Optineurin deficiency disrupts phosphorylated tau proteostasis and clusterin expression in human neurons

Abstract

Optineurin (OPTN) is an autophagy adaptor protein involved in selective autophagy, including aggrephagy and mitophagy. Pathogenic mutations in OPTN have also been linked to amyotrophic lateral sclerosis, frontotemporal dementia, and glaucoma, supporting its role in the etiology of neurodegenerative diseases. Despite its established biological roles, knowledge about its potential contribution to Alzheimer's disease (AD) pathology and neuronal functioning is lacking. AD is characterized by the accumulation of extracellular amyloid-β plaques and intracellular phosphorylated tau (pTau) tangles, with dysfunction in the autophagy-lysosomal pathway exacerbating tau pathology and impairing proteostasis. To investigate the role of OPTN in neuronal proteostasis and AD, we utilized induced pluripotent stem cell-derived neuron (iN) and astrocyte (iA) models. Analyses revealed a significant negative correlation between OPTN and specific pTau epitopes in neurons, as well as a decrease in OPTN protein abundance in brain tissues of individuals with AD. Given these findings, we generated OPTN knockout (KO), heterozygous, and wildtype iNs and iAs using CRISPR/Cas9 editing of iPSCs in two genetic backgrounds. Loss of OPTN in iNs increased specific pTau proteoforms without substantially affecting autophagy processes or mitochondrial respiration. Despite no clear effect on mitochondrial function, several mitochondrial proteins, including OXCT1, were enriched in an unbiased analysis of the OPTN interactome in iNs, as well as proteins involved in intracellular trafficking. Proteomic analyses further identified intracellular clusterin, an AD risk gene, as significantly upregulated in OPTN KO iNs, suggesting OPTN may influence its intracellular processing. Our model system demonstrates modest roles for OPTN in certain neuronal biological processes and potential implications for AD pathogenesis. These findings also suggest that OPTN may exhibit functional redundancy with other autophagy adaptor proteins in human neurons, leading to relatively mild phenotypic changes with complete loss of OPTN.

Keywords: Alzheimer’s disease; Autophagy; Clusterin; Optineurin; Tau.

Fig. 1

OPTN is downregulated in AD brain and significantly correlates to AD-relevant pTau abundance within neurons. A Protein from 30 ROSMAP human brains was sequentially extracted using TBS, SDS, and urea. Each brain extract was run on a western blot (WB) and probed for OPTN. OPTN abundance was normalized to total protein using REVERT staining. A representative WB of 5 non-cognitively impaired (NCI) and 2 AD brains is shown. B, C OPTN abundance was quantified from the WB across SDS and urea sequential extracts and compared between AD (n = 10) and NCI individuals (n = 20). OPTN abundance was normalized to total protein using REVERT staining. D–F Quantification of high molecular weight (HMW; ~250 kDa) phosphorylated-tau levels normalized to total pTau (HMW + Major) in iNs derived from ROSMAP NCI (n = 35) and AD (n = 18) individuals. The tau measurements for this data as well as sample WBs were published previously [21]. G OPTN protein abundance across NCI (n = 35) and AD (n = 18) ROSMAP iN lines. Data derived from a larger TMT-MS dataset published previously [21]. H–J Sample Spearman correlations of OPTN protein abundance (derived from G) and phosphorylated HMW pTau proteoforms (derived from D–F) is shown with AD lines in red and NCI lines in blue (n = 53 ROSMAP neurons). For B–G, statistical analysis was performed using an unpaired t-test and error bars reflect the standard error of the mean (SEM). For all comparisons: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns = not significant

Fig. 2

Loss of OPTN modestly increases specific phosphorylated tau species independent of an overall change in LC3B-II accumulation and lysosomal degradation. A–C OPTN WT, HET, and KO iNs were produced using CRISPR/Cas9 technology targeting exon 8 of OPTN in both BR24 and BR33 ROSMAP iPSCs. Representative western blot (WB), and immunocytochemistry images staining for DAPI (blue) and beta-III-tubulin (red) are shown as well as the quantification of OPTN knockdown (n = 4 independent wells; unpaired one-way ANOVA with Sidak’s post-hoc testing). D–F The ratio of aggregated phosphorylated tau to total phosphorylated tau (HMW pTau/(HMW + Major pTau)) was calculated after probing for relevant tau forms, pTau^202/205 and pTau²¹⁴. Representative WBs are shown (n = 6 differentiations across 2 genetic backgrounds with 4 technical replicates per differentiation; normalized to WT levels within each experiment. Each dot corresponds to a single differentiation with technical replicates averaged). G, H OPTN WT, HET, and KO iN LC3B-II accumulation was measured after 24 h of 40 µM chloroquine treatment, as assayed by WB in chloroquine treated samples compared to vehicle (n = 8 differentiations across 2 genetic backgrounds with 4 technical replicates per differentiation. Each dot corresponds to a single differentiation with technical replicates averaged). I Live cell imaging of average DQ Red BSA signal was measured in OPTN WT and KO iNs and normalized to iN cell area (n = 4 differentiations across 2 genetic backgrounds with 3–6 technical replicates per differentiation. Each dot corresponds to a single differentiation with technical replicates averaged). J, K Phagophore/autophagosome associated LC3B-II was quantified via WB in OPTN WT and KO iNs following chloroquine treatment for 24 h immediately followed by 10 min exposure to 0.3% saponin. Protein abundance was normalized to beta-III-tubulin (Tuj1; n = 4 differentiations across 2 genetic backgrounds with 3 technical replicates per differentiation. Each dot corresponds to a single differentiation with technical replicates averaged). For D–H, statistics were performed using paired one-way ANOVA analyses, with Sidak’s post-hoc testing. For I, J, paired t-tests were used to determine significance. For all comparisons: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001, ns = not significant

