Genetic Suppression of mTOR Rescues Synaptic and Social Behavioral Abnormalities in a Mouse Model of Pten Haploinsufficiency

Abstract

Heterozygous mutations in PTEN, which encodes a negative regulator of the mTOR and β-catenin signaling pathways, cause macrocephaly/autism syndrome. However, the neurobiological substrates of the core symptoms of this syndrome are poorly understood. Here, we investigate the relationship between cerebral cortical overgrowth and social behavior deficits in conditional Pten heterozygous female mice (Pten cHet) using Emx1-Cre, which is expressed in cortical pyramidal neurons and a subset of glia. We found that conditional heterozygous mutation of Ctnnb1 (encoding β-catenin) suppresses Pten cHet cortical overgrowth, but not social behavioral deficits, whereas conditional heterozygous mutation of Mtor suppresses social behavioral deficits, but not cortical overgrowth. Neuronal activity in response to social cues and excitatory synapse markers are elevated in the medial prefrontal cortex (mPFC) of Pten cHet mice, and heterozygous mutation in Mtor, but not Ctnnb1, rescues these phenotypes. These findings indicate that macroscale cerebral cortical overgrowth and social behavioral phenotypes caused by Pten haploinsufficiency can be dissociated based on responsiveness to genetic suppression of Ctnnb1 or Mtor. Furthermore, neuronal connectivity appears to be one potential substrate for mTOR-mediated suppression of social behavioral deficits in Pten haploinsufficient mice. Autism Res 2019, 12: 1463-1471. © 2019 International Society for Autism Research, Wiley Periodicals, Inc. LAY SUMMARY: A subgroup of individuals with autism display overgrowth of the head and the brain during development. Using a mouse model of an autism risk gene, Pten, that displays both brain overgrowth and social behavioral deficits, we show here that that these two symptoms can be dissociated. Reversal of social behavioral deficits in this model is associated with rescue of abnormal synaptic markers and neuronal activity.

Keywords: PTEN; macrocephaly; network activity; social behavior; synapses.

Figure 1.

Brain overgrowth in Pten conditional heterozygous mice is rescued in Pten and Ctnnb1 double conditional heterozygous mice. Graph showing brain mass in Control, Emx1-Cre⁺; Pten^loxp/+, Emx1-Cre⁺; Pten^loxp/+; Mtor^loxp/+, and Emx1-Cre⁺; Pten^loxp/+; Ctnnb1 ^loxp/+ mice. One-way ANOVA and Tukey’s posthoc tests were used. F(3,12) = 7.89, P = 0.004. *P < 0.05, and **P < 0.01. n.s. indicates no significant difference. N = 4 animals per genotype.

Figure 2.

Three-chamber social approach deficits in Pten conditional heterozygous mice are corrected in Pten and Mtor double conditional heterozygous mice. (A) Graph showing percent time female mice spent in each chamber. Paired t-tests were used to compare time spent in the chambers containing a mouse in a tube (Mouse chamber, M) and an empty tube (Object chamber, O) separately for each genotype. *P < 0.05, and **P < 0.01. n.s. indicates no significant difference. (B) Example heat maps showing the time spent in each chamber. M indicates mouse chamber, and O indicates object chamber. (C) Graph showing the velocity of the mice during the 10 min sociability test. Velocity is not altered across genotypes. One-way ANOVA was used. F(3,41) = 0.32, P = 0.81. N = 12 Control, 12 Emx1-Cre⁺; Pten^loxp/+, 9 Emx1-Cre⁺; Pten^loxp/+; Mtor^loxp/+, and 12 Emx1-Cre⁺; Pten^loxp/+; Ctnnb1 ^loxp/+ mice.

Figure 3.

Increased vGluT1 puncta in Pten conditional heterozygous mice is corrected in Pten and Mtor double conditional heterozygous mice. (A) Cartoon and an image showing the staining of the vesicular glutamate transporter 1 (vGluT1) and the vesicular GABA transporter (vGAT) in the prelimbic cortex (PrL). vGluT1 and vGAT puncta were quantified separately in outer and inner layer 1. Scale bar: 50 μm. (B) Representative images of vGluT1 and vGAT staining in the PrL of each genotype. Scale bar: 10 μm. (C–F) Quantification of vGluT1 and vGAT puncta in outer (C and D) and inner (E and F) layer 1 of PrL. N = 4 animals per genotype. (G) Cartoon and an image showing the staining of the vGluT1 and vGAT in the basolateral amygdala (BLA). Scale bar: 50 μm. (H and I) Quantification of vGluT1 (H) and vGAT (I) puncta in the BLA. N = 4 animals each genotype. (J) Representative images showing fluorescent Nissl staining of layerV cells in the primary somatosensory cortex. Scale bar: 50 μm. (K) Quantification of layer V cell soma size. N = 5 animals per genotype. For all quantification graphs, one-way ANOVA was used, and Tukey’s post hoc tests were used where appropriate. *P < 0.05, and **P < 0.01. n.s. indicates no significant difference. F(3, 12) = 8.25, P = 0.003 (C); F(3, 12) = 1.01, P = 0.421 (D); F(3, 12) = 0.641, P = 0.603 (E); F(3, 12) = 0.337, P = 0.799 (F); F(3, 12) = 1.95, P = 0.176 (H); F(3, 12) = 1.94, P = 0.177 (I); F(3, 16) = 7.986, P = 0.002 (K).

Figure 4.

Neuronal hyperactivity in Pten conditional heterozygous mice is corrected in Pten and Mtor double conditional heterozygous mice. (A) Schema showing the procedure of social exposure. (B) Representative images of c-fos staining in the prelimbic cortex (PrL) and basolateral amygdala (BLA) of socially exposed mice. Scale bar: 100 μm. (C and D) Quantification of c-fos⁺ neurons in the PrL(C) and BLA (D) of socially exposed mice. One-way ANOVA and Tukey’s post hoc tests were used. *P < 0.05. n.s. indicates no significant difference. N = 7 animals per genotype. F(3, 24) = 8.25, P = 0.0006 (C); F(3, 24) = 7.462, P = 0.0011 (D).