Systemic delivery of a mitochondria targeted antioxidant partially preserves limb muscle mass and grip strength in response to androgen deprivation

Abstract

Muscle mass is important for health. Decreased testicular androgen production (hypogonadism) contributes to the loss of muscle mass, with loss of limb muscle being particularly debilitating. Androgen replacement is the only treatment, which may not be feasible for everyone. Prior work showed that markers of reactive oxygen species and markers of mitochondrial degradation pathways were higher in the limb muscle following castration. Therefore, we tested whether an antioxidant preserved limb muscle mass in male mice subjected to a castration surgery. Subsets of castrated mice was treated with resveratrol (a general antioxidant) or MitoQ (a mitochondria targeted antioxidant). Relative non-castrated control mice, lean mass, limb muscle mass, and grip strength were partially preserved only in the castrated mice treated with MitoQ. Independent of treatment, markers of mitochondrial degradation pathways remained elevated in all castrated mice regardless of treatment. Therefore, a mitochondrial targeted antioxidant may partially preserve limb muscle mass during hypogonadism.

I. INTRODUCTION

Maintaining a critical amount of muscle mass is important for physical function and overall health (Powers, Lynch, Murphy et al., 2016,Srikanthan and Karlamangla, 2014,Srikanthan, Horwich and Tseng, 2016). In males, a decrease in testicular androgen production (hypogonadism) contributes to the loss of muscle mass that occurs during aging and various pathological conditions (Bhasin, Storer, Berman et al., 1997,Bhasin, Woodhouse, Casaburi et al., 2005). Indeed, a longitudinal study showed that older males with higher levels of testosterone maintained lean mass over a 4.5 year period (LeBlanc, Wang, Lee et al., 2011). While hypogonadism impacts various muscle groups, the loss of limb muscle is particularly important as these muscles comprise a majority of total muscle mass (Kim, Wang, Heymsfield et al., 2002), and they are the primary muscle involved with functional tasks such as walking and climbing stairs. Androgen replacement is the only effective therapy to blunt the loss of limb muscle in hypogonadal individuals (Ferrando, Sheffield-Moore, Yeckel et al., 2002,Ferrando, Sheffield-Moore, Paddon-Jones et al., 2003), but it is not a universal option due to side effects (e.g. enhanced malignant tumor growth and adverse cardiac events) (Metzger and Burnett, 2016,Fowler and Whitmore, 1982,Amos-Landgraf, Heijmans, Wielenga et al., 2014). Therefore, it is imperative to define new treatments that can preserve limb muscle mass in response to androgen deprivation.

It is thought that androgens regulate muscle mass by signaling through the androgen receptor (Ophoff, Van Proeyen, Callewaert et al., 2009,Serra, Sandor, Jang et al., 2013). While this appears to be true for certain muscles (e.g. levator ani) (Serra et al., 2013), the androgen receptor is dispensable for regulating mass of the limb muscles (Altuwaijri, Lee, Chuang et al., 2004,Ueberschlag-Pitiot, Stantzou, Messéant et al., 2017). For example, the presence of androgens themselves, not a functional androgen receptor, regulated mass of the tibialis anterior (TA) muscle in rodents (Ueberschlag-Pitiot et al., 2017). More recently, myofiber-specific deletion of the androgen receptor did not prevent androgen-mediated growth of limb muscles in female mice (Sakakibara, Yanagihara, Himori et al., 2021). While androgens do not appear to regulate limb muscle mass via the androgen receptor, the pathways by which androgens mediate limb muscle mass, particularly when androgen production is compromised, remain almost completely unknown, limiting therapeutic options.

Because the TA muscle mass is sensitive to androgens in an androgen receptor independent manner (Ueberschlag-Pitiot et al., 2017,Sakakibara et al., 2021), our laboratory has characterized many of the intramuscular signaling events that change in this muscle in response to androgen deprivation (Steiner, Fukuda, Rossetti et al., 2017,Rossetti, Steiner and Gordon, 2018,Rossetti, Esser, Lee et al., 2019,Rossetti, Tomko and Gordon, 2020). Specifically, we showed that markers of impaired mitochondrial quality were elevated in the TA in response to androgen deprivation including higher reactive oxygen species (ROS) content (e.g. H₂O₂) and ROS reactive byproducts (e.g. 4-hydroxynonenol; 4HNE). These markers coincided with markers of mitochondrial degradation pathway activation comprising BCL2/adenovirus E1B 19 kDa protein-interacting protein 3 (BNIP3) and the PTEN-induced kinase 1 (PINK1)/PARKIN pathways, which clear damaged or dysfunctional mitochondria via the autophagy/lysosomal system (Serra et al., 2013,Rossetti et al., 2018,Rossetti et al., 2019). Accordingly, we and others have shown markers of autophagy/lysosomal pathway were also elevated in the limb muscle in response to androgen deprivation, including an increase in the ratio of the lipidated (II) to non-lipidated (I) forms of microtubule-associated protein light chain 3 (LC3) and a decrease in p62 protein content (Steiner et al., 2017,Rossetti et al., 2019,Rossetti and Gordon, 2017). Although these markers of impaired mitochondrial quality and increased ROS were not present at all time points throughout the day, these markers were inversely related to TA mass (Rossetti et al., 2018), suggesting they could be contributing to limb muscle atrophy. The markers of increased ROS were of particular interest given that high levels of ROS, especially ROS generated by mitochondria, contribute to muscle atrophy in other conditions (e.g. limb muscle immobilization and mechanical ventilation) (Min, Smuder, Kwon et al., 2011,Powers, Hudson, Nelson et al., 2011). Therefore, the purpose of this study was to determine whether systemic administration of an antioxidant preserved limb muscle mass in response to androgen deprivation. We provide evidence that systemic administration of an antioxidant that targets the mitochondria partially preserved limb muscle mass and grip strength in response to androgen deprivation.

