Glycosylated clusterin species facilitate Aβ toxicity in human neurons

doi:10.21203/rs.3.rs-1439113/v1

Download PDF

Research article

Glycosylated clusterin species facilitate Aβ toxicity in human neurons

https://doi.org/10.21203/rs.3.rs-1439113/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: Clusterin (CLU) is one of the most significant genetic risk factors for late onset Alzheimer’s disease. However, the mechanisms by which CLU contributes to AD development and pathogenesis remain unclear. Numerous studies have demonstrated that CLU-AD mutations and amyloid-β (Aβ) treatment alter the trafficking and localisation of glycosylated CLU proteins, which may contribute to AD pathogenesis. However, the roles of non-glycosylated and glycosylated CLU proteins in mediating Aβ toxicity have not been studied in human neurons.

Methods: iPSCs with altered CLU trafficking were generated following the removal of CLU exon 2 by CRISPR/Cas9 gene editing. Neurons were generated from unedited wildtype, control unedited and exon 2 -/- edited iPSCs and were incubated with aggregated Aβ peptides. Changes in cell death and neurite length were quantified to determine if altered CLU protein trafficking influenced neuronal sensitivity to Aβ. Finally, RNA-Seq analysis was performed to identify key transcriptomic differences between CLU exon 2 -/- and CTR neurons.

Results: The removal of CLU exon 2, and the ER-signal peptide located within, abolished the presence of glycosylated CLU and increased the abundance of intracellular, non-glycosylated CLU. Aβ_25-35 treatment was demonstrated to alter glycosylated CLU trafficking in control neurons, while non-glycosylated CLU levels were unaltered in both, control and exon 2 -/- neurons. Partial protection against Aβ-induced cell death and neurite retraction was demonstrated in exon 2 -/- neurons. Transcriptome analysis identified downregulation of multiple extracellular matrix (ECM) related genes in exon 2 -/- neurons, potentially contributing to their reduced sensitivity to Aβ toxicity.

Conclusions: This study identifies a crucial role of glycosylated CLU in facilitating Aβ toxicity in human neurons. The loss of these proteins reduced both, cell death and neurite damage, two key consequences of Aβ toxicity identified in the AD brain. Strikingly, transcriptomic differences between exon 2 -/- and CTR neurons were small, but a significant and consistent downregulation of ECM genes and pathways was identified in exon 2 -/- neurons. This may contribute to the reduced sensitivity of these neurons to Aβ, providing new mechanistic insights into Aβ pathologies and therapeutic targets for AD.

Clusterin (CLU)

Alzheimer’s disease (AD)

amyloid beta (Aβ)

human neurons

CRISPR/Cas9

iPSCs

extracellular matrix (ECM).

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder and is the most common cause of dementia (1). AD is characterised by the presence of two pathological hallmarks: extracellular Aβ aggregates known as Aβ plaques and intracellular neurofibrillary tau tangles (NFTs) formed of hyperphosphorylated tau proteins. A prevailing hypothesis of AD is the Amyloid Cascade Hypothesis that postulates Aβ plays a central role in initiating AD development (2–4). Aggregation of Aβ may trigger downstream events, including synaptic dysfunction, neurite damage, formation of NFTs and cognitive decline that ultimately result in AD. Major support for this hypothesis arises from the genetics of familial AD (fAD) (2) where several mutations in genes encoding proteins involved in Aβ production, including amyloid precursor protein (APP), presenilin 1 and 2 (PSEN 1/2) (5–12) have been identified. However, fAD represents only a small number of AD cases (< 1%), and it is not clear if fAD offers mechanistic insight into sporadic AD (sAD). Nevertheless, such studies provide important insight into a crucial role of Aβ in AD development. Numerous risk genes have been identified that alter an individual’s susceptibility to sAD, of which CLU is the third most significant genetic risk factor after APOE and BIN1 (13–15). However, little is known of the mechanisms by which single nucleotide polymorphisms (SNPs) and mutations in CLU alter AD risk (13,14,16).

In addition to CLU acting as a genetic risk factor for sAD, much interest has been placed on CLU’s relationship with Aβ. Secreted CLU binds to Aβ altering its solubility and aggregation (17) and contributes to the clearance of Aβ by astrocytes (Nuutinen et al 2007) and through the blood-brain barrier (18). In contrast, data from animal studies have identified a role for CLU in facilitating the deposition of Aβ and the formation of plaques (17,19,20), suggesting CLU may facilitate AD pathology in vivo. It therefore remains unclear if CLU proteins facilitate or protect against Aβ toxicity in AD.

Clusterin is a highly glycosylated, cleaved and secreted protein (sCLU) that is considered to be cytoprotective (21–23). During its biogenesis, sCLU is first translated as an immature precursor protein that undergoes phosphorylation, glycosylation and cleavage in the ER and Golgi apparatus (24–26).These processing and maturation events result in the formation of a mature, heterodimeric secreted protein consisting of an α and β chain. Although typically considered secreted, glycosylated CLU proteins have also been identified within cells, typically in the cytoplasm, although the function, origin and biogenesis of intracellular, glycosylated CLU proteins remains debated (27). Stress has been demonstrated to induce localisation and trafficking of glycosylated CLU into the cytoplasm. This may be mediated through the binding of sCLU to cell surface receptors resulting in its uptake back into cells (28,29), or via premature release of CLU from the secretory pathway prior to secretion (22), both of which result in reduced secretion and increased intracellular retention of CLU proteins (22,29–33). Interestingly, trafficking and localisation of glycosylated CLU proteins are altered by AD related factors too, including Aβ treatment of cells in vitro (29,30) and CLU-AD mutations (34,35), indicating altered CLU trafficking contributes to AD pathogenesis.

Although often referred to as a cytoprotective protein, CLU has also been shown to play a pro-apoptotic role in the human breast cancer cell line MCF-7 (36,37). Alternative splicing of CLU generates a minor CLU variant lacking exon 2 (also referred to as CLU variant 1 Δ exon 2) (36–38). Translation of this variant generates a non-glycosylated, uncleaved CLU protein that resides intracellularly, typically in the cytoplasm and nuclei of cells (36–38). Although this shorter CLU isoform has low abundance in cells (37,38) or is not expressed at all (39), expression is upregulated in cancer cell lines by stress and following induction of the cell death cascade, suggesting a role for these proteins in mediating cellular response to stress (37,38). However, their role in human neurons and AD has not yet been explored.

In this study, we used CRISPR/Cas9 genome engineering and human iPSC technologies to examine the relationship between CLU proteins and Aβ induced toxicity in human neurons in vitro. We generated human neurons lacking glycosylated CLU proteins, which displayed altered sensitivity to Aβ. We also used RNA-Seq analysis to identify gene pathways that may underly this altered sensitivity and contribute to our understanding of AD pathogenesis and identification of novel AD therapeutics.

CRISPR/Cas9 gene editing to generate exon 2 -/- iPSC lines

CTR M3 36S (CTR) iPSCs were generated from human keratinocytes obtained from a healthy, neurotypical adult male, as previously described (40). Cells were reprogrammed by exogenous expression of OCT3/4, SOX2, cMYC and KLF4 using a Sendai virus vector (41). To generate exon 2 -/- iPSCs (A4 and D1 clones), a dual single guide RNA (sgRNA) approach was chosen to recruit Cas9-2NLS (Synthego) to cut within intron 1 and 2 to remove CLU exon 2 (Additional file 1, table S1) (42). Transfection of CTR iPSCs was performed using Nucleofection and an optimised protocol (Synthego). Briefly, ribonucleotide protein complexes (RNPs) were assembled by mixing sgRNA and Cas9 in a 7.5:1 ratio: 3µl (300pmol) sgRNA was added to 2µl (40pmol) Cas9 in a total volume of 5µl per well. CTR iPSCs were detached from wells using Accutase, and 500,000 cells were transferred to a falcon tube for transfection. Cells were centrifuged at 115g for 3 minutes and resuspended in 20µl Lonza P3 Nucleofector™ solution (Lonza). 20µl of cells were mixed with 5µl of pre-prepared RNP complex solution and the total volume (25µl) was transferred to a Nucleocuvette™ strip and placed into a 4D-X Core unit. Electroporation was performed using protocol ‘CA137’. Following transfection, cells were returned to a well of a 6-well plate into pre-warmed E8 containing 10µM Rock inhibitor. Cells were incubated overnight, and media was replaced daily until cells reached 70–80% confluence and quality control assays were performed.

