Stem Cell

Combined small-molecule treatment accelerates maturation of human pluripotent stem cell-derived neurons – Nature Biotechnology

Time Stamp: January 1, 2024 7:00 PM
Source Node: 439980

Experimental details are summarized in Supplementary Table 2.

Cell culture

Human PSCs

PSCs, both embryonic and induced, were maintained in Essential 8 medium (Thermo Fisher Scientific) on vitronectin-coated plates as previously described⁶⁴. All stem cell work was conducted according to protocols approved by the Tri-Institutional Stem Cell Initiative Embryonic Stem Cell Research Oversight Committee (Tri-SCI ESCRO).

Human PSC-derived excitatory cortical neurons

Cortical neurons were generated using a protocol based on the previously described dual-SMAD inhibition paradigm⁶⁵. Briefly, hESCs were dissociated into single cells with Accutase and seeded at 250,000 cm⁻² on to Matrigel-coated plates in Essential 8 medium with 10 μM Y-27632. During days 1–10 of the protocol, medium consisted of Essential 6 (Thermo Fisher Scientific) with 10 μM SB431542 (Tocris) and 100 nM LDN193189 (Stemgent). Wnt inhibitor XAV-939 at 2 μM was included from day 1 to day 3 to improve anterior patterning⁶⁶. On days 11–20, medium consisted of N-2-supplemented Dulbecco’s modified Eagle’s medium and Ham’s F-12 nutrient mixture (DMEM/F12; Thermo Fisher Scientific). Cells received daily medium exchanges throughout the differentiation. On day 20 cells were dissociated in Accutase for 30 min and plated on poly(l-ornithine) and laminin-coated (PLO/Lam) plates, in low-glucose (5 mM) Neurobasal A medium supplemented with 2% B27 and 1% GlutaMAX (Thermo Fisher Scientific) or cryopreserved in STEM-CELLBANKER solution (Amsbio). Neurons received medium exchanges twice a week. During the first 7 d after plating, medium was supplemented with notch-inhibitor DAPT (N-(N-(3,5-difluorophenacetyl)-l-alanyl)-s-phenylglycinet-butyl ester, a γ-secretase inhibitor) at 10 μM (ref. ⁶⁷). Long-term cultures were maintained with BDNF (10 ng ml⁻¹ of PeproTech), glial cell line-derived neurotrophic factor (GDNF; 10 ng ml⁻¹, R&D Biosystems), dibutyryl cAMP (100 μM, Sigma-Aldrich) and ascorbic acid (100 μM, Sigma-Aldrich). However, owing to the activation of immediate early genes by BDNF and cAMP, the initial screen and validation experiments involving IEG induction were done in the absence of those factors.

Primary embryonic rat cortical neurons

Cells were obtained from Thermo Fisher Scientific, thawed following the vendor’s instructions and maintained in the same manner as hPSC cortical neurons.

SMNs

Motoneuron derivation was adapted from a previously described protocol⁶⁸. In brief, Accutase-dissociated hESCs were seeded at 600,000 cm⁻² on to Geltrex-coated plates and underwent dual-SMAD inhibition in the presence of CHIR99021 and Smoothened agonist. On day 11, spinal progenitors were collected and plated on PDL/Lam/FN plates and maintained in N-2/B27 medium containing Smoothened agonist, retinoic acid, BDNF, GDNF, ciliary neurotrophic factor (CTNF) and DAPT. On day 24, SMNs were re-plated on PDL/Lam/FN and maintained in Neurobasal medium supplemented with 2% B27, ascorbic acid, retinoic acid, BDNF, GDNF and CTNF. Treatment with GENtoniK or DMSO was initiated the day after re-plating.

Dorsal forebrain organoids

Organoids were derived from a previously reported protocol⁶⁹. Briefly, 10,000 EDTA-dissociated hPSCs were plated per well of a 96-well, V-bottomed, low-attachment plate (S-bio). Cells were allowed to self-aggregate in hPSC growth medium overnight. From day 1 to day 8, medium was changed every 2 d with Essential 6 medium supplemented with 10 μM SB431542, 100 nM LDN193189 and 2 μM XAV-939. On day 8, the medium was switched to organoid growth medium consisting of a 50:50 mixture of Neurobasal and DMEM/F12 medium with 1% NeuroBrew 21 (Miltenyi), 0.5% N2, 1% GlutaMAX, 0.5% minimal essential medium (MEM) nonessential amino acid (NEAA) solution, 0.1% 2-mercaptoethanol and 1 μM recombinant human insulin (Sigma-Aldrich). On day 14 organoids were transferred to 10-cm dishes at roughly 20 organoids per dish and placed on an orbital shaker set to gentle motion to prevent organoid fusion.

