Free and protein-bound glutamine have identical splanchnic extraction in healthy human volunteers

Julio J. Boza, Martial Dangin, Denis Moënnoz, Franck Montigon, Jacques Vuichoud, Andrée Jarret, Etienne Pouteau, Gerard Gremaud, Sylviane Oguey-Araymon, Didier Courtois, Alfred Woupeyi, Paul-André Finot, Olivier Ballèvre

Abstract

The objectives of the present study were to determine the splanchnic extraction of glutamine after ingestion of glutamine-rich protein (15N-labeled oat proteins) and to compare it with that of free glutamine and to determine de novo glutamine synthesis before and after glutamine consumption. Eight healthy adults were infused intravenously in the postabsorptive state with l-[1-13C]glutamine (3 μmol · kg−1 · h−1) andl-[1-13C]lysine (1.5 μmol · kg−1 · h−1) for 8 h. Four hours after the beginning of the infusion, subjects consumed (every 20 min) a liquid formula providing either 2.5 g of protein from 15N-labeled oat proteins or a mixture of free amino acids that mimicked the oat-amino acid profile and contained l-[2,5-15N2]glutamine andl-[2-15N]lysine. Splanchnic extraction of glutamine reached 62.5 ± 5.0% and 66.7 ± 3.9% after administration of 15N-labeled oat proteins and the mixture of free amino acids, respectively. Lysine splanchnic extraction was also not different (40.9 ± 11.9% and 34.9 ± 10.6% for15N-labeled oat proteins and free amino acids, respectively). The main conclusion of the present study is that glutamine is equally bioavailable when given enterally as a free amino acid and when protein bound. Therefore, and taking into consideration the drawbacks of free glutamine supplementation of ready-to-use formulas for enteral nutrition, protein sources naturally rich in this amino acid are the best option for providing stable glutamine.

  • lysine
  • oats
  • nitrogen-15 intrinsic labeling

Historically, glutamine has not been used as a nutritional supplement because it is considered a nonessential amino acid. In healthy humans there appears to be no need for glutamine supplementation. However, studies conducted in animal models of infection have shown that oral or parenteral glutamine supplementation was correlated with survival rates, decreased bacterial translocation, and enhanced gut morphology (11, 17, 23). In humans, during stress associated with injury, sepsis, etc., there is a marked increase in glutamine consumption by the gastrointestinal tract, immunologic cells, inflammatory tissue, and kidneys. Requirements for glutamine by these tissues may outstrip the synthetic capacity of the skeletal muscle. Thus, in these situations, the intracellular pools of glutamine in muscle are markedly reduced (29). Next, as tissue stores become depleted, plasma concentrations are reduced. Finally, as the deficiency state is manifested, alterations in tissue function are observed, and these changes are associated with alterations in the protein economy (e.g., negative protein balance) (23). In recent studies, it has been shown that the addition of glutamine to parenteral formulas maintained concentrations of this amino acid in blood, improving nitrogen balance and cell proliferation (13,14). However, evidence of the effects of glutamine supplementation of enteral formulas in clinical nutrition remains scarce (6, 18).

Free glutamine has an unfavorable chemical property that hampers its use in routine clinical ready-to-feed enteral formulas, namely, instability, especially during heat sterilization and prolonged storage. This limitation led to an intensive search for new alternative substrates. The best way of providing stable glutamine is to feed proteins or peptides rich in this particular amino acid that can be mixed with milk proteins to obtain a balanced amino acid profile. However, there is not enough evidence in the literature about the metabolic fate of protein-bound versus free glutamine. The objectives of the present study were to determine the splanchnic extraction of glutamine after ingestion of glutamine-rich protein (15N-labeled oat proteins) and to compare it with that of free glutamine and to determine the de novo glutamine synthesis before and after glutamine consumption.

MATERIALS AND METHODS

Isotopes.

KNO2 (50% 15N), (NH4)2SO4 (50% 15N), and Ca(NO3)2-4 H2O (50%15N) from Euriso-Top (St. Aubin, France) were used for the intrinsic labeling of the oats. For the metabolic experiments,l-[2,5-15N2]glutamine (99%15N), l- [2-15N]lysine (99%15N), l-[1-13C]glutamine (99%13C), and l-[1-13C]lysine (99%13C) were purchased from MassTrace (Woburn, MA). Chemical and isotope purities were determined by gas chromatography-mass spectrometry (GC-MS).

