The threonine dehydrogenase (TDG) pathway is a significant route of threonine degradation, yielding glycine in experimental animals, but has not been accurately quantitated in humans. Therefore, the effect of a large excess of dietary threonine, given either as free amino acid (+Thr) or as a constituent of protein (+P-Thr), on threonine catabolism to CO2 and to glycine was studied in six healthy adult males using a 4-h constant infusion ofl-[1-13C]threonine and [15N]glycine. Gas chromatography-combustion isotope ratio mass spectrometry was used to determine [13C]glycine produced from labeled threonine. Threonine intakes were higher on +Thr and +P-Thr diets compared with control (126, 126, and 50 μmol ⋅ kg−1 ⋅ h−1, SD 8, P < 0.0001). Threonine oxidation to CO2increased threefold in subjects on +Thr and +P-Thr vs. control (49, 45, and 15 μmol ⋅ kg−1 ⋅ h−1, SD 6, P < 0.0001). Threonine conversion to glycine tended to be higher on +Thr and +P-Thr vs. control (3.5, 3.4, and 1.6 μmol ⋅ kg−1 ⋅ h−1, SD 1.3, P = 0.06). The TDG pathway accounted for only 7–11% of total threonine catabolism and therefore is a minor pathway in the human adult.
- threonine oxidation
- threonine flux
- stable isotopes
- plasma threonine concentration
threonine is an indispensable amino acid with a complex degradative pathway. Two major pathways for the degradation ofl-threonine are known to occur in mammals. Threonine is either catabolized by threonine dehydratase (EC 188.8.131.52; TDH) to and 2-ketobutyrate, which is rapidly and irreversibly converted to CO2, or by threonine dehydrogenase (EC 184.108.40.206; TDG) to form 2-amino-3 ketobutyrate, which is mainly cleaved by 2-amino-ketobutyrate CoA ligase to form glycine and acetyl-CoA (4, 8).
The relative importance of the metabolic pathways initiated by TDH and TDG and the effect of dietary threonine on this partitioning are not well defined in humans. Zhao et al. (38) measured threonine oxidation by the production of labeled CO2 during a constant infusion of l-[1-13C]threonine in healthy adult males fed varying intakes of threonine. As threonine intake increased above the minimum requirement, labeled13CO2 increased linearly but not as rapidly as for other indispensable amino acids (38). The labeling of [13C]glycine in circulating plasma glycine was not detected during the infusion ofl-[1-13C]threonine, and the authors concluded that the conversion of threonine to glycine through the TDG pathway was not important if it exists at all in humans (38). This finding in human adults is in contrast to animal data showing that TDG is the major pathway, accounting for 80% of threonine oxidation in growing pigs (2, 21) and rats (4, 25). Furthermore, the methodology used by Zhao et al. (38) to detect labeling of glycine froml-[1-13C]threonine [gas chromatography (GC)-quadrupole mass spectrometry (MS)] may not have been sensitive enough given the relatively high glycine flux rate, which would dilute the label. For this reason, we have applied a novel and more sensitive methodology, that of GC-combustion isotope ratio MS to detect labeled [13C]glycine in circulating plasma glycine, to quantify the TDG pathway. Isotope ratio MS is ∼100-fold more sensitive than quadrupole MS (33). Using this methodology, we have recently demonstrated that the TDG pathway does exist in human infants and accounts for 44% of total threonine oxidation (9). Insight into the effect of age on threonine metabolism would be gained by quantifying the TDG pathway in the human adult fed excess threonine.
In contrast to other indispensable amino acids, threonine oxidation to CO2 in response to excess dietary threonine appears to be limited in the human infant (9) and adult (38) as well as in rats (18), and plasma threonine concentration increases dramatically. In experimental animals, TDH activity is not inducible by its substrate threonine (6, 20, 25), and the hepatic uptake of threonine in liver is known to be low compared with the uptake of other indispensable amino acids (5, 27). Increasing dietary protein in the rat leads to increased TDH activity (6), enhanced hepatic extraction of threonine (25), and a lowering of plasma threonine concentration (25). There is no study to our knowledge that has examined the effect of increasing either dietary threonine or protein intake on in vivo rates of threonine oxidation to CO2 and to glycine in the human adult.