2. MATERIALS and METHODS

2.1. Experimental Design and Castration Surgery

2.1.1. Study #1: Antioxidant Gene Expression

The TA muscle samples analyzed for diurnal antioxidant gene expression were generated from a study previously conducted by our laboratory (Rossetti et al., 2019). In brief, male C57Bl/6NHsd mice (14 weeks of age) were purchased from Envigo (Indianapolis, IN) and subjected to either sham or castration surgery. Mice recovered for 8 weeks prior to sacrifice. At sacrifice, tissues were harvested from a subset of mice from each treatment group every 4 hr beginning at the onset of the dark cycle (N=3/group/time point). Food and water were consumed ad libitum throughout the data collection. For sacrifice during the dark cycle, mice were transported in a black Tupperware container to the surgical suite, where they were lightly anesthetized with isoflurane under dim red light. Mice were euthanized via cervical dislocation, lights were turned on, the TA muscles were extracted, and flash frozen in liquid nitrogen. The Animal Care and Use Committee at Florida State University approved these procedures. This data set allowed us to observe expression patterns of antioxidant genes across the day as opposed to measurement at a single time point.

2.1.2. Study #2: Antioxidant Treatment Study

Male, C57Bl/6NHsd mice (12 weeks of age) were purchased from Envigo (Indianapolis, IN). Upon arrival, all mice were housed individually on a reverse 12:12 hr light-dark cycle for 10 days with ad libitum access to food (5001 rodent chow (LabDiet, St. Louis, MO) and water. Mice remained on the reverse light-dark cycle throughout the experiment. After the acclimation, baseline forelimb grip strength was assessed, and mice were randomized into 4 groups of equal body weight (N=9-10 per group). Three groups were subjected to a castration surgery to effectively stop testicular androgen production while the fourth group was subjected to a sham surgery where testicular androgen production remained intact. All mice were given buprenorphine (0.05 mg/kg) immediately following surgery and again 5 hr later to reduce postoperative pain. Mice recovered for 3 days prior to beginning treatment. One group of castrated mice began consuming a resveratrol enriched diet (resveratrol manufactured by Chengdu Wagott Bio-Tech Co., Ltd., kindly provided by Barrington Nutritionals, and incorporated into 5001 rodent chow; LabDiet, St. Louis, MO) at a concentration of 0.8 mg/g of food, resulting in a daily consumption of 15.9 ± 0.24 mg/kg/day. This dose has been shown to act as an antioxidant in skeletal muscle (Jackson, Ryan, Hao et al., 2010). A second group of castrated mice began consuming the mitochondrial targeted antioxidant, MitoQ (Kindly provided by MitoQ Limited, Auckland, New Zealand) at a concentration of 250 μM diluted in the drinking water (Gioscia-Ryan, LaRocca, Sindler et al., 2014). The MitoQ group consumed 5.05 ± 0.5 ml of water per day on average, which is in line with normal values for mice of this strain (Bachmanov, Reed, Beauchamp et al., 2002,Gelineau, Arruda, Hicks et al., 2017). The remaining two groups (one castrated and one sham) received untreated food and water as controls. All mice were allowed ad libitum access to food and water. Three weeks after beginning treatment, all four groups of mice received an intraperitoneal injection of 99% deuterium oxide (D₂O, cat. #151882, Sigma Aldrich, St. Louis, MO) equivalent to 5% of the body water pool followed by continued consumption of 8% D₂O in the drinking water until sacrifice for measurement of myofibrillar and mitochondrial protein fractional synthetic rates (Miller, Reid, Price et al., 2020). Six weeks after beginning treatment, all mice were tested for body composition (EchoMRI, Houston, TX) and retested for grip strength. Seven weeks post-surgery, mice were euthanized by cervical dislocation between 1300-1500 hr (dark cycle) under light isoflurane anesthesia in the freely fed state. Sacrifice in the freely fed state was chosen because our previous work showed that nutrient consumption increased markers of mitochondrial degradation pathway activation in androgen deprived male mice. These markers were increased when castrated mice were refed after an overnight fast (Steiner et al., 2017,Rossetti et al., 2018), or when these pathways were assessed in the muscle of freely fed castrated mice at the time of day when food is being consumed (i.e. the dark cycle) (Rossetti et al., 2019). Similar outcomes were also observed for markers of general autophagy under the same experimental paradigms (Steiner et al., 2017,Rossetti et al., 2018,Rossetti et al., 2019,Rossetti et al., 2020). Sacrificing mice in the freely fed state is also not expected to affect our measure of myofibrillar/mitochondrial protein synthetic rates as this measure assessed incorporation of deuterium labelled amino acids into the muscle protein pool across multiple weeks. Whole blood was collected in EDTA treated syringes, and the plasma fraction was isolated by centrifugation at 2000 x g for 10 min at 4 °C. Tissues were then extracted, frozen in liquid nitrogen, and stored at −80°C until analysis. The Animal Care and Use Committee at Florida State University approved these procedures.

