QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ball, R. O. Articles by Pencharz, P. B. Search for Related Content PubMed PubMed Citation Articles by Ball, R. O. Articles by Pencharz, P. B.

---

Supplement: 6^th Amino Acid Assessment Workshop: SESSION 2

Nutritional Consequences of Interspecies Differences in Arginine and Lysine Metabolism^1–3,

⁴ Department of Agricultural, Food and Nutritional Science, University of Alberta, Edmonton, AB, Canada T6G 2P5; ⁵ The Research Institute, The Hospital for Sick Children, Toronto, ON, Canada M5G 1X8; and the ⁶ Departments of Paediatrics and Nutritional Science, University of Toronto, Toronto, ON, Canada M5G 1X8

^* To whom correspondence should be addressed. E-mail: ron.ball{at}ualberta.ca.

	ABSTRACT

TOP ABSTRACT LITERATURE CITED

 
Differences in lysine and arginine requirements among various species such as omnivores (humans, pigs, rats, dogs), carnivores (cats), herbivores (rabbits, horses), ruminants (cattle), poultry, and fish, are covered in detail in this article. Although lysine is classified as an indispensable amino acid across species, the classification of arginine as either an indispensable or dispensable amino acid is more ambiguous because of differences among species in rates of de novo arginine synthesis. Because lysine is most often the limiting amino acid in the diet, its requirement has been extensively studied. By use of the ideal protein concept, the requirements of the other indispensable amino acids can be extrapolated from the lysine requirement. The successful use of this concept in pigs is compared with potential application of the ideal protein concept in humans. The current dietary arginine requirement varies widely among species, with ruminants, rabbits, and rats having relatively low requirements and carnivores, fish, and poultry having high requirements. Interspecies differences in metabolic arginine utilization and reasons for different rates of de novo arginine synthesis are reviewed in detail, as these are the primary determinants of the dietary arginine requirement. There is presently no dietary requirement for humans of any age, although this needs to be reassessed, particularly in neonates. A thorough understanding of the factors contributing to the lysine and arginine requirements in different species will be useful in our understanding of human amino acid requirements.

Definition of the scope and terms used in this article

The discussion in this article focuses primarily on interspecies differences in metabolism during growth because this is where the most data are available. Dietary requirements at maintenance are also discussed, particularly in species that spend the majority of their lifespan at maintenance, specifically humans, dogs, and cats. In the case of dietary indispensable amino acids, e.g., lysine, the dietary amino acid requirement is equivalent to the metabolic amino acid requirement. In this article, the use of the term metabolic amino acid requirement represents the obligatory amino acid demand defined by Millward (1), which represents the amount of amino acid needed to support net protein synthesis, obligatory levels of oxidation, and the synthesis of nonprotein products, including carnitine for lysine, and creatine, nitric oxide, polyamines, and urea for arginine. Therefore, there is a metabolic, but not necessarily a dietary, requirement for all amino acids. This point is emphasized because it is particularly important with regard to arginine metabolism and variation in the dietary requirement both among and within species.

The criteria used to define amino acid requirements vary among species and make comparisons more difficult. The new dietary reference intake system in humans uses the estimated average requirement (EAR; the median requirement for the nutrient),⁷ the recommended dietary allowance (RDA; nutrient intake level that meets the requirements of 97% of the population), or adequate intake (AI; intake that appears to be adequate when not enough information is available to calculate an RDA) to express amino acid requirements (2). In most animals, the requirements provided by the National Research Council (NRC) publications are mean population requirements, which is roughly equivalent to the EAR in humans. The recently released dog and cat requirements (3) use a system of expressing requirements that more closely resembles the human RDA, provides a recommended allowance that is greater than the minimum requirement, takes into account the fact that nitrogen balance is relatively insensitive in adult animals and can occur over a range of intakes, and considers that the bioavailabilities of amino acids vary among feed ingredients. The bioavailability of dietary amino acids is considered in most animal species [see Nutrient Requirements of Swine (4) for examples of application of amino acid bioavailability] except humans. These differences in expression of dietary requirements are often a source of confusion and misinterpretation when species are compared because many readers are not aware of these differences.

Finally, the metabolic uses and specifics of metabolism of lysine and arginine have been thoroughly discussed, including other reviews in the present Supplement. Therefore, the species differences are discussed with the assumption that the reader has an understanding of the details of lysine and arginine metabolism.

Interspecies differences in lysine metabolism and requirements

Lysine is a dietary indispensable amino acid in all species that have been studied. It is usually the most limiting dietary amino acid for body protein synthesis, which explains why there is such a vast quantity of literature, compared with other amino acids, on lysine requirements in different species. In growing animals, the lysine requirement accounts for 3.6% (broiler breeders, 3–6 wk) to 6.1% (rats) of the recommended crude protein (CP) intake (5,6). Human requirements during growth are also within this range with the exception of the newborn, where a lysine AI may be slightly higher at 7.0% of the AI of protein (2). During maintenance, the lysine requirement (as a percentage of the CP intake requirement) varies from 1.7% in cats (3) to 4.8% in humans (2).

Because lysine is the first limiting amino acid in most grain- and cereal-based animal diets, it also defines the protein intake required to meet the animals' amino acid requirements. The protein requirement published for most species is therefore actually the protein intake required to satisfy the dietary need for lysine. Comparisons among species show that, on a CP basis, the requirements for lysine are remarkably similar. Implications of this are discussed below.

The importance of lysine as the first limiting amino acid has been used in the development of the ideal protein concept: the expression of the requirement of all the amino acids relative to the lysine requirement. The concept of ideal protein is that there is an optimal pattern of dietary amino acids that corresponds to the amino acid requirements of the animal. This is a fundamental concept in animal nutrition and has been found to apply to all species in which it has been tested. In growing mammals, the ideal amino acid pattern determined experimentally has been found to be similar to the amino acid patterns in milk protein and body tissue protein within that species. The primary body protein is muscle, and muscle is structurally, biochemically, and compositionally similar across species (although there are many types of muscle, when the amino acid composition of all the muscle in the mammalian body is compared, the values are strikingly similar across species), including nonmammalian species (Table 1) (2,4,7,8). Although there are some differences among species, there are many more striking similarities. This is one of the reasons that the lysine requirement expressed per mass CP and the ideal amino acid pattern are also similar across many mammalian species.

