BACKGROUND OF THE INVENTION

This invention relates to compounds and methods f the treatment and diagnosis of cancer in animals such as humans. More particularly, this invention relates to ident fication and isolation of human amino acid transporters, including transporters such as glutamine transporters that are common to tumour cells but which are generally not foun or are less active, or are present in lower quantities in most non-tumour cells. This invention also relates to reco binant products and methods useful for preparing such trans porters and other biological products such as antibodies, t diagnostic products and assays useful for detecting the pre ence of such transporters in animals, and to compositions o matter such as antibodies to the transporter, anti-glutamin compounds and glutamine analogues which can successfully inhibit the passage of amino acids such as glutamine throug such transporters, or which interfere with the glutamine metabolism in the cells, and various therapeutic compositio comprising such inhibitors. This invention further relates to methods for treating cancer by administering such compos tions, to vaccines comprising such transporters, fragments subunits thereof, and to methods for immunization using suc vaccines.

Current methods for treating cancer generally involve administering to patients one or more drugs from three major classes of compounds, namely: (1) al ylating agents such as chlorambucil, busulphan, melphalan and cyclo phosphamide; (2) antimetabolites such as methotrexate, 6- mercaptopurine, 6-thioguanine, 5-fluorouracil, cytosine arabinoside and acivicin; and (3) antibiotics such as actin mycin D, mithra ycin, mitomycin C, bleomycin and daunoru- bicin. Unfortunately, the use of such compounds can presen problems The major problem is that such compounds are gene ally cytotoxic to both tumour and non-tumour cells and thus can cause serious damage to the patient's healthy tissue an

organs. And with few exceptions, there are no currently- available methods for selectively directing such cytotoxic agents only to tumour cells.

Included among the above chemotherapeutic compou are analogues of glutamine such as β-diazo-5-oxo-L-norleuc (DON), O-diazo-acetyl-L-serine (Azaserine), and α-amino-3- chloro— ,5- dihydro-5-isoxazoleacetic acid (Acivicin), whi interfere with intracellular biosynthesis. However, such compounds have apparently proven too toxic and thus are no currently marketed for the treatment of cancer.

The lack of knowledge concerning the biological basis of malignant cell behavior has hindered the developm of drugs designed to exploit consistent and significant di ferences between normal and malignant cells. One such dif ference is the amount of glutamine required for cell metab lism. Glutamine is a major metabolite of cancer cells, having an importance equal to that of glucose in the suppl of energy, as well as being a critical supplier of acid groups for DNA biosynthesis. However, whereas normal cell contain glutamine synthetase which enables them to synthes the glutamine they require, glutamine synthetase is essen¬ tially inactive (or decreased in amounts) in tumour cells. Instead, glutaminase, which is the hydrolytic enzyme that destroys glutamine, is highly active. Hence, tumour cells depend upon the surrounding blood supply and tissues for their glutamine.

Glutamine is transported into tumour cells by gl tamine transporter proteins. The capacity of such glutami transporters to move glutamine across the plasma membranes tumour cells is thus vital to their growth and developmen In this regard, it has been demonstrated that the glutami transporters in tumour cells are considerably more active than the glutamine transporters of normal cells.

Researchers have tried to exploit the tumour ce need for glutamine by lowering the serum concentration of glutamine in cancer patients. This has been accomplished for example, using glutamine-restricted diets and/or by

administering glutaminase or asparaginase to destroy glutamine in the bloodstream. Deprivation of the glutamine can possibly weaken the cancer cells and thereby render the more susceptible to conventional chemotherapy or radiation therapy. Again, however, this therapy can ultimately resul in the indiscriminate destruction of both tumour and normal cells.

In summary, while it is known that tumour cells require an increased uptake of glutamine from surrounding blood and tissues, no significant progress has been made in utilizing this increased uptake to design compounds or methods for selectively acting on tumour cells. Likewise, this difference has not been exploited in the area of cance diagnosis, where reliable assays for diagnosing the presence of cancer are clearly desireable.

SUMMARY OF THE INVENTION

Using a kinetic analysis developed by the present inventors, three glutamine transporters that are present in tumour cells have been discovered. It has been further dis¬ covered that at least one of the transporters is common to other solid-type tumours and lymphomas. Using the technique developed to identify and isolate the tumor glutamine trans¬ porters, another human amino acid transporter, an alanine transporter, has also been identified.

The identification, isolation and characterization of tumour glutamine transporters, as well as other human and mammalian amino acid transporters, have thus made possible a wide new range of diagnostic, therapeutic and other related compositions of matter and methods which exploit the differ¬ ence in glutamine uptake between tumour and non-tumour cells and which can be used to detect tumours, or selectively act on tumour cells while producing little or no effect on norma cells.

Accordingly, a first embodiment of this invention provides substantially purified human amino acid transporter including glutamine and alanine transporters found in tumour cells

SUBSTITUTE SHEET

Other embodiments of this invention provide metho for screening compounds to determine their ability to bind binding sites on a tumour glutamine transporter, or the ability of compounds to inhibit the transport of glutamine into tumour cells.

Another embodiment of this invention provides laboratory equipment for carrying out the binding and trans port inhibition assays of this invention.

Other embodiments provide diagnostic products, an methods which use such diagnostic products, for detecting tumour cells jLn vivo or in. vitro. Also provided is a metho for monitoring the progress of cancer by repeated, periodic assays in accordance with this invention.

Other embodiments of this invention provide bio¬ logical products which are useful for producing transporter proteins of this invention by recombinant DNA methods, and for producing monoclonal antibodies which specifically bind to the transporters.

Other embodiments provide anti-glutamine compound and glutamine analogues, and methods for using such compoun and analogues in the treatment of cancer.

Other embodiments of this invention provide metho for treating cancer, including methods in which the transpo of glutamine into tumour cells is inhibited, using anti- glutamine compounds, glutamine analogues, or antibody compo sitions. Such methods may be used in combination with othe conventional anti-cancer therapy such as chemotherapy or radiation therapy.

This invention also provides vaccines comprising least one antigenic human amino acid transporter or fragmen or subunit thereof.

Accordingly, it is an object of this invention to provide amino acid transporters, including tumour glutamine transporters, which are useful in the diagnosis and treatme of cancer.

It is another object of this invention to provide screening methods which can identify compounds useful in the * ^■ treatment of cancer.

A further object is to provide diagnostic products ¹ and methods for diagnosing and monitoring the progression of cancer.

A still further object of this invention is to provide biological products for diagnosing and treating cancer, and for preparing quantities of purified transporter or antibodies which bind to such transporters.

Yet another object of this invention is to provide compounds and compositions for use in treating cancer, and methods for treating cancer comprising administering such compounds and compositions.

A further object of this invention is to provide methods for treating cancer by inhibiting transport of glu¬ tamine into tumour cells, or by using the transport mecha¬ nisms of tumour cells to selectively transport cytotoxic compounds into the cells. Methods are included wherein the cell is weakened by the inhibition of glutamine uptake into the cell, and thereafter killed by conventional methods such as chemotherapy, radiation therapy, or any other therapy directed against the cell.

A still further object is to provide vaccines whic can be used for the prevention of cancer.

A further object is to provide compounds and methods which effect the proliferation of lymphocytes. . Additional objects and advantages of the invention will be set forth in part in the description that follows,

•* and in part will be apparent from the description, or may be learned by practice of the invention. The objects and the advantages of this invention may be realized and obtained by means of the compositions of matter and processes partic¬ ularly pointed out in the appended claims.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 illustrates a polyacrylamide gel showing purification of the glutamine transport proteins, lane 1

- 6 -

containing molecular weight standards, lane 2 containing eluted protein, and lanes 3-8 containing whole membrane samples;

FIG. 2 is a graph illustrating glutamine uptake into HeLa cells;

FIG. 3 is a graph illustrating titration of glu¬ tamine uptake into HeLa cells with NEM;

FIG. 4 is a graph illustrating the effect of dif¬ ferent NEM incubation times on glutamine uptake into HeLa cells;

FIG. 5 is a graph illustrating the effect of 1 mM NEM on glutamine uptake into HeLa cells;

FIG. 6 is a graph illustrating the effect of anti serum on glutamine transport into HeLa cells;

FIGS. 7a-7j are graphs illustrating glutamine transport into red blood cells (RBC) and various cell lines

FIG. 8 is a graph illustrating a binding of ³ H-NE to HeLa cell plasma membranes;

FIG. 9 is a graph illustrating the binding of ³ H- NEM to material obtained from peaks A-E of HeLa cell plasma membranes defined in FIG. 8;

FIG. 10 is a graph illustrating titration of glu¬ tamine transport with ³ H-NEM binding to material obtained from peaks A-D defined in FIG. 8;

FIG. 11 is a graph illustrating the theoretical curve for binding of ^S H-NEM to material obtained from peak defined in FIG. 8;

FIG. 12 is a graph illustrating protection of ³ H- NEM binding with excess substrate;

FIG. 13 is a graph illustrating the effect of dif ferent NEM incubation times on glutamine uptake into HeLa cells;

FIG. 14 is a graph illustrating the differential labelling of HeLa cell plasma membranes with ³ H-NEM;

FIG. 15 is an autoradiograph of ³ H-NEM labelled HeLa cell plasma membrane proteins, lanes 1 and 2 show separated proteins labelled with 1 mM ³ H-NEM, and lanes 3

and 4 show separated proteins labelled with 1 mM ³ H-NEM in the presence of 100 mM glutamine; FIG. 16 is a graph illustrating protection of

¹⁴ C-NEM binding with excess substrate;

* FIG. 17 illustrates a polyacrylamide gel of HeLa cells membrane proteins extracted into four fractions, lanes 1 and 2 containing fraction 1, 3 and 4 containing fraction 2 lanes 5 and 6 containing fraction 3, and lanes 7 and 8 con- taining fraction 4;

FIG. 18 illustrates a polyacrylamide gel of HeLa cell membrane proteins stained for carbohydrate by Schiff- Periodate staining;

FIG. 19 illustrates a polyacrylamide gel of HeLa cell membrane proteins under reducing and non-reducing condi tions, lane 1 containing molecular weight standards, lane 2 containing membrane protein under non-reducing conditions, and lanes 3 and 4 containing membrane protein under reducing conditions;

FIG. 20 is a graph illustrating the effect of sodium on ³ H-NEM binding to HeLa cell plasma membranes;

FIG. 21 is a graph illustrating the binding of ³ H- NEM to outer plasma membrane sites (UPM);

FIG. 22 is a graph illustrating the binding of ³ H- NEM to inner plasma membrane sites (LPM);

FIG. 23 is a graph illustrating the effect of seru on ³ H-NEM binding to HeLa cell plasma membranes;

FIG. 24 is a graph illustrating a Lineweaver-Burk

_* plot for binding of representative compounds;

FIG. 25 illustrates silver stained 4% - 30% SDS-

_* polyacrylamide gradient gel showing the position of glutami transport proteins;

FIG. 26 illustrates a representative immunoblot showing the response to anti-serum by various cell lines, lane 1 containing molecular weight standards, lane 2 contai ing HeLa cells, lane 3 containing Detroit 562 cells, lane 4 containing Molt-3 cells, and lane 5 containing Molt-4 cells;

SUBSTITUTE SHEET

FIG. 27 is a graph illustrating protection of ³ H- NEM binding to protein bands with excess alanine, or withou alanine, prior to and during exposure to ³ H-NEM;

FIG. 28 is a graph illustrating inhibition of alanine uptake into HeLa cells by NEM correlated with bindi of ³ H-NEM to peaks A-E;

FIG. 29 is a graph illustrating corrected values for inhibition of alanine uptake into HeLa cells by NEM correlated with binding of ³ H-NEM to peak D;

FIG. 30 is a graph illustrating a comparison of t concentration dependent glutamine uptake curve in HeLa cell and bovine lymphocytes under zero-trans conditions;

FIG. 31 is a graph illustrating the time course f ¹ C-glutamine uptake by normal lymphocytes subsequent to stimulation with concanavalin A;

FIG. 32 is a graph illustrating the initial rate glutamine uptake into bovine lymphocytes in the presence an absence of concanavalin A;

FIGS. 33-38 are graphs illustrating the survival lymphocytes measured by the dye exclusion method and relati mitosis measured by the MTT assay plotted against a range o concentrations of glutamine analogues;

FIG. 39 is an inverse velocity plot of label upta against substrate concentration for one single degree syste

FIG. 40a is an inverse velocity plot of label uptake using two single degree systems;

FIG. 40b is an inverse velocity of label uptake plot of three single degree systems;

FIG. 41 is an inverse velocity of label uptake pl of three single degree systems;

FIG. 42 illustrates the inverse velocity of label uptake of a multidegree system followed by a first degree system;

FIG. 43 is a plot of a single degree system fol¬ lowed by a multi-degree system;

FIG. 44 is a plot of two multi-degree systems;

FIG. 45 is a plot of inverse velocity of uptake of three systems, two of which are multi-degree;

FIG. 46 is a plot of uptake velocity of glutamine

a plot of 1/v * against concentration; a final curve calculated from the final

a plot illustrating glutamine uptake in HeLa cells under conditions of Na -deficiency;

FIG. 50 is a plot of inverse velocity of label uptake against concentration;

FIGS. 51-59 comprise Reaction Schemes 1-9, respec¬ tively, which illustrate syntheses useful for making anti- glutamine compounds and glutamine analogues in accordance with this invention;

FIG. 60 is a graph illustrating the growth of HeLa cells in 1 mM histidine, the rate of growth quantitated by measuring total cellular protein present in multiple samples (n = 6) every 24 hours;

FIG. 61 is a graph illustrating alanine uptake into HeLa cells;

FIG. 62 is a graph illustrating a Lineweaver-Burk analysis of alanine uptake into HeLa cells, the HeLa cells having been pre-incubated with 1 mM NEM, or without NEM;

FIG. 63 is a graph illustrating titration of alanine uptake into HeLa cells with varying NEM concentrations;

FIG. 64 is a graph illustrating the effect of 1 mM NEM on alanine uptake into HeLa cells which were pre- incubated with, or without NEM prior to radioactive incubation;

FIG. 65 is a flow diagram illustrating methods for preparing upper plasma membranes (UPM) and lower plasma membranes (LPM) in accordance with this invention;

FIG. 66 is a schematic illustration of a six-well assay plate which is placed upside down on an inverted assay plate cover in acqegdanoa wit this invention;

FIG. 67 is a schematic view of the assembly of FI 66 inverted so that the contents of the lid are transferred to wells in the plate;

FIG. 68 illustrates an inverted 6 well cover plat in accordance with this invention;

FIG. 69 illustrates the manner in which a 6 well bottom plate is inverted and placed over an inverted cover plate which contains solution;

FIG. 70 illustrates a 6 well plate and cover plat assembled for transfer of solution;

FIG. 71 illustrates a 96 well cover plate in acco dance with this invention;

FIG. 72 illustrates the 96 well cover plate showi channels into which solution is placed;

FIG. 73 illustrates the manner in which a 96 well plate is inverted and placed over an inverted cover plate containing solution;

FIG. 74 illustrates a 96 well plate and cover assembled for transfer of solution; and

FIGS. 75-86 illustrate Tables 4-15, respectively, of this invention.

DESCRIPTION OF THE PREFERRED EMBODIMENTS I. DEFINITIONS AND ABBREVIATIONS

Provided below are definitions and abbreviations for some basic terms used throughout this disclosure.

Anti-glutamine compound - refers to any compound that enters the cell via the glutamine transporters or by a other route, and/or acts on glutamine uptake, and/or metabo lism, biosynthesis, intracellular degradation (e.g., enzy¬ matic), etc., and/or prevents the binding of glutamine to t transporter of any other substance which binds glutamine.

Anti-receptor - refers to agents which bind selec tively to the receptor of a ligand-receptor pair. For example, if the ligand is an antigen and the receptor is a antibody, for example, a mouse IgG antibody, the anti- receptor may be an antibody against IgG. The anti-recepto may also be a substance which binds selectively with a moi

- 11 -

conjugated to the receptor. For example, the moiety may be hapten and the anti-receptor an antibody against the hapten. Fragments and derived products are also encompassed.

Glutamine analogue - refers to any compound which bears a structural similarity to glutamine.

Human Amino Acid Transporter - refers to a molecu located within the membrane of a cell derived from human tissue which transports amino acids into and out of the cel

Ligand - refers to a target analyte being assayed This term includes any member of an immunological binding pair such as an antibody or an antigen. Alternatively, a ligand can be, for example, an amino acid or its transporte an enzyme or a nucleic acid sequence such as RNA or DNA probe.

Receptor - refers to a molecule which binds directly with the target ligand. For example, if the ligan is an antigen, the receptor may be an antibody, preferably monoclonal antibody. If the target ligand is an antibody, the receptor may be an antigen or anti-antibody. If the ligand is an enzyme, the receptor may be substrate for the enzyme. If the ligand is a nucleic acid sequence, the receptor may be a complementary sequence. Fragments and derived products are also encompassed.

Tumour Glutamine Transporter (TGT) - refers to a molecule located within the membrane of a tumour cell which is capable of transporting the amino acid glutamine across the membrane, from external medium into the cell's cytoplas and vice versa.

Ac - Acetyl

ACS - Aqueous counting scintillant

ATV - Antibiotic trypsin versene

Bm - Maximum binding constant

Bn - Benzyl

BSA - Bovine serum albumin

Bu - Butyl

Ci - Curie

DCC - Dicyclohexylcarbodiimide

DCHA - Dicyclohexylamine

DCM - Dichloromethane

DMF - Dimethylformamide

DMSO - Dimethylsulphoxide dp - Distintegrations per minute

EDTA - Ethylenediamine tetraacetic acid

ELISA - Enzyme linked immunosorbent assay

Et - Ethyl

FBS - Feotal bovine serum h - Hill coefficient

HBSS - Hank's Balanced Salt Solution

HEPES - N-2-hydroxyethylpiperazine-N'-2- ethanesulphonic acid IgG - Immunoglobulin G Km - Michaelis Menten constant Ks' - Apparent dissociation constant LPM - Lower plasma membrane M199 - Media 199 Me - Methyl

M - Relative molecular mass NEM - N-ethylmaleimide PBS A - Phosphate balanced saline without calci or magnesium. PEG - Polyethylene glycol PMSF - Phenylmethylsuphonyl flouride rpm - Revolutions per minute RPMI 1640 - Rosewell Park Memorial Institute cell culture medium 1640. SDS - Sodium dodecyl sulphate

SDS-PAGE - SDS polyacrylamide gel electrophoresis t - Tertiary

TBST - Tris buffered saline with Tween TEN - Tris-EDTA-NaCl TFA - Trifluoroacetic acid THF - Tetrahydrofuran

Tosyl - Para-toluenesulphonyl

Tris - Tris-(hydroxymethyl)aminoethane

UPM - Upper plasma membrane v/v - Volume by volume

V - Maximum velocity w/v - Weight by volume

Z - Carbobenzyloxy

II. IDENTIFICATION OF AN AMINO ACID TRANSPORTER PROTEI The first embodiment of this invention comprises human amino acid transporter proteins. Of special interest are the transporters which are present in tumour cells, but which are not present, or are not active, or are present in lower quantities in non-tumour cells. Three such trans¬ porters have been discovered, namely TGT I, TGT II and TGT

III, which are present in tumour cells but do not appear to be present or active in non-tumour cells. Alternatively, TGT I - III may be present in normal cells, but only in low quantities. The method developed for isolating TGT I, TGT II and TGT III is illustrative of the method by which other transporters, especially those which are specific to tumour cells, can be identified and isolated. This method is based on a set of procedures (Hayes et al., Biochem. J. , 214, (1983) 489 - 495) used to isolate an alanine carrier from hepatocytes, however, there are some important differences. The tumour glutamine transporters of this invention were isolated from HeLa cells, which are derived from an estab¬ lished human tumour cell line. The cell line originated from a biopsy of a cervical carcinoma from Henrietta Lacks in 1951.

A. Procedures for Separating HeLa Cell Proteins Procedures used for initially preparing a monolayer culture for HeLa cells and other adherent cell lines, and for preparing microcarrier culture cells were as follows.

1. Monolayer culture for HeLa cells and other adherent cell lines Cells were grown as onolayers on the surface of plastic flasks or glass bottles (Flow Laboratories) and when

confluence was attained the cells were harvested using antibiotic trypsin versene (ATV) . The media in actively growing cultures was changed every two days for fresh media (M199 or RPMI 1640) containing 5% (v/v) foetal bovine serum (FBS) .

2. Microcarrier Culture

Adhesive cells were grown on the surface of glass beads (Sigma Chemical Company) by an adaptation of the methods described by Pharmacia (1981). Beads (0.5 g) were rinsed three times with fresh medium containing 10% (v/v) F and then mixed with approximately 10 ^β cells which were harvested from monolayer cultures with ATV and diluted to 2 ml with fresh medium containing 10% (v/v) FBS. The inoculu was then transferred to a microcarrier vessel (500 ml capacity, Techne, Cambridge, U.K.) and placed on a magnetic stirrer at 37°C which stirred the culture at 25 rpm for two minutes every half hour for the first 24 hours and then stirred continuously at 35 rpm for the remainder of the growth period. The medium was changed every two days by allowing the beads with attached cells to settle and then decanting off the old medium and replacing it with fresh medium containing 5% (v/v) FBS. The beads became confluent after 5 days and were then used for plasma membrane preparation.

Identification and purification of the glutamine transporters of HeLa cells was then carried out as follows.

3. Plasma membrane preparation from HeLa cells

Cells were grown as microcarrier cultures until t growth on the beads was fully confluent. The plasma mem¬ branes were then prepared in the following way:

(i) The beads with their attached cells were allowed to settle and the supernatant was discarded. The beads were washed three times with fresh medium and then resuspended in 20 ml of hypotonic buffer (10 mM tris-HCl, 7.4) for 10 minutes at 4°C to cause swelling and lysis of cells.

(ii) The cell suspension was vortexed vigorously for 8 seconds followed by sonication with the microtip of an MSE ultrasonic disintegrator at maximum setting for 8 seconds.

(iii) The cell suspension was washed three times with hypotonic buffer and then resuspended in 10 ml of hypotonic buffer containing 1 mM EDTA and sonicated for 30 seconds.

(iv) The released membranes were collected by dif¬ ferential sedimentation and pelleted by centrifuging in a Damon/IEC model B-20A centrifuge for 15 minutes at 10000 g.

(v) The pellet was resuspended in 1 ml of hypotonic buffer containing 0.25 M sucrose, pH 7.4 and store at -20°C. Enzyme assays and binding experiments were carrie out on freshly prepared membrane fractions. See Gotlib et al. , Biochem. et Biophys. Acta, 602, (1980) 201-212.

4. Sodium dodecyl sulphate polyacrylamide σel electrophoresis a. Selective extraction of membrane proteins Membrane proteins were selectively extracted into four groups by a procedure which is based on the unique properties of the detergent Triton X-114 (Thomspon et al.. , Biochemistry. 26, (1987) 743-750). Monolayer cultures were rinsed three times with 10 ml of phosphate buffered saline (PBS A) and the first group of membrane proteins was extracted from the surface of intact HeLa cells by incubatin with 2 ml of 1 mM EDTA, 0.15 mM NaCl and 10 mM 4-(2- hydroxyethyl)-l-piperazine-ethanesulphonic acid (HEPES) pH 7.4 (fraction 1). The cells were scraped from the surface o the flask with a rubber policeman and collected by centrif¬ ugation at 500 g for 5 minutes in a conical polypropylene tube. The cell pellet was treated with 100 μl of 2% (v/v) Triton X-114 in 10 mM HEPES, pH 7.4 and insoluble material was removed by centrifugation at 11600 g for 10 minutes. Th upper detergent-free phase contained hydrophobic proteins (fraction 2) and the lower phase contained hydrophillic

- 16 -

proteins (fraction 3). The insoluble pe.let was solubilized in SDS solubilization buffer (fraction 4). The proteins of fractions 1-3 were precipitated by adding acetone to a final concentration of 90% (v/v) and pelleted at 11600 g in an Eppendorf ultracentrifuge (model 5414) for 10 minutes. The acetone precipitates were dried under vacuum before solubi- lizing in 100 μZ of SDS solubilization buffer and separat¬ ing by SDS-PAGE. b. Separation of proteins

Membrane preparations (approximately 1 mg protein) were solubilized in 1 ml of SDS solubilization buffer (25 mM tris-HCl, 2% (w/v) SDS, 2% (w/v) ficoll, 1% (v/v) 2-mercapto ethanol, 2 mM EDTA, 5 mM PMSF, pH 6.8) by vortexing vigor¬ ously three times for 1 minute and then placing at -20°C overnight. Proteins were separated on 12% polyacrylamide gels using the Bio-Rad Mini-Protean II apparatus (see Laemmli, Nature, 227, (1970) 680-685). Solubilized membrane (1 mg protein/ml) were mixed with sample buffer (62.5 mM tris-HCl pH 6.8, 2% (w/v) SDS, 2% (w/v) ficoll, 5% (v/v) mercaptoethanol, 0.00125% bromophenol blue, 25 mM EDTA) at a ratio of 1:2 and heated at 70°C for 10 minutes. Samples of 30 μl were applied to the wells of 0.75 mm gels and a constant current of 50 mAmps applied for 1.5 hours. The gel were stained in either of the following ways. (See Ohsawa e al.. Anal. Biochem.. 135. (1983) 409-415).

(i) Coomassie blue staining

The gels were removed from the electrophoresis apparatus and placed in a solution of 70% (w/v) polyethylene glycol (PEG) 2000 overnight. The gel which had become opaqu was then stained in a solution of 1% (w/v) coomassie blue fo 15 minutes followed by destaining in several changes of 10% acetic acid.

(ii) Silver staining

The gels were shrunk in 70% (w/v) polyethylene glycol 2000 overnight and then transferred to a silver nitrate solution (20% (w/v)AgN0 ₃ , 35% (v/v) ammonia, 4% (w/ v) NaOH, which was used within five minutes) for 20 minutes

with gentle agitation. The gel was rinsed three times in distilled water for 1 minute and then transferred to devel- oper solution (0.002% (v/v) formaldehyde, 0.005% (w/v) citr acid) until bands became visible. Development could be ϊ stopped with 10% (v/v) acetic acid and the gel could be stored in the PEG solution. c. Radioactivity counting

The stained gels were placed on a clear plastic cutting surface which was placed on a transilluminator (visible wavelength) so the bands could be easily seen. Stained protein bands were sliced out of the gel using a scalpel and placed into 4 ml capacity scintillation vials. The gel slices were solubilized in 100 μl of H ₃ 0 ₃ and 100 μl of 1.8 mM CuS0 ₄ overnight at room temperature: then 4 ml of aqueous counting scintillant (ACS) was added and the vials were counted for radioactivity. (See Donato et al.. , Biochem. Biophvs. Methods. 15. (1988) 331-336.) d. Autoradiography

The radiolabelled polyacrylamide gels containing resolved membrane proteins were incubated in Amplify (Amersham) for 30 minutes at room temperature and then dried in a Pharmacia gel dryer (model GSD-4) for 2 hours at 60°C. A piece of plastic was placed between the gel and the lid of the drier to avoid sticking. The dried gel was then placed in contact with Fuji X-ray film for 6 days at -65°C. At the end of the exposure period the gel was allowed to come to room temperature and developed by placing in a solution of i Kodak Developer for two minutes with gentle agitation every

15 seconds followed by washing in tap water for 2 minutes an then placing in Kodak Fixer solution until the film became translucent. e. Reducing and non-reducing electrophoresis

Plasma membrane preparations were solubilized in buffer containing 5% (v/v) mercaptoethanol or in buffer lack ing mercaptoethanol. (Allore et al.. , Mol. Immunol. , 20, (1983) 383-395.) The solubilized membranes were mixed with

sample buffer with or without mercaptoethanol and the proteins were then separated by SDS-PAGE and stained in the manner described above. f. Glycoprotein staining

Membrane proteins were separated by SDS-PAGE as described above and then stained for carbohydrate. (Zacharius et al., Anal. Biochem.. 30, (1969) 148-152.) The gel was fixed in 7.5% (v/v) acetic acid for 1 hour at 4°C and then placed in a solution of 0.2% (w/v) periodic acid for 45 minutes at room temperature. The gel was then placed immediately in Schiff's reagent (0.5 g Fuchsin, 9.0 g Na ₂ S0 ₃ in 500 ml H ₂ 0: then 10 ml HCl was added and the solution was protected from light) for 45 minutes at 4°C. The gel was destained overnight in 2-3 changes of 10% (v/v) acetic acid.

5. Electroelution

Electroelutions were carried out in the Bio-Rad Electroeluter model 422. Polyacrylamide gels of membrane proteins were lightly stained (5 minutes) in 0.1% (w/v) coomassie blue and destained in several washes of 10% (v/v) acetic acid. The stained band corresponding to the glutamine transport protein was sliced out of the gel using a scalpel and placed into the glass tube connected to a dialysis mem¬ brane. The tank was filled with volatile elution buffer (0.05 M NH ₄ HC0 ₃ 1% (w/v) SDS) and a constant current of 6 r mAmps/tube was applied for 24 hours. The eluted protein was collected from the surface of the dialysis membrane and stored in elution buffer containing 50 μl of 0.1 M PMSF at -20°C. The eluted protein samples were concentrated using Centricon microconcentrators with a molecular weight cutoff of 10000. The microconcentrators were centrifuged in a 45° fixed angle rotor (IEC/Damon model B-20A centrifuge) at 5000 g for 30 minutes to give a final volume of 50 - 100 μl . A silver stained polyacrylamide gel showing purificatin of the eluted protein band with M 42000 is shown in FIG. 1.

The microconcentrators could also be used to dialyse the protein splufcion against several washes of PBS A

in order to remove salts and SDS. Purity of the eluted protein was detected by equilibrating 50 μl of the concen¬ trated protein with 200 μl of SDS sample buffer followed by SDS-PAGE and staining with silver. The amount of protein present in each sample was determined by an adaptation of t MicroLowry method using 200 μl of protein solution and reading the absorbance at 660 nm. (Markwell et al_. , Anal. Biochem. , 87, (19,78) 206.)

