The Journal of Nutrition Vol. 128 No. 11 November 1998, pp. 1961-1968

The Current ¹⁵N-Leucine Infusion Technique Is Not Suitable for Quantitative Measurements of Ileal Endogenous Amino Acid Flows in Pigs^1,2,3

Pascal Leterme⁴, Bernard Sève^*, and André Théwis⁵

Faculté Universitaire des Sciences Agronomiques, Unité de Zootechnie, 5030 Gembloux, Belgium and ^* INRA Station de Recherches Porcines, F-35590 Saint-Gilles, France

	ABSTRACT

Abstract Introduction Methods Results Discussion References

The current ¹⁵N-leucine infusion technique may overestimate the ileal endogenous nitrogen losses in pigs. To determine the reason, we infused four cannulated pigs intravenously, fed them a pea-based diet with ¹⁵N-leucine, and examined some methodological variables. Neither the blood sampling time nor the choice of precursor pool (total N or amino acid N of deproteinized plasma) or the method of estimation of the isotopic equilibrium level significantly affected the results. On the other hand, the ¹⁵N-enrichment of purified mucin, isolated from ileal digesta, was higher than that of the plasma amino acid pool (0.114 vs. 0.077 atom % excess). The endogenous proportion of the labeled amino acids (Ala, Gly, Ile, Leu and Val) in the ileal digesta ranged from 23 (Leu) to 74% (Ala), compared with 70% for total N. The low value of leucine was ascribed to the constant marker infusion condition. In pigs infused with ¹³C-leucine, a similar endogenous proportion was obtained for lumenal leucine with ¹³C-leucine and ¹⁵N-leucine infusion. However, the ¹³C-enrichment of the leucine bound to mucin was markedly lower than that of plasma leucine (38%). The endogenous amino acid flows were also estimated by combining the ileal N flow measured with ¹⁵N and the endogenous amino acid profile obtained by means of an N-free diet. They were different from those obtained with the ¹⁵N-amino acid dilution technique. We conclude that the precursor pool currently used (plasma total N or total -amino acid N pools) is a poor indicator of the enrichment of the secretions and that the infusion of one labeled amino acid is not sufficient to extend the method at the amino acid level.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

The isotope dilution technique using ¹⁵N is increasingly used to distinguish between endogenous and dietary nitrogen in the digestive tract of pigs. The endogenous N is labeled after intravenous infusion of ¹⁵N-leucine because the latter transaminates and spreads its ¹⁵N to other amino acids (AA),⁶ thereby increasing the representativity of the isotope as a marker of the endogenous AA pool. The method allows the establishment of a hierarchy of feedstuffs, according to their effect on the endogenous N losses (Grala et al. 1998).

The method is based on a series of hypotheses that have never been validated thoroughly. For example, nitrogen from leucine and from the AA labeled through transamination with ¹⁵N-leucine is considered to be representative of the whole endogenous AA pool, and the different protein secretions are assumed to be uniformly labeled, their ¹⁵N-enrichment matching that of the precursor pool for their synthesis, which is supposed to be that of the plasma free AA.

Finally, the method is believed to overestimate the ileal endogenous N losses in pigs because the true protein digestibilities calculated with these losses were in fact always very high (Gabert et al. 1997, Leterme et al. 1996a, Lien et al. 1997a and 1997b). Lien et al. (1997a and 1997b) suspected the influence of some methodological variables. Gabert et al. (1997) also claimed effects of feeding frequency, infusion protocol, rate of tracer infusion or sampling procedure. However, other parameters require attention. For example, the digestive secretions (Souffrant et al. 1993) and the AA (Hess et al. 1997, Lien et al. 1997a and 1997b) are heterogeneously labeled. Leucine, the infused AA, remains highly enriched in blood and in the secretions, whereas a number of AA are poorly enriched or not labeled at all.

Attempts have been made to obtain results for each AA either by measuring the ¹⁵N dilution in the individual AA (Gabert et al. 1997, Lien et al. 1997a) or by combining the values of ileal endogenous N flow measured with ¹⁵N, with an endogenous AA profile determined by means of an N-free diet (de Lange et al. 1990, Leterme et al. 1996a).

