Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 May 1.
Published in final edited form as: Exp Cell Res. 2025 Apr 4;448(1):114540. doi: 10.1016/j.yexcr.2025.114540

Increased Oxidative Stress and Autophagy in NGLY1 Patient iPSC-derived Neural Stem Cells

Zeenat A Shyr 1,*, Soukaina Amniouel 1, Kofi Owusu-Ansah 1, Mitali Tambe 1, Joshua Abbott 1, Matthew Might 2, Wei Zheng 1,*
PMCID: PMC12070446  NIHMSID: NIHMS2073199  PMID: 40189184

Abstract

NGLY1 (N-glycanase) is a de-glycosylating enzyme that promotes clearance of misfolded glycan proteins. NGLY1 deficiency leads to a disease pathology with varied symptoms, including severe neurological defects. There are no therapeutic options currently available for the treatment of this rare disease. With the goal of finding potential therapeutic avenues, we performed comprehensive characterization of aberrant cellular stress pathways in a patient relevant model of NGLY1 deficiency. For a more accurate study of NGLY1 deficiency without other confounding factors, we compared differences between iPSC-derived neural stem cells carrying the commonly occurring nonsense mutation c.1201A>T (p.R401X) and their genetically similar CRISPR-corrected isogenic controls. Our findings demonstrate that NGLY1 deficiency in neural stem cells leads to an upregulation of ER stress, increased autophagic flux and significant signs of oxidative stress. These results provide new insights into the cellular dysfunctions associated with this disorder. Moreover, they point to better establishing reliable high throughput phenotypic assays that can be utilized for drug discovery.

Keywords: NGLY1, neural stem cells, cellular stress, autophagy

Graphical Abstract

graphic file with name nihms-2073199-f0001.jpg

INTRODUCTION

NGLY1 (N-glycanase) deficiency is an ultra-rare genetic disorder with fewer than 100 known cases worldwide that is caused by mutations in the NGLY1 gene. The NGLY1 enzyme de-glycosylates N-glycoproteins, which contributes to a non-lysosomal mediated clearance of misfolded glycans through the endoplasmic reticulum-associated degradation pathway (ERAD) [1, 2]. NGLY1 deficiency was first described through exome sequencing on patients with undiagnosed disorders with phenotypes reminiscent of congenital disorders of glycosylation (CDG) [3]. Some typical symptoms of the disease include global developmental delay, hypolacrima/alacrima, movement disorders and hypotonia [4, 5]. There are several known pathogenic variants in NGLY1 with both homozygous and heterozygous mutations [6] but the nonsense mutation c.1201A>T (p.R401X) is most common [5]. Thus far, it is the only known congenital disease of de-glycosylation (CDDG), while there are multiple cases of CDG.

The presence of ER stress has been described in several CDGs [7] [8]. This is not surprising, since any disruptions in ER homeostasis, such as impairment in protein glycosylation, can lead to ER stress [9]. There has been much speculation about CDDG, and whether the loss of NGLY1 could also lead to ER stress due to accumulation of misfolded proteins. In response to ER stress, an adaptive response called the unfolded protein response (UPR) is initiated that promotes ER efficiency by upregulating molecular chaperons to increase protein folding capacity, attenuating protein translation, and upregulating ERAD and autophagy pathways [10]. The UPR transduction pathways are mediated through three transmembrane proteins - inositol-requiring enzyme 1α (IRE1α), protein kinase RNA-like endoplasmic reticulum kinase (PERK) and activating transcription factor 6 (ATF6). Under normal conditions, these proteins are bound to the chaperone protein, heat shock protein family A (Hsp70) member 5 (BiP/HSPA5) but during ER stress, BiP associates with misfolded or unfolded proteins and dissociates from IREα, PERK, or ATF6, which activates the UPR pathway [11]. Activation of IRE1α leads to the splicing of Xbp1 mRNA, producing the active transcription factor, spliced XBP1 (sXBP1), which in turn upregulates genes in the UPR and ERAD pathways [12]. The PERK-EIFα pathway reduces global protein translation and induces the production of the transcription factor ATF4, which upregulates genes for amino acid metabolism, oxidative stress resistance, autophagy, and with prolonged ER stress, pro-apoptotic factors such as DNA damage inducible transcript 3 (CHOP or DDI3T) [11]. With ER stress, ATF6 translocates from the ER to the Golgi, where it is cleaved from its full-length form to its transcriptionally active form that further allows for the activation of genes involved in the UPR [11].

Despite the assumption that NGLY1 deficiency would lead to ER stress, data from studies in different NGLY1disease models have yielded conflicting results. Some in vivo as well as pluripotent stem cell (iPSC)-derived cell models have shown little to no evidence of overt ER stress [13–15]. In contrast, others have observed increase in components of the UPR pathway and/or ER stress in the absence of NGLY1 in human immortalized cell lines [16], mouse embryonic fibroblasts [17], and iPSC-derived neurons [18]. These contradictory results underscore the need for better models to accurately characterize the prevalence of ER stress as well as associated cellular stresses that may culminate with NGLY1 deficiency.

To address this gap, we used patient-derived iPSCs carrying the common nonsense mutation c.1201A>T (p.R401X) and their CRISPR-corrected isogenic controls to generate neural stem cells (NSCs) for all assessments. Neurological defects are a significant feature of NGLY1 disease, and we chose to perform all screens in NSCs, which have inherent similarities to differentiated mature neurons, are available in large quantities, and are amenable to high-throughput screening (HTS) [19]. With this approach, we could assess potential disease mechanisms underlying NGLY1 deficiency in a patient-relevant model that minimizes confounding genetic differences that otherwise lead to difficulties in data interpretation. Therefore, the novelty of our study lies in the extensive characterization of cellular stresses in a patient relevant model where NGLY1 deficiency is directly compared to their genetically similar NGLY1 replete cells. By performing omics analysis and phenotypic assays, we demonstrate that NGLY1 deficiency in p.R401X NSCs leads to an upregulation in components of UPR, increased autophagic flux, as well as distinct signs of oxidative stress and lipid peroxidation. These observations help further our understanding of potential disease mechanisms and enable us to find suitable and reliable phenotypic assays that can be utilized for drug discovery. Currently, there are no treatments available for this rare disease, and HTS assays can provide avenues to identify potential therapeutic targets in a cost and time effective manner.

METHODS

NSC induction and culture

Patient-derived iPSCs and their corresponding isogenic controls were generated as previously described [20]. Cells were plated in 6 well Matrigel (Corning)-coated plates at 30,000 cells per well in StemFlex media (Gibco, Catalog # A3349401) and grown at 37°C, 5% CO2. To form NSCs, iPSCs were grown in neural induction medium (Gibco, Cat#A1647801) for 7-days. Neural induction media was replaced every 24 hours. NSC differentiation was characterized by evaluating NESTIN, SOX2, SOX1, and PAX6 using the Human Neural Stem Cell Immunocytochemistry Kit (Life Technologies, Cat#A24354). Immunofluorescence staining was performed according to the manufacturer’s instructions and images were obtained with the Opera Phenix Plus High-Content Screening System (Revvity, Waltham, MA). For subsequent passages, NSCs were maintained and propagated in StemPro NSC SFM media (Gibco, Catalog# A1050901) supplemented with 10% Glutamax (Gibco, Cat# 35050061) on Geltrex (ThermoFisher Scientific, Cat#A1569601)-coated plates.

RNAseq and Proteomics protocol and analysis

Total RNA isolation of NSCs and RNA-seq as well as total protein extraction and proteomics were carried out as previously described [21]. All secondary analyses were performed in the R statistical environment (R version 4.4.1).

Read counts of RNA-seq were normalized across all samples using the default normalization method in the DeSeq2 package in R (v1.46.0) [22]. Principal component analysis (PCA), a widely employed technique in RNAseq studies, was carried out to find outlying samples. Gene expression changes were evaluated through linear modeling with DeSeq2 in R, setting the isogenic group as the reference point for comparison. A false discovery rate (FDR) or p-adjusted <0.1 with the Benjamini-Hochberg method was used for further analysis.

The proteomics dataset was preprocessed using the DEP package in R (v1.28.0). This package offers a robust framework to identify differentially expressed proteins (DEPs) from quantitative proteomics experiments. The workflow included filtering proteins with insufficient valid intensity measurements, followed by quantile normalization and log2 transformations to ensure comparability across samples. Statistical analysis was conducted using the limma package (v3.62.1), using the empirical Bayes moderated t-test to detect significant changes in protein expression between experimental conditions. The differential data with logFC of |1| and adjusted p-value < 0.1 were considered significant and used for further analysis.