PCR amplification was used to confirm the homozygous knockout (KO) of CLU exon 2 in A4 and D1 iPSCs. For each clonal iPSC line, genomic DNA was extracted using the QuickExtract™ DNA Extraction Solution (Lucigen, QE09050) and was amplified using the AmpliTaq Gold™ 360 Master Mix according to manufacturer’s instructions (ThermoFisher, 4398881) with 200nM forward and reverse primers (Additional file 2 table S2). PCR amplifications were performed using the following parameters: 95°C for 10 min, 95°C for 30 seconds, 55°C for 30 seconds, 72°C for 45 seconds and a final extension at 72°C for 7 minutes, a total of 39 cycles were used. PCR products were then used in agarose electrophoretic analysis by running on a 1% agarose gel and imaged using a BioRad Gel Doc (Additional file 3 supplementary Fig. 1). Following confirmation that A4 and D1 iPSCs were homozygous exon 2 KO clonal lines, quality control experiments were performed to check karyotype and pluripotency of the iPSC lines. Synthego provided karyotype data for A4 and D1 iPSCs, which were confirmed to be normal for both lines (Additional file 4 supplementary Fig. 2). For pluripotency analysis, A4 and D1 iPSCs were plated on 96-well plates, fixed, and stained with nuclear marker DAPI, and immunostaining for pluripotency markers OCT4 and SSEA4 to confirm the retention of pluripotency of A4 and D1 iPSCs following removal of CLU exon 2 (Additional file 5 figure S3).

Neuronal differentiation

Human iPSCs were maintained in Essential 8 media (Life Technologies) on Geltrex coated plates (Nunc) until 100% confluent, at which point cells were differentiated according to a protocol adapted from Shi and colleagues (43,44). iPSCs were cultured in 50% Neurobasal (Life Technologies) and 50% DMEM-F12 media (Life Technologies) supplemented with 1% N2 (Life Technologies), 2% B27 (Life Technologies), 1% Glutamax (Life Technologies), 10µM SB-431542 (ApexBIO) and 1µM LDN-193189 (Sigma) for 7 days with daily 100% media changes. Cells were dissociated with Accutase (Life Technologies) and replated on days 7, 12, 15 and 18 in 50% Neurobasal and 50% DMEM-F12 media supplemented with 1% N2 (Life Technologies), 2% B27 (Life Technologies) and 1% Glutamax (Life Technologies). On day 21, neural progenitor cells were replated on poly-ornithine and laminin coated plates at a density of 10,000 cells/well (96-well plate) and 200,000 cells/well (12-well plate) in B27 media (100% Neurobasal media supplemented with 2% B27 and 1% Glutamax). Media was changed after 24 hours to remove cell debris. A full media change was performed on day 25 when 0.1µM cytosine arabinoside (Sigma) was added to inhibit the growth of glial cells. A further full media change was performed 3 days later, and a half media change was performed another 3 days later. For all experiments, control and exon 2 -/- iPSCs were differentiated three independent times and neuronal identity was assessed by immunostaining (details of antibodies used can be found in additional file 6 table S3). The expression of neuronal marker MAP2 and cortical lineage marker CTIP2 was confirmed in all lines, and the numbers of MAP2 positive neurons and of CTIP2-positive MAP2 neurons were quantified to determine the percentage of neuronal cells in control and exon 2 -/- neuron cultures that were CTIP2 positive (Additional file 7 figure S4).

Aβ₂₅₋₃₅ oligomer preparation

Aβ₂₅₋₃₅ Trifluoroacetate salt (BACHEM AG, Bubendorf, Switzerland, #4030205) was supplied at 5mg with a purity of 97% (molecular weight 1060.28). Nuclease-free water was added to make up a solution of 2mg/ml and stored at -20°C. Prior to use, aliquots were defrosted on ice at room temperature (RT). For measuring Aβ toxicity, three concentrations of Aβ₂₅₋₃₅ (500nM, 2µM, 20µM) was added to cells for 24 hours prior to experimentation. For Western blotting and RNA-Seq, cells were treated for 24 hours with 20µM Aβ₂₅₋₃₅ only.

Western blotting

Prior to cell lysis, media samples were collected on ice and concentrated 45 times using Amicon Ultra-2ml 10K Centrifugal Filter Units (Sigma, Aldrich, UFC201024) according to the manufacturer’s instructions. Protein content in the concentrated media samples was then measured following the addition of 150µl Pierce 660nm™ reagent (ThermoFisher, 22662). Protein content was measured relative to samples of Bovine Serum Albumin of known concentrations on a CLARIOstar Plus Plate reader (BMG LABTECH). For total cell lysates, 4X 200µl of laemmli buffer (BioRad) supplemented with HALT™ inhibitor (ThermoFisher, 78429) was added directly to cells. Lysates were collected, vortexed and boiled at 95°C for 5 minutes at which point the supernatant was collected. For each sample, 4µg of concentrated media was prepared in DPBS (Life Technologies, 14190094) and equal volumes of cell lysate and laemmli buffer were separated by SDS-PAGE electrophoresis on 10% gels and transferred onto nitrocellulose membranes. The membranes were blocked with 5% BSA (made in 1X TBST, tris-buffered saline and 0.1% Tween-20) for 1 hour in agitation at RT. After blocking, membranes were incubated at 4°C overnight in agitation with goat anti-clusterin primary antibody in 5% BSA (Santa Cruz, SC-6420, diluted 1:500). Membranes were then washed three times with 1X TBST, incubated with secondary antibody for 2 hours at RT in agitation (goat anti-donkey, IRDye® 800 Li-COR Biosciences 926-32214, diluted 1:10,000 in 5% BSA). Membranes were again washed three times and imaged using an infrared scanner (LI-COR Biosystems). To confirm equal protein loading across samples, membranes were finally incubated with a rabbit anti-β-actin primary antibody (Cell Signalling Technology, D6A8, diluted in 1:1000 in 5% BSA) for 1 hour at RT in agitation. Membranes were washed and incubated for 1 hour at RT in agitation with a secondary antibody prior to imaging (donkey anti-rabbit, IRDye® 680, Li-COR Biosciences 926-68073, diluted 1:10,000 in 5% BSA).

Cell death assay

Cell death was quantified using the CytoTox-Glo™ Cytotoxicity Assay (Promega, G9291) according to manufacturer's instructions. For all treatment conditions, cell death was compared relative to neurons that were untreated for 24 hours. Following Aβ_25− 35 treatment (500nM, 2µM, 20µM) 50µl AAF-Glo™ was added to each well for 15 minutes. Luminescence as a result of apoptosis induced loss of membrane integrity was read on a PHERAstar FSX Microplate reader (BMG LABTECH) giving a measure of cell death induced by Aβ_25− 35 treatment. Then, 50µl lysis reagent containing digitonin and AF-Glo™ was added to wells for 15 minutes. The addition of these reagents was used to induce death in remaining live cells in each well, quantified as the total dead count. Luminescence was again recorded and was used as a measure of the total number of cells in each well. Dead/Total Dead Ratios were then calculated and compared to the ratios of untreated neuron cultures; an increase in this ratio following treatment with Aβ_25− 35 would reflect an increase in the number of dead cells in the culture indicating Aβ_25− 35 treatment had caused significant increases in apoptosis and cell death.

Neurite length assay

To quantify neurite damage following Aβ₂₅₋₃₅ treatment (500nM, 2µM, 20µM), neurons were fixed with 4% paraformaldehyde-DPBS at RT for 15 minutes. Cells were washed with DPBS three times prior to immunostaining. Cells were simultaneously permeabilised and blocked in DPBS containing 0.25% Triton-X and 10% donkey serum (Life Technologies) for 1 hour at RT. Chicken anti-MAP2 (Abcam, ab5392, 1:1000) primary antibodies were diluted in DPBS containing 0.25% Triton-X and 5% donkey serum. Cells were washed three times with TBST and incubated for 1 hour at RT in secondary antibody goat anti-chicken 647 diluted in blocking solution (Life Technologies, A-21449, 1:500). Cells were then washed once with Hoechst 33342 for 30 seconds (Thermo Fisher, 0.01µg/ml diluted in DPBS). Cells were then washed a further three times with DPBS prior to imaging on an Opera Phenix High Content Screening System (Perkin Elmer) under 20X magnification. Immunofluorescent images were analysed using the Harmony High-Content Analysis Software (Perkin Elmer) using a neurite length assay developed in the analysis interface. First, intact nuclei were identified following positive staining by Hoechst. Neurites protruding from cell bodies stained positive for MAP2 in the 647 channel were then traced and quantified.

