The Fate of [U-¹³C]Palmitate Extracted by Skeletal Muscle in Subjects With Type 2 Diabetes and Control Subjects Free

We studied seven healthy lean male subjects and seven obese male subjects with type 2 diabetes (diabetes duration 2 years, range 0.5–8 years). Data on skeletal muscle fatty acid utilization in these subjects have been previously published (7). The diabetic subjects were treated with diet alone (n = 2) or diet in combination with oral blood glucose−lowering agents (low dosages of sulfonylureas, which were withheld for 2 days before the experiments; n = 5). Beside this, no other medications were used. None of the subjects had a serious health problem apart from diabetes. Normal resting electrocardiogram and blood pressure were prerequisites for participation. Subject characteristics are given in Table 1. All subjects engaged in sports ≤3 h per week, and none had a physically demanding job. The study protocol was approved by the Medical Ethical Review Committee of Maastricht University, and all subjects gave written informed consent.

Body weight was determined on an electronic scale, accurate to 0.1 kg. Waist and hip circumference measurements to the nearest 1 cm were made with the subject standing upright. Body composition was determined by hydrostatic weighing with simultaneous lung volume measurement (Volugraph 2000; Mijnhardt, Bunník, the Netherlands). Body composition was calculated according to Siri’s method (10).

The present study was performed to determine skeletal muscle ¹³C-palmitate kinetics during baseline and intravenous infusion of the nonselective β-agonist ISO. Subjects arrived at the laboratory at 8:00 a.m. by car or by bus after an overnight fast (at least 12 h). They were studied while resting supine on a comfortable bed in a room kept at 23–25°C.

Forearm skeletal muscle metabolism was studied under baseline conditions and during intravenous infusion of ISO by means of the forearm muscle balance technique with continuous intravenous infusion of the stable isotope tracer [U-¹³C]palmitate. Before the start of the experiment, three cannulas were inserted, as follows: one cannula was inserted under local anesthesia in the radial artery of the forearm for sampling of arterial blood; in the same arm, a second cannula was inserted in a forearm vein for the infusion of ISO and the stable isotope tracer; in the contralateral arm, a third catheter was inserted in the retrograde direction in an antecubital vein for the sampling of deep venous blood, draining from forearm muscle. Measurements were done during the last 30 min of a 90-min baseline period (0–90 min) and a subsequent 60-min period of intravenous infusion of ISO (90–150 min) given at 20 ng · kg⁻¹ lean body mass · min⁻¹.

After taking background blood and breath samples (see below), an intravenous priming dose of 0.085 mg/kg NaH¹³CO₃ was given. Then a constant-rate continuous infusion of [U-¹³C]palmitate was begun (0.011 μmol · kg⁻¹ body wt · min⁻¹) and continued during the entire period via a calibrated infusion pump (IVAC 560 pump; IVAC, San Diego, CA). The concentration of palmitate in the infusate was measured for each experiment (see biochemical methods, below) so that the exact infusion rate could be determined. The palmitate tracer (60 mg of the potassium salt of [U-¹³C]palmitate, 99% enriched; Cambridge Isotope Laboratories, Andover, MA) was dissolved in heated sterile water and passed through a 0.2-μm filter into 5% warm human serum albumin (Central Blood Bank, the Netherlands) to make a 0.670 mmol/l solution (mean ± SD, 0.668 ± 0.016 mmol/l).

Arterial and deep venous blood samples and a breath sample were obtained before the start of the experiment to determine background isotopic enrichment. Expired-air samples were obtained by having the subjects breath normally for 3 min into a mouthpiece connected to a 6.75-l mixing chamber and then collecting a sample into a 20-ml vacutainer tube. At time points 10, 20, 30, 40, 50, 60, 75, and 90 min during the baseline period and 110, 120, 130, 145, and 160 min during ISO infusion, breath samples were taken to determine the enrichment of CO₂ (¹³C/¹²C ratio) in expired air. During the entire experiment, CO₂ exchange was determined with an open-circuit ventilated hood system (Oxycon Beta, Jaeger, Breda, the Netherlands).

After the 60, 75, and 90 min of the baseline period, and 30, 45, and 60 min of ISO infusion, forearm blood flow, arterial and venous concentrations, and ¹³C/¹²C ratios of glucose, palmitate, glutamine, glutamate, and CO₂ were determined.

Total forearm blood flow was measured by venous occlusion plethysmography with a mercury strain gauge (Periflow 0699; Janssen Scientific Instruments, Belgium), as reported previously (11).

