- Email: [email protected]
Neuropharmacological properties of neurons derived from human stem cells
Neurochemistry International 59 (2011) 404–412
Contents lists available at ScienceDirect
Neurochemistry International journal homepage: www.elsevier.com/locate/neuint
Neuropharmacological properties of neurons derived from human stem cells Leanne Coyne a,1, Mu Shan a, Stefan A. Przyborski b, Ryoko Hirakawa a, Robert F. Halliwell a,* a b
School of Pharmacy, University of the Pacific, Stockton, CA, USA School of Biological & Biomedical Sciences, University of Durham, England, UK
A R T I C L E I N F O
A B S T R A C T
Article history: Received 15 October 2010 Accepted 19 January 2011 Available online 16 February 2011
Human pluripotent stem cells have enormous potential value in neuropharmacology and drug discovery yet there is little data on the major classes and properties of receptors and ion channels expressed by neurons derived from these stem cells. Recent studies in this lab have therefore used conventional patchclamp electrophysiology to investigate the pharmacological properties of the ligand and voltage-gated ion channels in neurons derived and maintained in vitro from the human stem cell (hSC) line, TERA2.cl.SP12. TERA2.cl.SP12 stem cells were differentiated with retinoic acid and used in electrophysiological experiments 28–50 days after beginning differentiation. HSC-derived neurons generated large whole cell currents with depolarizing voltage steps ( 80 to 30 mV) comprised of an inward, rapidly inactivating component and a delayed, slowly deactivating outward component. The fast inward current was blocked by the sodium channel blocker tetrodotoxin (0.1 mM) and the outward currents were significantly reduced by tetraethylammonium ions (TEA, 5 mM) consistent with the presence of functional Na and K ion channels. Application of the inhibitory neurotransmitters, GABA (0.1–1000 mM) or glycine (0.1–1000 mM) evoked concentration dependent currents. The GABA currents were inhibited by the convulsants, picrotoxin (10 mM) and bicuculline (3 mM), potentiated by the NSAID mefenamic acid (10–100 mM), the general anaesthetic pentobarbital (100 mM), the neurosteroid allopregnanolone and the anxiolytics chlordiazepoxide (10 mM) and diazepam (10 mM) all consistent with the expression of GABAA receptors. Responses to glycine were reversibly blocked by strychnine (10 mM) consistent with glycine-gated chloride channels. The excitatory agonists, glutamate (1–1000 mM) and NMDA (1– 1000 mM) activated concentration-dependent responses from hSC-derived neurons. Glutamate currents were inhibited by kynurenic acid (1 mM) and NMDA responses were blocked by MgCl2 (2 mM) in a highly voltage-dependent manner. Together, these findings show that neurons derived from human stem cells develop an array of functional receptors and ion channels with a pharmacological profile in keeping with that described for native neurons. This study therefore provides support for the hypothesis that stem cells may provide a powerful source of human neurons for future neuropharmacological studies. ß 2011 Elsevier Ltd. All rights reserved.
Keywords: Human neurons Stem cells Voltage-gated ion channels Ligand-gated receptors GABA Glutamate
1. Introduction Stem cells offer enormous potential value in regenerative medicine and drug discovery yet the number of published studies at this time utilizing stem cells in drug discovery, toxicity screening & high-content screening is exceedingly small (Razvi and Oosta, 2010). The paucity of data in these areas, in part, reflects the complex and time-consuming requirement to determine the functional proper-
* Corresponding author at: Department of Physiology & Pharmacology, TJ Long School of Pharmacy & Health Sciences, University of the Pacific, Stockton, CA 95211, USA. Tel.: +1 209 946 2074; fax: +1 209 946 2857. E-mail address: rhalliwell@pacific.edu (R.F. Halliwell). 1 Present address: California Northstate College of Pharmacy, 10811 International Drive, Rancho Cordova, CA 95670, USA. 0197-0186/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuint.2011.01.022
ties of the phenotypes derived from stem cells if they are to be used in pharmacological, toxicological or drug discovery studies. Over the past decade, some progress has been made in delineating the physiological and pharmacological properties of neurons derived from rodent stem cells. Several studies, for example, have reported that neurons from rodent embryonic stem (ES) cells develop voltage-activated calcium, sodium, and potassium currents and display spontaneous synaptic currents that are blocked by GABA or glutamate antagonists (Cai et al., 2004; Ma et al., 2004; Lang et al., 2004; Pagani et al., 2006; Biella et al., 2007; Nakagomi et al., 2009). Such cells also show rises in intracellular calcium in response to several neurotransmitters, including acetylcholine, N-methyl-D-aspartate (NMDA) and adenosine triphosphate (ATP) (Cai et al., 2004; Ma et al., 2004; Lang et al., 2004; Pagani et al., 2006). In a more recent study, Khaira et al. (2009) have
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
also demonstrated that mouse ES derived ‘GABAergic’ neurons elevate intracellular Ca2+ levels through activation of P2X2, P2X4 and P2Y1-like purinoceptors expressed in these cells. Functional studies of human stem cell (hSC) derived neurons, in contrast, are still in their infancy (Pouton and Haynes, 2007). Cho et al. (2002) for example, reported that an immortalized clonal stem cell line HB1.F3, derived from human embryonic telencephalon, differentiated into neurons expressing voltagegated potassium and sodium channels. Westerlund et al. (2003) later reported that multipotent neural stem cells resected from patients undergoing neurosurgery for epilepsy could be grown as neurospheres and subsequently differentiated to neurons and glia. The neurons developed functional IK and IA-like potassium channels and TTX-sensitive sodium channels after about 3 weeks in culture. Moe et al. (2005) similarly described the isolation and culture of multipotent progenitor cells from the temporal cortex of patients with intractable epilepsy. These cells could be grown as neurospheres in vitro, passaged and subsequently differentiated to glia and neurons. Electrophysiological recordings showed that the neurons were able to generate TTX-sensitive action potentials and spontaneous synaptic activity that was sensitive to GABAA and glutamate receptor antagonists. Human induced pluripotent stem cells (iPS) have also recently generated significant interest as sources of human neurons for neuropharmacological studies and for regenerative medicine. In support of this expectation, two recent studies have indicated that human iPS cells can differentiate into spinal-like (HB9/ISL1/2+) motor neurons able to fire action potentials and generate some spontaneous synaptic activity after 5–8 weeks in vitro (Hu et al., 2010; Karumbayaram et al., 2009). At the present time, however, there is little data on the pharmacology of the ion channels and receptors expressed by any hSC derived neurons. This study has therefore begun to explore the pharmacological properties of the ligand and voltage-activated currents recorded from neurons derived from human stem cells using patch-clamp techniques. We have utilized the human embryonal carcinoma (hEC) clonal stem cell line TERA2.cl.SP12 which displays an enhanced propensity to form neurons and glia and which express several molecular markers and several functional receptors and ion channels similar to those described in native neurons, making them an excellent experimental model (Stewart et al., 2004; Przyborski, 2001; Ulrich and Majumder, 2006). 2. Methods 2.1. Cell culture TERA2.cl.SP12 EC stem cells were prepared, maintained and differentiated according to methods previously described (e.g. Stewart et al., 2004). Briefly, stem cells were maintained in 75 ml tissue culture flasks at 37 8C, 5% CO2, 100% relative humidity in DMEM supplemented with foetal bovine serum (10%, v/v; Invitrogen), L-glutamine (2 mM final concentration; Sigma–Aldrich) and penicillin–streptomycin (100 U/ml/100 mg/ml). When cells reached approximately 80% confluence, they were detached from the flask base using micro-glass beads (Fisher Scientific), removed and differentiated either as monolayers or suspension aggregates. For monolayer cultures, stem cells were re-seeded into 35 mm tissue culture dishes (at a density of 2 104 cells/cm2) on sterile glass coverslips pre-coated with poly-Dlysine (10 mg/ml). Retinoic acid (10 mM Sigma–Aldrich) was now also included in the media. After 7 days, the mitotic inhibitors, cytosine-D-arabinoside (1 mM), fluoroxyuridine (10 mM) and uridine (10 mM) (all Sigma–Aldrich) were included in the culture media for a further 7 days. For suspension aggregates 1.25 106 stem cells were seeded into sterile 90 mm bacteriological Petri dishes and cultured in retinoic acid (RA) containing media, as described above. Cells were re-seeded every 3–4 days into new Petri dishes with fresh RA media. Cell aggregates were then harvested after 28 days differentiation, washed and plated onto sterile glass coverslips pre-coated with poly-D-lysine in 35 mm culture dishes. Cell aggregates were maintained on coverslips for a further 7–14 days in DMEM medium in the absence of retinoic acid but in the presence of the mitotic inhibitors (for the first 7 days only).
