Characterisation of neurons derived from a cortical human neural stem cell line CTX0E16

Abstract

Introduction: Conditionally immortalised human neural progenitor cells (hNPCs) represent a robust source of native neural cells to investigate physiological mechanisms in both health and disease. However, in order to recognise the utility of such cells, it is critical to determine whether they retain characteristics of their tissue of origin and generate appropriate neural cell types upon differentiation. To this end, we have characterised the conditionally immortalised, cortically-derived, human NPC line, CTX0E16, investigating the molecular and cellular phenotype of differentiated neurons to determine whether they possess characteristics of cortical glutamatergic neurons.

Methods: Differentiated CTX0E16 cells were characterised by assessing expression of several neural fates markers, and examination of developing neuronal morphology. Expression of neurotransmitter receptors, signalling proteins and related proteins were assessed by q- and RT-PCR and complemented by Ca(2+) imaging, electrophysiology and assessment of ERK signalling in response to neurotransmitter ligand application. Finally, differentiated neurons were assessed for their ability to form putative synapses and to respond to activity-dependent stimulation.

Results: Differentiation of CTX0E16 hNPCs predominately resulted in the generation of neurons expressing markers of cortical and glutamatergic (excitatory) fate, and with a typical polarized neuronal morphology. Gene expression analysis confirmed an upregulation in the expression of cortical, glutamatergic and signalling proteins following differentiation. CTX0E16 neurons demonstrated Ca(2+) and ERK1/2 responses following exogenous neurotransmitter application, and after 6 weeks displayed spontaneous Ca(2+) transients and electrophysiological properties consistent with that of immature neurons. Differentiated CTX0E16 neurons also expressed a range of pre- and post-synaptic proteins that co-localized along distal dendrites, and moreover, displayed structural plasticity in response to modulation of neuronal activity.

Conclusions: Taken together, these findings demonstrate that the CTX0E16 hNPC line is a robust source of cortical neurons, which display functional properties consistent with a glutamatergic phenotype. Thus CTX0E16 neurons can be used to study cortical cell function, and furthermore, as these neurons express a range of disease-associated genes, they represent an ideal platform with which to investigate neurodevelopmental mechanisms in native human cells in health and disease.

Fig. 1

Characterisation of undifferentiated and differentiated CTX0E16 hNPCs. a, b Representative images of undifferentiated (days differentiated (DD) 0) CTX0E16 cells immunostained for the neural progenitor cell (NPC) markers nestin (a) and Sox2 (b). c At DD 0, the majority of CTX0E16 NPCs were positive for the cell proliferation marker, KI67. d By DD 5, the number of KI67-positive CTX0E16 cells had greatly reduced. e Quantification of the number of KI67 positive cells as a percentage of total cells; n = approximately 1,500 cells from three independent experiments carried out in triplicate; error bars represent SD; ***p <0.001 (Student’s unpaired t test). f, g Only a few cells were positive for the astrocyte marker S100β at DD 0 (f) or DD 28 (g). h, i At DD 0 very few cells expressed the neuronal marker Tau (h). However, after 28 days of differentiation, the majority of cells expressed Tau (i). j, k Similarly, very few cells were positive for the neuronal marker MAP2 at DD 0 (j), but by DD 28 the majority of cells were positive for MAP2, indicating that the vast majority of cells at this time point had differentiated into neurons. l Number of proliferative (DD 0) or differentiated (DD 28) cells positive for S100β, Tau or MAP2; N = approximately 1,500 cells from three independent experiments carried out in triplicate; error bars represent SD; ***p <0.001 (two way analysis of variance with Tukey post hoc analysis). m Quantitative PCR (q-PCR) analysis revealed an increase in the expression of dorsal forebrain marker, EMX1 and cortical neuron marker TBR1 in DD 28 CTX0E16 neurons compared to DD 0 cells; n = 4 independent experiments carried out in triplicate; error bars represent SD; ***p <0.001 (Student’s unpaired t test). Scale bar, 50 μm. DAPI 4′,6-diamidino-2-phenylindole

