Lack of Effect of Ageing on Acetate Oxidation in Rat Skeletal Muscle During Starvation: A ¹³C NMR Study

Mitochondrial alterations have been observed with ageing in various tissues, including skeletal muscle, in both animals and humans. It has been suggested that these changes contribute significantly to ageing and to age-associated degenerative diseases and may compromise the ability of the cell to adapt to various physiological stresses (for a review, see Shigenaga et al., 1994).

Despite the remaining activated cellular defences such as catalase and superoxide dismutase, an increase in mitochondrial production of free radicals (superoxide, H₂O₂, hydroxyl radical) with age has been observed (Beckman and Ames, 1998). Oxidants damage cellular macromolecules including mitochondrial DNA (Randerath et al., 1996), proteins (Levine and Stadtman, 1996) and lipids (Tappel, 1975). Indeed, the increase in mitochondrial membrane lipid peroxidation results in an impairment of membrane fluidity (Tappel, 1975). Moreover, abnormalities in electron transport systems (Lee et al., 1998) and the decreased content and activity of mitochondrial enzymes (Hansford and Castro, 1982) contribute to the decline in mitochondrial oxidative capacity (Shigenaga et al., 1994) with age.

¹³C nuclear magnetic resonance (NMR) spectroscopy is a powerful tool for probing intermediary metabolism in perfused organs (Bailey et al., 1981; Cohen, 1987; Desmoulin et al., 1985; Neurohr et al., 1983; Chauvin et al., 1994, 1997; Künnecke, 1995) and in vivo (Mason et al., 1992; Cerdan et al., 1990). Because of the low natural abundance of ¹³C (1.1 %), most intermediary metabolites must become sufficiently enriched in ¹³C for detection by NMR. This has been accomplished by using ¹³C-labelled substrates such as glucose, lactate or acetate. High-resolution ¹³C spectra displaying carbon–carbon couplings of the easily detected glutamate pool are now widely used for the analysis of substrate selection and of contributions from competing carbon sources that supply the tricarboxylic acid (TCA) cycle within biological samples. Furthermore, studies have shown that the relative flux through various metabolic pathways associated with the citric acid cycle can be derived from a single ¹³C spectrum collected at metabolic and isotopic steady states (Malloy et al., 1988, 1990a,b; Szczepaniak et al., 1996). Recently, ¹³C NMR isotopomers were used to study skeletal muscle metabolism in vivo (Szczepaniak et al., 1996; Roussel et al., 1996; Bertocci et al., 1997). However, to our knowledge, there is no report on the effects of ageing on skeletal muscle metabolism.

The β-oxidation of free fatty acids may be impaired with ageing (Shigenaga et al., 1994), and the activity of citrate synthase, a major enzyme in flux-generating reactions in the TCA cycle, may decline with advancing age (Rooyackers et al., 1996). Mitochondrial alterations would suggest changes in flux in pathways through the TCA cycle. On the basis of the ability of ¹³C NMR to probe the TCA cycle, we analyzed the pattern of acetate oxidation in adult and old rat skeletal muscle using ¹³C NMR. Acetate was chosen for this study because of its simple metabolism (Szczepaniak et al., 1996) and its role in stimulation of fluxes through citrate synthase (Dugelay et al., 1999). Fasted animals were used because an excess of fatty acids suppresses glucose oxidation (Randle et al., 1963; Fuller and Randle, 1984) and thus promotes the utilization of acetate. We have examined the ¹³C fractional enrichment of acetyl-coenzyme A (FC₂) entering the TCA cycle and the ratio of anaplerotic flux to TCA flux by analyzing glutamate isotopomers using a method that assumes steady-state conditions. To confirm the latter, FC₂ was also determined using a method appropriate for non-steady-state conditions.

Adult (6–8 months old) and old (22 months old) female Wistar rats (Iffa-Credo, L’Arbresle, France) were housed under controlled environmental conditions (temperature 22 °C; 12 h:12 h L:D photoperiod, dark period starting at 20:00 h) and were given free access to commercial laboratory chow and water.

