Autism-Associated Insertion Mutation (InsG) of Shank3 Exon 21 Causes Impaired Synaptic Transmission and Behavioral Deficits

Abstract

SHANK3 (also known as PROSAP2) is a postsynaptic scaffolding protein at excitatory synapses in which mutations and deletions have been implicated in patients with idiopathic autism, Phelan-McDermid (aka 22q13 microdeletion) syndrome, and other neuropsychiatric disorders. In this study, we have created a novel mouse model of human autism caused by the insertion of a single guanine nucleotide into exon 21 (Shank3(G)). The resulting frameshift causes a premature STOP codon and loss of major higher molecular weight Shank3 isoforms at the synapse. Shank3(G/G) mice exhibit deficits in hippocampus-dependent spatial learning, impaired motor coordination, altered response to novelty, and sensory processing deficits. At the cellular level, Shank3(G/G) mice also exhibit impaired hippocampal excitatory transmission and plasticity as well as changes in baseline NMDA receptor-mediated synaptic responses. This work identifies clear alterations in synaptic function and behavior in a novel, genetically accurate mouse model of autism mimicking an autism-associated insertion mutation. Furthermore, these findings lay the foundation for future studies aimed to validate and study region-selective and temporally selective genetic reversal studies in the Shank3(G/G) mouse that was engineered with such future experiments in mind.

Keywords: Phelan–McDermid syndrome; Shank3; autism; behavior; postsynaptic density; synaptic plasticity.

Figure 1.

Schematic diagram of the exon 21 insertion mutation (insert G) mouse model. A gene repair targeting vector (Shank3-GRV) was created via insertion of an insertion mutant exon 21 followed by a NeoStop cassette flanked by loxP sites (“floxed” E21^G-NeoStop cassette) upstream of the wild-type exon 21. The resulting Shank3-GRV was targeted in mouse ES cells into the wild-type Shank3 gene (Shank3-wt). This resulted in the creation of mice constitutively expressing a human Shank3 exon 21 insertion mutation associated with autism, the Shank3^G mouse model. In future experiments, the Shank3^G mice can be genetically reversed to wild-type Shank3 whenever and wherever cre-recombinase is activated or expressed. Nde1 denotes restriction sites with size of diagnostic restriction fragments indicated between the Nde1 sites below.

Figure 2.

Expression of Shank3 and associated proteins at hippocampal synapses. A, Western blots with both C-terminal and N-terminal Shank3 antibodies reveal that all three major Shank3 isoforms are absent in hippocampal synaptosomes from Shank3^G/G mice. B, The Shank3 N-terminal antibody reveals the appearance of one novel Shank3 isoform and a trend toward enhanced expression of another isoform of low molecular weight in Shank3^G/G mice compared with WT. Expression of synaptic and scaffolding proteins known to associate with Shank3 is unchanged in hippocampal synaptosomes (C) and in whole hippocampal lysates (D) from Shank3^G/G mice. n = 8 mice per genotype. E, Ionotropic glutamate receptor expression remains unchanged in the hippocampal PSD preparations in the Shank3^WT/G and Shank3^G/G mice. n = 6 sets of hippocampi pooled from 4 mice per genotype. F, A decrease in phosphorylation levels of GluN2B at tyrosine-1472 was observed in the hippocampal synaptosomes of Shank3^G/G mice. n = 6 sets of hippocampi pooled from 4 mice per genotype. *p < 0.05. ***p < 0.001.

Figure 3.

Shank3^G/G mice exhibit impairments in behavioral tasks. A, Latency to fall from or to go one full revolution on the rotarod task. Shank3^G/G mice exhibit motor coordination impairments in 8 trials of rotarod test conducted over 2 h. Legend in A applies to B, E. B, C, Width and height of nest built as a function of time in a nest-building task. Shank3^G/G mice exhibit impairments in nest building behavior over a 90 min period. D, Number of marbles buried during a 30 min marble-burying task. Shank3^G/G mice show impaired marble burying behavior. Legend in D applies to F–M. E, Locomotor activity as measured by number of photobeam breaks during successive 5 min intervals over a 2 h period. Shank3^G/G mice exhibit normal locomotor habituation over the full 2 h period. F, Number of photobeam breaks during the initial 5 min of the locomotor task shown in E. Shank3^G/G mice show decreased activity, initially suggesting abnormal locomotor response to novelty. G, Total distance traveled during the 5 min elevated plus maze task. Shank3^G/G mice have decreased locomotor activity in the elevated plus maze task. H, Total distance traveled during the 10 min open field task. Shank3^G/G mice have decreased locomotor activity in the open field. I, Number of photobeam breaks during the 10 min dark/light task. Shank3^G/G mice have decreased locomotor activity in dark/light. J, Ratio of time spent in the center to time spent in the periphery in an open field task. No differences were observed among genotypes. K, Ratio of time in the open arms versus time in other arms in the elevated plus maze task. Shank3^G/G mice spend the same time in open versus closed arms compared with littermate controls. L, Latency to enter the light side of the dark/light apparatus. Shank3^G/G mice exhibit an increased latency to enter the light chamber. M, Time spent in dark and light sides of the dark/light apparatus. No difference was observed in the total time spent in the dark versus the light side of the chamber: WT, n = 23; Shank3^WT/G, n = 19; Shank3^G/G, n = 19 for all panels. *p < 0.05. **p < 0.01. ***p < 0.001.

Figure 4.