Although the primary use of dietary lysine is for body protein synthesis, the metabolic requirement for lysine also includes carnitine synthesis and obligatory oxidation. The most important role of carnitine is in transporting long-chain fatty acids into the mitochondria for ß-oxidation and subsequent energy production via the citric acid cycle. Carnitine biosynthesis involves the methylation of protein-bound lysine, after which the carnitine moiety is released from the protein [as recently reviewed by Vaz and Wanders (13)]. Although carnitine serves an important metabolic role, data from rats indicate that <1% of dietary lysine is actually converted to carnitine when lysine is not limiting in the diet (14,15). However, this study also found that with a diet limiting in lysine, there was a decline in extrahepatic levels of carnitine (14); therefore, although carnitine synthesis uses only a small amount of dietary lysine, it is influenced by dietary lysine intake. Because carnitine is not found in vegetarian diets and is lower in milk and eggs than in meat (13), humans consuming vegan and lactoovovegetarian diets had lower levels of circulating carnitine than omnivorous humans (16). Furthermore, the humans consuming vegetarian-type diets also had carnitine levels that were very highly correlated to plasma lysine concentrations, suggesting that when carnitine is not provided in the diet, there is a greater dependence on endogenous carnitine synthesis from lysine. Therefore, in domestic animals and humans consuming primarily grain-based diets, the importance of dietary lysine intake for carnitine synthesis cannot be overlooked.

The obligatory oxidation of dietary lysine, an important component of the lysine requirement, responds to deficiency differently than other amino acids. A study by Moehn et al. (17) in growing pigs found that even when the dietary intake of lysine was well below (60–90%) its requirement for maximum protein synthesis, the rate of lysine oxidation was relatively stable until lysine intake was 60% or less of the required intake (17). This is contrary to the results observed for other dietary indispensable amino acids, whose degradation/oxidation declines, often to the basal or obligatory rate, when intake is below requirement (18). We have recently demonstrated in vitro (19,20) that lysine oxidation is relatively constant under a variety of conditions and also occurs in several different tissues of the pig, including a physiologically significant level of oxidation in the intestine. The implication of these results for nutritionists is that a slight decrease in lysine intake below requirement is not readily compensated for by decreasing oxidation. Therefore, protein synthesis by the animal is relatively more sensitive to a deficiency of lysine than of many other amino acids, where a decrease in intake below requirement results in a decrease in catabolism of the amino acid. So far, these results have been observed only in pigs and poultry (21), and research in other species is required to determine whether this is a unique or widespread phenomenon.

Interspecies differences in arginine metabolism and dietary requirements

The metabolic arginine requirement is primarily influenced by demand for protein deposition, as for all amino acids. However, different species have different capacities for endogenous arginine synthesis, and therefore, the dietary arginine requirement varies widely among species during growth, ranging from 1.4% of CP intake in 100-kg swine (4) to over 5% in young broiler chicken (5) and Pacific salmon (22). Therefore, the effects of dietary arginine intake, from deficient to excess, on endogenous synthesis are also highly species dependent. A representation of the metabolic requirement for arginine and its partitioning into dietary contribution and endogenous synthesis is shown in Figure 1. The metabolic requirement is the total use of arginine for all functions, net of recycling. This representation demonstrates the relation between maximum and minimum rates of endogenous synthesis and dietary requirement. This representation assumes that there is a minimum obligatory arginine synthesis rate, although this has not been measured or looked for in most species. This minimum also includes the probable contribution of arginine synthesized by intestinal microflora and absorbed by the animal, which also needs to be quantified. The difference between metabolic requirement and the rate of maximum endogenous synthesis corresponds to the minimum dietary arginine intake that is required. This is the intake of arginine identified in most requirement experiments. However, additional dietary intake of arginine will spare the utilization of energy and nitrogen for arginine synthesis by the amount that equals the difference between the maximum and minimum endogenous synthesis rates. We studied the metabolic response to arginine intake in piglets over the range from deficient to more than adequate and showed that there was a slow but steady decline in plasma ammonia with intakes greater than the minimal dietary requirement (23). Another consideration is that in many species, such as piglets as a example, there is a minimum arginine intake for survival (24) that is less than the minimum dietary requirement. For example, enterally fed piglets receiving 0.20 g·kg^–1·d^–1 of arginine were able to survive, but they had elevated plasma ammonia as well as urea concentrations and low plasma arginine concentrations despite maximum rates of endogenous arginine synthesis (24). Therefore, to normalize all of these parameters, the minimum dietary arginine requirement must be greater than the minimum arginine requirement for survival. Investigation in clinical and disease states of possible metabolic advantages to increasing arginine intake above the minimum dietary requirement is worth considering. Quantification of all these levels in the different species would aid greatly in completing our understanding of arginine metabolism.

View larger version (32K): [in this window] [in a new window]   FIGURE 1  Partitioning of the metabolic and dietary arginine requirement in animals.

View larger version (7K): [in this window] [in a new window]   FIGURE 2  Metabolic responses to dietary arginine intake relative to changes in protein synthesis. Solid lines: 3 possible response patterns to arginine intake for arginine use in other metabolic functions (i.e., nitric oxide synthesis), based on the assumption that they demonstrate an inflection at the same intake as protein synthesis. Dashed lines: Other possible response patterns to arginine intake for arginine use in other metabolic functions, where the arginine intake required for the maximum response in arginine use is different from the requirement necessary for maximum protein synthesis. Many combinations of different slopes and inflection points exist as described in the text. In addition, some responses, such as plasma ammonia and urea concentrations (see text), decrease with increasing arginine intake.

    Arginine requirements and metabolism in pigs. Arginine is considered a conditionally indispensable amino acid in pigs. There is a dietary requirement in the neonate, but in the healthy adult, endogenous synthesis is adequate to meet all metabolic requirements (28). However, there is recent experimental evidence that in some physiological instances arginine supplementation may be beneficial. Arginine intake from sow's milk, which has been estimated at 0.42 g·kg^–1·d^–1 in wk-old piglets (29), is similar to the estimated dietary arginine requirement of 3- to 5-kg suckling piglets (0.35 g·kg^–1·d^–1) (4). However, the factorial estimate of the metabolic arginine requirement of wk-old piglets is 1.1 g·kg^–1·d^–1 (29), and the arginine requirement of parenterally fed piglets, where de novo arginine synthesis is very low (25,30), is 1.2 g·kg^–1·d^–1 (23). Therefore, to meet the entire metabolic arginine requirement, there must be substantial reliance on de novo arginine synthesis in suckling piglets. This was confirmed in neonatal piglets given an injection of gabaculline, an inhibitor of ornithine amino transferase (OAT; E.C. 2.6.1.13), where there were sharp declines in plasma arginine, citrulline, and ornithine concentrations and an increase in plasma ammonia concentration (31). A study by our group (24) found that the maximal rate of arginine synthesis, determined in enterally fed piglets receiving the lowest intake of dietary arginine that could be given without inducing hyperammonemia (0.20 g·kg^–1·d^–1), was 0.68 g·kg^–1·d^–1. Adding this to the estimated intake in sow's milk (24) gives a value not different from the estimated total metabolic requirement of arginine (0.42 + 0.68 = 1.1 g·kg^–1·d^–1). Despite the fact that dietary intake and de novo synthesis appear to be adequate to meet the metabolic arginine requirement in young piglets, piglet growth was reported to be enhanced by feeding a milk replacer with additional arginine (32). Therefore, endogenous arginine synthesis and normal milk intake may not be sufficient to meet the total metabolic requirement for arginine in piglets. In this arginine supplementation study, however, the arginine intake in the control diet (0.50 g·kg^–1·d^–1) was based on the author's measured arginine content in sow's milk, which was reported as 7.69 g/kg dry matter (32). However, in a separate study, the ileal digestible arginine content of sow's milk on a dry matter basis was found to be substantially greater at 12.4 g/kg (33). It is therefore possible that the reason for the growth response to the supplemental arginine in the previous study reflected an abnormally low arginine intake in the milk received by the control group of piglets (32). Additional research on arginine content of sow's milk and potential benefit of supplementation in neonates is required to confirm these results. Supplementation of the gestation diet of sows with 1% arginine was recently shown to increase the number of piglets born alive by upward of 2 piglets (34,35), presumably through improved placental transfer of nutrients as a result of the role of arginine as a precursor for nitric oxide and polyamines. The implications of these data with respect to neonatal and gestating humans should be considered. Additional discussion may be found in the section on Implications of Interspecies Comparisons for Human Nutrition.

