Mitochondrial superoxide flashes: metabolic biomarkers of skeletal muscle activity and disease

Abstract

Mitochondrial superoxide flashes (mSOFs) are stochastic events of quantal mitochondrial superoxide generation. Here, we used flexor digitorum brevis muscle fibers from transgenic mice with muscle-specific expression of a novel mitochondrial-targeted superoxide biosensor (mt-cpYFP) to characterize mSOF activity in skeletal muscle at rest, following intense activity, and under pathological conditions. Results demonstrate that mSOF activity in muscle depended on electron transport chain and adenine nucleotide translocase functionality, but it was independent of cyclophilin-D-mediated mitochondrial permeability transition pore activity. The diverse spatial dimensions of individual mSOF events were found to reflect a complex underlying morphology of the mitochondrial network, as examined by electron microscopy. Muscle activity regulated mSOF activity in a biphasic manner. Specifically, mSOF frequency was significantly increased following brief tetanic stimulation (18.1 ± 1.6 to 22.3 ± 2.0 flashes/1000 μm²·100 s before and after 5 tetani) and markedly decreased (to 7.7 ± 1.6 flashes/1000 μm²·100 s) following prolonged tetanic stimulation (40 tetani). A significant temperature-dependent increase in mSOF frequency (11.9 ± 0.8 and 19.8 ± 2.6 flashes/1000 μm²·100 s at 23°C and 37°C) was observed in fibers from RYR1(Y522S/WT) mice, a mouse model of malignant hyperthermia and heat-induced hypermetabolism. Together, these results demonstrate that mSOF activity is a highly sensitive biomarker of mitochondrial respiration and the cellular metabolic state of muscle during physiological activity and pathological oxidative stress

Figure 1.

Properties and spatial morphologies of mSOF activity in FDB fibers from mt-cpYFP transgenic mice. A) Representative 2-photon image of mt-cpYFP fluorescence of an FDB fiber obtained from a muscle specific mt-cpYFP transgenic mouse. Boxed area indicates a region of a rod-shaped flash that was maximal during acquisition of the displayed x,y image. B) Time course of the rod-shaped flash even in the boxed area shown in A. Top panels: series of pseudocolor time-lapse images within the boxed region shown in A (each image separated by 1.1 s). Bottom panel: time course of mt-cpYFP fluorescence within the boxed region shown in A. C) Average ± se mSOF frequency in the absence (solid bars) and presence (shaded bars) of ETC blockers rotenone (5 μM, n=10–12 fibers from 2 mice) and oligomycin (5 μM, n=10 fibers from 2 mice). *P < 0.05. D–I) Representative sd images showing the different types of mSOF morphologies observed in FDB fibers from mt-cpYFP transgenic mice (left panels) and corresponding mitochondrial disposition as observed in EM images of FDB fibers from different experiments (right panels). D) Punctate (arrows). E) Rod-shaped. F) U-shaped. G) String-like along one side of the Z line. H) String-like across the Z line. I) Patches. Arrows indicate mitochondria stretching along the A band to connect different I bands longitudinally (E, I), adjacent sarcomeres transversally (H, I), or across the Z line to connect different mitochondria within the same I band, but located on opposite sides of the Z line (F, H). Fluorescent images: scale bars = 10 μm (A); 2 μm (D–I). EM images: scale bars = 2 μm (D, E, I); 0.5 μm (F–H). Fluorescence and EM images in panels D–I were obtained from different muscle fibers. Since EM enables higher-resolution images compared to confocal microscopy, the EM images are shown at higher magnification in order to provide increased resolutions of mitochondrial positioning and ultrastructure.

Figure 2.

