Introduction

During aging regulatory mechanisms of carbohydrate and lipid homeostasis undergo distinct alterations that can be modulated by nutritional or pharmacological manipulations.

The mechanisms of body mass regulation have become an important object of animal and human studies in the context of both ‘epidemic of obesity’ and increase of the elderly population (Colleluori and Villareal 2021). In old people, there is an increasing prevalence of elevated body mass index (BMI), impaired glucose tolerance, insulin resistance leading to type 2 diabetes, other features of metabolic syndrome, and cardiovascular disease. In the treatment of obesity regular physical activity and reduction of the caloric intake are being advocated for the majority of patients (Zubrzycki et al. 2018). Nonetheless, weight loss in the elderly carries the risk of concomitant loss of lean body mass and bone mineral density (Colleluori and Villareal 2021). Long-term moderate calorie restriction (CR) in humans improves metabolic health and slows biological aging (Kebbe et al. 2021). However, even short-term CR may result in positive metabolic outcomes, including improved cardiovascular risk factors, insulin sensitivity, and antioxidant capacity (Kanikowska et al. 2021; Vion et al. 2021).

Sirtuins, which are NAD+-dependent protein deacetylases with critical roles in metabolism, stress responses, and aging processes, are also key players in CR-induced metabolic improvements (Hebert et al. 2013; Tauriainen et al. 2011). Among the seven mammalian sirtuins (SIRT1–SIRT7), the primarily mitochondrial SIRT3 deacetylates multiple mitochondrial enzymes in response to nutrient stresses such as fasting and CR (Hebert et al. 2013; Hirshey et al. 2010). In this way, SIRT3 regulates mitochondrial fatty acid β-oxidation (Hirschey et al. 2010), ketone body production (Shimazu et al. 2010), the urea cycle (Hallows et al. 2011), antioxidant capacity (Qiu et al. 2010), and multiple other processes (Hebert et al. 2013).

Importantly, SIRT3 activity may mitigate metabolic dysfunctions that are often associated with aging in the cardiac and skeletal muscles (Cao et al. 2022; Vargas-Ortiz et al. 2019). Aging of the cardiac muscle is associated with impairment of mitochondrial abundance and function, decreased fatty acid oxidation, cardiac lipotoxicity, and accumulation of reactive oxygen species (ROS), leading to increased risk of cardiovascular disease (Sithara and Drosatos 2021). Age-related changes in skeletal muscles include the loss of muscle mass and strength (sarcopenia), decreased protein synthesis and increased protein degradation, mitochondrial dysfunction, chronic low-grade inflammation, and insulin resistance (Bilski et al. 2022).

In the heart and skeletal muscles, energy metabolism is strictly regulated and depends on the amount and type of substrates. Peroxisome proliferator-activated receptor-γ coactivator-1α (PGC-1α) is involved in glucose and fatty acid metabolism, fibre type switching in skeletal muscle, cardiac diseases, and other processes (Cheng et al. 2018). In both types of muscles, PGC-1α co-activates peroxisome proliferator-activated receptor beta/delta (PPAR-β/δ) to modulate the transcription of genes involved in fatty acid oxidation and oxidative phosphorylation, as well as glucose uptake (Arany et al. 2005; Kleiner et al. 2009; Park et al. 2020; Wang et al. 2010). Studies in mice with inducible knockout of Ppar-β/δ revealed that in the adult heart PPAR-β/δ deficiency decreased mitochondrial biogenesis and reduced tissue levels of PGC-1α, resulting in attenuated fatty acid and glucose oxidation (Wang et al. 2010). Activation of PPAR-β/δ increases the transcription of genes involved in fatty acid catabolism in skeletal muscles of mouse (Dressel et al. 2003) and human (Krämer et al. 2007). PGC-1α and PPAR-β/δ co-activate the transcription of many genes, including the one encoding carnitine palmitoyltransferase- I (CPT-I), a crucial enzyme for oxidation of fatty acids as the primary energy source in muscles (Liu et al. 2007). CPT-I, located in the mitochondrial outer membrane, catalyzes the transfer of long chain acyl groups from coenzyme A to carnitine. In this form, the acyl groups pass into the mitochondrial intermembrane space, and further into matrix where they may be broken down via β-oxidation (Liu et al. 2007).

Aging is associated with increased glucose utilization as energy substrate for the heart (Sithara and Drosatos 2021). Glucose uptake in the cardiac and skeletal muscles takes place predominantly through the insulin-dependent glucose transporter type 4 (GLUT4) (Aerni-Flessner et al. 2012; Kern et al. 1990). Insulin stimulation translocates intracellular GLUT4 storage vesicles to the plasma membrane (Kern et al. 1990). Moreover, skeletal muscles play a critical role in glucose homeostasis. Since muscles comprise about 42% of body mass in normal-weight adults, they are responsible for most of the whole body insulin-induced glucose disposal (Bilski et al. 2022).

