Holly Van Remmen, PhD, The Sam and Ann Barshop Institute for Longevity and Aging Studies, University of Texas Health Science Center at San Antonio, 15355 Lambda Drive, San Antonio, Texas 78245, USA. Tel.: 210 562-6141; fax: 210 562 6110; e-mail:firstname.lastname@example.org
In a previous study, we reported that a deficiency in MnSOD activity (approximately 80% reduction) targeted to type IIB skeletal muscle fibers was sufficient to elevate oxidative stress and to reduce muscle function in young adult mice (TnIFastCreSod2fl/fl mice). In this study, we used TnIFastCreSod2fl/fl mice to examine the effect of elevated oxidative stress on mitochondrial function and to test the hypothesis that elevated oxidative stress and decreased mitochondrial function over the lifespan of the TnIFastCreSod2fl/fl mice would be sufficient to accelerate muscle atrophy associated with aging. We found that mitochondrial function is reduced in both young and old TnIFastCreSod2fl/fl mice, when compared with control mice. Complex II activity is reduced by 47% in young and by approximately 90% in old TnIFastCreSod2fl/fl mice, and was found to be associated with reduced levels of the catalytic subunits for complex II, SDHA and SDHB. Complex II-linked mitochondrial respiration is reduced by approximately 70% in young TnIFastCreSod2fl/fl mice. Complex II-linked mitochondrial Adenosine-Tri-Phosphate (ATP) production is reduced by 39% in young and was found to be almost completely absent in old TnIFastCreSod2fl/fl mice. Furthermore, in old TnIFastCreSod2fl/fl mice, aconitase activity is almost completely abolished; mitochondrial superoxide release remains > 2-fold elevated; and oxidative damage (measured as F2- isoprostanes) is increased by 30% relative to age-matched controls. These data show that despite elevated skeletal muscle–specific mitochondrial oxidative stress, oxidative damage, and complex II-linked mitochondrial dysfunction, age-related muscle atrophy was not accelerated in old TnIFastCreSod2fl/fl mice, suggesting mitochondrial oxidative stress may not be causal for age-related muscle atrophy.
Elevated oxidative stress (Lopez et al., 2000; Wanagat et al., 2001; Fulle et al., 2004; Mansouri et al., 2006; Muller et al., 2007) and decreased mitochondrial function (Marcinek et al., 2005; Short et al., 2005; Mansouri et al., 2006) have been shown to be associated with age-related skeletal muscle atrophy, yet a causal role for increased oxidative stress and mitochondrial dysfunction has not been established. Knockout mouse models with alterations in mitochondrial antioxidant defense provide a mechanism for directly testing the effect of elevated mitochondrial oxidative stress and mitochondrial dysfunction on skeletal muscle atrophy. Mice deficient in MnSOD, the first line of antioxidant defense against superoxide radicals in the mitochondrial matrix, are an excellent model to study the effects of elevated mitochondrial oxidative stress. However, complete deletion of Sod2−/− leads to lethality within days to just a few weeks after birth, depending on genetic background (Huang et al., 2006), and therefore, Sod2−/− are not a useful model for aging studies.
To circumvent the lethality of complete deficiency of MnSOD, investigators have generated mouse models with targeted deletion of MnSOD in specific tissues using Cre-Lox approaches. Young mice with a combined heart and skeletal muscle deletion of Sod2 (H/M-Sod2−/− mice, Nojiri et al., 2006) or skeletal muscle–specific MnSOD knockout mice (muscle-Sod2−/−; Kuwahara et al., 2010) have been shown to have elevated mitochondrial oxidative stress in the affected tissues. Despite the elevated mitochondrial oxidative stress in the skeletal muscle of these two mouse models, there was no indication of muscle atrophy in either animal model (Nojiri et al., 2006; Kuwahara et al., 2010). Similarly, in a previous study, we reported that young adult TnIFastCreSod2fl/fl mice that have a 70% reduction in MnSOD content specifically in type IIB fibers, exhibit elevated mitochondrial oxidative stress, and oxidative damage but show no loss of muscle mass (Lustgarten et al., 2009). Thus, collective evidence from the H/M-Sod2−/−, muscle-Sod2−/−, and TnIFastCreSod2fl/fl mice suggests that elevated mitochondrial oxidative stress and mitochondrial dysfunction are not sufficient to initiate muscle atrophy, at least in young mice. However, it is possible that over the lifespan chronic elevation of mitochondrial dysfunction and oxidative stress might contribute to muscle atrophy in older mice. To address this question, we asked whether elevated mitochondrial dysfunction and oxidative stress resulting from MnSOD deficiency targeted to type IIB enriched skeletal muscle in TnIFastCreSod2fl/fl mice would lead to an accelerated loss of muscle mass during aging.