Fig. 3

Loss of OPTN causes modest proteome-wide changes, but significantly alters intracellular mature clusterin (CLU). A, B Volcano plots of OPTN HET versus WT and OPTN KO versus WT, respectively, show differentially expressed proteins as measured by TMT-MS (red dots; q < 0.05). Data shown includes BR24 and BR33 combined, with line ID regressed (n = 3 wells per genetic background). C Gene (RNAseq) and protein (TMT-MS) level Pearson correlations with OPTN were plotted and each dot represents data for a single gene (n = 53 ROSMAP iN lines). Proteins in the upper right and lower left quadrant with cutoff r =  ±0.4 were used for additional analyses. Dots labeled blue are from the 11 DEPs that overlap between both the HET vs WT and KO versus WT proteomic analysis (A, B). Dots labeled green are AD GWAS hits [4]. Dots labeled purple are from the 31 genes identified from the pipeline used in D. Dots labeled red are prioritized genes that overlapped across multiple categories and included CLU and ANK3. D Within the TMT-MS for OPTN KO versus WT, 11,302 proteins were identified in both genetic backgrounds. A selection criterion was conceived to winnow out relevant protein changes. First, proteins with nominal p value < .001 in both BR24 and BR33 were selected. Of these 345 proteins identified, we then selected those proteins that had a corresponding protein and RNA correlation to OPTN of r > 0.4 or r < −0.4 (identified in C). Both protein and RNA had to make the r value cutoff and have the same directionality to be considered. With this pipeline, we identified 31 proteins of interest that were up or downregulated in OPTN KO iNs. E The 31 proteins that passed our selection criterion laid out in D are displayed using their z-score in this heatmap (nominal p < 0.001). F A sample spearman correlation between OPTN and CLU protein abundance is shown (n = 53). Data derived from a TMT-MS dataset published previously [21]. G Mature intracellular CLU protein was assayed via western blot (WB) from RIPA OPTN WT and KO iN lysates and normalized to total GAPDH protein abundance (n = 6 differentiations across 2 genetic backgrounds with 4 technical replicates per differentiation). H Representative WB of mature intracellular CLU is shown. I Quantification of secreted CLU in 48 h iN conditioned media, as measured by the MSD U-PLEX Human CLU Assay, corrected for total protein REVERT in iN lysates (n = 6 differentiations across two genetic backgrounds; 3–4 technical replicates per differentiation). Each dot corresponds to a single differentiation with technical replicates averaged. Statistics were performed using a paired t-test. For all comparisons: **p < 0.01, ns = not significant

Fig. 4

The neuronal OPTN interactome identifies several mitochondrial and microtubule proteins. A BR24 and BR33 WT and KO neurons were harvested in 0.1% CHAPSO lysis buffer and subjected to OPTN antibody immunoprecipitation (IP) using Protein G Dynabeads. OPTN was largely cleared from the lysate following immunoprecipitation (~72 kDa). Samples were then sent for DIA- mass spectrometry. Volcano plot displaying proteins differentially enriched in WT IP elution (q < 0.05) (n = 2 genetic backgrounds, with 3 replicates each). B Representative WB displaying the immunoprecipitation of OPTN across BR24 and BR33 WT and KO iNs. C The top 20 most significant OPTN-interacting proteins by nominal p-value are shown, and mitochondrial proteins are highlighted in red. D Results of gene ontology (GO) analysis performed on the significant OPTN-interacting proteins derived from the IP-MS.

Fig. 5

OPTN loss does not alter mitochondrial morphology or function in human iNs. A Confocal microscopy images of OPTN (WT and KO) iNs. Right: 3D rendering of mitochondrial morphology from OPTN WT and KO iNs after immunostaining for mitochondria (ATPB). Scale bar = 10 µm (left), 5 µm (center). B, C Quantification of individual mitochondrial objects (volume and sphericity). (n = 4 differentiations across 2 genetic backgrounds with 2 technical replicates per differentiation, 6 micrographs per technical replicate). D MitoTracker Deep Red (300 nM for 45 min) was used with Hoescht to quantify relative mitochondrial content in OPTN WT and KO iNs. MitoTracker intensity was normalized to number of Hoescht positive nuclei (n = 6 differentiations across 2 genetic backgrounds with at least 4 technical replicates per differentiation). E Quantifications of S-OPA1 as a percentage of total OPA1 (normalized to WT) are shown (n = 2 differentiations across 2 genetic background and 3–4 technical replicates per differentiation). F Representative  WB of long (L-OPA1) and short (S-OPA1) isoforms in OPTN WT and KO iNs. G Oxygen consumption rate (OCR) was assayed using the Seahorse Mito Stress Test and normalized to total protein. H, I Basal and maximal OCR were graphed across OPTN WT, HET, and KO iNs (n = 2 differentiations per genetic background with 6–12 technical replicates per differentiation. Each dot corresponds to a single differentiation with technical replicates averaged). Statistics were performed using a paired t-test (Fig. 3B–D), unpaired t-test (Fig. 3E) or paired one-way ANOVA (Fig. 3H, I) and data is displayed using standard error of the mean. For all comparisons: ns = not significant