insufficiencygutl993;34:1112-1119 intestinalglutamineandammoniametabolism...

Gutl993;34: 1112-1119

Intestinal glutamine and ammonia metabolismduring chronic hyperammonaemia induced by liverinsufficiency

Cornelis H C Dejong, Nicolaas E P Deutz, Peter B Soeters

AbstractDuring liver insufficiency, besides porta-systemic shunting, high arterial glutamineconcentrations could enhance intestinalglutamine consumption and ammonia genera-tion, thereby aggravating hyperammonaemia.To investigate this hypothesis, portal drainedviscera (intestines) fluxes and jejunal tissueconcentrations of ammonia and glutaminewere measured in portacaval shunted rats witha ligated bile duct, portacaval shunted, andsham operated rats, seven and 14 days aftersurgery, and in normal unoperated controls.Effects of differences in food intake wereminimised by pair feeding portacaval shuntedand sham operated with portacaval shuntedrats with biliary obstruction. At both timepoints, arterial ammonia was increased in thegroups with liver insufficiency. Also, arterialglutamine concentration was raised in ali

operated groups compared with normalunoperated controls. At both time points,ammonia production by portal drained viscerawas reduced in portacaval shunted rats withbiliary obstruction, portacaval shunted, andsham operated rats compared with normalunoperated controls, and no major differenceswere found between these operated groups. Atday 7 in ali operated groups glutamine uptakeby portal drained viscera was lower than innormal unoperated controls, but no majordifferences were found at day 14. Theseexperiments show that ammonia generation byportal drained viscera remains unchanged inrats with chronic liver insufficiency despitealterations in arterial glutamine concentrationsand intestinal glutamine uptake. The hyper-ammonaemia seems to be mainly determinedby the portasystemic shunting.(Gut 1993; 34: 1112-1119)

Department ofSurgery, University ofLimburg, Maastricht,The NetherlandsC H C DejongN E P DeutzP B SoetersCorrespondence to:Dr N E P Deutz, University ofLimburg, Department ofSurgery, Biomedical Center,PO Box 616, NL-6200 MDMaastricht, The Netherlands.

Accepted for publication24 November 1992

Ammonia is still considered to be of crucialimportance in the pathogenesis of hepaticencephalopathy.' The gut and kidney are

generally considered to be the most importantsites of ammonia production.2'- In the gut, themain sources of ammonia production are thebacterial breakdown of urea2367 and the mucosalutilisation of glutamine as an energy sub-strate.2 37I In the physiological situation virtuallyall ammonia generated in the gut is immediatelycleared by hepatic urea synthesis, and thushepatic venous ammonia concentrations are

lower than arterial concentrations. Therefore, inhealthy subjects, systemic ammonia concentra-tions are probably mainly set by the interactionbetween renal ammonia production and ammonia

consumption by other organs. During livercirrhosis induced by chronic liver disease thissituation changes drastically, because ammoniagenerated in the intestines bypasses hepaticclearance by intra or extrahepatic portasystemicshunts.910 Also, the diminished urea synthesisand glutamine synthetase capacity9' may reduceammonia detoxification in the liver. Thesecombined factors probably cause the systemichyperammonaemia found during liverinsufficiency.

It has been suggested that systemic hyper-ammonaemia during chronic liver failure leads toenhanced ammonia detoxification via alternativepathways.'2 It is generally believed that theenergy dependent synthesis of glutamine fromequimolar amounts of glutamate and ammonia isthe most important of these pathways.'3 Thisglutamine synthesis in organs containing gluta-mine synthetase, mainly skeletal muscle andbrain,'4 probably contributes to the raisedarterial glutamine concentrations during hyper-ammonaemia induced by liver insufficiency.'2Because intestinal glutamine uptake is concen-tration dependent,7 these increased arterialglutamine concentrations in turn could increaseintestinal glutamine utilisation and subsequentliberation of ammonia.A considerable amount of research has been

performed concerning several aspects ofammonia and glutamine metabolism duringacute and chronic liver insufficiency (see forexample,4612I 15-18 and also our laboratory'9 20).Despite the well known effects of food intake onnitrogen metabolism in general and more specifi-cally on glutamine and ammonia metabolism,2'few studies have been performed under con-ditions of controlled food intake during hyper-ammonaemia induced by liver insufficiency.The hypothesis underlying the present study

was that during chronic liver insufficiency,besides portasystemic shunting, high arterialglutamine concentrations could aggravatehyperammonaemia by enhanced intestinalglutamine consumption and ammonia genera-tion. To test this hypothesis the exhange ofammonia, glutamine, and several other aminoacids across the portal drained viscera (gut) aswell as jejunal tissue concentrations weremeasured in two models of hyperammonaemiainduced by chronic liver insufficency in rats:portacaval shunting and portacaval shuntingcombined with bile duct ligation, as well as inappropriate controls. Portacaval shuntingcombined with biliary tract ligation was recentlysuggested to be a reliable, new model for hyper-ammonaemia induced by chronic liver in-sufficiency and to induce a more pronounced

112

on March 27, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.34.8.1112 on 1 A

ugust 1993. Dow

nloaded from

http://gut.bmj.com/

Intestinal glutamine and ammonia metabolism during chronic hyperammonaemia induced by liver insufficiency

clinical degree of hepatic encephalopathy thanportacaval shunting alone.22 To minimise effectsof differences in food intake, the portacavalshunted and the sham operated groups were pairfed with the anorectic portacaval shunt andbiliary obstruction group.