405
2.2. Immunocytochemistry For immunolabeling, stem cells and their neural derivates were cultured on sterile glass coverslips in 35 mm culture dishes, as described above. All cells were first washed in PBS and then fixed with 4% paraformaldehyde in PBS. The fixative was carefully removed, and the cells rinsed with PBS and placed overnight (at 4 8C) in a blocking solution, consisting of 5% normal donkey serum, 0.3% triton X-100 in PBS. Thereafter, cells were placed in primary antibodies (from Millipore) diluted in blocking solution and incubated overnight at 4 8C. The primary antibody concentrations used were mouse anti-Oct-4, 1:100, and mouse anti-b-III Tubulin 1:500. 24 h later, cells were washed with PBS and also twice with the blocking solution. At the completion of the final wash, cells were left in blocking solution for a further 30 min. Thereafter, cells were incubated with donkey anti-mouse IgG (Cy3-or FITC-conjugated) secondary antibody (Millipore) for 2 h at 1:250 dilution at room temperature and then washed with PBS and counter-stained with DAPI (100 ng/ml; Sigma–Aldrich) before mounting over chamber slides using anti-fade mounting solution. (Prolong1 Gold Antifade Reagent; Invitrogen). Immunolabeled cells were visualized through an inverted fluorescent microscope (Nikon Eclipse TE200) with an attached CCD camera (Microfire; Optronics). Fluorescence was observed using filter sets appropriate for each label. The percentage of immunofluorescent cells was determined by counting DAPI-positive cells and then immuno-positive for b-III Tubulin plus DAPI in 3–6 fields (at 50 magnification) across the slide. Data was gathered from 3 independently replicated experiments. 2.3. Electrophysiology Agonist and voltage-evoked currents were recorded from cells with a neuronallike morphology (phase bright with neurite processes), differentiated for 28– 50 days, using the whole-cell configuration of the patch-clamp technique as described previously (e.g. Coyne et al., 2007). Briefly, patch electrodes were made from borosilicate glass pipettes (Harvard Apparatus, Holliston, MA, USA) on a Narishige PB-7 electrode puller (East Meadow, NY, USA) and had tip resistances of 1–4 MV when filled with internal solution. Currents were recorded using an Axopatch 200B amplifier and headstage (Axon Instruments, Foster City, USA) and low-pass filtered at 10 kHz before digitization via a National Instruments DAQ card and a National Instruments BNC-2090 interface board (Austin, TX, USA) and then stored on a PC running WinWCP software (University of Strathclyde, UK). Whole cell currents were continuously monitored on the desktop computer. Series resistance and pipette and whole-cell capacitance were cancelled electronically. Cells were perfused with a bath solution containing the following (in mM): NaCl (140.0), KCl (5), MgCl2 (2.0), CaCl2 (1.0), HEPES (10.0). The internal solution contained the following (in mM): CsCl or KCl (140.0), MgCl2 (2.0), EGTA (11), HEPES (10), Mg-ATP (2.0). All solutions were titrated to pH 7.2 and internal solutions were filtered before use using 0.2 mm disposable filters (Whatman, Maidstone, England). All electrophysiology experiments were carried out at ambient room temperature (23–25 8C). Neural derivatives were voltage-clamped at a holding potential of 60 mV, unless otherwise stated. For determination of agonist-evoked current-to-voltage relationships, the membrane potential was stepped between 140 mV and 60 mV in 20 mV increments. To activate voltage-gated currents, 20 ms voltage steps were applied from 80 to 30 mV in 10 mV increments from a holding potential of 80 mV. For recording potassium currents, CsCl was replaced with KCl (140 mM) in the internal solution. All bath solutions for studying voltage-activated currents also contained 2 mM CoCl2, to prevent possible calcium channel activation. 2.4. Drugs and their application All drugs were obtained from Sigma–Aldrich (St. Louis, MO, USA) except TTX which was purchased from Tocris Bioscience (Ellisville, MO, USA). Stock solutions of bicuculline methyl bromide was made up in bath solution; pentobarbital-Na was dissolved in H2O; mefenamic acid was dissolved in NaOH (1 M), picrotoxin was dissolved in ethanol; chlordiazepoxide, diazepam and allopregnanolone were first dissolved in 100% dimethylsulphoxide (DMSO). The final bath concentrations of these solvents did not exceed 0.1% of the recording solutions and in control experiments had no effects on the responses. Agonists and other drugs were applied directly to cells under voltage-clamp from the tip of a 250 mm pipette connected to a Y-tube (fabricated in-house). Fresh bath solution was also perfused through the bath (at 1–2 ml/min) using a gravity-feed system (also manufactured in-house) to ensure that there was no build up of drug solutions in the bath. At least 3 control responses were recorded before the addition of agonists. Drugs were washed off once a clear asymptotic effect was observed. Control responses were re-established before further drug applications. 2.5. Data analysis Agonist- and voltage-evoked currents were measured at their peak amplitude. Current responses in the presence of drugs are expressed as a percentage of the control response (S.E.M. of n experiments). For concentration–response data, agonist-evoked currents were normalized to the maximal response evoked by a saturating concentration of agonist. This data was plotted and fitted, by a least squares,
[()TD$FIG]
406
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
Fig. 1. Human EC stem cells differentiate into neuronal-like cells. (A) Phase contrast image of stem cells in vitro. (B) Immuno-fluorescence image of stem cells positive for Oct-4 (green). (C) Phase-contrast image of hEC stem cell derived neurons at 28 days differentiation. (D) Immuno-fluorescence image of neurons positive for bIII-tubulin (red). Images are from 4 different dishes. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.) non-linear regression analysis (Graphpad Prism 4TM) to the logistic equation. Agonist reversal potentials were determined for each cell from the data line intercepting the xaxis at 0 current on the current to voltage (I–V) plot.
3. Results 3.1. Immunolabeling Undifferentiated TERA2.cl.SP12 stem cells were positive for the pluripotent stem cell marker, Oct4 (see Fig. 1a and b). Addition of RA (10 mM) induced differentiation of the EC stem cells into phenotypes with the morphological features of neurons after 15– 20 days in vitro. Cells with phase-bright cell bodies and processes (presumed dendrites and axons) emanating from the soma labeled positive for the neural marker, b-III Tubulin from day 7 of differentiation and were negative for Oct4 (see Fig. 1c and d). Notably however, pilot experiments indicated limited responses of neuronal-like cells to electrical or pharmacological stimulation up to day 28 in vitro; we therefore focused our recordings from cells 28 to 50 days of differentiation in vitro. 3.2. Voltage-gated currents Electrophysiological recordings were made from cells that were phase-bright and had neuronal (pyramidal or multi-polar) somas with neurite processes emanating from the cell body. The membrane potential recorded from a random sample of cells was 39 2 mV (n = 11); the cell membrane capacitance was 14.95 3.5 pF (n = 4) and the membrane resistance was 1.02 0.3 GV (n = 7). These values are consistent with those reported for neural cells derived from human stem cells by other groups (e.g. Westerlund et al., 2003; Moe et al., 2005). A series of experiments in which the membrane holding potential was stepped between 80 and +30 mV evoked a fast
activating, fast inactivating inward current and a slower activating, slowly deactivating outward current (see Fig. 2). Further investigation revealed that addition of TTX (0.1 mM) to the bath solution reversibly blocked the inward current consistent with activation of Na channels. The current–voltage relationship and reversible potential (which was greater than 50 mV) for the inward current was also consistent with voltage-gated sodium channels (see Fig. 2). Addition of tetraethylammonium (5 mM) reduced the outward current in keeping with activation of a K current; the current–voltage relationship was also consistent with voltagegated potassium channels. Inclusion of CsCl in the patch pipette abolished the outward (potassium) current (see Fig. 2). 3.3. Ligand-gated currents GABA is the major inhibitory neurotransmitter in the mammalian central nervous system (CNS) with a complex molecular pharmacology (e.g. Moult, 2009). Application of GABA (1–1000 mM for 1–2 s on average) to neural-like cells evoked concentration-dependent currents in 85% of cells tested, with an EC50 of 46 mM [37–59, 95% CI] and a Hill slope of 1.5 0.2 (n = 6). The GABA (10 mM) current– voltage relationship showed slight outward rectification and reversed at 2.3 3.3 mV (n = 4), consistent with the equilibrium potential of 0.8 mV for chloride ions under our recording conditions (Fig. 3). Addition of bicuculline (3 mM) or picrotoxin (10 mM) reversibly inhibited the GABA (EC20) currents to 12 1% (n = 3) and 35 3% (n = 4) of control, respectively. Conversely, the benzodiazepines diazepam (10 mM) and chlordiazepoxide potentiated the GABA currents to 132 7% (n = 3) and 161 30 (n = 5) of control, respectively. The barbiturate, pentobarbital (100 mM) and the NSAID, mefenamic acid (100 mM) reversibly potentiated the GABA currents to 229 71% (n = 4) and 250 58% (n = 3) of control, respectively (see Fig. 4). Similarly the neurosteroid allopregnanolone (0.2 mM), potentiated the GABA responses to 206 30% (n = 3) of control. These data
[()TD$FIG]
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
407
Fig. 2. Voltage-activated whole cell currents in hSC-derived neurons. (A) An example of whole-cell currents recorded from a single neuron (evoked by a series of voltage steps (shown schematically at the top) between 80 and +30 mV. (B) Addition of tetra ethyl ammonium ions (TEA, 5 mM) to the bath solution, and cesium chloride (CsCl) in the patch pipette, essentially only leaves inward (presumed INa) currents intact. (C) Tetrodotoxin (TTX, 0.1 mM), inhibits INa leaving only the outward (presumed IK) currents intact. (D) Plot of the peak inward currents against voltage step in the absence (INa, *) and presence of TTX (INa+TTX, *) and also for the peak outward currents in the absence (IK, &) and presence of TEA (IK+TEA, &). The data points are averaged from 3 to 5 cells; for clarity the S.E.M.’s are not shown.
show that the GABA current is mediated by activation of a GABAA-gated chloride channel similar to that described for native rodent and human GABAA receptors (Johnston, 2005). Glycine is also an inhibitory neurotransmitter in mammalian neurons (Hernandes and Troncone, 2009). Application of glycine (1–1000 mM) to neurons under whole-cell voltage clamp evoked concentration-dependent currents (see Fig. 4) in more than 67% of cells tested. The glycine EC50 was 51 mM [44–60, 95% CI] (n = 3) and the Hill slope was 1.6 0.2 (n = 3). The reversal potential was
approximately 2.7 mV (n = 3), close to the chloride equilibrium potential of 0.8 mV; glycine evoked currents (EC30) were also reversibly inhibited by strychnine (10 mM) to 3 3% (n = 3) of control (see Fig. 4). Together these data are similar to those reported for strychnine-sensitive glycine responses in the central nervous system (Hernandes and Troncone, 2009) (Fig. 5). Glutamate is the major excitatory neurotransmitter in the CNS (Moult, 2009; Traynelis et al., 2010). Application of glutamate (1– 1000 mM) evoked low amplitude but concentration-dependent
[()TD$FIG]
408
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
Fig. 3. GABA evokes concentration-dependent (Cl ) currents in hSC derived neurons. (A) Actual recording of GABA (3–1000 mM) currents from a single neural cell. (B) Plot of the GABA concentration–response curve showing IGABA on the y-axis, normalized (norm) to the maximal response (usually seen with 1 mM) against the GABA concentration, given on the x-axis. (C) The GABA current–voltage plot with IGABA (10 mM) on the y-axis against the holding potential (VhmV) on the x-axis. Note the GABA reversal potential is approximately 0 mV, consistent with the chloride equilibrium potential under our recording conditions. The data plotted in (B) and (C) are the mean S.E.M. from 3 to 5 different cells.