The specific objectives of this study were to use the recently developed and more sensitive GC-combustion isotope ratio MS to quantify the TDG pathway in human adults and to determine the effect of a large increment of dietary threonine, whether provided as free amino acid or as a constituent of protein, on threonine kinetics, plasma threonine concentrations, and rates of threonine catabolism to CO2and to glycine.
MATERIALS AND METHODS
Six healthy adult male volunteers (mean age 34.0 ± 7.0 yr) were studied on an outpatient basis in the Clinical Investigation Unit at The Hospital for Sick Children (Toronto, ON, Canada). None of the subjects had a history of chronic disease, recent weight loss, unusual dietary practices, endocrine disorders, pharmacological therapy, or hormonal treatment. A summary of each subject's characteristics is shown in Table 1.
The purpose of the study and the potential risks and discomforts were fully explained to each subject, after which their written consent was obtained. The subjects were paid for their participation in the study. The protocol was approved by the University of Toronto Human Experimentation Committee and the Human Subjects Review Committee of the Hospital for Sick Children.
Feeding regimens and experimental design.
The subjects each received three isocaloric feeding regimens that varied with respect to protein and threonine level during three separate study periods. Individual daily energy intakes were determined by performing indirect calorimetry (2900; SensorMedics, Yorba Linda, CA) using a ventilated hood system on each fasted subject to measure resting energy expenditure (REE). These values were multiplied by an activity coefficient of 1.7 to ensure weight maintenance throughout the study (3, 36). Subjects were studied under conditions of a control diet (control), a high-protein diet (+P-Thr), and a high free threonine diet (+Thr; Table 2). The control diet consisted of Ensure (Ross Laboratories, Montreal, QC), which is a complete liquid formula diet. The high-protein diet consisted of Ensure with added Promod (Ross Laboratories), which is a whey protein powder. Promod was added to the complete liquid formula diet to achieve a protein content of 26% of total energy from protein. The volume of Ensure was reduced to maintain the same energy content as the control diet. The +Thr diet was produced by adding l-threonine (allo-free; United States Biochemical, Cleveland, OH) to the control diet at a level that matched the threonine content of the +P-Thr diet. The diets were divided into four equal portions that were consumed daily at 0800, 1200, 1600, and 2000. Subjects were fed the control and +Thr diets for 2 days before undergoing the isotope tracer studies on the morning of the 3rd day. Because of the significant increase in protein intake, subjects were fed the +P-Thr diet for 3 days before having the isotope tracer study performed on the 4th day (31). Subjects received the control and +Thr diets in random order and always received the +P-Thr diet last, since the +P-Thr diet was expected to cause the most perturbation to amino acid metabolism (6, 14).
l-[1-13C]threonine (99% 1-13C, allo-free; Cambridge Isotope Laboratories, Woburn, MA), [15N]glycine (99% 15N; Merck Sharp & Dohme, Montreal, QC), and NaH13CO3(90%; Merck Sharp & Dohme) were used. Chemical purity was verified by the companies by means of NMR, TLC, and gas liquid chromatography using a chirasil column to confirm the absence of the d- and allo-isomers of threonine. Stock solutions of [13C]bicarbonate (3.30 mg/ml) and priming doses of l-[1-13C]threonine (9.40 mg/ml) and [15N]glycine (5.10 mg/ml) were made up in saline (4.5 g/l). Constant-infusion doses ofl-[1-13C]threonine (1.44 mg/ml) and [15N]glycine (1.13 mg/ml) were made up in saline (9.0 g/l). Solutions were sterilized by passage through a 0.22-μm Millipore filter (Millipore, Bedford, MA) under a laminar flow hood and were transferred to single-dose vials. Each batch was shown to be sterile and pyrogen free.
Isotope study procedures and sample collection.
On the days of the infusion studies, the subject's 0800 and 1200 meals were divided into six equal meals, which the subjects consumed hourly beginning 2 h before the start of the tracer infusion. Subjects remained in a reclined and relaxed position during the 6-h infusion study.
