Figure 1 Pathways and sources of substrates entering the TCA cycle

¹³C magnetic resonance spectroscopy (MRS) techniques allow calculation of relative flux rates through multiple pathways as well as permitting detection of successive biochemical events (Sherry & Malloy, 1996). Analysis of the ¹³C isotope isomers (isotopomers) of glutamate has been central in the development of complex analysis models for calculating TCA cycle flux and the rate of flux through anaplerotic pathways (Sherry & Malloy, 1998). The use of ¹³C analysis methods to determine anaplerotic flux through combined anaplerotic reactions relative to TCA cycle flux (Y) is well established (Jeffrey et al. 1996; Jones et al. 1997; Malloy et al. 1987, 1988; Szczepaniak et al. 1996). The Y value reflects flux through metabolic pathways other than citrate synthase entering the TCA cycle, as well as flux through cataplerotic reactions exiting the TCA cycle, which are equal at metabolic steady state. Calculation of Y requires that the tissue being measured has achieved ¹³C isotopic steady state (ISS), which refers to the metabolic condition when the quantity of ¹³C label entering the TCA cycle is equivalent to ¹³C label exiting the TCA cycle. In tissue undergoing metabolic transition (e.g. due to a change in substrate availability or the onset of contraction), ISS will occur after TCA cycle fluxes and TCAI concentrations have reached their new metabolic steady state (Malloy et al. 1988). In ¹³C studies, ISS is indicated by no change in ¹³C spectra at sequential time points, and no difference in non-steady-state versus steady-state isotopomer analysis at a given time point (Malloy et al. 1987; Szczepaniak et al. 1996).

Previous studies in resting skeletal muscle of mixed fibre type have reported relative anaplerotic flux values (Y) ranging from 0.18 to 0.26 (Szczepaniak et al. 1996; Jucker et al. 1997; Mezzarobba et al. 2000). However, relative anaplerotic flux has not been determined in muscle fibre types of differing oxidative capacity. Additionally, the effect of muscle contraction upon levels of relative anaplerotic flux in muscle of differing fibre composition is unknown. TCA cycle intermediate concentrations rise at the onset of exercise (Aragón & Lowenstein, 1980; Gibala et al. 1997, 1998), presumably due to increased anaplerotic influx. It is not known if anaplerotic flux remains high throughout the exercise period or if this flux returns to basal (rested) values once the muscle has achieved its new contracting metabolic steady state. We hypothesized that absolute anaplerotic flux would be higher during sustained steady-state contraction than during rest, which would be reflected in a similar or higher relative anaplerotic flux value (Y) in contracted muscle. To address these issues, we evaluated relative anaplerotic flux in muscle of differing fibre types at rest and during muscle contraction in the intact rat.

All experimental procedures were approved by the Institutional Animal Care and Use Committee at the Veterans Affairs North Texas Health Care System (VANTHCS) and conformed with national guidelines.

Initially, 220 mm unlabelled sodium octanoate (Aldrich, Milwaukee, WI, USA) was infused at a rate of 2.67 ml h⁻¹. At time zero (after approximately 30 min of unlabelled octanoate infusion), the infusate was changed to 220 mm sodium [2,4,6,8-¹³C₄]octanoate (Cambridge Isotope Laboratories (CIL), Andover, MA, USA) and the muscles of the left limb were contracted via direct electrical stimulation to the sciatic nerve. The muscles of the non-contracted right limb were considered resting and results in these muscles were compared to the contralateral contracted muscles.

