Activation of eIF4E‐binding‐protein‐1 rescues mTORC1‐induced sarcopenia by expanding lysosomal degradation capacity

Abstract Background Chronic mTORC1 activation in skeletal muscle is linked with age‐associated loss of muscle mass and strength, known as sarcopenia. Genetic activation of mTORC1 by conditionally ablating mTORC1 upstream inhibitor TSC1 in skeletal muscle accelerates sarcopenia development in adult mice. Conversely, genetic suppression of mTORC1 downstream effectors of protein synthesis delays sarcopenia in natural aging mice. mTORC1 promotes protein synthesis by activating ribosomal protein S6 kinases (S6Ks) and inhibiting eIF4E‐binding proteins (4EBPs). Whole‐body knockout of S6K1 or muscle‐specific over‐expression of a 4EBP1 mutant transgene (4EBP1mt), which is resistant to mTORC1‐mediated inhibition, ameliorates muscle loss with age and preserves muscle function by enhancing mitochondria activities, despite both transgenic mice showing retarded muscle growth at a young age. Why repression of mTORC1‐mediated protein synthesis can mitigate progressive muscle atrophy and dysfunction with age remains unclear. Methods Mice with myofiber‐specific knockout of TSC1 (TSC1mKO), in which mTORC1 is hyperactivated in fully differentiated myofibers, were used as a mouse model of sarcopenia. To elucidate the role of mTORC1‐mediated protein synthesis in regulating muscle mass and physiology, we bred the 4EBP1mt transgene or S6k1 floxed mice into the TSC1mKO mouse background to generate 4EBP1mt‐TSC1mKO or S6K1‐TSC1mKO mice, respectively. Functional and molecular analyses were performed to assess their role in sarcopenia development. Results Here, we show that 4EBP1mt‐TSC1mKO, but not S6K1‐TSC1mKO, preserved muscle mass (36.7% increase compared with TSC1mKO, P < 0.001) and strength (36.8% increase compared with TSC1mKO, P < 0.01) at the level of control mice. Mechanistically, 4EBP1 activation suppressed aberrant protein synthesis (two‐fold reduction compared with TSC1mKO, P < 0.05) and restored autophagy flux without relieving mTORC1‐mediated inhibition of ULK1, an upstream activator of autophagosome initiation. We discovered a previously unidentified phenotype of lysosomal failure in TSC1mKO mouse muscle, in which the lysosomal defect was also conserved in the naturally aged mouse muscle, whereas 4EBP1 activation enhanced lysosomal protease activities to compensate for impaired autophagy induced by mTORC1 hyperactivity. Consequently, 4EBP1 activation relieved oxidative stress to prevent toxic aggregate accumulation (0.5‐fold reduction compared with TSC1mKO, P < 0.05) in muscle and restored mitochondrial homeostasis and function. Conclusions We identify 4EBP1 as a communication hub coordinating protein synthesis and degradation to protect proteostasis, revealing therapeutic potential for activating lysosomal degradation to mitigate sarcopenia.


Polysome profiling
Polysome profiling was modified from previous published work [4][5][6] . Briefly, limb muscles from 2-mo-old male mice were homogenized with a Dounce homogenizer in ice-cold lysis buffer [5] and the homogenate was then centrifuged at 5,000 ×g for 15 min at 4°C. The supernatant from centrifugation was measured at OD260 and equal volumes of samples were layered on the top of a pre-cooled 15-60% sucrose gradient [6]. Samples were then centrifuged in a SW41Ti rotor at 39,000 rpm for 4 h at 4°C. Gradients were collected from the top and profiles were monitored at 254 nm.

Differential expression analysis
Genes differentially expressed in each genotype were identified using DESeq2 version 1.26.0 [9]. Genes with reads less than 10 were removed for the DESeq2 analysis. Genes with false discovery rate (FDR)-adjusted P value < 0.05 and fold change ≧ 1.5 were considered to be differentially expressed. Variance stabilizing transformation (VST) was applied to the gene counts before the principal component analysis (PCA).

