Search

Patients with cancer often undergo a specific loss of skeletal muscle mass, while the visceral protein reserves are preserved. This condition known as cachexia reduces the quality of life and eventually results in death through erosion of the respiratory muscles. Nutritional supplementation or appetite stimulants are unable to restore the loss of lean body mass, since protein catabolism is increased mainly as a result of the activation of the ATP-ubiquitin-dependent proteolytic pathway. Several mediators have been proposed. An enhanced protein degradation is seen in skeletal muscle of mice administered tumour necrosis factor (TNF), which appears to be mediated by oxidative stress. There is some evidence that this may be a direct effect and is associated with an increase in total cellular-ubiquitin-conjugated muscle proteins. Another cytokine, interleukin-6 (IL-6), may play a role in muscle wasting in certain animal tumours, possibly through both lysosomal (cathepsin) and non-lysosomal (proteasome) pathways. A tumour product, proteolysis-inducing factor (PIF) is produced by cachexia-inducing murine and human tumours and initiates muscle protein degradation directly through activation of the proteasome pathway. The action of PIF is blocked by eicosapentaenoic acid (EPA), which has been shown to attenuate the development of cachexia in pancreatic cancer patients. When combined with nutritional supplementation EPA leads to accumulation of lean body mass and prolongs survival. Further knowledge on the biochemical mechanisms of muscle protein catabolism will aid the development of effective therapy for cachexia.

The effect of 3-methyl(temozolomide) and 3-ethyl (CCRG 82019) substituted imidazotetrazinones on cytosine methylation has been studied in the human lymphoblastoid cell line GM892. There was a decrease in the 5-methylcytosine content of newly synthesized DNA in cells treated with the 3-methyl and a small increase in cells treated with the 3-ethyl analogue, which was maximal 4 days after drug treatment. There was a progressive decrease in nuclear DNA methyltransferase after treatment with temozolomide with complete inhibition at 11–12 hr after drug addition, followed by a re-establishment of enzyme levels towards control values. While the free drugs had no effect on DNA methyltransferase activity in vitro , DNA isolated from GM892 cells previously treated with temozolomide inhibited the transfer of methyl groups from S -adenosyl- l -methionine to M. lysodeiktious DNA. The maximum effect was observed at 6 hr after drug addition and was proportional to the concentration of temozolomide to which the cells had previously been exposed. These results suggest that temozolomide may induce a block in cellular replication as a result of an indirect inhibition of DNA methylation and cells which escape this block progress with hypomethylated DNA.

The effect of anti-tumour alkylating agents on the adenosine 3′,5′-monophosphate (cyclic AMP) levels of sensitive and resistant Walker carcinoma cells in tissue culture has been investigated. Chlorambucil caused a two-fold elevation of cyclic AMP, in the sensitive line only, 8 hr after treatment with a dose of 5 μg/ml. A comparable increase is produced by the cyclic nucleotide phosphodiesterase inhibitor aminophylline at a dose which produces the same degree of inhibition of cell growth. The therapeutically-inactive monofunctional N -ethyl analogue of chlorambucil had no effect on the cyclic AMP level of the sensitive cells at a dose of 250 μg/ml after 8 hr, while the highly selective monofunctional alkylating agent, CB 1954, at 1 μg/ml caused an elevation of cyclic AMP in the sensitive line 24 hr after treatment. This is not a non-specific effect caused by an inhibition of cell growth for the antimetabolite methotrexate had no effect on the intracellular cyclic AMP of sensitive Walker cells at doses which produced complete inhibition of cell growth. The effect of the alkylating agents on cyclic AMP levels is probably due to a specific inhibition of the cyclic nucleotide phosphodiesterase with a low K m value, since only this form of the enzyme is inhibited, and only in the sensitive cells, when an effect on cell growth and cyclic AMP content is observed. In the case of CB 1954, however, there was no inhibition of either form of the phosphodiesterase at a time when cyclic AMP levels were elevated. A linear relationship exists between the reciprocals of the intracellular cyclic AMP and the percentage growth inhibition produced by a given dose of chlorambucil.

A comparison has been made between the ability of long-chain triglycerides (LCT) and medium-chain triglycerides (MCT) to prevent weight loss induced by the cachexia-inducing colon adenocarcinoma (MAC16) and to reduce tumour size. There was no difference in calorie consumption or nitrogen intake between the various groups. When compared with a normal control high carbohydrate, low fat diet, animals fed MCT showed a reduced weight loss and a marked reduction in tumour size. In contrast neither weight loss nor tumour size differed significantly from the controls in animals fed the LCT diet. An elevated plasma level of 3-hydroxybuturate was found only in the animals fed the MCT diets. Administration of LCT caused an increase in the plasma level of FFA, which was not observed in the MCT group. These results suggest that diets containing MCT would provide the best ketogenic regime to reverse the weight loss in cancer cachexia with a concomitant reduction in tumour size.

