Search bioRxiv⌕ Search

Biology subjects

Lepagnol-Bestel, A.-M.

Publications and source records attributed to Lepagnol-Bestel, A.-M..

5 recordsLinked to original sources

AUTS2 expression within mammalian lineage: a predictor of neural networks involved in Autism Spectrum Disorders

Autism susceptibility gene 2 (AUTS2) is a neurodevelopmental regulator associated with an autosomal dominant neurological syndrome with ASD-like (Autism Spectrum Disorder-like) features. The AUTS2 syndrome phenotype includes borderline to severe intellectual disability (ID), microcephaly and mild dysmorphic traits. Specific ASD-like features, including obsessive or ritualistic behaviours, are frequently displayed, although sociability is largely unaffected. Syndrome severity is worse when mutations are present in the 3 region (exons 9-19) of the AUTS2 gene. AUTS2 is also associated with alcohol consumption, heroin dependence, schizophrenia and dyslexia, using GWAS studies. Our working hypothesis is that AUTS2 expression during mammalian evolution can be instrumental to predict phenotypes linked to the disease. For that, we studied brain In Situ Hybridization (ISH) data in mouse, marmoset, human embryo and fetuses and sequences data of Neanderthal, Denisovan and modern humans. Developmental mouse brain expression is found in neocortex, hippocampus, thalamus, tegmentum and cerebellum suggesting a link linked with cognition, stereotypies and perseverative behaviors. Expression in amygdala and claustrum found in marmoset extend these phenotypes to anxiety and higher-order cognitive processes. In human embryos (8 weeks) and fetuses (15- and 19-weeks), we found AUTS2 expression in similar regions but also in medial, lateral and caudal ganglionic eminence, that are involved in the diversity of interneurons found in neocortex, hippocampus and striatum. Sequence comparison of AUTS2 locus in Neanderthal, Denisova and modern human reveals novel sites in regulatory regions of AUTS2. Altogether, these results suggest that ISH distribution along mammals is a novel phenotype-related biomarker useful for translational research.

neuroscience↗

Chr21 protein-protein interactions: enrichment in products involved in intellectual disabilities, autism and Late Onset Alzheimer Disease

Intellectual disability (ID) found in Down syndrome (DS), which is characterized by an extra copy of 234 genes on Chr21 is poorly understood. We first used two DS mouse models that either display an extra copy of the Dyrk1A gene or of the mouse Chr16 syntenic region. Exome sequencing of transcripts deregulated in embryonic hippocampus uncovers enrichment in genes involved in chromatin and synapse respectively. Using large-scale yeast two-hybrid screen (154 distinct screens) of human brain library containing at least 107 independent fragments, we identified 3,636 novel protein-protein interactions with an enrichment of direct interactors of both Chromosome 21(Hsa21) baits and rebounds in ID-related genes. Using proximity ligation assays, we identified that Hsa21-encoded proteins are located at the dendritic spine postsynaptic density in a protein network located at the dendritic spine post synapse. We located Hsa21 DYRK1A and DSCAM that confers a [~] 20-fold increase in Autism Spectrum Disorders (ASDs), in this postsynaptic network. We found that a DSCAM intracellular domain binds either DYRK1A or DLGs that are multimeric scaffolds for the clustering of receptors, ion channels, and associated signaling proteins. The DYRK1A-DSCAM interaction is conserved from drosophila to humans. The identified postsynaptic.network is enriched in ARC-related synaptic plasticity, ASDs and Late-Onset Alzheimer Disease. Altogether, these results emphasize links between DS and brain diseases with complex genetics.

neuroscience↗

Multi-hit autism genomic architecture evidenced from consanguineous families with involvement of FEZF2 and mutations in high-risk genes

