NFκB Regulates Muscle Development and Mitochondrial Function

Abstract

Nuclear factor (NF)κB is a transcription factor that controls immune and inflammatory signaling pathways. In skeletal muscle, NFκB has been implicated in the regulation of metabolic processes and tissue mass, yet its affects on mitochondrial function in this tissue are unclear. To investigate the role of NFκB on mitochondrial function and its relationship with muscle mass across the life span, we study a mouse model with muscle-specific NFκB suppression (muscle-specific IκBα super-repressor [MISR] mice). In wild-type mice, there was a natural decline in muscle mass with aging that was accompanied by decreased mitochondrial function and mRNA expression of electron transport chain subunits. NFκB inactivation downregulated expression of PPARGC1A, and upregulated TFEB and PPARGC1B. NFκB inactivation also decreased gastrocnemius (but not soleus) muscle mass in early life (1–6 months old). Lower oxygen consumption rates occurred in gastrocnemius and soleus muscles from young MISR mice, whereas soleus (but not gastrocnemius) muscles from old MISR mice displayed increased oxygen consumption compared to age-matched controls. We conclude that the NFκB pathway plays an important role in muscle development and growth. The extent to which NFκB suppression alters mitochondrial function is age dependent and muscle specific. Finally, mitochondrial function and muscle mass are tightly associated in both genotypes and across the life span.

Chronic low-grade inflammation, changes in mitochondrial function, and oxidative stress are proposed mechanisms underlying the aging process and the pathophysiology of aging-associated diseases such as sarcopenia (1–3). Inflammatory proteins and oxidative stress increase with age in mice and humans (4,5), resulting in the activation of pro-inflammatory transcriptional regulators such as nuclear factor (NF)κB (6). Activation of the canonical NFκB pathway is mediated by the degradation of inhibitor of NFκB α/β (IκBα/β), which liberates the p50/p65 heterodimer allowing for translocation to the nucleus where it regulates gene transcription. Increased muscle NFκB activity is thought to be involved in the muscle loss seen with cancer cachexia, muscle disuse, and denervation (7–10). NFκB activity also increases during aging in muscle and may contribute to sarcopenia (4,11). Yet, the role of NFκB in aging-related changes to skeletal muscle is questionable; we recently reported that, contrary to our prediction, NFκB inactivation resulted in a muscle atrophy phenotype (12).

Similar to alterations in NFκB activity, disruptions in mitochondrial bioenergetics are associated with reduced skeletal muscle mass and function under pathological conditions (13–15). For instance, activation of upstream NFκB pathway regulators such as toll-like receptor 4 and tumor necrosis factor (TNF) receptor reduces mitochondrial respiration and inhibits the switch from glycolytic to a more oxidative phenotype in muscle cells (14,15). Genetic activation of the canonical NFκB pathway in myotubes by adenoviral mediated IKK-β overexpression decreases levels of mitochondrial proteins, fatty acid oxidative enzymes, and nuclear receptors (16). Consistent with these in vitro studies, in vivo administration of the toll-like receptor 4 agonist endotoxin, decreases skeletal muscle peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC1α) protein and citrate synthase (CS) activity, a marker of mitochondrial content (17). Collectively, these findings suggest that NFκB may have a negative affect on mitochondrial function in muscle. To address this question, we examined the effects of persistent NFκB inactivation on mitochondrial function (oxygen consumption and reactive oxygen species [ROS] production) in permeabilized fibers from different muscles across the life span. We hypothesized that suppression of NFκB signaling would protect against perturbations to mitochondrial function seen with aging.

Methods

Animal Experiments

Initial generation and characterization of the muscle-specific IκBα super-repressor (MISR) mice was conducted by Cai and colleagues (7). MISR mice were on a C57BL/6J background, and littermates without the transgene serve as their controls. Mice were fed standard chow and housed in an animal room maintained at 23°C with a 12 hours light/12 hours dark cycle for all studies. Male mice were studied and no difference in life span was seen between wild-type (WT) and MISR mice (Supplementary Figure 4). All procedures were approved by the Institutional Animal Care and Use Committee at the University of Texas Health Science Center at San Antonio.

