Oxidative stress-mediated mitochondrial dysfunction facilitates mesenchymal stem cell senescence in ankylosing spondylitis

Abstract

Ankylosing spondylitis (AS) is a chronic inflammatory disease possessing a morbid serum microenvironment with enhanced oxidative stress. Long-term exposure to an oxidative environment usually results in cellular senescence alone with cellular dysfunction. Mesenchymal stem cells (MSCs) are a kind of stem cell possessing strong capabilities for immunoregulation, and senescent MSCs may increase inflammation and participate in AS pathogenesis. The objective of this study was to explore whether and how the oxidative serum environment of AS induces MSC senescence. Here, we found that AS serum facilitated senescence of MSCs in vitro, and articular tissues from AS patients exhibited higher expression levels of the cell cycle arrest-related proteins p53, p21 and p16. Importantly, the levels of advanced oxidative protein products (AOPPs), markers of oxidative stress, were increased in AS serum and positively correlated with the extent of MSC senescence induced by AS serum. Furthermore, MSCs cultured with AS serum showed decreased mitochondrial membrane potential and ATP production together with a reduced oxygen consumption rate. Finally, we discovered that AS serum-induced mitochondrial dysfunction resulted in elevated reactive oxygen species (ROS) in MSCs, and ROS inhibition successfully rescued MSCs from senescence. In conclusion, our data demonstrated that the oxidative serum environment of AS facilitated MSC senescence through inducing mitochondrial dysfunction and excessive ROS production. These results may help elucidate the pathogenesis of AS and provide potential targets for AS treatment.

Fig. 1. Compared with HD serum, AS serum facilitated MSCs senescence.

a MSCs treated with AS serum displayed stronger SA-β-gal staining (black arrow) than those treated with HD serum (scale bar = 50 µm). b The percentages of SA-β-gal-positive cells among treated MSCs were 38.23 ± 7.47% and 15.66 ± 5.16% for AS serum and HD serum, respectively; n = 20, p < 0.001. c, d The expression of the cell cycle arrest-related proteins p53, p21, and p16 was detected by Western blotting, and the levels were higher in MSCs treated with AS serum than in those treated with HD serum. e The signal for p-H2A.X measured by immunofluorescence was more obvious in MSCs treated with AS serum than in those treated with HD serum (scale bar = 10 µm). The results are presented as the means ± SD (n = 20, determined by independent-sample t tests). All experiments were performed three independent times, *p < 0.05, **p < 0.01.

Fig. 2. MSCs located in the articular soft tissue of AS patients displayed remarkable characteristics of senescence.

a Tissues from AS patients displayed stronger SA-β-gal staining (black arrow) than those from non-AS donors (scale bar = 100 µm). b–d Immunofluorescence detection indicated increased levels of p53 (b), p21 (c), and p16 (d) in MSCs in articular soft tissue samples from AS patients than in those from non-AS patients (scale bar = 40 µm). All experiments were performed three independent times (n = 4).

Fig. 3. AS serum contained elevated AOPPs levels and could stimulate ROS production in MSCs.

a The levels of AOPPs were higher in AS serum (108.0 ± 11.69 µmol/L) than in HD serum (44.63 ± 9.41 µmol/L); p < 0.001, n = 20. b–e The levels of AOPPs in AS serum were positively correlated with the level of CRP (r = 0.6055, p < 0.01) (b), BASDAI scores (r = 0.5635, p < 0.01) (c), the percentage of SA-β-gal-positive cells after treatment with AS serum (r = 0.7148, p < 0.01) (d), and the expression of the protein p53 in MSCs treated with AS serum (r = 0.65458, p < 0.01) (e–g). Flow cytometry detection revealed that the total ROS level was enhanced in MSCs treated with AS serum compared with those treated with HD serum. h AS serum-treated MSCs displayed a stronger signal for ROS than HD serum-treated MSCs under a fluorescence microscope (scale bar = 20 µm). The results in a, f, and g are presented as the means ± SD (n = 20, determined by independent-sample t tests). The data in b–e were determined by Pearson correlation and linear regression analysis (n = 20). All experiments were performed three independent times, *p < 0.05, **p < 0.01.

Fig. 4. ROS inhibition via NAC rescued MSCs from the senescence caused by AS serum.

a, b NAC reduced the ROS level in AS serum-treated MSCs to the level in HD serum-treated MSCs. c, d NAC rescued the extent of SA-β-gal staining of MSCs (black arrow) induced by AS serum (scale bar = 50 µm). e, f NAC decreased the expression levels of the cell cycle arrest-related proteins p53, p21, and p16 in AS serum-treated MSCs to the levels in HD serum-treated MSCs. The results are presented as the means ± SD (n = 10, determined by independent-sample t tests). All experiments were performed three independent times, *p < 0.05, **p < 0.01.

Fig. 5. AS serum-mediated mitochondrial dysfunction in MSCs.

a The mitochondrial membrane potential detected by the probe JC-1 showed reduced fluorescence of JC-1 polymers and elevated fluorescence of JC-1 monomers in MSCs treated with AS serum, which represented a reduced ∆Ψm in the AS serum-treated MSCs (scale bar = 10 µm). b The fluorescence index analyzed by ImageJ showed that the ratio of JC-1 monomers (green) to polymers (red) was higher in MSCs treated with AS serum than in those treated with HD serum, which implied a reduced ∆Ψm in the AS serum-treated MSCs. c ATP detection showed that AS serum treatment obviously reduced the ATP production of MSCs. d, e Mitochondrial oxygen consumption rate detection revealed that compared to HD serum, AS serum impaired the basal respiration, spare respiration and ATP production of MSCs. The results are presented as the means ± SD (n = 10, determined by independent-sample t tests). All experiments were performed three independent times, *p < 0.05, **p < 0.01.

Fig. 6. Mitoquinone decreased the ROS level and reversed the senescence of MSCs caused by AS serum.

a Mitochondrial ROS evaluation with MitoSOX showed that AS serum increased the mitochondrial ROS levels of MSCs and that MQ pretreatment eliminated this effect (scale bar = 40 µm). b, c The total ROS level detected by the probe DCFH-DA showed that MQ reduced the total ROS level of AS serum-treated MSCs to the level of HD serum-treated MSCs. d, e The SA-β-gal staining assay indicated that MQ decreased the percentage of SA-β-gal-positive MSCs (black arrow) treated with AS serum (scale bar = 50 µm). f, g Western blot analysis revealed that MQ reversed the expression of the cell cycle arrest-related proteins p53, p21, and p16 in MSCs treated with AS serum. The results are presented as the means ± SD (n = 10, determined by independent-sample t tests). All experiments were performed three independent times, *p < 0.05, **p < 0.01.