Ethics statement

All human material (blood RNA, primary microglia RNA, iPSCs) used in this study was derived after signed informed consent: for blood, according to University of Oxford OHS policy document 1/03; all procedures related to the use of the primary microglia followed established institutional (McGill University, Montreal, QC, Canada) and Canadian Institutes of Health Research guidelines for the use of human cells; for iPSC, with approval from the South Central Berkshire Research Ethics Committee, U.K. (REC 10/H0505/71). The blood RNA and primary microglia RNA samples have been published previously26, as have the iPSC lines (see below).

Four healthy control iPSC lines, SFC840-03-03 (female, 67years old,35), SFC841-03-01 (male, 36,18), SFC856-03-04 (female, 78,36), OX3-06 (male, 49,37), generated from skin biopsy fibroblasts and characterized as described before, were used in this study. Additionally, the previously reported26 line AH016-3 Lenti_IP_RFP (male, 80years old), which constitutively expresses Red Fluorescent Protein (RFP) under continuous puromycin selection, was used for some live-imaging experiments.

iPSCs were cultured in mTeSR1 (StemCell Technologies) or OXE8 medium38 on Geltrex (Thermo Fisher)-coated tissue culture plates with daily medium changes. Passaging was done as clumps using EDTA in PBS (0.5mM). Cells were initially expanded at low passage to create a master stock, which was used for all experiments to ensure consistency. Cells were regularly tested negative for mycoplasma using MycoAlert Mycoplasma Detection Kit (Lonza).

iPSCs were differentiated to MNs according to our previously published protocol18,19,27. Briefly, neural induction of iPSC monolayers was performed using DMEM-F12/Neurobasal 50:50 medium supplemented with N2 (1X), B27 (1X), 2-Mercaptoethanol (1X), AntibioticAntimycotic (1X, all ThermoFisher), Ascorbic Acid (0.5M), Compound C (1M, both Merck), and Chir99021 (3M, R&D Systems). After two days in culture, Retinoic Acid (RA, 1M, Merck) and Smoothened Agonist (SAG, 500nM, R&D Systems) were additionally added to the medium. Two days later, Compound C and Chir99021 were removed from the medium. After another 5days in culture, neural precursors were dissociated using accutase (ThermoFisher), and split 1:3 onto Geltrex-coated tissue culture plates in medium supplemented with Y-27632 dihydrochloride (10M, R&D Systems). After one day, Y-27632 dihydrochloride was removed from the medium, and then the cells were cultured for another 8days with medium changes every other day. For terminal maturation, the cells were dissociated on day in vitro (DIV) 18 using accutase and plated onto coverslips or tissue culture plates coated with polyethylenimine (PEI, 0.07%, Merck) and Geltrex or tissue culture dishes coated with PDL (Sigma-Aldrich)/ Laminin (R&D Systems)/ Fibronectin (Corning). For this step, the medium was additionally supplemented with BDNF (10ng/mL), GDNF (10ng/mL), Laminin (0.5g/mL, all ThermoFisher), Y-27632 dihydrochloride (10M), and DAPT (10M, R&D Systems). Three days later, Y-27632 dihydrochloride was removed from the medium. After another three days, DAPT was removed from the medium. Full medium changes were then performed every three days.

For MNs differentiated in co-culture medium alone, all steps were performed similarly until three days after the terminal re-plating (D21). MNs were then cultured in co-culture medium as described below.

