Additional file 1 of Alzheimer’s disease protective allele of Clusterin modulates neuronal excitability through lipid-droplet-mediated neuron-glia communication

Additional file 1. Figure S1: Bioinformatic and experimental validation of the regulatory effect of rs1532278 on TF-binding and CLU expression, related to Fig. 1. (A) Multiz alignment and phyloP conservation (470 mammals) around rs1532278 (from UCSC hg38 genome browser). (B) JASPAR predicted TF binding sites at rs1532278 and TF expression levels in iGlut (10/23 predicted TFs can be detected by RNA seq). CPM, counts per million reads. (C) Representative images (CD07 line) of iGlut of all three genotypes are also shown, related to Fig. 1D (bottom panel); GFAP and HuNu (human nuclear antigen) staining shows specificity of HuNu and MAP2 staining for iGlut in iGlut-mAst co-cultures. (D) No difference of iGlut differentiation efficiency were found between T/T and CC carriers in iGlut-mAst co-cultures, and none proliferating cells were observed in these neurons indicated by negative staining of Ki67. HuNu +, human cells. n=5 coverslips per group from one differentiation of both CD05 and CD07 lines (2-3 clones per line, one coverslip per clone and 4-5 images per coverslip). (E) DRGX ChIP-qPCR for iGlut-mAst co-cultures of CD07 line on day 30. n=3 biological replicates per group (one clone with 3 biological replicates from the CD07 line) from one independent differentiation. (F-G) ISL2 siRNA knockdown in day-30 pure iGlut (C/C) cultures. Samples of 72 hours post-siRNA transfection were used for qPCR. n=3 biological replicates from one clone per line in one independent differentiation. (H) CLU mRNA levels in iGlut pure cultures. n=6 biological replicates per group (2-3 clones per line and 2-3 biological replicates for each clone) from two independent differentiations of each line (I) sCLU levels detected by ELISA from the supernatant of iGlut pure cultures. n=4 biological replicates per group (2 clones per line and 2 biological replicates for each clone) from two independent differentiations of each line. (J) CLU mRNA levels of mAst in iGlut-mAst co-cultures. n=4 biological replicates per group (2 clones per line and 2 biological replicates for each clone) from two independent differentiations of each line. (K) Three qPCR primer sets that can capture the majority of CLU transcript isoforms (14/17) are used in (L) to detect CLU mRNA levels. Only iGlut-mAst co-cultures of the CD07 line was used. n=6 biological replicates per group (3 biological replicates per clone for 2 clones) from two independent differentiations. (M) rs1532278 did not alter the expression of SCARA3, a CLU-adjacent gene, in iGlut (or when co-cultured with mAst) of the isogenic pairs of both CD05 and CD07 lines. n=10 biological replicates per group (2 biological replicates per clone and 2-3 clones per line) from two independent differentiations of both CD05 and CD07 lines. (N) Representative images (CD05 line) of CLU staining, related to Figure 1K and 1L. Data = mean±SEM. * p<0.05, ** p< 0.01, *** p< 0.001, and **** p< 0.0001. Figure S2. Quality control of the CRISPR/Cas9-edited iPSC lines, related to Fig. 1.(A) Sanger sequencing of top off-target sites (Benchling prediction) with 2 and 3 mismatched sites with gRNA in one clone of the edited alleles in both CD05 and CD07 lines. Note no off-target editing was found. (B) eSNP-karyotyping using RNA-seq data of iGlut of the two isogenic pairs of CRISPR-edited lines (CD05 and CD07; only the iPSC clones used for cellular functional assays were analyzed). Left panel, moving average of the SNP allelic ratio (RNA-seq reads of each allele) along the genome; right panel, stretches of SNP heterozygosity of all common SNPs for each chromosome. Note that no chromosomal abnormalities were found. (C) CLU, ISL2, and DRGX expression in postmortem brain transcriptomic dataset hub ( https://cellxgene.cziscience.com/ ). (D-E) CLU mRNA Levels in iAst differentiated from the isogenic pairs of CRSIPR/Cas9-edited iPSC lines (for donor lines CD05 and CD07). Representative immunofluorescence staining (CD05 line) of S100β and vimentin shows the identity and purity of astrocytes. n=10 biological replicates per group (2-3 clones per line and 2 replicates per clone) from two independent differentiations of both CD05 and CD07 lines. Data, mean±SEM. Figure S3. Synaptic and electrophysiological