Gene Set Enrichment Analysis (GSEA) was performed using ClusterProfiler (v4.14.4) to identify the biological process (BP). A rnk file was built using the ranked list of DEGs obtained with DESeq2 or limma analysis as input data. These categories were determined to be of interest with a Benjamini–Hochberg false-discovery rate below 10%

Ingenuity Pathway Analysis (IPA) software (QIAGEN, Redwood 185 City, CA) was employed for functional analysis of the identified DEGs and DEPs between the two groups. Canonical pathway analysis, which was used to predict the downstream effects of the DEGs and DEPs based on the observed gene expression changes, was also on the strength of the canonical information embedded in the IPA database. The calculated Z-score was used to infer the activation states (“increased” or “decreased”) of implicated biological processes. Fisher’s exact test was used to calculate a p-value to determine the probability that the association between genes or proteins in the datasets, and the biological process could be explained by chance alone.

Quantitative PCR analysis

RNA was extracted using the RNeasy Plus Mini Kit (Qiagen) and was converted to cDNA by utilizing the QuantiTect Reverse Transcription Kit (Qiagen). FAM-MGB TaqMan Probes (Applied Biosystems) specific to genes of interest were used to perform qPCR analysis with ViiA 7 Real-Time PCR System (Applied Biosystems). Ten ng of cDNA was used for each condition. The relative expression of all genes was normalized to GAPDH and calculated using the 2−ΔΔCt method.

Western blot analysis

NSCs were detached from the plate surface using Accutase (Stemcell Technologies, Cat#07920), washed with PBS and lysed in RIPA Buffer (Pierce, Cat#89900) supplemented with Protease Inhibitor Cocktail (Roche, #50–100-3301). Total protein was quantified by using Pierce BCA assay (ThermoFisher Scientific). Protein lysates were prepared with Laemmli sample buffer (Bio-Rad) and 10μg of total protein lysate was loaded per lane. Protein samples were resolved by 4–20% TGX gel (Bio-Rad) and transferred to nitrocellulose membrane by the trans-blot transfer system (Bio-Rad). Blots were incubated overnight with the following antibodies: β-actin (1:2000; CST #3700S), NGLY1 (1:1000; Sigma-Aldrich #HPA036825), ATF4 (1:1000, CSTD4B8), ATF6 (1:1000, CSTD428V), BiP (1:1000; CST #3177S), phospho-Eifα (1:1000; CST #3398S), ERO1α (1:1000, CST#3264T), phosphor-IRE1α (1:1000; Invitrogen #PA1–16927), LC3B (1:1000; CST #3868S), p62 (1:1000, CSTD5L76), phospho-PERK (1:1000; CST #3179S), XBP1 (1:1000; CST #12782S). Blots were washed and probed with IRDye secondary antibody conjugates (1:10,000; LI-COR Biosciences). Fluorescence intensity was quantified by Empiria Studio software (LI-COR Biosciences).

Oxidative stress/ lipid peroxidation assays

8000 to 10,000 NSCs were cultured in geltrex-coated 96-wells overnight before performing any assays. For oxidative stress assays, cells were incubated with DMSO or Menadione at 37°C for one hour. A 5uM final volume of cellROX Deep Red Reagent (Thermo Scientific, #C10422) and Hoechst 33342 dye (Thermo Scientific) was added to cells and incubated for another 30 minutes. Cells were washed with pre-warmed Live Cell Imaging Solution (Invitrogen Cat# A14291DJ) and fluorescence was measured on the Pherastar Microplate Reader (BMG Labtech) using a 635 filter.

For lipid peroxidation aassy, the Image-iT™ Lipid Peroxidation Kit (Invitrogen, Cat#C10445) was used. DMSO or cumene hydroperoxide (as a positive control) was added to cells at a final concentration of 25 μM at 37°C for 2 hours. In the last 30 minutes of incubation, lipid peroxidation sensor as well as Hoechst 33342 dye (Thermo Scientific) was added at 10 μM concentration to all wells. Media was replaced with pre-warmed Live Cell Imaging Solution (Invitrogen Cat# A14291DJ) and the plate was imaged immediately with the Opera Phenix Plus High-Content Screening System (Revvity Waltham, MA) using 590 nm and 510 nm emission filters. For quantification of lipid peroxidation, a ratio of 590 and 510 channels readings were calculated.

Immunofluorescence assays

5000–8000 NSCs were plated on geltrex-coated clear bottom 96 well plates. After 24 hours, cells were fixed in 4% PFA, permeabilized with 0.1% Triton-X (Sigma) and blocked in Cell Staining Buffer (Biolegend) before incubating overnight with primary antibodies: 4-Hydroxynonenal (1:50; ThermoFisher Scientific #MA5–27570), LAMP1 (1:100; CST #15665), LC3B antibody (1:500, CST #3868S), Cleaved Caspase 3, CC3 (1:1000; CST #9661S) at 4°C. For CC3, cells were treated with 2uM Staurosporine (Selleck chemicals, Cat#S1421) for 90 minutes prior to fixation. After overnight incubation, cells were washed with PBS before incubating with Alexa Fluor 488 or Alexa Fluor 594 secondary antibody (1:1000; ThermoFisher Scientific) for one hour at room temperature. Cells were rinsed with PBS and incubated with 4μM Hoechst 33342 (ThermoFisher Scientific) for 30 minutes at room temperature. Wells were washed with PBS and sealed before imaging with the Opera Phenix Plus High-Content Screening System (Revvity, Waltham, MA) or the ImageXpress (Molecular Devices, San Jose, CA)

Tandem-LC3B assay

6000 NSCs were plated on geltrex coated-clear bottom 96 well plates for 24 hours before viral transductions. Tandem RFP-GFP-LC3B sensors containing BacMam 2.0 chimeric reagent (Primo™ Autophagy Tandem Sensor RFP-GFP-LC3B Kit, Thermo Fisher Scientific) was added to cells at different MOIs along with 5μg/mL polybrene (Sigma-Aldrich, Cat# TR-1003). The plate was sealed with parafilm and spun at 800rcf at 37° C for 60 mins. The existing media was aspirated and replaced with fresh media before placing the plate in the incubator (37° C, 85% RH, 5% CO2). After further 48 hours, cells were fixed with 4% PFA for 20 minutes, washed with PBS before further incubating with Hoechst (1:5000, Thermo Fisher Scientific) for 30 minutes at room temperature. The cells were washed twice with PBS before imaging. Cells were imaged with 10x air or 20x water objective with ImageXpress (Molecular Devices).

Overexpression of WT-NGLY in p.R401X NSC lines

NSCs were plated at 400,000 cells/well in a 6-well plate. After overnight culture, fresh media containing 5ug/mL polybrene (Sigma) was added to all wells. Subsequently, lentivirus expressing wild type NGLY1 gene (pcDH-CMV- NGLY1-EF1alpha-GreenPuro, Systems Biosciences and Azenta) was added to the appropriate wells at MOI 40. The plates were parafilm sealed, followed by spinfection (800rcf at 37°for one hour). Existing media was aspirated, and fresh media was added. Plates were incubated at 37°/5% CO2. Cell pellets were collected after 72 hours.

Image analysis

Images obtained were transferred to Columbus Image Data Storage and Analysis System (Revvity) for batch analysis. Cells were first identified using nuclei staining, followed by identification of proteins in the region of interest. For LC3B tandem assay, cells were first identified using nuclei staining, followed by total RFP spots. GFP spots were calculated by restricting counts to only those that were colocalized with RFP spots.

Statistical analysis

All data analysis was performed using GraphPad Prism version 9 for Windows, GraphPad Software, Boston, Massachusetts USA, www.graphpad.com. Data are expressed as mean +/− standard error of mean (SEM). For comparisons between NGLY1-p.R401X and corrected lines, statistical significance was determined by Student’s t test. For group analysis, either a one-way ANOVA or two-way ANOVA with Tukey’s post hoc test was used.