A short series of α-substituted analogues of serine, designed as enzyme-activated, irreversible inhibitors of serine hydroxymethyltransferase, has been prepared. (±)-α-Allyl- and (±)-α-prop-2-ynyl-serine were synthesised by appropriate alkylation of the anion derived from ethyl acetamidocyanoacetate, followed by selective reduction of the ester function and hydrolysis of protecting groups. Acetoxymethylation of the anion derived from methyl 2-(benzylideneamino)but-2-enoate gave (±)-α-vinylserine after deprotection. These novel analogues of serine were largely inactive as inhibitors of the enzyme, except that (±)-α-vinylserine showed weak competitive inhibition.

Treatment of K562 human erythroleukaemia cells with 8-carbamoyl-3-methylimidazo[5,1-d]-l,2,3,5-tetrazin-4-(3H)-one (CCRG 81045) caused a concentration-dependent increase in the number of cells producing haemoglobin after 3 days of treatment. The ethyl analogue (CCRG 82019) was inactive in the induction of erythroid characteristics. The concentration of 5-methyl-cytosine in the DNA of CCRG 81045 treated cells decreased 3 days after treatment, and was directly proportional to the number of benzidine-positive cells in the cultures, suggesting a direct correlation between hypomethylation of DNA and the induction of haemoglobin synthesis. Although the mechanism of this druginduced hypomethylation of DNA is not known, the methyl analogue (CCRG 81045) also appeared to reduce template activity of isolated DNA to a greater extent than the ethyl analogue, suggesting that the extent or position of alkylation of DNA bases be important in inhibiting DNA-recognition enzymes.

Both ascorbic and dehydroascorbic acids were found to be reversible inhibitors of both the high and low affinity forms of the cyclic adenosine 3′,5′-monophosphate phosphodiesterase from Walker carcinoma. In contrast only the high affinity form of the enzyme from rat liver was inhibited by dehydroascorbic acid.

An attempt has been made to reverse cachexia and to selectively deprive the tumour of metabolic substrates for energy production by feeding a ketogenic regime, since ketone bodies are considered important in maintaining homeostasis during starvation. As a model we have used a transplantable mouse adenocarcinoma of the colon (MAC 16) which produces extensive weight loss without a reduction in food intake. When mice bearing the MAC16 tumour were fed on diets in which up to 80% of the energy was supplied as medium chain triglycerides (MCT) with or without arginine 3-hydroxybutyrate host weight loss was reduced in proportion to the fat content of the diet, and there was also a reduction in the percentage contribution of the tumour to the final body weight. The increase in carcass weight in tumour-bearing mice fed high levels of MCT was attributable to an increase in both the fat and the non-fat carcass mass. Blood levels of free fatty acids (FFA) were significantly reduced by MCT addition. The levels of both acetoacetate and 3-hydroxybutyrate were elevated in mice fed the high fat diets, and tumour-bearing mice fed the normal diet did not show increased plasma levels of ketone bodies over the non-tumour-bearing group despite the loss of carcass lipids. Both blood glucose and plasma insulin levels were reduced in mice bearing the MAC16 tumour and this was not significantly altered by feeding the high fat diets. The elevation in ketone bodies may account for the retention of both the fat and the non-fat carcass mass. This is the first example of an attempt to reverse cachexia by a diet based on metabolic differences between tumour and host tissues, which aims to selectively feed the host at the expense of the tumour.

(±)-α-Vinylserine has previously been found to be a competitive inhibitor of serine hydroxymethyltransferase, an important target in anti-tumour chemotherapy. The crystal structure of (±)-α-vinylserine was determined by X-ray diffraction techniques, and molecular modelling was used to build the amino acid pyridoxal 5′-monophosphate Schiff's base. Molecular-mechanics calculations indicate that the reacting conformation for the cleavage of the α–β bond of the compound when bound to pyridoxal 5′-monophosphate is achievable. However, this compound, when incubated with homogeneous rabbit-liver serine hydroxymethyltransferase, was not dehydroxymethylated under any conditions, nor could any quinonoid intermediates be detected. Circular dichroism spectroscopy indicates that this molecule is not an effective enzyme inhibitor because it is not forming an imine base with the pyridoxal phosphate at the active site of the enzyme.

The activities of glyoxalase I and glyoxalase II were determined in human promyelocytic leukaemia HL60 and erythroleukaemia K562 cells in culture. The activity of glyoxylase I is ca 10–20 times greater than the activity of glyoxalase II under assay conditions. When HL60 and K562 cell lines were incubated with N -methylformamide and 8-carbamoyl-3-methylimidazo-[5,1-d]-1,2,3,5-tetrazin-4(3H)-one (CCRG 81045), substantial changes in the glyoxalase activities were induced. With HL60 cells, treatment with N -methylformamide and CCRG 81045, both of which induce functional differentiation of this cell line, there is a dose-dependent decrease in glyoxalase I activity and a concomitant dose-dependent increase in glyoxalase II activity, both of which are directly proportional to the number of differentiated cells. With K562 cells, N -methylformamide and CCRG 81045 induce an increase in both glyoxalase I and glyoxalase II activities, although only CCRG 81045 induces the appearance of haemoglobin producing cells. N -Methylformamide and CCRG 81045 do not activate or inhibit the activities of glyoxalase I and glyoxalase II from HL60 and K562 cells when studied in cell-free systems. The changes in the glyoxalase activities of HL60 and K562 cells during the incubations therefore appear to be due to alteration in the synthesis and/or regulatory modification of the glyoxalase enzymes induced by N -methylformamide and CCRG 81045. Despite the apparent disparity of the effect of differentiation on the glyoxalase system in the two cell lines, in both cases the glyoxalase I/glyoxalase II activity ratio decreases with the appearance of differentiated cells. Since glyoxalase II catalyses the rate-determining step in the glyoxalase system, this suggests that immature cells have an impaired capacity to metabolise S- d -lactoylglutathione.

