CANCER: RESTRICTION OF CHEMICAL ENERGY BY FATTY ACIDS AS THE COMMON PATHWAY WHEREBY VARIOUS ANTI-TUMOUR PROCEDURES SELECTIVELY DAMAGE MALIGNANT CELLS IN SITU

G R.N JONES. Department of Biochemistry, Royal College of Surgeons of England, 35/43 Lincoln’s Inn Fields, London, WC2A 3PN, England.

SUMMARY Harmful effects may be inlXcted upon cancer cells in vivo by a variety of means, for example, by the administration

of bacterial endotoxins or alkylating agents, and by radiation treatment. These different procedures appear to act on tumours indirectly, and are also capable of inducing shock. The suggestion is made that the deleterious actions on cancer cells are mediated by the tissues of the host, and involve raising the levels of non-esterified fatty acids (NEFAs) in the plasma, Fatty

acids may restrict the availability of chemical energy within cancer cells in at least two ways, either by modulating the activity of the mitochondrial adenine nucleotide translocase through the agency of their coenzyme A derivatives (LCFACoAs), or by uncoupling oxidative phosphorylation through the activation of phospholipase Ail. The effactive- ness of fatty acids in the former respect is theoretically and qualitatively examined on the basis of the affinity constants for LCFACoAs of the translocase and of certain enzymes concerned with P-oxidation. The development of a new form of cancer chemotherapy by selectively harnessing and stimulating the lipotropic response of the host to a tumour is discussed in terms of the administration of fatty acids which possess specific properties while at the same time raising plasma NEFA concentrations and maintaining them at an elevated level.

During the past century numerous reports have appeared in the medical and biological literature describing ex- periments in which injury of varying intensity has been inflicted on cancerous tissues in vivo by diverse procedures. In certain cases clinically useful treatments have been dev- eloped, while in others the claims have not been sub- stantiated. It now appears that the effects of these diverse procedures may have a common metabolic basis; in con- sequence, the possibility emerges whereby the effects might be explained in terms of a single fundamental response of the host, regardless of the nature of the original procedure.

Delete&ous effects on cancers Among the earliest of these reports was that of Fehleisen

(1 I), who recorded diminutions of tumour size in two patients with carcinoma and in another with fibrosarcoma in response to infection with Streptococcus erysipelas, although other cases of neoplastic disease failed to respond. Mouillin (40) noted that disappearance of sarcomata following treatment with mixed preparations of Strep- tococcus erysipelas and Serratia marcescens was not due to inflammation, but to an intensely rapid form of fatty degeneration. Early clinical studies have been reviewed (42). Haemorrhage (5 1) and haemorrhagic necrosis (12) have been observed in transplanted mouse tumours in response to the administration of purified preparations of endotoxin from Escherichia cofi (12, 5 1) and from other gram- negative organisms (12). In human patients, irradiation with preparations of radium has been extensively used against tumours in situ (27, 47, 54). The powerful local necrotising action (35, 62) of b&(2-chloroethyl) sulphide (mustard gas) suggested its use in the treatment of neoplasia (I), and led to the application of nitrogen mustards (45) to the problem (61; see also refs. 14 and 49 for citations of early work). Haemorrhagic necrosis was reported in a mouse sarcoma when the host animal% were subjected to an anaphylaxis- inducing procedure that resulted in shock (2).

Less success against tumours has been obtained by other

means; these include treatments involving colloidal preparations or salts of various elements, chiefly heavy metals. Here the early literature contains such a plethora of claims and refutations that objective review is out of the question. The delayed release of hind-limb tourniquets in mice totally abolishes mitotic activity in the epidermis of the ear (5); in contrast, cellular division in a mouse carcinoma and in a rat sarcoma is affected to a much lesser extent (la), and the eflect on the rat tumour is not always reproducible ( 15).

