The glycolytic metabolism of tumour tissue 5 has received considerable attention since the 1920s and the first reports of high rates of lactate-production noted by Otto Warburg. Since then, many changes have been reported in the pathways of glucose import and glycolysis in tumour cells compared with the 0 corresponding normal cell. These include alterations to the glucose transporter (Kawai et al., 1971; Birnbaum et al . , 1987; Flier et al . , 1987 Bramwell et al . , 1978 and White et al . , 1981), inap; ropriate isoenzyme displays for glycolytic enzymes (Weinhouse, 5 1972; Van Veelen et al . , 1978 and Ibsen et al . , 1982), incidence of tyrosine phosphorylation of glycolytic enzymes (Presek et al . , 1980 and Cooper et al . , 1983) and alterations in maximal activity for some enzymes catalysing non-equilibrium reactions (Shatton et al . , 0 1969 for hexokinase and Ibsen et al . , 1982 for pyruvate kinase) .

Attention, during the present work, has focussed on the stage of glucose-phosphorylation, the first non-equilibrium reaction in the glycolytic 5 pathway from intracellular glucose in tumour cells. Normally, in tissues other than liver and pancreatic β-cells, a combination of isoenzymes I, II and III of hexokinase catalyses this reaction. In the first mentioned two tissues, a fourth isoenzyme is present, 0 hexokinase IV or glucokinase. However, available evidence suggests that hexokinase activity may assume a new and important regulatory role in tumour cells (Board et al . , 1990(a)) . Indications are that the in vivo activity of this enzyme may approach its maximal

35 in vitro activity (Board, 1990 (b) ) . Moreover, given

the high intracellular concentrations of glucose in some tumour cells, hexokinase may approach saturation with its pathway substrate (Board, 1990 (b) ) . In this case, hexokinase may be the flux-generating step for glycolysis in the tumour cell. This would be an unusual situation, since for tissues such as muscle, it is considered that the step of glucose import into the cell constitutes the flux-generating step (Newsholme and Board, 1991 ) .

The altered role for hexokinase in tumour cells raises many interesting possibilities. Factors which act on the enzyme in such a way as to affect its in vivo activity will alter the rate of flux through glycolysis and this process will be independent of the intracellular glucose concentration.

It has now been found that hexokinase IV (glucokinase) activity is present in tumour cells. This has been demonstrated in more than twenty cultured tumour cell-lines. ^Tne invention provides a method of inhibiting the growth of tumour cells comprising contacting the cells with an inhibitor of glucokinase.

The invention also provides the use of an inhibitor of glucokinase for inhibiting the growth of tumour cells.

The inhibitor of glucokinase may be any compound which suppresses or inhibits the activity of the glucokinase enzyme, either directly or indirectly. The inhibitor is preferably otherwise substantially non-toxic. Compounds which inhibit glucokinase directly include certain ^' sugars and sugar analogues, such as some substituted or unsubstituted hexoses and heptoses. Glucokinase inhibitors include the following compounds and their analogues and

derivatives :

(1) D-Mannoheptulose (D-mannoketoheptose) and related compounds such as mannoheptose, both of which are commercially available.

(2) D-Glucoheptose (D-glycero-D-guloheptose, commerci ly available) and derivatives.

(3) A β-methylamide derivative of D-glucose (N- acetyl-β-D-glucopyranosyl amine) .

(4) A diethyl ester substituent of D-glucose.

(5) Derivatives of D-mannose, generε ^{^} y.

(6) N-acetyl glucosamine (commercially available) and related compounds.

(7) 1 -methyl-N-acetylglucopyranosyl amine.

The structures and formulae of some compounds of the type (1) to (7) above are given in Table 7. All of these have been tested and found to inhibit glucokinase activity.

Indirect inhibitors of glucokinase will include for example compounds which reduce or block transcription of the glucokinase gene or translation of the mRNA, such as oligonucleotide sequences or oligonucleotide-analogue sequences which are capable of hybridising to part or all of the DNA or RNA sequences which encode glucokinase. It is expected that abnormal derepression of the glucokinase gene in tumour cells is causing the abnormal production of the glucokinase enzyme. Thus, interruption of transcription- or

translation-related events is one way to inhibit glucokinase in an indirect fashion. The amino acid sequence of glucokinase and the sequence of the glucokinase gene are known and published. (Andreone et al., 1989; Magnuson et al. , 1989).

A further type of glucokinase inhibitor uses the "magic bullet" approach and involves antibodies directed to eg. glucokinase or glucokinase mRNA. Such antibodies can be raised by known methods, including the development of hybridomas which secrete monoclonal antibodies of the desired specificity. These antibodies can then be delivered to tumour cells to inhibit growth.