The animals were fasted for 5 days before the experiments but had free access to water. The initial body masses of adult and old rats were 369±11 and 400±17 g, respectively (means ± S.E.M., N=4–6; not significantly different). The respective body mass loss due to food deprivation was 45±3 and 54±3 g, respectively (means ± S.E.M., N=4–6; not significantly different). Pilot studies have shown that the percentage of body fat in old rats (28±10 % body mass) was approximately twice that of adults (16±9 % body weight; means ± S.E.M., N=6–9). Fasting for 5 days did not significantly modify the proportion of body fat (23±10 % in old rats versus 13±9 % in adults; means ± S.E.M., N=6–9).

Rats were anaesthetized by intraperitoneal injection of urethane (1.5 g kg⁻¹ body mass). The right jugular vein was dissected and cannulated with a polyethylene catheter for administration of [2-¹³C]sodium acetate (99.9 % enriched, Masstrace Inc., Woburn, MA, USA). The animals received a bolus (110 μmol 100 g⁻¹ min⁻¹ over a period of 8 min) and a constant infusion at 10 μmol 100 g⁻¹ min⁻¹ over the subsequent 2 h. Some animals (four for each age group) were infused with NaCl (0.9 %) and used as controls for enzymatic determinations. After 2 h of infusion, skeletal muscle from the hindlimb (approximately 1 g of mixed biceps femoris and tibialis anterior muscles) was quickly freeze-clamped in situ, removed, powdered using a mortar and pestle, and extracted with 4 % perchloric acid (w/v). A portion of this extract was neutralized with 3.5 mol l⁻¹ K₂CO₃ (to pH 7) and stored at −20 °C until analysis. Muscle glutamate concentration was assayed as described previously (Meynial-Denis et al., 1996). The remainder of the acid extract was used for NMR analysis. Acid extracts were neutralized to pH 7 with 5 mol l⁻¹ KOH, filtered, freeze-dried and redissolved in 500 μl of 0.1 mol l⁻¹ D₂O phosphate buffer (pH 7.4) containing 3 mmol l⁻¹ EDTA (to remove paramagnetic ions present in the sample) and 20 mmol l⁻¹ dioxane as both an internal standard and a chemical shift reference. Samples (400 μl) were pipetted into 5 mm NMR tubes. High-resolution NMR spectra of these extracts were recorded as reported below.

All high-resolution spectra of muscle extracts were performed on a Bruker AMX 400 (9.4 T) wide-bore spectrometer operating at 100.6 MHz for carbon and 400.13 MHz for proton. ¹³C NMR spectra were acquired by summing 46 800 free induction decays. Signals were collected using a 60 ° pulse angle, a 21 kHz sweep width, a 32K memory size, a 1 s repetition time and broadband Waltz-16 proton decoupling turned on during the acquisition. T₁ relaxation times of both dioxane and acetate were measured using the inversion–recovery method (Breitmaier and Voelter, 1987). ¹H acquisition parameters were as follows: spectral width, 6000 Hz; flip angle, 40 °data size, 32K; repetition time, 9.5 s; number of scans, 128.

¹³C NMR data were analyzed by using NMR1 data-analysis software on a Sun computer. The free induction decay values were processed with 1 Hz line-broadening Fourier transformation. Dioxane resonance was used as a reference and assigned at 67.4 p.p.m. relative to tetramethylsilane (TMS). Areas of resonance were determined by fitting with a sum of Lorentzian and Gaussian curves. The areas of dioxane and acetate were corrected by T₁ saturation factors (×5.83 and ×3.26, respectively).

Fractional ¹³C enrichments of acetate in muscle perchloric extracts were determined from the ¹H NMR spectra. The ratios of the area of the ¹³C satellites to the total area of the corresponding proton resonances were calculated. Acetate concentration was then determined as described by Badar-Goffer et al. (1990).

Results are presented as means ± S.E.M. (N=4–6) and were compared using the standard Student’s t-test. Differences were considered significant at P<0.05.

