JOURNAL
OF SURGICAL
RESEARCH
49, 8-13 (1990)
Amino Acid Metabolism
in Isolated Perfused Rat Liver
J.P. DE BANDT,PHARM.D.,* L.CYNOBER, PH.D.,* F.BALLET,PH.D.,P C. COUDRAY-LUCAS,PH.D.,* C. REY,~ AND J. GIBOUDEAU, PH.D.* *Laboratoire
de Biochimie
A, TINSERM Submitted
U 181, H6pital St. Antoine,
for publication
MATERIALS
AND
METHODS
Liver Perfusion Male Sprague-Dawley rats (Janvier Laboratories; 244 + 30 g) were used. The animals were fasted overnight before experiments but had free access to a glucose solution (50 g/liter). Under sodium pentobarbital anesthesia (60 mg/kg intraperitonealy) the livers were prepared according to Miller’s technique, as previously described [4]. Briefly, after cannulation of the bile duct, 1 ml of saline containing 1000 IU of heparin was injected into the saphenous vein and the portal vein was then cannulated. The liver was perfused immediately with 30 ml of the perfusion solution and excised. The liver was then perfused through the portal vein with a recirculating system (180 ml) in a thermostatically controlled cabinet; the perfusate flowed freely back into the reservoir via the hepatic veins. Portal pressure was maintained at the physiological value of 13 cm Hz0 by overflow of the perfusate into the reservoir. The perfusate was composed of whole rat blood diluted in Krebs-Ringer buffer (1:3, v/v) with 1700 UI/ liter of heparin, 30 g/liter of bovine serum albumin, 1.5 g/liter of glucose, and 2 mmole/liter of Ca2+. The AA composition of the medium varied according to the experiment.
INTRODUCTION
The study of hepatic amino acid (AA) extraction is important for the investigation of liver metabolism. However, this parameter is currently approached using in uiuo studies which investigate splanchnic rather than strictly hepatic exchanges. Among the different methods used to study hepatic metabolism, the isolated perfused rat liver (IPRL) is a very convenient tool because of the flexibility of the system. In addition, the structural and functional integrities of the organ are preserved, an essential condition for the study of the metabolism of some AA such as glutamine
Inc. reserved.
15, 1989
[ 11. The amino acid composition of the perfusion medium is critical for hepatic AA metabolism, as shown by the work of Woodside and Mortimore on the control of hepatic proteolysis [2]. An inadequate AA supply can thus explain the very unphysiological results reported by a number of authors, e.g., the release of alanine by the liver during pyruvate loading reported by Parilla and Goodman [3], something which never occurs in physiological situations. The aim of the present study was thus to investigate the variations in hepatic AA metabolism according to the AA supply and then to determine perfusion conditions which restore near-physiological metabolism.
Conflicting evidence concerning hepatic amino acid (AA) metabolism in the isolated perfused rat liver (IPRL) led us to investigate the response of IPRL using perfusates with various AA contents. Perfusion (n = 4) with whole rat blood diluted in Krebs buffer (1:3, v/v) led to acute proteolysis on account of AA deprivation, as shown by the large release of AA (- 1400 nmole in 120 min), especially branched-chain AA (BCAA) (e.g., Leu, 35.4 f 10.4 nmole . mine1 * g-l the first hour, 34.3 f 5.5 nmole . min-’ . g-’ the second hour). In a first attempt to prevent proteolysis, livers (n = 4) were perfused with the previous medium supplemented with AA known for their antiproteolytic activity, at twice their physiological concentrations. Results during the first hour showed uptake of several AA (mainly alanine, glutamine, and proline), reduced release of BCAA (leucine, 12.5 _t 6.3 nmole . min-’ . g-l), and an increase in glucose and urea production. However, during the second hour, because of the use of a recirculating system, progressive AA depletion induced a reappearance of proteolysis. A two-step AA loading technique, i.e., the addition of antiproteolytic AA at the beginning of the perfusion and the addition of a balanced AA mixture at 60 min caused a further decrease in proteolysis during the 2 hr of perfusion (n = 6). Under these conditions, most AA were taken up by the liver with uptake values comparable to those observed in vivo. 0 1990 Academic Press, Inc.
0022-4804/90 $1.50 Copyright 0 1990 by Academic Press, All rights of reproduction in any form
February
75571 Paris Ceden 12, France
8
DE BANDT
Experimental
ET AL.: AMINO
ACID
METABOLISM
Design
RAT
Methods
For AA analysis in the perfusate, the sample was immediately deproteinized using sulfosalicylic acid (50 mg/ ml). Individual free AA concentrations were measured by ion-exchange chromatography on a Chromakon 500 analyzer (Kontron, Switzerland) [S]. Glucose and urea concentrations were measured with routine enzymatic methods using glucose oxidase and urease, respectively.
