All animal experiments were performed in compliance with protocols approved by the Institutional Animal Care and Use Committee at the University of Illinois at Urbana-Champaign (#19255). Wild-type C57BL/6N male mice were maintained in an EcoFlo ventilation cage system (Allentown Inc., Allentown, NJ, United States) at 23°C with a 12-h light/dark cycle and fed ad libitum. Mice at 8 and 27months of age received denervation surgery as described below, and any mice exhibiting signs of abnormality during the recovery period, such as lethargy and reduced body weight, were excluded from the study. The animals were anesthetized with isoflurane during all surgical and euthanasia procedures.

Unilateral denervation of the hindlimb muscles was performed by excising an approximately 0.5-cm-long segment of the sciatic nerve from the right leg. The left contralateral control leg was subjected to a sham surgery that involved only an incision in the skin where the denervation surgery would have occurred. After the surgery, the skin incision was closed with a 3-0 polysorb suture.

In vivo measurements of protein synthesis rates were performed with the surface sensing of translation (SUnSET) technique (Goodman et al., 2011a). Specifically, puromycin (75mM stock in diH₂O) was diluted in 200μl of PBS and intraperitoneally injected into mice at the concentration of 0.04μmol/g bodyweight. Muscles were collected exactly 30min after injection and analyzed by Western blotting for the amount of puromycin-labeled peptides per mg protein or per muscle.

Collected muscles were immediately cut in half cross-wise and each was weighed and frozen in liquid nitrogen for storage. The proximal halves of the muscles were homogenized with a Polytron in ice-cold buffer containing 20mM Tris (pH 7.4), 0.3% Triton X-100, 2mM EGTA, 2mM EDTA, 0.1mM Na₃VO₄, 25mM NaF, 25mM β-glycerolphosphate, and 1×protease inhibitor cocktail (MilliporeSigma, P8240; Burlington, MA, United States). Protein concentrations of the homogenates were measured with the DC protein assay kit (Bio-Rad, Hercules, CA, United States), and total protein content in each whole muscle was calculated.

An equal amount of protein from each sample was boiled in Laemmli buffer, resolved by SDS-PAGE, transferred onto PVDF membrane, blocked with 5% milk in PBS-T (PBS with 0.5% Tween 20), and probed with appropriate primary and secondary antibodies listed below. After washing in PBS-T, blots were visualized and captured with the SuperSignal West Pico PLUS Chemiluminescent Substrate and an iBright CL1000 Imaging System (both from ThermoFisher Scientific, Waltham, MA, United States). The quantification of the blots was performed using ImageJ 1.53 (NIH, Bethesda, MD, United States; https://imagej.nih.gov/ij/).

The antibodies used in this study are: anti-ubiquitin (Cat# sc-8017, RRID: AB_2762364), anti-IGF-1Rβ (Cat# sc-713, RRID: AB_671792), and anti-phospho-IRS-1 (Tyr632; Cat# sc-17196, RRID: AB_669445) from Santa Cruz Biotechnology (Dallas, TX, United States); anti-puromycin (Cat# MABE343, RRID: AB_2566826) from MilliporeSigma; anti-K48-linkage specific polyubiquitin (Cat# 8081, RRID: AB_10859893), anti-pan-actin (Cat# 8456, RRID: AB_10998774), anti-phospho-p70 S6 kinase (Thr389; Cat# 9234, RRID: AB_2269803), anti-p70 S6 kinase (Cat# 9202, RRID: AB_331676), anti-phospho-S6 (Ser240/244; Cat# 5364, RRID: AB_10694233), anti-S6 (Cat# 2317, RRID: AB_2238583), anti-phospho-ERK (Ser240/244; Cat# 4370, RRID: AB_2315112), anti-ERK (Cat# 9102, RRID: AB_330744), anti-phospho-Akt (Thr308; Cat# 13038, RRID: AB_2629447), anti-phospho-Akt (Ser473; Cat# 9271, RRID: AB_329825), anti-Akt (Cat# 9272, RRID: AB_329827), anti-phospho-glycogen synthase kinase-3β (GSK-3β; Ser9; Cat# 5558, RRID: AB_10013750), anti-GSK-3β (Cat# 9832, RRID: AB_10839406), anti-phospho-IGF-1Rβ (Tyr1135/1136; Cat# 3024, RRID: AB_331253), anti-IRS-1 (Cat# 2382, RRID: AB_330333), anti-phosphatidylinositol 3-kinase (PI3K) p85 (Cat# 4292, RRID: AB_329869), and anti-phosphatase and tensin homologue deleted on chromosome ten (PTEN; Cat# 9552, RRID: AB_10694066) from Cell Signaling Technology (Danvers, MA, United States); peroxidase-conjugated anti-rabbit IgG (Cat# 115-036-003, RRID: AB_2617176), anti-mouse IgG (Cat# 111-036-003, RRID: AB_2337942), and anti-mouse IgG2a (Cat# 115-035-206, RRID: AB_2338514) from Jackson Immuno Research Laboratories (West Grove, PA, United States); and peroxidase-conjugated anti-goat IgG (Cat# A27014, RRID: AB_2536079) from ThermoFisher Scientific.