Materials and methods

ANIMALSMale Wistar rats (n=56, weight 300 (SEM 25)g)were used throughout. They were housed understandard conditions (12/12 hour light/dark cycle)and received standard food and water ad libitumuntil surgery. Rats were maintained andhumanely cared for according to the recom-mendations of the guide for the care and use oflaboratory animals as applied in our institute.

GROUPSFour groups were studied: (1) PCSBDL group,portacaval shunting (PCS) combined with bileduct ligation (BDL) in one surgical session; (2)PCS-PF group, Portacaval shunting; (3) PFgroup, laparotomy and manipulations as in PCS-PF rats, but without shunting (sham operation);(4) normal group, sampling without previoussurgery; normal unoperated control rats receiv-ing food ad libitum.

Before surgery, the rats were randomlyassigned to one of the groups. Surgical pro-cedures were of equal duration in PCS-PF andPF rats and their individual PCSBDL mate.After resuscitation from surgery, PCSBDL,

25-

20

C

V 10 -00

0)

400

ao) 20

0Vm 200

0 5 10 15

Time (days)

Figure 1: Food intake (upper panel) and body weight (lowerpanel) in PCSBDL (-), PCS-PF (0) andPF (0) ratsduring 14 days ofpairfeeding. Curves are means (SEM).n= 14 to 16 in thefirst seven days and six to eight rats betweendays 7 and 14.

PCS-PF, and PF rats were placed in metaboliccages. The PCSBDL rats had free access tostandard pellet food. In these rats, daily foodintake was recorded and this amount wasadministered to fixed PCS-PF and PF mates tominimise effects of differences of food intake(pair feeding: PF). In the PCSBDL, PCS-PF,and PF groups blood was sampled seven or 14days after surgery (see later). Body weight wasmeasured at the intervals indicated in Fig 1. Allgroups were allowed ad libitum water.

BEHAVIOURBefore they were used in experiments and beforesampling, the behaviour of all rats was studiedduring a 5 minute period to estimate the degreeof hepatic encephalopathy. Specific attentionwas paid to appearance, presence of lethargy,spontaneous locomotor activity and exploratorybehaviour, presence of tiptoe gait, toeing out,and arched back, reaction to tail pinching, andpresence of ataxia. Because the PCSBDL ratswere jaundiced, this staging procedure could notbe performed blinded.

SURGERYSurgery and sampling were performed underether anaesthesia. Portacaval shunting was bythe button technique.923 Biliary obstruction wasachieved by ligating the common bile duct nearthe liver and near the duodenum and removingthe piece between the ligatures. In PCS-PF andPF rats the bile duct was manipulated but notdissected, to avoid stenosis.

SAMPLINGBefore sampling, rats were fasted overnight.Sampling procedures were performed as des-cribed elsewhere.'92'24 Briefly, rats wereanaesthetised 30 minutes before sampling andthe portal vein and a tertiary branch of thesuperior mesenteric vein were cannulated withneedle bearing microcatheters fixed in place withcyanoacrylate adhesive. The right commoncarotid artery was catheterised with polyethylenetubing (PE 50). For flow determinations, after apriming dose, a 5 mM iso-osmolar pH adjustedp-aminohippuric acid (PAH) solution was con-tinuously infused into the mesenteric veincatheter (25 ,ul min-'), allowing for a 20 minuteperiod to attain steady state concentrations.2025Blood was slowly and simultaneously aspiratedfrom the portal vein and carotid artery and put inheparinised tubes on ice. Finally, about 10 cmdistal to Treitz's ligament, a 10 cm piece ofjejunum was excised and frozen in liquidnitrogen.24 Rats were killed by an overdose of theanaesthetic. These procedures permit simul-taneous determination of blood flow and arterio-venous concentration difference (flux) across theportal drained viscera (intestines),'9"2 24 as well asthe tissue concentration measurements.

ANALYSISBlood samples were kept on ice duringprocessing. Plasma for determination of aminoacids, ammonia, and urea was obtained by

1113



j.com/


ugust 1993. Dow

nloaded from

http://gut.bmj.com/

Dejong, Deutz, Soeters

TABLE I Behaviour

Day 7 Day 14DayONormal PF PCS-PF PCSBDL PF PCS-PF PCSBDL

"Gerbil" appearance 0/10 4/8 5/8 8/8 q 1/8 5/8 6/6Toeing out (hindlimbs) 0/10 0/8 0/8 6/8 0/8 6/8 r 6/6Tiptoe gait 0/10 0/8 1/8 3/8 0/8 1/8 6/6"Duck" gait 0/10 0/8 1/8 3/8 0/8 2/8 6/6Arched back 0/10 0/8 1/8 3/8 0/8 3/8 5/6Ataxia 0/10 0/8 0/8 2/8 2/8 6/8 6/6

All items were scored on a dichotomic basis (present or absent). Data are rats with item/total. Fischer'sone tailed exact test: v PF qp<0*05; 'p<0-01; 'p<0-001. v PCS-PF tp<005, 'p<001, 'p<O 001.

centrifugation of whole blood at 4°C for fiveminutes at 8800 g. Packed cell volume wasobtained with a microfuge. Plasma and tissuewere stored at -70°C. For tissue ammonia andamino acid determinations, 5% (w/v) sulpho-salicylic acid extracts were prepared.'9 Tissuewater content was determined as describedpreviously.20 Ammonia, urea, lactate and glucosewere determined enzymatically,'9 and aminoacids by high performance liquid chromato-graphy.26 P-Aminohippuric acid (PAH) wasdetermined as described previously.'9 Totalbilirubin was measured in plasma with com-mercial kits (Roche, Basel, Switzerland). Thecoefficient of variance for all determinations wasbelow 4%. 1926