The invention also provides novel compounds which inhibit glucokinase, such as compounds (3), (4) and (7) in table 7 below. Other novel sugars, particularly substituted hexose and heptose sugars, which can be made by known methods, are also expected to work as glucokinase inhibitors. These novel compounds will be suitable for inhibiting the growth of tumour cells. The compounds may be useful in therapy, particularly for treatment of tumours. The invention also provides compositions comprising such compounds, for inhibiting the growth of tumour cells. Inhibitors having I values, (I _5Q being the concentration required to inhibit enzyme activity by 50%) of between <1μM and 250mM, are expected to be effective in tumour inhibition. Inhibitors having an

I,- _n in the range 100μM to 35mM will be most useful. A further aspect of the invention is the treatment of patients with tumours. The invention provides a method of treatment of a tumour in a patient comprising administering an effective amount of a glucokinase inhibitor to the patient; or alternatively, the use of a glucokinase inhibitor for the treatment of

a tumour in a patient. Administration may be by any route, such as oral, intravenous, intraperitoneal or direct injection into the tumour. Most probably, compounds will be given orally or by intravenous infusion. The invention also provides the use of a glucokinase inhibitor in the manufacture of a medicament for the treatment of a tumour in a patient. A biopsy sample showing glucokinase activity would indicate the likelihood that this mode of treatment would be successful.

Further provided by the invention is a pharmaceutical formulation comprising a glucokinase inhibitor and a pharmaceutically acceptable carrier.

A scheme envisaged for treatment of patients involves administration of mannoheptulose, which is thought to be free of toxic side-effects and is a naturally-occurring compound, or of other glucokinase inhibitors, to patients between bouts of radiotherapy or chemotherapy. This scheme, including comparison with a control group receiving a placebo, would form the basis of clinical trials. For the purposes of the trials, the mannoheptulose (or other glucokinase inhibitor) and control groups will be matched for age, state of progression of the disease and state of health discounting the cancer. An upper limit of dosage of mannoheptulose would be in the region of 200mg/kg. This reduces to an approximate equation for other inhibitors of

I _ (M) x 4.6/70 x molecular weight (g) giving a dose in mg/kg for an average-sized person.

This is intended to give only a very rough indication of the upper limit of dosage and not to be a clinical guideline. In practice, much smaller doses seem to work very well in the experimental animal. In aspects of the invention which involve a

glucokinase inhibitor, a combination of two or more glucose inhibitors may be used.

Still another aspect of the invention concerns the diagnosis of a tumour. In this aspect, the invention provides a method of diagnosing a tumour comprising detecting the presence of glucokinase activity, or of glucokinase protein or messenger RNA (mRNA) , or expression of the glucokinase gene, in a sample suspected of containing malignant or pre- malignant cells. The invention also provides compounds which bind specifically to glucokinase or glucokinase mRNA for use in the diagnosis of tumours.

Diagnostic tests may require a small biopsy sample, or in the case of non-solid tumours such as leukemia, a blood sample could be used. Examples of possible diagnostic tests are as follows:

(1) Probing for glucokinase protein or messenger RNA with a radiolabelled probe in a population of extraheptatic and extapancreatic cells would show the presence of glucokinase in tumorigenic cells. The gene sequence of glucokinase is known and 45% of it has been published. This is a technique of great sensitivity and would be expected to give positive results for a population of cells containing only a small proportion of tumorigenic cells. Since the suppressed cells of the present study (see Examples) also show glucokinase activity, it is expected that pre-malignant cells would also test positive enabling preventive treatment to be administered before the tumour cells begin to proliferate.

(2) Culture of cells in 96 well plates over a period of 72 hours is a standard laboratory procedure for measuring growth rates by means of incorporation

of radiolabelled thymidine or a similar agent. This could be adapted to show malignant or pre-malignant states of cell-populations by inclusion of mannoheptulose or another inhibitor of glucokinase. Malignant or pre-malignant cells will have their rates of growth inhibited under these conditions, whereas normal cells will not.

Diagnosis may involve using a probe as noted above, labelled with a suitable label. Or, labelled antibodies, in particular monoclonal antibodies raised against the glucokinase nRNA or the glucokinase molecule itself, may provide suitable compounds for use in diagnosis.

The magnitude of glucokinase activity may relate to malignancy of .he tumour and if so, will be useful in prognosis.

The glucokinase inhibitors demonstrated in the Examples act on glucokinase by an unknown mechanism. Once their mode of action has been determined then it will be easier to design new glucokinase inhibitors, which IΓHV be based on hexose or heptose sugars or which may be entirely different types of compound.