Production of 15N-labeled oat protein concentrate.

Spring oats (variety MH FD 2–6; Momont) were used. Seeds (2,500) were grown in an inert support medium without any nitrogen content. The oats were grown in a greenhouse with artificial light (13:11-h light-dark cycle) at 26–28°C in a hydroponic culture containing a nutritive solution. The oats were grown for 4 mo in these conditions, and 222 g of oat grains were harvested. The protein content of the grains was 11.8%, and the total 15N enrichment was 50.6%. The 15N-labeled oat grains were dehulled, milled, and digested with the aid of amylase. After digestion at 68°C for 120 min, the hydrolysate was heat treated at 90°C for 30 min, then cooled at 4°C before separation by centrifugation at 10,000 rpm for 20 min. The supernatant was discarded, and the solids were washed, separated by centrifugation, and freeze-dried.

The 15N-labeled oat protein concentrate was mixed with unlabeled oat protein concentrate to obtain sufficient amounts of the appropriate enrichments (∼5% 15N). This mixture is referred to as 15N-oats. The final glutamine and lysine enrichments and contents were checked by GC-MS after enzymatic hydrolysis. The amino acid profile is shown in Table1. The chemical composition of the15N-oats was as follows: 35.2% protein, 4.2% fat, 2.7% ash, 14.7% fiber, 41% carbohydrates, and 2.2% moisture. Similarly, a free amino acid mixture mimicking the amino acid profile of the oat protein concentrate was prepared from l-amino acids (Ajinomoto) and is referred to as 15N-AA. Five percent of the glutamine and lysine in this amino acid mixture was represented asl-[2,5-15N2]glutamine (99%15N) and l-[2-15N]lysine (99% 15N), respectively. To resemble the oat protein concentrate composition in terms of macronutrients, this free amino acid mixture was added to tapioca starch, sucrose, and soy oil. Finally, citron flavor (0.7%) and citric acid (0.6%) were also added to improve the taste.

View this table:
Table 1.

Amino acid composition of 15N-labeled oat protein concentrate

Subjects.

Eight healthy adults (4 men and 4 women, 26–44 yr old) participated in the study. Subjects were defined as healthy after medical history and physical examinations. No subject was taking any nonroutine medication 1 wk before the study or had suffered any recent gastrointestinal, cardiovascular, or infectious disease. No subject was suffering from any chronic or metabolic disease (diabetes, hypertension, celiac disease). They were instructed to maintain their usual levels of dietary intake and physical activity 1 wk before and during the study. The age, weight, height, and body mass index of the subjects are given in Table 2. The protocol was approved by the Ethical Committee of the NestléResearch Centre, and informed written consent was obtained from the subjects.

View this table:
Table 2.

Age, weight, height, and body mass index of volunteer subjects

Experimental protocol.

Before each study, sterile and apyrogen (Limulus lysate assay) tracer solutions were prepared by accurately weighing each compound and dissolving it in normal saline. The solutions were filtered through 0.22-μm filters.

Two separate protocols were performed in the eight volunteers in random order, with a 2-wk gap. The protocol is shown in Fig.1. After an overnight fast lasting 10 h, a catheter was inserted in a retrograde fashion into a dorsal vein of the hand for arterialized blood sampling after introduction of the hand into a 55°C heated, ventilated box (4). A second catheter was inserted into a vein of the contralateral arm for tracer infusion. A primed (60 × infusion rate/min) continuous infusion ofl-[1-13C]glutamine (3 μmol · kg−1 · h−1) andl-[1-13C]lysine (1.5 μmol · kg−1 · h−1) began at 7:00 AM, and continued for 8 h. After 4 h, the protein meal (containing 30 g of protein or protein equivalent from15N-oats or 15N-AA dissolved in 500 ml of mineral water) was offered, divided into 12 equal repeated meals, one every 20 min, to simulate a constant oral infusion. Blood samples were taken before any infusion (basal); every 15 min between hours 3 and 4, corresponding with the isotopic plateau of the intravenous tracer before the meal (referred to as fasted state); every 30 min after meal ingestion; and every 15 min between hours 7 and 8, corresponding with the isotopic plateau of the intravenous and oral tracer (referred to as fed state). Plasma was immediately separated in a refrigerated centrifuge and kept at −80°C until analysis.

