Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Sep 1.
Published in final edited form as: Mol Cell Neurosci. 2015 Aug 16;68:244–257. doi: 10.1016/j.mcn.2015.08.007

hVGAT-mCherry: a novel molecular tool for analysis of GABAergic neurons derived from human pluripotent stem cells

Brooke A DeRosa 1, Kinsley C Belle 1, Blake J Thomas 1, Holly N Cukier 1, Margaret A Pericak-Vance 1, Jeffery M Vance 1, Derek M Dykxhoorn 1,2
PMCID: PMC4593758  NIHMSID: NIHMS718808  PMID: 26284979

Abstract

Background

GABAergic synaptic transmission is known to play a critical role in the assembly of neuronal circuits during development and is responsible for maintaining the balance between excitatory and inhibitory signaling in the brain during maturation into adulthood. Importantly, defects in GABAergic neuronal function and signaling have been linked to a number of neurological diseases, including autism spectrum disorders, schizophrenia, and epilepsy. With patient-specific induced pluripotent stem cell (iPSC)-based models of neurological disease, it is now possible to investigate the disease mechanisms that underlie deficits in GABAergic function in affected human neurons. To that end, tools that enable the labeling and purification of viable GABAergic neurons from human pluripotent stem cells would be of great value.

Results

To address the need for tools that facilitate the identification and isolation of viable GABAergic neurons from the in vitro differentiation of iPSC lines, a cell type-specific promoter-driven fluorescent reporter construct was developed that utilizes the human vesicular GABA transporter (hVGAT) promoter to drive the expression of mCherry specifically in VGAT-expressing neurons. The transduction of iPSC-derived forebrain neuronal cultures with the hVGAT promoter-mCherry lentiviral reporter construct specifically labeled GABAergic neurons. Immunocytochemical analysis of hVGAT-mCherry expression cells showed significant co-labelling with the GABAergic neuronal markers for endogenous VGAT, GABA, and GAD67. Expression of mCherry from the VGAT promoter showed expression in several cortical interneuron subtypes to similar levels. In addition, an effective and reproducible protocol was developed to facilitate the fluorescent activated cell sorting (FACS)-mediated purification of high yields of viable VGAT-positive cells.

Conclusions

These studies demonstrate the utility of the hVGAT-mCherry reporter construct as an effective tool for studying GABAergic neurons differentiated in vitro from human pluripotent stem cells. This approach could provide a means of obtaining large quantities of viable GABAergic neurons derived from disease-specific hiPSCs that could be used for functional assays or high-throughput screening of small molecule libraries.

Keywords: human induced pluripotent stem cells, neural differentiation, GABAergic neurons, vesicular GABA transporter, fluorescent reporter construct

INTRODUCTION

The use of human pluripotent stem cells (hPSCs) has proven to be a powerful approach for studying human development and disease. The impact of hPSC-based differentiation strategies has been of particular importance in modeling diseases of the central nervous system (CNS), including many neurodevelopmental, neuropsychiatric and neurodegenerative disorders, due to the lack of readily available primary tissues from affected individuals. Where available, this tissue is restricted to post-mortem samples. A great deal of progress has been made in the development of methods for the differentiation of hPSCs into specific neuronal cell types (Chambers et al., 2009). Although effective at enriching for the specific neuronal type, these protocols result in the production of heterogeneous cultures containing multiple neuronal cell types, as well as, underlying supporting cells such as astrocytes and glial cells (Bilic and Izpisua Belmonte, 2012; Maherali and Hochedlinger, 2008; Narsinh et al., 2011). As a result, approaches are needed for the identification and isolation of specific cell types of interest from the complex mixture of cells.

Recent studies have supported an important role for inhibitory neurons, such as cortical interneurons (GABAergic neurons), in the development of the cortex. In addition, impairment in this cell population has been linked to neuropsychiatric disorders such as autism and schizophrenia (Acosta and Pearl, 2003; Lewis and Levitt, 2002; Rubenstein and Merzenich, 2003; Spencer et al., 2004). Although GABAergic interneurons make up a relatively small fraction of the total number of cells in the neocortex, even small changes in the balance of excitation and inhibition could have profound effects on key neurological functions, including cognition, sensory perception, language and spatial reasoning (Lui et al., 2011). This suggests that GABAergic interneurons play an important role in not only regulating the degree of excitation of the neocortex but also in the fine-tuning of neural networks and the quality of information processing across the different regions of the cortex.

The vesicular γ-aminobutyric acid (GABA) transporter (VGAT; also called SLC32A1) is a well-known marker for GABAergic neurons and is specifically expressed in inhibitory neurons (Gasnier, 2000; Sagne et al., 1997). GABA is the principal inhibitory neurotransmitter in the mammalian CNS and functions by binding to specific transmembrane receptors on both pre- and post-synaptic neuronal processes (Watanabe et al., 2002). It is synthesized from glutamate through the actions of two glutamate decarboxylases (GAD65 and GAD 67) and loaded into synaptic vesicles by the Vesicular GABA transporter (VGAT or SLC32A1) (Jin et al., 2003). In transgenic mouse and rat lines, the selective fluorescent labeling of nearly the entire GABAergic neuron population in the neocortex (>95%) has been accomplished using bacterial artificial chromosome (BAC) constructs that allow the expression of the fluorescent marker Venus, to be driven from the VGAT gene promoter (Uematsu et al., 2008; Wang et al., 2009). Given the results of these studies, we reasoned that a human VGAT promoter-driven fluorescent reporter construct might serve as a useful tool in the identification and isolation of the population of GABAergic cortical neurons generated in vitro through the differentiation of hiPSCs.

To begin to identify inhibitory GABAergic interneurons from complex hPSC-derived neuronal cultures, an ~ 1,800 base pair region of the VGAT promoter was cloned upstream of the fluorescent marker mCherry. Lentiviral-mediated delivery of the VGAT promoter driven fluorescent construct (pLV-hVGAT-mCherry) resulted in sustained expression of mCherry in GABAergic interneurons that co-localized with endogenous VGAT expressing cells. Although variable levels of expression were observed, the majority of hVGAT promoter-mCherry positive cells stained positively for GAD67 and GABA. In addition, these VGAT promoter mCherry positive cells could be purified by fluorescent activated cell sorting (FACS) using a novel protocol resulting in a highly pure and viable population of mCherry-expressing cells. These hVGAT-mCherry expressing cells stained positively for GABA and survived for over 3 weeks post FACS purification. This population of hVGAT positive cells included a variety of GABAergic interneuron subtypes with unique developmental origins.

METHODS

Development of the Human Inhibitory Neuron-Specific hVGAT-mCherry Fluorescent Reporter Construct

To target human stem cell-derived GABAergic neurons in heterogeneous populations of differentiated cells, a lentiviral-based fluorescent reporter system was developed that uses the human VGAT (solute carrier family 32 (GABA vesicular transporter), member 1, SLC32A1) gene promoter to drive the expression of the fluorescent protein mCherry (Figure 1). A 1,865 bp region of the VGAT gene including 262 bp downstream of the transcription start site (TSS) and 1603 bp upstream of the TSS was PCR amplified from genomic DNA purified from foreskin fibroblast cells. This region overlaps with peaks for several markers of promoter activation, histone H3K4 monomethylation (H3K4me1), histone H3K4 trimethylation H3K4me3, and RNA polymerase II (Pol2) binding (Figure 1A). The purified PCR product was digested with BamHI and EcoRI and ligated into the corresponding sites in the pENTR4 no ccdB vector (Addgene plasmid 17424) into which the mCherry gene had been previously cloned (Campeau et al., 2009). The inserted region of the VGAT promoter was confirmed by Sanger sequencing and the VGAT promoter mCherry cassette was transferred to the lentiviral vector pLentiX1 puro DEST (Addgene plasmid 17297) using the LR Clonase II enzyme mix (Invitrogen) to produce the pLV-hVGAT-mCherry vector. The pLV-hSyn-RFP vector was obtained from Addgene (Addgene plasmid 22909) (Nathanson et al., 2009).

Figure 1. Construction of human VGAT promoter driven mCherry reporter construct.

Figure 1

Open in a new tab

A) Region of the VGAT gene PCR cloned from genomic DNA upstream of the mCherry fluorescent reporter. UCSC genome browser tracts for markers of transcriptional activation (H3K4me1, H3K4me3, and RNA polymerase II (Pol2)) are shown. B) Composition of the lentiviral hVGAT-mCherry promoter reporter construct. C) Structure of the hVGAT promoter driven mCherry fluorescent reporter construct integrated (proviral) into the target cell genome.

Production of Lentiviral Expression Particles

The pLV-hVGAT-mCherry was packaged by cotransfection with the psPAX2 lentiviral packaging plasmid (Addgene plasmid 12260) and the Vesicular stomatitis virus envelope glycoprotein expressing pCMV-VSV-G plasmid (Addgene plasmid 8454) in lenti-X 293T cells (Clontech) using jetPRIME® (Polyplus-transfection™). The culture supernatant was harvested after 48h and the viral particles concentrated using the Lenti-X™ concentrator (Clontech) according to the manufacturer's protocol. The concentrated virus was resuspended in DMEM:F12 media, aliquoted, and stored at −80°C. The well characterized reporter construct expressing red fluorescent protein (RFP) from the neuron-specific synapsin I gene promoter (pLV-hSynapsin-RFP, hSYN-RFP) lentiviral construct has been used throughout as a positive control for neurons. Lentiviral transduction of the reporter constructs were carried in culture medium supplemented with 4 μg/mL polybrene using a multiplicity of infection (MOI) of 5.