6. ³ H-NEM labelling experiments a. Labelling with different NEM concentrations

NEM solutions ranging from 0.25 to 10 mM in 10 ml HBSS were prepared, with each containing 0.1 ml ³ H-NEM stoc solution. Plasma membrane preparations (0.2 mg protein/ml) were incubated with 200 μl of the NEM solutions for 1 hour at 37°C: then labelling was stopped by washing in ice-cold 0.9% NaCl. Membranes were pelleted at 11600 g, and then solubilized in 200 μl of solubilization buffer as described above and stored at -20°C. b. Protection of binding with excess substrate

Plasma membrane preparations (0.2 mg protein/ml) were pre-incubated with the D or L isomer of 100 mM glutami in HBSS for 20 minutes at 37°C before labelling with 1 mM ³ H-NEM containing 100 mM glutamine for a further .20 minutes. Labelling was stopped by washing in ice-cold 0.9% NaCl and these membranes were then solubilized in solubilization buffer. c. Differential labelling of glutamine binding sites

Plasma membrane preparations (0.2 mg protein/ml) were incubated with 100 μl of various glutamine concentra¬ tions ranging from 0.5 to 10 mM for 5 minutes at 37°C. NEM (100μi) at 2 mM made up in the test glutamine concentration was added to give a concentration of 1 mM NEM and the mem¬ brane suspension was incubated for a further 10 minutes at 37°C. At the end of this incubation the membranes were

centrifuged for 1 minute at 11600 g rpm and the pellet was washed with Hanks balanced salt solution (HBSS). The plasm membranes were then incubated with 1 mM ³ H-NEM for 1 hour a 37°C. The labelling was stopped by washing with ice-cold 0.9% NaCl. Membranes were solubilized in solubilization buffer and stored at -20°C. d. Labelling upper and lower plasma membranes

The flow diagram in FIG. 65 describes the methods for preparing upper plasma membrane (UPM) and lower plasma membrane (LPM) fractions and labelling with ³ H-NEM. This procedure can be applied to any adherent cell line in addi¬ tion to HeLa, and can be used for any type of assay where t inner and outer faces of the same cell membrane are require A radioactive solution of 1 mM NEM was prepared by adding 1 μl of ³ H-NEM stock solution (0.5 mCi/ml) to 10 ml of 1 mM NEM in HBSS, pH 7.4. e. Na -dependence

Aliquots of HeLa cell plasma membranes containing approximately 100 μg/ml of protein were used for experi¬ ments. All solutions were prepared in sodium-free HBSS and radioactive NEM was prepared by adding 100 μl of ³ H-NEM stock solution (0.5 mCi/ml) to 1 ml of 1 mM NEM in HBSS. Membranes were pre-incubated with 10 mM glutamine for 5 minutes in the presence of either 100 mM NaCl or 100 mM choline chloride at 37°C. A solution of NEM (containing glutamine and sodium or choline at the appropriate concen¬ trations) was added to the incubating membranes to a final NEM concentration of 1 mM for a further 10 minutes at 37°C. The membranes were then pelleted at 11600 g in an Eppendorf ultracentrifuge for 5, minutes then resuspended in 1 mM ³ H- NEM (containing 100 mM sodium or choline) for 1 hour at 37° At the end of the labelling period the membranes were washe with HBSS and then solubilized in 100 μl of SDS solubiliza¬ tion buffer. The solubilized membrane samples were applied to a 12% polyacrylamide gel and subjected to 50 mAmps of constant current for 1 hour. The separated proteins were

stained with coomassie blue then sliced out of the gel and counted for radioactivity. f. The effect of serum on glutamine transporters

Prior to the isolation of membrane fractions, the HeLa cells were treated in a variety of ways in order to induce migration of the glutamine transporters into the mem¬ brane. HeLa cells were incubated in fresh Medium 199 con¬ taining 20% (v/v) FBS or lacking FBS 2 hours prior to membrane preparation. Membranes were isolated in the usual way and then labelled with ³ H-NEM. All solutions were pre¬ pared in HBSS and radioactive NEM was prepared by adding 100 μl of a stock solution of ³ H-NEM (0.5 mCi/ml) to 1 ml of 1 mM NEM in HBSS. The membranes were preincubated in 10 mM glutamine for 5 minutes at 37°C and then NEM was added to a final concentration of 1 mM for a further 10 minutes. The membranes were pelleted at 11600 g for five minutes in an Eppendorf ultracentrifuge and then resuspended in ³ H-NEM for 1 hour at 37°C. The membranes were washed in 0.9% NaCl and solubilized in 100 μl of SDS solubilization buffer for SDS- PAGE analysis. The separated proteins were sliced out of the gel and counted for radioactivity.

B. Identification and Properties of the Isolated Glutamine Transporters 1. Identification of carrier proteins a. ³ H-NEM labelling experiments

HeLa cell plasma membranes were incubated with ³ H- NEM, dissolved in SDS detergent and the proteins were sepa¬ rated by polyacrylamide gel electrophoresis. A number of proteins were found to bind ³ H-NEM as shown in FIG. 8 and they were labelled A-E. This pattern of binding was consis¬ tently observed in several membrane preparations. Table 1 below shows the approximate molecular weights of the peaks A- E and the binding of ³ H-NEM to each peak which was calcu¬ lated knowing that the specific activity for the ³ H-NEM solution was 2.22 x 10 ^δ dpm/nmole. Molecular weights were

PEAK

A B C D E

In order to establish that a particular protein was involved in the NEM-sensitive transport of glutamine it was necessary to demonstrate that the titration of NEM binding t this protein was parallel with the inhibition of transport b NEM. The NEM binding curves for each of the peaks A-E are shown in FIG. 9. Saturation of peak C occurred at approxi¬ mately 2 mM NEM and the binding of ³ H-NEM was shown to be half maximal at 1 mM NEM, which agrees with the fact that glutamine transport is inhibited by 50% at this level of NEM (see FIG. 5). The amount of ³ H-NEM bound to peak B was slightly higher than that for peak C at 1 mM while peaks A, D, and E had a much greater amount of ³ H-NEM bound. A plot of NEM binding to peaks A-E at various NEM concentrations against glutamine transport at the same NEM concentrations i shown in FIG. 10. A linear plot with one to one correlation was obtained for peak C indicating that the inhibition of transport was parallel with the binding of the inhibitor to this protein. Binding of ³ H-NEM to peaks B-E does not show one to one correlation with inhibition of glutamine transpor by NEM. A similar plot for peak C alone is given in FIG. 11 where the rate of glutamine uptake at 100% saturation of ³ H- NEM binding was assumed to be equal to the non-saturable com ponent of total glutamine uptake and was deducted from the uptake rates measured at each NEM concentration. The line drawn is the theoretical line for agreement between percent saturation of the carrier protein with ³ H-NEM and percent control for glutamine uptake. There is good agreement between the observed values and the theoretical line which supports* the implication of peak C in glutamine transport

into HeLa cells. Further evidence for the involvement of peak C in the NEM-sensitive transport of glutamine was pro¬ vided by showing that the transport substrate could protect the transport protein from binding NEM. Membranes were pre- incubated with a large excess of L-glutamine or D-glutamine before incubating with unlabelled NEM. The excess glutamine and NEM were removed and the proteins separated by gel elec¬ trophoresis. FIG. 12 shows that there was clear protection of peak C by L-glutamine whereas D-glutamine provided no pro tection from binding of ³ H-NEM. Peak B also showed substan¬ tial protection from binding ³ H-NEM indicating that this protein may also be involved in glutamine transport. b. Differentiation between high affinity and low affinity carriers A new approach was employed in order to differen¬ tiate between the roles that peaks B and C play in the transport of glutamine into HeLa cells, in particular to identify the high affinity and low affinity carriers. The effect of different pre-incubation times with NEM on glu¬ tamine transport is shown in FIG. 13. It can be seen that a pre-incubation time of six minutes gives some inhibition of low affinity transport, but has no effect on high affinity transport. However, a pre-incubation time of twelve minutes appeared to give inhibition of both systems. The new exper¬ iment was based on these observations and involved using unlabelled NEM for six minutes, which would bind to all sul- phydryl groups in the plasma membrane. However, increasing concentrations of glutamine were also added to protect the glutamine binding sites. The glutamine and excess NEM were washed out, exposing the glutamine binding sites. ³ H-NEM was then added for twelve minutes to specifically label the glutamine binding sites. It was assumed that as the concen¬ tration of glutamine used to protect the binding sites was increased, the amount of ³ H-NEM subsequently attaching to the glutamine binding site would increase proportionately. High affinity binding sites would be protected by 1 mM glutamine since this is the level at which high affinity IT TE SHEET

transport operates. Low affinity binding sites would require a higher level of protecting glutamine before being labelled with ³ H-NEM since low affinity glutamine transport operates at levels above 1 mM. The results of this type of differen¬ tial labelling experiment are shown in FIG. 14 and peak C with M 42000 behaves exactly as would be expected of the high affinity transport protein becoming labelled with ³ H- NEM at 1 mM glutamine and above. Peak B, on the other hand, does not become labelled with ³ H-NEM until 5 mM glutamine and above, suggesting that this protein with M 53000 is involved with the low affinity transport of glutamine. The position of peaks B and C in a polyacrylamide gel is shown i FIG. 1. c. Autoradiography HeLa cell plasma membranes were incubated with ¹⁴ C-NEM in the presence or absence of excess transport sub¬ strate. The membrane proteins were resolved by SDS-gel elec trophoresis and the gel was then exposed to X-ray film. The autoradiograph in FIG. 15 shows that the band at M 42000 is radioactively labelled with NEM in the absence of the pro¬ tecting substrate glutamine (lanes 1 and 2) but when glu¬ tamine is present in the incubation medium the band is pro¬ tected from becoming radioactively labelled (lanes 4 and 5). These results indicate that the mode of action of NEM is by direct competition with glutamine for a site on the transpor protein and allows the specific protein responsible for glutamine transport to be identified. A duplicate gel of membrane proteins which had been labelled with ¹⁴ C-NEM was also sliced and counted for radioactivity. The results obtained are shown in FIG. 16 and support the results shown in FIG. 12 for the tritiated membrane proteins. A number of peaks were labelled with ¹⁴ C-NEM, but only peak C with M 42000 was significantly protected from becoming labelled wit ¹⁴ C-NEM in the presence of excess glutamine. The strength of the radioactive signal was much enhanced compared to ³ H- NEM labelling because the energy of ^X C is much higher which in turn gives much higher specific activity values.

2. Fractionation of membrane proteins

Membrane proteins were selectively extracted into four fractions depending on their hydrophobic nature and location within the cell membrane. The stained gel in FIG. 17 demonstrates that there was little overlap of proteins between the groups with fraction 1 containing the surface or extrinsic membrane proteins. Fraction 2 contained the hydro phillic membrane proteins and fraction 3 contained the hydro phobic membrane proteins. The final fraction 4 contained integral membrane proteins which could only be solubilized i the presence of the ionic detergent SDS. The stained gel shows that the high affinity glutamine transport protein wit M 42000 appeared mainly in the fourth fraction with slight overlap into the hydrophobic protein fraction. The low affinity transport protein with M 53000 appears predom¬ inantly in the third fraction indicating that it is a hydrophobic membrane protein.

3. Characterization of glutamine transporter by SDS-PAGE

HeLa cell membrane proteins which had been separated by electrophoresis were stained for carbohydrate using the Schiff method. The gel in FIG. 18 indicates that number of proteins were glycosylated including the glutamine transport protein with M 42000. Membrane proteins were separated by SDS-PAGE under reducing and non-reducing condi¬ tions in order to detect the presence of intra- and inter- molecular disulphide bonds. The gel shown in FIG. 19 indicates that the glutamine transport protein with M 42000 does not appear in the non-reducing sample (lane 2) but is present in the reducing sample (lane 4) suggesting that this protein represents a subunit of a protein complex. The intermediate lane (lane 3) was affected by the diffusion of reducing agent on the side closest to the reducing sample causing the partial visualization of the band.

4. Na -dependence of NEM binding The binding of ³ H-NEM to HeLa cell plasma membranes was investigated in the presence or absence of

SUBSTITUTE SHEET

- 26 -

sodium ions in order to establish if binding of NEM to the glutamine transport protein is sodium dependent. The graph in FIG. 20 shows that the specific binding of NEM to the hi affinity glutamine transport protein with M_ 42000 followin protection of the glutamine binding site with excess sub¬ strate was decreased by 30% in the absence of sodium ions. The inhibition of NEM binding in the absence of sodium ions suggests that binding is dependent on the presence of sodiu ions at or near the active site on the glutamine transport protein. Binding of NEM to the low affinity glutamine tran port protein with M 53000 was not affected by the presence or absence of sodium ions. These results agree with the kinetic findings which show that the Km of the Na -dependen transporter II is increased from 0.17 mM to 1.5 mM in the absence of Na whereas transporter III is not affected.

5. Labelling inner and outer plasma membrane faces A method was developed for the selective labellin of glutamine transport sites on the inner and outer faces o the plasma membrane (described above in Section H.A.6.). The binding of ³ H-NEM to the outer face of the plasma mem¬ brane is shown in Fig 21 and indicates that the peaks with 42000 and 53000 are specifically labelled with the isotope following protection of the glutamine binding site with excess glutamine. As expected, binding to these peaks was proportional to the concentration of glutamine used for pro tection of the binding site.

Inner plasma membranes were also specifically labelled with ³ H-NEM and the binding pattern shown in FIG. 22 is markedly different from that obtained for outer plasm membranes. Although the magnitude of the binding to peaks with M 42000 and 53000 was still comparable to that for th outer plasma membranes, there was no apparent correlation between the amount of binding and the concentration of glu tamine used for protection. The most significant differen was the appearance of another band (designated peak F) whi had not been shown to bind ³ H-NEM in previous experiments

but apparently showed a good positive correlation between binding and glutamine concentration when the inner face of the membrane was exclusively labelled. Also-, peak E, which had previously given an inverse correlation with glutamine concentration, now showed a direct correlation between bind¬ ing of ³ H-NEM and glutamine concentration. The reason for the significant difference in binding patterns between the inner and outer faces of the plasma membrane is unclear but may be related to the regulation of glutamine transport by sites containing reactive sulphydryl groups.

6. Translocation of glutamine transporters FBS contains growth factors and hormones which potentiate the active proliferation of cells in culture. HeLa cells were incubated in the presence or absence of FBS prior to the preparation of plasma membranes which were sub¬ sequently labelled with ³ H-NEM after protection of the glutamine active site of the transport protein with excess substrate. The graph in FIG. 23 shows the binding activity of HeLa cell membrane proteins prepared from normal cells and serum-depleted cells. Binding of NEM to the high affinity glutamine transporter was decreased by 50% in the absence of serum. A number of other proteins which normally show sig¬ nificant binding of NEM, namely peaks A, B, D, and E, also exhibited a decreased binding of NEM in the absence of serum. Re-addition of serum to the medium following incubation with serum-free medium resulted in a small 10% increase in the binding of ³ H-NEM to the glutamine transport protein.

7. Kinetic analysis a. Derivation of kinetic equations At an early stage, it was recognized that the meta¬ bolic importance of glutamine uptake into tumour cells made necessary the investigation of the nature of this transport in quantitative terms. The kinetics of glutamine transport in HeLa cells was therefore investigated in detail. In order to obtain parameters which were as closely related as pos¬ sible to in vivo conditions, all the transport experiments

were carried out under conditions of optimum trans- stimulation, and by this means it was possible to obtain dat under in vitro conditions which should apply over the range of glutamine concentration normally found in blood. Further more the complexity of the glutamine uptake curve found in HeLa cells, made it necessary to consider or develop new analytic procedures to clarify as far as possible the major components of the transport mechanism. The use of unmodifie double reciprocal plots (Lineweaver & Burk, J. Am.Chem. Soc. 56, (1934) 658-666) in investigations of membrane transport kinetics is still a common practice even though their lack o reliability when applied to unweighted data from enzyme experiments has been frequently discussed (Dowd & Riggs, J. Biol. Chem. 240, (1965) 863-869; Wong, Kineticss of Enzyme Mechanisms, (1975) p. 29, Academic Press, London; Roberts, Enzyme Kinetics, (1977) p. 287, Cambridge University Press). Furthermore, these plots are seen to be even more unreliable when the systems under investigation do not appear to confor to simple first degree- hyperbolae as represented by the Michaelis-Menten equation (Blangy, et al . , J. Mol Biol. 31, (1968) 13-35). There is also an additional complication in cellular transport, which is that the presence of a single uncomplicated transporter is a rarity. Thus, in the presenc of multiple transporters, the Lineweaver-Burk plot becomes increasingly difficult to handle as the concentration range increases, due to the fact that the I/S variate (where S = substrate concentration) reverses the relative distances between plot points. This results in constant increments in S at the top end of the range producing diminishing changes in the value of 1/S, making it very difficult to have a single plot with adequate separation between significant points at the top end of the concentration range, and thereb placing emphasis on the plot points at the top end of the 1/ range (Roberts, 1977), which are also the points with the smallest velocity values, and, therefore likely to be the least accurate of the entire plot. The main effect of this is seen in the double reciprocal plot for glutamine uptake i

HeLa cells where the main inflexion between 1.5 and 5mM is compressed into 1/400 of the 1/S range. In the case of glutamine transport in HeLa cells it was found necessary to plot the velocity against a glutamine concentration range of 0.005-5mM in order to include all the components of the transport mechanism. When these results were transformed to a Lineweaver-Burk plot it became impossible to show a clear separation of significant plot points on a single graph. Furthermore, the size of the error span at 0.01 mM was 32 times that at ImM, thereby making any attempt at a meaningful plot a matter of guesswork. In order to overcome these dif¬ ficulties a new plotting procedure was developed specifically for experiments involving label transfer kinetics. The fol¬ lowing equation for unidirectional flux of label under conditions of constant concentration of label has been derived from the Michaelis-Menten equation: v* = ■&*/(•* ^■ __. + S ₁ ) (1) where v* = v.S* ₁ /S ₁

= uptake of label from face 1, S.* = the constant label concentration at face 1

= the 'label velocity-substrate concentration coefficient' . On inverting equation 1 we get:

1/v* •= (K _m + S _χ ) /L* (2)

Thus the plot of 1/v* against S gives a gradient 1/L* , and since the concentration of label is known, V^. can be estimated. (As a matter of convenience, the label con¬ centration in these experiments was taken as being the lowest concentration of substrate used in a given set of experiments) . The intercept at the S axis is equal to

-K , and the intercept at the 1/v* axis is equal to m K /L* . In the case of multiple single degree transport m systems, the parameter relationships are as follows:

SUBSTITUTE SHEET

The intercept at the 1/v* axis (S = 0)

= 1/ (L*/K _{m(1 )} ) (3) where the lowest K = K . . . m m(l) and the rest of the transporters are numbered con¬ secutively in the order of the increasing value of each K . All subsequent uses of ∑ for sums of series are for n terms commencing with i = 1. The intercept at the S axis (1/v* •*= 0)

-S = Σ {L*/Km)'/∑ { L*/Km ^~ )' (4) '

The final gradient as S - ∞ ,

= 1/ΣL* (5)

The intercept of the above gradient at the S axis

= -Σ(tf _m .L*)/∑L* (6) and the intercept of the above gradient at the 1/v* axis

~- Σ(iC _m .J.*)/∑L* ² (7)

The following equation is derived from the Hill equation, and applies to those multi-degree systems which conform to the Hill relationshi :

On inverti

Thus, if the Hill coefficient can be approximately esti¬ mated, a linear relationship can be obtained which makes p ^c ossible the estimation of the Vm and the Km. It will be shown that, in the case of multi-transporter systems, equations 3 and 9 are extremely useful, the first in the case of multi first degree transporter systems and the second in the case of multi multi-degree transporter systems. It will be immediately apparent that equation 2 is related to the Hanes equation (Hanes, Biochem. J. 26. (1932) 1406) since it has the form -

^(l / ^S * ^)(S /v ⁾ = ⁽¹ / ^s * ⁾⁽ V ^V m ^{+ S} V < ¹² > and is therefore the Hanes equation multiplied by 1/S*. When the Lineweaver-Burk plot was compared with that

derived from equation 2 for glutamine uptake in HeLa cells, it was shown that the 1/v* error span _d ecrease _d 1.6 times between the concentration of 0.01 and ImM using equation 2 compared to 32 times for 1/v using the Linewaver-Burk plot. Also, since the distribution of points in the new plot is directly related to the sub _¬ strate concentrations used, there is no difficulty in obtaining a reasonable separation of significant points over a wide range pf substrate concentration.

In the case of multi-degree relationships where- the velocity of uptake can be represented by the Hill equation:

.*&-. , *_, h , „h , ⁼ < ^si <V ^{+ SΛ)} (13)

^T ransformat ⁱ on of this equation for the velocity of la _b el uptake gives :

= ( S ^Λ - ^J .S*)/(* _m ^Λ + S ^Λ ) (14) On inversion equation 14 becomes:

1/v* = (K _m ^h /L* ) . l/S ^h'~~ + S/L* (15) Differentiation of the above equation with respect to dS gives: (l/v*)/dS = (1-Λ)* _m ^h /L*.l/S ^Λ + 1/L* (16) When the above gradient approaches a minimum (d(l/v)/dS - 0), then d ⁽ l/v*)/dS = 0 = (l - h)K ^h /L* . 1/S ^h ₊ 1/L* and (1 - h)K _m ^h /s ^h = -l which gives the result:

V ^{S =} < ^! / ⁽ * " 1)> ^{I Λ} (17) Thus the relationship between the minimum (d(l/v*)/dS = 0) and the K _m is simply dependent upon the Hill coefficient and is given by the expression {1/(Λ - l ) y ^{1 h} . _It should be noted that this expression has some interesting and useful properties. First, as h approaches 1 the expression goes to infinity; second, when h equals 2 the expression equals 1; and third, as h gets larger the expression decreases to abou 0.7 and returns to 1 at infinity.

It can be demonstrated that for h values between 2 and 10 the expression varies between 1 and 0.7. This small

variation in the ratio between the concentration at the minimum plot point and the K^, for values of h = 2 and above means that an approximate estimate of the K . can be obtained directly from the concentration at the plot minimum. This value can then be used in Table 2 which provides a list of h values in relation to velocity ratios at different concentra _¬ tions relative the K___ . Paired ratios of velocities at con _¬ centrations which are equidistant from the K value are used to get duplicate estimates of h from the Table. The values of h obtained from the paired readings should not differ by more than 1. If the difference between the paired readings is significant then the K _m should be adjusted downwards if the higher concentration is giving the lower h value, or upwards if the reverse is true. The K adjustment is con _¬ tinued until the h values are within 1 unit of one another, which then provides a new estimate of K ., . The mean of the

__ _ . ~~^ ) two h values is then taken as an estimate of \ This estimate of h can then be used in Table 3 (ratios of K /S, where S = the concentration at the plot minimum for different values of h) to further specify the value of the K . Having obtained a good estimate of h it is then possible to plot 1/ v* against S (equation 9) which provides a further specification of V _m and K _m . It can also be shown (Fig. 5, see Appendix I) that the above minima can be seen as inflexion points between pairs of positive gradients in multi-component systems when h is equal to or greater than 2.

Finally, the estimation of the approximate V for the separate transporters in multi-variate systems can also be directly derived from the plot gradients. Thus the plot of 1/v* against substrate concentration provides an initial indication of the number of transporters present, the range of concentration over which they operate, and whether the type of uptake is represented by single or multi-degree Michaelis-Menten (Hill) equations. This then provides a 'structural hypothesis' which covers the essential charac _¬ teristics of the transporters present, and which can be used as a starting point for a more precise determination of the

parameters of all the transporters appearing in the structural hypothesis. The development of a structural hypothesis is of course dependent on a reasonable separation of the K values, and the absence of extreme differences in the V _m values. These requirements are covered by a set of rules in Appendix I where the details of the calculation procedures are dealt with. Application of the above cal¬ culations to glutamine uptake in HeLa cells led to the determination of the parameters for three main transporters in Hela cells. The two transporters with the lowest Km values being Na-dependent (transporters I and II) and the on with the highest K (transporter III) being Na-independent. Appendix 1 shows the analytical procedures used to demon¬ strate the Na -ion dependency of transporters I and II, and the consequential parameter changes arising from Na -ion deficiency. Transporter II was shown to be the transporter which has a concentration velocity curve which covers the range of glutamine. concentration in blood. It is therefore assumed that this transporter is the one which is physiologi cally significant, and is therefore usually referred to here as the main Na-dependent transporter.

The Structural Hypothesis: General description of models used.

Table 11 shows the parameters of the models chosen for testing the plotting procedure. In this analysis each K is assumed to represent a single transporter. Model 1.

Figure 39 shows the plot for model 1 (one single degree transporter) . The axes intercepts are:

The intercept at 1/v* the axis = KL/L* the gradient = 1/L*. Model 2♦ Figure 39 shows the plot of model 2 (two single degree transporters) . This plot is non-linear and is a composite of two linear relationships separated by a single inflexion, the first linear relationship being the initial

SUBSTITUTE SHEET

gradient which approaches 1/1* * and the the second being the final gradient which is equal to 1/L * .

Figure 40 shows that there is a marked departure from linearity (relative to the final gradient as S - ∞ with the downward inflexion starting near Km(,„2.) and inter- ceptmg the 1/v* axis at a point which is 30% below the extrapolated final slope (S - oo), a difference which should be very obvious in most transport experiments.

On the other hand in the Lineweaver-Burk plot for

Model 2 the departure from linearity is negligible and can be observed only between 1/K _m m _t _..) and 0, and at 1/32Tm( _/ 2 ₀ . ) the divergence is only 9% of the extrapolated line which would be difficult to distinguish from experimental error in most experiments, since a very small adjustment in the slope of the line would be sufficient to include the points below 1/

■K ^" ₂ . In the Eadie plot (Eadie, 1942; Hofstee, 1952) for the same model, the departure from linearity is even more marked than that in Figure 40, and it is possible to obtain values for Vm(t.) and Vπι(.l..)/ _m_,(l..) + V ^m (.2-.) IKm(2 _n ) from the extrapolated intercepts at the v and v/S axes respectively. Model 3.

Figure 41 shows the inverse label uptake plot for a multiple transport mechanism which consists of three single degree transporters. This curve is similar to that obtained in Figure 40, except that there are three linear relation¬ ships linked by two inflexions. The gradient of the first linear relationship (where S is close to 0) intercepts at the 1/v* axis to give the relationship shown in equation 3. This is usually an accurate intercept and can be used to determine once - ^m (-- _n *-)/V-LK*- _n (l,) has been estimated. The curve is first analyzed by obtaining the slopes Δ(1/V*)/ΔS from one lower point to the next. In multi-component single degree systems this will reveal portions of the plot where the change in the slopes from one point to the next are at a minimum. These sections of the plot which are near linear alternate with more entensive sections which extend from each near linear section up to the next K _m value. The near linear

SUBSTITUTE SHEET

sections start at or just above the successive Km values (se

Calculations). The lowest one on the concentration scale to be seen being that of the if ₁ . The first gradient is an underestimate of 1/Z, * (overestimate of Vm(,l, ) _. )', and the intercept at the S axis as 1/v* - 0 is an overestimate of

-2T .... The last slope before the second minimal change in slope (see Model 3 data) provides an estimate of 1/(1. * +

Ly* ) and so on. The last gradient provides a close approach to 1/Lt*. Thus ea,ch Vm can be obtained by ^J subtraction of th previous estimate moving along the plot from the lowest concentration. Model 4.

Figure 42 shows the characteristic curve obtained when there are two transporters present with transporter 1 being multi-degree relationship (all the multi-degree rela¬ tionships considered here are of the Hill type) and trans¬ porter 2 a single degree relationship. This plot displays a negative gradient starting from zero concentration with a sharp minimum near -K _/ i *. *•* as previously indicated by the application of equations 2 and 17 in the main text. In the case of curves which conform to the Monod, Wyman and Changeaux (1965) model, the 1/v* plot will again show a clea minimum with an inflection p ^c oint at the 0.57m ( ^v Km )' ■ The major difference between the two types of plot is in the nature of the negative gradients starting at zero concentra¬ tion. Considering the Monod type of relationship, let the dissociation constant for the T _Q form ^■ = K„, and that for the R _Q form 4, c = 0.01 and L = 10,000, then, applying the Monod equatio to calculate v* we can plot 1/v* against S. This plot will show the following diagnostic indications:

1. Starting from zero concentration there is a negative gradient which initially increases to produce a maximum negative gradient in the region of K-.. The gradients are much more gradual than those seen for a simple Hill type relationship, which produces very steep gradients which rapidly decrease towards the minimum inflexion point. In th

SUBSTITUTE SHEET \

case of the Hill equation where h = 4, Km = 0.15 and Vm = 20 the gradient seen at a concentration of 0.05 is -540, wherea that for the plot of the Monod equation was -38.

2. In the plot of the Monod equation the negativ gradient eventually bottoms out to give an inflexion point near the Km (0.5Vm)', and this was found to be true for all values of L above 100.

3. Decreasing L results in a decrease in the position of the inflexion point on the concentration scale, and the inflexion broadens to give only small changes in the negative gradient. As L approaches 1, there is very little difference between the type of plot produced by the Monod equation and that derived from the Hill equation when h < 1.5. Model 5.

Figure 43 shows the type of curve given by a singl degree transporter followed by a multi-degree transporter. In this case there is a positive gradient followed by a shar inflexion towards the final gradient. The first gradient provides an estimate of 1/Lι * and the second gradient pro¬ vides an estimate of 1/L * . The plot minimum at 1.4mM (inflexion point) gives an approximate estimate of ₍ ~ . , and the intercept of the extrapolated plot as S - 0 at the S axis is an estimate of -Km( .l.) .

Models 6 and 7.

These are shown as contrasting plots in Figure 44.

Line 1 depicts the plot for model 6, and this repeats the essential features of model 4 over the first segment of the curve: these being a negative gradient followed by a sharp minimum just above the Km(,l,.) value. This curve then follow a different course to that of model 4 in that there is a sharp increase in the gradient which provides an estimate o 1/L.* followed by an inflection over towards the final gradient which as before gives an estimate of 1/L * . The curve shown as line 2 demonstrates one type of curve which very difficult to analyze and quantify. The negative gradient is once again indicative of a multi-degree equatio

for transporter 1. However, the approach to the minimum is much more gradual so that the inflexion point is not clearly indicated. This suggests that the value of h is close to or below 1.5, or that we have a Monod type of relationship with

L near to 1, and the slow roll-over to the final gradient indicates that the following transport system is a multi- degree with a significantly higher value for the V , or that the true Km( .2«.) is above 1.

Model 8.

Figure 45 shows the 1/v* plot for a mixed three component system. The diagnostic analysis of the previous plots should be sufficient to indicate the nature of the transport mechanism indicated by this plot. The three gradients representing the three transporters are indicated in the Figure. Transporter 1 is a single degree relation¬ ship, and the plot shows a minimum gradient between trans¬ porters 1 and 2 indicating that transporter 2 is a multi- degree relationship. Finally, the roll-over from transporter 2 to transporter 3 is an indication that transporter 3 is also a multi-degree relationship. The roll-over is very gradual, so that the estimation of the minimum by eye is very approximate.

1. Rules specifying parameter limits in multi- component systems. Introductory notes:

The transporters are always numbered 1 to n com¬ mencing at the lowest K . All plots were tested as regres¬ sion lines on the H-P 15C. The following data is prepared before commencing the analysis: velocity against concentra¬ tion, 1/v* against concentration and Δ(1/V*)/ΔS. In the examples considered below the maximum number of transporters present is never greater than three. In general the middle transporter provides the model for the treatment of all intermediate transports irrespective of number.

All concentrations are in mM, and all velocities are in nmoles.min ^" .mg protein. Rules:

SUBSTITUTE SHEET

1. (i) In general the difference between successive K values should be greater that four-fold (in th case of multi-degree systems this can be close to two-fold) .

(ii)' That V values should not differ from on another more than five-fold, and that as a general rule, -

^(V mL ^V mH ⁾ ^m(x ₊ l) ^iξ m(x)) ⁴ < ¹⁹ > where V mL_ = the lower of the two Via. values,

^mH ⁼ higher V___ value, m(x+l) ^{= the u} P ^per m ^value ' and Jm(.x). = the lower Km value

(iii)That plot points must go down to a con¬ centration which is at least one fifth that of Km(,l, .) in orde to ensure reasonable approximations of the parameters of tha transporter, with a minimum of at least three points below m(l) ^'

(iv) That plot points must go at least as hig as 2Km(,n.) (the hig ^~~ hest Km)', with a minimum of at least three p ^c oints above Km( .n.) .

(v) That there should be a minimum of at least four plot points between each pair of consectutive K values.

2. Rules for developing a structural hypothesis. (i) (a) Initial gradient is positive:

Transporter I is a single degree relationship. Estimate

Vm(.l..) from the final g ⁼ radient as S - 0, and Km(., .) from the intercept of this gradient with the S axis = -■ ,-. » or from the first position of minimum change in gradient.

(b) Initial gradient is negative: Transporter I is a multi-degree relationship. The concen¬ tration at which the initial gradient reaches its minimum value gives the first estimate of -R _{T X Γ} and the slope at th top of the first positive gradient gives an estimate of 1/L. and hence Vm(,l, .) .

