This study investigates the effects of increased NH3 or amino acid supply on glutamine utilisation and production by the splanchnic tissues of fed sheep. Six sheep, prepared with vascular catheters in the aorta, mesenteric, portal and hepatic veins, were fed grass pellets to 1·1×energy maintenance requirements. Each treatment involved a 4 d abomasal infusion, of either ammonium bicarbonate (AMM; 23·4 μmol/kg0·75 per min), water (CONT), or a mixture of amino acids that excluded glutamine and glutamate (AA; 46·8 μmol amino acid-N/kg0·75 per min). The treatments simulated nutritional extremes in terms of the balance of absorbed N. Kinetics across the whole gut and the liver were monitored during an intra-jugular infusion of [5-15N]glutamine. Blood flow across the whole gut or liver were unaffected by treatment. Both AMM and AA infusions doubled the hepatic release of urea-N compared with CONT (P<0·02). AA infusion decreased arterial glutamine concentration by 26 % (P<0·01) and 23 % (P<0·05) compared with AMM and CONT respectively. Despite this, whole-body glutamine flux was not affected by treatment. In contrast, AMM infusion increased hepatic glutamine production by 40 % compared with CONT (P<0·02). This provided a mechanism to ensure NH3 supply to the periphery was maintained within the normal low physiological levels. Hepatic glutamine utilisation tended to increase during AA infusion, probably to ensure equal inflows of N to the ornithine cycle. Between 6 and 10 % of NH3 absorbed across the digestive tract was derived from the amido-N of glutamine. Overall, splanchnic glutamine utilisation accounted for 45–70 % of whole-body glutamine flux.

The effect of abomasal infusion of glucose (120 kJ/d per kg body weight (BW)0·75, 758 mmol/d) on urea production, plasma alanine-N flux rate and the conversion of alanine-N to urea was studied in sheep offered a low-N diet at limited energy intake (500 kJ/d per kg BW0·75), based on hay and grass pellets. Glucose provision reduced urinary N (P = 0·040) and urea (P = 0·009) elimination but this was offset by poorer N digestibility. Urea-N production was significantly reduced (822 v. 619 mmol/d, P = 0·024) by glucose while plasma alanine-N flux rate was elevated (295 v. 342 mmol/d, P = 0·011). The quantity of urea-N derived from alanine tended to be decreased by glucose (127 v. 95 mmol/d) but the fraction of urea production from alanine was unaltered (15 %). Plasma urea and alanine concentrations (plus those of the branched chain amino acids) decreased in response to exogenous glucose, an effect probably related to enhanced anabolic usage of amino acids and lowered urea production.

The mass transfers of O2, glucose, NH3, urea and amino acids across the portal-drained viscera (PDV) and the liver were quantified, by arterio–venous techniques, during the last 4 h of a 100 h infusion of 0 (basal), 150 or 400 μmol NH4HCO3/min into the mesenteric vein of three sheep given 800 g grass pellets/d and arranged in a 3 × 3 Latin-square design. Urea irreversible loss rate (ILR) was also determined by continuous infusion of [14C]urea over the last 52 h of each experimental period. PDV and liver movements of glucose, O2 and amino acids were unaltered by NH4HCO3 administration, although there was an increase in PDV absorption of non-essential amino acids (P = 0·037) and a trend for higher liver O2 consumption and portal appearance of total amino acid-N, glucogenic and non-essential amino acids at the highest level of infusion. PDV extraction of urea-N (P = 0·015) and liver removal of NH3 (P < 0·001), release of urea-N (P = 0·002) and urea ILR (P = 0·001) were all increased by NH4HCO3 infusion. Hepatic urea-N release (y) and NH3 extraction(x) were linearly related (R2 0·89), with the slope of the regression not different from unity, both for estimations based on liver mass transfers (1·16; SE 0·144; PB ≠ 1 = 0·31) AND [14C]UREA (0·97; se 0·123; Pb ≠ 1 = 0·84). The study indicates that a sustained 1·5 or 2·4-fold increase in the basal NH3 supply to the liver did not impair glucose or amino acid supply to non-splanchnic tissues; nor were additional N inputs to the ornithine cycle necessary to convert excess NH3 to urea. Half of the extra NH3 removed by the liver was, apparently, utilized by periportal glutamate dehydrogenase and aspartate aminotransferase for sequential glutamate and aspartate synthesis and converted to urea as the 2-amino moiety of aspartate.

Hepatic NH3 detoxification by ureagenesis requires an input of aspartate-N, originating either from amino acid-N or NH3-N. The relative importance of these two routes may depend on the nutritional state. To test this, four volunteers were given a liquid diet for 2 d and then on day 3 were either fed every 20 min or fasted. Doses of 15NH4Cl were taken orally every 20 min for 6 h (total 1.5 g) and blood was sampled hourly. Urea-N elimination under fasted conditions was only 0.75 of that for the fed state. Considering the increase in body urea pool during feeding, ureagenesis during fasting was probably closer to 0.6 of that during feeding. Since the [14N15N] urea enrichment was not different between the fed and fasted states, the proportion of the 15NH3 dose converted to urea during fasting was also 0.6 of that during the fed condition. No change in [14N15N] urea and [amide-15N] glutamine enrichment suggested that NH3 enrichment was also not affected by nutritional state. Enrichment of [14N15N] urea was approximately 0.05 that of [14N15N] urea which indicates that 15NH3 can also enter the aspartate route, the importance of which is yet unknown. Both [15N15] urea and [amino-15N]glutamine enrichment in the fasted state were approximately 1.7 times that in the fed state, indicating increased labelling of precursors and/or increased NH3 flux through the aspartate route. Glutamate, valine, leucine and isoleucine showed comparable increases in enrichment during fasting. Arginine enrichment was unaltered by nutritional state, but was lower than [14N15N] urea, indicating incomplete equilibration with the arginine pool in periportal hepatocytes. The present study indicates that hepatic NH3 detoxification may use the aspartate route, gaining importance in the fasted state. The majority of urea was supplied with only one N atom from NH3, thus provision of the other may have consequences for alternative substrates, in particular amino acids.

The effects of either low (25 μmol/min) or high (235 μmol/min) infusion of NH4Cl into the mesenteric vein for 5 d were determined on O2 consumption plus urea and amino acid transfers across the portal-drained viscera (PDV) and liver of young sheep. Kinetic transfers were followed by use of 15NH4Cl for 10 h on the fifth day with simultaneous infusion of [1-13C]lleucine to monitor amino acid oxidation. Neither PDV nor liver blood flow were affected by the additional NH3 loading, although at the higher rate there was a trend for increased liver O2 consumption. NH3-N extraction by the liver accounted for 64–70% of urea-N synthesis and at the lower infusion rate the additional N required could be more than accounted for by hepatic removal of free amino acids. At the higher rate of NH3 administration additional sources of N were apparently required to account fully for urea synthesis. Protein synthesis rates in the PDV and liver were unaffected by NH3 infusion but both whole-body (P < 0·05) and splanchnic tissue leucine oxidation were elevated at the higher rate of administration. Substantial synthesis of [15N]glutamine occurred across the liver, particularly with the greater NH3 supply, and enrichments exceeded considerably those of glutamate. The [15N]urea synthesized was predominantly as the single labelled, i.e. [14N15N], species. These various kinetic data are compatible with the action of ovine hepatic glutamate dehydrogenase (EC 1.4.1.2) in periportal hepatocytes in the direction favouring glutamate deamination. Glutamate synthesis and uptake is probably confined to the perivenous cells which do not synthesize urea. The implications of NH3 detoxification to the energy and N metabolism of the ruminant are discussed.