Melanocytes

Melanocyte differentiation was executed as previously reported⁷⁰. In brief, the day before differentiation, hPSCs were plated on Matrigel at 200,000 cells cm⁻² in Essential 8 medium with 10 μM Y-27632. From day 0 to day 11 of the protocol, cells received daily exchanges of Essential 6 medium containing: 1 ng ml⁻¹ of bone morphogenetic protein (BMP)4, 10 μM SB431542 and 600 nM CHIR99021 (days 0–2); 10 μM SB431542 and 1.5 μM CHIR99021 (days 2–4); 1.5 μM CHIR99021 (days 4–6); and 1.5 μM CHIR99021, 5 ng ml⁻¹ of BMP4 and 100 nM EDN3 (days 6–11). On day 11, cells were dissociated into single cells with Accutase for 20 min and cKIT⁺ melanoblasts were sorted using an allophycocyanin-conjugated antibody (Thermo Fisher Scientific) on a BD-FACS Aria cell sorter at the Memorial Sloan–Kettering Cancer Center (MSKCC) Flow Cytometry Core Facility, and re-plated on to dried PO/Lam/FN dishes. Cells were fed with melanocyte medium (Neurobasal medium supplemented with: 50 ng ml⁻¹ of stem cell factor, 500 μM cAMP, 10 ng ml⁻¹ of fibroblast growth factor (FGF)2, 3 μM CHIR99021, 25 ng ml⁻¹ of BMP4, 100 nM EDN3, 1 mM l-glutamine, 0.1 mM MEM NEAA solution, 2% B27 and 2% N-2) every 2–3 d and passaged using Accutase at a ratio of 1:4 once a week.

Pancreatic β-cells

β-Cell differentiation was performed using INS^GFP/W MEL-1 cells. Cells were maintained on Matrigel-coated plates in StemFlex medium (Thermo Fisher Scientific) at 37 °C with 5% CO₂. MEL-1 cells were differentiated using a previously reported strategy⁷¹. Briefly, on day 0, cells were exposed to basal Roswell Park Memorial Institute (RPMI) 1640 medium (Corning) supplemented with 1× GlutaMAX (Thermo Fisher Scientific), 50 μg ml⁻¹ of Normocin, 100 ng ml⁻¹ of activin A (R&D systems) and 3 μM CHIR99021 (Cayman Chemical) for 24 h. The medium was changed on day 2 to basal RPMI 1640 medium supplemented with 1× GlutaMAX, 50 μg ml⁻¹ of Normocin, 0.2% fetal bovine serum (FBS; Corning) and 100 ng ml⁻¹ of activin A for 2 d. On day 4, the resulting definitive endoderm cells were cultured in MCDB131 medium supplemented with 1.5 g l⁻¹ of sodium bicarbonate, 1× GlutaMAX, 10 mM glucose, 2% bovine serum albumin (BSA), 50 ng ml⁻¹ of FGF7 and 0.25 mM ascorbic acid for 2 d. On day 6, the cells were differentiated in MCDB131 medium supplemented with 2.5 g l⁻¹ of sodium bicarbonate, 1× GlutaMAX, 10 mM glucose, 2% BSA, 0.25 mM ascorbic acid, 2 μM retinoic acid, 0.25 μM SANT1, 50 ng ml⁻¹ of FGF7, 200 nM tetraphenyl butadiene (TPB), 200 nM LDN193189 and 0.5× insulin–transferrin–selenium–ethanolamine (ITS-X) supplement for 2 d to pancreatic progenitor stage 1 cells. On day 8, the cells were directed to pancreatic progenitor stage 2 cells in MCDB131 medium supplemented with 2.5 g l⁻¹ of sodium bicarbonate, 1× GlutaMAX, 10 mM glucose, 2% BSA, 0.25 mM ascorbic acid, 0.2 μM retinoic acid, 0.25 μM SANT1, 2 ng ml⁻¹ of FGF7, 100 nM TPB, 400 nM LDN193189 and 0.5× ITS-X supplement for 3 d. On day 11, the cells were directed to insulin-expressing cells in MCDB131 medium supplemented with 1.5 g l⁻¹ of sodium bicarbonate, 1× GlutaMAX, 20 mM glucose, 2% BSA, 0.1 μM retinoic acid, 0.25 μM SANT1, 200 nM LDN193189, 1 μM triiodothyronine (T₃), 10 μM ALKi5, 10 μM zinc sulfate, 10 μg ml⁻¹ of heparin and 0.5× ITS-X for 3 d. On day 14, the cells for static or dynamic KCl-stimulated insulin secretion (KSIS) analysis were scraped off from plates and relocated on to 24-mm insert and 3.0-μm polycarbonate membrane, six-well tissue culture, trans-well plate into hemispherical colonies, and the cells for insulin content analysis and flow cytometry analysis were kept on the original plates. All the cells were then further matured in MCDB131 medium supplemented with 1.5 g l⁻¹ of sodium bicarbonate, 1× GlutaMAX, 20 mM glucose, 2% BSA, 100 nM LDN193189, 1 μM triiodothyronine (T3), 10 μM zinc sulfate, 10 μg ml⁻¹ of heparin, 100 nM GS and 0.5× ITS-X for 7 d. Then cells were further matured in MCDB131 medium supplemented with 1.5 g l⁻¹ of sodium bicarbonate, 1× GlutaMAX, 20 mM glucose, 2% BSA, 1 μM T₃, 10 μM zinc sulfate, 10 μg ml⁻¹ of heparin, 1 mM acetylcysteine, 10 μM Trolox, 2 μM R428 and 0.5× ITS-X with GENtoniK or control treatment for 7 d.