Mitochondrial function is reduced in young TnIFastCreSod2fl/fl mice
We have previously shown that mitochondria isolated from type IIB enriched skeletal muscle in young TnIFastCreSod2fl/fl mice have a > 80% decrease in MnSOD activity, elevated intramitochondrial superoxide, elevated mitochondrial superoxide release, and increased lipid oxidative damage, relative to wild type (Lustgarten et al., 2009). However, the role of elevated type IIB enriched skeletal muscle mitochondrial oxidative stress on mitochondrial function in TnIFastCreSod2fl/fl mice was not addressed. To determine the effect of the targeted reduction of MnSOD in type IIB muscle fibers on muscle mitochondrial function, we measured mitochondrial oxygen consumption and ATP production in skeletal muscle mitochondria isolated from type IIB enriched skeletal muscle in young adult (6–8 month) TnIFastCreSod2fl/fl and wild-type mice. We found no significant difference in the glutamate plus malate (GM)-supported basal rate of mitochondrial oxygen consumption (in the absence of ADP) in mitochondria isolated from type IIB enriched skeletal muscle from TnIFastCreSod2fl/fl when compared with wild-type mice (Fig. 1A). Glutamate plus malate-stimulated maximal respiration (in the presence of exogenously added ADP) was reduced by 32% in mitochondria isolated from type IIB enriched muscle from TnIFastCreSod2fl/fl mice, but this difference was not statistically significant (Fig. 1B). Succinate plus rotenone (SR)-linked basal and maximal respiration were reduced by 65% (Fig. 1C) and 74% (Fig. 1D), respectively, in type IIB enriched muscle from TnIFastCreSod2fl/fl mice when compared with wild type. The rate of mitochondrial ATP production by mitochondria isolated from type IIB enriched skeletal muscle respiring on several substrates including GM (Fig. 1E), α-ketoglutarate (Fig. 1F), and α-glycerol phosphate (Fig. 1G) was not significantly different between TnIFastCreSod2fl/fl and wild-type mice. In contrast, the rate of mitochondrial ATP production with SR used as substrate was reduced by 39% in type IIB enriched muscle from TnIFastCreSod2fl/fl mice (Fig. 1E), indicating a complex II-specific defect in type IIB enriched skeletal muscle from TnIFastCreSod2fl/fl. Thus, mitochondrial function with complex II-linked substrate was altered in young mice with a targeted reduction of MnSOD.
Activities of Complex I and II are reduced in TnIFastCreSod2fl/fl mice
The electron transport chain (ETC.), consisting of five protein complexes: complex I (NADH-ubiquinone oxidoreductase), complex II (succinate-ubiquinone oxidoreductase), complex III (ubiquinol-cytochrome c reductase), complex IV (cytochrome c oxidase), and complex V (ATP synthase), is located in the inner mitochondrial membrane. Therefore, the lack of MnSOD could potentially increase oxidative stress–induced damage to the ETC complexes. As shown in Fig. 2, the activities of complexes I and II were significantly reduced by 32% and 47%, respectively, in mitochondria isolated from type IIB enriched skeletal muscle in TnIFastCreSod2fl/fl compared with muscle mitochondria from wild type. Activities of complex III and IV were not different in the two mouse models. In addition, there was no difference in the activities of complexes I, II, III, or IV in the homogenates of type I fiber enriched skeletal muscle (soleus) from TnIFastCreSod2fl/fl or wild-type mice (data not shown). We confirmed the changes in ETC activity using histological analysis of gastrocnemius muscle sections. As shown in Fig. 3, complex IV activity is not changed, but clear zones indicative of reduced complex II activity are evident in muscle from TnIFastCreSod2fl/fl mice, when compared with muscle from wild-type mice.
SDHA and SDHB protein contents are reduced but levels of SdhA and SdhB mRNA are not significantly different in young TnIFastCreSod2fl/fl mice
Complex II is composed of two hydrophilic (SDHA and SDHB) and two hydrophobic membrane-localized subunits (SDHC and SDHD). To determine whether the reduced activity of complex II is associated with a reduction in the level of complex II protein, we measured the level of SDHA and SDHB in mitochondria isolated from type IIB enriched muscle in TnIFastCreSod2fl/fl and wild-type control mice. SDHA and SDHB protein contents were significantly reduced by 32% and 70%, respectively (Fig. 4A–C). SDHA and SDHB protein levels were not different in the homogenates of type I fiber enriched soleus muscles isolated from TnIFastCreSod2fl/fl and wild-type mice (data not shown). We also measured SdhA and SdhB mRNA content to determine whether the deficit in SDHA and SDHB protein content found in TnIFastCreSod2fl/fl was associated with a decrease in gene transcription. The mRNA content for SdhA and SdhB found in type IIB enriched skeletal muscle was not significantly different in muscle isolated from TnIFastCreSod2fl/fl and wild-type mice (Fig. 4D).
Activities of other TCA cycle enzymes are not decreased in young TnIFastCreSod2fl/fl mice
To determine whether the reduction in the enzymatic activities of aconitase and succinate dehydrogenase (SDH) was selective, we measured the activity of the mitochondrial matrix–localized citric acid cycle (TCA) enzymes citrate synthase, isocitrate dehydrogenase, malate dehydrogenase, and fumarase. No significant difference was found for the enzymatic activity of citrate synthase, isocitrate dehydrogenase, malate dehydrogenase, or fumarase either in type IIB enriched skeletal muscle (Table 1) or in the type I enriched soleus muscle (data not shown) when comparing the values obtained in TnIFastCreSod2fl/fl with wild-type mice. These data indicate that the TCA cycle enzymes aconitase and SDH are preferentially sensitive to MnSOD deficiency.
