Aerobic and resistance exercise training reverses age-dependent decline in NAD+ salvage capacity in human skeletal muscle

Abstract

Aging decreases skeletal muscle mass and strength, but aerobic and resistance exercise training maintains skeletal muscle function. NAD⁺ is a coenzyme for ATP production and a required substrate for enzymes regulating cellular homeostasis. In skeletal muscle, NAD⁺ is mainly generated by the NAD⁺ salvage pathway in which nicotinamide phosphoribosyltransferase (NAMPT) is rate-limiting. NAMPT decreases with age in human skeletal muscle, and aerobic exercise training increases NAMPT levels in young men. However, whether distinct modes of exercise training increase NAMPT levels in both young and old people is unknown. We assessed the effects of 12 weeks of aerobic and resistance exercise training on skeletal muscle abundance of NAMPT, nicotinamide riboside kinase 2 (NRK2), and nicotinamide mononucleotide adenylyltransferase (NMNAT) 1 and 3 in young (≤35 years) and older (≥55 years) individuals. NAMPT in skeletal muscle correlated negatively with age (r² = 0.297, P < 0.001, n = 57), and VO₂ peak was the best predictor of NAMPT levels. Moreover, aerobic exercise training increased NAMPT abundance 12% and 28% in young and older individuals, respectively, whereas resistance exercise training increased NAMPT abundance 25% and 30% in young and in older individuals, respectively. None of the other proteins changed with exercise training. In a separate cohort of young and old people, levels of NAMPT, NRK1, and NMNAT1/2 in abdominal subcutaneous adipose tissue were not affected by either age or 6 weeks of high-intensity interval training. Collectively, exercise training reverses the age-dependent decline in skeletal muscle NAMPT abundance, and our findings highlight the value of exercise training in ameliorating age-associated deterioration of skeletal muscle function.

Keywords: NAMPT; Aging; NAD+ salvage pathways; exercise training; skeletal muscle.

Figure 1

NAD⁺ salvage pathway in mammals. NAM is the major NAD⁺ precursor in mammals. NA and NR can also be used to synthesize NAD⁺. NAMPT is the rate‐limiting step for the synthesis of NAD⁺ from NAM. Once NMN from NR and NAM or deamido‐NAD from NA are formed, it is converted to NAD⁺ by the action of NMNAT or NAD synthetase, respectively. The NAD⁺ generated can then be used for cellular redox reactions or as substrate for the activity of PARPs and sirtuins. NA, nicotinic acid; NAD⁺, nicotinamide adenine dinucleotide; NAMPT, nicotinamide phosphoribosyltransferase; NAM, nicotinamide; NR, nicotinamide riboside; NRK, nicotinamide riboside kinase; NPT, nicotinic acid phosphoribosyltransferase; NMN, nicotinamide mononucleotide; NMNAT, NMN adenylyltransferase; NNMT, nicotinamide‐N‐methyltransferase; NaMN, nicotinic acid mononucleotide; PRPP, phosphoribosyl pyrophosphate; ATP, adenosine triphosphate; PPi, inorganic pyrophosphate; PARPs, poly‐ADP ribose polymerases.

Figure 2

NAMPT protein levels in human skeletal muscle declines with increasing age and body fat. Correlation of the baseline skeletal muscle NAMPT with A, age (n = 57, r ² = 0.297, P < 0.001)**; B, BMI (n = 57, r ² = 0.151, P < 0.01)*; C, body fat (n = 57, r ² = 0.305, P < 0.001)*; D, VO₂peak (n = 57, r ² = 0.403, P < 0.001)*; E, lean body mass (n = 57, r ² = 0.139, P < 0.01)**; F, total blood cholesterol (n = 57, r ² = 0.113, P < 0.05)*; G, blood LDL cholesterol (n = 57, r ² = 0.112, P < 0.05); H, M‐value (n = 57, r ² = 0.172, P < 0.01)*. *Pearson correlation coefficient, r was used. Significance of Pearson coefficient was tested using the t‐distribution. **Spearman correlation coefficient, r _s was used.

Figure 3

NAMPT protein levels increase with aerobic and resistance exercise training in human skeletal muscle. (A) Skeletal muscle NAMPT in response to aerobic exercise for 12 weeks, ** indicates main effect of training versus baseline (P < 0.01) ## indicates main effect of age versus young (P < 0.01), no interaction effect. (B) Skeletal muscle NAMPT after 12 weeks of resistance exercise training, ** indicates main effect of training versus baseline (P < 0.01), tendency of effect of age P = 0.077, no interaction effect. (C) Skeletal muscle NRK2 after aerobic exercise (n = 9). (D) Skeletal muscle NRK2 of resistance exercised individuals. (E) Skeletal muscle NMNAT1 in aerobic exercise. (F) Skeletal muscle NMNAT1 in resistance exercise. (G) NMNAT3 in aerobic exercise, # indicates main effect of age versus young (P < 0.05). (H) NMNAT3 protein levels in resistance exercise. Number in bar graph indicates number of participants. Statistical test performed: 2‐way RM ANOVA with Tukey as post hoc.

Figure 4

Plasma eNAMPT levels decrease with fasting plasma glucose and insulin levels, but are unaffected by exercise training. (A) Correlation of fasting plasma glucose and eNAMPT levels (n = 39, r ² = 0.212, P < 0.01)*. (B) Correlation of insulin and eNAMPT levels (n = 39, r ² = 0.141, P = 0.053)**. (C) Correlation of HOMA‐IR and eNAMPT levels (n = 39, r ² = 0.163, P < 0.05)**. *Pearson correlation coefficient, r was used. Significance of Pearson coefficient was tested using the t‐distribution; **Spearman correlation coefficient, r _s was used; eNAMPT levels in young (≤35 years old) and older (≥55 years old) participants before and after D, aerobic and E, resistance exercise training. Number in bar graph indicates number of participants. Statistical test performed: 2‐way RM ANOVA with Tukey as post hoc.

Figure 5

NAD⁺ salvage enzyme levels in adipose tissue of both young and older individuals are unaltered by exercise training. Subcutaneous adipose tissue protein levels of NAD⁺ salvage enzymes at baseline and after high‐intensity interval training (HIIT) for 6 weeks: (A) NAMPT. (B) NRK1. (C) NMNAT1 and (D) NMNAT2. Number in bar graph indicates number of participants. Statistical test performed: 2‐way RM ANOVA with Tukey as post hoc.