RNA-Seq and analysis

For RNA-Seq experiments, neurons were treated with 20µM Aβ₂₅₋₃₅ for 24 hours. Total RNA was collected using the Qiagen RNeasy Mini PLUS Kit (Qiagen) according to the manufacturer’s instructions from untreated and Aβ₂₅₋₃₅ treated control and exon 2 -/- neurons from three independent differentiations. 100ng of RNA were used for library preparation and analysis (Wellcome Trust Centre for Human Genetics, University of Oxford). Only samples with an RNA integrity number greater than 7 were used for subsequent sequencing. TruSeq RNA Library Preparation Kit (Illumina) was used, and mRNA was enriched by polyA selection. The mRNA fraction was selected from total RNA and converted to cDNA. cDNA was then end83 repaired, A-tailed and adaptor ligated. Samples were sequenced at 150 paired ends on a NovaSeq6000 for PolyA- RNA-Seq. The quality control (QC) of the reads was conducted (FastQC version 0.11.5), low-quality adapters and reads were removed (Cutadapt version 1.13), and then aligned (BWA-MEM version 0.7.15) to the human reference genome GRCh37.EBVB95-8wt.ERCC. Reads that overlap multiple features were resolved and summarised to the gene-level using the `summarizeOverlaps` function of the (GenomicAlignments R package) (45), with the follow settings: i) mode = Union, ii) singleEnd = F, iii) ignore.strand = T, iv) fragments = T and features were summarised to the gene-level. This yielded an Ensembl gene ID by sample counting matrix of the class S4 (R-object). The latter counting matrix was modelled using the DESeq2 pipeline which performs i) size factors estimation, ii) dispersion estimation and iii) Gamma- Poisson Generalized Linear Model (GLM) fitting (46). We report differentially expressed (DE) genes based on log2 fold-change (abs(FC) > 1) and p-values (p < 0.05) adjusted with Benjamini-Hochberg false discovery rate (FDR). Over-representation analysis (ORA) of ontology and pathway terms performed with hypergeometric tests (clusterProfiler, PMID:22455463) using gene sets encompassing Gene Ontology (GO) subsets such as biological process (BP), molecular function (MF), and cellular component (CC). Also, incorporating biological pathways gene sets from KEGG, WikiPathways and Reactome. Topological signalling pathway enrichment performed in SPIA (47).

Statistical analysis

Statistical significance was set as p < 0.05 for all assays and experiments used in this study. Statistical tests were performed using GraphPad Prism (v. 7.02). A normal distribution in cell populations was assumed and two-tailed parametric tests were used. For experiments examining changes in CLU protein abundance in control neurons, analysis was performed using an unpaired t-test (Fig. 1). For comparisons of CLU protein abundance between untreated and Aβ treated control and exon 2 -/- neurons analysis was performed by One-way ANOVA with Tukey’s Multiple Comparisons post hoc test (Fig. 1). Additionally, One-way ANOVA with Tukey’s Multiple Comparisons post hoc test was used to analyse differences in CTIP2 expression in control, A4 and D1 neurons (Additional file 7 supplementary Fig. 4d-e). For comparisons of changes in cell death and neurite length between untreated and Aβ₂₅₋₃₅ treated control, A4 and D1 neurons, analysis was performed using a Two-way ANOVA with Tukey’s multiple comparisons post hoc test (Figs. 2 and 3). Data are shown as mean ± standard error of the mean (SEM) unless otherwise indicated.

Glycosylated CLU protein trafficking is altered by Aβ₂₅₋₃₅ in human neurons

Previous studies have demonstrated that a variety of stressors induce alterations in CLU trafficking, resulting in reduced secretion and increased intracellular retention (22,29,31,33). Therefore, our first aim was to determine if Aβ₂₅₋₃₅ treatment alters trafficking of glycosylated CLU in human neurons. Control (CTR) and exon 2 -/- neurons were treated with 20µM Aβ₂₅₋₃₅ for 24 hours, after which media and cell lysates were collected, ran on a western blot and immunostained for CLU. The abundance of CLU proteins was compared between treated and untreated neurons, as well as between exon 2 -/- and CTR neurons to examine the effect of both, Aβ₂₅₋₃₅ and the loss of CLU exon 2.

Exon 2 -/- neurons did not express glycosylated CLU proteins, confirming previous reports that CLU exon 2 is required for the generation of mature, cleaved and glycosylated CLU proteins (Fig. 1a and 1c), demonstrating that the ER signal peptide containing exon 2 is required to produce mature, glycosylated CLU. To examine if Aβ₂₅₋₃₅ altered glycosylated CLU trafficking, we compared levels of sCLU and iCLU expression between untreated and 20µM Aβ₂₅₋₃₅ treated CTR neurons and showed that intracellular CLU was significantly increased (Fig. 1a and c, p = 0.0221*) while secretion of CLU was significantly decreased (Fig. 1b and d, p = 0.0432*). This response was absent in exon 2 -/- neurons. This is the first demonstration that Aβ₂₅₋₃₅ exposure alters CLU trafficking in human neurons and supports previous studies that demonstrated CLU trafficking is altered in a variety of cell types/lines including rodent neurons, U251, SH-SY5Y, N2a and LNCaP cell lines following the induction of several stressors including Aβ₂₅₋₃₅ treatment and ER stress (induced by MG132 proteasome inhibitor) (22,29,31,33).

The translation of CLU variant 1 Δ exon 2 mRNA generates intracellular CLU proteins that are neither cleaved nor glycosylated, due to the absence of the ER-signal peptide (37,38,48). Abundance of these CLU proteins is increased following stress (37), and is thought to facilitate cell death (36,37). Therefore, we next aimed to examine whether Aβ₂₅₋₃₅ treatment induces an upregulation of these proteins in human neurons. First, abundance of non-glycosylated CLU proteins was compared between CTR and exon 2 -/- neurons prior to treatment with Aβ₂₅₋₃₅. Unsurprisingly, abundance of non-glycosylated CLU proteins was significantly higher in exon 2 -/- neurons than in CTR neurons following the removal of CLU exon 2 (Fig. 1a, e and f). Next, the effect of Aβ₂₅₋₃₅ treatment on non-glycosylated CLU abundance was examined. 20µM Aβ₂₅₋₃₅ did not alter the abundance of non-glycosylated CLU in either CTR or exon 2 -/- neurons (Fig. 1a, e and f). This contrasts with previous research that demonstrated that stress in cancer cell lines induces an upregulation of non-glycosylated CLU proteins. Our data show that stress induced by Aβ₂₅₋₃₅ treatment in human neurons has a specific effect on the trafficking of glycosylated CLU proteins, but not on the overall abundance of the non-glycosylated CLU proteins.

Exon 2 -/- neurons display less Aβ₂₅₋₃₅ induced cell death than CTR neurons

Having demonstrated that Aβ_25− 35 induces alterations in CLU trafficking in CTR neurons but does not alter abundance of non-glycosylated CLU, we next aimed to determine if this resulted in altered sensitivity of exon 2 -/- neuron to Aβ_25− 35. To do this, we first treated CTR and exon 2 -/- neurons with increasing concentrations of Aβ_25− 35 for 24 hours (500nM, 2µM or 20µM, respectively). After treatment, cell death was quantified by luminescence using the CytoTox-Glo™ Cytotoxicity Assay.

Previous studies have identified a potential role for non-glycosylated CLU in facilitating cellular response to stress and apoptosis (36,37). Additionally, overexpression of non-glycosylated CLU has been demonstrated to sensitise cells to stress (36). Therefore, we aimed to determine if the loss of CLU exon 2 and the increased abundance of non-glycosylated CLU proteins at basal conditions in exon 2 -/- neurons altered their sensitivity to Aβ₂₅₋₃₅ induced cell death. Dead/Total Dead Ratios were compared across all treatment conditions and between genotypes to determine if exon 2 -/- neurons were more sensitive to Aβ₂₅₋₃₅ induced cell death. In untreated conditions, Dead/Total Dead Ratios of exon 2 -/- neurons did not significantly differ to that of the CTR neurons, indicating that the proportion of dead neurons at basal conditions was not influenced by the genotype. No significant increase in ratios was observed at either 500nM or 2µM Aβ₂₅₋₃₅ treatment for either CTR or exon 2 -/- neurons, indicating that these concentrations were not sufficient to induce a significant increase in cell death. However, a significant increase in ratios, reflecting a significant increase in the number of dead cells, was observed using 20µM Aβ₂₅₋₃₅ treated neuron cultures. This was observed in both, CTR and exon 2 -/- neuron cultures. Interestingly higher amounts of cell death was quantified in CTR cultures treated with 20µM Aβ₂₅₋₃₅ compared to exon 2 -/- cultures, since the Dead/Total Dead ratios of the latter where significantly lower. This finding indicates significantly fewer exon 2 -/- neurons underwent cell death as a result of Aβ₂₅₋₃₅ treatment. These data demonstrate that the loss of CLU exon 2 and the increased abundance of non-glycosylated CLU proteins in human neurons reduces sensitivity to Aβ₂₅₋₃₅.