[()TD$FIG]
Fig. 4. The pharmacology of the GABA responses in stem cell-derived neurons is consistent with the expression of human GABAA receptors. Sub-maximal GABA (30 mM) currents are inhibited by (A) bicuculline (BIC, 3 mM), but potentiated by (B) diazepam (DIAZ, 10 mM) and (C) mefenamic acid (MFA, 100 mM). (D) A histogram summarizing similar experiments examining the effects of pentobarbital (PTB, 100 mM), chlordiazepoxide (CDZ, 10 mM), diazepam (DIAZ, 10 mM), mefenamic acid (MFA, 100 mM), allopregnanolone (ALLO, 0.2 mM), picrotoxin (PTX, 10 mM) and bicuculline (BIC, 3 mM) on the GABA (EC20) currents. The bars represent the mean S.E.M. of 3–5 experiments. The holding potential was 60 mV.
[()TD$FIG]
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
409
Fig. 5. Glycine evokes strychnine-sensitive currents in hSC-derived neurons. (A) The glycine concentration response curve mean (mean S.E.M., of 3 cells) and (B) The glycine (IGly, 30 mM) current voltage-curve (mean S.E.M. of 3 cells). (C) At the top, sub-maximal glycine (Gly, 30 mM) activated currents before, in the presence of and following washout of strychnine (Strych, 10 mM) from the bath solution; at the bottom is a histogram of 3 such experiments. The holding potential was 60 mV.
responses in 33% of cells tested. The glutamate EC50 was 16 mM [13–20, 95% CI] and the Hill slope was 1 0.1 (n = 7). The current to voltage relationship was approximately linear with a reversal potential of 10 4 mV (n = 4). Addition of the broad-spectrum glutamate receptor anatagonist, kyunrenic acid (1 mM), reduced
[()TD$FIG]
currents elicited by glutamate (10 mM) in a reversible manner to 14 3.5% of control (n = 3, see Fig. 6). These observations are consistent with activation of AMPA/kainate cation channels described for mammalian neurons in culture (Traynelis et al., 2010).
Fig. 6. Glutamate evokes concentration-dependent currents in hSC-derived neurons. (A) The glutamate concentration response curve (mean S.E.M. of 7 cells) and (B) the glutamate (Iglut) current–voltage curve (from 4 cells). (C) Glutamate (10 mM) currents before in the presence of and following washout of kynurenic acid (1 mM) from the bath solution; at the bottom is a histogram of 3 such experiments. The holding potential was 60 mV.
[()TD$FIG]
410
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
Fig. 7. NMDA evokes concentration-dependent currents in hSC-derived neurons. (A) The NMDA concentration response curve (mean S.E.M. from 3 cells). (B) The NMDA current is inhibited by magnesium ions (2 mM, Mg2+) when the cell is voltage-clamped at 60 mV (left) but it has little effect when the cell is voltage-clamped at +60 mV (right). (C) A histogram illustrating the results of 3 experiments where the NMDA current is plotted on the y-axis in the absence and presence of (Mg2+) in cells voltage-clamped at 60 mV and +60 mV.
NMDA-type glutamate receptors are both voltage and ligand gated ion channels and implicated in synaptic plasticity and learning and memory (Moult, 2009; Traynelis et al., 2010). Application of NMDA (1–1000 mM) to cells in a bath solution containing glycine (1 mM) but zero magnesium ions evoked responses in approximately 22% of cells sampled. The NMDA EC50 was 68 mM [56–82, 95% CI] and the Hill slope was 1.2 0.1 (n = 3). Addition of Mg2+ to the bath solution rapidly blocked the NMDA (30 mM) current in cells held at 60 mV but had little or no effect when the cells were held at +60 mV (see Fig. 7); this observation is consistent with voltage-dependent block of the NMDA channel, first described by Nowak et al. (1984) in mouse central neurons. Together, these data show that neurons derived from stem cells express AMPA/Kainate and NMDA-type glutamate ion channels. 4. Discussion Human stem cells offer enormous promise in regenerative medicine and biomedical sciences (Trounson, this issue). This study investigated some of the physiological and pharmacological properties of neurons derived from the human EC stem cell line, TERA2.cl.SP12. Our results show that these stem cells express the pluripotent marker, Oct4 and when exposed to retinoic acid, differentiated into cells with a neuronal morphology and labeled for the pan neural antigen, b-III tubulin. These data are consistent with previous studies (e.g. Przyborski et al., 2004). Depolarizing voltage steps in these neural cells evoked a biphasic current response: the fast inward component had a current–voltage relationship consistent with a sodium flux and that was blocked by TTX, supporting the notion that neural derivatives expressed voltage-gated Na channels similar to those described in CNS neurons (Hille, 2001). The outward,
slowly inactivating component was inhibited by TEA and displayed a current to voltage relationship resembling the delayed rectifier potassium current (IKDR) described in CNS neurons (Storm, 1990), although further experiments are needed to fully characterize the potassium currents. Together, however, these electrophysiological characteristics are typical of polarized, excitable cells (Hille, 2001). Moreover, in preliminary experiments, these neurons were also able to fire action potentials when electrically stimulated but few were spontaneously active, perhaps because of their low (circa 40 mV) resting membrane potentials (Shan et al., unpublished observations). Our data are therefore supportive of the development of neurons from human EC stem cells and consistent with studies of mouse (Cai et al., 2004; Ma et al., 2004; Lang et al., 2004; Pagani et al., 2006; Biella et al., 2007) and human adult stem cell derived neurons (Cho et al., 2002; Westerlund et al., 2003; Stewart et al., 2004; Moe et al., 2005). The two major inhibitory neurotransmitters in mammalian CNS, GABA and glycine, each evoked concentration-dependent currents in many of the neurons tested: GABA responses were reversibly inhibited by the GABAA antagonists, bicuculline and picrotoxin, and potentiated by the allosteric GABAA receptor modulators, chlordiazepoxide, diazepam, pentobarbital, allopregnanolone and mefenamic acid. Sensitivity to the benzodiazepines, chlordiazepoxide and diazepam, is conferred by the presence of the g2 receptor subunit in the GABAA receptor complex; similarly, sensitivity to mefenamic acid is conferred by the presence of the b2/or b3 subunits in the receptor complex (Halliwell et al., 1999; Johnston, 2005; Coyne et al., 2007). We can therefore deduce from our electrophysiological data that the GABAA receptors expressed by the hSC-derived neurons in this study were likely composed of axb2/3g2 subunits (where x is one of several possible a subunits) to enable the rich and complex pharmacological responses
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
observed to these diverse agents. The glycine-gated currents were blocked by strychnine in keeping with the native glycine chloride channel described in CNS neurons (Lynch, 2009). Taken together, our data suggest that neurons derived from (hEC) stem cells may be useful models for the investigation of human GABAA and glycine receptor pharmacology. This study also revealed that neural cells derived from human EC stem cells expressed functional glutamate and NMDA-gated cation channels. However, responses to glutamate and NMDA were much smaller in amplitude than those to GABA, observed in fewer cells, and normally only in cells differentiated for longer periods (35 days) suggesting that expression of glutamate receptors occurs later in neural development. Overall our data and that of other labs shows that, in vitro, voltage-gated K channels and GABA receptors are expressed early in neurogenesis followed by voltagegated Na channels, and later excitatory (glutamate) receptors (Westerlund et al., 2003; Lang et al., 2004; Ma et al., 2004; Moe et al., 2005; Pagani et al., 2006; Biella et al., 2007). Significantly, this pattern resembles that reported in vivo (Maric et al., 2000). Our data thus support the idea that neurons developing from stem cells in vitro recapitulate the morphological and electrophysiological steps to mature functional neurons that occurs in vivo (e.g. Ma et al., 2004; Moe et al., 2005). Consistent with our pharmacological observations, a recent study using microelectrode arrays to record from populations of hES cell-derived neurons in culture dishes, reported spontaneous spike-like activity that was inhibited by TTX and suppressed by the glutamate antagonists, CNQX and D-AP5 (Heikkila et al., 2009). In addition, adult mouse germ line stem cells have also been differentiated into functional neural cells able to generate spontaneous synaptic currents that can be blocked by the glutamate antagonists, NBQX and D-AP5 (Glaser et al., 2008). Similarly, neural cells derived from rat cortex progenitors cells grown in 3D collagen gels developed spontaneous synaptic-like currents