RNA Sequencing

Procedures for RNA isolation, assessment of purity, fragmentation, library preparation, sequencing, and alignment were described previously (12).

Muscle Mass and Cross-Sectional Area Measurements

Mice were anesthetized with isoflurane and euthanized by cervical dislocation. Muscles of the hind limb were dissected with tendons and fascia removed before weighing. Muscle wet-weights were normalized to body weight. For cross-sectional area measurements, paraffin-embedded muscles were sectioned from the midbelly at 10 µm and stained with laminin (L9393, Sigma, St. Louis, MO). Images were visualized and captured with Nikon Element software (Nikon Inc., Melville, NY). Cross-sectional area was quantitated with ImageJ64 software (ImageJ, National Institutes of Health). All experiments were conducted in a blinded fashion.

Transmission Electron Microscopy

Initial fixation was in phosphate buffered formaldehyde/1% glutaraldehyde for 24 hours. The samples were then washed in 0.1 M phosphate buffer followed by secondary fixation in 1% osmium tetroxide (1% Zetterqvist’s buffer) for 60 minutes. The samples were washed in 0.1 M phosphate buffer. Dehydration was accomplished by treating the samples with a graded series of ethyl alcohols followed by immersion in propylene oxide. Infiltration of resin into the samples was accomplished by initially placing the specimens in a 1:1 mixture of propylene oxide/resin overnight. The specimens then were placed in 100% resin for 120 minutes under 25 psi vacuum. The specimens were placed in plastic vials containing the resin and polymerized in an 85°C oven for 90 minutes or an incubator at 70°C overnight. Thick (1 µm) samples were cut and stained with toluidine blue dye to select appropriate areas for thin sectioning. Ultrathin sections were placed on copper grids and stained with uranyl acetate followed by lead citrate. The sections were viewed using a JEOL JEM 1230 transmission electron microscope. Mitochondria were counted in 10,000× magnification images from ~14 month old WT and MISR mice. Between one and three images were counted per animal and mean mitochondrial number was used for statistical analysis. Mitochondrial morphology was assessed in a blinded manner.

High-Resolution Respirometry and ROS Production

High-resolution respirometry (HRR) was conducted on a G-model Oxygraph-2k machine (Oroboros Instruments, Innsbruck, Austria). Reagents, muscle fiber preparation, and Substrate Uncoupler Inhibitor Titration (SUIT) protocols used for HRR have been previously described (18) with some modifications (Supplementary Figure 3) To allow for detection of ROS, MiRO5 (+creatine) running media that lacks catalase was used. Also, antimycin A was sufficient to block complex 3 activity; therefore, myxothiazol was not added in our SUIT protocol. ROS production was detected using the O2k-Fluo (Oroboros Instruments) and a modified Amplex Red assay. After daily background correction, chambers were closed and 20 µM final concentration Amplex Red ultra (Life Technologies, A36006), 1 U/mL horseradish peroxidase (Sigma-P8250), and 5 U/mL superoxide dismutase (Sigma-S8409) were added to the wells. After the amp-slope stabilized 0.1 µM H₂O₂ (Sigma-95321) was added for calibration. Throughout the entire assay, repeated injections of 0.1 µM H₂O₂ were used for calibration including one injection after the muscle tissue was added to correct for autofluorescence. All subsequent 0.1 µM H₂O₂ injections were used to control for changing concentrations of Amplex Red during the assay. Samples for HRR and ROS were conducted in duplicate and means were used for the final analyses. To eliminate confounds from tissue damage or hyperpermeabilization, only samples with stable oxygen flux and Raw-amps were included in the final analysis.