(ii) In the case of multi-component single degree relationships estimate the remaining transporters by obtaining the successive K values from the positions of minimum change in slope (near linear stretches) and the

successive 1/ΣL* values from the slopes immediately below the stretches of minimum change in gradient. The final gradient supplies 1/L * .

(iii) In the case of multi-component multi- degree relationships estimate the successive K values from the inflexion points between pairs of positive gradients. The inflexion points are indicated by the positions at which negative changes in the rates of gradient change switch to positive changes in the rates of gradient change. The cumu¬ lative 1/L* gradients are obtained from the gradients at the top of each stretch of positive gains in gradient. As in (ii) above the 1/L _fc * is given by the final gradient.

3. Rules for detailed analysis of parameters in multi- component single degree relationships

(assuming 3 components ) .

(i) Use equation 3 (main text) and the ini¬ tial estimates of Km__,( _t 1 _l .) and L1- * to estimate Km(.2,.)/,12.*, and hence the estimated intercept of the net 1/v-* values on the 1/v* axis. The estimated net v, values are obtained by cal¬ culating the v., values using the initial transporter I parameters and subtracting from the v . values. The points are taken in the estimated range Km(.1...) to 2Km _I (1.) or 0.0 IK m( , 3 -, ) _* to 0.02J? ^" , ₃ . In this range the influence of the transporter

III velocities would be negligible, assuming a more than four-fold g ³ ap ^c in the successi*ve Km values. l/v2«* is then plotted against S and the intercept at the 1/v axis is estimated from the regression line. This value for the intercept is then compared with that estimated from equation 3 above. It is likely that the plotted intercept will be well above that estimated by equation 3, that is the ratio of the plotted result over that of equation 3 (R) >1. Iterate the procedure by continuously decreasing L- * in approximate 10% steps until R becomes constant. At this point the K _{f l} *. is then decreased by 10% in a single step. Decreasing the L, * is then repeated until once again R is constant. This procedure is continued until R = 1. This then provides a means for simultaneously solving the parameters of both ~

transporters I and II. Transporter II parameters can _b e ^f urther confirmed from the final net 1/v* against S regres _¬ sion line.

(ii) The above procedure is now repeated with the net v^ ₃ values. The previous estimates of the trans _¬ porter II parameters are put into equation 3 and the inter _¬ cept of the net l/v ₃ * estimated. The transporter II parameters are now adjusted as in the case of the transporte

I parameters above, and the final values for the transporter

II and the transporter III values are obtained.

(iii) To confirm the results of the analysis the net v _χ values are obtained and l/v _χ * is plotted against S in the range 0 to about 3K___ _± , and the correlation coef _¬ ficient obtained. This should be above 0.98.

4. Rules for detailed analysis in multi- componen multi- degree relationships.

(i) If transporter I is a single degree relationship, estimate the net v ₂ values using the trans _¬ porter I parameters estimated as part of .the structural hypotheses by subtracting the estimated v, from v . . Plot i obs l/v ₂ * against S from 0.5iT _m(2) to 2J^ ₍₂₎ . A negative gradient indicates that transporter II is a multi-degree system. The concentration when 1/v* is at a minimum then indicates the

than 1. If the difference between the paired readings is significant then the K _m should be adjusted downwards if the higher concentra-tion is giving the lower h value, or upwards if the reverse is true. The #__. adjustment is continued until the h values are within 1 unit of one another, which then provides a new estimate of *^ _{m( 2 ) '} ^{τhe mβan of} the two h values is then taken as an estimate of h~ . Table 13 (using equation 17) then indicates the correction required from the first esti-mate using the plot minimum. The two adjusted K _m *

values should agree to within about 20% (it should be noted that the Table 13 correction is based on a single plot point esti-mate). Carry out a 1/ v* plot against S ^h . The trans¬ porter II values are then used to obtain net v. values and the 1/v * is plotted against S in the range below K . . This plot provides a reestimation of the transporter I parameters, which are then used to reestimate the transporte II parameters, and the process above is repeated until con¬ stant values of one set of the transporters is obtained. Th final plot of net 1/ v * against S in the range of S from below Km( _/ _.. ) .to 0.2ϋTm(._-j,). p^rovides an estimate of the trans- porter II parameters.

(ii) If transporter I is a multi-degree relationship, the procedure is the same as that given for transporter II in (i) above. The h value is then used to plot 1/ v* against S to obtain V .- . and further confirm K ..... These parameters are then used to obtain net v ₂ values, and, assuming that transporter II is a multi-degree relationship, the procedure continues as for transporter ii in fi . above.

These parameters are then used to obtain a cor¬ rected assessment of the transporter I parameters, and the process of net determination of transporters I and II is iterated until the parameters of one or the other become constan . Estimation of the transporter II parameters is obtained by the final 1/ ₂ v* plot against S .

(iii) The net v- parameters are now estimated using the transporter I and II parameters previously esti¬ mated, and the net /v ₃ * values are plotted against S in the rang ^~~ e . ) to 2Km(.3._.) . h3- and -m_(.3.,.) are estimated as above for the other multi-degree systems, and brought to constancy on iteration. Plot 1/ v ₃ * against S, and estimate

Km( .,.) and V_m(.,.) .

(iv) The round of net estimations is completed by the estimation of the net transporter I (that the trans¬ porter with the lowest K ) velocities by subtraction of the calculated V2 + V3 values from v _ODS , and plotting 1/v* or

1/ v* over a range of concentrations where the contribution of the transporters with higher K values is minimal.

(v) Fine tune by checking v , against calcu¬ lated v values at the lowest concentrations, the highest concentrations, ' at the Km values and midway ^■ * ^■ between the Km values. Only minor adjustments should be required to bring the majority of observed velocities onto the calculated curve. Model 3.

Transporter Km Vm

1 2 3

Data: Concentrations are mM, and velocities are nmoles.min ^"1 .mg ^-1 protein.

S.mM. 0.005 0.01 0.02 0.03 . .

TITUTE SHEET

Notes:

1. A constant rate of decrease in the slopes between points occurs between points occurs between K values. These sections of the curve are underlined.

2. In the vicinity ^• * of the Km values the curves become almost linear (minimal change in slope). These sections occur just above the Km values and are bolded.

3. S* = 0.005mM. Structural hypothesis:

Rule 2(i) a. Regression line through 0.005-0.02mM

1/v* intercept = 0.49, Gradient = 12.57. m(l) - °' ^04mM ' m(l) 15.9.

Rule 2(ii) m(l) ^<0 ' ⁰⁴ ' m(2) - °' ⁴ ' Vm(2) = 16.4m, Km(3)

" ² ' ⁵ ' ^y m(3) " ²⁰ - ⁶ - Parameters from the structural hypothesis:

Transporter Km V_m

1 1 2 1 3 1

Rule 3(i) Inverting equation 3:

^L ~ * ^/K -~ _\ ( l _. ^{+ L} * * ^K m( 2 ₎ ⁼ !/In ercep = 2.04 0.08/0.04 + x = 2.04 and x = 0.04, 1/x = 25 (initial intercept much too high) . Testing the ^L ι */K _{( 1} . ratios using net l/v ₃ * plots:

Intercepts at the 1/v* axis agree.

^{L~ *} " °' ^05< ^m(l) ^{= 20} ' ^and m(l) ⁼ °' ⁰³ -

Net l/v ₂ * intercept at the 1/v* axis is 2.66, the gradient is = 7.93

Then ^f m_(,2 ₂ ) = 0- ^• .—33 ^r , a ---n-d-- Vm _m (2 ₂ ) = 25.2

Rule 3 , ii ..

The above parameters for transporters I and II are used to estimate net v ₃ values for use in the application of equation 3.

Equation 3: ^L - *^ ^K i 2 λ ^{+ x =} °* ³⁷⁶ ' therefore

0.126/0.33 + x = .376, x will be too high.

Testing ^L~ ** ^iR _r _ _t 2 \ ^ra *tios using net l/v ₃ * plots:

L2* = 0.10, therefore Vm(.-_.) = 20, and Km(.2«). =

0.30mM. l/v ₃ * plot: 1/vs* intercept = 23.95, gradient = 7.97. m( ) ^m (3) Rule 3fiii . :

Confirm transporter parameters: 0.005 0.01 0.02 0.03 0.04 0.05mM

1/vi* 0.7 0.8 1.00 1.2 1.4 1.6

SUBSTITUTE SHEET

1/vι* axis intercept = 0.60, gradient = 20.0

Structural hypothesis:

Rule 2(i) a. Positive gradient: Transporter I single degree relationship.

SUBSTITUTE SHEET

Regression line through 3 lowest points:

1/v* axis intercept = 0.615, gradient = 15.29

Rule 2(iii). S Vucceie)d ⁼ ing° ^' p ⁰⁴ ar'am^et(eir)s ⁼ ar ¹ e ^{3 <} indicated by a decreasing gradient which reaches a minimum at O.lmM and a new increasing gradient commences at 0.15mM = K ^' , _2\ ' ³ - ^nc - ^- ^ε completed with a final gradient of 0.59 at 0.5mM therefore

VmL _/ (l _{ι \} ) ^{+ V} IΛ _. (, - _> ) _. ⁼ 33.9, g'iving V ^m (..2,.) = 20.9.

The next minimum is at 1.5mM with the increasing gradient commencing at 1.75 mM = ^ _{m (} ' ^anα * ^t * ^ιe fin l gradient is 3.61, therefore V_... = 55.4, which makes

Structural hypothesis parameters:

Transp ^orter Km Vm h

1 0.04 13 1

2 0.15 20.9 >1

1.75 21.5 >1

Rule 4.i.

Estimate net v ₂ value parameters using the estimated parameters for transporter 1 above: JT ... = 0.04 i ^{= 13} -

■ξm,( _{. *} * _. ) _. = 0.15. v ₂ at 0.15 = *"m,,(i-.) = 8.07,

" 0.11 = 0.75J. ..J. = 5.20*,

" 0.19 = 1.25JTm(.l..,) = 10.04*.

♦Estimated by interpolation.

Velocity ratios (% at the respective concentration expressed as a proportion of K ^' ' 0.15K /K = 65%, and

Applying the ratios to Table 12 shows that

Using Table 13, K /0.15 = 0.85. Therefore

*m(2) ^■ 0- 3.

SUBSTITUTE SHEET

Carry out 1/ v* plot of net v ₂ values. First net v ₂ plot.

S ^2,5 0.0032 0.0087 0.018 0.031 0.049 l/ ^A v*flst.Est. 392 610 956 1421 2035

1/v* axis intercept = 296, gradient = 35,785.

^K - _{t ( 2 .} ⁼ 0-15, ^v _{m (} 2 ⁼ 5-9, h ₂ = 2.5, correlation coefficient (r) = 1.0.

Use the above parameters to reestimate transporter I: First net vi plot.

S.mM. 0.005 0.01 0.02 v _χ 1.44 2.55 4.17 1/v* 0.69 0.78 0.96

1/v* axis intercept = 0.6, Gradient = 18.

Useuthe ⁼ n °e'w ⁰³ t ³ r'anspoDrte ⁼ r ¹¹ I'p ¹ a'ra ^r me ⁼ te ^λ r's ⁰ 'to replot v ₂ .

Second net l/v ₂ * plot.

S.mM. 0.05 0.075 0.10 0.15 0.20 0.25 v ₂ 1.56 3.43 5.5 9.23 11.97 13.84

1/v* 6.4 4.37 3.63 3.25 3.34 3.61

Km( ,2~ .) = 0.15 from the plot minimum. v ₂ at 0.15 = K _m , _{2 )} ⁼ 9*23,

" 0.11 = 0.75iT _m/2 . = 6.25*,

" 0.19 = 1.25JTm _/ (2.) = 11.29*.

♦Estimated by interpolation.

Velocity ratios (% at the respective concentrations expressed as a proportion of K : ⁰ _' ⁷⁵ * ^κ /-^ _m ⁼ 68% (Ratio 1), and Km /1 . 25Km = 81% (Ratio 2).

Table 12 shows: h = 2.25 for ratio 1, h = 2.13 for ratio 2.

The above h values are close to identical, there¬ fore no further adjustment is necessary, the mean value for h = 2.19.

The Km( _/ 2) = 0.15mM from the plot minimum, on correction for the h value in Table 13 gives 0.15 x 0.96 = 0.144mM.

SUBSTITUTE SHEET

2nd. net 1/ v ₂ * plot.

S.mM. ( l 0.15 0.20 0.25 0.30

S2.2 0.0063 0.015 0.029 0.047 0.071

.

Therefore Km(.2). = 0.15, and Vm(.2,..) = 18.4

Use the above parameters to replot l/v_* 2nd. net 1/vι* plot.

S.mM, 0.005 0.01 0.02 0.03

Vi 1.44 2.54 4.13 5.24 1/vι* 0.70 0.79 0.97 1.14

1/vι* intercept = 0.61, Gradient = 17.9. ^11*1 - Rule 4fiii ..

Estimate v ₃ using the parameters for transporters I and II.

Transporter Km V_m__ h

1 0.034 11.1 1

2 0.15 18.4 2.2

Plot l/v ₃ *:

Estimated •£_,.,. ⁼ 2.0 from plot minimum. v ₃ at 2.0 « K m( ,3,). = 19.50

" 1.5 = 0.15K m(.3-.,) = 12.96

" 2.5 = l_25.tr m(.3,.) = 21.76.

Velocity ratios (% at the respective concentrations expressed as a proportion of K _: ^0, (Ratio 1), and iT m/1.25Jrm = 90% (Ratio 2).

Table 12 shows: h = 2.5 for ratio 1, h - 1.0 for ratio 2.

The low value for h on the up ^•IΓ p ^*** er side of the Km value suggests that the initial K estimate is too high. Fine tune -K _{/ T} *. until h values are equal. K . ₃ . was finally adjusted to 1.6mM. Then ratio 1 = 53.6% and ratio 2 = 74%. Table 12 shows the h = 3.5 for ratio 1 and 3.5 for ratio 2.

Km( .3-. .) = 1.6 and h = 3.5.

Plot 1/ ^J V ₃ *:

1.0 1.25 1.5 1.75 2.0 3.0 4.0 5.0

.3.5 1.0 2.18 4.13 7.1 11.3 46.8 128 280

1/ ^Λ v ₃ *xl0 ^"6 24 28.8 36.1 48.0 65.6 220 579 1250

1/ v* Intercept = 17.26 x 106, gradient =

4.4 x 10 ^β .

Therefore Km( _/ .) = 1.48, Vm(._.) = 25.7, and h = 3.5.

Confirm Km___(.3-.) _* using Table 13: ifm(,3,,) = 2.0 x 0.775 = 1.55mM.

At this stage the estimated parameters are:

Transporter Km Vm h

Rule 4.iv..

Finally estimate net v..

l/v ₁ * intercept = 0.60, gradient - 18.57.

Therefore Km _m ,(l. .) = 0.032, and Vm _m ( _/ l.) = 10.8.

On completion of the detailed analysis, the estimated parameters are:

- 50 -

Transporter m Vm

Rule 4(v)

Fine tune: Check observed and calculated veloci¬ ties at the Km values, and at selected below the lowest, above the highest, and between.

No further adjustments required. Final parameters (actual values are in parentheses):

Transporter m V_m

1 0.032 (0.03) 10.8(10.0) 1 (1) 2 0.15 (0.15) 18.4 (20) 2.2 (2.0) 3 1.48 .1.50. 25.7 .25. 3.5 .4.0.

ANALYSIS OF THE CURVE FOR GLUTAMINE UPTAKE IN HeLa CELLS.

Figures 46 and 47 show the uptake curve and the inverse label plot respectively. Data: . 0 .01 0.015 0.025 0.03 0.05 0.10

1 . 75 2 . 0 2 . 5 3 .

Structural hypothesis: Rule 2(i) a. Positive gradient: Transporter I single degree relationship.

Regression line through 3 lowest points:

1/v* axis intercept = 0.74, gradient = 20.

Km _t (1. .) = 0.04, Vm(1) = 10.

Rule 2(iii). There is a zero gradient between 0.025 and O.lmM, and a new increasing gradient commences at 0.2 indi¬ cating the approximate value of J_m(2). This increasing gradient is completed with a final slope of 5.4 at 0.5mM, therefore Vm___( .l. .) + Vm,(2~). = 37, g ⁼ iving ^~~ V_m__(.2-.) = 27. The next minimum increase in the slope occurs at 2.0mM with a new increasing gradient commencing at 2.5mM = -K / . ^{and tιe} final gradient is 3.9, therefore m.t.. = 51.3 which makes

Estimate net v ₂ value parameters using the estimated parameters for transporter 1 above: •K-m(l) = 0.04, V _m(1) = 10.

S.mM. 0.05 0.1 0.20 0.25 0.50

Jm(. _T ,) = 0.20. vj at 0.15 ^■ Jm(„.) = 13.7,

" 0.11 = 0.75Jr _m(2) ^« 10.5*,

" 0.19 = 1.25Jm,(,2.) = 15.4*.

^♦ Estimated by interpolation.

SUBSTITUTE SHEET

Velocity ratios (% at the respective concentrations expressed as a proportion of Km: 0 . 15Km /Km = 73.3 = Ratio 1, and Km /1 . 25Km = 89% ^~~ Ratio 2.

Applying the ratios to Table 12 shows that h ₂ for

Ratio 1 = 2.0, and hi for Ratio 2 = 1.1. This indicates that the estimation of Km is too hig ³ h.

The final adjustment of K to 0.15 gives: v ₂ at 0.15 = J? _m , ₂ = 10.05,

" 0.075 = 0 .5K ₂ = 3.675,

" 0.225 = 1.52Cm(.2,,). = 14.55.

Velocity ratios (% at the respective concentrations expressed as a proportion of Km : 0 .5Km IKm = 37% = Ratio 1, and Km/1.5Km = 69% *■= Ratio 2.

Table 12 shows that for ratio 1, Λ ₂ = 2.25, and for ratio 2, h = 2.25. This confirms the Km(.2«). at 0.15mM.

Use the above hi value to correct the initial

K _{m (} 2 _\ estimate from the plot minimum as shown in Table 13:

K . . = 0.2 x 0.92 = 0.18mM. ^m ( ² ) _h

Carry out 1/ v* plot of net v ₂ values.

First net v ₂ plot.

S ² ' ²⁵ 0.0056 0.027 0.044 0.021

1/ ^Λ v* 132 294 432 1597

1/v* axis intercept = 103, gradient = 7127. ^' ₂ . = 0.15, ^v _{m ( \} ⁼ 21.1, h ₇ = 2.2, correlation coefficient (r) = 0.9999.

This confirms transporter II parameters.

Use the above parameters to reestimate transporter I: First net i plot. S.mM. 0.005 0.01 0.015 0.025 0.03 0.05 i

1/v*

1/v* axis intercept = 0.74, Gradient = 26.48.

Km _f (l) _. " 0.028, m(....) - 7.55, r = 0.9783.

Use the new transporter I parameters to replot v ²

m(2) ⁼ °' ¹⁵ - ^v ~

♦Estimated by inte

Velocity ratios (% at the respective concentrations expressed as a proportion of K : 0 .5K /K = 41% (Ratio 1), and Km /1 .5Km = 71% (Ratio 2)'.

Table 12 shows: Λ ₂ = 2.0 for ratio 1, Λ ₂ = 2.0 for ratio 2.

The above h values are close to identical, there¬ fore no further adjustment is necessary, the mean value for h = 2.0.

Use the above h ₂ value to correct the initial ^K _{( 7 \} ^est imate from the plot minimum as shown in Table 13: ■ (2) = 0.15 x 1.0 = 0.15mM.

2nd. net 1/ ^Λ v ₂ * plot.

1/ Intercept = 36.36, Gradient = 1677, r = 0 . 9999 . 23 . 9 , and

Proceed to transporter III. Rule 4(iii) .

Estimate v ₃ using the parameters for transporters I and II.

Transp ^£ orter Km Vm h

1 0.028 7.55 1 2 0.15 23.9 2.0

SUBSTITUTE SHEET

vs a^ . * r m( .3..) = 1 .

" 1.5 = 0.75iT _{m 3} = 6.22,

" 2.5 = 1.25iT m( .3..) = 19.5.

Velocity ratios (% at the respective concentrations expressed as a proportion of K : ^.75iT /JT = 38% (Ratio 1), and Km /1.25Km = 85% (Ratio 2)'.

Table 12 shows: Λ ₃ = 5.0 for ratio 1, h ₃ = 1.6 for ratio 2.

Fine tune K .,. until h values are equal. m(3)

Reestimate Λ ₃ : Final adjustment of

Velocity ratios (% at the respective concentrations expressed as a proportion of K : 0.75JT /JT = 47% (Ratio 1), and JTm/1.25J-m1 = 63% ( ^v Ratio 2)'.

Table 12 shows: Λ- = 4.0 for ratio 1 , and h~ = 6.0 for ratio 2. Let h^ = 5

Use the above Λ 3- value to correct the initial Km(,3_ _> .) estimate from the plot minimum as shown in Table 13: K_ m__(.3~) =

2.0 x 0.71 = 1.4mM. Plot 1/ ^Λ v ₃ ^:

1/V Intercept = 0.245 x 10 ¹³ , gradient = 4.4 x 10 ¹³ .

Therefore K _m , _{3 )} - 1.77, ^ _m/3v = 20.9, and h = 5.0 Rule 4(iv) .

Reestimate net transporter I using the above parameters for transporters II and III. Data: mM 0.005 0.01 0.015 0. 2 .

Notes:

1 Bolded numbers are numbers requiring adjustment. 2 v(l) are results from calculated parameters. 3 v(2) adjustments are: K___ . . to 0.03mM, V . . . ^v ' m(l) m(1) to 6.8, Vm(2) to 25 and V__ to 22. m(3) 4. v(3) adjustment is h~ to 6

Final parameters:

SUBSTITUTE SHEET

Transporter

1 0.03 6.8 1 2 0.15 25 2 3 1.8 22 6

Figure 48 shows the calculated curve in relation to the means of the plot points.

Note: Experiments were carried out at different times using 2 different parameters for velocity determina¬ tion. Numerically these two sets of parameters turn out to be almost identical in that 1 fmol/min per cell = 1.12 nmol/ min per mg protein.

Progression chart: a = Structural hypothesis, b = detailed analysis, c = fine tuning.

Transporter

ANALYSIS OF THE CURVE FOR THE Na -DEFICIENT UPTAKE OF GLUTAMINE INTO HeLa CELLS.

Figures 49 and 50 show the uptake curve and the inverse label plots respectively for normal uptake and Na - deficient uptake. Data: S 0.025 0.05 0.10 0.15 0.25 0.30 v 0.30 0.70 1.50 2.40 4.20 5.15 1/v* 16.7 14.2 13.3 12.5 11.9 11.7

Δ(1/V*)/ΔS oo -18 -16 -6.0 -4.0

0.35 0.40 0. 0.60 0. .7

Structural hypothesis: Rule 2f and iii.b.

Initial gradient is negative therefore the first degree transporter I has been eliminated. The gradient minimum occurs minimum near 0.5mM = Km(,2~) _. . and the final gradient at the end of the following increasing gradients is at 0.75mM, the gradient = 12.0 = 1/L ₂ * = -~ ^/v _m(2\ ^x 0-005) therefore V_, ₂ » = 17nmol.min~ ¹ .mg-_ protein. The next minimum in minimum change in slope occurs at 2.5mM = K .-.., and the maximum slope following this minimum (at 5mM) gives a value for mw(2i*). ⁺ _ ^v ~m.t( ~ 69 nmol.min ^-1 .mg protein. Thus V , ₃ . = 52nmol.min ^"'1 . ¹ protein. The shape of the curve is typical for that of two multi-degree systems (Figure 44) Structural hypothesis parameters:

Transporter m Vm

1 2 0.5 17 >1

3 2.5 52 >1

E T

Rule 4.1..

Estimate Km(.2.) and h2: Final adj ^J ustment of Km(,-2). =

0.5mM. v ₂ at 0.5mM (K _{m 2 )} ) = 9.0 " 0.375mM(0.75iT _{m 2} ) = 5.44, " 0.625mM(1.25JT _{m 2} ) = 18.40. Velocity ratios (% at the respective concentrations expressed as a proportion of K_: 0.15K /K = 72% (Ratio 1), and jrm/1.25._rm = 80%( ^x Ratio 2)'.

Table 12 shows: h ₂ - 2.0 for ratio 1, and h ₂ =

2.3 for ratio 2. Let h ₂ = 2.15.

Use the above h ₂ value to correct the initial jrm(.2). estimate from the plot minimum as shown in Table 13:

■* ^m _Λ (-2 _> ) _. = _h 0.5 x 0.95 = 0.48mM.

Plot 1/ v ₂ *:

l/ ^h v * Intercept = 730, gradient = 6578

Therefore Km(.2). = 0.36, Vm( ,2.) = 13.5, and h = 2.15, r = 0.9861

2.5mM.

Ve oc ty rat os ( at t e respect ve concentrat ons expressed as a proportion of K - ' ^56% (Ratio 1), and Km/1.25Km = 69%( ^v Ratio 2)'.

Table 12 shows: h _s - 3.25 for ratio 1, and h _s = 4.25 for ratio 2. Let h* ^~ = 3.75.

Use the above Λ ₃ value to correct the initial

Km( .3 -..) estimate from the plot minimum as shown in Table 13

Km,(3-,,) = 2.5 x 0.79 = 2.0mM.

1st. net 1/ v ₃ * plot.

S.mM. 1.25 1.5 2.0 2.5 3.0 3.5 4.0 5.0

.3.75 2.31 4.57 13.45 31.1 61.5 110 181 418

1/ v ₃ *xlθ ^" 8.77 12.1 19.9 31.7 52.75 90.5 149 338

1/ v ₃ * Intercept = 6.74 x 10 ⁷ , Gradient = 0.789 x

10 ⁷ , r = 0.9999.

Therefore K _m i • m _i (3* .) = 1.77mM, Vm(.3,.) = 53.9nmoles.min ^"

"^protein, ^{and h~ = 3} - ¹⁵ ' Rule 4(iv, .

1st. net v ₂ plot.

S.mM. 0.15 0.25 0.30 0.35 0.40

Net v ₂ 2.4 4.2 5.1 5.9 6.8

Decrease in net v ₂ values between 0.6 and 1.OmM suggests that either Λ ₃ or ■*• ^* , _3* is too low. Increase Λ ₃ from 3.75 to 5.0, and replot net l/v ₂ * against S to check whether values of net v ₂ now increase or approach a maximum between 0.6 and 1.OmM. 1st. net l/v ₂ * plot: S.mM. 0.15 0.25 0.30 0.35 0.40

SUBSTITUTE SHEET

Plot

expressed as a proportion of K : 0.752C /K = 71% (Ratio 1), and K /1 . 25Km = 86% ( ^v Ratio 2)'.

Table 12 shows: h ₂ = 2.0 for ratio 1, and h

1.5 for rattiioo 22.. LLeett hh ₂ ~ == 11..7755..

Using Table 13, the Km correction shows Km(.«2.)

0.42. Good agreement! 1st. net 1/V* plot.

S mM 0.15 0.25 0.30 0.35 0.4 0.50

1/V* Intercept = 135, Gradient = 752, r =

0.9658.

Therefore 2Tm,,(,.) = 0.375mM, Vm( .2.. .) = 14.1 nmoles.min~ ¹ mg.- ₁ . . , and h ₂ ^~~ 1.75.

Rule 4(ii) .

2nd estimate of net v ₃ :

S.mM. 1.25 1.50 2.0 2.5 3.0 3.5 4.0

Net v ₃ 11.2 16.0 28.6 41.4 49.26 51.2 52.1 1/v..* 22.3 18.7 14.0 12.1 12.2 13.7 15.3

Estimate K , -. . and h ₃ : Initial value for K ._.. m(3) πι(3)

= 2.5mM.

Final adjustment of 2Tm,( _/ 3) = 2.OmM. v ₂ at 2.OmM (•^^• = 28.6, " 1.5mM(0.75JT _{m 3} = 16.0, " 2.5mM(1.252^(3) = 41.4.

Velocity ratios (% at the respective concentrations expressed as a proportion of Km: 0.752? ^" /K = 56% (Ratio 1)',' and K /1 . 25Km = 69% (Ratio 2) '.

Table 12 shows: h ₃ = 3.25 for ratio 1 , and h ₃ = 4.25 for ratio 2. Let h ₃ = 3.75.

Use the above h ₃ value to correct the initial ^K m ₍ ~ _{\ .} ^anc ~ ^est imate from the plot minimum as shown in Table 13: =^(3) = 2.5 x 0.79 = 2.OmM.

2nd. net l/ ^h v * plot.

S.mM. 1.25 1.5 2.0 2.5 3.0 3.5 4.0

S ^5,0 3.05 7.59 32.0 97.7 243 525 1024

Net v ₃ 12.35 17.3 30.1 43.0 51.0 53.0 54.0

1/ ^Λ v ₃ *xl0 ^"10 8.7 15.2 35.8 75.5 158 328 629

1/V* Intercept = 12.3 x 10 ¹⁰ , Gradient = 0.602 x 10 ¹⁰ , r = 0.9999.

Therefore Km(.3..) = 1.83 mM, Vm(.3..) = 53.1 moles.min mg-iprotein/ ^a ~- h ^~ = 5 .

Final net l/ ^h v ₂ * plot.

S.mM. 0.15 0.25 0.30 0. .4

1/ ^Λ v ₂ * Intercept = 143, Gradient = 756, r = 0.9957

Th..erefore Km _ι (2.) = 0.39mM, V_m__(.2~) ^~~ 14.1 nmoles.min ^{" m} 9~ι _Dro tein' ^nd ~~~ ^{= >7} *5'

Parameters on completion of detailed analysis:

I SUBSTITUTE SHEET

Transporter K_m Vm

1 2 0.39 14.1 1.75 3 1.83 53.1 5.0

Rule 4fv) . Fine tune:

The results show that the values at 1.25, 2.0 and 5.0 require an adjustment of around 10% to match the observed values. This would be achieved by lowering the value of h ₃ . However, since this adjustment must lie between 3.75 and 5, it was considered that any further fine turning was not justified.

Final parameters are those given for the detailed analysis above.

Figure 502 shows the calculated curve in relation to the means of the plot points.

Note: K values are in mM and velocities are in nmol . -l. -1 ^m .mm mg protein.

Progression chart: a = Structural hypothesis, b = detailed analysis, c = fine tuning.

Transporter Progression Km Vm

Na ₊ -deficient parameters (in parentheses) compared with those obtained in the presence of Na -ions:

Transp ^c orter Km vm

1 0.03 (-) 6.8 (-) 1.0

2 0.15 (0.39) 25.0(14.1) 2.0 (1.75) 3 1.8 .1.8. 22.0.55.0. 6 .5.

The results show that in relation to normal Na - dependent function, Na -deficiency results in the complete elimination of transporter I, and that transporter II has been inhibited to the extent that the Km has been more than doubled and the V nearly halved. It is suggested that the residual transport is possibly due to a facilitated system, since the uptake of the residual transport is inhibited by leucine, whereas the Na -dependent transport is not. The uptake transport seen is almost certainly due to label exchange as the concentrations of glutamine insede and out equalize.

It should be emphasized that the parameters given for the Na -ion deficient uptake are very approximate, and can be expressed by a series of curves depending on the weighting given to particular sections of the plot. Fur¬ thermore, there is no evidence as to the nature of the mechanism(s) which are responsible for the residual activity. The significance of the above data is that the figures show that the concentration range of activity normally occupied by transporter II is highly inhibited under the condition of Na -ion deficiency.

