![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The Journal of Nutrition Vol. 128 No. 9 September 1998,
pp. 1487-1494
Laboratoire des Maladies Métaboliques et Micronutriments, INRA, Centre de Recherches en Nutrition Humaine de Clermont/Theix, F-63122 St-Genès Champanelle, France
 |
ABSTRACT |
|---|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
Glutamine release by the liver constitutes a process of nitrogen salvage through the recycling of a part of the nitrogen, which prevents irreversible nitrogen losses as urea. The aim of this work was to study the nitrogen cycling in the splanchnic bed under different nutritional conditions: fed state, postabsorptive state (16 h food deprivation) or prolonged starvation (24 or 40 h). Rats were adapted to a 15% casein diet for 15 d and then sampled. The digestive, hepatic and splanchnic balances of glucose, lactate, ketone bodies, urea and amino acids were determined. There was a net release of lactate and alanine by the digestive tract, due to the high rate of glycolysis and glutaminolysis. During prolonged starvation, ketone bodies became major energy fuel for the intestine. In fed rats, there was a net uptake of most amino acids by the liver, except for glutamine and glutamate. Urea, glutamine and glutamate released represented 33, 24 and 6% of total nitrogen taken up by the liver, respectively. In postabsorptive rats, compared with fed rats, there was a significant reduction of ureagenesis, and glutamine became the major form of nitrogen released by the liver. In fact, nitrogen cycling in the form of glutamine or glutamate in the liver may be interpreted as a nitrogen salvage process, rather than as an acid-base control process. In the splanchnic area, in parallel with a highly active cycling of glucose as lactate, there exists a nitrogen cycling involving opposite fluxes of glutamine and alanine.
KEY WORDS: alanine · amino acids · glutamine · nitrogen sparing
Nitrogen homeostasis requires a strict control of the mechanisms of N salvage and N elimination. In the body, there is a permanent protein turnover, involving concomitant processes of protein synthesis and degradation. Although many amino acids are recycled during these processes, a permanent catabolism and irreversible N losses exist, especially in the form of urea excreted in urine. The liver plays a key role in the control of N catabolism, by channeling it towards ureagenesis or glutamine (Gln) synthesis. Gln is used by many tissues, especially by intestine or kidneys in the case of acidosis (Welbourne 1987 Together with Glu, Gln is the only amino acid released by the liver. Because there is a permanent N flux from the digestive tract to the liver, chiefly as alanine and ammonia, it was interesting to assess the importance of this N cycling as a nitrogen salvage process. Production of Gln by liver or muscle is important because it has physiologic effects as a precursor for purine or pyrimidine (Meister 1979 Glucose, like Gln, is a major fuel substrate for enterocytes; glucose utilization results in lactate release and Gln utilization in Ala release (Rémésy et al. 1997 Thus, the liver production of Gln might fulfill intestinal requirements when protein intake is insufficient. Large amounts of Gln are metabolized by the small intestine (Souba 1991 The aim of this work was thus to investigate the mechanisms of N sparing via opposite fluxes of Gln and Ala in the splanchnic area during the transition from fed status to starvation.
Animals and diets.
Male Wistar rats (IFFA-CREDO, L'Arbresle, France) weighing 150-160 g were adapted for 15 d to a semipurified diet providing 15 g/100 g casein. This diet, described by Rémésy et al. (1988) Sampling procedures.
Rats were sampled under different nutritional conditions as follows: fed (8 h after food distribution), postabsorptive (16 h after food removal) or starved (24 h or 40 h after food removal). Animals were anesthetized with pentobarbital sodium (40 mg/kg). Blood (1 mL) was withdrawn in heparinized syringes from the hepatic vein (inside the left hepatic lobe), the portal vein and then the abdominal aorta. Blood from each animal was placed in a plastic tube containing heparin and centrifuged at 10,000 × g for 15 min. After centrifugation, 0.5 mL of plasma was removed and kept at Analytical procedures.
