Abstract
Marked inhibition of the growth of the Walker rat carcinoma 256 is produced by administration of 4-aminostilbene and of 4-dimethylaminostilbene. From similar experiments with the Crocker sarcoma 180, the carcinoma C63, and spontaneous mammary cancer, it appears that the growth inhibitory action of these compounds is much less pronounced in mice. In a dosage of 200 to 250 mg. /kg. in the rat, both 4-amino- and 4-dimethylaminostilbene produce (
a
) gastro-intestinal submucous haemorrhage most evident in the pyloric portion of the stomach, (
b
) haematuria, and (
c
) haemolymph changes, while the former compound induces methaemoglobinaemia in addition. Using the Walker rat carcinoma 256 as the biological test object, a series of derivatives of 4-aminostilbene has been examined to determine the relationship between the growth-inhibitory effect and chemical constitution. The great majority of the active compounds can be defined as stilbenes with a basic substituent, the position of which is of paramount importance; thus
o
-dimethylaminostilbene is very much less active than the
p
-isomeride, and the
m
-compound is completely inactive. A further essential feature is the ethylene bridge, since activity disappears when either of its hydrogen atoms is substituted, when the bridge is reduced, when it is extended to contain three or four carbon atoms, or when either methine group is replaced by a nitrogen atom. Compounds in which the ethylene bridge is absent, or is replaced by oxygen or sulphur, are also inactive, and activity is further dependent upon the
trans
configuration of the molecule about the ethylenic bond, and to a large extent upon a free
p'
-position. These and other facts have suggested the working hypothesis that one of the features required for activity is an unbroken conjugation of the amino group with both nuclei, enabling the compound to assume some dipolar quinonoid character, which depends, among other things, on the co-planar arrangement of the two benzene nuclei which characterizes the
trans
form of the stilbenes. When evidence concerning steric conditions in the molecules (obtained mainly from the ultraviolet spectroscopy of 4-dimethylaminostilbene and thirteen of its alkyl derivatives) is compared with the biological activities of these compounds, a close parallel is suggested between lack of growth-inhibitory power and buckling of the molecule. Thus in the 4-dimethylaminostilbene derivatives with substituents on the x- and ?-carbon atoms of the ethylenic double bond, or with methyl groups at two ortho positions in a phenyl group, steric factors reduce the planarity of the molecule, thus affecting the conjugation resonance characteristic of the whole molecule. These compounds had previously been found to be non-inhibitory. All the evidence from diagrams, models and spectra suggests that steric interference with the planar configuration of molecules in this series varies continuously from the planar 4-dimethylaminostilbene to the highly buckled a/?-diethyl derivative. Inhibitory activity within this series appears to depend upon a conjunction of such factors as molecular size and shape, and the apposition of a planar molecule to a hypothetical adsorbing surface. In view of the previously suggested connexion between growth-inhibitory activity and tumour production, a number of selected aminostilbenes has been tested for carcinogenicity. Of seventy-two rats exposed to the action of 4-amino-, 4-acetamido-, 4-dimethylamino-, or 4-diethyIaminostilbene, twenty-three developed a total of eight sarcomata and thirty distant tumours mostly comprising squamous keratinizing carcinomata of the external acoustic duct, mammary adenomata, and cholangiomata. In a second series, the compounds 4-amino-, 4-dimethylamino- and 2,-methyl- 4-dimethylaminostilbene, and 1-(4'-dimethylaminophenyl)-2-(T-naphthyl)ethylene, were tested in male and female mice and rats. Of 120 rats exposed, forty-eight developed a total of twenty sarcomata at the site of injection and fifty-one tumours in organs distant from the site of injection, while of 120 mice similarly treated only four sarcomata and two distant tumours were recorded in five. The nature and distribution of the distant tumours induced in the rat strongly suggest some common feature in the carcinogenic action of aminostilbenes and 2-acetamidofluorene. Other points discussed include (
a
) the association between haemolymphatic changes and carcinogenesis, (
b
) the possible significance of the nitrogenous analogue 2-(4'-dimethylaminostyryl) quinoline in relation to the carcinogenic action of ‘styryl 430’, (
c
) a comparison of the biological properties of 4-dimethylaminostilbene and 4-dimethylaminoazobenzene, and (
d
) the dependence for manifestation of the growth-inhibiting action of the aminostilbenes upon a sufficiently low intake of dietary protein.