Weaning, the transition from the ingestion of maternal milk to solid foods, seems to be a particularly vulnerable period for mammals, with increased incidences of malnutrition, intestinal infections and poor growth (Blecha et al. 1983, Wilson et al. 1989). Numerous nutrients [i.e. arginine, (n-3) fatty acids, nucleotides and glutamine] affect immune response (Alexander 1995), but glutamine seems particularly relevant because of its requirement by the gastrointestinal epithelium and its associated lymph tissues. During hypermetabolic states, exogenous glutamine supplementation preserves intestinal mucosal structure and function (O'Dwyer et al. 1989), supports normal immunologic responses of lymphocytes, macrophages and neutrophils (Ogle et al. 1994, Wallace and Keast 1992) and improves nitrogen balance (Babst et al. 1993).

The predominance of glutamine in milk samples of various species, including humans and pigs (Davis et al. 1994, Wu and Knabe 1994), suggests that glutamine may play an important role in the development and growth of young mammals, but little is known about glutamine requirements at weaning. Although amino acid requirements have been described for 10-kg pigs (Chung and Baker 1992), the ideal amino acid profile does not include glutamine. We hypothesized that the addition of glutamine to weanling diets would maintain normal skeletal muscle glutamine status and immune function in early weaned pigs.

METHODS AND MATERIALS

The experiment involved 36 animals (4 pigs per litter). Four piglets from each litter were randomly assigned to weaning diets supplemented with glutamine (40 g glutamine/kg diet; +Gln) or nonessential amino acid nitrogen (Gln) (Table 1). The essential amino acid (EAA)⁶ profiles of both diets (defined as grams of amino acids/100 g lysine) were based on the Illinois ideal amino acid pattern (IIP) described by Chung and Baker (1992) (Table 2). The chemically defined elemental diets were formulated to meet or exceed NRC (1988) nutrient requirements for 5 to 10-kg swine and were calculated to be isonitrogenous and isoenergetic. Throughout the 7-d experimental period, the pigs had free access to diet and water. At 26 d of age, one piglet from each diet was randomly selected and injected intraperitoneally with Escherichia coli [serotype 078; prepared as described by Samuels and Baracos (1992); +Ecoli] at a concentration of 0.5 × 10⁸ colony-forming units/kg body weight (CFU/kg BW). The other pig on the same diet was injected intraperitoneally with PBS (Ecoli). Rectal temperatures and clinical signs of shock were assessed at 6, 12 and 24 h after the intraperitoneal injection. At 28 d of age, the pigs were transferred to surgery and anesthetized using 2-5% halothane with oxygen (1 L/min). Blood samples (10 mL) were collected by cardiac puncture and muscle biopsies (~ 2-3 g) of longissimus dorsi (LD) were taken and snap frozen in liquid nitrogen. The pigs were killed by injection of a commercial euthanasia cocktail (T-61; Hoerchst Canada, Regina, Canada).

Table 1. Composition of the chemically defined glutamine-supplemented (+Gln) and glutamine-free (Gln) diets1 [View Table]

Table 2. Amino acid profiles of the glutamine-supplemented (+Gln) and glutamine-free diets (Gln)1 [View Table]

The mitogen response assay was conducted in triplicate on microtiter plates as previously described (Field 1995), using 200 µL of suspended lymphocytes (3 × 10⁵ cells/well). The cells were incubated up to 96 h without mitogens or with one of each of the following mitogens: Con A (5 mg/L) or PMA (40 µg/L) with Iono (400 µg/L). These concentrations and times were determined from preliminary experiments. Each well was pulsed with 1.85 kBq of [³H]-thymidine, and the cells were harvested on glass fiber filters 18 h later using a multiwell cell harvester (Skatron, Lier, Norway). Thymidine incorporation into the fraction retained on the filter was determined using Ecolite scintillation cocktail in a Beckman betacounter (LS 5801, Beckman Instruments, Mississauga, Canada). Stimulation indices (SI) were calculated as follows:

Table 3. Plasma amino acid concentrations of 28-d-old pigs weaned to glutamine-supplemented (+Gln) or glutamine-free (Gln) diets at 21 d of age and intraperitoneally injected with Escherichia coli (+Ecoli) or PBS (Ecoli) at 26 d of age1 [View Table]

Table 4. Longissimus dorsi intramuscular amino acid concentrations of 28-d-old pigs weaned to glutamine-supplemented (+Gln) or glutamine-free (Gln) diets at 21 d of age and intraperitoneally injected with Escherichia coli (+Ecoli) or PBS (Ecoli) at 26 d of age1 [View Table]