AA Composition
1
of the Perfusion
Unsupplemented medium (A) (pmole/liter) Taurine Threonine Serine Glutamate Glutamine Proline Glycine Alanine Cysteine Methionine Valine Leucine Isoleucine Tyrosine Phenylalanine Tryptophan Ornithine Lysine Histidine Arginine Aspartate
9
LIVER
TABLE
To investigate variations in hepatic AA metabolism according to the AA supply and to determine the AA composition of the perfusion medium able to restore physiological hepatic AA metabolism, three sets of experiments were performed at various AA concentrations. In the first set of experiments, four livers were perfused with an unsupplemented medium, i.e., the medium previously described with no added AA. AA concentrations were thus those of diluted fasted rat blood (Table 1, column A). In the second set of experiments, four livers were perfused with an “antiproteolytic medium,” i.e., the above medium to which nine AA (glutamine, proline, alanine, methionine, leucine, phenylalanine, tryptophan, histidine, and taurine) were added at the start of the perfusion at twice their physiological concentration (Table 1, column B): eight were chosen for their antiproteolytic properties [5, 61 and the ninth (taurine) for its ability to conjugate bile acids 173. In the third set of experiments, six livers were submitted to a two-step AA loading: livers were first perfused with the antiproteolytic medium and, after 60 min, a balanced mixture of AA was added to the perfusate. The AA composition of the perfusion medium after the addition of the balanced AA mixture is given in Table 1, column C. AA were concentrated solutions prepared either by ourselves (antiproteolytic AA) or from commercially available solutions (balanced AA mixture, Sigma MEM B 6766 and BME 7145). Perfusate samples (2 ml) were taken for biochemical analysis at the portal inflow and at the venous effluent at the beginning of the perfusion and, unless otherwise stated, every 30 min until the end of the perfusion. Hepatic blood flow was measured at each time. Analytical
IN PERFUSED
Antiproteolytic AA medium (B) (~mole/liter)
372 9 70 It 11 52 I? 17 41+ 20
236f 30 65 3~10 46 f 12 50 + 15
163 + 27 50 2 11
1018 + 65
75 t 16 93 4 20 9+ 3 16+
Medium
4
67ic 6 52 AI 10 37* 7 21+ 5 21% 6 15k 6 202 3 92+ 4
280 +_35 79 * 14 64Ok60 8f 3 90 It 17 55f 5 287 IL 40 36f 7 3Ok 9 92 +_15 91* 7 13+ 4 86 +_18
182
6
91 + 14
262
4
33 f 12
ND
ND
Balanced AA medium (C) (fimole/liter)
102 + 381-c 210 + 476 +
18 40 35 32
591 k 45
2'71f 161 k 292 k 81t
36 24 45 32
110 +_30
363 + 48 539 f 60 381250 145 f 18 239 t- 27 83 i 16 32 k 12 304 k43 62 + 17 134 f 25 113 * 30
Note. ND, not determined; A and B, AA concentrations in the perfusion medium at t = 0; C, AA concentrations in the perfusion medium at t = 65 min.
a given AA at times tl and t2, respectively. Results are expressed as nanomoles per minute per gram of liver (wet weight). This mode of expression, currently used for such experiments [g-11], was preferred to hepatic extraction since the volume of the perfusate (180 ml) and portal blood flow (2.92 + 0.12 ml * min-’ * g-‘) were identical in all experiments. Urea and glucose production rates (R) were calculated from R = Q(Co - C,), where Q is hepatic blood flow and Co and C, are concentrations in venous outflow and portal inflow, respectively. Results are expressed as micromoles per minute per gram of liver (wet weight). All results are given as the mean _t SD. Data were analyzed using Student’s t test. RESULTS
Calculations
AA Exchanges in Unsupplemented
Rates (R) of AA uptake or release by the liver were calculated according to the formula
Livers perfused with the unsupplemented medium released large amounts of AA, as shown by the sharp increase (133%) in the total AA concentration during the perfusion period (2387 1- 551 ymolelliter at t = 120 min vs 1023 +_132 pmole/liter at t = 0; P < 0.01). Release was grossly linear during the perfusion. Almost all AA were released, but the phenomenon was most striking for BCAA (leucine, valine, isoleucine; Table
R =
- Ct1 x _v t2 - tl P ’
Ct2
where V is the volume of perfusate, P is the liver wet weight, and Ct, and Ct2 are the portal concentrations of
Medium
10
JOURNAL
TABLE Liver
OF SURGICAL
with
First hour
Second hour
-7.3 -5.2 1.1 1.6 2.2 2.6 4.7 5.4 6.7 7.3 7.3 7.4 9.5 13.3 21.5 30.2 32.1 33.1 35.3 35.4
10.2 -2.2 -0.5 1.8 2.1 2.7 -0.1 4.7 -1.8 8.4 5.4 6.2 0.9 18.5 17.8 34.5 42.5 26.0 43.2 34.3
-t 3.9 + 4.3 -+ 2.8 ? 0.9 f 1.8 + 1.2 _+ 1.3 + 2.7 f 3.2 + 4.4 + 5.2 k 4.8 + 1.9 f 10.2 -t 6.9 f 12.0 + 10.4 + 11.4 + 25.3 f 10.4
VOL. 49, NO. 1, JULY
1990
crease in urea production (in both the first and the second hour) and in glucose production was noted, contrasting with results for the unsupplemented medium (Table 3). However, during the second hour, the net uptake of AA was reduced or converted to a net release. A concommitant increase in the release of BCAA (data not shown), reflecting significant proteolysis, and a decrease in urea and glucose production (Table 3) also occurred. Glutamine showed a very different pattern from other AA, being actively taken up by the liver during the first hour and released during the second hour (data not shown).
2
AA Exchanges during Perfusion Unsupplemented Medium
Alanine Taurine Arginine Tryptophan Tyrosine Cysteine Methionine Ornithine Serine Threonine Lysine Histidine Phenylalanine Proline Isoleucine Glycine Glutamate Valine Glutamine Leucine
RESEARCH:
t 6.4 f 0.5 f 0.2 k 0.3 + 1.3 + 0.9 5 1.4 3~ 2.3 f 1.6 + 2.4 f 2.2 + 2.4 f 2.3 f 13.5 -t 6.3 f 12.3 Z!T 5.4 + 7.2 + 16.7 -t 5.5
Metabolic Effect of Two-Step AA Loading During the 60 min of perfusion with the antiproteolytic medium, results were similar to those described above. However, during the second hour (following the addition of the second AA load), a linear decrease in total AA concentrations was observed (Fig. 3). The uptake of most AA was increased, and the release of BCAA was lower than that during the corresponding period with the antiproteolytic medium alone (Fig. 4). There was also a net decrease in glucose production, with no modification in urea production (Table 3).