2.2. In Vivo Grip Strength

Forelimb grip strength was tested using a Chatillon force gauge (Ametek Inc., Largo, FL) in a manner similar to that described previously (Gordon, Rossetti and A, 2021). The protocol included 2 days of acclimation and 1 day of testing. Each session consisted of 3 successful trials per mouse with a ~10-15 s recovery period between attempts. Trials were considered successful if the mouse used an overhand grip of both forelimbs to generate force. If a trial was unsuccessful, results were discarded, and the trial was repeated. To ensure consistency, the same researcher conducted the acclimation and testing, which were always conducted at the same time of day (beginning of the active cycle). The experimental groups were blinded to the researcher during testing. The mean force (N) generated from 3 successful trials was used for analysis. Grip strength was expressed relative to lean mass and as a percent change pre- to post-treatment.

2.3. Determination of Myofibrillar and Mitochondrial Fractional Synthetic Rates

The tibialis anterior (TA) muscles were fractionated as previously described (Drake, Peelor, Biela et al., 2013,Drake, Bruns, Peelor et al., 2014). The TA was homogenized 1:10 in isolation buffer (100 mM KCl, 40 mM Tris–HCl, 10 mM Tris base, 5 mM MgCl₂, 1 mM EDTA, 1 mM ATP, pH 7.5) with phosphatase and protease inhibitors (Halt, Thermo Fisher Scientific, Waltham, MA, USA) using a bead homogenizer (Next Advance Inc., Averill Park, NY, USA). Subcellular fractions were then isolated by differential centrifugation (Drake et al., 2013,Drake et al., 2014). Protein pellets were isolated and purified, then incubated in 250 μl of 1 M NaOH for 15 min at 50°C while slowly mixing. Protein was hydrolyzed via incubation for 24 hr at 120°C in 6 M HCl. The pentafluorobenzyl-N,N-di(pentafluorobenzyl) derivative of alanine was analyzed on an Agilent 7890A Gas Chromatograph (Thermo Fisher Scientific) coupled to an Agilent 5975C Mass Spectrometer (Thermo Fisher Scientific) as previously described (Drake et al., 2013,Drake et al., 2014). The newly synthesized fraction of proteins was calculated from the true precursor enrichment using plasma analyzed for D₂O enrichment and adjusted using a mass isotopomer distribution analysis (Busch, Kim, Neese et al., 2006). Body water enrichment of D₂O was performed using published procedures (Brown, Lawrence, Ahn et al., 2020). Briefly, 120 μL of serum was placed into the inner well of O-ring screw cap and inverted on an 80°C heating block for overnight distillation. Distilled samples were diluted 1:300 in ddH₂O and analyzed on a liquid water isotope analyzer (Los Gatos Research, Los Gatos, CA, USA) against a standard curve prepared with 0–12% D₂O (Brown et al., 2020).

2.4. RNA Extraction, cDNA Synthesis and RT-PCR

TA muscles (~15-20 mg) were homogenized in 600 μl of Zymo Tri Reagent (Irvine, CA), and RNA was isolated using a Zymo RNA Miniprep extraction kit (cat. #R2071) with on-column DNase treatment (Irvine, CA). RNA quantity was determined spectrophotometrically by the 260-to-280 nm ratio. cDNA was synthesized from 1 μg of total RNA using a High Capacity cDNA Reverse Transcription Kit (cat. #4368814; Thermo Fisher Scientific, Waltham, MA). RT-PCR was conducted on a QuantStudio3 (Thermo Fisher Scientific) RT-PCR thermal cycler using PowerUp Sybr Green Master Mix (cat. #A2742; Thermo Fisher Scientific) or TaqMan Fast Advanced Master Mix (cat. #4444557; Thermo Fisher Scientific). The conditions for RT-PCR with Sybr Green included an initial 2 min at 50 °C and 2 min at 95 °C, followed by 40 cycles that included a 15 s denaturation step at 95 °C, a 15 s annealing step at 55 °C, and a 1 min extension step at 72 °C within each cycle. A melt curve analysis was performed for each primer pair to ensure that a single product was amplified, and the product sizes for each primer pair were verified via agarose gel electrophoresis prior to experimentation. Primer sequences for all Sybr Green RT-PCR reactions are listed in Table 1. Measurement of transcription factor A, mitochondrial (Tfam) (assay ID Mm00447485_ml), nuclear respiratory factor 1 (Nrf1) (assay ID Mm01135606_m1), and Pink1 (assay ID Mm00550827_m1) were quantified using TaqMan predesigned primer probes according to the manufacturer-recommended conditions for the QuantStudio3. Relative expression levels of all genes were normalized using the delta-delta Ct (2^−ΔΔct) method. Ribosomal Protein Lateral Stalk Subunit P0 (Rplp0) was used as the internal control as Rplp0 expression was not affected by either castration or treatment.