Blood samples were taken simultaneously from the radial artery and deep forearm vein after the blood flow measurement while the hand circulation was still occluded. Duplicate 1-ml blood samples were immediately injected with a needle through the rubber stopper of preweighed vacutainer tubes, without disturbing the vacuum, for the determination of ¹³CO₂/¹²CO₂. After being weighed again, 1 vol of 1 mol/l H₂SO₄ was injected through the rubber stopper into the tubes to direct all blood CO₂ into the gaseous head space. The tubes were weighed again to determine the dilution factor. The gaseous head space was finally brought to atmospheric pressure with helium. The same procedure was applied to bicarbonate standards of known concentration. The coefficient of variation (CV) for this method with CO₂ concentrations in the 15–30 mmol/l range is 0.4%. The CV between duplicate measurements of CO₂ concentrations in blood is 0.09%.

All other blood samples were collected in tubes containing EDTA kept on ice. These samples were immediately centrifuged at 4°C, and the plasma was put in liquid nitrogen until storage at −80°C.

Breath and blood samples were analyzed for their ¹³C/¹²C ratio and CO₂ content by injecting 20 μl of the gaseous head space into a gas chromatography (GC) continuous flow isotope ratio mass spectrometer (IRMS; Finnigan MAT 252, Bremen, Germany).

For the determination of plasma palmitate, FFAs were extracted from plasma, isolated by thin-layer chromatography, and derivatized to their methyl esters. Isotope enrichment of palmitate was determined by GC-IRMS after on-line combustion of the fatty acids to CO₂. The methyl ester of palmitate contains 17 carbon atoms; therefore, the tracer/tracee ratio of palmitate was corrected for the extra methyl group.

Palmitate concentrations were determined on an analytical GC with ion-flame detection using heptadecanoic acid as the internal standard. On average, it comprised 24.5 ± 0.6% of the total FFA concentration. Plasma glucose, glutamine, and glutamate concentrations (deproteinized with 3.5 wt/vol % sulphosalicyclic acid [SSA]) were measured using standard enzymatic techniques automated (Cobas Fara centrifugal analyzer) at 340 nm. Isotopic enrichment of plasma glutamine and glutamate was determined as the MTBSTFA derivative using GC combustion IRMS (MAT 252; Thermo Finnigan, Bremen, Germany). The resulting derivative contained 23 carbon atoms, 5 of which were from glutamine or glutamate; the ¹³C/¹²C ratio was therefore corrected by a factor of 23/5. To determine plasma glucose enrichment, glucose was extracted with chloroform-methanol-water, and derivatization occurred with butylboronic acid and acetic anhydride, as previously described (12). The resulting derivative contained 16 carbon atoms, 6 of which were from glucose; the ¹³C/¹²C ratio was therefore corrected by a factor of 16/6. The glucose derivatives were analyzed by GC combustion IRMS.

Fractional recovery of the palmitate label in breath CO₂ was calculated as follows:

where F is the infusion rate of palmitate in micromoles per minute, TTR_CO2 − TTR_bkg equals the increase in tracer/tracee (¹³C/¹²C) ratio in expired air during infusion (compared with background), and V_CO2 is the expired CO₂ in micromoles per minute. The number 16 in the denominator is to correct for the number of ¹³C atoms in palmitate.

The exchange of metabolites or tracer-labeled metabolites across forearm muscle was calculated by multiplying the arteriovenous concentration difference of metabolites (in micromoles per liter) or ¹³C-labeled atoms in metabolites (¹³C/¹²C ratio × number of C-atoms in molecule × metabolite concentration) by forearm plasma flow (ml · 100 ml⁻¹ forearm muscle · min⁻¹) or by total forearm blood flow (for CO₂ exchange). Forearm plasma flow was calculated by multiplying forearm blood flow with (1−hematocrit)/100. A positive exchange indicates uptake.

Data are expressed as means ± SE, unless otherwise indicated. To compare baseline and isoprenaline-induced responses between groups, a two-factor repeated measures ANOVA was performed. P < 0.05 was regarded as statistically significant.

The fractional recovery of [U-¹³C]palmitate tracer in expired ¹³CO₂ was lower in type 2 diabetic subjects versus control subjects during both baseline conditions and ISO infusion (P < 0.001) (Fig. 1).