RESULTS

Characterization of NGLY1-p.R401X and isogenic control neural stem cells

We have previously described the generation of an iPSC line from human fibroblasts of a patient with a homozygous nonsense mutation (R401X) in the NGLY1 gene that leads to a severe form of NGLY1 deficiency [20]. Here we differentiated the iPSCs carrying the R401X mutation as well as their homozygous CRISPR-corrected lines to neural stem cells (NSCs), which were characterized by commonly known NSC-specific markers such as Nestin, PAX6, SOX1, and SOX2 (Figure 1A). NSCs are multipotent cells that are dedicated to form cell types of the central nervous system that can be produced in large quantities with high reproducibility and have been used in drug discovery for high throughput screening (HTS). We then utilized the NSCs to perform all screening experiments to elucidate disease phenotypes of NGLY1 deficiency. As expected, we found that the NGLY1 gene (Figure 1B) and protein expression (Figure 1C) in the NGLY1-p.R401X cells were significantly lower compared to isogenic control cells. These results confirmed the expression of NGLY1 in relatively normal isogenic control NSCs and the dramatic reduction in NGLY1-p.R401X cells.

Figure 1. Generation of NGLY1-R401X and isogenic control NSCs.

Figure 1.

Open in a new tab

(A) Representative immunofluorescent images for markers specific for Neural Stem cells (NSCs) including SOX1, SOX2, Pax6 and Nestin. NGLY1-p.R401X and homozygous corrected NSCs were analyzed by (B) Quantitative PCR and (C) Western blot analysis. Left, representative blot showing NGLY1 protein detected at 74 kDa and β-actin used as housekeeping protein. Right, quantification; n=4–6 biological replicates each. **P < 0.01, ****P < 0.0001 by students t-test.

NGLY1 deficient NSCs results in transcriptional and proteomics changes

RNA-seq analysis and quantitative proteomic profiling were performed to comprehensively explore the differences between NGLY1-p.R401X and their isogenic corrected cells Normalized counts for all replicates of the neural stem cell lines were analyzed by PCA and hierarchical clustering (Figures 2A, 3A, 3B). For both RNAseq and proteomics data, NGLY-1 R401X and isogenic controls were clustered separately. Genes and proteins were determined to be differentially expressed with a cutoff of ≥ 1- fold change at FDR 5%. Deseq2 identified 5793 genes that were significantly differentially expressed between the two groups (Figure 2B). Limma, on the other hand, identified 427 significantly differentially expressed proteins (Figure 3C). Importantly, NGLY1 was found to be significantly downregulated for the disease group in both RNAseq and proteomics data.

Figure 2. RNA-seq data analysis as evaluated by.

Figure 2.

Open in a new tab

(A) Principal Component Analysis (PCA) which illustrates the separation between NGLY1-p. R401X and Isogenic-control groups, each with three replicates for each condition (B) Enhanced Volcano plot demonstrating significantly differentially expressed genes based on fold-change of |1| and adjusted p-values of less or equal 0.1. Colors indicate the significance: grey represents non-significant genes, blue represents genes with significant p-values but not fold change, green indicates significant fold change but not p-value, and red represents gene significant for both p-value and fold change. (C) Gene Set Enrichment Analysis (GSEA) plot showing results of Gene Ontology (GO) category: Biological Process (BP). (D) Ingenuity Pathway Analysis (IPA) bubble plot which visualizes canonical pathways, annotated with log(p-value), gene ratio, and z-score. The color of bubbles indicates the predicted z-score: red suggests an increase in activity, while blue suggests a decrease. The size of the bubbles corresponds to the gene ratio for each pathway.

Figure 3. Proteomics data analysis as evaluated by.

Figure 3.

Open in a new tab

(A) Sample correlation heatmap which demonstrates pairwise correlation across all samples, with red indicating higher correlation and blue indicating lower correlation. (B) Principal Component Analysis (PCA) illustrating the separation between NGLY1-p. R401X and Isogenic-control groups, each with three replicates for each condition. (C) Enhanced Volcano plot demonstrates significantly differentially expressed proteins based on fold-change of |1| and adjusted p-values of less or equal 0.1. Colors indicate the significance: grey represents non-significant proteins, blue represents proteins with significant p-values but not fold change, and red represents proteins significant for both p-value and fold change. (D) Gene Set Enrichment Analysis (GSEA) plot showing results across the Gene Ontology (GO) - Biological Process (BP). (E) Ingenuity Pathway Analysis (IPA) bubble plot visualizing canonical pathways. The color of bubbles indicates the predicted z-score: red suggests an increase in activity, while blue suggests a decrease. The size of the bubbles corresponds to the gene ratio for each pathway.

Gene ontology (GO) analysis of the upregulated genes from RNAseq data demonstrated a strong enrichment for genes involved in digestion (GO:0007586), cholesterol homeostasis (GO:0042632), ribonucleoprotein complex biogenesis (GO:0022613), ribosome biogenesis (GO:0042254), and negative regulation of blood coagulation (GO:0030195). GO analysis of the downregulated genes showed enrichment for genes involved in ribonucleoprotein complex biogenesis (GO:0022613), ribosome biogenesis (GO:0042254), aerobic respiration (GO:0009060), oxidative phosphorylation (GO:0006119), and proton motive force-driven ATP synthesis (GO:0015986) (FIG 2C). Although RNAseq provides a detailed view of transcriptional activity, it does not account for the functional state of proteins, which is shaped by post-transcriptional modifications, degradation, and interactions. To complement these findings, we conducted GO analysis of proteomics data, focusing on enriched proteins in the same conditions. Proteomics GO analysis revealed significant enrichment in processes such as myofibril assembly (GO:0030239), sarcomere organization (GO:0045214), cellular component assembly involved into morphogenesis (GO:0010927), postsynaptic cytoskeleton organization (GO:0099188), and heart morphogenesis (GO:0003007) (Figure 3D)

While several studies have linked NGLY1 deficiency to the aforementioned biological processes [14, 18, 23], our research was focused mostly on specific pathways such as ERAD, UPR, cellular stress, and autophagy, guided by the hypothesis that proteostasis-related functions play a critical role in the pathogenesis of NGLY1 deficiency in humans [5, 24–26]. To put our RNAseq and proteomics data into biological context and understand if any of these pathways are perturbed in NGLY-1 R401X group, the DEGs and DEPs were uploaded into ingenuity pathways analysis (IPA) software. Genes identified through RNA-seq and proteins from proteomics analysis were mapped to gene objects in the IPA knowledge Base and analyzed to identify canonical pathways. Ingenuity Pathway Analysis classifies canonical pathways based on Z-score and log(p-value). Among the thousands of identified canonical pathways, 274 from RNA-seq data and 536 from proteomics data were statistically significant according to both Z-score and log (p-value). The canonical pathways such as ERAD, autophagy, oxidative stress, unfolded protein response (UPR), protein folding, and ferroptosis were identified to be upregulated in either RNA-seq (Figure 2D) or proteomics (Figure 3E) data, if not both. Most of these canonical pathways were statistically significant according to both Z-score and log(p-value), except unfolding protein response.

ER Stress- Related Canonical Marker Validation through qPCR and Western Blot Analysis

We performed qPCR analysis for specific markers for ER stress and compared NGLY1-p.R401X NSCs to their isogenic control (Figure 4A). We found that the genes in the UPR pathway such as sXBP1 and EIF2s1 were significantly increased in NGLY1-p.R401X cells. There was no change in PERK, ATF6 or BiP but expression of IRE1 was decreased in NGLY1-p.R401X. It is well known that if UPR does not restore ER homeostasis, then chronic ER stress can lead to an increase in cell death [9], and both the IRE1α and PERK pathways can promote cell death by activating genes such as CHOP [12]. Both CHOP and BiP that are generally elevated in response to ER stress, were either significantly decreased or not changed in NGLY1-p.R401X cells. To better understand whether apoptosis was increased in NGLY1-p.R401X compared to their isogenic controls, cleaved caspase 3 levels (CC3) were quantified in the NSCs by immunofluorescence analysis (Figure S1). At baseline, there was no difference between NGLY1-p.R401X cells and their isogenic controls suggesting that prolonged ER stress-induced apoptosis was not present in NGLY1-p.R401X cells. However, NGLY1-p.R401X cells showed significantly increased CC3 intensity and number when exposed to staurosporine for a short duration compared to their isogenic controls (Figure S1). This suggested that NGLY1-p.R401X cells may be more susceptible to cellular insults compared to their isogenic controls.

Figure 4. Canonical ER stress markers as assessed by.

Figure 4.

