The Interplay Between Endoplasmic Reticulum Stress and Oxidative Stress in Chondrocyte Catabolism

Abstract

Objective

Oxidative stress and endoplasmic reticulum (ER) stress play pivotal roles in disrupting the homeostasis of chondrocytes by producing catalytic proteases and enhancing chondrocyte senescence, consequently contributing to the progression of osteoarthritis (OA). Despite their close interaction, the underlying molecular mechanisms remain poorly understood. Here, we show that ER stress and oxidative stress reciprocally modulate each other to promote cartilage degradation.

Methods

Primary chondrocytes were obtained from the articular cartilage of 5-day-old C57BL/6J mice by excising distal femur and proximal tibia. Tunicamycin was applied to induce ER stress in primary chondrocytes. Surgical OA was induced in 12-week-old male C57BL/6J mice by destabilizing the medial meniscus (DMM).

Results

Tunicamycin-induced ER stress led to an increase in the production of reactive oxygen species (ROS) and catalytic proteases, including MMP13 and Adamts5, in primary chondrocytes, and it was primarily dependent on the NADPH oxidase (NOX) system. ER stress directly increased the expression of NOX2, NOX3, NOX4, and p22phox. Specifically, the protein kinase RNA-like ER kinase (PERK) pathway is involved in the expression of NOX4 and p22phox, the inositol-requiring enzyme 1 alpha (IRE1α) pathway in NOX2 and NOX3 expression, and the activating transcription factor 6 (ATF6) pathway influences NOX3 expression in chondrocytes. Conversely, inhibiting NOX function significantly reduced both ER stress sensor–related signaling and chondrocyte catabolism, thereby decelerating the progression of surgically induced OA in vivo.

Conclusions

Our findings highlight the positive feedback loop between ER stress and oxidative stress in OA pathogenesis, suggesting that targeting NOX isoforms is a promising therapeutic strategy for OA.

Keywords

osteoarthritis NADPH oxidase oxidative stress ER stress chondrocyte

Introduction

Chondrocytes appear to remain in a quiescent state with insufficient regenerative capacity within the dense, hypoxic extracellular matrix (ECM) environment.¹ However, they are highly active in producing a diverse range of ECM proteins, notably type II collagen (Col2) and aggrecan.^1-3 These proteins are essential for providing structural support, tensile strength, and elasticity to the cartilage, thereby maintaining cartilage homeostasis and offering the necessary framework for proper joint function.^2,3 Recent evidence has shown that chondrocytes from lesional cartilage exhibit accelerated protein synthesis compared with chondrocytes from non-lesional cartilage in the same osteoarthritis (OA) patient.^3,4

The protein synthesis requires intricate processes, such as protein-folding and post-translational modifications, which take place in the endoplasmic reticulum (ER).⁵ However, in the presence of cellular stress, the protein-folding capacity of the ER may become insufficient to meet the increased demand, resulting in an accumulation of misfolded or unfolded proteins within the ER.⁶ This accumulation leads to ER stress, which subsequently triggers the activation of the unfolded protein response (UPR).^6,7 UPR is mainly mediated by 3 ER transmembrane sensors, called protein kinase RNA-like ER Kinase (PERK), activating transcription factor 6 (ATF6), and inositol-requiring enzyme 1 α (IRE1α).⁶ The primary goal of the UPR is to restore ER homeostasis by improving the protein-folding capacity, reducing protein synthesis, and facilitating the degradation of misfolded proteins.⁷ However, when the ER stress persists, the UPR can have detrimental effects on chondrocytes, shifting the balance toward catabolic processes.^7,8 Indeed, while mild ER stress can be protective for chondrocytes by activating autophagy via the 78-kDa glucose-regulated protein (GRP78) pathway, excessive ER stress can lead to chondrocyte apoptosis, ultimately contributing to the progression of OA.^9,10 Therefore, chondrocytes have evolved mechanisms to counteract excessive ER stress. For instance, the AMP-activated protein kinase (AMPK) signaling can inhibit CHOP-mediated chondrocyte apoptosis,¹¹ and the sirtuin-1 attenuates the PERK-eIF-2α-CHOP axis by deacetylating PERK, thereby maintaining cartilage homeostasis.¹²

The pathophysiologic stresses that can induce ER stress and the UPR include hyperglycemia, hyperlipidemia, hypoxic stress, oxidative stress, calcium imbalance, and nutrient deprivation.^8,13 Among them, increased oxidative stress impairs the protein folding process and disulfide bond formation, ultimately leading to ER stress, which can be mitigated by antioxidants, like N-acetyl-cysteine.^14,15 Not only does reactive oxygen species (ROS) induce ER stress, but recent evidence highlights that ER stress can in turn enhance ROS production.¹⁶ This bidirectional relationship between ER stress and ROS forms a positive feedback loop that can amplify cellular stress responses and promote catabolic processes, synergistically contributing to cartilage degeneration in OA.^17,18 However, there is a lack of research exploring the detailed mechanisms through which ER stress increases ROS production. In our previous study, we reported that the disruption of the ECM in a chondroitin sulfate-based 3D hydrogel increases oxidative stress in chondrocytes, potentially involving the NADPH oxidase (NOX) and mitochondrial ROS systems.^19,20 These findings suggest that the breakdown of the cartilage ECM due to excessive mechanical stress may result in increased ROS production, leading to elevated ER stress in the context of OA.

In this study, we aimed to investigate the molecular mechanisms through which ER stress enhances ROS production in chondrocytes. We found that NOX system plays an important role in ROS production and catabolic response in chondrocytes under ER stress condition. Then, we investigated which ER stress sensor signaling pathways control the expression of NOX isoforms. Finally, we examined the role of inhibiting NOX isoforms both in the ER stress sensor signaling and catabolic response in chondrocytes, as well as in an in vivo, surgically induced OA model. By elucidating the molecular mechanisms underlying the crosstalk between ER stress and ROS in chondrocytes, we aim to provide new insights into the pathophysiology of OA and identify potential therapeutic targets for the treatment of this debilitating disease.