CALCULATIONSPlasma PAH concentration was calculated from

whole bloodPAH concentrations with the packedcell volume. Plasma flow was calculated from theformula FPDV=I/([PAH]p - [PAHIA) whereFPDV is the portal drained viscera plasma flow(ml.min-'), I is the infused PAH (iimole.min-'),and [PAH]p and [PAH]A are the plasma PAHconcentrations (>tM) in the portal vein andcarotid artery respectively. Portal drained viscerafluxe (nmole.100 g body weight-'.min-') wascalculated as plasma flowx portal venous-arterialconcentration difference. Positive figures meannet efflux, negative values net uptake. Fractionalglutamine extraction was calculated as thearterial-portal vein concentration differencedivided by the arterial concentration ((A-PV)/A)*100%). Urea values were corrected forammonia. a Amino nitrogen (aAN) wascalculated as the sum of the individual aminoacids measured.26

STATISTICSCalculations were with the SPSS/PC+ statisticalsoftware package, version 3.1 (SPSS Inc,Chicago, IL, USA). Data are presented as means(SEM), p values <0'05 were considered signifi-cant. The tests used were: analysis ofvariance forgroup effects (ANOVA) and time effects withingroups (ONEWAY); the non-parametric Mann-WhitneyU test for differences between groups atspecific time points; Wilcoxon's non-parametrictest for differences from zero; and Fischer'sexact test for differences in encephalopathy

TABLE II Arterial concentratons

Day 7 Day 14Day ONormal PF PCS-PF PCSBDL PF PCS-PF PCSBDL

Bilirubin 1-3 (0 1) 1-2 (0-1) 2-4 (0 4) h 119 9 (7 6) 1-3 (0-1) 2-6 (0 3) - 112-9 (7-3) cfhGlucose 7'9(0'3) 9-4(0-7) 8-9(0 5) 7-1 (0 2) 9 4(0 4) 8-0(0 4) 7-9(0 3) cLactate 3-4(0 3) 4-5 (1-0) 9 0(0 7) ' 10-5 (0 8) 53 (0-7) 9-2 (0-9) c 99 (08) cAmmonia 59(6) 59(12) 199(19) h 1% (20) 68(7) 181(26) ch 168 (25)'-Urea 7-7(0-4) 7-1(0-4) 6-1(0-4) 6-1(0-2) ' 7-7(0-2) 64(03) " 63(03) chGlutamate 82(8) 70(7) 79(14) 55(5) ' 61(7) 66(7) 40(4) b"kPGlutamine 554 (12) 592 (20) 644 (23) 713 (40) 604(24) 609 (17) 599 (27)Glycine 285 (13) 270(16) 265 (19) 278 (23) 276 (15) 252 (10) 248 (15)Citrulline 43(2) 41(2) 45(5) 55(3) b 42(1) 47(2) 54(5) ddgAlanine 301(13) 264(21) 393(46) 6 409(39) h 263(25) 385 (19) c 333(21) "Taurine 153 (9) 169(9) 266 (29) ' 510(24) 203 (11) ' 215 (16) 439 (31)BCAA 404 (17) 394 (26) 337 (37) 318 (27) K 394 (18) 354 (20) 273 (17)AAA 138 (7) 125 (9) 167(10) ' 221 (10) 131 (7) 173 (8) c 186 (6) -hpacAmninonitrogen 3016(84) 2831(91) 3201 (170) 3593(156) 2835 (95) 3069(104) b 3030(106)

Data are expressed as means (SEM) in FM, except urea, glucose, and lactate (LAC) which are expressed in mM. BCAA=branched-chain amino acids (leucine, isoleucine, and valire); AAA=aromatic amino acids (tyrosine and phenylalanine). ANOVA for group effects;t=7 to t= 14 days: v PF op<0 05, hp<0-01, 'p<0-001; v PCS-PF dp<0-05, 'p<0-01, 'p<0-001. Mann-Whitney U test for differencesbetween groups: v PF sp<005, 'p<0-01, ip<0-001; v PCS-PF ip<005, kp<0-01, 'p<0-001; v normal -p<005, 'p<0-01, p<0-00l.Wilcoxon test; tnot significantly different from zero. ONEWAY procedure for time effects with groups pp<O-05.

TABLE III Metabolitefluxes and plasma flows in portal drained viscera

Day 7 Day 14Day 0Nornal PF PCS-PF PCSBDL PF PCS-PF PCSBDL

Flow 3-1(0-9) 0-8(0-2)n 2-3(1-4) 1-7(0-9) 2-4(1-2) 2-7(1-1) 2-3(1-0)Ammonia 413(62) 99(33)' 221(47) 160(37)' 295(74)P 328 (81) 260(140)Glutamate 20 (14)1 5 (1) -1 (15)' 13 (3) 40(9)' 16 (7) " 34(8) rGlutamine -486(49) -127(12) -351(48)h -312 (19)i -566 (53) r -386 (64) -308 (31) h"Glycine 47(19) 19(8) 33(19)t 3(7) t 118(17) p 57(12) - 4(20) -h