The frozen liver samples were crushed in 5 vol of 0.6 mol/L perchloric acid and neutralized with K2CO3; the plasma samples were deproteinized with sulfosalicylic acid. Amino acids were determined on a Chromakon 500 autoanalyzer (Kontron, Zurich, Switzerland) by using lithium buffers and postcolumn ninhydrin detection. For PAH determination, 100 µL of heparinized blood was added to 500 µL of trichloroacetic acid (100 g/L), vortexed and centrifuged at 8000 × g at 4°C. PAH was determined on the trichloroacetic supernatant by using a colorimetric procedure (Desbordes and Samorcq 1963 Enzyme assays.
The activity of carbamoylphosphate synthase (CPS, EC 6.3.4.16) was determined on liver mitochondrial fractions prepared as previously described (Rémésy et al. 1988 Calculations.
The digestive balance was determined as follows: [portal vein Statistical analysis.
Values are means ± SEM; where appropriate, significance of the differences among means was determined by one-way ANOVA coupled with the Student-Newman-Keuls test (StatView, Abacus, Berkeley, CA). Differences with P < 0.05 were considered significant.
Compared with fed rats, liver weight was reduced by ~2.7 g in postabsorptive rats (P > 0.05); in starved rats (24 or 40 h), the decrease of hepatic mass was more pronounced than that of body weight (Table 1).
Arterial metabolite concentrations.
In postabsorptive rats, there was a significant lowering of plasma urea, Gln and Ala compared with fed rats, whereas glucose and lactate concentrations were not significantly altered (Table 2). More prolonged periods of food deprivation (24 or 40 h) led to a decline of glucose, lactate and Ala concentration that was grossly proportional to the duration of the starvation. In contrast, Gln concentration remained steady at 0.60-0.65 mmol/L. Urea concentration exhibited a minimum at ~1.5 mmol/L during the postabsorptive period, then markedly rose to a value (3.7 mmol/L) higher than that found in fed rats. Plasma ketone bodies were almost undetectable in fed rats and were still relatively low in rats starved short term. Overt ketonemia (>2 mmol/L) developed in rats starved for 24 or 40 h. The HCO
Liver metabolite concentration and liver enzyme activities.
Glycogen storage was particularly high due to the high carbohydrate diet, but it rapidly declined after 16 h of food deprivation and was practically exhausted after 24 h of starvation (Table 3). The liver concentration of the major glucogenic substrates (lactate, Ala) progressively declined in starved rats, whereas Gln was significantly depressed only in rats starved for 40 h, at a still substantial value (~6 mmol/L). Liver ammonia underwent little change, remaining at ~1 mmol/L, except in rats starved for 40 h (1.43 mmol/L).
Dietary modulation of the digestive, hepatic and splanchnic fluxes of glucose and lactate.
The metabolic fluxes in the digestive tract, and hepatic and splanchnic areas are shown in Table 5. Glucose was absorbed at a high rate in rats sampled during the absorptive period, whereas the digestive balance of glucose shifted to negative values during the postabsorptive period. There was a permanent release of lactate by the digestive tract, due to a high rate of glycolysis. In fed rats, this release flux corresponded to 11% of absorbed glucose. In postabsorptive rats, lactate release in the portal vein represented ~70% of glucose taken up by the intestine. Figure 1 presents the evolution of the digestive fluxes of glucose and lactate in the other nutritional situations. In rats subjected to longer starvation periods, glucose utilization by the digestive tract was markedly reduced, but the apparent recovery of its carbon chain as lactate was still very high (~70%).
Dietary modulation of the fluxes of ketone bodies.
Ketogenesis was activated relatively early because 16 h of food deprivation (almost physiologic) was sufficient to elicit a significant rise in circulating acetoacetate and
Dietary modulation of the digestive and hepatic fluxes of glutamine, alanine, ammonia and urea.