Fig. 1.

Protocol design of studies with the labeled meals. After the first isotopic plateau, corresponding to the Fasted state, the protein meal [15N-oats or 15N-free amino acids (AA)] was offered in 12 equal repeated meals (arrows), 1 every 20 min (Fed). Blood samples (arrows) were obtained before any infusion (Basal), at the isotopic plateau of the intravenous infusion (Fasted), and after meal consumption (Fed).

Analytical procedures.

Plasma (0.75 ml) was deproteinized with 0.6 ml of sulfosalicylic acid (5% wt/vol) and centrifuged at 10,000 g for 2 min. Supernatants were acidified with 0.1 ml of 1 M HCl and desalted on cation-exchange resin (Bio-Rad 50W-X8, 100–200 mesh, H+) 1-ml columns. Water washes (2 × 5 ml) were discarded, and amino acids were eluted with 5 ml of 4 M NH4OH. The eluate was then dried under vacuum conditions (2). Plasma 13C and 15N enrichments of glutamine and lysine were determined as their tertiary butyldimethylsilyl derivative on a Finnigan MAT 252 isotope ratio mass spectrometer (Bremen, Germany) hooked to a HP5890 series II gas chromatograph (Hewlett Packard, Palo Alto, CA) via a gas chromatograph combustion interface. The interface consisted of a NiO-CuO-Pt combustion furnace reactor (940°C) and a copper reduction furnace (600°C) (5). Mean 15N and 13C enrichments of plasma glutamine and lysine at the two plateaus for the two protocols (15N-oats and 15N-AA) are shown in Table 3. Plasma amino acid concentrations were determined by ion exchange liquid chromatography (LS 6300, Beckman) (5).

View this table:
Table 3.

Mean 15N and 13C enrichments of plasma glutamine and lysine at protocol plateaus

Protein degradation.

Whole body protein catabolism was calculated from the endogenous rate of appearance of lysine, Endo Ra-Lys, which was the only source of lysine apart from that coming from the oral nutrition. Therefore, from the rate of protein breakdown (Endo Ra-Lysin μmol · kg−1 · h−1) and the abundance of this essential amino acid in body proteins (720 μmol lysine/g body proteins) (22), protein degradation in grams of protein per kilogram per hour can be estimatedProtein breakdown=EndoRa­Lys/720 Equation 1

Glutamine and lysine turnover.

The estimation of the de novo synthesis of glutamine was calculated as follows. Because lysine is an essential amino acid, its rate of appearance is entirely derived from protein breakdown in the postabsorptive state. However, the endogenous rate of glutamine appearance (Endo Ra-Gln) has two inflow components: protein breakdown (BGln) and de novo synthesis (DGln)EndoRa­Gln=BGln+DGln Equation 2Therefore, BGln is estimated with the assumption that the release of an essential amino acid (lysine) depends on the rate of protein breakdown and the content of this amino acid in body protein (92.5 mmol glutamine/100 g protein and 72.0 mmol lysine/100 g protein) (22)BGln=EndoRa­Lys×(92.5/72.0) Equation 3ThenDGln=EndoRa­GlnBGln Equation 4The total rate of appearance of glutamine (Total Ra-Gln) was calculated as follows (8)TotalRa­Gln=i13C­Gln×[Ei(13C­Gln)/Ep(13C­Gln)] Equation 5where i13C­Gln is the glutamine tracer infusion rate (in μmol · kg−1 · h−1), Ei(13C-Gln) is [13C]glutamine enrichment in the intravenous infusion (mole percent excess; MPE) and Ep(13C-Gln) is the average [13C]glutamine enrichment in plasma (MPE) during isotopic equilibrium. The endogenous rate of appearance of glutamine (Endo Ra-Gln) expressed in micromoles per kilogram per hour isEndoRa­Gln=TotalRa­GlnExoRa­Glni13C­Gln Equation 6where Exo Ra-Gln is the amount of glutamine arising from the protein meal (in μmol · kg−1 · h−1). During the postabsorptive state, Exo Ra-Gln is 0; i13C­Gln  is the glutamine tracer infusion rate (in μmol · kg−1 · h−1).