Note. Results (mean + SD) are expressed in nmole * min-’ * g-‘. Negative values, uptake; positive values, release.
DISCUSSION 2), whose release, given their poor hepatic catabolism [12], is considered a marker of protein catabolism [13]. Furthermore, in contrast to a stable rate of urea production, glucose production stopped during the second hour (Table 3). AA Exchanges in the Antiproteolytic
The results of this study point out the strong variations in AA metabolism in the perfused rat liver according to the AA content of the medium. The massive release of AA by livers perfused with an unsupplemented medium is in accordance with the report of Lacy [14] and with that of Pii@ and Mortimore [6] concerning AA deprivation-induced proteolysis. The strong release of BCAA, indicating acute proteolysis, together with abnormally low gluconeogenic AA metabolism and the high rate of glucose production which both indicate acute glycogenolysis, and the release of AA such a phenylalanine usually taken up by the liver, led us to question the relevance of such perfusion conditions to the in vivo situation.
Medium
Under these conditions, high AA uptake occurred during the first hour, as demonstrated by the decrease in the total AA concentration in the per&sate (Fig. 1). Measurements of individual AA indicated a significant uptake of gluconeogenic AA, mainly glutamine, alanine, and proline, with a concommitant large release of glutamate. The release of BCAA was significantly lower than that with the unsupplemented medium (Fig. 2). An in-
TABLE Effect
of AA Content
of the Perfusion
3
Medium
on Glucose
and Urea
Production
Glucose
Unsupplemented medium Antiproteolytic medium Two-step AA loading
Urea
First hour
Second hour
First hour
Second hour
0.433 + 0.206 0.577 f 0.210 0.585 _t 0.180
-0.191 f 0.154 0.341 5 0.158” 0.147 +- 0.035c
0.137 * 0.062 0.639 + 0.350* 0.619 -t 0.199
0.107 f 0.031 0.387 rf: 0.225’ 0.419 f 0.166
Note. Results (mean f SD) are expressed in mmole * min-’ * g-‘. a and 5, significantly different < 0.01, respectively). c, significantly different from antiproteolytic medium (P < 0.01).
from unsupplemented
medium (P < 0.05 and P
DE BANDT
0.
60 Time tmim
30
0
FIG. 1. Variations (m) and antiproteolytic
ET AL.: AMINO
ACID
METABOLISM
120
90
in total AA concentrations perfusate (A).
in unsupplemented
As P&o et al. [5, 61 have shown that eight AA (glutamine, proline, methionine, leucine, phenylalanine, tyrosine, tryptophan, and histidine) inhibit AA deprivation-
TAU
CLU
* ** --I--I----II
CLN
PRO
CLY
ALA
MET
VAL
LEU
ILE
PHE
l *
**
l
l
*
*
l
NS
*
HIS
IN PERFUSED
RAT
11
LIVER
induced proteolysis and that a ninth AA, alanine, acts as a coregulator in this inhibitory effect, we first attempted to prevent proteolysis by increasing the concentration of antiproteolytic AA in the perfusion medium. This has a dramatic effect on amino acid exchanges, converting a net release into a net uptake. These results are in accordance with those of Mondon and Mortimore [ 151 obtained under similar conditions: at low AA concentrations, a high release of AA was observed, whereas during perfusion with high AA concentrations (8.2 mmole/l) a net uptake occurred. AA uptake was most marked for gluconeogenic AA, particularly alanine whose rate of uptake represented approximately one-third that of total AA. These results are similar to findings in uiuo by the measurement of splanchnic AA exchanges; in effect, the amount of alanine and glutamine taken up by the liver each represent onethird of total AA uptake, while the amount of glutamate released is equivalent to approximately half the uptake of glutamine [16]. The increase in AA uptake and in glucose and urea production shows an improvement in hepatic AA metabolism during the first hour of perfusion with the antiproteolytic medium. However, during the second hour the net uptake of AA was reduced or even converted to a net release. Given the role of alanine in the regulation of hepatic proteolysis [5,6] it seems reasonable to assume that there was a renewal of protein degradation due to progressive AA depletion because of the use of a recirculating
*
IC
5
-lC
-2c
0 0 FIG. 2. Uptake (negative values) and release (positive values) of individual amino acids during the first hour of perfusion with unsupplemented medium (open bars) and antiproteolytic medium (hatched bars). Values are expressed as nmole * mini * g-i (mean -+ SD): Significant (P < 0.05) uptake or release; l P < 0.05 and l *P < 0.01 between unsupplemented medium and antiproteolytic medium.
30
60
Time
SO
1zu
(min)
FIG. 3. Variations in total amino acid concentrations using twostep AA supplementation. The arrow indicates the introduction of the balanced amino acid mixture into the perfusate. *P < 0.05, t = 120 vs t = 65 min.