Culture of hiPSCs and In Vitro Differentiation of Ventral Forebrain-Like Neurons

Human iPSCs (System Biosciences) were maintained on mitomycin c treated mouse embryo fibroblasts (MEFs) (EmbryoMax® Primary Mouse Embryo Fibroblasts, Strain CF1; Millipore) in mTESR1 culture medium (STEMCELLTechnologies) supplemented with 10 μM CHIR99021 (STEMGENT®), 1 μM PD325901 (STEMGENT®), 1 μM thiazovivin (STEMGENT®), and 10 μM Y27632 (STEMGENT®). The culture medium was changed daily and hiPSC colonies were enzymatically passaged with StemPro® Accutase® Cell Dissociation Reagent (Thermo Fisher Scientific) at a 1:4–1:6 split ratio every 4 to 7 days. If identified, spontaneously differentiated cells were mechanically removed prior to passaging. All cultures described here were kept in a 37°C incubator with 5% CO2.

Prior to inducing neural differentiation, hiPSCs were isolated from MEF feeder layer cells through magnetic column separation using MEF-specific antibodies (anti-mEF-SK4) coupled to paramagnetic beads (Miltenyi Biotec). Magnetic separation of the cells was carried out according to the manufacturer's instructions with few exceptions. Briefly, hiPSC colonies were dissociated into a single cell suspension through a 10-minute treatment with StemPro® Accutase® Cell Dissociation Reagent. In lieu of MACS buffer, mTeSR1 containing 2 μM thiazovivin and 20 μM Y27632 was used for both incubation of the cells in MEF-specific antibodies and for column washes during magnetic separation.

Cortical inhibitory neuron differentiation of hiPSCs was carried out using a chemically defined system similar to the previously described B27+5F method (Nicholas et al., 2013). The concentration of recombinant growth factors and small molecule compounds used in our differentiation scheme, in addition to the days they were added to the differentiation medium are listed in Table S1. The diagram shown in Figure 2A outlines the methods used for cortical inhibitory neuron differentiation. Neural induction was initiated through the formation of neural aggregates using AggreWell™800 plates (STEMCELLTechnologies) according to the manufacturer's protocol. Briefly, 3-4.5 × 106 cells (10,000 to 15,000 cells per neural aggregate) were added to each well of an Aggrewell™800 plate in neural induction media (NIM; STEMCELL technologies) supplemented with 10 μM Y2 6 2, 10 μM SB4 1542 (STEMGENT®), 1 μM dorsomorphin (STEMGENT®), and 1 μM thiazovivin.

Figure 2. Differentiation of hiPSCs into VGAT-expressing neurons with stable mCherry reporter expression over prolonged periods of time in vitro.

Figure 2

Open in a new tab

(A) Differentiation scheme used to derive VGAT-expressing cortical inhibitory neurons from hiPSCs. (B) Live cell images of day 70 hiPSC-derived neurons expressing mCherry from the human VGAT promoter (top images) and RFP from the human synapsin I promoter (bottom images). The cells shown in (B) were transduced with the respective lentiviral particles carrying the fluorescent reporter constructs on day 35. Scale bar: 100 μm.

After 5 days, neural aggregates were collected from the AggreWell™800 plate and transferred to 6-well plates coated with Poly-L-Ornithine (15 μg/mL; Sigma) and laminin (10 μg/mL (Trevigen) and cultured for another 5 days in NIM. On day 12, the differentiation medium was transitioned from NIM to a modified Enriched Neurobasal (ENB) medium7 consisting of Neurobasal™ medium (minus phenol red) (Life Technologies), 2% B-27® Supplement (Life Technologies), 1% GlutaMAX (Life Technologies), 1% Insulin-Transferrin-Selenium-A (Life Technologies), 1% Penicillin-Streptomycin (Life Technologies), 30 ng/mL tri-iodothyronine, 40 ng/mL thyroxine, 100 μg/ml bovine-serum albumen (BSA), 60 ng/ml progesterone, 16 μg/mL putrescine, 5 μg/mL N-acetyl-L-cysteine (NAC), and 5 μM forskolin (all Sigma). Between days 12-35, neural progenitor cells (NPCs) were expanded by enzymatically passaging the cells with StemPro® Accutase® Cell Dissociation Reagent every 4-7 days and replated at a density of ~ 4-8 × 104 cells/cm2. On day 35, the cells were plated for terminal differentiation on culture plates coated with Poly-D-Lysine (100 μg/mL; Sigma), laminin (20 μg/mL), and fibronectin (10 μg/mL; Sigma) (PDL/L/F) at a density of 1-2 ×104 cells/cm2.

Immunocytochemistry and Fluorescence Imaging

For immunocytochemistry (ICC) and imaging, cells plated for terminal differentiation or after FACS were grown in either 4-well or 8-well Millicell EZ SLIDES (Millipore) coated with PDL/L/F. Cultures were fixed with 4% formaldehyde (Thermo Scientific) for 15 minutes at room temperature and washed with PBS. The slides were simultaneously blocked and permeabilized for 45 minutes at room temperature in 20% normal donkey serum (NDS) or Normal Goat Serum (NGS; both from Jackson ImmunoResearch) and 0.2% Triton X-100 (Sigma) in antibody buffer (150 mM NaCl, 50 mM Trizma base, 1% BSA, 100 mM L-Lysine, 0.04% sodium azide, pH 7.4, 0.3 M glycine; Sigma) (Blackmore et al., 2012). Table S2 lists the primary antibodies used in this study. Cultures were incubated in primary antibody solutions overnight at 4°C then washed extensively in ICC wash buffer (PBS /1% BSA) and incubated with the appropriate Alexa-fluor conjugated secondary antibody (Life Technologies) for 1h at RT. After the incubation in the secondary antibody, the cells were washed extensively in ICC wash buffer and incubated with DAPI (NucBlue Fixed Cell Stain; Life Technologies) and mounted with glass coverslips using DAKO Fluorescence Mounting Medium (DAKO). Images shown in Figures 3, 5, and 7D were acquired using a Zeiss LSM 710 microscope (Carl Zeiss Microscopy). Live cell images shown in Figures 2B and 7C were acquired using an Eclipse TE2000U fluorescence inverted microscope (Nikon).

Figure 3. Characterization of hVGAT-mCherry expression in human iPSC-derived GABAergic neurons.

Figure 3

Open in a new tab

Representative images of neuronal cultures transduced with hVGAT-mCherry and immunostained with anti-VGAT (A), anti-GAD67 (B), anti-GABA (C), or anti-GFAP (D) antibodies. DAPI staining was included in all samples imaged to visualize cell nuclei (A-D). Scale bar: 50 μm (A-C) and 100 μm (D).

Figure 5. Expression of hVGAT-mCherry reporter in GABAergic neuron subtypes.

Figure 5

Open in a new tab

Differentiation cultures were transduced with pLV-hVGAT-mCherry then stained with antibodies specific for cortical inhibitory neuron subtype markers on day 130. (A-C) Confocal images showing the colocalization of hVGAT-mCherry and inhibitory neuron subtype marker-specific antibody immunoreactivity. First panel (left): anti-mCherry staining; second panel: antibody-specific staining of CR (A), PV (B), and SST (C); third panel: DAPI staining of cell nuclei; fourth panel (right): merged image of anti-mCherry, subtype marker staining, and DAPI. Arrows indicate colocalization of mCherry and GABAergic subtype antibody staining (A-C). (D) Percentage of the total number of cells stained positive for the cortical inhibitory neuron subtype marker (CR, PV, or SST) co-labeled with hVGAT-mCherry. Percentages were calculated by dividing the number of cells exhibiting colocalization of mCherry and CR, PV, or SST antibody-specific staining by the total number of cells stained positive for each individual subtype marker. The percentages reflect the combined count of 3 wells for each chemical marker and the standard error reflects the mCherry labeling distribution treating each well as an independent observation. For each subtype marker, over 1000 cells were counted amongst individual wells independently labeled with CR, PV, and SST specific antibodies. A one-way ANOVA was used to test for differences in mCherry expression among CR, PV, and SST cell populations. The percentage of cells expressing mCherry was not significantly different across the three GABAergic cortical neuron subtypes, F (df: 2, 6) = 1.8, p = 0.24. Abbreviations: calretinin (CR), parvalbumin (PV), and somatostatin (SST). Scale bar: 100 μm.

Figure 7. FACS-purified hVGAT-mCherry labeled-cells remain viable after sorting and continue to express GABAergic neuron-related markers.