Procedures for tracer infusion and biological sample collection were those that have been previously reported by this laboratory (10, 19,35). Briefly, three baseline samples of blood and breath and one urine sample were collected immediately before infusing the isotope tracers. Stable isotope tracers were administered via a 23-gauge butterfly needle inserted in the antecubital fossa vein of the right arm using an aseptic sterile procedure. The priming doses of isotope solutions were infused from individual syringes (Becton-Dickinson, Rutherford, NJ) in the order 1.2 μmol/kg [13C]bicarbonate, 7.8 μmol/kg l-[1-13C]threonine, and 6.7 μmol/kg [15N]glycine (34) and were followed by a saline wash over a 2-min period. The constant-infusion solutions were delivered immediately after the priming dose at rates of 4.8 μmol ⋅ kg−1 ⋅ h−1for l-[1-13C]threonine and 3.0 μmol ⋅ kg−1 ⋅ h−1for [15N]glycine (38). After finding that the enrichment of [15N]glycine in plasma for the first two subjects was not as high as desired, the continuous [15N]glycine infusion rate was increased to 5.7 μmol ⋅ kg−1 ⋅ h−1for the other four subjects. The solutions were infused over a 4-h period in the intravenous line by means of a calibrated syringe pump (IVAC Syringe Pump 710; Eli Lilly, San Diego, CA).
The isotope design employed in the present study was based on the assumption that any 1-13C labeling of glycine would not be detectable by GC-quadrupole MS, and hence only the [15N]glycine would be detected. This assumption was based on the earlier work of Zhao et al. (38) in adult volunteers in which they were unable to detect any 13C labeling in glycine. Similarly, in our neonatal work, we could not detect any13C labeling in urinary glycine (9). Finally, in the present experiment, as shown in Table 3, the enrichments of [1-13C]glycine were below the detection limit of 0.1 mole percent excess (MPE) for GC-quadrupole MS; therefore, any labeling detected was from [15N]glycine.
Blood was sampled from a 21-gauge butterfly needle that was placed in a superficial dorsal vein in the left hand and kept patent between sampling by flushing with 100 USP units/ml of heparin. Three milliliters of venous blood , having been arterialized by the hand-warming procedure (37), were drawn into heparinized syringes (Aspirator Marquest Medical Products, Englewood, CO) at 30-min intervals during the entire 6-h study. Blood samples were then stored on ice until centrifugation at 4°C. The plasma was then stored at −80°C until analysis.
Urine was collected in polypropylene bottles before starting the tracer infusion and then during the last 2 h of infusion, with a final urine sample collected at the end of the tracer infusion. Expired breath samples were collected for a 6-min period every 30 min throughout the 6-h study by means of a mouthpiece connected to a 100-liter neoprene bag via a Rudolf Valve (no. 1400; Rudolf Valve, Kansas City, MO). Complete trapping of CO2 from the breath samples in a solution of 1 N NaOH and the determination of CO2production rate were carried out as described previously (17).
Isotopic enrichment of [13C]threonine and [15N]glycine in plasma and in urine was determined by GC-MS analysis (model 5840A GC and quadrupole MS model 5985; Hewlett-Packard, Mississauga, ON) under conditions of negative chemical ionization and selected ion monitoring. Amino acids in plasma and urine were derivatized to their N-heptafluorobutyrylO-isobutyl ester derivatives (12). The isotopic enrichment of threonine and glycine was analyzed by separate injections. Selected ion chromatographs were obtained by monitoring the mass-to-charge ratio of 351 and 352 for threonine and 307 and 308 for glycine, corresponding to the unenriched and enriched ion peaks, respectively. Areas under the peaks were integrated by a Hewlett-Packard 1000E series computer.
To determine the proportion of threonine catabolized through the TDG pathway, glycine and hippurate (HA) were extracted from 2 ml of urine according to the method described by Ballevre et al. (2). Isotopic enrichments of urinary free [13C]glycine and [13C]glycine derived from [13C]HA at baseline and at the end of the tracer infusion were determined by analysis of theirN-propyl,N-acetyl derivative (33) using an Orchid system consisting of a Europa Scientific 20–20 MS with gas chromatograph-combustion interface and an HP5890 series II gas chromatograph (Europa Scientific, Crewe, UK).
The isotopic enrichment of 13C in breath CO2was measured on a dual-inlet isotope ratio mass spectrometer (VG Micromass 602D, Cheshire, UK) using techniques described previously (17). Breath 13CO2 enrichments were expressed as atoms percent excess (APE) over a reference standard of compressed CO2 gas.
