Novel Shank3 mutant exhibits behaviors with face validity for autism and altered striatal and hippocampal function

Abstract

Mutations/deletions in the SHANK3 gene are associated with autism spectrum disorders and intellectual disability. Here, we present electrophysiological and behavioral consequences in novel heterozygous and homozygous mice with a transcriptional stop cassette inserted upstream of the PDZ domain-coding exons in Shank3 (Shank3^E13 ). Insertion of a transcriptional stop cassette prior to exon 13 leads to loss of the two higher molecular weight isoforms of Shank3. Behaviorally, both Shank3^E13 heterozygous (HET) and homozygous knockout (KO) mice display increased repetitive grooming, deficits in social interaction tasks, and decreased rearing. Shank3^E13 KO mice also display deficits in spatial memory in the Morris water maze task. Baseline hippocampal synaptic transmission and short-term plasticity are preserved in Shank3^E13 HET and KO mice, while both HET and KO mice exhibit impaired hippocampal long-term plasticity. Additionally, Shank3^E13 HET and KO mice display impaired striatal glutamatergic synaptic transmission. These results demonstrate for the first time in this novel Shank3 mutant that both homozygous and heterozygous mutation of Shank3 lead to behavioral abnormalities with face validity for autism along with widespread synaptic dysfunction. Autism Res 2017, 10: 42-65. © 2016 International Society for Autism Research, Wiley Periodicals, Inc.

Keywords: Phelan-McDermid syndrome; Shank3; autism spectrum disorder; grooming; mouse model; social interaction.

Figure 1

Generation of Shank3^E13 mice. (A) The schema showing the location of the Neo-Stop cassette within intron 12 (Bgl II) near WT exon 13 of Shank3, wild-type Shank3 allele (Shank3-wt), product of homologous recombination (Shank3^E13*) and hypothetical genetic reversal (Shank3^E13*rev) by cre recombinase. The position of a Shank3 5′ flanking probe along with the size of diagnostic restriction fragments identifying the wild type, Shank3^E13* and Shank3^E13*rev are indicated. The restriction sites are: Sca1 and Mfe1. The black arrowheads represent loxP sites. (B) Genotyping analysis of WT (W), HET (H), and KO (K). WT mice produced a single 332 bp band while the KO produced a 576 bp band. HET produced bands of both sizes. A sample with no DNA was loaded as a control (C). “L” designates the 100bp ladder. (C) RT-PCR of cDNA from brain tissue of 4-week old (WT): wildtype, (e4–9):Shank3^e4–9 (Jaramillo et al., 2015a), (e13):Shank3^e13, (G/G):Shank3^G/G (Speed et al., 2015), ΔC/ΔC: Shank3^ΔC/ΔC (Kouser et al., 2013) mice. The Shank3 gene contains five functional domains; Ankyrin binding domain from exons 4 to 9, SH3 domain from exons 11 to 12, PDZ domain from exons 14 to 16, Proline rich domain from exon 21, and SAM domain from exon 22. Shank3^e13 mice show a loss of bands from exons 13 to 19 which indicates a lack of PDZ domain-encoding exons. Shank3^e4–9 shows a loss of bands from exons 4 to 9. Shank3^G/G and Shank3ΔC/ΔC show a loss of band containing exons 21–22. (D) Whole brain cell lysates from WT, HET, and KO Shank3^E13 mice probed with C-terminal Shank3 antibody.

Figure 2

Social test in Shank3^E13 mutant mice. (A) In the 3-chamber test of sociability, none of the three mouse groups displayed an aberrant chamber preference in the initial trial. In second trial of this task, Shank3^E13 HET mice did not display a preference for the social target over the inanimate object (B). In the third trial of this task, neither Shank3^E13 HET nor KO mice displayed a preference for social novelty (C). Shank3^E13 KO mice also displayed a significant decrease in approach of a novel, caged social target in the caged conspecific test (D). In the genotype and sex-matched social interaction test, Shank3^E13 KO mice displayed a significant reduction in social interaction time (E), and a decrease in number of social bouts (F). Both HET and KO mice display deficits in social recognition in the social interaction with juvenile test of social memory (G). (*P<0.05; **P<0.01, ***P< 0.001, ****P< 0.0001; n = 20 WT, 20 HET, 15 KO).

