Modification of the effects of blood on amino acid metabolism by intravenous isoleucine

GASTROENTEROLOGY

1991;101:1613-1620

Modification of the Effects of Blood on Amino Acid Metabolism by Intravenous Isoleucine NICOLAAS E. P. DEUTZ, PETRONELLA L. M. REIJVEN, MAY C. F. BOST, CHARLES L. H. VAN BERLO, and PETER B. SOETERS Department of Surgery and Centralized Limburg, Maastricht, The Netherlands

Animal

Facilities,

Biomedical

Center,

University

of

The absence of isoleucine in the hemoglobin molecule has been suggested to contribute to increased urea production after a blood meal. To unravel the underlying mechanism, the effects of isoleucine infusion after blood ingestion in the healthy pig were studied. The isoleucine dose was chosen to induce an arterial isoleucine increase comparable to those observed for leucine or valine after blood ingestion. For the experiments, 10 female overnight-fasted pigs (20-25 kg) received 250 mL bovine erythrocytes intragastrically 1 week after catheter implantation for measuring hepatic, splanchnic, portal-drained viscera, and hindquarter fluxes of amino acids, urea, and ammonia. After the administration of erythrocytes, isoleucine or saline was administered IV for 6 hours. The data obtained show that the increase in arterial levels of urea and almost all amino acids was significantly greater in the control group (P < 0.001)than in the isoleucine group. The net efflux of nearly all amino acids by the portaldrained viscera increased significantly less (P < 0.001) in the isoleucine group. The liver uptake of amino acids increased after the blood meal, but the difference was not significant except for glutamine (P < 0.001). Hindquarter amino acid net influx increased marginally. Splanchnic urea production increased more in the control group than in the isoleucine group (P < 0.05). The data strongly suggest that lV administration of isoleucine enhanced the biological value of a blood meal, possibly by promoting amino acid retention in the portal drained viscera.

meal, the plasma isoleucine concentration decreased to extremely low levels, while the concentrations of leucine, valine, and several other amino acids reached extremely high levels. Furthermore, an increase in plasma urea concentration was observed, which was much higher than when isonitrogenous amounts of normal protein present in regular food were given. These metabolic changes were suggested to underlie the pathophysiology of the metabolic disturbances observed in patients suffering from gastrointestinal bleeding. The observed high concentrations of the branchedchain amino acids (BCAAs) Val and Leu in the administered blood meal is expected to lead to stimulation of BCAA oxidation (2). Because of the low levels of Ile in the blood meal, this process will lead to a decrease in concentration of the BCAA Ile in plasma, a phenomenon called BCAA antagonism (2-6). In this study we aimed to determine whether the observed metabolic changes induced by enterally administered blood were caused by the low content of Ile in the administered blood protein. To this end, Ile was infused IV together with the enterally administered blood in healthy pigs. During the experiment, ammonia, urea, glucose, lactate, and amino acid fluxes were measured across the portal-drained viscera (PDV), liver, and hindquarter. The data obtained show that IV administration of Ile, inducing arterial concentrations comparable to Leu and Val, reversed the effects of a blood meal on amino acid metabolism in the healthy pig. Furthermore, it was shown that the intestine plays a major

E

Abbreviations usedin thispaper: BCAA, branched-chain amino acid; PAH, p-aminohippuric acid; PDV, portal-drained viscera. o 1991 by the American Gastroenterological Association 6016-5005/91/$3.QQ

nteral administration of large amounts of blood or has been reported to affect amino erythrocytes acid metabolism in the healthy pig (1).After a blood

1614 DELJTZETAL.

role in the changes enteral administration

GASTROENTEROLOGYVol. lOl.No.6

in amino acid of blood.