In growing pigs, the dietary arginine requirement decreases as the pigs get older and larger, with 4-kg piglets requiring 2.4% of their CP intake as arginine and 100-kg pigs requiring only 1.4% of CP as arginine (4). Arginine is considered a dispensable amino acid following puberty and during gestation in breeding females based on nitrogen retention, urinary urea, citrate and orotate excretion, and plasma urea concentrations (28). When the arginine requirement of pigs over 20 kg, per kilogram per day, is plotted against whole-body protein accretion, also per kilogram per day, there is a decrease in the dietary arginine requirement with increased pig size despite the large increase in rate of protein accretion (Fig. 3). Although endogenous arginine synthesis has been measured in wk-old (2.5-kg) piglets (24,36,37), it has not been measured for any other age of pig. Therefore, there is no definitive evidence whether the decreasing dietary arginine requirement results from an increase in endogenous arginine synthesis or a decrease in the metabolic arginine requirement. However, based on Figure 3, an increase in endogenous arginine synthesis probably explains the decrease in dietary requirement with weight and age, despite the increases in protein accretion.

View larger version (7K): [in this window] [in a new window]   FIGURE 3  The relation among dietary arginine requirement, pig body weight, and the rate of protein accretion in pigs from 20 to 100 kg. Protein accretion (g/d) was calculated using protein accretion = [0.47666 + 0.02147*body weight + 0.00023758*(body weight)2 + 0.000000713*(body weight)3] x 127.5 (4).

The enzymes necessary for glutamate, glutamine, and proline to be precursors for arginine synthesis have been detected in vitro (38,39), although in vivo research by our group has shown that proline is the major arginine precursor in wk-old piglets and not glutamine/glutamate (24,25). A diet containing ample amounts of glutamate, but not proline, was unable to prevent hyperammonemia in either enterally or parenterally fed piglets receiving an arginine-free diet (25). Furthermore, when a radioactive glutamate isotope was intragastrically infused into enterally fed piglets receiving either an arginine-deficient or generous diet, none of the label was recovered in arginine (24). The enzyme alanine aminotransferase (E.C. 2.6.1.2), an enzyme that can convert glutamate and pyruvate to alanine and -ketoglutarate, was present in the intestinal mucosa of piglets of all ages (42), and the small intestinal mucosa was also able to oxidize -ketoglutarate. Furthermore, in enterally fed piglets, only 5% of an intragastrically administered glutamate isotope appeared in the portal blood, and 50% of this dietary glutamate was oxidized to CO₂ (43). This explains why glutamate was a poor arginine precursor in enterally fed piglets. We have also shown in vivo that in wk-old enterally fed piglets citrulline synthesis is limiting to whole-body arginine synthesis because citrulline addition was as effective as arginine addition, whereas ornithine or proline were not (37). Both citrulline and arginine addition to the arginine-deficient diet spared the use of proline for arginine synthesis (37), demonstrating in vivo that the partitioning of proline use for arginine synthesis, vs. other activities, is dependent on the demand for endogenous arginine synthesis. Metabolism in growing pigs is similar to that in the young pig; in growing pigs receiving an arginine-deficient diet, citrulline was more effective than ornithine at promoting pig growth and efficiency (44).

To verify the enzyme and in vitro work, we investigated the sites of arginine synthesis in wk-old piglets and found that, in vivo, 40–60% of whole-body arginine synthesis occurred with first-pass intestinal metabolism, regardless of arginine intake (24). First-pass hepatic metabolism does not contribute to whole-body arginine synthesis in the neonate (36). The remaining 40–60% of whole-body arginine synthesis was from the metabolism of circulating precursors (36), and we have experimental evidence showing that the intestinal metabolism of arterial proline may be another major contributor to whole-body arginine synthesis (30). From our in vivo isotope data, it was not possible to conclude that first-pass intestinal metabolism or the intestinal metabolism of circulating precursors were responsible for the entire conversion of proline to arginine, only that they were necessary for the conversion. Intestinal metabolism is critical in the neonate for ornithine synthesis from proline (39,45). However, it is possible that renal or extraintestinal metabolism may also be involved in the remainder of the arginine synthetic pathway, even in wk-old piglets. This requires experimental confirmation.

    Arginine requirements and metabolism in rats. Much of the pioneering research relating to arginine and urea cycle metabolism was conducted in rodents, and many review articles summarize the results of rodent research (46–49); however, there are a few instances and some information that is different or unique in arginine metabolism and requirements in rats compared with other species. When rat nutrient requirements, including amino acids, are compared with those of other species, it must always be considered that rats practice coprophagy (50,51). Coprophagy provides an unknown amount of microbial amino acid to the diet, with the effect that most dietary studies in rats produce an underestimate of actual dietary requirement.

Arginine is classified as an indispensable amino acid for rats during growth (52) but not for adult rats (53). However, in adult rats formerly subjected to protein malnutrition, increasing increments of dietary arginine resulted in a positive linear response with regard to weight gain and nitrogen retention and a negative linear response for urinary orotate (54); therefore, in some cases arginine may be indispensable for mature rats. Although this research has implications for clinical treatment of humans, this effect has not been tested, to our knowledge, in humans or other species.

The dietary arginine requirement for growing rats is based on a crude protein intake of 15% (6). As in other species, the arginine requirement for tissue accretion and normal levels of metabolite excretion in rats increases as the protein content of the diet increases (55). This is likely because of arginine's role in the urea cycle and the fact that increasing levels of dietary protein have been shown to increase the activity of urea cycle enzymes (56,57). Therefore, the protein content of the diet, particularly the excess amino acid intake, must also be taken into account when determining if arginine intake is adequate.