iPSCs were differentiated to macrophage/microglia precursors as described previously20,21. Briefly, embryoid body (EB) formation was induced by seeding iPSCs into Aggrewell 800 wells (STEMCELL Technologies) in OXE838 or mTeSR1 medium supplemented with Bone Morphogenetic Protein 4 (BMP4, 50ng/mL), Vascular Endothelial Growth Factor (VEGF, 50ng/mL, both Peprotech), and Stem Cell Factor (SCF, 20ng/mL, Miltenyi Biotec). After four days with daily medium changes, EBs were transferred to T175 flasks (~150 EBs each) and differentiated in X-VIVO15 (Lonza), supplemented with Interleukin-3 (IL-3, 25ng/mL, R&D Systems), Macrophage Colony-Stimulating Factor (M-CSF, 100ng/mL), GlutaMAX (1X, both ThermoFisher), and 2-Mercaptoethanol (1X). Fresh medium was added weekly. After approximately one month, precursors emerged into the supernatant and could be harvested weekly. Harvested cells were passed through a cell strainer (40M, Falcon) and either lysed directly for RNA extraction or differentiated to microglia in monoculture or co-culture as described below.

Three days after the final re-plating of differentiating MNs (DIV21), macrophage/microglia precursors were harvested as described above and resuspended in co-culture medium comprised of Advanced DMEM-F12 (ThermoFisher) supplemented with GlutaMAX (1X), N2 (1X), AntibioticAntimycotic (1X), 2-Mercaptoethanol (1X), Interleukin-34 (IL-34, 100ng/mL, Peprotech), BDNF (10ng/mL), GDNF (10ng/mL), and Laminin (0.5g/mL). MNs were quickly rinsed with PBS, and macrophage/microglia precursors re-suspended in co-culture medium were added to each well. Co-cultures were then maintained for at least 14days before assays were conducted as described below. Half medium changes were performed every 23days.

For comparisons between co-cultures and monocultures, MNs and monocultured microglia were also differentiated alone in co-culture medium.

Cells cultured on coverslips were pre-fixed with 2% paraformaldehyde in PBS for 2min and then fixed with 4% paraformaldehyde in PBS for 15min at room temperature (RT). After permeabilization and blocking with 5% donkey/goat serum and 0.2% Triton X-100 in PBS for 1h at RT, the coverslips were incubated with primary antibodies diluted in 1% donkey/goat serum and 0.1% Triton X-100 in PBS at 4C ON. The following primary antibodies were used: rabbit anti-cleaved caspase 3 (1:400, 9661S, Cell Signaling), mouse anti-ISLET1 (1:50, 40.2D6, Developmental Studies Hybridoma Bank), mouse anti-TUJ1 (1:500, 801201, BioLegend), rabbit anti-TUJ1 (1:500, 802001, BioLegend), chicken anti-TUJ1 (1:500, GTX85469, GeneTex), rabbit anti-IBA1 (1:500, 019-19741, FUJIFILM Wako Pure Chemical Corporation), goat anti-IBA1 (1:500, ab5076, abcam), rabbit anti-synaptophysin (1:200, ab14692, abcam), goat anti-ChAT (1:100, ab114P, abcam), rat anti-TREM2 (1:100, MAB17291-100, R&D Systems), rabbit anti-TMEM119 (1:100, ab185337, abcam), rat anti-CD11b (1:100, 101202, BioLegend).

After three washes with PBS-0.1% Triton X-100 for 5min each, coverslips were incubated with corresponding fluorescent secondary antibodies Alexa Fluor 488/568/647 donkey anti-mouse/rabbit/rat/goat, goat anti-chicken (all 1:1000, all ThermoFisher). Coverslips were then washed twice with PBS-0.1% Triton X-100 for 5min each and incubated with 4,6-diamidino-2-phenylindole (DAPI, 1g/mL, Sigma-Aldrich) in PBS for 10min. After an additional 5min-washing step with PBS-0.1% Triton X-100, the coverslips were mounted onto microscopy slides using ProLong Diamond Antifade Mountant (ThermoFisher). Confocal microscopy was then performed using an LSM 710 microscope (Zeiss).

For the analysis of neuronal and MN markers after differentiation, three z-stacks (2m intervals) of randomly selected visual fields (425.1425.1m) were taken for each coverslip at 20magnification. The ratios of TUJ1-positive, ChAT-positive, ISLET1-positive, ChAT-positive/ TUJ1-positive, and ISLET1-positive/ TUJ1-positive cells were then quantified using Fiji in a blinded fashion.