Open in a new tab

(A) Gene expression by qPCR, n=6 biological replicates each. (B) Western blot analysis for proteins levels. Left, representative blot with related quantification; n=6–8 biological replicates each. Samples were run on the same blot but not on contiguous lanes. *P < 0.05, **P < 0.01, ****P < 0.0001 by students t-test.

Because mRNA levels do not always reflect changes in protein expression, we performed western blot analysis on NGLY1-p.R401X and isogenic controls for the canonical markers of UPR that get upregulated in response to ER stress (Figure 4B). Select markers of the UPR pathway such as ATF4, phospho-EIFα, phospho-PERK and s-XBP1 were significantly elevated in NGLY1-p.R401X cells. Other markers such as ATF6 (whole and cleaved), BiP and phospho-IRE1α were unchanged. CHOP could not be detected in either cell line by several antibodies tested. The results revealed that for EIFα and s-XBP1, both mRNA as well as protein levels were increased in NGLY1-p.R401X cells. However, for some components of the UPR pathway such as ATF4 and phospho-PERK, protein levels were elevated without changes in their mRNA expression. The apparent discrepancies between mRNA and protein expression could stem from differences in many processes such as translational efficiencies, post-translational modifications, other protein-protein interactions and protein degradation. Indeed, in cells that are not under steady state conditions, a strong correlation between mRNA and protein expression is usually attenuated [27].

Lipid peroxidation is increased in NGLY1 deficient NSCs

Protein disulfide isomerase family A member 2 (PDIA2/PDI) and endoplasmic reticulum oxidoreductase 1 alpha (ERO1α) are ER resident chaperone proteins that also get upregulated during ER stress conditions [28]. We found that while ERO1α was unchanged, PDI was significantly decreased in NGLY1-p.R401X cells (Figure 4B). The decrease in PDI may not support elevation in ER stress but suggests that protein folding capacity within the ER may be decreased. Since these proteins are also involved in the maintenance of oxidative homeostasis in the cell, this was suggestive of increased signs of oxidative stress in NGLY1-p.R401X cells.

Some of the main causative factors leading to oxidative stress is excess amounts of reactive oxygen species (ROS), which includes superoxide (O2•−), hydrogen peroxide (H2O2) and hydroxyl radical (•OH) [29]. We found that with a fluorescent assay using CellRox dye (Invitrogen), there was no difference in basal or menadione-induced superoxide ROS levels between NGLY1-p.R401X and control cells (Figure S2A). Quantifying oxidative stress can be difficult, and lipid peroxidation may be a more sensitive and useful method to assess oxidative stress, particularly in neurodegenerative disease models [29, 30]. We found that there was a significant increase in lipid peroxidation in NGLY1-p.R401X cells, as assessed by a commercially available kit that specifically detects peroxidation of lipids in live cells (Figure 5A). Oxidative stress-induced increase in lipid peroxidation in NGLY1-p.R401X cells was very apparent, both at baseline as well as when challenged with a hyperoxide compound. These results indicated that NGLY1-p.R401X NSCs may also be more susceptible to cellular oxidative insults compared to isogenic controls. Further evidence for increased lipid peroxidation levels in NGLY1-p.401X cells was found by staining for 4-Hydroxynonenal (4HNE), which is a well-known stable by-product of lipid peroxidation. 4HNE levels were significantly elevated in NGLY1-p.401X cells, as assessed by immunofluorescence analysis (Figure 5B).

Figure 5. Lipid peroxidation measurement.

Figure 5.

Open in a new tab

(A) Left, representative images of lipid peroxidation measurement in live cells at baseline and treated with 25μM Cumene Hyperoxide. Right, quantification of 590/510 (Gold/green) ratio from 18 well replicates and 25 fields in each well. NGLY1-p.R401X showed increased in lipid peroxidation, as assessed by decrease in 590/510 ratio (B) Left, representative images of NSCs stained for the lipid peroxidation marker, 4-HNE (green). Cells were delineated by cellmask (deep red) and Hoechst (blue) stains. Right, quantification of mean fluorescence intensity of 4-HNE per cell from n=4 well replicates and 15 fields in each well. ****P < 0.0001 by two-way ANOVA with Tukey’s post hoc test or **P < 0.01 by students t-test.

Flux assessment shows significant increase in autophagy in NGLY1 deficient NSCs

RNAseq and proteomics analysis also revealed significant differences in the autophagy pathway. Since some components of the UPR pathway were also elevated, it is possible that these elevations could be geared towards using autophagy as a process for protein clearance. Specifically, an elevation in PERK-EIFα-ATF4 pathway as well as sXBP1 is known to lead to an increase in autophagy [12, 31]. The gold standard for measuring autophagy is by assessing the state of microtubule associated protein 1 light chain 3 beta (MAP1LC3B/LC3B). When autophagy is activated, the cytosolic protein LC3BI is incorporated into autophagosomes, lipidated and a concomitant increase in LC3II level is observed. In addition, sequestosome 1 (SQSTM1/p62) is a cargo protein that is expected to decrease when the autophagosomes fuse with lysosomes. We found that there was a significant increase in LC3II and a significant decrease in p62 protein levels in NGLY1-p.R401X cells compared to their isogenic controls (Figure 6A). While our data suggested that autophagy is likely enhanced in NGLY1-p.R401X cells, an increase in LC3BII is not sufficient to definitively establish that. This is because, paradoxically, a significant increase in LC3II could also indicate a block in autophagy due to accumulation of autophagosomes. To delineate whether autophagy is enhanced or blocked, an autophagy flux assay was performed next. Chloroquine (CQ) blocks the fusion of autophagosomes with lysosomes, and a significant increase in both LC3BII as well as p62 should be observed if autophagy is enhanced. There was a significant increase in both LC3BII as well as p62 over respective basal levels in both NGLY1-p.R401X cells and controls (Figure 6B), confirming that autophagy is likely increased in NGLY1-p.R401X cells. Furthermore, the lysosome specific protein LAMP1 showed an increase in perinuclear and nuclear overlapping localization in the NGLY1-p.R401X cells (Figure 6C), which is supportive of enhanced autophagy [32]. We also tested the tandem LC3-RFP-GFP dual reporter in NGLY1-p.R401X and control cells. We found that viral transduction efficiency in NGLY11-p.R401X cells was significantly lower than controls at 25, 50 or 100 MOI in three independent experiments (Figure S3). This is exemplified by significantly lower mean red-fluorescent spots (RFP) in the NGLY1-p.R401X cells. However, the acid-sensitive green-fluorescent spots (GFP) were calculated to only include those that were colocalized with RFP. Therefore, although transduction efficiency was low in NGLY1-p.R401X cells, by restricting the GFP spots to RFP areas only, the ratio of GFP to RFP could be calculated. The ratio of RFP to GFP spots was significantly increased in NGLY1-p.R401X cells, indicating that LC3BII was localized in the lysosomes. Taken together, we report here that in NSCs, NGLY1 deficiency leads to an enhancement in autophagy induction.

Figure 6. Autophagy assessment.

Figure 6.

Open in a new tab

(A) Western blot analysis for LC3BII and p62; Left, representative blot and right, quantification; n= 6 biological replicates. (B) Protein levels of p62 and LC3BII in the presence or absence of chloroquine, CQ to assess autophagy flux; left, representative image and right, quantification; n=6 per condition. **P < 0.01 by two-way ANOVA with Tukey’s post hoc test. (C) Staining for lysosome specific protein, LAMP1 (green), LC3B (red) and nuclei (blue). Left, representative image and right, quantification of LAMP1 localization by distance to the nuclei and overlap with nuclei. Data was quantified from an average of 3–6 wells with nine fields per well.

Next, we wondered whether exogenous expression of NGLY1 protein in patient-derived NSCs would restore LC3BII to control levels. To this end, we used transduced NGLY1-p.R401X NSCs with lentivirus expressing wild type NGLY1 and measured LC3BII protein levels (Figure 7A). We noted a significant decrease in LC3BII levels to control levels when NGLY1 protein was introduced to patient-derived 1 cells, suggesting that autophagy induction is indeed caused by NGLY1 deficiency in patient-derived NSCs (Figure 7B).

Figure 7. Assessment of LC3BII with introduction of WT-NGLY1 in mutant cells.

Figure 7.

Open in a new tab

(A) Overexpression of WT-NGLY1 in Ngly1-R401X mutant cells. (B) Representative western blot (top) and quantification (below) for LC3BII in homozygous corrected control, NGLY1-R401X and Ngly1-R401X+WT-NGLY1; n=3 in each group. **P < 0.01 and *P<0.05 by one-way ANOVA analysis with Tukey’s post hoc test.

