Strain-dependent oxidant release in articular cartilage originates from mitochondria

Abstract

Mechanical loading is essential for articular cartilage homeostasis and plays a central role in the cartilage pathology, yet the mechanotransduction processes that underlie these effects remain unclear. Previously, we showed that lethal amounts of reactive oxygen species (ROS) were liberated from the mitochondria in response to mechanical insult and that chondrocyte deformation may be a source of ROS. To this end, we hypothesized that mechanically induced mitochondrial ROS is related to the magnitude of cartilage deformation. To test this, we measured axial tissue strains in cartilage explants subjected to semi-confined compressive stresses of 0, 0.05, 0.1, 0.25, 0.5, or 1.0 MPa. The presence of ROS was then determined by confocal imaging with dihydroethidium, an oxidant sensitive fluorescent probe. Our results indicated that ROS levels increased linearly relative to the magnitude of axial strains (r(2) = 0.87, p < 0.05), and significant cell death was observed at strains >40%. By contrast, hydrostatic stress, which causes minimal tissue strain, had no significant effect. Cell-permeable superoxide dismutase mimetic Mn(III)tetrakis (1-methyl-4-pyridyl) porphyrin pentachloride significantly decreased ROS levels at 0.5 and 0.25 MPa. Electron transport chain inhibitor, rotenone, and cytoskeletal inhibitor, cytochalasin B, significantly decreased ROS levels at 0.25 MPa. Our findings strongly suggest that ROS and mitochondrial oxidants contribute to cartilage mechanobiology.

Figure 1

Mechanical loading devices. (a) Device for static compressive stress shown in a low oxygen incubator. The inset shows an osteochondral explant submerged in culture medium in the housing under the indenter. (b) Hydrostatic device shown in a water bath. The inset shows an explant sealed in a plastic bag containing culture medium. The bag is submerged under water in the unit.

Figure 2

Representative z-axis stacked confocal images (20× objective) showing oxidant production (Red) and live cells (Green) after static compressive stress with 0 MPa (a), 0.05MPa (b), 0.1 MPa (c), 0.25 MPa (d), 0.5 MPa (e), and 1.0 MPa (f). The white bar indicates 40 microns.

Figure 3

Percentage of cells stained with oxidative marker, dihydroethidium (DHE), and cell death marker, EthD-2 (DEATH), under various static stresses. Statistical significance markers are denoted (p<0.05) as α indicating significance vs. unloaded control, β vs. 0.05MPa, γ vs. 0.1MPa, δ vs. 0.25MPa, and ε vs. 0.5MPa for compressive stress. For hydrostatic stress, ω represents significance compared to the corresponding compressive stress and stain.

Figure 4

DHE staining and bulk tissue strain after stress completion (a). Regression analysis revealed a strong linear relationship between DHE and strain. Cell death and bulk tissue strain (b). Each point represents the mean percentage from the three image sites of each explant.

Figure 5

Effects of MnTMPyP, rotenone, and cytochalasin B on stress-induced DHE staining. Sequestering superoxide produced significant reductions in DHE staining at both 0.5MPa and 0.25MPa (p<0.05). Inhibition of mitochondrial electron transport with rotenone decreased DHE staining after 1 hour of 0.25MPa static compressive stress (p<0.05). Cytoskeletal dissolution with cytochalasin B significantly decreased DHE staining at 0.25MPa (p<0.05). Statistical significance markers (p<0.05) are denoted as α indicating significance vs. 0.25MPa untreated specimens, β vs. 0.5MPa untreated specimens, γ vs. 0.5MPa rotenone treatment, and δ vs. 0.5MPa cytochalasin B treated.