Citrulline 51(9) 10(2) n 45 (14)6 31(4) h 67 (10) r 57 (12) 34(14)Alanine 169 (57) 74(20) 215 (69) 123 (14) 403 (45)' 256 (57)' 105 (52)t"hTaurine 30(15) t 11(9) t 9(30) t -28(6) 83(31) P 72(35) -50(27) t-hkBCAA -26 (31) t 14 (9) t 27 (27) t -6 (4) t 75 (24)' 44 (23) t -11(12)t"hAAA -11 (15) t 2 (3) t 5 (9) t -4 (5) t 19 (6) P 8 (5)' -19(7) "",a Amino nitrogen -315 (180)' 16 (46)' -24 (174) -229(35) 399 (101) P 186 (126)' -285 (159)"h,FEGLN 28-1(2-5) 27-9(2-5) 23-7(3-7) 26-3(1-9) 39-5(3-1) "P 23-6(3-9) h 22-9(2-8) hi

Data are means (SEM) in nmole. 100 g body wt-'.min-', except plasma flows, which are in ml. 100 g body wt-'.min-'; FEGLN=fractionalglutamine extraction; for other abbreviations and significance symbols, see table II.

1114



j.com/


ugust 1993. Dow

nloaded from

http://gut.bmj.com/

Intestinalglutamine and ammonia metabolism during chronic hyperammonaemia induced by liver insufficiency

TABLE IV jejunal concentrations and water content

Day 7 Day 14Day 0Normal PF PCS-PF PCSBDL PF PCS-PF PCSBDL

%H20 76-9 (0 6) 77 5 (0 7) 77-6 (0 6) 76-9 (0 4) 77-3 (0 5) 78-0 (0 9) 77-1(0 6)Ammonia 1436 (83) 1213(69) 1688(146) 1681(153) 1391(91) 1541 (83)b 1634(152) bAspartate 752 (47) 950 (50) 853 (60) 862 (42) 865 (38) 749 (52) 738 (53) -

Glutamate 2579(74) 2628(75) 2872(146) 2853(96) 2683(61) 2533(145) 2839(111) -

Glutamine 693(46) 778(34) 827(16) 870 (35) - 693 (25) 740(48) 856(56)Glycine 1112 (46) 1071 (62) 1171(164) 958 (70) 1220 (67) 1035 (45) ' 919 (40)Citrulline 171(10) 177(6) 192(20) 199(11) 190(9) 185(12) 215(14)'Alanine 1041(52) 1022 (75) 1180 (96) 1171(76) 1268 (93)P 1147 (51) 1011 (72)'Taurine 14821 (645) 15666(745) 17105 (661) 17641(518)' 17142 (469) 16957(802) 18292(339) bBCAA 510 (24) 506(44) 542 (58) 348(25) 682 (73) P 534 (78) 314(9) 'AAA 219 (25) 198 (21) 300 (38) g 272 (19) 262 (27) 287 (33) - 234 (14)atAminonitrogen 23404(773) 24311(663) 26662(771)' 26551(480)' 26657(614)P 25655(834) 26777(519)'

Data are expressed as mean (SEM) in ptmole.kg wet weight'. For abbreviations and significance symbols, see table II.

300

200-

100

0L

750

500

250

2000

1500 [LT

1000

01 0 70 7 14

Time (days)Figure 2: Ammonia: arterial concentrations (upper panel;pM), portal drained viscera fluxes (middle panel; nmole. 100g body wt'.min-'), andjejunal tissue concentrations (lowerpanel; ptmole. kg wet wr') in normal control rats, shamoperated pairfed rats (PF), portacaval shunted rats (PCS-PF), and in rats with portacaval shunts and bile duct ligation(PSCBDL). Data are means (SEM). n=6 to 10 per group.

stages (one tailed, binomial testing). Signifi-cances are indicated in Tables I-IV.

Results

GENERALFood intake and body weight (Fig 1) were similarin PCSBDL, PCS-PF, and PF rats. Specificallyin PCSBDL rats, no signs of systemic or localinflammation or impaired wound healing werefound. These rats seemed to be more encephalo-pathic as evidenced by abnormalities in gait,decreased spontaneous locomotor activity, andexploratory behaviour (Table 1). Total arterialbilirubin concentrations (Table II) wereincreased in PCS-PF compared with PF rats,as noted previously,27 and in PCSBDL ratscompared with PCS-PF rats. In bothhyperammonaemic groups, arterial glucoseconcentrations were decreased and lactate con-centrations were raised in comparison with PFrats (Table II). Portal drained viscera plasmaflow (Table III) was decreased in the PF group atday 7 when compared with normal unoperatedcontrol rats, but no further differences in portalplasma flow were found between the PCSBDL,PCS-PF, and PF groups.

AMMONIAArterial ammonia concentrations were raisedequally in PCSBDL and PCS-PF rats on bothday 7 and day 14. No major differences inammonia production by portal drained viscerawere found between PCSBDL, PCS-PF, and PFrats, and ammonia production by portal drainedviscera in the hyperammonaemic groups neverexceeded that in normal control rats. In fact, itwas lower in all operated groups at day 7compared with normal unoperated controls.Although no major changes occurred in ammoniarelease by portal drained viscera ammonia con-centrations in jejunal tissue were raised in bothhyperammonaemic groups (Fig 2; Tables II-IV).

