Linking genotype to phenotype in autism: mechanisms of cell-type specific presynaptic dysfunction in Chd8 haploinsufficiency

Autism is a neurodevelopmental disorder for which there is a severe lack of effective treatments for those who need them. A major problem in the development of new treatments is our poor understanding of its underlying biology. While there is clearly a strong genetic component, we now know that there are hundreds of different gene mutations that increase the risk of developing autism. Many of these risk genes are involved in synapse function or formation, suggesting that synapses (the site of communication between neurons in the brain) may be a key part of autism biology. However, an even larger group of autism risk genes simply regulate the expression of other genes. What happens downstream of these genes? Might they also affect synaptic function or formation? In this proposal, we aim to identify the molecules and pathways that link gene mutation in one of these 'regulator genes' to changes in synapse function, in a part of the brain that is well known to be affected in autism, the prefrontal cortex.

We will focus on a specific gene called Chd8, which is one of the highest confidence risk genes for autism, and where we have already identified a robust synaptic phenotype in prefrontal cortex. In particular, there is increasing evidence that the effects of missing a copy of Chd8 (and other autism risk genes) depends critically on which type of cell is affected. We will use transgenic technology to selectively delete a copy of Chd8 from excitatory neurons or different types of inhibitory neurons and find out exactly how synapse function and/or formation is affected. We will then collect individual cells of each type - excitatory and inhibitory - and use single nuclei sequencing to discover which genes show either increased or decreased expression when that cell type is missing a copy of Chd8. While this kind of approach usually results in many different genes that show changes, we will focus on those that we already know are involved in synaptic function / formation. We will carry out extensive validation of potential targets, including testing whether simply increasing or decreasing expression of the candidate in cortical neurons in a dish leads to the expected alteration to synaptic phenotype. Finally, we will take advantage of recent advances in CRISPR/Cas9 technology (the so-called genetic 'scissors' that allow researchers to edit DNA) to bring expression levels of candidate genes back to normal in the living brain, and see whether this can rescue the synaptic phenotypes.

This will allow us to work out the molecules and pathways that link the original genetic mutation to changes in cell and synapse function, thereby helping to fill a major gap in our understanding of this disorder. These molecules and pathways may provide novel targets for therapeutic intervention, potentially opening up new avenues in the effort to develop new treatments.

Genetic studies have identified hundreds of autism risk genes, but how mutations in these genes results in phenotypic changes remains very poorly understood. Our goal in this proposal is to identify the cell-type specific molecules and pathways that link haploinsufficiency of the autism risk gene, Chd8, to cellular phenotypes. We have previously identified a specific presynaptic phenotype from layer 5 (L5) excitatory synapses onto deep layer pyramidal neurons in the Chd8 haploinsufficient prefrontal cortex (PFC), and have evidence of presynaptic alterations at inhibitory synapses.

Our first aim is to define the precise cell-type specific synaptic alterations at excitatory and inhibitory synapses onto these neurons, using whole-cell patch clamp electrophysiology, cell-type specific Cre driver lines (to delineate interneuron subtypes) and a conditional Chd8 haploinsufficient line, structural imaging and optogenetics.

To identify the molecules and pathways downstream of Chd8 haploinsufficiency, we will use single nuclei RNA sequencing of PFC for excitatory and inhibitory neurons separately. Analysis of dysregulated excitatory and inhibitory transcriptomes will focus on known presynaptic genes, which will be validated using standard approaches as well as by testing whether upregulation / downregulation (as seen in Chd8 haploinsufficiency) alters presynaptic function in cortical neurons in vitro. Finally, we will test whether in vivo rescue of excitatory and inhibitory gene alterations (using CRISPRa/i AAVs) can rescue synaptic phenotypes in the Chd8 haploinsufficient PFC. This will shed light on the mechanisms downstream of mutation in transcriptional regulator genes such as Chd8 that lead to identifiable phenotypes, and provide potential new targets for therapeutics.