The indirectness of the effects on cancer tissue The question as to whether these deleterious actions on

cancer cells are produced directly or indirectly has not been neglected. A toxic carbohydrate complex isolated in a highly purified form from Salmonella enteritidis caused severe haemorrhagic necrosis in both sarcoma and carcinoma from rats, but failed to induce any effect on the same tumours in tissue culture at 2000 to 5000 times the minimum effective dose in viva (50). Likewise, clinical results tend to be more satisfactory when radium needles are inserted in the healthy tissue surrounding a neoplastic growth than when set into the growth itself (27, 44). In addition, if mouse tumours previously exposed to doses of radiation sufficient to cause complete regression are promptly excised and transplanted into another mouse, growth will frequently occur as well as if no irradiation had taken place. Radiation doses necessary to destroy neoplastic cells in vitro are enormous, and are out of all proportion to those required for cellular destruction in situ (54). The amount of %-labelled &s-(2-chloroethyl) sulphide taken up by the nucleic acids of mouse ascites cells is but a tiny fraction of the amount injected (29); similar situations exist with regard to other alkylating agents (58, 64). The available evidence tends to favour the conclusion that damaging effects of the various procedures are not exerted directly, but arise indirectly from the tissues of the host as a consequence of the treatment.

50

Control of translocation

A common feature of most of these clinical and experimental procedures when taken to extremes is the induction of shock, which has been recently ascribed to an attenuation of the energised movement of essential meta- bolites across the inner mitochondrial membrane of cells in target organs (23, 24). The sequence of events has been discussed in detail (24); in brief, severe injury -+ elevated, sustained secretion of catecholamines -+ cyclic AMP generation --+ lipolysis in adipose tissue ---+ increase in circulating non-esterilled fatty acid (NEFA) levels ---+ uptake of NEFAs by target organs - -+ thioesterification with coenzyme A derivatives (LCFACoAs), with concomitant inhibition of the enzymi- tally-catalysed exchange of cytoplasmic ADP for mito- chondrial ATP. It should perhaps be added that NEFAs can be mobilised by other agents besides catecholamines (55). Work in progress indicates that at least 41% of adenine nucleotide translocase activity in mouse liver mitochondria is inhibited in vivo; preliminary experi-

ments have revealed greater levels of inhibition (56%) in we&preserved mitochondria from shocked livers (26). If the chain of events outlined above is indeed responsible for the injuries inflicted upon tumours, then the administration of catecholamines themselves would be expected to exert damaging effects. Reicher (48) and his associate Engel (10) have described a “powerful necrotising action” in a range of well-developed rat sarcomas when adrenaline was injected subcutaneously into the hosts. The effects of a range of compounds chemically related to adrenaline were also investigated, but no relationships between the necrotising action and raising the blood pressure or vasoconstriction could be demonstrated. (10).

The question then arises as to the precise manner in which fatty acids might be involved in tumour damage and destruction. In theory, the availability of chemical energy within neoplastic tissues may be restricted by fatty acids in at least two ways, the first of which could be to uncouple oxidative phosphorylation (46) by activating mitochondrial phospholipase A2. However, the isolation of poorly-coupled mitochondria from tumour tissues has been ascribed to procedural artefact and to other causes (9), but uncoupling may fulfil an important role in vivo, and provide some index of the !endogenous (damage effected by the host. The elevated temperatures reported for tumour tissue in situ provide support for this concept. Alternatively, fatty acids in the form of their CoA thioesters might accumulate in amounts sufficient to inhibit competitively (43) the enzymic translocation of mitochondrial ATP to the cytoplasm (20). Inhibition of the enzymically-catalysed movement of adenine nucleotides across the inner mitochondrial membrane can actually result in cellular death. This was demonstrated by Tocco (57), who showed that potassium atractylate, which inhibits the translocase (20), produced a local necrotising effect when injected subcutaneously in neutral aqueous solution. The range of neoplasia susceptible to the treatments outlined above may have been limited to tumours capable of taking up NEFAs from the blood- stream. Uncoupling of oxidative phosphorylation and in- hibition of translocation need not necessarily be regarded as

mutually exclusive processes; indeed, both my occur simultaneously.

