Skin biopsies from three female and three male subjects diagnosed with CDD, and their first-degree-related healthy control (mother or father, respectively) were gently donated through a collaboration with the International Foundation for CDKL5 Research. The study was reviewed and approved by the University of California San Diego IRB/ESCRO committee (protocol 141223ZF).
Fibroblasts from CDD patients and their respective non-affected controls were reprogrammed using non-integrative methods, either by transduction with six episomal plasmid vectors (Sox2, Klf4, Oct3/4, Lin28, p53 shRNA, and L-Myc) or by Sendai virus vector-mediated expression of Oct4, Sox2, Klf4, and c-myc (CytoTune-iPS Kit, Life Technologies), as described elsewhere [65, 66]. iPSCs from each subject were isolated and transferred to feeder-free Matrigel (BD Biosciences) coated plates ~4 weeks after initial transduction. Colonies were propagated in mTeSR1 (Stem Cell Technologies) and manually passaged as small colonies. Standard G-banding karyotype of iPSC clones was performed by the Stem Cell Core Facility at USC (Los Angeles, CA, USA), in collaboration with the Children’s Hospital Los Angeles. Analysis of copy-number variation in iPSC clones was performed by the DRC Genomics and Epigenetics Core at the University of California San Diego.
Generation of NPCs and 2D neurons
iPSCs were differentiated into NPCs as previously described [45, 46]. Briefly, iPSCs colonies were cultured for 2 days in the presence of DMEM/F12 1:1 with 1x Glutamax (Life Technologies), 1x N2 NeuroPlex (N2; Gemini Bio-products), 1x penicillin-streptomycin (PS; Life Technologies), 10-μm SB431542 (SB; StemGent), and 1-μm dorsomorphin (Dorso; Tocris Biosciences). The colonies were lifted off and kept in suspension, under rotation (95 rpm) for 7 days to form embryoid bodies (EB). EBs were gently disrupted, plated onto Matrigel-coated plates, and cultured in DMEM/F12 1:1 with 1x HEPES, 1x Glutamax, 1x PS, 0.5x N2, 0.5x Gem21 NeuroPlex (Gem21; Gemini Bio-products), supplemented with 20-ng/mL basic fibroblast growth factor (bFGF; Life Technologies) for 7 days. Next, neural rosettes were manually collected, dissociated with Accutase (Thermo Fisher), and NPCs were plated onto poly-L-ornithine/laminin-coated plates. NPCs were expanded in the presence of bFGF and fed every other day. Neural differentiation was promoted by bFGF withdrawn from the medium; ROCK inhibitor (Y-27632; Calbiochem, Sigma-Aldrich) was added at 5 μm for the first 2 days. Medium was changed twice a week, and cells were allowed to differentiate for as long as needed.
Generation of cortical organoids
Cortical organoids were generated as previously described . Briefly, fully-grown iPSCs colonies were dissociated for ~10 min at 37 °C with Accutase in PBS (1:1), and centrifuged for 3 min at 150 × g. The cell pellet was resuspended in mTeSR1 supplemented with 10-µM SB and 1-µM Dorso. Approximately 4 × 106 cells were transferred to one well of a 6-well plate and kept in suspension under rotation (95 rpm) in the presence of 5-µM ROCK inhibitor for 24 h to form free-floating spheres. After 3 days, mTeSR1 was substituted by Media1 [Neurobasal (Life Technologies) supplemented with 1x Glutamax, 1x Gem21, 1x N2, 1x MEM nonessential amino acids (NEAA; Life Technologies), 1x PS, 10-µM SB and 1-µM Dorso] for 7 days. Cells were then maintained in Media2 [Neurobasal with 1x Glutamax, 1x Gem21, 1x NEAA, and 1x PS] supplemented with 20-ng/mL bFGF for 7 days, followed by 7 additional days in Media2 supplemented with 20 ng/mL of FGF2 and 20-ng/mL EGF (PeproTech). Next, cells were transferred to Media3 [Media2 supplemented with 10 µg/mL of BDNF, 10 µg/mL of GDNF, 10 µg/mL of NT-3 (all from PeproTech), 200-µM L-ascorbic acid, and 1-mM dibutyryl-cAMP (Sigma-Aldrich)]. After 7 days, cortical organoids were maintained in Media2 for as long as needed, with media changes every 3–4 days.