Glucokinase has a much higher specificity for glucose than the other hexokinase enzymes. Hexakinases I, II and III are capable of phosphorylating many different sugars, but glucokinase is not. Glucose has a high Km (which describes how high the concentration of the substrate needs to be before an enzyme will work), of about 1OmM for glucose. This is useful for liver cells as it allows ^" high levels of glucose to accumulate. Other hexakinases are low Km enzymes, which work in the presence of a much lower concentration of substrate.

The invention will now be further illustrated

in the Examples which follow.

EXAMPLES

Materials and Methods

Cell-Lines Included in the Study

Cell-Lines of Human Origin

H.Ep 2 : epithelial cell from primary carcinoma of the larynx (morphologically indistinguishable from the HeLa cell) (Moore, 1955). ESH TR1-2 and ESH p6 : hybrid matched pair resulting from somatic cell hybridisation of a HeLa cell derivative, D98, and a human keratinocyte (Peehl et al. , 1981). 2B1 TG and 2B1 Coll: hybrid matched pair from fusion of D98 with a human fibroblast (Klinger, 1980). RVH 421 : melanoma cell-line (McCormick et al. , 1983). HT29 : carcinoma of the colon (epithelial) (Marshall, 1977). RT112 : carcinoma of the urinary epithelium (Marshall, 1977).

T24/83 : bladder carcinoma (Bubenik et al. , 1973). MRC 5 : diploid fibroblast from foetal lung of 14 weeks gestation. MRC 5 persists for 48 passages before the onset of senescence (Jacobs, 1970). Hs578T: breast carcinoma cell line (Hackett et al) . ZR-75-1 and T47D: breast carcinoma cell lines (Engel and Young, 1978). MCF-7: breast carcinoma cell line (Moscow et al, 1988).

Human melanoma * : tissue excised from a patient and frozen in liquid nitrogen before assay of enzyme activity.

Cell-Lines of Rat Origin

The rcc 1 group of cell-lines was established in the William Dunn School of Pathology, Oxford

(Moraes, 1989). Cells from a new-born rat heart were

cultured and three normal clones isolated and propagated, rcc 1 c2, c3 and c4, along with the heterogeneous parent cell-line, rcc 1. After 25 passages, clones with primary tumorigenic activity could be isolated from rcc 1 and these were cultured separately as rcc 1 t. After 52 passages, clones with metastatic activity could be isolated from the tumorigenic culture and these were propagated as rcc 1m.

LLC WRC 256 : epithelial carcinoma (Dunham, 1953) 0

Culture and Harvesting of Cells

Cells were cultured in minimal essential medium containing 5% foetal calf serum, 3.5 mM L-

15 glutamine, streptomycin and penicillin, at 37°C with 5% CO- in air. Cells were removed from the plastic substrate by incubation for 10 minutes at 37"C with phosphate buffered saline (PBS), 0.4 % in EDTA, after removal of the growth medium. Cells were collected by 0 centrifugation (3 minutes at 2000 rpm), resuspended in 5 volumes of the appropriate extraction buffer and homogenised on ice in a glass hand-held homogeniser. The homogenate was stored on ice for the duration of the experiment.

- -

Assay of Glucokinase Activity

The enzyme was assayed by an adaptation of the method of Stanley et al. (1984). Measurements

30 depend on the separation of radiolabelled glucose-6- phosphate, formed from U- 14C-glucose by action of glucokinase, from remaining glucose. Th: is achieved with the aid of ion-exchange discs which bind and retain the charged glucose-6-phosphate, but not

35 glucose. Cells were extracted and homogenised in a medium containing 50 mM TRA-HC1, 300 mM sucrose, 100 mM

KC1 and 1 mM EDTA at pH 7.5. Aliquots (90μl) of assay medium containing 50 mM Tris, 10 mM MgCl ₂ , 5 mM ATP, 10 mM glucose-6-phosphate, 100 mM glucose (incorporating 4 μCi/ml of D-U- 14C-glucose) , pH 8.0, were pre-incubated at 30°C. Homogenate was pre-incubated separately. At zero time, 20 μl of homogenate was added to each aliquot of assay medium and the reaction stopped with

0.25 ml ice-cold methanol 12 minutes later. The resulting mixture was cooled on ice. Aliquots of medium (10 μl) were spotted onto each DEAE-cellulose ion-exchange disc and allowed to dry. This procedure was repeated up to a total volume of 40 μl per disc. The discs were left to dry for 30 minutes before being washed on a microfilter holder with distilled water. The resulting washed discs were allowed to dry for 48 hours and the radioactivity was measured with aqueous scintillation fluid (dispersions per minute) in a Beckman scintillation counter.