The extent of macromolecular alkylation by three imidazotetrazinones, 8-carbamoyl-3-(2-chloroethyl)imidazo[5,1- d ]-1,2,3,5-tetrazin-4 (mitozolomide) and the 3-methyl CCRG 81045) and 3-ethyl (CCRG 82019) analogues has been studied both in intact cells and with isolated DNA, RNA and protein. Towards isolated DNA and RNA CCRG 81045 was about twice as reactive as mitozolomide and 5–10-fold more reactive than CCRG 82019. Two cell lines were chosen to study macromolecular alkylation, GM892A and Raji, the latter being 10–20-fold less sensitive to mitozolomide and CCRG 81045 than the former, but only one-and-a-half-fold less sensitive to CCRG 82019. Drug uptake into both cell lines was shown to be by a rapid diffusion process with a cell medium distribution ratio not far from unity. For all three agents intracellular radioactivity became associated with macromolecules, and the level found at any time is a balance between the rate of alkylation and the rate of alkyl group removal by repair processes. Both CCRG 81045 and CCRG 82019 produced approximately the same level of alkyl groups bound to DNA, RNA and protein over a 24-hr period, whereas mitozolomide produced a greater extent of alkylation. All three agents left more alkyl groups bound to DNA and RNA in GM892A than in Raji cells, but there was no difference in the level of alkyl groups remaining bound to proteins. However, in GM892A cells the overall level of alkylation of DNA by CCRG 81045 exceeded that of CCRG 82019 only after 24 hr of drug incubation despite the twenty-fold difference in potency of these agents. These results suggest that specific base alkylations rather than total macromolecular alkylation may be more important in determining relative cytotoxicity.

The ability of a number of anti-tumour agents to elevate the intracellular level of adenosine 3′,5′-monophosphate (cyclic AMP) in Walker carcinoma has been investigated. Of these agents HN2, chlorambucil, merophan, CB 1954, and cis-dichloro diammine Pt II [cis-Cl2(NH3)2 Pt II) caused appreciable elevations of the cyclic nucleotide. These compounds are all regarded as having a typical alkylating agent spectrum of action. A linear relationship was obtained between the intracellular cyclic AMP level produced by a given dose of cis-Cl2(NH3)2 Pt II and the observed percentage inhibition of cell growth. There was a rapid elevation of cyclic AMP in response to chlorambucil which reached its peak within 1 hr of treatment. This preceded the inhibition of thymidine incorporation which reached 50 per cent by 7 hr. 1,3-Bis(2-chloroethyl)1-nitrosourea (BCNU) and (±) 1,2-bis-(3,5-dioxopiperazin-1-yl)propane (ICRF 159) had no effect on the intracellular level of cyclic AMP in Walker carcinoma at any dose level studied. The cyclic 3′,5′-nucleotide phosphodiesterase from bone marrow and intestinal mucosa, two tissues susceptible to alkylating agents, has been investigated. This enzyme behaves kinetically as if two separate activities exist, one with a low affinity for the substrate and the other with a high affinity. The distribution of the two forms of the enzyme from these tissues agrees with the correlation previously obtained between sensitivity to alkylating agents and a high percentage of the low Km form of the enzyme.

The uptake and incorporation of 8-carbamoyl-3-(2-chloroethyl) (6- 14 C-imidazo)[5,1-d]-1,2,3,5-tetrazin-4-(3H)-one (Mitozolomide) into TLX5 mouse lymphoma cells has been studied in vitro . Uptake was rapid, reaching a cell/medium distribution of approximately unity in 1 min at 37° and 10 min at 4°, directly proportional to drug concentration and was unaffected by metabolic inhibitors. These results are consistent with a simple diffusion mechanism. No difference in uptake was observed between drug sensitive and resistant TLX5 lymphoma cells. Cellular radioactivity was found to be progressively accumulated into acid-insoluble material. Acid hydrolysis of this precipitate followed by hplc analysis of the DNA and RNA bases showed that the radioactivity was associated solely with adenine and guanine bases. Mitozolomide was unstable in tissue culture medium and over a 24 hr period about 80% of the drug was converted into 5-aminoimidazole-4-carboxamide (AIC). Non-radioactive AIC suppressed the incorporation of radioactivity into nucleic acids, but had no effects on the initial rate of uptake of mitozolomide into the cell. These results suggest that the radioactivity in nucleic acids arises as a result of salvage of AIC, formed by intracellular decomposition of mitozolomide.

Prostaglandins are secreted by a variety of animal and human tumours. A higher prostaglandin production is sometimes associated with a decreased tendency to metastasis, though there are features of the metastatic cascade where prostaglandins might be expected to facilitate dissemination. Prostaglandins may aid metastases through the mediation of proteolytic enzyme production, by prostaglandin-induced neovasculanzation or by subversion of the immune response. Inhibitors of platelet aggregation such as prostacyclin on thromboxane A2 synthetase inhibitors may be useful in the prevention of metastatic spread of cancer.