Growing rats receiving an arginine-free or arginine-deficient diet have elevated urinary orotate, citrate, and urea excretion (55,58,59) and decreased nitrogen retention (58,60). It is of interest, however, that improvements in nitrogen retention, in response to supplemental arginine, appear to occur at a lower intake of arginine than changes in the metabolite (such as citrate, urea, and orotic acid) excretions (59). Furthermore, in adult, mature, female nonpregnant rats, although there was no weight change in response to feeding an arginine-free diet, there was an 27-fold increase in orotate excretion and an 8-fold increase in citrate excretion relative to rats receiving a control, arginine-containing, diet (60). We have shown in piglets that there was a breakpoint in plasma urea and ammonia with increasing arginine intake but that these continued to decline at a rate significantly different from zero until the entire metabolic requirement (1.2 g·kg^–1·d^–1) was provided by the diet (23). Based on these findings, it is clearly inappropriate to define arginine as dispensable or indispensable based on weight gain and nitrogen retention alone because metabolic perturbations may continue to exist even when growth and/or nitrogen balance is apparently maximized.

The capacity of growing rats to endogenously synthesize arginine for use in tissue accretion was demonstrated over 70 y ago (61), when it was discovered that the amount of arginine accreted in tissue was 2- to 3-fold greater than the amount provided in the diet. However, to our knowledge, the actual in vivo arginine synthesis rate has never been quantified in rats under various conditions. Although the total metabolic arginine requirement does not appear to have been studied in rats, it is clearly greater than the dietary requirement (2.9% of CP) (6), as illustrated by the fact that a diet containing 1% arginine (over twice the dietary recommendation) was still not enough to support optimal rat growth in the absence of endogenous synthesis (62). The rat was the main experimental animal model used to define the intestinal-renal axis of endogenous arginine synthesis in weaned mammals. The perfused rat intestine shows a substantial uptake of luminal and circulating glutamine and glutamate (63) and a release of citrulline (63). The intestine was shown to be the primary organ responsible for citrulline production (64,65) but was unable to metabolize the citrulline to other metabolites (64). The kidney was identified as the organ responsible for the majority of whole-body arginine synthesis in vivo in several different perfusion (64), mass balance (66), and ligation (67) studies. The release of arginine by the kidney was directly related to citrulline uptake by the kidney (66); therefore, if circulating citrulline concentrations were increased, there was a subsequent increase in renal arginine release (66). In rats, unlike piglets (24), arginine intake did not affect the renal release of arginine (68).

The endogenous synthesis of citrulline by the rat intestine appears to be critical for the growth of rats. Citrulline, unlike arginine (69) and ornithine (70), was not extensively metabolized by the liver or the intestine (64); therefore, supplemental citrulline provided in an arginine-free diet fed to rats resulted in an increase in circulating citrulline and arginine concentrations (70). Moreover, when the intestinal activity of ornithine transcarbamoylase (OTC; E.C. 2.1.3.3) was inhibited, using an infusion of a glycyl-glycyl derivative of -N-(phophonacetyl)-L-ornithine, the rats lost weight or gained less than when OTC activity was present (62). The growth retardation was only partially reversed by the addition of 1% arginine to the casein hydrosylate diet and was completely reversed by adding 1% citrulline (62). The main implication of this is that, in rats, the endogenously synthesized arginine is the primary source of arginine used for growth, whereas the dietary arginine seems to be used mainly as a precursor for intestinal ornithine or citrulline formation. The partitioning of the metabolism of endogenously synthesized vs. dietary arginine in other species still requires investigation. Interestingly, ornithine addition to an arginine-deficient growing rat diet supported the same growth rates and plasma ammonia and urea concentrations as either citrulline or arginine addition (62,70), but plasma arginine concentrations were still lower than if either citrulline or arginine were added to the diet (70). Therefore, in growing, postweaning rats, if there is the capacity for endogenous arginine synthesis, then growth and urea cycle function can be maintained by the addition of ornithine, citrulline, or arginine to an arginine-free diet (62,70,71). This is in contrast to other species, where citrulline addition is clearly advantageous compared with ornithine addition. The importance of proline as an arginine precursor and its ability to spare a portion of the dietary arginine requirement has not been investigated in rats.

    Arginine metabolism and requirements in cats. Cats were unable to synthesize enough ornithine, and therefore citrulline, to satisfy their metabolic arginine requirement, resulting in the rapid onset of hyperammonemia when arginine-free diets were fed (26,72). On a per-kilogram body weight basis, the small intestinal mucosa of cats only had 5% of the pyrroline-5-carboxylate (P5C) synthase (E.C. number not assigned) activity (73) of the rat intestinal mucosa. Furthermore, the activity of OAT per gram mucosa in cats was only 23% of the activity in rats (73). Perfusion studies in rats showed that the primary site of citrulline synthesis was the small intestine (63,64), and studies in pigs found that P5C synthase activity is exclusively small intestinal and OAT activity is primarily small intestinal (74). The very low intestinal activities of these enzymes in cats (73) results in very low de novo synthesis of ornithine, and subsequently citrulline, compared with other species. These differences in enzymes explains why arginine is an indispensable amino acid in cats of all ages. In addition to not being able to synthesize ornithine from glutamate, cats also have low intestinal alanine aminotransferase activity compared with rats (58%) (73). Therefore, unlike other mammals, cats cannot tolerate diets that are high in glutamate (75), likely because of an inability to convert glutamate nitrogen to alanine nitrogen.

Compared with other species, cats also have very low renal arginine synthesis from citrulline: 86 µmol/d (76) vs. over 350 µmol/d in rats fed a standard diet (66). The low plasma citrulline concentrations in cats (>10 µmol/L) (76) compared with rats, rabbits, or pigs (60–120 µmol/L) (44,76), explains the limited amount of renal arginine synthesis from citrulline in cats (76). For example, when kittens were fed an arginine-free diet with supplemental citrulline, plasma citrulline concentrations were 250 µmol/L, and plasma arginine concentrations were not different from those in kittens receiving an arginine-sufficient diet (77). These observations demonstrate an increase in renal arginine synthesis with increased plasma citrulline concentration (77), which is in agreement with rat research where the plasma citrulline concentration increased 3.9 times and the rate of renal arginine release increased 3.4 times (66). Therefore, in cats the primary limitation in de novo arginine synthesis is the low levels of intestinal enzymes needed for ornithine synthesis, which limits citrulline formation and circulating citrulline concentrations and subsequently renal arginine synthesis.

