Genetic contribution to microglial activation in schizophrenia

Abstract

Several lines of evidence indicate the involvement of neuroinflammatory processes in the pathophysiology of schizophrenia (SCZ). Microglia are brain resident immune cells responding toward invading pathogens and injury-related products, and additionally, have a critical role in improving neurogenesis and synaptic functions. Aberrant activation of microglia in SCZ is one of the leading hypotheses for disease pathogenesis, but due to the lack of proper human cell models, the role of microglia in SCZ is not well studied. We used monozygotic twins discordant for SCZ and healthy individuals to generate human induced pluripotent stem cell-derived microglia to assess the transcriptional and functional differences in microglia between healthy controls, affected twins and unaffected twins. The microglia from affected twins had increased expression of several common inflammation-related genes compared to healthy individuals. Microglia from affected twins had also reduced response to interleukin 1 beta (IL1β) treatment, but no significant differences in migration or phagocytotic activity. Ingenuity Pathway Analysis (IPA) showed abnormalities related to extracellular matrix signaling. RNA sequencing predicted downregulation of extracellular matrix structure constituent Gene Ontology (GO) terms and hepatic fibrosis pathway activation that were shared by microglia of both affected and unaffected twins, but the upregulation of major histocompatibility complex (MHC) class II receptors was observed only in affected twin microglia. Also, the microglia of affected twins had heterogeneous response to clozapine, minocycline, and sulforaphane treatments. Overall, despite the increased expression of inflammatory genes, we observed no clear functional signs of hyperactivation in microglia from patients with SCZ. We conclude that microglia of the patients with SCZ have gene expression aberrations related to inflammation response and extracellular matrix without contributing to increased microglial activation.

Fig. 1. Characterization of human iMGLs derived from monozygotic twins discordant for SCZ.

a Timeline for iMGL differentiation. Scale bar 50 µm in all. Generated iMGLs expressed (b) on protein level pan-macrophage marker IBA1, and microglia-specific markers P2Y12, TREM2 and CX3CR1 (nuclei on blue, scale bar 50 µm), and (c) on transcriptomic level AIF1 (IBA1; H(3) = 10.08, p = 0.0035), ITGAM (CD11b; H(3) = 10.93, p = 0.001; HPC vs. iMGL U = 0, p = 0.0286; iMGL vs. iMGL + LPS U = 0, p = 0.0286), P2RY12 (H(3) = 8.717, p = 0.014; HPC vs. iMGL U = 0, p = 0.0286; iMGL vs. iMGL + LPS U = 0, p = 0.0286) and TREM2 (H(3) = 9.175, p = 0.009; HPC vs. iMGL U = 0, p = 0.0286; iMGL vs. iMGL + LPS U = 0, p = 0.0286). Kruskal–Wallis test, Mann–Whitney test, n = 3–4 CTRL lines. d Illustration of patient cohort created with BioRender. e RNA expression of AIF1, TREM2, PTPRC (CD45) and TMEM119 based on bulk RNA sequencing. CTRL healthy individuals, ST affected twin, HT healthy twin, HPC hematopoietic stem cell, iMφ hiPSC-macrophage, iMGL hiPSC-microglia like cell, iMGL + LPS LPS-treated iMGL.

Fig. 2. Inflammation, cytokine release and phagocytosis in iMGLs.

a Gene expression of inflammatory related HLA-DRA, TSPO and IL1β based on RNA sequencing. b Representative image after 4-h migration assay. Cells were masked with magenta for analysis. Scale bar 800 µm. c The number of migrated cells normalized to unstimulated condition for each group. Performed Kruskal–Wallis test (H(2) = 2.371, p = 0.3305; H(2) = 0.7829, p = 0.6927). d Release of IL6 (CTRL: χ²(3) = 10.33, p = 0.0030; ST: χ²(3) = 1.114, p = 0.8307; HT: χ²(3) = 5.864, p = 0.1132), IL8 (CTRL: χ²(3) = 8.400, p = 0.0190; ST: χ²(3) = 7.500, p = 0.0517; HT: χ²(3) = 7.800, p = 0.0443) and MCP1 (CTRL: χ²(3) = 5.100, p = 0.1897; ST: χ²(3) = 11.10, p = 0.0009; HT: χ²(3) = 9.72, p = 0.0120) cytokines after 24 h treatment with either 20 ng/ml IL1β, 20 ng/ml TNFα or 20 ng/ml IFNγ. Friedman test and Mann–Whitney test. e Gene expression of TLR2 and TLR4. f Representative image from phagocytosis after 6 h (Scale bar 100 µm) and (g) phagocytosis of pHrodo-zymosan bioparticles. h Phagocytosis between the groups after 6 h (H(2) = 0.7800, p = 0.7027). Kruskal–Wallis test. n = 4–5 cell lines.