All cellular cultures were routinely tested for mycoplasma by PCR. Media supernatants (with no antibiotics) were collected, centrifuged, and resuspended in saline buffer. Ten microliters of each sample were used for a PCR with the following primers: Forward: GGCGAATGGGTGAGTAAC; Reverse: CGGATAACGCTTGCGACCT. Only negative samples were used in the study.
Proteomics and phosphoproteomics analysis
Cell lysates from NPCs, neurons, and cortical organoids were prepared with RIPA buffer or 100-mM TEAB with 1% SDS. After reduction (10-mM TCEP) and alkylation (50-mM chloroacetamide), MeOH/CHCl3 precipitation was performed. Pellets were dissolved with 6-M urea in 50-mM TEAB, and LysC/Tryp (Promega) was added at 1:25 (w/w) ratio. After 3–4-h incubation at 37 °C, reaction mixture was diluted with 50-mM TEAB for urea to be <1 M. After overnight digestion, peptides were labeled with TMT 10-plex (Thermo Fisher), followed by quenching with hydroxylamine. All reaction mixtures were pooled together and dried using SpeedVac. One hundred μg of peptides were separated for total protein analyses, and the remaining mixtures were used for phosphoproteomic analyses. After desalting using Pierce peptide desalting spin columns (Thermo Fisher), phosphopeptides were enriched using the High-Select TiO2 Enrichment Kit (Thermo Fisher). Resulting eluates were dried in SpeedVac immediately after the enrichment. Peptides to be analyzed in both total and phosphoprotein analyses were fractionated using Pierce High pH Reversed-phase Peptide Fractionation Kit (Thermo Fisher) and then dried in SpeedVac. Dried peptides were dissolved with buffer A (5% acetonitrile, 0.1% formic acid), and each fraction was injected directly onto a 25 cm, 100-μm-ID columns packed with BEH 1.7-μm C18 resin (Waters). Samples were separated at a flow rate of 300 nL/min on nLC 1000 (Thermo Fisher). A gradient of 1–25% buffer B (80% acetonitrile, 0.1% formic acid) over 200 min, an increase to 50% B over 120 min, an increase to 90% B over another 30 min and held at 90% B for a final 10 min of washing was used for 360-min total run time. Column was re-equilibrated with 20 μL of buffer A prior to the injection of sample. Peptides were eluted directly from the tip of the column and nanosprayed directly into the mass spectrometer Orbitrap Fusion by application of 2.8-kV voltage at the back of the column. Fusion was operated in a data-dependent mode. Full MS1 scans were collected in the Orbitrap at 120k resolution. The cycle time was set to 3 s, and within this 3 s, the most abundant ions per scan were selected for CID MS/MS in the ion trap. MS3 analysis with multi-notch isolation (SPS3)  was utilized for detection of TMT reporter ions at 60k resolution. Monoisotopic precursor selection was enabled, and dynamic exclusion was used with exclusion duration of 10 s. Tandem mass spectra were extracted from the raw files using RawConverter  with monoisotopic peak selection.
The spectral files from all fractions were uploaded into one experiment on Integrated Proteomics Applications (IP2, Ver.5.1.3) pipeline. Proteins and peptides were searched using ProLuCID and DTASelect 2.0 on IP2 against the UniProt H. sapiens protein database with reversed decoy sequences (UniProt_Human_reviewed_05-05-2016_reversed.fasta). Precursor mass tolerance was set to 50.0 ppm, and the search space allowed all fully-tryptic and half-tryptic peptide candidates without limit to internal missed cleavage and with a fixed modification of 57.02146 on cysteine and 229.1629 on N-terminus and lysine. Peptide candidates were filtered using DTASelect parameters of -p 1 (proteins with at least one peptide are identified) -y 1 (partial tryptic end is allowed) -pfp 0.01 (protein FDR < 1%) -DM 5 (highest mass error 5 ppm) -U (unique peptide only). Quantification was performed by Census on IP2. The expression value for each protein was calculated by adding the peptide-level reporter ion intensities normalized to total intensity of each channel to remove the variances caused by the different loading amount or labeling efficiency for different channels. For phosphoproteome analysis, search parameters included differential modification of 79.966331 on serine, threonine, and tyrosine, and DTASelect parameters -p 1 -y 1 -pfp 0.01 -DM 5 -m 0 (only phophorylated peptides). To detect the phosphorylation-specific changes, peptide-level reporter ion intensities were normalized to total protein intensities.