that were sensitive to CNQX and bicuculline, consistent with the development of functional excitatory and inhibitory synapses (Ma et al., 2004). Moreover, Lang et al. (2004) reported that mouse ES-derived neurons generated spontaneous synaptic currents sensitive to CNQX and NMDA currents that were blocked in a voltage-dependent fashion by magnesium ions. Together our work and the studies outlined above therefore support the potential value of neural cells derived from human and rodent stem cells for neurophysiological and neuropharmacological studies. One additional and important corollary observation in this study was that stem cells exposed to retinoic acid labeled for the neuronal marker, b-III Tubulin (Tuj1) within 5–7 days of differentiation but our electrophysiological experiments indicated these cells had limited functional receptors or ion channels until at least 15–20 days differentiation in vitro. This (un-quantified) observation indicates a mismatch between immuno-staining and function. Several other groups using mouse and human embryonic and adult stem cells to derive neurons have also reported that neural markers (such as b-III Tubulin) do not confirm the presence of functional neurons (see Pagani et al., 2006; Westerlund et al., 2003; Moe et al., 2005;). These data therefore suggest that caution should be exerted in defining cells derived from stem cells as neurons unless or until functional data is also combined with their molecular and morphological characteristics. In conclusion, our results show the generation of neurons from human adult (tissue) stem cells with physiological and pharmacological properties similar to those reported for native neurons. These data thus support their potential value in drug discovery and neuropharmacological research. Further work however is clearly required to define the exact conditions to derive specific neuronal
411
types and to characterize the rich and complex array of receptors and ion channels expressed by human stem cell derived neurons. Acknowledgements This work was supported by grants from the Center for Alternatives to Animal Testing (CAAT); the authors would also like to thank Ms Bonnie O’Hearn for excellent administrative support. MS is the recipient of a Pharmaceutical & Chemical Sciences Program Graduate Student Scholarship. References Biella, G., Di Febo, F., Goffredo, D., Moiana, A., Taglietti, V., Conti, L., Cattaneo, E., Toselli, M., 2007. Differentiating embryonic stem-derived neural stem cells show a maturation-dependent pattern of voltage-gated sodium current expression and graded action potentials. Neuroscience 149 (1), 38–52. Cai, J., Cheng, A., Luo, Y., Lu, C., Mattson, M.P., Rao, M.S., Furukawa, K., 2004. Membrane properties of rat embryonic multipotent neural stem cells. J. Neurochem. 88 (1), 212–226. Cho, T., Bae, J.H., Choi, H.B., Kim, S.S., McLarnon, J.G., Suh-Kim, H., Kim, S.U., Min, C.K., 2002. Human neural stem cells: electrophysiological properties of voltagegated channels. Neuroreport 13 (11), 1447–1452. Coyne, L., Su, J., Patten, D., Halliwell, R.F., 2007. Characterization of the interaction between fenamates and hippocampal neuron GABA(A) receptors. Neurochem. Int. 51 (6–7), 400–406. Glaser, T., Opitz, T., Kischlat, T., Konang, R., Sasse, P., Fleischmann, B.K., Engel, W., Nayernia, K., Brustle, O, 2008. Adult germ stem cells as a source of functional neurons and glia. Stem Cells 26, 2434–2443. Halliwell, R.F., Thomas, P., Patten, D., James, C.H., Martinez-Torres, A., Miledi, R., Smart, T.G., 1999. Subunit-selective modulation of GABA-A receptors by the non-steroidal anti-inflammatory agent, mefenamic acid. Eur. J. Neurosci. 11, 2897–2905. Heikkila, T.J., Yla-Outinen, L., Tanskanen, J.M.A., Lappalainen, R.S., Skottman, H., Suuronen, R., Mikkonen, J.E., Hyttinen, J.A.K., Narkilahti, S, 2009. Human embryonic stem cell-derived neuronal cells form spontaneously active neuronal networks in vitro. Exp. Neurol. 218, 109–116. Hernandes, M.S., Troncone, LR, 2009. Glycine as a neurotransmitter in the forebrain: a short review. J. Neural Transm. 116 (12), 1551–1560. Hille, B., 2001. Ion channels of excitable membranes, 3rd ed. Sinauer Associates Inc., ISBN:0-87893r-r321-2. Hu, B.Y., Weick, J.P., Yu, J., Ma, L.X., Zhang, X.Q., Thomson, J.A., Zhang, S.C., 2010. Neural differentiation of human induced pluripotent stem cells follows developmental principles but with variable potency. Proc. Natl. Acad. Sci. U.S.A. 107 (9), 4335–4340. Johnston, G.A.R., 2005. GABAA receptor channel pharmacology. Curr. Pharm. Des. 11 (15), 1867–1885. Karumbayaram, S., Novitch, B.G., Patterson, M., Umbach, J.A., Richter, L., Lindgren, A., Conway, A.E., Clark, A.T., Goldman, S.A., Plath, K., Wiedau-Pazos, M., Kornblum, H.I., Lowry, W.E., 2009. Directed differentiation of human-induced pluripotent stem cells generates active motor neurons. Stem Cells 27 (4), 806–811. Khaira, S.K., Pouton, C.W., Haynes, J.M., 2009. P2X2 P2X4 and P2Y1 receptors elevate intracellular Ca2+ in mouse embryonic stem cell-derived GABAergic neurons. Br. J. Pharmacol. 158 (8), 1922–1931. Lang, R.J., Haynes, J.M., Kelly, J., Johnson, J., Greenhalgh, J., O’Brien, C., Mulholland, E.M., Baker, L., Munsie, M., Pouton, C.W., 2004. Electrical and neurotransmitter activity of mature neurons derived from mouse embryonic stem cells by Sox-1 lineage selection and directed differentiation. Eur. J. Neurosci. 20 (12), 3209– 3221. Lynch, J.W., 2009. Native glycine receptor subtypes and their physiological roles. Neuropharmacology 56 (1), 303–309. Ma, W., Fitzgerald, W., Liu, Q.Y., O’Shaughnessy, T.J., Maric, D., Lin, H.J., Alkon, D.L., Barker, J.L., 2004. CNS stem and progenitor cell differentiation into functional neuronal circuits in three-dimensional collagen gels. Exp. Neurol. 190 (2), 276– 288. Maric, D., Maric, I., Chang, Y., barker, JL, 2000. Stereotypical physiological properties emerge during early neuronal and glial lineage development in the embryonic rat neocortex. Cereb. Cortex 10, 729–747. Moe, M.C., Varghese, M., Danilov, A.I., Westerlund, U., Ramm-Pettersen, J., Brundin, L., Svensson, M., Berg-Johnsen, J., Langmoen, I.A., 2005. Multipotent progenitor cells from the adult human brain: neurophysiological differentiation to mature neurons. Brain 128 (Pt 9), 2189–2199. Moult, PR, 2009. Neuronal glutamate and GABA receptor function in health and disease. Biochem. Soc. Trans. 37 (Pt 6), 1317–1322. Nakagomi, T., Taguchi, A., Fujimori, Y., Saino, O., Nakano-Doi, A., Kubo, S., Gotoh, A., Soma, T., Yoshikawa, H., Nishizaki, T., Nakagomi, N., Stern, D.M., Matsuyama, T., 2009. Isolation and characterization of neural stem/progenitor cells from poststroke cerebral cortex in mice. Eur. J. Neurosci. 29 (9), 1842–1852. Nowak, L., Bregestovski, P., Ascher, P., Herbert, A., Prochiantz A, 1984. Magnesium gates glutamate-activated channels in mouse central neurones. Nature 307 (5950), 462–465.
412
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
Pagani, F., Lauro, C., Fucile, S., Catalano, M., Limatola, C., Eusebi, F., Grassi, F., 2006. Functional properties of neurons derived from fetal mouse neurospheres are compatible with those of neuronal precursors in vivo. J. Neurosci. Res. 83 (8), 1494–1501. Pouton, C.W., Haynes, J.M., 2007. Embryonic stem cells as a source of models for drug discovery. Nat. Rev. Drug Discov. 6 (8), 605–616. Przyborski, S.A., 2001. Isolation of human embryonal carcinoma stem cells by immuno-magnetic sorting. Stem Cells 19, 500–504. Przyborski, S.A., Christie, V.B., Hayman, M.W., Stewart, R., Horrocks, G.M., 2004. Human embryonal carcinoma stem cells: models of embryonic development in humans. Stem Cells Dev. 13 (4), 400–408. Razvi, E.S., Oosta, G.M., 2010. Stem cells for cellular therapy space. Drug Discovery World Spring, 37–40. Stewart, R., Coyne, L., Lako, M., Halliwell, R.F., Przyborski, S.A., 2004. Human embryonal carcinoma stem cells expressing green fluorescent protein form
functioning neurons in vitro: a research tool for co-culture studies. Stem Cells Dev. 13 (6), 646–657. Storm, J.F., 1990. Potassium currents in hippocampal pyramidal cells. Prog. Brain Res. 83, 161–187. Traynelis, S.F., Wollmuth, L.P., McBain, C.J., Menniti, F.S., Vance, K.M., Ogden, K.K., Hansen, K.B., Yuan, H., Myers, S.J., Dingledine, R., Sibley, D., 2010. Glutamate receptor ion channels: regulation and function. Pharmacol. Rev. 62 (3), 405– 496. Ulrich, H., Majumder, P., 2006. Neurotransmitter receptor expression and activity during neuronal differentiation of embryonal carcinoma and stem cells: from basic research towards clinical applications. Cell Prolif. 39 (4), 281–300. Westerlund, U., Moe, M.C., Varghese, M., Berg-Johnsen, J., Ohlsson, M., Langmoen, I.A., Svensson, M., 2003. Stem cells from the adult human brain develop into functional neurons in culture. Exp. Cell Res. 289 (2), 378–383.