CS Activity

After HRR experiments, tissues were snap frozen and lyzed using a TissueLyser II (Qiagen) in 100 µL lysis buffer (20 mM Tris, pH 7.5, 5 mM EDTA, 10 mM Na₃PO₄, 100 mM NaF, 2 mM Na₃VO₄, 1% Nonidet P-40, 10 μM leupeptin, 3 mM benzamidine, 10 μg/mL aprotinin, and 1 mM phenylmethylsulfonyl fluoride). Protein concentration was measured by Bradford assay and samples diluted in 0.1 M TRIS (Sigma-T3253) pH 8.1. Stock solutions of 1 mM 5,5-dithio-bis-(2-nitrobenzoic acid) DTNB (Sigma-D8130) and 10 mM oxalo acetic acid (Sigma-O4126) were prepared in 0.1 M TRIS pH 8.1. Assay buffer consisted of 0.25 mg/mL acetyl-CoA (Sigma-A2181), 1 mL DTNB stock solution, and 8 mL 0.1M TRIS. To each well, 10 µL of diluted samples and 150 µL of assay buffer were added. The reaction was started with the addition of 10 µL of oxalo acetic acid stock solution and absorbance was measured immediately at 412 nM and every 30 seconds for 5 minutes after the initial reading. All samples were run in triplicate.

Data Analyses and Statistical Tests

All data were analyzed by t-test, chi-square, or 2-way between-subject ANOVA with Tukey post hoc comparisons depending on the number of variables and groups in each analysis. HRR and ROS production were analyzed by 2-way (genotype × complex activity) between-subject ANOVA. Survival curves were analyzed using Log-rank (Mantel-Cox) Test. Data were analyzed using Prism 7 (GraphPad Software, La Jolla, CA) and are represented as means and standard error of the means. All experiments and data analyses were conducted in a blinded fashion.

Results

Canonical NFκB Pathway Inactivation Reduces Early Life Muscle Mass Accrual

Using mice that overexpress a muscle-specific IκBα super-repressor transgene (MISR), we previously reported that NFκB inactivation altered the expression of many genes involved in muscle development including MET, MAP3K14, MYOD1, IGFBP5, and MSTN (12). Therefore, the current study tested whether MISR mice have abnormal muscle mass in early life. At 1 month of age, wet muscle mass of quadriceps, gastrocnemius, extensor digitorum longus, and tibialis anterior muscles was significantly smaller in MISR mice, but no difference in soleus muscle weights were observed between genotypes (Figure 1A–E). At 4 months of age, no muscle groups measured were statistically different between genotypes except gastrocnemius muscles which were still significantly smaller in MISR mice (Supplementary Figure 1 and Figure 1F). In line with these findings, the cross-sectional area of MISR mice gastrocnemius muscles was significantly reduced at 3–6 months of age (Figure 1G and H).

Figure 1.

Suppression of canonical NFκB signaling results in smaller muscles early in life. (A–E) Muscle weights at 1 month of age were normalized to body weight (n = 14–15 per group). Data were analyzed by unpaired t-test. (F) Gastrocnemius muscle wet-weights normalized to body weights during aging. (G) Representative laminin stain and (H) quantification of cross-sectional area (CSA) from gastrocnemius muscles of WT and MISR mice during aging (n = 4 per group). Data were analyzed by 2-way between-subject ANOVAs with Tukey post hoc comparisons and are represented as means and SEM. * indicates p < .05; ** indicates p < .01. MISR = muscle-specific IκBα super-repressor; WT = wild type.

Mitochondrial Function in Muscle Declines During Aging

We assessed mitochondrial function in different muscles throughout life (ages 4, 13–15, and 30–33 months) using HRR with simultaneous measurements of ROS production rates. In WT mice, oxygen consumption rates decreased with aging in gastrocnemius and soleus muscles (Figure 2A and C). The reduced oxygen consumption rates were accompanied by increased substrate-driven ROS production at CI+CII_P (Figure 2B and D), whereas ROS production after inhibition of the electron transport chain (ETC; with rotenone, malonic acid, and antimycin A) was significantly higher in young compared to middle-aged and old animals (Figure 2B and D).

Figure 2.