Table 5. Hematologic variables of 28-d-old pigs weaned to glutamine-supplemented (+Gln) or glutamine-free (Gln) diets at 21 d of age and intraperitoneally injected with Escherichia coli (+Ecoli) or PBS (Ecoli) at 26 d of age1 [View Table]

RESULTS

Plasma glycine, serine and taurine concentrations (Table 3) were significantly higher in pigs fed the Gln diet, which contained these amino acids, compared with the +Gln diet, which was devoid of these three amino acids. Plasma concentrations of alanine, citrulline, glutamine, glycine, histidine, methionine, ornithine, threonine, tryptophan and tyrosine were significantly lower in infected pigs independent of diet. Phenylalanine was the only plasma amino acid to be significantly greater in infected pigs.

Intracellular glutamine and citrulline concentrations of LD muscle were significantly greater in the glutamine-supplemented pigs (Table 4) and glutamine concentrations of infected pigs were less than those of the noninfected controls. Infection was associated with reduced intracellular histidine and threonine concentrations. Intracellular concentrations of branched-chain amino acids were unaffected by diet or infection, whereas asparagine concentrations were significantly higher in infected pigs than in noninfected controls.

White blood cell counts, red blood cell counts, hemoglobin concentration and hematocrit were significantly higher in pigs supplemented with glutamine compared with pigs fed the glutamine-free diet (Table 5). However, when hematologic values were corrected (divided by) hematocrit concentrations to adjust for possible differences in hydration associated with fever, only the white blood cell counts were significantly affected by glutamine supplementation (data not shown). Infection significantly elevated white blood cells and mean corpuscular hemoglobin. The higher total leukocyte counts observed in infected pigs compared with the noninfected controls were also associated with significantly higher numbers of monocytes, neutrophils and band cells (Table 6). When differential white blood cell counts were expressed as a percentage of total white blood cells, infected piglets had significantly higher percentages of band cells and lower lymphocyte percentages than the healthy animals.

Table 6. Hematologic variables of 28-d-old pigs weaned to glutamine-supplemented (+Gln) or glutamine-free (Gln) diets at 21 d of age and intraperitoneally injected with Escherichia coli (+Ecoli) or PBS (Ecoli) at 26 d of age1 [View Table]

Spontaneous (unstimulated) [³H]-thymidine uptakes (3, 18 ± 2.52 kBq/3 × 10⁵ cells) of isolated lymphocytes did not differ among the experimental groups after in vitro incubations of 0, 24, 48 and 96 h (data not shown). In vitro, lymphocytes of infected piglets had significantly lower responses to PMA + Iono at 24- and 72-h timepoints (Fig. 2A). The average peak response (stimulation index) to PMA + Iono was significantly lower in lymphocytes isolated from infected compared with noninfected piglets (Fig. 2B).

Lymphocytes isolated from glutamine-supplemented pigs had greater responses to Con A than did those from animals fed glutamine-free diets at 24 h (Fig. 3A and B). Average peak mitogen responses to Con A of lymphocytes isolated from +EcoliGln animals were significantly lower, whereas those from +Ecoli+Gln pigs did not differ from noninfected pigs (Fig. 3C).

DISCUSSION

The enterotoxigenic E. coli peritonitis model is an effective reproducible model with which to assess the metabolic effects of infection (McBurney et al. 1994, Samuels and Baracos 1992). Weanling pigs infected with <1 × 10⁸ CFU/kg BW experience a mild-to-moderate infection in a dose-dependent manner, with feed intakes and rectal temperatures returning to normal within 24 h of infection (McBurney et al. 1994). Using similar elemental diets, Chung and Baker (1992) reported normal feed intakes and body weight gains in pigs during the 3-wk postweaning period, but feed intakes during the wk 1 of weaning were not reported. The persistence of infection, measured by elevated rectal temperatures, observed in this study compared with previous studies with practical swine diets (McBurney et al. 1994) may be attributed to low feed intakes.

The decline in plasma glutamine concentrations observed with infection may reflect increased hepatic glutamine utilization to support gluconeogenesis, glutathione production and acute phase protein synthesis (Austgen et al. 1991) or glutamine utilization by stimulated lymphocytes (Wu et al. 1992). The reduced plasma concentrations of alanine, citrulline and ornithine, metabolites of glutamine metabolism by enterocytes (Wu et al. 1994), may also reflect decreased glutamine metabolism by the intestine as reported in infected rats by Salloum et al. (1991). Similar to reports concerning postsurgical patients receiving intravenous nutrition (Petersson et al. 1994), plasma glutamine concentrations were unchanged by glutamine supplementation, whereas intramuscular glutamine concentrations were increased. These results are also consistent with the effects of dietary glutamine supplementation and infection on intramuscular glutamine concentrations in rats (Askanzia et al. 1980, Shewchuk et al. 1997).