Rats were weighed and anaesthetized with an intraperitoneal injection of 1–3 ml kg⁻¹ ketamine-xylazine mixture (6 mg ml⁻¹ xylazine, 94 mg ml⁻¹ ketamine). After a tracheotomy was performed, 1–3 % isoflurane (Isotec 3 vaporizer, Matrx Medical Inc., Buffalo, NY, USA) in a 100 % O₂ mixture was used to maintain anaesthesia and was delivered via a positive pressure ventilator (model 683 Small Animal Ventilator, Harvard Apparatus Inc., South Natick, MA, USA) at a rate of 1.25 ml min⁻¹, ~100 breaths min⁻¹. The right carotid artery was catheterized and connected to a fluid-filled pressure transducer (model P23XL Gould Transducer, Future Tech, Birmingham, AL, USA) to continuously monitor heart rate and blood pressure throughout the surgical procedure. Systemic arterial blood gases were monitored using a blood gas analyser (ABL-4, Radiometer, Westlake, OH, USA). The gas mixture and/or ventilation rate was adjusted to maintain physiological blood pH (end pH 7.3 ± 0.09) and arterial O₂ saturation to > 99%. Following arterial cannulation, the jugular vein was cannulated and connected to an infusion pump (model 22, Harvard Apparatus Inc., South Natick, MA, USA) for the administration of labelled and unlabelled octanoate. Temperature was maintained at 37 °C with an external heating pad placed under the rat, in combination with the heat radiating from the incandescent light source illuminating the surgical field. Both calcaneal tendons and the left sciatic nerve pocket were surgically isolated. The left sciatic nerve was bathed in mineral oil and isolated from surrounding tissue. The left calcaneal tendon was attached to a force transducer (Grass model FT-10, New Astromed Inc., West Warwick, RI, USA) and the output was recorded using PolyVIEW software (Grass, New Astromed Inc., West Warwick, RI, USA). Nerve viability was maintained by regular topical application of mineral oil during the remainder of the experiment. Custom-built silver electrodes connected to an electrical stimulator and isolation unit (Grass models S88 and S7, New Astromed Inc., West Warwick, RI, USA) were used to deliver 100 ms trains of pulses (100 Hz, 0.2 ms duration, no delay) at the voltage required to elicit maximal force output (3.2–5.0 V) at 0.5 Hz for 90 or 105 min. Force drop-off to 85 ± 17 % of the original value occurred by the end of the experiment. At the end of 90 or 105 min of ¹³C-octanoate infusion, gastrocnemius and soleus muscles were isolated from the contracted hindlimb, and then from the contralateral resting hindlimb. Isolated tissue was immediately frozen in liquid nitrogen and stored at −80 ° C until extraction. The rat was killed by removal of the heart under deep anaesthesia.

Figure 2 13C spectra from rested and contracted gastrocnemius and soleus muscles

Fractional enrichment of acetyl-CoA pools was calculated using non-steady-state and steady-state isotopomer analysis for 90 and 105 min ¹³C-labelled octanoate infusions (Table 1). Metabolism of ¹³C-labelled octanoate accounted for about half of acetyl-CoA production (F_C2) in soleus (mainly type I muscle fibres) and predominantly red gastrocnemius muscle (mostly type IIa muscle fibres) during both rest and contraction; this was higher than the contribution of 25–30 % in rested and contracted white gastrocnemius muscle (primarily type IIb fibres) (Table 1). There were no differences in F_C2 or the fractional contribution of endogenous unlabelled substrates (F_C0) to the acetyl-CoA pool using the non-steady-state versus steady-state analysis at either the 90 or 105 min infusion times (Table 1). Additionally, no differences were found using non-steady-state or steady-state isotopomer analysis in the 90 versus 105 min infusion times in any of the muscles examined (Table 1). These combined data establish that ISS occurs by 90 min in rested and contracted gastrocnemius and soleus muscles.

Table 1 Non-steady-state and steady-state analyses after infusion of [2,4,6,8-13C4]octanoate for 90 or 105 min

Table 2 Relative anaplerotic flux (Y) in gastrocnemius and soleus muscles after infusion of [2,4,6,8-13C4]octanoate

The main findings of this study were: (1) flux through anaplerotic pathways was 18–41 % of TCA cycle flux at rest, with higher relative anaplerotic flux in oxidative versus glycolytic fibre types (Table 2), (2) fibre type differences in rates of relative anaplerotic flux persisted during muscle contraction with higher levels in soleus compared to predominantly red and white gastrocnemius (Table 2), and (3) during contraction, Y is maintained in soleus and gastrocnemius muscle (Table 2), indicating that absolute rates of anaplerotic flux rise in proportion to increased oxidation rates during contraction.