The way in which LCFACoA accumulation might arise merits closer examination. Measurements of respiratory quotients indicate that various tumours readily oxidise endogenous fatty acids (8); later work with I-14C-labelled palmitic acid has confirmed this finding (33), and revealed that exogenous palmitate is also oxidised (63). Although by no means impossible, it therefore seems less likely that an inhibition of the adenine nucleotide translocase so extensive as to cause cellular necrosis would arise exclusively from an accumulation of acyl CoA derivatives of the commoner fatty acids.

In P-oxidation, the initial thio-esterification step with coenzyme A proceeds outside the mitochondrial matrix. The inner mitochondrial membrane is impermeable to the resulting LCFACoAs; accordingly, thio-esterification is followed by a reversible transfer reaction catalysed by a carnitine long-chain acyl transferase, thus:-

LCFACoA + carnitine ~1 fatty acyl carnitine + free coenzyme A. The carnitine ester can traverse the inner mitochondrial membrane, where the reaction is reversed and P-oxidation can proceed (17).

Sherratt (52) has drawn attention to the importance of the specificity of the acyl transferase located on the outside of the matrix. If unusual fatty acids are present, various factors could favour an inhibition of adenine nucleotide translocation. First high Kmvalues (i.e., low tinities) or low Vmax values of carnitine acyl transferase for the cor- responding LCFACoAs would favour an accumulation of CoA thioesters outside the inner mitochondrial membrane. Second, low inhibitory constants (i.e., high affinities) of the translocase with regard to unusual LCFACoAs would result in an increased susceptibility of adenine nucleotide movement to inhibition. Inhibitory constants (Kis) have been determined for an extensive range of LCFACoAs with regard to ADP entry into rat liver mitochondria; they display wide variations, depending on the chemical identity of the acyl moiety (39). A further complication is provided by the finding that carnitine utilisation by the transferase itself is competitively and powerfully inhibited by palmitoyl CoA concentrations even lower than the Km value (4); the inhibitory capacity of unusual LCFACoAs appears to be quite unknown. Third, LCFACoAs might also accumulate intramitochondrially if a stage of P-oxidation is reached where high Km or low Vmax values for the acyl CoA dehydrogenase obtain with regard either to the original substrates or to their catabolic thioester inter- mediates; again, low Ki values would need to obtain with respect to the translocase.

Success in the non-surgical management of tumours might therefore have depended to some extent on the ability of the procedure to maintain NEFA concentrations, perhaps of specific fatty acids, at elevated levels for periods of time sufficiently long to permit uptake consistent with the production of irreversible damage in tumour cells. For- tuitously, the fatty acid composition of adipose tissue tends to reflect the chemical composition of dietary fat (30, 3 1). These five factors (magnitude and duration of increase in NEFA levels, extent of NEFA uptake by neoplastic cells,

51

chemical identity of circulating fatty acids, and diet) provide extensive scope for variable responses to standard&d cancer therapies.

“Damaged respiration” as a consequence of defective trans- location

The classical finding that tumour slices produce lactate when incubated in the presence of oxygen (38) has been interpreted in terms of a lowering of cytochrome oxidase content associated with the change from the normal to the neoplastic condition. However, valid comparisons can only be made when the same cell type is studied in the two different states. The liver, which on a weight basis consists almost entirely of parenchymal cells, has often provided the experimental model for investigations of this kind, but the cytological composition shows striking changes in response to feeding the frequently-used liver carcinogen 4- dimethylaminoazobenzene to rats. The dominant response to this substance is an increase in the proportion of extensively-proliferating cells of the bile-duct and connective tissue (7), in which the activity of cytochrome oxidase is much weaker than in parenchymal cells (6). On the other hand, the capacities of liver cell metastases located in rat lung to oxidise reduced cytochrome c, as studied by a quantitative histochemical procedure, were essentially normal, even when cellular damage was histologically apparent (22). Dickens and Simer (8) showed that a wide variety of tumours, including two of human origin, oxidised exogenous pyruvate freely, but, in contrast with normal tissues, were incapable of oxidising exogenous lactate. The failure of tumour tissue to oxidise carbohydrate completely can therefore be pinpointed at the stage of lactate con- version to pyruvate. Measurements of lactate de- hydrogenase activity in nineteen different tumours and in twenty normal tissues in the mouse indicate unambigously that this enzyme is unlikely to be a rate-limiting factor (34).