Materials and Methods

Primary Chondrocyte Cultures

All animal experiments were approved by the Animal Care Committee of Kyungpook National University (Approval No. KNU-2022-0203) and were performed in accordance with the guidelines of United States’ Public Health Service (PHS) Policy on Use of Laboratory Animals.²¹ Primary chondrocytes were obtained from the articular cartilage of 5-day-old C57BL/6J mice by excising the distal femur and proximal tibia, which were then minced into small pieces. The cartilage pieces underwent a 15-minute incubation in 0.25% trypsin-EDTA at 37°C with gentle shaking and then were digested in 3 mg/mL of Collagenase P (Sigma-Aldrich) within complete Dulbecco’s Modified Essential Media (DMEM) for 4 hours. After digestion, the cell suspension was filtered through a 40-μm nylon mesh to eliminate undigested tissue. The chondrocytes were centrifuged at 300 x g for 5 minutes, washed twice with phosphate-buffered saline (PBS), and then resuspended in a 3:2 F12:DMEM-based culture medium with 10% fetal bovine serum (FBS), 0.25% l-glutamine, and 0.25% penicillin/streptomycin. Isolated primary chondrocytes were seeded at a density of 1 x 10⁵ cells/cm² in a 6-well plate. To reduce de-differentiation risks, only chondrocytes at passage 0 were used in experiments.

To investigate the effects of ER stress on primary chondrocytes, cells were treated with tunicamycin at concentrations of 0.5, 1.0, and 2 μg/mL or with equivalent volume of dimethyl sulfoxide (DMSO) (0.1% v/v) as control, and incubated for 6 hours in RNA isolation and 24 hours in protein isolation and staining. To investigate the cellular origin of ROS under ER stress conditions, cells were pre-treated with an ER stress inhibitor, tauroursodeoxycholic acid (TUDCA), and various ROS system inhibitors, including 2-deoxy-d-glucose (2-DG, a glycolysis inhibitor), IACS-010759 (an oxidative phosphorylation inhibitor), 1400 W (an inhibitor of inducible nitric oxide synthase, iNOS), and APX-115 (a pan-NOX inhibitor), all of which were pre-treated for 30 minutes prior to the administration of tunicamycin (2 μg/mL). To assess the effect of distinct signaling pathways of ER stress, specific inhibitors for PERK (MKC8866; 10 μM), IRE1α (GSK2606414; 10 μM), as well as ATF6α pathway (Caepins; 10 μM) were pre-treated for 30 minutes prior to tunicamycin (2 μg/mL).

Analysis of Oxidative Stress: Detection of Intracellular ROS and Oxidative DNA Damage

Intracellular and mitochondrial ROS production, as well as oxidative DNA damage in primary chondrocytes, were assessed using dihydroethidium (DHE) and mitoSOX Red (ThermoFisher Scientific, #M36008) fluorescent dyes for ROS, and the 8-oxo-dG antibody for DNA damage, respectively. Primary chondrocytes were fixed with 4% paraformaldehyde for 10 minutes, permeabilized using 0.25% Triton X-100, and then rinsed 3 times in PBS. To assess intracellular ROS, chondrocytes were incubated with 5 μM DHE for 30 minutes at 37°C, or incubated with 2.5 μM mitoSOX Red for 10 minutes at 37°C, and washed with PBS, and then counterstained with DAPI. Images were captured using a KI-3000F fluorescence microscope (Korealabtech, South Korea). For immunofluorescence staining of 8-oxo-dG, cells were incubated with the primary 8-oxo-dG antibody (sc-66036; Santa Cruz Technology, Beverly, MA) overnight at 4°C. Then, cells were incubated with a secondary antibody for 2 hours, counterstained with DAPI, and imaged using a fluorescence microscope. Fluorescence-positive cells were quantified using ImageJ software (version 1.8.0, National Institutes of Health, Bethesda, MD).

Quantitative Real-Time Reverse Transcription-PCR

From cultured primary chondrocytes, total RNA was extracted using TRIzol (Invitrogen, Carlsbad, CA), followed by the generation of the first-strand cDNA with Superscript III reverse transcriptase (Invitrogen). A ViiA 7 Real-Time PCR System (Applied Biosystems, CA) and SYBR Green Master Mix (Applied Biosystems) were employed for real-time qPCR. Primers for real-time qPCR can be found in Suppl. Table S1. Target gene expression levels were calculated using the 2−ΔΔCT method and normalized to the geometric mean of Gapdh, which served as an internal control. All quantitative real-time reverse transcription-PCR (qRT-PCR) analysis were carried out in triplicate, repeated 3 to 5 times, and the average results for each were presented.

Western Blot Analysis and Quantification

Proteins from primary chondrocyte cultured in 6-well plates were extracted using 300 μL RIPA buffer supplemented with protease and phosphatase inhibitors (Roche Diagnostics, Indianapolis, IN). Total cell lysates containing 10 to 20 μg of protein were subjected to 10% SDS-polyacrylamide gel electrophoresis and subsequently transferred onto polyvinylidene difluoride membranes (Immobilon-P; Millipore Corporation, Billerica, MA). Membranes were blocked using 5% skimmed milk in PBS with 0.25% Tween-20 (PBST) and incubated with the primary antibodies overnight at 4°C detailed in Suppl. Table S2. After washing with PBST, membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies at room temperature for 2 hours and then blots were developed using enhanced chemiluminescence Western blotting detection reagent (Thermo Fisher Scientific) and examined with the MicroChemi system (DNR Bio-imaging Systems). All analysis was biologically triplicated, and the Western blot band intensities were quantified using ImageJ software (version 1.8.0, NIH). Band intensities of target proteins were normalized to the β-actin band intensity of respective lanes.

