Truncating mutations in SHANK3 associated with global developmental delay interfere with nuclear β‐catenin signaling

Abstract

Mutations in SHANK3, coding for a large scaffold protein of excitatory synapses in the CNS, are associated with neurodevelopmental disorders including autism spectrum disorders and intellectual disability (ID). Several cases have been identified in which the mutation leads to truncation of the protein, eliminating C-terminal sequences required for post-synaptic targeting of the protein. We identify here a patient with a truncating mutation in SHANK3, affected by severe global developmental delay and intellectual disability. By analyzing the subcellular distribution of this truncated form of Shank3, we identified a nuclear localization signal (NLS) in the N-terminal part of the protein which is responsible for targeting Shank3 fragments to the nucleus. To determine the relevance of Shank3 for nuclear signaling, we analyze how it affects signaling by β-catenin, a component of the Wnt pathway. We show that full length as well as truncated variants of Shank3 interact with β-catenin via the PDZ domain of Shank3, and the armadillo repeats of β-catenin. As a result of this interaction, truncated forms of Shank3 and β-catenin strictly co-localize in small intra-nuclear bodies both in 293T cells and in rat hippocampal neurons. On a functional level, the sequestration of both proteins in these nuclear bodies is associated with a strongly repressed transcriptional activation by β-catenin owing to interaction with the truncated Shank3 fragment found in patients. Our data suggest that truncating mutations in SHANK3 may not only lead to a reduction in Shank3 protein available at postsynaptic sites but also negatively affect the Wnt signaling pathway.

Targeting of δ-catenin to postsynaptic sites through interaction with the Shank3 N-terminus

“…S1). This is in agreement with recent work where we showed that Shank3 directly binds to β-catenin via its PDZ domain [ 23 ].…”

“…In contrast, the postsynaptic level of β-catenin is not altered in the postsynaptic density of these Shank3 αβ deficient mice. This is in agreement with our previous work [ 23 ], as interaction of β-catenin with Shank3 is not dependent on the N-terminus of Shank3, but requires the PDZ domain (Fig. 5 c, d).…”

“…1 ). As observed previously [ 23 , 24 ], Shank3 forms large clusters in these cells, which were found partly in cytosolic structures but also to a larger extent in small nuclear clusters. Coexpression with α-catenin did not change this pattern, and we observed little colocalization of α-catenin with Shank3 as α-catenin was found mainly in diffuse cytosolic, as well as membrane attached structures.…”

“…Coexpression with α-catenin did not change this pattern, and we observed little colocalization of α-catenin with Shank3 as α-catenin was found mainly in diffuse cytosolic, as well as membrane attached structures. We have described before that β-catenin, in agreement with its known nuclear role, was found mostly in nuclear clusters where it was almost perfectly colocalized with Shank3 [ 23 ]. Finally, coexpression with δ-catenin led to a relocalization of Shank3 to a plasma membrane associated pattern, which was not seen in the absence of δ-catenin.…”

“…As different catenin proteins interact with each other directly, and also indirectly via their association with cadherins [ 22 ], it was initially unclear whether this was due to direct binding of Shank3 to any of these proteins. In addition, we have recently reported that the central PDZ domain of Shank3 binds to β-catenin [ 23 ]. It was therefore important to identify which catenin variant binds directly to the Shank3 N-terminal domains.…”