Plasma amino acid concentrations were determined by ion-exchange chromatography with postcolumn ninhydrin reaction and visible colorimetric detection, using the Beckman system 7300 high-performance amino acid analyzer (Beckman Instruments, Mississauga, ON).
Threonine metabolism was evaluated according to a stochastic model, using a constant infusion approach to determine the rate of threonine oxidation to CO2 (38) and the rate of threonine catabolism to glycine (2). Kinetic calculations were performed on the data collected during the last 2 h of tracer infusion. During this period, isotopic steady state in the metabolic pool was represented by plateaus in breath 13CO2 and in plasma [13C]threonine and [15N]glycine enrichments. Plateau for breath and plasma isotopic enrichments was defined as a coefficient of variation (CV) <5% and absence of a significant slope. Plateau in urine enrichment of [13C]threonine and [15N]glycine was indicated by a nonsignificant difference between enrichments of two urine samples collected during the final 2 h of the tracer infusion, by paired Student'st-test. The mean ratio of the enriched peak to the unenriched peak in plasma for both baseline and plateau samples was used to calculate MPE.
Amino acid (threonine or glycine) flux (Q Thr orQ Gly) was calculated from the dilution ofl-[1-13C]threonine or [15N]glycine infused in the plasma at isotopic steady state using the following equation (21) wherei is the rate of amino acid infused (μmol ⋅ kg−1 ⋅ h−1), Ei is the enrichment of amino acid infused, and Ep is the enrichment of plasma amino acid at isotopic steady state (MPE). The −1 removes the contribution of the tracer to the flux.
The rate of 13CO2 released by thel-[1-13C]threonine tracer (μmol13CO2 ⋅ kg−1 ⋅ h−1) was calculated by the following equation (24) where Fco 2 is the carbon dioxide production rate (cm3/min), Eco 2 is the13CO2 enrichment in expired breath at isotopic steady state (APE), and W is the weight of the subject (kg). The constants 44.6 μmol/cm3 and 60 min/h convert Fco 2 to micromoles per hour, and the factor 100 changes APE to a fraction. The factor 0.82 accounts for13CO2 retained in the body due to bicarbonate fixation (15) and is a mean value for the fed state.
The rate of threonine oxidation to CO2(OThr→CO2) (μmol ⋅ kg−1 ⋅ h−1) was calculated as follows The rate of threonine oxidation to glycine (OThr→Gly) (μmol ⋅ kg−1 ⋅ h−1) was derived as follows whereQ Gly is estimated from the primed constant infusion of [15N]glycine; EGly is the enrichment of [1-13C]glycine in urinary free glycine or in glycine from urinary HA; EThr is the plasma enrichment of [1-13C]threonine, estimated from the primed constant infusion ofl-[1-13C]threonine; and iThr is the rate of infusion of the threonine tracer. The term [Q Thr/(iThr +Q Thr)] corrects for the contribution of the threonine tracer to OThr→Gly, as discussed by Thompson et al. (30) for the phenylalanine hydroxylation model. The location of TDG enzyme in humans, whether it is mainly hepatic or mainly extrahepatic, is uncertain; therefore, both plasma glycine enrichment and liver glycine enrichment, measured by HA enrichment, were used to estimate the glycine pool derived from threonine. Urinary [13C]glycine enrichment was used to reflect plasma [13C]glycine enrichment; we have previously shown urinary and plasma enrichments to be equal (34). Because HA is synthesized from liver mitochondrial glycine (13), the isotopic enrichment of glycine from HA represents the hepatic glycine pool that is labeled froml-[1-13C]threonine (2).
Differences among the three diet groups were evaluated by ANOVA with repeated measures using PC SAS (version 6.04; SAS Institute, Cary, NC). Post hoc comparison of differences between groups was performed using the Student-Newman-Keul's multiple range test. Isotopic enrichment of corresponding plasma and urine samples was compared using the paired Student's t-test. Results are expressed as means ± SD.
The characteristics of the six subjects are shown in Table 1. Body weight did not change significantly over the three diet periods. REE measured at the start of the study was on average 105 ± 13% of predicted values (11).