Although short-term calorie restriction (SCR) has been advocated as a weight-loss regimen in overweight or obesity (Kanikowska et al. 2021; Vion et al. 2021; Zubrzycki et al. 2018), its molecular effects may also be affected by aging. The metabolism of cardiac and skeletal muscles strongly affects health and is subjected to age-related changes. Therefore, we decided to assess the effects of 60% CR lasting 30 days in these muscle types in young and old male rats. The duration of CR was chosen to approximately reflect 4–6 month-long CR regiments in humans (Zubrzycki et al. 2018). We measured the blood serum concentrations of key metabolites and hormones reflecting whole-body metabolic outcomes. Next, we focused on SCR’s effects on cardiac and skeletal muscle expression levels of key genes involved in mitochondrial fatty acid β-oxidation (the rate-limiting enzyme Cpt-I and its transcriptional regulator Ppar-β/δ) and glucose uptake (Glut4 and its transcriptional regulator Pgc-1α). In addition, we analysed the gene and protein expression of SIRT3 as a crucial regulator of mitochondrial metabolism and effector of CR. We also measured the effects of refeeding SCR-subjected rats with full-diet for two or four days. Our results provide new data on aging-related differences in the response to SCR and refeeding in skeletal and cardiac muscles of male rats.

Materials and methods

Animals and tissue collection

The experiment was performed on young (4-month-old at the onset of the experiment) and old (24-month-old) male Wistar-Han rats maintained at Specific Pathogen Free conditions in the Academic Laboratory Animal Centre in Gdansk, Poland, at 22 °C under a 12 h/12 h light/dark cycle. Littermates were housed in cages individually or in pairs of similar weight. All animals were fed a standard, sterilized chow, which provided 52% of metabolic energy from carbohydrates, 11% from fat, and 37% from proteins (Labofeed B; Wytwornia Pasz Morawski, Kcynia, Poland). Rats had free access to water throughout the experiment. Food intake was measured every second day during 3-week-long pre-trial period to determine the baseline food intake, which was approximately 21 g/day and 25 g/day for the young and old rats, respectively. Rats from control groups (C; n = 16 and n = 12, young and old, respectively) were fed ad libitum during the whole experiment. Calorie-restricted animals received pre-weighted portions of chow equivalent to 60% of the baseline food intake, for 30 days on a daily basis, 2 h after lights on, and it was found that they consumed entire daily portions. At the end of the SCR period some animals were euthanized (SCR group; n = 20/14, young/old) while others were refed ad libitum for 2 (SCR + 2; n = 14/14) or 4 days (SCR + 4; n = 14/14). Under full anaesthesia (90 mg/kg ketamine and 10 mg/kg xylazine, i.p.), blood was collected from the heart, followed by decapitation. Tissue samples of cardiac muscle and vastus lateralis muscle (mixed fast-twitch oxidative-glycolytic fibres) were collected, immediately frozen in liquid nitrogen, and stored at − 80 °C until analyses. All animal experimental procedures had been authorized by the Local Ethical Committee in Gdansk (protocol no 11/2011) and were conducted in agreement with the institutional and European Community guidelines.

Gene expression

Frozen tissues samples of the cardiac and skeletal muscles were homogenised (MagNALyser homogenizer, Roche Diagnostics, Indianapolis, IN, USA) and total RNA was extracted from homogenates using the Total RNA kit (A&A Biotechnology, Gdynia, Poland). 2 µg of RNA were reverse transcribed to cDNA with the RevertAid kit (Thermo Fischer Scientific, Fitchburg, WI, USA) and oligo(dT)18 primers (Sigma-Aldrich, Munich, Germany). Relative quantification of Cpt-I, Glut4, Pgc-1α, Ppar-β/δ, and Sirt3 genes’ expression was performed by real-time PCR (StepOne Plus thermal cycler, Applied Biosystems-Life Technologies, Grand Island, NY, USA), using SybrGreen Master Mix (A&A Biotechnology). The mRNA levels were assessed relative to the mean expression of housekeeping genes: acidic ribosomal phosphoprotein P0 (36B4) and cyclophilin A (CycloA), according to the 2−ΔΔCT method (Livak and Schmittgen 2001). The primers were designed using Primer3Plus software based on BLAST, ENSMBL, and AceView databases. The primer pairs were (5′→3′): Cpt-I ATGTTTGACCCAAAGCAGTACCCC and TCGCCTGCGATCATGTAGGAAAC; Glut4 AGGCCGGGACACTATACCCTATTC and AAACTGAAGGGAGCCAAGCACAG; Pgc-1α CACGTTCAAGGTCACCCTACAGC and TAAATCACACGGCGCTCTTCAAT; Ppar-β/δ ACAAGGCCTCAGGCTTCCACTAC and TCCGATCGCACTTCTCGTACTTG; Sirt3 AAGCTGGTTGAAGCTCATGGGTC and TCCAGGGAGGTCCCAAGAATGAG; 36B4 CTCAGTGCCTCACTCCATCA and GGGGCTTAGTCGAAGAGACC; CycloA TGTCTCTTTTCGCCGCTTGCTG and CACCACCCTGGCACATGAATCC.

