Additional file 1 of TREM2 Alzheimer’s variant R47H causes similar transcriptional dysregulation to knockout, yet only subtle functional phenotypes in human iPSC-derived macrophages

Additional file 1 : Figure S1. Validation of R47H genotype. (A) CRISPR single guide RNA used for insertion of R47H mutation by Bioneer. (B) Chromatograms from sequencing of WT line BIONi010-C and R47H TREM2 line BIONi010-C-7. Red asterisk indicates the R47H mutation, black asterisks are silent mutations added by Bioneer to prevent re-cutting. Figure S2. SNP microarray of iPSCs. Chromosome karyograms from Illumina microarray SNP analysis, showing (A) BIONi010-C line, (B) BIONi010-C-7 R47H TREM2 line, (C) BIONi010-C-17 TREM2 KO line. Figure S3. Validation of R47H TREM2 and TREM2 KO pMac. (A) Macrophage surface markers CD11b, CD14, and CD45 measured by flow cytometry. Median fluorescence intensity (MFI) for each sample was normalized to the relevant isotype IgG, and then to the average for the three genotypes. Histogram shows means ± SEM, for n=3-4 harvests. 1-way ANOVA with Dunnett’s post-hoc test, comparisons to WT line. ** p < 0.01, *** p < 0.001, **** p < 0.0001, all unannotated comparisons are not significant. (B) Total levels of TREM2 protein shown in a representative western blot (WB). (C-D) Surface TREM2 measured by immunofluorescence staining (IF): live pMac were stained with TREM2 antibody, followed by fluorescent secondary antibody, and subsequently fixed. Images are maximum projections from a z-stack of 5 slices, 1-5 μm, taken on an Opera Phenix microscope (Perkin Elmer). Quantified mean fluorescence (per μm2), for triplicate wells, was normalised to the average for the three genotypes, and then expressed as a ratio of whole-cell TREM2 staining from separate permeabilised wells on the same plate (D). Means ± SEM, for N=3 harvests, p = 0.047 in one-tailed paired t-test. (E-F) Kinetics of pMac calcium responses to 0.5 mM ATP (E), and 10 μg/mL TREM2 antibody (F). Means ± SEM, for N=3-5 harvests. Figure S4. Validation of antibodies for TREM2 immunocytochemistry. Fixed and permeabilized WT, R47H, and TREM2 KO pMac were stained for 1 hour at RT with three different TREM2 antibodies at the concentrations indicated, followed by staining with Alexa Fluor 488-conjugated secondary antibody (1:1000, Invitrogen). Cells were counterstained with DAPI nuclear dye and imaged on an EVOS FL Auto automated microscope (Thermo Fisher). Ab209814 showed cytoplasmic staining in all three genotypes, 13,483–1-AP showed nuclear staining in all three genotypes, whereas AF1828 stained cytoplasm and plasma membrane in WT and R47H TREM2 pMac but not TREM2 KO pMac. Scale bar is 100 μm. Figure S5. Validation of dead SH-SY5Y phagocytosis assay. (A) Freshly-fixed SH-SY5Ys stain uniformly for phosphatidylserine exposure (annexin V-FITC), but have limited cell permeability (propidium iodide). Live SH-SY5Ys do not stain for annexin V-FITC or propidium iodide, except for focal staining present on the few dead cells in culture. (B) No TREM2 expression in an SH-SY5Y not undergoing phagocytosis, marked with a white arrow. (C) No RAB9 expression in non-engulfed SH-SY5Ys, marked with a white arrow. (D) Dose-dependent uptake of dead SH-SY5Ys after 5 hours of phagocytosis with WT line BIONi010-C, means quantified from three independent experiments for % of spot positive (phagocytic) cells per well. Means ± SEM, for N=3 harvests. (E) Phagocytosis of 3 hours is inhibited with 10 μM cytochalasin D, 1 μM bafilomycin A1, 1 μM jasplakinolide, all with 1 hour pre-treatment, and 13 μg/mL recombinant annexin V added simultaneously to the dead SH-SY5Ys. Data was normalized to mean for each genotype per experiment. Means ± SEM, for N=3-6 harvests and with two WT cell lines (SFC840-03-03, the characterisation of this line is described in Fernandes et al [32], and BIONi010-C). 