Amino acid starvation
NPCs were grown to 80–90% confluency in complete culture medium before amino acid withdraw. Next, medium was replaced by a glucose-containing starvation buffer, Earle’s Balanced Salt Solution (Thermo Fisher), and the cells were incubated for 10, 30, 60, 120, or 240 min before protein extraction.
Protein was extracted using the RIPA Lysis and Extraction buffer (Thermo Fisher) containing cOmplete ULTRA mini protease inhibitor (Roche) and PhosSTOP phosphatase inhibitor (Roche). Twenty microgram of protein lysates were separated on a 4–12% Bis-Tris protein gel (Novex), and transferred onto a nitrocellulose membrane (Novex) using the iBlot2 Gel Transfer device (Thermo Fisher). Following blockage with Rockland Blocking Buffer (Rockland), the membrane was incubated with primary antibodies overnight at 4 °C. Next, the membrane was washed five times (5 min each) with 0.1% Tween 20 in PBS and incubated with secondary antibodies for 2 h at room temperature. Antibodies used in this study can be found in Table S9. Odyssey CLx imaging system (Li-Cor) was used for signal detection, and semi-quantitative analysis was performed using Odyssey Image Studio software.
iPSCs, NPCs, and 2D neurons were fixed with 4% paraformaldehyde for 15 min. After three washes with PBS, cells were permeabilized with 0.25% Triton X-100 for 15 min, blocked with 3% bovine serum albumin (BSA) and incubated overnight at 4 °C with primary antibodies diluted in 3% BSA. The following day, cells were washed and incubated with the secondary antibodies for 1 h. Antibodies used in this study can be found in Table S9. For nuclei staining, DAPI solution (1 μg/mL) was used. The slides were mounted using ProLong Gold antifade reagent and analyzed under a fluorescence microscope (Z1 Axio Observer Apotome, Zeiss).
Synaptic puncta quantification
After 8 weeks of differentiation, 2D neurons were fixed and stained. The number of pre-synaptic VGLUT1+ and post-synaptic HOMER1+ puncta co-localization was blindly quantified. Only puncta overlapping MAP2+ processes were scored. Images were taken randomly from two independent experiments.
DNA fragmentation analysis
NPCs were harvested to single-cell suspension in PBS, fixed by addition of 70% ethanol and stored for 24 h at 4 °C. Next, cells were washed with PBS, resuspended in DAPI staining solution (0.1% (v/v) Triton X-100, 1-µg/mL DAPI in PBS) and incubated for 5 min at 37 °C. Samples were analyzed on the NC-3000 Advanced Image Cytometer (Chemometec), using the preoptimized DNA Fragmentation Assay. The amount of high molecular weight DNA retained in the cells was quantified in order to detect apoptotic cells with fragmented DNA (subG1 population).
Caspase activity was assessed using the Green FLICA Caspases 3 & 7 Assay Kit (ImmunoChemistry Technologies, LLC) according to manufacturer’s protocol. Briefly, NPCs were harvested to single-cell suspension, washed with PBS, and stained with 1X carboxyfluorescein Fluorochrome Inhibitor of Caspase Assay (FAM-FLICA) reagent, 10-µg/mL Hoechst 33342 and 10-µg/mL propidium iodide (PI). Samples were analyzed on the NC-3000 Advanced Image Cytometer (Chemometec) using the preoptimized Caspase Assay. PI stained the nonviable cell population whereas FAM-FLICA stained cells with caspase activity for apoptosis analysis.