Induction of Surgical OA Model and APX115 Treatment

Surgical OA was induced in 7 male 12-week-old C57BL/6 mice per group by destabilization of the medial meniscus (DMM) surgery under general anesthesia. All mice were purchased from Hana Biotech Co. (Pyeongtaek-si, South Korea) and were housed at a temperature of 23°C ± 1°C with a 12-hour light/dark cycle. They were kept under specific pathogen-free conditions in the animal facilities of Kyungpook National University Chilgok Hospital. The sample size was chosen based on our prior study that demonstrated statistically significant effect sizes.²⁰ Before starting the treatment, a total of 14 mice that underwent DMM surgery were randomly allocated into 1 of 2 groups. They were housed 2 per plastic cage, and the placement of the cages in the room was also randomized. The treatment group received intraperitoneal injections of APX-115 at a dosage of 30 mg/kg twice weekly, whereas the control group received an equivalent volume of DMSO in the same frequency. The dosage of APX-115 was based on a previous study that reported significant anti-inflammatory effects through the oral route in diabetic mice, as well as on our prior research.^20,22 Eight weeks after surgery, the mice were euthanized via cervical dislocation, and knee joints were collected for histological examination, and all specimens were used for analysis. Treatments were administered by a separate technician who was aware of the groupings, but all assessments, measurements, and analyses on the mice were performed by researchers unaware of the mice’s treatment assignments. This study was conducted in accordance with ARRIVE guidelines 2.0, and the associated ARRIVE guideline checklist has also been offered in the Supplementary sections.

Safranin-O Staining and Immunohistochemistry of Mouse Knee Joint

Mouse knee joints were fixed in 4% paraformaldehyde for 24 hours, decalcified in 10% EDTA for 3 weeks, and then embedded in paraffin. The embedded blocks were sectioned at a thickness of 6 μm and stained with Safranin-O/Fast Green. Cartilage destruction in the total all 4 quadrants of joint (grade 0-24) was scored by 2 observers under blinded conditions using the Osteoarthritis Research Society International (OARSI) score system.²³ For immunohistochemical staining, the rehydrated sections underwent antigen retrieval in sodium citrate buffer (10 mM sodium citrate, 0.05% Tween 20, pH 6.0) or by boiling for 15 minutes at temperature ranging from 95°C to 100°C. After blocking the sections with 2% bovine serum albumin (BSA) in PBS for 1 hour, they were incubated overnight at 4°C with primary antibodies for MMP13, CHOP, Grp78, or normal rabbit IgG diluted in 1% BSA (Suppl. Table S2). Subsequently, the sections were incubated with a biotinylated anti-rabbit secondary antibody (BA-1000; Sigma-Aldrich) and visualized with the Vector Red Alkaline Phosphatase Kit (AK-5200, Vector Labs, CA). Sections were mounted with anti-fade mounting solution (Vector Labs) and imaged and quantified under the KI-3000F fluorescence microscope. The immunostaining images of MMP13, CHOP, and Grp78 were quantified in the upper part of the tide mark of lateral tibial plateau.

Statistical Analysis

All data are presented herein as mean ± standard deviation (SD). Statistical analyses to compare the mean values of 2 groups were performed using the Mann-Whitney U test, a nonparametric test selected due to the small sample size and the lack of a normality assumption. A P-value of less than 0.05 was considered statistically significant. All statistical analyses were carried out using Prism software version 8.0 (GraphPad Software, La Jolla, CA).

Results

ER Stress Increases Oxidative Stress, Primarily via the NOX System

First, we established the tunicamycin-induced ER stress condition. Tunicamycin was treated at various dosages, and the expression of UPR marker proteins was assessed. Tunicamycin increased the protein levels of CHOP, XBP1s, and Grp78 in a dose-dependent manner, with the most significant effects observed at a concentration of 2 μg/mL ( Fig. 1A and B ). To investigate whether ER stress increases oxidative stress in chondrocytes, we treated the cells with tunicamycin and assessed ROS production using DHE staining, as well as ROS-induced DNA damage using 8-oxo-dG. Tunicamycin increased DHE and 8-oxo-dG levels in primary chondrocytes in a dose-dependent manner ( Fig. 1A and B ). This increase was accompanied by elevated expression of oxidative stress marker genes, specifically heme oxygenase (Hmox) and ferritin heavy chain 1 (Fth1), whereas thioredoxin (Txn) remained unchanged ( Fig. 1C ). Moreover, tunicamycin-induced ER stress upregulated the expression of catabolic proteases such as Mmp13 and Adamts5 in primary chondrocytes ( Fig. 1D ).

Figure 1.

Tunicamycin-induced ER stress leads to oxidative stress in primary chondrocytes. (A) Western blot images showing the tunicamycin-induced increase of ER stress marker proteins. Primary chondrocytes were treated with the indicated dose of tunicamycin for 24 hours and subjected to Western blot analysis. (B) The band intensity was quantified using ImageJ program, and normalized to that of β-actin. Data are presented as mean ± SD. *P < 0.05 compared to control, Mann-Whitney U test, N = 3. (C) Representative images of fluorescent staining for DHE and 8-oxo-dG. Primary chondrocytes, which were treated with tunicamycin at the indicated doses for 24 hours, were stained with dihydroethidium (DHE; red) as a fluorescent probe for ROS and immunostained with 8-oxo-dG (green) as a marker for oxidative DNA damage. Nuclei of cells were counterstained with 4’,6-diamidino-2-phenylindole (DAPI; blue). Scale bar represents 200 μm. (D) DHE and 8-oxo-dG positive cells were counted in 5 fields of view from 3 separate experiments. The counts of these cells were expressed as a percentage of the DAPI-positive cells. Data are presented as mean ± SD. *P < 0.05 compared to control, #P < 0.05, Mann-Whitney U test, N = 3. (E, F) RT-qPCR analysis of genes for oxidative stress markers, such as Hmox, Fth, and Txn, and catabolic proteases, MMP13 and Adamts5. Primary chondrocytes were treated with indicated doses of tumicamycin for 6 hours, followed by RNA extraction. RNA expression levels are presented as fold change over control cells. Data are presented as mean ± SD of 3 independent experiments. **P < 0.01 compared to control, #P < 0.05, ##P < 0.01, Mann-Whitney U test, N = 3.