SIRT3 protein measurement by western blotting

Semi-quantification of SIRT3 protein levels was performed in duplicate for every young and old rats’ group. Protein samples were obtained through tissue lysis with Mammalian Cell Extraction Kit (BioVision, Milpitas, CA, USA), then separated by 10% SDS-PAGE, transferred to PVDF membranes (Invitrogen-Life Technologies, Carlsbad, CA, USA), and blocked with 5% non-fat milk in TBS. The membranes were incubated with anti-SIRT3 antibodies (SAB5700222, Merck Millipore, Saint Louis, MI, USA) at 4 °C overnight. Following washing, the membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (A9169, Merck Millipore, Burlington, MA, USA) for 2 h at room temperature. GAPDH protein levels were measured as loading control (G9295, Merck Millipore). Bands were visualized using Chemiluminescent Peroxidase Substrate (Merck Millipore) and developed against light-sensitive X-ray films. QuantityOne software (Bio-Rad, Hercules, CA, USA) was used for densitometric analysis.

Determination of carnitine palmitoyltransferase I activity

CPT-I activity was measured in cardiac and skeletal muscles of control and SCR rats (5 animals per group). 100 mg tissue samples were minced and homogenised in a Dounce homogenizer in 1 ml of cold buffer (pH 7.4), containing 250 mM sucrose, 10 mM TRIS, and 0.5 mM EDTA. Homogenates were centrifuged at 300 g for 3 min at 4 °C. CPT-I activity and total protein content were measured in the supernatant at 412 nm at 37 °C using a spectrophotometer with a thermostatic holder (Beckman Coulter Inc., USA). The reaction mixture was composed of 1 M Tris–HCl (pH 8.0), 1 mM DTNB, 100 µM palmitoyl-CoA, and 50–100 µl of supernatant. The enzyme activity was calculated as the difference in the absorbance of two measurements: with or without 1.67 mM l-carnitine, in relation to the total amount of protein in the tissue homogenates (nmol/ min/ mg protein).

Measurements of serum concentrations of metabolites

Blood samples were allowed to clot and were subsequently centrifuged (2000 g, 15 min, 4 °C). Sera were frozen at − 80 °C until analyses. Measurements of the concentrations of glucose, free fatty acids (FFA), triglycerides (TG), total cholesterol (Ch), HDL-cholesterol (HDL), urea, and albumin in rat blood sera were performed by commercially available standard diagnostic tests at the Clinical Laboratory of the Medical University of Gdansk, Gdansk, Poland.

Serum leptin and insulin concentrations

The concentrations of leptin and insulin in blood sera were measured by rat-specific ELISA tests (Merck Millipore, St Charles, MO, USA).

Statistical analysis

Statistical analysis was performed using GraphPad Prism v.6 (GraphPad Software, San Diego, CA, USA). Because not all sets of data followed normal distribution (Shapiro-Wilk test), analyses of group pairs were performed using the Mann–Whitney test. Correlation coefficients between two groups were calculated using Spearman rank correlation test. Data are presented as mean ± SEM. Statistical significance was considered for p < 0.05.

Results

Body mass and concentrations of metabolites, leptin and insulin in blood serum

Mean body mass was substantially higher in old vs. young animals (526 ± 11.2 g and 375 ± 6.1 g, respectively, p < 0.001). The body mass decreased after SCR by 6.1% and 10.9% in young and old rats, respectively (p < 0.001 for both). Young rats restored the pre-SCR body mass after 2 days of ad libitum feeding, whereas the body mass of old rats was lower than the pre-SCR by 7.4% (p < 0.01) after 4 days of refeeding.

Table 1 presents the blood serum levels of major biochemical factors and hormones in control and treated rats. In old as compared to young control rats, glucose and insulin concentrations were comparable, TG, FFA, Ch, HDL, total protein, and leptin concentrations were higher, whereas urea and albumin concentrations were lower in old rats (Table 1).

Table 1 Concentrations of metabolites and hormones in blood serum of young and old rats after short-term calorie restriction and refeeding. Data are means ± SEM of 6–20 animals per group
Full size table

SCR decreased serum glucose concentration by 42% and 32% in young and old rats, respectively (Table 1). Subsequent ad libitum feeding increased the concentration in young rats; however, in old rats glucose remained below control level throughout 4 days of refeeding. Insulin concentration in young rats decreased after SCR and reverted to control level after 2 days of refeeding. In old rats, insulin concentration decreased after SCR, returned to control values after 2 days of refeeding, but decreased again after 4 days.

