Toby N. Behar

Neurotrophins Stimulate Chemotaxis of Embryonic Cortical Neurons

During mammalian cortical development, neuronal precursors proliferate within ventricular regions then migrate to their target destinations in the cortical plate, where they organize into layers. In the rat, most cortical neuronal migration occurs during the final week of gestation (Bayer et al., 1991; Jacobson, 1991). At this time (E15–E21), reverse transcriptase‐polymerase chain reaction demonstrated that cortical homogenates contain mRNA encoding brain derived neurotrophic factor (BDNF) and the catalytic form of its high‐affinity receptor, TrkB. lmmunocytochemistry and in situ hybridization of sections revealed that the catalytic TrkB receptors predominantly localize to regions containing migratory cells. Many TrkB+ cells exhibited the classic morphology of migrating neurons, suggesting that TrkB ligands play a role in cortical neuronal migration. We analysed whether TrkB ligands influence the motility of embryonic cortical cells (from E15–E21) using a quantitative in vitro chemotaxis assay. High‐affinity TrkB ligands (BDNF and NT4/5) stimulated chemotaxis (directed migration) of embryonic neurons at concentrations ranging from 1 to 100 ng/ml. NT‐3, a low‐affinity TrkB ligand, only stimulated significant migration at high concentrations (±100 ng/ml). Peak migration to BDNF was observed at gestational day 18 (E18). BDNF‐induced chemotaxis was blocked by either tyrosine kinase inhibitor, K252a, or the Ca2+‐chelator, BAPTA‐AM, suggesting that BDNF‐induces motility via autophosphorylation of TrkB receptor proteins and involves Ca2+‐dependent mechanisms. BDNF‐stimulation of increased cytosolic Ca2+ was confirmed with optical recordings of E18 cortical cells loaded with Ca2+ indicator dye. Thus, signal transduction through the TrkB receptor complex directs neuronal migration, suggesting that, in vivo, BDNF exerts chemotropic effects that are critical to morphogenesis of the cortex.

Platelet‐derived growth factor induces chemotaxis of neuroepithelial stem cells

The ability of differentiating cells to migrate within the developing central nervous system (CNS) depends on extrinsic guidance signals, some of which are growth factors. In this study we have investigated the chemotactic response of cultured stem cells from the embryonic rat cortex to platelet‐derived growth factor (PDGF). Nestin‐positive stem cells from the developing CNS can be maintained and expanded in vitro under serum‐free conditions in the presence of basic fibroblast growth factor (bFGF). Northern blot analysis of PDGF receptor expression revealed both α‐ and β‐receptors on bFGF‐treated neural stem cells. Both PDGF‐AA and PDGF‐BB readily induced directed migration of cultured neuroepithelial cells as measured in a microchemotaxis assay. Blocking of the migratory response was achieved by incubation with PDGF isoform‐specific antibodies. More than 90% of the migrating cells were nestin‐positive and incorporation of BrdU was also seen suggesting the cells to be immature and not yet committed to a specific cell lineage. These findings suggest a role for PDGF in cell migration in the developing cortex. J. Neurosci. Res. 53:521–530, 1998.

Quantitative analysis of transient GABA expression in embryonic and early postnatal rat spinal cord neurons

GABA expression was investigated using biochemical analysis of spinal cord homogenates and immunocytochemical analysis of cells acutely dissociated from the embryonic and postnatal rat spinal cord. gamma-Aminobutyric acid (GABA) was detected by both methods as early as embryonic day 13 (E13). At E13, the percentage of neurons that were GABA+ was 0.5%. This value increased during embryogenesis, peaked during the first two postnatal weeks to just over 50%, and declined to approximately 20% by the third postnatal week emphasizing the transient nature of GABA expression. At E17 there was a pronounced, positive ventro-dorsal and rostro-caudal gradient of GABA+ cells that persisted until just before birth. At this time the gradients reversed in cervical and lumbosacral regions indicating that GABA immunoreactivity in discrete anatomical regions is also a transient phenomenon. During the embryonic period GABA immunoreactivity was diffusely distributed throughout cell bodies and proximal processes. At E21, both GABA and synaptophysin were present in the same cells. However the two antigens did not co-localize point for point. By postnatal day 21 GABA immunoreactivity appeared in puncta that co-localized entirely with puncta of synaptophysin immunoreactivity. The sizable percentage of neurons that transiently express GABA during development, and the fact that it can be detected prior to the synaptic form of glutamic acid decarboxylase (GAD65), suggest that the amino acid may play a significant role during differentiation before it functions as an inhibitory neurotransmitter.

Analysis of the anatomical distribution of GAD67 mRNA encoding truncated glutamic acid decarboxylase proteins in the embryonic rat brain.