Table 1. Citric acid cycle (TCA) enzymatic activity in type IIB enriched skeletal muscle isolated from young (6–8 month) TnIFastCreSod2fl/fl and wild type mice
TCA enzymatic activity was measured as described in Experimental Procedures. For citrate synthase, aconitase, isocitrate dehydrogenase, fumarase, and malate dehydrogenase (n = 4). For succinate dehydrogenase (n = 5). Data are expressed as Kilo Units (KU, 103) per min per mg protein for citrate synthase, succinate dehydrogenase and fumarase. Data are expressed as Mega Units (MU, 106) per min per mg protein for aconitase, isocitrate dehydrogenase and malate dehydrogenase. Values represent means ± SEM. *A significant difference (P < 0.05) from wild type values.
12.8 ± 0.1
0.41 ± 0.02
6.7 ± 0.4
8.7 ± 0.7
12.3 ± 0.3
3.8 ± 0.5
11.8 ± 0.9
0.18 ± 0.02*
6.1 ± 0.5
4.1 ± 0.4*
12.6 ± 0.2
3.7 ± 0.3
SDH is sensitive to inactivation by superoxide
Aconitase inactivation was previously shown to be because of superoxide-mediated Fe-S cluster damage (Gardner et al., 1995). The SDHB subunit of complex II contains three Fe-S clusters (Hagerhall, 1997). Therefore, we asked whether the decrease in SDH activity found in TnIFastCreSod2fl/fl muscle was related to superoxide-mediated inactivation. As shown in Fig. 5A, SDH activity is reduced in the presence of potassium superoxide in vitro, demonstrating that SDH is also susceptible to inactivation by superoxide. In contrast, the activities of isocitrate dehydrogenase (Fig. 5B) and fumarase (Fig. 5C) are not altered by exposure to potassium superoxide.
Oxidative stress and oxidative damage are elevated in old TnIFastCreSod2fl/fl mice
In a previous study, we demonstrated that type IIB enriched muscle from the TnIFastCreSod2fl/fl mice had increased levels of oxidative stress as measured by reduced aconitase activity, increased superoxide release, and increased oxidative damage (measured as F2- isoprostanes; Lustgarten et al., 2009). We hypothesized that a lifelong reduction in MnSOD content would lead to an increase in mitochondrial matrix–localized superoxide and potentially exacerbate oxidative stress in older animals. To ask whether mitochondrial oxidative stress is elevated during aging in muscle from TnIFastCreSod2fl/fl mice, we measured aconitase activity in young and old wild-type and TnIFastCreSod2fl/fl mice. As shown in Fig. 6A, aconitase activity was reduced by approximately 90% in old (27–34 month) TnIFastCreSod2fl/fl mice when compared with the value obtained either in young TnIFastCreSod2fl/fl or in old wild-type mice. Furthermore, aconitase activity is decreased by more than 35% during aging in wild-type mice. Interestingly, the aconitase protein level is also reduced by more than 50% in type IIB enriched muscle from old TnIFastCreSod2fl/fl mice when compared with young TnIFastCreSod2fl/fl mice (Fig. 6B). This effect is specific to type IIB enriched skeletal muscle in the TnIFastCreSod2fl/fl mice as we found no significant difference in aconitase protein in type I fiber enriched soleus muscle from old mice of either genotype (Fig. 6B). These data demonstrate an increase in type IIB enriched skeletal muscle mitochondrial matrix oxidative stress during aging that occurs in old wild-type mice but to a greater extent in old TnIFastCreSod2fl/fl mice.
Because aconitase activity is an indirect indicator of mitochondrial matrix superoxide content, we directly measured mitochondrial superoxide release with use of electron paramagnetic resonance (EPR). Mitochondria isolated from type IIB enriched skeletal muscle from old TnIFastCreSod2fl/fl or old wild-type mice do not release more superoxide than mitochondria isolated from their young, genotype-matched counterparts (Fig. 6C,D). However, mitochondria isolated from both young and old TnIFastCreSod2fl/fl mice release > 2-fold more superoxide than mitochondria isolated from age-matched wild-type mice in the presence of either GM or SR as respiratory substrates. Thus, the reduction in MnSOD results in elevated levels of superoxide that can be detected extramitochondrially.
To determine whether a reduction in MnSOD content leads to an increase in oxidative damage during aging in TnIFastCreSod2fl/fl mice, we measured lipid peroxidation (F2-Isoprostanes) in the tibialis anterior muscle (35% type IIB fibers) isolated from 21-month-old TnIFastCreSod2fl/fl mice. We had previously shown that young adult TnIFastCreSod2fl/fl mice have elevated levels of F2-isoprostanes (Lustgarten et al., 2009). Here, we found a significant elevation (30%) of F2-isoprostanes in tibialis anterior muscle isolated from middle-aged TnIFastCreSod2fl/fl mice, when compared with the value obtained in middle-aged wild-type mice (Fig. 6E).