Exon 2 -/- neurons are protected against Aβ₂₅₋₃₅ induced neurite retraction

The observed reduction in Aβ₂₅₋₃₅ induced cell death in exon 2 -/- neurons suggests that the lack of glycosylated CLU proteins renders neurons less sensitive to Aβ₂₅₋₃₅ induced toxicity. To further explore this, we examined Aβ₂₅₋₃₅ induced neurite retraction by measuring changes in neurite length in CTR and exon 2 -/- neurons.

As before, CTR and exon 2 -/- neurons were treated with one of three concentrations of Aβ₂₅₋₃₅. After 24 hours, neurons were fixed, immunostained and imaged. Neurons that were positively stained for nuclear marker DAPI and neuronal marker MAP2 were imaged (Fig. 3a-c) and an automated pipeline was used to identify MAP2 positive neurite processes and quantify their length. Neurite length parameters were quantified and compared between genotypes and Aβ₂₅₋₃₅ treatment conditions (Fig. 3d-f).

Previously, we demonstrated that cell death prior to treatment was comparable between CTR and exon 2 -/- neurons, indicating that loss of CLU exon 2 did not impact neuronal death in the absence of an Aβ challenge. Accordingly, we aimed to determine if this was similar in the case of neurite length. Surprisingly, two neurite length parameters measured were significantly reduced in exon 2 -/- neurons prior to treatment with Aβ₂₅₋₃₅, suggesting that loss of CLU exon 2 may impact neurite length even at basal conditions.

Next, we aimed to determine what concentration, if any, is sufficient to induce neurite damage in human neurons resulting in significant neurite retraction. For all parameters, 20µM Aβ_25-35 consistently caused a significant reduction in neurite length in CTR neurons, but no significant changes in neurite length were identified in CTR neurons treated with either 500nM or 2µM Aβ_25-35. Finally, we explored whether the observed changes in neurite length in CTR neurons could also be detected in exon 2 -/- neurons. Interestingly, exon 2 -/- neurons did not demonstrate significant neurite retraction at any concentration of Aβ_25-35. Taken together, our data show that exon 2 -/- neurons are protected against both, Aβ_25-35 induced neurite retraction and Aβ_25-35 induced neuronal death.

CLU exon 2 removal downregulates extracellular matrix (ECM) pathways

Having identified differences in the responses of exon 2 -/- and CTR neurons to Aβ₂₅₋₃₅ treatment, we aimed to use RNA-Seq analysis to identify changes in gene expression that may contribute to the altered sensitivity of exon 2 -/- neurons to Aβ₂₅₋₃₅ induced toxicity. Differences in gene expression were compared between CTR and exon 2 -/- neurons and between untreated and Aβ₂₅₋₃₅ treated neurons giving a total of 7 comparisons (table 1 and additional file 8 table S4). UMAP analysis revealed two cell clusters corresponding to Aβ₂₅₋₃₅ treated or untreated neurons (additional file 9, figure S5). Differentially expressed genes in each comparison were grouped according to GO pathway enrichment to identify significantly altered pathways (Fig. 4a). Nine pathways were differentially activated between exon 2 -/- and CTR neurons in basal conditions prior to treatment with Aβ₂₅₋₃₅ (res 4 and 5). Genes implicated in extracellular matrix (ECM) organisation, extracellular structure organisation, homophilic cell adhesion via plasma membrane adhesion molecules, cell-cell adhesion via plasma membrane adhesion molecules and several chondrocyte morphogenesis pathway were all were downregulated in exon 2 -/- neurons. Several ECM implicated genes were identified in a heatmap displaying the top 20 most highly variable genes (Fig. 4b), including COL1A1, COL1A2, COL3A1 and PCDHGA3. These were demonstrated to be more highly expressed in CTR neurons than in A4 and D1 neurons, implicating a depression of ECM-related gene expression. Functional enrichment analysis was performed (additional file 10, table S5). Functional directionality analysis of KEGG pathway enrichment (Fig. 4c) identified focal adhesion and extracellular matrix receptor interaction pathways and pointed to a general downregulation of ECM in exon 2 -/- neurons. Importantly, this was unaltered by Aβ₂₅₋₃₅ in exon 2 -/- neurons. Topology analysis performed on KEGG pathway enrichment data supported these findings (Fig. 4d) and we could identify a suppression of ECM receptor interaction in exon 2 -/- neurons independent of treatment condition. Additionally, this analysis identified suppression of PI3K-Akt signalling in exon 2 -/- neurons, also independent of the treatment condition. These data suggest that loss of exon 2 and the subsequent reduction in CLU protein abundance in human neurons may result in reduced expression of genes encoding proteins that perform functions in the ECM and in the PI3K-Akt pathway. These findings indicate that the loss of CLU exon 2 and glycosylated CLU proteins is sufficient to reduce the sensitivity of human neurons in vitro to Aβ₂₅₋₃₅ induced toxicity and downregulate genes performing ECM functions.

Identification of gene expression changes induced by Aβ₂₅₋₃₅ treatment

Having identified transcriptomic changes arising in neurons following removal of CLU exon 2, we next used our RNA-Seq dataset to identify changes in gene expression arising in neurons following treatment with Aβ₂₅₋₃₅ that may be regulating neurite damage and cell death induced by Aβ₂₅₋₃₅. Using topology-based analysis of KEGG pathway changes, several pathways were found altered in both, CTR and exon 2 -/- neurons, including apoptosis (activated), MAPK signalling (activated), p53 signalling (activated), cholinergic synapse (activated), glutamatergic synapse (inhibited) and axon guidance (inhibited) (table 2 and additional file 11 table S5). Given these pathways were regulated by Aβ₂₅₋₃₅ in CTR and exon 2 -/- neurons, it is likely that these contribute to cell death observed in both lines. For example, it is reasonable to suggest the observed activation in p53 signalling, apoptosis and MAPK signalling are related to Aβ₂₅₋₃₅ induced cell death in both, exon 2 -/- and CTR neurons. However, it is not clear why cell death is induced by exon 2 -/- neurons but only to a lesser extent than in CTR neurons. Interestingly, although neurite retraction was absent in Aβ₂₅₋₃₅ treated exon 2 -/- neurons, like in CTR neurons, axon guidance and glutamatergic synapse pathways were inhibited while cholinergic synapses were activated. These observations would be predicted in CTR neurons displaying neurite damage but since neurite retraction is absent in exon 2 -/- neuron it remains unclear why these pathways appear altered. These two findings together suggest that the extent of the activation and inhibition of the pathways highlighted by our analysis is significantly reduced in exon 2 -/- neurons and induces less cell death and no retraction of neurite processes in exon 2 -/- neurons.

Finally, no pathway was identified as significantly altered in both A4 and D1 neurons that were not altered in CTR neurons following Aβ₂₅₋₃₅ treatment. This suggests that exon 2 -/- neurons’ altered sensitivity to Aβ₂₅₋₃₅ induced toxicity may arise due to an absence of pathway activation/inhibition observed in CTR neurons. As supporting evidence, three further pathways were altered in CTR neurons and not in exon 2 -/- neurons. Cell cycle, long term depression (LTD) and complement and coagulation cascades were all significantly activated in CTR neurons following Aβ₂₅₋₃₅ treatment. This could be an indication that one or more of these pathways plays a crucial role in mediating Aβ₂₅₋₃₅ induced neurite retraction, given that activation of these pathways was not observed in exon 2 -/- neurons in response to Aβ₂₅₋₃₅ treatment.

CLU has long been implicated in AD, however, its exact role in AD pathogenesis and disease development remains unclear. Our study demonstrates a central role for CLU proteins in facilitating Aβ₂₅₋₃₅ induced cell death and neurite damage in human neurons in vitro.

Previous studies have examined the cytoprotective role of secreted CLU and the pro-apoptotic role of non-glycosylated CLU proteins (21,23,37,49). However, these studies have focussed on the role of CLU proteins in cancer pathways in immortal cancer cell lines and few studies have directly manipulated the expression of non-glycosylated CLU proteins in human neurons to explore their importance in neurodegeneration. In contrast to previous work, we demonstrate that stress induced by Aβ₂₅₋₃₅ does not induce increased expression of non-glycosylated CLU in human neurons. Therefore, the previously identified pro-apoptotic role of non-glycosylated CLU proteins may not be universal and may be dependent on factors such as the type of cell and source of stress. Yang et al., demonstrated that overexpression of non-glycosylated CLU proteins in cancer cells sensitizes cells to stress and induces more death in basal conditions (36). This is not observed in human neurons where we have demonstrated that exon 2 -/- neurons displayed similar levels of cell death in basal conditions, indicating that increased expression of non-glycosylated CLU did not result in increased neuronal death. Exon 2 -/- iPSC-neurons displayed reduced sensitivity to Aβ₂₅₋₃₅ induced cell death and neurite damage, suggesting that it is the glycosylated version of CLU that facilitate Aβ₂₅₋₃₅ toxicity. We have therefore demonstrated that non-glycosylated CLU proteins likely do not play a critical role in mediating neurodegeneration in human neurons induced by Aβ₂₅₋₃₅.