Assay of Hexokinase I, II and III Activities

Cells were extracted in three to four volumes of extraction buffer comprising 50 mM TRA-HC1, 1 mM EDTA, 2 mM gCl ₂ and 26 mM mercaptoethanol at pH 7.5. The change in absorbance at 340 nm, due to the reduction of NADP , was followed on a recording spectrophotometer. The linear nature of the initial reaction was established over the time course of the experiments. Assays were performed in a medium (pH 7.5) containing 75 mM Tris, 7.5 mM MgCl_, 0.8 mM EDTA, 1.5 mM KC1, 4 mM mercaptoethanol, 0.4 mM NADP ⁺ , 2.5 mM ATP, 10 mM glucose, 0.05 % triton, 7.8 mM creatine phosphate, 0.175 U/ml creatine kinase and 7 U/ml glucose-6-phosphate dehydrogenase, to which homogenate was added to give a total volume of 1 ml.

Measurements of Rates of Glucose Consumption

Cells were harvested and washed 3 times with PBS (pH 7.2), before resuspension in PBS. PBS intended for the incubation medium was gassed with 100 % oxygen for 30 minutes prior to the experiment. To Erhlenmeyer flasks, previously treated with dimethyldichlorosilane (2 % in 1 , 1 , 1-trichloroethane) to give water-repellent properties, were added 0.5 mis of PBS, 0.2 mis of 10 % BSA, 100 μl of a 50 mM solution of glucose and cell- suspension to a final volume of 1 ml. Flasks were gassed with oxygen for 30 seconds and incubated, stoppered, for 120 minutes before the injection through the seal of 0.2 mis of 25 % perchloric acid and cooling on ice. The medium was neutralised to pH 7 with 40 % potassium hydroxide before being assayed for metabolites. Glucose concentration was estimated by the method of Bergmeyer et al. (1974).

Collection of ¹⁴ C0 ₂

14 C0 ₂ was collected and the radioactivity determined by the method of Leighton et al. (1985).

Estimation of Protein Concentration

Protein concentrations were determined according to the method of Bradford (1976).

Measurement of Cell-Population Numbers

Cells were cultured as normal in Minimal Essential Medium with foetal calf serum, glutamine and antibiotics and an appropriate concentration of mannoheptulose, sedoheptulose or galactose. Each flask was inocculated with an equal volume of cell-

suspension. Cells were extracted in phosphate buffered saline at 24 hour intervals and total cell numbers determined per flask by visual inspection of an aliquot of cell-suspension in a Neubauer Improved cell-counting chamber. Nutrients in the medium were not exhausted and the cell-population did not reach confluence over the 5 day period of the growth experiment.

3 Incorporation of H-Thymidine

Aliquots of cells (100 μl of a 1 x 10 per ml suspension in Minimal Essential Medium) were added to individual wells of a 96 well tissue-culture plate with 0.2 μCi per well of 6- H-thymidine. Appropriate volumes of sugar solution in growth medium were added to give final concentrations of 15 mM or 30 mM in a total volume of 200 μl per well. Plates were incubated at 37 ^* C for 72 hours before harvesting with a Skatron Combi cell harvester onto glass fibre discs. Radioactivity in the discs was measured using a Beckman scintillation counter.

Growth of Tumours in Host Animals

Genetically athymic nu/nu mice (female MF1 or

ICRF) were inocculated with 2 - 5 x 10 cells of the RT112/84, T24/83 or LLC WRC 256 cell-lines subcutaneously in the lumbar region. The animals developed visible tumours within 3 - 5 days which increased in size over the subsequent 15 - 20 days. The final tumour burden never exceeded 5 % of the animal's body weight. The animals ate and drank normally during the tumour-bearing period.

For studies on the inhibition of tumour growth by mannoheptulose, animals were injected intraperitoneally with 100 μl mannoheptulose solution

(1.7 mg/g body weight) or 100 μl saline. One injection was given daily over a 5 day period.

Results

Maximal glucokinase activities, in a range of cultured cell-lines and excised tumours from human, rat and somatic cell hybridisation sources, are presented in Table 1. Activities are on a par with or higher than that recorded for rat liver tissue measured during the present study and by Stanley (1984). An assay of rat heart tissue by the same method produced the expected absence of detectable glucokinase activity. A correlation of maximal glucokinase activity with growth rate (a quantity considered to be proportional to malignancy) reveals that the activity does not show a significant correlation with malignancy.

Maximal activities of hexokinase are presented for comparative purposes in Table 2. A wide variation in activity among the tumour cells is apparent (range 4 to 109 nmol/minute/mg protein), there being no correlation with malignancy.