Fig. 2

Differentiation of CTX0E16 cells generates glutamatergic cortical neurons. a-d Representative images of differentiated (DD 28) CTX0E16 neurons co-stained for 4′,6-diamidino-2-phenylindole (DAPI), MAP2 and either the deep cortical layer marker Ctip2 (a), the upper cortical layer marker Cux1 (b), GABAergic interneuron marker calretinin (c) or calbindin (d). Yellow arrows indicate neurons expressing these markers. e Quantification revealed that 41.6 % of MAP2-positive neurons expressed Ctip2, whereas less than 10 % of MAP2-labelled neurons were positive for Cux1, calretinin or calbindin; n = approximately 1,500 cells per condition from three independent experiments carried out in triplicate; error bars represent SD. (f, g) Images of DD28 CTX0E16 neurons co-stained for DAPI, MAP2 and the glutamatergic and excitatory markers VGlut1 (f) or CaMKIIα (g). h, i DD 28 CTX0E16 neurons were also co-stained for DAPI, MAP2 and the GABAergic markers GAD65/67 (h) or VGAT (i). j Quantification of F-I revealed that over 65 % of MAP2-psoitive neurons also expressed VGlut1 or CaMKIIα, whereas less than 45 % of MAP2-neurons were also positive for either GAD65/67 or VGAT; n = approximately 1,500 cells per condition from three independent experiments carried out in triplicate; error bars represent SD. Scale bar, 20 μm

Fig. 3

Differentiated CTX0E16 neurons display morphological features of pyramidal neurons. a-c Generation of small neurites in βIII Tubulin (Tuj1)-positive CTX0E16 cells after 2 or 4 days of differentiation. d-f Expression of green fluorescent protein (GFP) in young CTX0E16 neurons, reveals the development of neuronal morphology; at differentiation day (DD) 15, neurite processes extend from the cell’s soma (d). By DD 20, a single long and thin process can be seen emerging from the cell’s soma with a thicker single process also emerging from the opposite side. Note the pyramidal shape of the cell soma. f By DD 35, the dendritic process displays some level of arborisation; additional smaller processes protruding from the cell soma are also evident. g, h Double immunostaining of DD 35 CTX0E16 neurons for the trans-Golgi marker, GM130 and MAP2. In MAP2-positive neurons, GM130 is clearly seen orientated towards a single, typically the longest, dendrite (red arrows). This indicates the primary dendrite, and the formation of a polarized morphology (g). h High-magnification images reveal that the Golgi-network is present along the primary dendrites (red arrow). Scale bars, 20 μm (a-g) and 5 μm (h). DAPI 4′,6-diamidino-2-phenylindole

Fig. 4

CTX0E16 cells generate neurons that respond to neurotransmitter stimulation. a Expression of a subset of neurotransmitter receptors, synaptic and signalling proteins is increased in differentiation day (DD) 28 CTX0E16 neurons as compared to undifferentiated (DD 0) CTX0E16 neural stem cells (NPCs) indexed by quantitative PCR (q-PCR); n = 4 independent experiments carried out in triplicate; error bars represent SD; **p <0.01; ***p <0.001 (Student’s unpaired t test). b Representative image of neurons loaded with Fura-2 AM used for single cell Ca²⁺ imaging. c-e Representative traces of intracellular Ca²⁺ in response to various neurotransmitter receptor ligands. g Number (%) of total cells generating 5 %, 20 % or 50 % responses following treatment with neurotransmitter ligands; n = 3 independent experiments carried out in triplicate (600 cells in total). h Mean maximal response produced by DD 28 CTX0E16 neurons following application of neurotransmitter ligands; n = 3 independent experiments carried out in triplicate (600 cells in total); error bars represent SD