Contents lists available at ScienceDirect
Neurochemistry International journal homepage: www.elsevier.com/locate/neuint
Neuropharmacological properties of neurons derived from human stem cells Leanne Coyne a,1, Mu Shan a, Stefan A. Przyborski b, Ryoko Hirakawa a, Robert F. Halliwell a,* a b
School of Pharmacy, University of the Pacific, Stockton, CA, USA School of Biological & Biomedical Sciences, University of Durham, England, UK
A R T I C L E I N F O
A B S T R A C T
Article history: Received 15 October 2010 Accepted 19 January 2011 Available online 16 February 2011
Human pluripotent stem cells have enormous potential value in neuropharmacology and drug discovery yet there is little data on the major classes and properties of receptors and ion channels expressed by neurons derived from these stem cells. Recent studies in this lab have therefore used conventional patchclamp electrophysiology to investigate the pharmacological properties of the ligand and voltage-gated ion channels in neurons derived and maintained in vitro from the human stem cell (hSC) line, TERA2.cl.SP12. TERA2.cl.SP12 stem cells were differentiated with retinoic acid and used in electrophysiological experiments 28–50 days after beginning differentiation. HSC-derived neurons generated large whole cell currents with depolarizing voltage steps ( 80 to 30 mV) comprised of an inward, rapidly inactivating component and a delayed, slowly deactivating outward component. The fast inward current was blocked by the sodium channel blocker tetrodotoxin (0.1 mM) and the outward currents were significantly reduced by tetraethylammonium ions (TEA, 5 mM) consistent with the presence of functional Na and K ion channels. Application of the inhibitory neurotransmitters, GABA (0.1–1000 mM) or glycine (0.1–1000 mM) evoked concentration dependent currents. The GABA currents were inhibited by the convulsants, picrotoxin (10 mM) and bicuculline (3 mM), potentiated by the NSAID mefenamic acid (10–100 mM), the general anaesthetic pentobarbital (100 mM), the neurosteroid allopregnanolone and the anxiolytics chlordiazepoxide (10 mM) and diazepam (10 mM) all consistent with the expression of GABAA receptors. Responses to glycine were reversibly blocked by strychnine (10 mM) consistent with glycine-gated chloride channels. The excitatory agonists, glutamate (1–1000 mM) and NMDA (1– 1000 mM) activated concentration-dependent responses from hSC-derived neurons. Glutamate currents were inhibited by kynurenic acid (1 mM) and NMDA responses were blocked by MgCl2 (2 mM) in a highly voltage-dependent manner. Together, these findings show that neurons derived from human stem cells develop an array of functional receptors and ion channels with a pharmacological profile in keeping with that described for native neurons. This study therefore provides support for the hypothesis that stem cells may provide a powerful source of human neurons for future neuropharmacological studies. ß 2011 Elsevier Ltd. All rights reserved.
Keywords: Human neurons Stem cells Voltage-gated ion channels Ligand-gated receptors GABA Glutamate
1. Introduction Stem cells offer enormous potential value in regenerative medicine and drug discovery yet the number of published studies at this time utilizing stem cells in drug discovery, toxicity screening & high-content screening is exceedingly small (Razvi and Oosta, 2010). The paucity of data in these areas, in part, reflects the complex and time-consuming requirement to determine the functional proper-
* Corresponding author at: Department of Physiology & Pharmacology, TJ Long School of Pharmacy & Health Sciences, University of the Pacific, Stockton, CA 95211, USA. Tel.: +1 209 946 2074; fax: +1 209 946 2857. E-mail address: rhalliwell@pacific.edu (R.F. Halliwell). 1 Present address: California Northstate College of Pharmacy, 10811 International Drive, Rancho Cordova, CA 95670, USA. 0197-0186/$ – see front matter ß 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuint.2011.01.022
ties of the phenotypes derived from stem cells if they are to be used in pharmacological, toxicological or drug discovery studies. Over the past decade, some progress has been made in delineating the physiological and pharmacological properties of neurons derived from rodent stem cells. Several studies, for example, have reported that neurons from rodent embryonic stem (ES) cells develop voltage-activated calcium, sodium, and potassium currents and display spontaneous synaptic currents that are blocked by GABA or glutamate antagonists (Cai et al., 2004; Ma et al., 2004; Lang et al., 2004; Pagani et al., 2006; Biella et al., 2007; Nakagomi et al., 2009). Such cells also show rises in intracellular calcium in response to several neurotransmitters, including acetylcholine, N-methyl-D-aspartate (NMDA) and adenosine triphosphate (ATP) (Cai et al., 2004; Ma et al., 2004; Lang et al., 2004; Pagani et al., 2006). In a more recent study, Khaira et al. (2009) have
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
also demonstrated that mouse ES derived ‘GABAergic’ neurons elevate intracellular Ca2+ levels through activation of P2X2, P2X4 and P2Y1-like purinoceptors expressed in these cells. Functional studies of human stem cell (hSC) derived neurons, in contrast, are still in their infancy (Pouton and Haynes, 2007). Cho et al. (2002) for example, reported that an immortalized clonal stem cell line HB1.F3, derived from human embryonic telencephalon, differentiated into neurons expressing voltagegated potassium and sodium channels. Westerlund et al. (2003) later reported that multipotent neural stem cells resected from patients undergoing neurosurgery for epilepsy could be grown as neurospheres and subsequently differentiated to neurons and glia. The neurons developed functional IK and IA-like potassium channels and TTX-sensitive sodium channels after about 3 weeks in culture. Moe et al. (2005) similarly described the isolation and culture of multipotent progenitor cells from the temporal cortex of patients with intractable epilepsy. These cells could be grown as neurospheres in vitro, passaged and subsequently differentiated to glia and neurons. Electrophysiological recordings showed that the neurons were able to generate TTX-sensitive action potentials and spontaneous synaptic activity that was sensitive to GABAA and glutamate receptor antagonists. Human induced pluripotent stem cells (iPS) have also recently generated significant interest as sources of human neurons for neuropharmacological studies and for regenerative medicine. In support of this expectation, two recent studies have indicated that human iPS cells can differentiate into spinal-like (HB9/ISL1/2+) motor neurons able to fire action potentials and generate some spontaneous synaptic activity after 5–8 weeks in vitro (Hu et al., 2010; Karumbayaram et al., 2009). At the present time, however, there is little data on the pharmacology of the ion channels and receptors expressed by any hSC derived neurons. This study has therefore begun to explore the pharmacological properties of the ligand and voltage-activated currents recorded from neurons derived from human stem cells using patch-clamp techniques. We have utilized the human embryonal carcinoma (hEC) clonal stem cell line TERA2.cl.SP12 which displays an enhanced propensity to form neurons and glia and which express several molecular markers and several functional receptors and ion channels similar to those described in native neurons, making them an excellent experimental model (Stewart et al., 2004; Przyborski, 2001; Ulrich and Majumder, 2006). 2. Methods 2.1. Cell culture TERA2.cl.SP12 EC stem cells were prepared, maintained and differentiated according to methods previously described (e.g. Stewart et al., 2004). Briefly, stem cells were maintained in 75 ml tissue culture flasks at 37 8C, 5% CO2, 100% relative humidity in DMEM supplemented with foetal bovine serum (10%, v/v; Invitrogen), L-glutamine (2 mM final concentration; Sigma–Aldrich) and penicillin–streptomycin (100 U/ml/100 mg/ml). When cells reached approximately 80% confluence, they were detached from the flask base using micro-glass beads (Fisher Scientific), removed and differentiated either as monolayers or suspension aggregates. For monolayer cultures, stem cells were re-seeded into 35 mm tissue culture dishes (at a density of 2 104 cells/cm2) on sterile glass coverslips pre-coated with poly-Dlysine (10 mg/ml). Retinoic acid (10 mM Sigma–Aldrich) was now also included in the media. After 7 days, the mitotic inhibitors, cytosine-D-arabinoside (1 mM), fluoroxyuridine (10 mM) and uridine (10 mM) (all Sigma–Aldrich) were included in the culture media for a further 7 days. For suspension aggregates 1.25 106 stem cells were seeded into sterile 90 mm bacteriological Petri dishes and cultured in retinoic acid (RA) containing media, as described above. Cells were re-seeded every 3–4 days into new Petri dishes with fresh RA media. Cell aggregates were then harvested after 28 days differentiation, washed and plated onto sterile glass coverslips pre-coated with poly-D-lysine in 35 mm culture dishes. Cell aggregates were maintained on coverslips for a further 7–14 days in DMEM medium in the absence of retinoic acid but in the presence of the mitotic inhibitors (for the first 7 days only).