After 2 h of [2-¹³C]acetate infusion, the acetate concentrations in skeletal muscle were not significantly different in adult and old rats (41±10 and 54±6 μmol g⁻¹, respectively, determined from ¹H and ¹³C NMR spectra). Total muscle glutamate concentrations (measured enzymatically) were not affected by acetate infusion in either adult (0.54±0.22 μmol g⁻¹ in acetate-infused rats; 0.50±0.13 μmol g⁻¹ in control rats) or old (0.59±0.13 and 0.56±0.13 μmol g⁻¹, respectively) rats. However, [2-¹³C]acetate appeared to act as a carbon donor for glutamate synthesis in adult and old rats since enrichments in C2, C3 and C4 were detectable in muscle extracts by ¹³C NMR (see below).

Expanded regions of high-field proton-decoupled ¹³C NMR spectra of adult and old skeletal muscle after [2-¹³C]acetate infusion are shown in Fig. 1A,B. The regions containing glutamate C2, C3 and C4 are displayed. The glutamate multiplets were clearly resolved and were quite similar in both cases.

The pool of glutamate molecules consisted of a mixture of ¹³C isotopomers resulting from [2-¹³C]acetate metabolism in the TCA cycle. The pseudo-triplet observed for the glutamate C4 resonance arises from the superposition of a doublet and a singlet. The doublet comes from [3,4-¹³C]glutamate and [2,3,4-¹³C]glutamate, which both contain two neighbouring ¹³C carbons, while the singlet is derived from [4-¹³C]glutamate and [2,4-¹³C]glutamate, which each have only one ¹³C bound to a ¹²C neighbour.

For each individual C2, C3 and C4 glutamate multiplet, the peak areas as a fraction of the total peak area are given in Table 1. A comparison of a steady-state versus non-steady-state isotopomer analysis of the data is given in Table 2.

The steady-state isotopomer analysis showed that 76±4 % and 82±3 % (P=0.30, not significantly different) of the acetyl-CoA entering the TCA cycle was derived from [2-¹³C]acetate in adult and old rats, respectively. The relative contribution of anaplerotic (y) flux to the TCA cycle versus flux through citrate synthase was 21 % in both groups. Similarly, the non-steady-state analysis indicated that [2-¹³C]acetate infusion accounted for 84±3 % and 78±6 % of acetylcoA in adult and old rats, respectively. These values did not differ from those observed in the steady-state analysis, suggesting that steady-state conditions were achieved in our experiments (Szczepaniak et al., 1996).

It is known that acetate is readily taken up by skeletal muscle and converted to acetyl-CoA by acetyl-CoA synthase and that a significant proportion of the acetate infused into humans is oxidized by skeletal muscle (Vinay et al., 1987). Our results indicated that, after [2-¹³C]acetate infusion, the glutamate pool of skeletal muscle became sufficiently enriched for ¹³C NMR isotopomer analysis of tissue extracts in both adult and old rats. The analysis shows that 76±4 % and 82±3 % (difference not significant) of the acetyl-CoA oxidized by the skeletal muscle was derived directly from the infusion of [2-¹³C]acetate in adult and old rats respectively. This is in good agreement with the study of Szczepaniak et al. (1996), which demonstrated that, in rabbits infused with [2-¹³C]acetate, 87 % of the total oxidized substrate arises from acetate in the biceps femoris muscle.

The fraction of unlabelled acetyl-CoA entering the TCA (FC₀) represents 20 % of the total. Glycolysis and β-oxidation of endogenous fatty acids are the two major pathways that could be the source of unlabelled substrate. In the isolated rat heart perfused solely with [2-¹³C]acetate, FC₂ was approximately 88±2 %. However, when unlabelled pyruvate was added to the perfusate, FC₂ decreased by 20 %. This diminution was accounted for by competition between the two substrates and a reduction in the enrichment of acetyl-CoA through pyruvate incorporation (via pyruvate dehydrogenase) (Malloy et al., 1988). In our experimental conditions, pyruvate dehydrogenase activity was probably low for two reasons. First, flux through pyruvate dehydrogenase and the concentration of active pyruvate dehydrogenase complex diminish in skeletal muscle during prolonged fasting (Fuller and Randle, 1984; Hagg et al., 1976; Holness et al., 1989). Second, various studies in rat and human skeletal muscles have demonstrated that acetate infusion decreases pyruvate dehydrogenase activity, presumably by elevating muscle citrate and acetyl-CoA levels and the ratio of acetyl-CoA to free CoA (Putman et al., 1995; Jucker et al., 1997). Thus, the unlabelled acetyl-CoA in the present study probably arose mainly from lipolysis.