Figure 7

Open in a new tab

(A) Countess cell counter (Invitrogen) images of unsorted and sorted cells stained with trypan blue; dead cells are stained blue by the uptake of dye. Trypan blue exclusion test of cell viability was performed immediately after sorting. The unsorted and sorted cells are from the same sample; the unsorted sample was collected just prior to sorting and kept on ice during FACS. (B) Boxplot of unsorted vs sorted cell viability. Center lines show the medians; box limits indicate the 25th and 75th percentiles as determined by R software; whiskers extend 1.5 times the interquartile range from the 25th and 75th percentiles, outliers are represented by dots; data points are plotted as open circles. n = 3 sample points. (C) Phase-contrast and mCherry expression in sorted hVGAT-mCherry positive cells 5 days after replating. (D) Immunocytochemical analysis for mCherry, GABA, β-Tubulin III, and DAPI in sorted hVGAT-mCherry positive cells 20 days after post-FACS replating. hVGAT-mCherry positive cells remain viable for at least several weeks after sorting, and continue to express GABAergic-specific markers. Scale bar: 50 μm.

Quantitative Image Analysis

To quantify hVGAT-mCherry and GABAergic neuron marker antibody-labeled cells, fluorophores were imaged independently using a 10× or 20× objective to acquire images from a minimum of 3 separate wells per marker-specific antibody staining. Quantification of mCherry and GABAergic neuron marker positive cells was performed using Fiji (Schindelin et al., 2012) (Cell Counter plugin; Kurt De Vos, University of Sheffield). Fiji was also used to determine the total number of cells in each image field using the using DAPI to identify nuclei (ITCN plugin; Thomas Kuo and Jiyun Byun, University of California Santa Barbara). The percentages shown in Figure 5D indicate the percent total number of cells stained positive for the GABAergic subtype marker that were co-labeled with mCherry. Percentages for the total number of mCherry-positive and mCherry-negative cells stained positive for each GABAergic subtype marker are described in Supplemental Table 3. Analysis of variance (ANOVA) was used to identify significant differences amongst subtype cell populations in the percentage of cells expressing the reporter (Figure 5). The analysis of fluorescence intensity of VGAT antibody staining in pLV-hVGAT-mCherry treated cells was performed in Fiji. Briefly, a small circular region (50 pixels) was placed randomly over either a region (n=50) of the image staining positive for DAPI but no perceptible VGAT expression to establish the background fluorescence or cells that expressed mCherry from the pLV-hVGAT-mCherry vector but lacked apparent VGAT staining (n=25) (as above). The mean fluorescence intensity of each region was analyzed for the red and green channels. The average fluorescence intensity and standard deviation were calculated for each condition. Student t-test was used to examine the statistical significance of the difference in mean green fluorescence between the two groups. Pearson's correlation was used to examine the relationship between red fluorescence intensity (mCherry) and green fluorescence intensity (VGAT) in the pLV-hVGAT-mCherry positive cells (Supplemental Figure 2).

Purification of Fluorescent Reporter-Expressing hiPSC-Derived Neurons by FACS

All solutions and media used during single cell dissociation of aggregated neurons and cell sorting were supplemented with 0.132 M D-(+)-Trehalose dihydrate (Trehalose; Sigma)(Saxena et al., 2012). The hiPSC-derived neurons stably expressing hVGAT-mCherry were dissociated into a single cell suspension by incubating cultures in Accumax supplemented with Trehalose (5% w/v). The disaggregated cells were further triturated by gently pipetting with a 10 mL borosilicate glass serological pipet. An equal volume of an optimized ENB-Sorting (ENB-S) medium for sorting neurons comprised of ENB medium supplemented with 5% w/v Trehalose, 20 ng/mL BDNF, 20 ng/mL NT-3, 20 ng/mLβ-NG , 20 ng/mL G N , and 0.2% Anti-Clumping Agent (Life Technologies) was added to the cell. The cells were strained through a 70 μm nylon mesh cell strainer (B alcon) into a new 50 mL tube then centrifuged at 1,500 rpm for 10 minutes at 4°C. After centrifugation, cells were resuspended in pre-chilled ENB-S medium at a concentration of 2-10 × 106 cells/mL in 5 mL round bottom tubes with caps (BD Falcon) and kept on ice, in the dark, until sorted. A control well that was not subjected to flow cytometry was processed in the same manner and left on ice throughout the sorting procedure (unsorted cells). The cells were sorted at the University of Miami Miller School of Medicine Flow Cytometry Core Facility under sterile conditions on a FACS ARIA IIu using FACSDiva software (BD Biosciences). FlowJo software (Tree Star) was used to analyze and present the flow cytometry data shown in the results section (Figure 6). Viable cell populations were selected by forward (FSC) and side (SSC) scatter. RFP or mCherry-positive cells were isolated using 4-way purity precision gating. Optimized FACS conditions (Pruszak et al., 2007) were applied to sorting hiPSC-derived neurons using a 100-μm ceramic nozzle (BD Biosciences), sheath pressure of 20-25 pounds per square inch (PSI), and an acquisition rate of 1,000-3,000 events per second. Sorted cells were collected in 5 mL round bottom tubes containing 1 mL of ENB-S medium. The purity of cell populations isolated by FACS was examined by performing post-sort FACS analysis using FACSDiva software. The purity of extracted cell populations was assessed by the percentage of resorted cells having the same fluorescence intensity characteristic of cells selected for using the 4-way purity precision gate.

Figure 6. Enrichment of inhibitory neuron populations by FACS.

Figure 6

Open in a new tab

(A) Pre-sort FACS analysis of negative control cells (no reporter expression; indicated by blue dots) and cells transduced with pLV-hSYN-RFP and pLV-hVGAT-mCherry reporter constructs. To avoid the inclusion of cells with weak non-specific reporter expression (false positives; potential candidates indicated by cyan dots) in purified samples of hSYN-RFP and hVGAT-mCherry positive cells, FACS conditions involved the application of a purity gate that limited sorting to cells exhibiting a high degree of reporter expression (indicated by red dots bound by uppermost gate). (B-C) The purity of sorted cells was determined by performing post-sort FACS analysis on resorted samples of purified cell populations and assessing the percentages of cells in each gate. Sorting cells under the described conditions yielded over >90% pure RFP and mCherry positive cell populations (B) and 100% pure mCherry negative (gated using lower right quadrant) (C).

Post-sort assessment of cell viability and cell culture conditions

After FACS, the viability percentages of sorted and unsorted cells was determined by staining cells with trypan blue (Life Technologies) and counting the number of unstained (live) and trypan blue-stained (dead) cells using a Countess Cell Counter (Life Technologies). The cell viability data collected on sorted and unsorted cells immediately following FACS (Figure 7B) was analyzed and presented using the open-source application BoxPlotR(Spitzer et al., 2014) (http://boxplot.tyerslab.com/).

To evaluate effect of G N and β-NGF treatment on sorted neuron viability, Sorted and unsorted neurons were plated immediately after FACS at a cell density of 2 ×104 cells/cm2 and cultured in the presence or absence of 10 ng/mL G N and 20 ng/mL β-NGF in ENB medium supplemented with 5 μM Caspase-3/7 reagent (Essen Bioscience) to label nuclei of cells undergoing apoptosis. An IncuCyte ZOOM (Essen Bioscience) was used to automatically acquire kinetic data every 4 hours on the number of caspase-3/7 fluorescent objects for 60 hours post sorting. In addition, neuron cell-body area (mm2)/total surface area (mm2) was calculated every 4 hours for 200 hours. At the end of the experiment, the total number of cells was quantified following nuclei staining with Hoechst. The apoptotic index of the treated and untreated cells (Figure 8A) was calculated by dividing the number of caspase-3/7 fluorescent objects (multiplied by 100) by the total number of nuclei. SPSS statistical software was used to conduct a mixed-group ANOVA to compare the level of apoptosis of sorted neurons cultured in the presence or absence of G N and β-NGF and unsorted neurons (treatment; between-groups factor) at 0 hours and 60 hours post-plating (time; within-groups factor) using the apoptotic index as the dependent measure. Post-hoc analysis using Tukey's HS (alpha = 0.05) was used to identify differences in the apoptotic index of unsorted neurons and sorted neurons treated with or without G N and β-NGF (Table S4)

Figure 8. GDNF and β-NGF treatment increases viability of sorted neurons.

Figure 8

Open in a new tab

(A) Sorted and unsorted neurons were plated immediately after FACS and treated with Caspase-3/7 reagent in the presence or absence of GDNF and β-NGF. The unsorted and sorted cells are from the same sample; the unsorted sample was collected just prior to sorting and kept on ice during FACS. The number of cells undergoing apoptosis was determined by quantifying the number cell nuclei positive for Caspase-3/7 activity to determine the Apoptotic index. Kinetic data was acquired every 4 hours in an IncuCyte ZOOM (Essen Bioscience). (B) Confluence was measured every 4 hours for 200 hours and the Cell-body cluster area (mm2)/growth area (mm2) of sorted and unsorted cells cultured with or without G N and β-NGF was calculated.