Table 1.

Primer sequences for RT-PCR using Sybr Green

Gene Symbol	Forward (5’-3’)	Reverse (5’-3’)	Amplicon Size (bp)
Sod1	GAGACCTGGGCAATGTGACT	TTGTTTCTCATGGACCACCA	125
Sod2	CCGAGGAGAAGTACCACGAG	GCTTGATAGCCTCCAGCAAC	174
Catalase	TTGACAGAGAGCGGATTCCT	AGCTGAGCCTGACTCTCCAG	179
Gpx1	CCGACCCCAAGTACATCATT	TCGATGTCGATGGTACGAAA	133
Gsr	CACGACCATGATTCCAGATG	CAGCATAGACGCCTTTGACA	161
Prdx3	TCATCTTGCCTGGATCAACA	GACACTCAGGTGCTTGACGA	184
Txnip	ACCAGTGTCTGCCAAAAAGG	TTGGCAAGGTAAGTGTGTCG	198
Nox2	GCTTGTGGCTGTGATAAGCA	CTTGAGAATGGAGGCAAAGG	162
Nox4	TCTCAGGTGTGCATGTAGCC	TTCTGGGATCCTCATTCTGG	114
Pgc-1α	AAGACGGATTGCCCTCATTT	AGTGCTAAGACCGCTGCATT	191
Rplp0	CAACCCAGCTCTGGAGAAAC	GTTCTGAGCTGGCACAGTGA	169

2.5. Western blot Analysis

Western blotting was conducted as previously described (Steiner et al., 2017). Whole muscle protein from the TA was extracted by glass-on-glass homogenization in 10 volumes of buffer (10 μl/mg of muscle) consisting of 50 mM HEPES (pH 7.4), 0.1% Triton-X 100, 4 mM EGTA, 10 mM EDTA, 50 mM Na₄P₂O₇, 100 mM β-glycerophosphate, 25 mM NaF, 5 mM Na₃VO₄, and 10 μl/ml protease inhibitor cocktail (cat. #P8340, Sigma-Aldrich). Muscle extract was centrifuged for 10 min at 10,000 g at 4°C, and the soluble protein content in the supernatant was quantified via the Bradford method. After quantification, soluble proteins were diluted to the same concentration in 2X Laemmli buffer. Proteins in the supernatant were separated on 4-20% Bio-Rad Tris-Glycine Criterion precast gels (Hercules, CA) and transferred to PVDF membranes. Ponceau-S staining was used to assess effective transfer and equal protein loading (Gilda and Gomes, 2013). Membranes were blocked with 5% nonfat dried milk in Tris-buffered saline plus 0.1%-Tween20 (Tris-buffered saline-Tween 20). Membranes were then incubated overnight at 4°C with antibodies against BNIP3 (cat. no. 3769), PARKIN (cat. no. 2132), LC3B (cat. no. 2775), and p62 (cat. no. 5114), which were all purchased from Cell Signaling Technologies (Danvers, MA). After incubation with secondary antibody (Bethyl Laboratories; cat. no. A120–101P), the antigen-antibody complex was visualized by enhanced chemiluminescence using Clarity reagent (Bio-Rad) on a Bio-Rad ChemiDoc Touch imaging system. The pixel densities from all blots were quantified as the ratio of total protein to the 25-75 kD section of the Ponceau-S stained membrane using Image J software (National Institutes of Health, Bethesda, MD) (Rossetti, Tomko and Gordon, 2021).

2.6. Statistical Analysis

All data are presented as mean ± SD. One-way ANOVA or Kruskal-Wallis was used to compare differences among groups. Dunnets’s test was used post hoc to assess differences between groups if a significant F-value was obtained by One-way ANOVA. Fisher’s LSD was used post hoc only for assessment of body composition as the difference between the sham group and the untreated castrated group was negated by correction for multiple tests. Student’s t-test was used to assess differences between select groups. Analysis of select relationships was performed using Pearson product moment correlation. All analyses were performed using GraphPad Prism Software (La Jolla, CA). Significance was set at p ≤ 0.05 for all analyses.

3. RESULTS

3.1. Antioxidant defense gene expression is lower in the TA muscle following androgen deprivation.

The mRNA content of superoxide dismutase 1 and 2 (Sod1 & Sod2), which scavenge free superoxide radicals in the cytosol and mitochondria (Fukai and Ushio-Fukai, 2011), respectively, was overall lower in the TA muscle throughout the diurnal cycle following androgen deprivation (Fig. 1A & B; p≤0.001), The mRNA content of catalase and glutathione peroxidase 1 (Gpx1), which detoxify hydrogen peroxide predominantly in the cytosol, was also overall lower in the TA muscle throughout the diurnal cycle following androgen deprivation (Fig 1C & D; p≤0.0011). The mRNA content of other genes encoding proteins that detoxify ROS including glutathione-disulfide reductase (Gsr) and peroxiredoxin 3 (Prdx3) was overall lower in the TA following androgen deprivation (Fig. 1E & F; p≤0.006).