Cats must, as a result of lack of de novo synthesis, rely on their diet to meet the arginine requirement, which ranges from 0.54 g·kg^–1·d^–1 (4.3% of CP) in kittens to 0.12 g·kg^–1·d^–1 (3.9% of CP) in adult cats (3). The implication of this is that the dietary arginine requirement of the cat also closely represents the total metabolic requirement: the amount of arginine used for protein synthesis, support of urea cycle function, for the synthesis of other metabolites including creatine, polyamines, and nitric oxide (4,27), and for basal arginine catabolism. This is in contrast to other species where de novo synthesis provides significant, and often variable, proportions of the total metabolic requirement. This also means that, in comparison to other noncarnivorous ureotelic species, cats have a very high dietary arginine requirement. The high protein intake of the cat requires a high rate of nitrogen excretion and thus also contributes to the high dietary requirement for arginine. In the cat, the hepatic activities of most enzymes involved in amino acid degradation are constitutively high and unaffected by diet (78), relative to other species, meaning that the urea cycle must function at a high rate to detoxify the resulting ammonia. The effect of arginine intake on hepatic or renal arginase activity has not been quantified; however, in cats, unlike in rats (56), the hepatic activity of arginase was not affected by dietary protein intake (78). However, the hepatic activities of the urea cycle enzymes (OTC, ASS plus ASL, and arginase) in the liver of cats were 25–50% of the activities in the livers of rats fed a 90% protein diet (56,78) and therefore may more closely resemble the hepatic activity in rats receiving 40–50% protein diet. Because protein intake does not affect urea cycle enzyme activities in cats, arginine may need to be provided at a greater percentage of the total crude protein intake in cats receiving a low dietary protein intake compared with those fed a high-protein diet, to maintain the high basal level of urea cycle function. With high protein intakes, however, it is likely that arginine intake from the diet (assuming the diet has a balanced amino acid composition) will be adequate to support urea cycle function and that rate of urea cycle function would be similar to that of other mammals receiving a high-protein diet.

Although proline is a major precursor for arginine synthesis (10), to the best of our knowledge, the expression of proline oxidase (E.C. 1.5.99.8) has not been investigated in the cat. We predict that even if there is measurable activity of this enzyme in feline tissues, proline is unlikely to make a large contribution to whole-body arginine synthesis in cats because of the low OAT activity relative to other species.

    Arginine requirements and metabolism in dogs. Both growing (27,79) and adult (80) dogs have dietary requirements for arginine (0.36 g·kg^–1·d^–1 in young puppies and 0.06 g·kg^–1·d^–1 in mature dogs; 3.5% of CP in both ages of dogs) (3) that are intermediate to the requirements of the cat and the rat. Unlike cats, dogs are not obligate carnivores, and this may partly explain the lower arginine requirements in dogs than cats. To the best of our knowledge, the activities of the urea cycle/arginine synthetic enzymes have not been studied in dogs. The activities of these enzymes, particularly P5C synthase, have been suggested by others (81) to be intermediate between those of the cat and the rat, thereby explaining why their arginine requirement is also intermediate. This requires experimental confirmation.

    Arginine requirements and metabolism in rabbits and horses. Dietary arginine is necessary in growing rabbits for maximum growth and feed efficiency (82,83), and estimates of the arginine requirement are between 0.6 (82) and 1.23% (83) of diet. In comparison to other species, there has been little research on arginine requirements and metabolism in rabbits. Factors believed to affect the arginine requirement in rabbits included the crude protein content of the diet (increasing arginine requirement with increasing protein intake) (83) and the dietary concentrations of indispensable amino acids (increased arginine requirement when higher levels of indispensable amino acids are fed, even if diets were isonitrogenous) (82). The requirement of 0.60% of diet was based on growth, feed efficiency, [U-¹⁴C]arginine oxidation, and serum arginine concentrations (82). There were no effects of arginine intake on the plasma urea concentrations, urine creatinine excretion, the hepatic arginase activity, or the renal glycine transamidinase (E.C. 2.1.4.1) (82). Even when rabbits received an arginine-free diet for 7 d, they continued to grow, although at only 10% of the rate of those receiving an arginine-adequate diet (82). The authors did not observe any signs of hyperammonemia as a result of feeding the arginine-free diet, and they concluded that most aspects of arginine metabolism in growing rabbits were similar to those of the growing rat (82). They also concluded that de novo synthesis of arginine appeared adequate to meet the metabolic demands for arginine (82). However, the ability of the young growing rabbit to sustain growth in the absence of dietary arginine could also be the result of microbial synthesis of arginine in the ceca, subsequent excretion, and then intake of the microbial arginine via coprophagy. This has been recently described for microbially synthesized lysine in rabbits (84).

The rabbit must, at this time, be considered representative of other nonruminant herbivores because, to the best of our knowledge, there has been little or no research conducted in other nonruminant herbivores, such as horses. However, there may be unique aspects of arginine metabolism in horses. For example, mature horse milk has a high concentration of arginine (on a wt:wt total amino acid basis); compared with other mammalian species only feline milk has more arginine than mare's milk (85). Unlike other mammals, the arginine content of mare's milk increases 29% from colostrum to mature milk, whereas the arginine content of the milk in other mammals either decreases or remains unchanged (86). Arginine metabolism in horses is a clear opportunity for further investigation.

    Arginine metabolism and requirements in ruminants. There are no amino acid requirements published for ruminants such as growing beef (87) or dairy (88) cattle because it is assumed that suckling, preruminant offspring obtain enough amino acids from milk (or milk replacer formula) (89), and that once the rumen is developed, ruminants receive enough amino acids from the combination of diet and rumenal microbial amino acid synthesis. However, in a study that determined the lysine requirement of preruminant calves and then used the ratio of lysine to the other indispensable amino acids in calf body protein to estimate the amino acid requirements for the other amino acids, arginine intake was found to be only 60% of the estimated metabolic requirement (89). Furthermore, in milk replacer-fed calves, weaning weight and plasma arginine concentrations were greater when calves received an arginine-supplemented formula (90). The administration of arginine into the abomasum (similar to the stomach) of growing calves in conjunction with ammonium acetate improved nitrogen retention and lowered plasma ammonia concentrations relative to ammonium acetate administration alone (91). Therefore, similar to other young mammals, arginine should probably be reclassified as a conditionally indispensable amino acid for young ruminants. The estimated dietary arginine requirement of 8.5 g/d in a 500-kg cow (0.17 g·kg^–1·d^–1) (89) is very low compared with the dietary requirements of other growing mammals (0.30–0.50 g·kg^–1·d^–1, as discussed previously). However, because this requirement was based only on amino acid composition of body protein, it does not reflect the entire need for arginine to support urea cycle function; therefore, the requirement may be underestimated. The ruminant liver has detectable activity of all of the urea cycle enzymes (92), and cattle have a functional arginine-dependent urea cycle. Interestingly, however, unlike other mammalian species, an increase in arginine intake did not affect hepatic arginase activity (91). Ruminants extensively utilize nitrogen recycling; blood urea is secreted into the gastrointestinal tract, where it is used by the rumen microflora to resynthesize amino acids (93). Therefore, to ensure maximal amino acid efficiency in ruminants, a functional urea cycle, and therefore adequate whole-body arginine status, is critical to maintain normal nitrogen metabolism. These are examples, among others, of the many aspects of arginine, urea, and nitrogen metabolism that are unique to ruminants.