SCR decreased TG concentration by 48% and 63% in young and old rats, respectively (Table 1). Two days of ad libitum feeding increased the concentration by 22% and 53%, but did not restore the control levels throughout 4 days of refeeding. SCR and refeeding had no effect on serum FFA concentration in young rats (Table 1). In old rats the FFA concentration increased by 45% after SCR, and reverted to control level upon refeeding.

HDL-cholesterol decreased after SCR by 27% in young rats and by 31% in the old ones (p < 0.01 for both; Table 1). Refeeding of young rats increased the HDL-cholesterol to control levels, however, in old rats the concentration remained below control throughout 4 days of refeeding (p < 0.01). The TG/HDL ratio decreased after SCR by 27% (p < 0.05) and 38% (p < 0.001) in young and old rats, respectively, and increased upon refeeding up to control levels.

Cholesterol, urea, and albumin concentrations in blood serum were unaffected by SCR and refeeding in either age group (Table 1). SCR caused small but significant decreases in total protein concentrations in blood serum: by 4.3% (p < 0.001) in young rats and by 3.1% (p < 0.05) in old animals. Control protein level was restored by refeeding in old rats, but not in the young ones.

SCR reduced serum leptin concentrations by 73% and 42% in young and old rats, respectively (Table 1). Refeeding elevated leptin levels, however, did not restore the pre-SCR values throughout 4 days of refeeding.

Age-related differences in gene expression between cardiac and skeletal muscles

In cardiac muscle, Cpt-I gene expression was higher by 75% in control old than young rats (Fig. 1a), and CPT-I activity tended to increase in old rats (p = 0.06) (Fig. 1e). There were no differences in the mRNA levels of Ppar-β/δ (Fig. 1c), Glut4, Pgc-1α (Fig. 2a, c), and Sirt3 (Fig. 3a). In skeletal muscle, Cpt-I gene expression was lower in control old rats (Fig. 1b), even though CPT-I enzymatic activity was unaffected by aging (Fig. 1e). Ppar-β/δ expression did not differ between young and old animals (Fig. 1d), whereas Glut4 and Pgc-1α (Fig. 2b, d) mRNA levels were 2-fold lower in control old rats. Sirt3 expression was lower by 44% in skeletal muscle of old as compared to young rats (Fig. 3b).

Fig. 1
figure 1

Effects of short-term calorie restriction and refeeding on gene expression of Cpt-I and Ppar-β/δ (a–d) and CPT-I activity (e) in the cardiac and skeletal muscles. The mRNA expression levels were normalised to young control animals (expression level = 1); data are means ± SEM of 6–20 rats per group. C control rats, SCR short-term calorie restricted animals, SCR + 2 calorie-restricted and refed for 2 days, SCR + 4 calorie-restricted and refed for 4 days. Statistical significance: a vs. control rats (of the same age), b vs. calorie restricted animals, c vs. animals subjected to calorie restriction and refed for 2 days, *p < 0.05, **p < 0.01, ***p < 0.001

Full size image
Fig. 2
figure 2

Effects of short-term calorie restriction and refeeding on gene expression of Glut4 (a, b) and Pgc-1α (c, d) in the cardiac and skeletal muscles. The mRNA expression levels were normalised to young control animals (expression level = 1); data are means ± SEM of 6–20 animals per group. C control rats, SCR short-term calorie restricted animals, SCR + 2 calorie-restricted and refed for 2 days, SCR + 4 calorie-restricted and refed for 4 days. Statistical significance: a vs. control rats (of the same age), b vs. calorie restricted animals, c vs. animals subjected to calorie restriction and refed for 2 days, *p < 0.05, **p < 0.01, ***p < 0.001

Full size image
Fig. 3
figure 3

Effects of short-term calorie restriction and refeeding on sirtuin 3 gene (a, b) and protein (c, d) expression in the cardiac and skeletal muscles. a, b Sirt3 mRNA expression levels were normalised to young control animals (expression level = 1); data are means ± SEM of 6–20 animals per group. c, d SIRT3 protein levels were measured in duplicate relative to GAPDH and analysed semi-quantitatively through densitometry; representative blots are shown. C control rats, SCR short-term calorie restricted animals, SCR + 2 calorie-restricted and refed for 2 days, SCR + 4 calorie-restricted and refed for 4 days. Statistical significance: *p < 0.05, **p < 0.01, ***p < 0.001

Full size image

Effects of SCR and refeeding on gene expression in cardiac and skeletal muscles of young and old rats

In the cardiac muscle of young rats, SCR increased Cpt-I expression by 90% (Fig. 1a) and CPT-I activity by 100% (Fig. 1e). In old SCR animals, Cpt-I expression (Fig. 1a) and the enzyme activity (Fig. 1e) in the cardiac muscle were unaffected by the diet, while 4-day refeeding decreased Cpt-I expression by 30%. In skeletal muscle of young rats, Cpt-I mRNA level and CPT-I activity were unaffected by SCR, while refeeding for 2 days decreased Cpt-I expression by 46% (Fig. 1b, e). In the old rats’ skeletal muscle SCR and refeeding did not change Cpt-I expression (Fig. 1b) or enzymatic activity (Fig. 1e).