8. Exposure of cells to glutamine a. Procedure

The procedure consisted of a double-labelling experiment in which the cells were first exposed to 2 mM ³ H- glutamine (50 μCi/ml) for 10 minutes to measure uptake of label into protein (subsequently used in determination of cell number) and were then exposed to - *C-glutamine (50 μCi/ml) for 1 minute to determine uptake rates. In all replicated samples the same labelling solutions were used and TE SHEET

therefore a 50 μl sample was removed after each incubation and counted to correct for label dilution. Concentration dilution did not occur because each labelled, incubation was preceded by a 3 minute incubation with unlabelled glutamine at the same test concentration as that of the labelled glutamine sample. All incubations were carried out at 37°C and before each experiment the cell cultures were incubated with fresh medium 199 containing 10% (v/v) FBS for two hours. The medium was then exchanged for unlabelled 2 mM glutamine in medium 199 and incubated for 10 minutes. This medium was then exchanged for ³ H-glutamine in medium 199 and incubated for exactly 30 minutes. The cultures were washed three times with HBSS followed by incubation with test glutamine solu¬ tions (0.25 - 4.0 mM) in HBSS for three minutes. Immediately following exposure to ¹⁴ C-glutamine for exactly 1 minute, uptake was stopped by washing the cells three times with ice- cold 0.9% (w/v) NaCl followed by three washes with ice-cold acetone. The cultures were then placed in a desiccator under vacuum overnight. b. Separation of Protein and Soluble Amino Acids

The cells were removed from the surface of the culture bottle using a spatula and acetone and the protein was precipitated by coagulation at 100°C (McEvoy-Bowe et, al.. , J. Chromatoσ. , 347, (1985) 199-208). The upper aqueous laye was removed by pasteur pipette and kept. This fraction con¬ tained cytoplasmic amino acids and a 0.2 ml sample was taken to which 0.3 ml of H ₂ 0 and 4 ml of ACS (Amersham) were added for radioactive counting.

The lower layer was re-extracted with another 0.5 ml of ammonium acetate and the upper layer was this time dis carded. The chloroform was evaporated and the protein pre¬ cipitate was washed in 2 ml of 5% (w/v) sulphosalicylate and then two times in 2 ml of H ₂ 0. After each wash, the tubes were centrifuged at 600 g for 10 minutes. The protein pelle was dried at 56°C overnight then 1 ml of 0.1 M NaOH was adde to the dried protein and the tubes were tightly capped then

placed at 100°C overnight to digest the protein. Protein content was determined using BSA as a standard and reading absorbancies at 500 nm (Lowry, et al. , J. Biol. Chem. , 193, (1951) 265-275). Radioactive samples were counted on the LKB Wallac 1215 Rackbeta Liquid Scintillation counter, which was programmed for double label counting of ACS quenched samples. c. Calculations for glutamine uptake The number of cells present was determined by incorporation of ³ H-glutamine into cellular protein during 10 minute incubations. The amount of glutamine incorporated into cellular protein per minute is converted into cell number using the following relationship: Cell number = (2.2 + 24.6 x nmoles gln/min) x 10 ^s . The amount of glutamine incorporated into the cytoplasm in one minute was calculated from the specific radioactivity of the medium glutamine and the results were expressed in terms of cell number or protein content, either method giving almost identical uptake curves. Uptake rates for inhibition experiments involving pre- incubation periods of longer than 3 minutes were expressed in terms of protein content only. d. Laboratory apparatus used in transport experiments

Another embodiment of this invention relates to transport assay plate assemblies useful in carrying out the assays of this invention. 2_nother embodiment comprises a procedure for using such assemblies.

In tissue culture or the growing of cell cultures in clinical chemistry, it is common practice to use assay plates that have a plurality of small recesses or wells for receiving a specimen. A conventional multi-well plate may take the form of a one piece plastic moulding shaped to provide anything from six to ninety six wells positioned in a rectangular array. The plate is often provided with a flat cover which simultaneously closes off the top of all the wells. Individually removable covers may also be provided to allow separate access to each well. In U.S. Patent

SUBSTITUTE SHEET

4,599,314, there is a disclosure of this type of multiple well plate and lid structure.

Transport experiments on tissue or cell cultures frequently require a large number of samples and conditions of accurate timing in order to obtain reliable results. Tissue culture cells which grow as a monolayer on the surfac of culture plates provide a population of homogeneous cells which may be exposed to changes in culture conditions instan taneously and simultaneously thereby allowing physiological conditions to be approximated. Multi-well plates provide th opportunity to provide transport experiments on a large number of samples at the same time, but there is a need for method of transferring pre-incubation and radioactive solu _¬ tions into each of the wells at the same time.

This invention provides a cover for a transport multi-well assay plate comprising a lid with an array of downwardly projecting open mouthed enclosures arranged to be coincident with the wells in the assay plate, whereby when the cover is placed onto the assay plate and the lid abuts the plate, the mouths of the enclosures extend into the well cavities of the plate.

This invention also provides a transport assay plate assembly comprising an assay plate with a plurality of spaced apart wells therein and a removable cover adapted to be free standing on a flat support surface when inverted, th cover having an array of downwardly projecting open mouthed enclosures coincident with the wells in the plate whereby, when the cover is placed over and in abutting contact with the plate, the mouths of the enclosures project into the wel cavities of the assay plate.

This invention also provides a transport assay procedure comprising growing cells on the base of a pluralit of wells in an assay plate, inverting a cover comprising an array of a plurality of enclosures coincident with the wells in the assay plate, adding an incubation medium to each enclosure in the cover, inverting the assay plate and placin it over the cover, and inverting the assembly to ensure that E

the incubation medium in each enclosure is transferred to the adjacent well in the assay plate. Embodiments of the present invention will now be described, by way of example only, with reference to FIGS. 66 and 67.

The transport assay cover described herein is adapted for use with a variety of sizes of multi-well assay plates. In the two Examples described below (Examples 80 and 81), the covers are adapted for use with a six well plate and a ninety six well plate. However, it is understood that this invention includes other covers that would be equally suit¬ able for a variety of multi-well specimen plates. Further¬ more, although the examples describe manufacture of the cover from readily available laboratory equipment, it is understood that the covers could be moulded from suitable sterilizable plastics.

Figures 66 and 67 illustrate an assembly of a cover 10 and a transport assay plate 20 comprising a rectangular base structure 21 with six flat bottomed wells 22 spaced in two rows of three. The cover 10 comprises a substantially rectangular lid 11 of the same cross section as the plate with downwardly extending peripheral flanges 12. A plurality of cylindrical enclosures 13 are formed to extend downwardly from the lid in an array corresponding with the array of wells 22 in the plate 20. Each enclosure 13 is in the form of an inverted cylinder closed at one end 14 and with a mouth 15 at the other end that projects downwardly from the plane of the lid 11. When the cover 10 is placed on the six well plate 20, the lid 11 abuts the planar upper surface 25 of the plate 20 and the mouth 15 of each enclosure 13 projects into the adjacent well cavity.

In use, cells are grown on the bottom of the cul¬ ture plate to a confluency of 75% to 90% over a time period of two to three days. At the beginning of the transport experiment, the culture medium is discarded by tipping the medium from the assay plate and a new incubation medium is added to the wells by using the cover described above. iny number of different solutions can be placed into the

- 68 -

enclosures of an upturned cover prior to the commencement of the experiment. The solution may be kept at the required temperature by placing the inverted cover in a water bath at a predetermined temperature. A few seconds prior to the required time, the culture plate containing cells that have adhered to the base of the flat bottomed wells, is inverted and fitted over the corresponding solutions in the enclosures defined by the cover (see FIGS. 69 and 73). At the required time the entire assembly is then inverted and the solutions are transferred to the corresponding wells without any spillage from one well to another (see FIGS 70 and 74). The transfer is instantaneous for all the wells, which is both desirable in the sense of reducing the time to perform experiments, but also improving the accuracy of the experiments.

As can be seen from FIGS. 49 and 50, the cross section of the enclosures 13 in the cover 10 are considerably smaller than the cross section of the flat bottomed wells 22 in the assay plate 20. While it is not essential that the cross section of the enclosures be considerably smaller, it is advantageous because it reduces the likelihood of spilling or sloshing of the contents of the cover when the transfer takes place. Alternatively, the cross section of the enclosures can be virtually identical with the cross section of the wells.

Although not shown in the drawings, a second embodiment embraces a ninety six enclosure cover for use with a ninety six well assay plate. In this embodiment the cross section of the enclosures can be smaller or virtually identical with the cross section of the wells, and the cover is arranged to be a close abutting fit on the assay plate to again reduce the likelihood of spillage. Suitable clamping means such as clips may be used to improve the close fit between the cover and the plate.

The six well cover is designed so that the enclo¬ sures accommodate between 1 and 3 ml of solution, although the design can be modified to accommodate more solution. It

is important that the mouths of the enclosures project from the underside of the lid by at least about 5 mm to thereby protrude into the well cavity. The ninety six well cover can be designed with enclosures that accommodate between 100 and 300 μl or more of solution. Both the six well and ninety six well cover are designed with flat bottomed enclosures to ensure that the cover is free standing when inverted on a flat surface.

9. Results Glutamine transport into HeLa cells was concentra¬ tion dependent and saturable. The uptake rates are expressed per mg protein which reflects the number of viable cells present in a harvested sample as opposed to cell number. FIG. 2 shows that the uptake of glutamine into HeLa cells is divided between two high affinity transport systems (trans¬ porters I and II) and a low affinity system (transporter III). Uptake of glutamine below 0.2 mM (transporter I) showed a small degree of cooperativity whilst uptake via transporter II obeyed Michaelis-Menten kinetics between the concentration range of 0.2 mM - 1 mM. Transporter III exhibited cooperative kinetics for 1.5 - 3.0 mM and the nonsaturable component was 3.3 nmol/min/mg protein, which was deducted from all velocity determinations. Full kinetic analysis of glutamine uptake into HeLa cells (see Table 2 below) revealed that although the K for transporter I is low at 0.03 ± 0.0015 mM, the Vma„x is also very ^• " low at 6.8 ± 0.035 nmol/min/mg protein and the contribution of this transporter to glutamine uptake is relatively insignificant. The K for transporter II is 0.15 ± 0.075 mM with a _maχ of 25 ± 2 nmol/ min/mg protein over the substrate concentration of 0.2 -1.0 mM. Over the glutamine concentration range of 1.0 - 3.0 mM transp ^c orter III has a Km of 1.8 + 0.10 mM with a V ^" max of 22 ±

1.1 nmol/min/mg protein. The specificity of each of the transporters for their substrate was calculated. The specificity of transporter II was very high at 167 ml/min/mg protein whereas the specificity of transporter III is very low, 12.2 ml/min/mg protein. As expected, the specificity of EET

transporter I is extremely high at 227 ml/min/mg protein indicating that this transporter is highly specific for the substrate glutamine.

Table 2 Kinetic parameters of glutamine uptake into HeLa cells Trans- Vmax porter Km (mM) fnmol/min/mσ protein ) Hill Coefficient

1 . 0 ± 0..05 2. 0 ± 0. 2 6. 3 ± 0. 31

In addition the their kinetic values, the following information is also known about TGT I-III. TGT I mainly responsible for transporting when the glutamine concentration is below 0.2 mM obeys Michaelis-Menton kinetics sodium dependent slight acceptance of lithium ions high degree of stereospecificity accepts N-methyl amino acids not inhibited by 2-amino-2-norbornane- carboxylic acid does not undergo adaptive regulation - stimulated trans effect probably amino acid transporting system A.

TGT II mainly responsible for transporting when the glutamine concentration is between 2.0 and 1.0 mM co-operative carrier sodium dependent does not accept lithium ions high degree of stereospecificity does not accept N-methyl amino acids inhibited by 2-amino-2-norbornanecarboxylic acid

SUBSTITUTE SHEET

does not undergo adaptive regulation stimulated trans effect inhibited by mersalyl is an integral membrane glucoprotein which contains disulphide bridges relative molecular weight of dissociated protein is 42000 probably amino acid transporting system ASC.

TGT III mainly responsible for transporting when the glutamine concentration is above 1.0 mM co-operative carrier sodium independent does not accept lithium ions high degree of stereospecificity does not accept N-methyl amino acids inhibited by 2-amino-2-norbornanecarboxylic acid does not undergo adaptive regulation stimulated trans effect inhibited by mersalyl hydrophobic membrane protein probably amino acid transporting system L. relative molecular weight of dissociated protein is 53000 Table 4 (FIG. 75) classifies glutamine transporters according to Na -dependency, the velocity of uptake at 0.75 . mM glutamine concentration, the nature of the major known

» inhibitor, and the capacity to be derepressed in glutamine deficient medium.

Titration of glutamine transport at 1 mM glutamine with NEM ranging from 0.25 to 10.0 mM is shown in FIG. 3. The pre-incubation time used was twenty minutes which was found to be the optimum incubation time for maximum effect of 1 mM NEM on glutamine transport at 1 mM, as shown in FIG. 4.

The effect of the sulphydryl reagent N- ethylmaleimide on glutamine transport into HeLa cells over

the concentration range of 0.25 - 4.0 mM glutamine is shown in Fig 5. After twenty minutes pre-incubation with NEM the monolayers were exposed to radioactive glutamine and uptake measured for 1 minute. Uptake via all transporters I, II, and III was inhibited by 1 mM NEM. The inhibitory action of NEM was found to be by direct competition with glutamine for a site on the carrier by including a large excess of the substrate in the pre-incubation mixture which should protect the carrier from binding NEM. Table 3 shows that an excess of L-glutamine partially protected glutamine transport at 1 mM from inhibition by 1 mM NEM. D-glutamine however exerted no protective effect.

Table 3 Substrate protection of the inhibition of glutamine uptake by NEM

INCUBATION CONDITIONS GLUTAMINE UPTAKE (% of Control )

Control 100

NEM inhibited (1 mM) 53

Protected with L-glutamine (100 mM) 75

Protected with D-σlutamine (100 mM) 53

It is recognized that the isolated receptors (e.g., M 42000) may be subunits. The conditions used to isolate the transporters, particularly using SDS (which breaks non- covalent bonds) and, in combination with 2 mecaptoethanol (which breaks disulphide bonds), has the capacity of reducing large functional molecules to their lowest molecular weight. In this light, it is possible that the transporter exists as many subunits of size 42000 bonded either by disulphide bonds or by non-covalent forces. It may also be bonded to other smaller or larger molecules that are undetected in this system. The combination of such subunits, which together make up the function of the glutamine transporter, is also encompassed by this invention.

C. Identification and Isolation of Other Human 2-mino Acid Transporters

_# Other human amino acid transporters can be isolated in a manner similar to that described above for tumour glu- ta ine transporters. Of particular interest are the trans-

^* . porters which are responsible for the uptake of amino acids into tumour cells, especially if such transporters are not present or are not active in non-tumour cells. For example, an alanine transporter found in HeLa cells was also isolated using the following procedure.

1. Identification and isolation of alanine transporter in HeLa cells

Alanine uptake into HeLa cells was measured over the concentration range of 1.0 - 10.0 mM alanine and appeared to obey Michaelis-Menten type kinetics over this range as shown in FIG. 61. In FIG. 61, HeLa cells were exposed to ¹⁴ C-alanine for one minute and the amount of label appear¬ ing in the cytoplasm was quantitated in terms of total cell protein. Lineweaver-Burk analysis of alanine uptake is shown in FIG. 62, revealing the Km for alanine uptake was 1.33 mM with a Vmax of 19.6 nmol/min/mg protein. In FIG. 62, HeLa cells were pre-incubated with 1 mM NEM or without NEM prior to radioactive incubation. The rate of diffusion of alanine into these cells was 0.625 nmol/min/mg protein and was sub¬ tracted from the saturable component for kinetic analysis.

The titration of alanine transport at 1 mM alanine with NEM concentrations ranging from 0.25 mM - 10.0 mM is « shown in FIG. 63. In FIG. 63, HeLa cells were pre-incubated

» with varying NEM concentrations prior to exposure to ¹⁴ C- alanine to measure the rate of alanine uptake at 1 mM. Inhibition of alanine transport reached a maximum after 2 mM NEM where the rate of alanine transport was 40% of the con¬ trol value. The effect of 1 mM NEM on alanine transport into HeLa cells that were pre-incubated for 15 minutes with 1 mM NEM is shown in FIG. 64. The mode of action of the sulphy- dryl reagent was shown to be by direct competition with alanine for a site on the carrier by including a large excess

SUBSTITUTE SHEET

of substrate in the pre-incubation mixture. Table 14 (FIG. 85) shows that 100 mM alanine partially protected the inhibi¬ tion of alanine uptake by 1 mM NEM while the non-specific stereoisomer D-alanine failed to protect the carrier from NEM inhibiton.

Lineweaver-Burk analysis of the inhibition of alanine uptake by NEM (shown in FIG. 62) indicated that, although NEM is an irreversible inhibitor, its mode of action appears to be manifested in a lowering of the K suggesting it is a competitive inhibitor.

Identification of the alanine transport protein of HeLa cells was based on the same principles used for identi¬ fication of the glutamine transport proteins. The use of NEM to label the alanine carrier was a likely possibility because alanine transport has been reported to be highly sensitive to NEM inhibition. HeLa cells plasma membranes were incubated with ³ H-NEM in the presence or absence of excess transport substrate alanine and the membrane proteins were resolved by SDS gel electrophoresis. A number of proteins were found to bind ^~ H-NEM ^" as shown in FIG. 27, but only peaks C and D were found to be protected by the presence of L-alanine. In FIG. 27, HeLa cell plasma membranes were pre-incubated with 100 mM alanine, or without alanine, prior to and during incubation with ³ H-NEM. The non-transported stereoisomer D-alanine failed to protect either of these peaks from binding the label.

Plasma membranes which had been pre-incubated with excess alanine, followed by incubation with ¹⁴ C-NEM, were solubilized in SDS and resolved by SDS-gel electrophoresis. The gel was then exposed to X-ray film and the autoradiogram demonstrated that pre-incubation with transport substrate L- alanine resulted in specific radiolabelling of a band with M 31000 (lanes 3 and 4), whereas pre-incubation with D-alanine did not.

The binding of increasing concentrations of ³ H-NEM to these peaks was titrated with the'inhibition of alanine transport at the same levels of NEM. A one to one

correlation was found to exist for the binding of ³ H-NEM to peak D with alanine transport at the various NEM concentra¬ tions as shown in Fig 28. The rate of uptake at 100% sat¬ uration of the carrier with ³ H-NEM was assumed to be equal to the diffusion component of alanine uptake and was deducted from the uptake rates measured at each NEM concentration to yield the titration graph for peak D alone in FIG. 29. There was good agreement between the theoretical line for a 1:1 correlation between NEM binding to the protein with M 31000 and inhibition of alanine uptake indicating that this protein was the carrier responsible for the transport of alanine into HeLa cells.

2. Results

Alanine transport into HeLa cells appears to have distinct kinetic characteristics which distinguishes it from glutamine transport in these cells. The uptake curve con¬ forms to simple Michaelis-Menten kinetics, reaching maximum velocity at 15 nmol/min/mg protein with a Km of 5.0 mM. The presence of 1 mM NEM in the pre-incubation medium resulted in a 50% inhibition of alanine uptake at 1 mM alanine. Correla¬ tion of transport inhibition with binding of NEM to membrane proteins as well as substrate protection of binding indicated that a protein band of M 31000 is involved in alanine trans¬ port into HeLa cells. The protein with M 31000 could likely be isolated in the same manner as described above for TGT II. Likewise, the discussions in this specification relating to preparation of antibodies, recombinant DNA cloning methods, diagnostic and therapeutic products, compounds and methods, etc., are also applicable to the alanine transporter, or to any other transporters in accordance with this invention. Further, upon reading this specification, those skilled in the art will recognize other methods for isolating human amino acid transporters, and in particular, tumour amino acid transporters.

While the identification and isolation of amino acid transporters by using NEM has been described, those skilled in the art will appreciated that other reagents could

be used. For example, for transport systems that are sensi¬ tive to NEM, thereby indicating the presence of thiol groups at the active site, other thiol-reagents such as mersalyl, azaserine, DTNB, and organomercurials could likely also be used. For transport systems that are not sensitive to NEM, the particular amino acid that is transported could possibly be labelled and used to identify the transporter.

D. Identification of Transporters In Other Cell Lines

Having isolated a human amino acid transporter, it may be desirable to determine whether that same or related transporters are present in other types of cells. For example, in the case of tumour glutamine transporter isolated from HeLa cells, it would be desirable to know whether the same transporter is found in other tumour cells. As will be discussed more fully below, it has been discovered that TGT II is present in other solid-type tumour cell lines and lymphoma cell lines. The obvious benefit rom this knowledge is that it will now be possible to diagnose and selectively treat many different types of cancer using the same diagnos¬ tic and therapeutic compositions.

1. Screening of cell lines with anti-serum

The following experimental procedure was used to determine whether antibodies raised against TGT II would react with proteins present in other solid-type tumour cell lines, lymphoma cell lines, leukaemia cell lines and normal human cell lines. a. Antiserum production in mice

Male Balb/c mice were used to raise antiserum by injecting 500 μl of antigen (250 μl purified glutamine transport protein (250 μg/ml) and 250 μl of Freunds com¬ plete adjuvant) intramuscularly. The mice were boosted every month with antigen made up in Freunds incomplete adjuvant for three months before being bled through the tail to collect serum. The blood was allowed to clot overnight at room tem¬ perature and the serum was transferred to a microfuge tube

and centrifuged at 1200 g (Eppendorf ultracentrifuge) for 10 seconds to remove red blood cells. The antiserum was stored at -20°C. Control serum from uninjected mice was also obtained. b. Antiserum producting in rabbits Male rabbits were used to raise antiserum by injecting 2 ml of antigen (250 μg/ml mixed 1:1 with Freund's complete adjuvant) intramuscularly followed by booster shots subcutaneously every two weeks using Freund's incomplete adjuvant over a period of three months. Blood was collected by bleeding the rabbits through the peripheral ear vein and allowing the blood to clot overnight at 4°C. The serum was collected and centrifuged at 400 g for 10 minutes to remove red blood cells. The serum was stored at -20°C in 500 μl aliquots. Control serum was collected from the rabbits prior to the first injection. c. Standardization for ELISA screening of antiserum

A 96-well microtitre plate (Flow Laboratories) was coated with 50 μl of a doubling dilution series of antigen (50 μg/ml) solutions in coating buffer (35 mM NaHC0 ₃ , 15 mM Na ² C0 ³ , pH 9.6) into the first five rows of 12 wells. PBS A (50 μl ) was added to the wells of the final three rows. The plate was incubated for 2 hours at room temperature (20°C). The antigen/PBS solution was flicked out of the wells and the plate was washed three times in washing buffer (0.05% (v/v) Tween 20 in PBS A). The plates were dried by v slapping them face down on wads of paper towels. The wells

» were blocked by incubating with 0.2% (w/v) casein in TEN buffer (0.5 M Tris-HCl, 0.01 M EDTA, 1.5 M NaCl, pH 9.0) for 2 hours at room temperature followed by washing three times in washing buffer as before. A series of doubling dilutions of the antiserum starting at 1:10 was made in TEN buffer containing 0.2% (w/v) casein and 100 μl of the antiserum dilutions was added to 11 columns of 8 wells. Control antiserum was added to the wells of column 12 and the plate was incubated for 1 hour at room temperature. The antiserum

SUBSTITUTE SHEET

solutions were flicked out of the wells and the plate was washed three times in washing buffer as before. A 1/1000 dilution of the anti-mouse/rabbit horseradish peroxidase IgG conjugate (Bio-Rad) was made in TEN buffer with 0.2% (w/v) casein and 100 μl was added to each well of the plate which was incubated for 1 hour at room temperature. The plate was washed three times in washing buffer as before and 100 μi of substrate was added to each well. The substrate was 1 mM JU3TS dissolved in 0.O5 M sodium citrate, 0.15 M sodium phosphate containing 2% H ₂ 0 ₃ . The colour reaction was quantified by reading the difference in absorbance at 414 nm and 492 nm of each well on a Flow plate reader. d. Screening antiserum

Once the optimum conditions for the ELISA assay were obtained from the standardization, a screening assay was developed for further screening of the antisera. In general, plates were coated with a 1/50 dilution of antigen (50 g/ml) and probed with a 1/100 dilution of antiserum. The conjugate dilution used was 1/1000 and plates were read as described above. e. Western blotting

Because there was a low antibody titer in the antiserum, the more sensitive Western blotting assay was used to confirm an immunoreaction. A Bio-Rad electrophoretic transfer cell was used to transfer proteins onto nitrocel¬ lulose. (See Towbin et al., Proc. Nat. Acad. Sci. U.S.A., 76, (1979) 4350.) Following electrophoresis, gels were soaked in transfer buffer (0.025 M tris-HCl, 0.192 M Glycine, 20% (v/v) methanol) for 30 minutes at room temperature. Nitrocellulose cut to size was wetted slowly in transfer buffer and a sandwich was made with the gel at the cathode and the nitrocellulose on the anode side. Transfer was carried out at a constant voltage of 60 volts for 3 hours. The nitrocellulose was removed and placed onto a piece of filter paper wetted with tris-buffered saline (10 mM tris- HCl,150 mM NaCl, pH 8.0) containing 0.05% (v/v) Tween 20 (TBST) and 4% (v/v) FBS and incubated overnight at 4°C. The

nitrocellulose was then transferred to a solution of antiserum (1/100 dilution in TBST) and incubated for 1 hour at room temperature followed by washing three times in TBST. A 1/10000 dilution of anti-mouse/rabbit alkaline phosphatase IgG conjugate (Promega) was made in TBST and incubated with the nitrocellulose for 30 minutes at room temperature. The nitrocellulose blot was washed three times in TBST followed by addition of the substrates (0.37 mM BCIP, 0.5 mM NBT) dissolved in alkaline phosphatase buffer (100 mM tris-HCl, 100 mM NaCl, 5 mM MgCl ₂ , pH 9.5). Colour development was stopped by washing the blot with Stop solution (20 mM tris- HCl, 5 mM EDTA, pH 9.5) when bands appeared dark enough. f. Screening of cell lines

(i) Plasma Membrane Preparation Plasma membranes from adhesive cell lines were prepared as described above for HeLa cell plasma membranes. Suspension cultures were grown to a confluence of approxi¬ mately 10 ⁸ cells/ml and then a plasma membrane fraction was prepared by sucrose gradient centrifugation (see Segal et al. , J. Cell. Phvsiol., 100, (1979) 109-118). All manipula¬ tions were carried out at 4°C where the cells were first pelleted at 200 g for 5 minutes (Clements 2000 benchtop centrifuge) and then washed two times with 0.9% (w/v) sodium chloride. The pellet was resuspended in 10 ml of lysis solution (1 mM NaHC0 ₃ 0.5 mM CaCl ₂ , pH 7.4) and homog¬ enized with 30 strokes of a tight fitting Dounce homogenizer (maximum setting) . The cell homogenate was centrifuged at 500 g for 20 minutes in a swinging arm rotor (SW 25.1) of a Beckman centrifuge and the supernatant was collected and kept. The pellet was re-homogenized in another 10 ml of lysis solution and pelleted as described above. The super¬ natants were pooled and then sedimented at 12800 g for 20 minutes. The membrane pellet was resuspended in 5 ml of lysis buffer and mixed with 15 ml of 40% (w/v) sucrose (made up in lysis solution) to make a 30% solution. A 20 ml syringe fitted with a long piece of tubing was used to care¬ fully layer 15 ml of 40% (w/v) sucrose underneath the 30%

SUBSTITUTE SHEET

sucrose membrane solution in the centrifuge tube. The tube was then centrifuged at 54450 g for 4 hours after which time a white membrane layer was visible at the interface of the gradient. The membrane layer was collected with a pipette and sedimented at 45000 g for 1 hour. The pellet was resus¬ pended in 10 mM tris-HCl, pH 7.4 and stored at -20°C.

(ii) Screening procedure

Purified plasma membranes from a variety of cell lines (Table 9 (FIG. 80)) were solubilized in SDS-solubili- zation buffer and applied to separated wells of a 12% (w/v) polyacrylamide gel. The separated proteins were blotted onto nitrocellulose and probed for reaction with glutamine trans¬ porter anti-serum as described above.

2. Measurement of glutamine uptake in other cell lines

The rate of glutamine uptake into the other cell lines was also determined using the following procedure. a. Glutami e uptake procedure in other tissues and analogue inhibition experiments

A new procedure for measuring uptake into adhesive cells was developed for investigating the effects of glu¬ tamine analogues on the rate of glutamine uptake into HeLa cells and for measuring the rate of glutamine uptake into other adhesive cell types. A separate procedure is described for measuring glutamine uptake into suspension cultures.

(i) Adhesive Cells

Adhesive cells were grown in 6-well culture plates (Flow) until confluence was reached. All incubations were carried out at 37°C and each new solution was transferred into the wells using a modified culture plate lid (described above) which had 6 x 4 ml scintillation vials inserted into the lid directly above each well. Normal procedure was to pour out the old solution from the culture wells, invert the plate and clamp it to the lid with the new solution in place in the scintillation vials. The clamped assembly is then inverted, and the culture wells are simultaneously supplied EET

with the new solution. Two hours before each experiment the cell cultures were incubated with fresh medium 199 containing 10% (v/v) FBS. The medium was then exchanged for unlabelled 2 mM glutamine in medium 199 and incubated for 20 minutes. This medium was removed and replaced with glutamine in HBSS at the test concentration (0.25 - 5.0 mM) for three minutes followed by ¹⁴ C-glutamine (62.5 μCi/ml) at the test con¬ centration in HBSS for exactly 1 minute. Uptake was stopped by washing the wells three times with ice-cold 0.9% NaCl. Soluble amino acids were removed by incubation with 0.5 ml of 5% (w/v) trichloroacetic acid and counted for radioactivity on the scintillation counter. The remaining protein was solubilized in 1 M NaOH for 120 hours and the amount of protein present was determined. Results were expressed as nmoles of glutamine incorporated per minute per mg protein.

(ii) Suspension cultures Uptake experiments in suspension cultures were carried out using the methods described by Bradford and McGivan, Biochimica et Biophvsica Acta, 698, (1981) 55-56. The cells were pre-loaded with 2 mM glutamine for 20 minutes followed by pre-incubation with glutamine at the test con¬ centration for 3 minutes. Radioactive glutamine (0.5 μCi/ ml) was then added to the cell suspension for exactly 1 minute after which time uptake was stopped by spinning the cells through silicone oil.

In a better procedure for measuring glutamine uptake in suspension cultures, cells were resuspended in HBS Ϊ containing tritiated water (final concentration of 5 nCi/ml)

* and 2 mM glutamine for approximately 20 minutes. At time = 0, 200 μl aliquots of the cell suspension were added to 1.5 ml microfuge tubes containing an appropriate concentration of glutamine solution (in HBSS) to bring the 2 mM concentration to the test concentration. When time = 3 minutes, 200 μl of - *C-glutamine was added to the microfuge tubes and during the next minute, 100 μl aliquots of this was added to 300 μl microfuge tubes prewarmed to 37°C and containing 20 μl of }2% (v/v) perchloric acid above which was layered

SUBSTITUTE SHEET

50 _μl of silicone oil. At time = 4 minutes, the 300 μl microfuge tubes were spun for a minute and then placed in the freezer overnight.

The bottoms of the microfuge tubes were amputated into scintillation vials which contained 200 μl of PBS A and were shken vigorously. Scintillation fluid was added, the vials were again shaken, and then counted for radio¬ activity on a Liquid Scintillation Counter. Inhibitor/ analogue was added during the 20 minute incubation with the HBSS/2 mM glutamine for transport inhibition experiments.

