Gut: first published as 10.1136/gut.18.9.688 on 1 September 1977. Downloaded from Gut, 1977, 18, 688-691 Can hepatic coma be caused by a reduction of brain noradrenaline or dopamine? L. ZIEVE AND R. L. OLSEN From the Department ofMedicine, Minneapolis Veterans Hospital, University of Minnesota, Minneapolis, and Department of Chemistry, Hamline University, St. Paul, Minnesota, USA SUMMARY Intraventricular infusions of octopamine which raised brain octopamine concentrations more than 20 000-fold resulted in reductions in brain noradrenaline and dopamine by as much as 90% without affecting the alertness or activity of normal rats. As this reduction of brain catechol- amines is much greater than any reported in hepatic coma, we do not believe that values observed in experimental hepatic failure have aetiological significance for the encephalopathy that ensues. Though catecholaminergic nerve terminals represent dopamine and noradrenaline had no discernible only a small proportion of brain synapses (Snyder et effect on the state of alertness of the animals. al., 1973), the reduction in brain dopamine or nor- adrenaline by the accumulation of false neuro- Methods transmitterssuch as octopamine or of aromaticamino acids such as phenylalanine or tyrosine has been ANIMALS suggested as a cause of hepatic coma (Fischer and Male Sprague-Dawley rats weighing between 300 and Baldessarini, 1971; Dodsworth et al., 1974; Fischer 350 were g prepared by the method of Peterson and http://gut.bmj.com/ and Baldessarini, 1975; Munro et al., 1975). The Sparber (1974) for intraventricular infusions of following data have been cited in support of this octopamine. Rats of 250-400 g were used as controls. hypothesis: a three to five-fold increase in rat brain All animals received regular rat chow and water ad octopamine in acute experimental hepatic coma libitum. After treatment and observation rats were (Fischer and Baldessarini, 1971; Fischer and James, killed by decapitation. Brains were removed within a 1971; Dodsworth et al., 1974), a five-fold increase in few seconds and frozen immediately in liquid nitro- urinary octopamine excretion in patients with hepatic gen. Care was taken to exclude the ventricles from coma (Fischer and Baldessarini, 1971), a significant the brain tissue that was analysed. on September 26, 2021 by guest. Protected copyright. correlation between serum octopamine concentra- tion and severity of hepatic encephalopathy in CHEMICALS patients (Lam et al., 1973; Manghani et al., 1975), a Octopamine hydrochloride, noradrenaline, and reported decrease in brain noradrenaline in rats with dopamine used as standards were obtained from acute experimental hepatic coma (Dodsworth et al., Sigma Chemical Co. The enzyme phenylethanol- 1974), and a reported improvement in hepatic amine-N-methyltransferase used in the assay for encephalopathy when aromatic amino acids are octopamine was also obtained from Sigma Chemical decreased in plasma relative to branched chain amino Co. Pargyline was obtained from Abbott Labora- acids (Fischer et al., 1976). In order to test the tories. Dihydroxybenzylamine was prepared from hypothesis that reduction in brain dopamine or nor- dimethoxybenzylamine by treatment with hydro- adrenaline per se will cause coma, we cannulated the bromic acid (Refshauge et al., 1974). S-adenosyl-L- lateral ventricles of normal rats and infused octo- methionine (methyl-3H) was obtained from New pamine in amounts sufficient to raise the brain England Nuclear. Other chemicals were of usual octopamine more than 20 000-fold. The resulting reagent grade. marked reduction in the concentrations of brain APPARATUS Address for reprint requests: Dr L. Zieve, VA Hospital, Minneapolis, MN 55417, USA. For the determination of dopamine and nor- adrenaline a liquid chromatograph with electro- Received for publication 24 February 1977 chemical detection (LCEC) was assembled using a 688 Gut: first published as 10.1136/gut.18.9.688 on 1 September 1977. Downloaded from Can hepatic coma be caused by a reduction of brain noradrenaline or dopamine ? 689 Milton-Roy pump (model 396-31), a 5 ul slider octopamine. Three of the rats were infused with a injection valve and a 2 mm x 50 cm glass column, all total of 3-2 mg octopamine as described above. Three from Laboratory Data Control Co., Riviera Beach, additional rats were given a continuous intra- Fla. The column was packed with Dupont Zipax ventricular infusion of octopamine over a period of SCX, a strong cation exchanger. A model LC-2 20 hours. The rate of infusion was approximately 0 5 electrochemical system (Bioanalytical Systems, Inc., /l per minute, and the total dose of octopamine was West LaFayette, IN 47906) was used for detection. 