    Arginine requirements and metabolism in chicks. The dietary essentiality of arginine in the diet of chicks was first demonstrated in the 1930s (94,95). Further investigation revealed that arginine was a dietary indispensable amino acid in chicks because of the lack of detectable mitochondrial carbamoyl phosphate synthetase I (CPS I; E.C. 6.3.4.16) in all tissues investigated, including the liver, kidney, pancreas, and spleen (96). Although chick kidney, but not liver, has detectable activity of the enzymes OTC, ASS, and ASL, in comparison to rats (56), the renal activities of these enzymes in the chick are low. These low renal ASS and ASL activities have physiological significance, however, because citrulline addition to an arginine deficient diet was shown to be equally effective at promoting chick growth as arginine addition (97). The ability of citrulline to spare the arginine requirement in chicks was further confirmed when label from [ureido-¹⁴C]citrulline was found in chick body protein as arginine (98). Neither ornithine (95,97–99) nor bicarbonate (98) was a precursor for arginine, supporting the enzymatic data finding no detectable mitochondrial CPS I activity (96).

Unlike mammals, the metabolic arginine requirement of chicks does not include the support of urea cycle function. The main nitrogenous waste product in birds is uric acid, and its formation is not arginine-dependent. Because the urea cycle is not necessary for nitrogenous waste excretion, this is believed to be the reason for the almost complete lack of urea cycle enzymes (which are also arginine synthetic) in the chick liver (96).

Dietary arginine in chicks is therefore used only for protein synthesis and the synthesis of metabolically important molecules such as creatine. Creatine concentrations are related to arginine intake and availability in chicks (100–102). Therefore, although only a small amount of dietary arginine is used for creatine synthesis, adequate arginine intake is important for the synthesis of creatine.

Chicks have the enzymatic capacity for arginine degradation via arginase (E.C. 3.5.3.1) activity. However, unlike mammals, the arginase activity of the kidney is 30 times higher than that in the liver (96). Despite the fact that there is no functional urea cycle in chicks, chicks do excrete some urea, and this is a measure of arginine degradation via arginase. In laying hens, renal arginase activity and urea excretion were both increased by increasing arginine or protein intake (103), and urea production was related to plasma arginine concentration (103). Therefore, similar to other animals, arginine concentrations are regulated by arginine degradation in chickens. However, although arginine administration increased urea excretion in hens, the same was not observed with ornithine infusion (103), further demonstrating that ornithine is an ineffective arginine precursor in poultry. Studies in chicks showed that when renal arginase activity was altered by either changing dietary factors (increasing either lysine, arginine, or tyrosine content of the diet) or using strains of birds that were selected to have different levels of renal arginase expression, as arginase activity increased, there was also an increase in the arginine requirement (104). Therefore, the dietary requirement for arginine in chicks represents the metabolic arginine requirement for protein and metabolite synthesis plus that required to replace the arginine that is degraded by renal arginase.

The chick requirement for dietary arginine, on both a dietary concentration (1.10–1.25 g/100 g) and a percentage of CP basis (5.4–5.5%), is among the highest of any of the species studied despite not requiring arginine for urea cycle function. This high dietary requirement results from 1) lack of endogenous synthesis, 2) high rate of protein deposition because of the very rapid growth rate of meat-type chickens (50 g BW to 2 kg BW in 6 wk), and 3) the metabolic interaction between dietary lysine and arginine (discussed below). In comparison to most other indispensable amino acids, excess dietary arginine (up to 4 g/100 g of the diet) was well tolerated in young chicks, and only a small reduction in weight gain (9%) relative to a control diet was observed (105).

    Arginine requirements and metabolism in fish. Arginine is a dietary indispensable amino acid in fish. The activities of the urea cycle enzymes in various types of fish and the differences in urea metabolism among fish species have been reviewed (106). Therefore, this discussion focuses mainly on arginine and urea metabolism in rainbow trout, in comparison to mammals and chicks.

Similar to chicks, the limitation in arginine synthesis in fish is the synthesis of carbamoyl phosphate. Unlike in mammals, the enzyme that catalyzes the synthesis of carbamoyl phosphate in fish is carbamoyl phosphate synthetase III (CPS III; E.C. 6.3.5.5). Similar to CPS I, CPS III is mitochondrial and requires N-acetylglutamate as a cofactor, but instead of ammonia as the substrate, CPS III requires glutamine (107). Studies in rainbow trout have shown that there is no hepatic activity of CPS III and very low hepatic OTC activity (108). The muscle of trout has the greatest activity of CPS III (108); however, in comparison to the activity of CPS I in rat liver (56), its activity is very low (300 µmol of product formed·g liver·h^–1 in rats vs. 0.0144 µmol of product formed·g muscle·h^–1 in rainbow trout), and thus there is a very limited capacity for carbamoyl phosphate synthesis and subsequently de novo arginine synthesis in fish. There is detectable activity in tissues (specifically liver, kidney, muscle, and intestine) of rainbow trout of all of the other enzymes involved in arginine synthesis (108,109). Isotopic evidence indicates that trout are capable of some de novo arginine synthesis; injection of [1-¹⁴C]ornithine resulted in 10% of the label being recovered in arginine (109). There must be carbamoyl phosphate formation in order for ornithine to be used for arginine synthesis. Therefore, although the activity of CPS III may be low (108), it is of physiological relevance. However, de novo arginine synthesis is not sufficient to sustain optimal trout growth; addition of an equimolar amount of either arginine or citrulline to a low-arginine basal diet resulted in greater rates of gain, better feed efficiency, and higher plasma arginine concentrations than the addition of either glutamate or ornithine (109).

Many types of fish produce urea, but, as in the chick, this is largely believed to be caused by the degradation of dietary arginine (106,110). Arginase activity, primarily mitochondrial (8,108), has been detected in the liver, kidney, muscle, and intestine of rainbow trout (108,109), with the kidney, followed by the liver, having the greatest activity of arginase per gram tissue (108,109). In rainbow trout, hepatic arginase activity was not affected by arginine intake (8); however, in turbot fish, there was a strong correlation between arginine intake and urea production, which the authors concluded was a result of increasing hepatic arginase activity (110). The relation between arginine intake and whole-body arginine degradation via arginase activity, therefore, may be species-specific in fish. The extrahepatic contribution to arginine degradation in fish requires further investigation.

The main nitrogenous excretory product in fish is ammonia, which can diffuse through the gills into the aquatic environment [as outlined by Wright and Land (106)]. Therefore, unlike mammals, the metabolic and dietary requirement for arginine in fish does not include nitrogenous waste excretion. The dietary arginine requirement has been determined in many species of fish (8,111–117) using growth, feed efficiency, and plasma or serum arginine concentrations as the endpoint measurements. One study also used arginine oxidation (8). The importance of arginine for the synthesis of other metabolic products, to the best of our knowledge, has not been studied in fish. However, the low estimated maintenance requirements for arginine (calculated to be 10% or less of the arginine requirement for gain) (110,113), suggest that protein synthesis for growth is the primary use for arginine in fish.

Despite differences in the arginine biosynthetic enzymes in different species of fish (106), the arginine requirement of fish, 1.20–1.50 g/100 g of diet and 3.7–3.9 g/100 g of CP, is relatively similar among fish species, with the exception of the Pacific salmon (2.04 and 5.4, respectively) This similarity suggests that the observed differences in de novo arginine synthesis among fish species may not be nutritionally important under normal conditions in any of these species of fish.