Fig. 1.

The antioxidant gene expression is lower in the limb muscle following androgen deprivation. Diurnal expression patterns of (A–F) antioxidant and (G–I) pro-oxidant genes. The diurnal mRNA expression pattern of (A) Sod1 (B) Sod2, (C) Catalase, (D) Gpx1, (E) Gsr, (F) Prdx3, (G) Txnip, (H) Nox2, and (I) Nox4 were determined by RT-PCR on homogenates of TA muscle from adult male mice 8 weeks after castration (Castrated) or control sham operation (Sham). N=3/group per time point. Two-way ANOVA was used to assess changes in circadian expression patterns. ME: Main Effect p ≤ 0.05 for all analyses.

Analysis of genes whose protein products promote ROS generation and/or accumulation showed that the mRNA content of thioredoxin interacting protein (Txnip), a protein that inhibits antioxidative function (Saxena, Chen and Shalev, 2010), was actually lower in the TA following androgen depletion (Fig. 1G; p=0.003). The mRNA content of NADPH oxidase 2 (Nox2) and NADPH oxidase 4 (Nox4), which are catalytic subunits for the ROS producing NADPH oxidase complex, were unaffected by androgen deprivation (Fig. 1H & I). In all, these data suggest the antioxidant capacity in the limb muscle may be reduced following androgen deprivation.

3.2. A mitochondria targeted antioxidant partially preserved lean mass and grip strength following androgen deprivation.

Phenotypic data from Study #2 are represented in Tables 2 and 3. The efficacy of the castration surgery was confirmed by a substantial decrease in the mass of the seminal vesicle in all castrated mice. There was no difference in final body weight or average daily food intake among all groups. However, when average daily food intake from all castrated mice was collapsed together, the mean daily intake was significantly lower than that observed in sham mice (P=0.003). Similar to previous findings from our laboratory (Rossetti and Gordon, 2017,Rossetti, Fukuda and Gordon, 2018), heart mass was lower (p≤0.03) and spleen mass was higher (p≤0.0002) in castrated mice, and mass of these organs was not affected by resveratrol or MitoQ. There were no differences among groups for epididymal fat pad mass or tibia length.

Table 2.

Body Weights and Daily Food Intake for Study #2

	N	Initial Body Weight (g)	Final Body Weight (g)	Daily Food Intake (g)
Sham	10	29.0 ± 0.9	30.3 ± 1.0	4.2 ± 0.9
Cast	10	28.9 ± 0.6	28.1 ± 0.7	3.6 ± 0.35
Cast + RSV	10	28.8 ± 0.5	27.4 ± 0.6	3.5 ± 0.17
Cast + MQ	9	28.8 ± 0.5	28.2 ± 0.7	3.7 ± 0.23

Cast; Castration, RSV; Resveratrol, MQ; MitoQ

Table 3.

Tissue Weights for Study #2

	Seminal Vesicle (mg)	Heart (mg)	Spleen (mg)	Epididymal Fat (mg)	Tibia Length (mm)
Sham	273.6 ± 14.1	131.0 ± 4.3	69.4 ± 2.4	563.1 ± 75.4	17.4 ± 0.2
Cast	10.1 ± 4.1*	118.7 ± 2.4*	95.9 ± 2.2*	520.2 ± 110.8	17.2 ± 0.1
Cast + RSV	9.1 ± 0.8*	116.5 ± 2.5*	116.5 ± 2.5*	474.2 ± 61.5	17.3 ± 0.1
Cast + MQ	10.8 ± 2.1*	119.7 ± 2.1*	119.7 ± 2.1*	547.8 ± 96.8	17.3 ± 0.1

Cast; Castration, RSV; Resveratrol. MQ; MitoQ.

^*Significantly different than Sham by one-way ANOVA.

Absolute lean mass was lower in both untreated castrated mice and the resveratrol treated mice relative to the sham group (p≤0.03), but lean mass was not significantly different between MitoQ treated mice and the sham group (Fig. 2A; p=0.053). Absolute fat mass and total body water were not affected by castration or treatment (Fig. 2B & C). Relative grip strength (N/g of lean mass) was lower in the untreated and resveratrol treated castrated mice (p≤0.019), but treatment with MitoQ partially blunted this effect as the difference relative to the sham group was no longer significant (Fig. 2D; (p=0.10). The percent change in grip strength pre-to-post treatment was also lower in untreated and resveratrol treated castrated groups (p≤0.016), but the difference between the MitoQ treated group and the sham group was not significant (Fig. 2E; p= 0.09).

Fig. 2.

A mitochondria targeted antioxidant partially preserves lean mass and grip strength following androgen deprivation. Effect of antioxidant treatment on (A–C) body composition and (D–E) grip strength. Adult male mice were subjected to either castration (Castrated) or control sham operation (Sham); subsets of castrated animals were treated with either resveratrol (Cast + RSV) or MitoQ (Cast + MQ) for 6 weeks. Absolute (A) lean mass, (B) fat mass, and (C) total body water were determined by EchoMRI. (D) Grip strength normalized to lean mass. (E) Percent change in grip strength pre- to post-treatment. One-way ANOVA was used to assess differences among groups. N=9-10/group. * Significant difference compared to sham group. p ≤ 0.05 for all analysis.