JOURNAL
12 GLN .
GLU
OF SURGICAL
RESEARCH:
CYS
VOL. 49, NO. 1, JULY
1990
SER
VAL
LEU
ILE
GLY
THR
ORN
TYR
NS
*lt
NS
Irk
lrlr
NS
L--‘-~L..2L.J(IIIIIul--luIrt * Ir
LYS
ARC
HIS
TRP
MET
PHE
PRO
ALA
Ir
k
NS
Ir
*
t
*
NS
-1c (III-----NS t
FIG. 4. acids at
t
Release and uptake of amino acids by the liver during the second hour of perfusion. (0) No addition of amino acids at t = 60 ) Addition of a balanced amino acid mixture at t = 60 (n = 6). Significantly different from 0: l P < 0.05, ‘*P < 0.01. Addition of amino = 60 vs no addition: l P < 0.05, l *P < 0.01.
perfusion system. With regard to the variations in glutamine release during the course of the experiments, this could be explained by the compartimentalization of hepatic glutamine metabolism as demonstrated by Sies and HBussinger [ 11: at high portal glutamine levels, periportal hepatocytes metabolize glutamine, whereas at low portal levels, perivenous cells synthetize glutamine from glutamate. In summary, these data show that the use of an antiproteolytic medium does not allow stable metabolic status to be attained in prolonged experiments. Since portal flow increases progressively to reach a plateau during the first hour in an organ perfusion system, experiments are usually performed during the second hour [17, 181. We thus evaluated the effects of adding a balanced AA mixture at the end of the first hour of perfusion. The increase in AA uptake, observed under these conditions, together with the stable rate of urea production indicate an increase in protein synthesis, while the decrease in glucose production is probably related to a modification of the glycogen balance, i.e., an activation of glycogenesis or an inhibition of ‘glycogenolysis. This double AA supplementation regimen (t = 0, antiproteolytic medium; t = 60, balanced AA mixture) thus prevented the metabolic disorders observed during the first hour and reestablished the equilibrium between protein synthesis and degradation during the second hour.
In conclusion, two-step AA loading enabled us to obtain near-physiological hepatic AA metabolism, as attested by the correspondance between our results and those obtained in uiuo. This model will be useful in studies of liver metabolism in physiological or pathological conditions and in the investigation of hepatic response to hormones, pharmacological compounds, and various modifications of the nutrient supply. REFERENCES 1.
2.
3. 4.
5.
6.
Sies, H., and Hliussinger, D. Hepatic glutamine and ammonia metabolism. In D. Hliussinger and H. Sies, (Eds.), Glutamine Metabolism in A4ammalian Tissues. Berlin: Springer-Verlag, 1984. Pp. 78-91. Woodside, K. H., and Mortimore, G. E. Suppression of protein turnover by amino acids in the perfused rat liver. J. Biol. Chem. 247: 6474, 1972. Parilla, R., and Goodman, M. N. Nitrogen metabolism in the isolated perfused rat liver. Biochem. J. 138: 341, 1974. Ballet, F., Chr&ien, Y., Rey, C., and Poupon, R. Norepinephrine: a potential modulator of the hepatic transport of taurocholate. A study in the isolated perfused rat liver. J. Pharmacol. Exp. Ther. 240: 303, 1987. P&ii, A. R., Wert, J. J., and Mortimore, G. E. Multifunctional control by amino acids of deprivation-induced proteolysis in liver. J. Biol. Chem. 257: 12114, 1982. P&G, A. R., and Mortimore, G. E. Requirement for alanine in the amino acid control of deprivation-induced protein degradation in liver. Proc. Natl. Acad. Sci. USA 81: 4270, 1984.
DE BANDT
ET AL.: AMINO
ACID
METABOLISM
7. Wright, C. E., Tallan, H. H., and Lim, Y. Y. Taurine: Biological update. Annu. Rev. Bivchem. 55: 427,1986. 8. Cynober, L., Coudray-Lucas, C., Ziegler, F., and Giboudeau, J. High performance ion-exchange chromatography of amino-acids in biological fluids using chromakon 500. Performance of the apparatus. J. Autom. Chem. 7: 201,1985. 9. Mortimore, G. E. Effect of insulin on release of glucose and urea by isolated rat liver. Amer. J. Physiol. 204: 699, 1963. 10. Frolich, J., Scholmericb, J., Hoppe-Seyler, G., Maier, K. P., Talke, H., Schollmeyer, P., and Gerok, W. The effect of acute uraemia on gluconeogenesis in isolated perfused rat liver. Eur. J. Clin. Invest. 4: 453, 1974. 11. Ball&, C., and Jungermann, K. Control of urea production, glutamine release and ammonia uptake in the perfused rat liver by the sympathetic innervation. Eur. J. Biochem. 158: 13, 1986. 12. Harper, A. E., Miller, R. H., and Block, K. P. Branched-chain amino acid metabolism. Annu. Rev. Nutr. 4: 409,1984.
IN PERFUSED
RAT
LIVER
13
13.
Mortimore, G. E., and Mondon, C. E. Inhibition by insulin of valine turnover in liver. J. Biol. Chem. 245: 2375,197O.
l4
Lacy, W. W. Effect of acute uremia on amino acid uptake and urea production by perfused rat liver. Amer. J. Physiol. 216: 1300,1969.
’ 15.
Mondon, C. E., and Mortimore, G. E. Effects of insulin on amino acid release and urea formation in perfused rat liver. Amer. J. Physiol. 212: 173,1967.
16.
Felig, P. Amino acid metabolism 933,1975.
17.
Woodside, K. H., Ward, W. F., and Mortimore, G. E. Effects of glucagon on general protein degradation and synthesis in perfused rat liver. J. Biol. Chem. 249: 5458, 1974.
18.
