Sirt6 Mono-ADP-Ribosylates YY1 to Promote Dystrophin Expression for Neuromuscular Transmission

Abstract

The degeneration of the neuromuscular junction (NMJ) and the decline in motor function are common features of aging, but the underlying mechanisms have remained largely unclear. This study reveals that Sirt6 is reduced in aged mouse muscles. Ablation of Sirt6 in skeletal muscle causes a reduction of Dystrophin levels, resulting in premature NMJ degeneration, compromised neuromuscular transmission, and a deterioration in motor performance. Mechanistic studies show that Sirt6 negatively regulates the stability of the Dystrophin repressor YY1 (Yin Yang 1). Specifically, Sirt6 mono-ADP-ribosylates YY1, causing its disassociation from the Dystrophin promoter and allowing YY1 to bind to the SMURF2 E3 ligase, leading to its degradation. Importantly, supplementation with nicotinamide mononucleotide (NMN) enhances the mono-ADP-ribosylation of YY1 and effectively delays NMJ degeneration and the decline in motor function in elderly mice. These findings provide valuable insights into the intricate mechanisms underlying NMJ degeneration during aging. Targeting Sirt6 could be a potential therapeutic approach to mitigate the detrimental effects on NMJ degeneration and improve motor function in the elderly population.

Keywords: Dystrophin; NMJ; Sirt6; YY1.

Figure 1

Sirt6 ablation in skeletal muscle impairs neuromuscular transmission and motor functions. (A) Representative fluorescent images show the fragmented NMJs in aged mouse muscles. Tibialis anterior (TA) muscles of indicated ages were stained with R‐BTX (red) and DAPI (blue). Arrows indicate the fragmented NMJs. (B) Heat map of RNA‐seq data shows the expression of Sirtuin family members in young and old mouse trachea samples (NCBI GSE55162). Note that Sirt6 mRNA is specifically reduced in aged samples. (C) Immunoblot shows the reduction of Sirt6 protein levels in aged muscles. TA muscles from mice of different ages were analyzed. W, week; M, month. (D) Enrichment of Sirt6 proteins in the synaptic region (SR) of diaphragm muscles in 4‐month‐old mice. Neurofilament (NF) serves as a positive control for proteins in the synaptic region. Samples were from 3 mice. (E) Immunoblot shows Sirt6 ablation in gastrocnemius muscles in HSA‐Cre; Sirt6^floxp/floxp mice (HSA‐Sirt6 cKO, 2‐month‐old). Histone and ponceaus staining indicate loading. (F) Reduced grip strength in HSA‐Sirt6 cKO mice (6‐month‐old). Grip strength was normalized to the body weight. n = 18 male mice in control and n = 19 male mice in the HSA‐cKO group. (G) Impaired rotarod performance in HSA‐Sirt6 cKO mice (6‐month‐old). n = 14 male mice in control and n = 15 male mice in the HSA‐cKO group. (H) Impaired neuromuscular transmission in HSA‐Sirt6 cKO mice (5‐month‐old). Left, ten compound muscle action potential (CMAP) traces were stacked in succession for better comparison. CMAPs were recorded in the gastrocnemius in response to a train of 10 submaximal stimuli at different frequencies. CMAP traces between two genotypes at the 1st, 2nd, and 10th stimuli. Middle, CMAP amplitudes at 30 Hz with different stimulation times. Two‐way ANOVA with Sidak's post hoc test for multiple comparisons. Stimulation time: F (9, 60) = 3.934. Amplitude: F (1, 60) = 121.8. Interaction, *p < 0.05; Right, CMAP amplitudes of the tenth stimulation at different stimulation frequencies. n = 4 mice in control and n = 4 mice in the HSA‐cKO group. Two‐way ANOVA with Sidak's post hoc test for multiple comparisons. Frequency: F (5, 36) = 24.05. Amplitude: F (1, 36) = 63.01. Interaction, ****p < 0.0001. (I) Reduced miniature endplate potential (mEPP) amplitude in HSA‐Sirt6 cKO mice (5‐month‐old). Left: representative mEPP trace; Middle: frequency; Right: amplitude and its cumulative curve. n = 53 cells from 6 mice in the control and n = 49 cells from 7 mice in the HSA‐Sirt6 cKO group. Unless otherwise specified, at least three independent experiments were performed. Mean ± SEM; *p < 0.05, ***p < 0.001, and ****p < 0.0001; t‐test in (F), (G), and (I); two‐way ANOVA in (H).