In fed rats, absorption of amino acids by the digestive tract was very high, but the absorbed fluxes were very different among amino acids (Table 5). The largest fluxes were observed for Ala and proline as well as, to a lesser extent, for serine and glycine. Glu absorption was very low, and there was a net uptake of arterial Gln by the digestive tract, in spite of its relatively high percentage in dietary casein. Absorption of the other amino acids (branched-chain amino acids, methionine, lysine and histidine) was also substantial. As for Gln, a part of urea produced by the liver was used in the intestinal area, thereby contributing to the production of NH3 in the digestive tract. In postabsorptive rats, Ala production was dramatically depressed, whereas the apparent utilization of Gln was slightly enhanced. Absorption of the other amino acids was very low, frequently 10% of that found in fed rats, except for glycine. In postabsorptive rats, there was also a slight reduction of ammonia appearance in the digestive tract compared with fed rats, far less than differences in amino acid absorption. The consumption of urea by the digestive tract was strongly diminished during the postabsorptive period. In rats starved for 24 or 40 h, Gln uptake by the digestive tract showed a progressive decline (Fig. 3). There was a parallelism between Gln utilization and Ala release; the latter represented ~90% of the Gln flux. Urea uptake was very low in food-deprived rats and, in connection with the reduced Gln uptake, there was a decrease in digestive ammonia production.
Nitrogen cycling balance.
In fed rats, large amounts of N are available for liver metabolism (protein synthesis, ureagenesis). Urea, Gln and Glu release by the liver represent 33, 24 and 6% of total N uptake, respectively. In fact, N was essentially released as urea (53%) or Gln (38%). A part of N was also used for net protein synthesis (the weight of the liver was 2.7 g greater after the meal) (Fig. 4), and this synthesis could be estimated, by difference, at 37% of N taken up by the liver. In postabsorptive rats, there was a relative stoichiometry between the digestive utilization of Gln and Ala release. Under such conditions, the hepatic utilization of Ala is unlikely to provide the quantities of N required to sustain Gln synthesis. In fact, other N sources are available, especially ammonia and various amino acids provided by proteolysis. In these rats, the major part of N released by the liver was in form of Gln (51%) vs. 38% as urea and 11% as Glu. In fact, N release exceeded uptake, which is in keeping with active proteolysis. If Gln release were to be compared with N uptake as amino acids or ammonia from the blood, N cycling could be estimated to be in the range of 80% (Fig. 4). For longer starvation periods, arteriovenous differences in amino acids other than Ala or Gln were too low to be measured accurately, and it was not possible to calculate a reliable value for apparent N cycling. During prolonged fasting, Gln release was slightly higher than that of urea.
The capacity to take up or release Gln is a characteristic of liver metabolism; when Gln is utilized by the liver, it is essentially channeled toward ureagenesis. The actual physiologic importance of Gln release by the liver is still a matter of discussion. When there is an excess of amino acids supplied to the liver, ureagenesis is strongly stimulated by the generation of Glu, Asp and ammonia. In turn, ammonia concentrations are liable to activate glutaminase (MacGivan et al. 1984), which results in a net utilization of Gln by the liver (Rémésy et al. 1988 Liver metabolism.
When food is available for a limited period (8 h), the hepatic uptake of amino acids is particularly high in fed rats. The present approach affords an evaluation of the respective utilization of N toward protein synthesis or toward catabolic processes (37 and 63%, respectively). In fed rats, the abundance of N from various origins (including ammonia) induced important irreversible N losses as urea. But N cycling as Gln and Glu was observed in parallel. It is noteworthy that the hepatic production of Gln was higher than its apparent digestive utilization; as a result, there was a net supply of Gln and Glu for various tissues from the splanchnic area. After 16 h of food deprivation, although N availability was severely depressed, the liver released Gln at a rate similar to that observed in fed rats. Because there was a parallel reduction of ureagenesis, Gln was then the major form of N released by the liver. Therefore, the rate of N cycling was markedly higher in postabsorptive rats than in fed rats adapted to a well-balanced protein diet. During prolonged starvation, Gln release was slightly higher than ureagenesis, but there was a progressive decline of both metabolic processes.
Intestinal metabolism.