DISCUSSION

NGLY1 enzyme activity and human protein sequence were found in early 1990s and 2000 but its biological significance remained relatively unknown until 2012 when the first case of NGLY1 deficiency was diagnosed by whole exome sequencing [33]. Additional mutations of NGLY1 mutations were later reported [5, 34]. ER stress has been speculated as a major phenotype of NGLY1 deficiency. Since misfolded proteins are retro-translocated into the cytoplasm for degradation through the ERAD pathway, and some proteins require de-N-glycosylation for this process [35], an accumulation of misfolded or unfolded proteins could occur in the absence of NGLY1, which in turn, could lead to ER stress. However, to our knowledge, direct evidence of ER stress in patient cells has not been found. Here we used CRISPR-corrected isogenic control cells to compare with NGLY1-p.R401X cells to determine whether ER stress responses were activated, and whether other aberrations in related pathways could be used to identify phenotypic assays for a HTS.

Not surprisingly, the ERAD pathway was upregulated in NGLY1-p.R401X cells, as evidenced from both RNA-seq (Figure 2) and proteomics data (Figure 3). In addition, we also found evidence of increased ER stress by these pathway analyses. Experimental validation experiments showed that while some canonical ER stress markers such as CHOP and BiP were not elevated in NGLY1-p.R401X cells, the sXBP1 and the PERK-EIFα-ATF4 pathway of the UPR was significantly upregulated. These mediators of the UPR are also known to increase autophagy [31]. In agreement with this finding, we have shown by several methods that autophagy is indeed increased in NGLY1-p.R401X cells compared to their isogenic controls. Moreover, NGLY1-p.R401X cells also exhibited a significant increase in oxidative stress.

Undoubtedly, oxidative and ER stress and autophagy pathways are interconnected, but the crosstalk between these pathways is complicated. These cellular responses are dynamic processes that can affect one another depending on different factors such as cell type and other conditions that work to maintain cellular homeostasis. For instance, it is still not yet established whether ER stress enhances or inhibits autophagy. Most studies suggest that mild ER stress enhances autophagy, whereas chronic and prolonged ER stress could lead to inhibition of autophagic flux and eventual cell death [31]. Further, autophagy induction could itself be protective or detrimental to cellular homeostasis, which is likely dependent on the duration and/or intensity of ER stress [31]. In our model of NGLY1 deficient iPSC-derived NSCs, autophagy was increased and overexpressing NGLY1 in those cells reduced the levels of LC3BII to those of isogenic controls. Similar to our findings, several others have reported an increase in autophagy when NGLY1 is inhibited or with chemical activation of autophagy in the absence of NGLY1 in immortalized cell lines [36, 37] or in iPSC-derived neurons [18]. An increase in autophagosomes was also noted in NGLY1 deficient iPSC-derived brain organoids, suggesting that enhanced autophagy may persist beyond NSCs [21]. This phenomenon is also reflected in vivo in a mouse model of NGLY1 deficiency (Ngly1−/−), whereby increase in autophagy could compensate to remove ubiquitinated proteins when the ERAD pathway is disrupted [38]. Interestingly, a recent study showed that in the absence of NGLY1, cells failed to activate autophagy in response to proteotoxic stress in mouse cell lines [39]. These seemingly contradictory results could be attributable to species-specific differences, or due to differences in the experimental approaches used. Additionally, we cannot exclude the possibility of a cell-type-specific role for NGLY1 in mediating an autophagic response to stress. While most misfolded proteins are degraded by the ERAD pathway in a process by which proteins are ubiquitylated at the ER and marked for protein degradation by the ubiquitin-proteosome system, autophagic processes can also contribute to ER protein degradation [40]. In NGLY1-p.R401X cells, the increase in PERK-EIFα-ATF4 pathway as well as sXBP1 and lack of apoptotic cell death suggests that protective autophagy was induced in response to a relatively mild ER stress response. Other in vitro models of CDG and CDDG disease also report that these cells exhibit lower chronic ER stress response and an increase in sXBP1 with a concomitant increase in autophagy [16]. Such an increase in sXBP1 with protective autophagy has also been observed in response to mild ER stress in vivo [41]. It is also known that blocking autophagy can consequently lead to increase in ER stress and cell death [42, 43]. Therefore, in our model of NGLY1 deficiency, any early induction in full-blown ER stress may be averted due to enhanced autophagy-mediated clearance of misfolded proteins accumulation.

ER stress can lead to oxidative stress through processes such as disruption in calcium homeostasis and/or increased ROS production, and oxidative stress can exacerbate ER stress by disrupting normal protein folding [44]. Our omics analysis also found oxidative stress pathways to be upregulated in NGLY1-p.R401X cells. Inhibition of the ER resident protein PDI is known to promote accumulation of misfolded proteins and ER stress, as well as increase in oxidative stress [28, 45].Inhibition of PDI can also mediate clearance of misfolded or unfolded proteins through increased autophagy [46, 47]. PDI was decreased in NGLY1-p.R401X cells, which further supports the increase in oxidative stress. Moreover, the increase in autophagy may be another coping mechanism in NGLY1 deficient cells to attenuate oxidative stress. Others have reported increased oxidative stress or reduction in glutathione metabolism in different models of NGLY1 deficiency [15, 48, 49]. Using a CellRox dye for oxidative stress detection did not show differences between NGLY1-p.R401X and their isogenic control (Figure S2A). This lack of difference could be due several reasons such as inadequate sensitivity of the dye in our cells, lack of optimal conditions for the test performed, or other general difficulties in capturing accurate measurements of ROS [50]. Measurement of lipid peroxidation as a biomarker for oxidative stress is thought to be a useful tool for assessment of oxidative stress because •OH is the most reactive form of ROS and can initiate lipid peroxidation by attacking polyunsaturated fatty acids [29]. Therefore, increased lipid peroxidation can lead to tissue and cell membrane damage, which compromises the cell integrity. We found that NGLY1-p.R401X exhibited significantly increased levels of lipid peroxidation (Figure 5). Closely related to this, our RNAseq and proteomics data showed that ferroptosis was increased in NGLY1-p.R401X cells (Figures 2, 3 and S2B). Ferroptosis is a type of iron-dependent, ROS-mediated cell death that is independent from other forms of cell death such as apoptosis and necrosis [51]. Interestingly, ferroptosis may not be independent from autophagy-mediated cell death since several components of the autophagic machinery can contribute to ferroptosis [52]. Increased oxidative stress and impaired iron metabolism have also been noted in several neurodegenerative diseases [30], and could possibly also be a feature of NGLY1 deficiency. A study performed in human cancer cell lines showed that disruptions in NGLY1 sensitized the cells to ferroptosis, likely through a reduction in glutathione peroxidase 4 (GPX4) [53]. While we did not measure levels of iron or glutathione or GPX4 activity, our RNAseq and proteomics results indicate that ferroptosis is likely increased in NGLY1-p.R401X cells (Figure S2).

It has been shown by others that NGLY1 deficiency may enhance sensitivity to perturbations in cellular stress [37]. This is also apparent in our study, as there was a significant elevation in an apoptotic phenotype within 90 minutes of treatment with a known cell death inducer in NGLY1-p.R401X cells even though apoptosis was not increased under baseline conditions. This was in contrast to the NGLY1 replete CRISPR corrected lines, which did not show increased evidence of apoptotic cell death within the same timeframe. Another study similarly found that apoptosis in CDG and CDDG cell lines were low, and that such low levels were indicative of lower chronic ER stress responses [16]. We also found that NGLY1-p.R401X cells exhibit an increased susceptibility to lipid peroxidation beyond the existing elevated levels at baseline. Redox and proteotoxic stress with abnormal protein aggregates and mitochondrial damage are associated with neurodegenerative diseases [54]. Autophagy is also critical for alleviating oxidative stress, particularly in neural cell types, by clearing the accumulation of misfolded protein aggregates as well as damaged organelles such as mitochondria that contribute to excess ROS [55]. Our results indicate that in NGLY1-p.R401X, the increase in autophagic flux is likely an adaptation mechanism to prevent accumulation of excess misfolded or unfolded protein aggregates and cell death through apoptosis. In this model, the autophagic response might be sufficient to prevent a severe ER stress response and apoptotic cell death but might not be adequate to mitigate the extent of oxidative stress.