1-way ANOVA with Dunnett’s post-hoc test, comparisons to untreated cells. * p < 0.05, *** p < 0.001. Figure S6. Validation of synaptosome phagocytosis assay. (A) Two whole synaptosomes surrounded by cell debris in the cryopreserved prep, visualised by negative staining electron microscopy. White asterisks label the pre-synaptic termini, with many pre-synaptic vesicles, whereas purple asterisks label the post-synaptic termini. A dark post-synaptic density can be seen between connected pre- and post-synaptic termini. (B) Synaptosomes stain uniformly for phosphatidylserine exposure (annexin V-FITC), comparison is with unstained synaptosomes. An area magnified by 5X is shown inset. (C) Dose-dependent uptake of dead SH-SY5Ys after 3 hours of phagocytosis with WT line BIONi010-C, reaching saturation above 30 μg. (D) Phagocytosis in BIONi010-C pMac is inhibited by 10 μM cytochalasin D and 1 μM bafilomycin A1, and increased by prior opsonisation of synaptosomes for 30 minutes with 20% human serum. Data was normalized to mean for each genotype per experiment, and is represented as sum of spot areas (μm2) per cell. Means ± SEM, for N=3-4 harvests. 1-way ANOVA with Dunnett’s post-hoc test, comparisons to untreated cells. * p < 0.05, ** p < 0.01. Figure S7. Validation for cytokine ELISAs and transwell chemotaxis assay. Cytokine ELISAs: (A) Secretion of TNF in response to 4 hours of 0.1-1 μg/mL LPS. (B) Secretion of IL-6 in the same supernatants as (A). Means ± SEM, for N=3 harvests. 2-way ANOVA with Dunnett’s post-hoc test. Comparisons with the coloured annotations are stimulations versus untreated cells (None) for each genotype. Comparisons with the black annotations are R47H or KO versus the WT line for each stimulation, all unannotated comparisons are not significant. Transwell chemotaxis assay: (C) Migration of WT pMac in transwell chemotaxis assay in the presence of four concentrations of ADP or C5a, for 6 hours. (D) Migration of WT pMac for 6 hours in the presence of 30 μM ADP is attenuated by 30 minutes pre-treatment with a P2RY12-selective inhibitor (PSB0739), but not a P2RY1 (MRS2179) or P2RY13 (MRS2211) inhibitor. (E) Migration of WT pMac for 6 hours in the presence of 3 nM C5a is attenuated by 30 minutes pre-treatment with a C5aR inhibitor (PMX-53), or a Syk inhibitor (OXSI-2, 3 μM). Means ± SEM, for N=3-4 harvests. 1-way ANOVA with Dunnett’s post-hoc test, pairwise comparisons to control. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. Figure S8. RNA-seq differentially-expressed genes (DEGs). Volcano plots shown for DEGs relative to WT: (A) TREM2 KO and (B) R47H TREM2 pMac. Dashed lines show cut-offs at log2-fold-change=2 and p=0.001. Enrichment of Gene Ontology (GO) terms in significant (adjusted p value <0.05) DEGs relative to WT: (C) TREM2 KO, and (D) R47H TREM2 pMac. Top 30 terms shown in order of the adjusted p value, the relative R–score represents the average direction of change. Circle size corresponds with number of DEGs. Figure S9. Flow cytometry of integrins. Representative flow cytometry histograms for Figure 6e. Figure S10. Proportions of TREM2 transcript splice variants. TREM2 differential transcript usage analysis was performed on the RNA-seq data with the DRIMSeq bioconductor package, using Salmon-quantified transcript counts. DRIMSeq assumes an Dirichlet Multinomial model (DM) for each gene, where the total count for the gene is considered fixed, and the quantity of interest is the proportion for the transcript within a gene for each sample. A likelihood ratio test was used to test for gene and transcript level DTU between R47H vs WT, and no significant difference in transcript usage was found. Transcripts are described in Del-Aguila et al [62]: ENST00000373113 is the canonical TREM2 transcript and the longest, with five exons; ENST00000373122 is the second longest and lacks the 5’ exon but includes the transmembrane domain; ENST00000338469 is the shortest and excludes the transmembrane domain. The bars represent three sequential ages of iPSC-macrophages. Figure S11. Correlation of TREM2 KO DEGs with Claes et al (2019). Log-2-fold change values for the significant DEGs (adjusted p value <0.05) of the current study (purple) plotted against significant DEGs from Claes et al (orange). Overlapping significant DEGs shown in green.