Whatever the dietary protein level, the digestive tract can almost completely metabolize the Glu and Gln available after absorption. It must be noted that net Gln absorption is observed only with high protein diets or supplementation of normal diets with a Gln load (Moundras et al. 1993
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
). Since the work of Atkinson and Bourke (1984), N excretion duality has been interpreted to function in acid-base balance. It has recently been shown that Gln and glutamate (Glu) release by the liver may be part of a nitrogen salvage process, rather than an acid-base control process (Rémésy et al. 1997). In fact, compared with glucose cycling as lactate, these mechanisms of N salvage have received less attention.
), oxidative fuel for enterocytes (Windmüeller and Spaeth 1978) and cells of the immune system (Newsholme et al. 1987), modulator of protein synthesis or degradation (Jepson et al. 1988) and as key intermediate for acid-base balance (Welbourne 1987). Together with alanine (Ala), Gln therefore plays a major role in interorgan N and carbon transport (Nurjhan et al. 1995, Ruderman and Berger 1974).
). Depending on the dietary protein level, the liver may take up or release Gln (Fafournoux et al. 1990). During starvation, this organ releases both glucose and Gln. Lactate and Ala constitute major sources of carbon-chain units for liver under most nutritional conditions. Hepatic lactate metabolism represents a salvage and recycling process of glucose carbon-chain units (Cori 1981), whereas the physiologic role of liver Gln production and of N return as Ala and ammonia for N salvage is poorly documented.
); in the absence of effective N cycling, N depletion would rapidly occur during starvation or in subjects fed a low protein diet.
MATERIALS AND METHODS
Abstract
Introduction
Methods
Results
Discussion
References
, contained (g/100 g) the following: wheat starch (Louis François, Paris, France) 73, peanut oil (C.I.O., Genay, France) 5, salt mixture 6 and vitamin mixture 1 (both mixtures were purchased from Usine d'Alimentation Rationnelle, Villemoisson/Orge, France). For each experimental group, eight rats were used. Animals had free access to the food from 0900 to 1700 h, with a 12-h dark (0900-2100 h):light (2100-0900 h) cycle. Animals were handled according to the recommendations of the Institutional Ethics Committee (University of Clermont-Ferrand).
20°C. Portal and hepatic blood flows were determined by an indicator-dilution method, with p-aminohippurate (PAH)3 as indicator (Demigné et al. 1986). The liver was excised, and a portion of the liver (1.0 g) was immediately frozen in liquid nitrogen and stored at 80°C.
). Glucose, lactate, urea, ammonia, 
-hydroxybutyrate and acetoacetate were determined spectrophotometrically on neutralized perchloric acid extracts by enzymatic methods, as previously described (Bergmeyer 1974). The pH and PCO2 were measured with a blood-gas analyzer (Radiometer ABL 30, Copenhagen, Denmark). The concentrations of HCO
3 were calculated from pH and PCO2 using the Henderson-Hasselbalch equation as previously described (Bushinsky et al. 1983).
). CPS was measured by conversion of the carbamoylphosphate formed to citrulline by the coupled reaction with ornithine transcarbamylase according to Lutsy (1978); then citrulline was determined colorimetrically using diacetyl monoxime (Boyde and Rahmatullah 1980). Carbamoylphosphate activity was expressed as nmol citrulline formed/(min·mg mitochondrial protein). The activity of Gln synthase (GS) was measured on the supernatant obtained after centrifugation (16,000 × g for 30 min at 0°C) of a 10% liver homogenate in 150 mmol/L KCl, pH 7.2, 2.5 mmol/L dithiothreitol and 10 mmol/L EDTA. The enzyme activity was determined essentially as described by Rowe et al. (1970), by measuring the formation of 
-glutamylhydroxamate when hydroxylamine is substituted for ammonia (synthase assay). The activity of GS was expressed as nmol -glutamylhydroxamate formed/(min·mg protein).
artery ] × portal blood flow. The liver afferent concentration was calculated using portal vein and artery concentrations and their respective blood flow (hepatic artery blood flow was calculated as the difference between hepatic and portal blood flow). The hepatic balance was calculated as follows: [hepatic vein afferent] × hepatic blood flow. The splanchnic balance was calculated as the algebraic sum of the digestive and hepatic balances.