3.3. A mitochondria targeted antioxidant partially preserved limb muscle mass following androgen deprivation.

TA mass was lower in the untreated and resveratrol treated castrated mice relative to the sham group (p≤0.01), but MitoQ treatment partially preserved TA mass in the MitoQ treated group as they were not significantly different than the sham group (Fig. 3A; p=0.08). This outcome remained even when TA mass was normalized to tibia length (data not shown). A direct relationship was also observed between lean mass and TA mass (Fig. 3B). Though not significant by ANOVA, independent Student’s t-tests showed that the mass of the lower limb muscles (sum of the gastrocnemius, plantaris, soleus, and TA) was significantly lower in the untreated and resveratrol treated castrated groups compared to sham (p≤0.028), but not between the sham group and the MitoQ treated group (Fig. 3C; p=0.20). Similar to the TA, a direct relationship of similar strength was observed between lean mass and lower limb muscle mass (Fig. 3D). The partial preservation of TA mass with MitoQ treatment was not due to changes in long-term myofibrillar fractional synthetic rates (at least across the time points analyzed) or RNA content at the time of sacrifice (a marker of translational capacity) (Fig. 3E & F). Though we did not measure RNA content at multiple timepoints throughout the labeling period in our study, normalizing long-term myofibrillar FSR to the final RNA content per TA muscle at sacrifice showed that translational efficiency (synthesis per ribosomal unit) may have been higher in the TA of the untreated castrated mice (Fig. 3G; p=0.03). However, this measure was not significantly increased in the TA of MitoQ treated mice (Fig. 3G; p=0.09), suggesting a protein metabolism that more closely resembled the sham condition. The partial preservation of TA mass by MitoQ likely occurred in the insoluble protein fraction of the muscle as the soluble protein content per TA was significantly lower in all castrated groups regardless of treatment (Fig. 3H; (p≤0.044).

Fig. 3.

A mitochondria targeted antioxidant partially preserves limb muscle mass following androgen deprivation. Effect of antioxidant treatment on (A-D) TA and lower limb muscle mass and (E–H) protein metabolism. Animals are the same as in Fig. 2 after treatment for 7 weeks. (A) TA mass and (B) the relationship between TA mass and lean mass determined by Pearson product moment correlation. (C) The combined mass of the gastrocnemius, soleus, plantaris and TA; collectively termed lower limb muscle mass. (D) The relationship between lower limb muscle mass and lean mass determined by Pearson product moment correlation. (E) Myofibrillar fractional synthetic rate (FSR) in the TA. (F) Total RNA content per TA. (G) Myofibrillar FSR normalized to total RNA. (H) Soluble protein per TA. One-way ANOVA was used to assess differences among groups. Student’s t-test was used to assess differences for lower limb muscle mass. N=9-10 per group. * Significant difference compared to sham group by one-way ANOVA. # Significant difference compared to sham group by independent Student’s t-test. Cast, Castrated; RSV, Resveratrol; MQ, MitoQ; FSR, Fractional Synthetic Rate, p ≤ 0.05 for all analysis.

3.4. A mitochondria targeted antioxidant did not mitigate markers of mitochondrial degradation pathway activation following androgen deprivation.

Our previous work showed that nutrient consumption initiates the turnover of BNIP3 protein in the TA muscle of castrated mice (Steiner et al., 2017,Rossetti et al., 2018,Rossetti et al., 2020). Accordingly, BNIP3 protein content was lower in the TA muscle of all castrated mice at the time of day when nutrients are being consumed (i.e. dark cycle) and this was independent of treatment (Fig. 4A & F; p≤0.004). There was no difference in PARKIN protein content between groups at this time point (Fig. 4B & F), though Pink1 mRNA was higher in the untreated and MitoQ treated mice (Fig. 4C; p≤0.016). While not significant by ANOVA, independent Student’s t-tests showed that the LC3 II/I ratio was higher in the TA muscle of untreated and resveratrol treated castrated mice compared to sham values (p≤0.026), with a strong trend in the MitoQ treated mice (Fig. 4D & F; p≤0.06). Protein content of the complimentary marker of autophagy, p62, was lower in the TA of all castrated animals independent of treatment (Fig. 4E & F; p≤0.044). Despite sustained elevations in these markers of mitochondrial degradation pathway activation following androgen deprivation, mitochondrial protein fractional synthetic rates and the mRNA expression of Nuclear Respiratory Factor 1 (Nrf1) and the Mitochondrial transcription factor A (Tfam), transcription factors that regulate expression of mitochondrial genes, were unchanged at the analyzed time point (Fig. 5A–C). This occurred even though the mRNA content of a regulator of mitochondrial biogenesis, PPAR gamma coactivator 1 alpha (Pgc-1α), was lower in all castrated groups (Fig. 5D; p≤0.042).

Fig. 4.