NPC proliferation was assessed by cell counting. Briefly, a pre-determined number of NPCs was plated onto poly-L-ornithine/laminin-coated plates (day 0). After 4 h, the plates were transferred to a Viva View FL Incubator Microscope (Olympus), and allowed to acclimatize for 30 min (T0). Next, the cellular proliferation was monitored and contrast-phase images were taken after 48 h (T48; day 2). The images were processed, and the number of cells at T0 and T48 was determined using the Cell Counter plugin on the Fiji platform. The difference between days 2 and 0 was used to estimate the NPC proliferation rate.
Disruption of the mitochondrial transmembrane potential (Dym) is usually associated with early stages of apoptosis. We measured Dym in NPCs using the cationic dye JC-1. Briefly, NPCs were harvested to single-cell suspension in PBS containing 2.5 μg/mL of a JC-1 solution, and cells were incubated for 10 min at 37 °C. After washes with PBS, DAPI (1 μg/mL) was added for nuclei staining. Samples were analyzed on the NC-3000 Advanced Image Cytometer (Chemometec) using the preoptimized Mitochondrial Potential Assay. At high concentrations, JC-1 forms aggregates and become red fluorescent, while in apoptotic cells the mitochondrial potential collapses and JC-1 localizes to the cytosol in its monomeric green fluorescent form. The number of cells with collapsed mitochondrial membrane potential was quantified and the mitochondrial depolarization estimated as a decrease in the red/green fluorescence intensity ratio.
Neuronal spine-like dynamics
Dendrites from 8-week-old neurons stably transfected to express enhanced GFP by Synapsin1 promoter self-inactivating lentivirus were recorded using a Z1 Axio Observer Apotome (Zeiss). Images were taken every 30 s for 1 h, and analyzed using the NeuronJ plugin on the Fiji platform. Only neurons that displayed at least two visible neurites at t = 0 and had changes in spine dynamics during the 60-min time course were analyzed.
Three-week old spheres were treated for 3 additional weeks with 1 µM of selected compounds. Cellular migration was evaluated 8 days after spheres were plated onto poly-L-ornithine/laminin-coated plates. The outward radial migration was measured using NeuronJ plugin on the Fiji platform.
Quantitative multiplex co-immunoprecipitation (QMI)
QMI analysis was performed as previously described [53, 54]. Briefly, cortical organoids were homogenized in lysis buffer [150-mM NaCl, 50-mM Tris (pH 7.4), 1% NP-40, 10-mM NaF, 2-mM sodium orthovanadate + protease/phosphatase inhibitor cocktails (Sigma)] using a glass tissue homogenizer, incubated for 15 min, centrifuged at high speeds to remove nuclei and debris, and protein concentration was determined using a Pierce BCA Kit (Thermo Fisher). A master mix containing equal numbers of each antibody-coupled Luminex bead class was prepared and distributed into post-nuclear cell lysate samples in duplicate. Protein complexes were immunoprecipitated from samples containing equal amounts of protein overnight at 4 °C, washed twice in an ice-cold Fly-P buffer [50-mM tris (pH 7.4), 100-mM NaCl, 1% BSA, and 0.02% sodium azide], and distributed into as many wells of a 96-well plate as there were probes, on ice. Biotinylated detection antibodies were added and incubated for 1 h, with gentle agitation at 500 rpm in a cold room (4 °C). Following incubation, microbeads and captured complexes were washed three times in the Fly-P buffer using a Bio-Plex Pro II magnetic plate washer in a cold room. Microbeads were then incubated for 30 min with streptavidin-PE on ice, washed three times, and resuspended in 125 μL of an ice-cold Fly-P buffer. Fluorescence data were acquired on a customized, refrigerated Bio-Plex 200 instrument according to the manufacturer’s recommendations.
Data preprocessing and inclusion criteria XML output files were parsed to acquire the raw data for use in MATLAB while XLS files were used for input into R statistical packages. For each well from a data acquisition plate, data were processed to (i) eliminate doublets on the basis of the doublet discriminator intensity (>5000 and <25,000 arbitrary units; Bio-Plex 200), (ii) identify specific bead classes within the bead regions used, and (iii) pair individual bead PE fluorescence measurements with their corresponding bead regions. This processing generated a distribution of PE intensity values for each pairwise protein co-association measurement. ANC adaptive nonparametric analysis with empirical alpha cutoff (ANC)  was used to identify high-confidence, statistically significant differences (corrected for multiple comparisons) in bead distributions on an individual interaction basis. ANC was conducted as described elsewhere .