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
View this table:
Table 1.
Physiologic characteristics of fed, postabsorptive and food-deprived rats1
3 concentration remained unchanged during the postabsorptive period (near 25 mmol/L). Then, during starvation (24 and 40 h), there was a slight drop in this concentration to 19 mmol/L.
View this table:
Table 2.
Arterial concentrations of metabolites of fed, postabsorptive and food-deprived rats1
View this table:
Table 3.
Liver concentrations of glycogen, lactate, glutamate, glutamine, alanine and ammonia of fed, postabsorptive
and food-deprived rats1
View this table:
Table 4.
Hepatic activity of carbamoyl phosphate synthetase and glutamine synthetase of fed, postabsorptive and food-deprived rats1
View this table:
Table 5.
Metabolic fluxes of ketone bodies, glucose, lactate, urea, ammonia and amino acids in fed and postabsorptive rats1
View larger version (22K):
[in a new window]
Fig 1.
Glucose and lactate fluxes in portal-drained viscera and liver of fed, postabsorptive and food-deprived rats. Values are means ± SEM, n = 8. P < 0.05 vs. Fed. Fluxes were calculated as described in Materials and Methods; positive values correspond to a net release (and/or absorption in the case of digestive tract) and negative values to a net uptake.
-hydroxybutyrate (Fig. 2). The digestive tract preferentially removed -hydroxybutyrate; thus, the splanchnic balance was markedly higher for acetoacetate than for -hydroxybutyrate.
View larger version (21K):
[in a new window]
Fig 2.
Ketone bodies (acetoacetate and -hydroxybutyrate) fluxes in portal-drained viscera and liver of fed, postabsorptive and food-deprived rats Values are means ± SEM, n = 8. *P < 0.05 vs. Postabsorptive. Fluxes were calculated as described in Materials and Methods; positive values correspond to a net release (and/or absorption in the case of digestive tract) and negative values to a net uptake.
View larger version (31K):
[in a new window]
Fig 3.
Glutamine and alanine fluxes in portal-drained viscera and liver of fed, postabsorptive and food-deprived rats. Values are means ± SEM, n = 8. 
P < 0.05 vs. Fed and *P < 0.05 vs. Postabsorptive. Fluxes were calculated as described in Materials and Methods; positive values correspond to a net release (and/or absorption in the case of digestive tract) and negative values to a net uptake.
View larger version (32K):
[in a new window]
Fig 4.
Nitrogen recycling in liver of fed and postabsorptive rats. Values are means ± SEM, n = 8. Fluxes were calculated as described in Materials and Methods.
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
). In contrast, in animals fed a low protein diet, Gln is released by the liver, which reflects a high rate of Gln synthesis together with a low rate of ureogenesis (Rémésy et al. 1997). In anesthetized rats, during the postabsorptive period, a net release of Gln by the liver has been systematically found (Moundras et al. 1993). However, Welbourne et al. (1986) observed that the liver of conscious rats took up a small amount of Gln. However, it is still uncertain whether these differences in the Gln balance resulted from the sampling conditions (anaesthetized or conscious rats) because, in the Welbourne study, the rats were provided ammonia, a patent glutaminase inducer.
). The low rate of ketogenesis observed after 16 h of food deprivation (0.4 mmol/L ketone bodies in blood plasma) is unlikely to have a large effect on the acid-base equilibrium. Thus, during short-term starvation, Gln release seems independent of acid-base balance. In contrast, for longer starvation periods, there is a slight decrease in plasma bicarbonate, and an effect of acidosis on Gln metabolism could not be ruled out. A recent work (Rémésy et al. 1997) has shown that the hepatic production of Gln constitutes a major process of N sparing to fulfill the needs of the intestine. N cycling may vary considerably, from 60 to 14%, in rats adapted to an 11 or 22% casein diet. It was thus important to investigate the changes in N cycling in fed rats, as well as in those that were food deprived. One major aim of this study was to show whether there is a N sparing involving Gln/Ala opposite fluxes in the splanchnic area.
