INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

UNLIKE IN THE ADULT, ARGININE SYNTHESIS occurs primarily in the neonatal small intestine; indeed, the net synthesis of arginine by the small intestine of the piglet has been demonstrated using arteriovenous differences across the gut (18, 23). Because of low pyrroline-5-carboxylate (P-5-C, EC no. not yet assigned) synthase activity in the neonatal small intestine, glutamate and glutamine are not significant precursors for arginine synthesis in the neonate (14, 22). As a result, proline is believed to be the primary dietary precursor for de novo synthesis of arginine by the gut. Indeed, in a previous experiment employing arginine deficiency-induced hyperammonemia as the primary outcome, we demonstrated that arginine and proline are coindispensable in intragastrically fed piglets (7). In that study, we concluded that 1) the neonatal piglet cannot synthesize sufficient arginine to maintain the urea cycle (and dispose of ammonia); 2) the piglet cannot synthesize sufficient proline to maintain plasma concentrations; and 3) arginine synthesis from proline is dependent on gut metabolism, which is diminished as a result of gut atrophy during parenteral feeding. Although these observations were convincing, they were indirect.

In another study (5), we demonstrated that the free pools of the P-5-C amino acids (arginine, ornithine, citrulline, proline, glutamate, and glutamine) were dramatically altered when identical diets were fed to piglets via the stomach, central vein, or portal vein. In particular, we observed dramatic changes in the ornithine pools in liver, small intestinal mucosa, kidney, and plasma. We hypothesized that gut atrophy due to parenteral feeding caused lower nitrogen retention because of lowered conditionally indispensable amino acid (i.e., arginine and proline) synthesis by the atrophied gut (4). Taken together, these results led to the present multi-isotope experiment, which was designed to demonstrate that proline and arginine interconversion via ornithine is dependent on first-pass metabolism by the gastrointestinal tract.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals and surgical procedures. Intact male Yorkshire piglets were obtained from the University of Alberta's swine unit at 1-3 days of age and 1.4-1.8 kg body wt. Piglets were removed from the sow and transported to the laboratory, where the piglets immediately underwent surgery to implant catheters. All procedures used in this study had been approved by the Animal Care Committee of the University of Alberta.

1

1

1

1

1

1

1

1

Constant tracer infusions. Ten piglets were blocked by body weight between the two treatments: IP- or IG-infused piglets. On the morning of day 5, arginine kinetics were determined by a primed [111 kBq (3 µCi)/kg], constant infusion [185 kBq (5 µCi) · kg¹ · h¹] of L-[guanido-¹⁴C]arginine. The urea product pool was also primed [463 kBq (12.5 µCi)/kg] with [¹⁴C]urea via the femoral catheter (n = 4); as a result of preliminary evidence of overpriming, this prime was reduced to 370 kBq (10 µCi)/kg for the remaining pigs (n = 6). The constant infusion continued for 6 h, with blood sampled every 30 min. On the morning of day 6, ornithine kinetics were determined by a primed [481 kBq (13 µCi)/kg], constant infusion [370 kBq (10 µCi) · kg¹ · h¹] of L-[U-¹⁴C]ornithine. Simultaneously, proline kinetics were determined by a primed [370 kBq (10 µCi)kg], constant infusion [370 kBq (10 µCi) · kg¹ · h¹] of L-[2,3-³H]proline. All priming doses were determined in a pilot experiment in similar pigs; specific radioactivity data after bolus isotope infusions were fitted to a monoexponential function, and turnover time (i.e., prime-to-constant ratio) was calculated (21).

Analytical procedures. Plasma amino acids and the specific radioactivity (SRA) of plasma glutamate, hydroxyproline, glutamine, citrulline, arginine, proline, and ornithine were measured by reverse-phase HPLC using phenylisothiocyanate derivatives as previously described (10). A 400-µl plasma sample was mixed with two internal standards: norleucine and L-[U-¹⁴C]phenylalanine (NEN Life Science Products, Boston, MA). Radioactive derivatives were collected postcolumn in 1-ml fractions; 5 ml of scintillant were added, and fractions were counted on a liquid scintillation counter by use of the dual-isotope counting protocol for ³H and ¹⁴C. Plasma urea concentrations were determined using a spectrophotometric assay kit (Sigma Chemical, St. Louis, MO); radioactivity associated with urea was determined by collecting the underivatized urea fraction by HPLC. The rate of ¹⁴CO₂ expiration was determined for each collection period by quantifying the total radioactivity collected in the ¹⁴CO₂ absorber with liquid scintillation counting of a 1-ml aliquot of absorber mixed with 5 ml of scintillant (Atomlight; Dupont Canada, Mississauga, ON).