Figure 2

Muscle Sirt6 is required for Dystrophin expression and NMJ maintenance. (A) Representative fluorescent images show fragmented NMJs in diaphragm of HSA‐Sirt6 cKO mice (P195, indicated by white arrows). R‐BTX, red; DAPI, blue. (B) Statistical results of NMJ fragmentation in (A). n = 4 mice per group. (C) Real‐time PCR shows aberrant expression of AChR subunits in TA muscles of HSA‐Sirt6 cKO mice (6‐month‐old). (D) H&E staining reveals centralized nucleus in TA muscles of 7‐month‐old HSA‐Sirt6 cKO mice (yellow arrows). Right: quantification results. (E) Heat map of real‐time PCR results shows reduced mRNA levels of Dystrophin in the Dystrophin‐Glycoprotein complex (DGC) in TA muscles of HSA‐Sirt6 cKO mice (6‐month‐old). (F) Immunoblot reveals reduced Dystrophin protein levels in TA muscles of HSA‐Sirt6 cKO mice (6‐month‐old). n = 3 mice per group. (G) Immunoblot showing the reduction of Dystrophin protein levels in TA muscles of aged mice (24‐month‐old). n = 3 mice per group. Unless otherwise specified, at least three independent experiments were performed. Mean ± SEM; **p < 0.01, ***p < 0.001, and ****p < 0.0001; t‐test in (B‐D).

Figure 3

Sirt6 negatively regulates the Dystrophin repressor YY1. (A) Heat map of quantitative proteomic analysis shows differentially expressed proteins in TA muscles of control and HSA‐Sirt6 cKO mice (7‐month‐old). Note that YY1 is upregulated in HSA‐Sirt6 cKO samples. n = 3 mice in each group. (B) Volcano plot of quantitative proteomic analysis illustrates upregulated YY1 protein levels in HSA‐Sirt6 cKO mice (7‐month‐old). (C) Immunoblot shows enhanced YY1 protein levels in TA muscles of HSA‐Sirt6 cKO mice (7‐month‐old). (D) Immunoblot shows enhanced YY1 and reduced Dystrophin protein levels in Sirt6 knockout C2C12 cells (sgSirt6). (E) Immunoblot indicate that Sirt6 decreases the chromatin‐bound YY1 levels in C2C12 cells. (F) ChIP assay reveals that Sirt6 negatively regulates the interaction of YY1 with the Dystrophin promoter. The cartoon illustrates YY1 binding to the Dystrophin promoter and repressing its transcription. Bottom: C2C12 cells were treated with the Sirt6 inhibitor OSS‐128167 (Sirt6i, 100 µM, overnight) and subjected for ChIP analysis. (G) Immunoblot shows the requirement of YY1 in Sirt6‐regulated Dystrophin expression. The reduction of Dystrophin levels in Sirt6 knockout C2C12 cells (sgSirt6) was partially rescued in Sirt6 and YY1 double knockout cells (sgSirt6; sgYY1). (H) Immunoblot shows less YY1 and more Dystrophin protein levels in the synaptic region (SR), compared to those in non‐synaptic region (NSR) in the diaphragm of adult mice (6‐month‐old). NF served as a positive control for proteins in the synaptic region. Unless otherwise specified, at least three independent experiments were performed. Mean ± SEM; *p < 0.05; t‐test in (F).

Figure 4

Sirt6 promotes the SMURF2 E3 ligase to destabilize YY1. (A) Real‐time PCR shows no difference of YY1 mRNA levels in 7‐month‐old control and HSA‐cKO TA muscles (left), or vehicle and Sirt6i‐treated C2C12 cells (right). Sirt6i, the Sirt6 inhibitor OSS‐128167. (B) Chase analysis shows the shorter half‐life of YY1 proteins in Sirt6‐expressed C2C12 cells. Cells were treated with cycloheximide (CHX, 10 µM) for the indicated hours before harvesting for immunoblot. (C) Immunoblot indicates that Sirt6‐mediated YY1 reduction depends on the ubiquitination‐proteasome system rather than the autophagy‐lysosome pathway. C2C12 cells were pretreated with Sirt6a (10 µM, overnight), followed by treatment of MG132 (10 µM, 2 hours) or Chloroquine (CQ, 10 µM, 2 hours) before harvesting.^{[ 23 , 26 ]} Sirt6a, the Sirt6 agonist MDL‐800; MG132, the inhibitor of ubiquitination‐proteasome system; CQ, the inhibitor of autophagy‐lysosome pathway. (D) Immunoprecipitation and immunoblot show that the Sirt6 inhibitor reduces the ubiquitination of YY1 in C2C12 cells. (E) Peptide information of SMURF2 E3 ligase in the proteomics result. C2C12 cells were lysed and subjected to IP‐MASS (anti‐YY1 antibodies) to identify its interacting proteins. (F) Co‐immunoprecipitation shows the interaction between YY1‐Flag and SMURF2‐Myc proteins in C2C12 cells. (G) Co‐immunoprecipitation shows the interaction between endogenous YY1 and SMURF2 proteins in adult mouse TA muscles. (H) Immunoprecipitation and immunoblot show that SMURF2 promotes YY1 ubiquitination in C2C12 cells. (I) Immunoblot indicates that Sirt6 activity regulates the interaction between YY1 and SMURF2 in C2C12 cells. (J) Co‐immunoprecipitation shows that Sirt6 is required for the interaction between YY1 and SMURF2 in mouse TA muscles (7‐month‐old). Unless otherwise specified, at least three independent experiments were performed. Mean ± SEM; t‐test in (A).

Figure 5

Sirt6 mono‐ADP‐ribosylates YY1 to release it from the Dystrophin promoter for degradation. (A) The cartoon and table illustrate the defects in enzyme activities of Sirt6 mutants. (B) ChIP assay shows that the activity of Sirt6 mono‐ADP‐ribosyltransferase inhibits its binding on the Dystrophin promoter. C2C12 cells were transfected with different Sirt6‐HA mutants and subjected to ChIP analysis using anti‐YY1 antibodies. WT, wild type. One‐way ANOVA with Tukey's multiple comparisons test. F (5, 12) = 64.30. (C) Immunoblot shows that the activity of Sirt6 mono‐ADP‐ribosyltransferase is required for Dystrophin expression in C2C12 cells. (D) Immunoprecipitation and immunoblot show that Sirt6 promotes the mono‐ADP‐ribosylation of YY1 in C2C12 cells. (E) Immunoprecipitation and immunoblot show that Sirt6‐G60A cannot mono‐ADP‐ribosylate YY1 in C2C12 cells. (F) Immunoblot indicates that the activity of Sirt6 mono‐ADP‐ribosyltransferase is required for YY1 ubiquitination. (G) Immunoblot shows the reduction of YY1 mono‐ADP‐ribosylation and ubiquitination in TA muscles of HSA‐Sirt6 cKO mice (7‐month‐old). (H) Peptide information of YY1 mono‐ADP‐ribosylation sites in the proteomics result (top). The consensus of YY1 mono‐ADP‐ribosylation sites (E206/E208/K409 in the human sequence) is indicated in red (bottom). YY1‐Flag‐tansfected C2C12 cells were lysed and subjected to IP‐MASS (anti‐Flag antibodies) to identify its modification sites. (I) Immunoblot shows a dramatic reduction in mono‐ADP‐ribosylation in the YY1‐E206A/E208A/K409A mutant. C2C12 cells were transfected with Sirt6‐HA and YY1‐Flag (WT or mutant), and their lysates were subjected to detection of YY1 mono‐ADP‐ribosylation. (J) Immunoblot shows an enhanced interaction between YY1 with chromatin when the mono‐ADP‐ribosylation sites in YY1 were mutated. C2C12 cells were transfected with YY1‐Flag (WT or mutant), and their lysates were separated into chromatin and non‐chromatin fractions before immunoblot. (K) Chase analysis reveals a longer half‐life of YY1 proteins when the mono‐ADP‐ribosylation sites in YY1 were mutated. C2C12 cells were transfected with YY1‐Flag (WT or mutant) and treated with CHX (10 µM) for the indicated hours before harvesting for immunoblot. (L) Immunoprecipitation and immunoblot show that mono‐ADP‐ribosylation of YY1 is necessary for the interaction between YY1 and SMURF2 proteins. (M) Immunoblot shows that mono‐ADP‐ribosylation of YY1 is required for Dystrophin expression. C2C12 cells were transfected with YY1‐Flag (WT or mutant) and harvested for immunoblot. Unless otherwise specified, at least three independent experiments were performed. Mean ± SEM; one‐way ANOVA in (B).

Figure 6

NMN (β‐Nicotinamide mononucleotide) treatment increases mono‐ADP‐ribosylation of YY1 and alleviates motor defects in aged mice. (A) Immunoprecipitation and immunoblot show that NAD⁺ promotes mono‐ADP‐ribosylation of YY1. C2C12 cells were treated with NAD⁺ (100 µM, overnight)^{[ 64 ]} and harvested for mono‐ADP‐ribosylation of YY1 analysis. (B) The experiment procedure of NMN treatment. 18‐month‐old WT or HSA‐Sirt6 cKO mice were orally administered with NMN (1.5 mg/ml in H₂O) for 6 months before behavior analysis, electrophysiology, and immunoblot (Created in BioRender. Zhang, Z. (2020) BioRender.com/j41q476). (C) Immunoprecipitation and immunoblot show that NMN treatment enhances mono‐ADP‐ribosylation and ubiquitination of YY1 in TA muscles of aged mice (24‐month‐old). (D) Immunoblot shows that NMN treatment increases the levels of Dystrophin protein in TA muscles of aged WT mice (24‐month‐old). It is noteworthy that NMN has little effect in aged HSA‐Sirt6 cKO mice. (E) Representative fluorescent images show that NMN treatment reduces NMJ fragmentation in aged mouse muscles (24‐month‐old, above). TA muscles were stained with R‐BTX (red) and DAPI (blue). Statistical results (below). n = 4 mice per group. One‐way ANOVA with Tukey's multiple comparisons test. F (3, 12) = 48.90. (F) Electromyography analysis shows that NMN treatment increases the reduction of CMAPs amplitude in aged WT mice (24‐month‐old). WT (vehicle), n = 5 mice; WT (NMN treatment), n = 5 mice; HSA‐Sirt6 cKO (vehicle), n = 5 mice; HSA‐Sirt6 cKO (NMN treatment), n = 4 mice. Two‐way ANOVA with Sidak's post hoc test for multiple comparisons. Stimulation time: F (9, 150) = 6.702. Amplitude: F (3, 150) = 36.93. (G) Improved motor performance in NMN‐treated aged WT mice (24‐month‐old). Above: rotarod performance; Below: grip strength. WT (Vehicle), n = 9 mice; WT (NMN treatment), n = 10 mice; HSA‐Sirt6 cKO (Vehicle), n = 9 mice; HSA‐Sirt6 cKO (NMN treatment), n = 11 mice. One‐way ANOVA with Tukey's multiple comparisons test. Rotarod: F (3, 35) = 18.70, Grip strength: F (3, 35) = 43.32. (H) The working model (Created in BioRender. Zhang, Z. (2021) BioRender.com/f79z500). Unless otherwise specified, at least three independent experiments were performed. Mean ± SEM; *p < 0.05, **p < 0.01, and ***p < 0.001; one‐way ANOVA in (E) and (G); two‐way ANOVA in (F).