For the analysis of microglial markers in monoculture and co-culture, three z-stacks (1m intervals) of randomly selected visual fields (212.55212.55m) were taken for each coverslip at 40magnification. The expression of CD11b, TMEM119, and TREM2 in IBA1-positive cells in monoculture and co-culture was then quantified using Fiji.

For the analysis of apoptosis in neurons, five z-stacks images of randomly selected visual fields (212.55212.55m) were taken at 40magnification for each coverslip. The ratios of cleaved caspase 3/ TUJ1-positive cells were then quantified for neurons in monoculture and co-culture in a blinded fashion. For the analysis of apoptosis in microglia, three z-stacks images of randomly selected visual fields (212.55212.55m) were taken at 40magnification for each coverslip. The ratios of cleaved caspase 3/ IBA1-positive cells were then quantified for microglia in monoculture and co-culture.

For the analysis of microglial ramifications, five z-stacks images of randomly selected visual fields (212.55212.55m) were taken at 40magnification for each coverslip. To analyze the branching of IBA1-positive microglia in monoculture and co-culture, the average branch length, number of branch points and number of branch endpoints was determined using 3DMorph39, a Matlab-based script for the automated analysis of microglial morphology.

From the same harvest, macrophage precursors (pMacpre) were either lysed directly or differentiated to microglia in monoculture (pMGL) or microglia in co-culture with MNs (co-pMG) for 14days. pMGL were rinsed with PBS and directly lysed in the dish. For both pMacpre and pMGL, RNA was extracted using an RNAeasy Mini Plus kit (Qiagen) according to the manufacturers instructions. Co-cultures were first dissociated by 15min incubation with papain (P4762, Sigma-Aldrich) diluted in accutase (20 U/mL) and gentle trituration based on a previously published protocol40. The cell suspension was then passed through a cell strainer (70m, Falcon) to remove cell clumps. To extract co-pMG, magnetic-activated cell sorting (MACS) was then performed using CD11b-MACS beads (130093-634, Miltenyi Biotec) according to the manufacturers instructions. The panned cell population was lysed for RNA extraction using an RNAeasy Micro kit (Qiagen) according to the manufacturers instructions. In addition, RNA from human fetal microglia and blood monocytes from three different healthy genetic backgrounds wasre-used from our previous study26.

RNA from the four different healthy control lines (listed earlier) per condition (pMacpre, pMGL, co-pMG) was used for RNA sequencing analysis. Material was quantified using RiboGreen (Invitrogen) on the FLUOstar OPTIMA plate reader (BMG Labtech) and the size profile and integrity analysed on the 2200 or 4200 TapeStation (Agilent, RNA ScreenTape). RIN estimates for all samples were between 9.2 and 9.9. Input material was normalised to 100ng prior to library preparation. Polyadenylated transcript enrichment and strand specific library preparation was completed using NEBNext Ultra II mRNA kit (NEB) following manufacturers instructions. Libraries were amplified (14 cycles) on a Tetrad (Bio-Rad) using in-house unique dual indexing primers (based on41). Individual libraries were normalised using Qubit, and the size profile was analysed on the 2200 or 4200 TapeStation. Individual libraries were normalised and pooled together accordingly. The pooled library was diluted to~10nM for storage. The 10nM library was denatured and further diluted prior to loading on the sequencer. Paired end sequencing was performed using a NovaSeq6000 platform (Illumina, NovaSeq 6000 S2/S4 reagent kit, v1.5, 300 cycles), generating a raw read count of a minimum of 34M reads per sample.