Immunohistochemistry and Immunofluorescence staining (IF)
Hematoxylin and eosin (H&E), Oil-Red-O, and muscle fiber type staining were performed as previously described [2]. For Periodic acid-Schiff (PAS) staining, a PAS staining kit was used for detection of aldehydes and mucosubstances, in concordance with the manufacturer's instructions. For immunohistochemistry of succinate dehydrogenase (SDH) activities, following defrosting, slides were then rehydrated in PBS for 5-min, then incubated in SDH solution for 2-3 min at room temperature (RT). Subsequently, slides were washed with PBS and mounted with aqueous mounting medium. For p62, LAMP1, and laminin IF staining, following defrosting, tissues were fixed in 4% PFA (10 min). Following with PBS washing, slides were boiled in Target Retrieval Solution for antigen retrieval (10 min), then left to cool at RT (10 min). After a washing step, slides were blocked at RT (1 hr) and then incubated with primary antibody at 4°C overnight. After washing with PBS, the secondary antibody solution was added and incubated (1 hr) at RT. Samples were subsequently washed with PBS and then mounted with Prolong Gold antifade reagent. In a separate experiment, muscle was stained with Wheat Germ Agglutinin (WGA) instead of laminin to outline the muscle fiber. For Dihydroethidium (DHE), LysoTracker Red, LysoSensor Green and Magic Red staining, following defrosting, slides were washed with PBS. The fluorescence probes were then added on the slides (10 min) in a dark chamber. All the staining solution contained DAPI and WGA. The reaction was stopped by three washes in PBS.
Slides were mounted in Prolong Gold antifade reagent.
Slides were imaged at 20× magnification with stitching into whole sections using a TissueFAXS Slide Scanner (TissueGnostics), with additional viewing at high magnification (60x) using an Olympus FV3000 Confocal Microscope. Serial unstained sections for IF or fluorescence probe with only DAPI and WGA was imaged as a negative control ( Figure S1). For fiber type staining, whole tissue sections were analyzed to determine myofiber size of MyHC isoform populations using the Myosoft plugin of Fiji according to published methods [12]. For other IF staining, staining intensity per myofiber was determined by thresholding in ImageJ and fibers were also counted manually to determine number of positively stained fibers.
To analyse autofluorescence, following defrosting, tissues were then fixed in 2% PFA for 10 min. After a washing step, tissues were mounted with Antifade Mounting Medium. Slides were imaged using an Olympus DP72 upright microscope. Autofluorescence intensity was determined by mean fluorescence of five 20X images using ImageJ. Table S1. Transgenic mouse lines used in this study.

Mouse line Source Reference
Ckmm-Cre Jackson Lab. 006475 [13,14] TSC1flox Jackson Lab. 005680 [15] 4EBP1mt Our lab [2] S6K1flox Dominic J Wither's lab [16]         Percentage of total fibers in each intensity class is shown as mean ± SD. N=3/genotype; data was analysed by two-way ANOVA with Sidak's multiple comparison test (across rows). *P<0.05; ***P<0.001, ****P<0.0001 indicates significance between age in control mice.    Figure 3A. (I) The blot with additional samples accompanying for Figure 3A of OXPHOS proteins from 12mo-old male mouse QM, where each representative protein from each mitochondrial complex is indicated, and the quantification is shown in Table S4. Brief: C, Control; T, TSC1mKO; ST, S6K1-TSC1mKO; ET, 4EBP1mt-TSC1mKO.   Figure 5A. Immunoblotting of proteins in insoluble fraction of protein lysate from 12-mo-old male mouse QM; Ponceau S membrane staining was used as a loading control.
The detailed quantification of Figure 5A and S5B is presented in Table S5.    Figure 7A with quantification in Figure 7B and Table S6