metabolism

after

Materials and Methods Animals Female crossbred (Yorkshire x Dutch Landrace) piglets (n = 10; weighing 20-25 kg), about 8 weeks old, were used. They were housed individually inside galvanizedbar runs (2 x 3 m) equipped with an automated watering device and polyvinyl chloride-coated floors. Before surgery, the animals were allowed to adapt to this bar run for at least 1 week. The piglets were fed sow feed (pregnant sow feed; Landbouwbelang, Roermond, The Netherlands; 12% protein) to obtain the normal growth rate of the crossbred piglets. Surgical Procedure The method used is a modification of the model described by van Berlo et al. (1,7). Animals were fasted for 24 hours. Thirty minutes after premeditation (azoperone, 8 mg/kg), anesthesia was induced with thiopental sodium (25 mg/kg) via an ear vein and ampicillin (50 mg/kg IV) was administered as antimicrobial prophylaxis. Anesthesia was maintained after oral endotracheal intubation with an N,O, 0, (1:2), and halothane (0.8%) mixture. During surgery, Ringer’s lactate (500-1000 mL) was administered via the ear vein. First, through a midline abdominal incision, 80-cm-long catheters (ID, 0.040 in; OD, 0.070 in; Tygon, Westvaco, Cleveland, OH) were inserted via the left and right iliac circumflex profunda arteries into the abdominal aorta with the tip t5 cm above the bifurcation (Al)and above the level of the right renal artery (A2), respectively. Second, catheters were placed via the left and right iliac circumflex profunda veins into the inferior caval vein with the tip 25 cm above the bifurcation (VI) and above the level of the right renal vein (V2), respectively. Third, catheters were placed into the colonic vein (C), exposed between two coils of the colon using a purse-string suture in the vein, into the portal vein with the catheter tip in the liver hilus (P) via the splenic vein using a purse-string suture, and into the left hepatic vein (H) by direct puncture between the left and middle liver lobe as described by Imamura and Clowes (8). The positions of the tips of all catheters were checked during operation using x-ray radiography and an iodine-containing contrast solution. After proper positioning of the catheters was ascertained, the insertion places of the catheters and the stitches used to fix the catheters at the abdominal wall were glued with cyanoacrylate. Finally, a gastrostomy catheter was placed. All catheters were led through a small tunnel in the abdominal wall to the skin. The outside portions of the catheters were connected to a small closable valve (Bently Lab, Uden, The Netherlands). The abdominal incision was closed with continuous 2-Mersilene (Ethicon, Norderstedt, Germany) in the fascia, continuous 2-O catgut subcutaneously, and 2-Mersilene in the skin. Each pig wore a canvas harness to protect the catheters and to allow easy handling of the animal. The animals were then allowed to

recover. Twice a week, the catheters were flushed and filled with heparinized saline (50 U/mL).

Experimental

Protocol

Experiments were started at least 1 week after surgery because postoperative changes in blood flow and amino acid fluxes were found to be stabilized within this period (9). During the recovery phase, food intake returned to normal and the animals showed a normal weight gain. All animals used appeared healthy and without signs of infection. The experiments were performed in unrestrained, conditioned, conscious animals standing in a small movable cage (0.9 x 0.5x 0.3m) 16 hours after their last meal. After sampling, the catheters were filled with the heparinized (50 U/mL) saline.

Solutions p-Aminohippuric acid (PAH) (A 1422; Sigma, St. Louis, MO), 25 mmol/L (5 g/L), made isoosmolar by addition of 137.5 mmol NaCl/L and adjusted to pH 7.4 by addition of l.ON NaOH, and L-isoleucine (Ajinomoto, Raleigh, NC), 135 mmol/L (17.709 g/L), made isoosmolar by addition of 82.5 mmol NaCl/L and adjusted to pH 7.4 by addition of l.ON NaOH, were filter sterilized.