Further processing of the raw data was then performed using an in-house pipeline. For comparison, the RNA sequencing data (GSE89189) fromAbud et al.28 and the dataset (GSE85839) fromMuffat et al.29 were downloaded and processed in parallel. Quality control of fastq files was performed using FastQC (https://www.bioinformatics.babraham.ac.uk/projects/fastqc/) and MultiQC42. Paired-end reads were mapped to the human GRCh38.p13 reference genome (https://www.gencodegenes.org) using HISAT2 v2.2.143. Mapping quality control was done using SAMtools44 and Picard (http://broadinstitute.github.io/picard/) metrics. The counts table was obtained using FeatureCounts v2.0.145. Normalization of counts and differential expression analysis for the comparison of pMGL and co-pMG was performed using DESeq2 v1.28.146 in RStudio 1.4.1103, including the biological gender in the model and with the BenjaminiHochberg method for multiple testing correction. Exploratory data analysis was performed following variance-stabilizing transformation of the counts table, using heat maps and hierarchical clustering with the pheatmap 1.0.12 package (https://github.com/raivokolde/pheatmap) and principal component analysis. Log2 fold change (log2 fc) shrinkage for the comparison of pMGL and co-pMG was performed using the ashr package v2.2-4747. Genes with |log2 fc|>2 and adjusted p value<0.01 were defined as differentially expressed and interpreted with annotations from the Gene Ontology database using clusterProfiler v3.16.148 to perform over-representation analyses.

Equal amounts of RNA (30ng) were reverse-transcribed to cDNA using the High-Capacity cDNA Reverse Transcription Kit (ThermoFisher) according to the manufacturers instructions. Quantitative real-time PCR was performed with Fast SYBR Green Master Mix (ThermoFisher) according to the manufacturers instructions using a LightCycler 480 PCR System (Roche). The following primers (ChAT from Eurofins Genomics, all others from ThermoFisher) were used:

Quantification of the relative fold gene expression of samples was performed using the 2Ct method with normalization to the GAPDH reference gene.

AH016-3 Lenti-IP-RFP-microglia were co-cultured with healthy control motor neurons in PEI- and Geltrex-coated glass bottom dishes for confocal microscopy (VWR). The RFP signal was used to identify microglia in co-culture. To visualize microglial movement, images of the RFP signal and brightfield were taken every~30s for 1h (22 stitched images, 20magnification) using a Cell Observer spinning disc confocal microscope (Zeiss) equipped with an incubation system (37C, 5% CO2). To image phagocytic activity, co-cultures were rinsed with Live Cell Imaging Solution (1X, ThermoFisher), and pHrodo Green Zymosan Bioparticles Conjugates (P35365, ThermoFisher) diluted in Live Cell Imaging Solution (50g/mL), which become fluorescent upon phagocytic uptake, were added. The dish was immediately transferred to the spinning disc confocal microscope, and stitched images (33, 20magnification) were acquired every 5min for 2h.

To induce pro-inflammatory (M1) or anti-inflammatory (M2) microglial phenotypes, cells were treated with Lipopolysaccharides (LPS, 100ng/mL, Sigma) and Interferon- (IFN-, 100ng/mL, ThermoFisher), or Interleukin-4 (IL-4, 40ng/mL, R&D Systems) and Interleukin-13 (IL-13, 20ng/mL, Peprotech), respectively, for 18h. Vehicle-treated (co-culture medium) cells were used as an unstimulated (M0) control.

To analyze the clustering of microglia upon pro-inflammatory and anti-inflammatory stimulation, RFP-positive microglia were imaged directly before the addition of M1/M2 inducing agents, and at 9h and 18h post-stimulation using the Cell Observer spinning disc confocal microscope (55 stitched images, 10magnification). The number of individual microglial cells and size of microglial clusters was quantified using the analyze particle function in Fiji.

After stimulation with M1/M2-inducing agents, culture supernatants were collected and spun down at 1200g for 10min at 4C. Pooled samples from three different healthy control lines for each cell type were analyzed using the Proteome Profiler Human XL Cytokine Array Kit (R&D Systems) according to the manufacturers instructions. The signal was visualized on a ChemiDoc MP imaging system (Bio-Rad) and analyzed using ImageStudioLite v5.2.5 (LI-COR). Data was then plotted as arbitrary units using the pheatmap 1.0.12 package in RStudio 1.4.1103.

