Irisin Mitigates Oxidative Stress, Chondrocyte Dysfunction and Osteoarthritis Development through Regulating Mitochondrial Integrity and Autophagy

Abstract

Compromised autophagy and mitochondrial dysfunction downregulate chondrocytic activity, accelerating the development of osteoarthritis (OA). Irisin, a cleaved form of fibronectin type III domain containing 5 (FNDC5), regulates bone turnover and muscle homeostasis. Little is known about the effect of Irisin on chondrocytes and the development of osteoarthritis. This study revealed that human osteoarthritic articular chondrocytes express decreased level of FNDC5 and autophagosome marker LC3-II but upregulated levels of oxidative DNA damage marker 8-hydroxydeoxyguanosine (8-OHdG) and apoptosis. Intra-articular administration of Irisin further alleviated symptoms of medial meniscus destabilization, like cartilage erosion and synovitis, while improved the gait profiles of the injured legs. Irisin treatment upregulated autophagy, 8-OHdG and apoptosis in chondrocytes of the injured cartilage. In vitro, Irisin improved IL-1β-mediated growth inhibition, loss of specific cartilage markers and glycosaminoglycan production by chondrocytes. Irisin also reversed Sirt3 and UCP-1 pathways, thereby improving mitochondrial membrane potential, ATP production, and catalase to attenuated IL-1β-mediated reactive oxygen radical production, mitochondrial fusion, mitophagy, and autophagosome formation. Taken together, FNDC5 loss in chondrocytes is correlated with human knee OA. Irisin repressed inflammation-mediated oxidative stress and extracellular matrix underproduction through retaining mitochondrial biogenesis, dynamics and autophagic program. Our analyses shed new light on the chondroprotective actions of this myokine, and highlight the remedial effects of Irisin on OA development.

Keywords: FNDC5; autophagy; chondrocyte; irisin; mitochondria; osteoarthritis.

Figure 1

Histological analyses of articular cartilage in human end-stage knee OA. Erosion and fragmentation were present in osteoarthritic cartilage together with significant increases in OARSI scores (a); scale bar, 40 μm. Increased chondrocytes showed 8-OHdG immunostaining (b) and TUNEL staining (c) (scale bar, 20 μm), whereas few chondrocytes displayed LC3 (d) and FNDC5 immunostaining (e); scale bar, 20 μm. Articular cartilage damage was quantified using OARSI scales. TUNEL, LC3-II and FNDC5 were probed using immunohistochemistry. Data are mean ± standard errors calculated from 11 patients. Asterisks * indicate significant differences from the non-OA group. OA, osteoarthritic cartilage; non-OA, healthy cartilage.

Figure 2

Analyses of Irisin actions to OA development and gait profiles of DMM-injured knee joints. Purified recombinant proteins showed Irisin immunoreactivity (a). Schematic drawing for intra-articular injection of Irisin recombinant protein in DMM-injured knee joints (b). Irisin treatment compromised cartilage destruction and synovitis in DMM-injured knees (c) (upper panel scale bar, 160 μm; middle and lower panels scale bar, 40 μm), as well as reduced OARSI and synovitis scores (d). Irisin improved the DMM-induced loss of FNDC5, Sox9 and collagen II in injured knee joints (e). DMM-mediated irregular footprint histograms were reversed upon Irisin treatment (f). Irisin attenuated DMM-induced dysregulated footprint area, maximum foot contact, duty cycle and swing speed (g). Joint damage, chondrocyte markers and gait profiles were probed using histomorphometry, RT-qPCR and the Catwalk system. Data are mean ± standard errors calculated from 5 mice. Asterisks * indicate significant differences between groups.

Figure 3

Immunohistochemical analyses of articular cartilage in sham and DMM-injured knees. Few chondrocytes showed FNDC5 (a), LC3 (b) or PCNA (c) immunostaining in DMM-injured cartilage (scale bar, 20 μm), whereas increased chondrocytes displayed TUNEL staining (d); scale bar, 20 μm. FNDC5, LC3 and PCNA immunoreactions were reversed and TUNEL staining was downregulated in Irisin-treated knee joints. Data are mean ± standard errors calculated from 5 mice. Asterisks * indicate significant differences between groups.

Figure 4

Analyses of growth and ECM production of IL-1β and Irisin-treated chondrocytes. Alcian blue-stained glycosaminoglycan production in IL-1β and Irisin-treated chondrocyte micromass cultures (a); scale bars, 5 mm. Irisin dose-dependently improved ECM production of inflamed chondrocyte micromass (b). Irisin reversed IL-1β-mediated loss of FNDC5 (c), cell growth (d), collagen II, aggrecan and Sox9 expression (e), as well as repressing MMP9 and VEGF (f) expression. Glycosaminoglycan production, cell growth and chondrocyte markers were probed using Alcian blue staining, WST-1 uptake, and RT-qPCR. Data are mean ± standard errors calculated from 4–6 experiments. Asterisks * indicate significant differences between groups.

Figure 5

Analyses of autophagy and apoptosis in IL-1β and Irisin-treated chondrocytes. Irisin repressed IL-1β-mediated loss of Atg4, Atg12 and p62 (a), as well as improving LC3-II conversion (b). Irisin improved the IL-1β-induced loss of monodansylcadaverin-stained autophagic puncta formation (c) (scale bar, 10 μm) and compromised apoptosis (d) (scale bar, 30 μm) in chondrocytes. The autophagic markers, autophagic puncta and apoptosis in chondrocytes were probed using RT-qPCR, laser confocal microscopy and TUNEL staining. Data are mean ± standard errors calculated from 4–5 experiments. Asterisks * indicate significant differences between groups.

Figure 6

Analyses of mitochondrial dynamics and mitophagy in chondrocytes. Irisin reversed the IL-1β-mediated suppression of Mfn1, Drp1 (a), MitoSoxRed-stained morphology (b) (scale bar, 10 μm) and mitochondrial fusion (c), as well as improving PINK1, Parkin (d), Mitophagy dye-stained mitophagic puncta (e) and mitophagosome formation (f); scale bar, 10 μm. Mitochondrial dynamics markers, mitochondrial morphology, mitophagy markers and mitophagosome were probed using RT-qPCR, fluorescence MitoSoxRed and fluorescence Mitophagy dye together with Lyso dye. Data are mean ± standard errors calculated from 4–5 experiments. Asterisks * indicate significant differences between groups.

Figure 7

Analyses of mitochondrial regulators and antioxidants in chondrocytes. Irisin attenuated IL-1β-induced loss of PGC-1α and Tfam expression (a), ATP production (b), membrane potential depolarization (c) and reactive oxygen species production (d), as well as reversing UCP-1, Sirt3 and catalase protein expression (e). Irisin improved Sirt3 and UCP-1 (f), whereas 8-OHdG (g) immunostaining was increased in articular chondrocytes in DMM-injured knees (8 scale bar, 20 μm). Mitochondrial markers, ATP production, membrane potential and reactive oxygen species were probed using RT-qPCR, Mitochondrial ATP kits, JC-1 probe. Data are mean ± standard errors calculated from 4–5 experiments. Asterisks * indicate significant differences between groups.