Experimental

Design

To permit continous flow calculation, an infusion of PAH with a rate of 30 mL/h per catheter was started on the morning of the experiment (8 AM) and continued throughout the experiment into the C and Al catheters after an initial bolus of 5 mL PAH solution. After 1 hour, steadystate PAH concentrations were obtained (data not shown). Hereafter, blood samples were taken from the A2, P, H, and Vl catheters. Bovine erythrocyte packed cells (one batch) were prepared by a single centrifugation step and stored at -20°C. Before use, 250 mL of erythrocytes at room temperature were administered through the gastric catheter (t = 0 hours), and the catheter was then flushed with 20 mL water. Subsequently, the piglets received an infusion either with 2 mL . h-’ . kg-’ body wt of the Ile solution (1.620 mmol kg-’ . 6 h-‘; 0.213 g. kg-’ . 6 h-‘) or with saline at the same infusion rate as saline in the V2 catheter (Figure 1). The dose of infused Ile (total, *40 mmol/pig over 6 hours) was chosen to reach an arterial Ile concentration similar to that of Leu and Val. Each piglet was assigned randomly to receive the saline solution (control group) or the Ile solution [treatment, Ile group) and vice versa 1 week later when the experiment was repeated. For 6 hours after administration of the erythrocytes, blood was sampled hourly from the A2, P, H, and Vl catheters and immediately stored on ice. Blood-Sample

Processing

Within 20 minutes after sampling, the samples were centrifuged at 4°C for 5 minutes at 8500g. Hematocrit was measured. For glucose and lactate determinations, 200 )*L of

December

BLOOD MEAL AND IV ISOLEUCINE IN THE PIG

1991

Female

pig

Branched-chain amino acid levels are the sum of Val, Leu, and Ile levels. Blood flow was calculated as (12) Infusion

lmplantatton

catheters >

Sampling

250

270

mM

ml

at

bovine

I

2

F1ow = [PAH],“,, -

1 week

i h postabsorptlve

16

ml/kg

erys

h

1.4.

bw/h

0.9% NaCl

lie

Sampling

t=O

at

t= 1,2,3,4,5,6h

Figure 1. Experimental set-up. Every pig received, in a cross-over design, both the Ile solution and the control solution.

whole blood was centrifuged in NaF-containing tubes. Ammonia, urea, and amino acid levels were determined in plasma. For amino acid (sulfosalicylic acid) and PAH (trichloracetic acid and whole blood) analysis, the sample was deproteinized. All samples were frozen in liquid nitrogen and stored at -70°C until analysis. Biochemical

Analysis

Ammonia, glucose, lactate, and urea concentrations were determined spectrophotometrically by standard enzymatic methods on a centrifugal analyzer system (Cobas Bio; Roche Diagnostics, Hoffmann-La Roche, Basle, Switzerland) using commercial kits. Urea values were corrected for ammonia. Plasma amino acids were determined using a fully automated high-performance liquid chromatography system after precolumn derivatization with o-phthaldialdehyde as described previously (10). For PAH determination, the method according to Brun (11) was adapted for small volumes. In short, samples were deacetylated (lOO”C, 45 minutes) and measured (Cobas Bio; 465 nm). The amino acid contents of the bovine erythrocytes were determined in triplicate using an automated amino acid analyzer and lithium buffers after acid hydrolysis. Calculations the

cY-Amino-nitrogen sum of the amino

1615

concentration was calculated as acid concentrations measured.

rate . 1PAH infused]

P’AW,,,,. Ht,,,,,1Ht,w’

where [PAH infused] is the concentration of PAH in the infused solution, [PAHI,,,, is the concentration in blood after passing through the organ, and [P,4Hjllreis the concentration before. No significant changes between the two groups (P > 0.05) were observed in the hepatic, PDV, and hindquarter flows at any time interval (data not shown]. Fluxes were calculated using the mean flow values per time point per group. The PDV flux was calculated by multiplication of the portal-arterial (P-A2) difference with the portal flow. The splanchnic flux was calculated by multiplication of the hepatic-arterial (H-A2) difference and the liver flow. The liver flux was calculated by subtracting the PDV flux from the splanchnic 11ux. The hindquarter flux was calculated by multiplying the hindquarter flow with the caval vein-arterial (Vl-A2) difference. The apparent absorption coefficient of a particular amino acid, being the total net amount of a given amino acid in the portal blood divided by the total amount containing the blood meal, is calculated as total PDV flux of the amino acid over 6 hours divided by the amino acid content of the blood ingested. Because of occlusion of some Iof the catheters, the flux data presented are from fewer animals, as indicated in the figure legends. Statistical Analysis Results are presented as means t SEM. Levels of significance were set at P < 0.05 unless stated otherwise. For statistical analysis (131, the following tests were used: analysis of variance (ANOVA) for effects in time and comparison of differences between control and Ile groups, and the Mann-Whitney test (NPAR) for individual comparison of the group means. Part of the data are used to generate the figures presented. To keep the report legible, all data are kept on file and can be obtained from the corresponding author. Results Arterial Levels After gastric administration of blood, the plasma Ile concentration decreased from 159 to 33 kmol/L in arterial blood. With IV infusion of Ile, the arterial ILE concentration [Figure 2) increased significantly (P< 0.001) to a level similar to that of Leu and Val in the control group. The Ile infusion induced significantly lower increases in Tyr, Leu, cx-amino-nitrogen (Figure 3), Val, Asn, Ser, Thr, His, Arg, Met, Phe, Trp, and Lys concentrations (data not shown). The increases in the arterial concentrations of the metabolites ammonia (P < 0.05) and urea (P < 0.001) were also reduced (Figure 4). However, Ile infusion did not