Calculations. The following calculations were used for both plasma amino acids and urea. SRA was calculated as

SRA for plasma amino acids for each time point during the infusion study were plotted. Plateau values were calculated as the means of the plasma SRA at the time points within the plateau. Plateaus were visually determined and then verified using regression analysis by determining that the slope was not different from zero. All plateaus included 4 time points.

1

1

Statistical analyses. IG and IP treatments (means ± SD) were compared using a paired sample t-test (two-tailed) (version 6.02, Corel Quattro Pro, Ottawa, ON, Canada) and were considered significantly different if P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Piglet performance. The piglets remained healthy and were interested in the environment during the course of all protocols. The weight upon arrival (1.57 kg, pooled SD: 0.15), final weight (2.52 kg, pooled SD: 0.22), and rate of body weight gain (178 g/day, pooled SD: 25) was not significantly different between IG and IP treatments.

Fate of constant radiotracer infusions. In one of the pigs in the IP infusion group, the femoral sampling catheter was not patent during the [guanido-¹⁴C]arginine infusion protocol, and blood could not be collected; thus the IP group consists of four pigs for the arginine infusion data. However, the catheter was patent during the [¹⁴C]ornithine and [³H]proline infusion protocol, and these data included five pigs for both groups.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Typical specific radioactivity (SRA) plateau curves in intragastrically fed piglets after (A) primed, constant intragastric infusions of [14C]ornithine and [3H]proline or (B) a primed, constant intragastric infusion of [guanido-14C]arginine, including an overpriming intravenous dose of [14C]urea; [14C]urea overprime was resolved in remaining pigs by reducing bolus prime dose by 20% (not shown).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In our previous study, the IP infusion of diet was developed as a new and unique way to isolate small intestinal metabolism by maintaining hepatic first-pass metabolism (4). It is important to note that the plexus of anastomotic channels, which function as a ductus venosus in the fetal piglet, closes before birth (8), so that portal infusions via the umbilical vein are completely metabolized by the liver on first pass. To our knowledge, this method of in vivo separation of small intestinal metabolism has not previously been used to study amino acid kinetics. Thus, in the present study, we infused amino acid tracers into the portal vein or stomach, thus separating the respective absence or presence of intestinal first-pass metabolism. Unlike previous studies in which "splanchnic metabolism" was estimated by tracer infusion into a central vein rather than the stomach (2, 14), the present experiment reduced splanchnic metabolism to "gut metabolism," thereby excluding the substantial effects of hepatic metabolism. This more focused model is critical in establishing the role of the small intestine in P-5-C amino acid metabolism, because very different metabolic activities occur in the small intestine and the liver (11, 13). Therefore, using an in vivo model in intragastrically fed piglets, we have elucidated the role of small intestinal first-pass metabolism on P-5-C amino acid kinetics, including the net interconversions among proline, ornithine, and arginine.

This study has clearly demonstrated that small intestinal metabolism is necessary for the net conversion of proline to arginine. Indeed, no net conversion was evident when gut first-pass metabolism was bypassed, whereas 16% of arginine flux [42% (proline to ornithine) × 39% (ornithine to arginine), Table 2] or 48 µmol proline · kg¹ · h¹ (Table 4) were converted when gut first-pass metabolism was maintained. The net conversion of proline to arginine appears to occur during first-pass metabolism by the gut, because our models do not preclude the arterial extraction by the gut of precursors (i.e., proline) for the synthesis of arginine. Neonatal mammals are in a continuously fed state due to suckling, and so their dependence on first-pass metabolism of dietary precursors by the gut would not be a limiting scenario for the synthesis of conditionally indispensable amino acids.