EM cross-section showing complex mitochondrial reticulum within the I band that can account for large mSOF patch activity observed with confocal microscopy. A) Representative cross-section of a single FDB fiber: mitochondria are almost exclusively seen within the I band, on both sides of the Z line, where they form an extensive network. B) Higher magnification of the area marked with a dashed box in A. Arrowheads show long mitochondria running between myofibrils and being connected to other branches. C) Representative sd map generated from time-lapse images (0.7 s/frame, 150 frame total) showing large mSOF patch activity that stretches along the I band, spanning multiple sarcomeres. Red arrows indicate spatial spread of the observed patch flash activity. Such a large area of activity could reflect an extensively interconnected mitochondrial network like that shown in the EM images (A, B). Scale bars = 2 μm (A, C); 1 μm (B).

Figure 3.

mSOF activity does not require CypD-dependent mPTP functionality. A) Representative mt-cpYFP confocal image of an FDB fiber obtained from a WT mouse electroporated with mt-cpYFP cDNA; boxed region indicates an area containing mSOF activity. B) Time course of mt-cpYFP fluorescence within the boxed region shown in A. Top panels: series of pseudocolor time-lapse images. Bottom panel: time course of simultaneously recorded mt-cpYFP (green) and TMRE (red) fluorescence for the individual flash event. C, D) Same as A, B except for an FDB fiber obtained from a CypD^−/− mouse electroporated with mt-cpYFP cDNA. E, F) Average ± se basal mSOF frequency (E, left), amplitude (E, right), FDHM (F, left) and decay rate (F, right) in FDB fibers obtained from WT and CypD^−/− mice (n=22–23 fibers from 2 WT and 2 CypD^−/− mice) and WT FDB fibers in the presence of 1 μM CsA (n=14–17 fibers from 2 mice). G) Rotenone (5 μM) and oligomycin (5 μM) similarly inhibit mSOF frequency in FDB fibers from WT (left) and CypD^−/− (right) mice (n=8–12 fibers from 2 WT and 2 CypD^−/− mice). Scale bars = 10 μm. *P < 0.05.

Figure 4.

Biphasic regulation of mSOF activity by repetitive high-frequency tetanic stimulation. A) sd maps (100 images, 1.24 s/frame) for a representative FDB fiber before and immediately after 5 tetanic stimuli (500 ms, 100 Hz, 0.2 duty cycle). B) sd maps (100 images 1.24 s/frame) for a representative FDB fiber before, 30 s after, and 5 min after 40 tetanic stimuli (500 ms, 100 Hz, 0.2 duty cycle). Magenta borders indicate identified events. Scale bars = 10 μm. C–E) Schematic registry of individual mSOF events in representative control FDB fibers from mt-cpYFP transgenic mice before and 30 s after 2 tetani (C), before and 30 s after 5 tetani (D), and before and 30 s to 2 min after and 5–8 min after 40 tetani (E). Calculated flash frequencies F_f are indicated at left of each registry. F–H) Average ± se mSOF frequency (F), amplitude (G), and FDHM (H) before and after 2, 5, and 40 tetani (n=9–24 fibers from 7 mice). Data are normalized to control frequency, amplitude, and duration from consecutive time series taken in control fibers in the absence of stimulation in order to control for time/exposure-dependent changes in flash activity. Although no significant change in flash frequency, amplitude, and FDHM was seen for any of the 4 consecutive control time series, data were normalized against the appropriate control time series value. *P < 0.05.

Figure 5.

Temperature dependence of mSOF production in FDB fibers from RYR1^Y522S/WT MH mice. A, B) sd maps (150 images, 0.69 s/frame) for a representative FDB fiber obtained from a RYR1^Y522S/WT mouse at 23°C (A) and 37°C (B). Magenta borders indicate identified events. Scale bars = 10 μm. C) Schematic flash registry over time for the fiber depicted in A at 23°C (upper) and 37°C (lower). Calculated flash frequency F_f for each condition is indicated above the corresponding registry trace. D–F) Average ± se mSOF frequency (D), amplitude (E), and FDHM and decay rate (F) at 23°C and 37°C for FDB fibers obtained from WT and RYR1^Y522S/WT mice (n=18–48 fibers from 5 WT and 5 RYR1^Y522S/WT mice). *P < 0.05 vs. appropriate control; ^†P < 0.05 vs. all other conditions.