The aims of this study were to verify some of the hypotheses on which the method is based and to determine whether it is possible to obtain results at the AA level as follows: 1) the effect of some methodological parameters (choice of the precursor pool, sampling procedure, estimation of the ¹⁵N plateau enrichment level) was first verified; 2) the ¹⁵N-enrichment of an isolated intestinal secretion (mucin) was determined and compared with the blood amino acid pool; 3) the ¹⁵N-enrichment of leucine and other AA was measured at the plasma and lumenal level to calculate the individual losses using the ¹⁵N-AA isotope dilution method; 4) some pigs were also infused with ¹³C-leucine, and the ¹³C-enrichments were determined on leucine at the lumenal, plasma and mucin levels; 5) this was compared with calculation of AA losses through combining the endogenous N flow, determined with the ¹⁵N dilution technique, with an endogenous AA profile obtained by means of an N-free diet, supplemented or not with a highly digestible protein.

	MATERIAL AND METHODS

Abstract Introduction Methods Results Discussion References

Animals.  Eight male Large White pigs (Genes diffusion, Douai, France; initial body weight, 55 ± 3 kg) were fitted with a post-valve T-cannula as described by van Leeuwen et al. (1991). After recovery (7 d), the pigs were fitted with two permanent catheters (Silclear tubing; Degania, Israel; i.d., 1.5 mm) in the jugular veins (12 cm inside the vein) for blood sampling and labeled leucine infusion, respectively. The latter started the following day. The experiments were conducted under the guidelines of the Belgian Ministry of Agriculture for animal research.

Diets.  Winter peas (Pisum sativum L. cv Frisson) were ground through a 1.5-mm mesh screen. Pea hulls, inner fibers and starch used for the N-free diets were isolated from pea seeds and provided by Provital (Warcoing, Belgium). The isolation process and the composition of the fiber sources were previously described (Leterme et al. 1996b). Egg yolk (Belovo, Bastogne, Belgium) was roughly defatted with hexane.

A diet was formulated to contain peas as the sole protein source (Table 1). A second diet was composed of pea starch and pea isolated fibers (hulls and inner fibers) and sucrose, to be protein-free (N-free diet). The third diet was similar to the second but supplemented with defatted egg yolk (egg yolk diet) as a totally digestible protein source, at the expense of sucrose (Table 1). All diets were supplemented with chromium oxide (3 g Cr₂O₃/kg dry matter).

Experimental procedure.  Leucine infusion. The eight pigs received daily, in two meals (0800-1600 h), 90 g dry matter/kg metabolic weight of the pea diet, mixed with water (1:1). They were randomly allocated into two groups and received a blood infusion of a sterile saline solution containing either L-¹⁵N-leucine (98 % enrichment) or L-[1-¹³C]-leucine (9% enrichment) for 9 d. The solution was pumped with a peristaltic pump (P1; Pharmacia, Uppsala, Sweden) at the rate of 500 mL/d through the first catheter to provide 20 mg leucine/(kg body weight · d). Two blood samples (20 mL) were collected daily by means of the second catheter, 2 and 6 h postprandially; a third sample was also collected 4 h after the meal on the last 3 d (d 7 to 9). Samples were immediately centrifuged (2000 × g) and the plasma (supernatant) was kept at 18°C. The ileal digesta were collected on the last 3 d from 0900 to 1700 h, with plastic bags attached to the cannula. The samples were pooled per day and immediately frozen at 18°C.

Protein-free diets.  After the first experiment, the catheters were removed and the pigs randomly allocated into two other groups. After a 7-d rest, they received, for 5 d in two meals per day, 90 g dry matter/kg metabolic weight of an N-free diet, either not supplemented (N-free diet) or supplemented (egg yolk diet) with defatted egg yolk (Table 1). The ileal digesta were collected from 0900 to 1700 h on the last 3 d, pooled and frozen immediately after collection.

Chemical analyses.  The diet ingredients were analyzed for nitrogen, starch, AOAC dietary fiber, neutral and acid detergent fibers, as described by Leterme et al. (1996b). The AA were obtained through acid hydrolysis (6 mol/L HCl, 22 h, 110°C) and separated through HPLC using the Pico-Tag method of Waters (Millipore, Bedford, MA) with phenyl-thiocarbamyl derivatives and fluorimetric detection. The same method was used for cysteine; methionine was determined by the same method after oxidation with performic acid before hydrolysis. Tryptophan was not analyzed.