Activities of glucokinase in the presence of the specific inhibitor of this enzyme, mannoheptulose, are presented in Table 3. Inhibition is marked for every cell-line showing glucokinase activity except for the weakly malignant product of somatic cell hybridisation, 2B1 Coll . Inhibition of activity ranges from 20 to 100 %.

Rates of consumption of glucose in the presence and absence of ^" mannoheptulose are presented in Table 4. The presence of the inhibitor effects a severe reduction in the rate of glucose consumption for the tumorigenic cell-lines. Percentage inhibition is 25 and 47 % for HEp 2 and ESH TR1-2 cell-lines, respectively, at 30 mM mannoheptulose.

Rates of growth of cells cultured in minimal essential medium in the presence of mannoheptulose are shown in figures 1 to 4. Concentrations of inhibitor used were 0, 20 and 100 mM. A severe inhibition of the rate of growth is achieved. Percentage inhibition ranges from 75 to 91 % at 100 mM mannoheptulose and 35 to 65 % at 20 mM, as the curve presented in figure 5 shows. Other sugars, galactose and sedoheptulose, which are known to have no effect on the activity of glucokinase were included as controls. These sugars have no effect on rates of growth of either tumorigenic or non-tumorigenic cell-lines.

A similar inhibition of proliferation was recorded when criteria used were reduction in cell numbers, in DNA synthesis or in protein synthesis (Table 5) .

The present study on human tumour cell-lines shows that many possess glucokinase activity and that the proliferation of these cell-lines can be inhibited by the presence of mannoheptulose. Furthermore, the growth of two human bladder tumour cell-lines in nude mice is dramatically reduced by only one injection per day of mannoheptulose (Table 6) . Proliferation of rat tumour cells from the LLC WRC 256 cell-line is only weakly inhibited either in culture or in the nude mouse model by the presence of mannoheptulose. This finding is consistent with the low level of glucokinase activity measured in this rat cell-line.

Table 8 shows that glucokinase activity is present in samples of malignant melanoma tumour, surgicallly removed from patients in a British hospital. The range of activities (1.67 to 9.07 nmol/minute/mg protein) is smaller than for the cultured cell-lines which suggests that the magnitude of the activity may be a characteristic of a given

tumour-type (in this case melanoma), regardless of source.

Glucokinase activities from all melanoma samples are inhibited by 1OmM mannoheptulose (a very low dose) and this suggests that this dose would inhibit rates of growth.

Table 10 shows the concentrations of various of t'-*? compounds from Table 7 required to bring about a 50% reduction in growth rate of a range ~>f cultured human cell-lines. All compounds so far tested which inhibit glucokinase activity also inhibit growth rates of tumour cells. The compounds listed on Table 10 have all been screened for inhibition of growth of the normal, non-tumorigenic cell-line, MRC5 (a diploid fibroblast from a human embryonic lung) . None of the compounds affected growth of this cell-line at concentrations of up to 150 mM.

Discussion

The pronounced inhibition in the growth rate achieved in the presence of the glucokinase inhibitor, mannoheptulose, is entirely specific for tumorigenic and suppressed cells. Suppressed cells, in this context, may represent a pre-malignant state. These cells have acquired immortality in culture, but fail to show the uncontrolled proliferation necessary for tumour production. Growth rates of normal, non- tumorigenic cell-lines are unaffected by this compound. In the case of the non-tumorigenic cell-line used during the present work, ^" a human diploid fibroblast, this is almost certainly due to the absence of detectable glucokinase activity. The failure of the control sugars, galactose and sedoheptulose, to affect growth rates makes it a more secure conclusion that the mannoheptulose effect is due to inhibition of

glucokinase activity.

The severe inhibition of growth rate attained with mannoheptulose makes this a potential therapeutic agent. It may be especially significant in conjunction with established forms of radiotherapy and chemotherapy, the debilitating side-effects of which mean that extensive recovery periods are necessary between bouts of treatment. Tumour regrowth often occurs during these periods and the availability of a non-toxic tumour growth inhibitor, such as mannoheptulose, could reinforce the more conventional treatments.

It is possible to envisage the appearance of glucokinase activity as a diagnostic aid for extrahepatic cancers (the majority of human cancers are epithelial carcinomas) . The assay of this activity is sensitive and can be performed quickly with unsophisticated equipment. It may be of particular application to the increasing problem of ultraviolet- induced melanoma (skin cancer) . A biopsy sample can be readily obtained in a simple operation under local anaesthetic in this condition. The excised human melanoma included in the present study showed a consistent and high detectable glucokinase activity. Mannoheptulose inhibits growth of human tumour cell-lines both in culture and of the two bladder cancer cell-lines studied in the nude mouse. The specificity of this effect in vivo is indicated by the much weaker inhibition of growth of the rat carcinoma which is predicted by the lower level of glucokinase activity evinced by these cells.