We then explored the cellular origin of ROS under ER stress conditions in chondrocytes. For this, we used various inhibitors targeting different ROS systems, including TUDCA (an ER stress inhibitor), 2-deoxy-d-glucose (2-DG, a glycolysis inhibitor), IACS-010759 (an oxidative phosphorylation inhibitor), 1400W (an inhibitor of inducible nitric oxide synthase, iNOS), and APX-115 (a pan-NOX inhibitor). The treatment of TUDCA significantly reduced the number of cells positive for DHE, 8-oxo-dG, and MitoSox, suggesting that tunicamycin-induced ROS production is largely dependent on ER stress ( Fig. 2A and B ). ER stress-induced ROS production, as assessed by DHE and MitoSox, was profoundly suppressed by APX115, a pan-NOX inhibitor, with effects comparable to those of TUDCA. A similar pattern was observed in the expression of 8-oxo-dG ( Fig. 2A and B ). Moreover, the expression of oxidative stress target genes such as Fth, Txn, and Hmox, induced by tunicamycin, was most significantly suppressed by APX115, in a similar degree to that observed with TUDCA ( Fig. 2C ). The expression of MMP13 and Adamts5 was also effectively reduced by APX115 ( Fig. 2D ). These results indicate that the NOX system plays a crucial role in oxidative stress under ER stress conditions in chondrocytes.

Figure 2.

ROS production under tunicamycin-induced ER stress is primarily dependent on the NOX and the mitochondrial electron transport chain system. (A) Representative fluorescent microscopy images of DHE, 8-oxo-dG, MitoSox, and DAPI. Primary chondrocytes were treated with 2 μg/mL tunicamycin, with or without TUDCA (ER stress inhibitor), 2-DG (glycolysis inhibitor), IACS010759 (mitochondrial complex I OxPhos inhibitor), 1400 W (iNOS inhibitor), and APX-115 (pan-NOX inhibitor) for 24 hours, and then fixed and stained with DHE (red fluorescence), MitoSox (red fluorescence), and 8-oxo-dG (green). Nuclei were counterstained with DAPI (blue). The scale bar indicates 200 µm. (B) Quantitative representations of DHE, 8-oxo-dG, and MitoSox, fluorescence-positive cells, expressed as a percentage to DAPI-positive cells. Fluorescence-positive cells were counted in 5 fields of each slide from 3 separate experiments. (C, D) RT-qPCR gene expression analysis of oxidative stress marker genes such as Fth, Txn, and Hmox, and catabolic proteases such as MMP13 and Adamts5 under tunicamycin-induced ER stress in the presence of various inhibitors. Primary chondrocytes were treated with 2 μg/mL tunicamycin with various inhibitors for 6 hours. (B, C, D) Data are presented as mean ± SD, N = 3, *P < 0.05, **P < 0.01 compared to tunicamycin-treated group (red bar), #P < 0.05 and ##P < 0.01 by Mann-Whitney U test.

ER Stress Upregulates the Expression of NOX2, NOX3, NOX4, and p22phox, but Not NOX1

Based on the findings of NOX-mediated ROS production under ER stress conditions, we investigated the expression patterns of different NOX isoforms in primary chondrocytes. ER stress induced by treatment with 2 μg/mL of tunicamycin led to the upregulation of RNA levels for NOX2 (gp91phox), NOX3, NOX4, and p22phox, a shared subunit of NOX1-4. However, the expression of NOX1 remained unaffected ( Fig. 3A ). At the protein level, tunicamycin induced a dose-dependent increase in the levels of NOX2, NOX3, NOX4, and p22phox, but did not affect NOX1 expression ( Fig. 3B ).

Figure 3.

Tunicamycin-induced ER stress increases the expression of NOX2, NOX3, NOX4, and p22phox, but not NOX1. (A) RT-qPCR analysis of NOX isoforms under tunicamycin-induced ER stress. Primary chondrocytes were treated with the indicated doses of tunicamycin for 6 hours, then RNA was extracted. Data are presented as mean ± SD, N = 5, **P < 0.01, compared to control group (gray bar), ##P < 0.01 by Mann-Whitney U test. (B) Representative Western blot images of NOX isoforms and p22phox. Primary chondrocytes were treated with 0.5 and 2 µg/mL tunicamycin for 24 hours and then proteins were extracted. Data are representative of 3 experiments. Densitometric analysis of the blot images was carried out using the ImageJ program, and the results are presented as the relative ratio normalized to β-actin level. Data are presented as mean ± SD, N = 3, *P < 0.05, **P < 0.01, compared to the control group (gray bar) by Mann-Whitney U test.

The PERK Pathway Upregulates NOX4 and p22phox, while the IRE1α Pathway Upregulates NOX2 and NOX3 under ER Stress Conditions

The 3 main ER stress sensor signaling pathways—PERK, IRE1α, and ATF6—are activated under ER stress conditions. We investigated which of these pathways regulate the increase of NOX isoforms under ER stress conditions. Upon treating the cells with the IRE1α inhibitor MKC 8866 and the PERK inhibitor GSK2606414, we observed significant reduction of their respective target proteins, XBP1s for the IRE1α pathway and CHOP for the PERK pathway in Western blot analysis. The 78-kDa glucose-regulated protein (GRP78) expression, known to be regulated by the ATF6 pathway,²⁴ was significantly suppressed by both ceapins and GSK2606414 ( Fig. 4 ). These results demonstrated that the target pathways were appropriately suppressed by the drugs. In terms of NOX expression, NOX2 was suppressed by MKC8866, while NOX4 and p22phox were suppressed by GSK2606414. NOX3 was suppressed by both GSK2606414 and ceapins. However, NOX1was not affected by the ER stress sensor signaling pathways. With regard to the anabolic effect on chondrocytes, treatment with MKC 8866 and GSK2606414 completely rescued the decrease of Sox9 caused by tunicamycin-induced ER stress compared with control. However, ceapins were not able to inhibit the decrease of Sox9. The protein level of MMP13 was significantly suppressed by both GSK2606414 and ceapins, while that of Adamts5 was suppressed only by GSK2606414 under tunicamycin-induced ER stress conditions ( Fig. 4 ). These results suggest that the expression of NOX isoforms, as well as the anabolic and catabolic factors in chondrocytes, is differentially regulated by 3 ER stress sensor signaling pathways.