The digesta were analyzed for ¹⁵N-enrichment (see below), N, AA and chromium contents. The last mentioned was obtained by titration with Mohr's salt after nitric acid/perchloric acid digestion, as described by François et al. (1978). A small fraction of ileal digesta was also used for mucin isolation by putting 3 g digesta in 25 mL of a 0.15 mol/L NaCl solution. After centrifugation (12000 × g, 30 min, 4 °C), 15 mL of the supernatant was added to 22 mL ethanol (0°C), kept for one night at 20 °C and centrifuged (1400 × g, 10 min, 4°C). The precipitate was recovered in 15 mL of the NaCl solution, treated again and freeze-dried. It was composed mainly of raw mucus contaminated by noncovalently bound proteins. The latter were discarded by treatment with proteases (Pronase), and mucin was then purified by two gel filtrations, as described by Mantle and Allen (1981).

The plasma samples were treated twice with a 0.6 mol/L trichloroacetic (TCA) solution for protein precipitation. The TCA-soluble fraction (deproteinized plasma) was divided into two parts, the first for ¹⁵N analysis on total N and the second for isolation of the free AA, through passage on a cation-exchange resin (Amberlite IR 120, Sigma, St. Louis, MO). The AA were recovered with NH₄OH (4 mol/L). The samples were rinsed twice with distilled water and boiled, for ammonia removal. The isolated fractions were freeze-dried before ¹⁵N analysis. The ¹⁵N-enrichments of the digesta, TCA-soluble fraction and blood free AA samples were determined with an elemental analyzer coupled to an isotope-ratio mass spectrometer (EA-IRMS; Micromass UK, Manchester, UK). The ¹⁵N-enrichments of leucine, isoleucine, valine, alanine and glycine were also determined in the blood free AA and in the digesta with a gas chromatograph coupled to an IRMS through an interface composed of two successive ovens, combustion and reduction (GC-C-IRMS; VG Isochrom, Micromass UK). The samples were first hydrolyzed (6 mol/L HCl + 1% phenol; 24 h at 110°C in glass tubes under N atmosphere) and the AA derivatized in their ethylchloroformate derivatives, as described by Husek (1991).

The ¹³C-enrichment was measured on blood free leucine, lumenal leucine and leucine of mucin by means of the GC-C-IRMS, without reduction oven, using procedures similar to those used for ¹⁵N. For the measurements on lumenal leucine, the digesta were pooled per day and for the mucin leucine, the samples were pooled per animal. Because only carbon 1 was labeled, the ¹³C-enrichments were recalculated in mol % excess.

Statistical analyses and calculations.  Results are presented as means ± SD for isolated data or SEM when different treatments are compared. ANOVA was used, followed where necessary by the Student-Newman-Keuls test for comparison of means. For the comparison of the two estimation methods of plateau ¹⁵N-enrichment, a paired Student's t test was used.

	RESULTS

Abstract Introduction Methods Results Discussion References

Precursor pool and sampling procedure.  The time courses of ¹⁵N-enrichment in the total N and -amino acid N of deproteinized plasma are illustrated in Figure 1. The patterns of both blood fractions collected 2 h after the meal were identical (Fig. 1A), whereas those collected 6 h postprandially were slightly different, namely, at plateau level (Fig. 1B).

View larger version (17K): [in this window] [in a new window]   Fig 1. Time course of the 15N-enrichment (atom % excess, APE) of total N and amino acid N of deproteinized plasma in pigs receiving a continuous blood infusion of 15N-leucine [20mg/(kg bodyweight · d)]. Blood samples were collected 2 h (A) and 6 h (B) after the meal. Values are means + SD, n = 4.

The ¹⁵N-enrichments at plateau were estimated either by calculating the average enrichment of the last 3 d of infusion or by using the regression equations presented in Figure 1. Both values were used independently to calculate the proportion of endogenous N in the ileal digesta and the ileal flows of endogenous N (Table 2). There was no significant difference (P > 0.05) between the two methods of estimation. Blood sampling time (2 or 6 h after the meal) had no significant influence on the endogenous N flow estimation, with only one exception, i.e., plasma total N 6 h after the meal, presumably due to a higher enrichment of non--amino compared with -amino acid N. The ¹³C-enrichment of plasma leucine was also not affected by the blood sampling time.

