Authors :
Wei Niu, PhD – University of Michigan; Lu Deng, MD – University of Michigan; Sandra Mojica-Perez, BS – University of Michigan; Andrew Tidball, PhD – University of Michigan; Roksolana Sudyk, BS – University of Michigan; Miranda Walker, PhD – University of Michigan; Jack Parent, MD – University of Michigan
This abstract has been invited to present during the Basic Science Poster Highlights poster session.Rationale: Protocadherin-19 (PCDH19) clustering epilepsy (PCE) is a severe X-linked developmental and epileptic encephalopathy. PCE exclusively affects females and mosaic males, while hemizygous males are spared. Mosaic PCDH19 expression due to random X-inactivation is thought to cause "cellular interference" with altered cell-cell interactions between mutant and wildtype (WT) PCDH19-expressing cell populations to produce the disease. Some progress has been made in understanding “cellular interference” mechanisms, but how PCDH19 variants result in epilepsy is largely unknown. The goal of this study is to develop PCE human cortical organoid (hCO) models to investigate the function of PCDH19 during cortical development, to understand how its mutations alter neurodevelopment and lead to seizure-like activity, and eventually to explore potential therapeutic strategies specific to PCE.
Methods: We first generated H9 female human embryonic stem cells (hESCs) with a HA-FLAG tagged PCDH19 (WT) and homozygous PCDH19 knockout (KO) H9 hESCs using CRISPR/Cas9 genome editing. We then created a virtual "PCE patient" model in which isogenic GFP-labeled H9 hESCs with a HA-FLAG tagged PCDH19 allele are mixed with RFP-labeled KO cells. We generated hCOs after 1:1 mixing of KO and WT cells (PCE model), WT and WT cells (healthy control), or KO and KO cells (asymptomatic male carrier). In early-stage PCE organoids, we examine cell segregation and cortical lamination using live cell imaging and immunolabeling. In late-stage brain organoids, we examine synaptogenesis by quantifying numbers of excitatory synapses and examine network activity using multielectrode array (MEA) recordings.
Results: HA-FLAG tagged PCDH19 hESCs allowed us to use a HA antibody to detect PCDH19 expression. We found that PCDH19 and N-Cadherin co-localize to the apical junctions of neural rosettes, and PCDH19 labels the cell bodies of dividing radial glial cells (Phospho-Vimentin positive cells) in ventricular zone (
VZ)-like regions in early hCOs (~day 20). In early PCE hCOs, we found abnormal cell sorting in the VZ-like region.
We also found an altered expression pattern of N-cadherin and abnormal early cortical lamination demonstrated by disorganized CTIP2 expression in progenitor regions. No clear abnormalities were observed in organoids generated by mixing only WT or only KO labeled cells. These data suggest that PCDH19 is a critical cell adhesion molecule during early human brain development and altered cell-cell interactions between KO and WT PCDH19-expressing cell populations contribute to the disease phenotype.
Conclusions: PCE hCOs reliably model mosaic PCDH19 expression in vitro and produce robust PCE-related phenotypes during cortical development. Future assays such as examining synapse function, and MEA recordings of network activity in late-stage hCOs and/or fused cortical-ventral organoids will be exploited to determine epilepsy-like phenotypes.
Funding: PCDH19 Alliance, Insieme PCDH19, NIH (NINDS) U54NS117170, Simons Foundation