These findings are difficult to rationalise purely on the basis of lowered cytochrome oxidase activity. Lactate accumulation can be explained more readily in terms of defective operation of the malate-aspartate shuttle (3), the chief means by which reducing equivalents of cytosolic origin may be transferred to mitochondria for oxidation. In other words, the aerobic production of lactate by tumours could more convincingly indicate an attenuation of energised translocation as opposed to a lowered oxidative capacity in the electron transport chain. Various additional facts may be cited; first, that the malate-aspartate shuttle operates but weakly in the Morris hepatoma (28). Second, certain other tissues, such as the decidual placenta, areas of tissue injury, and the prepared site for the local Shwartz- mann reaction, not only produce lactate aerobically, but also respond to the intravenous injection of endotoxin by developing haemorrhagic necrosis (see ref. 56 for details). Third, a line of Ehrlich ascites cells, which had developed resistance in vitro to the nitrogen mustard nitromin, not only produced less lactate from glucose under aerobic conditions, but also demon”strated a higher oxidative capacity towards lower fatty acids (59). Fourth, blood lac- tate levels can rise substantially under aerobic conditions in response to fructose (65) or ethanol (21) administration, effects which may be correlated with momentary disturb-

ances to translocation (see 25 for discussion).

Lipid alterations in tumour-bearing hosts Certain changes that occur in the body of the tumour-

bearing host are consistent with the concept of NEFA mobilisation as a defence mechanism against neoplastic growth. Total lipid material was extensively depleted in rats with a transplanted sarcoma (53), the content of carcass lipid falling substantially as tumour weight increased (19, 36, 37). Losses of fatty substances exceeded those caused by falls in dietary intake (37). The disappearance of sub- cutaneous, mesenteric and retroperitoneal fat stores from a substantial proportion of human subjects at operation or autopsy is well known (32). In advanced cancer, the mean NEFA level was 66% higher than in normal subjects, and 61% higher than in diseased non-cancerous patients. (41). Involvement of the endocrine system has been suggested (I 9); adrenal hyperplasia has been reported in carcinoma- bearing rats (13, 19), yet lipid mobilisation occurred despite hypophysectomy (13).

General implications If the foregoing hypothesis is correct, then the fact that

a substantial number of treated patients still succumbs to cancer could reflect the bluntness of the body’s existing weaponry. Once the identity of the main mechanism res- ponsible for the induction of damage can be established, the weaponry might be sharpened by seeking fatty acids to restrict specifically the availability of high-energy phos- phate within tumour cells to a point at which selective necrosis would follow. A new chemotherapy might be based on the administration, preferably by dietary means, of such fatty acids, singly or in combination, followed by an appropriate substance, perhaps with sympathomimetic activity, capable of raising circulating NEFA levels while producing the minimum of side-effects.

The ability of cancerous cells to take up fatty acids is an important factor that also requires attention. The capacity of numerous transplanted rodent tumours to assimilate fatty acids of dietary origin is well-established (18, 33, 66), but the familiar refractoriness to treatment of brain tumours might now be explained in terms of the limited ability of brain itself, relative to other organs, to take up palmitic acid (60). Similarly, whereas the phospholipid composition of tumours in both rat and mouse showed variations in response to alterations in the chemical composition of dietary fat, that of phospholipid in mouse brain appeared independent of diet (66). Different neoplastic tissues, depending on cell type and organ of origin, might respond more effectively to different fatty acids; in addition, stimulation of NEFA uptake by tumour cells may prove amenable to manipulation, though probably not in all cases. The questions whether the putative treatment outlined might increase susceptibility to thrombosis, and whether normal tissues, notably the heart and other organs at risk in shock, might be affected adversely or not should receive very care- ful attention during the evaluation of new clinical procedures.