Precursor pool and AA isotope dilution technique.  The ¹⁵N of leucine was partly transferred to other AA such as isoleucine, valine, alanine and glycine (Table 3). The other dispensable AA, namely, glutamine and glutamic acid, were not considered here. The ¹⁵N-enrichment of plasma leucine was 16 times higher than that of the whole plasma amino acid pool and had two and four times the enrichment of isoleucine and valine, respectively. The endogenous proportion of each AA in the ileal digesta was estimated by measuring the ¹⁵N dilution at the AA level. The proportions were quite variable from one AA to another, but the values obtained for isoleucine, valine and alanine were comparable to those obtained for total N. On the contrary, that of leucine reached only 28% of that of total N. It is noteworthy that the proportion of endogenous leucine in the digesta measured with ¹³C was comparable to that obtained with ¹⁵N.

Because of the low amounts and the low N content of available mucin, the purified mucin samples of the pigs were pooled for ¹⁵N determination. The ¹⁵N-enrichment measured on mucin isolated from the ileal digesta was higher than that of the plasma AA pool [0.114 vs. 0.077 atom % excess (APE)], assumed to be its main precursor pool. When the enrichment of mucin was taken as reference for the endogenous N pool, the estimated endogenous proportion was markedly lower than that obtained with the enrichment of the plasma AA pool (Table 3). On the contrary, the ¹³C-enrichment of the leucine in mucin was lower than that of plasma leucine (0.19 vs. 0.50 APE). The contribution of endogenous leucine in the digesta was limited when the ¹³C-enrichments of plasma were taken as reference (20%). When the ¹³C-enrichment of mucin leucine was the reference, the relative contribution of endogenous leucine was markedly greater (65%).

Endogenous AA profile and true digestibility.  The AA profile of the endogenous secretions was estimated by the N-free diet technique (Table 4). The addition of a highly digestible protein source (egg yolk) increased the total ileal AA flow significantly (P < 0.05), and the difference was also significant for 12 of the 17 AA studied. The highest increase was monitored for serine, whereas a decrease was observed for proline. The AA profile (individual AA/ AA ratio) was slightly modified for some of the dispensable AA, namely, proline and serine.

These AA profiles were used to extrapolate the value of ileal total endogenous N flow measured by the ¹⁵N dilution method in pigs fed a pea diet, to the flow of each AA (Table 4). The conversion was based on the ratio of the ileal N flow obtained with the N-free diet to that obtained with the ¹⁵N dilution method (see footnote Table 4). The flow measured with the N-free diet corresponds to the basal losses, whereas that measured with ¹⁵N corresponds to the total losses (basal + specific of pea intake). In this case, the effect of protein addition was more limited. The difference was significant for only seven AA and not different (P = 0.072) for the sum of the AA. As for the N-free diet, the flow of proline was lower when egg yolk was added but, in this case, the difference was significant (P < 0.05). The ileal flows obtained were quite different from those measured with the ¹⁵N-AA dilution method (Table 3), i.e., the flow of glycine measured with the ¹⁵N-AA dilution method was lower than that obtained with the N-free diet.

As expected, the real AA digestibilities calculated with the endogenous losses estimated by the ¹⁵N dilution technique (total losses) were higher than the true ones, calculated with the data provided by the N-free technique (basal losses). In both cases, the effect of protein supplementation was limited, with the exception of serine and proline (Table 5).

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Methodology.  With the current ¹⁵N-dilution technique through ¹⁵N-leucine infusion, the main methodological problem is to determine an accessible reference pool of AA in which N labeling is the same as in the endogenous proteins. Deproteinized plasma N, composed mainly of AA, is considered to approach the precursor pool for secreted protein synthesis and, consequently, to have the same ¹⁵N-enrichment as that of the secretions. The method is suspected of providing overestimated values of endogenous N flows; Lien et al. (1997b) suggest that the nonprotein N fraction present in plasma, which is less enriched than the amino acid N, would dilute the ¹⁵N of the latter, thereby inducing an overestimation of the ileal endogenous N flows. In this experiment, we observed very limited differences between both blood fractions (Fig. 1; Table 2). Although the -amino acid N seems preferable to the total N of plasma, no evidence of any improvement brought by the isolation of the free AA was given here. Our results suggest that total N enrichment may increase, through non--amino acid N labeling, 6 h after the meal. This may be a consequence of increased oxidation of leucine and other AA at this time, but will certainly not improve the representativity of plasma total N enrichment for that in secreted proteins.