Glinsmann, W. H., and Mortimore, G. E. Influence of glucagon and 3’,5’-AMP on insulin responsiveness of the perfused rat liver. Amer. J. Physiol. 215: 553, 1968.
in man. Anna Reu. Biochem. 44:
OF SURGICAL
RESEARCH
49, 8-13 (1990)
Amino Acid Metabolism
in Isolated Perfused Rat Liver
J.P. DE BANDT,PHARM.D.,* L.CYNOBER, PH.D.,* F.BALLET,PH.D.,P C. COUDRAY-LUCAS,PH.D.,* C. REY,~ AND J. GIBOUDEAU, PH.D.* *Laboratoire
de Biochimie
A, TINSERM Submitted
U 181, H6pital St. Antoine,
for publication
MATERIALS
AND
METHODS
Liver Perfusion Male Sprague-Dawley rats (Janvier Laboratories; 244 + 30 g) were used. The animals were fasted overnight before experiments but had free access to a glucose solution (50 g/liter). Under sodium pentobarbital anesthesia (60 mg/kg intraperitonealy) the livers were prepared according to Miller’s technique, as previously described [4]. Briefly, after cannulation of the bile duct, 1 ml of saline containing 1000 IU of heparin was injected into the saphenous vein and the portal vein was then cannulated. The liver was perfused immediately with 30 ml of the perfusion solution and excised. The liver was then perfused through the portal vein with a recirculating system (180 ml) in a thermostatically controlled cabinet; the perfusate flowed freely back into the reservoir via the hepatic veins. Portal pressure was maintained at the physiological value of 13 cm Hz0 by overflow of the perfusate into the reservoir. The perfusate was composed of whole rat blood diluted in Krebs-Ringer buffer (1:3, v/v) with 1700 UI/ liter of heparin, 30 g/liter of bovine serum albumin, 1.5 g/liter of glucose, and 2 mmole/liter of Ca2+. The AA composition of the medium varied according to the experiment.
INTRODUCTION
The study of hepatic amino acid (AA) extraction is important for the investigation of liver metabolism. However, this parameter is currently approached using in uiuo studies which investigate splanchnic rather than strictly hepatic exchanges. Among the different methods used to study hepatic metabolism, the isolated perfused rat liver (IPRL) is a very convenient tool because of the flexibility of the system. In addition, the structural and functional integrities of the organ are preserved, an essential condition for the study of the metabolism of some AA such as glutamine
Inc. reserved.
15, 1989
[ 11. The amino acid composition of the perfusion medium is critical for hepatic AA metabolism, as shown by the work of Woodside and Mortimore on the control of hepatic proteolysis [2]. An inadequate AA supply can thus explain the very unphysiological results reported by a number of authors, e.g., the release of alanine by the liver during pyruvate loading reported by Parilla and Goodman [3], something which never occurs in physiological situations. The aim of the present study was thus to investigate the variations in hepatic AA metabolism according to the AA supply and then to determine perfusion conditions which restore near-physiological metabolism.
Conflicting evidence concerning hepatic amino acid (AA) metabolism in the isolated perfused rat liver (IPRL) led us to investigate the response of IPRL using perfusates with various AA contents. Perfusion (n = 4) with whole rat blood diluted in Krebs buffer (1:3, v/v) led to acute proteolysis on account of AA deprivation, as shown by the large release of AA (- 1400 nmole in 120 min), especially branched-chain AA (BCAA) (e.g., Leu, 35.4 f 10.4 nmole . mine1 * g-l the first hour, 34.3 f 5.5 nmole . min-’ . g-’ the second hour). In a first attempt to prevent proteolysis, livers (n = 4) were perfused with the previous medium supplemented with AA known for their antiproteolytic activity, at twice their physiological concentrations. Results during the first hour showed uptake of several AA (mainly alanine, glutamine, and proline), reduced release of BCAA (leucine, 12.5 _t 6.3 nmole . min-’ . g-l), and an increase in glucose and urea production. However, during the second hour, because of the use of a recirculating system, progressive AA depletion induced a reappearance of proteolysis. A two-step AA loading technique, i.e., the addition of antiproteolytic AA at the beginning of the perfusion and the addition of a balanced AA mixture at 60 min caused a further decrease in proteolysis during the 2 hr of perfusion (n = 6). Under these conditions, most AA were taken up by the liver with uptake values comparable to those observed in vivo. 0 1990 Academic Press, Inc.
0022-4804/90 $1.50 Copyright 0 1990 by Academic Press, All rights of reproduction in any form
February
75571 Paris Ceden 12, France
8
DE BANDT
Experimental
ET AL.: AMINO
ACID
METABOLISM
Design
RAT
Methods
For AA analysis in the perfusate, the sample was immediately deproteinized using sulfosalicylic acid (50 mg/ ml). Individual free AA concentrations were measured by ion-exchange chromatography on a Chromakon 500 analyzer (Kontron, Switzerland) [S]. Glucose and urea concentrations were measured with routine enzymatic methods using glucose oxidase and urease, respectively.
