Unraveling the pathology of different disease severities in human cerebral organoid models of LIS1-lissencephaly

Abstract

The human neocortex is greatly expanded and exhibits a highly organized and extensively folded (gyrencephalic) structure. Model systems gave a fundamental understanding about how the cortex is generated although the applied models often involve species with a smooth (lissencephalic) brain surface, such as mice. Thus, key cellular events that impact human-specific brain expansion and our understanding of how disease-linked mutations disrupt human cortical development remains elusive. Lissencephaly is a malformation of cortical development which is characterized by a smooth brain and a disorganized cortex. Heterozygous deletions or mutations in the LIS1 gene, encoding a microtubule-associated protein in humans, were identified to cause lissencephaly with diverse clinical phenotypic variations ranging from mild pachygyria (broad gyri) to severe agyria (no gyri) resulting in epilepsy and intellectual disabilities. While the clinical severity generally correlates with the degree of agyria, the location and type of mutation in the LIS1 gene does not. From LIS1 mouse models we know that LIS1 regulates the microtubule motor cytoplasmic dynein and by that dynein-dependent processes such as neuronal migration, nucleokinesis, interkinetic nuclear migration and mitotic spindle orientation. Even though the observed LIS1-deficiency-associated phenotypes appeared drastically milder in murine systems compared to humans these studies suggest that LIS1 gene dosage is relevant for the phenotypic severities. However, why a specific mutation within the LIS1 gene as identified in LIS1-lissencephalic patients (LIS1-patients) leads to different disease severities and whether human-specific processes during cortical development are differentially affected by the specific mutations could, due to a lack of adequate model systems, so far not been investigated. Here, I explore the ability to recapitulate different disease severities of LIS1-lissencephaly using LIS1-patient-specific iPS cells and thereof derived forebrain-type cerebral organoids. To do so, I selected from a LIS1-patient cohort comprising 63 cases 7 patients who cover the whole spectrum of gyrification alterations of LIS1-lissencephaly ranging from Dobyns grade 5 (mild) to 1 (severe). Each patient harbors a different molecular characterized heterozygous mutation in the LIS1 gene. To analyze the consequences of each LIS1 mutation on human brain development a 3D cell culture forebrain-organoid protocol was developed. Following reprogramming of patient-derived somatic cells and basic characterization (2 clones each) the iPS cells were applied to the organoid protocol. Organoids reproduced, in correlation with the patient’s severity, alterations in organoid cytoarchitecture and premature neurogenesis. To assess the direct consequences of the patient-specific mutations on LIS1 microtubule stabilizing function I investigated the stability of the cytoskeleton of apical (a) RG cells within the cortical ventricular-like zone (VZ) structures and found a progressive collapse of tubulin strand stability with increasing patient disease severity leading to a disruption of cellular organization. These phenotypic alterations could in part be reversed by stabilizing the microtubule array using the FDA-approved drug EpothiloneD. In addition, organoids from individuals with severe but not mild disease showed a non-random aRG cell division switch from proliferative to neurogenic division. As an underlying molecular cause, WNT-signaling alterations were identified, most prominently in severe conditions. To test to what extend perturbed WNT-signaling contributes to the observed patient-specific alterations, organoids were exposed to the GSK3s inhibitor CHIR99021 leading to a significant rescue of non-random aRG cell division switch in severe organoids and to enlarged VZ diameters as well as reduced neurogenesis in all patient derived organoids. The here demonstrated research underlines the capability of cerebral organoids to sensitively model individual disease severities, a so far not addressed major challenge of the system. My data show that different patient-specific mutations in the LIS1 gene have divergent direct impact on microtubule stability, which directly and/or indirectly lead to perturbed human corticogenesis providing the missing link between the patient-specific LIS1 mutation and the clinical severity grade. Future applications analyzing individual diseases have the potential to advance personalize medicine and improve the understanding of individual pathology for personalized therapy.