Lien et al. (1997a) observed differences mainly between blood samples collected before or after feeding; in fed pigs such as ours, however, the differences were more limited (Table 2). On the other hand, the blood sampling site could have some influence because differences of ¹⁵N-enrichment of total N or the individual AA have been observed between the jugular, hepatic and portal veins (Hess et al. 1997, Yu et al. 1990). Efforts to improve the methods are necessary but our results lead us to conclude that the methodological variables mentioned here do not jeopardize its validity and that more fundamental parameters must be studied.

¹⁵N dilution and precursor pool.  With regard to the representativity of leucine N for total N, both in the plasma free amino acid pool and in the endogenous proteins, transamination is often presented as an advantage because the ¹⁵N of leucine is spread to isoleucine, valine and the dispensable AA. Transamination is the initial step for transferring ¹⁵N from leucine to other AA, but other mechanisms are involved afterwards.

However, in agreement with others (Hess et al. 1997, Lien et al. 1997a), ¹⁵N was found to be heterogeneously distributed among the labeled free AA in plasma and in the secretions (Table 3). Moreover, most other AA, among them all those with more than one atom of N, are not labeled, and those present at the highest levels in plasma are among the less enriched, if enriched at all. Therefore, the main limit to the representativity of the N from leucine and other labeled AA will be the differences in the AA profile between the plasma free amino acid pool and the secretions. For example, the branched-chain AA (Ile, Leu, Val, highly labeled) represent nearly 20% of the AA of the pancreatic secretions (Corring and Jung 1972) and threonine (not labeled) about 26% of pig intestinal mucin (Mantle and Allen 1981), whereas their respective contribution to the plasma AA pool reaches only 4.5 and 0.9%, respectively (Rérat et al. 1988). Thus, a gap in the technique could be first the lack of representativity of N from leucine and the metabolically related AA for N of the other AA.

An important question is whether the plasma free AA pool is a reliable reflection of the true precursor pool for synthesis of secreted proteins. There is an intracellular pool that exchanges AA with blood and the tissues and contains some AA released from the breakdown of proteins. The splanchnic tissues take also part of the AA absorbed from the lumen for their own syntheses.

The contribution of each precursor will vary from one secretion to another and depend on the rate of protein turnover. For example, if they were not the consequence of heterogeneous labeling and composition of their AA, the higher ¹⁵N-enrichments measured in pancreatic enzymes (Souffrant et al. 1993) and in mucin (this study, Table 3), compared with the plasma amino acid pool, could be attributed to a fast turnover that required direct incorporation of the blood free AA into proteins without prior equilibrium with the free intracellular pool. However, at the epithelial level, we do not know to what extent the lumenal N dilutes the ¹⁵N-enrichment coming from blood.

We attempted here to see whether the ¹⁵N-enrichment of an individual secretion (mucin) could be taken as reference for the whole endogenous N in the digesta. In this case, the estimated proportion of endogenous N in the digesta was lower than that obtained with the blood amino acid N (47 vs. 70% endogenous N; Table 3). The value was closer to that obtained with ¹⁵N-labeled peas (55%) in another experiment in which pigs were fed a similar diet based on peas (Leterme et al. 1996a). However, the different secretions have different ¹⁵N-enrichments. For example, that of pancreatic juice was found to be twice that of the blood amino acid pool (Souffrant et al. 1993). Therefore, it will be difficult to determine a pool in which the ¹⁵N-enrichment is representative of that of the whole endogenous N in the ileal digesta, all the more because we do not know to what extent the dietary factors can modify the proportion of the different secretions recovered at the ileal level. At the current stage, the ¹⁵N-enrichment of the blood free amino acid pool seems to be an unsatisfactory surrogate measure of that of the ileal endogenous N fraction.

¹⁵N-amino acid dilution technique.  One of the objections to the ¹⁵N dilution technique is that ¹⁵N labeling will be affected by the heterogeneous labeling of the AA combined with different AA composition in plasma and secreted proteins. The direct measurement of endogenous AA through the ¹⁵N-AA dilution technique could provide useful data to analyze more specifically the problem of the reference pool. We wanted to compare the ¹⁵N-enrichment of the leucine of mucin to that of plasma leucine, but the samples were too small for accurate analysis by GC-C-IRMS. The latter was possible for the ¹³C-enrichment although the difficulty in purifying mucin may have affected the accuracy of the results, as shown by the high SD (Table 3).