The distal halves of the muscles were homogenized with a ribonuclease-free disposable pestle in 500μl of ice-cold TRIzol (Invitrogen), and 200μl of the upper aqueous phase was isolated for total RNA extraction. Total RNA was extracted using GeneJET RNA Purification Kit (ThermoFisher Scientific) with the first and second elution volume of 60 and 150μl, respectively, and the concentration of RNA eluates was measured by NanoDrop2000 (ThermoFisher Scientific). The gross amount of total RNA was normalized by the muscle weight. To measure ribosomal RNA (rRNA) content, RNA samples from equivalent amounts of muscle were run on 1.2% agarose gels, and the 28S and 18S rRNA were quantified by densitometry using ImageJ 1.53.

Total RNA was reverse-transcribed into cDNA using qScript cDNA Synthesis Kit (Quanta Bioscience, Beverly, MA, United States). Gene expression was analyzed by qPCR on a StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, United States) using SYBR green and primers listed in Table 1. TATA-binding protein (Tbp) was used as an internal control.

All values were presented as mean±SEM. A sample value that deviated more than three times from the mean in a given group was excluded as an outlier. Statistical significance (p<0.05) was determined by a two-tailed unpaired t test or two-way mixed ANOVA followed by a Tukey’s post hoc analysis. All statistical analyses were performed upon the verification of the test assumptions using the OriginPro 2021b software (OriginLab, Northampton, MA, United States).

In this study, we compared 27-month-old mice (aged) to 8-month-old mice (adult) to investigate aging effects. The aged mice had a similar body weight as adult control mice (Supplementary Figure S1), but they showed apparent signs of sarcopenia as revealed by reduced muscle mass and total protein content in tibialis anteria (TA, Figure 1A), gastrocnemius (GAST, Figure 1B), extensor digitorum longus (EDL, Figure 1C), and soleus (SOL, Figure 1D) muscles. After 7days of denervation, adult mice lost a significant amount of muscle mass and total protein content in the TA; however, these denervation-induced atrophic responses did not occur prominently in aged mice (Figure 1A). In other muscles except for EDL, denervation reduced muscle mass and total protein content to a similar degree in adult and aged mice (Figures 1B–D). These results indicate that aged muscle is not more susceptible to denervation atrophy than younger muscle and rather exhibits atrophy resistance in a specific type of muscle.