Acknowledgements I thank Professor C. Long for his interest and encourage-

52

ment; numerous friends and colleagues, especially Dr. H. S. A. Sherratt. for helpful and constructive criticism of the precirculated manuscript; Miss Heather Watson for typing; and the Freemasons’ 250th Anniversary Fund for financial support.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

Il.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23. 24.

25.

:;:

28.

29.

REFERENCES Adair FE. Bagg HJ. Experimental and clinical studies on the treat- ment of cancer by dichlorethylsulphide (mustard gas). Ann Surg 93, 190-199. 1931 Barrett MK. Anaphylaxis and haemorrhage in transplanted tumours. J Nat1 Cancer Inst 2, 625-630, 1942 Borst P. Interrelations between cytoplasmic and mitochondrial diphosphopyridine nucleotide in Ehrlich ascites tumour cells. In: Proc 5th Internat Congr Biochem, Moscow, 1961, Int Union Biochem Symp Ser VO! 22, Oxford, Pergamon, p 233-247, 1963 Bremer J. Norum KR. The mechanism of substrate inhibition of palmitoyl coenzyme A: carnitine palmitoyltransferase by palmitoyl coenzyme A. J Biol Chem 242, 1744-1748, 1967 Bullough WS. Green HN. Shock and mitotic activity in mice. Nature 164, 795-796, 1949 Chayen J. Aves EK, Barron ALE, Silcox AA. A histochemical study of respiratory enzymes in carcinogenesis. Ann Histochim 7, Suppl. 1 I I I 13. 1962 Daoust R. Canter0 A. The numerical proportions of cell types in rat liver during carcinogenesis by 4-dimethylaminoazobenzene (DAB). Cancer Res 19, 757-762, 1959 Dickens F. Simer F. The metabolism of normal and tumour tissue. II. The respiratory quotient. and the relationship of respiration to glycolysis. Biochem J 24, 1301-1326, 1930 Emmelot P. Long-chain fatty acids in rat hepatoma homogenates ;;d ;t;moperties of hepatoma mitochondria. Cancer Res 22, 38

Engel H. Chemotherapeutische Versuche mit Adrenalin und iihnlich constituirten Stoffen bei tumorkranken Thieren. Zeits exp Path01 11, 9-39, 1912 Fehleisen. On Erysipelas, trans Ogilvie L. In: Recent essays by various authors on bacteria in relation to disease, ed WW Cheyne. London. New Sydenham Society, p 261-286, 1886 Fogg LC. Effect of certain bacterial products upon the growth of mouse tumour. Pub1 Hlth Reps, US Publ Hlth Service 51,56-64, 1936 Frederick GL. Begg RW. A study of hyperlipaemia in the tumour- bearing rat. Cancer Res 16, 548-552, 1956 Gilman A. Philips FS. The biological actions and therapeutic applications of the p-chloroethyl amines and sulphiis. Science 103, 409:415, 1946 Green HN. Shock and neoplasia. Acta Unio Internat contra Cancrum 7, 466-468. 1951 Green HN. Savigear M. Shock and neoplasia. Brit Med J I, 498. 500. 1951 Greville GD. Tubbs PK. The catabolism of long chain fatty acids in mammalian tissues. In: Essays in Biochemistry, ed PN Campbell, GD Greville. vol 4. London, Academic Press, p 155-212, 1968 Haven FL. The phospholipid metabolism of tumours. J Biol Chem 118, Ill 121. 1937 Haven FL. Bloor WR. Randall C. Lipids of the carcass, blood plasma. and adrenals of the rat in cancer. Cancer Res 9, 5 1 I-5 14. 1949 Heldt HW. Jacobs H, Klingenberg M. Endogenous ADP of mitochondria, an early phosphate acceptor of oxidative phos- phorylation as disclosed by kinetic studies with Ci4-labelled ADP and ATP and with atractvloside. Biochem Bioohvs Res Commun

