IN a comprehensive joint investigation of the action of various compounds upon the growth of experimental tumours, attention was paid to that heterogeneous class of substances which have from time to time been reported as producing effects upon the mitotic cycle ("le syndrome mitoclasique"1) either in animals or plants. Among these substances, the caryoclastic poisons of Dustin and his school2, ethyl phenylcarbamate ('phenylurethane') (I) and its ?.NH.CO.O.C2H5 I: Ethyl phenylcarbamate ('phenylurethane')derivatives were selected for special study, particularly on account of the results obtained in another investigation, carried out in the laboratories of Imperial Chemical Industries, Ltd., by Templeman and Sexton3., of the effect of arylcarbamic esters and related compounds upon cereals and other plant species. This study confirmed the observations of Lefevre4 on'the morphological effects of ethyl phenylcarbamate on wheat seedlings, and extended them to include oats, barley and rye. Lefevre had described these morphological changes as bulbous hypertrophies of the coleoptile, the mesocotyl and the root-tip, with, progressive retardation or even arrest of growth: the underlying cytological alterations he interpreted as a blocking of mitosis in pseudo-metaphases, leading to irregular formation of monstrous restored nuclei. In comparisons of the potency of some fifty related arylcarbamic esters and thiocarbamates, Templeman / \ NH.CO.O.CH.(CH3)2II: Isopropyl phenylcarbamate and Sexton found isopropyl phenylcarbamate (II) to be approximately three times as active, in this sense, as ethyl phenylcarbamate: they regarded its effect on nuclear division as probably similar to that of colchicine, and in view of the low concentrations sufficient to arrest growth, suggested the use of this compound in the eradication of graminaceous weeds. Among compounds possessing no such activity were included methyl carbamate, ethyl carbamate (ure-thane), and formanilide. Apart from the references already cited, there is of course an extensive literature dealing with the suppressive action of urethane, phenylurethane and other carbamates on growth-processes, as in bacteria, protozoa, sea urchin eggs, plant tissues, and animal tissues growing in vitro; while other effects studied include the curious property of urethano in inducing pulmonary adenomata in mice5, its action in increasing the potassium content of bronchial secretion in the cat6, its toxic action upon the rat liver7, and an alleged antagonism towards the action of sulphanilamide in luminous bacteria8. It is perhaps rather surprising, therefore, that so little attention appears to have been given to the action of urethane and related substances on the growth of tumours.In our earliest experiments, the administration of phenylurethane was seen to produce the following undoubted effects in animal tissues. First was observed a transient increase of the mitotic count in the crypts of Lieberkiihn in the mouse gut: so far as could be judged, this effect was not attributable to retardation of the mitotic cycle at any single phase, and it was clearly distinguishable from the characteristic effect produced in the same tissues by colchicine. Although Lefevre had claimed that phenylurethane produces mitotic effects exactly similar to those described for colchicine, our impressions are more in keeping with the recent conclusion of Deysson9: "le phenylure* thane et la colchicine sont actuellement les deux corps agissant sur les cineses de la fagon la plus voisine, et cependant nous venons de montrer qu'il existe de nombreuses differences dans la modalite" de leur action". Secondly, phenylurethane (and also isopropyl phenylcarbamate) was found to cause a significant retardation of the growth of spontaneous mammary cancer in the mouse, an effect which persisted only during administration, and passed off rapidly when the drug was withdrawn. Fig. 1 shows the retardation produced by phenylurethane in the growth of one such tumour, which had earlier been treated with aβ-dipiperonylethylamine hydrochloride, but with negative result, in another investigation (with Prof. G. A. K. Kon) of various diphenylethylamines and related substances stated by Lettre and Fernholz10 to mimic the action of colchicine.Thirdly, both ethyl and isopropyl phenylcarbamates produced a similar retardation in the growth of the Walker rat carcinoma 256. That there is no parallelism between these results and the findings of Templeman and Sexton in relation to cereals was indicated by the somewhat unexpected observation that all these effects are also produced by