We required that hits be present in at least six of eight replicates at an adjusted p < 0.05. The α-cutoff value required per experiment to determine statistical significance was calculated to maintain an overall type I error of 0.05 (adjusted for multiple hypothesis testing with Bonferroni correction), with further empirical adjustments to account for technical errors. CNA: bead distributions used in ANC were collapsed into a single median fluorescent intensity (MFI), which was averaged across duplicate samples and input into the WGCNA package for R . Data were filtered to remove weakly detected interactions (“noise,” MFI < 100), and batch effects were removed using the COMBAT function for R , with “experiment number” as the “batch” input. Post-Combat data were log2-transformed prior to CNA analysis. Closely related protein co-associations were assigned to arbitrary color-named modules by the WGCNA program. Modules whose eigenvectors significantly (P < 0.05) correlated with the genotype were considered significantly as described previously [54, 69] to produce a high-confidence set of interactions that were both individually significantly different in comparisons between experimental groups, and that belonged to a larger module of co-regulated interactions that was significantly correlated with experimental group. Hierarchical clustering was performed using pvlcust in R .
Human iPSC-derived NPCs were stably transfected to express EGFP by Synapsin1 promoter self-inactivating lentivirus. Ten to twenty thousand cells were injected per site, 1 mm from the midline between the Bregma and Lambda and 1–2-mm deep into the cortex and striatum of newborns immunosuppressed NOD/SCID mice. Briefly, newborns (P0–P2) were anesthetized by hypothermia and then placed in a contoured Styrofoam mold. Two microliters of NPCs were injected into both hemispheres using a 5-ml Hamilton syringe with a 32-gauge needle. After 6 months, injected animals were anesthetized and perfused. Entire brains were sliced using a cryomicrotome, and immunohistochemical analysis was carried out on free-floating mice brain slices to identify and evaluate the efficiency of the transplantation.
Postmortem brain specimens and cortical sampling
Four postmortem brains that were gender, age, and hemisphere-matched were used. Specifically, postmortem brain tissue from a 5-year-old female CDD patient with a Pro719CysfsX66 mutation and, a 30-year-old female CDD patient with a deletion comprehending exons 1–3 in the CDKL5 gene; and two female individuals with no described genetic alteration (control), respectively at 6- and 30-year-old. All brain specimens were harvested within a postmortem interval of 15–36 h and had been immersed and fixed in 10% formalin for <3 years. For the purpose of the present experiments, samples were obtained from anatomically well-identified cortical areas in a consistent manner across specimens, comprehending the primary somatosensory cortex (Brodmann area 3), the primary motor cortex (Brodmann area 4) and the secondary visual area (Brodmann area 18). Details about the tissue-processing protocol is provided online by the NIH NeuroBioBank. We focused specifically on these parts of the cortex because pathologies in dendritic morphology in these areas have been reported in other neurodevelopmental disorders [46, 72,73,74]. In addition, pyramidal neurons in the selected areas reach their mature-like morphology early in development and start displaying dendritic pathologies sooner than high integration areas, such as the prefrontal cortex, allowing for a comparison of postmortem findings with iPSC-derived neurons in early stages of development [75, 76]. Samples were obtained from Harvard Brain Tissue Resource Center, the University of Maryland Brain and Tissue Bank, and the University of Miami Brain Endowment Bank, which are Brain and Tissue Repositories of the NIH NeuroBioBank.
Postmortem brain tissue processing
Processing and staining of brain tissue samples were performed by Neurodigitech LLC (San Diego, CA) using the semi-rapid Golgi technique . Briefly, specimens were immersed in a solution of 1% silver nitrate for 10 days. Blocks were then sectioned on a vibratome, perpendicular to the pial surface, at a thickness of 110–120 μm. Golgi sections were cut into 100% ethyl alcohol and transferred briefly into toluene, mounted onto glass slides, and cover-slipped.
