Absence of strong strain effects in behavioral analyses of Shank3-deficient mice

Abstract

Haploinsufficiency of SHANK3, caused by chromosomal abnormalities or mutations that disrupt one copy of the gene, leads to a neurodevelopmental syndrome called Phelan-McDermid syndrome, symptoms of which can include absent or delayed speech, intellectual disability, neurological changes and autism spectrum disorders. The SHANK3 protein forms a key structural part of the post-synaptic density. We previously generated and characterized mice with a targeted disruption of Shank3 in which exons coding for the ankyrin-repeat domain were deleted and expression of full-length Shank3 was disrupted. We documented specific deficits in synaptic function and plasticity, along with reduced reciprocal social interactions, in Shank3 heterozygous mice. Changes in phenotype owing to a mutation at a single locus are quite frequently modulated by other loci, most dramatically when the entire genetic background is changed. In mice, each strain of laboratory mouse represents a distinct genetic background and alterations in phenotype owing to gene knockout or transgenesis are frequently different across strains, which can lead to the identification of important modifier loci. We have investigated the effect of genetic background on phenotypes of Shank3 heterozygous, knockout and wild-type mice, using C57BL/6, 129SVE and FVB/Ntac strain backgrounds. We focused on observable behaviors with the goal of carrying out subsequent analyses to identify modifier loci. Surprisingly, there were very modest strain effects over a large battery of analyses. These results indicate that behavioral phenotypes associated with Shank3 haploinsufficiency are largely strain-independent.

Keywords: 22q13; Autism spectrum disorders; Behavior; Genetic modifier; Mouse strain; Phelan-McDermid syndrome; Shank3.

Fig. 1.

General health and sensory-motor performances in Shank3-deficient mice. (A) Distribution of genotype. A deficit in the number of knockout (KO) mice was observed in all strains. (B) Weight of mice. Knockout mice were significantly smaller than wild type (WT) and heterozygotes (Het) (the main effect of genotype across the three strains, P=0.001). 129, 129SVE. (C) Tail-flick assay. A higher tail-flick latency in response to a hot stimulus was observed during the first trial for the C57 strain knockout mice. (D) Beam walking. A significant decrease of the latency to the start of crossing the beam (left panel) and a shorter crossing time (middle panel) was observed in the C57 strain knockout mice. The distance traveled on the beam (right panel) was significantly shorter for 129SVE heterozygous mice, mainly because of an increase in the number of falls and freezing behavior; knockout animals on the 129SVE background performed similar to controls. (E) Spontaneous locomotion. The distance traveled over a 1-hour session in the open-field was similar for all genotypes; however, during the first ten minutes, a significant increase in locomotion was observed in 129SVE heterozygous and knockout animals. Data represents the means±s.e.m. *P<0.05, **P<0.01, ***P<0.001. In C and E: * WT versus KO, ^o WT versus Het, ⁺ Het vs KO.

Fig. 2.

Anxiety-like behavior in Shank3-deficient mice. (A–D) Open-field. A decrease of thigmotaxis was observed in 129SVE (129) heterozygous (Het) and C57 knockout (KO) mice, as shown by a decrease of (A) the total time spent in the center, (B) the time spent resting in the center and (C) the number of times the mice entered into the central region. (D) An increase in the number of times the mice reared was observed for FVB knockout mice. (E–H) Elevated zero-maze. A reduction of anxiety characterized by an increase in (E) the total moving time, (F) the time spent in the open arcs and in the (G) open:closed arc ratio was observed in both the C57 and FVB Shank3 mutant animals. (H) The number of entries into an open arc was increased in 129SVE and C57 knockout mice. Data represents means±s.e.m. *P<0.05, **P<0.01, ***P<0.001. WT, wild type.

Fig. 3.

Social behavior in Shank3-deficient mice. (A,B) Three-chambered social approach test. Normal sociability, characterized by significantly more time spent with another mouse compared with an object, was found for mice from all genotypes and all strains in the sociability test (A). C57 and FVB mice spent significantly more time with the unfamiliar mouse in the social novelty and recognition test. The 129SVE (129) knockout (KO) mice showed a trend in their preference (B). (C) Nest building. No differences between the genotypes were found for any of the mouse strains in the nest building assay. The graph on the left shows the average score for the nests built for each genotype, the three graphs on the right shows more detailed analysis of this data. Data represents means±s.e.m. *P<0.05, **P<0.01, ***P<0.001. Het, heterozygous; WT, wild type.

Fig. 4.

Learning and memory in Shank3-deficient mice. (A) Fear conditioning. Training (left column) was altered in Shank3 heterozygous and knockout mice on the 129SVE (129) background, but no differences between the genotypes were found in either contextual (middle column) or cued recall (right column) phases. No differences between the genotypes were observed for C57 mice. All FVB mice were strongly impaired in all phases of the fear conditioning test. During cued recall, the percentage of mice that exhibited freezing behavior immediately after the first tone was lower in FVB Shank3-deficient animals than in wild-type. The x-axes indicate the time (minutes) for which the mice were recorded. (B) Y-maze spontaneous alternation. No differences were noted for the genotypes of 129SVE, C57 or FVB mice. The data represents means±s.e.m. ^oP<0.05, ^ooP<0.01, ^oooP<0.001 for wild-type (WT) versus heterozygous (Het) mice; *P<0.05, **P<0.01, ***P<0.001 for WT versus knockout (KO) mice; ^#P<0.05, ^##P<0.01, ^###P<0.001 for Het versus KO mice.

Fig. 5.

Stereotypies, repetitive behavior and perseverance in Shank3-deficient mice. (A,B) Repetitive behaviors. No significant differences were observed between the genotypes for (A) the number of revolutions in the open-field (the left panel shows the total number of revolutions, and the right panel shows the percentage of the total revolutions that were performed in a clockwise direction) or (B) perseveration (the number of times an animal choose the same arm of the maze for three consecutive times) in the Y-maze. (C,D) Grooming and barbering. No significant differences between the genotypes were observed for (C) the number of grooming bouts (left panel) or time (right panel) and (D) the percentage of animals with barbering. The data represent means±s.e.m. 129, 129SVE.