Restoring Shank3 in the rostral brainstem of shank3ab\u2212/\u2212 zebrafish autism models rescues sensory deficits

Robert A Kozol,Julia E Dallman,Stephan Züchner,David M James,Sureni H Sumathipala,Ivan Varela

doi:10.1038/s42003-021-02920-6

Abstract

People with Phelan-McDermid Syndrome, caused by mutations in the SHANK3 gene, commonly exhibit reduced responses to sensory stimuli; yet the changes in brain-wide activity that link these symptoms to mutations in the shank3 gene remain unknown. Here we quantify movement in response to sudden darkness in larvae of two shank3 zebrafish mutant models and show that both models exhibit dampened responses to this stimulus. Using brain-wide activity mapping, we find that shank3−/− light-sensing brain regions show normal levels of activity while sensorimotor integration and motor regions are less active. Specifically restoring Shank3 function in a sensorimotor nucleus of the rostral brainstem enables the shank3−/− model to respond like wild-type. In sum, we find that reduced sensory responsiveness in shank3−/− models is associated with reduced activity in sensory processing brain regions and can be rescued by restoring Shank3 function in the rostral brainstem. These studies highlight the importance of Shank3 function in the rostral brainstem for integrating sensory inputs to generate behavioral adaptations to changing sensory stimuli.

Highlights

People with Phelan-McDermid Syndrome, caused by mutations in the SHANK3 gene, commonly exhibit reduced responses to sensory stimuli; yet the changes in brain-wide activity that link these symptoms to mutations in the shank[3] gene remain unknown
We generated two zebrafish Phelan McDermid Syndrome (PMS) models, shank3abΔN with mutations truncating both the Shank[3] a and b proteins in the ankyrin repeat domains and shank3abΔC with mutations truncating both the Shank[3] a and b proteins near the proline-rich domain[14] (Fig. 1a; Supplementary Fig. 1, Supplementary Data 1). These models mimic the most common types of SHANK3 mutations found in people with PMS2 and, by having two models, we control for genetic background
Consistent with the Mitogen-Activated Protein (MAP)-mapping experiments, these results indicate Shank3ab function in rostral brainstem is sufficient for wild type (WT) levels of light-evoked activity

Summary

Results and discussion

In contrast to the single SHANK3 gene in people, the shank[3] gene is duplicated in zebrafish; to generate zebrafish models of PMS we used CRISPR/Cas[9] to mutate both the shank3a and shank3b (shank3ab) gene paralogs. Shank3:Zb-T transplants rescued anatomical scaling of both the locus coeruleus and medulla oblongata (Fig. 4a, b, Supplementary Data 1, Sheets 31–38) These findings suggest that both neurodevelopmental and synaptic deficits could contribute to altered sensory processing in shank[3] mutant zebrafish. We used whole-brain activity mapping and chimeric rescue experiments to identify the rostral brainstem as a region that requires Shank[3] function to generate behavioral responses to visual stimuli To do this we generated two independent zebrafish shank3abΔN and shank3abΔC mutant models of Phelan McDermid Syndrome, both of which, like humans with Phelan McDermid Syndrome, showed dampened responses to visual stimuli. Colors defined by 2-2-4 LUT color scheme in Fiji

Conclusion

Methods

Code availability