3. Results The rate of glutamine uptake into a variety of cell lines was investigated over the concentration range of 0.25 - 4.0 mM glutamine. The uptake curves for the different cell lines are given in FIGS. 7a-g. A summary of the results for uptake rates at 0.75 mM glutamine (physiological concentra¬ tion ₎ is given in Table 6 (FIG. 77) with the inclusion of the results from the literature for a variety of tissues. All of the solid-type tumour cell lines investigated had high uptake rates. Most of the normal and lymphocyte-derived cell lines (both normal and cancer) exhibited low uptake rates with the exception of the normal cell line MRC5 and the lymphoblastic leukaemic cell line MOLT 3.

Table 9 (FIG. 80) lists the immunoblot analysis of the cell lines with either mouse or rabbit anti-serum raised against the glutamine transport protein of HeLa cells. A positive antibody response was indicated by a coloured band appearing on the blot in a position analogous the molecular weight of 42000. A representative immunoblot is given in FIG. 26 showing both positive and negative results. A numbe of other bands reacted non-specifically with the anti-serum which suggests some degree of cross reactivity was occurring Table 10 (FIG. 81) lists the cell lines with their response to immunoblot analysis with either mouse or rabbit anti-seru as well as their glutamine uptake rates at 0.75 mM. Only cell lines with complete data are listed.

^C omparison of the results given in Ta _b le 9 _(F I _G . 80 ⁾ for the specificity of a variety of cell types for the glutamine transport protein with the results in Table 6 ₍ FIG 77) for glutamine uptake rates in the various cell lines reveals a correlation between antibody specificity an _d glu _¬ tamine uptake for the cell lines, as shown in Table 10 ₍ FIG. 81 ⁾ . The cell lines which gave a negative response to antiserum also had low rates of glutamine uptake. In a two- way contingency test on the results in Table 10 (FIG. 81), the hypothesis of independence or lack of antibody speci _¬ ficity for tumour cells was rejected at the 0.005 probability level.

One purpose of these studies was to determine whether a clear correlation existed between high rates of glutamine uptake and malignancy. Table 6 (FIG. 77) shows that all of the 6 solid tumour type tissues had rates of glutamine uptake above 20 nmol/min/mg protein. The Burkitt lymphomas which all grew in suspended cultures seem to occupy a position which is intermediate between solid tumours and the leukemia/normal cell group in that two out of the three produced a positive antigen-antibody reaction against the HeLa cell glutamine transporter antibody (Tables 9 ₍ FIG. 80 ₎ and 10 (FIG 81)), and all of them had low rates of glutamine uptake. One of the normal cultures produced a high rate of glutamine uptake (MRC5, Table 9 (FIG. 80)). This is ascribed to one of the natural hazards of this type of work, in that normal cells have rates of growth in vivo which are much slower than those expected in vitro,, and the transformation from slow growth and limited life of the culture to fast growth and extended life, for obvious reasons, more than likely involves a switching up of glutamine transporter activity.

As a consequence of the above findings, it is sug _¬ gested that solid tumours have a depressed glutamine uptake because of the fact that their blood supply is usually inadequate, and they therefore need to take up all the glu _¬ tamine that is available to them. On the other hand, cells

which normally exist in the blood or the lymph have an ample supply of readily available glutamine and therefore the derepression mechanism may not be switched on. Among the 3 lymphoblastic leukemias and the 3 lymphomas for which information is complete (Table 10 (FIG. 81)), the 3 leukemias all gave a negative reaction to the transporter antibody, while one of them had a high rate of glutamine uptake (the only example so far of this relationship); and among the lymphomas, two gave a positive response to the transporter antibody while one did not. All three lymphomas had low rates of glutamine uptake. The Bordin Epstein-Barr trans¬ formed lymphocytes were placed with the lymphomas, as being the closest culture equivalent. Thus, one of the lymphomas demonstrated the reverse of the situation with one of the leukemias.

These results suggest that there may be two or more steps in the transformation of cells to the full solid tumour state, which may account for the two intermediate states dis¬ cussed above, one being due to a partial switching up and the other to a partial switching down of the glutamine uptake mechanism. Student's t test on the data in the Table 6 (FIG. 77) for significance in the mean difference between the 6 solid type tumour cells (surface adhering cells as against cells which grow in suspension) and the 15 other tissues gave a mean for the tumours of 39.4 nmol/min/mg protein with s = ±14.8, and the other tissues gave a mean of 8.0 nmol/min/mg protein with s = ±8.0. This gave t = 6.0, which for 19 degrees of freedom was well beyond the 0.005 level of prob¬ ability for no difference between the two means.

The possible use of polyclonal or monoclonal anti¬ bodies in cancer therapy and diagnosis relies on the differ¬ ent antigenicity of the glutamine transport protein in normal and tumour cells. It can be seen from Table 9 (FIG. 80) that the 6 solid tumour type cells were all positive, that two of the three Burkitt lymphomas were positive, the five leukemias were all negative, and the four normal type cells were all negative. The uniformly positive response of the solid-type

tumours when probed with mouse or rabbit antiserum raised against the purified glutamine transporter protein suggests that the clinical use of the glutamine transporter antibodies would be effective against solid-type tumours and at least many lymphomas. Table 10 (FIG. 81) shows that the antibody reaction results correlate excellently with high rates of glutamine uptake, reinforcing the suggestion made above that the derepressed glutamine transporter is a likely specific marker for solid tumours. III. ANTIBODIES TO A HUMAN J^MINO ACID TRANSPORTER

Other embodiments of this invention comprise mono¬ clonal and polyclonal antibodies which recognize human amino acid transporters in accordance with this invention. The procedure for preparing polyclonal antibodies to TGT II (dis¬ cussed above) can also be followed for preparing polyclonal antibodies to other human amino acid transporters.

In addition to polyclonal antibodies, it may also be desirable to have monoclonal antibodies that are specific to epitopes on a human amino acid transporter. For example, because TGT II is common to other tumour cells, monoclonal antibodies to epitopes of TGT II may be valuable for use in diagnosing the presence of cancer. Monoclonal antibodies can also be used in recombinant cloning procedures to isolate the gene which codes for TGT II. Monoclonals for any transporter can be prepared using a procedure similar to the following procedure for preparing monoclonal antibodies which recognize TGT.

1. Mice, rats or other animals are immunized against the glutamine transport protein using, for example, either purified dissociated TGT, purified native TGT, a preparation of crude membrane proteins in native form, or peptides made from the amino acid sequence of the transporter.

2. The spleen of an immunized mouse is removed and fused with cultured myeloma cells in fusion experiments.

3. The fused cells are seeded into culture dishes under conditions which select for hybridoma cell growth.

4. Once the hybridoma colonies have begun growing, screening of the hybridomas is commenced.

Several types of screening procedures may be used to obtain monoclonal antibodies which bind to the tumor glutamine transporter. One screening procedure is an ELISA assay comprising the following steps:

1. Plates (e.g., 96 well) are coated with purified transport protein.

2. Superhatant from the hybridoma colonies is added to the coated wells (positive antibodies will bind to the corresponding epitope on the transport protein) .

3. Excess antibody is washed away.

4. IgG conjugate (anti-mouse antibody which is conjugated to an enzyme or other reporter molecule) is added to the wells.

5. Excess conjugate is washed away.

6. The substrates for the enzyme are added and a colour reaction is seen in wells which were incubated with supernatant from positive clones.

7. Positive clones are grown in large numbers by culturing.

8. Positive clones may be grown in bulk by injecting into ascites fluid of mice.

However, due to the unique nature of TGT, the hybridomas can be screened using a glutamine transport inhibition assay. Monoclonal antibodies which bind to the transporter will inhibit transport of glutamine and, in this way, can also be identified. The screening procedure comprises the following steps:

1. Plates (e.g., 96 well) are inocculated with HeLa cells which are grown until 75% confluent.

2. The HeLa cells are pre-loaded with a high concentration of glutamine (2-3 mM) .

3. Supernatant from the growing clones is added to the wells (positive monoclonal antibodies will bind at the active site on the transport protein exposed on the outer membrane surface) .

4. Excess antibody is washed out and the cells are incubated with the test concentration of glutamine containing ^{~ 4} C-glutamine for an exact period of time.

5. Radioactive uptake of glutamine is stopped by washing with ice-cold 0.9% (w/v) NaCl.

6. The amount of label present within the cytoplasm is quantitated by radioactive counting (the soluble components are extracted using trichloracetic acid) .

7. The total amount of protein present is estimated by Lowry analysis and the results are expressed as nmoles glutamine incorporated per minute per mg protein.

8. The results for each clone tested are compared with controls which used control serum (lacking glutamine transport protein antibodies) and samples which show greater than 20% inhibition of glutamine uptake relative to controls are considered to be positive.

9. Positive clones are grown in large numbers either by culturing or by injecting into ascites fluid of mice.

It is also possible to produce antibodies to regions of the glutamine transporter which are distant from the glutamine binding sites. While such antibodies are equally valuable in diagnosing the presence of tumour cells, they cannot be screened for by the inhibition assay described above. Rather, they can be identified in other ways. First, if the receptor has been isolated, then antibodies will bind but will not block the binding of glutamine. Second, they could be examined by using cells which are high in the glu¬ tamine receptor, and the results compared with other cells which have absent or low amounts of receptor. A differential reactivity will indicate the likelihood that the antibodies are to the glutamine receptor.

The search for such monoclonal antibodies either to receptor molecules themselves or to the binding site on such molecules is common and usually a variety of diagnostic tests are used to analyze to what molecules the antibody bind and to which part of these molecules. For the most part, these

tests involve straightforward serological assays such as cytotoxicity or flow cytometry using differential reactions to indicate the likely molecules to which the cells bind. Such studies can then be followed up, for example, by immuno precipitation using radiolabelled cell membranes or by per¬ forming western (immunoblot) blot type of analyses when the molecular species can be identified. These tests identify the molecules with which the antibodies bind but do not indicate which part of the molecule is reactive. Follow-up tests can be done wherein the binding of the radiolabelled glutamine to the receptor molecules is blocked by antibody. As stated above these tests should be straightforward and ar quite standard in the analysis of antibodies binding to functional molecules.

Obviously, the antibodies that are obtained will depend upon the protein that is used to immunize the mouse o other animal. For example, if TGT II is obtained from SDS- PAGE, the protein will be in its "dissociated" or "denatured form. Antibodies which depend upon the three dimensional configuration of the transporter protein will not be formed. Thus, in order to obtain monoclonal (and polyclonal) anti¬ bodies which depend upon the three dimensional configuratio of, for example, TGT II, an animal must be immunized with t native form of the protein. This can be accomplished in at least two ways.

First, the animal can be immunized against portio of the tumour cell membranes which contain TGT II. Mono¬ clonal antibodies to the native form can then be screened f in the same manner as described above. Alternatively, the animal can be immunized with purified native protein obtain from recombinant methods as described below.

In short, the type of monoclonal antibody obtaine will depend not only on the amino acid transporter used, bu also whether the transporter is in its native or dissociate form. Similarly, the antibodies obtained may depend upon whether the entire protein or only a fragment or subunit is used.

Monoclonal antibodies can be produced to the glutamine transporter, and these can be produced in mice, rats, other species and eventually in humans. All of these varieties of monoclonal antibodies are within the scope of this invention if they can recognize parts of the glutamine transporter or subunits thereof. Additionally, it is now recognized that antibodies can be used whole, in part, (e.g., Fab'2 or Fab), or, indeed, single chain antibodies are now being described wherein the specificity is due to the reten¬ tion of the hypervariable region. Additionally, such mono¬ clonal antibodies can be genetically engineered so that segments of the sequence from one species can be transferred to the genes of the same or another species to provide hybrid chimeric molecules. All such antibodies which have the capacity of binding to the glutamine transporter (in part or in whole) are within the scope of this invention. IV. PRODUCTION OF TRANSPORTER USING RECOMBINANT METHODS

The N-terminal and partial amino acid sequence of the transporter (or its subunit) can be obtained, oligonu¬ cleotide probes can be synthesized, and both cDNA and genomic clones can be obtained by standard procedures involving the probing of both cDNA and genomic libraries of different sources and different species. The isolation of the glu¬ tamine transporter described herein will naturally lead to the cloning of both cDNA and genomic clones in one species (human), but should also lead to the isolation of the trans¬ porter in other species using cross-species hybridization. The methods used for gene cloning are numerous and usually several methods are tried which lead to the isolation of the gene. These could include the use of polyclonal antisera or monoclonal antisera, and an expression system; the isolation in cells where the antigens expressed on the cell surface lead to the isolation of cDNA; transfection methods; mes¬ senger RNA selection procedures; identification of restric¬ tion fragment length polymorphisms leading to the cloning of the gene or translocations. These methods are now standard and used in many laboratories throughout the world and need

not be described in detail. An important feature is the nucleotide sequence obtained from the cDNA and in genomic clones and variations thereof due to genetic polymorphism, tissue polymorphism, gene rearrangement or alternative splicing, or any other procedure leading to variations in the structure of the glutamine transporter. jE. coli cells, other bacteria, yeast and higher eucaryotic cells may be made to synthesize foreign proteins, such as the glutamine transport protein, by cloning cDNA into expression vectors. This is achieved by first isolating (total) mRNA from tissue known to be expressing the protein of interest and generating cDNA from the mRNA, using reverse transcriptase to generate the first strand, degrading the RNA template with RNase H, and second strand synthesis with DNA polymerase. The double-stranded cDNAs are then cloned into suitable expression vectors, either plasmids or phage. The cloning sites in these vectors are either (1) in the 5' coding region of a regulated gene, resulting in synthesis of a fusion protein, or (2) insertion immediately downstream of a suitable promoter/ribosome binding site, resulting in synthesis of non-fused protein. Colonies (plasmid vectors) or plaques (phage vectors) can then be screened for produc¬ tion of the protein of interest with a suitable probe, and clones producing the desired protein product isolated and grown in bulk. Thus, one can isolate, for example, several lambda gtll clones which synthesize proteins reacting with polyclonal antibodies raised to the purified transport protein.

A. Cloning Experiments

The availability of cDNA clones capable of express- ing the glutamine transport protein (or parts of it) in prokaryotic systems offers the potential for production of large quantities of the protein. The pure material can be used in diagnostic tests or for therapeutic purposes. Addi¬ tionally, it can be used for raising antibodies (both mono¬ clonal and polyclonal), as well as making the screening of

new chemical compounds (or new monoclonals) for anti- glutamine uptake activity much simpler than at present. As the first step towards production of cloned transport protein, one can, for example, screen a representative HeLa cell cDNA library with polyclonal anti-serum raised to the purified transport protein, and isolate and purify several positive clones.

The following sections describe characteristics of a cDNA library and a screening system that can be employed. As described above, this is only one of many different clon¬ ing methods which can be used in accordance with this inven¬ tion; this invention being explicitly not limited to any one method. Those of ordinary skill in this art will immediately recognize such other methods, which are likewise within the scope of this invention.

1. Characteristics of the HeLa cell cDNA library

Titre: 1-9 x 10 ⁹ pfu/ml

Volume: 0.2 ml mRNA source: HeLa derived D98-AH2 cells (HPRT in phenotype)

Vector: Lambda gtll

Cloning site: EcoRI

Screening criteria (recombinants versus non-recombinants) : clear plaques versus blue plaques.'

Percentage of clones which are recombinants by this criterion: 81%

Number of independent recombinant (i.e., clear) plaques:

1.3xlO ^β

Average insert size: 0.86 kb (range: 0.48 to 3.1 kb).

2. Screening the HeLa cells cDNA library with antibody preparations

The basis of one screening method which can be used is as follows. Cloning cDNAs into the EcoRI site of lambda gtll results not only in insertional inactivation of the (phage-encoded) ^-galactosidase but, if the insert is in the correct orientation and reading frame, the cDNA insert can be expressed as a fusion protein, which can be detected with TIT TE HEET

antibodies specific for the cDNA-encoded protein(s). Phage are induced using isopropyl b-D-thiogalactopyranoside (IPTG) and fusion proteins bound to a nitrocellulose filter lift. The filters are then screened by soaking briefly in dilute primary antibody (in this case the polyclonal antiserum raised to purified transport protein in mouse hosts) , unbound primary antibody is washed off, and then bound primary anti¬ body is detected using a secondary antibody to the primary host, which has been conjugated to an enzyme. In this case, the secondary antibody conjugate can be affinity purified using goat anti-mouse IgG (H+L) alkaline phosphatase conju¬ gate. Phosphatase conjugated secondary antibody bound is then detected using the coloured phosphatase substrate cock¬ tail 5-bromo-4-chloro-3-indolyl phosphate (BCIP) plus nitro blue tetrazolium (NBT) . a. Screening protocol The following steps can be used:

1. Grow plating cells. Streak out E.. coli Y1090r for single colonies on LB agar plates at pH 7.5 containing 100 μg/ml ampicillin.. Incubate at 37°C overnight. Starting with a single colony from that plate, grow Y1090r to satura¬ tion in LB (pH7.5) at 37°C with aeration.

2. Infect the cells with the cDNA library. For screening, one can use 0.2 ml of the overnight bacterial culture (Y1090r) plus 0.1ml of phage diluent (lOmM Tris-HCl, pH7.5, containing lOmM MgCl2 ) which contained 3 x 10 ⁴ pfu of the library per plate. Four plates are sufficient to give several positives using the presently available antibody preparations.

3. Plate cells. To the phage plus bacteria mix, add 3.5 ml (per plate) of LB soft agar, pH 7.5, and pour onto two day old LB agar plates, pH7.5. Incubate for 3.5 hours at 42°C.

4. Overlay nitrocellulose filter. Place plates in a 37°C incubator. Pre-soak a nitrocellulose filter disc in 10 mM (aqueous) isopropyl ø-D-galactopyranoside (IPTG).

Overlay the plates with dried IPTG-treated filters and incubate for 3.5 hours at 37°C.

5. Process filter for screening. Remove plates to room temperature, apply marks to filter to allow subse¬ quent alignment of filter with plate, and carefully remove filter from the agar. Wash the filters in 100 mM Tris-HCl, pH 8.0, containing 150 mM NaCl and 0.05% (v/v) Tween 20 (TBST solution) to remove any adherent agar. To saturate non¬ specific protein binding sites incubate the filters in TBST solution containing 1% bovine serum albumin for 30 minutes, using 7.5 ml of solution for each filter.

6. Pretreatment of primary antibody. Primary antibody solution (ie. mouse anti-transport protein) is treated with E,. coli extract immediately prior to use of the serum for screening nitrocellulose filters. For 1/10000 dilution of antiserum 100 μg/ml of E . coli extract can be used, and the incubation can be for 30 minutes.

7. Incubation in primary antibody. Filters are incubated in TBST containing primary antibody at 1/100 dilu¬ tion for 30 minutes, using 7.5 ml of solution per filter.

8. Wash filters. Filters are washed at least three times in 20ml aliquots of TBST for 10 minutes.

9. Incubation in secondary antibody. Filters are then transferred to fresh TBST containing secondary antibody conjugate (affinity-purified goat anti-mouse IgG (H+L) alka¬ line phosphatase conjugate) diluted 1/7500, and incubated for 30 minutes at room temperature using 7.5 ml per filter.

10. Wash filters. Repeat step 8 above.

11. Incubation with phosphatase substrates. Fil¬ ters are blotted dry on filter paper and then transferred to the colour development solution of substrates. For 5 ml of substrate solutionadd 33 μl of nitro blue tetrazolium (NBT) stock solution (50 mg/ml in 70% dimethylformamide) to 5ml of 100 mM Tris-HCl, pH 9.5, containing 100 mM NaCl and 5 mM MgCla . Mix, then add 16.5 μl of 5-bromo-4-chloro-3- indolyl phosphate (BCIP) stock solution (50 mg/ml in dimethylformamide) and mix again. Incubate for 1-4 hours

away from strong light. Positive clones appear as purple plaques on the .filters.

12. Stop colour development. When plaques are visible and before the background becomes dark, stop the phosphatase reaction by discarding the substrate solution and transferring the filters to 20 mM Tris-HCl, pH 8.0, contain¬ ing 5 mM EDTA. Then store the filters either dry or in this solution. b. Further screening

Primary screening generally results in an average of about 2 positives per plate, i.e., the overall frequency of positives on the primary screen is about 1 per 1.5 x 10 ⁴ plaques. Putative positive clones from the primary screening plates are picked and transferred to secondary screening plates of the host bacterium E . coli Y1090r. Positive plaques are aligned on a grid pattern template, and negative (control) plaques are included on each plate. The plates are treated as in the primary screen (steps 4-12). Approximately 75% of the putative positives from the primary screen are positive on the secondary screen. This purification process is repeated twice; on the tertiary screening plates the fre¬ quency of positives is about 80-100%, and on the quaternary screen all plaques can be positives. Positives from the quaternary screen can be grown in bulk and from them DNA prepared by standard methods (Davis et al., 1986, Basic Methods in Molecular Biology, Elsevier Publishers, New York).

The cloned, native glutamine transporter molecule or fragments thereof can then be used to produce both mono- clonal and polyclonal antibodies.

Alternatively, the dissociated protein obtained from SDS-PAGE could be sequenced to obtain an amino acid sequence of the transporter. From the amino acid sequence, a DNA probe can be prepared to probe the cDNA library for com¬ plimentary strands of DNA. By using cDNA clones or oligonu¬ cleotide probes, or other techniques known to those skilled in the art, genomic clones can be isolated from genomic libraries. Additionally, homologous cDNAs or genes in humans

or other species can be isolated using these reagents. In this way, the gene which codes for production of the native protein can be obtained. Expression is carried out as pre¬ f viously described.

It must be emphasized that there are several different cloning methods which can be tried. If one method, such as that described in detail above, fails, then one skilled in the art will recognize others which may work. Guidance can be found in publications such as: DNA Cloning, A Practical Approach, Vols. I-III, ed. M.D. Glover, I.R.L. Press, Washington, D.C. (1985); Current Protocols In Molecular Biology, Vols. I and II, ed. F.M. Ausubel et al. , Wiley & Sons (1987); and Molecular Cloning, A Laboratory Manual, T.Maniatis et. al. , Cold Spring Harbor Laboratory (1982).

B. Sequence Variation

The nucleotide sequence encoding the transporter can be variable for the reasons described below.

1. The degeneracy of the genetic code nucleotide change does not necessarily bring about a change in the amino acid encoding, e.g., the codon GUU specifies a valine residue as do the codons GUC, GUA, GUG, each being different by a single nucleotide.

2. Two or three nucleotide changes can give rise to the same amino acid, e.g., codons UUA, UUG, CUU, CUC, CUA and CUG all encode Leucine. Codons AGU, UCC, UCU, UCA and UCG encode serine.

3. Changing one or two nucleotides may result in a conservative amino acid change unlikely to greatly affect the function of the protein, e.g., codon UUC specifies leucine and AUU specifies isoleucine. Also UCG specifies tryptophan and UUU specifies phenylalanine - all conservative changes.

4. Allelic variations. Variations in nucleotide sequence and amino acid sequences of the encoded protein as well as resultant may occur between individual members of the same species. These variations arise from changes in the

SUBSTITUTE SHEET

nucleotide sequences encoding the protein. Thus, different forms of the same gene (called alleles) give rise to protein of slightly different amino acid sequence but still have the same function.

5. Variation can occur as the result of dif¬ ferential mRNA splicing where one gene composed of many different segments (exons) of coding sequence - DNA encoding the mature protein - gives rise to a RNA that is spliced such that the portion of the RNA derived from certain exons are removed. Selection of exons may be different in different cell types.

6. Proteins having the same function, e.g., major histocompatibility proteins may arise from related genes. Many protein gene families have been described, e.g., immunoglobulms which have nucleotide and amino acid sequence variation but retain their primary function of antigen bind¬ ing. Such homologous proteins are encoded by homologous genes. These genes arise by duplication of one original gene or by gene conversion.

Variation may be intentionally introduced by:

1. Mutating cloned cDNA or genomic DNA by point mutation, rearrangement or insertion of related or unrelated DNA into the cDNA or genomic clones encoding the functional protein. Such mutated (variant) clones can be used to gen¬ erate variant proteins or peptides which in the context of this specification may have glutamine transport function.

2. Enzymatic cleavage of the protein (from eithe in vitro synthesis or normal cell synthesised protein) with or without repair/rearrangement of the cleavage products.

3. Chemical modification.

4. Irradiation.

The present invention also includes the use of seg ments of the glutamine transporters (peptides) and variant peptides synthesised or genetically engineered. Other trans porters are within and between species that are homologous a the nucleic acid, protein and functional levels. Because of

STIT TE SHEET

substantial sequence homologies, the cDNA clones described herein would enable the isolation of related sequences. V. ANTI-GLUTAMINE COMPOUNDS i-VND GLUTAMINE ANALOGUES Another embodiment of this invention provides anti-glutamine compounds and glutamine analogues suitable fo the treatment of cancer, which have the general Formula (I) :

R ⁴ 0

V W X CH ₂ - C ¹ — C" R ¹ (I)

I

NR ³ R ² or a physiologically acceptable salt thereof, wherein: X is selected from CR ⁵ R ⁶ , NH, NOH and 0; p 0 0 II II

W is selected from C , S , i?

8 Av

R ⁸ R ¹⁰ / / V is -N , C R ^{1 ~}

R- R ¹²

R ¹ is selected from 0CH ₂ Ph, OR ¹³ , NH ₂ and NHNH ₂ , wherein R ¹³ is selected from H and a C_-s sub¬ stituted or unsubstituted, cyclic or acyclic, saturated or unsaturated hydrocarbon group;

R ² is selected from groups defined by R ¹³ , NH ₂ , (C=0)R ¹³ and (C=0)0R ¹³ ;

R ³ is selected from groups defined by R ¹³ ;

R ⁴ is selected from groups defined by R ¹³ , halogen and (C=0)R ^x ;

R ^δ and R ^β can be independently selected from groups defined by halogen and R ¹³ ;

R ⁷ is selected from groups defined by OR ¹⁴ and NR ¹⁵ R ^lβ , wherein R ¹⁴ , R ^lδ and R ^lβ are independently selected from H, C1-5 substituted or unsubstituted, satu¬ rated or unsaturated, cyclic or acyclic hydrocarbon group (and a Cs-β aromatic or heteroaromatic ring which may be substituted or unsubstituted;

R ⁸ is selected from groups defined by R ¹ , OR ¹⁴ , amino, -* i t ^~ n . - ^~ r. _s anri Ύ-Γ. - .

R ⁹ is selected from groups defined by R ¹⁴ ; or R ⁸ and R ⁹ taken together with the nitrogen atom repre¬ sent a 3- to 8-membered, substituted or unsubstituted, saturated or unsaturated heterocycle;

R ¹⁰ is selected from groups defined by R ¹⁴ , OR ¹⁴ , amino and amido; and

R ^{1 ~} and R ^{1 ~} are independently selected from groups defined by R ¹⁴ and amido.

Alternatively, V W X CH ₂ can represent the group;

wherein R ¹⁷ is selected from groups defined by R ¹⁴ , S0 ₂ C1, and (C=0)-R ¹⁴ ; and

R ¹⁸ and R ¹⁹ are independently selected from groups defined by R ¹ * , halogen and R ¹⁴ groups substituted at the carbon α to the 4- membered ring by a leaving group (exemplified by, but not limited to, OMs, OTs and halogen);

In another alternative, V — w — X represents the group;

R ²¹

wherein R ²⁰ is selected from groups defined by R ¹⁴ , OR ¹⁴ and NHR ¹⁴ and (C=0)R ¹⁴ ; or wherein ZR ²¹ is halogen;

Z is selected from 0, S and NR ¹⁴ ;

R ^{2 ~} is selected from groups defined by mesyl, tosyl, (C=0)-R ¹⁴ and a C1-5 substituted or unsubstituted, cyclic or acyclic, saturated or unsaturated hydrocarbon group;

SUBSTITUTE SHEET

or wherein R ² ° and R ² taken together with the group

form a 5- or 6- membered cyclic ring which may be fur¬ ther substituted or unsaturated.

Classes of compounds defined by Formula (I) which are of interest include, but are not limited to compounds of the general Formula (II):

or a physiologically acceptable salt thereof, wherein X is selected from CH ₂ , NH, NOH or 0, and wherein

R ¹ , R ² , R ⁸ , and R ⁹ are as defined above.

Classes of compounds defined by Formula (II) which are of interest include, but are not limited to compounds of the general Formula (III):

or a physiologically acceptable salt thereof, wherein X is CH ₂ , NH, NOH or 0;

R ²³ is H when R ⁸ is selected from groups defined by OH, NHR ¹⁴ , (C=0)NHR ¹⁴ and (C^JNfR ¹ * )0H; R ²³ is OH when R ⁸ is selected from groups defined by (C=0)NHR ¹⁴ , and (C=0)N(R ¹⁴ )0H; or R ³ and R ⁸ are independently selected from groups pre¬ viously defined by R ^{: 4} ; or R ²³ and R ⁸ taken together with the nitrogen atom form a 5-or 6-membered heterocyclic ring which may be saturated or unsaturated, and substituted or unsubstituted.

Classes of compounds defined by Formula (III) which are of interest include, but are not limited to:

A: Compounds of the formula:

SUBSTITUTE SHEET

or physiologically acceptable salts thereof.

wherein R ⁸ and R ⁹ are as previously defined. B: Compounds of the the formula:

or physiologically acceptable salts thereof, wherein R ²⁶ is H, OH or NO, and

R ²⁴ and R ²⁵ are independently selected from H, OR ¹³ SR ¹³ , NR ¹³ ₂ , N0 ₂ , CHO, C0 ₂ H or halogen, wherein R ^1S is as previously defined. C: Compounds of the formula:

or physiologically acceptable salts thereof, wherein R ²⁷ and R ²⁸ are independently selected from H and

OH;

D: Compounds of the formula:

or physiologically acceptable salts thereof, wherein R ²⁹ is selected from:

wherein X is OH, OMs, OTs or halogen, and R ³ ° and R ³ ~ are independently selected from groups as previously defined by R ¹³ .

The term "d- ₅ substituted or unsubstituted, saturated or unsaturated, cyclic or acyclic hydrocarbon group" includes alkyl, alkenyl or alkynyl, inclusive of straight and branched chain groups and cyclic groups of one to five carbon atoms, which may be substituted with sub- stituents which assist the binding of such compounds to the glutamine transporter protein of tumour plasma cell mem¬ branes, or which modify the action once the compounds have been transported.

The term "3- to 8-membered cyclic hydrocarbons" includes 3- to 8-membered saturated and unsaturated hydro¬ carbon rings, which hydrocarbon rings may contain 1 to 3 heteroatoms (N, 0 or S) and may be substituted with sub- stituents which assist the binding of such compounds to the glutamine transporter protein of tumour plasma cell membranes or which modify the action once the compounds have been transported.

Preferably, the "3- to 8-membered cyclic hydro¬ carbons are selected from Cs-e aromatic or heteroaromatic hydrocarbons substituted by one or more polar groups such as OH, SH, NHR ¹³ , N0 , CHO, COOR ¹³ , or halogen, wherein R ¹³ is as previously defined.

The term "amino" includes primary secondary and tertiary amino groups, NR ⁸ R ⁹ , wherein R ⁸ and R ⁹ are as previously defined.

The term "amido" includes C(0)NR ⁸ R ⁹ wherein R ⁸ and R ⁹ are as previously defined.