Proline synthesis from arginine was also dependent on small intestinal metabolism, as indicated by the fractional net conversion differences for the pathway from ornithine to proline (Table 2). The fractional net conversion of arginine to ornithine (via arginase; EC 3.5.3.1), measured as arginine to urea as in Table 2, was not different between IP and IG infusions, probably because arginase activity is negligible in the neonatal piglet small intestine (6, 22) and is exceedingly abundant in the neonatal liver (12). However, the conversion of ornithine to proline appeared to be localized to the gut, because a 75% lower fractional net conversion was observed during IP infusion (Table 2). Indeed, the molar rate of conversion to proline was 41.5 µmol arginine · kg¹ · h¹ in IG infused pigs, which was reduced to 8.5 µmol arginine · kg¹ · h¹ when gut first-pass metabolism was bypassed (Table 4). The importance of proline in the rapidly growing piglet was demonstrated by the fractional net conversions to hydroxyproline. Although proline net conversion to hydroxyproline was not affected by route of isotope infusion, the finding that hydroxyproline SRA reached plateau rapidly during either [³H]proline or [¹⁴C]ornithine infusions indicates a very rapid turnover of collagen in the neonatal piglet. Because hydroxyproline synthesis is posttranslational and irreversible, this rapid collagen turnover must be a considerable proline "sink" in the neonatal pig; indeed, proline has been shown to be indispensable in the piglet (1). Furthermore, this proline loss may be extreme in situations that lead to increased collagen turnover, such as during wound or burn healing; such situations would increase proline requirements and should be quantified.

Several data in this study strongly support the conclusion that the synthesis of arginine from proline is more dependent on first-pass metabolism than is the synthesis of proline from arginine. In particular, the contribution of the gut to the fractional net conversion of ornithine to proline (75%, Table 2) was not as profound as that of proline to ornithine (100%, Table 2). In addition, the overall molar flux from proline to arginine (48.1 µmol proline · kg¹ · h¹, Table 4) was entirely dependent on the gut, whereas the arginine-to-proline molar conversion attributable to the gut amounted to 33.0 µmol arginine · kg¹ · h¹. More revealing are the arginine balance data (to and from ornithine), which indicate a large net synthesis of arginine (i.e., 40.9 µmol · kg¹ · h¹) when gut metabolism is included, and a large net catabolism of arginine (i.e., 29.1 µmol · kg¹ · h¹) when gut metabolism is bypassed. These data suggest a net synthesis of 70.0 µmol arginine · kg¹ · h¹ from ornithine during first-pass gut metabolism and a net degradation of arginine by the rest of the body. These data are in contrast to those for proline balance, when equal amounts of proline were synthesized from ornithine regardless of gut metabolism; indeed, the large degradation of ornithine by the gut (70.8 µmol · kg¹ · h¹) was entirely accounted for by the higher synthesis of arginine during gut metabolism. Despite the greater dependence of arginine synthesis on gut metabolism, it is obvious from the combined data that proline synthesis also relies on gut metabolism and that ornithine is central to these conversions.

One of the goals of our multiple tracer infusion protocol was to identify the limiting step in the pathway that results in arginine synthesis being dependent on gut metabolism. As shown in Fig. 2, P-5-C is the central metabolite for the interconversions among ornithine, proline, and glutamate. The pathway from ornithine to proline involves the two enzymes, ornithine aminotransferase (OAT, ornithine to P-5-C) and P-5-C reductase (EC 1.5.1.2, P-5-C to proline); the reverse pathway from proline to ornithine involves proline dehydrogenase (EC 1.5.99.8, proline to P-5-C) and OAT (P-5-C to ornithine). The limiting step in the pathway for proline synthesis from arginine was the net conversion from ornithine to proline (4.5 vs. 18%). Furthermore, the limiting step for arginine synthesis from proline was the net conversion of proline to ornithine (0 vs. 42%). The interconversion between ornithine and proline appears to be the pathway that is localized to the small intestine. Therefore, the gut-localized enzymes are either OAT or P-5-C reductase and proline dehydrogenase. Proline dehydrogenase is also the first enzyme in the pathway from proline to glutamate; however, despite the relatively high concentrations of proline dehydrogenase in the neonatal pig small intestine (17, 24), this net conversion was not dependent on the gut, because IP and IG infusions gave similar rates (22 vs. 18%, respectively, P = 0.20). However, OAT is also involved in the conversion from ornithine to glutamate, and this net conversion was lower when gut metabolism was bypassed (4.2 vs. 6.7%, P = 0.04). Consequently, we conclude that, in the neonatal pig, OAT must be predominantly localized to the small intestine, and this localization therefore explains the negligible net interconversion between arginine and proline when gut metabolism is bypassed.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Pathways demonstrating the localization of ornithine aminotransferase to the gut. Fractional net conversions from [3H]proline to [3H]ornithine and [3H]glutamate (solid arrows and boxes) and from [14C]ornithine to [14C]proline and [14C]glutamate (dashed arrows and boxes) with respective enzymes for each conversion pathway. P5C, pyrroline-5-carboxylate. Data (mean ± SD) were from intragastric (IG) or intraportal (IP) infusions of [3H]proline and [14C]ornithine in intragastrically fed piglets. Underlined values were significantly greater (P < 0.05).