ethyl carbamate (urethane) itself (NH2.CO.O.C2H5), the inhibitory activity of this substance being actually greater than that of either of the phenylcarbamates when tested against the Walker carcinoma. Some indication of the degree of this effect is shown in the accompanying table. (It should, however, be pointed out that although urethane showed no effect on plants such as was obtained with the phenylurethanes, differences due to such factors as cell permeability or metabolic changes in the drugs in vivo have not been considered, and hence there may well be a common factor in the chemical mechanism underlying the biological effects produced in plants and animals.) EFFECT OF ETHYLCARBAMATE ON THE GROWTH OF THE WALKERRAT CARCINOMA 256. INDIVIDUAL TUMOUR WEIGHTS (GM.) 16 DAYS FOLLOWING IMPLANTATION.Control series Treated series (12 intra-peritoneal injections (12 intra-peritoneal injections of solvent over 14 days) of 50 mgm. ethylcarbamate in aq. solution over 14 days)39-7 8-7 34-8 8-032-3 6-7 32-0 4-330-9 4-0 27-1 3-927-0 3-8 24-2 3-322 -3 2 -9 17-9 0312-0 0-2 5-7 This action of urethane upon the growth of the Walker carcinoma was next found to be accompanied by a profound modification in the histological structure of the tumour, of a kind which had not thus far been encountered in very numerous comparable experiments representing a large number of compounds of other chemical types. From Fig. 2 (A and B) it will be seen that under the influence of urethane the characteristic cellular texture gives place to a more fibrous structure, with spindle-cells and a distinctly abundant stroma. While precise interpretation is difficult, the impression was gained that the alteration involves, in part at least, some attempted differentiation, and is not altogether to be ascribed to secondary changes involving the vessels and stroma only.Fig. l. SPONTANEOUS MAMMARY CANCER IN MOUSE. Although the growth effects above described were in no way dramatic, their reproducibility, and the interest of the circumstance that they might be brought about by a known substance as simple and as readily available as urethane, suggested the advisability of testing their action in advanced and inoperable or otherwise intractable cancer in the human subject. Clinical trials, using urethane and isopropyl phenylcarbamate, were accordingly startedFig. 2. HISTOLOGY OF THE WALKER RAT CARCINOMA G (A) NORMAL APPEARANCE, X 300; AND (B) AFTER 12 X 50 MGM. URETHANE IN 14 DAYS, X 270. in 1943, first at the Royal Cancer Hospital (Free), and afterwards in collaboration with the Christie Hospital and Holt Radium Institute, Manchester. The results of these early trials, mainly referring to carcinoma of the breast and a miscellaneous group including other types of malignant disease, are summarized in a paper elsewhere11. Briefly, while indications of slight amelioration attributable to treatment were observed in a small proportion (for example, in three cases of carcinoma of the breast, and in single cases of lymphosarcoma, Hodgkin's disease, and recurrent mixed salivary tumour of the ant-rum), the results were mainly negative. It was, however, decided to pursue the investigation for a further period. In that part conducted at the Christie Hospital it was next observed, by Dr. Edith Paterson, that urethane produced a fall in the leucocyte count in some of these cases, following which the clinical trial was extended to include examples of leukaemia and other lymphadeno-pathies. From the clinical results it is obvious that although the resuls of urethane therapy are not entirely restricted to leukaemia, they are vastly more pronounced in that disease, whether myeloid or lymphatic, as is shown in favourable cases by the rapid decline in the white-cell count to normal limits, by the tendency for the differential count to revert at least towards a more normal composition, and by diminution in spleen size and glandular masses. These effects, while they may occasionally persist for a variable time after urethane withdrawal, are, however, essentially reversible, agreeing in that respect with those produced in various animal tumours as described above. Possible Modes of Action of UrethaneThe remarkable effects of urethane in leukaemia, and indications from other experiments (to be published elsewhere) that the growth-inhibitory influence of urethanes upon the Walker rat carcinoma is restricted to a few such substances, and is certainly not exhibited by the great majority of a long series of related compounds, raise interesting problems as to its mode of action. Although these inhibitory effects appear not to be