Possible salts of compounds of the general formulas (I), (II) and (III) include, for example, all of the acid and base addition salts. Physiologically acceptable salts may be generally derived from inorganic or organic acids or bases. Physiologically unacceptable salts, for example, which may initially be obtained as process products, for example in the preparation of the compounds according to the invention on an

industrial scale, can be converted into physiologically acceptable salts by known processes. Examples of physio¬ logically acceptable salts are water-soluble and water- insoluble acid or base addition salts, such as the hydrochloride, hydrobromide, hydroiodide, phosphate, nitrate, sulfate, acetate, citrate, gluconate, benzoate, butyrate, sulfosalicylate, maleate, laurate, malate, fumarate, succinate, oxalate, tartrate, stearate, tosylate, mesylate, salicylate, sodium, potassium and ammonium.

A. Synthesis of Anti-Glutamine Compounds and Glutamine Analogues Compounds of the general Formula I - III may be synthesised by a number of different methods as outlined below. The method of synthesis will generally depend upon the nature of X, W and V. Protecting groups will generally be required, and the choice of protecting groups for precusors will depend upon the functional groups desired in the final product. The use of protecting groups, as well as the methods for protection and deprotection, are well known to those skilled in the art. (For example, see T.W.Greene, "Protecting Groups in Organic Synthesis", John Wiley & Sons, 1981.)

Similarly, the various groups which can comprise R ¹ to R ⁴ are also well known and can be introduced using reported syntheses or simply by exercise of skill following this disclosure, together with reported syntheses. The order of substitution and the time of placement in the synthesis may change depending upon the nature of R ¹ to R ⁴ and the other substituents. In short, one of ordinary skill will recognize various ways to prepare the compounds of this invention, and will have to tailor the synthesis depending upon the particular groups chosen for X, W, V and R ¹ to R ⁴ .

Compounds of Formulas I - III in which X is CH ₂ and W is C=0 may be synthesised, as illustrated in Schemes 1 and 2 (FIGS. 51 and 52, respectively), from suitably pro¬ tected derivatives of glutamic acid. These protected derivatives include the pyroglutamic acid derivatives 2_ and 3_

- 103 -

as well as the glutamic acid half-esters, 6_ and 1_. While several of these compounds are commercially available, it may be less expensive to synthesise them as needed.

As shown in Scheme 2, activation of compounds 6_ or to render them suitable for nucleophilic substitution, e.g. by amines, R ⁸ R ⁹ NH, may be achieved by two main methods, viz., by forming a mixed anhydride, e.g. 8_, or an activated ester, e.g. 9_. Both methods are reported in the literature (for example, see Goodman et. al.. (1962) J. Am. Chem. Soc, 84, 1279; Anderson et al. (1967) J. Am. Chem. Soc, .89., 5012; Dutta et al. (1971) J. Chem. Soc. (Perkins C), 2896; Kim et al. (1975) J. Chem. Soc. Chem. Commun., 473) and are illustrated in the Examples provided below.

Compounds of the general Formula I in which X represents CH ₂ and W is other than C=0, may be prepared by reacting a suitably protected derivative of dehydroalanine e.g. iO. with an -phosphonyl carbanion JL1, or -sulphonyl carbanion .12.. Alternatively, as exemplified in Scheme 3 (FIG. 53) , the reaction of diethyl acetylaminomalonate .13. with a suitably elaborated vinyl phos-phonate .14. or vinyl sulphonate 15, may be used to synthesise these compounds (for example, see: Maier et al. (1983) Phosphorus Sulfur, 17, 1). Where W is S0 ₂ , and X is CH ₂ , derivatives may also be synthesised by elaboration of homocysteine or homocysteic acid.

Compounds of Formulas I - III in which X is 0 and W is C=0 may generally be synthesised by acylation of a suit¬ ably protected derivative of serine e.g. J5. (Scheme 4; FIG. 54). In cases where V is represented by NR ⁸ R ⁹ , wherein R ⁸ and R ⁹ are as previously defined, reaction of phosgene with the protected serine will form a chloroformate. This product would then be reacted with the amine nucleophile R ⁸ R ⁹ NH to yield the carbamate 11_ as exemplified in Scheme 4. (FIG. 54). In cases where W is other than C=0, compounds of Formula I can likely be synthesised by reacting a suitable sulphonic acid or phosphonic acid or activated derivatives (e.g., the acid halides) with a serine derivative.

Compounds of the Formulas I - III, wherein X is N or N-OH and W is C=0, may be synthesised by reacting a suit¬ ably protected form of 3-aminoalanine JB or 3- hydroxy- laminoalanine .19. with a suitable isocyanate 2___\ or acyl halide 21 as shown in Scheme 5 (FIG. 55). In cases where W is other than C=0, compounds of the Formula I may be synthesised by procedures similar to those shown in Scheme 5. In these cases, a suitably protected derivative of 3-aminoalanine JL8. or 3-hydroxylaminoalanine .19. could be reacted with a phos- phoryl or sulphonyl halide which is appropriately substituted (Scheme 6; FIG. 56).

Compounds of the Formula I in which V-W-X-CH ₂ represents the group:

wherein R ¹⁷ , R ¹⁸ , R ¹⁹ are as previously defined, may be synthesised from a suitably protected alkenyl glycine .22. by "[2 + 2] cycloaddition" with a substituted isocyanate 23 under thermolytic conditions, as exemplified in Scheme 7 (FIG. 57).

Compounds of the Formula I in which V-W-X repre¬ sents the group:

in which Z is O, as previously defined, may be synthesised by reaction of a suitably pro¬ tected form of glutamine e.g. .24. with tosyl chloride and pyridine under dehydrating conditions to form the nitrile 25. The nitrile may then be subject to partial solvolysis with an alcohol, R ²¹ 0H, to yield the imino ether 2^, as exemplified in Scheme 8 (FIG. 58). See, for example, Hirotsu et al. (1970) Biochim. Biophys. Acta, 222, 540. The imino ether may

be further elaborated by alkylation or acylation of the imine nitrogen. Alternatively, a substituted amide 2_1_ may be acylated or alkylated at the amide oxygen to yield the imino ether or imino ester 28.

The nitrile _______ or the simple imino ethers _______

(R ¹ =Me or Et) may also serve as precursors to compounds in which Z is S or N. Thus, reacting 25 ^or 6. (R ²¹ =Me or Et) with thiols (R ²¹ SH) or amines (R ²¹ R ¹⁴ NH), wherein R ^{2 ~} and R ¹⁴ are as previously defined, would be expected to afford the derivatives .29. and J30. respectively (Scheme 9; FIG. 59).

Similar methodology may be applied to the synthesis of compounds of Formula I wherein R ² ° and R ² ~ are taken together with the group:

to form a 5- or 6-membered heterocyclic ring which may be further substituted or unsaturated. For example, reacting 2J5. or J26. (R ²¹ =Me or Et) with a β-amLno alcohol, thiol or amine (e.g. NH ₂ CH ₂ CH ZH; Z= 0, S, NH) would be expected to give the dihydrooxazole, Δ ² -thiazoline or imidazoline derivatives respectively (.31.; Z= 0, S, NH) . (For example, see: "Comprehensive Heterocyclic Chemistry", Eds. A.R.Katritzky & C.W.Rees, Pergamon Press, 1984, V.5, 469; V.6, 228-229, 309). These derivatives may be further elaborated e.g. by dehydrogenation to the corresponding oxazole, thiazole or imidazole derivatives (JJ2.; Z= 0, S, NH), as illustrated in Scheme 9 (FIG. 59). This methodology is capable of extension to the corresponding six-membered heterocyclic compounds and to further substituted examples.

B. Anti-Tumour Design Strategy

The program of chemical synthesis should be highly integrated with the biochemistry program and can be guided by computer-aided molecular modelling. As discussed below, biological assays viz. binding parameters (Bin, Ks_'), together with glutamine transport inhibition and tumour growth

measurements, have enabled identification of at least three classes of anti-glutamine compounds and glutamine analogues which have potential in cancer therapy.

Preliminary molecular modelling studies can make gross structural comparisons as a means of identifying potential synthetic target molecules. As further testing data for newly synthesised compounds becomes available, quantitative structure-activity relationships for each of the above-mentioned classes can be developed. Such relationships correlate observed biological activities with a variety of physical and chemical properties of the compounds, e.g., minimum energy conformations, atom juxtapositions, critical dipole moments, electrostatic potentials etc., which are available from molecular modelling and/or molecular orbital calculations undertaken within the program.

This approach has successfully identified struc¬ tural modifications of the -amido function of glutamine, which still allow the resulting analogue to be efficiently and selectively transported by the glutamine transporter of tumour cells. Appropriate levels of cytotoxicity can then be incorporated into such analogues. VI. SCREENING METHODS FOR ANTI-GLUTAMINE COMPOUNDS AND GLUTAMINE ANALOGUES Another embodiment of this invention comprises methods for screening anti-glutamine and glutamine analogues to determine binding parameters, transport inhibition with respect to the transport of glutamine in tumour cells and lymphocytes, and the effect of such compounds on cell growth. From the results obtained (discussed below), potentially useful anti-cancer compounds can be identified. It will be appreciated that the same types of screening methods can be used to screen other compounds with respect to other human amino acid transporters.

A. Binding Parameters

The following experimental procedure was used to determine binding parameters for the anti-glutamine compounds and analogues.

1. Solutions

All solutions were made up in HBSS at a pH of 7.2 adjusted with 5% (w/v) sodium bicarbonate. Pre-incubation test solutions contained a range of anti-glutamine compound and analogue concentrations from 0.25 mM to 20 mM and the control pre-incubation solution contained 1 mM glutamine. The NEM incubation solutions contained the same concentra¬ tions of analogue or control glutamine as well as 1 mM NEM. The radioactive solutions were 0.1 mM NEM containing 0.05 mCi of ³ H-NEM /ml.

2. Procedure

Membrane preparations containing 0.2 mg protein/ml were centrifuged for 2 minutes at 11600 g (Eppendorf micro¬ fuge) and the pellet was resuspended in 100 μl of pre- incubation test solution or control solution for 10 minutes at room temperature with gentle agitation. At the end of 10 minutes, 100 μl of NEM incubation solution was added to each sample for a further 10 minutes, at room temperature. The membrane suspension was then centrifuged for 2 minutes at 11600 g and the pellet was washed twice in 1 ml of fresh HBSS. The membrane pellet was resuspended in 100 μl of ³ H-NEM solution for 1 hour minutes at room temperature with gentle agitation. At the end of the labelling period the membrane suspension was centrifuged for 2 minutes at 11600 g, followed by three washes in 1 ml of ice-cold 0.9% (w/v) NaCl. The pellet was solubilized in 200 μl of SDS sample buffer by vortexing for 1 minute every 5 minutes over aperiod of 15 minutes and stored at -20°C. The total protein content of each sample was determined by Lowry analysis and 30 μl of each sample was applied to separate lanes of a 12% poly¬ acrylamide gel and the membrane proteins separated by SDS- PAGE and stained with coomassie blue. The stained proteins of interest, including the glutamine transport protein, were sliced out of the gel, solubilized in a 1:1 solution of 30% H ₂ θ ₂ and 0.9 mM CuS0 overnight at room temperature and counted for radioactivity. The results were expressed as the

amount of label bound to the protein band in nmoles ³ H-NEM/ mg total protein.

3. The estimation and derivation of binding parameters

This procedure assesses the two major binding parameters of the anti-glutamine compounds and glutamine analogues in terms of their capacities to protect the glu¬ tamine binding sites on incubation of the membranes for 10 minutes with 1 mM NEM in HBSS. The membranes were first incubated with a range of glutamine or analogue concentra¬ tions (ligands) in the presence of NEM and since the reaction of the NEM with the thiol sites is covalent and therefore irreversible, the extent to which the NEM reaction with the _. glutamine sites was impeded by the ligands was determined by washing the membranes and then re-exposing them to the same concentration of tritiated NEM (NEM*). The NEM* then occupies the sites which were protected by the ligand molecules in the first incubation. The use of a range of concentrations of the ligands against a fixed concentration of NEM in the first incubation enables one to obtain an estimate of the relative dissociation constant (K ' ) , and the maximum binding constant (B ) by making use of a double reciprocal plot (FIG. 24). Reaction Scheme 10 below shows that the former is a relative value, since each ligand binding reaction is carried out in competition with NEM, and that the latter should give absolute values for the maximum binding so long as the binding of the ligands is reversible. In the case where the binding of the ligands is irreversible, they will form the tightly bound or covalent complex (CS) through the rate constant k ³ :

+ k-

NEM _NEM *

C. EM _c _ _NEM *

The ratio of CS to total carrier (C ₀ ) present at the end of the first incubation is provided by the equation below:

CS/Co = S/{K _g (l + k ₂ .[NEM]) + S(l + k ₃ )> where K__ = k-_/κ_.

From this it is possible to derive a new constant:

V ⁼ X-. . ^{1 +} k ₂ [NEM]), where [NEM] is kept constant in the incubations.

In basic terms, B _m represents the ability of a compound to reversibly bind to the binding sites of a glutamine transporter, with B _m = 3.45 (glutamine) being the highest possible value, and indicating that the compound can bind and then become unbound. A low value (e.g. less than about 1.0) indicates that the compound irreversibly binds to the glutamine transporter. K__ ' indicates how well the compound fits the transporter. A low value, e.g., less than about 1.5, indicates a relatively good fit, whereas a higher value indicates a fit that is not as good. B. Transport Inhibition Assay Whether a given anti-glutamine compound or glu _¬ tamine analogue can inhibit transport of glutamine can be determined as described above in Section II.C.2.a. Experi _¬ ments involving the inhibition of transport required the presence of the appropriate concentration of glutamine analogue in all glutamine test solutions used for incor _¬ poration of glutamine into cytoplasm. The normal pre- incubation period was extended from three minutes to ten minutes.

SUBSTITUTE SHEET

C. Growth Inhibition Assay

The effect of the anti-glutamine compounds on the growth of HeLa cells in culture was investigated by following the change in total protein content of the culture over a period of time. HeLa cell cultures were inoculated to give a confluence of 20% in 25 cm ² plastic flasks. After an ini- tial establishment period of 24 hours the media was changed to fresh medium containing 1 mM of the test compound. Fresh medium was added to the cultures every day and the condition of the cultures was carefully monitored by microscopy and photography. Cultures (n = 3) were collected every 24 hours over a period of three days and the cells were harvested in the following way:

1. The cultures were washed three times in 0.9% (w/v) NaCl and then 5 ml of 0.9% (w/v) NaCl was added to the flask for 10 minutes at room temperature.

2. The cells were scraped from the surface of the flask into a glass test tube using a rubber policeman.

3. The tubes were centrifuged at 200 g for 10 minutes (Clements bench-top centrifuge) and the pellet was washed two times with 0.9% (w/v) NaCl.

4. The cell pellet was digested in 1 ml of 0.1 M NaOH overnight at 100°C and the total protein content of each sample was determined by Lowry analysis.

5. The results were expressed as the amount of protein present relative to the control samples which had not been exposed to the analogue.

In experiments where a combination of two anti- glutamine compounds were used in growth inhibition studies three sets of controls were required: a control for the effect of each of the compounds separately and another control for normal growth.

D. Glutamine Transport in Lymphocytes Lymphocytes require a significant amount of glu¬ tamine during mitogenesis. The glutamine uptake of indi¬ vidual lymphocytes is actually low, but the large number of lymphocytes present means that a significant amount is taken

- Il l -

up during periods of activity. Glutamine thus plays an important role in the immune system due to its metabolic role in lymphocyte biology. Therefore, the potential effect of anti-glutamine compounds and glutamine analogues on lympho¬ cytes should also be checked through screening. For example, compounds which are toxic to both cancer cells and lympho¬ cytes may not be desirable. Furthermore, those compounds which enhance mitogenesis may prove useful in stimulating the immune system to combat immune-related diseases such as AIDS and cancer. Still further, compounds which inhibit mitogene¬ sis, but which are not toxic to lymphocytes, may be useful, e.g., as immunosuppressive agents.

A comparison of the glutamine uptake in HeLa cells and in lymphocytes is shown in FIG. 30. Glutamine transport into HeLa cells is biphasic over the concentration range 0.25 mM - 4 mM and is 50% inhibited after 20 minutes incubation with the sulfhydryl reagent NEM (ImM). This result indicates that sulfhydryl groups are located at the active site on the glutamine transporter protein and modification of these groups leads to inhibition of glutamine transport.

1. Materials and Methods

Lymphocytes were routinely obtained from fresh whole blood obtained by jugular vein puncture from cattle, and in some experiments from fresh human blood obtained by venepuncture. Lymphocytes were separated from erythrocytes by density gradient centrifugation using standard techniques involving Ficoll Paque. Tissue culture materials were ob¬ tained from Flow Laboratories, USA and U- ¹ C- glutamine was from Amersham, Australia.

Freshly isolated cells were resuspended in RPMI 1640 medium at concentrations of 10 ⁷ cells/ml. Concanavalin A was added to a final concentration of 2.5 μg/ml of the suspension. The mixture was gassed with a 5% C0 ₂ : 95% air mixture and incubated at 37°C for 48 hours.

For the measurement of glutamine transport, the lymphocytes were resuspended in PBS A containing tritiated water to a final concentration of 5 nCi/ml for 30 minutes.

SUBSTIT TE SHEET

Fifty microliters of the samples were taken in quadruplicate and added each to a 0.5 μl microfuge tube prewarmed to 37°C and containing 20 μl sucrose solution (10% w/v) above which was layered 50 1 of silicone oil (Versilube 5-50, General Electric, N.Y. ) . At a noted (zero) time a further 50 μl of ¹ C-glutamine was added to each tube. The level of radio¬ activity was kept constant in each solution while the concen¬ tration of glutamine varied from 0.1 to 20 mM.

At the end of the exposure period, transport was terminated by centrifuging the cells through the silicon oil layer at 11600 g using an M.S.E. Microcentaur centrifuge for 60 seconds. The great majority of the lymphocytes appeared in the sucrose layer within the first 15 seconds. ³ H was used as an indicator of cell numbers appearing in the sucrose layer.

The tubes were frozen by immersion into liquid N ₂ and the bottom sucrose compartment was amputated into plastic mini scintillation vials prior to counting. To ensure dis¬ persion of the lymphocyte button the vials containing scin- tillant and the base of the amputated tubes were shaken over¬ night in an orbital shaker. The ratio of ¹⁴ C/ ³ H was calculated from the results and was used as a measure of glutamine uptake per cell (lymphocyte).

2. Inhibition studies using glutamine analogues

In these experiments, the suspension of lymphocytes was incubated in PBS A with a series of concentrations of the analogue for 20 minutes prior to measurements of uptake being made as above.

FIG. 31 shows the time course for - C-glutamine uptake by normal lymphocytes subsequent to stimulation with concanavalin A. FIG. 32 shows the initial rate of glutamine uptake into bovine lymphocytes expressed as nmoles/mg protein/min, in the presence and absence of concanavalin A. FIG. 32 also shows uptake due to passive diffusion.

3. Effects of anti-glutamine compounds or glutamine analogues on lymphocyte metabolism

'* a. Chemosensitivity testing of normal cell lines using the MTT assay ** Chemosensitivity tests were performed using reduc¬ tion of a tetrazolium salt (MTT) as the assessable end point. This particular assay has been automated and further modified in reported literature to allow for better solubilisation of the formazan product for absorbance measurement. The MTT assay was performed using microtitre plates and scanning plate reader to measure absorbance in individual wells, thus allowing replicate testing of several drugs at multiple con¬ centrations on several cell lines. Previous workers have found that the MTT test gave response curves which were highly correlated with clonogenic and dye exclusion assays. Thus, for any given cell line, the optical density of the solubilised formazan product obtained after incubation of the cells with MTT, is directly proportional to the number of viable cells per well. . Method The cell suspension was diluted to 4 x 10 ⁶ cells/ ml using RPMI 1640 with 10% (v/v) FBS. 100 μl of the cell suspension was added per well followed by 50 μl con A (final concentration of 2.5 μg/ml) and 50 μl of the solution of the analogue.

The plate was incubated at 37°C for 48 hours in 5% C0 ₂ / 95% air and 20 μl of MTT was added per well. Incu- _* bation was continued at 37°C for a further 4 hours. The supernatant was removed, and 50 μl of DMSO was added to dissolve the crystals. The solution was read at 560 nm.

The capacity of cells to exclude the dye Trypan Blue is a commonly used indicator of their viability. The lymphocyte suspension was diluted so that approximately 100 cells appeared in the viewing field of a Neubauer haemocytom- eter. The ratio of cells which excluded dye to those which

SUBSTITUTE SHEET

took up the dye was assessed several times by direct counting. c. Stimulation of in vitro lymphocyte mitogenesis by inhibitory glutamine analogues It was observed that at sub-inhibitory concentra¬ tions of the glutamine analogues, the extent of lymphocyte mitogenesis which could be induced by exposure to concanava¬ lin A was regularly stimulated. Generally it was observed that at concentrations of the order of 1/10 that for sig¬ nificant inhibition of mitogenesis, there was commonly a regular 50% increase in the rate of cell division. Figures 33-38 show survival of lymphocytes measured by the dye exclu¬ sion method and relative mitosis measured by the MTT assay plotted against a range of concentrations of glutamine ana¬ logues.

We regard this as a particular example of the general phenomenon known as "hormesis", i.e., the stimulus of growth by a sub-inhibitory concentration of a toxic sub¬ stance. It is possible that this phenomenon can be exploited in such a way that the delivery of the drug to the tumour (which must inevitably be accompanied by some "leakage" to other cells of the body) , may in fact be at such a concen¬ tration that the natural immunity of the body is stimulated rather than depressed. Thus, compounds, compositions and formulations as described herein, and dosages thereof may be designed to encourage hormesis with the immune system rather than inhibition, and to encourage growth inhibition rather than hormesis within tumours. d. Inhibition of lymphocyte mitogenesis by inhibitory glutamine analogues The results of MTT assay were compared with results obtained under identical conditions by dye exclusion assay: the principle MTT assay follows mitochondrial activity while dye exclusion measures viability.

It was observed that all of the analogues tested showed an inhibitory concentration range using MTT which was ^*

preceded by a stimulating range. The inhibitory range in turn always preceded cell death, thus there is a range of concentrations for some analogues, where the cell is first stimulated towards division. At a higher range the cells are pushed into a "resting" stage, while at even higher concen¬ trations, they are killed. It is intended to exploit all of these ranges of the drugs' behaviour in cancer and other disease treatment.

E. Results of Screening Tests 1. Binding parameters

Tables 5-8 (FIGS. 76-79) summarize the data with respect to the binding parameters of the various ligands tested, and how these relate to glutamine transport and cellular growth. Thus, under conditions where S is satu¬ rating, the amount of C.NEM* found at the end of an exper¬ iment is inversely proportional to } ^~ ₃ and where k ₃ approaches zero, the equation reduces to S/{S + K _s (l ⁺ k ₂ NEM), and all of the carrier is converted to CS giving rise to maximal formation of C.NEM*. The list of ligands in Table 8 (FIG. 79) has been divided into four classes (A-D) . Class A is composed of compounds 1-3, and represents those ligands with B values which approach those of glutamine (>than 2.5 nmol/mg protein). Because of their ability to protect the glutamine sites under saturation conditions (indicated by maximum binding of NEM* on washing), it is assumed that Class A ligands bind reversibly with the trans¬ porter molecule. Such compounds are transported on the glutamine carrier in exactly the same way as glutamine itself, and this has been demonstrated in other experiments using ~ ⁴ C-histidine. Both of the compound appearing in this class have an effect on HeLa cell growth.

Class B ligands have B^ values between 1.0 and 2.5 nmol/mg protein. These are compounds which, either bind to the transporter protein in such a way that the reaction of the ligand at the binding site is a poor fir and they are thereby unable to saturate the site in competition with the NEM, or, they are compounds which are only moderately

reversible, and compete to an intermediate degree with the NEM, thereby producing an intermediate degree of blocking of the available sites for the second incubation with NEM*. It can be seen that one of these compounds, i.e., glutamic acid- ₇ -hydrazide, is the most powerful inhibitor of glutamine transport so far tested, and that it has only a relatively small effect on HeLa cell growth.

Class C ligands consist of compounds which have B values which are less than 1.0 nmol/mg protein and which significantly inhibit glutamine transport (more than 20%). The mechanism suggested in Scheme 10 indicates that this group of compounds should consist of irreversible binders. The fact that three known covalent binders (β-chloroalanine, azaserine and DON) appear in this class provides confirmatory evidence for the mechanism above and the present system of classification. All three compounds critically inhibit growth of the HeLa cells. The other tight binders strongly inhibit transport but have little effect on growth, in par¬ ticular compound C.2 should have low toxicity and could be an excellent blocker of glutamine transport when used in the presence of a more toxic agent.

The final class of ligands (D) consists of those compounds which either bind poorly at the glutamine site or bind actively at sites on the transporter other than the glutamine site. Compounds D.9 and 12 may move out of this class when the transport and growth experiments have been completed. Compound D.14 is being investigated further, since, if it is being transported through the glutamine transporter, it is getting into the cells in extremely small amounts, which leads to the conclusion that this compound may be highly toxic. Note should also be made of the marked contrast in properties exhibited by compounds A.3 and D.14 where the addition of a methyl group to glutamic acid hydroxamate results in a loss of the ability to inhibit transport but which has retained the capacity to inhibit growth, though with only 50% of the effect of the glutamic acid hydroxamate. Compound D.10, N-acetylglutamine, was

found to activate glutamine transport, and it has therefore been assumed that this compound does not bind at the glutamine site but somewhere very close where it may act as a positive effector.

2. Growth inibition assay

*_ The results for the effects of the anti-glutamine compounds on the growth of HeLa cells in culture are given in Table 4 (FIG. 75). The results are expressed as the amount of inhibition produced by the compound after 48 hours expo¬ sure relative to control cultures which had not been exposed to the compound. The established antitumour compounds, i.e., azaserine, methotrexate, DON and β-chloroalanine, all had significant inhibition of cell growth. Of the new compounds tested, only 2 had significant effects on growth, namely glutamic acid- ₇ -monohydroxamate, and glutamic acid- ₇ - methylmonohydroxamate. None of the anti-glutamine compounds used in combination with methotrexate produced any appre¬ ciable synergistic action.

3. Transport inhibition assay a. The effect of anti-serum on glutamine transport in HeLa cells

The effect of glutamine transport protein anti¬ serum which had been raised in either rabbits or mice is shown in FIG. 6. At 1 mM glutamine the rate of glutamine uptake into HeLa cells was inhibited by 25% in the presence of 1% anti-serum. b. The effect of anti-glutamine compounds and glutamine analogues

I on glutamine transport in HeLa cells

The effect of anti-glutamine compounds and glu- _λ tamine analogues on the rate of glutamine uptake into HeLa cells is shown in Table 5 (FIG. 76). A number of the com¬ pounds were found to inhibit the rate of glutamine uptake at 1 mM glutamine, including glutamic acid-(2-hydroxyethyl) amide, glutamic acid derivative (see Table 2), glutamic acid-

SUBSTITUTE SHEET

-diethylamide, cyclized glutamic acid- ₇ -tris(hydroxy¬ methyl)methylamide, glutamic acid- ₇ -bis(hydroxyethyl)amide and glutamic acid- ₇ -p-nitroanilide.

The compounds appearing in Table 8 (FIG. 79) represent a wide range of chemical functionality. In this survey, all measurements used have been conducted on the basis of a preliminary screening, which will point the way to the final chemical strategies. Thus, in most cases, the inhibition of transport, for example, has been conducted with both the glutamine and the analogue at 1 mM because that represents the optimum concentration of glutamine to be expected in human blood. The results, therefore, provide an initial indication of physiological response. More detailed dose response experiments can be implemented after there is sufficient information to develop investigative strategies based on chemical functionality.

The division of the compounds into 4 classes (A-D) is related to the working hypothesis proposed for the mecha¬ nism describing reversible and irreversible binding of ligands against NEM and ³ H-NEM as shown in Scheme 10. This classification is based on two main measurable variates, the maximum binding test for ³ H-NEM and transport inhibition. Class A compounds are compounds which have levels of H-NEM binding close to that of glutamine. They are considered to react at the glutamine site with a degree of reversibility approximating that of glutamine. Class B compounds are com¬ pounds which have intermediate levels for ³ H-NEM binding and are therefore assumed to have intermediate reversibility. Class C compounds are those compounds which exhibit very low levels of ³ H-NEM binding, but which also significantly inhibit transport (>20%). Such compounds are considered to be capable of irreversibly binding at the glutamine trans¬ porter site. Class D compounds are those compounds which also have low levels of ³ H-NEM binding, but which also inhibit transport minimally (<20%). Such compounds are considered to have minimal binding capacity for the transporter site.

The compounds which exerted the greatest effect on transport and growth were the three known anti-tumour com¬ pounds: α-chloroalanine, azaserine and DON (Compounds Cl, 3 and 5). All of these compounds had very low values for the

Bm which is accounted for by the fact that all of them react covalently at the active site and therefore are irreversible competitors for the active site. This confirms their clas¬ sification as C type compounds. Since these compounds are bound covalently at the active site of the transporter, their capacity to bring about cell death within 48 hours is some¬ what problematical. However, it should be realised that the covalent binding occurring under these conditions is rela¬ tively slow, so that transport of the ligand will proceed through the initial stages of the binding reaction until all the sites are fixed with covalently bound ligands, at which point further transport, except for non-saturable transport, should cease.

The work of Huber et al. , Int. J. Cancer, 41, 752- 755, supports this hypothesis in that they have shown that the uptake of ¹ C-D0N in mouse P388 leukaemia cells approached a maximum within approximately 3 minutes. Since DON does not appear to be metabolised in these cells, this final state must either represent the resultant steady state between influx and efflux, or the point at which further influx and efflux ceases, and the final concentration of DON represents molecules which have been trapped in the cells as a consequence of saturation of the transport sites by covalently bound ligands. If the former is true, then the efflux experiment should have shown near zero retention of the ¹⁴ C-D0N at near final cellular concentration. In fact there was a drop of about 50% in the first 4 minutes, and at the end of 10 minutes there was still 40% of the DON present in the cells, indicating that effluxing of the DON had essentially ceased. Thus it must be concluded that the remaining DON has either been bound internally or that it represents DON which has been trapped inside the cell by saturation of the transport sites with bound ligand. The

latter appears to be the more likely explanation because in the same paper it was demonstrated that the internal DON on release from the cells was freely separated on a thin-layer chromatogram.

Furthermore -chloroalanine, azaserine and DON all had ' values (dissociation constant in competition with NEM) which were more than 3 times that of glutamine (K_' = 1.33 mM), suggesting that these compounds were making a poor fit at the glutamine binding site, and that specificity for the tumour glutamine binding sites could be considerably improved. The lowest K ' was that of -chloroalanine (4.34 mM) and the highest was DON (7.1 mM), in agreement with the finding that the latter produced the lowest inhibition of glutamine transport. These results with DON are again sup¬ ported by the work of Huber et al. , who showed that the uptake of ¹⁴ C-D0N in mouse P388 leukaemia cells was mediated at least in part by the "L" transport system, thus confirming a lack of specificity for the glutamine transport site. The possibility that the main transporter for DON is the leucine preferring system is not surprising in view of the structure of DON (norleucine derivative) . It has been found that Na -dependent glutamine transport in HeLa cells is not inhibited by leucine, and is therefore not mediated by an "L" type transporter.