Using the same rationale, the total and endogenous rates of appearance in plasma of lysine (Ra-Lys), in micromoles per kilogram per hour, were calculated from their respective tracer infusion rates (I13C-Lys), the enrichment (MPE) of the intravenously enriched tracer [Ei(13C-Lys)], and Ep(13C-Lys) is the average [13C]lysine enrichment in plasma (MPE) during isotopic equilibrium. The endogenous rate of appearance of lysine (Endo Ra-Lys), expressed in micromoles per kilogram per hour, was calculated from the total rates of appearance in plasma of lysine, as well as from the amounts of lysine during the protein meal administration (also in μmol · kg−1 · h−1; Exo Ra-Lys)TotalRa­Lys=i13C­Lys×[Ei(13C­Lys)/Ep(13C­Lys)] Equation 7 EndoRa­Lys=TotalRa­LysExoRa­Lysi13C­Lys Equation 8

Splanchnic extraction.

The appearance rates of glutamine and lysine were calculated as explained in Glutamine and lysine turnover, either when the intravenous route was used to infuse the tracers or when the enteral route was used. However, when the latter route is used, some fraction of the tracers is taken up by the gut or liver during its first pass. Splanchnic extraction of glutamine and lysine (SEGln and SELys) therefore represents the fraction sequestered by the splanchnic bed on the first pass, which never mixes with the systemic blood. It was calculated as follows (20)SEGln=1({Ep­o(15N­Gln)/[Ei­o(15N­Gln)×io(15N­Gln)]}/{Ep­iv(13C­Gln) Equation 9 ÷[Ei­iv(13C­Gln)×iiv(13C­Gln)]}) where Ep-o(15N-Gln) is the average [15N]glutamine enrichment in plasma (MPE) of the oral tracer, Ei-o(15N-Gln) is the [15N]glutamine enrichment in the oral tracer (MPE), io(15N-Gln) is the oral tracer infusion rate of [15N]glutamine (μmol · kg−1 · h−1), Ep-iv(13C-Gln) is the average [13C]glutamine enrichment in plasma (MPE) of the intravenous tracer, Ei-iv(13C-Gln) is the [13C]glutamine enrichment in the intravenous tracer (MPE), and iiv(13C-Gln) is the intravenous infusion rate of [13C]glutamine (μmol · kg−1 · h−1)SELys=1({Ep­o(15N­Lys)/[Ei­o(15N­Lys)×io(15N­Lys)]}/{Ep­iv(13C­Lys) Equation 10 ÷[Ei­iv(13C­Lys)×iiv(13C­Lys)]}) where Ep-o(15N-Lys) is the average [15N]lysine enrichment in plasma (MPE) of the oral tracer, Ei-o(15N-Lys) is the [15N]lysine enrichment in the oral tracer (MPE), io(15N-Lys) is the oral tracer infusion rate of [13C]lysine (μmol · kg−1 · h−1), Ep-iv(13C-Lys) is the average [13C]lysine enrichment in plasma (MPE) of the intravenous tracer, Ei-iv(13C-Lys) is the [13C]lysine enrichment in the intravenous tracer (MPE), and iiv(13C-Lys) is the intravenous infusion rate of [13C]lysine (μmol · kg−1 · h−1).

Statistics.

13C and 15N enrichments and concentrations of glutamine and lysine in plasma for each subject represent the average of five blood samples taken during isotopic equilibrium. Values are means ± SD. Comparison between groups was assessed by a paired Student's t-test. Differences were considered statistically significant if P < 0.05.

RESULTS

The mean values of the plasma amino acid concentrations are shown in Table 4. Compared with the values in the fasted state (before meal administration), nutrient delivery caused a significant elevation (P < 0.05) of most plasma amino acids, with no differences related to the type of meal (oats vs. free amino acids).

View this table:
Table 4.

Mean plasma amino acid concentrations before and after continuous meal ingestion

The rate of glutamine appearance in plasma increased significantly (P < 0.05) during enteral feeding compared with the fasting state. The fraction of the rate of appearance of glutamine in plasma as a result of protein breakdown was estimated from the glutamine and lysine content of body protein as stated in thematerials and methods section. As expected from measurements performed in lysine, the release of glutamine resulting from protein breakdown decreased during enteral feeding (17% decrease, but not significant for 15N-oats; 27% decrease for15N-AA) compared with fasting values. The fraction of the rate of appearance of glutamine that was not accounted for by protein breakdown can be attributed to de novo glutamine synthesis under conditions in which there is no exogenous supply of glutamine. During enteral nutrition, de novo glutamine synthesis decreased significantly (32% 15N-AA and 35% 15N-oats) compared with values obtained during the fasting state (P < 0.05; Table 5). Figures2 and 3show, respectively, the rate of appearance of glutamine resulting from protein breakdown (BGln) and the de novo synthesis of glutamine before (fasted state) and after (fed state) the continuous15N-labeled meal ingestion (15N-Oats or15N-AA) for all volunteers.