The lower ¹³C-enrichments measured in mucin leucine, compared with that of plasma leucine (38%) could indicate that plasma is not the unique source for mucin synthesis. This confirmed the data of Lien et al. (1997a) who found an ¹⁵N-enrichment of mucin leucine lower than that of plasma leucine (55%), whereas those of isoleucine and valine were nearly the same in both fractions. The differential response of leucine compared with other AA may be related in part to constant ¹³C- (or ¹⁵N)-leucine infusion. Thus, the high level of the tracer in the plasma samples (Table 3) may contribute to the underestimation of the ileal flow of endogenous leucine.

The ¹⁵N-enrichment of AA other than leucine must be close to that of the intracellular pool because transamination occurs at this level. It does not mean that this pool would better reflect the enrichment of the secretions. Vanvenrooij et al. (1973) showed in vitro that proteins in the rat submandibular gland were synthesized from an extracellular amino acid pool rather than from intracellular AA.

The fact that the splanchnic region, the main provider of endogenous proteins released in the lumen, uses substantial amounts of dietary AA for protein syntheses provides further evidence that the enrichment of the plasma AA may not be sufficiently representative of that of the AA incorporated into the secretions. The cells in the crypts extract AA mainly from blood, whereas those near the villus tip take up mainly lumenal AA (Alpers 1972). Simon et al. (1983) showed that intracellular leucine in the stomach mucosa is mainly of lumenal origin and is used for synthesis of secretory proteins. Recently, Stoll et al. (1998) determined, during short-term (6 h) intragastric infusion of ¹³C-labeled protein in piglets, that 18, 21, 18 and 12% of the total first-pass metabolism of lysine, leucine, phenylalanine and threonine, respectively, was recovered in mucosal protein. They conceded that the real rate could have been underestimated as a result of the short period of infusion. On the other hand, they also estimated that only 4.6% of the arterial flux of leucine was taken up by the portal-drained viscera. Furthermore, in starved men, Gaudichon et al. (1994) found that the ¹⁵N-enrichment of lumenal protein-bound leucine reached only one third of that of plasma free leucine. This will significantly influence the enrichment of the AA bound to the digestive secretions. Moreover, we do not know to what extent the distribution of ¹⁵N among the blood free AA is representative of that in the secretions because of all of the metabolic processes that happen in the tissues. The enrichment of the plasma free AA is also affected by feeding (Hoerr et al. 1991, Ljungqvist et al. 1997, Matthews et al. 1982).

On the other hand, the AA profile of the endogenous secretions and of the indigestible proteins can markedly influence the relative contribution of the endogenous AA in the total AA recovered at the ileal level. This may explain in part the smaller contribution of endogenous leucine to total leucine in digesta compared with that of other AA.

Ileal endogenous AA losses and true AA digestibilities.  The extrapolation of the endogenous N flow to that of the individual AA by using an endogenous AA profile determined with an N-free diet is based on the ratio of the ileal endogenous N flow obtained with ¹⁵N to that obtained with the N-free diet. However, the endogenous N compounds excreted at the ileum after N-free diet intake contain only 78% of amino acid N, on average (Seve and Leterme 1997). Moreover, the N-free diet provides the profile of the basal endogenous AA fraction, i.e., a constant fraction, independent of the diet. The latter is normally composed of dietary factors able to modify the ileal endogenous N losses and possibly the endogenous AA profile.

In two previous experiments, we observed no major modification of the AA profile of the endogenous N losses in pigs fed different pea fiber fractions, with the exception of some dispensable AA, i.e., glutamic acid, glycine, proline and serine (Leterme et al. 1996b and 1998). These AA, together with aspartic acid, account for >50% of the basal losses measured with the N-free diet and have also the most variable excretion. Proline is hyperexcreted because of modification of the mucosal metabolism (Seve and Leterme 1997). This variation can significantly affect the total endogenous N excretion and thus the estimation of the individual AA excretion (Table 4).

