Efficient generation of human CA3 neurons and modeling hippocampal neuronal connectivity in vitro (single cells)
ABSTRACT: Despite widespread interest in using human stem cells in neurological disease modeling, a suitable model system to study human neuronal connectivity is lacking. Here, we report a protocol for efficient differentiation of hippocampal pyramidal neurons and an in vitro model for hippocampal neuronal connectivity. We developed an embryonic stem cell (ESC)- and induced pluripotent stem cell (iPSC)-based protocol to differentiate human CA3 pyramidal neurons from patterned hippocampal neural progenitor cells (NPCs). This differentiation induces a comprehensive patterning and generates multiple CA3 neuronal subtypes. The differentiated CA3 neurons are functionally active and readily form neuronal connection with dentate granule (DG) neurons in vitro, recapitulating the synaptic connectivity within the hippocampus. When we applied this neuronal co-culture approach to study connectivity in schizophrenia, we found deficits in spontaneous activity in patient iPSC derived DG–CA3 co-culture by multi-electrode array recording. In addition, both multi-electrode array recording and whole cell patch clamp electrophysiology revealed a reduction in spontaneous and evoked neuronal activity in CA3 neurons derived from schizophrenia patients. Altogether these results underscore the relevance of this new model in studying diseases with hippocampal vulnerability. Overall design: 4 technical replicates were used and later pooled together for the bioinformatic analysis.
Project description:Despite widespread interest in using human stem cells in neurological disease modeling, a suitable model system to study human neuronal connectivity is lacking. Here, we report a protocol for efficient differentiation of hippocampal pyramidal neurons and an in vitro model for hippocampal neuronal connectivity. We developed an embryonic stem cell (ESC)- and induced pluripotent stem cell (iPSC)-based protocol to differentiate human CA3 pyramidal neurons from patterned hippocampal neural progenitor cells (NPCs). This differentiation induces a comprehensive patterning and generates multiple CA3 neuronal subtypes. The differentiated CA3 neurons are functionally active and readily form neuronal connection with dentate granule (DG) neurons in vitro, recapitulating the synaptic connectivity within the hippocampus. When we applied this neuronal co-culture approach to study connectivity in schizophrenia, we found deficits in spontaneous activity in patient iPSC derived DG–CA3 co-culture by multi-electrode array recording. In addition, both multi-electrode array recording and whole cell patch clamp electrophysiology revealed a reduction in spontaneous and evoked neuronal activity in CA3 neurons derived from schizophrenia patients. Altogether these results underscore the relevance of this new model in studying diseases with hippocampal vulnerability. Overall design: 8 samples, 2-3 technical replicates per sample
Project description:Human doublecortin (DCX) mutations are associated with severe brain malformations leading to aberrant neuron positioning (heterotopia), intellectual disability and epilepsy. The Dcx protein plays a key role in neuronal migration, and hippocampal pyramidal neurons in Dcx knockout (KO) mice are disorganized. The single CA3 pyramidal cell layer observed in wild type (WT) is present as two abnormal layers in the KO, and CA3 KO pyramidal neurons are more excitable than WT. Dcx KO mice also exhibit spontaneous epileptic activity originating in the hippocampus. It is unknown however, how hyperexcitability arises and why two CA3 layers are observed. Transcriptome analyses were performed to search for perturbed postnatal gene expression, comparing Dcx KO CA3 pyramidal cell layers with WT. Gene expression changes common to both KO layers indicated mitochondria and Golgi apparatus anomalies, as well as increased cell stress. Intriguingly, gene expression analyses also suggested that the KO layers differ significantly from each other, particularly in terms of maturity. Layer-specific molecular markers and BrdU birthdating to mark the final positions of neurons born at distinct timepoints revealed inverted layering of the CA3 region in Dcx KO animals. Notably, many early-born ‘outer boundary’ neurons are located in an inner position in the Dcx KO CA3, superficial to other pyramidal neurons. This abnormal positioning likely affects cell morphology and connectivity, influencing network function. Dissecting this Dcx KO phenotype sheds light on coordinated developmental mechanisms of neuronal subpopulations, as well as gene expression patterns contributing to a bi-layered malformation associated with epilepsy. Expression profiling by array Overall design: 15 laser microdissection samples analyzed corresponding to 5 wild type mouse brain hippocampi CA3 region, 5 doublecortin knockout mouse brain hippocampi internal part of CA3 region and from the same 5 knockout animals, mouse brain hippocampi external part of CA3 region