related to narcotic properties, at any rate in any simple fashion, it might be expected that assistance could be obtained on this point from the vast pharmacological literature dealing with urethanes and narcosis generally. It was recorded by Warburg12 so long ago as 1910 that if phenylurethane (c. 1/2,000 N) is added to sea water containing eggs of the sea urchin Strongylocentrotus lividus, cell division and nuclear division are suppressed, while the oxygen consumption is only very slightly reduced; and Clark13 found the inhibition of cell division in similar material by urethane to be an all-or-none effect, in contrast with a graded inhibitory action on respiration. Later, Warburg14 advanced the general theory that narcosis is due to adsorption of the narcotic in unimolecular layers on the catalytic surfaces involved in oxidation: but others have the contrary opinion, that urethane narcosis cannot be attributed to simple adsorption15. Even so it is clear, as has been pointed out by Quastel16, that if the character of a drug of this kind determines its entry into a cell, or its orientation at a surface, this alone can give no obvious clue to its mechanism of action. The next interpretation., an attempt to relate both narcosis and inhibition of cell division with a general inhibition of dehydrogenases, is faced with the difficulty that the concentrations required to induce narcosis, and often to suppress growth, or likely to be met with in vivo, are often far smaller than those required to inhibit enzyme reactions. Keilin and Hartree17 found that urethane and other narcotics, which in presence of biological reducing systems inhibit the reduction of the cytochrome components a, a3 and c, inhibit, on the contrary, the oxidation of b; and they suggested that the effect of urethane consists in bringing about the formation of a not easily dissociable complex composed of dehydrogenase, substrate and cytochrome b, and so making it inaccessible to the portion of the system reacting with oxygen. Quastel also put forward the qualified view that the inhibition of respiration of brain tissue obtained by low concentrations of narcotics under aerobic conditions is due not to competition of the narcotic with substrates for their dehydrogenases, but to the affinity of the narcotic to a special component playing an important part in the complete respiratory process of the cell-the narcotic effect being restricted at low concentrations to a tissue component which is possibly a flavoprotein. It is of interest here that the 'activity system' of Fisher and Stern18 (that is, that portion of the overall respiration which is most sensitive to inhibition) appears to be associated with the maintenance of normal mitosis, for when it is differentially suppressed by appropriate urethane concentrations, abnormal nuclear figures appear19. In a special case, Johnson et al.20 have recently systematized our present information regarding the relation of luminescent oxidation to the respiratory pathway in bacteria, and the inhibitory influence of urethane thereon: the luminescent system shows a characteristic sensitivity to urethane and is affected to a much greater extent than total oxygen consumption.When our earlier results were being considered, it was suggested (by Prof. A. R. Todd) that urethane might conceivably act by competing with some natural amine involved in the biosynthesis of nucleo-tides: such a possibility is clearly of interest in connexion with the belief of Plentl and Schoenheimer21., that since neither purines nor pyrimidines in the diet are utilized for the synthesis of nucleoproteins, these substances may be built up from smaller molecules within the nucleus. The possibility that urethane may act by inducing a deviation in purine synthesis becomes even more suggestive, at least as a tentative hypothesis, in view of the long recognized aberration of purine metabolism in leukaemia2 2, and speculations regarding a relationship between leukaemia and uratosis in subjects carrying a latent tendency to gout23. Here may be mentioned several papers of more than forty years ago, which are highly intriguing in the present context, and also in view of current interest in the significance of nucleoproteins for "ellular differentiation. Following a description by v. Moraczewski of the metabolism in a case of chronic leukaemia showing marked nitrogen- and phosphorus-retention, Wliite and Gowland Hopkins21 carried out a similar investigation. They thought it likely that when large numbers of new cells are present (for example, myelocytes), some of the diminished excretion of phosphorus