Phase separation of autism-linked POSH–SHANK2/3 in spine development

Researchers led by Prof. XU Zhiheng from the Institute of Genetics and Developmental Biology of the Chinese Academy of Sciences, revealed the important roles of POSH phase separation and its co-condensates with SHANK2/3 in the synaptic targeting of SHANK2/3 and synaptogenesis.

This work was published in Protein & Cell.

Autism spectrum disorder (ASD) is a highly heritable neurodevelopmental disorder. SHANK2/3 copy-number variants and truncating mutations account for ~1 % of ASD cases. POSH (plenty of SH3s, also known as SH3RF1) is also a high-risk gene for ASD. Previous studies have shown that Posh gene deletion significantly reduces SHANK2/3 protein levels at postsynaptic density (PSD), impairs dendritic spine maturation, and induces autism-like behaviour with learning and memory deficits in conditional knockout (cKO) mice (Yao et al., 2022). However, the organisational mechanisms of POSH and SHANK2/3 at PSD and the mechanisms regulating synaptic development remain unclear.

In this study, the researchers demonstrated that POSH undergoes liquid-liquid phase separation (LLPS) and co-condenses with SHANK2/3. Functional rescue experiments in Posh-deficient hippocampal neurons revealed that the co-condensates of POSH and SHANK2/3 are essential for postsynaptic targeting of SHANK2/3 and spine development.

Key findings from the study include:

POSH undergoes LLPS on its own and recruits SHANK2/3, whose subsequent enrichment further reinforces POSH condensation.

LLPS of POSH is driven by proline-rich motifs (PRMs) within its intrinsically disordered region and tandem SH3 domains. SHANK2/3 carries cognate PRMs that allow selective integration into POSH condensates, whereas the PRM-deficient partner PSD-95 is excluded.

Re-introducing wild-type POSH—but not phase-separation mutants—into Posh-deficient neurons fully restored SHANK2/3 synaptic clustering, spine density and morphology.

Because adult-onset re-expression of SHANK proteins can rescue synaptic and behavioral deficits in Shank-mutant mice, researchers believe that POSH-mediated LLPS not only represents a new mechanism for SHANK2/3 synaptic targeting and synaptic development, but also provides new insights into the pathological mechanisms of ASD and may become a potential therapeutic strategy targeting co- condensation.

This research was supported by the National Natural Science Foundation of China, and the Ministry of Science and Technology of China, etc.

Yao et al. demonstrated that POSH undergoes liquid-liquid phase separation and co-condenses with SHANK2/3, revealing a novel mechanism for organizing SHANK2/3 at synapses and driving spine development.