RESULTS

Design of the human VGAT-mCherry reporter construct

To target human stem cell-derived inhibitory neurons in heterogeneous populations of differentiated cells, a lentiviral-based fluorescent reporter system was developed that uses the human VGAT (solute carrier family 32 (GABA vesicular transporter), member 1, aka: SLC32A1) gene promoter to drive the expression of the fluorescent protein mCherry (Figure 1). The region directly upstream of the VGAT transcription start site (TSS) is enriched for markers of active genes. Specifically, peaks of monomethylated (H3K4me1) and trimethylated (H3K4me3) histone occurs directly upstream of the TSS (Barski et al., 2007, Heintzman et al, 2007) (Figure 1A). In addition, this region contains binding sites for RNA polymerase II indicating active recruitment of the transcriptional machinery to the VGAT TSS(Hager et al., 2009). The promoter region was cloned into the pENTR4 no ccdB vector(Campeau et al., 2009) (Addgene plasmid 17424) upstream of the mCherry gene and transferred to the lentiviral construct pLentiX1 puro DEST (Addgene plasmid 17297) to produce the pLV-hVGAT-mCherry vector (which will be abbreviated as hVGAT-mCherry in the remainder of the text; Figure 1B). The organization of the hVGAT promoter and mCherry gene in the integrated proviral DNA is shown in Figure 1C.

Characterization of hVGAT-mCherry expression in hiPSC-derived ventral forebrain neurons

To characterize the expression of hVGAT-mCherry in human GABAergic cortical-like neurons, human induced pluripotent stem cells (hiPSCs) were differentiated using a protocol that drives the development of ventral forebrain neurons according to the schematic in Figure 2A. The differentiating GABAergic neurons were transduced with lentiviral expression particles carrying either hVGAT-mCherry or hSYN-RFP vectors between days 55 and 97 of the neuronal differentiation scheme. Expression of mCherry from the VGAT promoter or RFP from the Synapsin I promoter was monitored by fluorescent microscopy beginning at 48h post-lentiviral transduction. As expected, the Synapsin I promoter drove strong expression of RFP which was visible by 48h post treatment. In contrast, there was only a weak signal from the mCherry at 48h post transduction which gradually increased over the next several days.

Next, we examined the stability of reporter expression by determining if labeled cells retained hVGAT-mCherry expression upon further in vitro differentiation. After the transductions, in vitro differentiation was continued under the same conditions for up to 75 days post transduction. We found that both hVGAT-mCherry and hSYN-RFP maintained robust expression of their reporters and that, within individual cells, there was little to no variability in expression level of the reporters over the time frame measured (Figure 2B). From this, we conclude that mCherry is stably expressed from the hVGAT promoter reporter construct at consistent levels for at least 75 days post-transduction.

To establish the specificity of the hVGAT-mCherry fluorescent reporter construct, the virally transduced cultures of differentiated neurons were stained with antibodies that recognize endogenous VGAT (Figure 3A), the GABAergic neuron-specific marker GAD67 (Figure 3B), the neurotransmitter GABA (Figure 3C), the neuron-specific marker β-tubulin III (Supplemental Figure 1), or the glial cell marker GFAP (Figure 3D). The cells that were expressing mCherry from the VGAT promoter showed a significant co-localized with those that stained positive for the endogenous VGAT protein (Figure 3A). Quantitative image analysis was used to assess the degree of overlap between the hVGAT-mCherry+ cells and the endogenous VGAT stained cells. Based on the automated cell counter plug in on the Fiji imaging software, 72% of the cells expressing hVGAT-mCherry stained positively for the VGAT protein (Figure 4A). Further analysis was performed on the hVGAT-mCherry positive cells in which endogenous VGAT expression was not detected by the automated cell counter. Using a 50-pixel window, the fluorescence intensity in both the green and red channel was assessed on multiple regions that stained positive for DAPI but which lacked VGAT expression. This criteria was used since it is possible that there would be cells which stained positive for VGAT expression but were not transduced by the fluorescent reporter construct. This same window was then applied to analyze the level of fluorescence in hVGAT-mCherry positive cells in which endogenous VGAT appeared not to be expressed. This analysis showed that there was low but statistically significant level of endogenous VGAT expression in these cells (Figure 4B and C). There was a positive correlation (Pearson's correlation=0.5 , p-value=0.007) between mCherry expression from the hVGAT-mCherry vector and the endogenous VGAT levels even in these low VGAT expressing cells (Supplemental Figure 2). Therefore, these results show a strong co-relation between mCherry expression from the hVGAT-mCherry vector and endogenous VGAT expression. There were cells in the culture that stained positively for VGAT but which lacked mCherry expression. Although high levels of lentiviral transduction can be achieved (>85% transduced using a CMV-driven reporter construct) (data not shown), there are cells within the culture that have failed to be transduced by the hVGAT-mCherry vector and, as a result, lack mCherry expression.

Figure 4. Quantitation of the colocalization of GABAergic neuron markers and hVGAT-mCherry expression.

Figure 4

Open in a new tab

(A) pLV-hVGAT-mCherry transduced cells were stained for the expression of GAD67, VGAT, or GABA and the number of hVGAT-mCherry positive cells staining for the respective marker was calculated using the cell counter plug in for the Fiji software. Standard deviation (SD) of individual well counts (n = 4). The last column indicates the total number of cells counted amongst the wells. (B) Detailed analysis of the hVGAT-mCherry positive/endogenous VGAT-negative cells showed modest but statistically significant levels of VGAT fluorescence compared to background fluorescence when mean fluorescence intensity was measured using a restricted window (50-pixels). (C) Scatter plot of the distribution of red fluorescence intensity (hVGAT-mCherry positive cells) and green fluorescence intensity (endogenous VGAT) from the regions analyzed in (B).

In addition to VGAT expression, the percentage of hVGAT-mCherry positive cells that stained positively for GAD67 and GABA was analyzed. Over 80% of the hVGAT-mCherry positive cells stained positively for GAD67 expression, while 66% of the cells stained positively for GABA (Figure 3B and C and Figure 4A). This lower level of GABA expression in hVGAT-mCherry transduced cells could be a result of several factors, including the developmental heterogeneity of in vitro differentiated neuronal cultures and the high level of background in GABA staining in these heterogeneous cultures. Of note, unlike the heterogeneous cultures, significant GABA staining was seen in the FACS purified hVGAT-mCherry positive cells (Figure 7). We observed that nearly all of the mCherry-labeled cells showed positive staining of β-Tubulin III, signifying that hVGAT reporter expressing cells are neuronal (Figures 3A, C). Furthermore, the mCherry-positive β-Tubulin III-expressing cell population displayed a wide range of neuronal morphologies. These results show that GABAergic and hVGAT-mCherry cells consist predominantly of overlapping populations of neurons in vitro, suggesting that the cells expressing the hVGAT-mCherry reporter are indeed GABAergic. Consistent with these results, no overlap in signal was seen between hVGAT-mCherry expressing cells that were stained with the glial marker GFAP (Figure 3D).

To further confirm the specificity of the hVGAT-mCherry reporter construct, the viral particles were used to transduce non-neuronal human embryonic kidney (HEK) 293T cells. HEK293T cells were treated with lentiviral particles expressing mCherry from the ubiquitously expressed Cytomegalovirus (CMV) promoter, hVGAT-mCherry, or hSYN-RFP. Over 97% of the cells treated with the CMV-mCherry expression construct were positive for mCherry expression as measured by flow cytometry compared to untreated cells (Supplemental Figure 3). On the other hand, there were only a very small number (0.2%) of cells showing mCherry expression in the hVGAT-mCherry transduced or RFP in the hSYN-RFP transduced cells. In addition, the analysis of the mean fluorescence intensity (MFI) of the CMV-mCherry treated cells was significantly increased from a background fluorescence (1,185 ± 27 fluorescence units (fu)) to over 190,000 fu. On the other hand, the hVGAT-mCherry and hSYN-RFP treated cells showed only a modest increase in the MFI (1,661 ± 4 fu and 1,401 ± 21 fu, respectively).

hVGAT-mCherry expression in cortical inhibitory neuron subtypes

Next, we characterized the distribution of hVGAT-mCherry+ cells from several different subtypes of cortical inhibitory neurons in the differentiation cultures. This was accomplished by quantitatively examining the colocalization of mCherry with antibody-specific staining of the cortical interneuron subtype markers: Calretinin (CR), Parvalbumin (PV), or Somatostatin (SST) (Figure 5A-C). The expression of hVGAT-mCherry in cortical inhibitory neuron subtypes was quantified in day 130 neurons by identifying cells exhibiting co-localization of mCherry and cortical inhibitory neuron subtype-specific markers. Immunocytochemical (ICC) staining of each marker (i.e. CR, PV, and SST) was carried out independent of one another. hVGAT-mCherry was expressed in the majority of CR, PV, and SST antibody-labeled cells. On average the pLV-hVGAT promoter drove expression of mCherry in 90.2% ± 1.7%, 80.5% ± 3.7%, and 75.7% ± 15.9% (mean ± standard deviation) of CR, PV, and SST inhibitory neuron subtypes, respectively (Figure 4D). A one-way ANOVA was used to test for the significance of the differential expression of mCherry among CR, PV, and SST cell populations. The percentage of cells expressing mCherry was not significantly different across these three GABAergic cortical neuron subtypes (F = 1.8, p-value = 0.24). These results are consistent with reports of the overlap of CR, PV, and SST antibody-immunoreactivity with Venus expression in cortical slices of VGAT-Venus transgenic rat lines (Uematsu et al., 2008). In addition, the proportion of GABAergic subtypes from the total population of mCherry positive and mCherry negative population was calculated (Supplemental Table 3).