’ ‘ I~

18, 174-179. 1965 Himwich HE. Nahum LH. Rakieten N, Fazikas JF, Du Bois D, Gildea EF. The metabolism of alcohol. JAMA 100, 65 l-654, 1933 Jones GRN. Quantitative histochemistry: levels of certain oxidative enzymes in secondary neoplasia found in a rat after feeding 4- dimethylaminoazobenzene. Brit J Cancer 19, 360-369. 1965 Jones GRN. Ionic shuttles in shock. Lancet 1, 905. 1974 Jones GRN. The shocking consequences of defective ionic trans- location in mitochondria. Med Hypoth 1, 118-127, 1975 Jones GRN. Metabolic studies of isolated canine hearts oerfused at 4OC for up to 96 hours. with assessment of viability. C’ryobiology 12, 238-254. 1975 Jones GRN, Kramer R. Klinnenberg M. In preparation. Keynes G. The radium treatment ofcarcinom’a df the breast. Brit J Surg 19, 415-480, 1932 LaNoue KF. Hemington JG. Ohnishi T, Morris HP, Williamson JR. Defects of anion and electron transport in Morris hepatoma mitochondria. In: Hormones and Cancer, ed KW McKems. New York. Academic Press. p I3 I- 167, 1974 Lawley. PD. Brookes P. The action of sulphur mustard on RNA and DNA irl viva. Brit Empire Cancer Campaign 37th Ann Rept: Part II. p 70. 1959

30. Longenecker HE. Deposition and utilization of fattv acids. I. Fat synthesis from high carbohydrate and high rats. J Biol Chem 128. 645-658. 1939

proteinhiets in fasted

31.

32.

33.

Longenecker HE. Deposition and utilization of fatty acids. II. The non-preferential utilization and slow replacement of depot fat con- sisting mainly of oleic acid and linoleic acids; and a fatty acid analysis of corn oil. J Biol Chem 129, 13-22. 1939 Mays ET. Serum lipids in human cancer. J Surg Res 9, 273-277, 1969

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

Medes G. Paden G. Weinhouse S. Metabolism of neoplastic tissues. XI. Absorption and oxidation of dietary fatty acids by implanted tumours. Cancer Res 17, 127-133, 1957 Meister A. Lactic dehydrogenase activity of certain tumours and normal tissues. J Nat1 Cancer Inst IO, 1263-1271, 1950 Meyer V. Medicinisch-chemische Notizen. Chem Ber 20, 1725- 1731. 1887 Mider GB, Tesluk H. Morton JJ. Effects of Walker carcinoma 256 on food intake, body weight and nitrogen metabolism of growing rats. Acta Unio Internat contra Cancrum 6, 409-420, 1948 Mider GB. Sherman CD, Morton JJ. The effect of Walker carcinoma 256 on the total lipid content of rats. Cancer Res 9, 222-224. 1949 Minami S. Versuche an iiberlebendem Carcinomgewebe. [Atmung und Glykolyse]. Biochem Z 142, 334-350, 1923 Morel F. Lauquin G, Lunardi J, Duszynski J, Vignais PV. An appraisal of the functional significance of the inhibitor effect of long chain acyl-CoAs on mitochondrial transport. FEBS Lett 39, 133- 138, 1974 Mouillin CM. The treatment of sarcoma and carcinoma by in- jections of mi.xed toxins. London, John Bale, Sons and Danielsson 1898: cited by Nauts HC, Swift WE, Coley BL in ref. 42. Mueller PS, Watkin DM. Plasma unesterified fatty acid con- centrations in neoplastic tissue. J Lab Clin Med 57, 95-108, 1961 Nauts HC. Swift WE, Coley BL. The treatment of malignant tumours by bacterial toxins as developed by the late William B. Coley, MD. reviewed in the light of modern research. Cancer Res 6, 205-216. 1946 Pande SV. Blanchaer MC. Reversible inhibition of mitochondrial adenosine diphosphate phosphorylation by long chain coenzyme A esters. J Biol Chem 246, 402-411, 1971 Pannett CA. The response of malignant tumours to radiation. Lancet I, 829, 1957

47.