Project description:Neurodegenerative brain disorders become more common in the aged. Most of these disorders are associated with or caused by selective death of certain neuronal subpopulations. The mechanisms underlying the differential vulnerability of certain neuronal populations are still largely unexplored and few neuroprotective treatments are available to date. Elucidation of these mechanisms may lead to a greater understanding of the pathogenesis and treatment of neurodegenerative diseases. Moreover, preconditioning by a short seizure confers neuroprotection following a subsequent prolonged seizure. Our goal is to identify pathways that confer vulnerability and resistance to neurotoxic conditions by comparing the basal and preconditioned gene expression profiles of three differentially vulnerable hippocampal neuron populations. Hippocampal CA1 and CA3 pyramidal neurons are highly susceptible to seizures and ischemia, whereas dentate gyrus granule cells are relatively resistant. A brief preconditioning seizure confers protection to the pyramidal cells. We will first determine gene expression profiles of untreated rat CA1 and CA3 pyramidal cells, and dentate granule cells, using laser capture microscopy to obtain region-specific neuronal mRNA. We will then determine the effect of a brief preconditioning seizure, which is neuroprotective in CA1 and CA3, on these expression profiles. We hypothesize that common molecular mechanisms exist in neurons that determine their susceptibility to seizure-induced injury. Intrinsic differences in gene expression exist between hippocampal glutamatergic CA1 and CA3 pyramidal neurons, on the one hand, and dentate granule cells on the other, which contribute to the greater susceptibility of pyramidal neurons to degeneration in experimental stroke and epilepsy. We specifically hypothesize that differences in basal energy metabolism genes may confer differential susceptibility to neurodegeneration produced by seizures and ischemia. Anesthetized animals will be sacrificed by decapitation, and frozen 10 micron sections will be lightly stained with cresyl violet to identify cell layers in the hippocampus. Approximately 1000 neurons from each of the three cell layers will be isolated by LCM. Poly-A RNA will be amplified using a modified Eberwine protocol. The quality of our aRNA will be evaluated by quantitative RT-PCR of GluR6 and KA2 mRNA levels before we send the samples to the Center for labeling and hybridization to Affymetrix rat 230A arrays. We will provide a one-round amplification cDNA product to the center for labeling and hybridization. This protocol is identical to a previously approved study by Jim Greene in our laboratory.
Project description:Here, we used next-generation RNA sequencing (RNA-seq) to produce a quantitative, whole genome atlas of gene expression for every excitatory neuronal class in the hippocampus; namely, granule cells and mossy cells of the dentate gyrus, and pyramidal cells of areas CA3, CA2, and CA1. Moreover, for the canonical neurons of the trisynaptic loop, we profiled transcriptomes at both dorsal and ventral poles, producing a cell class- and region-specific transcriptional atlas for these canonical populations. Overall design: Hippocampal RNA profiles were generated by deep sequencing on Illumina HiSeq 2500, with three biological replicates per population
Project description:Complete global brain ischemia (CGBI) and reperfusion occur following resuscitation from cardiac arrest. Different brain neurons are selectively vulnerable to CGBI: pyramidal neurons of hippocampal CA3 survive 10 min CGBI but those of CA1 die at 3 days following 10 min CGBI. CA3 neurons are expected to have more robust stress responses and repair responses than CA1 neurons. We used microarrays to compared total and polysome-bound mRNAs in CA1 and CA3 at 8 hr reperfusion after 10 min CGBI in Long Evans male rats to ascertain differences in total vs polysome-bound gene expression. Male Long Evans rats were subjected to (1) sham operation (non-ischemic control, NIC) or normothermic CGBI of 10 min followed by 8 hr reperfusion (8R). Hippocampal CA1 and CA3 were dissected. n = 5 CA1 or CA3 were pooled to give a single replicate and there were 3 or 4 replicates per group. Post-mitochondrial supernatant (PMS) was prepared. Twenty percent of PMS was TRIzol extracted to give total RNA. The remainder was run on a 20% sucrose pad to isolate polysome pellets, which were also TRIzol extracted to give polysome RNA. Total and polysome RNA were then run on Affymetrix Rat Gene 2.0 microarrays.