The high arginine requirements in fish, expressed as grams per 100 g of diet, are comparable to the requirements in the chick, another nonureotelic species. These similarities probably exist because protein synthesis is the major contributor to the dietary arginine requirement in both species. Amino acid composition of muscle protein, specifically with regard to arginine, is also similar among these species (Table 1).

    General observations on arginine metabolism across species. The research on arginine metabolism in all of the species covered in this article raises many interesting points for discussion. First, it is commonly accepted that it is the ASS step of the urea cycle that is limiting for hepatic synthesis of arginine and therefore urea synthesis (56,118). However, in all species studied, citrulline was a more effective arginine precursor than either ornithine or proline (37,44,70,77,79,97,98), which shows that it is citrulline formation and not the conversion of citrulline to arginine that is the limiting step for endogenous arginine synthesis. The reason for the limitation in citrulline synthesis varies among species depending on the enzymes present. For example, the limitation in chicks and fish is because they do not excrete urea as their nitrogenous waste and lack mitochondrial CPS I activity (96,108), whereas in carnivores the reason is low P5C synthase and OAT activity (73). In other mammals, the reason that citrulline formation is limiting to arginine synthesis has not been conclusively determined, but recent research in neonatal piglets has suggested that it may be N-acetylglutamate synthesis, which in turn limits carbamoyl phosphate synthesis (119). Furthermore, these comparisons indicate that ASS and ASL enzyme function in arginine synthesis is fairly ubiquitous and similar across many species, including mammals, birds, and fish.

In piglets, proline is the major dietary precursor for endogenous arginine synthesis (24), and therefore, it is reasonable to assume that proline partially spares the dietary arginine requirement. In piglets receiving an adequate arginine diet, or an arginine-deficient diet supplemented with citrulline, the use of proline for arginine synthesis was reduced, showing that arginine can partially spare the proline requirement (37). Experimental confirmation is still required in other species to determine the ability of proline and arginine to spare the dietary requirement of the other amino acid.

Arginine deficiency symptoms are most severe in carnivores, with severe hyperammonemia occurring after a single arginine-free meal in cats (26,77), followed by omnivores, and are least severe in herbivores, with rabbits continuing to grow on an arginine-free diet, albeit at a reduced rate (82). Omnivores display a wide range of deficiency symptoms, with dogs appearing more sensitive to arginine deficiency than rats (27,55,58–60,80). Deficiency symptoms were also more severe in young vs. older growing mammals (25,27,44,80). The order of sensitivity to arginine deficiency follows approximately the same pattern as the dietary arginine requirements on the bases of grams per 100 g of diet, grams per kilogram per day, and grams per 100 g of CP. The species with virtually no capacity for endogenous arginine synthesis, fish and chicks, have the highest dietary arginine requirements despite not requiring arginine for urea cycle function. Therefore, the capacity for endogenous arginine synthesis is the major determinant of the dietary arginine requirement. This implies that all future research on arginine requirements and metabolism, regardless of species, should include a measurement of in vivo endogenous synthesis rate.

Because arginine requirements are highest in species without urea cycle function, this clearly shows that the major metabolic use of arginine is for protein synthesis. In piglets, it has been estimated that 70% of the daily arginine use is for protein synthesis (29). Therefore, although the other metabolic functions of arginine are critical for growth and health, protein synthesis is still the primary component of the metabolic arginine requirement.

The potential contribution to the animal of microbial synthesis of arginine in the small intestine, cecum, and large intestine does not appear to have been investigated thoroughly. In the young rabbit this source of arginine may be part of the reason that complete dietary deficiency does not result in the drastic response observed in other monogastric species. Rabbits and rodents may acquire significant intake of amino acids from coprophagy, and this is seldom considered when estimating dietary requirements. In contrast, there is evidence that microbially derived lysine may be used by the animal (120) in the absence of coprophagy. Therefore, additional research on microbial contribution of amino acids is required with all species.

To date, very little research has specifically been conducted, in any of the species discussed, to examine the effects of superphysiological intakes of arginine. In rats, for example, the daily intraperitoneal administration of 3.5 g/kg of body weight arginine induced an experimental form of pancreatitis (121). The effect of high arginine intake relative to dietary requirements in pigs is discussed in more detail below under lysine-arginine antagonism; however, it appears that the main effect of high arginine intake, relative to the amino acid profile, is a reduction in feed intake to compensate for the amino acid imbalance (122–125). In neonatal piglets, the twice-daily supplementation of arginine resulted in a decrease in plasma arginine and histidine concentrations relative to piglets receiving either no supplement, water, or alanine administration (119), which could have the negative implications of decreasing weight gain because of a limitation, in particular, in lysine from lysine-arginine antagonism. However, when the arginine was added to the milk-replacer and not given as a twice-daily bolus dose, there was no effect of arginine intake on plasma concentrations of either histidine or lysine, and piglets receiving the supplemental arginine grew better (32). In cattle, an increase in CP intake is associated with an increase in rumen ammonia concentrations as a result of microbial metabolism (126). Extremely high arginine intake in cattle may have effects, with regards to ammonia production, compared with an extremely high intake of other amino acids, because the microbial degradation of the arginine, which contains 4 nitrogen atoms, in the rumen would result in a greater rate and quantity of ammonia production than from the degradation of other amino acids. In species such as fish (110) and chicks (103), tissue arginase expression was generally induced by increasing arginine intake. Therefore, high arginine intake increases arginine degradation and urea excretion and produces no adverse effects. The limit to this adaptive degradation does not appear to have been defined in all species. Intakes of 4 to 5 times the dietary requirement have been used in requirement experiments, apparently without adverse effects based on lack of comments by the authors. Although future research is warranted to elucidate species-specific differences regarding the effects of excess arginine intake, it appears that arginine toxicity is unlikely to be a concern in any of the species examined.