Figure 4.

ER stress sensor signaling pathways regulate the expression of NOX isoforms and p22phox. (A) Representative Western blot images of markers for the 3 ER stress signaling, NOX isoforms, p22phox, Sox9, MMP13, and Adamts5. Primary chondrocytes were treated with 2 µg/mL tunicamycin along with MKC8866 (an IRE1α inhibitor), GSK2606414 (a PERK inhibitor), and Ceapins (an ATF6α inhibitor) for 24 hours. Data are representative of 3 experiments. (B) Densitometric values from blot images were quantified using the ImageJ program, and the relative ratios normalized to β-actin levels are displayed as bar graphs. Data are presented as mean ± SD, N = 3, *P < 0.05, compared to the control group (gray bar), #P < 0.05 and ##P < 0.01 by Mann-Whitney U test.

NOX Inhibition Significantly Suppresses ER Stress Sensor Signaling and Chondrocyte Catabolism under ER Stress Condition

To determine the effect of NOX isoforms on oxidative stress under ER stress condition, we inhibited NOX isoforms using various inhibitors, including ML171 (NOX1 inhibitor), GSK2795039 (NOX2 inhibitor), GLX351322 (NOX4 inhibitor), GKT137831 (NOX1/4 inhibitor), and APX115 (pan-NOX inhibitor). When assessing ROS production using DHE, the pan-NOX inhibitor APX115 most effectively suppressed ROS production by tunicamycin, while inhibitors targeting NOX1, NOX2, NOX4, and NOX1/4 yielded similar, albeit lesser, levels of suppression. The expression of 8-oxo-dG was most profoundly suppressed by GSK2795039, a NOX2 inhibitor, followed by the NOX4 and pan-NOX inhibitors, both of which showed similar suppression levels of 8-oxo-dG ( Fig. 5A ).

Figure 5.

NOX-mediated oxidative stress is pivotal for ER stress sensor signaling. (A) Representative fluorescence images of DHE, 8-oxo-dG, and DAPI staining. Primary chondrocytes were treated with 2 µg/mL tunicamycin along with ML171 (NOX1 inhibitor; 10 µM), GSK2795039 (NOX2 inhibitor; 10 µM), GLX351322 (NOX4 inhibitor; 10 µM), GKT137831 (NOX1/4 inhibitor; 10 µM), or APX-115 (pan-NOX inhibitor; 10 µM) for 24 hours. The scale bar represents 200 µm. DHE and 8-oxo-dG fluorescence-positive cells were quantified in 5 fields per slide from 3 separate experiments. The numbers were expressed as a ratio to DAPI-positive cells and presented as bar graph on the right side. Data are presented as mean ± SD, N = 3, *P < 0.05, **P < 0.01 in comparison to the tunicamycin group (red bar), #P < 0.05 and ##P < 0.01 by Mann-Whitney U test. (B) Representative Western blot images of the 3 ER stress signaling markers, Sox9, MMP13, and Adamts5. Primary chondrocytes were treated with 2 µg/mL tunicamycin with or without APX-115 (10 and 50 µM) for 24 hours. Densitometric analysis was used to quantify the blot images and normalized to β-actin levels, which were presented as bar graphs. Data are shown as mean ± SD, N = 3, *P < 0.05 relative to the tunicamycin group (red bar) by Mann-Whitney U test.

We further examined the effects of NOX inhibition on the activation of ER stress sensor signaling and chondrocyte catabolism in the context of tunicamycin-induced ER stress. Treatment with the NOX2 inhibitor GSK2795039, NOX4 inhibitor GLX351322, and the pan-NOX inhibitor APX115 significantly suppressed the expression of both CHOP and XBP1s. Grp78 expression was significantly suppressed only by a higher dose (20 μM) of the pan-NOX inhibitor, but not by NOX2 or NOX4 inhibitor alone. Moreover, the pan-NOX inhibitor led to a significant increase in Sox9 expression sufficiently to counteract the decrease by interleukin (IL)-1β, but NOX2 or NOX4 inhibitor failed to increase Sox9. The IL-1β-mediated increase of MMP13 was significantly suppressed by NOX2, NOX4, and pan-NOX inhibitor. In addition, Adamts5 was significantly reduced when treated with 20 µM APX115, whereas NOX2 or NOX4 inhibitor alone failed to induce a notable change ( Fig. 5B ). These results suggest that an increase in ROS through NOX isoforms plays a crucial role in activating ER stress sensor signaling as well as anabolic and catabolic response of chondrocytes under ER stress conditions.

NOX Inhibition Attenuates the Progression of Surgically Induced OA and Reduces the Expression of ER Stress Markers in Articular Cartilage

To prove the role of NOX-mediated oxidative stress in ER stress and its catabolic function, APX115, a pan-NOX inhibitor, was administered intraperitoneally to DMM-induced mice OA model. APX115 treatment significantly reduced both the total and tibial plateau OARSI grade, as well as the subchondral bone thickness, compared with the DMM controls ( Fig. 6A ). In addition, it markedly decreased the expression of CHOP, Grp78, and MMP13 in the articular cartilage of DMM-induced OA ( Fig. 6B ). These findings indicate that mitigating oxidative stress via NOX inhibition can diminish ER stress and slow the progression of OA.

Figure 6.

Inhibition of NOX-mediated oxidative stress not only prevents ER stress in the articular cartilage but also halts the progression of surgically induced OA. (A) Representative images of Safranin-O staining. DMM surgery was performed on 12-week-old male C57BL/6 mice, followed by an intraperitoneal injection of APX-115, a pan-NOX inhibitor at 30 mg/kg twice a week for 8 weeks. Scale bar indicates 200 µm. The OARSI grades of total and medial tibial plateau, and the thickness of subchondral bone plate was quantified and expressed as bar graph. (B) Representative images of immunohistochemistry staining for MMP13, CHOP, and Grp78. The number of 3,3’-diaminobenzidine (DAB)-staining positive cells and the overall stained area were quantified semi-quantitatively using the ImageJ software. The relative count of DAP-positive cells is presented as bar graphs in the right column. (A, B) Data are presented as mean ± SD, (A) N = 7, and (B) N = 3, *P < 0.05, **P < 0.01 in comparison to the sham group (gray bar); #P < 0.05, ##P < 0.01 by Mann-Whitney U test.