AA Composition
1
of the Perfusion
Unsupplemented medium (A) (pmole/liter) Taurine Threonine Serine Glutamate Glutamine Proline Glycine Alanine Cysteine Methionine Valine Leucine Isoleucine Tyrosine Phenylalanine Tryptophan Ornithine Lysine Histidine Arginine Aspartate
9
LIVER
TABLE
To investigate variations in hepatic AA metabolism according to the AA supply and to determine the AA composition of the perfusion medium able to restore physiological hepatic AA metabolism, three sets of experiments were performed at various AA concentrations. In the first set of experiments, four livers were perfused with an unsupplemented medium, i.e., the medium previously described with no added AA. AA concentrations were thus those of diluted fasted rat blood (Table 1, column A). In the second set of experiments, four livers were perfused with an “antiproteolytic medium,” i.e., the above medium to which nine AA (glutamine, proline, alanine, methionine, leucine, phenylalanine, tryptophan, histidine, and taurine) were added at the start of the perfusion at twice their physiological concentration (Table 1, column B): eight were chosen for their antiproteolytic properties [5, 61 and the ninth (taurine) for its ability to conjugate bile acids 173. In the third set of experiments, six livers were submitted to a two-step AA loading: livers were first perfused with the antiproteolytic medium and, after 60 min, a balanced mixture of AA was added to the perfusate. The AA composition of the perfusion medium after the addition of the balanced AA mixture is given in Table 1, column C. AA were concentrated solutions prepared either by ourselves (antiproteolytic AA) or from commercially available solutions (balanced AA mixture, Sigma MEM B 6766 and BME 7145). Perfusate samples (2 ml) were taken for biochemical analysis at the portal inflow and at the venous effluent at the beginning of the perfusion and, unless otherwise stated, every 30 min until the end of the perfusion. Hepatic blood flow was measured at each time. Analytical
IN PERFUSED
Antiproteolytic AA medium (B) (~mole/liter)
372 9 70 It 11 52 I? 17 41+ 20
236f 30 65 3~10 46 f 12 50 + 15
163 + 27 50 2 11
1018 + 65
75 t 16 93 4 20 9+ 3 16+
Medium
4
67ic 6 52 AI 10 37* 7 21+ 5 21% 6 15k 6 202 3 92+ 4
280 +_35 79 * 14 64Ok60 8f 3 90 It 17 55f 5 287 IL 40 36f 7 3Ok 9 92 +_15 91* 7 13+ 4 86 +_18
182
6
91 + 14
262
4
33 f 12
ND
ND
Balanced AA medium (C) (fimole/liter)
102 + 381-c 210 + 476 +
18 40 35 32
591 k 45
2'71f 161 k 292 k 81t
36 24 45 32
110 +_30
363 + 48 539 f 60 381250 145 f 18 239 t- 27 83 i 16 32 k 12 304 k43 62 + 17 134 f 25 113 * 30
Note. ND, not determined; A and B, AA concentrations in the perfusion medium at t = 0; C, AA concentrations in the perfusion medium at t = 65 min.
a given AA at times tl and t2, respectively. Results are expressed as nanomoles per minute per gram of liver (wet weight). This mode of expression, currently used for such experiments [g-11], was preferred to hepatic extraction since the volume of the perfusate (180 ml) and portal blood flow (2.92 + 0.12 ml * min-’ * g-‘) were identical in all experiments. Urea and glucose production rates (R) were calculated from R = Q(Co - C,), where Q is hepatic blood flow and Co and C, are concentrations in venous outflow and portal inflow, respectively. Results are expressed as micromoles per minute per gram of liver (wet weight). All results are given as the mean _t SD. Data were analyzed using Student’s t test. RESULTS
Calculations
AA Exchanges in Unsupplemented
Rates (R) of AA uptake or release by the liver were calculated according to the formula
Livers perfused with the unsupplemented medium released large amounts of AA, as shown by the sharp increase (133%) in the total AA concentration during the perfusion period (2387 1- 551 ymolelliter at t = 120 min vs 1023 +_132 pmole/liter at t = 0; P < 0.01). Release was grossly linear during the perfusion. Almost all AA were released, but the phenomenon was most striking for BCAA (leucine, valine, isoleucine; Table
R =
- Ct1 x _v t2 - tl P ’
Ct2
where V is the volume of perfusate, P is the liver wet weight, and Ct, and Ct2 are the portal concentrations of
Medium
10
JOURNAL
TABLE Liver
OF SURGICAL
with
First hour
Second hour
-7.3 -5.2 1.1 1.6 2.2 2.6 4.7 5.4 6.7 7.3 7.3 7.4 9.5 13.3 21.5 30.2 32.1 33.1 35.3 35.4
10.2 -2.2 -0.5 1.8 2.1 2.7 -0.1 4.7 -1.8 8.4 5.4 6.2 0.9 18.5 17.8 34.5 42.5 26.0 43.2 34.3
-t 3.9 + 4.3 -+ 2.8 ? 0.9 f 1.8 + 1.2 _+ 1.3 + 2.7 f 3.2 + 4.4 + 5.2 k 4.8 + 1.9 f 10.2 -t 6.9 f 12.0 + 10.4 + 11.4 + 25.3 f 10.4
VOL. 49, NO. 1, JULY
1990
crease in urea production (in both the first and the second hour) and in glucose production was noted, contrasting with results for the unsupplemented medium (Table 3). However, during the second hour, the net uptake of AA was reduced or converted to a net release. A concommitant increase in the release of BCAA (data not shown), reflecting significant proteolysis, and a decrease in urea and glucose production (Table 3) also occurred. Glutamine showed a very different pattern from other AA, being actively taken up by the liver during the first hour and released during the second hour (data not shown).
2
AA Exchanges during Perfusion Unsupplemented Medium
Alanine Taurine Arginine Tryptophan Tyrosine Cysteine Methionine Ornithine Serine Threonine Lysine Histidine Phenylalanine Proline Isoleucine Glycine Glutamate Valine Glutamine Leucine
RESEARCH:
t 6.4 f 0.5 f 0.2 k 0.3 + 1.3 + 0.9 5 1.4 3~ 2.3 f 1.6 + 2.4 f 2.2 + 2.4 f 2.3 f 13.5 -t 6.3 f 12.3 Z!T 5.4 + 7.2 + 16.7 -t 5.5
Metabolic Effect of Two-Step AA Loading During the 60 min of perfusion with the antiproteolytic medium, results were similar to those described above. However, during the second hour (following the addition of the second AA load), a linear decrease in total AA concentrations was observed (Fig. 3). The uptake of most AA was increased, and the release of BCAA was lower than that during the corresponding period with the antiproteolytic medium alone (Fig. 4). There was also a net decrease in glucose production, with no modification in urea production (Table 3).