405
2.2. Immunocytochemistry For immunolabeling, stem cells and their neural derivates were cultured on sterile glass coverslips in 35 mm culture dishes, as described above. All cells were first washed in PBS and then fixed with 4% paraformaldehyde in PBS. The fixative was carefully removed, and the cells rinsed with PBS and placed overnight (at 4 8C) in a blocking solution, consisting of 5% normal donkey serum, 0.3% triton X-100 in PBS. Thereafter, cells were placed in primary antibodies (from Millipore) diluted in blocking solution and incubated overnight at 4 8C. The primary antibody concentrations used were mouse anti-Oct-4, 1:100, and mouse anti-b-III Tubulin 1:500. 24 h later, cells were washed with PBS and also twice with the blocking solution. At the completion of the final wash, cells were left in blocking solution for a further 30 min. Thereafter, cells were incubated with donkey anti-mouse IgG (Cy3-or FITC-conjugated) secondary antibody (Millipore) for 2 h at 1:250 dilution at room temperature and then washed with PBS and counter-stained with DAPI (100 ng/ml; Sigma–Aldrich) before mounting over chamber slides using anti-fade mounting solution. (Prolong1 Gold Antifade Reagent; Invitrogen). Immunolabeled cells were visualized through an inverted fluorescent microscope (Nikon Eclipse TE200) with an attached CCD camera (Microfire; Optronics). Fluorescence was observed using filter sets appropriate for each label. The percentage of immunofluorescent cells was determined by counting DAPI-positive cells and then immuno-positive for b-III Tubulin plus DAPI in 3–6 fields (at 50 magnification) across the slide. Data was gathered from 3 independently replicated experiments. 2.3. Electrophysiology Agonist and voltage-evoked currents were recorded from cells with a neuronallike morphology (phase bright with neurite processes), differentiated for 28– 50 days, using the whole-cell configuration of the patch-clamp technique as described previously (e.g. Coyne et al., 2007). Briefly, patch electrodes were made from borosilicate glass pipettes (Harvard Apparatus, Holliston, MA, USA) on a Narishige PB-7 electrode puller (East Meadow, NY, USA) and had tip resistances of 1–4 MV when filled with internal solution. Currents were recorded using an Axopatch 200B amplifier and headstage (Axon Instruments, Foster City, USA) and low-pass filtered at 10 kHz before digitization via a National Instruments DAQ card and a National Instruments BNC-2090 interface board (Austin, TX, USA) and then stored on a PC running WinWCP software (University of Strathclyde, UK). Whole cell currents were continuously monitored on the desktop computer. Series resistance and pipette and whole-cell capacitance were cancelled electronically. Cells were perfused with a bath solution containing the following (in mM): NaCl (140.0), KCl (5), MgCl2 (2.0), CaCl2 (1.0), HEPES (10.0). The internal solution contained the following (in mM): CsCl or KCl (140.0), MgCl2 (2.0), EGTA (11), HEPES (10), Mg-ATP (2.0). All solutions were titrated to pH 7.2 and internal solutions were filtered before use using 0.2 mm disposable filters (Whatman, Maidstone, England). All electrophysiology experiments were carried out at ambient room temperature (23–25 8C). Neural derivatives were voltage-clamped at a holding potential of 60 mV, unless otherwise stated. For determination of agonist-evoked current-to-voltage relationships, the membrane potential was stepped between 140 mV and 60 mV in 20 mV increments. To activate voltage-gated currents, 20 ms voltage steps were applied from 80 to 30 mV in 10 mV increments from a holding potential of 80 mV. For recording potassium currents, CsCl was replaced with KCl (140 mM) in the internal solution. All bath solutions for studying voltage-activated currents also contained 2 mM CoCl2, to prevent possible calcium channel activation. 2.4. Drugs and their application All drugs were obtained from Sigma–Aldrich (St. Louis, MO, USA) except TTX which was purchased from Tocris Bioscience (Ellisville, MO, USA). Stock solutions of bicuculline methyl bromide was made up in bath solution; pentobarbital-Na was dissolved in H2O; mefenamic acid was dissolved in NaOH (1 M), picrotoxin was dissolved in ethanol; chlordiazepoxide, diazepam and allopregnanolone were first dissolved in 100% dimethylsulphoxide (DMSO). The final bath concentrations of these solvents did not exceed 0.1% of the recording solutions and in control experiments had no effects on the responses. Agonists and other drugs were applied directly to cells under voltage-clamp from the tip of a 250 mm pipette connected to a Y-tube (fabricated in-house). Fresh bath solution was also perfused through the bath (at 1–2 ml/min) using a gravity-feed system (also manufactured in-house) to ensure that there was no build up of drug solutions in the bath. At least 3 control responses were recorded before the addition of agonists. Drugs were washed off once a clear asymptotic effect was observed. Control responses were re-established before further drug applications. 2.5. Data analysis Agonist- and voltage-evoked currents were measured at their peak amplitude. Current responses in the presence of drugs are expressed as a percentage of the control response (S.E.M. of n experiments). For concentration–response data, agonist-evoked currents were normalized to the maximal response evoked by a saturating concentration of agonist. This data was plotted and fitted, by a least squares,
[()TD$FIG]
406
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
Fig. 1. Human EC stem cells differentiate into neuronal-like cells. (A) Phase contrast image of stem cells in vitro. (B) Immuno-fluorescence image of stem cells positive for Oct-4 (green). (C) Phase-contrast image of hEC stem cell derived neurons at 28 days differentiation. (D) Immuno-fluorescence image of neurons positive for bIII-tubulin (red). Images are from 4 different dishes. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.) non-linear regression analysis (Graphpad Prism 4TM) to the logistic equation. Agonist reversal potentials were determined for each cell from the data line intercepting the xaxis at 0 current on the current to voltage (I–V) plot.
3. Results 3.1. Immunolabeling Undifferentiated TERA2.cl.SP12 stem cells were positive for the pluripotent stem cell marker, Oct4 (see Fig. 1a and b). Addition of RA (10 mM) induced differentiation of the EC stem cells into phenotypes with the morphological features of neurons after 15– 20 days in vitro. Cells with phase-bright cell bodies and processes (presumed dendrites and axons) emanating from the soma labeled positive for the neural marker, b-III Tubulin from day 7 of differentiation and were negative for Oct4 (see Fig. 1c and d). Notably however, pilot experiments indicated limited responses of neuronal-like cells to electrical or pharmacological stimulation up to day 28 in vitro; we therefore focused our recordings from cells 28 to 50 days of differentiation in vitro. 3.2. Voltage-gated currents Electrophysiological recordings were made from cells that were phase-bright and had neuronal (pyramidal or multi-polar) somas with neurite processes emanating from the cell body. The membrane potential recorded from a random sample of cells was 39 2 mV (n = 11); the cell membrane capacitance was 14.95 3.5 pF (n = 4) and the membrane resistance was 1.02 0.3 GV (n = 7). These values are consistent with those reported for neural cells derived from human stem cells by other groups (e.g. Westerlund et al., 2003; Moe et al., 2005). A series of experiments in which the membrane holding potential was stepped between 80 and +30 mV evoked a fast
activating, fast inactivating inward current and a slower activating, slowly deactivating outward current (see Fig. 2). Further investigation revealed that addition of TTX (0.1 mM) to the bath solution reversibly blocked the inward current consistent with activation of Na channels. The current–voltage relationship and reversible potential (which was greater than 50 mV) for the inward current was also consistent with voltage-gated sodium channels (see Fig. 2). Addition of tetraethylammonium (5 mM) reduced the outward current in keeping with activation of a K current; the current–voltage relationship was also consistent with voltagegated potassium channels. Inclusion of CsCl in the patch pipette abolished the outward (potassium) current (see Fig. 2). 3.3. Ligand-gated currents GABA is the major inhibitory neurotransmitter in the mammalian central nervous system (CNS) with a complex molecular pharmacology (e.g. Moult, 2009). Application of GABA (1–1000 mM for 1–2 s on average) to neural-like cells evoked concentration-dependent currents in 85% of cells tested, with an EC50 of 46 mM [37–59, 95% CI] and a Hill slope of 1.5 0.2 (n = 6). The GABA (10 mM) current– voltage relationship showed slight outward rectification and reversed at 2.3 3.3 mV (n = 4), consistent with the equilibrium potential of 0.8 mV for chloride ions under our recording conditions (Fig. 3). Addition of bicuculline (3 mM) or picrotoxin (10 mM) reversibly inhibited the GABA (EC20) currents to 12 1% (n = 3) and 35 3% (n = 4) of control, respectively. Conversely, the benzodiazepines diazepam (10 mM) and chlordiazepoxide potentiated the GABA currents to 132 7% (n = 3) and 161 30 (n = 5) of control, respectively. The barbiturate, pentobarbital (100 mM) and the NSAID, mefenamic acid (100 mM) reversibly potentiated the GABA currents to 229 71% (n = 4) and 250 58% (n = 3) of control, respectively (see Fig. 4). Similarly the neurosteroid allopregnanolone (0.2 mM), potentiated the GABA responses to 206 30% (n = 3) of control. These data
[()TD$FIG]
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
407
Fig. 2. Voltage-activated whole cell currents in hSC-derived neurons. (A) An example of whole-cell currents recorded from a single neuron (evoked by a series of voltage steps (shown schematically at the top) between 80 and +30 mV. (B) Addition of tetra ethyl ammonium ions (TEA, 5 mM) to the bath solution, and cesium chloride (CsCl) in the patch pipette, essentially only leaves inward (presumed INa) currents intact. (C) Tetrodotoxin (TTX, 0.1 mM), inhibits INa leaving only the outward (presumed IK) currents intact. (D) Plot of the peak inward currents against voltage step in the absence (INa, *) and presence of TTX (INa+TTX, *) and also for the peak outward currents in the absence (IK, &) and presence of TEA (IK+TEA, &). The data points are averaged from 3 to 5 cells; for clarity the S.E.M.’s are not shown.
show that the GABA current is mediated by activation of a GABAA-gated chloride channel similar to that described for native rodent and human GABAA receptors (Johnston, 2005). Glycine is also an inhibitory neurotransmitter in mammalian neurons (Hernandes and Troncone, 2009). Application of glycine (1–1000 mM) to neurons under whole-cell voltage clamp evoked concentration-dependent currents (see Fig. 4) in more than 67% of cells tested. The glycine EC50 was 51 mM [44–60, 95% CI] (n = 3) and the Hill slope was 1.6 0.2 (n = 3). The reversal potential was
approximately 2.7 mV (n = 3), close to the chloride equilibrium potential of 0.8 mV; glycine evoked currents (EC30) were also reversibly inhibited by strychnine (10 mM) to 3 3% (n = 3) of control (see Fig. 4). Together these data are similar to those reported for strychnine-sensitive glycine responses in the central nervous system (Hernandes and Troncone, 2009) (Fig. 5). Glutamate is the major excitatory neurotransmitter in the CNS (Moult, 2009; Traynelis et al., 2010). Application of glutamate (1– 1000 mM) evoked low amplitude but concentration-dependent
[()TD$FIG]
408
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
Fig. 3. GABA evokes concentration-dependent (Cl ) currents in hSC derived neurons. (A) Actual recording of GABA (3–1000 mM) currents from a single neural cell. (B) Plot of the GABA concentration–response curve showing IGABA on the y-axis, normalized (norm) to the maximal response (usually seen with 1 mM) against the GABA concentration, given on the x-axis. (C) The GABA current–voltage plot with IGABA (10 mM) on the y-axis against the holding potential (VhmV) on the x-axis. Note the GABA reversal potential is approximately 0 mV, consistent with the chloride equilibrium potential under our recording conditions. The data plotted in (B) and (C) are the mean S.E.M. from 3 to 5 different cells.