A mitochondria targeted antioxidant does not affect markers of mitochondrial degradation pathway activation in the TA muscle following androgen deprivation. Animals are the same as in Fig. 2. (A) BPIN3 protein content and (B) PARKIN protein content were determined by Western blot analysis on TA muscle homogenates. The mRNA content of (C) Pink1 was determined by RT-PCR. The (D) LC3 II/I ratio and (E) p62 protein content were determined by Western blot analysis. (F) Representative Western blots. One-way ANOVA was used to assess differences among groups. Student’s t-test was used to assess differences for LC3 II/I ratio. N=9-10 per group. * Significant difference compared to sham group. # Significant difference compared to sham group by independent Student’s t-test. Cast, Castrated; RSV, Resveratrol; MQ, MitoQ; FSR, Fractional Synthetic Rate, p ≤ 0.05 for all analysis.

Fig. 5.

A mitochondria targeted antioxidant does not affect mitochondrial synthetic rates in the TA muscle following androgen deprivation. Animals are the same as in Fig. 2. (A) Mitochondrial protein fractional synthetic rate (FSR) in the TA. The mRNA of (B) Nrf1, (C) Tfam, and (D) Pgc-1α were determined by RT-PCR. One-way ANOVA was used to assess differences among groups. * Significant difference compared to sham group. Cast, Castrated; RSV, Resveratrol; MQ, MitoQ; FSR, Fractional Synthetic Rate. p ≤ 0.05 for all analysis.

4. DISCUSSION

The decrease in androgen production that occurs in response to various pathological conditions contributes to the long-term atrophy of limb skeletal muscles (Ferrando et al., 2003,Steiner et al., 2017,White, Puppa, Narsale et al., 2013,White, Gao, Puppa et al., 2013). Similar to previous work in other atrophic conditions (Min et al., 2011,Powers et al., 2011), we show that systemic administration of a mitochondria targeted antioxidant can partially preserve limb muscle mass and grip strength following androgen deprivation. Our laboratory previously reported increased H₂O₂ content in the TA-muscle of castrated mice following refeeding (~5%) (Rossetti et al., 2018), which is in line with the quantity of mass and grip strength that was partially preserved in the present study. While the partial preservation of muscle mass by a mitochondria targeted antioxidant is consistent with prior work (Min et al., 2011,McClung, Whidden, Kavazis et al., 2008,Min, Kwon, Smuder et al., 2015,Powers, Morton, Ahn et al., 2016,Talbert, Smuder, Min et al., 2013), it contrasts with reports from other laboratories that showed no effect of oral- or genetically-induced antioxidant treatment [Trolox or mitochondria-targeted catalase (MCAT) transgene] on muscle atrophy or functional deficits in response to atrophic stimuli such as hind-limb unloading (Eshima, Siripoksup, Mahmassani et al., 2020,Rosa-Caldwell, Lim, Haynie et al., 2020). The discrepancies between our work and these other reports could be that mitochondrial ROS was not increased in the limb muscle by hindlimb unloading (Eshima et al., 2020), or that the effect of MitoQ in the present study was due to effects in non-muscle tissue. Indeed, both androgens and MitoQ target various organs that could alter muscle mass including the liver, kidneys, and the brain (Kurita, Horie, Yamazaki et al., 2016,Sinclair, Grossmann, Gow et al., 2015,Zitzmann, 2006,Rehman, Liu, Krishnasamy et al., 2016,Maiti, Saha, More et al., 2017,Xiao, Xu, Zhang et al., 2017). Moreover, there was no difference in skeletal muscle 4HNE content between groups in the present study (data not shown). This may be because 4HNE content was only higher in the TA of castrated mice during the light cycle (Rossetti et al., 2019), and therefore, markers of increased ROS in the muscles from the current study would not be apparent because the samples were harvested in the dark cycle.

Regardless of MitoQ’s mode of action, the partial preservation of limb muscle mass by MitoQ was likely due to reduced rates of limb muscle protein breakdown during the time points where muscle mass was changing. While it is also possible that MitoQ increased myofibrillar synthetic rates at the initial onset of changes in muscle mass, previous work implies the decrease in limb muscle mass that occurs in response to hypogonadism is mediated largely through increased rates of muscle protein breakdown. Treating hypogonadal men with testosterone resulted in a net protein balance that was less negative within 1 month of treatment that persisted through at least 6 months (Ferrando et al., 2002,Ferrando et al., 2003). It was concluded that this positive shift in protein balance was due to decreased rates of protein breakdown as rates of synthesis were not affected at those time points (Ferrando et al., 2002,Ferrando et al., 2003). Other laboratories also did not find a difference in limb muscle rates of protein synthesis in castrated rodents compared to those rodents with intact testicular androgen production (Steiner et al., 2017,Davidyan, Pathak, Baar et al., 2021,Jiao, Pruznak, Huber et al., 2009). Other than the elevated markers of autophagy, which may not be involved in the regulation of muscle mass in response to androgen deprivation, we didn’t observe any changes in other markers of protein breakdown in the TA muscle at the time of sacrifice (i.e. ubiquitylated proteins or active Calpain-1; data not shown). This was likely because muscle mass is stable at the 7 week post castration time point (Pan, Singh, Sahasrabudhe et al., 2016,Pan, Jaiswal Agrawal, Zulia et al., 2020). Given we didn’t observe a measurable changes in myofibrillar synthetic rates between groups during our labeling period, protein breakdown should also be equal. However, other laboratories showed that the activity of the ubiquitin proteasome was higher in the limb muscle of hypogonadal humans as well as castrated animals (Ferrando et al., 2002,Jiao et al., 2009), particularly at time points closer to the castration surgery (Davidyan et al., 2021). Therefore, we speculate that MitoQ blunted one or more protein degradative processes in the initial time points following the castration surgery when muscle mass was changing.