Optimized methods for cell sorting of viable, hiPSC-derived neuronal subtypes from heterogeneous cell populations

The hVGAT-mCherry reporter construct allows for VGAT+ cells to be identified in heterogeneous neuronal differentiation cultures derived from hiPSCs. These cells are viable and can be monitored for extended periods of time (over 7 weeks) in culture following transduction with the viral constructs. In addition to facilitating the identification of live VGAT+ cells in culture, the hVGAT-mCherry reporter construct can also be used to isolate enriched populations of VGAT+ cells using FACS. Previously, the sorting of neuronal cells by FACS has been hindered by technical challenges, including significant cell loss resulting from the treatment. Purification of specific neuronal populations with FACS is known to result in poor isolation rates of viable cells. Compared to other cell-types, neurons are particularly vulnerable to the physical stress inflicted by enzymatic and manual dissociation of cells and high pressure sorting. Although a number of variables factor into this loss of viability of sorted neurons, it is largely related to the high level of cellular stress experienced by neurons when dense networks of interconnected neuronal processes are damaged (Pruszak et al., 2007). Therefore, we sought to develop a method that would allow for highly purified population of intact and viable GABAergic neurons expressing mCherry from the hVGAT promoter to be isolated by FACS.

Saxena et al. described the use of trehalose supplementation to increase viability in sorted eGFP-positive neurons obtained from whole cortices of transgenic PVALB-eGFP reporter mice (Saxena et al., 2012). Trehalose is a disaccharide that is well documented for its role in maintaining viability of cells exposed to environmental stress, including cryopreservation (Buchanan et al., 2004) and heat stress (Carninci et al., 1998). Pruszak et al. reported methods that included gentle FACS conditions that enabled them to increase cell survival rates in Synapsin-GFP+ sorted hESC-derived neurons (Pruszak et al., 2007). To demonstrate the utility of the hVGAT-mCherry construct in isolating hiPSC-derived GABAergic neurons via FACS, we developed an approach that incorporates both the treatment with trehalose with the “gentle FACS” settings to maximize the viability of the purified VGAT+ neurons (Pruszak et al., 2007; Saxena et al., 2012).

Day 55 differentiating neurons were transduced with either the hSYN-RFP or hVGAT-mCherry lentiviral vectors. The hSYN-RFP+ or hVGAT-mCherry+ cells were purified from their respective cultures 7 days post-transduction by FACS on a FACS Aria IIu (BD Biosciences, Figure 6A). The purity of isolated RFP+ and mCherry+ cells populations was determined by post-sort FACS reanalysis. Using this method, we were able to obtain highly purified (>90% purity) samples of reporter-expressing cells (Figure 6B). As expected, reanalysis of the mCherry-negative cell population show a depletion of mCherry expressing cells in this cell fraction (Figure 6C).

Viability of sorted mCherry-positive cells

To evaluate viability, sorted cells and unsorted cells (which were kept on ice but not sorted) were simultaneously stained with trypan blue and the number of viable (unstained) and dead (stained blue) cells was counted using a Countess Cell Counter (Life Technologies) (Figure 7A and B). We did not observe a significant difference between the percentages of trypan blue-stained dead cells in sorted samples compared to the unsorted samples. Figure 7A shows a representative image of sorted mCherry positive cell or unsorted cells stained with trypan blue. The counts from three samples per condition were averaged and the interquartile range of the percentage of viable cells present in each sample was calculated. The first to third quartiles ranged from 76.5% to 87.5% cell viability in samples collected just prior to FACS and kept on ice during the sort procedure (unsorted) and 76.5% to 91.5% cell viability in samples obtained through FACS (sorted) (Figure 7B). The slight increase in the percentage of live cells in the FACS population is due to the gating applied during the FACS that distinguishes the live and dead cells based on their light scattering properties (forward scatter). An average yield of 2.5 × 105 viable, reporter positive cells were obtained from ~ 5 × 106 cells (counted prior to FACS).

Previous studies have shown the utility of sorting neurons expressing cell-type specific fluorescent reporters as a means to obtaining specific neuronal cell populations for gene expression studies (Lobo et al., 2006; Nelson et al., 2006). This strategy could also prove useful in obtaining large quantities of viable human neurons with distinct cellular phenotypes for functional and pharmacological studies. We propose that our methods could be used as a means of providing sufficient quantities of viable, highly purified neuronal cells for functional studies aimed at exploring the role of inhibitory neurons in human development and disease (Xu et al., 2014). We tested the hypothesis that the plating of sorted mCherry-positive cells would remain viable for extended periods of time in vitro and continue to express GABAergic neuron-related markers (Figure 7C and D). In preliminary experiments, we observed that mCherry-positive sorted cells remained viable for ~2 weeks, after which the viability of the cells began to decline. One possible explanation for this decrease in viability is the absence of supportive or neurotrophin-secreting cell-types from these highly purified GABAergic cell cultures. Therefore, we tested whether supplementing the culture media of the isolated cells would help to maintain their health and viability. A mixed-groups ANOVA was performed to compare the apoptotic index of FACS-isolated neurons cultured in the presence or absence of GDNF and β-NGF to unsorted neurons (no GDNF/β-NGF treatment) at 0 hours and 60 hours post-plating after FACS. Our results show there was a main effect for treatment group with the apoptotic index of neurons at the examined time points (F(2,3) = 73.667, p-value = 0.00 , ηp2 = 0.98). Moreover, post-hoc analysis using Tukey's HSD indicated that FACS-isolated cells grown in media supplemented with GDNF and β-NGF do not significantly differ in levels of apoptosis compared to unsorted neurons (p = 0.77), (Figure 8A). On the other hand, isolated hVGAT-mCherry expressing neurons grown in the absence of GDNF and β-NGF showed a significant increase in the apoptotic index during the first 60h post plating compared to unsorted neurons (p = 0.003) and sorted neurons treated with GDNF and β-NGF (p = 0.007) (Figure 8A and Supplemental Table 4).

In addition, the isolated hVGAT-mCherry expressing neurons that were grown in the absence of these neurotropic factor showed elevated cell loss as measured by the decrease in confluence of the plated cells (i.e. decreased cell body area (mm2) per total plates area (mm2) measured in the Incucyte Zoom (Essen Biosciences) compared to the GDNF and β-NGF supplemented cells and the unpurified hVGAT-mCherry cells (Figure 8B). Therefore, supplementation of the culture media with GDNF and β-NGF helps to maintain the viability and health of the isolated hVGAT-mCherry cells to levels comparable to cells grown in the heterogeneous cultures.

Finally, we aimed to confirm that FACS-purified hVGAT-mCherry-positive cells were indeed neuronal and, importantly, retained the ability to synthesize the inhibitory neurotransmitter GABA when cultured as part of a cell population comprised almost entirely of inhibitory neurons. ICC staining of FACS-purified hVGAT-mCherry cells 20 days after sorting showed positive co-staining of β-Tubulin III and GABA, in addition to maintaining continuous expression of mCherry through use of the human VGAT reporter promoter (Figure 7D). Although GABA expression was difficult to assess from the mixed cultures due to high background staining, the isolated hVGAT-mCherry expressing neurons showed significant and readily identifiable GABA staining (Figure 7D).

Discussion

Patient-specific iPSC-based in vitro models of human disease has become an important tool in the field of neurological disease research. To this end, much attention has been placed on improving current techniques for in vitro differentiation of specific neuronal subtypes from hiPSCs, including the refinement of methods used to analyze targeted populations of cells17,30-36. Notably, recent advancements in protocols for the derivation of GABAergic cortical interneurons from human embryonic stem cells (hESCs) have provided a platform for studying the multifaceted roles of these neurons in the developing and adult brain. Moreover, due to the importance of GABAergic neurons in a number of human neurodevelopmental disorders, including autism spectrum disorder and schizophrenia, these newly established methods could prove to be of great value if extended to patient-specific iPSC-based models of these diseases (Insel, 2010; Lewis, 2014; Pizzarelli and Cherubini, 2011).

Despite the recent progress, there are several technical limitations that need to be addressed before the full potential of this model can be realized. Current methods for the in vitro differentiation of human pluripotent stem cells into inhibitory neurons, or other types of neuronal cells, ultimately generates a population of cells containing a wide variety of neuronal and glial cell-types, as well as other unknown cell-types that spontaneously differentiate in culture (Kozubenko et al., 2010). Although a cell culture containing a mixed population of neurons and glia, is arguably a better representation of the in vivo cellular environment of the brain, heterogeneity in samples of differentiated cells complicates the analysis of specific cell-types of interest, potentially leading to misinterpretation of results.