have shown that ammonia may be readily utilized for amino acid synthesis such as for Glu or Gln. Furthermore, according to Cooper et al. (1988), N exchanges very rapidly between Glu and Gln. In fact, in fed animals, NH2 provision is not limiting for Gln synthesis.
all the more because Gln contains 1.7-fold more carbon than Ala. In fact, in fed rats, there is an ample supply of pyruvate-yielding substrates leading to 
-ketoglutarate via the Krebs cycle. The carbon supply for Gln synthesis in starved rats is more questionable, i.e., these data indicate that Gln release deprives the glucogenic pathway of carbon units. In fact, some Gln carbon arises from Ala, but a major part probably comes from amino acids provided by proteolysis. As was the case for carbon, the sum of alanine N and ammonia N is not sufficient to ensure the observed Gln synthesis; here again, proteolysis is probably an additional source of N in postabsorptive or starved rats.
). The present results suggest that the end products of Gln metabolism are chiefly Ala and NH3, rather than citrulline and proline (Windmüeller 1982). More recent investigations (Watford 1994) have confirmed that Gln was essentially metabolized as three carbon compounds such as Ala, with glycolysis as the probable major source of pyruvate for Ala synthesis by transamination (Cremin and Fleming 1997).
). The removal of arterial Gln after 16 h of food deprivation seems markedly higher than during the absorptive period. However, because enterocytes of fed rats are also provided both Gln and Glu of dietary origin, the total utilization of Gln is certainly higher in fed rats than in postabsorptive rats. With long-term starvation, there was a significant decrease of Gln utilization, consistent with a relative atrophy of intestinal villi. Accordingly, during refeeding or enteral nutrition, Gln could facilitate enterocyte recovery (Frankel et al. 1993).
. In fact, it seems unlikely that Gln metabolism is extensively channeled toward CO2 production, whereas glucose would provide the totality of pyruvate for lactate and Ala synthesis. Nevertheless, Gln would be less effective as an energy fuel if there were no glucose to provide pyruvate as an ammonia acceptor.
). In blood plasma, Gln metabolized by the small intestine may arise from muscles (Welbourne 1987). In the same way, Gln of hepatic origin may be used for kidneys or muscle metabolism (Wu et al 1991). However, our results suggest that, in term of net Gln fluxes, the liver fulfills most of the Gln requirement for intestinal metabolism in fed as well as in postabsorptive rats. For longer fasting periods, a part of Gln can certainly be used for renal gluconeogenesis, which supports the view of the role of Gln synthesis in sparing bicarbonate necessary for urea synthesis (Bean and Atkinson 1984).
). The question arises then about an adaptive response of liver metabolism because Gln could be considered a conditionally essential amino acid (Lacey and Wilmore 1990).
| |
FOOTNOTES |
|---|
Manuscript received 23 September 1997. Initial reviews completed 29 October 1997. Revision accepted 5 May 1998.
| |
ACKNOWLEDGMENT |
|---|
We thank Pierre Lamby for expert technical assistance.
| |
LITERATURE CITED |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Watford Glutamine and Glutamate Metabolism across the Liver Sinusoid J. Nutr., April 1, 2000; 130(4): 983 - 983. [Abstract] [Full Text]
|
|
|
|
 |
|
 C. M. Croniger, C. Millward, J. Yang, Y. Kawai, I. J. Arinze, S. Liu, M. Harada-Shiba, K. Chakravarty, J. E. Friedman, V. Poli, et al. Mice with a Deletion in the Gene for CCAAT/Enhancer-binding Protein beta Have an Attenuated Response to cAMP and Impaired Carbohydrate Metabolism J. Biol. Chem., January 5, 2001; 276(1): 629 - 638. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. T. Brosnan, M. E. Brosnan, M. Yudkoff, I. Nissim, Y. Daikhin, A. Lazarow, O. Horyn, and I. Nissim Alanine Metabolism in the Perfused Rat Liver. STUDIES WITH 15N J. Biol. Chem., August 17, 2001; 276(34): 31876 - 31882. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