is due to retention "to provide for the nuclein of the newly formed corpuscles". Milroy and Malcolm25., in a study of excretion in leukaemia, and the intracellular 'nuclein' metabolism of the granular leucocytes, found "more phosphorus being taken up than given off by these cells", but regarded as hypothetical the view that the diminished excretion of phosphorus was due to its retention for nuclein synthesis in the newly formed cells. In a later study of the origin of uric acid, Plimmer, Dick and Lieb26 quote an observation by Hor-baczewski, which they were able to confirm, of a relation between the number of leucocytes in the blood, and the output of uric acid: it was found (in a normal subject) that the uric acid output was high when the white cell count was raised, and that disappearance of the leucocytes from the blood was associated with a fall in the output of uric acid. They did not., however, consider purines (arising from the white cell nuclei) as the precursors of the uric acid, as had been supposed by Horbaczewski. It is a further question whether the exceptional response of leukaemia to urethane may reflect either a special sensitivity of leukaemic cells to inhibition (thus Stamer27 states that the carcinogenic hydrocarbon 9: 10-dimethyl-l: 2-benzanthracene, which has only slight retarding action when tested against many animal tumours, is in fact capable of eradicating transplanted leukaemia in mice: he also finds that the leukaemic cells are more sensitive to this substance than are normal leucocytes), or unusual metabolic properties in leucocytes whether normal or malignant: here it must be borne in mind that while normal white cells are certainly susceptible to the action of urethane11, they are relatively much more refractory than are leukaemic cells. It was found by Fleisch-mann28 that when respiration was fully inhibited by hydrocyanic acid, the phagocytic and amoeboid functions of leucocytes were fully retained, the necessary energy being derived from glycolysis ('anoxybiotic activity'); and Glover et al.29, although describing a relation between oxygen consumption of leukaemic cells and the degree of their maturity, believed there is no fundamental difference between the metabolism of leukaemic cells and that of normal white blood cells. A more recently discovered finding, which may be significant, is the presence of abnormal amounts of thiamine in leukaemic cells, with contrasts between normal and leukaemic cells in their metabolism of pyruvate30. Thus much of the information at present available is only vaguely suggestive as to possible mechanisms through which urethane may produce growth-inhibitory effects, as manifested most strikingly in its action upon leukaemic cells. Investigations of the purine metabolism in several of the cases reported by Paterson et al. have not as yet been sufficient to yield any useful data, largely because in none of them was the uric acid content of the blood unduly raised before treatment. It is, however, a striking circumstance that colchicine, which was earliest employed for its action upon gout, then observed to produce effects upon the bone marrow31, later found to have its remarkable influence on mitosis, and since known to have some action in leukaemia32, should be imitated by arsenic both in the treatment of leukaemia33 and in effects on the mitotic cycle34, and now to some extent by urethane in leukaemia and on cell division. It is an important problem whether the primary action of urethane is upon the growth mechanism or upon differentiation processes-to the extent that these are separable. That the latter is more likely is indicated from the marked histological and cytolog-ical changes in the Walker rat carcinoma which are produced by urethane. Although these alterations are associated with a moderate degree of inhibition of growth, it is certain that a much more intense retardation, repeatedly observed in independent experiments following administration of many other compounds quite unrelated to urethane, is not accompanied by any such marked effects upon cellular differentiation. Hence these effects may be regarded as due to a characteristic property of urethane, and it is perhaps reasonable to suspect that the drug may act on growth primarily through the processes of maturation, and in leukaemia by tending to remedy some deficiency in these processes. It is likely that more precise information must await the application, for this problem, of recent improvements in cytolog-ical technique as they affect the study of mitosis and differentiation in the cells of the marrow and blood35.