Project description:Recent advances in single-cell RNAseq technologies are enabling new cell type classifications. For neurons, electrophysiological properties traditionally guide cell type classification but correlating RNAseq data with electrophysiological parameters has been difficult. Here we demonstrate RNAseq of electrophysiologically and synaptically characterized individual, patched neurons in the hippocampal CA1-region and subiculum, and relate the resulting transcriptome data to their electrical and synaptic properties. In this analysis, we explored the hypothesis that precise combinatorial interactions between matching cell-adhesion and signaling molecules shape synapse specificity. In analyzing interneurons and pyramidal neurons that are synaptically connected, we identified two independent, developmentally regulated networks of interacting genes encoding cell-adhesion, exocytosis and signal-transduction molecules. In this manner, our data allow postulating a presumed cell-adhesion and signaling code, which may explain neuronal connectivity at the molecular level. Our approach enables correlating electrophysiological with molecular properties of neurons, and suggests new avenues towards understanding synaptic specificity. Overall design: These data include 15 tissue samples (including 3 independent replicas in 5 developmental stages) as well as 93 single-cell samples (including CA1 cholecystokinin, parvalbumin, and pyramidal neurons as well as subiculum burst and regular firing pyramidal neurons).
Project description:The long-lasting changes in synaptic connectivity that underlie long-term memory require new RNA and protein synthesis for their persistence. To elucidate the temporal pattern of gene expression that gives rise to long-lasting, learning-related neuronal plasticity, we profiled RNAs in mouse hippocampal CA3-CA1 slices following induction of late phase long-term potentiation (LTP), analyzing differential expression (DE) specifically within pyramidal excitatory neurons by Translating Ribosome Affinity Purification RNA sequencing (TRAP-seq). We detected time-dependent changes in up- and down-regulated ribosome-associated mRNAs over the two hours following LTP induction, with minimal overlap of DE transcripts between time points. TRAP-seq revealed greater numbers and amplitudes of LTP-induced changes than RNA-seq of all cell types in the hippocampus. Transcripts that were DE by TRAP-seq but not RNA-seq were enriched in mRNAs encoding cytoskeletal and cell adhesion proteins, while RNA-seq identified DE in many non-neuronal mRNAs. Together our results highlight the importance of considering both the time course and the cell-type specificity of activity-dependent gene expression during memory formation. Overall design: RNA-seq and TRAP-seq were performed on RNA isolated from acute hippocampal slices following a chemical LTP induction, which utilizes forskolin followed by elevated levels of Ca2+/K+ in 0 Mg2+ artificial cerebral spinal fluid. We crossed the RiboTag line of mice with a Camk2a-cre line to generate mice heterozygous for both cre and RiboTag, which expresses an HA-tagged ribosomal protein L22 in excitatory neurons. Total RNA from all cell-types and TRAP RNA from excitatory neurons was isolated from the same hippocampal slice homogenate at 30 min, 60 min, and 120 min following LTP induction from LTP induced slices as well as time-matched vehicle-treated control slices. Both LTP and basal slices were generated from the same animals on the same day, with slices from each animal evenly distributed between basal and LTP. Thus, each basal replicate has a corresponding LTP treated replicate.