Neurons included in the morphological analysis did not display degenerative changes . The morphological analysis was performed on pyramidal neurons located in cortical layers V/VI, with fully impregnated soma, apical dendrites with present oblique branches, and at least two basal dendrites with second/third order segments. To minimize the effects of cutting on dendritic measurements, we included neurons with cell bodies located near the center of 120-μm thick histological sections, with natural terminations of higher-order dendritic branches present where possible [79, 80]. Inclusion of the neurons completely contained within 120-μm sections biases the sample toward smaller neurons, leading to the underestimation of dendritic length ; therefore, we applied the same criteria blinded across all control and CDD specimens, and we thus included the neurons with incomplete endings if they were judged to otherwise fulfill the criteria for successful Golgi impregnation. All neurons were oriented with apical dendrite perpendicular to the pial surface; inverted pyramidal cells as well as magno-pyramidal neurons were excluded from the analysis.
Postmortem brain neuronal tracing
Neuronal morphology was quantified along x-, y-, and z-coordinates using OPTIMAS Bioscan software (Media Cybernetics Inc., MD, USA) connected to a Zeiss Axio scope system, equipped with 100×/1.25 oil Plan Fluor objective and a CCD camera (Hamamatsu, ORCA-Flash4.0 V3), motorized X-, Y-, and Z-focus for high-resolution image acquisition and digital quantitation. Tracings were conducted on both apical and basal dendrites, and the results reflect summed values for both types of dendrites per neuron. Following the recommendation that the applications of Sholl’s concentric spheres for the analysis of neuronal morphology are not adequate when neuronal morphology is analyzed in three dimensions , we conducted dendritic tree analysis with the following measurements [79, 80]: (1) soma area—cross sectional surface area of the cell body; (2) dendritic length—summed total length of all dendrites per neuron; (3) dendrite number—number of dendritic trees emerging directly from the soma per neuron; (4) dendritic segment number—total number of segments per neuron; (5) dendritic spine/protrusion number—total number of dendritic spines per neuron; (6) dendritic spine/protrusion density—average number of spines/20 μm of dendritic length; and (7) branching point number—number of nodes (points at the dendrite where a dendrite branches into two or more) per neuron. Dendritic segments were defined as parts of the dendrites between two branching points—between the soma and the first branching point in the case of first order dendritic segments, and between the last branching point and the termination of the dendrite in the case of terminal dendritic segments. Since the long formalin-fixation time may result in degradation of dendritic spines, spine values may be underestimated and are thus reported here with caution. All of the tracings were accomplished blind to brain region and diagnostic status.
iPSC-derived neuronal tracing
The iPSC-derived samples consisted of EGFP-SYN1-positive 8-week-old neurons with pyramidal- or ovoid-shaped soma and at least two branched neurites (dendrites) with visible spines/protrusions. Protrusions from dendritic shaft, which morphologically resembled dendritic spines in postmortem specimens, were considered and quantified as dendritic spines in iPSC-derived neurons. The neurites were considered dendrites based on the criteria applied in postmortem studies: (1) thickness that decreased with the distance from the cell body; (2) branches emerging under acute angle; and (3) presence of dendritic spines. Only EGFP-positive neurons with dendrites displaying evenly distributed fluorescent stain along their entire length were considered. In addition, neurons included in the analysis had to exhibit the nuclei co-stained with CTIP2, indicative of layer V/VI neurons. The morphology of the neurons was quantified along x-, y-, and z-coordinates using Neurolucida v.9 software (MBF Bioscience, Williston, VT) connected to a Nikon Eclipse E600 microscope with 40× oil objective. No distinction was made between apical and basal dendrites, and the results reflect summed length values of all neurites/dendrites per neuron, consistent with what was done for the postmortem neurons. The same set of measurements used in the analysis of Golgi-impregnated neurons was applied to the analysis of iPSC-derived neurons, and all tracings were accomplished blind to the genotype.