View this table:
Table 5.

Glutamine kinetics in volunteer subjects

Fig. 2.

Rate of appearance of glutamine resulting from protein breakdown before (Fasted) and after (Fed) continuous15N-labeled meal ingestion in human volunteers.A: 15N-oats meal. B:15N-AA meal. Each line represents data from an individual volunteer.

Fig. 3.

De novo synthesis of glutamine before and after continuous meal ingestion in human volunteers. A:15N-oats meal. B: 15N-AA meal. Each line represents data from an individual volunteer.

As plasma lysine concentrations showed, total lysine rates of appearance in plasma showed a trend to increase when meal consumption began compared with the rates in the fasting state and were similar with both types of nutrition (Table 6). On the whole, nutrient meals caused a significant change in the rate of protein breakdown (P < 0.05), and this decrease did not differ between the two regimens (Table7).

View this table:
Table 6.

Lysine kinetics in volunteer subjects

View this table:
Table 7.

Protein degradation in volunteer subjects

Table 8 shows the splanchnic extraction of glutamine and lysine in human volunteers after oral consumption of either 15N-oats or 15N-AA. The splanchnic extraction of glutamine was 66.7 ± 3.9% (15N-AA) and 62.5 ± 5.0% (15N-oats).The splanchnic extraction of lysine was 34.9 ± 10.6% (15N-AA) and 40.9 ± 11.9% (15N-oats).

View this table:
Table 8.

First-pass splanchnic extraction of glutamine and lysine in human volunteers

DISCUSSION

The present study used isotope-labeled probes to assess the metabolism of glutamine and its response to exogenous glutamine delivered through the enteral route as free amino acid or as a glutamine-rich protein source. The main conclusion of the present study is that despite a 65% local utilization by the splanchnic bed, glutamine is bioavailable when given enterally, irrespective of the molecular form in which it is consumed (free amino acid or protein bound). Moreover, exogenous protein is able to modulate whole body glutamine metabolism by decreasing both de novo glutamine synthesis (32% or 35% depending on meal type) and glutamine resulting from protein breakdown (0% or 27% depending on meal type). To our knowledge, this study is the first to quantify the splanchnic extraction of glutamine provided as protein-bound glutamine as well as its influence on whole body glutamine metabolism. The results clearly show that oral administration of free glutamine and protein-bound glutamine (15N-oats) are equivalent in terms of splanchnic extraction.

Despite the high extraction of glutamine in the splanchnic bed, oral administration of 15N-oats or 15N-AA in repeated meals caused an 11–13% significant increase in the total rate of glutamine appearance in plasma compared with fasting values. As suggested from earlier studies (10, 17, 20), this demonstrates that the enteral route may be used effectively to supply the glutamine required for extrasplanchnic tissues. Although we failed to measure muscle glutamine concentrations, a rise in circulating glutamine is likely to stimulate rates of glutamine uptake by muscle (1) and could potentially correct depletion of the glutamine pool, a situation that occurs during stress and severe disease (15). On the contrary, protein administration led to a significant reduction of the de novo synthesis of glutamine, the main glutamine supplier during the postabsorptive state. This hypothesis has been proven in cultured muscle cells in which increasing glutamine concentrations inhibited glutamine synthetase activity (26). Likewise, the glutamine kinetic results are in agreement with those of Hankard et al. (15), who observed in human volunteers that enteral administration of glutamine produced a sharp reduction in de novo synthesis of glutamine. In very low birth weight infants, Darmaun et al. (9) observed similar results. It seems that there is a feedback mechanism between glutamine arriving in plasma from exogenous sources and that arriving by de novo glutamine synthesis. In the study by Hankard et al. (15), the oral administration of glutamine did not affect the appearance in plasma of glutamine resulting from protein breakdown.