    Implications of interspecies comparisons for human nutrition. Arginine metabolism and requirements in humans There are presently no dietary recommendations for humans for arginine during either growth or maintenance. There is a clear dietary arginine requirement for growth in all major experimental species; humans appear to be the exception. In light of the animal data summarized in this article, this position should be reconsidered, especially in neonates, and based on research in rats (88) during recovery from trauma or malnutrition at all ages. Arginine was concluded to be dispensable for neonates and children based on nitrogen retention and growth data (127,128); however, studies in rats have clearly showed that there can be metabolic aberrations related to inadequate arginine intake, even when growth is unaffected (59). Premature infants may be particularly susceptible to arginine deficiency (129), and poor arginine status has been associated with the onset of neonatal diseases including necrotizing enterocolitis (NEC) (130,131) and persistent pulmonary hypertension of the neonate (PPHN) (132,133). Supplemental arginine has been suggested to be protective against both NEC and PPHN (134–136). Furthermore, when intravenously fed infants with hyperammonemia were given additional arginine, the symptoms of hyperammonemia were alleviated (137,138). Together these observations suggest that even the combination of endogenous arginine synthesis and normal dietary arginine intake may not be enough for optimal neonatal health, particularly when health is already compromised by prematurity or intestinal disease. The question remains: what are the metabolic and dietary arginine requirements for neonates and growing children? Based on the recent research on arginine supplementation for suckling piglets (32), it should be considered whether arginine supplementation would be beneficial for preterm or low-birth-weight or normal-weight infants. Based on recent research in gestating sows (33), arginine supplementation may be a way to increase birth weight of infants with intrauterine growth retardation. Despite the important clinical implications relating to arginine intake in human neonates, there is presently no recommended adequate intake for arginine (2). Other young mammals clearly have a dietary requirement for arginine, which varies with a number of conditions. Therefore, it is particularly important that endogenous synthesis rates and metabolic/dietary requirements in human neonates and children be determined.

Although healthy adult males do not appear to be adversely affected by prolonged feeding of an arginine-free diet (139–142), in part because of the ability to decrease arginine catabolism (140), arginine metabolism is altered by pathological conditions such as sepsis (143), burn trauma (144), and end-stage renal disease (145), and therefore both metabolic and dietary arginine requirements must be investigated under each of these conditions. The metabolic requirement for arginine in adult humans following major trauma or surgery should also be determined because it is important for optimal clinical treatment (146,147).

The dietary arginine requirements for adult animals of species without appreciable amounts of endogenous arginine synthesis, specifically cats, fish, and chicks, may provide some insight into metabolic arginine requirements for humans. This is an example of an amino acid where application of the ideal protein concept could be used to recommend an adequate intake until experimental data are available. Neither chicks nor fish have an arginine requirement for urea cycle function, and dietary arginine is used primarily for protein synthesis. Therefore, their dietary arginine requirement may be comparable to the human arginine requirement for protein synthesis when corrected for the interspecies differences in rates of protein accretion. Cats appear to synthesize negligible amounts of arginine (76) but otherwise use arginine in the same basic manner as humans. Thus, their dietary arginine requirement could also be used as an initial estimate of the total metabolic arginine requirement in humans.

The closest approximation of the human neonatal arginine requirement is probably the pig. The pig is well recognized as an appropriate model to study human amino acid metabolism (148), and the amino acid content of porcine tissue is very similar to that of human tissue (7). Clearly these speculations require experimental confirmation, but the animal work has provided an excellent basis for the design of future studies in humans. Based on what has been learned from research in animals, the following must be considered when measuring arginine requirements in human populations: 1) the maximum rate of endogenous arginine synthesis under specific physiological conditions (i.e., illness, growth, pregnancy, lactation); 2) the metabolic arginine requirement for each physiological condition; 3) the shape of the response in metabolic function with intake varying from deficient to more than adequate; and 4) the bioavailabilty of dietary arginine and proline and the extent to which proline spares the dietary arginine requirement. In determining the ability of proline to spare arginine, the difference in molecular weights between proline and arginine must be recognized and taken into account in the calculations, similar to the process used to calculate cysteine sparing of methionine requirement (149–152). We have fully described how to calculate sparing capacity in our review of cysteine and methionine requirements (12).

Interspecies differences in the metabolic interaction between lysine and arginine

Amino acid antagonism vs. an amino acid imbalance must be clearly defined before it is possible to discuss the antagonism between lysine and arginine and its nutritional implications in different species. Both amino acid antagonisms and imbalances can cause reductions in rates of weight gain in growing animals. However, with an amino acid antagonism there is an interference in the metabolism of 1 amino acid caused by the intake of the other amino acid (153). Therefore, with an amino acid antagonism the negative effects on body weight are not only caused by a reduction in feed intake, which is also observed during cases of a dietary amino acid deficiency and imbalance, but by other effects on metabolism and utilization of the antagonized amino acid (153). Unlike the case of an imbalance, the effects of an amino acid antagonism can be reversed only by supplementing the diet with the amino acid that is being antagonized and not by supplementing the diet with the limiting amino acid(s) (153). The differences between amino acid antagonism and imbalance are illustrated in Figure 4.

View larger version (40K): [in this window] [in a new window]   FIGURE 4  Amino acid imbalance vs. amino acid antagonism. A diet with methionine as its limiting amino acid and excess amounts of threonine added, creating an amino acid imbalance (A). The effects of the amino acid imbalance in A are alleviated by adding the limiting amino acid, methionine, to the diet (B). Adding lysine to a diet that is limiting in methionine also has an adverse effect on daily gain because of an effect on arginine metabolism, which is an amino acid antagonism. Adding methionine to this diet will alleviate the limitation of methionine on protein synthesis but will not counteract the antagonism induced by the excess lysine intakes (C). The effects of the amino acid antagonism in C can be alleviated by adding arginine, the amino acid that was antagonized by the lysine, to the diet, although methionine is still the limiting amino acid (D).

Of the species studied, chicks (100,156–160), dogs (161), and rats (163) displayed evidence of a lysine-arginine antagonism. For chicks, antagonism was observed when the lysine content of the diet was 2–3.5% (157–160) or when there was a lysine-to-arginine ratio of 2.2–2.6:1 (156). In rats, when lysine content of the diet was >2.8%, there was a decrease in growth rate (163), whereas in dogs, the antagonism of lysine on arginine was not observed until lysine content was 4.9%, and not at either 1.9% or 2.9% (161). In all 3 species, the addition of arginine to the diet containing growth-depressing amounts of lysine resulted in increased growth and improved gain to feed ratios (156,161,163), substantiating an antagonistic relation of lysine on arginine metabolism.

The exact mechanisms for the lysine-arginine metabolism in chicks, dogs, and rats have not been conclusively elucidated. Lysine and arginine share and compete at both intestinal and renal transporters (164). However, in all 3 species there was no evidence to support that it was competition for intestinal absorption that was causing the observed antagonism (158,161,163). A deleterious effect on the digestion of arginine-containing protein by excess lysine was also eliminated; activities of pancreatic enzymes were not associated with differences in growth in rats (163), and diets containing free amino acids were equally growth-depressing for chicks as diets containing a similar amino acid composition as a casein protein (157). In chicks (157) and dogs (161), but not rats (163), there appears to be competition between lysine and arginine for renal reabsorption when growth-depressing intakes of lysine are fed. Thus, urinary arginine excretion was higher than when the control diet was fed (157,161). In chicks, renal arginase (E.C. 3.5.3.1) activity was markedly increased by excess lysine intake (157), which would cause increased arginine catabolism and thus explain the antagonism in chicks. However, increased arginase activity is not the only cause of the antagonism because this response is not apparent until after 2–4 d of feeding of the growth-depressing diet, despite significant decreases in plasma arginine concentrations within 6 h (157). Increased hepatic arginase activities were not associated with excess lysine intake in dogs (161<