Six-week-old cortical organoids were plated per well in poly-L-ornithine/laminin-coated 12-well MEA plates (Axion Biosystems). Cells were fed twice a week, and measurements were collected 24 h after the medium was changed, once a week, starting at 2 weeks after plating (8 weeks of organoid differentiation). Recordings were performed using a Maestro MEA system and AxIS Software Spontaneous Neural Configuration (Axion Biosystems) with a customized script for band-pass filter (0.1-Hz and 5-kHz cutoff frequencies). Spikes were detected with AxIS software using an adaptive threshold crossing set to 5.5 times the standard deviation of the estimated noise for each electrode (channel). The plate was first allowed to rest for 3 min in the Maestro device, and then 3 min of data were recorded. For the MEA analysis, the electrodes that detected at least 5 spikes/min were classified as active electrodes using Axion Biosystems’ Neural Metrics Tool. Bursts were identified in the data recorded from each individual electrode using an inter-spike interval (ISI) threshold requiring a minimum number of five spikes with a maximum ISI of 100 ms. At least ten spikes under the same ISI with a minimum of 25% active electrodes were required for network bursts in the well. The synchrony index was calculated using a cross-correlogram synchrony window of 20 ms. Independent experiments were performed with three cell lines with at least three technical replicates.
Whole-cell patch-clamp recordings were performed in cultured human iPSC-derived neurons at room temperature (~20 °C), as described previously [44, 82]. The extracellular solution for patch-clamp experiments contained the following: 130-mM NaCl, 3-mM KCl, 1-mM CaCl2, 1-mM MgCl2, 10-mM HEPES, and 10-mM glucose; pH 7.4 with 1-M NaOH (∼4 mM Na+ added). The internal solution for patch electrodes contained the following: 138-mM K-gluconate, 4-mM KCl, 10-mM Na2-phosphocreatine, 0.2-mM CaCl2, 10-mM HEPES (Na+ salt), 1-mM EGTA, 4-mM Mg-ATP, 0.3-mM Na-GTP; pH 7.4 with 1-M KOH (∼3-mM K+ added). The osmolarity of all solutions was adjusted to 290 mOsm. Electrodes for electrophysiological recording were pulled on a Flaming/Brown micropipette puller (Model P-87, Sutter Instrument) from filamented borosilicate capillary glass (1.2-mm OD, 0.69-mm ID, World Precision Instruments). The electrode resistances were 3–6 MΩ for whole-cell recordings. Patch-clamp experiments were performed with an Axon CV-4 headstage, Axopatch 200A amplifier and DigiData 1322A (Molecular Devices). Recordings were digitized at 10 kHz and low-pass filtered at 2 kHz. The spontaneous excitatory synaptic currents were recorded with the holding membrane potentials of −60 mV and last 3–5 min for each neuron. Evoked APs were measured from −60 mV under current-clamp configuration. Data were all corrected for liquid junction potentials (10 mV). Voltage-clamp synaptic currents (sEPSCs) were analyzed using Mini Analysis (Synaptosoft) and other electrophysiological results were analyzed using pCLAMP 10 software (Molecular Devices). The electrophysiological experiments and analyses were performed in a blinded manner to avoid bias.
CDD HTS platform development
An HTS platform for CDD was developed by StemoniX Inc. Briefly, StemoniX’s HTS system relies on spheroids derived from human iPSCs that comprise a balanced culture of cortical neurons and astrocytes. These spheroids display quantifiable, robust, and uniform spontaneous calcium oscillations, which correlated with synchronous neuronal activity . The CDD HTS platform uses cells derived from CDD1 and Control1 lines.
Chemical library and drug treatment
The compound library used in this study comprises a unique collection of 1112 compounds with biological activity used for neurologic research and associated assays; of which two-thirds are FDA-approved drugs (SelleckChem, Houston, TX, USA). Three-week-old CDD spheroids received chronic treatment three times per week, during 3 weeks. All compounds were tested at 1 µM in three technical replicates; vehicle controls (DMSO or water) were included in multiple replicates. The dose–response assay comprehended six concentrations ranging from 0.0003 to 1 µM, in four technical replicates.
HT calcium oscillation assay
To assess the intracellular calcium oscillations in CDD and control spheroids, cells were incubated with the FLIPR Calcium 6 Kit (Molecular Devices LLC, San Jose, CA, USA) as previously described . Briefly, cells were pre-loaded with Calcium 6 Dye for 2 h prior to the recording. The peaks observed correlate with synchronous neural activity in the spheroids. The kinetics of intracellular calcium oscillations was determined using the FLIPR Tetra High-Throughput Cellular Screening System (Molecular Devices LLC): the fluorescence intensity was set at 515–575 nm following excitation at 470–495 nm for 10 min at a frequency of 3 Hz; the exposure time per read was 0.4 s, the gain was set to 30, and the excitation intensity was set to 40%. The instrument temperature was kept at 37 °C.