Conflicting results were obtained in studies comparing citrate synthase activity in quiescent skeletal muscle during ageing in animals and humans (Rooyackers et al., 1996; McCully et al., 1993; Lawler et al., 1993; Coggan et al., 1993). Indeed, some studies showed a 30 % age-related decrease in citrate synthase activity, indicating a decline in mitochondrial oxidation (Rooyackers et al., 1996; McCully et al., 1993), while others suggested that citrate synthase activity was not altered during ageing (Lawler et al., 1993; Coggan et al., 1993). In control conditions in the present study, adult and old rats exhibited skeletal muscle glutamate pools of similar size, which were not affected by acetate infusion. In our steady-state conditions, analysis of [¹³C]glutamate isotopomers showed that the fraction of labelled acetyl-CoA taking part in the citrate synthase reaction (FC₂) was approximately 80 % in both groups. Moreover, the acetate concentration was not significantly different in muscle extracts from adult and old rats. These results suggest that the ability of skeletal muscle to oxidize the same proportion of exogenous carbon substrates in the TCA cycle was not altered with ageing.

Anaplerosis refers to the entry of carbon skeletons into the TCA cycle other than through citrate synthase and the generation of citric acid cycle intermediates (Malloy et al., 1988). The anaplerotic flux observed here in skeletal muscle was 21 % of the citrate synthase flux in both cases. In our experimental conditions, the carbon skeletons generated through the anaplerotic pathways originate from both pyruvate carboxylation and proteolysis. ¹³C NMR (Szczepaniak et al., 1996; Yang et al., 1992) or ¹⁴C radioisotope (Lee and Davis, 1979; Davis et al., 1980) studies have shown that pyruvate carboxylation could be an important anaplerotic pathway in skeletal muscle mitochondria. The flux through pyruvate carboxylase was 20–35 % of the flux through citrate synthase. Since muscle proteolysis is poorly regulated in old compared with adult rats (Dardevet et al., 1995; Mosoni et al., 1999), we suggest that the respective contributions of pyruvate carboxylase and proteolysis to the anaplerotic pathway could change with ageing.

Finally, our results clearly indicated that the relative flux through the TCA cycle in skeletal muscle was not altered with ageing. In particular, we have shown that relative acetate oxidation was similar in adult and old rats. It has been demonstrated that the mobilization of free fatty acids during fasting does not change during ageing (Gerber et al., 1999; Klein et al., 1986). We therefore hypothesize that oxidation of free fatty acids will not be altered with age. Similarly, Mosoni et al. (1999) demonstrated that, after 5 days of fasting, proteolysis was activated to the same extent in adult and old rats. However, muscle proteolysis was less well regulated in old rats than in adults during refeeding after fasting or glucocorticoid treatment (Dardevet et al., 1995; Mosoni et al., 1999). The metabolic responses to various stresses that occur during ageing might therefore be slower, but of similar magnitude, in old animals.

In summary, this study illustrates the use of [2-¹³C]acetate perfusion to investigate substrate oxidation by skeletal muscle via the Krebs cycle in adult and old starved rats. One particular advantage of this approach is its ability to yield information on the contribution of acetyl-CoA to skeletal muscle glutamate synthesis with ageing.

The authors would like to thank Didier Attaix for a critical review of the manuscript and Hélène Lafarge for her contribution to the bibliography. This study was supported by a grant (no. 95G0078) from the French Ministère de la Recherche et de la Technologie.