PFC in male WT and Ube3a2X mutants. Red-yellow coloring represents regions exhibiting fMRI connectivity with the seed region in control and Ube3a2X mice (WT, top panels; Ube3a2X, middle panels; one-sample t test, t > 3). Contrast maps are at the bottom of the panel (blue indicates reduced connectivity in Ube3a2X females, red indicates increased connectivity in Ube3a2X males, t test, t > 2). All statistics are FWE cluster-corrected. Quantification of connectivity strength between seeds and region of interest in the (E) female and (F) male groups (*P < 0.05, **P < 0.01, unpaired t test). Seed regions are indicated in red lettering. BF, basal forebrain; Hypo, hypothalamus; PFC, prefrontal cortex; Thal, thalamus; dHPC, dorsal hippocampus; Ins, insula; SS, somatosensory cortex; CPu, Caudate Putamen. *P < 0.05, **P < 0.01. Error bars indicate SEM, and each dot represents a mouse. Ube3a2X, n = 20 versus WT n = 20, n = 10 males and females within each group.

Fig. 3.. Ube3a dosage affects stereotyped behavior in a sex-dependent manner.

(A) Rotarod test to assess locomotor activity. (B) Quantification of latency to fall (Ube3a2X n = 26, n = 14 males and n = 12 females; WT n = 30, n = 14 males and n = 16 females). Sex*genotype interaction was not significant (ANOVA, F = 0.29, P = 0.59). Both sex and genotype factors were instead significant (F = 12.3, P < 0.001 and F = 16.2, P < 0.001, respectively), driven by decreased latency in transgenic males (Tukey’s post hoc test, ***P < 0.001). (C) Cohen’s d effect size for the latency to fall. (D) Schematics of the self-grooming test. (E) Quantification of time spent grooming (Ube3a2X n = 20, n = 10 males and females; WT n = 20, n = 10 males and females). Sex*genotype interaction was significant (ANOVA, F = 10.95, **P = 0.002) and driven by increased grooming in male mutants. (F) Habituation/dishabituation social interaction test. (G) Social interaction duration in the habituation/dishabituation test for all trials (WT n = 25, n = 12 males and n = 13 females; Ube3a2X n = 18, n = 10 males and n = 8 females). (H) Cumulative social interaction during the first four trials of the habituation/dishabituation test. Sex*genotype interaction was not significant (ANOVA, F = 0.21, P = 0.64). (I) Schematics of the three-chamber test. (J) Quantification of the sociability index (WT n = 30, n = 14 males and n = 16 females; Ube3a2X n = 26, n = 14 males and n = 12 females). Sex*genotype interaction was not significant (ANOVA, F = 0.11, P = 0.73). (K) Quantification of novelty index. Sex*genotype interaction was not significant (ANOVA, F = 0.33, P = 0.57).

Fig. 4.. Sex-specific PFC transcriptomic dysregulation by Ube3a overexpression and enrichment with autism-associated, dup15q, and sex-relevant mechanisms.

(A) Plots display log(CPM) for shank3 and gabra4, two examples of DE genes for the sex*genotype interaction (M+F−, light blue; M−F+, orange). High-confidence SFARI genes that belong to each of the two groups are at the bottom. (B) Heatmap showing enrichments with gene lists from dup15q (dup15q DE) (43), SFARI genes, private (pt) inherited likely gene disrupting (LGD) variants (58), autism spectrum disorders (ASD), schizophrenia (SCZ) and bipolar disorder (BD) (59), iPSC-derived neurons from dup15q (dup15q iPSC), and Angelman syndrome (AS iPSC) individuals (60). The symbols on the acronyms indicate down-regulated (minus) or up-regulated (plus) expression. (C) Plot showing the percentage of genes per each chromosome that are DE−. Color indicates the enrichment OR. The X chromosome is in red. The vertical dotted line indicates the FDR threshold. (D) Heatmap showing enrichments between sex*genotype interaction DE genes and genes relevant to sex hormones or sex-differential gene regulation. DHT DE+ and DHT DE− are genes that are up-regulated (plus) or down-regulated (minus) after dihydrotestosterone (DHT) manipulation (7). EST DE+ and EST DE− are genes that are up-regulated (plus) or down-regulated (minus) after treatment with estrogen (EST) (115). AR Targets are downstream target genes of the androgen receptor (AR) as defined by chromatin immunoprecipitation sequencing in human neural stem cells (36). ER Targets are downstream target genes of the estrogen receptor (ER) (37). Male DT, Female DT, and SexDiv DT genes are sex-differentially targeted by transcription factors (116). The numbers in each cell indicate the enrichment OR, and the color indicates the–log₁₀(P value) for the enrichment test. Cells outlined in black pass FDR. FDR q < 0.05 threshold for multiple comparisons correction.

Fig. 5.. Sex-specific PFC transcriptomic dysregulation by Ube3a overexpression affects convergent ASD-relevant biological systems, pathways, and cell types.

(A) PPI graph of PFC sex-by-genotype interaction genes. Nodes are colored according to a k-means clustering solution with k = 3. These clusters also segregate genes with GO BP enrichment terms specified in text next to each cluster. Nodes are circled in black if they are SFARI ASD genes. Nodes circled in turquoise are sex hormone–related proteins, whereas magenta circled nodes are sex hormone receptors. Ube3a is circled in orange. Fmr1 and Tsc2 are circled in red to indicate that these two translation and protein synthesis relevant genes also serve as gene connector hubs in the identified transcriptional network. (B) Cell type enrichment heatmap showing how M+F− or M−F+ PFC sex-by-genotype interaction gene sets (columns) are enriched in numerous cell type markers (specified on the rows) from the Allen Institute mouse scRNA-seq data (65). M−F+ genes strongly hit a variety of glutamatergic and GABAergic cell types, whereas M+F− genes show specific enrichments with astrocyte and oligodendrocyte cell types. The numbers in each cell indicate the enrichment OR, while the coloring indicates the–log₁₀(P value). Only enrichments significant at FDR q < 0.05 are shown.