Our laboratory and others previously showed that markers of increased ROS in the limb muscle in response to androgen deprivation coincided with elevated markers of autophagy (p62) and markers of mitochondrial degradation pathway activation (BNIP3/pink1)(Serra et al., 2013,Steiner et al., 2017,Rossetti et al., 2018,Rossetti et al., 2019,Rossetti and Gordon, 2017,Rossetti et al., 2021). These markers remained largely elevated in the TA of castrated mice despite MitoQ treatment, suggesting mitochondrial ROS may not be the factor initiating these changes, though this would have to be directly shown as skeletal muscle mitochondrial ROS was not assessed in this study. A more likely scenario is that androgen deprivation disrupts the cellular signals regulating mitochondrial quality control, thereby failing to maintain a quality mitochondrial pool. Accordingly, we have shown that the expression of various genes regulating mitochondrial dynamics (e.g. fission/fusion) were lower in the TA muscle of androgen deprived mice (Rossetti et al., 2019), and it is unlikely that MitoQ would affect expression of those genes. Although synthetic rates of mitochondrial proteins were not affected by either castration or MitoQ, we did observe a decrease in the mRNA expression of PGC-1α, which may be one of the androgen-sensitive changes accentuating alterations to mitochondrial quality, though the protein content of PGC-1α was not assessed. The extent to which androgen deprivation alters mitochondrial quality/function, and the contribution of changes in mitochondrial quality to the overall limb muscle atrophy, requires additional investigation.

Our data show the limb muscle antioxidant defense system may be compromised by androgen deprivation as the overall diurnal expression of various genes encoding enzymes involved with ROS quenching were lower in response to androgen deprivation, at least at the post-castration time point analyzed. This is consistent with previous reports that showed a reduction in ROS scavenging capacity in the plasma and prostate of castrated rats (Tam, Gao, Leung et al., 2003). Specifically, the mRNA expression of Sod2, Catalase, Gpx1, Gsr, as well as a member of the peroxiredoxin family, were all lower following castration (Tam et al., 2003). In contrast to our findings, that report also showed that the expression of genes encoding components of the NADPH oxidase system (e.g. Nox2 & Nox4) were higher in the prostate following castration (Tam et al., 2003), suggesting androgen deprivation increases the pro-oxidant system. Discrepancies between our study and the previous work could be due to differences in the tissues analyzed (skeletal muscle vs. prostate) or that NOX2 and NOX4 protein, or the activity of the NADPH oxidase system, are actually increased in the limb muscle in response to androgen deprivation even though the mRNA expression of each gene was not affected.

In conclusion, we show that a mitochondria targeted antioxidant partially blunts the loss of limb muscle mass and grip strength in response to androgen deprivation. Despite these favorable outcomes following MitoQ treatment, markers of mitochondrial degradation pathways remained elevated in the limb muscle. Because maintaining muscle mass has been linked to a favorable survival outcome, there is a need to further identify the intramuscular events by which androgens regulate muscle mass, including the potential role for impaired mitochondrial quality, in order to develop safe and effective treatments that blunt muscle atrophy in those that cannot receive androgen replacement.

HIGHLIGHTS.

• Markers of increased ROS were previously seen in limb muscle following castration.

• Markers of mitochondrial degradation pathways were higher following castration.

• These were inversely were related to limb muscle mass following castration.

• The antioxidant, MitoQ, partially preserved limb muscle mass following castration.

• Increased markers of mitochondrial degradation pathways were unaffected by MitoQ.

Abbreviations:

Cast

Castration

RSV

Resveratrol

MQ

MitoQ

TA

tibialis anterior

ROS

reactive oxygen species

4HNE

4-hydroxynonenol

BNIP3

BCL2/adenovirus E1B 19 kDa protein-interacting protein 3

PINK1

PTEN-induced kinase 1

LC3

microtubule-associated protein light chain 3

D₂O

deuterium oxide

Tfam

transcription factor A, mitochondrial

Nrf1

nuclear respiratory factor 1

Rplp0

ribosomal protein lateral stalk subunit P0

Gpx1

glutathione peroxidase 1

Sod1 & Sod2

superoxide dismutase 1 and 2

Gsr

glutathione-disulfide reductase

Prdx3

peroxiredoxin 3

Txnip

thioredoxin interacting protein

Nox2

NADPH oxidase 2

Nox4

NADPH oxidase 4

Pgc-1α

PPAR gamma coactivator 1 alpha