[()TD$FIG]
Fig. 4. The pharmacology of the GABA responses in stem cell-derived neurons is consistent with the expression of human GABAA receptors. Sub-maximal GABA (30 mM) currents are inhibited by (A) bicuculline (BIC, 3 mM), but potentiated by (B) diazepam (DIAZ, 10 mM) and (C) mefenamic acid (MFA, 100 mM). (D) A histogram summarizing similar experiments examining the effects of pentobarbital (PTB, 100 mM), chlordiazepoxide (CDZ, 10 mM), diazepam (DIAZ, 10 mM), mefenamic acid (MFA, 100 mM), allopregnanolone (ALLO, 0.2 mM), picrotoxin (PTX, 10 mM) and bicuculline (BIC, 3 mM) on the GABA (EC20) currents. The bars represent the mean S.E.M. of 3–5 experiments. The holding potential was 60 mV.
[()TD$FIG]
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
409
Fig. 5. Glycine evokes strychnine-sensitive currents in hSC-derived neurons. (A) The glycine concentration response curve mean (mean S.E.M., of 3 cells) and (B) The glycine (IGly, 30 mM) current voltage-curve (mean S.E.M. of 3 cells). (C) At the top, sub-maximal glycine (Gly, 30 mM) activated currents before, in the presence of and following washout of strychnine (Strych, 10 mM) from the bath solution; at the bottom is a histogram of 3 such experiments. The holding potential was 60 mV.
responses in 33% of cells tested. The glutamate EC50 was 16 mM [13–20, 95% CI] and the Hill slope was 1 0.1 (n = 7). The current to voltage relationship was approximately linear with a reversal potential of 10 4 mV (n = 4). Addition of the broad-spectrum glutamate receptor anatagonist, kyunrenic acid (1 mM), reduced
[()TD$FIG]
currents elicited by glutamate (10 mM) in a reversible manner to 14 3.5% of control (n = 3, see Fig. 6). These observations are consistent with activation of AMPA/kainate cation channels described for mammalian neurons in culture (Traynelis et al., 2010).
Fig. 6. Glutamate evokes concentration-dependent currents in hSC-derived neurons. (A) The glutamate concentration response curve (mean S.E.M. of 7 cells) and (B) the glutamate (Iglut) current–voltage curve (from 4 cells). (C) Glutamate (10 mM) currents before in the presence of and following washout of kynurenic acid (1 mM) from the bath solution; at the bottom is a histogram of 3 such experiments. The holding potential was 60 mV.
[()TD$FIG]
410
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
Fig. 7. NMDA evokes concentration-dependent currents in hSC-derived neurons. (A) The NMDA concentration response curve (mean S.E.M. from 3 cells). (B) The NMDA current is inhibited by magnesium ions (2 mM, Mg2+) when the cell is voltage-clamped at 60 mV (left) but it has little effect when the cell is voltage-clamped at +60 mV (right). (C) A histogram illustrating the results of 3 experiments where the NMDA current is plotted on the y-axis in the absence and presence of (Mg2+) in cells voltage-clamped at 60 mV and +60 mV.
NMDA-type glutamate receptors are both voltage and ligand gated ion channels and implicated in synaptic plasticity and learning and memory (Moult, 2009; Traynelis et al., 2010). Application of NMDA (1–1000 mM) to cells in a bath solution containing glycine (1 mM) but zero magnesium ions evoked responses in approximately 22% of cells sampled. The NMDA EC50 was 68 mM [56–82, 95% CI] and the Hill slope was 1.2 0.1 (n = 3). Addition of Mg2+ to the bath solution rapidly blocked the NMDA (30 mM) current in cells held at 60 mV but had little or no effect when the cells were held at +60 mV (see Fig. 7); this observation is consistent with voltage-dependent block of the NMDA channel, first described by Nowak et al. (1984) in mouse central neurons. Together, these data show that neurons derived from stem cells express AMPA/Kainate and NMDA-type glutamate ion channels. 4. Discussion Human stem cells offer enormous promise in regenerative medicine and biomedical sciences (Trounson, this issue). This study investigated some of the physiological and pharmacological properties of neurons derived from the human EC stem cell line, TERA2.cl.SP12. Our results show that these stem cells express the pluripotent marker, Oct4 and when exposed to retinoic acid, differentiated into cells with a neuronal morphology and labeled for the pan neural antigen, b-III tubulin. These data are consistent with previous studies (e.g. Przyborski et al., 2004). Depolarizing voltage steps in these neural cells evoked a biphasic current response: the fast inward component had a current–voltage relationship consistent with a sodium flux and that was blocked by TTX, supporting the notion that neural derivatives expressed voltage-gated Na channels similar to those described in CNS neurons (Hille, 2001). The outward,
slowly inactivating component was inhibited by TEA and displayed a current to voltage relationship resembling the delayed rectifier potassium current (IKDR) described in CNS neurons (Storm, 1990), although further experiments are needed to fully characterize the potassium currents. Together, however, these electrophysiological characteristics are typical of polarized, excitable cells (Hille, 2001). Moreover, in preliminary experiments, these neurons were also able to fire action potentials when electrically stimulated but few were spontaneously active, perhaps because of their low (circa 40 mV) resting membrane potentials (Shan et al., unpublished observations). Our data are therefore supportive of the development of neurons from human EC stem cells and consistent with studies of mouse (Cai et al., 2004; Ma et al., 2004; Lang et al., 2004; Pagani et al., 2006; Biella et al., 2007) and human adult stem cell derived neurons (Cho et al., 2002; Westerlund et al., 2003; Stewart et al., 2004; Moe et al., 2005). The two major inhibitory neurotransmitters in mammalian CNS, GABA and glycine, each evoked concentration-dependent currents in many of the neurons tested: GABA responses were reversibly inhibited by the GABAA antagonists, bicuculline and picrotoxin, and potentiated by the allosteric GABAA receptor modulators, chlordiazepoxide, diazepam, pentobarbital, allopregnanolone and mefenamic acid. Sensitivity to the benzodiazepines, chlordiazepoxide and diazepam, is conferred by the presence of the g2 receptor subunit in the GABAA receptor complex; similarly, sensitivity to mefenamic acid is conferred by the presence of the b2/or b3 subunits in the receptor complex (Halliwell et al., 1999; Johnston, 2005; Coyne et al., 2007). We can therefore deduce from our electrophysiological data that the GABAA receptors expressed by the hSC-derived neurons in this study were likely composed of axb2/3g2 subunits (where x is one of several possible a subunits) to enable the rich and complex pharmacological responses
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
observed to these diverse agents. The glycine-gated currents were blocked by strychnine in keeping with the native glycine chloride channel described in CNS neurons (Lynch, 2009). Taken together, our data suggest that neurons derived from (hEC) stem cells may be useful models for the investigation of human GABAA and glycine receptor pharmacology. This study also revealed that neural cells derived from human EC stem cells expressed functional glutamate and NMDA-gated cation channels. However, responses to glutamate and NMDA were much smaller in amplitude than those to GABA, observed in fewer cells, and normally only in cells differentiated for longer periods (35 days) suggesting that expression of glutamate receptors occurs later in neural development. Overall our data and that of other labs shows that, in vitro, voltage-gated K channels and GABA receptors are expressed early in neurogenesis followed by voltagegated Na channels, and later excitatory (glutamate) receptors (Westerlund et al., 2003; Lang et al., 2004; Ma et al., 2004; Moe et al., 2005; Pagani et al., 2006; Biella et al., 2007). Significantly, this pattern resembles that reported in vivo (Maric et al., 2000). Our data thus support the idea that neurons developing from stem cells in vitro recapitulate the morphological and electrophysiological steps to mature functional neurons that occurs in vivo (e.g. Ma et al., 2004; Moe et al., 2005). Consistent with our pharmacological observations, a recent study using microelectrode arrays to record from populations of hES cell-derived neurons in culture dishes, reported spontaneous spike-like activity that was inhibited by TTX and suppressed by the glutamate antagonists, CNQX and D-AP5 (Heikkila et al., 2009). In addition, adult mouse germ line stem cells have also been differentiated into functional neural cells able to generate spontaneous synaptic currents that can be blocked by the glutamate antagonists, NBQX and D-AP5 (Glaser et al., 2008). Similarly, neural cells derived from rat cortex progenitors cells grown in 3D collagen gels developed spontaneous synaptic-like currents