1616

DEUTZ ET AL.

GASTROENTEROLOGY

1500

A 1200

_L 900

-

600

-

_z-

/ 300

?? ’

- ’ / ‘-0.

,i-o-

’

2

1

0

.--

..-

Control

o-

0 0

/

2-

1’

-*-

+Ile

-~3

4

5

6

200

Vol. 101. No. 6

The splanchnic urea efflux (Figure 7) did not increase in contrast to the control group, although in the Ile group a significantly higher level at onset was seen (P < 0.05). The liver uptake of lactate (data not shown) decreased significantly (P < 0.001) in both groups, although less in the Ile group (P < 0.001). No changes were observed in the fluxes of other amino acids and of glucose. The amino acid PDV fluxes observed showed a pattern similar to that of the splanchnic fluxes (data not shown). Because the liver fluxes were only marginally different, it appeared that the PDV fluxes largely determine the splanchnic amino acid fluxes.

BI (is

150 ? ‘0

&

\ 100

Hindquarter d

Control

\

50

K,_

-

o-----

o-o-

3

4

,

The substances not taken up by the splanchnic region are offered to the peripheral tissues. To study the metabolism of these metabolites in peripheral tissues, hindquarter fluxes were measured. After Ile infusion, uptake of Ile (data not shown) was greater (P < O.OOl), coinciding with a lower uptake of the

I

OL

0

Flux

1

2

Time

5

6

/n/

Figure 2. The arterial Ile concentration (A)after a blood meal in the pig without treatment (0, expanded in II; n = 10) and after Ile infusion (0; n = 9).

affect the arterial concentration of Gln (Figure 4), some amino acids, glucose, and lactate (data not shown). Portal Drained Viscera Flux Portal drained viscera fluxes were measured to assess intestinal uptake and metabolism of the blood meal. After the blood meal, a small net PDV efflux of Ile was observed, which changed into net uptake in the Ile group (P < 0.001) (data not shown). The flux (Figure 5) of most of the amino acids (see cx-aminonitrogen) and ammonia increased significantly with time in both groups (P < 0.001) but was significantly less (P < 0.01) in the Ile group, suggesting an enhanced retention by the enterocytes. No changes were observed in the PDV fluxes of Gln (Figure 5), Glu, glucose, and lactate (not shown).

0

1 -o-

2

3

Control

4 -*-

and Liver Flux

After the components of the blood meal are taken up and released by the intestine, the liver is the next organ in the metabolic pathway. Infusion of Ile caused a significantly greater liver uptake of Gln (Figure 6) and a larger Glu efflux (data not shown).

0

1

1

2

3 Time

6

+Ile

oL Splanchnic

5

4

5

’ 6

fW

Figure 3. Arterial Tyr (A),Leu (B) and *amino-nitrogen (C) concentrations minus Ile after a blood meal in the pig without treatment (0; n = 10) and after Ile infusion (0; n = 9). The concentration of the amino acids is significantly different (P < 0.001) between the control and the Ile groups.