The compound which had the lowest K ' value to¬ gether with very tight binding (low B ) was compound C.2 (glutamic acid- -p-nitroanilide) • ^Tne structure of this compound makes it an automatic candidate for tight binding, once again fitting its position in the C class. The compound produced a significant inhibition of glutamine transport, but had very little effect on growth, giving rise to the likeli¬ hood that it was transported in only small amounts. Thus, such compounds would be expected to have relatively low toxicities, but would at the same time significantly diminish intracellular levels of glutamine (glutamine deficiency always results in a drastic decrease in glutamine concen¬ tration in HeLa cells), and thereby diminish cellular

resistance to compounds of high toxicity such as acivicin and the other active compounds in class C. Compound _A .2, the hydrazide derivative, produced the most active inhibition of glutamine transport but had only a small effect on growth. Since this compound exhibited a fairly high B value, it must be assumed that large quantities of the compound were trans¬ ported into the cells, and there had a minimal toxic effect. This latter finding agrees with an earlier reported observa¬ tion of the effect of the hydrazide on cancers in experi¬ mental animals. On the other hand Compound A.3, the hydroxa¬ mate, had a major effect on cellular growth, but a lesser effect on transport. Compounds B.2 and C.6 demonstrate an excellent reciprocal relationship as between K ' and inhibi _¬ tion of transport. Doubling the number of substituents on the 7-amido of group glutamine resulted in doubling the K ' s value and halving the inhibition of transport. Further investigation is required on compound D.14, which was appar _¬ ently minimally transported but managed to produce 22% inhibition of growth. If these results are confirmed, this compound would be the most powerful inhibitor or growth of all compounds tested.

The above discussion leads to the following con¬ clusions with regard to the activities of the various compounds tested:

1. The known anti-tumour compounds (compounds Cl, 3 and 5), which display covalent binding at the glu¬ tamine active site, and which are powerful inhibitors of both transport and growth, were shown to have relatively poor K ' values and therefore were not making the best possible fit at the transporter site. This conclusion has received strong support from the work of Huber et. al.. who demonstrated that DON transport is mediated in part by the "L" transport system. Thus, it is likely that at least one of these compounds, in particular DON, can be further modified to provide a structure which has a significantly greater affinity (lower K ') for the glutamine carrier sight in solid tumour cells.

SUBSTITUTE SHEET

2. Compound A.3 (glutamic acid- ₇ -hydroxamate) which was reversibly bound at the glutamine site, and proved to be an excellent inhibitor of both glutamine transport and growth. The properties of this compound must therefore be further investigated together with its related compound D.14, the methylhydroxamate, which did not appear to have any binding activity for the glutamine site, but nevertheless produced a significant effect on growth.

3. Compound B.l, the hydrazide derivative, had the highest effect oh glutamine transport of all the com¬ pounds investigated, but only a minimal effect on growth. Further investigations on the possible potentiating effect of this compound on other more toxic compounds should be under¬ taken, because, in the presence of this compound, the tumor cells would be glutamine deficient, and the requirements for tumour growth would be delicately poised.

4. Compound Cl, the nitroanilide, which is a tight binder and is active at μM levels, was a powerful inhibitor of glutamine transport. Because it is a tight binder, this compound will be active at levels well below its K ___> and should therefore have a negligible toxic effect. This compound should also be investigated for its effect on tumour growth in the presnece of more toxic agents.

5. Except for compound D.14, the remainder of the compounds from D.7 to D.18 have little or no inhibitory activity. Albizziin and acivicin are in this group, although further testing is required. There is one exception in this group which is compound D.10, the N-acetylglutamine. This compound activates the transporter III activity, and, because it is a normal metabolite, it is completely non-toxic It is proposed that this compound could be used to activate the transport system to further increase the rate of pumping suitable analogues into the tumor cells.

It will be appreciated that the screening tests and results presented herein enable evaluations to be made of the optimum properties that a glutamine analogue must have in

order to be carried into cancer cells by the glutamine transport system with maximum efficiency.

On the basis of the results discussed above, a structural model for anti-glutamine compounds or glutamine analogues suitable for the treatment of cancer can be developed. Thus, the results enable one to predict compounds that will be moved into cancer cells by the glutamine trans¬ port system of the plasma membranes of those cells, at the expense of the extracellular glutamine required to be moved into tumour cells. Once the anti-glutamine compound or glutamine-analogue is inside the cancer cell, it should inhibit most of the glutamine requiring enzymes, particularly those involved in DNA synthesis, and in this way bring about the death of the cell. It is not necessary for such a com¬ pound to be particularly toxic, because glutamine analogues can be designed which will be carried into cancer cells (the transport system is an active energy-requiring system) in very high quantities, so that if the analogues are not metabolized, they will be broken down to products which are highly toxic to the cell.

The evidence presented above also provides a basis for the future development of new glutamine analogues pos¬ sessing the essential property of being able to specifically target tumour cells and thereby produce directly or indi¬ rectly a critical inhibition of DNA and RNA synthesis. Such analogues offer the hope that their specificity for tumour cells will be such as to lower significantly the cytotoxicity for normal cells which commonly accompanies present cancer chemotherapy. It will be appreciated that applications of this invention are not restricted to use of a single ana¬ logue, but also two or more of the analogues could be used in concert.

4. Inhibition and metabolism studies using mitochondria

In the search for anti-glutamine compounds that inhibit glutamine transport in the cells, one can also check to see whether such compounds inhibit glutamine transport

SUBSTITUTE SHEET

- 124 -

into the mitochondria. Metabolism studies using mitochondria can also be performed. Previously, studies on tumour mito¬ chondria have been mainly restricted to Ehrlich ascites cells and hepatomas because of the relative ease with which ade¬ quate material for experimental purposes can be obtained. On the other hand, there are considerable difficulties involved in obtaining sufficient material for the preparation of mito¬ chondria from cultured tissues. The major limitation in the study of mitochondrial respiration and metabolism is the availability of large quantities of the cells. However, the large scale microcarrier techniques described below allowed sufficient quantities of mitochondria to be isolated from the cultures, thus enabling experiments to be performed.

HeLa cells obtained from Flow Laboratories arrived in a confluent state. These cells were grown in MEM con¬ taining FBS (10% v/v). The cells were passaged in a 1:2 split ratio, and placed in Medium 199 containing FBS (10% v/ v) . These cells took a few days to adjust to the new culture medium, after which they behaved similarly to cells cultured for long periods in Media ^" 199, as ascertained by comparison of glutamine uptake rate measurements.

Due to the adhesive properties of Cytodex 1 or 3 microcarrier beads, all glassware used should be siliconized prior to use. A 10% (v/v) solution of Surfasil in toluene was liberally applied to the glass surfaces. After the fluid was allowed to drain off, the glassware was dried overnight. When dry, the glassware was rinsed 10 times in distilled water, and placed in an oven to dry. The glassware was then ready for use. If an excessive number of beads attached to container surfaces, the containers were resiliconized.

Cytodex 1 or 3 microcarrier beads (2 g beads/l00 ml PBS A) were hydrated in siliconized bottles overnight. The next day the old PBS A was discarded, and 100 ml of fresh PBS A added to the beads. The beads were sterilized (121°C, for 20 minutes at 1.2 atm) prior to use.

The sterilized beads were allowed to settle and the supernatant discarded. They were resuspended in Medium 199

(50 ml/g beads). After a few minutes the beads were allowed to settle and the supernatant discarded. This was repeated a second time, before the beads were finally resuspended in Medium 199 (25 ml/g beads).

Cells harvested from 4-5 confluent 200 ml culture flasks were added to the flask containing the beads. Ten percent (v/v) FBS was added to neutralize the effects of ATV and to promote the attachment of the cells to the beads (Grinell, 1976). The beads and cells were transferred to a 250 ml microcarrier vessel (Techne). For large cultures a 1.5 I microcarrier vessel was used, in which cells from 16- 20 200 ml culture flasks were added to 6 g beads.

The cultures were placed on a microcarrier stirrer (Techne) and set on intermittent stirring (2 minutes on, 30 minutes off at 40 rpm) for 24 hours to allow for maximal cell attachment, after which, the beads were continuously stirred at 40 rpm. On the second day, a further 100 ml of Medium 199 containing FBS (10% v/v) was added to the culture. Every second day, half of the culture medium was discarded and replaced by fresh medium, ^" thereby ensuring that maximal growth rates were achieved.

Constant monitoring of medium pH was required due to the rapid cellular growth rates. The medium pH was main¬ tained by the addition of NaHC0 ₃ to the small culture flasks. NaHCθ3 was shown to be incapable of maintaining pH levels in the large culture flasks for long periods. This problem was overcome by using Hepes (20 mM) as the buffering agent. A small aliquot from the culture was taken daily so that the rate of cellular growth on the culture beads could be measured. The beads were placed onto a glass slide between two pieces of a broken coverslip (which acted as a support for the coverslip placed over the beads), allowing for easy observation of the cells.

Cells grown in microcarrier culture were used for whole cell respiration studies and in the preparation of mitochondria. The details of the harvesting techniques used are described below. The cell cultures and culture media

were regularly inspected for presence of biological con¬ taminants. Infected cultures were removed and autoclaved to destroy the source of contamination. HeLa cell mitochrondria for use in studying respiration and glutamine (glutamate) oxidation were obtained as follows.

Confluent HeLa cell microcarrier cultures were the source of material used for preparing mitochondria. Con¬ fluent cultures were harvested and the mitochondria isolated as follows. The beads were allowed to settle and the media decanted off, after which, the beads were given two further washes of PBS A (pH 7.6; 100 ml/g beads). After this, the beads were resuspended in PBS A containing (0.02% w/v) EDTA (pH 7.6) at 50 ml/g beads, and stirred for 5-10 minutes (30 rpm) . At the end of this period, the beads were allowed to settle, and the media decanted off.

The beads were resuspended in a PBS A containing EDTA/ATV mixture (50:50) at 50 ml/g beads, and transferred to a siliconized 250 ml conical flask. The flask was occasion¬ ally agitated and kept in a 37°C incubator to ensure that maximal cell detachment occurred. Samples were taken and observed microscopically to monitor the rate of cell detach¬ ment. The beads were allowed to settle and the supernatant with the suspended cells was removed and the cells were cen¬ trifuged down at 180 g for 5 minutes on a Damon IEC centri¬ fuge. The cells were finally resuspended in a small volume of H-medium plus EDTA.

The above treatment was repeated 3 times to obtain maximum yields. The cells were pooled, and the mitochondria were isolated as described by a literature method, except that the digitonin treatment step was omitted. The mito¬ chondria obtained were resuspended in 500 μl H-medium and stored on ice prior to use. Mitochondria were kept on ice for a maximum period of 12 hours before being discarded. VII. DIAGNOSTIC APPLICATIONS

By now it will be apparent to those skilled in the art that many of the biological products such as monoclonal antibodies, polyclonal antibodies, and transporter proteins

SUBSTITUTE

can be used in assays for cancer. It will also be recognized that there are a number of methods for the diagnosis of the presence of glutamine or glutamine transporter in the circu _¬ lation (blood, plasma) or on cells or in any bodily secre _¬ tions including urine, tears, milk, feces, spinal fluid and the like. The basis for these procedures lies first in using monoclonal or polyclonal antibodies to glutamine transporter and, second, in using the glutamine transporter or part thereof as antigen in the systems. In this way, many dif _¬ ferent diagnostic systems can be established based on exist¬ ing immunological techniques. For example, antigen (glu _¬ tamine transporter) can adhere to a plate wherein antibodies will bind to this. The presence of the transporter (or anti¬ bodies thereto) in serum or other secretions will inhibit this binding and this can be measured either in ELISA assays, radioimmunoassays, immunoradiometric assay or by chemilumi¬ nescence or by many other means. Modifications of this simple antibody reaction are common and include the use of 2 antibodies - one as a "catcher" and one as the readout system for a "capture-tag" assay to capture circulating or secreted glutamine transporter. In addition, the transporter can be directly measured in tissues using fluorescence, immuno- peroxidase or radioactive methods. Many of these methods use plastic plates, adherence to beads, chemiluminescence and the like. The antigen,, which is glutamine transporter, can be used in whole or part, and includes synthetic peptide frag _¬ ments of the receptor or genetically engineered parts of the transporter. Additionally, as discussed above, many anti¬ bodies can be made to this transporter.

Combinations of diagnostic products encompassed by this invention will likely be used, for example, in a sand¬ wich or competitive assay. It will be appreciated that the diagnostic products of this invention are applicable to a wide range of assays and are not limited to use in any par¬ ticular assay or format. Exemplary assays include any type of: binding assay, sandwich assay, enzymeimmunoassay, fluoroimmunoassay, radioimmunoassay, competitive assay.

♦

SUBSTITUTE SHEET

immunoradiometric assay, immunoenzymatic assay, immuno- fluorometric assay, or luminescence immunoassay.

The diagnostic product can be labelled with con¬ ventional labels such as an enzyme, a radioactive isotope, a particle, a fluorescent molecule, a free-radical, a chemi- luminescent molecule, a bioluminescent molecule, a phage or a metal. Other labels may also be apparent to those skilled in the art.

It will be appreciated that if an amino acid trans¬ porter such as a tumour glutamine transporter is used as the diagnostic product, the transporter can be in its native or dissociated form, or a whole transporter protein or a frag¬ ment or subunit thereof.

It will also be readily appreciated that diagnostic products such as monoclonal antibodies, glutamine or ana¬ logues thereof can be conjugated to radioactive isotopes and used in vivo. The diagnostic product can then be injected into the patient, where it will selectively bind to, or be transported into the tumour. After a suitable time, the patient can be scanned, for example, by radio-imaging, to locate the presence of any tumours.

The diagnostic products of this invention will also be useful in performing assays on biological tissue such as a biopsy. For example, the tissue sample can be contacted with labelled monoclonal antibodies to permit binding. If labelled antibody remains bound to the tissue sample, then tumour cells are present in the sample.

By repeating the assay process over a period of time, and quantitating the label detected, a process is thus provided for monitoring the progress of cancer in a patient. The diagnostic products of this invention may also be useful as biosensors, and in particular, as immunological biosensors useful in detecting the presence of immunological binding pair members. Biosensors are devices that employ a biological molecule or tissue in its native form, modified form, or functional fragments thereof as components of a sensing apparatus. These molecules can be associated in

various ways with the signal transducing device e.g., incorporated into lipid membranes, coated onto metal par¬ ticles, optical fibres, piezoelectric crystals, or by other methods and means. Thus, it is possible to sensitively measure the presence of different materials by, for example, potentiometric measurement, fibre optics, surface plasmon resonance and pieozoelectric measurement.

The Hela cells, Hela cell membranes, glutamine transporters or fragments thereof or antibodies (or fragments thereof) to the glutamine transporters, can be used in asso¬ ciation with a signal transducing device in a biosensor. Where Hela cells or Hela cell membranes (e.g., vesicles) are used, the presence of ions, amino acids, e.g., glutamine, growth factors, etc. could be detected and monitored. Where the glutamine transporter, fragment thereof or modified form thereof is used as a component of the biosensor, the presence of glutamine can be detected or monitored. Antibodies or fragments thereof recognizing glutamine transporters can be used to detect the presence of the transporter or fragment thereof in cells or various biological fluid samples. These biological materials can be used individually and in combination. VIII. THERAPEUTIC APPLICATIONS

This invention provides several different thera¬ peutic compositions and methods for treating a cancer patient. The compositions and methods can generally be categorized as either a transport inhibitor and/or a cyto- toxic agent.

A. Transport Inhibitors

The main function of transport inhibitors is to deprive the tumour cell of the glutamine that is needed for normal biosynthesis. It is believed that if the cell can be substantially deprived of glutamine, it will weaken and perhaps eventually die, and/or become more susceptible to cytotoxic agents or the body's own immune defenses. For example, it is believed that glutamine may have an effect on glutamine synthetase, and that inhibition of glutamine into

SUBSTITUTE SHEET

- 130 -

the cell could affect glutamine synthetase as well as other cellular enzymes, e.g., glutaminase and ₇ -glutamyl-trans- ferase. This could potentially cause many different problems, of which some might be unrelated to cancer growth.

Further, compounds other than those described below, which are shown to inhibit glutamine transport, can also be used in accordance with this invention. That is, another embodiment of this invention simply comprises the use of compounds which inhibit glutamine transport in order to substantially selectively treat tumors. Thus, for example, compounds such as ouabain and mersalyl could also be used, although extensive studies for all such compounds will ordinarily be required.

1. Anti-glutamine compounds and glutamine analogues

One class of compounds which can inhibit glutamine uptake is the anti-glutamine compounds and glutamine ana¬ logues previously described. Of particular interest are those compounds and analogues which exhibit high transport inhibition, i.e., at least about 20% inhibition, preferably at least about 40%, and more preferably at least about 60% inhibition. It is also desirable that such compounds exhibit low K ' values, i.e., less than about 1.5 mM and preferably less than about 1.0 mM. Such values indicate that the ana¬ logue is a relatively highly specific binder for the glu¬ tamine transporter. If the analogue has a B value greater than about 1.0 nmol/mg protein, then the analogue will likely be transported into the cell. Otherwise, the analogue will likely just bind to the transporter binding sites, thereby inhibiting transport of other molecules into the cell. By depriving the cell of glutamine, the cell may be weakened and thus become susceptible to other cytotoxic agents and/or the body's own immune defense.

2. Antibodies

Monoclonal or polyclonal antibodies which bind to the active transport sites of the transporter will likewise inhibit the transport of glutamine into the cell.

Combinations of monoclonal and polyclonal antibodies, or products derived therefrom, may be used, as well as com¬ binations of antibodies and glutamine analogues (or other amino acid analogues) which are also transport inhibitors.

Using recently developed recombinant DNA methods, modified antibodies in accordance with this invention can be made. The antibodies can, for example, be modified by taking one or more combining sites from one antibody and putting it/ them on another antibody. Alternatively, the antibodies could be modified to add new sites for conjugation with other compounds, or to incorporate protein toxins as part of the overall antibody molecules. Many such variations will be apparent to those skilled in the art, and such variations are encompassed by this invention.

3. Histidine and other amino acids

Histidine has been shown to be a highly competitive inhibitor of glutamine uptake into HeLa cells, causing a 50% inhibition of the rate of glutamine uptake at 1 mM. The presence of high levels of histidine in the medium of growing cells could severely inhibit the transport of glutamine from the medium into the cell's cytoplasm. Under normal physio¬ logical conditions, histidine levels are low and so glutamine uptake into HeLa cells is not affected. However, if high levels of histidine were present in the medium, the entry of glutamine into HeLa cells could be impeded, subsequently causing a state of glutamine starvation within the cell.

Inhibition of glutamine uptake into the tumour cell would lead to a state of glutamine starvation and so would likely result in weakened cells. Histidine is a highly com¬ petitive inhibitor of glutamine uptake in HeLa cells and so would be expected to interfere significantly with the entry of glutamine into these cells, if present at sufficiently high concentrations. In this investigation, histidine was added to the medium of HeLa cells in culture and the growth of the cells was monitored over a period of time. The results shown in FIG. 60 suggest that in the presence of 1 mM histidine, the growth of the HeLa cells in the first 48 hours

was arrested relative to controls which were not exposed to high histidine concentrations. However, after 48 hours the cells appeared to recover and resumed growth rates which were comparable with controls. The reasons for the recovery of the cells after 48 hours are not clear and may be due to a shift in metabolism of the cells to compensate for the state of glutamine-starvation. However, this investigation pro¬ vides evidence that blocking the entry of glutamine into tumour cells is sufficient to cause some degree of trauma to the tumour, resulting in cells of a relatively weakened con¬ dition. Again, it could then be possible to take advantage of this weakened condition by administering known or novel anti-cancer drugs and thereby cause an additive or synergis¬ tic effect in the cytotoxicity toward tumour cells. It is also possible that the levels of known or novel anti-cancer drug required could be lowered significantly in the presence of the blocking agent, thereby decreasing the unwanted cyto¬ toxicity towards normal cells during treatment.

As seen in Table 15 (FIG. 86), other amino acids can also inhibit uptake of glutamine into HeLa cells, and thus may also have therapeutic value as inhibitors. The inhibition could be mediated by inhibiting uptake in any mechanism, i.e., TGT I-III, diffusion, etc.

B. Cytotoxic Agents

Several new types of cytotoxic agents are also provided by this invention.

1. Anti-glutamine compounds and glιτt--*πnne analogues

Glutamine analogues which exhibit good growth inhibition may be good cytotoxic agents. That is par¬ ticularly true if such analogues possess a relatively low K ' thereby indicating that it is likely selectively binding to tumour cells instead of normal healthy cells. It is also desirable that the cytotoxic glutamine analogues exhibit high

Bm values which would tend to indicate a substantial amount of analogue is making its way into the cell. A low B value would likely indicate that the analogue is simply binding to

SHEET

the transport protein on the outside of the cell membrane, but not being transported in. The modification of glutamine or analogues thereof to render them toxic is also an embodi¬ ment of this invention. This could be accomplished, for example, by incorporating known or novel toxic molecules, atoms or groups into the glutamine or analogue structure.

2. Antibodies linked to cytotoxic agents Another embodiment of this invention provides mono¬ clonal or polyclonal antibodies to tumour glutamine trans¬ porter, which antibodies are conjugated to novel or known cytotoxic agents. The known cytotoxic agents can be, for example, chemotherapeutic agents which are otherwise insuf¬ ficiently selective, or radioisotopes. The antibodies could also be conjugated to the novel anti-glutamine compounds or analogues of this invention. When chemotherapeutic agents are linked to the antibodies of this invention, it is prefer¬ able that the chemotherapeutic be modified such that if the bond breaks between the monoclonal and the chemotherapeutic, the chemotherapeutic will not enter any untargeted cell. Thus, the chemotherapeutic would only be effective if it remained conjugated, and thus it would retain selectivity. Other similar approaches using "prodrugs" will be apparent to those skilled in the art and are likewise encompassed by this invention.

3. Agents linked to glutamine or analogues By linking known or novel cytotoxic agents to glu¬ tamine, anti-glutamine compounds or glutamine analogues, the possibility exists that the glutamine or other compound will selectively transport the agent into the tumor cell cyto¬ plasm, where it can kill or weaken the cell. Such agents can include radioactive isotopes, drugs or toxins. To be suc¬ cessful, the link and the agent must be designed so that they do not destroy the ability of glutamine or other compound to reversibly bind and be transported. The glutamine or ana¬ logue thus targets the toxic agents to cancer cells.

SUBSTITUTE SHEET

C. Other Compounds

It is recognized that a crutial function of tumour cells is the binding of glutamine to the glutamine trans¬ porter, transport of the glutamine across cells, and subse¬ quent metabolism of glutamine inside the cell. Several compounds, compositions and methods for inhibiting this process such as the use of glutamine analogues, antibodies, histidine, and cytotoxic agents, have been described. How¬ ever, those of ordinary skill in the art, upon reading this specification, will immediately realize other substances, including enzymes, which in any way inhibit the function of a glutamine transporter in its role of transporting glutamine or analogues into the cell. Such substances are accordingly within the scope of this invention. Also included with the scope of this invention are the inhibitors of the subsequent metabolism of glutamine inside the cell. It is recognized that many of the analogues described are likely to operate in this function, i.e., not merely blocking the binding of glu¬ tamine to the transporter, but also inhibiting the pathways of glutamine metabolism inside the cell.

It is also quite possible that other amino acids may be transported by the glutamine transporters, and that their analogues may be even more potent inhibitors or toxic agents than the glutamine analogues. For example, histidine was shown above to inhibit glutamine uptake. Accordingly, therapeutic products such as amino acids, analogues thereof, or other compounds that enter via the glutamine transporters, e.g., TGT I-III, are within the scope of this invention.

D. Combination Therapy

The therapeutic applications of this invention lend themselves to combinations of therapy, particularly where one type of compound is used to inhibit the transport of glu¬ tamine, thereby weakening the cell, and the remainder of the therapy comprises administering a known or novel cytotoxic agent, including those described above, which can take advantage of the weakened tumour cell.

E. Pharmaceutical Compositions and Methods of Treatment

Another embodiment of this invention comprises pharmaceutical compositions comprising one or more of the anti-glutamine compounds described above in association with a physiologically acceptable adjuvant or carrier. Such com¬ positions are produced by processes which are known to those skilled in the art. Thus, the physiologically active anti- glutamine compounds according to this invention can be used either as such, or preferably, in combination with suitable pharmaceutical auxiliaries, in the form of tablets, dragees, capsules, suppositories, emulsions, suspensions or solutions.

Those skilled in the art will also be familiar with the auxiliaries which are suitable for the desired pharmaceu¬ tical formulations. In addition to solvents, gelling agents, suppository bases, tableting auxiliaries and other excipients for active ingredients, it is also possible to use, for example, antioxidants, dispersing agents, emulsifiers, anti- foaming agents, flavor correctants, preservatives, solubiliz- ing agents and colorants. The optimum dosage and method of administration of the active anti-glutamine compounds, ana¬ logues or compositions required in each particular case can be determined by those skilled in the art through routine experimentation.

Anti-glutamine compounds, analogues or compositions in the form of pharmaceutical formulations according to this invention, can also contain one or more physiologically active members of other groups of medicaments, such as « steroidal and/or non-steroidal anti-inflammatory agents,

* immunosuppressants, sulfated glycosamines/glycans and other sulfated carbohydrates, analgesics and antipyretics.

The anti-glutamine compounds, analogues or compo¬ sitions of the present invention may be administered orally, rectally, or by injection, for example, by transdermal application for the treatment of skin cancer, in the form of pharmaceutical preparations comprising at least one active compound in association with a pharmaceutically acceptable

carrier, which may be a solid or semi-solid or liquid diluent or capsule or aerosol applicator. The active components may constitute between 0.1 and 99% by weight of a solid/semi- solid/liquid preparation, more particularly, between 0.5 and 20% by weight for preparations intended for injection, and between 2 and 50% by weight for preparations suitable for oral administration.

Dosage unit pharmaceutical preparations containing at least one compound or composition in accordance with the invention for oral application, may be prepared by mixing th selected compound or composition with a solid pulverulent carrier such as lactose, saccharose, sorbitol, mannitol, starches such as potato starch, corn starch or amylopectin, cellulose derivatives, or gelatine, and a lubricant such as magnesium stearate, calcium stearate, polyethylene glycol waxes, then compressed to form tables. Coated tables can be prepared by coating the tablets prepared as described above, with a concentrated sugar solution which may contain com¬ ponents such as gum arabic, gelatine, talcum, titanium dioxide, or the tablet can be coated with a lacquer dissolve in a readily volatile organic solvent or mixture of organic solvents.

Soft gelatine capsules can be prepared by enclosin the selected compound or composition, mixed with a vegetable oil, in a soft gelatine shell. Hard gelatine capsules may contain the selected compound or composition in admisture with solid, pulverulent carriers such as lactose, saccharose sorbitol, mannitol, starches such as potato starch or corn starch, or amylopectin, cellulose derivatives or gelatine.

Dosage unit preparations for rectal application ca be prepared in the form of suppositories comprising the active substance(s) in admisture with vegetable oil or paraffin oil.

Liquid preparations for oral application can be i the form or syrups or suspensions, such as solutions contai ing from about 0.2% to about 20% by weight of the selected

compound or composition, the balance being sugar and a mixture of ethanol, water, glycerol and propylene glycol.

Solutions for parenteral application by injection can be prepared as an aqueous solution of the selected com¬ pound or composition, preferably in a concentration of from about 0.5% to about 10% by weight. These solutions may also contain stabilizing agents and/or buffering agents and may conveniently be provided in various dosage ampoules.

Suitable transdermal daily-dose administration of the selected compounds or compositions in accordance with the invention can be 100-500 mg, preferably 200-300 mg, while weekly-dose administration can be in dosage of 25-2000 mg every 1 to 3 weeks.

The anti-glutamine compounds and analogues of this invention can be expected to be useful in the treatment of cancer. The invention thus also relates to a method of treating animals suffering from cancer. The method comprises administering to an animal a therapeutically active and phys¬ iologically acceptable amount of one or more of the therapeu¬ tic compounds or compositions described above, or combina¬ tions of therapeutic compounds and compositions as also described.

F. Immunosuppressants and Immunostimulants

Glutamine has been shown to be taken up by stimu¬ lated lymphocytes. Accordingly, the anti-glutamine compounds and analogues of this invention may also be useful as immuno¬ suppressants and immunostimulators. Such immunosuppressants and immunostimulators would be useful for treating graft rejection, graft vs. host disease, autoimmune diseases, immune diseases such as AIDS, etc. Those skilled in the art will readily appreciate many applications.

Because lymphocytes transport glutamine, the anti- glutamine compounds and glutamine analogues of this invention could thus be transported into the lymphocytes to either ablate proliferating lymphocytes, or to stimulate lympho¬ cytes, depending upon the desired effect. Anti-glutamine compounds and glutamine analogues would simply have to be

screened, as previously described, to determine their effects on lymphocytes. Lymphocyte glutamine transporter(s) could likely be isolated and identified as discussed above, e.g., for TGT II. Pharmaceutical formulations and methods of treating humans would be the same or similar to that described above.

IX. VACCINES

Another embodiment of this invention provides vaccines for immunization against cancer. The vaccine composition preferably comprises the entire transport protein in its native form, or at least the subunit thereof which is responsible for actively transporting glutamine. Several different transporters, fragments or subunits thereof can be co-administered to generate an immune response in the host to the several different transporters. For example, a triple vaccine can be administered which contains TGT I, TGT II and TGT III, or the fragments or subunits thereof responsible for active transport. The host would then develop an immune response to the transporters. At the first appearance of such transporters, the host would provide antibodies which would bind to the transporters, thereby inhibiting transport of glutamine into the tumour cell in its early stages. The tumour cell may then become vulnerable to the host's immune defenses.

X. EXAMPLES

The following Examples 1-79 illustrate, but do not limit in any manner, preparation of the anti-glutamine com¬ pounds and glutamine analogues of this invention. The general experimental procedures, instruments, solvents and reagents used in Examples 1-79 were as follows.

Diethyl ether and tetrahydrofuran were distilled from sodium, followed by a second distillation from sodium in the presence of a small amount of benzophenone. Pyridine was dried over potassium hydroxide pellets and distilled prior to use. Triethylamine was dried over potassium hydroxide pellets and distilled prior to use. The distilled product was stored over Linde type 13X molecular sieves in a dark bottle.

Methanol was dried by warming a small amount with magnesium turnings and iodine until all of the iodine had disappeared. More methanol was then added and the product distilled under dry nitrogen. For some experiments HPLC grade methanol was substituted for anhydrous methanol. Absolute ethanol, 95% ethanol, ethyl acetate (AR grade) and 30°-40°C b.p. petrol (AR grade) were used as supplied. Dimethylformamide was dried over Linde type 4A molecular sieves and distilled in vacuo (b.p. 76°C/40 mm). The distillation product was stored over molecular sieves in a dark bottle until used. Aniline was distilled in vacuo from potassium hydroxide pellets and stored over Linde type 4A molecular sieves in a dark bottle prior to use. Benzyl bromide was distilled in vacuo (b.p. 85-87°C/10 mm) and stored in bottles wrapped in aluminium foil at 4°C until used. Acetic anhydride was distilled (b.p. 138°C) prior to use. Distilled acetic anhydride was stored under nitrogen.

Chromatography

Thin layer chromatography (t.1.c. )

The ascending solvent system was employed using one of the systems outlined below. Standard samples and co-spots were run where possible and Rf values are quoted, (i). Commercial silica gel plates (Merck, DC-Plastikfolien Kieselgel 60F254); solvent systems are indicated in paren¬ theses. Plates were viewed under ultraviolet light and developed by submersion in 5% ethanolic phosphomolybdic acid followed by oven heating to 100°C

(ii). Commercial cellulose plates (Merck, DC-Plastikfolien Cellulose); solvent system n-butanol/pyridine/ 0.4% acetic acid in water (22:10:10). Plates were developed by submer¬ sion in 1% ninhydrin in acetone to which 1% pyridine had been added followed by heating to 100°C

Vacuum assisted chromatography (Flash chromatography) TLC silica gel (Merck, kieselgel 60G) absorbent was dry packed into columns fitted with a sintered glass disc. Vacuum assisted elution was employed in the usual manner.