properties of iGlut carrying TT or C/C alleles of rs1532278 at the CLU locus, related to Fig. 2. (A) Quantification of SYP and PSD-95 puncta density. n=15-17 neurons per group (1-2 neuron per coverslips, 5-6 coverslips per clone, and 2 clones per line) from two independent differentiations of each line. (B-C) Western blotting shows PSD-95 and SYP levels in day-30 iGlut pure cultures. n=4 biological replicates per group (2 clones per line and 2 biological replicates per clone) from two independent differentiations of each line. (D-E) Number of bursts and synchronicity analysis in MEA. AUNCC: Area Under Normalization Cross-Correlation. n=9-12 biological replicates per group (2-3 clones per line and 3-4 biological replicates per clone) from two independent differentiations of each line. Data, mean±SEM. * p<0.05, ** p< 0.01, *** p< 0.001, and **** p< 0.0001. Figure S4. Characterization of neurons carrying the rs1532278-flanking OCR deletion or with exogenous CLU overexpression, related to Fig. 3. (A) Sanger sequencing traces the CRISPR/Cas9-engineered homozygous OCR deletion (representative result of the CD07 line). (B) CLU mRNA levels (n=6 biological replicates per group from two independent differentiations, 2 clones per line, and 3 biological replicates for each clone) and sCLU levels (n=4 biological replicates per group from two independent differentiations, 2 clones per line, and 2 biological replicates for each clone) in iGlut pure co-cultures. (C) Immunofluorescence staining of neuronal CLU in iGlut pure cultures on days 23-25. (D) Quantification of CLU staining shown in panel (C). n=7 coverslips per group (2 clones per line, 3-4 coverslips for each clone, and 3-5 images per coverslip; shown are example images of CD05) from two independent differentiations of each line. (E-F) Western-blot for SYP in day-30 iGlut-mAst co-cultures. n=4 biological replicates per group (2 clones per line, and 2 biological replicates for each clone) from two independent differentiations of each line. (G-H) Number of neuron network bursts and synchronicity analysis in MEA. AUNCC: Area Under Normalization Cross-Correlation. n=5-10 biological replicates per group (2 clones per line and 2-5 biological replicates per clone) from two independent differentiations of each line (I) Immunofluorescence staining (CD07 line) of MAP2 and CLU (Flag tag) to determine AAV-hCLU transduction efficiency (~100%). (J) qPCR to confirm CLU overexpression. n=3 biological replicates per group from one clone of CD07 line in one independent differentiation. (K) Number of network bursts and synchronicity analysis in MEA. AUNCC: Area Under Normalization Cross-Correlation. n=3 biological replicates per group (all from one clone of line CD07) in one independent differentiation. Data, mean±SEM. Scale bars are indicated in corresponding images. * p<0.05, ** p< 0.01, *** p< 0.001, and **** p< 0.0001. Figure S5. Transcriptomic analysis of iGlut-mAst co-cultures and validation, related to Fig. 4. (A) Correlation analysis Salmon sorted iGlut data set to our previous published single-cell data sets iGlut (CD05 and CD07) and others published single cell data sets in postmortem brain with various cell types including ex-neuron, ih-neuron, microglia, astrocyte, oligodendrocytes, and OPC. Ast: astrocyte, ex: excitatory neuron, ih: inhibitory neuron, mic, microglia, Oli: oligodendrocyte, and OPC: oligodendrocyte progenitor cell. (B) MDS (Multidimensional scaling) analysis of all transcripts in iGlut and mAst, which are expressed in ≥75% of samples. n=9 biological replicates per group from two independent differentiations of each line, 2-3 biological replicates per clone, 2 clones per line, and 2 lines in total. (C) ELISA quantification of Aβ in the supernatants of iGlut-mAst co-cultures and iGlut pure cultures. n=4 biological replicates per group (2 clones per line and 2 replicates for each clone) from two independent differentiations of each line. (D) GO Biological process enrichment in significantly downregulated gene sets in iGlut. Blue stars indicate neuron morphology-related pathways. (E) Confirmation of SYP, PSD-95, and CLU expression levels in iGlut in RNA seq analysis. (F) Confirmation of human CLU expression levels by using in-silico qPCR approach to exclude mouse CLU reads interference. Violin plots showed data with median and interquartile range, and all other statistical graphs showed data with mean±SEM. * p<0.05, ** p< 0.01, *** p< 0.001, and **** p< 0.0001. Figure S6. The assessment of the Impact of mouse CLU on human CLU expression in iGlut-mAst co-culture system (related to Figures 1, 3, 4, and 6) and the Analysis Procedure of LipidTox staining in iGlut pure cultures (related to Fig. 5).