December

BLOOD MEAL AND IV ISOLEUCINE IN THE PIG

1993

1617

0 o

12

3

4

5

6

0

12

3

4

5

6

-4

-o-

-*-

Control

I_-1 o

I -o-

+Ile

2

3

Control

4

5

--•-

+lle

6

Cl

500

200’ 0

’ 12

3

4

5

6

o

P < 0.05).

other BCAAs, Val and Leu (P < 0.05). However, total BCAA uptake was comparable between the two groups. During the whole experiment, the net uptake of Glu, Ser, Arg, and Tyr (not shown) and o-amino-nitrogen [Figure 8) was less in the Ile group (P < 0.05). The significant (P < 0.05) net ammonia uptake seen in the control group was not altered by Ile infusion. The flux of Gln (Figure 8) and several other amino acids, glucose, and lactate [data not shown) was also unaffected.

Absorption Coefficient After the blood meal, the calculated apparent absorption coefficient (Table 1) of the different amino acids was found to be in the order of 4O%-50%. Infusion of Ile caused a decrease to about 20%. Regarding Ile, it appeared that a net release of Ile occurred in the control group (value > lOO%), suggesting loss of Ile from the enterocytes themselves,

2

3

4

5

6

Time fhl

Time fitI Figure 4. The arterial ammonia (A),urea (B), and Gln (C) concentration after a blood meal in the pig, without treatment (0; n = 10) and after Ile infusion (0; n = 6). A significant difference between the control and Ile groups was observed for urea (t = 0 to t = 4-6 hours; P < 0.001)and ammonia (t = 0 to t = 6 hours;

I

Figure 5. The PDV flux of ammonia (A),Gln (II), and *aminonitrogen (C) after a blood meal in the pig without treatment (0; n = 6) and after De infusion (0; n = 7). A significant difference between the control and De groups was observed in ammonia flux (t = 0 to t = 4-6 hours; P < 0.01) and a-amino-nitrogen flux (t = 0 tot = 4 hours; P< 0.01).

whereas a net uptake from arterial blood occurred the Ile group (negative value).

in

Discussion This study shows that the earlier reported hypoisoleucinemia and increased arterial urea concentration after gastric administration of erythrocytes in the healthy pig can be reversed by concomitant IV administration of Ile in a dose inducing arterial Ile concentrations in the same range as Leu and Val in the untreated group. The total amount of Ile infused in 6 hours appeared to be in the same order as Val and Leu present in the blood meal given (40 mmol Ile vs. 51 mmol Val and 72 mmol Leu). The observed decrease in the arterial Ile concentration after the blood meal could be due to the almost complete lack of Ile in blood proteins and the stimulatory effect of the large amounts of Leu and Val in blood proteins on branched-chain ol-keto dehydrogenase activity in muscle and other organs (5), leading

1618

DEUTZ ET AL.

GASTROENTEROLOGY Vol. 101,

-30’ 1

0

’ 2

3

4

’ 6

5

4

No. 6

BI

--O-~

Control

-@-

file

20 -B/

r I

0 -o-

2

3

4 -*-

Control

5

6

flk

-20

L.-~LL--__ 0

-50

’ 1

0

2

3 rime

4

5

/M

Figure 6. The liver flux of ammonia

(A), Gln (II), and a-aminonitrogen (C) after a blood meal in the pig without treatment (0; n = 5) and after Ile infusion (0; n = 3). A significant difference between the control and Ile groups was observed in Gin flux (t = 0 tot = 3-6 hours; P < 0.001).

to increased oxidation of the BCAA Ile. Furthermore, a stimulation of protein synthesis may have lowered the Ile concentration. Infusion of Ile after the blood meal counteracted the arterial Ile concentration de-

-25

I

’ 0

1

2

3

Time

4

1

2

3

Time

’ 6

5

6

In/

Figure 7. The splanchnic flux of urea after a blood meal in the pig without treatment (0; n = 7) and after Ile infusion (0; n = 5). A significant increase (P < 0.05) was observed between t = 0 and t = 1 hour in the control group.