Figure 3

Grooming, anxiety and motor behavior in Shank3^E13 mutant mice. (A) Shank3^E13 HET and KO mice displayed a significant increase in time spent grooming compared to WT littermate mice. In the open field test all genotypes spent similar time in the center (B). Additionally they all traveled similar distances (C). In the dark/light test all three genotypes had similar latencies to enter into the light chamber (D). All three genotypes spent a similar amount of time in light and dark chambers (E). In the elevated plus maze all three genotypes spent a similar percentage of time in the closed arms, however, there was a significant decrease in time spent in the open arm in Shank3^E13 HET mice when compared to WT mice (F). Shank3^E13 HET and KO mice also displayed less rearing over a 1hr time period (G). Female Shank3^E13 HET and KO mice displayed less rearing (H) while male mutant mice only displayed a trend toward less rearing (I). Total vertical rearing beam breaks was decreased in KO mice when analyzing the cohort. Further analysis revealed a decrease in female KO mice (J). In the marble-burying test Shank3^E13 KO mice buried significantly less marbles than the WT and HET mice (K). In the locomotor test all three genotypes displayed similar levels of habituation over a 2-hr time period (L). There was no difference in the total number of beam breaks during the 2-hr locomotor test (M). In the rotarod test KO mice displayed decreased latency to fall compared to WT and HET mice (N). Female KO mice showed a significant decrease in latency to fall compared to WT (O). Male KO mice showed a significant decrease in latency to fall compared to HET and KO mice (P). All mouse groups displayed a similar startle response to auditory stimuli ranging from 0 to 120 db (Q). All three groups displayed similar % inhibition in the prepulse inhibition test (R). (*P<0.05, **P<0.01, ***P<0.001, n=20 WT, 20 HET, 15 KO).

Figure 4

Spatial learning behavior in Shank3^E13 mutant mice. Shank3^E13 KO mice displayed increased latency to reach the hidden platform (A, asterisks in graph near symbols refer to WT vs. KO). KO mice also displayed an increase in distance traveled before reaching the hidden platform (B, asterisks in graph near symbols refer to WT vs. KO) and % of time in thigmotaxis (C, asterisks in graph near symbols refer to WT vs. KO) during the training phase of the Morris water maze test. In the probe trial KO mice did not show a preference for the target quadrant compared to the other quadrants (D, asterisks in graph near bars denote comparison to Target quadrant within each group). (*P <0.05; **P<0.01, ***P <0.001, ****P<0.0001; n = 20 WT, 20 HET, 15 KO).

Figure 5

Hippocampal synaptic function in Shank3^E13 mice. (A) Input/output (I/O) curves show no effect of genotype on fEPSP slope at stimulus intensities 0–350 µA. Inset: fEPSP slope (mV/ms) plotted against fiber volley amplitude (mV). WT: 9 slices/6 mice, HET: 12 slices/7 mice, KO: 12 slices/7 mice. (B) Paired-pulse ratio (PPR, slope 2/slope 1) is not affected by the Shank3^E13 mutation at interstimulus intervals 30–500 ms. WT: 8 slices/4 mice, HET: 14 slices/8 mice, KO: 17 slices/9 mice. (C) Long-term potentiation (LTP) induced by a single 1 sec, 100Hz train (arrow) is decreased in Shank3^E13 HET and KO mice. Each data point is shown as the average of five consecutive traces for clarity. (D) Mean normalized fEPSP at 55–60 min following conditioning stimulus showed decreased LTP in Shank3^E13 HET and KO mice. WT: 11 slices/7 mice, Het: 9 slices/8 mice, KO: 12 slices/9 mice. (E) mGluR5-dependent long-term depression induced by 10-min bath application of Group I mGluR agonist DHPG (100 µM, solid bar) is not affected in Shank3^E13 mutant mice. (F) Mean normalized fEPSP at 55–60 min following the start of DHPG washout showing normal mGluR-LTD in Shank3^E13 mice. WT: 7 slices/5 mice, HET: 13 slices/8 mice, KO: 9 slices/6 mice. *P<0.05.

Figure 6

Striatal synaptic transmission is altered in Shank3^E13 mutant mice. (A) Mean NMDA/AMPA ratio is decreased in Shank3^E13 HET and KO mice. (B) Ten consecutive raw traces (gray) with overlaid average traces (black) at −70 mV (bottom) and +40 mV (top) from WT (left), HET (middle), and KO (right) mice. WT: 28 cells/9 mice, HET: 18 cells/4 mice, KO: 25 cells/5 mice (*P < 0.05, **P< 0.01, ***P < 0.001).

Figure 7

Biochemical analysis of striatal and hippocampal synaptic proteins in Shank3^E13 mutant mice. (A) Quantification of whole cell lysates derived from striatal tissue shows significant decreases in GluN1, Homer 1 b/c, and PSD-95. Sample blots are shown on the left with grouped data on the right. (B) Quantification of synaptosome preparations from striatal tissue shows significant decrease in GluA2, GluA3, GluN2A, GluN2B, Homer 1b/c, PSD-95, and Shank3 HMW bands (*P<0.05, **P<0.01, ***P<0.001; for whole cell lysates n=8 WT, 3 HET, 8 KO; for Synaptosome lysates n=16 WT, 11 HET, 16 KO for all samples except blots for Shank3 protein, n=8 WT, 8 HET, 8 KO). (C) Quantification of synaptosome preparations from hippocampal tissue shows significant decrease in GluA1 and GluN1 in heterozygous mice. (*P<0.05 for synaptosome lysates n = 8WT, 8HET, 8KO).