(A-B) The mRNA expression levels, extracellular (secreted) levels, and intracellular levels of both hCLU and mCLU were analyzed across iGlut-mAst co-cultures and iGlut and mAst mono-cultures in the SNP isogenic pair (A) and CLU overexpression assays (B). To ensure comparability among different cultures, iGlut cells in both mono- and co-cultures were seeded, collected, and processed simultaneously, with the same approach applied to mAst mono- and co-cultures. n=3 biological replicates per group from one clone of CD05 line in one independent differentiation, and C/C of CD05 line was used in CLU overexpression assay. (C) The representative images from CD07 line showing LipidTox and NeuN co-staining analysis. NeuN signals were utilized to create a mask using the CellProfiler program, which excluded any contaminated signals from dead cells in both LipidTox and DAPI staining. The LipidTox signals were then quantified, and cell numbers were determined based on DAPI signals. Data, mean±SEM. Scale bars are indicated in corresponding images. * p<0.05, ** p< 0.01, *** p< 0.001, and **** p< 0.0001. Figure S7. Validation of neuronal CLU as a lipid shuttle in lipid transfer system, related to Fig. 5. (A) A schematic diagram illustrating the experimental setup: Flag-tagged CLU was overexpressed in iGlut using AAV (refer to Fig. 3G), followed by co-culture with mAst in the lipid transfer system (refer to Fig. 5E). Specifically, Day-5 iGlut were seeded on coverslips and infected with AAV-hCLU-Flag at a multiplicity of infection (MOI) of 105 on Day 8. From Day 26 to Day 28, iGlut were co-cultured with mAst pre-seeded on separate coverslips (positioned above with Parafilm separators). GFAP staining was performed to confirm mAst identity, while Flag and MAP2 staining verified successful AAV infection. Representative images from CD05 line. (B) Co-staining of BD493/563 and Flag in mAst was conducted to confirm the co-localization of neuronal CLU and lipid droplets (LDs) in mAst. (C) Validation of 2nd antibody (against Flag antibody) specificity in mAst from experiment (A). (D) Exclusion of gravity effects was tested by reversing the system in (A), placing the mAst coverslip at the bottom and the iGlut coverslip at the top. Representative images from CD05 line. (E) Validation of Flag antibody specificity in only mAst culture. Figure S8. Neuronal CLU facilitates lipid transfer to astrocytes and LD accumulation, related to Fig. 6. (A) The process of cell segmentation in iGlut-mAst co-cultures, related to Fig. 6A. (B) LDs staining and quantification by BD493/503 in day-30 iGlut-mAst co-cultures. n=5-6 coverslips per group (2 clones per line, 2-3 coverslips for each clone with 4-6 images per coverslip, and shown are example images of CD07) from one independent differentiation of each line. Violin plots are shown with data median and interquartile range. Other data, mean±SEM. Scale bars are indicated in corresponding images. * p<0.05, ** p< 0.01, *** p< 0.001, and **** p< 0.0001. Figure S9. Transcriptomic data provide mechanistic support to lipid accumulation and ROS homeostasis in mAst, related to Figs. 6 and 7.(A) Digitonin permeabilization (~0.01%) does not affect CellROX staining (representative images from CD05 co-cultures). (B) iGlut cell numbers in CellRox Staining. n=25-29 images per group (2 clones per line, 2-3 coverslips for each clone with 2-5 images per coverslip) from two independent differentiations of each line. (E) ELISA measurement of CLU level from supernatant after immunodepletion, related to Fig. 7C. n=3 technical replicates per group, all from one supernatant collection of CD07 line. (D) A Sankey plot depicts glutathione metabolism and oxidative stress related pathways in mAst from Wiki pathway analysis (E) Lysosome and autophagy related GO items are enriched in DE gene lists of iGlut and mAst. Violin plots are shown with data median and interquartile range. Other data, mean±SEM. Scale bars are indicated in corresponding images. * p<0.05, ** p< 0.01, *** p< 0.001, and **** p< 0.0001. Figure S10. Raw images for all western blot.