PERSPECTIVES AND CONCLUSION

Although a limitation to this study is that experiments were performed in vitro, results gained from this study could provide meaningful insights into mechanistic defects leading to NGLY1 disease pathology. Animal models and immortalized cell lines do not always recapitulate disease phenotype and often lead to high attrition rates for therapeutics at the clinical stage. For rare diseases, particularly those that include pathologies of a neurodevelopmental nature, iPSC-derived patient cells may be a better alternative for disease characterization and identification of suitable therapeutics since they offer a non-invasive, ethically sound, as well as disease and patient-specific solution [56]. We utilized CRISPR-corrected isogenic controls to ascertain phenotypic differences in patient iPSC-derived NSCs with NGLY1 deficiency due to p.R401X nonsense mutation. We observed significant upregulation in autophagic flux as well as lipid peroxidation/ferroptosis in NGLY1-p.R401X NSCs. Our data suggests that while there was evidence of ER stress, the increase in autophagy was able to prevent overt ER stress induced apoptotic cell death. This could also explain the lack of consistency in reports of ER stress in other models of NGLY1 deficiency.

Modulating one or more of these pathways in combination could mitigate cellular stress and help elucidate potential therapeutic options for this disease. For example, it may be beneficial to further enhance autophagy in these cells to overcome the adverse effects of oxidative stress through pharmacological interventions, some of which have shown promising results in preclinical models of neurodegeneration [57]. However, fine-tuning autophagic pathways to prevent deleterious effects of excessive autophagy might prove to be challenging in a clinical setting. Similarly, using ER stress inhibitors as a therapeutic option for neurodegenerative diseases is a valid approach; however these modulators still have several shortcomings, including unwarranted side-effects [58]. In vivo models of NGLY1 deficiency have shown that evidence for increased ER stress is lacking [13, 14]. In addition, variability in ER stress responses in different models and cell types indicates that developing phenotypic assays that utilize ER stress-based differences may not be currently feasible.

There appears to be good consensus on reports of increased oxidative stress in NGLY1 deficient in vivo models [48, 49]. Oxidative stress, including lipid peroxidation, is also known to be associated with common features of NGLY1 disease phenotype such as neurodegeneration and adverse neurodevelopmental effects on cognitive, sensory and motor functions [59, 60]. It is well known that neurons are highly susceptible to the adverse effects of lipid peroxidation and ferroptosis since they have high levels of polyunsaturated fatty acids and increased iron utilization [59]. Therefore, directly targeting avenues that alleviate oxidative stress and lipid peroxidation would be an appropriate therapeutic intervention. This could be achieved with treatment with antioxidants such as Coenzyme Q10, Vitamins A, C, K or compounds that reduce excess iron, or inhibitors of ferroptosis such as liproxstatins and ferrostatins [59, 61–63]. As an alternative, modulating the cell’s endogenous antioxidant system such as superoxide dismutase, catalase and glutathione peroxidase could be a practical option, since mimics for these antioxidant enzymes are already being tested in clinical trials [64]. In our patient relevant in vitro model, there was a strong indication that lipid peroxidation was increased. However, future studies in vivo will need to demonstrate whether increased lipid peroxidation and/or ferroptosis can also be observed with NGLY1 deficiency. In the meantime, these parameters could be used as a robust and reliable method for HTS and drug discovery.

Supplementary Material

1

Figure S1. Cleaved caspase staining in NSCs. Top, representative images. Below, quantification from 30 fields and 3–5 wells per condition for normalized CC3 intensity or number of spots. Magenta= cleaved caspase 3, and Cyan=Nuclei. **P < 0.01 by two-way ANOVA with Tukey’s post hoc test.

2

Figure S2. Oxidative stress and ferroptosis pathway in NSCs. (A) Menadione dose-dependently increased oxidative stress measurement in both NGLY1-p. R40X cells and their isogenic controls; n=4 biological replicates for each condition, as assessed by CellROX dye. (B) Heatmaps demonstrating markers from RNA-seq and proteomics data that are associated with ferroptosis pathway identified from the IPA knowledge base. Each heatmap represents the expression levels of the pathway-associated markers across samples, with color intensity reflecting relative expression or abundance. Rows corresponds to markers and columns correspond to samples.

3

Figure S3. Tandem LC3B fluorescence signal. Left, representative images. Right, quantification from 16 fields of view per well and 6 wells per condition. ****P <0.0001by students t-test

HIGHLIGHTS.

  • NGLY1 deficient neural stem cells show increase in components of the unfolded protein response

  • Lipid peroxidation is pronounced in NGLY1 deficient neural stem cells

  • Autophagy is upregulated in NGLY1 deficient neural stem cells

ACKNOWLEDGEMENTS

Graphical abstract was created with BioRender.com. This work was supported by the Intramural Research Programs of the National Center for Advancing Translational Sciences, National Institutes of Health. We would like to thank Atena Farkhondeh Kalat for providing management support for the NGLY1 project.

Footnotes

CONFLICT OF INTEREST

The authors declare no conflict of interest

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

DATA AVAILABILITY

The datasets presented in this study can be found in Zenodo data repository: https://doi.org/10.5281/zenodo.15079723