Mitochondrial function declines at different rates in skeletal muscles during aging. (A, C) Oxygen flux normalized to citrate synthase activity and (B, D) reactive oxygen species (ROS) production rates in gastrocnemius and soleus muscles from WT mice during aging, respectively (n = 7 per group). ETF_L is leak respiration. ETF_P is fat oxidation coupled to ATP production. CI_P is complex I and CI+CII_P is complex I and II linked respiration coupled to ATP production. CI+CII_E is uncoupled or maximum respiration. CII_E is complex II respiration in the uncoupled state. Y (4 months old), M (13–15 months old), O (30–33 months old). All data were analyzed by 2-way between-subject (age × complex activity) ANOVA with Tukey post hoc comparisons and are represented as means and SEM. * indicates p < .05.

NFκB Suppression Decreases Mitochondrial Respiration in Muscle From Young Mice, but Has Differential Effects During Aging

In gastrocnemius and soleus muscles from young mice (4 months old), NFκB inactivation reduced mitochondrial oxygen consumption (Figure 3A and D) without altering substrate-driven ROS production (Supplementary Figure 2). Changes in oxygen consumption were driven mainly by complex I (CI_P) and complex I-II (CI+CII) linked respiration in gastrocnemius (Figure 3A), but only CI+CII linked respiration in soleus muscles (Figure 3D).

Figure 3.

NFκB pathway inactivation decreases mitochondrial respiratory capacity in young muscles but has differential effects during aging. (A–C) Oxygen fluxes normalized to citrate synthase activity in gastrocnemius and (D–F) soleus muscles during aging (n = 7 per group). ETF_L is leak respiration. ETF_P is fat oxidation coupled to ATP production. CI_P is complex I and CI+CII_P is complex I and II linked respiration coupled to ATP production. CI+CII_E is uncoupled or maximum respiration. CII_E is complex II respiration in the uncoupled state. All data were analyzed by 2-way (genotype × complex activity) between-subject ANOVA with Tukey post hoc comparisons and represented as means and SEM. * indicates p < .05.

In contrast to the effects seen in young animals, NFκB suppression did not alter oxygen consumption rates in gastrocnemius (Figure 3B) and soleus (Figure 3E) muscles from middle aged (13–15 months old) mice. ROS production in both muscle groups was also unaffected by NFκB inactivation at these ages (Supplementary Figure 2). Similar findings were observed in gastrocnemius muscles from old (aged 30–33 months old) mice (Figure 3C and Supplementary Figure 2), whereas soleus muscles from MISR mice (30–33 months old) had increased oxygen consumption rates compared to WT (Figure 3F), without a concomitant increase in ROS production (Supplementary Figure 2).

Aging and NFκB Inactivation Independently Increase Markers of Mitochondrial Mass/Content in Muscle

To determine whether differences in mitochondrial function were attributed to changes in mitochondrial quality or content/mass, we measured CS activity. Aging was associated with higher CS activity in gastrocnemius (age main effect p = .006, Figure 4D), but not in soleus muscle (Supplementary Figure 3). NFκB suppression also led to elevated CS activity in gastrocnemius (genotype main effect p = .03, Figure 4D), but not soleus muscles (Supplementary Figure 3). Transmission electron microscopy images of gastrocnemius muscles revealed that MISR mice had an increase in mitochondria counts, and these mitochondria were modestly enlarged compared with WT mice (Figure 4A–C). Along with the increased CS activity, these findings are suggestive of increased mitochondrial mass/content. Gastrocnemius muscles from young MISR mice had a modest increase in mitochondrial content/mass (as measured by CS activity) despite lower respiration rates, indicating that differences in respiration were not a result of changes in content/mass.

Figure 4.

Markers of mitochondrial content and mass increase with aging and NFκB suppression. (A) Representative image of gastrocnemius muscles from 14-month-old WT and MISR mice using transmission electron microscopy (TEM). Arrows highlight differences in mitochondrial size and quantity between WT and MISR (B) Analysis of sarcomeric and mitochondrial morphology (n = 6–9 per group). Data were analyzed by chi-square and represented as percent of animals with each phenotype. (C) Quantification of mitochondria numbers in 10,000× TEM images for WT and MISR mice. Analyzed by unpaired t-test. (D) Citrate synthase (CS) activity in gastrocnemius muscles (n = 13–14 per group). Data were analyzed by 2-way (genotype × age) between-subject ANOVA with Tukey post hoc comparisons and are represented as means and SEM. * indicates p < .05. MISR = muscle-specific IκBα super-repressor; WT = wild type.