Discussion

Gene signatures linked to increased oxidative stress and ER stress are frequently observed in the cartilage tissue of OA patients.^25,26 This study confirmed the importance of NOX system in the increase of ROS associated with ER stress. Each pathway of the ER stress sensor signaling specifically acted on the increase of NOX isoforms. In particular, the IRE1α pathway primarily influenced the upregulation of NOX2 and NOX3, whereas the PERK pathway was crucial for augmenting NOX4 and p22phox. The increase of the NOX system and the subsequent rise in ROS levels were pivotal in the ER stress response. Inhibiting NOX in vivo in the DMM model resulted in a marked reduction of ER stress markers in articular cartilage, thereby demonstrating a protective effect against the progression of OA.

We found that the ROS production is upregulated in chondrocytes under ER stress conditions induced by tunicamycin, which is primarily mediated by the NOX system ( Figs. 1 and 2 ). Mitochondria have long been considered the main source of cellular ROS, based on the theory that ROS production is directly correlated with metabolic rate as the by-products of mitochondrial respiration.²⁷ However, a recent study has raised doubts about the mitochondria-centered ROS theory.²⁸ They demonstrated that the contribution of mitochondria is not always the primary source of cellular ROS. Instead, the NOX system contributed the majority of ROS production in various cell types under unstressed and stressed conditions.²⁸ While it may vary depending on the cell type or the context of the stimulus environment, the NOX system can play a significant role in ROS production induced by ER stress seen in the OA chondrocytes.²⁸ Notably, our data revealed that APX115, a pan-NOX inhibitor, more effectively suppressed ROS production as well as the expression of catabolism markers such as MMP13 and Adamts5, which was comparable to that observed with TUDCA, an ER stress inhibitor ( Fig. 2 ). These findings suggest that targeting the NOX system could be a promising strategy for ER stress-related catabolic conditions in OA.

ER stress can trigger 3 distinct signaling pathways, namely PERK, IRE1α, and ATF6 pathways. These pathways sense an accumulation of unfolded proteins in the ER lumen and relay signals to the cytosol, which are then directed toward the nucleus. While ATF6 directly functions as a transcription factor, the PERK and IRE1α pathways use downstream transcription factors like ATF4 and XBP1, respectively.²⁹ Most previous studies have focused on how NOX-mediated ROS amplifies ER stress and promotes catabolic responses, such as apoptosis.^30-33 In this study, we aimed to elucidate the interaction between ER stress and the NOX system. Our most notable finding provides a detailed understanding of molecular mechanisms by which tunicamycin-induced ER stress controls the expression of NOX isoforms in chondrocytes. Specifically, NOX2 is regulated through the IRE1α pathway, NOX3 through both the IRE1α and ATF6 pathways, and NOX4 and p22phox through the PERK pathway ( Figs. 3 and 4 ). However, this finding is not consistent with a previous study in smooth muscle cells which found that NOX4 is induced by 7-ketocholesterol through an IRE1α-JNK pathway.³² This discrepancy might arise from contextual differences, such as different cell types or the stimuli inducing ER stress. Despite this discrepancy, both our findings and previous data³² imply that inhibiting a specific ER stress sensor signaling can enhance the expression of a specific isoform of NOX in catabolic conditions.

The pivotal finding of our study highlights a positive feedback loop between the NOX system and ER stress signaling. The inhibition of NOX2 and NOX4 suppresses the 2 major ER stress sensor pathways of PERK, and IRE1α, and pan-NOX inhibition suppresses all 3 major ER stress pathways ( Fig. 5 ). Among NOX isoforms, NOX4 is well known for its involvement to ER stress.^31-33 NOX4 is topologically located in the ER and mitochondrial membrane and primarily produces hydrogen peroxide (H₂O₂) rather than superoxide (O₂−).³⁴ Given that H₂O₂ is permeable and can diffuse through cell membranes, ROS derived from NOX4 is likely to affect areas close to the ER membrane, thereby affecting signaling pathway of ER membrane.^31,33 Indeed, we showed that the inhibition of NOX4 reduced the PERK and IRE1α signaling in a dose-dependent manner. Interestingly, the inhibition of NOX2 yielded similar effects on PERK and IRE1α pathways even with their different cellular location and primary superoxide production ( Fig. 5 ). Compared with NOX4, the role of NOX2 on ER stress has not been well studied. NOX2 and CHOP have been reported to form a positive feedback loop, where CHOP induces NOX2 through calcium/calmodulin-dependent protein kinase II (CaMKII) activation.³⁰ Then, NOX2-dependent oxidative stress activates CHOP and consequently induces apoptosis.³⁰ Since CHOP can be induced via both the IRE1α and PERK-eIF2α pathways,³⁵ these results support our findings regarding the IRE1α-dependent regulation of NOX2. Our results indicate that the increase of the NOX system and the subsequent rise in ROS levels were pivotal in the ER stress response, irrespective of where the ROS is produced within the cell or whether it is superoxide or H₂O₂.

Our data showed that NOX2, 3, and 4 exhibited a dose-dependent increase in response to tunicamycin, while NOX1 maintained its expression without any increase in the presence of ER stress in chondrocytes ( Fig. 3 ). NOX1 is primarily found in the cells of the colon, prostate, uterus, and vasculature, where it is involved in cell growth and differentiation by regulating ROS-dependent signal transduction.^36,37 Based on our expression pattern of NOX1, it seems that NOX1 is constitutively expressed even under ER stress conditions and is involved in ROS production. However, when treated with the NOX1 inhibitor ML171, there was no difference in 8-oxo-dG compared with the control group even though there was a reduction of ROS production ( Fig. 5A ). This suggests that NOX1 might not be a primary source of ROS in chondrocytes under ER stress.