Daily amounts of energy, protein, threonine, and glycine ingested by the subjects during the three diet periods are shown in Table 2. Energy intakes were similar among diet groups by design. Protein intake was similar between the control and +Thr diets at 14.0% of energy intake and was significantly increased to 26.0% of energy intake in the +P-Thr diet, mainly at the expense of carbohydrate. Threonine intake was similar between the +Thr and +P-Thr groups and was 2.6-fold higher than the control group. Glycine intake was similar between the control and +Thr diet groups and was significantly increased in the +P-Thr diet group.
The effect of the diets on the concentrations of plasma amino acids is shown in Table 4. Threonine and 2-aminobutyrate (ABA) were the plasma amino acids to be significantly affected by the +Thr diet compared with the control diet. ABA is irreversibly produced from 2-ketobutyrate via aminotransferase through the TDH pathway. Mean plasma threonine and ABA concentrations both increased threefold on the +Thr diet compared with the control diet. On the +P-Thr diet, both plasma threonine and ABA concentrations increased 1.8-fold from control, which was significantly less than the rise observed on the +Thr diet. A significant correlation existed between plasma threonine and ABA concentrations (r 2 = 0.74,P < 0.0001) in the six subjects fed the three experimental diets. Threonine plasma concentrations in subjects fed the +Thr and +P-Thr diets were elevated beyond the age-appropriate range (79–246 μmol/l), which encompasses common physiological variables such as diet (fasting and nonfasting), gender, and time of day (28). Plasma glycine concentration significantly decreased in subjects fed the +P-Thr diet compared with the control diet. A number of other amino acids were significantly altered by feeding the +P-Thr diet compared with the control and +Thr diets, reflecting the higher intake of protein and amino acids in the former group. Specifically, plasma concentrations of the branched-chain amino acids (isoleucine, leucine, and valine; Table 4), as well as lysine, methionine, citrulline, proline, and tyrosine, increased to levels above their normal ranges (27) when subjects were fed the +P-Thr diet.
During the infusion of [1-13C]threonine and [15N]glycine, the isotopic enrichment of [13C]threonine and [15N]glycine in plasma and urine and of13CO2 in breath reached plateau by 120–150 min and was maintained in all subjects until the end of the 4-h study. Plasma [13C]threonine enrichment was significantly lower in subjects fed the +Thr and +P-Thr diets compared with the control diet (Table 3). The +Thr diet produced the greatest dilution of [13C]threonine in plasma. Plasma [15N]glycine enrichment was not significantly altered by the diets. Plateau enrichments of urinary free [13C]glycine did not differ among diet groups. Enrichment of glycine from HA was not measurable in 6 of the 18 infusion studies due to low urinary concentrations of HA glycine. In those studies with measurable HA [1-13C]glycine enrichments, there was no difference within subjects between mean urinary free [13C]glycine enrichment and HA [1-13C]glycine enrichment (paired Student'st-test; time = 1.85, degrees of freedom = 11, P= 0.1). Breath CO2 enrichments of both threonine-supplemented groups were significantly greater than the control diet group and did not differ significantly from each other. CO2 production rates of the subjects (CV 1.3%) did not differ significantly among diet groups (results not shown).
Threonine kinetics and catabolism to glycine and CO2 are shown in Table 5. Threonine intake, which included the amount delivered by the tracer, was increased 2.5-fold in both threonine-supplemented groups compared with the control group. Threonine plasma flux significantly increased 2- and 1.5-fold in subjects receiving the +Thr and +P-Thr diets, respectively, compared with the control diet. Threonine flux was significantly related (r 2 = 0.77, P < 0.0001) to plasma threonine concentration. There was no significant effect of diet on plasma glycine flux in five subjects. One subject, for whom the [15N]glycine infusion rate was 3 μmol ⋅ kg−1 ⋅ h−1, had a low enrichment of plasma [15N]glycine that could not be measured accurately and was not included in the analysis of glycine flux and of threonine disposal to glycine. The rate of threonine disposal to glycine tended to be higher in both threonine-supplemented groups vs. control but was not influenced by protein intake. The rate of threonine oxidation to CO2 was increased significantly with threonine supplementation and was not affected significantly by the form of threonine supplementation, i.e., as free amino acid or as a constituent of protein. The rate of threonine disposal to glycine was estimated to be 7–10% of total threonine catabolism, whereas the oxidation of threonine to CO2 accounted for 90–93% of total threonine catabolism. The relationship between plasma threonine concentration and the two routes of threonine disposal to CO2 or to glycine is shown in Fig. 1. Threonine disposal to CO2 rose rapidly between threonine plasma concentrations of 130 and 250 μmol/l after which there was no further increase in threonine oxidation via the TDH pathway. There was a small but significant (P < 0.04) increase in threonine disposal to glycine via the TDG pathway.