GLUTAMINEArterial glutamine concentrations were alwayshigher in PCSBDL, PCS-PF, and PF than innormal rats. Also, at day 7 arterial glutamineconcentrations were raised in PCSBDL com-pared with PF rats, but no differences werefound at day 14. At day 7, glutamine uptake by

._

._

0EE

x

>0~

c0CU

C

4)

cJC)0

1115



j.com/


ugust 1993. Dow

nloaded from

http://gut.bmj.com/


I Normal

PF

PCS-PF

PCSBDL

800

T700 F

a1)

a)

CU>

40

0)600-

500 110,

500

450

400

350

300

250

500

500

400

a1) Xc :CU04

300

200

100

0

T T

fn

T

Figure 3: Glutamine:arterial concentrations(upper panel; FM), portaldrained viscera fluxes(middle panel; nmole. 100g body wr'.min-'), andjejunal tissue concentrations(lower panel; pmole.kg wetwt') in normal control rats,sham operated pairfed rats(PF), portacaval shuntedrats (PCS-PF), and in ratswith portacaval shunts andbile duct ligation(PCSBDL). Data aremeans (SEM). n=6 to 10per group.

1000 r

c0

Coca

0

0

900

800

700

1000

H0 7 14

Time (days)

portal drained viscera was enhanced in bothhyperammonaemic groups compared with shamoperated animals, mainly caused by a reducedportal vein plasma flow in the PF group. Nomajor differences were found at day 14. Althougharterial glutamine concentrations were alwayshigher in the hyperammonaemic groups,glutamine uptake by portal drained viscera neverexceeded that in normal controls. Simul-taneously, jejunal glutamine concentrationswere raised in PCSBDL compared with PF ratsat both time points. The changes in glutamineand ammonia handling make it of special interestto look at some key metabolites intimatelyrelated to the intestinal breakdown of glutamine- namely, alanine, glutamate, and citrulline (Fig3; Tables II-IV).

ALANINE

Arterial alanine concentrations were raised inboth chronic liver insufficiency groups at bothtimes. Alanine efflux from portal drained viscera

1500

o 1250

. T

) 1000

o(Io

( 750 L

100

0 7 14

Time (days)Figure 4: Alanine: arterial concentrations (upper panel;pM), portal drained viscerafluxes (middle panel; nmole.100 g body wr'. min'), andjejunal tissue concentrations (lowerpanel; imole. kg wet wt') in normal control rats, shamoperated pairfed rats (PF), portacaval shunted rats(PCS-PF), and in rats with portacaval shunts and bile ductligation (PCSBDL). Data are means (SEM). n=6 to 10 pergroup.

was decreased overall in the PCSBDL groupcompared with both PF and PCS-PF rats(ANOVA days 7 to 14). Also, at day 14 alaninerelease was lower in the PCS-PF group comparedwith PF rats. Despite these changes, no majordifferences in tissue alanine were found (Fig 4;Tables II-IV).

GLUTAMATEAt both time points arterial glutamate wasdiminished in PCSBDL rats compared with thatin PCS-PF and PF rats. Only minor amounts ofglutamine uptake could be accounted for byglutamate release, in agreement with publishedreports.7'28 Glutamate release by the portaldrained viscera was slightly lower in the PCS-PF

z

1116

as as



j.com/


ugust 1993. Dow

nloaded from

http://gut.bmj.com/

1117Intestinal glutamine andammonia metabolism during chronic hyperammonaemia induced by liver insufficiency

100

75

a) 50

25

0

0)

4-

coEco

&3

x

0

50

25

o

-25

Figure 5: Glutamate:arterial concentrations(upper panel; pM) portaldrained viscera fluxes(middle panel; nmole. 100 gbody wt' .min-'), andjejunaltissue concentrations (lowerpanel: pmole.kg wet wt') innormal control rats, shamoperated pair-fed rats (PF),portacaval shunted rats(PCS-PF), and in rats withportacaval shunts and bileduct ligation (PCSBDL).Data are means (SEM).n=6 to 10 group.

3250 r

c

0

r.)

0)

3000 [

2750 [

2500

1000

T

tissue were decreased in PCSBDL rats comparedwith all other groups. Arterial taurine concentra-tions were raised in PCS-PF compared with PFrats and in the PCSBDL group compared withall other groups. Taurine release by the portaldrained viscera in the PCS-PF and PF groupsreversed to uptake in the PCSBDL group,although taurine concentrations in jejunal tissuewere already raised (Tables II-IV).

0 7

Time (days)

14

group compared with all other groups, but no

further differences were found. Glutamateconcentrations in jejunal tissue were increased in

the PCSBDL compared with the PF group,

despite decreased arterial glutamate concentra-tions (Fig 5; Tables II-IV).

CITRULLINEArterial citrulline concentrations were raised inthe PCSBDL group compared with that in PCS-PF and PF rats. Citrulline production by the gutdid not show major differences between thegroups, although tissue citrulline concentrationswere raised in the PCSBDL group.

REMAINING AMINO ACIDS: MAIN FINDINGSArterial branched chain amino acid (BCAA)concentrations were decreased in both chronicliver insufficiency groups compared with the PFgroup at both time points. Although BCAAfluxes were not significantly different from zero

in most groups, BCAA concentrations in jejunal

DiscussionThe present experiments were designed to studyintestinal ammonia and amino acid metabolismduring hyperammonaemia induced by chronicliver insufficiency. Specifically, we wereinterested in whether raised arterial glutamineconcentrations during chronic liver insufficiencywould lead to enhanced intestinal glutamineuptake and subsequent ammonia release, therebyaggravating the already existing portasystemicshunting induced hyperammonaemia. To mini-mise the influence of differences in nutritionalstate, rats were studied under conditions of wellcontrolled food intake. Studies were performedone and two weeks after surgery, as we havepreviously shown that button portacaval shuntedrats recover from several of the metabolicdisturbances ofportacaval shunting after three tofour weeks, probably due to the development ofhepatopetal shunts.23