Shank3^G/G mice exhibit mild spatial learning impairment, normal social interaction, and normal grooming behavior. A–D, Training days for the Morris water maze task. For each day of training, data were averaged across four daily trials. A, Latency to reach hidden platform on successive water maze days. Shank3^G/G mice take longer to reach the submerged platform. Legend in A applies to B–D. B, Swim speed on successive water maze training days. Shank3^G/G mice swim faster in the water maze. C, Distance traveled before reaching the hidden platform on successive water maze training days. Shank3^G/G mice travel a longer distance before reaching the submerged platform. D, Percentage time spent in thigmotaxis on successive water maze training days. Shank3^G/G mice spend the same amount of time in the thigmotaxis region as their littermate controls. E, Time spent in target quadrant and other quadrants during probe trial in which target platform is removed. Shank3^G/G mice spend more time in the target quadrant versus other quadrants. Legend in E applies to F–J. F, Number of target location crossings and corresponding phantom platform location crossings in other quadrants during the probe trial. Shank3^G/G mice show a clear preference for the target platform location: WT, n = 22; Shank3^WT/G, n = 19; Shank3^G/G, n = 18. G, In the three chambered social interaction test, all groups show a normal preference for the social versus the inanimate target and (H) for the stranger mouse versus the familiar mouse: WT, n = 19; Shank3^WT/G, n = 15; Shank3^G/G, n = 15. I, Shank3^G/G mice show no change in total time spent grooming or (J) number of grooming bouts in a 10 min period: WT, n = 23; Shank3^WT/G, n = 19; Shank3^G/G, n = 19. *p < 0.05. **p < 0.01. ***p < 0.001.

Figure 5.

Hippocampal synaptic plasticity is impaired in Shank3^G/G mice. A, LTP induced by a single 1 s, 100 Hz train (arrow) is normal at 55–60 min after tetanus, summarized in B, although PTP in the first 5 min following conditioning stimulus is decreased in Shank3^WT/G and Shank3^G/G mice (WT, n = 13 slices/8 mice; Shank3^WT/G, n = 9 slices/5 mice; Shank3^G/G, n = 9 slices/5 mice). Inset, Average of 10 consecutive raw traces immediately preceding (black) and 60 min (gray) after conditioning stimulus for WT (left), Shank3^WT/G (middle), and Shank3^G/G (right). Calibration: 0.25 mV, 10 ms. C, LTP induced by four, 1 s, 100 Hz trains (arrow) is not affected by the Shank3^G mutation, as summarized at 55–60 min after induction (D) (WT, n = 7 slices/5 mice; Shank3^WT/G, n = 10 slices/5 mice; Shank3^G/G, n = 8 slices/4 mice). Inset, Average of 10 consecutive raw traces immediately preceding (black) and 60 min (gray) after conditioning stimulus for WT (left), Shank3^WT/G (middle), and Shank3^G/G (right). Calibration: 0.25 mV, 10 ms. E, mGluR-LTD induced by 10 min application of 100 μm DHPG (bar) is impaired at 55–60 min following DHPG washout (F) in Shank3^G/G mice (WT, n = 7 slices/5 mice; Shank3^WT/G, n = 13 slices/5 mice; Shank3^G/G, n = 8 slices/6 mice). Inset, Average of 10 consecutive raw traces immediately preceding (black) and 60 min (gray) after DHPG application for WT (left), Shank3^WT/G (middle), and Shank3^G/G (right). Calibration: 0.5 mV, 10 ms. For clarity, scatter plots are represented as the mean ± SEM of 5 consecutive data points. *p < 0.05. **p < 0.01.

Figure 6.

Hippocampal synaptic transmission is impaired in Shank3^WT/G and Shank3^G/G mice. A, I/O curves of stimulus intensity versus fEPSP slope indicate a decrease in basal synaptic strength at CA3-CA1 synapses. Inset, fEPSP slope (mV/ms) measured in relation to fiber volley amplitude (mV) (WT, n = 16 slices/8 mice; Shank3^WT/G, n = 12 slices/6 mice; Shank3^G/G, n = 12 slices/7 mice). B, PPR is not affected by the Shank3^G mutation at interstimulus intervals 30–500 ms (WT, n = 12 slices/7 mice; Shank3^WT/G, n = 8 slices/5 mice; Shank3^G/G, n = 15 slices/8 mice). C, NMDA/AMPA ratio is decreased in Shank3^G/G mice (WT, n = 22 cells/10 mice; Shank3^WT/G, n = 18 cells/6 mice; Shank3^G/G, n = 23 cells/8 mice). Inset, Average of 10 consecutive traces at −70 mV and at 40 mV holding potential from Shank3^WT (left), Shank3^WT/G (middle), and Shank3^G/G (right). Calibration: 200 pA, 25 ms. D, One minute traces of mEPSCs from WT (top), Shank3^WT/G (middle), and Shank3^G/G (bottom). Calibration: 20 pA, 1 s. Mean mEPSC frequency (E) is decreased in CA1 neurons from Shank3^G/G mice and is reflected in the rightward shift in the distribution of interevent frequency (F) in Shank3^G/G mice indicating longer interevent intervals in Shank3^G/G mice. Mean mEPSC amplitude (G) and distribution of mEPSC amplitudes (H) are not affected by the Shank3^G mutation (WT, n = 15 cells/5 mice; Shank3^WT/G, n = 21 cells/8 mice; Shank3^G/G, n = 14 cells/6 mice). *p < 0.05. **p < 0.01.

Figure 7.

Treatment with tamoxifen diet results in biochemical reversal of the Shank3 protein in Reversible-Shank3^GCre+ mice. Quantification and representative Western blot of whole-brain lysates showing reversal of Shank3 protein expression with C terminus antibody (JH3025) 2 weeks after administration of tamoxifen diet for 4 weeks (initiated at 8 weeks of age). Data are normalized to the β-actin control and then to the average of WT levels (N = 7). *p < 0.05.