Arterial palmitate concentrations were not significantly different between the two groups (Table 2), whereas plasma palmitate enrichment (¹³C/¹²C ratio) was higher in type 2 diabetic subjects (Fig. 2). Arterial glucose concentrations were higher in type 2 diabetic subjects and more decreased during ISO infusion in this group (P < 0.01). Arterial glucose enrichment (Fig. 2) and the concentration of ¹³C-glucose (defined as ¹³C/¹²C ratio glucose × glucose concentration × number of C-atoms in glucose) increased throughout the experiment (P < 0.001) and were not significantly different between groups. Glutamine concentration tended to be lower in the type 2 diabetic group (P = 0.10). Glutamine enrichment (Fig. 2) and the ¹³C-glutamine concentration (defined as ¹³C/¹²C ratio glutamine × glutamine concentration × number of C-atoms in glutamine) increased significantly throughout the experiment (P < 0.001); the ISO-induced increase in glutamine enrichment and ¹³C-glutamine was significantly more pronounced in type 2 diabetic subjects than in control subjects (P < 0.01). Glutamate and CO₂ concentrations were not significantly different between groups during baseline conditions or ISO infusion, although glutamate concentrations tended to be higher in type 2 diabetic subjects (P = 0.12). In both groups, glutamate concentrations significantly decreased during ISO infusion (P < 0.001) (Table 2). Glutamateenrichment increased throughout the experiment and was not significantly different between groups (Fig. 2). The concentration of ¹³C-glutamate (defined as ¹³C/¹²C ratio glutamate × glutamate concentration × number of C-atoms in glutamate) tended to be higher in type 2 diabetic subjects than in control subjects (P = 0.08).

Forearm plasma blood flow tended to be lower in type 2 diabetic than in control subjects (P = 0.095) (Table 3). Net muscle palmitate uptake was not significantly different between groups during baseline or ISO-stimulated conditions. The uptake of ¹³C-palmitate did not change as a result of ISO stimulation and tended to be lower in type 2 diabetic than in control subjects (P = 0.06). Glucose uptake tended to be lower in type 2 diabetic versus control subjects (P = 0.08), whereas there were no significant differences in the uptake of ¹³C-glucose across muscle between the two groups. Glutamine release was comparable in both groups during baseline conditions and ISO infusion. There was a significant release of ¹³C-glutamine in control subjects during ISO infusion, whereas no release could be detected in type 2 diabetic subjects (interaction group × ISO; P < 0.05). Muscle CO₂ release did not differ between groups during baseline conditions or ISO infusion, although values tended to be lower in type 2 diabetic subjects compared with control subjects (P = 0.07). There was significant muscle ¹³CO₂ production during ISO stimulation in control subjects, whereas we could not detect significant ¹³CO₂ release in type 2 diabetic subjects (interaction group × ISO; P < 0.05).

Figure 3 shows the ¹³C-labeled carbon balance across skeletal during ISO infusion. In both groups there was a significant uptake of ¹³C-palmitate (slightly lower in type 2 diabetic subjects; P = 0.06) and ¹³C-glutamate. In control subjects, there was significant ¹³CO₂ release (15% of ¹³C uptake from labeled palmitate) and a significant ¹³C-glutamine release (release of ¹³C-labeled atoms from glutamine was 6% of ¹³C uptake from labeled palmitate; P < 0.05). In contrast, in type 2 diabetic subjects, there was no release of ¹³C-labeled oxidation products.

The present study investigated the fate of [U-¹³C]palmitate extracted by forearm muscle in type 2 diabetic and control subjects. In skeletal muscle of control subjects, there was a significant release of oxidation products from ¹³C-palmitate in the form of ¹³CO₂ (15% of ¹³C uptake from labeled palmitate) and ¹³C- glutamine (release of ¹³C atoms from glutamine was 6% of ¹³C uptake from labeled palmitate). In contrast, in type 2 diabetic subjects, there was no detectable release of ¹³CO₂ and ¹³C-glutamine, despite a significant ¹³C label uptake from [U-¹³C]palmitate (60% of control value). These data indicate differences in FFA handling in skeletal muscle of control and type 2 diabetic subjects during infusion of the nonselective β-agonist ISO.

Pathways for label fixation are the loss of ¹³C label to glutamine and glutamate or glucose. Indeed, we found a significant increase in arterial plasma glutamine, glutamate, and glucose enrichment throughout the experiment, as previously reported (5). During baseline conditions, no significant release of ¹³C-glutamine from muscle could be detected in either group, despite a significant increase in glutamine enrichment and ¹³C-glutamine in arterial blood. These data indicate that there was a more rapid release of ¹³C-glutamine by other tissues than by skeletal muscle. It has been previously reported that the liver can both take up and release glutamine by the periportal parenchymal (at inflow site) and perivenous cells, respectively. In the perivenous liver cells (situated at liver outflow), a sodium-dependent glutamate transporter (responsible for glutamate uptake against a concentration gradient) and glutamine synthetase are exclusively expressed (13,14). For this reason, perivenous cells react similarly to muscle; they extract glutamate (from circulation or formed in the TCA cycle) and use it for glutamine synthesis. Because of the central role of the liver in glutamine/glutamate homeostasis and fatty acid metabolism, the liver seems to be the most likely site for the initial increase in arterial glutamine, glutamate, and glucose enrichments.