It is important to note that this extrapolation of plasma amino acid kinetic data to intracellular enzyme activity assumes that equilibration is achieved among the various pools. Although true enzyme activity cannot be determined from such data, the net in vivo consequence of these activities can be inferred, because all amino acid precursors and products reached kinetic equilibrium in the plasma. Nevertheless, the present data add to previous studies by measuring the net interconversions among the amino acids associated with OAT activity with and without small intestinal first-pass metabolism. Indeed, although all of the enzymes involved in arginine and proline interconversion have been shown to be present in the small intestine of neonatal pigs (17, 22), OAT activity is present predominantly in the small intestine (24). The critical role of intestinal OAT for arginine synthesis has also been demonstrated in mice and humans with intestinal OAT deficiency; this deficiency results in hypoornithinemia, hypocitrullinemia, hypoargininemia, hyperammonemia, and even death during the neonatal period (20). Furthermore, gabaculine (an inhibitor of OAT) results in decreased plasma concentrations of ornithine, citrulline, and arginine by 59, 52, and 76%, respectively, in 4-day-old suckling piglets (9). The present interpretation of plasma kinetic data, suggesting the localization of enzyme activity to the gut, is supported by these other studies and provides important in vivo verification of in vitro evidence.

The relative differences in fractional net conversions when the gut is or is not bypassed clearly indicate that interconversions among arginine, ornithine, and proline are dependent on first-pass gut metabolism. The quantitative importance of these conversions on whole body metabolism can be extrapolated from the flux data. On first pass, the small intestine extracted 40% of dietary arginine and 15% of dietary proline; although ornithine was not provided in the diet, 51% of the oral dose was also extracted on first pass (Table 3). The fate of extracted arginine and proline within the small intestine is oxidation, protein synthesis, and conversion to other metabolites. For extracted ornithine, which is not a component of protein, its fate is oxidation and conversion. Ornithine oxidation was significantly higher during IP infusion, suggesting that ornithine oxidation occurs primarily in the liver. However, during IG infusion of ornithine isotope, 49% of the ornithine dose reaches the liver (i.e., 51% extraction), and 35.7% of that dose is oxidized (i.e., IP infusion oxidation rate). This calculation predicts an ornithine oxidation rate of 17.5% during IG infusion, which is lower than the rate observed (i.e., 25.1%). Therefore, ~7.6% of the ornithine dose must be oxidized within the small intestine. Of the ornithine extracted by the small intestine, ~15% was oxidized, and the remaining 85% must have been converted. The flux data also indicate that 40% of dietary arginine was extracted by the gut on first pass. Because arginine-to-ornithine conversion was not dependent on the gut, intestinal arginine oxidation via ornithine was negligible. Therefore, the majority of this extracted arginine must have been either converted or incorporated into protein. Stoll et al. (18) estimated that threonine was incorporated into piglet mucosa at a rate of 2.2 µmol · kg¹ · h¹. With a relative proportion of 3.4% of protein, this threonine incorporation involves ~4.4 µmol arginine · kg¹ · h¹ and 5.3 µmol proline · kg¹ · h¹, with the assumption that piglet protein includes 6.8% arginine and 8.2% proline (25). Thus protein synthesis accounts for 6% of the 78 µmol arginine · kg¹ · h¹ and 9% of the 57 µmol proline · kg¹ · h¹ extracted by the gut on first pass. Therefore, the primary fate of the substantial amounts of arginine, proline, and ornithine extracted by the gut is conversion to other amino acids, as well as to other minor products such as nitric oxide, polyamines, and creatine.