Autism Spectrum Disorders (ASDs) are a heterogeneous collection of neurodevelopmental disorders with a strong genetic basis. Recent studies identified that a single hit of either a de novo or transmitted gene-disrupting, or likely gene-disrupting, mutation in a subset of 65 strongly associated genes can be sufficient to generate an ASD phenotype. We took advantage of consanguineous families with an ASD proband to evaluate this model. By a genome-wide homozygosity mapping of ten families with eleven children displaying ASD, we identified a linkage region of 133 kb in five families at the 3p14.2 locus that includes FEZF2 with a LOD score of 5.8 suggesting a founder effect. Sequencing FEZF2 revealed a common deletion of four codons. However, the damaging FEZF2 mutation did not appear to be sufficient to induce the disease as non-affected parents also carry the mutation and, similarly, Fezf2 knockout mouse embryos electroporated with the mutant human FEZF2 construct did not display any obvious defects in the corticospinal tract, a pathway whose development depends on FEZF2. We extended the genetic analysis of these five FEZF2-linked families versus five FEZF2 non-linked families by studying de novo and transmitted copy number variation (CNV) and performing Whole Exome Sequencing (WES). We identified damaging mutations in the subset of 65 genes strongly associated with ASD whose co-expression analysis suggests an impact on the prefrontal cortex during the mid-fetal periods. From these results, we propose that both FEZF2 deletion and multiple hits in the repertoire of these 65 genes are necessary to generate an ASD phenotype.\n\nSignificance StatementThe human neocortex is a highly organized laminar structure with neuron positioning and identity of deep-layer cortical neurons that depend on key transcription factors, such as FEZF2, SATB2, TSHZ3 and TBR1. These genes have a specific spatio-temporal pattern of expression in human midfetal deep cortical projection neurons and display mutations in patients with Autism Spectrum Disorder (ASD). Here, we identified a linkage region involving FEZF2 gene in five consanguineous families with an ASD proband. For these FEZF2-allele linked probands, we identified a four-codon deletion in FEZF2 and damaging mutations in other high-risk ASD genes, that exhibit regional and cell type-specific convergence in neocortical deep-layer excitatory neurons, suggesting a multi-hit genomic architecture of ASD in these consanguineous families.

neuroscience↗

DYRK1A up-regulation specifically impairs a presynaptic form of long-term potentiation

Chromosome 21 DYRK1A kinase has long been associated with a variety of psychiatric diseases including Down Syndrome. We previously showed that Dyrk1A interacts with SWI/SNF (SWItch/Sucrose Non-Fermentable) nucleosome remodeling complex inducing expression changes of genes encoding key neuronal proteins. However, the functional impact of this kinase at the synapse level remains unclear. We studied a mouse model that incorporated the YAC 152F7 (570 kb) encoding six chromosome 21 genes including DYRK1A. We found that DYRK1A Interacts with the key chromatin remodelers EP300 and CREBBP. Moreover, we observed changes in the transcriptional levels of genes encoding presynaptic proteins involved in glutamate vesicle exocytosis, namely Rims1, Munc13-1, Syn2, Rab3A. This result prompted us to investigate the two main forms of long-term potentiation (LTP) required for learning and memory: the (N-methyl d-aspartate) receptor-dependent postsynaptic form versus the glutamate release-dependent presynaptic form. Interestingly, extracellular electrophysiological recordings in hippocampal slices of the YAC mouse line revealed that only the presynaptic forms of plasticity were impacted, leaving the post-synaptic form of plasticity intact. T o refine our findings, we used a mouse BAC 189N3 (152 kb) line that only triplicate the gene Dyrk1A. Again, we found that this presynaptic form of LTP is also impaired in this mouse line. This result demonstrates that abnormal up-regulation of Dyrk1A alone is sufficient to inhibit specifically the presynaptic forms of LTP. Altogether, our results suggest that impairment of DYRK1A gene dosage may impact memory precision, and therefore reinforce our mechanistic understanding of the cognitive impairment detected in this mouse model.

neuroscience↗

BIN1 genetic risk factor for Alzheimer is sufficient to induce early structural tract alterations in entorhinal cortex-dentate gyrus pathway and related hippocampal multi-scale impairments

Genetic factors are known to contribute to Late Onset Alzheimers disease (LOAD) but their contribution to pathophysiology, specially to prodomic phases accessible to therapeutic approaches are far to be understood. To translate genetic risk of Alzheimers disease (AD) into mechanistic insight, we generated transgenic mouse lines that express a [~]195 kbp human BAC that includes only BIN1, a gene associated to LOAD. This model gives a modest BIN1 overexpression, dependent of the number of BAC copies. At 6 months of age, we detected impaired entorhinal cortex (EC)-hippocampal pathways with specific impairments in EC-dentate gyrus synaptic long-term potentiation, dendritic spines of granular cells and recognition episodic memory. Structural changes were quantified using MRI. Their whole-brain functional impact were analyzed using resting state fMRI with a hypoconnectivity centered on entorhinal cortex. These early phenotype defects independent of any changes in A-beta can be instrumental in the search for new AD drug targets.

neuroscience↗