that were sensitive to CNQX and bicuculline, consistent with the development of functional excitatory and inhibitory synapses (Ma et al., 2004). Moreover, Lang et al. (2004) reported that mouse ES-derived neurons generated spontaneous synaptic currents sensitive to CNQX and NMDA currents that were blocked in a voltage-dependent fashion by magnesium ions. Together our work and the studies outlined above therefore support the potential value of neural cells derived from human and rodent stem cells for neurophysiological and neuropharmacological studies. One additional and important corollary observation in this study was that stem cells exposed to retinoic acid labeled for the neuronal marker, b-III Tubulin (Tuj1) within 5–7 days of differentiation but our electrophysiological experiments indicated these cells had limited functional receptors or ion channels until at least 15–20 days differentiation in vitro. This (un-quantified) observation indicates a mismatch between immuno-staining and function. Several other groups using mouse and human embryonic and adult stem cells to derive neurons have also reported that neural markers (such as b-III Tubulin) do not confirm the presence of functional neurons (see Pagani et al., 2006; Westerlund et al., 2003; Moe et al., 2005;). These data therefore suggest that caution should be exerted in defining cells derived from stem cells as neurons unless or until functional data is also combined with their molecular and morphological characteristics. In conclusion, our results show the generation of neurons from human adult (tissue) stem cells with physiological and pharmacological properties similar to those reported for native neurons. These data thus support their potential value in drug discovery and neuropharmacological research. Further work however is clearly required to define the exact conditions to derive specific neuronal
411
types and to characterize the rich and complex array of receptors and ion channels expressed by human stem cell derived neurons. Acknowledgements This work was supported by grants from the Center for Alternatives to Animal Testing (CAAT); the authors would also like to thank Ms Bonnie O’Hearn for excellent administrative support. MS is the recipient of a Pharmaceutical & Chemical Sciences Program Graduate Student Scholarship. References Biella, G., Di Febo, F., Goffredo, D., Moiana, A., Taglietti, V., Conti, L., Cattaneo, E., Toselli, M., 2007. Differentiating embryonic stem-derived neural stem cells show a maturation-dependent pattern of voltage-gated sodium current expression and graded action potentials. Neuroscience 149 (1), 38–52. Cai, J., Cheng, A., Luo, Y., Lu, C., Mattson, M.P., Rao, M.S., Furukawa, K., 2004. Membrane properties of rat embryonic multipotent neural stem cells. J. Neurochem. 88 (1), 212–226. Cho, T., Bae, J.H., Choi, H.B., Kim, S.S., McLarnon, J.G., Suh-Kim, H., Kim, S.U., Min, C.K., 2002. Human neural stem cells: electrophysiological properties of voltagegated channels. Neuroreport 13 (11), 1447–1452. Coyne, L., Su, J., Patten, D., Halliwell, R.F., 2007. Characterization of the interaction between fenamates and hippocampal neuron GABA(A) receptors. Neurochem. Int. 51 (6–7), 400–406. Glaser, T., Opitz, T., Kischlat, T., Konang, R., Sasse, P., Fleischmann, B.K., Engel, W., Nayernia, K., Brustle, O, 2008. Adult germ stem cells as a source of functional neurons and glia. Stem Cells 26, 2434–2443. Halliwell, R.F., Thomas, P., Patten, D., James, C.H., Martinez-Torres, A., Miledi, R., Smart, T.G., 1999. Subunit-selective modulation of GABA-A receptors by the non-steroidal anti-inflammatory agent, mefenamic acid. Eur. J. Neurosci. 11, 2897–2905. Heikkila, T.J., Yla-Outinen, L., Tanskanen, J.M.A., Lappalainen, R.S., Skottman, H., Suuronen, R., Mikkonen, J.E., Hyttinen, J.A.K., Narkilahti, S, 2009. Human embryonic stem cell-derived neuronal cells form spontaneously active neuronal networks in vitro. Exp. Neurol. 218, 109–116. Hernandes, M.S., Troncone, LR, 2009. Glycine as a neurotransmitter in the forebrain: a short review. J. Neural Transm. 116 (12), 1551–1560. Hille, B., 2001. Ion channels of excitable membranes, 3rd ed. Sinauer Associates Inc., ISBN:0-87893r-r321-2. Hu, B.Y., Weick, J.P., Yu, J., Ma, L.X., Zhang, X.Q., Thomson, J.A., Zhang, S.C., 2010. Neural differentiation of human induced pluripotent stem cells follows developmental principles but with variable potency. Proc. Natl. Acad. Sci. U.S.A. 107 (9), 4335–4340. Johnston, G.A.R., 2005. GABAA receptor channel pharmacology. Curr. Pharm. Des. 11 (15), 1867–1885. Karumbayaram, S., Novitch, B.G., Patterson, M., Umbach, J.A., Richter, L., Lindgren, A., Conway, A.E., Clark, A.T., Goldman, S.A., Plath, K., Wiedau-Pazos, M., Kornblum, H.I., Lowry, W.E., 2009. Directed differentiation of human-induced pluripotent stem cells generates active motor neurons. Stem Cells 27 (4), 806–811. Khaira, S.K., Pouton, C.W., Haynes, J.M., 2009. P2X2 P2X4 and P2Y1 receptors elevate intracellular Ca2+ in mouse embryonic stem cell-derived GABAergic neurons. Br. J. Pharmacol. 158 (8), 1922–1931. Lang, R.J., Haynes, J.M., Kelly, J., Johnson, J., Greenhalgh, J., O’Brien, C., Mulholland, E.M., Baker, L., Munsie, M., Pouton, C.W., 2004. Electrical and neurotransmitter activity of mature neurons derived from mouse embryonic stem cells by Sox-1 lineage selection and directed differentiation. Eur. J. Neurosci. 20 (12), 3209– 3221. Lynch, J.W., 2009. Native glycine receptor subtypes and their physiological roles. Neuropharmacology 56 (1), 303–309. Ma, W., Fitzgerald, W., Liu, Q.Y., O’Shaughnessy, T.J., Maric, D., Lin, H.J., Alkon, D.L., Barker, J.L., 2004. CNS stem and progenitor cell differentiation into functional neuronal circuits in three-dimensional collagen gels. Exp. Neurol. 190 (2), 276– 288. Maric, D., Maric, I., Chang, Y., barker, JL, 2000. Stereotypical physiological properties emerge during early neuronal and glial lineage development in the embryonic rat neocortex. Cereb. Cortex 10, 729–747. Moe, M.C., Varghese, M., Danilov, A.I., Westerlund, U., Ramm-Pettersen, J., Brundin, L., Svensson, M., Berg-Johnsen, J., Langmoen, I.A., 2005. Multipotent progenitor cells from the adult human brain: neurophysiological differentiation to mature neurons. Brain 128 (Pt 9), 2189–2199. Moult, PR, 2009. Neuronal glutamate and GABA receptor function in health and disease. Biochem. Soc. Trans. 37 (Pt 6), 1317–1322. Nakagomi, T., Taguchi, A., Fujimori, Y., Saino, O., Nakano-Doi, A., Kubo, S., Gotoh, A., Soma, T., Yoshikawa, H., Nishizaki, T., Nakagomi, N., Stern, D.M., Matsuyama, T., 2009. Isolation and characterization of neural stem/progenitor cells from poststroke cerebral cortex in mice. Eur. J. Neurosci. 29 (9), 1842–1852. Nowak, L., Bregestovski, P., Ascher, P., Herbert, A., Prochiantz A, 1984. Magnesium gates glutamate-activated channels in mouse central neurones. Nature 307 (5950), 462–465.
412
L. Coyne et al. / Neurochemistry International 59 (2011) 404–412
Pagani, F., Lauro, C., Fucile, S., Catalano, M., Limatola, C., Eusebi, F., Grassi, F., 2006. Functional properties of neurons derived from fetal mouse neurospheres are compatible with those of neuronal precursors in vivo. J. Neurosci. Res. 83 (8), 1494–1501. Pouton, C.W., Haynes, J.M., 2007. Embryonic stem cells as a source of models for drug discovery. Nat. Rev. Drug Discov. 6 (8), 605–616. Przyborski, S.A., 2001. Isolation of human embryonal carcinoma stem cells by immuno-magnetic sorting. Stem Cells 19, 500–504. Przyborski, S.A., Christie, V.B., Hayman, M.W., Stewart, R., Horrocks, G.M., 2004. Human embryonal carcinoma stem cells: models of embryonic development in humans. Stem Cells Dev. 13 (4), 400–408. Razvi, E.S., Oosta, G.M., 2010. Stem cells for cellular therapy space. Drug Discovery World Spring, 37–40. Stewart, R., Coyne, L., Lako, M., Halliwell, R.F., Przyborski, S.A., 2004. Human embryonal carcinoma stem cells expressing green fluorescent protein form
functioning neurons in vitro: a research tool for co-culture studies. Stem Cells Dev. 13 (6), 646–657. Storm, J.F., 1990. Potassium currents in hippocampal pyramidal cells. Prog. Brain Res. 83, 161–187. Traynelis, S.F., Wollmuth, L.P., McBain, C.J., Menniti, F.S., Vance, K.M., Ogden, K.K., Hansen, K.B., Yuan, H., Myers, S.J., Dingledine, R., Sibley, D., 2010. Glutamate receptor ion channels: regulation and function. Pharmacol. Rev. 62 (3), 405– 496. Ulrich, H., Majumder, P., 2006. Neurotransmitter receptor expression and activity during neuronal differentiation of embryonal carcinoma and stem cells: from basic research towards clinical applications. Cell Prolif. 39 (4), 281–300. Westerlund, U., Moe, M.C., Varghese, M., Berg-Johnsen, J., Ohlsson, M., Langmoen, I.A., Svensson, M., 2003. Stem cells from the adult human brain develop into functional neurons in culture. Exp. Cell Res. 289 (2), 378–383.