In cardiac muscle of young rats, Ppar-β/δ expression increased by 70% after SCR, and decreased to control level upon refeeding, whereas in old rats there were no changes (Fig. 1c). In skeletal muscle of young rats SCR caused a 70% increase in Ppar-β/δ expression, which decreased upon refeeding to a level 79% lower than control after 4 days, while in old rats there were no changes (Fig. 1d).

In cardiac muscle of young rats, SCR increased Glut4 expression by 97%, and refeeding decreased the expression to control level (Fig. 2a). In contrast, the treatments did not affect the cardiac expression of Glut4 in old rats. In skeletal muscle of young rats SCR did not affect, while refeeding decreased Glut4 expression (Fig. 2b). In skeletal muscle of old rats there were no changes in Glut4 mRNA levels upon treatment (Fig. 2b).

In cardiac muscle of young rats, SCR increased Pgc-1α expression by 230%, and refeeding decreased the expression to control level (Fig. 2c). In old rats, Pgc-1α expression did not change upon SCR, but increased by 70% after refeeding for 2 days, and decreased to control level after 4 days (Fig. 2c). In skeletal muscle of young rats, SCR increased Pgc-1α expression by 75%, and refeeding decreased the expression below control level (Fig. 2d). In old rats, SCR decreased Pgc-1α expression 2.5-fold, and refeeding for 4 days restored the control level.

In cardiac muscle of young rats on SCR, the mRNA level of Pgc-1α strongly correlated with Glut4 expression (r = 0.82, p = 0.00002), however, not in the old SCR animals. In skeletal muscle of old SCR rats the expression of Pgc-1α positively correlated with the expression of Glut4 (r = 0.79, p = 0.02), however, no significant correlation was found for the young SCR animals.

In cardiac muscle, SCR increased Sirt3 mRNA expression by 288% and 45% in young and old rats, respectively (p < 0.05 and p = 0.18, respectively; Fig. 3a). Refeeding for 2 and 4 days restored control levels of Sirt3 expression in both age groups. Similarly, SIRT3 protein levels were upregulated by about 50% after SCR and reverted to control levels after refeeding of both young and old rats (Fig. 3c). In skeletal muscle of young animals, SCR decreased by 43% the Sirt3 mRNA, which remained below control level after 4 days of refeeding (Fig. 3b). In the old rats’ skeletal muscle, neither SCR nor refeeding affected Sirt3 gene expression. SIRT3 protein expression in skeletal muscle of young rats was elevated after SCR and refeeding for 2 and 4 days by 45%, 75%, and 50%, respectively, as compared to control. However, in old rats SCR and 2-day refeeding were associated with 50% lower SIRT3 protein expression in skeletal muscle (Fig. 3d).

In cardiac muscle of young SCR rats, Sirt3 expression correlated with those of Pgc-1α (r = 0.73, p = 0.0006) and Glut4 (r = 0.94, p = 10− 6). In the cardiac muscle of old rats Sirt3 expression correlated with those of Ppar-β/δ (r = 0.65, p = 0.029) and Glut4 (r = 0.64, p = 0.048). In skeletal muscle of old SCR rats, Sirt3 expression correlated with those of Ppar-β/δ (r = 0.86, p = 0.006) and Glut4 (r = 0.88, p = 0.004).

Discussion

Our study compared transcriptional-level changes in response to SCR followed by ad libitum feeding for up to 4 days in cardiac and skeletal muscles of young and old male rats. Additionally, we assessed gene expression and protein levels of sirtuin 3 that was shown to play an important role in the control of mitochondrial lipid and carbohydrate metabolism. The results point to age-related differences in the metabolic adaptations to SCR in these two types of striated muscle tissue (Fig. 4).