In the present study, compared with fasting values, the appearance in plasma of glutamine as a result of protein breakdown decreased 27% after 15N-AA consumption, and the difference was not significant after 15N-oats consumption, likely because of the low number of volunteers in each group, which led to a lack of statistical significance. Nevertheless, in the present study, subjects received test meals containing not only glutamine but whole protein or a mixture of free amino acids plus carbohydrates and fat. This could have inhibited protein breakdown and also affected the de novo synthesis of glutamine compared with fasting values as Table 7 shows for both test meals. On the contrary, in the study by Hankard et al. (15), volunteers received a nasogastric administration of glutamine only.

The most important goal of the present study was to compare the metabolic fate of enteral glutamine provided within a protein to that of free glutamine. With regard to the first-pass extraction of glutamine in human volunteers, our results are in accordance with other data in the literature (9, 15, 20). Unfortunately, all these studies were done with free glutamine, and the plasma glutamine response to enteral administration of protein-bound glutamine and its splanchnic extraction remain undetermined. The molecular form of the protein source can affect, per se, protein kinetics and the metabolic fate of amino acids as a result of several factors, such as gastric emptying and sites and rates of absorption. However, this was not the case.

According to our results, glutamine resulting from oats behaved the same way as free glutamine in terms of splanchnic extraction and plasma glutamine response in human volunteers. It should be emphasized, however, that all amino acids were labeled in the 15N-oats. It has been shown elsewhere (8) that branched-chain amino acids (BCAA) are the major donors of α-amino N for glutamine synthesis. Therefore, part of the 15N-labeled BCAA that result from 15N-oats could have contributed to de novo synthesized [15N]glutamine. Our isotope detection method (isotope ratio mass spectrometry) did not allow us to differentiate whether the enriched glutamine was labeled in one (from de novo synthesis by transamination mainly from 15N-labeled BCAA) or two nitrogens (glutamine present in 15N-oats). Estimating the 15N isotopic enrichment of plasma glutamine would lead to an overestimation of the labeled glutamine that appeared in plasma as a result of the test meal. This was not the case when volunteers received 15N-AA, a preparation that only contained two 15N-labeled amino acids, glutamine and lysine, the latter of which cannot be transaminated. Darmaun and Déchelotte (8) estimated that the contribution of leucine accounted for 9% of de novo glutamine synthesis in the postabsorptive state in human volunteers. However, the data of Darmaun and Déchelotte (8) were derived from determinations performed solely in plasma. Glutamine synthesis obviously occurs in the intracellular space. Therefore, estimating the isotopic enrichment in the intracellular space from measurements made in plasma clearly leads to an overestimation of the true precursor enrichment. We failed to measure 15N enrichments of plasma BCAA, which would have allowed us to estimate the contribution of transamination to15N enrichments of plasma glutamine, but we know the initial 15N enrichment of the meal containing the15N-labeled oat proteins, the relative content of each one of the BCAA in the oat protein, as well as the oat protein administration rate for each volunteer. From these numbers, and assuming similar amino acid fluxes, we can quantify theoretically the contribution of transamination from BCAA resulting from the labeled meal to the 15N enrichment of the amine group of glutamine. We have estimated this contribution to be 7.9 ± 1.9%. Therefore, taking into account that ∼8% of 15N enrichment in plasma glutamine comes from transamination from BCAA and not directly from glutamine present in labeled oats, splanchnic extraction of glutamine after administration of oat protein can account for 62.5 ± 5.0%, which is even closer to that seen in volunteers receiving the meal based on free amino acids (66.7 ± 3.9%).

Despite the limitations of the study, approximately one-third of the enterally administered glutamine eventually reached the systemic circulation, demonstrating that in the clinical setting the enteral route may supply glutamine efficiently for the extrasplanchnic tissues. Apart from that, however, the other two-thirds remained in the splanchnic tissues (especially the gut) for protein synthesis and oxidation. Because gut proteins are not especially rich in glutamine and because other amino acids have lower splanchnic extraction rates compared with that of glutamine, it is logical to think that the majority of the glutamine captured in the small intestine is used for purposes other than mucosal protein synthesis. Enterocytes, like other rapidly dividing cells, have been shown to utilize glutamine to a greater extent than any other fuel source (27). Glutamine requirements for the gut and the immune system are increased in critical illnesses such as sepsis or trauma. Therefore, an adequate supply of this amino acid at the right level is crucial, first, to maintain the functionality and the integrity of the small intestine and second, to limit the muscle protein catabolism to yield glutamine via transamination of BCAA (19). The splanchnic extraction of lysine was 34.9 ± 10.6% (15N-AA) and 40.9 ± 11.9% (15N-oats). These results are in agreement with those obtained previously by Hoerr et al. (16) who estimated the first-pass splanchnic uptake of lysine in young volunteers in the fed state. They quantified it to 35%. First-pass splanchnic uptake of lysine in young pigs (in the fed state) has been found to be 32% of the total lysine intake (P. Reeds, personal communication).

