Very recently, during the continuing study of the metabolism of acetaminofluorene, it was found that N-hydroxylation was a major metabolic pathway, and thus a new chelating metabolite was isolated. It may be the intermediate in the formation of the I- or 3-hydroxy derivatives, which are also chelating agents. This was the first isolation of an N-hydroxylation product in vivo. Precaution must be taken in looking for N-aro-matic hydroxylamines, for these compounds are not stable in air. Phenylhydroxylamine decomposes within two hours after it has been synthesized. N-hydroxy-N-acetylaminofluorene is carcinogenic as is its copper chelate, both of the compounds inducing tumors at lower levels than did the original acetylamino-fluorene. Other cases of N-hydroxy-compounds' having an effect on tumor are known. In the field of anti-tumor drugs, N-hydroxy-formamide was shown to be equally effective as N-methylforma-mide in retarding growth of the Ehrlich ascites tumor; the expected metabolic product of N-hydroxymethyl formamide was ineffective.

An argument against the chelation thesis is that some ortho-methoxy amines are carcinogens but not chelating agents. The extent to which demethylation takes place is not known, but it has been established that the splitting of methyl aromatic ethers is a possible biosynthetic pathway for some compounds. Thus xanthurenic acid, the 4,8-dihydroxyquinaldic acid, is not carcinogenic although it is a chelating agent of the 8-hydroxyquinoline family. The 8-methyl ether of xanthurenic acid is weakly carcinogenic. It is quite possible that the ether is more easily absorbed and then transported to a vital organ where a demethylation enzyme is encountered. The 8-hydroxyquinoline derivative may be liberated in situ.

Miscellaneous Carcinogens

Numerous other miscellaneous carcinogens are known, and a few will be mentioned. A perusal of the literature reveals that compounds like thiourea are carcinogenic, and compounds with this structure easily form complexes with trace metals. It would, therefore, be of interest to see if thioacetamide (XLIII) and thiobenzamide (XLIV), which also form complexes with bivalent ions, can also induce tumors.

Carcinogenic compounds of diverse structure always seem to be chelating or complexing agents. They may be polyamides, or polyalcohols or phenols like tannins, or even chemotherapeutic agents like nitrogen mustards or TEM. Isonicotinic acid hydrazide is also a chelating agent, and recently it has been shown to reduce the hepatoma incidence of DMAB, but, at the same time, has been reported to induce pulmonary tumors in mice after feeding experiments lasting two months; a later study brings doubts to these findings.

Diazomethane, a non-chelating but an alkylating agent, is carcinogenic; it may be liberated from dimethylnitrosoamine and N-methyl-N-nitrosourethane by tissue hydrolysis. This may explain the carcinogenicity of these two compounds. Once liberated, diazomethane combines with sulfhydryl groups. The diazomethane-cysteine derivative may bind metals; N-methyl-N-nitrosourethane has a chelation structure, and a prediction can be made that dimethylnitrosoamine metabolizes to a hydroxymethyl derivative.

β-Propiolactone and some monofunctional ethyleni-mines are potential chelating agents, and also carcinogenic.