During baseline conditions, there was no significant ¹³CO₂ production by skeletal muscle in either group, indicating there was no oxidation of plasma palmitate. However, we cannot exclude the possibility that ¹³CO₂ production could not be measured with the present infusion rate, as the recovery of ¹³CO₂ from a palmitate tracer was low (7% recovery in expired CO₂ after 90-min infusion) (Fig. 1) and increased linearly in time for periods of 10–11 h. This point has been previously discussed (7).

As indicated above, in type 2 diabetic subjects there was no detectable release of ¹³CO₂ and ¹³C-glutamine despite a significant uptake of ¹³C label from [U-¹³C]palmitate (60% of control value), whereas in control subjects there was a significant release of ¹³CO₂ and ¹³C-glutamine. Several mechanisms may be responsible for this finding. First, these data may indicate that the palmitate taken up by muscle was not converted to oxidation products in type 2 diabetic muscle. Then, the most likely alternative fate of palmitate in these subjects was synthesis of triglycerides and incorporation in the muscle triglyceride stores. On the other hand, we previously found an increased glycerol release in these subjects during both baseline conditions and ISO infusion (7), indicating an increased lipolysis of the triacylglycerol stores and thus possibly an increased intramuscular FFA pool. Therefore, it is also possible that the [U-¹³C]palmitate tracer was diluted to a larger extent in the intramuscular FFA pool in type 2 diabetic subjects as compared with control subjects, which may have made it more difficult to detect a significant ¹³C label fixation into oxidation products. Nevertheless, both explanations imply that there are differences in how type 2 diabetic and healthy control subjects handle fatty acids.

Interestingly, the ISO-induced increase in arterial glutamine enrichment and arterial ¹³C-glutamine was more pronounced in the diabetic group. In view of the reduced ¹³C-glutamine production by skeletal muscle in type 2 diabetic subjects, these data indicate that tissues other than skeletal muscle were responsible for the more pronounced increase in glutamine enrichment during ISO infusion. As discussed above, the most likely candidate for this release of ¹³C-glutamine is the liver, because the liver can both take up and release glutamine (high glutaminase and glutamine synthetase activity) (13,14) and also plays a central role in fatty acid metabolism. The higher increase in arterial glutamine enrichment in type 2 diabetic subjects as compared with control subjects indicated that more [U-¹³C]palmitate entered the liver in the former group, where more ¹³C label accumulated in glutamine. Support for this idea also comes from the finding that the concentration of ¹³C-glutamate tends to be higher in arterial blood of type 2 diabetic subjects, a finding most likely explained by an increased incorporation of ¹³C label from palmitate into glutamate in the liver. An increased FFA supply to the liver may promote hepatic gluconeogenesis and glucose output (15), decrease insulin binding to hepatocytes (16), and promote VLDL triglyceride production (17), all important factors in the etiology of insulin resistance, type 2 diabetes, and cardiovascular disease.

Thus the release of ¹³C-labeled oxidation products during [U-¹³C]palmitate infusion was lower in muscle of type 2 diabetic subjects than in control subjects. Mittendorfer et al. (18) reported that the ¹³C label recovery during [1,2-¹³C]acetate infusion was similar across muscle, across the splanchnic bed, and at the whole-body level in healthy subjects. In view of the differences in the exchange of ¹³C-substrates and products across muscle in type 2 diabetic subjects relative to control subjects, it would be worthwhile to study whether the above assumption also holds for type 2 diabetic subjects.

In summary, in muscle of type 2 diabetic subjects, there was no detectable release of ¹³C label into oxidation products during ISO infusion, despite a significant uptake of ¹³C label from [U-¹³C]palmitate; in contrast, in control subjects, there was a significant release of ¹³CO₂ and ¹³C-glutamine. We propose the following explanations for this finding: First, the lack of release of ¹³C-oxidation products may indicate that more palmitate was incorporated in the triglyceride stores in type 2 diabetic muscle. Because an increased intramuscular triglyceride content is strongly linked to insulin resistance in lean offspring of type 2 diabetic subjects (19), these data may indicate an important underlying biochemical mechanism in the development of insulin resistance in type 2 diabetic muscle. Second, an increased forearm lipolysis in the diabetic subjects (7) may have flooded the muscle with FFAs, resulting in a higher dilution of the [U-¹³C]palmitate in an increased intramuscular FFA pool, making it more difficult to detect any significant ¹³C-oxidation products. Third, the data also suggest that in type 2 diabetic subjects, more of the palmitate tracer is taken up and oxidized by the liver during ISO infusion.