H2O2-induced mitochondrial fragmentation in C2C12 myocytes

Abstract

In skeletal muscle and many other cell types, mitochondria exist as an elaborate and dynamic network in which "individual" mitochondria exist only transiently even under nonstimulated conditions. The balance of continuous mitochondrial fission and fusion defines the morphology of the mitochondrial reticulum. Environmental stimuli, such as oxidative stress, can influence fusion and fission rates, resulting in a transformation of the network's connectivity. Using confocal laser scanning microscopy of C(2)C(12) mouse myocytes, we show that acute exposure to the reactive oxygen species (ROS) hydrogen peroxide (H(2)O(2)) induces a slow fragmentation of the mitochondrial reticulum that is reversible over 24h. Although H(2)O(2) decomposes rapidly in culture medium, the full extent of fragmentation occurs 5-6h posttreatment, suggesting that H(2)O(2) affects mitochondrial morphology by modulating cellular physiology. Supraphysiological (>1 mM) concentrations of H(2)O(2) are cytotoxic, but lower concentrations (250 μM) sufficient to induce transient fragmentation do not lower cell viability. H(2)O(2)-induced mitochondrial fragmentation is preceded by decreases in inner mitochondrial membrane potential and maximal respiratory rate, suggesting a possible mechanism. Because H(2)O(2) is produced in contracting muscle, our results raise the possibility that ROS generation may contribute to exercise-induced changes in mitochondrial morphology in vivo.

Figure 1. H₂O₂-induced mitochondrial network fragmentation is reversible

C₂C₁₂ cells were treated with 250 μM H₂O₂ or 0 μM H₂O₂ (control) for 6 hours, then stained with MitoTracker Deep Red 633 for 30 minutes and visualized by CLSM immediately, or after a 24- or 48-hour incubation. For each image taken, the total number of cells and the number of cells displaying a reticular mitochondrial network were counted. A, representative images for each of the conditions. B, the percentage (mean ± SEM) of total imaged cells displaying a reticular mitochondrial network under each condition. An average of 235 cells was scored per condition. *Significantly different from all other bars, P < 0.001. Six-hour treatment with 250 μM H₂O₂ significantly decreased the percentage of cells having a reticular mitochondrial network, but the cells recovered within 24 hours.

Figure 2. Dose-dependence, cytotoxicity assessment, and time-dependence of H₂O₂-induced mitochondrial network fragmentation

A, C₂C₁₂ cells were treated with 0, 62.5, 125, 250, and 500 μM H₂O₂ for 3 hours, stained with MitoTracker Deep Red, visualized by CLSM, and categorized by network morphology. The percent of cells in the “fragmented” category is shown for each condition. Mock-treated (0 μM H₂O₂) myocytes showed a predominantly reticular network morphology, which became increasingly fragmented as H₂O₂ dose was increased. An average of 52 cells was scored per condition. Data are presented as percents of the total number of cells scored in each treatment group. Data from 2 independent experiments are shown. B, myocytes were treated with 0, 125, 250, 500, 1000, or 2000 μM H₂O₂ for 6 hours and assayed for cell viability with WST-8 immediately (triangles), or washed and returned to H₂O₂-free medium, and assayed 24 hours later (squares). Data are presented as means ± SD of 5-6 replicates. Notably, the H₂O₂ concentration used in subsequent experiments (250 μM) did not significantly lower cell viability. Student's t-test was used to test for statistically significant differences between each condition and the respective 0 μM H₂O₂ control. NS, not significantly different from mock-treated (0 μM H₂O₂); *P < 0.05; **P = 0.001; ***P < 10^-7. C, cells were incubated in medium II initially containing 250 μM H₂O₂ for 5 minutes or 1 to 6 hours, then stained with MitoTracker for 30 minutes and visualized by CLSM. The percent of cells in the “fragmented” category is shown for each time point. The occurrence of fragmented mitochondria appears to peak at 5.5 hours. An average of 51 cells was scored per time point. Data from 2 independent experiments are shown.

Figure 3. Mean hydrogen peroxide concentration in cell culture supernatant as a function of time

At time zero, 250 μM H₂O₂ was added to dishes with (solid diamond) or without (open triangle) C₂C₁₂ cells seeded. In a mock experiment (open square), H₂O₂-free medium was added to a dish of cells. Data are presented as means ± SD of 3 replicate dishes. Inset, the same data is plotted, but with the natural logarithm of [H₂O₂] on the ordinate axis. The disappearance of H₂O₂ followed first-order kinetics (t_1/2 = 6.3 min in the presence of C₂C₁₂ cells; 9.0 min in their absence).

Figure 4. 250 μM H₂O₂ pretreatment decreases JC-1 loading

C₂C₁₂ cells were pretreated with 250 μM H₂O₂ or 0 μM H₂O₂ (“mock”) for 1 or 5 hours, or were not pretreated (“untreated”). Subsequently, cells were stained in 2 μM JC-1 for 30 minutes, trypsinized, and analyzed by flow cytometry. Some “untreated” cells were treated with 10 μM CCCP before flow cytometry (denoted “CCCP”). A, two-parameter (log scale) pseudo-colored dot plots for fluorescence detected at 525 nm and 575 nm. B, the data from A shown as superimposed single-channel histograms. Notably, H₂O₂ pretreatment (whether for 1 hour or 5 hours) causes a marked shift toward lower 575 nm fluorescence. In A and B, data from 1 representative experiment of 3 are shown. C, quantification of JC-1 fluorescence at 575 nm (J-aggregates). The median fluorescence intensity of each single-channel histogram was expressed as a percentage of the “untreated” control of that experiment. Mean percentages ± SEM of 3 independent experiments are shown. *P < 0.01; **P < 0.05. One- and five-hour pretreatments with 250 μM H₂O₂ appear to have similar effects in decreasing median 575 nm fluorescence intensity.

Figure 5. H₂O₂-treated cells show depressed uncoupled-to-endogenous respiration ratios 1, 5, or 25 hours post-treatment

Representative respiration traces of intact C₂C₁₂ myocytes, measured using a Clark-type oxygen electrode, are shown. Untreated, mock-pretreated, or H₂O₂-pretreated cells were trypsinized and added (triangle) to temperature-equilibrated medium, and endogenous respiration was recorded for 15 minutes. Then, CCCP (final concentration 10 μM) was added (diamond) and uncoupled respiration was recorded for at least 10 minutes. The values of the slopes of the endogenous and uncoupled respiration traces are shown. Inset, the uncoupled respiration slope of each trace was divided by the endogenous respiration slope of that trace. Data are shown as means ± SEM of at least 6 independent experiments each. *P ≤ 0.01. The partial recovery seen 25 hours post-treatment suggests that the H₂O₂-induced depression was transient.