In a previous paper (Barnes & Magee, 1954) dimethylnitrosamine was shown to cause acute centrilobular necrosis of the liver in the rat, mouse, guinea pig, rabbit and dog, when given in doses of the order of 25 mg./kg. body weight. No gross abnormality was found in organs or tissues other than the liver. In long-term feeding experiments, when dimethylnitrosamine was added to the diet of rats at a level of 50 parts per million, a high incidence of malignant hepatic tumours occurred, with no evidence of primary tumours elsewhere. In similar experiments with rabbits no tumours were produced (Magee & Barnes, 1956).It appears that dimethylnitrosamine is a highly specific liver poison, and the present paper describes experiments which were performed in the hope of throwing some light on the mechanism of this specificity. A polarographic method for the estimation of dimethylnitrosamine in tissues and body fluids has been developed in this Laboratory (Heath & Jarvis, 1955) and this method was used to study the distribution of the compound in the animal body after administration, and its excretion.
METHODSThe animals used were albino rats of the Porton strain, mice of the C. 57 Chester Beatty strain and commercially obtained albino mice, and cross-bred rabbits. The rats and mice were maintained on M.R.C. diet 41 (Bruce & Parkes, 1949) and the rabbits on M.R.C. diet 18 (Bruce, 1947).Dimethylnitrosamine, when given orally, was diluted with water and administered by a metal cannula attached to a syringe. For parenteral administration dilution was with 0 9 % NaCl. Intravenous injection was by tail vein in rats and by ear vein in rabbits, both procedures being without anaesthesia, unless otherwise stated. The compound was estimated in tissues and body fluids by the polarographic method of Heath & Jarvis (1955). In determinations of the total body content of dimethylnitrosamine in rats the animals were killed by intraperitoneal injection of Na pentobarbital (Veterinary Nembutal, Abbott) and immediately frozen by immersion in liquid N.. The rats were then transferred to a bucket and fragmented by percussion with a heavy brass ram-rod. The fragments were disintegrated further in 2-5% (w/v) sulphosalicylic acid by means of an Atomix blender (Measuring and Scientific Equipment Ltd., London, S.W. 1), and the resulting tissue suspension was made up to a known volume and centrifuged. Dimethylnitrosamine was estimated in a suitable portion of the supernatant fluid. A similar procedure was used for whole mice except that they were killed by cervical dislocation, and it was found possible to disintegrate them in the blender without preliminary freezing. In experiments on the excretion of dimethylnitrosamine, rats were placed in conventional metabolism cages constructed ofmetal coated with bakelite. Urine and faeces were collected separately by means of a glass separator. Total hepatectomy was performed on rats by Dr K. K. Cheng, using a two-stage technique (Cheng, 1951). The second stage of the operation on the previously prepared anim...