Complex II-linked ATP production, complex II activity, and protein content are each reduced in old TnIFastCreSod2fl/fl mice
The rate of mitochondrial ATP production with GM used as substrate was not significantly different in mitochondria isolated from type IIB enriched skeletal muscle in old TnIFastCreSod2fl/fl mice, when compared with the corresponding value in age-matched wild-type mice (Fig. 7A). However, in response to SR as substrate, the rate of mitochondrial ATP production was reduced by 98% in old TnIFastCreSod2fl/fl mice, relative to age-matched wild-type (Fig. 7B). The activities of complexes I through IV were measured to investigate whether mitochondrial respiratory complex activity was impaired during aging in TnIFastCreSod2fl/fl. Complex II activity was selectively reduced by 88% (Fig. 7D), but the activity of the other three protein complexes was not different when comparing old mutant mice with age-matched wild-type. SDHA and SDHB protein contents were found to be reduced by more than 50% and 90%, respectively (Fig. 7G) in type IIB enriched muscle isolated from TnIFastCreSod2fl/fl mice, when compared with age-matched wild-type. In contrast, levels of SDHA and SDHB were not significantly different in the homogenates of type I fiber enriched soleus muscles isolated from old TnIFastCreSod2fl/fl, when compared with old wild-type mice (Fig. 7H).
Muscle atrophy is not greater during aging in TnIFastCreSod2fl/fl mice
The preceding experiments demonstrated an increase in oxidative stress and mitochondrial dysfunction in mitochondria isolated from type IIB enriched muscle in TnIFastCreSod2fl/fl mice during aging. If mitochondrial dysfunction and oxidative stress play a casual role in muscle atrophy during aging, we would predict an increase in muscle atrophy in muscle with high type IIB content (and therefore low MnSOD activity) in the TnIFastCreSod2fl/fl mice. As shown in Table 2, the mass of the type IIB rich gastrocnemius muscle (normalized to body weight, G/BW) and type I rich soleus muscle (S/BW) was examined in young (3–4 months) and old (27–34 months) TnIFastCreSod2fl/fl and wild-type mice. G/BW declined significantly during aging for both wild-type and TnIFastCreSod2fl/fl female but not male mice (data not shown). S/BW did not change during aging for either genotype or gender. Significant increases in body mass and the percentage of body fat were identified for both old wild-type and TnIFastCreSod2fl/fl female, but not male mice (data not shown) during aging. In addition, the percentage of lean mass significantly declined during aging for both wild-type and TnIFastCreSod2fl/fl female mice. However, lifelong elevated oxidative stress, oxidative damage, and mitochondrial dysfunction targeted to type IIB enriched skeletal muscle were not sufficient to further alter G/BW, S/BW, body mass, or the respective percentages of body fat and lean mass during aging in female or male (data not shown) TnIFastCreSod2fl/fl, when compared with age-matched wild-type mice.
Table 2. Phenotypic data in young and old TnIFastCreSod2fl/fl and wild type mice
Body mass (g)
Lean mass (%)
Body fat (%)
Sol./BW (mg g−1)
Gast./BW (mg g−1)
Shown are values for body mass, percent lean mass or body fat, and muscle mass normalized to body mass for soleus and gastrocnemius muscle during aging in young (6–8 months, females; n = 5) and old (27–34 months, females, n = 5) wild type and TnIFastCreSod2fl/fl mice. Quantitative magnetic resonance imaging (qMRI) was used for the determination of the percentages of body fat and lean mass based on Taicher et al. (2003), Tinsley et al. (2004) and as described by Lustgarten et al. (2009). Values represent means ± SEM. *A significant difference (P < 0.05) as a function of age.
21.3 ± 0.3
83.2 ± 0.6
16.5 ± 0.6
0.32 ± 0.01
5.9 ± 0.1
20.9 ± 0.3
81.7 ± 0.6
17.8 ± 0.6
0.33 ± 0.02
5.7 ± 0.1
23.4 ± 0.8*
78.2 ± 2.1*
21.2 ± 2.1*
0.34 ± 0.02
4.1 ± 0.2*
25.7 ± 1.6*
76.1 ± 1.9*
23.4 ± 1.9*
0.30 ± 0.02
4.4 ± 0.2*
The goal of the present study was to investigate the role of chronic elevations of mitochondrial oxidative stress, oxidative damage, and mitochondrial dysfunction in age-related muscle atrophy in a mouse model with targeted deletion of the mitochondrial antioxidant, MnSOD in type IIB muscle fibers. The data presented in this report show that increases in mitochondrial oxidative stress and oxidative damage and alterations in mitochondrial function caused by MnSOD deficiency in type IIB fibers does not lead to skeletal muscle fiber atrophy in either young or old mice. In young mice with a > 80% reduction in MnSOD activity in type IIB enriched skeletal muscle, we found reduced aconitase activity (Lustgarten et al., 2009), decreased rates of mitochondrial oxygen consumption and ATP production, and a specific defect in complex II activity and protein content. In old TnIFastCreSod2fl/fl mice, aconitase activity was further reduced, mitochondrial superoxide release was elevated, oxidative damage was increased, and mitochondrial function was decreased, but there was no acceleration in the rate of muscle atrophy, relative to the values found in age-matched wild-type mice. These results are consistent with previous studies in mice with a 50% reduction in MnSOD (Sod2+/− mice, Van Remmen et al., 2001, 2003; Mansouri et al., 2006), in mice deficient in MnSOD in both skeletal muscle and heart (H/M-Sod2−/−, Nojiri et al., 2006), and in mice lacking MnSOD in skeletal muscle (M-Sod2−/−, Kuwahara et al., 2010), as none of these models show increases in muscle atrophy despite elevated mitochondrial oxidative stress and mitochondrial dysfunction.
Skeletal muscle is a heterogeneous tissue composed of several different fiber types that are characterized on the basis of the content of myosin heavy chain type I, IIA, IIX, or IIB (Schiaffino et al., 1986; Schiaffino & Reggiani, 1994). The fiber types also differ in terms of metabolism. Mitochondrial content and oxidative metabolism are high in type I muscle fibers, while type IIB muscle fibers contain low levels of oxidative enzymes and mitochondria, but have a high abundance of glycolytic proteins (Pette & Staron, 1990). Type IIA and IIX fibers are intermediate in terms of combined oxidative and glycolytic capacity. Furthermore, mitochondria from type I and type II skeletal muscle have previously been shown to differ with respect to production of reactive oxygen species. Mitochondrial free radical leak (expressed as moles of H2O2 produced/mole O2 consumed), as found in the lateral portion of the gastrocnemius that is enriched in type IIB fiber content, was reported to be higher than mitochondrial free radical leak in muscles comprised mostly of type IIA fibers (medial gastrocnemius) or type I fibers (soleus) (Anderson & Neufer, 2006; Picard et al., 2008). Consistent with these findings, muscles comprised primarily of type II fibers have been shown to have increased levels of mitochondrial protein oxidative damage and decreased mitochondrial function relative to muscle (soleus) enriched in type I muscle fibers (Conley et al., 2007; Feng et al., 2008). Furthermore, type II muscle fibers have been shown to be more susceptible to age-related atrophy than type I skeletal muscle (Tomonaga, 1977; Larsson et al., 1978; Jakobsson et al., 1990; Lexell & Taylor, 1991; Fulle et al., 2004). Collectively, these data suggest that mitochondrial oxidative stress and damage and mitochondrial dysfunction in type II fibers may be important factors in age-related muscle atrophy. Use of the TnIFastCreSod2fl/fl mouse model targets the deletion of MnSOD specifically in type IIB muscle fibers and is therefore particularly relevant for studying age-related changes in skeletal muscle.
Our data also strengthen the direct link between reduced MnSOD content and specific deficits in the activity of aconitase (Gardner et al., 1994; Li et al., 1995; Longo et al., 1999; Powell & Jackson, 2003; Hinerfeld et al., 2004; Lustgarten et al., 2009), complex I (Longo et al., 1999; Powell & Jackson, 2003; Hinerfeld et al., 2004; Nojiri et al., 2006; Kuwahara et al., 2010), and complex II (Li et al., 1995; Longo et al., 1999; Powell & Jackson, 2003; Hinerfeld et al., 2004; Nojiri et al., 2006; Kuwahara et al., 2010). Complex I-supported respiration (GM) and activity were both decreased by 30%, but this amount was insufficient to adversely affect the rate of mitochondrial ATP production found in young TnIFastCreSod2fl/fl mice when compared with the corresponding value in wild type. One possible explanation involves the mitochondrial threshold effect (Mazat et al., 2001), in which 35–40% of complex I must be inactivated before mitochondrial function is affected (Barrientos & Moraes, 1999). In contrast, the reduction in complex II activity found in TnIFastCreSod2fl/fl mice was sufficient to negatively affect its respective rates of mitochondrial respiration, ATP and H2O2 production (Lustgarten et al., 2009). Succinate dehydrogenase is comprised of four nuclear-encoded subunits: SdhA, SdhB, SdhC, and SdhD. SDHA is a 70-kDA protein and contains the FAD-binding site. SDHB is a 30-kDA protein and contains three iron-sulfur clusters. Both SDHA and SDHB are soluble in the mitochondrial matrix and are anchored to the inner mitochondrial membrane by two 15-kDA transmembrane proteins, SDHC and SDHD (Hagerhall, 1997). The activity of complex II has been previously reported to be directly proportional to the content of its catalytic subunits, SDHA and SDHB (Hinerfeld et al., 2004). SDHA and SDHB protein contents found in type IIB enriched skeletal muscle were each significantly reduced, but no significant difference was observed for the mRNA levels of either SdhA or SdhB in TnIFastCreSod2fl/fl mice. These results are consistent with a previous study in mice lacking MnSOD in both heart and skeletal muscle (H/M-Sod2−/− mice, Nojiri et al., 2006). SDHB contains three iron-sulfur clusters, leaving open the possibility that SDHB is translated, oxidized, and subsequently degraded, as it may pass through the same metabolic fate as aconitase-reversible post-translational inactivation, release of a labile iron from the [4Fe-4S]2+ cluster, disassembly of the [4Fe-4S]2+ cluster, carbonylation, and protein degradation (Bulteau et al., 2003). SDHB contains one [4Fe-4S]2+ cluster, and the [3Fe-4S]+ cluster found in SDHB has been shown to be sensitive to oxidation (Beinert et al., 1977). Here, we show that purified SDH is sensitive to inactivation by superoxide, with significant time-dependent reductions in SDH activity following incubation with superoxide. Prior to our study, superoxide had been shown to be only a mild inactivator of SDH (Zhang et al., 1990). Inactivation by superoxide is not a general effect as the enzymatic activity of fumarase and isocitrate dehydrogenase was not sensitive to inactivation when assayed under the same experimental conditions used to test the superoxide sensitivity of SDH. Mammalian fumarase activity was previously shown to be insensitive to superoxide-mediated inactivation (Patel et al.,1996), and our data validate that result.
One possible explanation for the reduction in SDHA and SDHB protein content may involve mRNA oxidation. Mitochondria isolated from type IIB enriched muscle in TnIFastCreSod2fl/fl mice respiring on complex I or complex II-linked substrate release > 2-fold more superoxide than mitochondria isolated from wild-type mice (Lustgarten et al., 2009). As a result of potentially elevated cytosolic superoxide (or H2O2, after conversion by CuZnSOD), an increase in SdhA or SdhB mRNA oxidation may occur, thereby resulting in a reduction in protein expression (Shan et al., 2007), and may be responsible for the decrease in SDHA and/or SDHB protein content. Future studies aimed at investigating the association between a decrease in MnSOD content and the reduction in SDHA and SDHB subunit levels should investigate whether SdhA and SdhA mRNA are oxidized and are unable to be translated.
In summary, we report that mitochondrial oxidative stress and damage and mitochondrial dysfunction specifically targeted to type IIB skeletal muscle fibers over the lifespan of mice are not sufficient to accelerate age-related muscle atrophy. The fact that reductions in MnSOD activity of 50% in both young and old Sod2+/− mice (Mansouri et al., 2006), > 80% (present report, young and old), and 100% in both M-Sod2−/− (Kuwahara et al., 2010) and H/M-Sod2−/− mice (Nojiri et al., 2006) all result in an increase in mitochondrial oxidative stress and compromised mitochondrial function without affecting skeletal muscle mass strongly argues that mitochondrial oxidative stress and mitochondrial dysfunction are not involved in the maintenance of muscle mass in either young or old mice.
Young (6–8 months) and old female (27–34 months) wild-type and TnIFastCreSod2fl/fl mice (described previously in Lustgarten et al., 2009) were used for all experiments, with the exception of the assay of F2-isoprostanes, in which 21-month-old wild-type and TnIFastCreSod2fl/fl male mice were used. All procedures involving the mice were approved by the Subcommittee for Animal Studies at the Audie L. Murphy Veterans Administration Hospital and the University of Texas Health Science Center at San Antonio.
Measurement of mitochondrial function
Because the TnI promoter is expressed predominantly in type IIB muscle fibers (Hallauer & Hastings, 2002), we used portions of mouse hindlimb muscle that are predominantly type II in composition (the white portion of the quadriceps and the gastrocnemius) for the isolation of mitochondria for our experiments and refer to this as type IIB enriched muscle throughout the manuscript. Burkholder et al. (1994) reported the type IIB fiber composition of the gastrocnemius and quadricep group to range from 55% to 70%. Red portions of both muscles were removed to maximize type IIB fiber content in our muscle preparations. We have previously shown that MnSOD activity is reduced by > 80% in this combined muscle preparation (Lustgarten et al., 2009).
Mitochondria were isolated based on the method of King et al. (2007) and as described by Lustgarten et al. (2009). Mitochondrial respiration was measured using the method of Hynes et al. (2006). This assay is based on measuring the fluorescence of the oxygen-sensing probe (A65N-1; Axxora, San Diego, CA, USA) which is quenched in the presence of oxygen. Signal quench declines as oxygen is consumed, and the corresponding increase in fluorescence is directly proportional to the rate of mitochondrial oxygen consumption. Forty to eighty micrograms of mitochondria isolated from type IIB enriched muscle from wild-type and TnIFastCreSod2fl/fl mice was resuspended in 150 μL of respiration buffer containing 125 mm KCl, 10 mm MOPS, 5 mm MgCl2, 2 mm K2HPO4, pH 7.44, with substrates: GM (12.5 mm glutamate, 12.5 mm malate) or SR (25 mm succinate, 0.5 μm rotenone) and A65N-1 probe, 67 nm. One hundred microliters of mineral oil was used to seal each well from atmospheric oxygen. Measurements were performed in a Spectramax 384 spectrophotometer (Molecular Devices, Sunnyvale, CA, USA).
The rate of mitochondrial ATP production was measured in isolated muscle mitochondria as we have previously described (Lustgarten et al., 2009). The respiratory substrates GM (2.5 mm), SR (5 mm succinate, 0.5 μm rotenone), α-ketoglutarate (2.5 mm), and α-glycerol phosphate (2.5 mm) were added as indicated. Mitochondrial electron transport respiratory complex activity (ETC., complexes I-IV) was measured using spectrophotometric assays. Mitochondria isolated from type IIB enriched skeletal muscle isolated from wild-type and TnIFastCreSod2fl/fl mice were solubilized in buffer containing 750 mm 6-aminocaproic acid (hexanoic acid), 50 mm Bis Tris, and 1% L-dodecyl maltoside, pH 7.0. The samples were then placed on a rotator for 1 h at 4 °C, followed by centrifugation at 100 000 g for 15 min. The supernatant was then assayed for protein content and resuspended in respiration buffer (250 mm sucrose, 10 mm KH2PO4, 10 mm Tris, 1 mm EGTA, pH 7.4). Activity measurements for complexes I, II, and IV were taken on a Spectramax 384 Spectrophotometer (Molecular Devices). Complex III activity was measured via a Beckman spectrophotometer. For complex I, 3.3 μg of mitochondrial protein was added to 150 μL of respiration buffer containing 0.1 mm NADH, 0.05 mm decylubiquinone, and 0.05 mm dichloroindophenol (DCIP). Complex I activity was determined by monitoring the oxidation of NADH at 340 nm, at 30°C (Birch-Machin & Turnbull, 2001). A separate sample was measured in parallel with the addition of 640 nm rotenone. Subtraction of this data is indicative of rotenone-sensitive complex I activity. Complex I data were expressed as units of NADH oxidized per min per mg protein. For complex II, 3.3 μg of mitochondrial protein was added to 150 μL of respiration buffer containing 20 mm succinate, 50 um ubiquinone, 2 mm KCN, and 50 μm DCIP. Complex II activity was measured via the succinate-dependent reduction of DCIP, as measured at 600 nm, at 30°C (Boveris & Cadenas, 1975). A separate sample containing 3 mm malonate was measured in parallel. Subtraction of this data is indicative of malonate-sensitive complex II activity. Complex II data are expressed as units of DCIP reduced per min per mg protein. For complex III, 3.3 μg of mitochondrial protein was added to 1 mL of buffer containing 100 μm (Fe3+) cytochrome c, 100 μm decylubiquinol, and 2 mm KCN. Complex III activity was measured by following the reduction of cytochrome c at 550 nm, at 25°C. Complex III data were expressed as units of cytochrome c reduced per min per mg protein. Complex IV activity was measured by the addition of 1.65 μg of mitochondrial protein to 150 μL of respiration buffer containing 90 μm (Fe2+) cytochrome c. Complex IV activity was determined by following the oxidation of cytochrome c at 550 nm, at 30°C. Complex IV data are expressed as units of cytochrome c oxidized per min per mg protein.
Cryostat sections of gastrocnemius muscles isolated from young female wild-type and TnIFastCreSod2fl/fl mice were used for the histologic analysis of complex II and complex IV activities and were prepared as previously described by Lustgarten et al. (2009). Complex II activity was assayed by incubating sections for 1 h at 37°C in a buffer containing 20 um NaPO4, 40 μm succinate, 0.1% NBT, pH 7.4 (Nachlas et al., 1957). Sections were then rinsed in saline, followed by fixation in formalin-saline (10% neutral buffered formalin containing 10 mm NaCl) for 10 min. Sections were placed in 15% ethanol for 15 min, followed by addition of mounting media. A dark blue formazan color is indicative of muscle fibers rich in complex II activity. Complex IV activity was assayed by incubating sections for 1 h at room temperature in buffer consisting of 200 mm sucrose, 20 mm NaPO4, 1.3 mm 3, 3′ diaminobenzidine tetrahydrochloride (DAB), 80 μm cytochrome c, 8 nm catalase, pH 7.6. Sections were then washed three times with double-distilled water, followed by muscle section dehydration in ascending alcohols (50%, 70%, 80%, 95% × 2 washes, 100% × 2 washes). Sections were cleared with two additions of xylene. Following the addition of mounting media, sections were visualized for dark brown staining, which is indicative of muscle fibers with high complex IV activity (Seligman, 1968).
Measurement of SDH mRNA
SdhA and SdhB mRNA were isolated from type IIB enriched skeletal muscle, and the content was measured using quantitative real-time PCR as described by Dahia et al. (2005). For the determination of SdhA mRNA content, GTGCGGATTGATGAGTACGATT and CACATGCATGAGCTATTATACATAA were used as the forward and reverse primers, respectively. SdhB mRNA content was measured with use of CGACTCCAGAGACGACTTCAC and GCTCGCTTCTCCTTGTAGGTC as the forward and reverse primers, respectively.
Homogenates of type IIB enriched or type I enriched (soleus) skeletal muscle isolated from TnIFastCreSod2fl/fl and wild-type mice were prepared in assay buffer containing 50 mm Tris–HCl buffer with 150 mm NaCl, 1% Nonidet P-40, 0.25% sodium deoxycholate, and 1× protease inhibitor. Equivalent amounts of protein (40–80 μg) for each sample were resolved in 4–20% Tris–HCl SDS-PAGE gels in triplicate. After electrophoresis, the proteins were transferred to a polyvinylidene difluoride membrane. The membrane was then incubated in Tris-buffered saline, pH 7.4, with 0.05% Tween 20 (TBS-T) containing 10% nonfat milk for 1 h at room temperature. The blots were then reacted with antibodies specific for aconitase (1:2000, a kind gift from Luke Szweda, Oklahoma Medical Research Foundation, Oklahoma City, OK, USA), SDHA (1:1000; Invitrogen, Carlsbad, CA, USA), and SDHB (1:500, Invitrogen) at 4°C overnight. After washing with TBS-T, the blots were incubated with either goat anti-rabbit IgG horseradish peroxidase or goat anti-mouse IgG horseradish peroxidase (1:1000; Sigma-Aldrich, Natick, MA, USA) for 2 h at room temperature. The blots were washed five times with TBS-T, and the bands corresponding to each respective protein was visualized using chemiluminescent detection reagents obtained from Amersham Biosciences (Pittsburgh, PA, USA).