Therefore, the development of novel tools for the tracking and isolation of specific neuronal sub-types is needed. In this report, we describe the development of a cell-type specific fluorescent reporter construct based on the expression of mCherry from the human SLC32A1 (hVGAT) promoter that enables the identification of human GABAergic neurons derived from pluripotent stem cells in vitro. Concomitantly, we optimized FACS-based methods for isolating highly pure samples of viable reporter-expressing hiPSC-derived neurons from the heterogeneous population of cells that arise during in vitro differentiation of pluripotent stem cells.

The results of this study indicate that expression of the mCherry gene under control of the human VGAT promoter permits the labeling and FACS-based enrichment of human GABAergic neurons derived from pluripotent stem cells. Lentiviral transduction of hVGAT-mCherry in day ≥55 hiPSC-derived ventral forebrain neurons and subsequent immunocytochemical analysis showed mCherry expression was expressed GABAergic interneurons, with only a few cells exhibiting low level mCherry expression that did not stain positive for at least one GABAergic neuron markers (i.e. antibodies against GABA, GAD67, or VGAT). Although GABA staining showed the lowest level of colocalization with the hVGAT-mCherry expressing cells in the mixed cultures (Figure 3C and 4A), the purified hVGAT-mCherry expressing cells showed significant levels of GABA production. This suggests that the high level of background GABA staining in the mixed cultures may be reducing the ability to appropriately colocalize GABA expression in the hVGAT-mCherry expressing cells. In our initial analysis, approximately 72% of the hVGAT-mCherry expressing co-localized with endogenous VGAT protein. However, a closer examination of VGAT expression suggests that some of the hVGAT-mCherry positive cells in which VGAT did not appear to be colocalized actually expressed low but discernable levels of endogenous VGAT. Stem cell-derived neuronal cultures are heterogeneous not only for the variety of cell types found but also in the development stage of the cells. Although many of the neuronal specific markers are expressed in young neuronal cultures (e.g. day 55 neurons), many of the cells present are not yet functionally mature. As a result, we would expect the cells to have heterogeneous features, including the level of expression of GABAergic marker genes. Even in the low VGAT expressing cells there was a positive correlation between the level of hVGAT-mCherry expression and endogenous VGAT (Supplemental Figure 2). Although there was an overall positive correlation between hVGAT-mCherry expression and VGAT expression, there were cells in the culture that showed discernable hVGAT-mCherry expression with minimal to no endogenous VGAT expression. There may be several explanations for this phenomenon. First, since the cloned VGAT promoter takes a portion of the genomic DNA directly upstream of the TSS of the hVGAT gene, it may lack some additional regulatory elements such as enhancers or insulators that may be present further upstream or downstream of the promoter that may contribute to the regulation of endogenous VGAT. Since the majority of hVGAT-mCherry positive cells showed at least some expression of endogenous VGAT albeit weak in some cells the cloned promoter appears to contain the important regulatory elements. However, this lack of potential additional regulatory sequences may influence the level of VGAT expression and the timing of VGAT expression during development. In addition, any post-transcriptional regulatory mechanisms (egg. miRNA regulation) to which the endogenous VGAT transcript would be subject would be absent in this reporter construct. This could potentially explain the discrepancy in some cells between hVGAT-mCherry expression levels and that of the endogenous VGAT. In addition, elevated mCherry expression could be seen in the case in which specific cells had multiple hVGAT-mCherry vectors integrated into their genome. Lack of hVGAT-mCherry expression in cells positively identified as GABAergic neurons may result from an uneven distribution of infectious viral particles per cell. It is also possible that in some GABAergic neurons the expression of mCherry was silenced due to a positional effect, in which the chromosome sequences flanking the pLV-hVGAT-mCherry integration site imposed a negative regulatory effect on the transcription of mCherry (Bessis et al., 1995; Feng et al., 2001). Likewise, chromosomal position effects may explain anomalous expression of hVGAT-mCherry in some non-GABAergic cell-types(Clark et al., 1994).

Although we saw the sustained levels of hVGAT-mCherry expression in these studies, transgene expression from retroviral vectors have been shown to be silenced during the course of differentiation from stem and progenitor cells to the cell type of interest (For example, (Hong et al., 2007)). Therefore, delivery of the reporter construct at a time point closer to the time of analysis will help to abrogate potential silencing of the transgene during differentiation. The ability to transduce neuronal progenitor cells and neurons with lentiviral vectors with high efficiency and viability would help to circumvent this issue of transgene silencing. Furthermore, we demonstrate the utility of hVGAT-mCherry in labeling a variety of GABAergic cortical interneuron subtypes with different in vivo developmental origins. Finally, we outline methodological adaptations to FACS purification of hiPSC-derived neurons that enables >95% pure hiPSC-derived GABAergic neurons to be isolated from the complex mixture of cells that arise during in vitro differentiation of pluripotent stem cells. Furthermore, we show that our method yields relatively large quantities of purified human neurons that are 83% ± 15% viable on average, thus are likely to be amendable to functional screening assays. As a proof of principle, we demonstrate that sorted hVGAT-mCherry cells can be replated and further differentiated in vitro for several weeks after FACS purification and continue to express GABAergic neuron-related markers. This longevity and survival of the isolated hVGAT-mCherry expressing cells was augmented by the supplementation of the culture media with the neurotropic factors GDNF and β-NGF. In fact, the viability and growth of the isolated cells supplemented with GDNF and β-NGF was comparable to that seen in hVGAT-mCherry expressing cells grown in the heterogeneous cultures (Figure 8). The increase in sorted neuron viability with the addition GDNF and β-NGF is likely due to the neuroprotective influence of these factors on the cellular health of purified neurons being cultured in the absence of neurotrophin-secreting cell-types.

The use of cell-type specific promoter-driven fluorescent reporters has recently become a popular strategy to select specific types of neuronal cells from the wide variety of cell-types that arise during the directed differentiation of human pluripotent stem cells (hPSCs) into neurons. To date, transgenic labeling of human GABAergic neurons has involved using recombinant hESC reporter lines that have been designed for the purpose isolating cells expressing transcription factors that serve as medial ganglionic eminence (MGE)-specific determinants (e.g. NKX2.1) of cortical interneuron cell fate (Goulburn et al., 2011). The NKX2.1-GFP reporter hES cell line, and other reporter lines, are important tools with current applications in human-based studies aimed at paralleling in vitro cortical interneuron fate determinism with in vivo studies(Arber and Li, 2013; Germain et al., 2013). However, the use of this technology to study human cortical inhibitory neuron development and function is restricted to those specific cell lines and are not amenable for use in broader cell lines, such as studies looking at the role of inhibitory neurons in human genetic disease using multiple patient-specific hiPSC lines for example.

Another consideration is that in vivo, NKX2.1 expression spatially defines the MGE and the ventral caudal ganglionic eminence (CGE), distinguishing these regions from the lateral ganglion eminence (LGE) and the dorsal CGE (Arber and Li, 2013; Sussel et al., 1999). Because GABAergic cortical interneurons are a highly diverse population of cells comprised of over 20 known subtypes that originate from a number of different regions in the subpallium, the use of any single regional determinant as marker for cell sorting will only enable a proportion of the total GABAergic cortical interneuron population to be collected. Therefore, a panoptic marker may be useful in studies where the identification or isolation of the total population of GABAergic cortical neurons is desired.

In this study we demonstrate the ability of hVGAT-mCherry to effectively label hiPSC-derived GABAergic neurons. Notably, we show that the percentage of cells expressing mCherry was similar across several different GABAergic interneuron subtype populations with distinct in vivo developmental origins in the MGE (PV and SST) and CGE (CR). This implies that hVGAT-mCherry is capable of labeling a variety of cortical interneuron subtypes with different regional identities. Thus this reporter construct will provide a tool that has broad applications in studies of the functionality of GABAergic neurons derived from hPSCs both in normal neurodevelopment and in the study of neurological disorders using patient-specific iPSCs.

Because VGAT is also expressed in inhibitory glycinergic neurons, these cells should also be labeled by hVGAT-mCherry (Chaudhry et al., 1998; Dumoulin et al., 1999). Though the vast majority of forebrain inhibitory neurons are GABAergic, in other parts of the central nervous system, such as the in the hindbrain, the bulk of inhibitory neurons are glycinergic (Moore et al., 1996; Wang et al., 2009). Thus, pLV-hVGAT-mCherry could possibly serve as a tool for studying glycinergic neurons if other neuronal differentiation protocols are used which aim at specifying cellular identities associated with brain regions that are known to be dense with glycinergic cells.