Fig. 3. Expression of neurotrophic factors and neural activity in co-culture model.

a Neurotrophic factor BDNF, GDNF and MANF and (b) neurotransmitter receptor GRIN2D, GRIK2 and GABRA2 gene expression. c Timeline for co-culture design and MEA recording. d Representative images from CTRL iMGL co-cultures at 30 and 60 days in vitro (DIV) with stainings for NGN2-neurons (MAP2, red), rat astrocytes (GLT1, magenta), iMGLs (ABI3, green), and nuclei (DAPI). e Representative raster plots from weekly CTRL iMGL co-culture MEA recordings. Spikes detected on individual electrodes marked with black lines, bursts with blue lines and network bursts with purple boxes. f MEA recording results as Number of spikes (95% CI of diff. = −27,769 to −3067, adj.p = 0.0212; 95% CI of diff. = −11,671 to −1434, adj.p = 0.0104; 95% CI of diff. = −8259 to −331.2, adj.p = 0.0366; 95% CI of diff. = −19,980 to −4147, adj.p = 0.0113), Weighted mean firing rate (95% CI of diff. = −2.821 to −0.4223, adj.p = 0.0152; 95% CI of diff. = −1.418 to −0.07197, adj.p = 0.0288; 95% CI of diff. = −2.117 to −0.3807, adj.p = 0.0139), Network IBI coefficient of variation (95% CI of diff. = −1.499 to −0.1199, adj.p = 0.0222; 95% CI of diff. = −1.181 to −0.4205, adj.p = 0.0060), and Synchrony index (95% CI of diff. = −0.6299 to −0.09522, adj.p = 0.0125; 95% CI of diff. = −0.6945 to −0.09386, adj.p = 0.0147) from co-culture with and without iMGLs. n = 5–7 wells for no iMGL and n = 6–10 wells from two CTRL iMGL lines. g MEA recording results as Number of spikes, Weighted mean firing rate, ISI coefficient of variation, Number of bursts, Burst duration (95% CI of diff. = −0.1197 to −0.007484, adj.p = 0.0368; 95% CI of diff. = −0.2053 to −0.05687, adj.p = 0.0165), Mean ISI within burst (95% CI of diff. = 0.0004615 to 0.01288, adj.p = 0.0436; 95% CI of diff. = 0.002695 to 0.01507, adj.p = 0.0248; 95% CI of diff. = 0.005213 to 0.007635, adj.p = 0.0019), Network IBI coefficient of variation (95% CI of diff. = −1.574 to −0.3703, adj.p = 0.0197; 95% CI of diff. = −1.173 to −0.1141, adj.p = 0.0331), and Synchrony index (95% CI of diff. = −0.3168 to −0.1937, adj.p = 0.0030; 95% CI of diff. = −0.6867 to −0.03380, adj.p = 0.0414) from ST, HT and CTRL iMGL co-cultures. n = 3 lines per group. Dunnett’s (timepoints compared to week 4 timepoint) or Tukey’s (comparison between with/without iMGL groups) multiple comparisons tests were used for significance. CTRL healthy individuals, ST affected twin, HT healthy twin.

Fig. 4. iMGL expression profile and pathway analysis.

a Overview of differentially expressed genes (DEGs) in different comparisons. b Venn diagram of shared DEGs between HT vs. CTRL and ST vs. CTRL. Ingenuity Pathway Analysis (IPA) canonical pathways (c) from ST vs. CTRL and (d) HT vs. CTRL comparisons. Cutt-offs: Fold change > abs 2, adjusted p < 0.05. e Downregulated and (f) upregulated Gene Ontology (GO) Molecular function terms from ST vs. CTRL comparison and (g) downregulated terms from HT vs. CTRL comparison. CTRL healthy individuals, ST affected twin, HT healthy twin. RNAseq n = 4 lines in each group.

Fig. 5. iMGL responses to different drug treatments.

a Viability of iMGLs after 24-h treatment with clozapine (W = −21.00, p = 0.0313 all), minocycline and sulforaphane (W = −21.00, p = 0.0313). Tested with four independent experiments with two control, two HT and two ST lines (n = 6 lines) (Wilcoxon test). b NFκB pathway downstream gene FOS. Heatmaps of CTRL, ST and HT line gene expression changes after (c) 10 µM clozapine (d) 10 µM minocycline and (e) 5 µM sulforaphane treatments. No data from ST5 patient after clozapine and sulforaphane. f Gene expression differences after drug treatments in BDNF, HLA-DRA, TLR2 and IL1β. RNAseq n = 3–4 lines in each group. CTRL healthy individuals, ST affected twin, HT healthy twin.

Fig. 6. Drug treatment protection from LPS-induced inflammation in iMGLs.

a IL6 (H(2) = 0.9800, p = 0.6500), IL8 (H(2) = 1.8600, p = 0.4163) and MCP1 (H(2) = 1.220, p = 0.5824) secretion after 24-h 100 ng/ml LPS treatment between the groups. Kruskal–Wallis test. b LPS effect on pHrodo-zymosan phagocytosis after 24-h LPS treatment and 6 h after adding pHrodo bioparticles (CTRL: H(4) = 3.397, p = 0.4937; ST: H(4) = 3.094, p = 0.5422; HT: H(4) = 2.592, p = 0.6282). Kruskal–Wallis test. c IL6 (χ²(4) = 5.400, p = 0.1514), IL8 (χ²(4) = 6.360, p = 0.0933), MCP1 (χ²(4) = 3.480, p = 0.3720) and TNFα (χ²(4) = 3.735, p = 0.3141) secretion in CTRL iMGL cultures after 30 min pre-treatment with 10 µM clozapine, 10 µM minocycline or 5 µM sulforaphane and 24-h 100 ng/ml LPS treatment. Friedman test, n = 5 lines.