Although typically considered a secreted protein, glycosylated CLU proteins have been shown to localise within cells (22,29,33), however, the mechanisms leading to intracellular retention of sCLU have not been described. Numerous studies have demonstrated that trafficking of glycosylated CLU proteins is altered by stress in a variety of cell types/lines (22,30–33,38), resulting in reduced secretion and increased intracellular retention of CLU. Additionally, CLU-AD mutations and Aβ₂₅₋₃₅ treatment of rodent neurons have been shown to alter CLU trafficking, providing a potential mechanistic role of CLU in mediating neurodegeneration (29,34).

To date, few studies have attempted to dissect whether stress induced alterations in CLU trafficking act to facilitate cell stress or act to protect cells from further damage. Gregory et al. demonstrated altered CLU trafficking by ER stress in N2a cells resulting in increased binding of intracellular CLU and TDP-43, which reduced both the aggregation and toxicity of TDP-43, thus suggesting altered CLU trafficking in this context provides a protection to cells (33). In comparison, previous work by our group found evidence that altered CLU trafficking facilitates Aβ₂₅₋₃₅ induced cell death in rodent neurons and that Aβ₂₅₋₃₅ increased intracellular CLU retention and CLU knockdown provided protection from Aβ₂₅₋₃₅ induced neurodegeneration (29). It is therefore likely that stress-induced CLU trafficking alterations and subsequent increases in intracellular CLU as well as reduced secretion can be both protective and detrimental to cells dependent on the type of stress and cell. Understanding the importance of glycosylated CLU proteins is particularly important since several studies have identified plasma CLU as a promising marker for AD (50); higher plasma CLU levels are associated with increased hippocampal atrophy and increased rate of clinical progression (51,52).

Our data clearly show that loss of glycosylated CLU protein in human neurons in vitro provides partial protection against Aβ₂₅₋₃₅ induced toxicity, which we postulate may arise from ECM remodelling (Fig. 5). Although cell death was still induced by Aβ₂₅₋₃₅ in exon 2 -/- neurons, it was significantly less than control neurons and neurite damage was absent in exon 2 -/- neurons. What remains unclear is whether the observed protection arises because of the loss of intracellular glycosylated CLU proteins or the loss of secreted glycosylated CLU proteins, or both. Although previous studies have suggested a pro-apoptotic role of non-glycosylated CLU proteins in cells (36,37), little is really known of their physiological and pathological functions in cells. We demonstrate an increase in the abundance of non-glycosylated CLU proteins in exon 2 -/- neurons, which is not altered by Aβ₂₅₋₃₅ treatment. It would be crucial to determine if protection observed in exon 2 -/- is conferred by the increased abundance of these CLU proteins.

It is therefore crucial to determine the roles played by intracellular and secreted CLU proteins in Aβ₂₅₋₃₅ induced toxicity to determine the exact roles played by CLU proteins in Aβ₂₅₋₃₅ induced cell death and neurite retraction and the underlying mechanisms. Several studies have shown that CLU proteins bind directly to Aβ peptides (17,30) to regulate Aβ uptake and clearance by astrocytes (30). This role is attributed to secreted glycosylated CLU proteins. We hypothesise that in control neurons in response to Aβ₂₅₋₃₅ treatment, secreted CLU is taken back into cells, potentially as a method of protecting cells. However, this role may facilitate toxicity if concentrations of Aβ₂₅₋₃₅ peptides is high in the cellular media. sCLU proteins may bind to Aβ₂₅₋₃₅ peptides and facilitate the uptake of CLU- Aβ₂₅₋₃₅ complexes into neurons, thereby promoting the entrance of Aβ₂₅₋₃₅ peptides into cells. The intracellular accumulation of Aβ₂₅₋₃₅ peptides can then induce damage to neurons. Thus, in this context CLU is facilitating Aβ₂₅₋₃₅ toxicity. Further interrogation of this relationship is required. Several receptors for CLU have been identified, including the megalin receptor (18,28), but this interaction is not well described and has not been analysed in iPSC-neurons. In exon 2 -/- neurons, the absence of secreted CLU may result in reduced Aβ₂₅₋₃₅ uptake into neurons, resulting in a lower concentration of Aβ₂₅₋₃₅ peptides accumulating in the neurons and thus, less cell damage. Since we described only partial protection in exon 2 -/- neurons, other mechanisms must also facilitate the uptake of Aβ₂₅₋₃₅ peptides and Aβ₂₅₋₃₅ toxicity in human neurons, in addition to CLU proteins.

Changes to the ECM were demonstrated in exon 2 -/- neurons and we postulate that ECM remodelling contributes to the partial protection observed in exon 2 -/- neurons to Aβ₂₅₋₃₅ providing evidence for targeting the ECM as a therapeutic option for AD and neurodegeneration. The ECM is composed of a range of diverse proteins that serve a multitude of functions (53). Those include regulation of cell death and survival (54), albeit this is not clearly described in neurons, synaptic function (55,56) and neurite outgrowth (57,58), all three of which are significantly altered in AD and by Aβ₂₅₋₃₅. ECM remodelling is associated with several diseases including AD (59–62). Numerous ECM proteins have been described as upregulated or downregulated in AD brains (62). For example, Collagen V and fibronectin are upregulated in the cerebral cortex of AD patients in early stages of disease progression and may arise from a reduction in proteolytic activity and may contribute to early changes in the AD brain (63). Several ECM proteins play a crucial role in regulating synaptic transmission and thus, changes to these proteins may result in a disruption in synaptic function and result in abnormal synaptic activity, and impaired learning and memory (64,65). ECM proteins have also been shown to alter Aβ toxicity. Heparan and chondroitin sulfate proteoglycans bind to Aβ peptides reducing Aβ₂₅₋₃₅ induced toxicity in hippocampal cultures and PC12 cells (66,67). Indeed, reducing the expression of chondroitin sulfate proteoglycans enhances Aβ₁₋₄₂ toxicity in neurons (68). Other ECM proteins have been shown to promote Aβ aggregation and plaque formation (69,70). Most recently, a large scale deep multi-layer analysis study of AD brains identified the ECM to strongly correlate with AD traits (71),

providing strength to our observations that the ECM may play a key role in mediating Aβ₂₅₋₃₅ induced toxicity in human neurons. More work needs to be done to fully understand the role of the ECM in AD since both, protective and faciliatory roles have been identified. In our study it is apparent that the ECM directly contributes to toxicity induced by Aβ₂₅₋₃₅ in vitro in human iPSC-neurons since ECM remodelling is associated with neuroprotective properties to iPSC-neurons. Given the varied functions of ECM proteins in neurons, it is essential to identify the key proteins that are central to the partial protection observed in exon 2 -/- neurons to determine if they can be utilised as a therapeutic target to provide neuroprotection.

Our data show that glycosylated CLU proteins facilitate neurotoxicity induced by Aβ₂₅₋₃₅ in human neurons. In fact, Aβ₂₅₋₃₅ induced neurite retraction is dependent on the expression of glycosylated CLU proteins and the absence of glycosylated CLU species in human neurons provides partial protection against Aβ₂₅₋₃₅. Previous research has identified the role of non-glycosylated CLU proteins in promoting cell death and increasing cell sensitivity to stress. However, these studies had been carried out in cancer cell lines and not neurons. Our work clearly shows the importance of cellular context and clearly demonstrates that, in neurons, non-glycosylated CLU does not promote neuronal degeneration but that increased expression results in a protection against neurite loss and neuronal death. We have identified several genes and pathways that potentially underlie this neuroprotection including those that may play a critical role in regulating cell death and neurite damage induced by Aβ₂₅₋₃₅. Our data will facilitate the identification of key components underlying Aβ₂₅₋₃₅ toxicity and identify potential therapeutic targets to prevent neurodegeneration in AD.

Aβ

Amyloid-β

Alzheimer’s disease

CLU

Clusterin

CRISPR

Clustered regularly interspaced short palindromic repeats

CTR

Control iPSC-neuron line

ECM

Extracellular matrix

Endoplasmic reticulum

Exon 2 -/-

iPSC-neuron line lacking CLU exon 2

fAD

Familial AD

iCLU

Intracellular clusterin

iPSC

Induced pluripotent stem cells

gRNA

Guide RNA

NFT

Neurofibrillary tau tangles

sAD

Sporadic AD

sCLU

Secreted clusterin

sAD

Sporadic AD

Declarations

Not applicable

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of supporting data

RNA-Seq data generated and/or analysed during this study are included in the article and as additional files. Any additional information required is available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Funding

This work was supported by AstraZeneca as part of a CASE studentship and the Valat-Jones Foundation (Nigel and Françoise Jones).

