Oxidation of Intracellular and Extracellular Fatty Acids in Skeletal Muscle: Application of kinetic modeling, stable isotopes and liquid chromatography/electrospray ionization ion-trap tandem mass spectrometry technology

Abstract

Fatty acids are a major fuel for many tissues and abnormal utilization is implicated in diseases. However, tissue fatty acid oxidation has not been determined reliably in vivo. Furthermore, fatty acid oxidation has not been partitioned into intracellular and extracellular components. In this report, a one-pool model is described that enables direct quantitation of fluxes of intracellular and plasma fatty acids to mitochondria in skeletal muscle using dual stable isotopes and liquid chromatography/electrospray ionization ion-trap tandem mass spectrometry (LC/ESI-itMS²) technology. It is validated by the determination of palmitate oxidation by skeletal muscle in lean and obese rats and the regulation by insulin. Resting postabsorptive intramyocellular and plasma palmitate oxidation by gastrocnemius muscle was determined to be 3.47±0.8 and 2.06±0.5 nmol/g min in lean and 6.96±1.8 and 1.34±0.2 nmol/g min in obese rats, respectively. In obese rats, hyperinsulinemia (1 nmol/l) suppressed intramyocellular (by 59±5% to 2.88±0.3 nmol/g min P<0.05) but not plasma (1.41±0.14 nmol/g min, P>0.05) palmitate oxidation. The fractional turnover rate of palmitoylcarnitine (0.34±0.1/min vs. 0.83±0.2/min, P<0.05) was also suppressed by insulin. In obese and lean rats, there are 83% and 51%, respectively (P=0.08), of plasma fatty acids traverse triglyceride pool before being oxidized. The results demonstrated that the methodology is feasible and sensitive to metabolic alterations and thus can be used to study fatty acid utilization at tissue level in a compartmentalized manner for the firs time.

Keywords: LC-MS; beta-oxidation; compartmentalization; extracellular; intracellular; skeletal muscle; stable isotope.

Figure 1

Schematic representation of the one-pool model for quantifying mitochondrial β-oxidation. Intramyocellular fatty acid pool NEFA (non-esterified fatty acids) and imcTG undergo recycling constantly [44]. There might be also fatty acid exchanges between NEFA and PL (phospholipids). Palmitate released from imcTG may directly enter palmitoylcarnitine pool or via NEFA. These interactions are complex but irrelevant to this model. The model treats these intracellular fatty acid pools as a single, composite entity (black box) with the interactions within it irrelevant and thus ignored. Only is the net efflux of fatty acids from this box considered and quantified (m). The efflux of m comes out of this black box and then goes through the much smaller palmitoylcarnitine pool. Thus, the latter pool must turn over faster as described in the text. p represents plasma FFA flux, direct and indirect, to palmitoylcarnitine pool. m and p, and β, are unidirectional as acylcarnitines do not recycle (m+p=β). The glycerol moiety of imcTG and PL is not directly involved in β-oxidation and thus not shown.

Figure 2

Isotopic kinetic model of [¹³C₄]palmitoylcarnitine and [U-¹³C]palmitoylcarnitine during chase (t₀—t) for the separation of m and p. During pulse (120 min until t₀), imcTG was labeled by [¹³C₄]palmitate. At t₀, [¹³C₄]palmitate infusion stopped and thus [¹³C₄] label in imcTG becomes the sole source and starts to decay exponentially according to e^−kt kinetics (broken curvilinear descending slope). From this slope, fractional turnover constant of palmitoylcarnitine, k (thick solid descending slope) is derived using Eq 1 and adjusted as described in the text. Possible loss of the label to other unknown pathways (? sign) does not affect the calculation. Meanwhile, during chase [U-¹³C]palmitate equilibrates quickly in plasma FFA and [U-¹³C]palmitoylcarnitine pools (plateaus in these pools). Solid and broken arrows from plasma represent direct and indirect path, respectively, of plasma palmitate flux to palmitoylcarnitine. The turnover of [U-¹³C]palmitoylcarnitine label is the same as the entire palmitoylcarnitine pool (k). Thus, the flux of [U-¹³C]palmitoylcarnitine label is calculated from which the flux of total plasma palmitate to palmitoylcarnitine pool is determined (Eq 3). Note that during chase, [¹³C₄]palmitoylcarnitine and [U-¹³C]palmitoylcarnitine are independent of each other with distinct kinetics, namely decaying vs. steady, respectively. This is the basis that enables the segregation of imcTG-palmitate and plasma palmitate fluxes.

Figure 3

Basal postabsorptive total (β), imcTG (m) and plasma (p) palmitate oxidation in gastrocnemius of lean and higher fat-fed obese rats and the effect of insulin (clamp). From the same studies as in Table 1. β was also estimated using an independent technique with [9,10-³H]palmitate as a tracer in a separate group of obese rats (basal, n=4). For the detailed experimental procedures and data analysis, see METHODS. *, P<0.05, compared to the same compartments in the obese basal group.

Figure 4

Turnover of [¹³C₄]palmitoylcarnitine. Decay of [¹³C₄]palmitoylcarnitine label in gastrocnemius of rats determined at 3 sequential time points during chase following prelabeling (pulse). The raw data of [¹³C₄] enrichment (Y-axis) were transformed by natural logarithm plotted against regular scale in X-axis. The linear decay shows that palmitoylcarnitine has single-pool turnover kinetics according to e^−kt kinetics. No new labels come in as they have been cleared from the circulation (inset, where t₀=20 min, t=80 min). It is emphasized that decay curves with or without natural logarithmic transformation of the raw data appear similar at low k′s. However, without transformation, the error is significant, or 14–97% for k at 0.5–10% based on simulation.

Figure 5

Unidirectional movement of long chain fatty acids (LCFA) into mitochondria. Plasma free fatty acids and imcTG-fatty acids are sequentially activated to acyl CoA by ACS (acyl CoA synthase) and acylcarnitine by extra-mitochondrial CPT1 (carnitine-palmitoyl transferase-1). Upon formation, acylcarnitine is transferred into mitochondria by AT (acylcarnitine translocase) only where can it be reverted back to acyl CoA by the intramitochondrial CPT2 to enter β-oxidation cascade initiated by ACD (acyl CoA dehydrogenase). Note that the extra- and intra-mitochondrial acylcarnitine pools are isotopically identical and that the intra-mitochondrial pool does not accumulate. The thick arrow, or β-oxidation flux, stresses this unidirectionality but is identical stoichiometrically to the upstream and downstream steps.