The patterns of ileal excretion obtained by this method, based on the pattern of basal losses, are also quite different from those obtained with the ¹⁵N-AA dilution technique (Table 3). This may be due either to the heterogeneity in the profile of AA taken up from the arterial blood compared with those absorbed from the diet or to a modification in the proportions of the different endogenous N compounds, when the endogenous losses exceed the basal level. The results are thus essentially dependent on the method used.

The ¹⁵N-leucine infusion technique could be improved by a better control of methodology but the present results suggest that other parameters should be revised. The plasma amino acid N enrichment seems to be an inappropriate surrogate measure of the lumenal endogenous N enrichment, and the ¹⁵N of leucine can scarcely be representative of the whole endogenous N pool. Moreover, the distribution of ¹⁵N to other AA through transamination or some other path is too limited and excludes all of the important AA for the pig nutritionist (Lys, Met, Thr, Trp). The ¹⁵N tracer technique has undoubtedly high potential to improve our knowledge in protein nutrition, but further experiments are necessary to develop and improve the method, such as the use of multiple AA mixtures and the designation of an appropriate accessible reference pool.

	FOOTNOTES

Manuscript received 17 March 1998. Initial reviews completed 19 May 1998. Revision accepted 6 July 1998.

	ACKNOWLEDGMENTS

The authors are indebted to D. Tomé (Institut National Agronomique; Paris, France) for the use of the EA-IRMS of his department for ¹⁵N determinations. They also gratefully acknowledge J.-N. Thibault and Ph. Ganier (INRA, St Gilles, France) for the analysis of ¹³C and ¹⁵N by GC-C-IRMS. The experiment would have not be possible without the expert technical assistance of L. Givron, J.P. Haulotte and T. Monmart.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• Alpers D. Protein synthesis in intestinal mucosa: the effect of route of administration of precursor amino acids. J. Clin. Investig. 1972; 51:167-173
• Corring T., Jung J. The amino acid composition of pig pancreatic juice. Nutr. Rep. Int. 1972; 6:187-190
• de Lange C., Souffrant W., Sauer W. Real ileal protein and amino acid digestibilities in feedstuffs for growing pigs as determined with the ¹⁵N-isotope dilution technique. J. Anim. Sci 1990; 68:409-418[Abstract]
• François E., Thill N., Théwis A. Méthode rapide de dosage de l'oxyde de chrome dans les aliments, les feces et les contenus digestifs par titrage apres oxydation nitro-perchlorique. Ann. Zootec. 1978; 27:355-361
• Gabert V., Canibe N., Jorgensen H., Eggum B., Sauer W. Use of ¹⁵N-amino acid isotope dilution techniques to determine endogenous amino acids in ileal digesta in growing pigs. Acta Agric. Scand. Sect. A Anim. Sci. 1997; 47:168-177
• Gaudichon C., Laurent C., Mahé S., Marks L., Tomé D., Krempf M. Rate of ¹⁵N-leucine incorporation and determination of nitrogenous fractions from gastro-jejunal secretion in fasting humans. Reprod. Nutr. Dev. 1994; 34:349-359
• Grala W., Verstegen M., Jansman A., Huisman J., van Leeuwen P. Ileal apparent protein and amino acid digestibilites and endogenous nitrogen losses in pigs fed soybean and rapeseed products. J. Anim. Sci. 1998; 76:557-568[Abstract/Free Full Text]
• Hess, V., Thibault, J. N. & Seve, B. (1997) Apparent, true and real digestibility of intact or hydrolysed isolated pea protein. In: Proceedings of the 7th International Symposium on Digestive Physiology in Pigs ( Laplace, J. P., Février, C. & Barbeau, A., eds.), pp. 587-590. EAAP Publication no.88, St-Malo, France.
• Hoerr, A., Matthews, D., Bier, D. & Young, V. (1991) Leucine kinetics from [²H₃]- and [¹³C]-leucine infused simultaneously by gut and vein. Am. J. Physiol. 260 (Endocrin. Metab. 23): E111-E117.