Relative anaplerotic flux in the different muscles examined ranged from 0.18–0.41 at rest after infusion with ¹³C-octanoate (Table 2). Results from previous studies in resting skeletal muscle using ¹³C-acetate showed relative anaplerotic flux values of 0.18 in combined quadriceps and gastrocnemius muscle (Jucker et al. 1997), 0.21 in mixed biceps femoris and tibialis anterior (Mezzarobba et al. 2000), and 0.26 in biceps femoris (Szczepaniak et al. 1996), which are in the lower range of our results (Table 2). During rest, Y is higher in oxidative versus glycolytic fibre types (Table 2), thus the lower range reported previously in resting muscle of mixed fibre type is probably due to the preponderance of type IIb fibres in the muscles examined.

The metabolic basis of increased flux through anaplerotic pathways relative to TCA cycle flux in oxidative versus glycolytic muscle fibres is uncertain. Enzymes believed to be critical for normal anaplerosis are mitochondrial (e.g. pyruvate carboxylase) or have mitochondrial as well as cytoplasmic forms (e.g. alanine aminotransferase), and higher levels of mitochondria and mitochondrial enzymes are characteristic of more oxidative muscle fibres. However, there are no data to our knowledge that indicate that there is greater expression of these enzymes per unit of mitochondria in oxidative muscle. It may be that many of the anaplerotic enzymes are regulated by substrate availability in vivo. Maximal enzyme activity assays performed on muscle homogenates in vitro primarily reflect the quantity of the active enzyme present and are not sensitive to substrate regulation. It remains to be determined which intrinsic characteristics of the different skeletal muscles examined contribute to the observed fibre type-specific anaplerotic flux rates during rest.

Many studies have examined the increase in TCAI at the onset of exercise (Aragón & Lowenstein, 1980; Sahlin et al. 1990; Gibala et al. 1997, 1998) but none have quantified the anaplerotic flux that produces these dramatic changes in muscle. It has been speculated that the increase in TCAI is a function of pyruvate availability and not of much functional significance to TCA cycle flux (Gibala et al. 1998; Graham & Gibala, 1998; Constantin-Teodosiu et al. 1999). The initial increased utilization of anaplerotic pathways at the onset of exercise may be related to excess pyruvate availability. However, if pyruvate availability were the prevailing determinant of flux through anaplerotic pathways, it would be expected that, once a new metabolic steady state was reached, flux through anaplerotic pathways would revert to original values. Although our study does not examine events in the initial minutes of muscle contraction, maintained relative flux through anaplerotic pathways was observed in soleus and gastrocnemius muscle during a contracting metabolic steady state (Table 2). These results indicate there is a sustained increase in the absolute rate of anaplerotic flux that is proportional to the increase in oxidation rate during contraction.