Incubation of cytosolic fractions of some tumour and normal cell lines with 5'-methylthioadenosine (5'-MTA) resulted in the formation of methionine. Methionine formation only occurred in those cell lines possessing 5'-MTA phosphorylase. The kinetics of product formation indicated that 5'-MTA was first rapidly converted into 5-methylthioribose-1-phosphate, followed by its slower conversion into methionine. Methionine synthesis from 5'-MTA was increased in cells previously incubated in methionine-depleted medium supplemented with 0.1 mM L-homocysteine for 24 hr. For most cell lines methionine synthesis from 5'-MTA was linear for only a short time period, and was followed by a first order decrease in the rate of methionine synthesis. Methionine synthesized from 5'-MTA was extensively incorporated into cellular macromolecules suggesting that 5'-MTA may substitute for methionine as a one-carbon source. This was confirmed by growth experiments which showed that low concentrations of 5'-MTA could partially substitute for methionine for some, but not all, cell lines. Higher concentrations of 5'-MTA were growth inhibitory. It may be possible to use 5'-MTA to selectively 'rescue' cells from methionine deprivation produced by the enzyme L-methioninase.

Human promyelocytic leukaemia HL60 cells were incubated with the glyoxalase intermediate S- d -lactoylglutathione in culture. The effects on cell proliferation, maturation, viability and cell cycle were investigated. When HL60 cells (5 × 10 4 /ml) were incubated with 50–500 μM S- d -lactoylglutathione for two days, the rate of cell proliferation was decreased. This effect was maximal at 500 μM S- d -lactoylglutathione where the cell proliferation rate was only 16% of control levels. There was a concomitant decrease in cell viability but little differentiation. During the first day of treatment, there was a significant decrease in the percentage of cells in the G 2 -M phase of the cell cycle with a concomitant increase in the G 0 -G 1 phase. In contrast, when HL60 cells were incubated with 1.0–1.5 mM S- d -lactoylglutathione, the inhibition of cell proliferation was progressively lifted, with a concomitant increase in the percentage of differentiated cells (27% differentiation with 1.5 mM S- d -lactoylglutathione). The activities of glyoxalase II and γ-glutamyl transpeptidase were increased in these cells. S- d -Lactoylglutathione slowly entered the HL60 cells and was consumed over the period when changes in cell cycle distribution, growth arrest and decrease in cell viability were observed. The mechanism of inhibition of proliferation of HL60 promyelocytes by S- d -lactoylglutathione is unknown but it may be related to the ability of S- d -lactoylglutathione to stimulate the assembly of microtubules.

The intracellular level of guanosine 3′,5′-monophosphate (cGMP) has been measured in Walker carcinoma cells in tissue culture after treatment with various alkylating agents. At concentrations which caused a rise in the level of adenosine 3′,5′-monophosphate (cAMP) chlorambucil and 5-(1-aziridinyl)-2,4-dinitrobenzamide (CB 1954) produced only a small (35%) elevation of cGMP, while merophan had no such effect. This suggests that any effect of cAMP will not be outweighed by an equivalent rise in cGMP. Specific cytosolic binding of cGMP decreased with increasing resistance of Walker cells to alkylating agents, while the dissociation constant, KD, for binding increased. This was also observed with cAMP binding which suggests that the same protein is responsible for binding both nucleotides.

The adenosine 3′,5′-monophosphate phosphodiesterase of Walker rat carcinoma 256, ADJ/ PC6 plasma cell tumour, NK lymphoma. Sarcoma 180 and TLX5 lymphoma behaves kinetically as if two separate activities exist, one with a low affinity for the substrate and the other with a high affinity. The high K m values are 82·5, 566, 590, 1975 and 1075 μM for the enzyme from each tumour respectively, and the low K m values are 1·1, 17·7, 5·75, 7·1 and 4·4 μM. In the Walker carcinoma and PC6 plasma cytoma, tumours which are sensitive to alkylating agents, the apparent V max of the low K m forms are respectively 38 and 25 per cent of the total activity. In those tumours which are naturally resistant to the growth inhibitory properties of the alkylating agents, the apparent V max of the low K m form is less than 10 per cent of the total activity. Walker carcinoma showing a 20-fold resistance to chlorambucil[ p -(di-2-chloroethylamino) phenylbutyric acid] has the V max of the high affinity form decreased to 15 per cent of the total, while a 70-fold resistant line has the V max of this form of the enzyme decreased to 9 per cent of the total. This decrease in the activity of the high affinity form of the enzyme in the resistant tumours does not appear to be due to the presence of an endogenous inhibitor. While the high K m form of the enzyme in the Walker carcinoma is mainly confined to the cytosol, about half of the activity of the low K m form appears to be associated with particulate fractions. This subcellular distribution does not differ appreciably in the sensitive and resistant tumours. The possible reasons for such an isoenzyme shift are discussed.