Fig. 4
figure 4

The molecular effects of short-term calorie restriction on factors regulating energy substrate utilization differ in the cardiac and skeletal muscles and in young and old rats—summary of results

Full size image

Firstly, we compared young and old rats to identify the transcriptional changes evoked by aging. We found that aging of the cardiac muscle was associated with increases in Cpt-I mRNA expression and matching enzymatic activity, similarly to the results reported in old Sprague-Dawley rats (Zhao et al. 2014). On the contrary, in vastus lateralis skeletal muscle of old rats we observed decreased Cpt-I expression, though without corresponding decline in enzymatic activity. Other researchers have reported either decreased CPT-I activity in skeletal muscle of old Fischer 344 rats (Kim et al. 2009) and old men (Kaczor et al. 2006) or increased CPT-I mRNA and protein levels in the skeletal muscle of old Sprague-Dawley rats (Zhao et al. 2014). It should be noted that control of the muscle-type CPT-Ib (analysed in our study) is largely performed at transcriptional level, through fatty acid-induced activation of PPAR-β/δ in a PGC-1-dependent manner, as demonstrated in C2C12 skeletal muscle cell cultures (Dressel et al. 2003), and preferentially through PPARα in primary cultures of cardiomyocytes (Brandt et al. 1998). In contrast, the liver-type CPT-Ia has a much higher affinity to malonyl-coenzyme A, which strongly inhibits its activity (Brandt et al. 1998). Since we observed no age-related change in Ppar-β/δ expression, it is likely that the age-related changes in Cpt-I gene expression could be PPAR-β/δ-independent in both the cardiac and skeletal muscles. Because aging of skeletal muscle was associated with parallel decreases of Cpt-I and Pgc-1α mRNA levels, a lower PGC-1α availability could have contributed to the age-related reduction of Cpt-I expression.

We observed an age-related decrease of Glut4 expression in skeletal muscle, in line with other studies in rat (Larkin et al. 2001; Qiang et al. 2007) and mice (Sczelecki et al. 2014). Our finding of the age-related reduction of Pgc-1α expression, one of the transcriptional regulators of Glut4 expression (Leick et al. 2010), as well as the reduction of Sirt3, could both contribute to this result. Interestingly, in our SPF rat colony we found no difference in the blood glucose levels between young and old rats despite the fact that skeletal muscles are the major site of glucose disposal. In pulmonary hypertension rat model signalling through SIRT3 and AMPK in skeletal muscle was shown to be indispensable for the blood glucose-lowering effect of nitrite (Lai et al. 2016).

It has been shown that prolonged moderate CR without malnutrition (decrease of daily calorie intake by 25%) positively impacts human health and lifespan (Kebbe et al. 2021). However, long-term adherence to this form of therapy may be a major obstacle for the majority of human patients. A more feasible option is short-term CR (up to 6 months), which exerts positive effects beyond reduction of body weight in rats (Pires et al. 2014; Wang et al. 2016) and humans (Vion et al. 2021; Zubrzycki et al. 2018). The duration of SCR regimen in most animal studies varies from 21 days (e.g. Pires et al. 2014) to 2 months (e.g. Wang et al. 2016). With the aim of comparing sexually-mature young and old male rats, we used a 30-day SCR protocol, with additional 2 or 4 days of refeeding to mimic the effects of returning to full diet.

The SCR-associated lowering of glucose, insulin, and TG blood serum levels, as well as TG/HDL ratio in both age groups indicate that even this short-term diet may improve cardio-metabolic risk factors. SCR caused only mild increases in FFA levels, reflecting lipid mobilization from adipose tissue, which suggests that the SCR was a mild intervention. At the same time SCR decreased the serum concentrations of HDL-cholesterol, a potent anti-atherogenic, anti-oxidant and anti-inflammatory factor (Yang et al. 2019). Other authors reported no change in HDL-cholesterol level in obese Wistar rats after 28-day CR (de Castro et al. 2020) or increases after 12-week CR regimens in Wistar rats (Chen et al. 2015). The authors ascribed the HDL elevation to increased lipoprotein lipase-mediated catabolism of TG-rich lipoproteins and transfer of the resultant lipid moieties to HDL particles (Chen et al. 2015). However, HDL proteomic analyses in metabolic syndrome patients who underwent a 12-week lifestyle therapy indicated that there could be improvements in HDL function preceding any significant changes in plasma HDL levels (Mathew et al. 2018). Of note, a recent study showed that the metabolic outcomes of dietary interventions depend on the timing of food intake, since rats who were for 10 weeks intermittently fasted during the night (i.e. active phase) had reduced HDL-cholesterol levels, in contrast to those fasted during the day (Hanjani et al. 2021). Nevertheless, because the TG/HDL ratios were reduced after our SCR regimen in both studied age groups, it can be concluded that the diet resulted in overall improvement of cardio-metabolic risk.

Interestingly, we noted that SCR caused greater relative reduction of body weight in old than in young rats. Regrettably, we found no other published data directly comparing SCR in young and old animals. We hypothesise that the smaller reduction in body weight of young animals could reflect more efficient energy preservation. Alternatively, the result may be related to some degree of dehydration in old animals, despite water accessibility.