Expression Patterns of Genes Involved in Cellular Metabolism Change With Aging and NFκB Inactivation

To determine whether the differences observed in muscle mitochondrial function between WT and MISR mice may be the result of NFκB-mediated transcriptional regulation of metabolism genes, we performed RNA sequencing on quadriceps muscles from WT and MISR mice at 3–6, 12–18, and 33–36 months of age. Advancing age and NFκB suppression both caused distinct alterations to transcription of cellular metabolism genes. Aging decreased the expression of genes that encode for the nuclear receptor ERRA and the sirtuins SIRT3 and SIRT5, but increased expression of sirtuins SIRT1 and SIRT2, the AMPK subunits PRKAA2/PRKAB1 and the AMPK upstream kinase LKB1 (Figure 5A–C). NFκB inactivation decreased the expression of the transcriptional coactivator PPARGC1A (Figure 5A); however, other important mediators of mitochondrial function were increased in MISR mice including the transcriptional coactivator PPARGC1B, the transcription factor TFEB, the AMPK subunit PRKAG2, and the AMPK upstream kinase LKB1 (Figure 5A and C). Despite increased expression of some mitochondrial biogenesis genes and mitochondrial content/mass, aging reduced the expression of nearly all complex I-IV ETC subunits in WT and MISR mice (Figure 5D).

Figure 5.

Aging and NFκB inactivation alter the expression patterns of cellular metabolism genes in skeletal muscle. (A) mRNA expression for transcriptional regulators of cellular metabolism, (B) AMPK/LKB1 subunits and (C) Sirtuins in WT and MISR mouse quadriceps muscles during aging (n = 5–6 per group). Y (3–6 months old), M (12–18 months old), O (33–36 months old). Data were analyzed by 2-way between-subject ANOVAs for each gene with Tukey post hoc comparisons. $ indicates significant age main effect, & indicates significant interaction, # indicates significant genotype main effect, and * represents significant post hoc comparison p < .05. Data are represented as means and SEM. (D) Heat map with each cell as mean mRNA expression for electron transport chain subunits (ETC) in WT and MISR mice throughout age (n = 5–6 per group). MISR = muscle-specific IκBα super-repressor; WT = wild type.

Discussion

We have previously reported that muscle-specific suppression of canonical NFκB signaling (in MISR mice) accelerated muscle wasting during aging in vivo and pushed myogenic precursor cells toward differentiation in vitro (12). In this work, we found that MISR mice had smaller muscles during development and young adulthood, indicating that the loss of muscle mass and function previously reported in MISR mice is due in part to impaired muscle development. These findings are in line with previous research showing that deletion of the toll-like receptor adapter protein, MyD88, also causes reduced muscle mass in early life due to alterations in muscle development (19). Furthermore, Bakkar and colleagues (20) reported a shift in the distribution of myofiber size toward smaller fibers in 1-month-old mice with NFκB inactivation (mice null for both TNFα and p65).