In combination with the fractional conversion differences between routes of infusion, we conclude that interconversion among the P-5-C amino acids is dependent on first-pass metabolism by the piglet small intestine and amounts to a significant proportion of whole body fluxes of these amino acids. Indeed, of the 57 µmol proline · kg¹ · h¹ extracted by the gut from the diet, an estimated 84% (i.e., 48.1 µmol · kg¹ · h¹, Table 4) is converted to arginine; similarly, ~42% (i.e., 33.0 µmol · kg¹ · h¹, Table 4) of the extracted 78 µmol arginine · kg¹ · h¹ is converted to proline. In total, the gut releases 48.1 µmol of synthesized arginine and 117 µmol · kg¹ · h¹ of dietary arginine, which is in excess of the recommended arginine requirement for piglets of 96 µmol · kg¹ · h¹ (i.e., 2.3 mmol · kg¹ · day¹) (15). As a comparison, the gut releases 33 and 348 µmol · kg¹ · h¹ of synthesized and dietary proline, respectively, compared with an estimated requirement of 329 µmol · kg¹ · h¹ (i.e., 7.9 mmol · kg¹ · day¹) (1). In other words, the gut synthesizes 50% of the arginine required by the piglet but only 10% of the proline requirement. Therefore, the de novo synthesis of arginine from proline is not only more dependent on the gut, it is also more important in relation to whole body supply and requirement.

Because ornithine is not a component of protein or the diet, the only input to ornithine flux is de novo synthesis. However, there appears to be a substantial amount of ornithine synthesis (i.e., flux) that is not accounted for by molar inputs from arginine and proline (i.e., 149 µmol · kg¹ · h¹ in IG and 76 µmol · kg¹ · h¹ in IP). It is possible that some ornithine is synthesized from glutamate and glutamine via P-5-C synthase and OAT. However, in vitro data suggest that P-5-C synthase activity is negligible in intestinal tissues (14, 22) and thus probably does not account for the 55% of ornithine synthesis that is missing. Alternatively, it is probable that ornithine plasma flux estimates do not reflect all aspects of ornithine metabolism (2) and can be overestimated if corrections for recycling are not employed (19). For example, the conversion of ornithine to citrulline was over 100% of citrulline flux (Table 2), which suggests either tracer recycling, possibly via nitric oxide synthase, which bypasses ornithine, or more likely, preferential conversion to citrulline during first-pass intestinal metabolism (2). These data suggest that absolute flux rates and associated stochastic calculations need to be used with caution, given the inappropriateness of an imposed single-pool model to complicated amino acid metabolism. In the present study, the relative conversions, which are independent of flux data, provide the primary basis for our conclusions.

In conclusion, the data in this study support our previous findings that arginine and proline are coindispensable in gastrically fed piglets (7). In situations in which first-pass gut metabolism is bypassed or compromised (i.e., TPN, gut resection, intestinal disease, or weaning-induced gut stress), arginine and proline are individually indispensable because their biosyntheses are negligible. Therefore, the requirements for arginine and proline need to be established for parenterally fed or gut-stressed neonates. Because the net synthesis of arginine from proline was more dependent on small intestinal metabolism than the reverse net conversion, we speculate that the increase in arginine requirement in such neonates will be greater than the increase in proline requirement. There is a need to quantify the synthetic capacity of the piglet as well as the minimum dietary requirement. Because of their coindispensability, the arginine and proline requirements need to be determined together; then the maximum ratio of arginine to proline (and vice versa) should be determined, similar to the way requirements are expressed for phenylalanine/tyrosine or methionine/cysteine. Because of the role of the gut in arginine and proline metabolism, in situations of gut stress or bypass (i.e., parenteral nutrition), the arginine and proline "requirement ratio" will be very different from that in healthy orally fed neonates and should be determined separately.