Authors' contributions

SL, NJB and EMR conceived the project. MP provided external funding and support for the project. PH and ADV provided technical assistance during the development and optimisation of the project. EMF carried out experiments, performed the data analysis and wrote the manuscript. RNA-Seq analysis was performed by MF. EMF and NJB edited the manuscript.

All authors read and approved the final manuscript.

Acknowledgements

We thank members of Synthego Corporation generating the CLU exon 2 knockout iPSC lines and their support in this research.

Authors' information (optional)

Not applicable.

National Institute on Aging. Alzheimer’s Disease Fact Sheet [Internet]. 2019 [cited 2021 Jan 2]. Available from: http://www.nia.nih.gov/health/alzheimers-disease-fact-sheet
Hardy JA, Higgins GA. Alzheimer’s disease: the amyloid cascade hypothesis. Science. 1992 Apr 10;256(5054):184–5.
Musiek ES, Holtzman DM. Three Dimensions of the Amyloid Hypothesis: Time, Space, and “Wingmen”. Nat Neurosci. 2015 Jun;18(6):800–6.
Selkoe DJ, Hardy J. The amyloid hypothesis of Alzheimer’s disease at 25 years. EMBO Mol Med. 2016 Jun;8(6):595–608.
Goate A, Chartier-Harlin M-C, Mullan M, Brown J, Crawford F, Fidani L, et al. Segregation of a missense mutation in the amyloid precursor protein gene with familial Alzheimer’s disease. Nature. 1991 Feb;349(6311):704–6.
Murrell J, Farlow M, Ghetti B, Benson M. A mutation in the amyloid precursor protein associated with hereditary Alzheimer’s disease. Science. 1991 Oct 4;254(5028):97–9.
Mullan M, Crawford F, Axelman K, Houlden H, Lilius L, Winblad B, et al. A pathogenic mutation for probable Alzheimer’s disease in the APP gene at the N–terminus of β–amyloid. Nat Genet. 1992 Aug;1(5):345–7.
Sherrington R, Rogaev EI, Liang Y, Rogaeva EA, Levesque G, Ikeda M, et al. Cloning of a gene bearing missense mutations in early-onset familial Alzheimer’s disease. Nature. 1995 Jun;375(6534):754–60.
Scheuner D, Eckman C, Jensen M, Song X, Citron M, Suzuki N, et al. Secreted amyloid β–protein similar to that in the senile plaques of Alzheimer’s disease is increased in vivo by the presenilin 1 and 2 and APP mutations linked to familial Alzheimer’s disease. Nat Med. 1996 Aug;2(8):864–70.
Bertram L. The genetic epidemiology of neurodegenerative disease. J Clin Invest. 2005 Jun 1;115(6):1449–57.
Cruts M, Theuns J, Van Broeckhoven C. Locus-Specific Mutation Databases for Neurodegenerative Brain Diseases. Hum Mutat. 2012 Sep;33(9):1340–4.
Tanzi RE, Bertram L. Twenty Years of the Alzheimer’s Disease Amyloid Hypothesis: A Genetic Perspective. Cell. 2005 Feb;120(4):545–55.
Lambert J-C, Heath S, Even G, Campion D, Sleegers K, Hiltunen M, et al. Genome-wide association study identifies variants at CLU and CR1 associated with Alzheimer’s disease. Nat Genet. 2009 Oct;41(10):1094–9.
Harold D, Abraham R, Hollingworth P, Sims R, Gerrish A, Hamshere ML, et al. Genome-wide association study identifies variants at CLU and PICALM associated with Alzheimer’s disease. Nat Genet. 2009 Oct;41(10):1088–93.
Lambert J-C, Genetic and Environmental Risk in Alzheimer’s Disease (GERAD), Alzheimer’s Disease Genetic Consortium (ADGC), Cohorts for Heart and Aging Research in Genomic Epidemiology (CHARGE), Ibrahim-Verbaas CA, Harold D, et al. Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer’s disease. Nat Genet. 2013 Dec;45(12):1452–8.
Bettens K, Brouwers N, Engelborghs S, Lambert J-C, Rogaeva E, Vandenberghe R, et al. Both common variations and rare non-synonymous substitutions and small insertion/deletions in CLU are associated with increased Alzheimer risk. Mol Neurodegener. 2012;7(1):3.
Wojtas AM, Kang SS, Olley BM, Gatherer M, Shinohara M, Lozano PA, et al. Loss of clusterin shifts amyloid deposition to the cerebrovasculature via disruption of perivascular drainage pathways. Proc Natl Acad Sci. 2017 Aug 15;114(33):E6962–71.
Zlokovic V. CEREBROVASCULAR TRANSPORT OF ALZHEIMER’S AMYLOIDP AND APOLIPOPROTEINS J AND E: POSSIBLE ANTI-AMYLOIDOGENIC ROLE OF THE BLOOD-BRAIN BARRIER. 1996;59(18):15.
DeMattos R, O’dell M, Parsadanian M, Taylor JW, Harmony J, Bales K, et al. Clusterin promotes amyloid plaque formation and is critical for neuritic toxicity in a mouse model of Alzheimer’s disease. Proc Natl Acad Sci. 2002 Aug 6;99(16):10843–8.
DeMattos R, Cirrito J, Parsadanian M, May P, O’Dell M, Taylor J, et al. ApoE and Clusterin Cooperatively Suppress A␤ Levels and Deposition: Evidence that ApoE Regulates Extracellular A␤ Metabolism In Vivo. 2004;
Zhang H, Kim JK, Edwards CA, Xu Z, Taichman R, Wang C-Y. Clusterin inhibits apoptosis by interacting with activated Bax. Nat Cell Biol. 2005 Sep;7(9):909–15.
Nizard P, Tetley S, Le Dréan Y, Watrin T, Le Goff P, Wilson MR, et al. Stress-Induced Retrotranslocation of Clusterin/ApoJ into the Cytosol. Traffic. 2007 May;8(5):554–65.
Trougakos IP, Lourda M, Antonelou MH, Kletsas D, Gorgoulis VG, Papassideri IS, et al. Intracellular Clusterin Inhibits Mitochondrial Apoptosis by Suppressing p53-Activating Stress Signals and Stabilizing the Cytosolic Ku70-Bax Protein Complex. Clin Cancer Res. 2009 Jan 1;15(1):48–59.
Choi-Miura N-H, Oda T. Relationship between multifunctional protein “clusterin” and Alzheimer disease. Neurobiol Aging. 1996 Sep 1;17(5):717–22.
Jenne DE, Tschopp J. Molecular structure and functional characterization of a human complement cytolysis inhibitor found in blood and seminal plasma: identity to sulfated glycoprotein 2, a constituent of rat testis fluid. Proc Natl Acad Sci U S A. 1989 Sep;86(18):7123–7.
Kirszbaum L, Bozas SE, Walker ID. SP-40,40, a protein involved in the control of the complement pathway, possesses a unique array of disulphide bridges. FEBS Lett. 1992;297(1–2):70–6.
Foster EM, Dangla-Valls A, Lovestone S, Ribe EM, Buckley NJ. Clusterin in Alzheimer’s Disease: Mechanisms, Genetics, and Lessons From Other Pathologies. Front Neurosci [Internet]. 2019 [cited 2019 May 10];13. Available from: https://www.frontiersin.org/articles/10.3389/fnins.2019.00164/full
Ghiso J, Matsubara E, Koudinov A, Frangione B. The cerebrospinal-fluid soluble form of Alzheimer’s amyloid beta is complexed to SP-40,40 (apolipoprotein J), an inhibitor of the complement membrane-attack complex. Biochem J. 1993;1(293):27–30.
Killick R, Ribe EM, Al-Shawi R, Malik B, Hooper C, Fernandes C, et al. Clusterin regulates β-amyloid toxicity via Dickkopf-1-driven induction of the wnt–PCP–JNK pathway. Mol Psychiatry. 2014 Jan;19(1):88–98.
Nuutinen T, Huuskonen J, Suuronen T, Ojala J, Miettinen R, Salminen A. Amyloid-β 1–42 induced endocytosis and clusterin/apoJ protein accumulation in cultured human astrocytes. Neurochem Int. 2007 Feb;50(3):540–7.
Li N, Zoubeidi A, Beraldi E, Gleave ME. GRP78 regulates clusterin stability, retrotranslocation and mitochondrial localization under ER stress in prostate cancer. Oncogene. 2013 Apr 11;32(15):1933–42.
O’Sullivan J, Whyte L, Drake J, Tenniswood M. Alterations in the post-translational modification and intracellular trafficking of clusterin in MCF-7 cells during apoptosis. Cell Death Differ. 2003 Aug;10(8):914–27.
Gregory JM, Whiten DR, Brown RA, Barros TP, Kumita JR, Yerbury JJ, et al. Clusterin protects neurons against intracellular proteotoxicity. Acta Neuropathol Commun. 2017 Dec;5(1):81.