The ability to differentiate patient-derived iPSCs into the various types of cells which become functionally compromised in neurological disease has, at least in part, spurred the development of highly efficient protocols for deriving specific types of neurons in vitro. Despite the recent methodological improvements, the differentiation of hPSCs in vitro inevitably results in cultures containing wide variety of neuronal and glial cell types. Considering the cellular heterogeneity observed in these cultures, removing superfluous cell-types prior to transcriptional or functional analysis would increase the credibility of results obtained from them. FACS has previously been applied in the isolation and analysis of relatively immature (<day 35) hPSC-derived neurons and progenitor cells. Utilizing FACS to purify neurons from cell populations that are more mature is known to result in low isolation rates of viable cells (Ladewig et al., 2008; Pruszak et al., 2007). Our results show that our modified FACS protocol can be applied to the purification of hPSC-derived neurons that have been continuously differentiated in culture for over 60 days. Using the methods described in this report, we were able to isolate highly pure populations of viable hiPSC-derived GABAergic neurons for prolonged study in vitro. Sorting day >60 cells collected from three to six wells of a 6-well culture plate would yield roughly 250,000 viable hVGAT-mCherry+ neurons. Naturally, greater cell yields could be obtained by scaling up the size of the culture to be sorted. Importantly, purified hVGAT-mCherry neurons remained viable in culture for over 30 days after sorting and expressed markers of GABAergic neurons.

Together with the techniques described for purifying hiPSC-derived neurons, we believe that the hVGAT-mCherry reporter construct provides a valuable tool to assist in examining the role inhibitory neurons play in pathogenesis across a broad range of neurological disorders. Furthermore, the results of this investigation suggest that hVGAT-mCherry-purified cells could possibly be used for the functional screening of pharmacological compounds for the identification of novel therapeutic treatments for neurological disease with impaired inhibitory neuron signaling.

Conclusion

The tremendous advancements in the stem cell-based modeling of neurological disorders has provided critical information to increase our understanding of a wide variety of neurological conditions. However, these experiments are hampered by the heterogeneity intrinsic to current in vitro differentiation approaches. Therefore, new tools that allow for the identification and isolation of specific cell types from these heterogeneous cultures are greatly needed. To that end, we describe the development of a novel cell type-specific fluorescent reporter systems that takes advantage of the human VGAT promoter to drive expression of the mCherry reporter gene. Coupled with this novel cell type-specific reagent, we have also developed a novel FACS protocol that allows for the isolation of a highly purified and viable hVGAT-mCherry positive population of cells. These purified cells are viable for extended lengths of time (>30 days) in culture making them amenable to a variety of functional and pharmacological experiments.

Supplementary Material

1
2
3
4
5
6
7

Highlights.

  • Development of human VGAT promoter-reporter lentiviral construct.

  • Expression of mCherry from the VGAT promoter was specific to GABAergic neurons.

  • Effective labeling of several different GABAergic interneuron subtypes.

  • High yields of viable GABAergic neurons using novel fluorescent activated cell sorting (FACS).