4

5

6

01)

Figure 8. The hindquarter flux of Gln (A) and a-amino-nitrogen (I?) after a blood meal in the pig without treatment (0; n = 5) and after Ile infusion (0; n = 6). o-Amino-nitrogen uptake was significantly (P < 0.05) less during the whole experiment in the Ile group.

crease and possibly reduced the described processes, resulting in a smaller increase of the arterial ammonia, urea, and amino acid concentrations. The appearance in the portal vein of almost all amino acids, except Gln, was lower in the Ile group. This could be the result of decreased gastric emptying, diminished or retarded absorption from the small intestinal lumen, or increased retention inside the small intestine cells. Reduced gastric emptying cannot explain our results because the uptake of blood proteins was almost completed during the experiments. A diminished or retarded absorption from the gut for neutral amino acids could be induced by Ile infusion because it could affect their transport systems (14-17) and therefore the resorption of amino acids (18). However, the appearance of all amino acids was decreased to a similar degree, making such a process unlikely. Reduced uptake by the small intestine would increase the total amount of protein in the lumen, leading to greater production of ammonia in the large intestine. Because PDV ammonia flux is lower after Ile infusion (Figure 5, after t = 3 hours), this possibility also seems to be excluded. Therefore, the data suggest that Ile infusion induces enhanced

December

BLOOD

1991

Table I, The Apparent Absorption Coeffcients Different Amino Acids Total (mmoli6 h)

250 mL erythrocytes mm01 Asp + Asn Glu + Gln Ser Pro Gly Thr His Ala Arg Tyr cys Val Met Ile Phe Leu Lys Sum AA

% of total

44.38

7.8

36.62 33.87 28.87 50.00 33.37 31.62 75.63 13.63 12.75 2.75 51.38 4.88 0.88 30.75 71.88 43.25

6.5 6.0 5.1

406.25

8.8 5.9 5.6 13.4 2.4 2.3 0.5 9.1 0.9 0.2 5.4 12.7 7.6

Control

Ile

of the

Absorption coefficient (%) Control

Ile

16.9

5.1

50

15

23.8 11.9 14.9 46.4 5.5 5.6

9.5 4.2 7.1 22.1 1.2 2.3

48 36 47 61 40 44

19 13 22 29 9 18

23.5 4.5 1.9 16.8 30.0 22.6 176.0

10.6 1.9 -8.4 8.7 14.6 11.2 76.4

46 92 >

21 39 <

55 42 52 43

28 20 26 19

NOTE. Data are calculated from the amino acid content of 250 mL bovine erythrocytes [?60 g protein (N . 6.25)] and the total PDV flux between t = 0 and t = 6 hours. Tryptophan is not measurable because of complete destruction by the hydrolysis process. Sum AA is the sum of Ser. Gly, Thr, His, Arg, Tyr, Val, Met, Phe, Leu, and Lys. The absorption coefficient is determined by dividing the total flux by the erythrocyte amino acid content.

amino acid retention in the small intestine, presumably by protein synthesis (l4,19). Protein synthesis in enterocytes by luminally delivered amino acids is considered an important mechanism in the uptake and metabolism of amino acids (20-22) and seems to occur in the rapidly dividing crypt cells of the mucosa (14,19,23-26). In the case of a protein meal lacking Ile, protein synthesis in the mucosal cells could be hampered by the imbalance of amino acids. Infusion of Ile replenishes the lacking Ile and may by this mechanism restore protein synthesis and lead to enhanced net (gut) protein retention. This process remains to be further established. Normalization of the imbalanced amino acid composition by Ile infusion decreased intestinal ammonia production after the blood meal, most likely by a reduction of the degradation of luminal or bloodborne amino acids, mainly Gln, from blood protein (1,27-29). Surprisingly, the uptake of amino acids except for Gln by the liver was not significantly different between the groups. Although it has been suggested that hepatic uptake of amino acids is controlled by the level of amino acid output from the gut (SO), the sensitivity of our model may not allow such conclusions. The net decrease in liver Gln uptake in the