Small-molecule treatment

A bioactive compound library containing 2,688 compounds was used for screening at a concentration of 5 μM (Selleck Bioactive Library, Selleck Chemicals). For confirmation of primary hits, compounds were extracted from the library plates with a JANUS (Perkin Elmer) liquid-handling platform and re-subjected to the high-content assay in triplicates at 5 μM. The 22 confirmed compounds were purchased from Selleck Chemicals, reconstituted in a suitable solvent and applied for dose–response validation in a concentration log scale (30 nM, 100 nM, 300 nM, 1000 nM, 3000 nM, 10,000 nM). GENtoniK cocktail was defined as a mixture of four small molecules—GSK2879552, EPZ-5676, Bay K 8644 and NMDA—applied at a working concentration of 1 μM each. Stocks of individual GENtoniK ingredients were reconstituted in DMSO to 10 mM (GSK2879552, EPZ-5676 and Bay K 8644) or in water to 50 mM (NMDA) and stored at −20 °C until the day of the experiments.

Immunostaining

Monolayer cultures

Cells were fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS) for 30 m, permeabilized for 5 min in PBS with 0.1% Triton X-100, and blocked for 30 min in PBS with 5% normal goat serum (NGS). Incubation with primary antibodies was performed overnight at 4 °C at the specified dilution in PBS with 2% NGS. After three washes with PBS, cells were incubated with fluorescently conjugated secondary antibodies (2 µg ml⁻¹) and DAPI (1 µg µl⁻¹) for 30 min at room temperature. For high-content experiments, all steps were assisted by automated liquid handling at the MSKCC Gene Editing and Screening Core Facility. A list of antibodies used in the present study is presented in Supplementary Table 3.

Forebrain organoids

Organoids were collected in 1.5-ml centrifuge tubes, washed in PBS and fixed with 4% paraformaldehyde solution in PBS overnight at 4 °C. Fixed organoids were rinsed in PBS and equilibrated in a solution of 30% w:v sucrose in PBS for 24 h or until they sank to the bottom of the tube. Organoids were embedded in optimal cutting temperature compound (Thermo Fisher Scientific) on cryomolds, flash frozen and sectioned to a thickness of 30 μm on a cryostat. Sections were collected in 1-ml centrifuge tubes, washed in Tris-buffered saline (TBS) with 0.3% Triton X-100 and blocked in the same solution with 10% NGS. Primary antibody incubation was done overnight in TBS with 0.5% Tween-20, followed by washes and secondary antibody incubation for 2 h at room temperature in the same buffer. Sections were mounted on slides with ProLong medium (Thermo Fisher Scientific) and imaged on a Zeiss microscope equipped with a ×20 high numerical aperture objective and an Apotome optical sectioning system (Zeiss). For quantification of SYN1 puncta, images were batch analyzed using the Synapse Counter ImageJ plugin⁷².

HCS

High-content maturity assay

Cortical neurons were seeded on PLO/Lam-coated, 384-well plates at a density of 5,000 per well and maintained as described. For bioactive compound screening, compounds were added in fresh medium 7 d after plating to a final concentration of 5 μM in replicate plates. After 7 d of treatment, cells were rinsed twice and maintained in plain medium for an additional 7 d. Before fixation, one replicate plate was stimulated with 50 mM KCl for 2 h. Immunostaining for FOS, EGR-1 and MAP2 and counterstaining with DAPI were performed as described above. Images (four fields per well at ×20 magnification) were captured through an InCell Analyzer 6000 HCA system (GE Healthcare).

Image analysis and quantification of screen results

Phenotypic analysis of screen images was conducted in automated and blinded fashion using the Columbus software (Perkin Elmer). Extracted parameters included total number of nuclei, nuclear area, nuclear roundness index (DAPI), total neurite length per nucleus (MAP2) and fraction of FOS⁺, EGR-1⁺ and double-IEG-positive nuclei (FOS⁺/EGR-1⁺). For IEG quantification, ratios of positive nuclei were calculated by applying a threshold to arbitrary fluorescence units (AFU) within DAPI-positive nuclei (that is, >800 AFU for FOS, >1,000 AFU for EGR-1). IEG nuclei ratios in unstimulated plates were then subtracted from KCl-stimulated plates to isolate the KCl depolarization-mediated response. Morphological variables (nuclear and neurite) were averaged between unstimulated and KCl plates. Sequential b-score and z-score normalization and PCA were performed in the KNIME analytics platform⁷³ with the High Content Screening Tools extension.

Synaptic marker analysis

Human PSC cortical neurons were plated on PLO/Lam 96-well plates. Drug treatment was initiated after 7 d and maintained for 21 d. Cells were fixed after an additional 7 d in plain medium. Immunostaining for SYN1, PSD95 and MAP2 was conducted as described above. Ten images per well were captured using the confocal modality of the InCell 6000 HCA system. A mask was applied to the area surrounding MAP2⁺ processes, and SYN1 and PSD95 puncta were quantified within the defined region. For quantification of pre- and postsynaptic marker apposition, a mask was applied to an area containing and immediately surrounding SYN1 puncta and PSD95 puncta localized within this region were counted. Synaptic puncta counts per field were normalized to total neurite length.

Cell viability assays

Cytotox Red assay

Live imaging for quantification of live/dead cells was conducted using an Incucyte S3 platform (Sartorius) in cortical neurons plated on 96-well plates. Small-molecule treatment and loading with 250 nM Cytotox Red dye were initiated on day 7 after plating and replenished every 7 d. Images were captured every 2 h for 21 d. Analysis was performed using the Incucyte software.

Presto Blue assay

Presto Blue viability reagent (Thermo Fisher Scientific) was diluted in Neurobasal medium at 1:10. Viability experiments were performed in 96-well, black wall/clear bottomed plates. Then, 85 μl of Presto Blue/NB solution per well was incubated at 37 °C 5% CO₂ for 2 h, after which 75 μl was transferred to a new plate for absorbance readings at 570 nm and 600 nm. Results were calculated by subtracting the 600-nm absorbance values from the 570-nm values, followed by subtracting background (600 − 570 nm) readings (read from blank wells) from experimental samples. GENtoniK and DMSO-treated neurons were compared at 1, 7 and 14 d of treatment to determine relative viability.

Western blotting

Cells were harvested in radioimmunoprecipitation assay (RIPA) buffer and protein content quantified with Precision Red Assay. Protein, 10 μg, was loaded per lane, separated using a sodium dodecylsulfate–polyacrylamide gel electrophoresis system, and transferred on to a poly(vinylidene fluoride) membrane. A Gel Doc imaging system was used to visualize the blots. For visualizing 4R tau, a SuperSignal West Femto Maximum Sensitivity ECL Substrate (Thermo Fisher Scientific) was used. Quantification of band densities was performed in Image Lab and ImageJ.

Shallow WGS and CNA analysis

Genomic DNA was collected using the Quick-DNA Miniprep Kit (Zymo) and submitted to the MSKCC Integrated Genomics Operation core for library preparation and shallow WGS. Paired-end reads were aligned to the GRCh19 reference human genome using BowTie2. Somatic CNA analysis was performed in R using the CNAclinic package (https://github.com/sdchandra/CNAclinic).

Electrophysiology

Whole-cell patch-clamp

Human PSC cortical neurons were plated on to PLO/Lam-coated 35-mm dishes at a density of 75,000 cm⁻². Treatment with GENtoniK or DMSO began 7 d after plating and maintained for 14 d. Recordings were initiated 7 d after treatment withdrawal, within 28–33 d of plating. Whole-cell recordings were performed at 23–24 °C while the cells were perfused in freshly made artificial cerebrospinal fluid (ACSF) containing (in mM): 125 NaCl, 2.5 KCl, 1.2 NaH₂PO₄, 1 MgSO₄, 2 CaCl₂, 25 NaHCO₃ and 10 d-glucose. Solutions were pH corrected to 7.4 and 300–310 mOsmol. Neurons were recorded with pipettes of 3- to 7-MΩ resistance filled with a solution containing (in mM): 130 potassium gluconate, 4 KCl, 0.3 (ethylenebis(oxonitrilo))tetra-acetate (EGTA), 10 Na₂ phosphocreatine, 10 Hepes, 4 Mg₂ ATP, 0.3 Na₂ GTP and 13 biocytin, pH adjusted to 7.3 with KOH and osmolarity to 285–290 mOsmol kg⁻¹. Recordings were performed on a computer-controlled amplifier (MultiClamp 700B Axon Instruments) and acquired with an AxoScope 1550B (Axon Instruments) at a sampling rate of 10 kHz and low-pass filtered at 1 kHz. Recordings for sEPSCs were performed at 28–30 °C in a standard oxygenated ACSF. Neurons were recorded with pipettes of 3- to 7-MΩ resistance filled with a solution containing (in mM): 120 CsMeSO₄, 8 NaCl, 0.3 EGTA, 10 tetraethylammonium Cl, 10 Hepes, 2 Mg₂ ATP, 0.3 Na₂ GTP, 13 biocytin and 3 QX-314-Cl, pH adjusted to 7.3 with KOH and osmolarity to 285–290 mOsmol kg⁻¹. Membrane potentials were held at −60 mV or 0 mV to isolate AMPAR- or NMDAR-mediated events, respectively.

Multielectrode array recording

The hPSC-derived SMNs were seeded on to poly(l-lysine)-coated, complementary, metal oxide semiconductor, multielectrode array (CMOS-MEA) probes (3Brain)⁷⁴. A 100-μl droplet of medium containing 200,000 neurons was placed on the recording area. After a 1-h incubation, 1.5 ml of medium was added to the probe and replaced every 3 d. Cells received treatment with GENtoniK or DMSO over days 3–9 after plating. Recordings were performed every 3 d for 18 d, 24 h after medium changes. Spontaneous activity (1 min) was sampled from 4,096 electrodes using the BioCAM system and analyzed using BrainWave 4 software. Spikes were detected using a precise timing spike detection algorithm⁷⁵ on the raw channel traces, applying a threshold for detection of 9 s.d. Network bursts were detected by applying a hard threshold of 1 spike s⁻¹ on the entire 4,096-channel array.

Calcium imaging

Forebrain organoids were collected in 0.6-ml centrifuge tubes and rinsed 3× with Hepes-buffered Hanks’ balanced salt solution. Organoids were incubated in dye-loading solution consisting of 5 μM Fluo-4 AM (Thermo Fisher Scientific) with 0.1% Pluronic F-127 (Sigma-Aldrich), at 37 °C on an orbital shaker for 1 h. After incubation, organoids were rinsed in imaging solution consisting of modified Tyrode medium, as described previously⁷⁶. Organoids were individually mounted on microscope slides fitted with customized adapters and sealed with no. 1.5 glass coverslips. Imaging was performed using a Nikon A1R HD25 confocal laser-scanning microscope equipped with a ×20 multiple immersion objective and a Tokai Hit stagetop incubator for temperature and CO₂ control. Frames were captured every 5 s for 20 min (240 frames). Data analysis was performed using the CALIMA open-source software⁷⁷.

Gene expression and chromatin profiling

RNA-seq

RNA was extracted using the Direct-zol RNA miniprep kit (Zymo). Total RNA samples were submitted to GENEWIZ for paired-end sequencing at 30–40 million reads. Analysis was conducted on the Galaxy platform⁷⁸. Transcript quantification was performed directly from adapter-trimmed FASTQ files using the Salmon quasi-mapping tool⁷⁹ referenced to GENCODE Release 36 (GRCh38.p13) transcripts. DESeq2 (ref. ⁸⁰) was used for differential expression analysis from Salmon-generated transcript per million (TPM) values. Differentially expressed genes with a Benjamini–Hochberg-adjusted P value <0.05 and a baseMean cutoff of 1,000 were applied to gene-set overrepresentation analysis using the Goseq tool⁸¹. For gene-set enrichment analysis (GSEA), all genes with a baseMean >1,000 were analyzed using the GSEA software⁸².

CUT&RUN

The hPSC-derived cortical neurons were collected 7 d after plating for CUT&RUN (cleavage under targets & release using nuclease) chromatin profiling using the standard protocol⁸³. Antibodies against H3K4me2 (Upstate), H3K79me2 (Active Motif) and mouse immunoglobulin (Ig)G (Abcam) were used at 1:100 for 100,000 cells per antibody. DNA was collected via phenol–chloroform extraction and submitted to the MSKCC Integrated Genomics Operation core for paired-end sequencing at 5 million reads. Analysis was performed in the Galaxy platform. After alignment to ENSEMBL GRCh38 genome build using Bowtie2 (ref. ⁸⁴), peaks were called using MACS⁸⁵ and visualized with ChIPSeeker⁸⁶ and deepTool2 (ref. ⁸⁷), with mouse IgG as a control for normalization.

Flow cytometry

The hPSC neurons were dissociated to single-cell suspensions using Accutase (Innovative Cell Technologies) supplemented with Neuron Isolation Enzyme for Pierce (Thermo Fisher Scientific) solution at 1:50. Single-cell suspensions were stained with Zombie UV Fixable Viability Kit (BioLegend) at 1:2,500 in PBS for 15 min (room temperature), followed by fixation in 4% paraformaldehyde for 10 min (4 °C), then permeabilized in 0.5% Triton X-100 for 10 min (4 °C) and blocked in 5% BSA for 10 min (4 °C). Cells were stained with H3K9me3-PE antibody (Cell Signaling Technologies) diluted 1:200 for 30 min at 4 °C and acquisition was performed on the Cytek Aurora Spectral Flow Cytometer, with data analyzed on FlowJo v.10.8.1.

Dot blot for melanocyte pigmentation

The hESC melanocytes were dissociated in Accutase, rinsed and collected in PBS. A pellet containing 1 M cells was lysed in 50 μl of RIPA buffer with sonication and centrifuged at 10,000g for 3 min. After discarding the supernatant, the insoluble fraction was resuspended in 80 μl of PBS. Then, 10 μl of this solution was applied to a nitrocellulose membrane, air dried and imaged with a standard office scanner to assess pigmentation.

Pancreatic β-cell maturation assays

Flow cytometry analysis

The hESC-derived cells were dissociated using Accutase, fixed and permeabilized using Fixation/Permeabilization Solution Kit (BD Biosciences) according to the manufacturer’s instructions. Briefly, cells were first fixed with fixation/permeabilization buffer for 30 min at 4 °C in the dark and then washed twice with washing buffer with a 10-min incubation each time at room temperature. Then, the fixed cells were incubated with primary antibody overnight at 4 °C and washed twice with washing buffer with a 10-min incubation each time at room temperature. After a 30-min incubation with fluorescence-conjugated secondary antibody at 4 °C, cells were washed twice with washing buffer with a 10-min incubation each time at room temperature and resuspended in PBS buffer for analysis. The following primary antibodies were used: anti-insulin (1:50, Dako) and anti-glucagon (1:100, Abcam). Samples were analyzed with an Accuri C6 flow cytometry instrument and the data were processed using FlowJo v.10 software.

Static and dynamic KSIS

On day 30, cells were starved in 2 ml of glucose-free, pancreatic β-cell maturation medium and followed by 2 ml of glucose-free DMEM (with GlutaMAX) for 1 h and an additional 1-h incubation in KRBH buffer (containing 140 mM NaCl, 3.6 mM KCl, 0.5 mM NaH₂PO₄, 0.2 mM MgSO₄, 1.5 mM CaCl₂, 10 mM Hepes, pH 7.4, 2 mM NaHCO₃ and 0.1% BSA) in a 5% CO₂/37 °C incubator. To perform static KSIS, cells were exposed sequentially to 100 μl of KRBH with 2 mM glucose or 2 mM glucose with 30 mM KCl; supernatants were collected after 60 min and spun down to eliminate the cells and debris. Supernatants were used for ELISA (Insulin Chemiluminescence ELISA Jumbo). To measure the total insulin levels in cells in each sample, cells were lysed in RIPA buffer supplemented with 1× Protease Inhibitor Cocktail (Thermo Fisher Scientific) with vortexing for 2 min at room temperature and flash freezing the samples in liquid nitrogen and thawing to help lysis and release the cellular insulin. Lysates were spun down and the supernatant was used for ELISA. Insulin secretion from cells in each condition was normalized to KRBH treatment. To perform dynamic KSIS, cells were embedded in chambers with the order of a filter paper–biogel P4 beads–cells–biogel P4 beads sandwich and then the chambers were installed on the biorep perfusion system (Biorep Technology) and first perfused with Krebs’ buffer containing 2 mM glucose at a flow rate of 100 μl min⁻¹ and followed by perfusion with 2 mM glucose + 30 mM KCl for 25 min. Insulin secretion from cells in each fraction in KCl stimulation was normalized to KRBH treatment.

Insulin content measurement

Day-30 hESC-derived β-like cells were dissociated using Accutase and resuspended in DMEM containing 2% FBS and 1 mM EDTA. INS-GFP⁺DAPI⁻ cells, 80,000, were FACS sorted by an ARIA2 instrument, washed once with PBS and lysed in 200 µl of RIPA buffer supplemented with 1× Protease Inhibitor Cocktail (Thermo Fisher Scientific). The insulin content was measured by ELISA.

Immunoelectron microscopy

To analyze granular ultrastructure, control or chemically treated, hPSC-derived, β-like cell clusters were washed with serum-free medium and fixed with 2.5% glutaraldehyde, 4% paraformaldehyde and 0.02 % picric acid in 0.1 M buffer. After three buffer washes, the cell clusters were fixed again using 1% OsO₄–1.5% potassium ferricyanide at room temperature for 60 min, followed by three buffer washes. After dehydration steps of 50%, 70%, 85%, 95%, 100%, 100% and 100% EtOH, the cell clusters were infiltrated with 100% EtOH mixed 1:1 with acetonitrile, followed by acetonitrile, acetonitrile 1:1 with Embed 812 epoxy resin, resin and, finally, embedded in fresh resin which was polymerized at 50 °C for 36 h. Sections were cut at 65 nm and picked up on nickel grids. Sections were washed with saturated sodium periodate, followed by 50 mM glycine and blocking buffer. Then, the sections were stained with anti-insulin antibody at the original dilution followed by 10-nm gold goat anti-guinea pig IgG (Aurion, 1:100). Samples were imaged with a JEOL JEM 1400 transmission ekectron microscope with an Olympus-SIS 2,000 × 2,000 Veleta CCD camera.

Statistical analysis

Averages are reported as arithmetic means ± s.e.m. unless otherwise indicated. Statistical significance was marked by asterisk notation as follows: NS: P > 0.05, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001. Biological replicates are defined as independent differentiations of a given hPSC line unless indicated otherwise. Statistical tests were performed using Graphpad Prism 9.1.