The production of cancer by pure hydrocarbons—IV

Abstract

Tests for carcinogenic activity on about 140 compounds were recorded in earlier parts of this series (Part I, Cook, Hieger, Kennaway and Mayneord 1932; Part II, Cook 1932; Part III, Barry, Cook, Haslewood, Hewett, Hieger and Kennaway 1935). These showed that the most active compounds are methylcholanthrene and 3:4-benzpyrene, in the sense that these compounds gave tumours which appeared more rapidly than in the case of any other compound. The carcinogenic activity of these two compounds is also of interest on account of the facts that methylcholanthrene is obtained by chemical means from the naturally occurring bile acids, cholic and deoxycholic acids, and 3:4-benzpyrene is present in coal tar. Experiments now recorded have shown that a similar high order of carcinogenic activity is likewise shown by certain other substances containing the same ring systems as these two compounds, although the activity may be readily destroyed by slight modification of molecular structure. We have also extended our general studies of the relationship between molecular structure and carcinogenic activity among compounds related to 1:2-benzanthracene. The technique has been described already, and no further details need be given. Except where otherwise stated the compounds were applied twice weekly in 0·3% solution in benzene to the skin of mice. Cholanthrene Derivatives The parent hydrocarbon of this group, cholanthrene, has been prepared synthetically by several different methods, and it is evident from the results summarized in Table I and fig. 1 that cholanthrene is no less potent than methylcholanthrene in producing malignant tumours of the skin of mice, when applied in dilute solution in benzene. A similar conclusion concerning the relative potencies of these two compounds was arrived at by Shear (1936), who studied the production of sarcomas by inserting pellets of cholesterol containing the hydrocarbon into the subcutaneous tissue of mice. In the light of our previous results with homologues of 1:2-benzanthracene it would seem likely that the carcinogenic activity of methylcholanthrene is determined mainly by the fact that this hydrocarbon is a 5:6-substituted derivative of 1:2-benzanthracene. From this standpoint it is surprising that cholanthrene should be so highly active, for this hydrocarbon lacks the substituent at position 6 of the benzanthracene ring system, and it has been found among the simple homologues of benzanthracene that the 5-monosubstituted derivatives, though carcinogenic (see p. 355), are less effective than the 5: 6-disubstituted derivatives (Part III). Thus, the second point of attachment of the five-membered ring of cholanthrene, at position 10 of the benzanthracene ring system, appears to reinforce the influence of substitution at position 5 to a greater extent than is the case with a substituent at position 6. On this view it might be expected that activity would be shown by 10-monosubstituted benzanthracenes. The only such compounds hitherto examined are 10-isopropyl- and 10-benzyl-1:2-benzanthracenes, which were completely inactive. In our experiments with our recently synthesized 10-methyl- 1:2-benzanthracene, which unfortunately had to be abandoned at an early stage on account of an epidemic among the mice, we have been anticipated by Fieser and his collaborators (1936,1937), who report that both 10-methyl-1:2-benzanthracene and 5:10-dimethyl-1:2-benzanthracene show the high degree of carcinogenic activity characteristic of cholanthrene and methylcholanthrene. Therefore, substitution at 5:10 appears to be more effective in promoting carcinogenic potency than substitution at positions 5:6, irrespective of whether or no these substituents form part of a new ring.