Acknowledgments

We are thankful to J. Van Baaren for technical support. This study was supported by grant R01 MH080647 (MAP-V) and P50NS071674 (JMV) from the NIH and funds from the Hussman Foundation (MAP-V, JMV, DMD).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Acosta MT, Pearl PL. The neurobiology of autism: new pieces of the puzzle. Current neurology and neuroscience reports. 2003;3:149–156. doi: 10.1007/s11910-003-0067-0. [DOI] [PubMed] [Google Scholar]
  2. Arber C, Li M. Cortical interneurons from human pluripotent stem cells: prospects for neurological and psychiatric disease. Frontiers in cellular neuroscience. 2013;7:10. doi: 10.3389/fncel.2013.00010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Barski A, Cuddapah S, Cui K, Roh TY, Schones DE, Wang Z, Wei G, Chepelev I, Zhao K. High-resolution profiling of histone methylations in the human genome. Cell. 2007;129:823–837. doi: 10.1016/j.cell.2007.05.009. [DOI] [PubMed] [Google Scholar]
  4. Bessis A, Salmon AM, Zoli M, Le Novere N, Picciotto M, Changeux JP. Promoter elements conferring neuron-specific expression of the beta 2-subunit of the neuronal nicotinic acetylcholine receptor studied in vitro and in transgenic mice. Neuroscience. 1995;69:807–819. doi: 10.1016/0306-4522(95)00303-z. [DOI] [PubMed] [Google Scholar]
  5. Bilic J, Izpisua Belmonte JC. Concise review: Induced pluripotent stem cells versus embryonic stem cells: close enough or yet too far apart? Stem cells. 2012;30:33–41. doi: 10.1002/stem.700. [DOI] [PubMed] [Google Scholar]
  6. Blackmore MG, Wang Z, Lerch JK, Motti D, Zhang YP, Shields CB, Lee JK, Goldberg JL, Lemmon VP, Bixby JL. Kruppel-like Factor 7 engineered for transcriptional activation promotes axon regeneration in the adult corticospinal tract. Proceedings of the National Academy of Sciences of the United States of America. 2012;109:7517–7522. doi: 10.1073/pnas.1120684109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Buchanan SS, Gross SA, Acker JP, Toner M, Carpenter JF, Pyatt DW. Cryopreservation of stem cells using trehalose: evaluation of the method using a human hematopoietic cell line. Stem cells and development. 2004;13:295–305. doi: 10.1089/154732804323099226. [DOI] [PubMed] [Google Scholar]
  8. Campeau E, Ruhl VE, Rodier F, Smith CL, Rahmberg BL, Fuss JO, Campisi J, Yaswen P, Cooper PK, Kaufman PD. A versatile viral system for expression and depletion of proteins in mammalian cells. PloS one. 2009;4:e6529. doi: 10.1371/journal.pone.0006529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Carninci P, Nishiyama Y, Westover A, Itoh M, Nagaoka S, Sasaki N, Okazaki Y, Muramatsu M, Hayashizaki Y. Thermostabilization and thermoactivation of thermolabile enzymes by trehalose and its application for the synthesis of full length cDNA. Proceedings of the National Academy of Sciences of the United States of America. 1998;95:520–524. doi: 10.1073/pnas.95.2.520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Chambers SM, Fasano CA, Papapetrou EP, Tomishima M, Sadelain M, Studer L. Highly efficient neural conversion of human ES and iPS cells by dual inhibition of SMAD signaling. Nature biotechnology. 2009;27:275–280. doi: 10.1038/nbt.1529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chaudhry FA, Reimer RJ, Bellocchio EE, Danbolt NC, Osen KK, Edwards RH, Storm-Mathisen J. The vesicular GABA transporter, VGAT, localizes to synaptic vesicles in sets of glycinergic as well as GABAergic neurons. The Journal of neuroscience : the official journal of the Society for Neuroscience. 1998;18:9733–9750. doi: 10.1523/JNEUROSCI.18-23-09733.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Clark AJ, Bissinger P, Bullock DW, Damak S, Wallace R, Whitelaw CB, Yull F. Chromosomal position effects and the modulation of transgene expression. Reproduction, fertility, and development. 1994;6:589–598. doi: 10.1071/rd9940589. [DOI] [PubMed] [Google Scholar]
  13. Dumoulin A, Rostaing P, Bedet C, Levi S, Isambert MF, Henry JP, Triller A, Gasnier B. Presence of the vesicular inhibitory amino acid transporter in GABAergic and glycinergic synaptic terminal boutons. Journal of cell science. 1999;112(Pt 6):811–823. doi: 10.1242/jcs.112.6.811. [DOI] [PubMed] [Google Scholar]
  14. Feng YQ, Lorincz MC, Fiering S, Greally JM, Bouhassira EE. Position effects are influenced by the orientation of a transgene with respect to flanking chromatin. Molecular and cellular biology. 2001;21:298–309. doi: 10.1128/MCB.21.1.298-309.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gasnier B. The loading of neurotransmitters into synaptic vesicles. Biochimie. 2000;82:327–337. doi: 10.1016/s0300-9084(00)00221-2. [DOI] [PubMed] [Google Scholar]
  16. Germain ND, Banda EC, Becker S, Naegele JR, Grabel LB. Derivation and isolation of NKX2.1-positive basal forebrain progenitors from human embryonic stem cells. Stem cells and development. 2013;22:1477–1489. doi: 10.1089/scd.2012.0264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Goulburn AL, Alden D, Davis RP, Micallef SJ, Ng ES, Yu QC, Lim SM, Soh CL, Elliott DA, Hatzistavrou T, Bourke J, Watmuff B, Lang RJ, Haynes JM, Pouton CW, Giudice A, Trounson AO, Anderson SA, Stanley EG, Elefanty AG. A targeted NKX2.1 human embryonic stem cell reporter line enables identification of human basal forebrain derivatives. Stem cells. 2011;29:462–473. doi: 10.1002/stem.587. [DOI] [PubMed] [Google Scholar]
  18. Hager GL, McNally JG, Misteli T. Transcription dynamics. Molecular cell. 2009;35:741–753. doi: 10.1016/j.molcel.2009.09.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Heintzman ND, Stuart RK, Hon G, Fu Y, Ching CW, Hawkins RD, Barrera LO, Van Calcar S, Qu C, Ching KA, Wang W, Weng Z, Green RD, Crawford GE, Ren B. Distinct and predictive chromatin signatures of transcriptional promoters and enhancers in the human genome. Nature genetics. 2007;39:311–318. doi: 10.1038/ng1966. [DOI] [PubMed] [Google Scholar]
  20. Hong S, Hwang DY, Yoon S, Isacson O, Ramezani A, Hawley RG, Kim KS. Functional analysis of various promoters in lentiviral vectors at different stages of in vitro differentiation of mouse embryonic stem cells. Mol Ther. 2007;15:1630–1639. doi: 10.1038/sj.mt.6300251. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Insel TR. Rethinking schizophrenia. Nature. 2010;468:187–193. doi: 10.1038/nature09552. [DOI] [PubMed] [Google Scholar]
  22. Jin H, Wu H, Osterhaus G, Wei J, Davis K, Sha D, Floor E, Hsu CC, Kopke RD, Wu JY. Demonstration of functional coupling between gamma -aminobutyric acid (GABA) synthesis and vesicular GABA transport into synaptic vesicles. Proceedings of the National Academy of Sciences of the United States of America. 2003;100:4293–4298. doi: 10.1073/pnas.0730698100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kozubenko N, Turnovcova K, Kapcalova M, Butenko O, Anderova M, Rusnakova V, Kubista M, Hampl A, Jendelova P, Sykova E. Analysis of in vitro and in vivo characteristics of human embryonic stem cell-derived neural precursors. Cell transplantation. 2010;19:471–486. doi: 10.3727/096368909X484707. [DOI] [PubMed] [Google Scholar]
  24. Ladewig J, Koch P, Endl E, Meiners B, Opitz T, Couillard-Despres S, Aigner L, Brustle O. Lineage selection of functional and cryopreservable human embryonic stem cell-derived neurons. Stem cells. 2008;26:1705–1712. doi: 10.1634/stemcells.2008-0007. [DOI] [PubMed] [Google Scholar]
  25. Lewis DA. Inhibitory neurons in human cortical circuits: substrate for cognitive dysfunction in schizophrenia. Current opinion in neurobiology. 2014;26:22–26. doi: 10.1016/j.conb.2013.11.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lewis DA, Levitt P. Schizophrenia as a disorder of neurodevelopment. Annual review of neuroscience. 2002;25:409–432. doi: 10.1146/annurev.neuro.25.112701.142754. [DOI] [PubMed] [Google Scholar]
  27. Lobo MK, Karsten SL, Gray M, Geschwind DH, Yang XW. FACS-array profiling of striatal projection neuron subtypes in juvenile and adult mouse brains. Nature neuroscience. 2006;9:443–452. doi: 10.1038/nn1654. [DOI] [PubMed] [Google Scholar]
  28. Lui JH, Hansen DV, Kriegstein AR. Development and evolution of the human neocortex. Cell. 2011;146:18–36. doi: 10.1016/j.cell.2011.06.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Maherali N, Hochedlinger K. Guidelines and techniques for the generation of induced pluripotent stem cells. Cell stem cell. 2008;3:595–605. doi: 10.1016/j.stem.2008.11.008. [DOI] [PubMed] [Google Scholar]
  30. Moore JK, Osen KK, Storm-Mathisen J, Ottersen OP. gamma-Aminobutyric acid and glycine in the baboon cochlear nuclei: an immunocytochemical colocalization study with reference to interspecies differences in inhibitory systems. The Journal of comparative neurology. 1996;369:497–519. doi: 10.1002/(SICI)1096-9861(19960610)369:4<497::AID-CNE2>3.0.CO;2-#. [DOI] [PubMed] [Google Scholar]
  31. Narsinh KH, Sun N, Sanchez-Freire V, Lee AS, Almeida P, Hu S, Jan T, Wilson KD, Leong D, Rosenberg J, Yao M, Robbins RC, Wu JC. Single cell transcriptional profiling reveals heterogeneity of human induced pluripotent stem cells. The Journal of clinical investigation. 2011;121:1217–1221. doi: 10.1172/JCI44635. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Nathanson JL, Yanagawa Y, Obata K, Callaway EM. Preferential labeling of inhibitory and excitatory cortical neurons by endogenous tropism of adeno-associated virus and lentivirus vectors. Neuroscience. 2009;161:441–450. doi: 10.1016/j.neuroscience.2009.03.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Nelson SB, Hempel C, Sugino K. Probing the transcriptome of neuronal cell types. Current opinion in neurobiology. 2006;16:571–576. doi: 10.1016/j.conb.2006.08.006. [DOI] [PubMed] [Google Scholar]
  34. Nicholas CR, Chen J, Tang Y, Southwell DG, Chalmers N, Vogt D, Arnold CM, Chen YJ, Stanley EG, Elefanty AG, Sasai Y, Alvarez-Buylla A, Rubenstein JL, Kriegstein AR. Functional maturation of hPSC-derived forebrain interneurons requires an extended timeline and mimics human neural development. Cell stem cell. 2013;12:573–586. doi: 10.1016/j.stem.2013.04.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Pizzarelli R, Cherubini E. Alterations of GABAergic signaling in autism spectrum disorders. Neural plasticity. 2011;2011:297153. doi: 10.1155/2011/297153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Pruszak J, Sonntag KC, Aung MH, Sanchez-Pernaute R, Isacson O. Markers and methods for cell sorting of human embryonic stem cell-derived neural cell populations. Stem cells. 2007;25:2257–2268. doi: 10.1634/stemcells.2006-0744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Rubenstein JL, Merzenich MM. Model of autism: increased ratio of excitation/inhibition in key neural systems. Genes, brain, and behavior. 2003;2:255–267. doi: 10.1034/j.1601-183x.2003.00037.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Sagne C, El Mestikawy S, Isambert MF, Hamon M, Henry JP, Giros B, Gasnier B. Cloning of a functional vesicular GABA and glycine transporter by screening of genome databases. FEBS letters. 1997;417:177–183. doi: 10.1016/s0014-5793(97)01279-9. [DOI] [PubMed] [Google Scholar]
  39. Saxena A, Wagatsuma A, Noro Y, Kuji T, Asaka-Oba A, Watahiki A, Gurnot C, Fagiolini M, Hensch TK, Carninci P. Trehalose-enhanced isolation of neuronal sub-types from adult mouse brain. BioTechniques. 2012;52:381–385. doi: 10.2144/0000113878. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A. Fiji: an open-source platform for biological-image analysis. Nature methods. 2012;9:676–682. doi: 10.1038/nmeth.2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Spencer KM, Nestor PG, Perlmutter R, Niznikiewicz MA, Klump MC, Frumin M, Shenton ME, McCarley RW. Neural synchrony indexes disordered perception and cognition in schizophrenia. Proceedings of the National Academy of Sciences of the United States of America. 2004;101:17288–17293. doi: 10.1073/pnas.0406074101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Spitzer M, Wildenhain J, Rappsilber J, Tyers M. BoxPlotR: a web tool for generation of box plots. Nature methods. 2014;11:121–122. doi: 10.1038/nmeth.2811. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Sussel L, Marin O, Kimura S, Rubenstein JL. Loss of Nkx2.1 homeobox gene function results in a ventral to dorsal molecular respecification within the basal telencephalon: evidence for a transformation of the pallidum into the striatum. Development. 1999;126:3359–3370. doi: 10.1242/dev.126.15.3359. [DOI] [PubMed] [Google Scholar]
  44. Uematsu M, Hirai Y, Karube F, Ebihara S, Kato M, Abe K, Obata K, Yoshida S, Hirabayashi M, Yanagawa Y, Kawaguchi Y. Quantitative chemical composition of cortical GABAergic neurons revealed in transgenic venus-expressing rats. Cerebral cortex. 2008;18:315–330. doi: 10.1093/cercor/bhm056. [DOI] [PubMed] [Google Scholar]
  45. Wang Y, Kakizaki T, Sakagami H, Saito K, Ebihara S, Kato M, Hirabayashi M, Saito Y, Furuya N, Yanagawa Y. Fluorescent labeling of both GABAergic and glycinergic neurons in vesicular GABA transporter (VGAT)-venus transgenic mouse. Neuroscience. 2009;164:1031–1043. doi: 10.1016/j.neuroscience.2009.09.010. [DOI] [PubMed] [Google Scholar]
  46. Watanabe M, Maemura K, Kanbara K, Tamayama T, Hayasaki H. GABA and GABA receptors in the central nervous system and other organs. International review of cytology. 2002;213:1–47. doi: 10.1016/s0074-7696(02)13011-7. [DOI] [PubMed] [Google Scholar]
  47. Xu X, Wells AB, O'Brien DR, Nehorai A, Dougherty JD. Cell type-specific expression analysis to identify putative cellular mechanisms for neurogenetic disorders. The Journal of neuroscience : the official journal of the Society for Neuroscience. 2014;34:1420–1431. doi: 10.1523/JNEUROSCI.4488-13.2014. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS718808-supplement-1.TIF (118.2KB, TIF)
2
NIHMS718808-supplement-2.tif (11.7MB, tif)
3
NIHMS718808-supplement-3.tif (5.5MB, tif)
4
NIHMS718808-supplement-4.tif (3.5MB, tif)
5
NIHMS718808-supplement-5.tif (5.8MB, tif)
6
NIHMS718808-supplement-6.tif (5.8MB, tif)
7
NIHMS718808-supplement-7.tif (4.4MB, tif)

RESOURCES