Absolute anaplerotic flux rates can be calculated using estimates of oxygen consumption based on previous results using a similar protocol (Hood et al. 1986). Oxygen consumption attributed to WG muscle was calculated to be 4-fold higher than that of the rested muscle, while oxygen consumption ascribed to RG muscle was calculated to increase approximately 20-fold during contraction (Hood et al. 1986). Blood flow and oxygen delivery were slightly higher in soleus versus RG muscle (Hood et al. 1986), therefore it is likely that soleus would have the same or higher oxygen consumption as RG muscle. Oxygen consumption is related to TCA cycle flux by a proportionality factor (R). R varies depending on the substrate being oxidized and considers the stoichiometry of O₂ consumed to the reducing equivalents generated in the production of a single acetyl-CoA unit (from a given substrate) and its subsequent oxidation by the TCA cycle. R for these experiments can be described by the equation: R_total=F_C0R₀+F_C2R₂ (Malloy et al. 1996), where the F_C0R₀ term reflects endogenous unlabelled substrate contribution and the F_C2R₂ term arises from ¹³C-labelled octanoate. For octanoate R₂ is 2.75 (Malloy et al. 1996). If all of the unlabelled endogenous substrate oxidized (F_C0) is considered to have come from carbohydrate (i.e. glucose, glycogen, and/or lactate), R₀ is 3. R_total was 2.88 and 2.89 for the predominantly red portion of the gastrocnemius muscle during rest and contraction, respectively. R_total was the same during rest and contraction for the white portion of the gastrocnemius and soleus (2.93 and 2.87, respectively). R values will be underestimated to the extent that substrate other than carbohydrate contributed to the unlabelled fraction (F_C0) of oxidative energy production. With these R values and fold oxygen consumption reported by Hood et al. (1986), Y values (Table 2) can be used to calculate increases in absolute anaplerotic flux. Flux through anaplerotic pathways increases similarly to that of TCA cycle flux and oxygen consumption, with a 3.3-fold increase in the white portion of the gastrocnemius, a 13.9-fold increase in the predominantly red portion of the gastrocnemius, and a 16.6-fold increase in soleus muscle. Recently, ¹³C MRS techniques have been developed to measure oxygen consumption in heart (Jeffrey et al. 1999). Although application of this methodology in rested and contracted skeletal muscle is technically more challenging, it could be employed in the future to allow changes in oxygen consumption to be unambiguously defined for each muscle. This would allow a more quantitative assessment of absolute increases in TCA cycle flux and flux through anaplerotic pathways during rest and contraction.

The possible role of anaplerosis in maintaining optimal levels of TCA cycle flux to support oxidative phosphorylation is controversial (Gollnick, 1985; Constantin-Teodosiu et al. 1999; Gibala et al. 2000). The observation that absolute levels of TCA cycle intermediates decline with fatiguing exercise, and the link between impaired TCA cycle expansion and impaired oxidative phosphorylation due to blocked glycogenolysis in muscle phosphorylase deficiency (Sahlin et al. 1990, 1995), support such a relationship. However, a recent study in healthy individuals demonstrated that, although the sum of TCAI increased almost 3-fold within the first few minutes of exercise, TCAI measurements after 60 and 90 min of sustained non-fatiguing exercise were not different from resting values (Gibala et al. 2002). Additionally, oxygen uptake in the exercising limb remained consistently high throughout the exercise period (Gibala et al. 2002). It may be that the ability to sustain flux through anaplerotic pathways is more important for normal oxidative metabolism in muscle than expansion of the TCAI pool per se. The purpose of anaplerotic flux into the TCA cycle may be to continually replenish key intermediates of low concentration such as oxaloacetate, while a function of TCA cycle efflux may be to aid in the removal of ammonia from exercising muscle. We consider that our finding of proportional increases in flux through anaplerotic pathways and oxidative carbon entry into the TCA cycle with muscle contraction supports the hypothesis that anaplerotic flux plays an important role in maintaining overall TCA cycle function and energy transduction during exercise.

In conclusion, fibre-type specific rates of relative anaplerotic flux were observed during both rest and contraction and correlated well with muscle oxidative capacity. Relative rates of anaplerotic flux were maintained during contraction, suggesting an association between TCA cycle oxidative and anaplerotic fluxes. Absolute rates of anaplerotic flux therefore rise in proportion to increased oxidation rates during contraction, which is consistent with a sustained increase in substrate entry into and exit from the TCA cycle through anaplerotic pathways during contraction.

This study was supported by an Associate Investigator award and Merit (No. 98–139) and VISN17 grants (No. 99–89) from the Department of Veterans Affairs.