Bettens K, Vermeulen S, Van Cauwenberghe C, Heeman B, Asselbergh B, Robberecht C, et al. Reduced secreted clusterin as a mechanism for Alzheimer-associated CLU mutations. Mol Neurodegener [Internet]. 2015 Jul 16 [cited 2019 May 10];10. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4502563/
Zhou Y, Hayashi I, Wong J, Tugusheva K, Renger JJ, Zerbinatti C. Intracellular Clusterin Interacts with Brain Isoforms of the Bridging Integrator 1 and with the Microtubule-Associated Protein Tau in Alzheimer’s Disease. PLOS ONE. 2014 Jul 22;9(7):e103187.
Yang C-R, Leskov K, Hosley-Eberlein K, Criswell T, Pink JJ, Kinsella TJ, et al. Nuclear clusterin/XIP8, an x-ray-induced Ku70-binding protein that signals cell death. Proc Natl Acad Sci. 2000 May 23;97(11):5907–12.
Leskov KS, Klokov DY, Li J, Kinsella TJ, Boothman DA. Synthesis and Functional Analyses of Nuclear Clusterin, a Cell Death Protein. J Biol Chem. 2003 Mar 28;278(13):11590–600.
Prochnow H, Gollan R, Rohne P, Hassemer M, Koch-Brandt C, Baiersdörfer M. Non-Secreted Clusterin Isoforms Are Translated in Rare Amounts from Distinct Human mRNA Variants and Do Not Affect Bax-Mediated Apoptosis or the NF-κB Signaling Pathway. Velasco G, editor. PLoS ONE. 2013 Sep 20;8(9):e75303.
Andersen CL, Schepeler T, Thorsen K, Birkenkamp-Demtröder K, Mansilla F, Aaltonen LA, et al. Clusterin expression in normal mucosa and colorectal cancer. Mol Cell Proteomics MCP. 2007 Jun;6(6):1039–48.
Cocks G, Curran S, Gami P, Uwanogho D, Jeffries AR, Kathuria A, et al. The utility of patient specific induced pluripotent stem cells for the modelling of Autistic Spectrum Disorders. Psychopharmacology (Berl). 2014 Mar;231(6):1079–88.
Nishimura K, Sano M, Ohtaka M, Furuta B, Umemura Y, Nakajima Y, et al. Development of Defective and Persistent Sendai Virus Vector. J Biol Chem. 2011 Feb;286(6):4760–71.
Blayney J, Foster EM, Jagielowicz M, Kreuzer M, Morotti M, Reglinski K, et al. Unexpectedly High Levels of Inverted Re-Insertions Using Paired sgRNAs for Genomic Deletions. Methods Protoc [Internet]. 2020 Jul 29 [cited 2021 Jan 6];3(3). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7565582/
Shi Y, Kirwan P, Livesey F. Directed differentiation of human pluripotent stem cells to cerebral cortex neurons and neural networks. Nat Protoc. 2012 Oct;7(10):1836–46.
Shi Y, Kirwan P, Smith J, Robinson H, Livesey F. Human cerebral cortex development from pluripotent stem cells to functional excitatory synapses. Nat Neurosci. 2012 Mar;15(3):477–86.
Lawrence M, Huber W, Pagès H, Aboyoun P, Carlson M, Gentleman R, et al. Software for Computing and Annotating Genomic Ranges. PLOS Comput Biol. 2013 Aug 8;9(8):e1003118.
Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014 Dec 5;15(12):550.
Tarca AL, Draghici S, Khatri P, Hassan SS, Mittal P, Kim J, et al. A novel signaling pathway impact analysis. Bioinformatics. 2009 Jan 1;25(1):75–82.
Reddy KB, Jin G, Karode MC, Harmony JAK, Howe PH. Transforming Growth Factor β (TGFβ)-Induced Nuclear Localization of Apolipoprotein J/Clusterin in Epithelial Cells ^†. Biochemistry. 1996 Jan;35(19):6157–63.
Trougakos IP, Lourda M, Agiostratidou G, Kletsas D, Gonos ES. Differential effects of clusterin/apolipoprotein J on cellular growth and survival. Free Radic Biol Med. 2005 Feb 15;38(4):436–49.
Liang H-C, Russell C, Mitra V, Chung R, Hye A, Bazenet C, et al. Glycosylation of Human Plasma Clusterin Yields a Novel Candidate Biomarker of Alzheimer’s Disease. J Proteome Res. 2015 Dec 4;14(12):5063–76.
Thambisetty M, Simmons A, Velayudhan L, Hye A, Campbell J, Zhang Y, et al. Association of plasma clusterin concentration with severity, pathology, and progression in Alzheimer disease. Arch Gen Psychiatry. 2010 Jul;67(7):739–48.
Thambisetty M, Simmons A, Hye A, Campbell J, Westman E, Zhang Y, et al. Plasma Biomarkers of Brain Atrophy in Alzheimer’s Disease. PLoS ONE [Internet]. 2011 Dec 21 [cited 2020 Dec 30];6(12). Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3244409/
Barros CS, Franco SJ, Muller U. Extracellular Matrix: Functions in the Nervous System. Cold Spring Harb Perspect Biol. 2011 Jan 1;3(1):a005108–a005108.
Makino H, Sugiyama H, Kashihara N. Apoptosis and extracellular matrix–cell interactions in kidney disease. Kidney Int. 2000 Sep;58:S67–75.
Ferrer-Ferrer M, Dityatev A. Shaping Synapses by the Neural Extracellular Matrix. Front Neuroanat. 2018 May 15;12:40.
Dityatev A, Schachner M. The extracellular matrix and synapses. Cell Tissue Res. 2006 Nov 1;326(2):647–54.
Letourneau PC, Condic ML, Snow DM. Extracellular matrix and neurite outgrowth. Curr Opin Genet Dev. 1992 Jan 1;2(4):625–34.
Srinivasan P, Zervantonakis IK, Kothapalli CR. Synergistic Effects of 3D ECM and Chemogradients on Neurite Outgrowth and Guidance: A Simple Modeling and Microfluidic Framework. Weaver AM, editor. PLoS ONE. 2014 Jun 10;9(6):e99640.
Bonneh-Barkay D, Wiley CA. Brain Extracellular Matrix in Neurodegeneration. Brain Pathol. 2009 Oct;19(4):573–85.
Morawski M, Brückner G, Jäger C, Seeger G, Matthews RT, Arendt T. Involvement of Perineuronal and Perisynaptic Extracellular Matrix in Alzheimer’s Disease Neuropathology: Perineuronal and Perisynaptic Matrix in AD. Brain Pathol. 2012 Jul;22(4):547–61.
Soria FN, Paviolo C, Doudnikoff E, Arotcarena M-L, Lee A, Danné N, et al. Synucleinopathy alters nanoscale organization and diffusion in the brain extracellular space through hyaluronan remodeling. Nat Commun. 2020 Dec;11(1):3440.
Sun Y, Xu S, Jiang M, Liu X, Yang L, Bai Z, et al. Role of the Extracellular Matrix in Alzheimer’s Disease. Front Aging Neurosci. 2021;13:554.
Lepelletier F-X, Mann DMA, Robinson AC, Pinteaux E, Boutin H. Early changes in extracellular matrix in Alzheimer’s disease. Neuropathol Appl Neurobiol. 2017;43(2):167–82.
Lindahl B, Eriksson L, Spillmann D, Caterson B, Lindahl U. Selective loss of cerebral keratan sulfate in Alzheimer’s disease. J Biol Chem. 1996 Jul 19;271(29):16991–4.
Buckley K, Kelly RB. Identification of a transmembrane glycoprotein specific for secretory vesicles of neural and endocrine cells. J Cell Biol. 1985 Apr;100(4):1284–94.
Woods AG, Cribbs DH, Whittemore ER, Cotman CW. Heparan sulfate and chondroitin sulfate glycosaminoglycan attenuate β-amyloid(25–35) induced neurodegeneration in cultured hippocampal neurons. Brain Res. 1995 Oct 30;697(1):53–62.
Pollack SJ, Sadler IIJ, Hawtin SR, Tailor VJ, Shearman MS. Sulfonated dyes attenuate the toxic effects of β-amyloid in a structure-specific fashion. Neurosci Lett. 1995 Sep 15;197(3):211–4.
Miyata S, Nishimura Y, Nakashima T. Perineuronal nets protect against amyloid β-protein neurotoxicity in cultured cortical neurons. Brain Res. 2007 May 30;1150:200–6.
Gupta-Bansal R, Frederickson RC, Brunden KR. Proteoglycan-mediated inhibition of A beta proteolysis. A potential cause of senile plaque accumulation. J Biol Chem. 1995 Aug 4;270(31):18666–71.
Bruinsma IB, te Riet L, Gevers T, ten Dam GB, van Kuppevelt TH, David G, et al. Sulfation of heparan sulfate associated with amyloid-beta plaques in patients with Alzheimer’s disease. Acta Neuropathol (Berl). 2010 Feb;119(2):211–20.
Johnson ECB, Carter EK, Dammer EB, Duong DM, Gerasimov ES, Liu Y, et al. Large-scale deep multi-layer analysis of Alzheimer’s disease brain reveals strong proteomic disease-related changes not observed at the RNA level. Nat Neurosci. 2022 Feb;25(2):213–25.