Two major isoenzymic forms of adenosine 3′,5′-monophosphate (cyclic AMP)-dependent protein kinase were resolved by DEAE cellulose chromatography of cytosols of normal human lymphocytes and of abnormal (colchicine ultrasensitive) lymphocytes from patients with chronic lymphocytic leukaemia (CLL). With only one exception the activity of type I protein kinase exceeded that of type II in normal lymphocytes, whilst in CLL the activity of the type II isoenzyme exceeded that of the type I. The total activity of protein kinase was approximately the same in normal and CLL lymphocytes. This difference in isoenzyme distribution may reflect the immaturity of CLL lymphocytes.

The extent of binding of 8-[ 3 H]adenosine 3′,5′-monophosphate (cyclic AMP) to specific sites was measured in Walker carcinoma cells of tissue culture lines, having different degrees of resistance to the cytotoxic effects of alkylating agents. When compared with sensitive cells, the resistant lines showed a loss of binding activity with increasing resistance, when measured at either pH 4.0 or pH 6.5. This applied to cell lines made resistant either to 5-aziridinyl-2,4-dinitrobenzamide (CB 1954) or chlorambucil, although the total loss of binding activity was greater in the former cell lines. Scatchard analysis of binding suggested the presence of two sites in all cell lines with K D I ∼ 1–5 × 10 −9 M and K D II ∼ 3 × 10 −8 M. The decreased binding activity was not due to an increased cAMP phosphodiesterase activity in the resistant cell lines, since the activity of the high affinity form of this enzyme was either the same as (CB 1954-resistant) or less than (chlorambucil-resistant) that found in sensitive cells. Walker cells with acquired resistance to either CB 1954 or chlorambucil showed a degree of cross resistance to the cytostatic effect of cAMP analogues and of other alkylating agents.

The specific activity of the enzyme serine hydroxymethyltransferase (EC 2.1.2.1) was determined in various murine, rat and human tumour cell lines. The activities of the enzyme were also investigated in tissues of non-tumour bearing DBA 2 mice and BALB c mice bearing the PC6 ascites tumour. The highest enzyme activity in the murine tissues was found in the liver and then the kidneys. The enzyme was present in all the tissues assayed. The activities of enzyme found in the tumours varied considerably, with the PC6 ascites, Walker 256 and Lewis lung cells, being the highest.

Transference of Walker carcinoma and TLX5 lymphoma from normal L -methionine-containing medium to medium containing limiting amounts of L -methionine, or L -homocysteine only, caused a 2-fold increase of S-adenosylmethionine decarboxylase activity. Kinetic analysis showed an increase in the V value of the enzyme from 22 to 53 pmol/min per mg protein in media containing only 0.1 mM L -homocysteine, without any alteration in the Km value (0.1 mM). The increase in enzyme activity does not result from (a) a reduction of the intracellular level of S-adenosylmethionine, since cycloleucine, an inhibitor of methionine adenosyltransferase, had no effect on enzyme activity; (b) an increase in intracellular adenosine 3′,5′ monophosphate (cyclic AMP), since high extracellular concentrations of N6-monobutyryl cyclic AMP had no effect on enzyme activity; (c) an alteration of polyamine levels, since addition of micromolar concentrations of exogenous putrescine, spermidine and spermine did not prevent the induction of S-adenosylmethionine decarboxylase activity in methionine-free media containing 0.1 mM L -homocysteine. The increased enzyme activity appears to be mainly due to enhanced stabilization, since the half-life was increased from 2.45 to 5.0 h in media containing only 0.1 mM L -homocysteine. Induction of enzyme activity is specific to the removal of L -methionine, since no increase occured in the absence of L -serine or L -glycine, or both, or by reduction of the serum concentrations in the medium.

Walker carcinoma cell lines sensitive or resistant to bifunctional alkylating agents have been found to contain multiple forms of cyclic AMP phosphodiesterase (3′:5′-cyclic AMP 5′-nucleotidohydrolase, EC 3.1.4.17). These activities have been resolved using Sepharose 6B gel filtration and their apparent molecular weights have been estimated. The enzyme appears to occur in four active forms of apparent mol. wts of > 62;1 000 000, 430 000, 350 000 and 225 000, when assayed at low substrate concentrations. Evidence has been obtained which suggests that all four forms of the enzyme are composed of subunits of mol. wt of approximately 15 000 and are interconvertible. While the ionic strength of the buffer affected the predominance of the different forms, the presence of cyclic AMP at 10 −6 M had no effect on aggregation or dissociation of the enzyme. An activity shift from high molecular weight forms of the enzyme to low molecular weight forms has been found in the resistant tumour at low substrate concentration. No change in elution profile between sensitive and resistant tumours was observed for the low affinity form of the enzyme. The pH optima of the enzymes with both high and low affinity for the substrate was found to be pH 8.0 in the sensitive line. In the resistant tumour the pH optima of the high affinity form is shifted to pH 8.4 while the low affinity form remains at pH 8.0. The high affinity forms of the phosphodiesterase in the sensitive and resistant tumour also differed in their inhibition by theophylline. In both cases inhibition was of the competitive type with K i values for the sensitive and resistant lines being 2.35 and 0.32 mM, respectively. There was no significant difference in the inhibition of the low affinity form between the sensitive and resistant tumour.