During development of the central nervous system (CNS) the gene that encodes the 67 kDa form of glutamic acid decarboxylase (GAD) undergoes alternative splicing. The alternatively spliced variants include an exon (referred to as ES, for embryonic stop) that contains a premature stop codon. The detection of mRNA containing the ES exon in embryonic rat brain has been previously reported (Proc. Natl. Acad. Sci., 87 (1990) 8771-8775). We have used in situ hybridization to identify the anatomical distribution of ES mRNA in the embryonic rat brain during two stages of development, embryonic day 17 (E17) and E20. At E17, GAD67 mRNA was expressed in several CNS regions that were destined to contain GABAergic neurons when mature. ES transcripts were predominantly localized to ventricular zones and other regions associated with populations of proliferative cells at E17 and E20. At both ages, however, the alternatively spliced variants were also detected in regions of brain associated with migratory or post-mitotic neurons. GAD67 transcripts that did not include the ES exon were localized to anatomical areas that contained post-mitotic, and often post-migratory neurons. The temporal and spatial disappearance of mRNA containing the ES exon generally followed a caudal-to-rostral gradient which paralleled neuronal terminal mitosis and differentiation.

Many spinal cord cells transiently express low molecular weight forms of glutamic acid decarboxylase during embryonic development

At early developmental stages in the rat spinal cord (embryonic day 13), when neuronal progenitors are still proliferating, most differentiating neurons express truncated forms of glutamic acid decarboxylase (GAD) (approximately 25 kDa) which are the products of alternative splicing of the GAD67 gene. These truncated proteins do not appear to synthesize gamma-aminobutyric acid (GABA). The amino acid is detected in cells only after alternative splicing of the GAD67 gene generates a full-length, 67 kDa enzymatically active form of GAD. Both the 67 kDa GAD and GABA colocalize and appear diffusely distributed in the cytoplasm of embryonic neurons. GABA does not appear associated with synaptic vesicles until after birth, when its intracellular distribution becomes punctate and it colocalizes with synaptophysin. At this time, it also colocalizes with an immunologically distinct 65 kDa GAD protein encoded by a second GAD gene (GAD65). Expression of different GAD-related proteins with distinct intracellular distributions during development suggests that GABA, the product of these enzymes, may have trophic or metabolic roles during spinal cord differentiation.

GABAergic cells and signals appear together in the early post-mitotic period of telencephalic and striatal development

Single cell suspensions derived from embryonic telencephala taken from embryos of gestational day 13 (E13) as well as rat striatal tissue from E14, 15 and 17 were prepared by tissue digestion with papain. Cell suspensions were analyzed by flow cytometry or plated onto poly-D-lysine-coated culture dishes for either nuclear staining or immunocytochemistry. Experiments on functional Na+ channels and GABAA receptor expression were carried out using a fluorescence-activated cell sorter (FACS) and a negatively charged fluorescent indicator dye (oxonol). FACS analysis of embryonic cell suspensions at E13-17 consistently revealed one major subpopulation accounting for 85-90% of the events and one minor subpopulation (10-15% of the total). When sorted, the major subpopulation consisted of phase-bright cells of 5-7 microns diameter some of which had neurites. The minor population consisted of phase-dark cells and resealed membranes of 0.5-4 microns diameter as well as debris. Almost all the cells obtained in the high FALS (forward-angle light scatter) subpopulation at E17 expressed 200-kDa neurofilament and tetanus toxin antigens while the small diameter cells seldom expressed tetanus toxin and particles never did. A small number of GABA-containing neurons were detected in the telencephalon at E13 (3%) and in the developing striatum at E14 (6%). All of the GABA-containing neurons expressed neurofilament. In the embryonic rat striatum, nanomolar concentrations of muscimol (GABAA agonist) induced depolarizing responses. A small number of cells in the high FALS subpopulation were responsive to muscimol starting at embryonic day 14, and the number of responsive cells increased at E15.(ABSTRACT TRUNCATED AT 250 WORDS)

Redox regulation of neuronal migration in a down syndrome model

Down Syndrome (DS), one of the major genetic causes of mental retardation, is characterized by disrupted corticogenesis produced, in part, by an abnormal layering of neurons in cortical laminas II and III. Because defects in the normal migration of neurons during corticogenesis can result in delayed cortical radial expansion and abnormalities in cortical layering, we have examined neuronal migration in murine trisomy 16 (Ts16), a mouse model for DS. Using an in vitro assay for chemotaxis, our data demonstrate that the number of acutely dissociated Ts16 cortical neurons migrating in response to glutamate or N-methyl-D-aspartate (NMDA), known chemotactic factors, was decreased compared to normal littermates, suggesting a defect in NMDA receptor- (NMDAR-) mediated events. Ts16 neurons did not lack NMDAR since expression of mRNA and protein for NMDAR subunits was observed in Ts16 cells. However, the number of cells that generated an observable current in response to NMDA was decreased compared to normal littermates. Similar to DS, Ts16 CNS demonstrated an inherent oxidative stress likely caused by the triplication of genes such as SOD1. To determine if the abnormal redox state was a factor in the failure of NMDAR-mediated migration in Ts16, we treated Ts16 neurons with either n-acetyl cysteine (NAC) or dithiothrietol (DTT), known antioxidants. The reduction in NMDAR-mediated migration observed in Ts16 neurons was returned to normal littermate values by NAC or DTT. Our data indicate that oxidative stress may play a key role in the abnormal glutamate-mediated responses during cortical development in the Ts16 mouse and may have an impact on neuronal migration at critical stages.