To our knowledge, this study is the first attempt to quantify the TDG pathway in human adults. The application of newly available GC-combustion isotope ratio MS permitted the measurement of a low [13C]glycine enrichment that would otherwise not be detectable by conventional GC-quadrupole MS methodology.
The TDG pathway accounted for only 10% of the total threonine degradation (sum of threonine oxidation to CO2 and to glycine) in subjects fed the control diet and therefore appears to be a minor pathway of threonine catabolism in the adult human. The present study also shows that the increase in threonine degradation in response to a large increment in dietary threonine occurred primarily through the production of CO2, presumably through the TDH pathway. Of the excess threonine provided by the +Thr diet compared with the control diet (76 μmol ⋅ kg−1 ⋅ h−1× 6 h), ∼50% was accounted for by the increase in rates of threonine disposal to CO2 and to glycine (37 μmol ⋅ kg−1 ⋅ h−1× 6 h). The remaining excess threonine could be accounted for by the significant expansion of the plasma free threonine pool size (353 μmol/l), some of which spills over into the urine. The urinary threonine concentration increased significantly from 27 to 119 mmol/mol creatinine in the adult subjects fed the +Thr diet compared with the control diet. The large expansion of the plasma threonine pool, in response to excess dietary threonine, was also observed in previous studies conducted in adults (38) and in neonates (9). Threonine oxidation to CO2 in the neonate appeared to be operating at a maximal rate (9), since increases in dietary threonine produced a doubling of plasma concentration but no significant change in the rate of oxidation. In the present study, threonine oxidation also seems to be operating at maximal capacity in the adults fed +Thr and +P-Thr diets. Studies in rats compared the metabolism and oxidation of threonine in comparison with histidine (18). Threonine oxidation was less responsive to substrate intake than was histidine oxidation. Furthermore, plasma and tissue threonine concentrations increased more dramatically in response to increased threonine intake than did the corresponding histidine concentrations to graded increases in histidine intake (18). Studies in fetal sheep have shown that growth-restricted fetuses accommodate to reduced placental transfer of threonine by reducing their oxidation of threonine, thereby routing threonine into fetal accretion (1).
When subjects consumed the higher-protein diet, plasma threonine flux and plasma threonine concentration were lower than on the +Thr diet, and rates of threonine oxidation to CO2 or to glycine were not different. The increased protein intake may have therefore enhanced both the hepatic uptake of threonine (25) and the utilization of threonine for protein synthesis during the fed state (26), resulting in a smaller expansion of the free threonine pool size. Increasing dietary protein has been shown to enhance TDH activity in rats (6); however, there was no difference in rates of threonine oxidation to CO2 in these human subjects fed the +Thr diet vs. the +P-Thr diet.