The growth inhibitory effects of N6-monobutyryl adenosine 3',5'-monophosphate (mbcAMP) and N6,O2'-dibutyryl adenosine 3',5' monophosphate (dbcAMP) towards Walker carcinoma in vitro are significantly potentiated by the addition of L-homocysteine to the culture medium. This effect is not seen with L-cysteine or when exogenous cAMP or prostaglandin E2(PGE2) replace the butyrylated cyclic nucleotide. Combinations of mbcAMP or dbcAMP and L-homocysteine significantly inhibit nucleic acid methylations. Both the butyrylated cyclic nucleotides cause an elevation of the intracellular level of S-adenosyl-L-homocysteine (SAH), a potent inhibitor of S-adenosyl-L-methionine (SAM) dependent methyltransferases, and this is significantly enhanced in combination with L-homocysteine. The increase in SAH level produced by such combinations is proportional to the inhibition of methyl group incorporation into 5-methyl cytosine and 7-methyl guanine. These results suggest that L-homocysteine potentiates accumulation of SAH in the presence of mbcAMP and dbcAMP and that the resultant inhibition of methylation accounts for the enhanced growth inhibition.

The sensitivity for recognition of cyclophosphamide and its metabolites by adenosine 3',5'-monophosphate (cAMP) specific proteins has been investigated. A 4-hydroxyl substituent in the 1,3,2-oxazaphosphorine ring is required for inhibition of cAMP binding to both cAMP phosphodiesterase and the regulatory subunit of the cAMP protein kinase holoenzyme. Binding to the latter causes an activation of the kinase and results in a dissociation into regulatory and catalytic subunits. The inhibitor constant Ki, for the inhibition of cAMP binding (0.19 mM) correlates well with that for inhibition of the low Km form of the phosphodiesterase (0.19mM). In both cases inhibition is of the competitive type. Although 4-ketocyclophosphamide resembles 4-hydroxycyclophosphamide in electron donating properties it is inactive with respect to binding to cAMP specific sites. This probably results from the difference in conformation of the rings of these two compounds.

8-Carbamoyl-3-(2-chloroethyl)imidazo[5,1-d]-1,2,3,5-tetrazin-4-(3H )-one- mitozolomide (CCRG 81010, M & B 39565, NSC 353451) is a potent inhibitor of the growth of a number of experimental tumours and can potentially decompose to give either an isocyanate or the monochloroethyltriazene (MCTIC). In vitro CCRG 81010 is not cross-resistant with the bifunctional alkylating agents against the Walker carcinoma. To investigate the mechanism of the antitumour activity of CCRG 81010 a comparison has been made with BCNU and MCTIC on precursor incorporation into macromolecules in TLX5 mouse lymphoma cells. Whereas BCNU produces a rapid and extensive inhibition of both (methyl 3H) thymidine and [5-3H]uridine incorporation into acid-insoluble material, neither CCRG 81010 or MCTIC have an early effect on precursor incorporation. Inhibition of precursor uptake is also not produced by concentrations of 2-chloroethylisocyanate that inhibit intracellular glutathione reductase activity. The potential carbamoylating activity of CCRG 81010 has also been assessed by comparing its effect with that of BCNU and 2-chloroethyl isocyanate on enzymes known to be inhibited by carbamoylation. Such enzymes, glutathione reductase, chymotrypsin and gamma-glutamyltranspepidase are not inhibited by CCRG 81010 under conditions where BCNU and 2-chloroethyl isocyanate show complete inhibition of enzyme activity, suggesting an absence of carbamoylating species. The results suggest that the most likely antitumour metabonate produced from CCRG 81010 is the triazene MCTIC.

The in vitro sensitivity of bone marrow cells from patients with leukaemia and from patients with non-malignant diseases to L-methionine removal by L-methioninase (L-methionine-alpha-deamino-gamma-mercaptomethane-lyase, EC 4.4.1.11) was determined using the incorporation of [methyl-3H]thymidine into acid-insoluble material as an index of survival. When compared with controls growing in medium containing 10 micrograms/ml of L-methionine, leukaemic cells showed a lower incorporation of [methyl-3H]thymidine after 24 h in the presence of 0.1 (normal 78 +/- 24%; leukaemic 26 +/- 18%, p less than 0.01) or 0.05 (normal 84 +/- 15%; leukaemic 50 +/- 21%, p less than 0.01) units of L-methioninase per ml. A similar differential sensitivity of leukaemic cells to L-methioninase was seen after 48 h of incubation. There was little effect on [methyl-3H]thymidine incorporation in the presence of boiled enzyme. Attempts to reverse L-methioninase toxicity with D-homocystine did not result in a differential effect on the normal cell population. The effects of L-methionine removal with L-methioninase were similar to those observed in L-methionine-depleted culture medium supplemented with 0.1 mM L-homocysteine. After 24 h in such depleted media leukaemic cells showed a lower incorporation of [methyl-3H]thymidine into acid-insoluble material (normal 88 +/- 17%; leukaemic 35 +/- 14%, p less than 0.01) and there was an elevation of the L-methionine-dependent enzymes: methionine adenosyltransferase, tRNA methyltransferase and S-adenosylmethionine decarboxylase. These results suggest the possibility of trying L-methioninase in the treatment of suitable leukaemias.