12 mg. During the infusion process, whether the material ASSAYS being infused was saline or the octopamine solution, To determine the catecholamines the rat brains were the spontaneous activity of the animals decreased weighed frozen, homogenised with 5 0 ml cold 0 05 and some appeared lethargic. This effect was more M perchloric acid containing 0 03 M sodium prominent in those rats pretreated with pargyline. bisulphite, and then centrifuged at 15 000 x g for 20 However, as soon as the infusion was stopped, the minutes. A portion of each supernate was reserved rats perked up promptly, appearing as alert and for octopamine analysis if appropriate. The re- active as before the infusion was begun. All were alert mainder was used for analysis by LCEC according just before decapitation. The effect of the infusion to the procedure of Refshauge et al. (1974), in which was less evident in the rats infused continuously for the supernate is treated with aluminium oxide which 20 hours, as the infusion rate--and thus intra- selectively adsorbs catecholamines. These were ventricular pressure-was lower. subsequently stripped from the alumina with acetic acid and separated chromatographically. Results Octopamine was determined in frozen, weighed rat brains essentially according to the radio-enzy- The brain octopamine concentrations in normal matic method of Saavedra (1974), the principal uninfused rats and in rats receiving intraventricular modification being the use of commercially prepared infusions of octopamine are given in the Table. phenylethanolamine-N-methyltransferase (50,ug pro- tein in each incubation tube). In the case of rats which had not been treated to increase brain levels of Table Brain octopanmine concentrations after octopamine the brains were homogenised in 20-30 intraventricular infusions of octopamine* http://gut.bmj.com/ volumes of 0-02 M Tris buffer (pH 8 6) containing No. Brain octopamine pargyline (50 tsg/ml) as specified by Saavedra. The (nglg) proteins were precipitated by heating in a boiling Control, no infusion 6 3-4 ± 0-6 water bath for three minutes and centrifuging at Control, saline infusion 5 45 ± 0 5 20 000 x g for 20 minutes. In the case of treated rats Octopamine infusion 3-2 mg over 6 hr 6 115 000 ± 27 000 the supernates from the catecholamine procedure Pargyline + oct. infusion contained high levels of octopamine and were 3-2 mg over 6 hr 3 507 000 ± 153 000 diluted with the 0 02 M Tris Pargyline + oct. infusion on September 26, 2021 by guest. Protected copyright. simply appropriately 3 847 000 ± 63 000 buffer mentioned above. 12 mg over 20 hr *Mean ± SEM. EXPERIMENTAL PROCEDURE Rats were infused intraventricularly with a 20 mg/ml solution of octopamine hydrochloride which was Infusion of saline in place of the octopamine approximately isotonic, slightly acidic but unbuf- solution had no effect on the brain octopamine fered. In one group of six rats, the infusions were concentration, nor did it alter the brain dopamine or carried out four times for 20 minutes each at 90 noradrenaline concentrations. Infusions of 3-2 mg minute intervals. The infusion rate was approxi- octopamine resulted in a 20 000-fold increase in mately 2 ,ul/min so that a total dose of 3-2 mg brain octopamine. Pretreatment with pargyline, a octopamine was administered over a period of 290 monoamine oxidase inhibitor that in itself leads to minutes. Controls were either untreated or were raised brain octopamine levels, resulted in a further infused on the above schedule with isotonic saline. marked increase in brain octopamine when 3 2 mg Seventy minutes after stopping the last infusion (six was infused intraventricularly, and an increase to the hours ffom beginning the first infusion) the animals fantastic levels of 847 000 ± 63 000 ng/g when 12 were killed by decapitation and brains assayed as mg was infused over 20 hours. In all instances the described above. Another group of six rats was pre- rats had recovered their preinfusion spontaneous treated with pargyline, 10 mg intraperitoneally, five activity and alertness before they were killed. In minutes before the intraventricular infusion of contrast, infusion of 0 5 mg (0-01 mmol) NH4CI into Gut: first published as 10.1136/gut.18.9.688 on 1 September 1977. Downloaded from 690 L. Zieve and R. L. Olsen the cerebral ventricle resulted in convulsions within a not associated with loss of consciousness in any of few seconds and coma within a minute or two. these rats. The brain noradrenaline (NA), and dopamine Acute experimental hepatic coma was produced in (DA) concentrations resulting from these enormous six rats by ligation of the hepatic artery 18-24 hours increments in brain octopamine are given in the after a portacaval shunt as described previously Figure. The control values shown are based upon a (Zieve et al., 1974). The rats lapsed into coma nine to combination of values obtained in four rats receiving 17 hours after acute massive hepatic necrosis result- an intraventricular saline infusion (NA 440 ± 37 and ing from the artery ligation and were killed as soon DA 1003 ± 107 ng/g) and six rats receiving no as coma became definite as judged by their failure to infusion (NA 364 ± 30 and DA 844 ± 50 ng/g).