Concentrations of mannoheptulose required for inhibition of growth rate in the cells of the present study are relatively high. Inhibition ceases to be consistently significant below a concentration of about 5 mM (equivalent to approximately 1 g per kg body

weight) . This is probably owing to the relatively low affinity of glucokinase for this inhibitor. Rat liver glucokinase has a K. for mannoheptulose of 12.5 mM. However, it is possible that modification of the chemical structure of mannoheptulose may result in more efficient inhibition. This would result in lower doses being required for inhibition of tumour growth rate while retaining the non-toxicity of the parent compound. Studies in this area, looking at the effects of structural analogues of mannoheptulose and related sugars on tumour proliferation, are already underway.

Methods of Assessing Inhibition of Glucokinase Activity

Requirements include a sample of tissue known to have glucokinase activity, such as rat liver or one of the cell-lines mentioned above. The sample is homogenised and glucokinase activity is assayed by the method described above in the section headed "Assay of Glucokinase Activity". Inclusion of an appropriate concentration of a putative inhibitor and comparison with the activity measured in its absence will give an indication of its inhibitory properties.

Table 1

Maximal Activities of Glucokinase in Normal. Suppressed and Tumorigenic Cell-Lines

Means are presented + standard error of the mean (S.E.M.). nd denotes glucokinase activity non-detectable.

Numbers of observations (in brackets) represent separate batches of cells.

* denotes tissue excised from the host animal.

$ data calculated from that given by Stanley et al. (1984).

Table 2

Maximal Activities of Hexokinase in Normal. Suppressed and Tumorigenic Cell-Lines

Malignant Cell-Line State

Means are presented ± S.E.M.s.

Numbers of observations (in brackets) represent different batches of cells.

* denotes tissue excised from the host animal. All other cell-lines were cultured.

GU STFji

Table 3

Effect of Mannoheptulose on Maximal Activities of Glucokinase in Tumorigenic and Suppressed Cell-Lines

Cell-Line Activity (nmol/minute mg protein) Percentage Inhibition

0 M mannoheptulose 20 mM mannoheptulose

rcc 1 m 8.06 nd (3) $ 100

5.63 ± 0.34 (7) $ 69

Means are presented ± S.E.M.S.

Numbers of observations (in brackets) represent different batches of cells.

Comparison of the means in the presence and absence of mannohepnilose, by Student's t test, yields the probability of the null hypothesis p < 0.001 for all data marked $. Other differences are not significant.

* denotes tissue excised from the host animal. All other cell-lines were cultured. nd denotes glucokinase activity non-detectable.

Table 4

Effects of Mannoheptulose on Rates of Utilisation of Glucose by Normal and Tumorigenic Cell-Lines

Cell-Line Utilisation (nmol/minute mg protein) Percentage Inhibition

0 mM 30 mM Mannoheptulose Mannoheptulose

H.Ep 2 4.94 3.71 ± 0.10 (3) $ 25 (Tumorigenic)

ESH TR1-2 5.29 2.80 + 0.14 (3) $ 47 (Tumorigenic)

MRC 5 2.21 8.47 ± 0.32 (3)

(Untransformed)

Means are presented ± S.E.M.S.

Numbers of observations (in brackets) represent different batches of cells.

Statistically significant data are marked $ (p < 0.001).

For details see Board (1990 (b)).

Table 5

Effects of Mannose. Mannoheptulose and Sedoheptulose on Proliferation of Human Tumour Cell-Lines and Mouse Fibroblasts in Culture

3T6

- denotes no inhibition of proliferation, + denotes inhibition of proliferation.

Proliferation was measured as increase in cell number, DNA synthesis (Η-thymidine incorporation) and protein synthesis.

Table 6

Effects of Mannoheptulose on Eventual Tumour Weight in Host Animals

CeH Line used for Weight of Tumour ( g) Percentage inocculum Inhibition

Saline Mannoheptulose

RT112/84 152 ± 39 (17) 32 ± 15 (13) 79

(Human Tumour)

T24/83 4.6 ± 0.4 (14) 1.6 ± 0.3 (15) 65

(Human Tumour)

LLC WRC 256 435 ± 83 (15) 366 ± 58 (14) 16 (Rat Tumour)

Mice (nu/nu) were injected daily with either saline (control) or mannoheptulose solution.

Means are presented ± S.E.M.s with numbers of observations as bracketed figures.

Table 7

Structures and Formulae of Inhibitors of Tumour Cell

Growth

(Figures in parentheses following compound names are I values for rat liver glucokinase enzyme activity, the I__ being the concentration required to cause 50% inhibition of enzyme activity)

(1) D-mannoheptulose, C ₇ H ₁₄ 0 ₇ (12.5m

(2) D-Glucoheptose, C H..0 (34mM)

(3) N-acetyl-β-D-glucopyranosyl amine, CgH.-C N (2.5mM)

(4) Diethyl ester substituent of D-glucose, C..H 0 ^{and S formS; O} .vimM and >20mM, respectively)

CH3.H2COCCH.CH3

11 0

Table 7 (continued)

(5) D-mannose, C-H^Og

(6) N-acetyl-D-glucosamine, CgH _^ Og (10mM)

(7) 1 -methyl-N-acetylglucopyranosyl amine (0.87mM)

Table 8

Activities of Glucokinase in Human Melanoma Samples

Sample Activity (nmol min/mg protein)

SW

JH

MM1

MA

MM2

ME

Means are presented ± S.E.M. of 5 assays of each tissue sample. AB samples were excised from patients suffering from malignant melanoma. Samples were frozen in liquid nitrogen after surgical removal and thawed 3 weeks later for assay of enzyme activity.

SUBSTITUTE SHEET (RULE 26;

Table 9

Inhibition of Glucokinase Activity from Human Melanoma Samples by 10 mM

Mannoheptulose

Sample Percentage Inhibition

SW 23 *

JH 37 #

MM1 5 -

MA A III

MM2 23 ^A

ME 34 #

Statistical significance by Student's t test :* 0.1>p>0.05; # 0.01>p>0.001; ^A 0.05>p>0.02; /// p<0.001. Samples obtained as for Table 8.

Table 10

la for Inhibition of Rates of Growth of Cultured Human Tumour Cells with Various Sugar

Analogues

Compound IJO (mM)

Mannoheptulose 21.4 ± 3.2

Glucoheptose 85 ± 10

N-Acetylglucosamine 85 ± 12

Mannoheptose 84 ± 3.3

Means are presented ± S.E.M. of 4 cultured human tumour cell-lines (RVH 421, H.Ep2, TR1-2, HT29). IJO represents the concentration required to elicit 50 % inhibition of growth rate in culture. All compounds inhibit glucokinase activity in the same cell-lines.

-.9

References

(1) Kawai, S. aαd Hanafusa, H. : "The Effects of

Reciprocal Changes in Temperature on the Transformed State of Cells Infected with a Rous Sarcoma Virus Mutant" Virology 4_β pp470 - 479 (1971) .

(2) Birnbaum, M. ^" ., Haspel, H.C. and Rosen, O.M. : "Transformation c. Rat Fibroblasts by FSV Rapidly Increases Glucose Transporter Gene Transcription" Science 2 3 pp1495 - 1498 (1987) .

(3) Flier, J.S., Mueckler, M.M., Usher, P. and Lodish, H.F. : "Elevated Levels of Glucose Transport and Transporter mRNA are Induced by ras or src oncogenes" Science 2 5_ pp1492 - 1495 (1987) .

( ⁴ ) Bramwell, M.E. and Harris, H. : "An abnormal membrane glycoprotein associated with malignancy in a wide range of different tumours" Proc. Royal Soc. B201 pp87 - 106 (1978) .

(5) White, M.K., Bramwell, M.E. and Harris, H. : "Hexose transport in hybrids between malignant and normal cells" Nature 294 pp232 - 235 (1981) .

(6) Weinhouse, S. : "Isoenzyme and Enzyme Regulation in Cancer" , Gann Monographs J_3, University of Tokyo

Press (Ed. Weinhouse and ^' Ono), p1 (1972) .

(7) Van Veelen, C.W.M., Verbiest, H., Vlug, A.M.C., Rijksen, G. and Staal, G.E.J. : "Isozymes of Pyruvate Kinase from Human Brain, Meningiomas and Malignant Glioroas" Cancer Research pp4681 - 4687 (1978) .

(8) Ibsen, K.H., Orlando, R.A., Garratt, K.N., Hernandez, A.M., Giorlando, S. and Nungaray, G. : "Expression of Multimolecular Forms of Pyruvate Kinase in Normal, Benign and Malignant Human Breast Tissue" Cancer Research 4_2 pp888 - 892 (1982).

(9) Presek, P., Glossmann, H., Eigenbrodt, E., Schoner, W., Rubsamen, H. and Friis, R.R. : "Similarities between a phosphoprotein (pp60 )- associated protein kinase of Rous sarcoma virus and a cyclic adenosine 3 ^' 5 ^' monophosphate-independent protein kinase that phosphorylates pyruvate kinase type M " Cancer Research 40. pp1733 - 1741 (1980).

(10) Cooper, J.A., Reiss, N.A. , Schwartz, R.J. and Hunter, T. : "Three glycolytic enzymes are phosphorylated on tyrosine in cells transformed by Rous sarcoma virus" Nature 302 pp218 - 223 (1983).

(11) Shatton, J. : Cancer Research 23. pp1161 - (1969).

(12) Board, M., Humm, S. and Newsholme, E.A. : "Maximum Activities of Key Enzymes of Glycolysis, Glutaminolysis, Pentose Phosphate Pathway and Tricarboxylic Acid Cycle in Normal, Neoplastic and

Suppressed Cells", Biochem. Journal 265 pp503 - 509 (1990) (a) .

(13) Board, M. : "A Study of Energy Metabolism in Neoplastic Cells", D.Phil Thesis, University of Oxford (1990) (b) .

(14) Newsholme, E.A. and Board, M. : "Application of Metabolic Control Logic to Fuel Utilisation and its Significance in Tumour Cells", Advances in Enzyme Regulation 21 pp225 - 246 (1991).

- 3 1 -

(15) Moore, A.E., Sabachowsky, L. and Toolen, H.W. : Cancer Research J_5 pp598-602 (1955).

(16) Peehl, D.H. and Stanbridge, E.J. :

"Characterisation of Human Keratinocyte x HeLa Somatic Cell Hybrids", International Journal of Cancer 27, pp625-635 (1981) .

(17) Klinger, H.P. : Cytogenet. and Cell Genetics 27 pp254 - 266 (1980) .

(18) McCormick, D., Wallace, I., Kirk, J., Dinsmore, S. and Allen, I. : "The establishment and characterisation of a cell-line and mouse xenografts from a human malignant melanoma", British J. Experimental Pathology pp103 - 115 (1983) .

(19) Marshall, C.J. : " Markers of Neoplstic Transformation in Epithelial Cell-Lines Derived from Human Carcinomas", J. National Cancer Institute jL£ _. pp1743 - 1747 (1977) .

(20) Bubenik, J., Baresova, M., Viklicky, V., Jakovbkova, J., Sainerova, H. and Donner, J. :

"Established cell-line of urinary bladder carcinoma (T24) containing tumour-specific antigen", International J. Cancer JJ_ pp765 - 773 (1973) .

(21) Jacobs, J.P., Jones, CM. and Bailie, J.P. . "Characteristics of a Human Diploid Cell Designated MRC5", Nature 221 pp168 - 170 (1970) .

(22) Dunham, L.J. : J. National Cancer Institute 113 pp1299 - 1303 (1953) .

(23) Moraes, O.M. : D.Phil Thesis, University of Oxford (1989) .

(24) Bergmeyer, H-U., Bernt, E., Schmidt, F. and Stork, . -. iinn MMeetthhooddss iinn EEnnzzyymmaattiicc AAnnaallyyssis (2 edition, ed Bergmeyer), 2 pp1196 - 1201 (1974) .

(25) Leighton, B., Budohoski, L., Lozeman, F.J., Challis, R.A.J. and Newsholme, E.A. : "The effect of prostaglandins E., E_ and F„ and indomethacin on the sensitivity of glycolysis and glycogen synthesis to insulin in stripped soleus muscles of the rat", Biochemical Journal 227 pp 337 - 340 (1988) .

(26) Bradford, M.M. : "A rapid and sensitive method for the quantitation of microgram quantities of protein utilising the principle of protein-dye binding", Analytical Biochemistry 12 pp 248 - 254 (1976) .

(27) Stanley, J.C., Dohm, G.L., McManus, B.S. and Newsholme, E.A. : "Activities of Glucokinase and Hexokinase in Mammalian and Avian Livers", Biochemical Journal 22A PP 667 - 671 (1984) .

(28) Andreone, T.L., Printz, R.L., Pilkis, S.J., Magnuson, M.A. and Granner, D.K. : "The amino acid sequence of rat liver glucokinase deduced from cloned cDNA" , Journal of Biological Chemistry, volume 264(1), pp363-369 (1989) .

(29) Magnuson, M.A. and Shelton, K.D. *. "An alternate promoter in the glucokinase gene is active in the pancreatic β cell", Journal of Biological Chemistry, volume 263 (27), pp15936-15942 (1989) .

(30) Hackett et al : Journal of the National Cancer Institute, 22 (6) •

(31) Engel and Young: Cancer Research, 22, p.4327-4339 (1978) .

(32) Moscow et al : Proc. National Academy Sciences 22 (17) p.6518-6522 (1988) .