In the current experiments, the similar foodintake and body weight in PCSBDL, PCS-PF,and PF rats suggests that differences betweenPCS-PF and PF rats are due to portacavalshunting. Similarly, differences betweenPCSBDL and PCS-PF rats are related to theaggravated hepatic parenchymal injury andcholestatic jaundice, induced by prolongedbiliary obstruction.2 2

Arterial ammonia concentrations were raisedequally in PCS-PF and PCSBDL rats, comparedwith both PF and normal control rats. This wasfound at day 14, when intestinal ammoniaproduction was similar in normal, PF, PCS-PF,and PCSBDL rats, as well as at day 7 whenammonia liberation was decreased in all operatedgroups compared with normal rats. The intesti-nal ammonia release in normal unoperatedcontrol rats was in close aggreement with datapreviously reported by our group.'9 From thesedata it seems that the (degree of) hyperammon-aemia in both chronic liver insufficiency groupsis mainly caused by the existence of a porta-systemic shunt and not by enhanced intestinalammonia production. Although the PCSBDLgroup undoubtedly has more severe paren-chymal damage than the PCS-PF group, arterialammonia concentrations were similar. This couldeasily by interpreted as evidence that the residualliver function is still able to detoxify total bodyammoniagenesis in this PCSBDL model. In asteady state situation as in our experiments,however, arterial ammonia concentrations donot give information about the rate of wholebody ammoniagenesis and detoxification, exceptthat these antagonistic processes are of equalmagnitude. Thus ammoniagenesis in otherorgans (for example, the kidney) may be reducedor alternatively, as we have found previously

I



j.com/


ugust 1993. Dow

nloaded from

http://gut.bmj.com/


(unpublished results), urinary ammoniaexcretion may be enhanced in the PCSBDLgroup.Atday 7 ammonia production by portal drained

viscera was less in PF than in PCS-PF rats. Thisdeserves some specific comment. The decreasein ammonia production by portal drained viscerain thePF group was mainly caused by a decreasedportal vein plasma flow, because portal vein-arterial ammonia concentration differences (notshown) were similar in all operated groups. Toexclude the possibility ofan artificially decreasedflow by coincidence or technical failure, werepeated portal plasma flow measurements in asimilar series of five PF rats, treated and fedidentically, and found the same value. Thus wemust assume that the decreased plasma flow inthe PF group at day 7 is a true finding for whichwe do not have an adequate explanation.

In PF as well as PCS-PF and PCSBDL rats,surgery and seven days ofsemistarvation resultedin an increase in arterial glutamine concentra-tions. No differences between the groups werefound, however, on day 14, despite substantialdifferences in arterial ammonia concentrations.These findings suggest that the regulation ofarterial glutamine concentrations is more relatedto nutritional factors or to surgical trauma thanto hyperammonaemia or liver insufficiencythemselves.

Although arterial glutamine concentrationswere always higher in the PF, PCS-PF, andPCSBDL groups than in normal control rats,glutamine uptake by portal drained viscera in thePF, PCS-PF, and PCSBDL groups neverexceeded that in normal control rats. Theglutamine uptake in portal drained viscera innormal controls was similar to data recentlyreported by our group.'9 Also, although theraised arterial glutamine concentrations at day 7in PCSBDL rats compared with PF rats resultedin increased glutamine uptake by portal drainedviscera, this was not accompanied by a signifi-cantly increased intestinal release of ammonia.Finally, the fractional glutamine extraction ofabout 30% in normal controls, closely resemblingthat reported by Windmueller and Spaeth,78decreased in the hyperammonaemic groups. Thehypothesis underlying this study was that hyper-glutaminaemia during chronic liver insufficiencywould enhance intestinal glutamine consumptionand subsequent ammonia release, thus aggravat-ing the already existing portasystemic shuntinginduced hyperammonaemia. From the presentdata we conclude that this hypothesis does notapply to these hyperammonaemia models in therat.There were several other interesting findings.

Intestinal release of taurine in all other groupsreversed to uptake in the PCSBDL group. Thiscould indicate that in rats with the bile ductligated, taurine is not excreted in the bile andtherefore cannot be taken up from the luminalside, but instead is taken up from the blood. Nodata are available concerning the free taurineconcentration of bile,33 but taurine can beassumed to be present in considerable amountsin bile as bile salt conjugates.-' It is likely thatthese bile salts can function as a source forintestinal taurine uptake. The fact that intestinal

glycine release in PCSBDL rats was significantlydecreased provides additional evidence that sucha mechanism may be operative, because the partof bile salts not conjugated with taurine consistsof glycine conjugates.34 The final metabolic fatefor the taurine taken up remains to be clarified.The finding that intestinal glutamine uptake

was less in the hyperammonaemic and shamoperated groups than in normal control ratsmakes it of special interest to consider the fluxesof the most important nitrogenous products(other than ammonia) of intestinal glutaminebreakdown - namely, glutamate, alanine, andcitrulline.728 35 36 Alanine release from theintestines decreased at day 14 in both hyper-ammonaemic groups compared with shamoperated rats, which seems to be compatible withthe reduced intestinal glutamine uptake. Alsointestinal release of citrulline did not changeappreciably in hyperammonaemic rats. Thiscould mean that other sources are also importantprecursors for citrulline release. It seems as ifcitrulline release is kept constant to provideprecursors for renal arginine biosynthesis.28Glutamate fluxes across portal drained visceradid not differ significantly from zero in severalgroups and could only account for minimalamounts ofglutamine uptake. This is compatiblewith published data suggesting that glutamate,whether taken up from the luminal side orderived from mucosal glutamine breakdown, iscompletely utilised by the intestines.8The changes in jejunal tissue concentrations

are difficult to interpret. The increased tissueglutamine concentrations in both hyperammon-aemic groups suggest that glutamine breakdownin the glutaminase reaction is diminished, whichseems supported by the decreased glutamineuptake and fractional glutamine extraction. Thisdoes not seem compatible with the simul-taneously raised concentrations ofglutamate andammonia, the products of the glutaminasereaction. Because intestinal glutamine synthetaseactivity is low,'428 probably the only plausiblealternative explanation for increased jejunalglutamine concentrations is enhanced intestinalbreakdown of protein. This should lead toenhanced intestinal release of essential aminoacids, however, which was not found either.Thus presumably the changes reflect the endresult of a new equilibrium, in which raisedintestinal ammonia concentrations'3 partiallyinhibit mucosal glutaminase activity. Theglutaminase reaction, yielding energy for thegut,28 can then only be driven at the expense ofraised tissue glutamine concentrations, providedby uptake of blood derived glutamine. Also inthe light of this, the decreased tissue BCAAconcentrations in the PCSBDL group could bedue to enhanced BCAA transamination,providing glutamate and a ketoglutarate forenergy. Firm conclusions are, however, notjustified based on the present data.

In the present experiments, portacavalanastomosis combined with common bile ductligation was introduced as a model to studynitrogen metabolism during chronic liverinsufficiency. Construction of a portacaval anas-tomosis alone induces, besides portasystemicshunting of gut derived blood, a relative and

118



j.com/


ugust 1993. Dow

nloaded from

http://gut.bmj.com/

Intestinalglutamine and ammonia metabolism duringchronic hyperammonaemia induced by liverinsufficiency 1119

absolute reduction of liver mass. 15-17 37 Ureasynthesis capacity is limited in portacavalshunted rats, probably related to the reducedliver mass.'5"6 To these effects of portacavalanastomosis alone, the effects of chronic bileduct ligation were added. Prolonged commonbile duct ligation induces periportal fibrosis,portal hypertension, and portasystemic shunt-ing293' as well as metabolic dysfunction of theliver.303' Although the resulting jaundice is of adifferent aetiology from that commonlyencountered in cirrhotic patients, it is also chole-static and therefore introduces impaired cellmediated immunity and Kupffer cell dysfunc-tion3131 in the model of portacaval shunting withbiliary obstruction. Because portacaval shuntedrats with biliary obstruction showed subtle signspointing to more pronounced hepatic encephalo-pathy, we think that the portacaval shunt andbiliary obstruction model could be useful infuture studies on the pathophysiology of chronicliver insufficiency and the related hepaticencephalopathy.

In conclusion, these experiments show thatunder conditions of standardised food intake,ammonia generation by portal drained visceraremained unchanged in rats with chronic liverinsufficiency despite alterations in arterialglutamine and intestinal glutamine uptake. Thissuggests that arterial ammonia concentrations inthese rats are mainly determined by the existenceof portasystemic shunting and not by arterialglutamine concentrations, intestinal glutamineuptake, or ammonia liberation. Finally, porta-caval shunting combined with common bile ductligation seems to be a useful model to studynitrogen metabolism during mild hyperammon-aemia induced by chronic liver insufficiency inrats.

We thank Mr H M H van Eijk, Mrs M A H van der Heijden, andMrs M A Janssen for laboratory determinations.

1 Record CO. Neurochemistry of hepatic encephalopathy. Gut1991; 32: 1261-3.

2 Weber FL, Veach GL. The importance of the small intestinein gut ammonium production in the fasting dog. Gastro-enterology 1979; 77: 235-40.

3 WeberFL, FriedmanDW, Fresard KM. Ammoniaproductionfrom intraluminal amino acids in canine jejunum. Am JPhysiol 1988; 254: G264-8.

4 Imier M, Schlienger J-L, Chabrier G, Comte F. Origine rmnalede l'hyperammoniemie provoquee par un regime hyper-protidique chez le rat normal ou porteur d'une strictureportale. Gastroenterol Clin Biol 1983; 7: 740-5.

5 Imler M, Schlienger J-L, Chabrier G, Simon C. Arterialanmmonia changes of renal origin induced in the rat by acidand alkaline diets. Res Exp Med 1986; 186: 353-63.

6 Imler M, Schlienger J-L. The effect of chronic uremia onportal and systemic ammonemia in normal and portal-strictured rats. J Lab Clin Med 1979; 94: 872-8.

7 Windmueller HG, Spaeth AE. Uptake and metabolism ofplasma glutamine by the small intestine. J Biol Chem 1970;249: 5070-9.

8 Windmueller HG, Spaeth AE. Intestinal metabolism ofglutamine and glutamate from the lumen as compared toglutamine from blood. Arch Biochem Biophys 1975; 171:662-72.

9 Zieve L. Pathogenesis of hepatic encephalopathy. Metab BrainDis 1987; 2: 147-65.

10 Meijer AJ, Lamers WH, Chamuleau RAFM. Nitrogenmetabolism and ornithine cycle function. Physiol Rev 1990;70: 701-48.

11 Kaiser S, Gerok W, Haussinger D. Ammonia and glutaminemetabolism in human liver slices: new aspects on thepathogenesis of hyperammonaemia in chronic liver disease.EurJ3 Clin Invest 1988; 18: 535-42.

12 Gjedde A, Lockwood AH, Duffy TE, Plum F. Cerebral bloodflow and metabolism in chronically hyperammonemic rats:Effect of an acute ammonia challenge. Ann Neurol 1978; 3:325-30.

13 Newsholme EA, Leech AR. Biochemistry for the medicalsciences. New York: John Wiley, 1983.

14 Lund P. A radiochemical assay for glutamine synthetase, andactivity of the enzyme in rat tissues. Biochem J1970; 118:35-9.

15 Brewer TG, Berry WR, Harmon JW, Walker SH, Dunn MA.Urea synthesis after protein feeding reflects hepatic mass inrats. Hepatology 1984; 4: 905-11.

16 Steele RD. Hyperammonemia and orotic aciduria in porta-caval shunted rats. J Nutr 1984; 114: 210-6.

17 Benjamin IS, Engelbrecht GHC, Saunders SJ, van HoornHickman R. Amino acid imbalance following portal diversionin the rat. The relevance of nutrition and ofhepatic function.J Hepatol 1988; 7: 208-14.

18 Lockwood AH, McDonald JM, Reiman RE, Gelbard AS,Laughlin JS, Duffy TE, et al. The dynamics of ammoniametabolism in man. Effects of liver disease and hyper-ammonemia. J Clin Invest 1979; 63: 449-60.

19 Dejong CHC, Kampman MT, Deutz NEP, Soeters PB.Altered glutamine metabolism in rat portal drained visceraand hindquarter during hyperammonemia. Gastroenterology1992; 102: 936-48.

20 Dejong CHC, Kampman MT, Deutz NEP, Soeters PB.Cerebral cortex ammonia and glutamine metabolism duringliver insufficiency-induced hyperammonemia in the rat.J Neurochem 1992; 59: 1071-9.

21 Dejong CHC, Deutz NEP, Soeters PB. Inter-organ nitrogenexchange during prolonged starvation in the rat. Journal ofClinical Nutrition and Gastroenterology 1991; 6: 176-83.

22 Maddison JE, Dodd PR, Morrison M, Johnston GAR, FarrellGC. Plasma GABA, GABA-like activity and the brainGABA-benzodiazepine receptor complex in rats with chronichepatic encephalopathy. Hepatology 1987; 7: 621-8.

23 de Boer JEG, Oostenbroek RJ, van Dongen JJ, Janssen MA,Soeters PB. Sequential metabolic characteristics followingportacaval shunts in rats. EurSurgRes 1986; 18: 96-106.

24 Deutz NEP, Dejong CHC, Athanasas G, Soeters PB. Partialenterectomy in the rat does not diminish muscle glutamineproduction. Metabolism 1992; 41: 1343-50.

25 Dejong CHC, Deutz NEP, Soeters PB. A simple new methodfor repeated in vivo cerebral cortex flux measurement in rats.Lab Anim Sci 1992; 42: 280-5.

26 van Eijk HMH, van der Heiiden MAH, van Berlo CLH,Soeters PB. Fully automated liquid-chromatographic deter-mination ofamino acids. Clin Chem 1988; 34: 2510-3.

27 Ribeiro J, Nordlinger B, Ballet F, Cynober L, Coudray-LucasC, Baudrimont M, et al. Intrasplenic hepatocellular trans-plantation corrects hepatic encephalopathy in portacavalshunted rats. Hepatology 1992; 15: 12-8.

28 Windmueller HG. Glutamine utilization by the small intestine.AdvEnzymeRegul 1982; 53: 210-37.

29 Kountouras J, Billing BH, Scheuer PJ, Prolonged bile ductobstruction: a new experimental model for cirrhosis in therat. British Journal of Experimental Pathology 1984; 65:305-11.

30 GreveJW, Gouma DJ, Soeters PB, Buurman WA. Suppressionof cellular immunity in obstructive jaundice is caused byendotoxins. A study in germ-free rats. Gastroenterology 1990;98:478-85.

31 Dunn CW, Horton JW, Megison SM, Vuitch MF. Contri-bution of portal systemic shunt to Kupffer cell dysfunctionin obstructive jaundice. J SurgRes 1991; 50: 234-9.

32 Roughneen PT, Gouma DJ, Kulkarni AD, Fanslow WF,Rowlands BJ. Impaired specific cell-mediated immunity inexperimental biliary obstruction and its reversibility byinternal biliary drainage. JSurg Res 1986; 41: 113-25.

33 Folsch UR, Wormsley KG. The amino acid composition of ratbile. Experientia 1977; 33: 1055-6.

34 Huxtable RJ. Physiological actions of taurine. Physiol Rev1992; 72: 101-63.

35 Windmueller HG, Spaeth AE. Respiratory fuels and nitrogenmetabolism in vivo in small intestine of fed rats. J Biol Chem1980; 225: 107-12.

36 Windmueller HG, Spaeth AE. Identification of ketone bodiesand glutamine as the major fuels in vivo for postabsorptiverats small intestine. J Biol Chem 1978; 253: 69-76.

37 Dunlop DS, Kaufman H, Zanchin G, Lajtha A. Proteinsynthesis rates in rats with portacaval shunts. J Neurochem1984; 43:1487-9.



j.com/


ugust 1993. Dow

nloaded from

http://gut.bmj.com/

insufficiencygutl993;34:1112-1119 intestinalglutamineandammoniametabolism...

Documents