Findings regarding the effects of aging on skeletal muscle mitochondrial function are inconsistent (21,22). The discrepancies may be attributed to methodological differences including muscles studied, fiber-type composition (oxidative or glycolytic), and techniques employed to measure mitochondrial function (eg, isolated mitochondria vs permeabilized fibers). Here, we combined RNA sequencing data from mixed fiber quadriceps muscles with HRR and simultaneous ROS production rate measurements in permeabilized fibers from the oxidative soleus and mixed fiber gastrocnemius muscles. Aging caused a decline in mRNA expression of nearly all mitochondrial ETC subunits in both WT and MISR mice quadriceps muscles, indicating that these changes are NFκB independent. Recent work has shown that ETC subunits are also reduced at the protein level during aging (23). Similarly, aging decreased mitochondrial respiration rates in gastrocnemius and soleus muscles from WT animals. The decline in respiration was more pronounced in soleus muscles (particularly in response to fatty acid substrates), suggesting that muscle and/or fiber type may be an important factor in the severity of decline in mitochondrial respiration during aging. This is in line with previous work showing age-dependent and fiber-type-specific differences in protein synthesis/translation, specific force, and muscle mass, in a rat model of hind limb unloading (24,25). NFκB inactivation enhanced respiration rates in soleus but not gastrocnemius muscles of old mice, indicating that NFκB may have a greater impact on mitochondrial function in oxidative compared to glycolytic fibers. We previously reported that fiber-type composition is unaltered by NFκB inactivation (12); therefore, differences in mitochondrial function between WT and MISR mice are not the result of alterations in the ratio of glycolytic to oxidative fibers.

Gastrocnemius and soleus muscles from young WT mice had the highest ROS production rates after ETC inhibition (with rotenone, malonic acid and antimycin A) compared to all other age groups (Figure 2). ROS produced under these conditions is primarily driven by the III_Qo site, which has the highest capacity for ROS production in mitochondria and is the only site that releases ROS into the cytosol (26). Cytosolic ROS can activate pro-longevity pathways such as NRF2 and FOXO, which have important roles in mitochondrial and protein homeostasis (27,28). Because ROS production at the III_Qo site is reduced during aging in muscles of WT mice, ROS-mediated activation of these pathways may be inhibited and contribute to aging-related declines in muscle mass and function. Contrary to ROS production at the III_Qo site, aging increased substrate-driven ROS production in both muscles regardless of genotype, demonstrating that these effects are also NFκB independent.

The decreased respiration rates and ETC subunits expression during aging occurred despite a modest increase in mitochondrial content/mass. Consistent with these findings, Leduc and colleagues (29) also observed increased mitochondrial mass/content in aged mice. These changes may have resulted from elevated expression of mitochondrial biogenesis regulators (LKB1, SIRT1/2, PRKAA2, and PRKAB1) during aging. Similarly, NFκB inactivation increased mitochondrial content/mass and biogenesis genes (TFEB, PPARGC1B, LKB1, and PRKAG2) while decreasing respiration rates. Increased skeletal muscle mitochondrial biogenesis may result as a compensatory mechanism to offset declining mitochondrial quality and function, similar to observations in heart muscle during aging (30). However, recent work suggests that these compensatory changes during aging (increased size and number) may lead to enlarged and dysfunctional mitochondria (31), not healthy mitochondrial expansion. Future studies also will be important to determine whether the changes in mitochondrial function and mass/content seen here are related to differences in mitochondrial protein synthesis and turnover.

We have previously reported elevated markers of autophagy in MISR mouse muscles (12). Here we found that these muscles also had higher expression of the master autophagy-lysosome regulator TFEB, suggesting that NFκB is a negative regulator of TFEB expression. Induction of the TFEB-autophagy pathway can result in muscle wasting (32–34). Therefore, elevated TFEB expression in MISR mice may contribute to their smaller muscle size.

In conclusion, canonical NFκB signaling is involved in normal skeletal muscle development and growth. The extent to which NFκB suppression alters mitochondrial function is age dependent (most pronounced in early life) and muscle specific. Measures of mitochondrial function and muscle mass are tightly associated in both genotypes and across the life span. Finally, these data suggest that targeting skeletal muscle NFκB activity in early life to prevent aging-related pathology may be harmful, although targeting it in late life (once muscles are fully developed) may not necessarily be detrimental.

Funding

N.M. was supported by grants from the National Institute of Health (R01-DK80157, R01-DK089229) and from the American Diabetes Association (7-13-GSK). J.M.V. was supported by the Biology of Aging T32 Training Grant (T32 AG021890). This research also was supported by the San Antonio Nathan Shock Center of Excellence on Aging Biology (P30 AG013319) and the San Antonio Claude D. Pepper Older Americans Independence Center (P30 AG044271).

Conflict of Interest

None declared.