REFERENCES

  • [1].Suzuki T, Huang C, Fujihira H, The cytoplasmic peptide:N-glycanase (NGLY1) - Structure, expression and cellular functions, Gene, 577 (2016) 1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [2].Fujihira H, Asahina M, Suzuki T, Physiological importance of NGLY1, as revealed by rodent model analyses, J Biochem, 171 (2022) 161–167. [DOI] [PubMed] [Google Scholar]
  • [3].Need AC, Shashi V, Hitomi Y, Schoch K, Shianna KV, McDonald MT, Meisler MH, Goldstein DB, Clinical application of exome sequencing in undiagnosed genetic conditions, J Med Genet, 49 (2012) 353–361. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [4].Tong S, Ventola P, Frater CH, Klotz J, Phillips JM, Muppidi S, Dwight SS, Mueller WF, Beahm BJ, Wilsey M, Lee KJ, NGLY1 deficiency: a prospective natural history study, Hum Mol Genet, 32 (2023) 2787–2796. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Enns GM, Shashi V, Bainbridge M, Gambello MJ, Zahir FR, Bast T, Crimian R, Schoch K, Platt J, Cox R, Bernstein JA, Scavina M, Walter RS, Bibb A, Jones M, Hegde M, Graham BH, Need AC, Oviedo A, Schaaf CP, Boyle S, Butte AJ, Chen R, Chen R, Clark MJ, Haraksingh R, Consortium FC, Cowan TM, He P, Langlois S, Zoghbi HY, Snyder M, Gibbs RA, Freeze HH, Goldstein DB, Mutations in NGLY1 cause an inherited disorder of the endoplasmic reticulum-associated degradation pathway, Genet Med, 16 (2014) 751–758. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [6].Stanclift CR, Dwight SS, Lee K, Eijkenboom QL, Wilsey M, Wilsey K, Kobayashi ES, Tong S, Bainbridge MN, NGLY1 deficiency: estimated incidence, clinical features, and genotypic spectrum from the NGLY1 Registry, Orphanet J Rare Dis, 17 (2022) 440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [7].Zdrazilova L, Rakosnikova T, Himmelreich N, Ondruskova N, Pasak M, Vanisova M, Volfova N, Honzik T, Thiel C, Hansikova H, Metabolic adaptation of human skin fibroblasts to ER stress caused by glycosylation defect in PMM2-CDG, Mol Genet Metab, 139 (2023) 107629. [DOI] [PubMed] [Google Scholar]
  • [8].Lecca MR, Wagner U, Patrignani A, Berger EG, Hennet T, Genome-wide analysis of the unfolded protein response in fibroblasts from congenital disorders of glycosylation type-I patients, FASEB J, 19 (2005) 240–242. [DOI] [PubMed] [Google Scholar]
  • [9].Lin JH, Walter P, Yen TS, Endoplasmic reticulum stress in disease pathogenesis, Annu Rev Pathol, 3 (2008) 399–425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [10].Oslowski CM, Urano F, Measuring ER stress and the unfolded protein response using mammalian tissue culture system, Methods Enzymol, 490 (2011) 71–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [11].Wodrich APK, Scott AW, Shukla AK, Harris BT, Giniger E, The Unfolded Protein Responses in Health, Aging, and Neurodegeneration: Recent Advances and Future Considerations, Front Mol Neurosci, 15 (2022) 831116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [12].Ghemrawi R, Khair M, Endoplasmic Reticulum Stress and Unfolded Protein Response in Neurodegenerative Diseases, Int J Mol Sci, 21 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Owings KG, Lowry JB, Bi Y, Might M, Chow CY, Transcriptome and functional analysis in a Drosophila model of NGLY1 deficiency provides insight into therapeutic approaches, Hum Mol Genet, 27 (2018) 1055–1066. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Asahina M, Fujinawa R, Nakamura S, Yokoyama K, Tozawa R, Suzuki T, Ngly1 −/− rats develop neurodegenerative phenotypes and pathological abnormalities in their peripheral and central nervous systems, Hum Mol Genet, 29 (2020) 1635–1647. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Rauscher B, Mueller WF, Clauder-Munster S, Jakob P, Islam MS, Sun H, Ghidelli-Disse S, Boesche M, Bantscheff M, Pflaumer H, Collier P, Haase B, Chen S, Hoffman R, Wang G, Benes V, Drewes G, Snyder M, Steinmetz LM, Patient-derived gene and protein expression signatures of NGLY1 deficiency, J Biochem, 171 (2022) 187–199. [DOI] [PubMed] [Google Scholar]
  • [16].Lebedeva IV, Wagner MV, Sahdeo S, Lu YF, Anyanwu-Ofili A, Harms MB, Wadia JS, Rajagopal G, Boland MJ, Goldstein DB, Precision genetic cellular models identify therapies protective against ER stress, Cell Death Dis, 12 (2021) 770. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Galeone A, Adams JM, Matsuda S, Presa MF, Pandey A, Han SY, Tachida Y, Hirayama H, Vaccari T, Suzuki T, Lutz CM, Affolter M, Zuberi A, Jafar-Nejad H, Regulation of BMP4/Dpp retrotranslocation and signaling by deglycosylation, Elife, 9 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [18].Manole A, Wong T, Rhee A, Novak S, Chin SM, Tsimring K, Paucar A, Williams A, Newmeyer TF, Schafer ST, Rosh I, Kaushik S, Hoffman R, Chen S, Wang G, Snyder M, Cuervo AM, Andrade L, Manor U, Lee K, Jones JR, Stern S, Marchetto MC, Gage FH, NGLY1 mutations cause protein aggregation in human neurons, Cell Rep, 42 (2023) 113466. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [19].Yu D, Swaroop M, Wang M, Baxa U, Yang R, Yan Y, Coksaygan T, DeTolla L, Marugan JJ, Austin CP, McKew JC, Gong DW, Zheng W, Niemann-Pick Disease Type C: Induced Pluripotent Stem Cell-Derived Neuronal Cells for Modeling Neural Disease and Evaluating Drug Efficacy, J Biomol Screen, 19 (2014) 1164–1173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Pavlinov I, Farkhondeh A, Yang S, Xu M, Cheng YS, Beers J, Zou J, Liu C, Might M, Rodems S, Baumgartel K, Zheng W, Generation of two gene corrected human isogenic iPSC lines (NCATS-CL6104 and NCATS-CL6105) from a patient line (NCATS-CL6103) carrying a homozygous p.R401X mutation in the NGLY1 gene using CRISPR/Cas9, Stem Cell Res, 56 (2021) 102554. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Abbott J, Tambe M, Pavlinov I, Farkhondeh A, Nguyen HN, Xu M, Pradhan M, York T, Might M, Baumgartel K, Rodems S, Zheng W, Generation and characterization of NGLY1 patient-derived midbrain organoids, Front Cell Dev Biol, 11 (2023) 1039182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Love MI, Huber W, Anders S, Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2, Genome Biol, 15 (2014) 550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].Yang K, Huang R, Fujihira H, Suzuki T, Yan N, N-glycanase NGLY1 regulates mitochondrial homeostasis and inflammation through NRF1, J Exp Med, 215 (2018) 2600–2616. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].He P, Grotzke JE, Ng BG, Gunel M, Jafar-Nejad H, Cresswell P, Enns GM, Freeze HH, A congenital disorder of deglycosylation: Biochemical characterization of N-glycanase 1 deficiency in patient fibroblasts, Glycobiology, 25 (2015) 836–844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Suzuki T, The cytoplasmic peptide:N-glycanase (Ngly1)-basic science encounters a human genetic disorder, J Biochem, 157 (2015) 23–34. [DOI] [PubMed] [Google Scholar]
  • [26].Fujihira H, Masahara-Negishi Y, Tamura M, Huang C, Harada Y, Wakana S, Takakura D, Kawasaki N, Taniguchi N, Kondoh G, Yamashita T, Funakoshi Y, Suzuki T, Lethality of mice bearing a knockout of the Ngly1-gene is partially rescued by the additional deletion of the Engase gene, PLoS Genet, 13 (2017) e1006696. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [27].Fan D, Cong Y, Liu J, Zhang H, Du Z, Spatiotemporal analysis of mRNA-protein relationships enhances transcriptome-based developmental inference, Cell Rep, 43 (2024) 113928. [DOI] [PubMed] [Google Scholar]
  • [28].Jha V, Kumari T, Manickam V, Assar Z, Olson KL, Min JK, Cho J, ERO1-PDI Redox Signaling in Health and Disease, Antioxid Redox Signal, 35 (2021) 1093–1115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [29].Ito F, Sono Y, Ito T, Measurement and Clinical Significance of Lipid Peroxidation as a Biomarker of Oxidative Stress: Oxidative Stress in Diabetes, Atherosclerosis, and Chronic Inflammation, Antioxidants (Basel), 8 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [30].Chen X, Guo C, Kong J, Oxidative stress in neurodegenerative diseases, Neural Regen Res, 7 (2012) 376–385. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [31].Rashid HO, Yadav RK, Kim HR, Chae HJ, ER stress: Autophagy induction, inhibition and selection, Autophagy, 11 (2015) 1956–1977. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [32].Tan KP, Ho MY, Cho HC, Yu J, Hung JT, Yu AL, Fucosylation of LAMP-1 and LAMP-2 by FUT1 correlates with lysosomal positioning and autophagic flux of breast cancer cells, Cell Death Dis, 7 (2016) e2347. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [33].Suzuki T, Yoshida Y, Ever-expanding NGLY1 biology, J Biochem, 171 (2022) 141–143. [DOI] [PubMed] [Google Scholar]
  • [34].Lam C, Ferreira C, Krasnewich D, Toro C, Latham L, Zein WM, Lehky T, Brewer C, Baker EH, Thurm A, Farmer CA, Rosenzweig SD, Lyons JJ, Schreiber JM, Gropman A, Lingala S, Ghany MG, Solomon B, Macnamara E, Davids M, Stratakis CA, Kimonis V, Gahl WA, Wolfe L, Prospective phenotyping of NGLY1-CDDG, the first congenital disorder of deglycosylation, Genet Med, 19 (2017) 160–168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [35].Pandey A, Adams JM, Han SY, Jafar-Nejad H, NGLY1 Deficiency, a Congenital Disorder of Deglycosylation: From Disease Gene Function to Pathophysiology, Cells, 11 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [36].Needs SH, Bootman MD, Grotzke JE, Kramer HB, Allman SA, Off-target inhibition of NGLY1 by the polycaspase inhibitor Z-VAD-fmk induces cellular autophagy, FEBS J, 289 (2022) 3115–3131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [37].Mueller WF, Jakob P, Sun H, Clauder-Munster S, Ghidelli-Disse S, Ordonez D, Boesche M, Bantscheff M, Collier P, Haase B, Benes V, Paulsen M, Sehr P, Lewis J, Drewes G, Steinmetz LM, Loss of N-Glycanase 1 Alters Transcriptional and Translational Regulation in K562 Cell Lines, G3 (Bethesda), 10 (2020) 1585–1597. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [38].Asahina M, Fujinawa R, Fujihira H, Masahara-Negishi Y, Andou T, Tozawa R, Suzuki T, JF1/B6F1 Ngly1(−/−) mouse as an isogenic animal model of NGLY1 deficiency, Proc Jpn Acad Ser B Phys Biol Sci, 97 (2021) 89–102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [39].Ward MA, Vangala JR, Kamber Kaya HE, Byers HA, Hosseini N, Diaz A, Cuervo AM, Kaushik S, Radhakrishnan SK, Transcription factor Nrf1 regulates proteotoxic stress-induced autophagy, J Cell Biol, 223 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [40].Mehrtash AB, Hochstrasser M, Ubiquitin-dependent protein degradation at the endoplasmic reticulum and nuclear envelope, Semin Cell Dev Biol, 93 (2019) 111–124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [41].Fouillet A, Levet C, Virgone A, Robin M, Dourlen P, Rieusset J, Belaidi E, Ovize M, Touret M, Nataf S, Mollereau B, ER stress inhibits neuronal death by promoting autophagy, Autophagy, 8 (2012) 915–926. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [42].Ghosh AK, Mau T, O’Brien M, Garg S, Yung R, Impaired autophagy activity is linked to elevated ER-stress and inflammation in aging adipose tissue, Aging (Albany NY), 8 (2016) 2525–2537. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [43].Ogata M, Hino S, Saito A, Morikawa K, Kondo S, Kanemoto S, Murakami T, Taniguchi M, Tanii I, Yoshinaga K, Shiosaka S, Hammarback JA, Urano F, Imaizumi K, Autophagy is activated for cell survival after endoplasmic reticulum stress, Mol Cell Biol, 26 (2006) 9220–9231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [44].Cao SS, Kaufman RJ, Endoplasmic reticulum stress and oxidative stress in cell fate decision and human disease, Antioxid Redox Signal, 21 (2014) 396–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [45].Perri ER, Thomas CJ, Parakh S, Spencer DM, Atkin JD, The Unfolded Protein Response and the Role of Protein Disulfide Isomerase in Neurodegeneration, Front Cell Dev Biol, 3 (2015) 80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [46].Lehtonen S, Jaronen M, Vehvilainen P, Lakso M, Rudgalvyte M, Keksa-Goldsteine V, Wong G, Courtney MJ, Koistinaho J, Goldsteins G, Inhibition of Excessive Oxidative Protein Folding Is Protective in MPP(+) Toxicity-Induced Parkinson’s Disease Models, Antioxid Redox Signal, 25 (2016) 485–497. [DOI] [PubMed] [Google Scholar]
  • [47].Wang R, Shang Y, Chen B, Xu F, Zhang J, Zhang Z, Zhao X, Wan X, Xu A, Wu L, Zhao G, Protein disulfide isomerase blocks the interaction of LC3II-PHB2 and promotes mTOR signaling to regulate autophagy and radio/chemo-sensitivity, Cell Death Dis, 13 (2022) 851. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [48].Han SY, Pandey A, Moore T, Galeone A, Duraine L, Cowan TM, Jafar-Nejad H, A conserved role for AMP-activated protein kinase in NGLY1 deficiency, PLoS Genet, 16 (2020) e1009258. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [49].Kong J, Peng M, Ostrovsky J, Kwon YJ, Oretsky O, McCormick EM, He M, Argon Y, Falk MJ, Mitochondrial function requires NGLY1, Mitochondrion, 38 (2018) 6–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [50].Murphy MP, Bayir H, Belousov V, Chang CJ, Davies KJA, Davies MJ, Dick TP, Finkel T, Forman HJ, Janssen-Heininger Y, Gems D, Kagan VE, Kalyanaraman B, Larsson NG, Milne GL, Nystrom T, Poulsen HE, Radi R, Van Remmen H, Schumacker PT, Thornalley PJ, Toyokuni S, Winterbourn CC, Yin H, Halliwell B, Guidelines for measuring reactive oxygen species and oxidative damage in cells and in vivo, Nat Metab, 4 (2022) 651–662. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [51].Rochette L, Dogon G, Rigal E, Zeller M, Cottin Y, Vergely C, Lipid Peroxidation and Iron Metabolism: Two Corner Stones in the Homeostasis Control of Ferroptosis, Int J Mol Sci, 24 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [52].Zhou B, Liu J, Kang R, Klionsky DJ, Kroemer G, Tang D, Ferroptosis is a type of autophagy-dependent cell death, Semin Cancer Biol, 66 (2020) 89–100. [DOI] [PubMed] [Google Scholar]
  • [53].Forcina GC, Pope L, Murray M, Dong W, Abu-Remaileh M, Bertozzi CR, Dixon SJ, Ferroptosis regulation by the NGLY1/NFE2L1 pathway, Proc Natl Acad Sci U S A, 119 (2022) e2118646119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [54].Redmann M, Darley-Usmar V, Zhang J, The Role of Autophagy, Mitophagy and Lysosomal Functions in Modulating Bioenergetics and Survival in the Context of Redox and Proteotoxic Damage: Implications for Neurodegenerative Diseases, Aging Dis, 7 (2016) 150–162. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [55].Ornatowski W, Lu Q, Yegambaram M, Garcia AE, Zemskov EA, Maltepe E, Fineman JR, Wang T, Black SM, Complex interplay between autophagy and oxidative stress in the development of pulmonary disease, Redox Biol, 36 (2020) 101679. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [56].Freel BA, Sheets JN, Francis KR, iPSC modeling of rare pediatric disorders, J Neurosci Methods, 332 (2020) 108533. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [57].Guo F, Liu X, Cai H, Le W, Autophagy in neurodegenerative diseases: pathogenesis and therapy, Brain Pathol, 28 (2018) 3–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [58].Hasan S-A-M, James AW, Fazili FM, Tarabishi S, Sheikh NM, Shah ZA, Endoplasmic Reticulum Stress in Neurodegenerative Diseases, Journal of Dementia and Alzheimer's Disease, 1 (2024) 87–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [59].Ryan SK, Ugalde CL, Rolland AS, Skidmore J, Devos D, Hammond TR, Therapeutic inhibition of ferroptosis in neurodegenerative disease, Trends Pharmacol Sci, 44 (2023) 674–688. [DOI] [PubMed] [Google Scholar]
  • [60].Nishimura Y, Kanda Y, Sone H, Aoyama H, Oxidative Stress as a Common Key Event in Developmental Neurotoxicity, Oxid Med Cell Longev, 2021 (2021) 6685204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [61].Teleanu RI, Chircov C, Grumezescu AM, Volceanov A, Teleanu DM, Antioxidant Therapies for Neuroprotection-A Review, J Clin Med, 8 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [62].Li Z, Zhang Y, Ji M, Wu C, Zhang Y, Ji S, Targeting ferroptosis in neuroimmune and neurodegenerative disorders for the development of novel therapeutics, Biomed Pharmacother, 176 (2024) 116777. [DOI] [PubMed] [Google Scholar]
  • [63].Skouta R, Dixon SJ, Wang J, Dunn DE, Orman M, Shimada K, Rosenberg PA, Lo DC, Weinberg JM, Linkermann A, Stockwell BR, Ferrostatins inhibit oxidative lipid damage and cell death in diverse disease models, J Am Chem Soc, 136 (2014) 4551–4556. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [64].Forman HJ, Zhang H, Author Correction: Targeting oxidative stress in disease: promise and limitations of antioxidant therapy, Nat Rev Drug Discov, 20 (2021) 652. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1