Citric acid cycle enzymatic activity
Citric acid cycle enzymatic activity was measured in mitochondria isolated from type IIB enriched skeletal muscle from wild-type and TnIFastCreSod2fl/fl mice. To obtain soluble mitochondrial proteins, mitochondria were resuspended in 30 mm potassium phosphate buffer containing 0.2% Triton-X-100, pH 7.4, followed by rotation for 45 min at 4°C and centrifugation at 16 100 g. The activity of citrate synthase (1 μg), isocitrate dehydrogenase (13 μg), fumarase (30 μg), and malate dehydrogenase (0.1) was measured by addition of mitochondrial extract (with the amount of protein added in parenthesis) to 200 μL of enzyme activity buffer as described by Robinson et al. (1987). All enzyme activity assays were conducted at 30°C.
Identification of proteins sensitive to superoxide
Purified Escherichia coli succinate dehydrogenase (SDH, a kind gift from Dr. Gary Ceccini, UCSF), fumarase (Sigma-Aldrich, USA), and isocitrate dehydrogenase (Sigma-Aldrich) were used to determine their potential sensitivity to inactivation in the presence of potassium superoxide. E. Coli SDH is structurally and functionally homologous to mammalian SDH (Cecchini et al., 2002). Potassium superoxide solution was prepared by adding 0.5 g of KO2 to 1.5 mL anhydrous dimethyl sulfoxide (DMSO). This solution was vortexed for 5 min, followed by centrifugation at 500 g for 5 min. The supernatant containing soluble KO2 was removed. Catalase was added (1/100) to degrade any hydrogen peroxide that had formed from the spontaneous dismutation of superoxide. Electron paramagnetic resonance was used to quantify the amount of superoxide present in the KO2 solution. The KO2 stock was then further diluted in DMSO (1/5) to obtain an EPR spectrum intermediate in magnitude to that produced by mitochondria isolated from both wild-type and TnIFastCreSod2fl/fl mice (Lustgarten et al., 2009). Nondiluted KO2 produces an EPR signal that is approximately 10–12 and 5–6 fold higher than mitochondria respiring on SR and isolated from wild type and TnIFastCreSod2fl/fl, respectively (data not shown). Addition of CuZnSOD (1 U μL−1) eliminated the observed EPR spectra (Lustgarten et al., 2009), thereby indicating the specificity of the EPR-derived signal for detecting superoxide. Individual enzymes were assayed for their sensitivity to inactivation in the presence of superoxide via room temperature incubation. For example, 10 μg of purified SDH was added to either 1 μL of diluted KO2 or 1 μL of DMSO in a final volume of 10 μL. For measurement of SDH activity, 200 mm succinate, 4 mm ubiquinone, and 3 mm DCIP were added to a reaction buffer consisting of 250 mm sucrose, 10 mm KH2PO4, 10 mm Tris, 1 mm EGTA, 0.4 U μL−1 CuZnSOD, and 7.5 U μL−1 catalase, pH 7.4. CuZnSOD and catalase were added to the reaction buffer to degrade any superoxide (and potentially, H2O2) that remained after incubation with SDH. At the indicated time points (0, 10, 20, 40 min), 0.6 μL of SDH incubated with either KO2 or DMSO was removed and added to 150 μL of reaction buffer. SDH activity was then measured based on the method of Boveris & Cadenas (1975).
To test the sensitivity of other mitochondrial matrix–localized proteins to inactivation by superoxide, we measured the enzymatic activity of fumarase and isocitrate dehydrogenase. Three micrograms of fumarase was incubated with 4 μL of diluted KO2 stock or 4 μL of DMSO in 50 μL of reaction buffer. At the indicated time points (0, 10, 20, 40 min), 10 μL of superoxide or DMSO-incubated fumarase was added to 150 μL of reaction buffer containing 100 mm malate, 0.4 U μL−1 CuZnSOD, and 7.5 U μL−1 catalase, pH 7.4. Fumarase activity was measured based on the method of Robinson et al. (1987). Similarly, 55 μg of isocitrate dehydrogenase was incubated with 3.75 μL of diluted KO2 or 3.75 μL of DMSO in 50 μL of reaction buffer. At the indicated time points (0, 10, 20, 40 min), 10 μL of superoxide or DMSO-incubated isocitrate dehydrogenase was added to 150 uL of reaction buffer containing 50 mm Tris pH 7.4, 1 mm MnCl2, 0.5 mm EDTA, 0.25 mm NADP+, 1.6 mm isocitrate, 0.4 U μL−1 CuZnSOD, and 7.5 U μL−1 catalase, pH 7.4. Isocitrate dehydrogenase activity was then measured based on the method of Robinson et al. (1987).
Measurement of oxidative stress status during aging in TnIFastCreSod2fl/fl mice