Wnt/β-catenin signaling stimulates the expression and synaptic clustering of the autism-associated Neuroligin 3 gene

Abstract

Synaptic abnormalities have been described in individuals with autism spectrum disorders (ASD). The cell-adhesion molecule Neuroligin-3 (Nlgn3) has an essential role in the function and maturation of synapses and NLGN3 ASD-associated mutations disrupt hippocampal and cortical function. Here we show that Wnt/β-catenin signaling increases Nlgn3 mRNA and protein levels in HT22 mouse hippocampal cells and primary cultures of rat hippocampal neurons. We characterized the activity of mouse and rat Nlgn3 promoter constructs containing conserved putative T-cell factor/lymphoid enhancing factor (TCF/LEF)-binding elements (TBE) and found that their activity is significantly augmented in Wnt/β-catenin cell reporter assays. Chromatin immunoprecipitation (ChIP) assays and site-directed mutagenesis experiments revealed that endogenous β-catenin binds to novel TBE consensus sequences in the Nlgn3 promoter. Moreover, activation of the signaling cascade increased Nlgn3 clustering and co- localization with the scaffold PSD-95 protein in dendritic processes of primary neurons. Our results directly link Wnt/β-catenin signaling to the transcription of the Nlgn3 gene and support a functional role for the signaling pathway in the dysregulation of excitatory/inhibitory neuronal activity, as is observed in animal models of ASD.

Fig. 1. Wnt/β-catenin signaling activates Nlgn3 transcriptional program in hippocampal cells.

a Top: Genomic context of human NLGN3 in the long arm of chromosome X and schematic exon-intron boundaries of the gene. White and gray boxes: 5′ and 3′ UTR and exons, respectively. Middle: Conservation profile of the human NLGN3 promoter sequence compared with similar genomic regions in Mus musculus and Rattus norvegicus (50–100%). Bottom: Schematic representation of potential TCF/LEF sites (TBE: CTTTG, circles) found in these species. ECR: Evolutionary conserved region. b Early expression levels of Nlgn3 and cMyc genes after 2 h treatment with increasing doses of either purified Wnt3a protein or LiCl in HT22 hippocampal cells or rat primary hippocampal neurons (RHN). Rpl13a was used as a reference gene. c Quantitative determination of Nlgn3 and cMyc mRNA levels after 2 h treatment with Wnt3a (200–400 ng/ml) protein or LiCl (10–20 mM) in HT22 hippocampal cells and RHNs. d TOP: protein levels of β-catenin after 48 h treatment with β-catenin-shRNA. shRNA of GFP was used as control and β-actin as a loading control. Bottom: expression levels of Nlgn3 and β- catenin. e Nlgn3 and β-catenin protein levels in HT22 cells and RHNs after 6 h treatment with 200 and 400 ng/ml of purified Wnt3a protein. β-actin was used as a loading control. In (c) and (d), data represent mean ± s.e.m., *P < 0.05, **P < 0.01, two-tailed Mann–Whitney test

Fig. 2. Contribution of Wnt/β-catenin responsive TBE sites to Nlgn3 promoter activity.

a Genomic context of mouse Nlgn3 promoter and different Nlgn3 chimeric promoter constructs (pNL3Mm). Circles indicate potential TBE consensus sequences in the predicted regulatory sequence of Nlgn3 gene. Distances are in reference to the transcription start site (TSS). b Transient transfection of increasing doses (30, 60, and 90 ng) of Nlgn3 chimeric promoter constructs pNL3Mm2.8, pNL3Mm1.4, and pNL3Mm0.3 in HEK293 cells for 24 h. c Transient co-transfection of pNL3Mm2.8 kb or pNL3Mm1.4 with increasing doses of constitutively active β-catenin (S33Y) for 24 h. d Effects of dominant negative delta-TCF4 (ΔTCF4) on β-catenin induced pNL3Mm1.4 promoter activity. In (c, d) SuperTOPFLASH (STF) activity was measured as a positive control for Wnt/β-catenin pathway activation. RLU: relative luciferase units. In (b, c) Data represent mean ± s.e.m., *P < 0.05, one-way ANOVA and Kruskal–Wallis post hoc test. In (d) Data represent mean ± s.e.m., **P < 0.01, two-tailed Mann–Whitney test. ns not significant

Fig. 3. Functional β-catenin responsive element in the Nlgn3 promoter.

a Top panel: Representation of the 1.4 kb mouse Nlng3 promoter construct depicting the five TBE sites analyzed by ChIP assays. Bottom panel: Endogenous β-catenin binding to TBE Sites in the mouse Nlgn3 promoter examined in HT22 cells after treatment with either Wnt3a-CM for 4 h (left) or 20 mM LiCl for 24 h (right). Data represent mean ± s.e.m., **P < 0.01, ***P < 0.001, two-tailed Mann–Whitney test. b TBE sites (II–VI) were similarly examined by ChIP assays in HT22 cells under control conditions or treated with 1 µM CHIR 98014 for 4 h. In (a and b) immunoglobulin G (IgG) was used as a control. c Top panel: Representation of changes introduced by site-directed mutagenesis in TBE Sites II and III in the context of the 1.4 kb mouse Nlng3 promoter construct. TBE consensus sequence: CTTTG; Mutated TBE sequence: CCTCG. Bottom panel: Transient transfection of increasing doses of mutant plasmids in the presence of constitutively active β-catenin (S33Y) in HEK293 cells for 24 h. RLU: relative luciferase units. Data represent mean ± s.e.m., *P < 0.05, one-way ANOVA and Kruskal–Wallis post hoc test. ns not significant

Fig. 4. Wnt/β-catenin signaling enhances Nlgn3 levels and synaptic clustering in hippocampal neurons.

a Left panel: Representative confocal images of dendritic processes showing MAP2 (blue), Nlgn3 (red), and PSD-95 (green) immunofluorescence signals in primary hippocampal neurons under control conditions or treated with 400 ng/ml of purified Wnt3a protein for 2 h. White bar represents 2.5 µm. Right panel: Total Nlgn3 intensity fluorescence. b Same as in (a) but after 24 h Wnt3a treatment. c Left panel: Three-dimensional isosurface rendering of dendritic processes (smoothness, 0.2 µm; quality level, 5) and florescence intensity of Nlgn3 and PSD-95 clusters. Right panel: Colocalization coefficients for Nlgn3–PSD-95 interaction in control versus Wnt3a-treated neurons (Pearson and Manders, respectively). Each figure corresponds to three independents experiments (a ctrl n = 44 and Wnt3a n = 21; b ctrl n = 44 and Wnt3a n = 22; c ctrl n = 16 and Wnt3a n = 12). Data represent mean, *P < 0.05, **P < 0.01, two- tailed Mann–Whitney test. ns not significant