Plasma and intracellular amino acid analyses have revealed intriguing differences in amino acid patterns among species. Excluding taurine, glutamine is the most abundant amino acid in plasma and intracellular pools of humans (i.e., 0.6 and 19.5 mmol/L, respectively; Bergstrom et al. 1974) and rats (1.1 and 9 mmol/L, respectively; Turinsky and Long 1990). Our results from pigs agree with those of Deutz et al. (1992) who reported glycine as the most abundant amino acid in plasma of piglets (0.650 mmol/L) with glutamine concentrations approximating 0.335 mmol/L. Intracellular arginine concentrations account for ~45-53% of total free amino acids in pigs vs. ~1-2% in humans and rats (Bergstrom et al. 1974). The high arginine concentrations may reflect the type of muscle studied because Turinsky and Long (1990) reported muscle-fiber type-dependent changes in free amino acid profiles in rats. Indeed, intracellular concentrations of arginine and glutamine in gastrocnemius muscle from weaned pigs are 4888 ± 1518 and 3772 ± 1505 nmol/g wet muscle tissues, respectively. These results suggest that the effects of muscle type on amino acid profiles must be studied further.

The hematologic variables of all pigs were within normal ranges for pigs weaned at 28 d (Friendship et al. 1984). The rise in white blood cell counts of the infected pigs was attributable to significant increases in monocytes, neutrophils and band cells (see Table 6). Increased numbers of monocytes and neutrophils are routinely observed with bacterial infection (Zimmerman and Ringer 1992), and the rapid escalation of phagocytes is associated with the release of immature neutrophils (or band cells) from the bone marrow (Bishop et al. 1968). In contrast to the increase in phagocytic cells, absolute lymphocyte counts did not change. The decreased proportion of white blood cells, which are lymphocytes in the infected pigs, reflects increased numbers of nonlymphoid cells, but the absolute lymphocyte counts were unchanged by diet or infection and were within normal ranges (McCauley and Hartmann 1984).

PMA + Iono are nonspecific mitogens that are believed to stimulate all mononuclear cells. Generally, PMA activates protein kinase C, whereas Iono increases intracellular calcium concentrations to mediate protein phosphorylation. Lymphocytes of infected pigs displayed reduced responses in vitro to PMA + Iono in this study. Con A, however, specifically activates T cells via binding to a specific membrane receptor (i.e., T-cell receptor:CD3 complex or TCR:CD3) and subsequently phosphorylating membrane-bound proteins. The proliferative response of lymphocytes from infected pigs fed a glutamine-free diet (+EcolGln) to Con A stimulation was depressed, whereas lymphocytes from +Ecol+Gln pigs responded similarly to those from noninfected controls. These results agree with reports that glutamine supplementation supports immune function during critical states (Inoue et al. 1993) but has no effect in healthy states (Shewchuk et al. 1997). Moreover, lymphocyte activation via TCR:CD3 complex has been shown to require glutamine (Horig et al. 1993), and rat lymphocyte proliferation in vitro is dependent upon exogenous glutamine (Wu et al. 1992). Although glutamine supplementation did not significantly alter plasma glutamine concentrations of the infected pigs, intramuscular glutamine concentrations were significantly higher in all pigs receiving dietary glutamine, suggesting improved glutamine status. In addition, other factors such as changes in the production or recognition of cell membrane receptors, modification in the function or relative numbers of antigen-presenting cells and T cells, and altered production of lymphokines may also influence the results of the mitogen response assay (Field 1996). For example, macrophages were not measured in this study, and glutamine has been shown to affect phagocytic activities and interleukin-1 production of macrophages (Wallace and Keast 1992). Further studies are required to understand the potential effect of glutamine on antigen-processing and presentation.

In conclusion, the results of this study with piglets were consistent with studies involving glutamine supplementation in traumatized humans and rats (Inoue et al. 1993, Petersson et al. 1992). Healthy, weaned pigs, eating less than normal nutrient requirements, did not seem to benefit from exogenous glutamine. However, when the pigs were exposed to moderate infection, glutamine supplementation maintained normal muscle intracellular glutamine concentrations, leukocyte populations and lymphocyte functions. The provision of glutamine in diets, formulated to meet the amino acid requirements of the weanling pig, was beneficial in moderately infected pigs. It remains to be determined if the addition of glutamine to standard diets would minimize morbidity associated with weaning in practical environments.