Nobiletin fortifies mitochondrial respiration in skeletal muscle to promote healthy aging against metabolic challenge

Abstract

Circadian disruption aggravates age-related decline and mortality. However, it remains unclear whether circadian enhancement can retard aging in mammals. We previously reported that the small molecule Nobiletin (NOB) activates ROR (retinoid acid receptor-related orphan receptor) nuclear receptors to potentiate circadian oscillation and protect against metabolic dysfunctions. Here we show that NOB significantly improves metabolic fitness in naturally aged mice fed with a regular diet (RD). Furthermore, NOB enhances healthy aging in mice fed with a high-fat diet (HF). In HF skeletal muscle, the NOB-ROR axis broadly activates genes for mitochondrial respiratory chain complexes (MRCs) and fortifies MRC activity and architecture, including Complex II activation and supercomplex formation. These mechanisms coordinately lead to a dichotomous mitochondrial optimization, namely increased ATP production and reduced ROS levels. Together, our study illustrates a focal mechanism by a clock-targeting pharmacological agent to optimize skeletal muscle mitochondrial respiration and promote healthy aging in metabolically stressed mammals.

Fig. 1

NOB improved physiological health and survival of aged mice under regular diet. a–d Young (Y) or aged (A) mice were fed with regular diet (RD) or regular diet with NOB (RD.N). a Glucose tolerance test and area under curve (AUC) (n = 10, 7, and 8 for Y.RD, A.RD, and A.RD.NOB, respectively). b Core body temperature (upper panel, n = 10) and cold tolerance (lower panel, n = 10, 7, and 5 for Y.RD, A.RD, and A.RD.NOB, respectively); *A.RD vs. A.RD.NOB; #A.RD vs. Y.RD; two-way ANOVA). c Wheel-running activity (WRA) analysis. Representative actograms (upper panels) are shown. Daily activity patterns (lower left, line graph) and daily averages (lower right, bar graph) are shown (ActiView) (n = 10, 5, and 6 for Y.RD, A.RD, and A.RD.NOB, respectively). d Mean sleep bout duration was analyzed with the PIEZO sleep analysis system (n = 9, 6, and 6 for Y.RD, A.RD, and A.RD.NOB, respectively). e Kaplan–Meier (KM) survival curves in 16-month-old mice fed with regular diet with or without 0.1% NOB supplement. Inset: median lifespan comparison. a–d *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA. e *p < 0.05, Log-rank and Mann–Whitney U-tests (see Supplementary Fig. 1h). Data are presented as mean ± SEM in bar and line graphs

Fig. 2

NOB improved adiposity, glucose homeostasis, and fitness in high-caloric conditions. Aged mice were fed with regular diet (RD), high-fat (HF) diet, or HF diet with NOB (HF.NOB). a End point (after 20–22 weeks of treatment) body weight and b total Visceral (Vis) fat mass (including perigonadal fat, perirenal fat, and mesenteric fat) were measured (n = 20, 16, and 17 for RD, HF, and HF.NOB, respectively). c Glucose tolerance test and area under curve (AUC) (n = 6, 5, and 7 for RD, HF, and HF.NOB, respectively). d Serum lipid homeostasis parameters were measured at indicated time points (n = 7, 4, and 6 for RD, HF, and HF.NOB, respectively). e In vivo heat production and oxygen consumption measured in metabolic chamber. Average values of day, night, and total are shown (n = 9, 10, and 10 for RD, HF, and HF.NOB, respectively). f Core body temperature (n = 6, 4, and 7 for RD, HF, and HF.NOB, respectively). g Cold tolerance test (n = 6, 5, and 7 for RD, HF, and HF.NOB, respectively; *HF vs. RD, †HF vs. HF.NOB, two-way ANOVA). h Hindlimb grip strength test (n = 5, 6, and 8 for RD, HF, and HF.NOB, respectively). i Running distances were measured with treadmill under HF conditions (n = 12, 11, and 12 for RD, HF, and HF.NOB, respectively). j Inflammation markers including TNFα (upper panel) and LBP (lower panel) (ZT6, n = 7, 6, and 7 for RD, HF, and HF.NOB, respectively; ZT18, n = 6, 8, and 7 for RD, HF, and HF.NOB, respectively). Color schemes are the same in all panels: gray, RD; blue, HF; red, HF.NOB. *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA; #p < 0.05, ##p < 0.01, ###p < 0.001, t-test. Data are presented as mean ± SEM in bar and line graphs

Fig. 3

NOB enhanced circadian rhythms in skeletal muscle. Aged mice were fed with regular diet (RD), high-fat (HF) diet, or HF diet with NOB (HF.NOB). a Wheel-running activity (WRA) analysis. Representative actograms are shown (n = 4, 6, and 6 for RD, HF, and HF.NOB, respectively). b Circadian wheel-running activity was calculated from a by using the ActiView software (n = 4, 6, and 6 for RD, HF, and HF.NOB, respectively). c Calf muscle mass were measured after 20 weeks of treatment (n = 20, 15, and 15 for RD, HF, and HF.NOB, respectively). d Triglyceride contents in calf muscle (n = 7, 6, and 4 for RD, HF, and HF.NOB, respectively). e Fiber size (n = 3, 4, and 5 for RD, HF, and HF.NOB, respectively). f Expression of Bmal1, Npas2, and Dec1 genes (ZT6, n = 11, 10, and 10 for RD, 10 HF, and 10 HF.NOB, respectively; ZT18, n = 9, 8, and 8 for RD, HF, and HF.NOB, respectively). g Representative BMAL1, RORα, RORγ, and REV-ERBα protein levels and quantification results are shown (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA. Data are presented as mean ± SEM in bar graphs

Fig. 4

NOB activated mitochondrial OXPHOS gene expression and function. Aged mice were fed with regular diet (RD), high-fat (HF) diet, or HF diet with NOB (HF.NOB). a Heat map of pairwise expression comparison for mitochondrial respiration complex genes from RNA-sequencing analysis (n = 3). The z-score indicates the number of SDs away from the mean of expression. b Expression of two mitochondrial complex V genes known to be targets of RORs were restored by NOB (ZT6, n = 11, 10, and 10 for RD, HF, and HF.NOB, respectively; ZT18, n = 9, 8, and 8 for RD, HF, and HF.NOB, respectively). c Oxygen consumption rate (OCR), ATP-linked OCR, and relative reserve capacity in mitochondria isolated from calf muscle after 20 weeks of treatment. Average and representative assay results are shown (ZT6, n = 5, 4, and 5 for RD, HF, and HF.NOB, respectively; ZT18, n = 4). *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA; #p < 0.05, t-test. Data are presented as mean ± SEM in bar graphs

Fig. 5

NOB improved mitochondrial OXPHOS function. a Real-time qPCR analysis revealed enhanced expression of RORE-containing anti-oxidant genes in skeletal muscle from aged mice treated with NOB (ZT6, n = 7, 6, and 6 for RD, HF, and HF.NOB, respectively; ZT18, n = 9, 9, and 7 for RD, HF, and HF.NOB, respectively). b–g Effects of NOB in C2C12 cells. b, c NOB-suppressed tert-butyl hydroperoxide (tBHP) induced ROS production (n = 6). d NOB rescued ATP production against tBHP-induced oxidative stress (n = 4). e NOB restored mitochondrial OXPHOS function (PortA; Vehicle or 1 mM tBHP, PortB; 2.5 μg/ml Oligomycin, PortC; 4 μM FCCP, PortD; 2 μM Antimycin A). Representative assay condition (line graph) and Basal OCR after tBHP treatment (left bar graph), ATP-linked OCR (middle bar graph), and maximum respiration capacity (right bar graph) (n = 5) are shown. f NOB effect was attenuated by RORs knockdown (n = 5). g MitoTracker deep red staining of control and ROR-deleted C2C12 cells (n = 3). Two-way ANOVA analysis shows statistically significant differences between treatments and genotypes. *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA; ^††p < 0.01, two-way ANOVA; #p < 0.05, ##p < 0.01, ###p < 0.001, t-test. Data are presented as mean ± SEM in bar graphs

Fig. 6

NOB regulates TCA and glycolysis flux. Aged skeletal muscle was subjected to unbiased metabolomic analysis. a, b PLS-DA analysis results are 3D plotted for ZT6 and ZT18, and VIP score plotting for overall are shown. Blue and red dots indicate individual HF and HF.NOB-fed mice (ZT6, n = 7 and 8 for HF and HF.NOB, respectively; ZT18, n = 7 and 6 for HF and HF.NOB, respectively). c TCA cycle metabolites in skeletal muscle (ZT6, n = 7, 7, and 8 for RD, HF, and HF.NOB, respectively; ZT18, n = 7, 7, and 6 for RD, HF, and HF.NOB, respectively). d Glycolysis and TCA pathway analysis between HF and HF.NOB in skeletal muscle at ZT18. Red and blue indicate fold change ≥ 1.1 and ≤ 0.9, respectively (n = 7 and 6 for HF and HF.NOB, respectively). e Complex II activity in isolated mitochondria from calf muscle were measured (ZT6, n = 7, 6, and 8 for RD, HF, and HF.NOB, respectively; ZT18, n = 6). f Glycolysis-related metabolites in skeletal muscle (ZT6, n = 7, 7, and 8 for RD, HF, and HF.NOB; ZT18, n = 7, 7, and 6 for RD, HF, and HF.NOB, respectively). g NAD+ /NADH in skeletal muscle (ZT6, n = 7, 7, and 8 for RD, HF, and HF.NOB, respectively; ZT18, n = 7, 7, and 6 for RD, HF, and HF.NOB, respectively). *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA; #p < 0.05, ##p < 0.01, t-test. Data are presented as mean ± SEM in bar graphs. For box-whisker plots, box edges correspond to 25th and 75th percentiles, lines inside the box correspond to 50th percentiles, and whiskers include extreme data points

Fig. 7

NOB restored mitochondrial supercomplex formation affected by HF feeding. Aged mice were fed with regular diet (RD), high-fat (HF) diet, or HF diet with NOB (HF.NOB). a Mitochondrial OXPHOS supercomplexes (SCs) analyzed by BN-PAGE. Representative BN-PAGE of digitonin-solubilized mitochondria purified from skeletal muscle of RD, HF, and HF-NOB-treated mice is shown. b Upper panel: enlarged view of the box area from a. Lower panel: quantification of SC5 band intensities (ZT6, n = 4, 4, and 3 for RD, HF, and HF.NOB, respectively; ZT18, n = 5, 5, and 4 for RD, HF, and HF.NOB, respectively). c In-gel activity assay with Complex IV and subsequently with Complex I substrates performed as described in Methods. Brown and violet color bands indicate CIV and CI activity, respectively. d Western blot analysis of the respiratory Complexes I, III and IV isolated from skeletal muscle mitochondria in aged mice. Western blotting was performed as described in Methods. Lane 1: antibody to CI subunit 8, pseudo-colored blue; lane 2: antibody to CIII, subunit 2 (QCR2, Cor2), pseudo-colored red; lane 3: antibody to CIV subunit I (Cox1), pseudo-colored green; lane 4: overlay of lanes 2 and 3. Of note, in SC labels, n reflects the number of individual complexes in SCs, not the level of their oligomerization. *p < 0.05, one-way ANOVA. Data are presented as mean ± SEM in bar graphs