In addition, to confirm the relative expression of Serpin E1 and CHI3L1 in cell culture supernatants, the Human Human Chitinase 3-like 1 Quantikine ELISA Kit (DC3L10) and Human Serpin E1/PAI-1 Quantikine ELISA Kit (DSE100, both R&D Systems) were used according to the manufacturers instructions.

pNeuron, pMGL and co-cultures were plated and maintained in WillCo-dish Glass Bottom Dishes (WillCo Wells) for 14days. Calcium transients were measured using the fluorescent probe Fluo 4-AM according to the manufacturers instructions (ThermoFisher). Cells were incubated with 20M Fluo 4-AM resuspended in 0.2% dimethyl sulfoxide for 30min at RT in Live Imaging Solution (ThermoFisher). After a washing step with Live Imaging Solution, cells were allowed to calibrate at RT for 1520min before imaging. Ca2+ images were taken by fluorescence microscopy at RT. The dye was excited at 488nm and images were taken continuously with a baseline recorded for 30s before stimulation. The stimuli used for calcium release were 50mM KCl (Sigma-Aldrich) for 30s, followed by a washing step for one minute. Microglial calcium release was stimulated by 50M ADP (Merck) under continuous perfusion for 1min, followed by a 1-min wash. Analysis of fluorescence intensity was performed using Fiji. Fluorescence measurements are expressed as a ratio (F/Fo) of the mean change in fluorescence (F) at a pixel relative to the resting fluorescence at that pixel before stimulation (Fo). The responses were analysed in 2040 cells per culture.

MNs on DIV 3345 were maintained in a bath temperature of 25C in a solution containing 167mM NaCl, 2.4mM KCl, 1mM MgCl2, 10mM glucose, 10mM HEPES, and 2mM CaCl2 adjusted to a pH of 7.4 and 300mOsm. Electrodes with tip resistances between 3 and 7M were produced from borosilicate glass (0.86mm inner diameter; 1.5mm outer diameter). The electrode was filled with intracellular solution containing 140mMK-Gluconate, 6mM NaCl, 1mM EGTA, 10mM HEPES, 4mM MgATP, 0.5mM Na3GTP, adjusted to pH 7.3 and 290mOsm. Data acquisition was performed using a Multiclamp 700B amplifier, digidata 1550A and clampEx 6 software (pCLAMP Software suite, Molecular Devices). Data was filtered at 2kHz and digitized at 10kHz. Series resistance (Rs) was continuously monitored and only recordings with stable<50 M and Rs<20% were included in the analysis. Voltage gated channel currents were measured on voltage clamp, neurons were pre-pulsed for 250ms with 140mV and subsequently a 10mV-step voltage was applied from 70 to+70mV. Induced action potentials were recorded on current clamp, neurons were held at 70mV and 8 voltage steps of 10mV, from 10 to 60mV, were applied. Data was analyzed using Clampfit 10.7 (pCLAMP Software suite).

Statistical analyses were conducted using GraphPad Prism 9 (GraphPad Software, San Diego, California USA, http://www.graphpad.com). Comparisons of two groups were performed by two-tailed unpaired t-tests and multiple group comparisons by one-way or two-way analysis of variance (ANOVA) with appropriate post-hoc tests as indicated in the figure legends. The statistical test and number of independent experiments used for each analysis are indicated in each figure legend. Data are presented as single data points and meansSEM. Differences were considered significant when P<0.05 (*P<0.05; **P<0.01; ***P<0.001; ns: not significant). GraphPad Prism 9 or RStudio 1.4.1103 were used to plot data. Final assembly and preparation of all figures was done using Adobe Illustrator 25.4.1.

Read this article:
Human iPSC co-culture model to investigate the interaction between microglia and motor neurons | Scientific Reports - Nature.com