NOX3 has been identified primarily in the vestibular and cochlear epithelia of inner ear, and the mice deficient in NOX3 exhibit a head-tilt phenotype.³⁸ However, the expression and role of NOX3 in tissues other than the inner ear are not well known. The datasheet for the commercial NOX3 antibody indicates a positive band in the mouse brain, pancreas, kidney, and liver tissues (novusbio, nox3-antibody_nbp2-94866). Our previous data revealed that primary chondrocyte from mice expresses NOX3, which is decreased by IL-1β treatment and OA articular cartilage.²⁰ Under ER stress condition examined in this study, NOX3 showed an increasing trend, which contrasts with catabolic situations such as OA or IL-1β treatment.²⁰ The function of NOX3 in cartilage physiology is still not well defined, and it warrants further in-depth research.

In conclusion, our study elucidates the interplay between ER stress and NOX-mediated oxidative stress in chondrocytes. The increase in unfolded proteins initiates ER stress, which activates 3 sensor signaling pathways, including PERK, IRE1α, and ATF6. Specifically, PERK signaling drives the expression of NOX4 and p22phox, IRE1α controls NOX2 expression, and both IRE1α and ATF6 jointly modulate NOX3. This enhanced expression of NOX isoforms further exacerbates ER stress via the amplification of oxidative stress ( Fig. 7 ). These findings highlight the pivotal interplay between ER stress and oxidative stress in OA pathogenesis. Consequently, interventions targeting NOX isoforms could offer a promising therapeutic strategy for OA.

Figure 7.

A summary of this study showing the interplay between ER stress and NOX-mediated oxidative stress in chondrocyte catabolism. ER stress by the unfolded protein response triggers 3 ER stress sensor signaling pathways. These pathways increase NOX isoforms: PERK affects NOX4 and p22phox, IRE1α affects NOX2 and NOX3, and ATF6 targets NOX3. The NOX-mediated oxidative stress, in turn, boosts ER stress, creating a feedback loop that might contribute to the deterioration of OA cartilage.

Supplemental Material

sj-pdf-1-car-10.1177_19476035241245803 – Supplemental material for The Interplay Between Endoplasmic Reticulum Stress and Oxidative Stress in Chondrocyte Catabolism

Supplemental material, sj-pdf-1-car-10.1177_19476035241245803 for The Interplay Between Endoplasmic Reticulum Stress and Oxidative Stress in Chondrocyte Catabolism by Yu Jung Kim, Jin Han and Seungwoo Han in CARTILAGE

Supplemental Material

sj-pdf-2-car-10.1177_19476035241245803 – Supplemental material for The Interplay Between Endoplasmic Reticulum Stress and Oxidative Stress in Chondrocyte Catabolism

Supplemental material, sj-pdf-2-car-10.1177_19476035241245803 for The Interplay Between Endoplasmic Reticulum Stress and Oxidative Stress in Chondrocyte Catabolism by Yu Jung Kim, Jin Han and Seungwoo Han in CARTILAGE

Footnotes

Author Contributions

Study conception and design: Y.J.K. and S.H. Acquisition of data: Y.J.K. and J.H. Analysis and/or interpretation of data: Y.J.K. and S.H. Wrote the paper: Y.J.K. and S.H. Critical revision: Y.J.K., J.H. and S.H. All authors read and approved the final manuscript.

Acknowledgments and Funding

We would like to express our sincere gratitude to Professor Ji-young Park, Pathology of Kyungpook National University, for her invaluable guidance, expertise, and unwavering support throughout this research. This research was supported by the National Research Foundation of Korea (NRF) grant funded by the Ministry of Education (grant number RS-2022-00243140).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethics Approval

All experiments were conducted in accordance with approved animal protocols and guidelines established by the Animal Care Committee of Kyungpook National University (Approval No. KNU-2018-62/54).

ORCID iD

Seungwoo Han

Data Availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Supplementary material for this article is available on the Cartilage website at .

References

Fukui

Purple

Sandell

. Cell biology of osteoarthritis: the chondrocyte’s response to injury. Curr Rheumatol Rep. 2001;3(6):496-505.

Kiani

Chen

Yee

Yang

. Structure and function of aggrecan. Cell Res. 2002;12(1):19-32.

Nelson

Dahlberg

Laverty

Reiner

Pidoux

Ionescu

, et al Evidence for altered synthesis of type II collagen in patients with osteoarthritis. J Clin Invest. 1998;102(12):2115-25.

Katsara

Attur

Ruoff

Abramson

Kolupaeva

. Increased activity of the chondrocyte translational apparatus accompanies osteoarthritic changes in human and rodent knee cartilage. Arthritis Rheumatol. 2017;69(3):586-97.

Kaufman

. Regulation of mRNA translation by protein folding in the endoplasmic reticulum. Trends Biochem Sci. 2004;29(3):152-8.

Hetz

. The unfolded protein response: controlling cell fate decisions under ER stress and beyond. Nat Rev Mol Cell Biol. 2012;13(2):89-102.

Wang

Kaufman

. Protein misfolding in the endoplasmic reticulum as a conduit to human disease. Nature. 2016;529(7586):326-35.

Bailly-Maitre

Reed

. Endoplasmic reticulum stress: cell life and death decisions. J Clin Invest. 2005;115(10):2656-64.

Meng

Jiao

Ren

Yang

Liu

, et al The endoplasmic reticulum stress induced by tunicamycin affects the viability and autophagy activity of chondrocytes. J Clin Lab Anal. 2020;34(10):e23437.

10.

Uehara

Hirose

Yamabe

Okamoto

Okada

Oyadomari

, et al Endoplasmic reticulum stress-induced apoptosis contributes to articular cartilage degeneration via C/EBP homologous protein. Osteoarthritis Cartilage. 2014;22(7):1007-17.

11.

Husa

Petursson

Lotz

Terkeltaub

Liu-Bryan

. C/EBP homologous protein drives pro-catabolic responses in chondrocytes. Arthritis Res Ther. 2013;15(6):R218.

12.

Kang

Yang

Wang

Xie

Feng

, et al Sirtuin-1 (SIRT1) stimulates growth-plate chondrogenesis by attenuating the PERK-eIF-2alpha-CHOP pathway in the unfolded protein response. J Biol Chem. 2018;293(22):8614-25.

13.

Hendershot

. The unfolding tale of the unfolded protein response. Cell. 2001;107(7):827-30.

14.

van der Vlies

Makkinje

Jansens

Braakman

Verkleij

Wirtz

, et al Oxidation of ER resident proteins upon oxidative stress: effects of altering cellular redox/antioxidant status and implications for protein maturation. Antioxid Redox Signal. 2003;5(4):381-7.

15.

Malhotra

Miao

Zhang

Wolfson

Pennathur

Pipe

, et al Antioxidants reduce endoplasmic reticulum stress and improve protein secretion. Proc Natl Acad Sci U S A. 2008;105(47):18525-30.

16.

Hasanain

Bhattacharjee

Pandey

Ashraf

Singh

Sharma

, et al Alpha-Solanine induces ROS-mediated autophagy through activation of endoplasmic reticulum stress and inhibition of Akt/mTOR pathway. Cell Death Dis. 2015;6(8):e1860.

17.

Kung

LHW

Mullan

Soul

Wang

Mori

Bateman

, et al Cartilage endoplasmic reticulum stress may influence the onset but not the progression of experimental osteoarthritis. Arthritis Res Ther. 2019;21(1):206.

18.

de Seny

Bianchi

Baiwir

Cobraiville

Collin

Deliege

, et al Proteins involved in the endoplasmic reticulum stress are modulated in synovitis of osteoarthritis, chronic pyrophosphate arthropathy and rheumatoid arthritis, and correlate with the histological inflammatory score. Sci Rep. 2020;10(1):14159.

19.

Jung

Park

Cho

Jang

Lee

Han

, et al Degrading products of chondroitin sulfate can induce hypertrophy-like changes and MMP-13/ADAMTS5 production in chondrocytes. Sci Rep. 2019;9(1):15846.

20.

Han

Park

Han

. Inhibition of NADPH oxidases prevents the development of osteoarthritis. Antioxidants (Basel). 2022;11(12):2346.

21.

Council

. Guide for the care and use of laboratory animals. 8th ed. Washington (DC): The National Academies Press; 2011. doi:10.17226/12910.

22.

Cha

Min

Kim

Ghee

Kim

, et al APX-115, a first-in-class pan-NADPH oxidase (Nox) inhibitor, protects db/db mice from renal injury. Lab Invest. 2017;97(4):419-31.

23.

Glasson

Chambers

Van Den Berg

Little

. The OARSI histopathology initiative—recommendations for histological assessments of osteoarthritis in the mouse. Osteoarthritis Cartilage. 2010;18(suppl 3):S17-23.

24.

Yoshida

Haze

Yanagi

Yura

Mori

. Identification of the cis-acting endoplasmic reticulum stress response element responsible for transcriptional induction of mammalian glucose-regulated proteins. Involvement of basic leucine zipper transcription factors. J Biol Chem. 1998;273(50):33741-9.

25.

Ruiz-Romero

Carreira

Rego

Remeseiro

López-Armada

Blanco

. Proteomic analysis of human osteoarthritic chondrocytes reveals protein changes in stress and glycolysis. Proteomics. 2008;8(3):495-507.

26.

Poulet

Beier

. Targeting oxidative stress to reduce osteoarthritis. Arthritis Res Ther. 2016;18:32.

27.

Turrens

. Mitochondrial formation of reactive oxygen species. J Physiol. 2003;552(pt 2):335-44.

28.

Zhang

Wong

. Are mitochondria the main contributor of reactive oxygen species in cells? J Exp Biol. 2021;224(pt 5):jeb221606.

29.

Walter

Ron

. The unfolded protein response: from stress pathway to homeostatic regulation. Science. 2011;334(6059):1081-6.

30.

Scull

Ozcan

Tabas

. NADPH oxidase links endoplasmic reticulum stress, oxidative stress, and PKR activation to induce apoptosis. J Cell Biol. 2010;191(6):1113-25.

31.

Chen

Kirber

Xiao

Yang

Keaney

Jr.

Regulation of ROS signal transduction by NADPH oxidase 4 localization. J Cell Biol. 2008;181(7):1129-39.

32.

Pedruzzi

Guichard

Ollivier

Driss

Fay

Prunet

, et al NAD(P)H oxidase Nox-4 mediates 7-ketocholesterol-induced endoplasmic reticulum stress and apoptosis in human aortic smooth muscle cells. Mol Cell Biol. 2004;24(24):10703-17.

33.

Liu

Terada

. Nox4-derived H2O2 mediates endoplasmic reticulum signaling through local Ras activation. Mol Cell Biol. 2010;30(14):3553-68.

34.

Nisimoto

Diebold

Cosentino-Gomes

Lambeth

. Nox4: a hydrogen peroxide-generating oxygen sensor. Biochemistry. 2014;53(31):5111-20.

35.

Liu

Shi

Luan

Wei

, et al ROR2 induces cell apoptosis via activating IRE1alpha/JNK/CHOP pathway in high-grade serous ovarian carcinoma in vitro and in vivo. J Transl Med. 2019;17(1):428.

36.

Suh

Arnold

Lassegue

Shi

Sorescu

, et al Cell transformation by the superoxide-generating oxidase Mox1. Nature. 1999;401(6748):79-82.

37.

Banfi

Maturana

Jaconi

Arnaudeau

Laforge

Sinha

, et al A mammalian H+ channel generated through alternative splicing of the NADPH oxidase homolog NOH-1. Science. 2000;287(5450):138-42.

38.

Paffenholz

Bergstrom

Pasutto

Wabnitz

Munroe

Jagla

, et al Vestibular defects in head-tilt mice result from mutations in Nox3, encoding an NADPH oxidase. Genes Dev. 2004;18(5):486-91.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.13 MB

0.00 MB

1.33 MB