In terms of limitations of the current methodology to estimate amino acid kinetics by the continuous-infusion method, the main difficulty lies in measuring the enrichment of the precursor pools for amino acid oxidation and incorporation into protein synthesis. Plasma threonine enrichment at isotopic steady state is considered to represent the whole body intracellular pool (38) and has been used as the precursor pool for both protein synthesis and oxidation. The precursor pool for oxidation ideally represents the intracellular fraction of the amino acid bound to the degradative enzyme; this is not readily accessible in in vivo human studies. In piglets, Stoll et al. (29) have shown that threonine labeling in apo-B-100, as an estimate of hepatic protein synthesis, was 33% higher than the labeling of the free threonine pool within the hepatocytes. In these studies, the label was given enterally; hence, first pass through the gut and liver has to be considered. In the present study, the labeled threonine was given intravenously, therefore avoiding initial passage through the splanchnic bed. Ballevre et al. (2) used hepatic threonine enrichment to calculate threonine disposal through the TDG and TDH pathways, since TDH and TDG pathways were considered to operate mainly in liver. However, Le Floc'h et al. (20) demonstrated that the pancreas is a major extrahepatic site of threonine oxidation to glycine. Given that a significant extrahepatic conversion of threonine to glycine occurs, Le Floc'h et al. (20) used glycine flux, calculated from plasma measurements, together with the fractional contribution of threonine to glycine, also calculated from plasma measurements, to estimate rates of threonine conversion to glycine in pigs. In a more recent paper, Le Floc'h et al. (22) extended their studies of the TDG pathway in pigs and concluded that the TDG enzyme is located only in the liver and pancreas. They showed a high level of labeled glycine (derived from labeled threonine) in the pancreas, which suggested a high rate of TDG activity in the pancreas. However, they estimated that, due to the larger mass of the liver, 90% of total TDG activity is in the liver, with the remainder being in the pancreas (22). Due to the necessarily noninvasive nature of the present study in humans, we estimated rates of conversion of threonine to glycine from plasma measurements. In the present study, although not significantly different, urinary free glycine 13C enrichment was not equal to urinary HA13C enrichment [mean ratio of [13C]glycine to HA [13C]glycine 1.18 ± 0.66 (SD), n = 12]. HA glycine enrichment is considered to be a reflection of hepatic glycine enrichment (7); therefore, the inconsistency between the two forms of glycine may reflect a variability among the subjects in terms of the site and proportion of threonine conversion to glycine.
Bird and Nunn (4) estimated that the in vivo contribution of TDG and TDH to the overall threonine catabolic activity of rat liver was 87 and 10%, respectively. Moundras et al. (25), however, questioned the assumptions made by Bird and Nunn (4) and suggested that the TDH pathway may have been underestimated. House et al. (16) recently provided strong evidence in rat hepatocytes, using a series of specific inhibitors, that 65% of threonine catabolism occurred via the TDH pathway. In vivo studies using a multitracer method quantified the partitioning of the TDH and TDG pathways in growing pigs (2). These data showed that ∼80% of threonine oxidation occurred through the TDG pathway, and 20% occurred through the TDH pathway (2). The partition of threonine oxidation between the two pathways is clearly very different in humans vs. rats and growing pigs. In human infants, we estimated that 44% of threonine oxidation occurred through the TDG pathway (9). In adult humans, the TDG pathway accounts for only 10% of total threonine oxidation. These differences due to age may be related to a higher metabolic requirement for glycine in infants compared with adults.
The current findings may have implications for the interpretation of the direct amino acid oxidation study to estimate threonine requirement in adult males (38). Because the sequestration of the carbon-1 ofl-[1-13C]threonine into glycine represents a minor route of disposal compared with CO2production and provided that this relationship holds at low levels of threonine intake, then the shape of the intake-oxidation curve would not be expected to change. The threonine requirement suggested by Zhao et al. (38) of 10–20 mg ⋅ kg−1 ⋅ day−1is also supported by the similar mean threonine requirement of 19 mg ⋅ kg−1 ⋅ day−1recently determined by the indicator amino acid oxidation method (32). This is strong support for the argument that the TDG pathway is of minor importance in the adult human.
In summary, threonine catabolism to glycine accounted for 7–10% of total threonine catabolism, and therefore TDG is a minor pathway of threonine catabolism in the human adult. Threonine oxidation to CO2 accounted for 89–93% of total threonine catabolism, was dependent on the level of threonine ingested, and was not independently influenced by protein intake.
This work was supported by Medical Research Council (MRC) of Canada Grant MT-12928. P. B. Darling was a recipient of an MRC studentship.
Address for reprint requests and other correspondence: P. B. Pencharz, Div. of GI and Nutrition, Hospital for Sick Children, 555 Univ. Ave, Toronto, ON, Canada M5G 1X8 (E-mail:).
This work was presented in part at Experimental Biology 97, April 6–9, 1997, New Orleans, LA, and was published in abstract form (FASEB J 11: A149, 1997).
Present address of P. B. Darling: St. Michael's Hospital, 30 Bond St., Toronto, ON, Canada M5B 1W8.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
- Copyright © 2000 the American Physiological Society