Our study proved that the aging process differently shapes the effects of SCR on the expression and enzymatic activity of CPT-I in rat cardiac and skeletal muscles. In line with the elevated TG levels, we found that SCR increased gene expression and activity of CPT-I in the cardiac muscle of young but not old rats. Moreover, we noted that SCR increased the expression of Ppar-β/δ (which activates the transcription of Cpt-I in the muscle (Dressel et al. 2003)) in cardiac and skeletal muscles of young but not old rats. However, in the skeletal muscle of both studied age groups Cpt-I expression and the enzyme activity did not change after the SCR diet. Similar results were observed in young male KKAy obese/diabetic mice after 8 weeks of 70% CR (Huang et al. 2006). On the contrary, increased abundance and activity of CPT-I in the skeletal muscle was reported in 17- and 24-month-old male Fisher 344 rats subjected to 3-week-long 60% CR (Kim et al. 2009), and in female rats fed for 27 weeks with high-fat diet and subjected for the last 7 weeks to 60% CR (Pattanakuhar et al. 2019). Thus, aging inhibits the induction of Ppar-β/δ and Cpt-I expression in response to SCR in rat cardiac muscle.

The SCR-associated decrease of blood glucose can be explained by greater glucose uptake by skeletal muscles of calorie-restricted animals (Wang et al. 2016). However, in our study in the skeletal muscle of young rats Glut4 expression was unaffected by SCR. This result falls in line with other reports showing that CR in rodents did not increase the muscle abundance of the GLUT4 glucose transporter (Argentino et al. 2005; Sequea et al. 2012; Wang et al. 1997). 20-day-long CR was shown to cause greater insulin-stimulated translocation of GLUT4 to cell surface membrane in rat epitrochlearis muscle, indicating increased insulin signalling (Dean et al. 1998). PGC-1α was reported to stimulate Glut4 expression in murine and rat cultured muscle cells and skeletal muscle (Michael et al. 2001). Our finding of increased expression of Pgc-1α upon SCR of young rats is similar to the up-regulation of Pgc-1α mRNA levels reported by Song et al. (2014) in the gastrocnemius muscle of young male Sprague-Dawley rats subjected for 8 weeks to 60% CR. In the skeletal muscle of old rats in our study, SCR exacerbated the age-related decline in Pgc-1α expression, with parallel (though insignificant) reduction of Glut4 mRNA. Interestingly, lifelong CR has been reported to increase Pgc-1α mRNA level in the skeletal muscle of old male B6D2F1 mice but not in the young and middle-aged animals, whereas PGC-1α protein expression was not affected by CR in any of the age groups (Miller et al. 2012). It should be noted that GLUT4 and PGC-1α are regulated mostly post-transcriptionally, e.g. through phosphorylation, acetylation, or interactions with other proteins (Bogan and Kandror 2010; Krämer and Handschin 2019; Rodgers et al. 2005).

In the cardiac muscle of young rats, SCR stimulated Glut4 expression, suggesting that SCR may potentiate the use of glucose in the heart of young, but not old rats. This could be achieved through the upregulation of SIRT3, whose endothelial expression was shown to regulate cardiomyocyte glucose availability in the mouse heart (Zeng et al. 2020). In support of this hypothesis, in our study the expression of Sirt3 in the heart of young SCR rats correlated with Glut4 expression. Moreover, Glut4 gene expression was induced by PGC-1α in murine C2C12 myoblasts, potentiating glucose transport into these cells (Michael et al. 2001). Both Sirt3 and Glut4 correlated with Pgc-1α mRNA levels, similarly to what was shown in mice (Wende et al. 2017). This may indicate that SCR may co-ordinately regulate these three factors affecting cardiac glucose metabolism. PGC-1α may also stimulate fatty acid oxidation in the heart (Arany et al. 2005), thus the observed Pgc-1α increase may contribute to the greater utilization of lipids, which was reported in the cardiac muscle of CR mice (Lee et al. 2002). Ageing blunted the cardiac response to SCR, as old rats showed little change in Glut4, Sirt3 or Pgc-1α expression.

SIRT3 is a key factor involved in the adaptation to nutrient stresses such as CR (van de Ven et al. 2017). We found that SCR stimulated SIRT3 gene and protein expression in the cardiac muscle of young rats. Other researchers demonstrated that induction of SIRT3 may be central to the beneficial outcome of CR, such as metabolic adaptations and antioxidant defence (Qiu et al. 2010; Tauriainen et al. 2011). Moreover, upregulation of SIRT3 by CR may potentiate the use of glucose in the cardiac muscle, as the endothelial expression of SIRT3 was shown to regulate cardiomyocyte glucose availability in the mouse heart (Zeng et al. 2020). In support of this hypothesis, in our cohort of young SCR rats Sirt3 correlated with Glut4 expression. Importantly, the intensity of CR may differently affect the cardiac SIRT3 expression, since SIRT3 protein levels in myocardium of female SpragueDawley rats increased after two months of 25% CR, but decreased after 45% CR (Yu et al. 2018). However, in our cohort of old Wistar-Han rats, SCR diminished SIRT3 induction in the cardiac muscle. Even though we did not measure SIRT3 activity, it should be noted that SIRT3 is activated through its deacetylation catalysed by SIRT1, whose levels drop upon ageing in many tissues (Kwon et al. 2017).