48.

49.

50.

51.

52. 53.

Prelog V. Dnza G. Hanousek V. Sur les bis-fi-halogenoithyl- amines. Coil Czech Chem Comm 3, 578584, 193 1 Pressman BC. Lardy HA. Effect of surface active agents on the latent ATPase of mitochondria. Biochim Biophys Acta 21, 458- 466, 1956 Rehns J. Salmon P. Action du radium sur les epitheliomes benins. CR Sot Biol 56, 313-315. 1904 Reicher K. Experimentelle Beitrlge zur Therapie maligner Tumoren. Dtsch Med Wchschr 36, 1356-1359, 1910 Rhoads CP. Nitrogen mustards in the treatment of neoplastic disease. JAMA 131, 656-658. 1946 Shapiro CJ. The effect of a toxic carbohydrate complex from S. enteritidis on transplantable rat tumours in tissue culture. Am J Hyg 31 (B), 114-126, 1940 Shear MJ. Andervont HB. Chemical treatment of tumours. III. Separation of haemorrhage-producing fraction of B. coli filtrate. Proc Sot Exp Biol Med 34, 323-325, 1936 Sherratt HSA. personal communication Smedley-Maclean I, Hume EM. Fat-deficiency disease of rats. The influence of tumour growth on the storage of fat and of polyun- saturated acids in the fat-starved rat. Biochem J 35, 996-1002, 1941

54. 55.

56.

57.

58.

59.

60.

61.

62.

Souttar HS. Radium and cancer. London: Heinemann, 1934 Sutherland EW, Butcher RW. Robison GA, Hardman JG. The role of adenosine 3’.5’monophosphate in hormone action. In: Wirkung- smechanismen der Hormone, 18th Colloq Gesell Biol Chemie, Mosbach, ed P Karlson. Berlin, Springer, pp l-29, 1967 Thomas L. The physiological disturbances produced by endotoxins. Ann Rev Physiol 16, 467-490. 1954 Tocco EL. Sull’ avvelenamento per Carlina gummifera. Nota IIL- Ricerche farmacologiche sul Carlinato di potassio. Arch Intern Pharmacodyn Ther 26, 291-295. 1922 Trams EG. Nadkarni MV. Smith PK. On the mechanism of action of the alkylating agents. 1. Interaction of alkylating agents with nucleic acids. Cancer Res 21, 560-566. 1961 Tsukada H. lnoue A, Onoe T. Hirota E, Onodera T. Metabolism of cancer ceils and its relation to the action mechanism of anti- cancer chemicals. 4. Biological and biochemical characteristics of nitromin-resistant Ehrlich ascites tumour cells. Gann (Suppl) 50, 92-93. 1959 Volk ME. Millington RH. Weinhouse S. Oxidation of endogenous fatty acids of rat tissues in vitro. J Biol Chem 195, 493-501, 1952 Ward K. Chlorinated trialkyl amines and method of producing, US Pat 2. 072.348. 2pp. 1937 Warthin AS. Weller CV. The pathology of the skin lesions produced by mustard gas (dichlorethylsulphide). J Lab Clin Med 3. 447~479. 19 17

63. W&house S. Millington RH. Wenncr CE. Occurrence of the crtric

53

acid cycle in turnours. J Am Chem Sot 72, 43324333, 1950 64. Wheeler GP. Studies related to the mechanisms of action of

cytotoxic alkylating agents: a review. Cancer Res 22, 651-688, 1962

65. Woods HF, Alberti KGMM. Dangers of intravenous fructose.

Lancet II, 1354-1357. 1972 66. Yasuda M. Lipid metabolism of tumours. II. Influence of food lipids

on the nature of tumour phospholipid. Proc Sot Exp Biol Med 28, 1074-1075. 1931

54