Fig. 5

Development of functional properties in CTX0E16 neurons. a Representative image of Fluo-4 AM-loaded CTX0E16 neurons used for single cell Ca²⁺ imaging. b Representative time series of 18 neurons displaying spontaneous Ca²⁺ transients. Spontaneous activity was classified as a somatic calcium event greater than 5 % ΔF/F0: 38.0 ± 6.89 % cells displayed spontaneous activity over an 80-second period of imaging (n = 155 cells from 14 coverslips). c, d Representative traces of intracellular Ca²⁺ in responses to 50 mM KCl (c) or 1 μM tetrodotoxin (TTX) (d). e Resting membrane potential (Vm) recorded in current clamp progressively becomes more negative as CTX0E16 neurons become more mature (day of differentiation (DD) 29−DD 61); error bars represent SD. f Representative action potential recorded in voltage clamp in the cell attached configuration, recorded from DD 50 CTX0E16 neuron. g Representative voltage clamp recording at a holding potential of −70 mV in DD 36 CTX0E16 neurons. The downward deflections indicate the presence of AMPA receptor-mediated spontaneous excitatory postsynaptic currents (EPSCs). h Example of a spontaneous N-methyl-D-aspartate (NMDA) receptor-mediated EPSC recorded in voltage clamp at +40 mV from a DD 33 CTX0E16 neuron; n = 3–6 cells from at least three independent coverslips

Fig. 6

Phosphorylation of ERK1/2 kinase in differentiated CTX0E16 neurons following activation of neurotransmitter receptors. a, b Treatment with glutamate results in a time-dependent increase in phospho-ERK1/2 levels in differentiation day (DD) 28 CTX0E16 neurons. By 60 minutes, phospho-ERK1/2 levels are still raised as compared to control levels (a). Conversely, GABA application results in a transient increase in phospho-ERK1/2 levels; after 50 minutes, p-ERK1/2 levels have returned to baseline (b). c-e Treatment with dopamine (c) or 5-HT (d) results in activation of ERK1/2; CTX0E16 neurons respond to dopamine with a slower temporal profile as compared to 5-HT. Phospho-ERK1/2 levels are elevated following application of acetylcholine, but not significantly compared to baseline; n = 3 independent experiments carried out in triplicate; error bars represent SEM; *p <0.05; **p <0.01; ***p <0.001 (one way analysis of variance with Tukey post hoc analysis)

Fig. 7

CTX0E16 neurons express pre- and post-synaptic proteins. a-e In differentiation day (DD) 35 CTX0E16 neurons, puncta for the post-synaptic proteins PSD-95 (a), GluA1 (b), GluA2 (c), GluN1 (d) and gephyrin (e) were observed along MAP2-positive dendrites (red arrows). f-j The excitatory pre-synaptic proteins bassoon (f), synapsin1 (g), VGlut1 (h), and the inhibitory pre-synaptic proteins VGAT (i) and GAD65/67 (j) were observed in punctate structures, either along or juxtaposed along MAP2-positive dendrites. Scale bars for a-j are 5 μm

Fig. 8

Differentiated CTX0E16 neurons display hallmarks of putative synapses and respond to activity-dependent stimulation. a-c Representative confocal images of differentiation day (DD) 35 CTX0E16 neurons immunostained for MAP2, PSD-95 and either bassoon (a), synapsin 1 (b) or VGAT (c). As previously observed, all synaptic proteins display punctate distribution along dendrites. In addition, a subset of PSD-95 puncta co-localised with all three pre-synaptic proteins (red arrow), indicating the presence of putative synapses; co-localisation is seen as white puncta. Not all pre-synaptic puncta co-localised with PSD-95 (white open arrowheads), suggesting that synaptogenesis was ongoing. d, e Representative confocal images of a distal dendrite of green fluorescent protein (GFP)-expressing ) CTX0E16 neurons (DD 35). Examination of dendrites revealed the presence of filopodia and dendritic spine-like structures, indicative of ongoing synaptogenesis. Yellow boxes indicate region used in high magnification in inset. f Western blotting of CTX0E16 neurons at different stages of differentiation demonstrate an increase in PSD-95 with maturation, consistent with an increase in synaptogenesis. Rat ctx, whole-cell lysate taken from rat cortex. g Representative binary images of GFP-expressing CTX0E16 neurons (DD 15) following treatment with control conditions (vehicle or osmotic control (30 nM NaCl)) or activity-dependent stimulation with 30 nM KCl for 7 hours. h Measurement of average neurite length in response to stimulation demonstrates that activity-dependent stimulation results in an increase in average neurite length; n = 14–22 neurons from 3–5 independent experiments carried out in triplicate; error bars represent SEM; *p <0.05 (one way analysis of variance with Tukey post hoc analysis). Scale bars for a-e are 5 μm; d and e inset, 1 μm; g, 50 μm