The in vitro l -methionine requirement of bone marrow cells from patients with leukaemia was compared with that from patients with non-malignant diseases by means of incorporation of (methyl 3 H) thymidine into acid-insoluble material. After 24 h in a medium in which l -methionine was substituted by l -homocysteine there was a lower incorporation of (methyl 3 H) thymidine by leukaemic marrow at either 0.1 mM (normal 84 ± 15; leukaemic 60 ± 11%, p l -homocysteine (normal 76 ± 13; leukaemic 58 ± 9%, p > 0.01). The response of bone marrow cells from leukaemic patients in regression was similar to that of normal marrow. Bone marrow cells from non-leukaemic patients required a lower concentration of l -methionine in the culture media for optimal incorporation of (methyl 3 H) thymidine than did those from leukaemic patients. Bone marrow cells from leukaemic patients had a decreased maximal initial rate of l -methionine transport ( V max ) when compared with marrow cells from normal subjects, though in all cases the Km value for l -methionine transport was lower in leukaemic cells. Saturation studies suggested that the size of the cytoplasmic pool for l -methionine was larger (about two times) in marrow cells from normal subjects. These results suggest a therapeutic strategy based on a reduction of the plasma l -methionine concentration.

A comparison has been made of the cachectic effects produced by the transplantable murine adenocarcinoma of the mouse colon (MAC16) with tumour necrosis factor-alpha (cachectin). Tumour necrosis factor-alpha (TNF-alpha) produced a dose-related weight reduction that was accompanied by a decrease in both food and water intake. The degree of weight loss was directly proportional to the decreased food and water intake. In contrast weight loss produced by the MAC16 tumour occurred without a reduction in fluid or nutrient intake. Both the MAC16 tumour and TNF-alpha produced hypoglycaemia and a reduction in the circulatory level of free fatty acids (FFA), but had opposite effects on the level of plasma triglycerides with the MAC16 tumour-induced cachexia causing a decrease and TNF-alpha producing an increase. The MAC16 tumour elaborated a lipolytic factor which caused an immediate release of FFA from adipose tissue. In contrast TNF-alpha had no effect on mobilization of adipose triglycerides over a short time period. Both TNF-alpha and extracts from the MAC16 tumour caused an enhanced release of amino acids from mouse diaphragm, which was suppressible with indomethacin and heat labile. No TNF was detected in the MAC16 tumour or in the serum of tumour-bearing animals. Both tumour and non-tumour-bearing animals responded with a similar elevation of their serum TNF levels 90 min after a single injection of endotoxin. It is concluded that weight loss produced by TNF-alpha arises from an anorexic effect and that this differs from the complex metabolic changes associated with cancer cachexia.

The effect of a number of anti-tumour agents on prostaglandin (PG) production from arachidonate by sheep seminal vesicles has been investigated. Of the drugs examined only those belonging to the alkylating agent type series showed inhibition of enzyme activity. Unlike most inhibitors of PG synthetase (EC 1.14.99.1) these agents caused an inhibition of prostaglandin E2 (PGE2) production, while unaffecting the formation of prostaglandin F 2alpha (PGF2 alpha). This suggests the site of inhibition is the isomerase converting prostaglandin H2 (PGH2) to PGE2. Kinetic studies indicated that merophan [DL-o-micron-(di-2-chloroethylamino)phenylalanine] inhibited the synthetase competitively with respect to substrate. The kinetics of the inhibition were also consistent with the formation of a reversible enzyme-alkylating agent complex prior to irreversible inactivation of the enzyme. The inactivation process could be described by the Main equation from which a dissociation constant (Kd) and a reaction rate constant (k2) were calculated. The inhibition of PG synthetase may be important in the anti-tumour effect of these agents.

Monobutyryl adenosine 3′,5′ monophosphate (mbcAMP) caused an inhibition of the thymidylate synthetase activity of Walker rat mammary carcinoma cells in tissue culture, which could be reversed by concomitant treatment with N2,O2′ dibutyryl guanosine 3′,5′ monophosphate (dbcGMP). There was no effect on thymidine kinase activity. The DNA polymerase activity of whole cells, but not broken-cell preparations was markedly inhibited by a dose of mbcAMP (100 μg/ml) having little effect on growth rate. This inhibition could be reversed to some extent by simultaneous treatment of the cells with caffeine. Treatment with mbcAMP produced a decrease in the level of dTTP and a concomitant rise in the levels of dATP, dGTP and dCTP. This situation was reversed in combination with dbcGMP, with levels of the deoxyribonucleoside triphosphates tending to revert back to control values. The effect of mbcAMP on thymidylate synthetase and DNA polymerase may account for its growth inhibitory effect.