Figure S1. Cleaved caspase staining in NSCs. Top, representative images. Below, quantification from 30 fields and 3–5 wells per condition for normalized CC3 intensity or number of spots. Magenta= cleaved caspase 3, and Cyan=Nuclei. **P < 0.01 by two-way ANOVA with Tukey’s post hoc test.

NIHMS2073199-supplement-1.tif (13.5MB, tif)
2

Figure S2. Oxidative stress and ferroptosis pathway in NSCs. (A) Menadione dose-dependently increased oxidative stress measurement in both NGLY1-p. R40X cells and their isogenic controls; n=4 biological replicates for each condition, as assessed by CellROX dye. (B) Heatmaps demonstrating markers from RNA-seq and proteomics data that are associated with ferroptosis pathway identified from the IPA knowledge base. Each heatmap represents the expression levels of the pathway-associated markers across samples, with color intensity reflecting relative expression or abundance. Rows corresponds to markers and columns correspond to samples.

NIHMS2073199-supplement-2.tif (3.3MB, tif)
3

Figure S3. Tandem LC3B fluorescence signal. Left, representative images. Right, quantification from 16 fields of view per well and 6 wells per condition. ****P <0.0001by students t-test

NIHMS2073199-supplement-3.tif (12.6MB, tif)

Data Availability Statement

The datasets presented in this study can be found in Zenodo data repository: https://doi.org/10.5281/zenodo.15079723

RESOURCES