We wondered whether the TA muscles of the 27-month-old mice had already undergone age-related denervation and became less sensitive to the surgical denervation than other muscles such as GAST. Hence, we measured several gene markers of denervation, including acetylcholine receptor (Chrn) subunits, muscle-specific kinase (Musk), Myogenin (Myog), and Histone Deacetylase 4 (HDAC4). Consistent with previous studies (Adams et al., 1995; Tang et al., 2009), all of these markers were transcriptionally elevated in response to denervation surgery (Figures 1E,F). However, none of those markers was increased in either TA or GAST muscles of aged mice in the sham condition (Figures 1E,F), suggesting that age-related denervation did not occur significantly before the surgical denervation. In fact, we found that the induction of Chrn subunits, Musk, Myog, and HDAC4 by denervation surgery was generally repressed in the aged muscles (Figures 1E,F). The potential implications of this observation will be discussed in the “Discussion” section.

The TA muscle-specific atrophy resistance in aged mice is interesting because its underlying mechanism could provide valuable information for developing therapies against denervation-induced muscle atrophy. To identify the potential mechanisms, we first asked if aging influenced protein degradation and protein synthesis in TA muscles during denervation. The levels of global and K48-linkage-specific polyubiquitination, which are readouts of ubiquitin-mediated protein degradation, were similarly increased in adult and aged mice in response to denervation (Figures 2A,B; Supplementary Figure S2A). On the other hand, the rate of protein synthesis as measured by SUnSET assay (Figure 2A) was elevated to a larger degree in aged mice than in adult mice when expressed per mg protein (Figure 2C) or elevated only in aged mice when expressed per muscle (Figure 2D). In GAST muscles, neither ubiquitin-mediated protein degradation nor protein synthesis rates were affected by aging during denervation (Figures 2E–H; Supplementary Figure S2B). These results indicate that aging promotes protein synthesis rates following denervation to confer resistance to denervation-induced atrophy in TA muscles.

We next asked if the TA muscle-specific effect of aging on protein synthesis was associated with rRNA biogenesis, an essential process of making the protein synthesis machinery, ribosome. Total RNA content, which is composed of >85% of rRNA (O’Neil et al., 2013), was increased by denervation to a greater degree in aged mice than in adult mice (Figure 3A). Measurements of 28S+18S rRNA confirmed a greater increase in rRNA in the aged (Figure 3B). Consistent with the rate of protein synthesis, rRNA biogenesis in GAST muscles was not affected by aging during denervation (Figures 3C,D). Thus, these results strongly suggest that during denervation aging induces TA muscle-specific increase in protein synthesis by enhancing rRNA biogenesis.

One of the key molecular regulators of rRNA biogenesis and protein synthesis is mTORC1 (Iadevaia et al., 2014). Signaling through mTORC1 has been reported to play an important role in the regulation of protein synthesis during denervation (Quy et al., 2013; You et al., 2021a). As such, we envisioned that mTORC1 may mediate the TA-specific effects of aging on rRNA biogenesis and protein synthesis during denervation. However, denervation-induced changes in mTORC1 signaling, revealed by the phosphorylation of the 70kDa ribosomal S6 protein kinase (p70^S6K) and of the S6 protein, were not affected by aging in the TA (Figures 4A,B). ERK signaling, another important regulator of rRNA biogenesis and protein synthesis (Zhao et al., 2003; Stefanovsky and Moss, 2008), was also not affected by aging during denervation (Figure 4C). Similar observations were also made in GAST muscles (Figures 4D–F).

Signaling through Akt is known to promote rRNA biogenesis and protein synthesis independently of its role in activating mTORC1 (Proud, 2006; Chan et al., 2011; Nguyen and Mitchell, 2013), which raises the possibility that Akt may be involved in the age- and TA-specific changes in the anabolic events during denervation. Indeed, in the TA, phosphorylation of Akt on Thr308 and Ser473, both indicating Akt activation, were increased only in the aged muscles during denervation (Figure 4G). Consistently, aging nearly significantly prevented the denervation-induced decrease in the phosphorylation of GSK-3β on Ser9, a direct substrate of Akt (Figure 4H). These age-dependent changes in Akt and GSK-3β phosphorylation did not occur in GAST muscles (Figures 4I,J). Combined, these results suggest that during denervation aging promotes rRNA biogenesis and protein synthesis specifically in the TA muscle through Akt signaling.