Note. Results (mean + SD) are expressed in nmole * min-’ * g-‘. Negative values, uptake; positive values, release.
DISCUSSION 2), whose release, given their poor hepatic catabolism [12], is considered a marker of protein catabolism [13]. Furthermore, in contrast to a stable rate of urea production, glucose production stopped during the second hour (Table 3). AA Exchanges in the Antiproteolytic
The results of this study point out the strong variations in AA metabolism in the perfused rat liver according to the AA content of the medium. The massive release of AA by livers perfused with an unsupplemented medium is in accordance with the report of Lacy [14] and with that of Pii@ and Mortimore [6] concerning AA deprivation-induced proteolysis. The strong release of BCAA, indicating acute proteolysis, together with abnormally low gluconeogenic AA metabolism and the high rate of glucose production which both indicate acute glycogenolysis, and the release of AA such a phenylalanine usually taken up by the liver, led us to question the relevance of such perfusion conditions to the in vivo situation.
Medium
Under these conditions, high AA uptake occurred during the first hour, as demonstrated by the decrease in the total AA concentration in the per&sate (Fig. 1). Measurements of individual AA indicated a significant uptake of gluconeogenic AA, mainly glutamine, alanine, and proline, with a concommitant large release of glutamate. The release of BCAA was significantly lower than that with the unsupplemented medium (Fig. 2). An in-
TABLE Effect
of AA Content
of the Perfusion
3
Medium
on Glucose
and Urea
Production
Glucose
Unsupplemented medium Antiproteolytic medium Two-step AA loading
Urea
First hour
Second hour
First hour
Second hour
0.433 + 0.206 0.577 f 0.210 0.585 _t 0.180
-0.191 f 0.154 0.341 5 0.158” 0.147 +- 0.035c
0.137 * 0.062 0.639 + 0.350* 0.619 -t 0.199
0.107 f 0.031 0.387 rf: 0.225’ 0.419 f 0.166
Note. Results (mean f SD) are expressed in mmole * min-’ * g-‘. a and 5, significantly different < 0.01, respectively). c, significantly different from antiproteolytic medium (P < 0.01).
from unsupplemented
medium (P < 0.05 and P
DE BANDT
0.
60 Time tmim
30
0
FIG. 1. Variations (m) and antiproteolytic
ET AL.: AMINO
ACID
METABOLISM
120
90
in total AA concentrations perfusate (A).
in unsupplemented
As P&o et al. [5, 61 have shown that eight AA (glutamine, proline, methionine, leucine, phenylalanine, tyrosine, tryptophan, and histidine) inhibit AA deprivation-
TAU
CLU
* ** --I--I----II
CLN
PRO
CLY
ALA
MET
VAL
LEU
ILE
PHE
l *
**
l
l
*
*
l
NS
*
HIS
IN PERFUSED
RAT
11
LIVER
induced proteolysis and that a ninth AA, alanine, acts as a coregulator in this inhibitory effect, we first attempted to prevent proteolysis by increasing the concentration of antiproteolytic AA in the perfusion medium. This has a dramatic effect on amino acid exchanges, converting a net release into a net uptake. These results are in accordance with those of Mondon and Mortimore [ 151 obtained under similar conditions: at low AA concentrations, a high release of AA was observed, whereas during perfusion with high AA concentrations (8.2 mmole/l) a net uptake occurred. AA uptake was most marked for gluconeogenic AA, particularly alanine whose rate of uptake represented approximately one-third that of total AA. These results are similar to findings in uiuo by the measurement of splanchnic AA exchanges; in effect, the amount of alanine and glutamine taken up by the liver each represent onethird of total AA uptake, while the amount of glutamate released is equivalent to approximately half the uptake of glutamine [16]. The increase in AA uptake and in glucose and urea production shows an improvement in hepatic AA metabolism during the first hour of perfusion with the antiproteolytic medium. However, during the second hour the net uptake of AA was reduced or even converted to a net release. Given the role of alanine in the regulation of hepatic proteolysis [5,6] it seems reasonable to assume that there was a renewal of protein degradation due to progressive AA depletion because of the use of a recirculating
*
IC
5
-lC
-2c
0 0 FIG. 2. Uptake (negative values) and release (positive values) of individual amino acids during the first hour of perfusion with unsupplemented medium (open bars) and antiproteolytic medium (hatched bars). Values are expressed as nmole * mini * g-i (mean -+ SD): Significant (P < 0.05) uptake or release; l P < 0.05 and l *P < 0.01 between unsupplemented medium and antiproteolytic medium.
30
60
Time
SO
1zu
(min)
FIG. 3. Variations in total amino acid concentrations using twostep AA supplementation. The arrow indicates the introduction of the balanced amino acid mixture into the perfusate. *P < 0.05, t = 120 vs t = 65 min.
