The reaction of mustard gas with nucleic acids in vitro and in vivo.

Mustard gas [di-(2-chloroethyl) sulphide] and analogous alkylating agents produce a wide range of biological effects (cf. review by Boyland, 1948). Some ofthese effects, such as induction ofmutations (Auerbach & Robson, 1946) and interference with mitosis and chromosome breakage (Darlington & Koller, 1947), stimulated interest in the reactions of alkylating agents with deoxyribonucleic acid. Previous studies of such reactions, mainly by physicochemical methods, have led to various suggestions regarding the chemical group or groups in nucleic acids involved. Chanutin & Gjessing (1946) showed by ultraviolet spectroscopy that the nitrogen mustards di-(2-chloroethyl)methylamine and tri-(2-chloroethyl)amine reacted with the bases of deoxyribonucleic acid, but the groups reacting were not specified. Elmore, Gulland, Jordan & Taylor (1948), on the basis ofelectrometric titration experiments, suggested that both basic and phosphate groups of deoxyribonucleic acid reacted with mustard gas. Press & Butler (1952) found that alkylation of deoxyribonucleic acid by di-(2chloroethyl)methylamine in sodium bicarbonate solution resulted in a decrease of primary amino nitrogen and of purine nitrogen precipitable as purine silver salt, but no specific sites of reaction were established. Wheeler, Morrow & Skipper (1955) concluded from experiments with methylated xanthines and mustard gas that alkylation of ring nitrogen atoms of purines could occur. Alexander, Cousens & Stacey (1957) postulated that phosphotriesters were produced by alkylation of deoxyribonucleic acid in neutral aqueous solution by a variety of agents. This was based on observations that alkylation was not accompanied by liberation of acid. Reiner & Zamenhof (1957), from experiments on the binding of protamine by deoxyribonucleic acid and paper chromatography of hydrolysed alkylated deoxyribonucleic acid, claimed that whereas dimethyl sulphate reacted exclusively with the purine moieties at N-7, diethyl sulphate reacted exclusively with primary phosphate groups. The only study of products formed by alkylation in vivo of nucleic acid is that by Wheeler & Skipper (1957). After administration of di-(2-chloroethyl)[14C]methylamine to rats and mice isotopic labelling of both ribonucleic acid and deoxyribonucleic acid was found, and it was shown that this was associated with the purine fraction. In most of the earlier chemical studies of alkylation of nucleic acids the conditions were very different from those which would occur in vivo, e.g. large concentrations of alkylating agents, and sometimes alkaline pH, were used. Further, the methods used were not such as to enable the sites of reaction to be specified. In view of this uncertainty and the fact that preliminary experiments (Lawley & Wallick, 1957; Lawley, 1957a, b) had indicated N-7 of guanine moieties to be the most reactive centre towards alkylation in nucleic acid, it was decided to reinvestigate this reaction with a typical alkylating agent. Mustard gas, di-(2-chloroethyl) [35S]sulphide, was chosen since it is known to be one of the most reactive of the alkylating agents and also is obtainable at a high specific radioactivity. The reaction of mustard gas with nucleic acid under mild conditions has therefore been studied chemically and compared with that resulting when tobacco mosaic virus, protoplasts of Bacillus megaterium, and the Ehrlich ascites tumour in the mouse are treated with this reagent.