Repeated meal feeding, either with native proteins or with free amino acids, led to a decrease in protein breakdown as assessed by Endo Ra-Lys. These results are consistent with data reported in numerous continuous feeding studies (3, 12, 24). This inhibition of protein breakdown is unlikely to be due to insulin because repeated meals are not able to produce detectable insulin increases (B. Beaufrère, personal communication). Increases in plasma amino acid concentrations per se are known to inhibit protein breakdown (7, 28), even in the presence of basal insulin levels (7). Although the plasma hyperaminoacidemia was not very high (Table 3), it was prolonged for 4 h. This would suggest that a significant duration of hyperaminoacidemia is necessary to inhibit protein breakdown. Boirie et al. (4) have suggested that there is a relationship between the regulation of postprandial protein kinetics and the rate of digestion and absorption of dietary proteins and thus amino acid appearance in plasma. They found that whey proteins are more rapidly digested than native micellar casein. In addition, postprandial leucine oxidation was significantly higher after the ingestion of whey proteins than of casein despite identical leucine intakes. The authors concluded that different digestion rates and amino acid profiles result in different protein kinetics.

In the present study, the only major difference between meals was the molecular form of the protein source (native protein or free amino acids), although the amino acid profile was identical. This fact, along with the experimental design that included 12 repeated meals, meant that plasma amino acid appearance was the same for both meals for almost the entire feeding period and, subsequently, had the same effect on whole body protein kinetics. Metges et al. (21) recently studied the kinetics ofl-[1-13C]leucine when ingested with free amino acids and unlabeled or intrinsically labeled casein in human volunteers in the fed state. They did not find that the form of dietary nitrogen or of the leucine tracer administered had an effect on estimates of splanchnic extraction. However, they found higher leucine oxidation and lower nonoxidative leucine disposal when thel-amino acid diet was given compared with the intrinsically labeled casein. Moreover, they found higher mean plasma leucine, isoleucine, and valine concentration during the ingestion of the free amino acid diet that may be causally related to the higher rate of leucine oxidation. In the present study, we did not find differences in lysine kinetics between 15N-oats and15N-AA.

Also, the question arises as to whether the nature of the specific labeled protein could have an effect on amino acid utilization. That is, would a different intrinsically labeled protein, as in our study, oats, have demonstrated a lower rate of amino acid oxidation or a higher nonoxidative amino acid disposal (21)? Thus a recent study (25) in elderly women showed that when dietary protein intake was increased through the addition of vegetable protein, postabsorptive protein breakdown was not inhibited to the same extent as that occurring when animal protein was given. Therefore, all these findings reveal that the immediate metabolic fate of absorbed amino acids is determined by a complex interaction of factors including the molecular form of the amino acid ingested, the amino acid profile, the composition of the meal, the level of intake, and the pattern of meal ingestion (21).

The main conclusion of the present study is that despite a two-thirds local utilization by the splanchnic bed, glutamine is bioavailable when given enterally, irrespective of the molecular form in which it is consumed (free amino acid or protein bound). Therefore, and taking into consideration the drawbacks of free glutamine supplementation of ready-to-use formulas for enteral nutrition, protein sources naturally rich in this amino acid are the best option for providing stable glutamine.

Acknowledgments

We thank Dr. C. Garcı́a-Ródenas for helpful discussion and help in the experimental design and Dr. C. Schindler (Service de Pharmacie-CHUV, Lausanne, Switzerland) for preparing the tracer solutions for intravenous perfusion as well as for performing the sterility and apyrogenicity tests of the labeled tracers.

Footnotes

  • Address for reprint requests and other correspondence: P.-A. Finot, Nestlé Research Cntr., Vers-chez-les-Blanc, 1000 Lausanne 26, Switzerland.

  • 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. Section 1734 solely to indicate this fact.

REFERENCES

View Abstract