HT cell viability
To determine any cytotoxic effect of the drug treatment on the neural culture, the CellTiter Glo 3D Cell Viability Assay (Promega, Madison, WI, USA) was performed according to the manufacturer’s instructions. Briefly, following the intracellular calcium oscillation recording, the spheroids were washed with PBS for removal of the calcium dye. Next, cells were incubated with CellTiter Glo Reagent for 30 min at room temperature and the number of viable cells was estimated based on the amount of ATP present in the culture. The luminescent signal was captured using a PHERAstar FSX Microplate Reader (BMG Labtech, Ortenberg, Germany).
HT 3D imaging for size measurement
To determine the size of CDD and control spheroids, contrast-phase images from the 384-well plates were weekly acquired using the ImageXpress Microscope-Micro Confocal System (Molecular Devices LLC). The instrument temperature was kept at constant 37 °C during image acquisition.
HT screening analysis
A multiparametric analysis of representative descriptors of the intracellular calcium oscillation was performed using StemoniX’s proprietary Assay AnalytiX Software. The features used include: (1) peak count, (2) average (Avg) peak height (amplitude), (3) peak height standard deviation (s.d.), (4) Avg peak width, (5) peak width s.d., (6) Avg peak spacing, (7) peak spacing s.d., (8) Avg peak rise time, (9) peak rise time s.d., (10) Avg peak decay time, (11) peak decay time s.d., (12) Class 1 peak count, (13) Class 2 peak count, (14) Class 3 peak count, (15) singular peak count, (16) irregular peak count, and (17) subpeak count. For peak classification, the relative height of each individual peak within the signal is classified based on percentage based binning of the max peak height into three classes, thus providing a count of peaks across the variation activation levels achieved by the system. Singular peaks are defined as uninterrupted calcium oscillations, meaning the activity from trough to peak and back to trough is continuous, without additional peaks occurring during the ascent or descent of the oscillation. Irregular defines the peaks which have been interrupted by such events, and subpeaks are the number of the interrupting peak events within a signal. To determine the Scalar Perturbation (SP), the data were normalized by dividing raw parameters by the median values of controls from the same plate. Each parameter was standardized by z-score of normalized and log-transformed control values. Next, calculating the Euclidean norm of selected parameters provides a SP for each replicate, which was then averaged among the replicates. For Parameter Recovery (PR), boundaries of rescue criteria were set using the parameter values extracted from vehicle control signals. These boundaries were used to calculate the number of rescued parameters for a given compound. Using the number of rescued parameters out of the total parameters considered, a percentage based PR was calculated and averaged across compound replicates.
To determine the correlation between compounds, pathways, and genes, we used the databases NCATS (National Center for Advancing Translational Sciences)–Inxight: Drugs portal (version 1.1, https://drugs.ncats.io/), CTD (Comparative Toxicogenomics Database) , KEGG (Kyoto Encyclopedia of Genes and Genomes), DGIdb (Drug Gene Interaction Database) . NCATS Inxight: drugs is a comprehensive portal for drug development information, and contains information on ingredients in medicinal products, including US-approved drugs, marketed drugs, and investigational drugs. CTD is a literature-based and manually curated associations between chemicals, gene products, phenotypes, diseases and environmental exposures. KEGG is a biological encyclopedia for the understanding of high-level functions and utilizes of a biological system and its parts (cell, molecules, genes) and, DBIdb is a drug-gene interaction database, composed by consolidated interactions and druggable genes that were extracted from peer-reviewed manuscripts, databases, and web resources.
Data are presented as mean ± s.e.m., unless otherwise indicated. No statistical method was used to predetermine the sample size. The statistical analyses were performed using GraphPad Prism v6; two-tailed Mann–Whitney U test or unpaired t-test with multiple-comparison correction was used as indicated. Significance was defined as p < 0.05(*), p < 0.01(**), p < 0.001(***), or p < 0.0001(****). Blinding was used for comparing affected and control samples.