• Husek P. Rapid derivatization and gas chromatographic determination of amino acids. J. Chromatogr. 1991; 552:289-299
• Leterme P., Froidmont E., Rossi F., Théwis A. The high water-holding capacity of pea inner fibers affects the ileal flow of endogenous nitrogen and amino acids. J. Agric. Food Chem. 1998; 46:1927-1934
• Leterme P., Théwis A., François E., van Leeuwen P., Wathelet B., Huisman J. The use of ¹⁵N-labeled dietary proteins for determining true ileal amino acid digestibilities is limited by their rapid recycling in the endogenous secretions of pigs. J. Nutr. 1996a; 126:2188-2198
• Leterme P., Théwis A., van Leeuwen P., Monmart T., Huisman J. Chemical composition of pea inner fibre isolates and their effect on the endogenous digestive secretions in pigs. J. Sci. Food Agric. 1996b; 72:127-134
• Lien K., Sauer W., Dugan M. Evaluation of the ¹⁵N-isotope dilution technique for determining the recovery of endogenous protein in ileal digestion of pigs: effect of pattern of blood sampling, precursor pools and isotope dilution technique. J. Anim. Sci. 1997a; 75:159-169[Abstract/Free Full Text]
• Lien K., Sauer W., Mosenthin R., Souffrant W., Dugan M. Evaluation of the ¹⁵N-isotope dilution technique for determining the recovery of endogenous protein in ileal digestion of pigs: effect of dilution in the precursor pool for endogenous nitrogen secretion. J. Anim. Sci. 1997b; 75:148-158[Abstract/Free Full Text]
• Ljunqvist, O., Persson, M., Ford, G. & Nair, K. (1997) Functional heterogeneity of leucine pools in human skeletal muscle. Am. J. Physiol.(Endocrin.& Metab. 36): E564-E570.
• Mantle M., Allen A. Isolation and characterization of the native glycoprotein from pig small intestinal mucus. Biochem. J. 1981; 195:267-275[Medline]
• Matthews D., Schwarz H., Yang R., Motil K., Young V., Bier D. Relationship of plasma leucine and -ketoisocarpoate during a L-[1-¹³C]leucine infusion in man: a method for measuring human intracellular leucine tracer enrichment. Metabolism 1982; 31:1105-1112[Medline]
• Rérat A., Jung J., Kandé J. Absorption kinetics of dietary hydrolysis products in conscious pigs given diets with different amounts of fish protein. 2. Individual amino acids. Br. J. Nutr. 1988; 60:105-120[Medline]
• Seve, B. & Leterme, P. (1997) Amino acid fluxes in the pig. In: Proceedings of the 7th International Symposium on Digestive Physiology in Pigs (Laplace, J. P., Février, C. & Barbeau, A., eds.), pp. 304-315. EAAP Publication no. 88, St-Malo, France.
• Simon, O., Zebrowska, T., Bergner, H. & Münchmeyer, R. (1983) Investigations on pancreatic and stomach secretion in pigs by means of continuous infusion of ¹⁴C-amino acids. Arch. Tierernähr. 33: 9-22.
• Souffrant W., Rérat A., Laplace J. P., Darcy-Vrillon B., Köhler R., Corring T., Gebhardt G. Exogenous and endogenous contributions to nitrogen fluxes in the digestive tract of pigs fed a casein diet. III. Recycling of endogenous nitrogen. Reprod. Nutr. Dev. 1993; 33:373-382
• Stoll B., Henry J., Reeds P., Yu H., Jahoor F., Burrin D. Catabolism dominates the first-pass intestinal metabolism of dietary essential amino acids in milk protein-fed piglets. J. Nutr. 1998; 128:606-614[Abstract/Free Full Text]
• van Leeuwen P., van Kleef D., van Kempen G., Huisman J., Verstegen M. The post-valve T-caecum cannulation technique: an alternative method for chyme collection in pigs. J. Anim. Physiol. Anim. Nutr. 1991; 65:183-193
• Vanvenrooij W., Kuijper-Lenstra A., Kramer M. Interrelationship between amino acid pools and protein synthesis in the rat submandibular gland. Biochim. Biophys. Acta 1973; 312:392-398[Medline]
• Yu, Y., Wagner, D., Tredget, E., Walaszewski, J., Burke, J. & Young, V. (1990) Quantitative role of splanchnic region in leucine metabolism: L-(1-¹³C, ¹⁵N)leucine and substrate balance studies. Am. J. Physiol 259 (Endocr. Metab. 22): E36-E51.

0022-3166/98 $3.00 ©1998 American Society for Nutritional Sciences