Tables 1 to 2 are available in the Supplementary Files section.

Table1.xlsx
TABLE 1: A list of RNA-Seq comparisons. Differences in gene expression were analysed between CTR, A4 and D1 neurons, both in untreated and Aβ_25-35treated neurons. The effect of Aβ_25-35treatment on gene expression was examined in res 1-3 comparisons, while the effect of CLU exon 2 KO was examined in res 4-7.
Table2.xlsx
TABLE 2: A list of pathways altered by Aβ_25-35in human neurons. Pathways significantly altered (activated/suppressed) by Aβ_25-35treatment in CTR and exon 2 -/- neurons, or CTR neurons alone.
Additionalfile1TableS1.docx
Additional file 1.: Table S1: Guide RNA sequences used in this study.
Additionalfile2TableS2.docx
Additional file 2.: Table S2: List of PCR primers used in this study to confirm exon 2 -/- genotype.
Additionalfile3FigureS1.docx
Additional file 3. : Figure S1. Agarose gel electrophoresis to confirm removal of CLU exon 2 from human iPSCs. (a) Primers positioned within introns 1 and 2 were initially used to confirm deletion of exon 2 from A4 and D1, indicating by the presence of a PCR product of smaller size in both these lanes. (b) Primers positioned within exon 2 were then used to confirm the removal of exon 2 from both alleles of A4 and D1 iPSCs.
Additionalfile4FigureS2.docx
Additional file 4. : Figure S2. Karyotyping of (a) A4 and (b) D1 iPSCs. No chromosomal abnormalities were identified in either line. A4 (46, X, Y) and D1 (46, X, Y). A standard metaphase karyotype (band resolution 475-525) was performed and identified no missing or additional chromosomes in either line.
Additionalfile5FigureS3.docx
Additional file 5.: Figure S3. Immunostaining of A4 and D1 iPSCs to confirm expression of pluripotency markers OCT4 (red) and SSEA4 (green). Expression was confirmed for both markers in A4 and D1 iPSCs and was confirmed to be nuclear (DAPI, blue) due to co-localisation with DAPI (purple).
Additionalfile6TableS3.docx
Additional file 6.: Table S3. List of immunocytochemistry antibodies used in this study.
Additionalfile7FigureS4.docx
Additional file 7.: Figure S4. Immunostaining of (a) CTR, (b) A4 and (c) D1 iPSC-neurons to confirm neuronal identity (MAP2) and cortical lineage (CTIP2). (d) The number of MAP2 positive neurons was similar between genotypes (p=0.1474; CTR vs. A4: p=0.3227, CTR vs. D1: p=0.1396, A4 vs. D1: p=0.7873). (e) The percentage of CTIP2 positive MAP2 neurons was significantly influenced by neuron genotype (p=0.0387*), however, no significant differences were revealed following multiple comparisons testing (CTR vs. A4: 0.0699, CTR vs. D1: p=0.0656 and A4 vs. D1: p=0.9957). The percentage of CTIP2 positive MAP2 neurons for all cultures were: CTR: 42.5 ± 5.6%, A4: 19.8 ± 5.2%, and D1: 20.5 ± 5.2%). Data are shown as mean ± SEM. Mean values were calculated from neuronal samples collected from three independent differentiations of neurons, of which each contained triplicate samples (n=3). Significance was tested using One-Way ANOVA with Tukey’s Multiple comparisons analysis P<0.05*.
Additonalfile8TableS4.xlsx
Additional file 8: : Table S4. Differentially expressed genes (DGEs) for all pairwise comparisons.
Additionalfile9FigureS5.docx
Additional file 9:: Figure S5. Dimensionality reduction of the normalized counts across samples. Principal component analysis (PCA) accounting for 75% (PC1-PC3) of the cumulative variance of uniform manifold approximation and projection (UMAP).
Additionalfile10TableS5.xlsx
Additional file 10: : Table S5. Enrichment analysis for all pairwise comparisons.
Additionalfile11FigureS6.docx
Additional file 11:: Figure S6. Topology analysis of KEGG pathway enrichment was performed to identify pathways activated or suppressed by Aβ_25-35treatment in CTR and exon 2 -/- neurons (res 1, 2 and 3). (a) Several relevant pathways were activated by Aβ_25-35treatment including apoptosis, p53, MAPK signalling, cholinergic synapses in neurons of both genotypes. (b) Fewer relevant pathways were inhibited by Aβ_25-35treatment including glutamatergic synapses and axon guidance in neurons of both genotypes. (c) Three pathways altered by Aβ_25-35treatment only in CTR neurons were identified: cell cycle (activated), long term depression (inhibited) and complement and coagulation regulation (activated). Comparisons: res1: untreated CTR vs Aβ_25-35treated CTR iPSC-neurons, res2: untreated A4 vs Aβ_25-35treated A4 iPSC-neurons and res3: untreated D1 vs Aβ_25-35treated D1 iPSC-neurons. tA: the observed total perturbation accumulation in the pathway, pPERT: probability to observe a total accumulation more extreme than tA only by chance. pPERT*<0.05, **<0.01 and ***<0.001.
ctranda4CLUmediarawfile.tif
ctranda4lysatesCLUrawfile.tif
ctranda4lysatesbetaactinrawfile.tif
ctrandd1CLUmediarawfile.tif
ctrandd1lysatesCLUrawfile.tif
ctrandd1lysatesbetaactinrawfile.tif

Download PDF

Version 1

posted

You are reading this latest preprint version

Glycosylated clusterin species facilitate Aβ toxicity in human neurons

Status:

Version 1

Abstract

Figures

Background

Materials And Methods

CRISPR/Cas9 gene editing to generate exon 2 -/- iPSC lines

Neuronal differentiation

Aβ₂₅₋₃₅ oligomer preparation

Western blotting

Cell death assay

Neurite length assay

RNA-Seq and analysis

Statistical analysis

Results

Glycosylated CLU protein trafficking is altered by Aβ₂₅₋₃₅ in human neurons

Exon 2 -/- neurons display less Aβ₂₅₋₃₅ induced cell death than CTR neurons

Exon 2 -/- neurons are protected against Aβ₂₅₋₃₅ induced neurite retraction

Identification of gene expression changes induced by Aβ₂₅₋₃₅ treatment

Discussion

Conclusions

Abbreviations

Declarations

References

Tables

Supplementary Files

Status:

Version 1

Glycosylated clusterin species facilitate Aβ toxicity in human neurons

Status:

Version 1

Abstract

Figures

Background

Materials And Methods

CRISPR/Cas9 gene editing to generate exon 2 -/- iPSC lines

Neuronal differentiation

Aβ25−35 oligomer preparation

Western blotting

Cell death assay

Neurite length assay

RNA-Seq and analysis

Statistical analysis

Results

Glycosylated CLU protein trafficking is altered by Aβ25−35 in human neurons

Exon 2 -/- neurons display less Aβ25−35 induced cell death than CTR neurons

Exon 2 -/- neurons are protected against Aβ25−35 induced neurite retraction

Identification of gene expression changes induced by Aβ25−35 treatment

Discussion

Conclusions

Abbreviations

Declarations

References

Tables

Supplementary Files

Status:

Version 1

Aβ₂₅₋₃₅ oligomer preparation

Glycosylated CLU protein trafficking is altered by Aβ₂₅₋₃₅ in human neurons

Exon 2 -/- neurons display less Aβ₂₅₋₃₅ induced cell death than CTR neurons

Exon 2 -/- neurons are protected against Aβ₂₅₋₃₅ induced neurite retraction

Identification of gene expression changes induced by Aβ₂₅₋₃₅ treatment