MEAL AND IV ISOLEUCINE

IN THE PIG

1619

control group seems to indicate that the balance between periportal Gln deamination and perivenous Gln synthesis is altered in the direction of glutamine synthesis (31). This could be explained by enhanced deamination and urea synthesis of amino acids after a blood meal with more spillover of ammonia from this process (31). The urea synthesis also would increase the hepatic precursor pool for gluconeogenesis. Infusion of Ile reduces urea synthesis, minimizing the ammonia spillover from urea synthesis and reducing the increase of the precursor pool for gluconeogenesis. This could explain the effect of Ile infusion on the liver: greater uptake of Gln and a lowering of the decrease of the lactate uptake. The increase in arterial urea concentration after the blood meal seems to be caused by the enhanced liver urea production (32). The net hindquarter uptake of amino acids after the blood meal was low, and Ile infusion did not induce large differences, suggesting that hindquarter amino acid uptake for protein synthesis is only marginally affected by a protein meal. IHowever, a small net hindquarter uptake of BCAA unaltered by Ile infusion was observed. However, the fate of the BCAA is unclear (33) and cannot be deduced from the present experiments. The Ile infusion-induced lower hindquarter uptake of Val and Leu and the higher Ile uptake could be explained by competition between substrates in the BCAA transamination (2). In conclusion, the present experiments show that Ile, infused IV, together with an orally ingested, isoleucine-poor, blood protein, administered in a dose sufficient to increase arterial IlIe in the same range as Val and Leu, enhanced the in vivo biological value of the protein, possibly by promoting amino acid retention in the PDV.

References 1. van Berlo CLH, van de Boogaard

2. 3.

4.

5.

6.

AEJM, van der Heijden MAH, van Eijk HMH, Janssen MA, Bost MCF, Soeters PB. Is increased ammonia liberation after bleeding in the digestive tract the consequence of complete absence of isoleucin in hemoglobin? A study in pigs. Hepatology 1989;10:315-323, Harper AE, Miller RH, Block KP. Branched chain amino acid metabolism. Annu Rev Nutr 1984;4:409-454. Rogers QR. Spolter PD, Harper AE. Effect of leucine-isoleucine antagonism on plasma amino acic pattern of rats. Arch Biothem Biophys 1962;97:497-504. Oestremer GA, Hanson LE, Meade RJ. Leucine-isoleucine interrelationship in the young pig. J Anim Sci 1973;36:674678. Aftring RP, Block KP, Buse MG. Leucine and isoleucine activate skeletal muscle branched c:hain alpha-keto acid dehydrogenase in vivo. Am J Physiol 19&6:250:E599-E604. Edmonds MS, Gonyou HW, Baker DH. Effects of excess levels of methionine, tryptophan, arginine. lysine or threonine on growth and dietary choice in the piq. J Anim Sci 1987:65:179185.

1620

DEUTZ ET AL.

7. van Berlo CLH, van den Bogaard AEJM, Bost MCF, Soeters PB.

8.

9.

10.

11.

12.

13. 14.

A technique to study splanchnic metabolism in the unrestrained conscious pig. Lab Anim Sci 1988;38:463-466. Imamura M, Clowes GHA. Hepatic blood flow and oxygen consumption in starvation, sepsis and septic shock. Surg Gynecol Obstet 1975;141:27-34. Souba WW, Wilmore DW. Postoperative alteration of arteriovenous exchange of amino acids across the gastrointestinal tract. Surgery 1983;94:342-350. van Eijk HMH, van der Heijden MAH, van Berlo CLH, Soeters PB. Fully automated liquid-chromatographic determination of amino acids. Clin Chem 1988;34:2510-2513. Brun C. A rapid method for the determination of poraaminohippuric acid in kidney function tests. J Lab Clin Med 1951;37:955-958. Katz ML, Bergman EN. Simultaneous measurements of hepatic and portal venous blood flow in the sheep and dog. Am J Physiol 1969;216:946-952. Norusis MJ, ed. SPSS/PC+ V3.1 Base manual for the IBM PCIXTIAT and PSl2. Chicago: SPSS, 1988. Alpers DH. Uptake and fate of absorbed amino acids and peptides in the mammalian intestine. Fed Proc 1986;45:2261-

2.267. 15. Munck BG. Transport

16.

17.

18. 19. 20.

21.

22.

of neutral and cationic amino acids across the brush border membrane of the rabbit ileum. J Membrane Biol 1985;83:1-13. Munck BG. Transport of amino acids and non-alpha amino acids across the brush border membrane of the rabbit ileum. J Membr Biol1985;83:15-24. Rerat A, Corring T, Laplace JP. Protein digestion and absorption. In: Cole DJA, Boorman KN, Buttery PJ, Lewis D, Neale RJ, Swan H, eds. Protein metabolism and nutrition. London: Butterworths, 1976:97-138. Webb KE. Amino acid and peptide absorption from the gastrointestinal tract. Fed Proc 1986;45:2268-2271. Lund P, Williamson DH. Inter-tissue nitrogen fluxes. Br Med Bull 1985;41:251-256. Istfan NW, Ling PR, Bistrian BR, Blackburn GL. Systemic exchangeability of enteral leucine: relationship to plasma flux. Am J Physiol1988;254:R688-R698. Fern EB, Hider RC, London DR. Studies in vitro on free amino acid pools and protein synthesis in rat jejunum. Eur J Clin Invest 1971;1:211-215. Yu YM, Wagner DA, Tredget EE, Walaszewski JA, Burke JF, Young VR. Quantitative role of splanchnic region in leucine metabolism: [L-1-13C,l5N]leucine and substrate balance studies. Am J Physiol1990;259:E36-E51.

GASTROENTEROLOGY Vol. 101. No. 6

23. Karlstad

24. 25.

26. 27.

28.

29.

30.

31.

32.

33.

MD, DeMichele S, Istfan N, Blackburn GL, Bistrian BR. Effect of burn and first-pass splanchnic leucine extraction on protein kinetics in rats. Am J Physiol 1988;255:R303-R309. Alpers DH, Kinzie JL. Regulation of small intestine protein metabolism. Gastroenterology 1973;64:471-496. Gill M, France J, Summers M, McBride BW, Milligan LP. Mathematical integration of protein metabolism in growing lambs. J Nutr 1989;119:1269-1286. Low AG. Nutrient absorption in pigs. J Sci Food Agr 1980;31: 1087-1130. Sugarbaker SP, Revhaug A, Wilmore DW. The role of the small intestine in ammonia production after gastric blood administration. Ann Surg 1987;206:5-17. Windmueller HG, Spaeth AE. Intestinal metabolism of glutamine and glutamate from the lumen as compared to glutamine from blood. Arch Biochem Biophys 1975;171:662-672. Weber FL, Friedman DW, Fresard KM. Ammonia production from intraluminal amino acids in canine jejunum. Am J Physiol 1988;254:G264-G268. Bloomgarden ZT, Liljenquist J, Lacy W, Rabin D. Amino acid disposition by liver and gastroentestinal tract after protein and glucose ingestion. Am J Physiol1981;241:E90-E99. Haussinger D, Sies H. Hepatic glutamine metabolism under the influence of the portal ammonia concentration in the perfused rat liver. Eur J Biochem 1979;101:179-184. Dahms WT, Owens RP, Kalhan SC, Kerr DS, Danish RK. Urea synthesis, nitrogen balance, and glucose turnover in growthhormone deficient children before and after growth hormone administration, Metabolism 1989;38:197-203. Gelfand RA, Glickman MG, Jacob R, Sherwin RS, DeFronzo RA. Removal of infused amino acids by splanchnic and leg tissues in humans. Am J Physiol 1986;250:E407-E413.

Received July 24, 1990. Accepted July 11, 1991. Address requests for reprints to: Nicolaas E. P. Deutz, M.D., Ph.D., Department of Surgery, Biomedical Center, University of Limburg, P.O. Box 616, NL-6200 MD Maastricht, The Netherlands. Supported by the Dutch Liver Gut Foundation. The authors thank M. A. Jansen, H. M. H. van Eijk, A. Gijsen, and M. A. H. van der Heijden for analytical help. A. E. J. M. van de Bogaard, D. V. M., is acknowledged for help in the management and maintenance of the animals and C. H. C. Dejong, M.D., for fruitful discussions. The authors are grateful to W. A. Buurman, Ph.D., for critically reviewing the manuscript.