The effect of some alkylating agents on the activity of the adenosine 3′,5′-monophosphate (cyclic AMP)-dependent protein kinase has been studied in Walker cells sensitive and resistant to the cytotoxic action of such agents. Chlorambucil (5 μg/ml) caused an activation of the cAMP-dependent protein kinase in sensitive Walker carcinoma cells which reached a maximum 1.5 hr after drug addition. Sephadex gel chromatography indicated that during this activation, the catalytic subunit of the protein kinase was released from the holoenzyme to the same extent as that measured in the crude supernatant of the tumour cells. The degree of activation was equivalent to that produced by 100 μg/ml of N 6 O 2′ -dibutyryl cAMP. In contrast, the monofunctional N -ethyl analogue of chlorambucil had no effect on the cAMP-dependent protein kinase at a dose of 250 μg/ml. The protein kinase activity ratio in sensitive cells increased with increasing doses of chlorambucil and reached a maximal activation at a concentration of 5 μg/ml, which was sufficient to cause complete inhibition of tumour cell growth. A much larger dose of chlorambucil (100 μg/ml) was required to cause activation of the kinase in Walker cells resistant to this agent. Chlorambucil (25 μg/ml) also caused an activation of the cAMP-dependent protein kinase in TLX5 cells, though the time scale of the activation differed for that found in Walker cells. Both merophan and 5-aziridinyl-2,4-dinitrobenzamide (CB 1954) caused an increase in the protein kinase activity ratio of sensitive Walker cells. The increase caused by CB 1954 could be abolished by 4-amino-2-phenylimidazole-5-carboxamide (2-phenyl-AlC). which reverses the tumour growth inhibitory action of CB 1954. The degree of stimulation of the cytosolic protein kinase by saturating concentrations of cAMP, and the apparent dissociation constant for cAMP bound to protein kinase decreased with increasing resistance of the cell lines to alkylating agents. These results sugest that the biological effect of the increase in cAMP in sensitive Walker cells induced by the alkylating agents is mediated through a protein kinase.

When injected into female NMRI mice tumour necrosis factor-alpha (TNF) produced a dose-related weight loss over the first 24 h, which was accompained by and directly proportional to a decrease in both food and water intake. Body composition analysis after the first 8 h revealed that the weight loss was associated with a decrease in body water without a change in the other body compartments. Since the TNF-induced weight loss was accompanied by hypophagia and a concomitant hypoglycaemia we attempted to reverse the weight loss by force-feeding either glucose or medium chain-triglycerides (MCT). Weight loss induced by TNF was reversed by both glucose and MCT and by administration of the equivalent volumes of water alone. The weight reversal was accompanied by an increased body water content. This suggests that the weight loss in mice induced by TNF is, at least in part, due to dehydration.

Although homocysteine was unable to support growth of Walker carcinoma in media lacking methionine it did enable some proliferation of TLX5 lymphoma. In both cell lines there was an increase in growth rate in the presence of homocysteine at limiting methionine concentrations. The proliferation rate of Walker carcinoma was proportional to the methionine concentration of the medium down to 0.5 μg/ml, whereas growth of TLX5 lymphoma was only slightly reduced at such methionine concentrations. The difference in proliferative ability between the two cell lines was reflected in the level of S- adenosyl- l -methionine under conditions of methionine deprivation. In both cases transferance to a media in which methionine was growth limiting caused a rapid increase in the activity of tRNA methyltransferases to levels six to seven-fold greater than the control. The initial increase in methylase activity was not prevented by cycloheximide, although after 4 h there was a progressive decrease in activity which approached control values within 24 h. The increase in tRNA methyltransferase activity on removal of the normal level of methionine in the medium was also seen with human embryonic fibroblasts, which are able to proliferate normally in methionine-deficient, homocysteine-supplemented media. These results suggest that methyltransferase activity may be regulated in part by the S-adenosylmethionine content of the cell.

The possible role of fluroacetate in the toxicity and antitumour activity of the fluroethylnitrosoureas, BFNU and FCNU has been studied in CBA mice bearing the TLX5 lymphoma either sensitive (TLXS) or resistant (TLXRT) to nitrosoureas. Treatment of mice bearing either TLXS or TLXRT tumours with either BFNU or FCNU caused an elevation in the citrate levels of heart, kidney and tumour, but not the liver, 24 hr after drug administration. Heart citrate levels were maximally elevated 10-fold, while the levels in kidney and tumour were increased 3- to 6-fold. Tissue levels of flurocitrate were determined by glc after conversion to the ethyl ester. This showed maximum levels of fluroacetate production in heart, with lower levels in kidney, tumour and liver. Treatment of K562 human erythroleukaemia cells in vitro with BFNU caused an inhibition in the production of 14CO2 from 14C palmitate and [U-14C] glucose. These results suggest that some of the effects of the fluroethylnitrosoureas may be related to fluroacetate production and the consequent blocking effect on aconitase. This effect is probably related more to the generalized toxicity of these agents than to their therapeutic efficacy.

MAC16 is a chemically induced, transplantable adenocarcinoma of the colon passaged in inbred NMRI mice. At small tumor burdens (less than 1% of the host weight), weight loss was observed without a reduction in food intake. As the tumor mass increased, weight loss also increased and reached 33% of host body weight in females and 20% in males when compared with the weight of age-matched controls. The reduction in host body weight was directly proportional to the tumor size and was reversible when the tumor was excised. There was a preferential loss of body fat in tumor-bearing animals with an increase in the plasma level of free fatty acids, although there was a minimal elevation of ketone bodies. Tumor growth was accompanied by progressive hypoglycemia and a reduction in the plasma insulin levels. The decrease in plasma insulin may have contributed to the catabolic effects of progressive tumor growth.