JOURNAL
12 GLN .
GLU
OF SURGICAL
RESEARCH:
CYS
VOL. 49, NO. 1, JULY
1990
SER
VAL
LEU
ILE
GLY
THR
ORN
TYR
NS
*lt
NS
Irk
lrlr
NS
L--‘-~L..2L.J(IIIIIul--luIrt * Ir
LYS
ARC
HIS
TRP
MET
PHE
PRO
ALA
Ir
k
NS
Ir
*
t
*
NS
-1c (III-----NS t
FIG. 4. acids at
t
Release and uptake of amino acids by the liver during the second hour of perfusion. (0) No addition of amino acids at t = 60 ) Addition of a balanced amino acid mixture at t = 60 (n = 6). Significantly different from 0: l P < 0.05, ‘*P < 0.01. Addition of amino = 60 vs no addition: l P < 0.05, l *P < 0.01.
perfusion system. With regard to the variations in glutamine release during the course of the experiments, this could be explained by the compartimentalization of hepatic glutamine metabolism as demonstrated by Sies and HBussinger [ 11: at high portal glutamine levels, periportal hepatocytes metabolize glutamine, whereas at low portal levels, perivenous cells synthetize glutamine from glutamate. In summary, these data show that the use of an antiproteolytic medium does not allow stable metabolic status to be attained in prolonged experiments. Since portal flow increases progressively to reach a plateau during the first hour in an organ perfusion system, experiments are usually performed during the second hour [17, 181. We thus evaluated the effects of adding a balanced AA mixture at the end of the first hour of perfusion. The increase in AA uptake, observed under these conditions, together with the stable rate of urea production indicate an increase in protein synthesis, while the decrease in glucose production is probably related to a modification of the glycogen balance, i.e., an activation of glycogenesis or an inhibition of ‘glycogenolysis. This double AA supplementation regimen (t = 0, antiproteolytic medium; t = 60, balanced AA mixture) thus prevented the metabolic disorders observed during the first hour and reestablished the equilibrium between protein synthesis and degradation during the second hour.
In conclusion, two-step AA loading enabled us to obtain near-physiological hepatic AA metabolism, as attested by the correspondance between our results and those obtained in uiuo. This model will be useful in studies of liver metabolism in physiological or pathological conditions and in the investigation of hepatic response to hormones, pharmacological compounds, and various modifications of the nutrient supply. REFERENCES 1.
2.
3. 4.
5.
6.
Sies, H., and Hliussinger, D. Hepatic glutamine and ammonia metabolism. In D. Hliussinger and H. Sies, (Eds.), Glutamine Metabolism in A4ammalian Tissues. Berlin: Springer-Verlag, 1984. Pp. 78-91. Woodside, K. H., and Mortimore, G. E. Suppression of protein turnover by amino acids in the perfused rat liver. J. Biol. Chem. 247: 6474, 1972. Parilla, R., and Goodman, M. N. Nitrogen metabolism in the isolated perfused rat liver. Biochem. J. 138: 341, 1974. Ballet, F., Chr&ien, Y., Rey, C., and Poupon, R. Norepinephrine: a potential modulator of the hepatic transport of taurocholate. A study in the isolated perfused rat liver. J. Pharmacol. Exp. Ther. 240: 303, 1987. P&ii, A. R., Wert, J. J., and Mortimore, G. E. Multifunctional control by amino acids of deprivation-induced proteolysis in liver. J. Biol. Chem. 257: 12114, 1982. P&G, A. R., and Mortimore, G. E. Requirement for alanine in the amino acid control of deprivation-induced protein degradation in liver. Proc. Natl. Acad. Sci. USA 81: 4270, 1984.
DE BANDT
ET AL.: AMINO
ACID
METABOLISM
7. Wright, C. E., Tallan, H. H., and Lim, Y. Y. Taurine: Biological update. Annu. Rev. Bivchem. 55: 427,1986. 8. Cynober, L., Coudray-Lucas, C., Ziegler, F., and Giboudeau, J. High performance ion-exchange chromatography of amino-acids in biological fluids using chromakon 500. Performance of the apparatus. J. Autom. Chem. 7: 201,1985. 9. Mortimore, G. E. Effect of insulin on release of glucose and urea by isolated rat liver. Amer. J. Physiol. 204: 699, 1963. 10. Frolich, J., Scholmericb, J., Hoppe-Seyler, G., Maier, K. P., Talke, H., Schollmeyer, P., and Gerok, W. The effect of acute uraemia on gluconeogenesis in isolated perfused rat liver. Eur. J. Clin. Invest. 4: 453, 1974. 11. Ball&, C., and Jungermann, K. Control of urea production, glutamine release and ammonia uptake in the perfused rat liver by the sympathetic innervation. Eur. J. Biochem. 158: 13, 1986. 12. Harper, A. E., Miller, R. H., and Block, K. P. Branched-chain amino acid metabolism. Annu. Rev. Nutr. 4: 409,1984.
IN PERFUSED
RAT
LIVER
13
13.
Mortimore, G. E., and Mondon, C. E. Inhibition by insulin of valine turnover in liver. J. Biol. Chem. 245: 2375,197O.
l4
Lacy, W. W. Effect of acute uremia on amino acid uptake and urea production by perfused rat liver. Amer. J. Physiol. 216: 1300,1969.
’ 15.
Mondon, C. E., and Mortimore, G. E. Effects of insulin on amino acid release and urea formation in perfused rat liver. Amer. J. Physiol. 212: 173,1967.
16.
Felig, P. Amino acid metabolism 933,1975.
17.
Woodside, K. H., Ward, W. F., and Mortimore, G. E. Effects of glucagon on general protein degradation and synthesis in perfused rat liver. J. Biol. Chem. 249: 5458, 1974.
18.
Glinsmann, W. H., and Mortimore, G. E. Influence of glucagon and 3’,5’-AMP on insulin responsiveness of the perfused rat liver. Amer. J. Physiol. 215: 553, 1968.
in man. Anna Reu. Biochem. 44:
