Glutathione as a mediator of cartilage oxidative stress resistance and resilience during aging and osteoarthritis

Abstract

Purpose: An underlying cause of osteoarthritis (OA) is the inability of chondrocytes to maintain homeostasis in response to changing stress conditions. The purpose of this article was to review and experimentally evaluate oxidative stress resistance and resilience concepts in cartilage using glutathione redox homeostasis as an example. This framework may help identify novel approaches for promoting chondrocyte homeostasis during aging and obesity.Materials and Methods: Changes in glutathione content and redox ratio were evaluated in three models of chondrocyte stress: (1) age- and tissue-specific changes in joint tissues of 10 and 30-month old F344BN rats, including ex vivo patella culture experiments to evaluate N-acetylcysteine dependent resistance to interleukin-1beta; (2) effect of different durations and patterns of cyclic compressive loading in bovine cartilage on glutathione stress resistance and resilience pathways; (3) time-dependent changes in GSH:GSSG in primary chondrocytes from wild-type and Sirt3 deficient mice challenged with the pro-oxidant menadione.Results: Glutathione was more abundant in cartilage than meniscus or infrapatellar fat pad, although cartilage was also more susceptible to age-related glutathione oxidation. Glutathione redox homeostasis was sensitive to the duration of compressive loading such that load-induced oxidation required unloaded periods to recover and increase total antioxidant capacity. Exposure to a pro-oxidant stress enhanced stress resistance by increasing glutathione content and GSH:GSSG ratio, especially in Sirt3 deficient cells. However, the rate of recovery, a marker of resilience, was delayed without Sirt3.Conclusions: OA-related models of cartilage stress reveal multiple mechanisms by which glutathione provides oxidative stress resistance and resilience.

Keywords: Osteoarthritis; cartilage; glutathione; oxidative stress; resilience; stress resistance.

Figure 1.. Extrinsic and intrinsic factors associated with increased risk of OA.

(A) Recent work by Wallace and colleagues (5) implicates numerous extrinsic behavioral and systemic biological factors that are common in our modern environment and may contribute to an observed increase in the prevalence of knee OA, even when adjusting differences in lifespan and obesity compared to pre-industrial human populations. Figure modified from (5). (B). Many ancestral environmental conditions are associated with intrinsic stress-reducing cellular and molecular processes; whereas, modern environmental factors may increase OA risk by stimulating intrinsic stress-inducing processes in chondrocytes.

Figure 2.. Resistance and resilience to oxidative stress: conceptual framework and example with glutathione.

(A). Under pro-oxidant producing stress conditions, such as inflammation and biomechanical loading, cellular oxidation increases. During homeostasis, the cellular redox potential returns to baseline following the removal of the stress. However, under conditions of impaired oxidative stress resistance or resilience, the cellular redox potential does not recover to baseline levels. Figure modified from reference (16). (B) Glutathione is a major regulator of cellular redox potential and oxidative damage. Glutathione exists in either a reduced (GSH) or oxidized (GSSG) form. Multiple pathways determine glutathione synthesis and redox balance. Deficits in these pathways may lead to impaired oxidative stress resistance or resilience.

Figure 3.. Experimental models to evaluate the role of glutathione in oxidative stress resistance and resilience in cartilage.

Model 1 compares multiple joint tissues from young and old rats to better understand how cartilage compares to other joint tissues in terms of whole-joint age-related changes in glutathione content and redox ratio. An ex vivo patella culture model was used to evaluate acute pro-inflammatory and anti-oxidant treatment effects on glutathione stress resistance. Model 2 compared different durations and "on - off" patterns of cyclic loading of cartilage explants to determine how temporal variables contribute to glutathione stress resistance and resilience concepts. Model 3 evaluated time-dependent changes in oxidized and reduced glutathione content in primary murine chondrocytes challenged with the pro-oxidant chemical menadione. The model compared wild type (WT) versus sirtuin 3 null (Sirt3 KO) mouse chondrocytes as an age-related feature because Sirt3 decreases with age in cartilage. We hypothesized that these three models would reveal multiple mechanisms by which stress resistance and stress resilience contribute to cartilage redox regulation and cellular homeostasis.

Figure 4.. Effect of aging on glutathione redox balance and content in rat joint tissues.

(A) Comparison of reduced (GSH), oxidized (GSSG), and total (GSH+GSSG) glutathione content normalized to protein in knee joint tissues from adult and aged F344BN F1 hybrid rats (left Y-axis). Ratio of reduced:oxidized glutathione content in knee joint tissues of adult and aged rats (right Y-axis). Values are mean ± sem, n=3 per age. *p<0.05, **p<0.01; unpaired, two-tailed t-test (Prism 8). (B) The same glutathione measurements in patella explant cultures treated with interleukin-1beta (IL-1b) and/or N-acetylcysteine (NAC) for 24 hrs. Values are mean ± sem, n=3 per age. *p<0.05, **p<0.01, ***p<0.001 for age or treatment comparisons; #p<0.05, ##p<0.01 for within age treatment effects versus control; 2-way ANOVA with Tukey's Multiple Comparisons test (Prism 8).

Figure 5.. Effect of continuous versus intermittent bouts of compressive cyclic loading on endogenous cartilage antioxidants.

(A) Healthy bovine cartilage explants were compressed with a 50 kPa 1Hz cyclic loading pattern continuously or in 12-hr "on-off" bouts for up to 60 hrs as indicated along the X-axis. Reduced (GSH), oxidized (GSSG), and total (GSH+GSSG) glutathione content and the GSH/GSSG ratio were plotted as percentages of unloaded site and animal-matched control explants. Values are mean ± sd, n=4-8 per loading group. #p=0.052, *p<0.05, **p<0.01, ***p<0.001 versus unloaded controls (one-sample t-test) or between two groups as indicated by bars (1-way ANOVA with Tukey's Multiple Comparisons test) (Prism 8). (B). Total antioxidant activity and catalase activity from cartilage explants harvested from the same joints and loading experiments as presented in panel A. Values are mean ± sd, n=4-8 per loading group. *p<0.05, **p<0.01, versus unloaded control value (one-sample t-test) (Prism 8).

Figure 6.. Effect of Menadione Pro-oxidant Challenge on Glutathione Redox Homeostasis in Wild-type and Sirt3 KO Primary Mouse Chondrocytes.

(A) Menadione treatment concentration was optimized by evaluating cell adherence and viability following a range of menadione treatments for 2 hrs. Cell adherence was reduced in Sirt3 KO cells, and the viability of adherent cells was reduced in WT and KO cells at ≥25 μM menadione. Values are mean ± sd in n=5 independent biological replicates per genotype (*p<0.05, **p<0.01; Dunnett’s posthoc test versus untreated condition, 2-way repeated-measures anova). (B) Comparison of reduced:oxidized glutathione ratio and protein normalized glutathione content in WT and Sirt3 KO cells under basal culture conditions. Values are mean ± sd for n=5 independent biological replicates. Statistical analysis used an unpaired two-tailed t-test with Welch’s correction. (C) GSH/GSSG ratio and GSH, GSSG and total glutathione levels normalized to average genotype-specific untreated value. Treated samples were collected immediately after 30 min menadione treatment. Additional time points indicate the duration (hrs) since menadione was removed. For data points where GSSG was not detected, the lowest limit of detection value was imputed for statistical purposes. Statistical analyses were evaluated by two-way ANOVA, with Dunnett’s posthoc test used to compare the effects of treatment (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001) and genotype (#p<0.05, ##p<0.01, ###p<0.001, ####p<0.0001). Values are mean ± sd for n=3 independent biological replicates per genotype with the following exceptions where n=2 due to a technical issue with sample processing (KO untreated, WT treated, KO treated). The same cell sources were used for all treatment conditions.