Differently than in the cardiac muscle, the skeletal muscle expression of Sirt3 decreased after SCR of young rats, even though SIRT3 protein levels were elevated. The results of our 4-week long CR model differ from that of 12-week-long CR, which stimulated Sirt3 gene expression in mixed-type skeletal muscle of young C57BL/6 male mice (Jing et al. 2011). As noted above, CR effect on SIRT3 levels depends on the diet intensity, and possibly genetic background of the subjects. In contrast to the young ones, in old rats SIRT3 protein was lower after SCR, though without corresponding mRNA changes. This discrepancy may be explained by the aging-related decline in the efficiency of protein synthesis or increased protein degradation (Bilski et al. 2022).

We investigated the studied regulatory factors after 2 or 4 days of refeeding to obtain more information about the changes in their expression upon transition from calorie-restricted to normal diet. Refeeding of young rats quickly reversed the SCR-induced transcriptional changes in the cardiac muscle. Only Cpt-I expression remained elevated above control level after 4 days of refeeding. Notably, in the cardiac muscle of old rats neither CR nor refeeding caused any significant transcriptional changes, with the exception of Pgc-1α expression transiently elevated upon 2-day-long refeeding. There is a scarcity of literature data on the transcriptional effects of refeeding in the cardiac muscle. Our data suggest that in the cardiac muscle of young rats, the transcriptional changes in genes regulating energy substrate utilization are not preserved upon return to normal feeding. In the cardiac muscle of old rats, the response to dietary manipulations appears to be blunted. However, we found no other reports that compared the refeeding response in various age groups of rodents.

In the skeletal muscle of young rats, refeeding reduced the studied genes’ expression below control level, corresponding to reported Pgc-1α mRNA reduction in tibialis anterior muscle of young male Sprague-Dawley rats (Zheng et al. 2012). The transcriptional changes observed in our study could contribute to the phenomenon of catch-up fat after calorie restriction, which favours disproportionate recovering of fat mass rather than lean mass in model animals and humans. It was reported that young male Sprague-Dawley rats, subjected to refeeding after 2 weeks of 50% CR, developed insulin resistance in skeletal muscles and insulin hyperresponsiveness in adipose tissue, leading to glucose redistribution from skeletal muscle to white adipose tissue (Cettour-Rose et al. 2005). With regard to old rats’ skeletal muscle, like in the cardiac muscle, the response to dietary manipulations was blunted.

We conducted our study only in male rats, since inclusion of females would require the assessment of the estrous cycle, supported by measurements of sex hormones. The results of this research cannot be extended to the female sex. Rat skeletal and cardiac muscles exhibit sexual dimorphism, which may manifest at different ages. For example, Vijay et al. (2015) reported that the cardiac expression of genes associated with oxidative phosphorylation was lower in male than in female Fisher 344 rats at 78 weeks of age, though not at 8 or 21 weeks. Moreover, analyses in isolated cardiac mitochondria showed higher mitochondrial efficiency of energy production in young female as compared to male Wistar rats (Colom et al. 2015). Another study, using DNA microarray analysis, revealed that expression of genes for myofibrillar and glycolytic proteins in skeletal muscle was higher in young male Sprague-Dawley rats compared to females, whereas females exposed to high fat diet showed greater induction of genes involved in oxidative metabolism and cellular defences (Oh and Yun 2012). The sexual dimorphism in skeletal and cardiac muscles contributes to different sex- and age-dependent outcomes of calorie restriction regimens (Bartke et al. 2019; Boldrin et al. 2017). It is therefore likely that the SCR effects revealed in our study in male rats would differently affect the skeletal and cardiac muscles in females.

In conclusion, changes in the expression of key genes involved in energy substrate utilization induced by SCR and refeeding suggest the presence of important age-related differences in the adaptive responses to nutritional challenge both in cardiac and skeletal muscles. SCR appears to increase the efficiency of glucose and fatty acid utilization in the cardiac muscle, however, this response is reduced in old rats. This may be a consequence of the aging process per se, or some other factors, such as increased adiposity of the old rats (Wronska and Kmiec 2012). The results of our study on the expression of factors controlling glucose and lipid metabolism suggest that SCR regimens might be much more effective in the young and middle-aged humans than in the elderly.