SUBSTITUTE SHEET

Ion exchange chromatography

Both anion and cation exchange media were employed to effect separations. These media were packed into ion exchange columns as slurries in water, and converted to the desired ionic form by treatment with a 1 molar solution of a soluble salt of the appropriate ion. The columns were then washed with distilled water and the mixture loaded. Eluent systems are indicated in parentheses. The ion exchange media referred to are listed below.

I. Amberlite IR-120 (analytical) ,BDH.

II. Amberlite IRC-50 (analytical) ,BDH.

III. Amberlite CG-120 (analytical),BDH.

IV. Amberlite CG-120 (analytical) ,BDH.

V. Bio-Gel Bio-rex 5 (analytical) ,Bio-Rad.

Gel filtration (ion exclusion) chromatography Sephadex LH-20 was stored in methanol prior to use, and loaded as a methanoic slurry into a gravity chromatogra _¬ phy column. The gel was washed with eluent until the bed had settled, then the mixture applied as a thin layer. Eluents were mixtures of dichloromethane and petrol; the proportions are indicated in parentheses.

High performance liquid chromatography (HPLC) A WATERS HPLC system consisting of a model 590 pump with U6K injector and R401/(402) differential refractometer as well as a model 481 variable wavelength uv/vis. detector was used. HPLC columns are referred to as listed below and solvent systems are indicated in parentheses. Gradient elu _¬ tion was effected with a Varian duel solvent pump and the same injector/detector configuration.

I. analytical silica

II. preparative silica

III. analytical reverse phase

The purity of the final amino acid derivatives were determined by HPLC, using a WATERS amino acid analyzer.

EET

Medium pressure liquid chromatography (MPLC)

MPLC was effected using a FMI-RP pump with a Rheodyne type 50 teflon injector and a WATERS R402 pre¬ parative differential refractometer. The columns used are referred to as listed below and the solvent system is indicated in parentheses.

I. Fertigsaule Kieselgel 60, size C (MERCK)

II. Lobar Fertigsaule Kieselgel 60, size B, 40-63 m (MERCK)

III. Lobar LiChromoprep RP-8, size A, 43-60 m (MERCK)

Infrared (IR. Spectra

All IR spectra were recorded using a Perkin-Elmer model 297 grating spectrophotometer. Samples were recorded as liquid films (l.f.) or nujol mulls (nujol) between sodium chloride discs, or as solutions in either carbon tetra chlo¬ ride (CC1 ) or deuterochloroform (CDC1 ₃ ) using 1 mm sodium chloride cavity cells. The major absorptions are reported in cm ^" .

Nuclear Magnetic Resonance (NMR) Spectra -Η. NMR spectra

Proton NMR spectra were recorded using a Bruker AM- 300 spectrometer (300.133 MHz). Chemical shifts are quoted as values relative to tetramethylsilane (TMS) internal stan¬ dard, in deuterochloroform (CDC1 ₃ ) solvent. Other solvents used were deuteromethanol (CD ₃ OD) and deuterium oxide (D ₂ 0) . Solvents and standards are indicated in parentheses. Coupling constants (J) are reported in Hertz, with signal multiplicity designated as singlet(s), doublet(d), triplet(t), quartet(q), quintet(qi), sextet(sx), multiplet(m) and broad(b). Demonstrated couplings are indicated diagram- matically. Nuclear Overhauser effect experiments (NOE) are indicated where performed.

¹³ C NMR spectra

Carbon-13 NMR spectra were recorded using a Bruker AM-300 spectrometer (75.47 MHz). Both completely ^~ E decoupled and DEPT or off-resonance spectra were run. All parameters are quoted in an analogous way to the proton

spectra. Two dimensional. Cosy and XY-correlation experiments are indicated where performed.

Melting Points (m.p.)

Melting points were determined using a Reichert hot stage apparatus and an Olympus microscope. All melting points are uncorrected and are quoted in degrees Centigrade.

Optical Rotations

The optical rotations of chiral compounds were measured using a Perkin-Elmer model 141 polarimeter. The D line of sodium was used for all measurements unless otherwise noted. Solvents are quoted in parentheses, as are concentra¬ tions in mg/ml. All measurements were carried out in a 1 ml jacketed quartz cell of 10 cm path length. Five measurements were taken for each sample and the average rotation calculated.

Miscellaneous

All reactions were carried out at room temperature o

(average 25 ) and stirred under an atmosphere of dry nitrogen (N ₂ ) unless otherwise indicated.

Compounds Nos. 1-78, which are prepared in Examples 1-79, are as follows:

Compound No, Name 1 iV- carbobenzyl oxy-L- gl u tami c anhydri de 2 Z-L-pyroglutamic acid 3 Z-L-Pyroglutamic acid t-butyl ester 4 Z-L-glutamic acid η -methyl ester 5 Z-L-glutamic acid α-t-butyl—η -methyl ester

6 Z-L-Glutamic acid α-t-butyl ester 7 Z-L-Glutamic acid -Benzyl ester 8 Z-L-Glutamic acid α-benzyl, ₇ -2'- pyridyl ester

9 Z-L-Glutamic acid η -hydroxamate 10 L-Glutamic acid η -hydroxamate 11 Z-L-glutamic acid η-2' -hydroxyethylamide 12 L-glutamic acid η- 2' -hydroxyethylamide

13 Z-L-Glutamic acid α-t-butyl ester η - methyl amide

14 Z-L-Glutamic acid ₇ - methyl amide 15 L-Glutamic acid ₇ -methylamide 16 Z-L-Glutamic acid α- t-butyl ester ₇ - hydrazide

1 7 Z-L-Glutamic acid α-t-butyl ester η-N' - methylhydrazide

18 Z-L-Glutamic acid η -N' -methylhydrazide 19 L-Glutamic acid η-N' -methylhydrazide 20 Z-L-Glutamic acid α-t-butyl ester η -2' - hydroxye thyl hydra zide

21 Z-L-Glutamic acid η -2' -hydroxy -ethyl- hydra zide

22 L-Glutamic acid η -2' -hydroxyethyl- hydrazide

23 Z-L-Glutamic acid α-t-butyl ester ₇ - hydroxamate

24 Z-L-Glutamic acid α-t-butyl ester ₇ - anili ^' de

25 Z-L-Glutamic acid η -anilide 26 L-Glutamic acid η-anilide 27 Z-L-Glutamic acid α-benzyl ester η -2' - hydroxyethyl amdi e

28 Z-L-Glutamic acid α-benzyl-η -2' amino¬ ethyl ester

29 L-glutamic acid η -2' -aminoethyl ester 30 Z-L-Glutamic acid α-benzyl ester

7- tris (hydroxymethyl) methylamide

31 L-Glutamic acid η -tris (hydroxymethyl) methylamide

32 L-Glutamic acid η -tris (hydroxymethyl ) methylamide - cyclized product .

33 Z-L-Glutamic acid α-benzyl ester η -bis (hydroxyethyl ) ami de

34 L-Glutamic acid η -bis (hydroxyethyl) amide

SUBSTITUTE SHEET

35 Z-L-Glutamic acid α-benzyl ester η- 1 ' , 1 ' -dimethyl-2 ^r -hydroxyethylamide

36 L-Glutamic acid η-1 ' , 1 ' -dimethyl- 2' - hydroxye thyl ami de

37 Z-L-Glutamic acid α-benzyl ester η- ethylamide

38 L-Glutamic acid η-ethylamide 39 Z-L-Glutamic acid α-benzyl ester η - diethylamide

40 ' L-Glutamic acid η-diethylamide

41 Z-L-Glutamic acid α-benzyl ester η -N- ( 2 ' hydroxyethyl ) piper azide

42 L-Glutamic acid η-N- ( 2' -hydroxyethyl ) piper zide

43 Z-L-Glutamic acid α-benzyl ester η- morpholide

44 L-Glutamic acid η-morpholide 45 Z-L-Glutamic acid α-benzyl ester η-O- methylhydroxamate

46 L-Glutamic acid η-O-methylhydroxamate 47 Z-L-Glutamic acid α-benzyl ester η-2' thioethylamide

48 L-Glutamic acid η-2' thioethylamide 49 Z-L-Glutamic acid α-benzyl ester η-2' methoxyethylamide

50 L-Glutamic acid η-2' methoxyethylamide 51 Z-L-Glutamic acid α-benzyl ester η-2 ^r chlorethylamide

52 L-Glutamic acid η-2' chloroethylamide 53 Z-L-Glutamic acid α-benzyl ester η-2' hydroxyani lide

54 L-Glutamic acid η-2' -hydroxyani lide 55 Z-L-Glutamic acid α-benzyl ester η-3' - chloroanilide

56 L-Glutamic acid η-3' chloroanilide 57 Z-L-Glutamic acid α-benzyl ester η-4' - chloranilide

⁵⁸ L-Glutamic acid η -4 ' -chloroanilide

⁵⁹ Z-L-Glutamic acid α-benzyl ester _η - piperidide

60 L-Glutamic acid η-piperidide 61 Z-L-Glutamic acid α-benzyl ester _η -3' - hydroxypiperi dide

62 L-Glutamic acid η -3' -hydroxypiperidide

63 Z-L-Glutamic acid α-benzyl ester _η -

[1 (4-methylpiperazino) Jamide

⁶⁴ L-Glutamic acid η -[ 1- ( 4-methylpipera¬ zino) Jamide

65 Z-L-Glutamic acid α-benzyl ester _η - (4- morph olino) ami de

66 L-Glutamic acid η - (4-morpho lino) amide

67 Z-L-Glutamic acid α-benzyl ester η-

2' fluoroethylamide

68 Z-L-Glutamic acid α-benzyl ester _η - 2' , -_" , 2' -trifluoroethylamide 69 L-Glutamic acid η -2' , 2' , 2' -tri fluoro¬ ethylamide 70 Z-L-Glutamic acid α-t-butyl ester η -O- methylhydroxima te

⁷¹ Z-L-Glutamic acid η -O-methylhydroximate

⁷² L-Glutamic acid η -O-methylhydroximate

⁷ 3 Z-L-Glutamic acid α-benzyl ester η - (2- pyridyl )methylamide

74 L-Glutamic acid η- ( 2' -pyr idyl ) methyl¬ amide

* 75 Z-L-Glutamic acid α-benzyl ester η- prop-2-enamide (allyl amide) -* 76 L-Glutamic acid η -propylamide

77 Z-L-Glutamic acid α-benzyl ester η - (2- mesyloxyethyl ) amide

78 Dihyrooxazole analogue Example 1

N-carbobenzyloxy-L-glutamic anhydride (1) (Z-L-glutamic anhydride) (see for example Le Quesne, W. J. & Young, G. T.

(1954) J. Chem. Soc , 1950. :Klieger, E. & Gibian H. (1961 ) Justus Liebig' s Ann . Chem. , 640, 145-157. )

A mixture of Z-L-glutamic acid (25.0 g, 0.089 mol) in acetic anhydride (150 mL) was stirred for 6 h, until all the Z-L-glutamic acid had dissolved. The reaction was worked up by removal of the excess acetic anhydride in vacuo (30°, 2mm) to yield an oil. The oil was washed with dry ether (2 x 50 mL) , followed by dry petrol (2 x 50 mL), and the pure product dried in vacuo to yield a white crystalline solid (23.1 g, 99%), m.p. 91-94° (lit. Klieger, E. & Gibian.H. (1961) Justus Liebig's Ann. Chem., 640, 145-157. m.p. 92-93°). Repetition of this prepartion yielded the desired product in 95-99% yield. No further attempt was made to purify the anhydride at this stage. Example 2

Z-L-pyroglutamic acid (2) (see for example: Klieger, E. & Gibian. H. (1962) Justus Liebig' s Ann. Chem. , 655, 195-211. )

A solution of the anhydride(l) (19.1 g, 0.072 mol) and dicyclohexylamine (14.5 mL, 0.088 mol) in dry ether (60 mL) and dry THF (40 mL) was stirred overnight under nitrogen. The solid formed in this reaction was filtered off and washed with ether, followed by recrystallisation from boiling methanol to yield the pure white DCHA salt of Z-L-pyroglu¬ tamic acid (21.0 g, 80%). A solution of the DCHA salt (21.0 g) in ethyl acetate (120 mL) and hydrochloric acid (IM, 120mL) was stirred overnight to form the free acid. DCHA hydrochloride was filtered off and the fixture extracted with ethyl acetate (3 x 100 mL) . The combined organic phases were dried over sodium sulphate and the solvent removed in vacuo to yield the white crystalline product (17.50 g, 68% over¬ all), m.p. 136-138° (lit. Klieger, E. & Gibian. H. (1962) Justus Liebig's Ann. Chem., 655, 195-21. 137-139°). Sub¬ sequent repetition of this reaction returned the desired product in 65-75% overall yield. Proton and ¹³ C NMR spectra compared well with literature values (Klieger, E. __ Gibian.H. (1962) Justus Liebig's Ann. Chem. 655, 195-211).

'H HMR (CDCls): δ 2.10 (lH m), 2.50 (3H ), 4.72 (1H, dd J=9.3,2.7), 5.30 (2H s), 7.35 (5H ) . ¹³ C NMR (CDCI3): δ 21.72 (CH2), 30.99 (CH2), 58.44 (CH), 68.56 (CH2), 128.04 (aromatic CH), 128.45 (aromatic CH), 128.59 (aromatic CH) , 134.88 (aromatic C), 151.09 (carbamate), 172.98 (amide), 175.62 (acid). Example 3

Z-L-Pyroglutamic acid t-butyl ester (3) (see for example: Klieger, E. & Gibian . H. (1962) Justus Liebig' s Ann . Chem. , 655, 195-211 . . 'Anderson , G. W. & Callahan . F.M. (1962) J. Am. Chem. Soc , 82, 3359-3363. ) Method A:

Z-L-Pyroglutamic acid (10 g, 0.038 mol) was dis¬ solved in dichloromethane (DCM)(120 mL) saturated with iso- butylene. Sulphuric acid (98%, 3 mL) was slowly added and isobutylene bubbled through the solution for 1 h. After this time no more starting material was observed by t.l.c (silica, 20% petrol/ether). The DCM solution was washed with water (2 x 50 mL) followed by sodium bicarbonate solution (10%, 2 x 50 mL) and dried over sodium sulphate. Removal of the solvent in vacuo yielded a mixture of two products which were separated by flash chromatography (20% petrol/ether) to yield the ester (3) (Rf 0.3, 6 g, 50%), m.p. 55-57°, [α ] ²⁰ -29.0 (c = 2.04, DMF) and α, ₇ -di-t-butyl-Z-L- glutamate (Rf 0.3, 6 g, 18%). Subsequent repetition of this reaction produced varying amounts of the di-ester by-product, with yields of the desired ester between 40 and 75%. Sub¬ stitution of perchloric or para-toluenesulphonic acids or boron trifluoride diethyl etherate for the sulphuric acid catalyst failed to improve the yield. Method B:

Z-L-pyroglutamic acid (20 g, 0.076 mol) was taken up in t-butyl acetate (300 mL) . To this solution was added perchloric acid (70%, 1 mL) and the mixture stirred at 0° in a stoppered flask overnight. The reaction was worked up by dilution with ethyl acetate (150 mL) and washing with water

SUBSTITUTE SHEET

(2 x 50 mL), sodium bicarbonate soution (10%, 2 x 50 mL) and saturated sodium chloride solution (1 x 50 mL) . The reaction solution was dried over sodium sulphate and solvents removed in vacuo to yield pure (3) (Rf 0.3 20% petrol/ether, 9.31 g, 39%), m.p. 57-59°. This compound was identical by NMR ( ^X H and ¹³ C) with the product of the isobutylene reaction. ^X H NMR (CDCLs): δ 1.38 (9H s) , 2.03 (1H dddd J=2.7, 3.2, 9.5, 13.1), 2.31 (1H dddd J=9.2, 9.3, 10.5, 13.1), 2.49 (lH ddd J=3.2, 9.2, 17.5), 2.63 (1H ddd J=9.5, 10.5, 17.5), 4.54 (1H d,d J=2.7, 9.3), 5.23, 5.29 (2H AB J=12.3), 6.50 (lH b) 7.36 (5H m) .

¹³ C NMR (CDC1 ₃ ): δ 21.58 (CH2), 27.52 (CH3), 30.70 (CH2), 59.14 (CH) , 67.85 (CH2), 82.18 (C), 127.86 (aromatic CH) , 128.09 (aromatic CH) , 128.25 (aromatic CH) , 134.90 (aromatic C), 150.64 (car- bamate), 169.86 (amide), 172.78 (ester). Example 4

Z-L-glutamic acid η-methyl ester (4) (see for example: Hanby, W. E. , Waley, S. G. , Watson, J. & Ambrose. E. J. (1950) J. Chem. Soc , 3239-3249. ) Method A:

Protection of L-glutamic acid ₇ -methyl ester with benzyl chloroformate. (see: Hanby, W.E., Waley, S.G. Watson, J. & Ambrose.E.J. (1950) J. Chem. Soc, 3239-3249.)

L-glutamic acid ₇ -methyl ester (5.0 g, 0.031 mol) was taken up in sodium bicarbonate solution (20%, 20 mL) and the pH adjusted to 9.0. To this solution was added benzyl chloroformate (4.84 mL, 0.034 mol) and the mixture cooled to 5°. The reaction mixture was allowed to stir overnight after which time no starting ester remained as shown by t.1.c (cellulose, Rf 0.24). The reaction was worked up by adjust¬ ment of the pH to 2.0 and extraction with ethyl acetate (3 x 60 mL) . The combined organic layers were washed with water (10 mL) and dried over sodium sulphate. Removal of the solvent in vacuo furnished a white crystalline solid (8.0 g, 87.5 %), m.p. 71-73° (lit. Hanby, W.E., Waley, S.G., Watson,

J. & Ambrose. E.J. (1950) J. Chem. Soc, 3239-3249. 72-73°), [ ] _D ²⁰ -14.3 (c = 10.60, methanol)(lit.5 [__] _D ²⁰ -15.3 (c = 7.46, 1.4M KHC0 ₃ )), which was pure by t.l.c. (cellulose, Rf f 0.80).

- H NMR (CDCls): δ 2.02 (1H dddd J= 6.6, 7.8,

* 14.3, 14.3), 2.23 (H dddd J= 5.5, 7.4, 14.3, 14.2),

2.45 (2H m), 3.65 (3H s) , 4.42 (1H ddd J=5.5, 7.8, 8.0), 5.11 (2H s), 5.61 (1H d J=8.0), 7.30 (5H m) , 8.53 (1H b) .

Method B: Attempted selective esterification of Z-L-glutamic acid.

Z-L-Glutamic acid (5.62 g, 0.02 mol) was taken up in anhydrous methanol (20 mL) . To this solution was added acetyl chloride (1.60 mL, 0.024 mol) and the mixture stirred overnight (see Hanby, W.E., Waley, S.G., Watson, J. & Ambrose.E.J. (1950) J. Chem. Soc, 3239-3249.). After this time the solution was pale yellow in colour. Pyridine (2 mL) was added and the mixture stirred for a further 24 h. T.l.c (silica, 5% petrol/acetic acid (0.1) in ether) indicated the major product was the dimethyl ester (Rf 0.58) which was confirmed by H NMR spectroscopy. Example 5

Z-L-glutamic acid α-t-butyl-methyl ester (5) (see for example: Tashner, E. , Wasielewski , C. , Sokolowska, T. & Biernat . J. F. (1961 ) Justus Liebig' s Ann . Chem. , 646, 127- 133. )

The methyl ester (4) (7.00 g, 0.024 mol) was treated with t-butyl acetate as outlined above to yield a mixture of

* two products as shown by t.l.c (silica, 60% ether/petrol, Rf 0.84, 0.22). These components were separated by flash chro- j matography (silica, 20% ether/petrol) to yield the α-t-butyl ester (5) (4.90 g, 58.7%) as a viscous oil (Rf 0.22, t.l.c silica, 60% ether/petrol) . The higher Rf component appeared to be the α, ₇ -di-t-butyl ester.

-Η. NMR (CDC1 ₃ ): δ 1.45 (9H s) , 1,97 (1H dddd J = 7.9, 8.1, 14.2, 14.3), 2.23 (lh bddd J = 6.5, 8.3, 14.3), 2.39 (2H m), 3.66 (3H s), 4.28 (1H ddd J =

SUBSTITUTE SHEET

- 150 -

6.5, 7.5, 7.9), 5.10 (2H s), 5.37 (1H d J = 7.5),

7.34 (5H m) . Example 6

Z-L-Glutamic acid α-t-butyl ester (6) (Tashner, E. , Wasielewski , C. , Sokolowska, T. & Biernat. J. F. (1961 ) Justus Liebig' s Ann . Chem. , 646, 127-133. ) Method A:

Attempted ring opening of (3) (see for example: Klieger, E. & Gibian.H. (1962) Justus Liebig's Ann. Chem., 655, 195-211.)

The t-butyl ester (3) (500 mg, 1.56 mmol) was take up in acetone (7 mL), to which was aded sodium hydroxide solution (IM, 1.9 mL) . The solution was refluxed for lh and monitored by t.l.c (silica 80% ether/petrol). After this time little starting material was observed, and the reaction was worked up by acidification to pH 1.5 with hydrochloric acid (0.2M) followed by extraction with DCM (3 x 25 mL) . Th combined organic layers were dried over sodium sulphate and solvents removed in vacuo to yield a mixture of two products (Rf 0.05, 0.78, silica 80% ether/petrol) . The mixture was separated by flash chromatography (silica, DCM followed by ethyl acetate) to yield starting ester (3) (220 mg, 44%) and the t-butyl ester (6) (328 mg, 65%). Repetition of this experiment in methanol as the solvent yielded the 5-methyl ester in 35-50% yield. Example 7

Method B: Hydrolysis of (5)(see for example: Tashner, E., Wasielewski, C, Sokolowska, T. & Biernat.J.F. (1961) Justu Liebig's Ann. Chem., 646, 127-133; Shin, C.G., Yonezawa, Y _ Wanatabe.E. (1985) Tetrahedron Lett., .26, 85-88.)

The diester (5) (4.90 g, 1.39 mmol) was dissolved in methanol (20 mL) and water (5 mL). To this solution was added lithium hydroxide monohydrate (0.64 g. 1.42 mmol) and the mixture stirred for lh at 0°. After this time the sol¬ vents were removed in vacuo (at ca. 20°) and the residue taken up in water (50 mL) . This solution was extracted wit

ethyl acetate (2 x 20 mL) then acidified to pH 1.5 and re- extracted with ethyl acetate (3 x 20 mL) . The latter extracts were dried over sodium sulphate and solvents removed in vacuo to yield the t-butyl ester (6) (3.27 g, 70%, Rf 0.50 t.l.c silica, 60% ether/petrol). Subsequent repetition of this preparation furnished (6) in 70-80% yield.

- H NMR (CDCls): δ 1.45 (9H s), 1.98 (1H dddd J = 7.9, 8.4, 14.2, 14.3), 2.20 (lH m), 2.40 (2H m), 4.30 (1H ddd J = 8.1, 7.9, 6.5), 5.09 (2H s), 5.59 (1H d J = 8.1), 7.31 (5H m) , 7.75 (1H bs) . Example 8

Z-L-Glutamic acid α-Benzyl ester ( 7) (see:Morley, J. S. (1967) J. Chem. Soc. (C). , 2410-2421 . )

Z-L-Glutamic acid (50.0 g, 0.178 mol) was dissolved in dimethylformamide (DMF) (50 mL) at 15°. To this solution was added triethylamine (24.64 mL, 0.178 mol) and benzyl bromide (23.24 mL, 0.195 mol). The reaction mixture was maintained at 15° for 6 h then allowed to warm to room tem¬ perature overnight. After this time, ice water (300 mL) was added and the mixture extracted with ethyl acetate (3 x 160 mL) . The combined extracts were washed with ice water (2 x 50 mL) then dried over sodium sulphate. Solvents were removed in vacuo and the residue taken up in ethyl acetate (180 mL) . To this solution was added dicyclohexylamine (38.67 g, 42.36 mL, 0.214 mol) and the mixture stirred for 6 h. After this time the resulting soid was collected by fil¬ tration and washed with cold ethyl acetate (60 mL) before being dried in a vacuum oven to yield the dicyclohexylamine salt as a hygroscopic white solid (97.8 g, 99% ) . The crude product was recrystallised from boiling ethanol (250 mL) to yield the pure salt (66.5, 68%) which was taken up in ethyl acetate (300 mL) and hydrochloric acid (3.5M, 70 mL, 2eq.) and stirred for 2 h. The mixture was filtered and the layers separated. The aqueous layer was extracted with ethyl acetate (2 x 50 mL) and the combined organic layers washed with chilled water (50 mL) and dried over sodium sulphate. Removal of the ethyl acetate in vacuo yielded a colourless

oil which crystallised on trituration with diethyl ether/ petrol (1:1, 25 mL) . This product was dried at the pump for 4 h to furnish the α-benzyl ester (7) (44.5 g, 99.5%, 67.5% overall), m.p. 96-98° (lit. Morley, J.S. (1967) J. Chem. Soc (C)., 2410-2421. m.p. 97-98°), [α] _D ⁰ -18.9 (c = 10.60, DMF) .

^X H NMR (CDCls): δ 1.98 (1H dddd J = 7.2, 8.1, 14.2, 14.3) 2.21 (lH m), 2.39 (2H m), 4.48 (1H ddd J=6.5, 7.2, 8.1) 5.10 (2H s) , 5.10, 5.14 (2H ab), 5.57 (1H δ J = 8.1), 7.33 (10H ) , 8.8 (1H b). ¹³ C NMR (CDCI3): δ 27.25 (CH2), 29.74 (CH2), 53.25 (CH), 67.25 (CH2-0), 67.34 (CH2-0), 127.98 (aromatic CH) , 128,13 (aromatic CH), 128.21 (aro¬ matic CH) , 128,44 (aromatic CH), 128.56 (aromatic CH) , 135.00 (aromatic C), 136.00 (aromatic C), 156.06 (carbamate), 171.68 (ester), 177.67 (acid). Example 9

Z-L-Glutamic acid α- benzyl, ₇ - 2' -pyridyl ester (8) (see: S. Kim, J. J.Lee and Y.K. Ko. , (1984) , Tetrahedron lett. , _2_5, no. 43, 4943 - 4946 or A. S. Dutta and J. S. Morley, (1981) , J. Chem. , soc. (C) , 2896-2902.

Z-L-Gluta ic acid α-benzyl ester (7) (881 mg, 2.37 mmol) was dissolved in anhydrous pyridine (7 mL) and reacted with 2-hydroxypyridine (405 mg, 4.26 mmol) and dicyclohexyl- carbodiimide (DCC) (645 mg, 3.08 mmol) for 24 h at 5°. After this time the reaction was worked up by removal of the pyri¬ dine in vacuo. The crude product was redissolved in ethyl acetate (4 mL) and filtered to remove the insoluble dicyclo- hexylurea. The ethyl acetate solution was extracted with sodium bicarbonate solution (2 x 3 mL) followed by saturated sodium chloride solution (1 x 3 mL) . Removal of the solvents in vacuo from the organic layer yielded crude (8). Chroma¬ tography of the crude product (silica; 50% petrol/ethyl acetate) yielded the 2'-pyridyl ester (8) (1.01 g, 91%) which still contained a small amount of dicyclohexylurea.

-H NMR (CDCls)s δ 2.12 (2H m), 2.34 (2H m), 4.55 (1H m), 5.11 5.18 (4H 2 x AB's), 5.60 (1H d J =

- 153 -

6.8), 7.04 (lH m), 7.20 - 7.74 (12H ), 8.37 (1H m). 1.3 Reaction of Intermediates with Nucleophiles

The following examples illustrate the preparation of compounds of this invention as defined by Formula I. Example 10 Z-L-Glutamic acid η -hydroxamate (9) .

Z-L-Pyroglutamic acid (500 mg, 1.78 mmol) was taken up in THF (2 mL). To this solution was added an aqueous solution of hydroxylamine (1.8 mL, 4 M, 7.12 mmol) which was prepared by dissolving hydroxylamine hydrochloride in water at 5° and adjusting the pH to 8.0 with sodium hydroxide solution. After 4 h the reaction was complete as indicated by t.l.c. (cellulose, Rf 0.49, starting material Rf 0.63). The mixture was worked up by adjusting the pH to 2.0 and removal of the solvents in vacuo at ca. 15° followed by extraction of the residue with THF (2 x 10 mL) and ethyl acetate (2 x 10 mL) . The solvents were removed from the combined extracts to yield a white solid. Recrystallisation of this material from chloroform afforded (9) as a white crystalline soid (535 mg 96%). Subsequent repetition of this experiment furnished the hydroxamate (9) in 80 - 95% yield. Example 11 L-Glutamic acid η-hydroxamate (10) .

Z-L-Glutamic acid ₇ -hydroxamate (0.50 g) was dissolved in HPLC grade methanol (4 mL) and 10% palladium on charcoal (0.05 g) added. This mixture was stirred under an atmosphere of hydrogen until a sample spotted on a silica chromatography plate showed no u.v activity. After this time (0.5 - 1 h) the reaction mixture was filtered through a kieselgel pad and the pagd washed with water (5 mL). The filtrate and wash were combined and the solvents removed in vacuo to yield the ₇ -hydroxamate (10) (270 mg, 98%), m.p. 145-146°, [α] _D ²⁰ -7.6 (c = 10.71, H ₂ 0) .

-H NMR (D ₂ 0/DC1): δ 1.92 (2H m) 2.12 (2H m) 3.83 (1H t J= 7.6)

E SHEET

The proton NMR spectrum and melting point of (10) were identical with those of the authentic compound purchased from Sigma. Example 12 Z-L-glutamic acid η-2' -hydroxyethylamide (11) .

Z-L-Pyroglutamic acid (2) (2.0 g, 7.6 mmol) was taken up in THF (20 mL) and water (1 mL). To this solution was added ethanolamine (1.4 mL) and the reaction stirred overnight. The reaction was worked up as outlined above to yield the crude product. This material was purified by chromatography on sephadex SP-C25, eluting with water. Removal of the water furnished (11) as an oil (2.0 g, 81%). -R NMR (D ₂ 0/DC1): 6 1.7 (1H m) 1.9 (1H m) 2.1 (2H m) 3.05 (2H q) 3.17 (1H t) 3.40 (2H t) 3.95 (1H m) 4.90 (2H s) 7.17 (5H m) . Example 13 L-glutamic acid η-2' -hydroxyethylamide (12)

Derivative (11) (600 mg) was taken up in HPLC grade methanol (5 mL) to which 10% palladium on charcoal (60 mg) had been added. This mixture was hydrogenated as described above to yield (12) as a white powder (340 mg, 97%). m.p. 210-210.5°, [α] _D ²⁰ + 7.3 (c= 10.96, water). HPLC analysis (amino acid analyzer) indicated the product was greater than 99% purity with less than 1% glutamic acid impurity.

^• K NMR (D ₂ 0/DC1): δ 2.17 (2H dddd J= 2.2, 4.7, 3.3, 14.8) 2.46 (2H ddd J= 3.3, 1.8, 14.8) 3.32 (2H t J= 5.3) 3.64 (2H t J= 5.4) 3.92 (1H dd J= 2.2, 4.8).

¹³ C NMR (D ₂ 0): δ 27.40 (CH2) 32.80 (CH2) 42.91 (CH2) 54.72 (CH) 61.33 (CH2) 174.11 (amide) 176.98 (acid).

Example 14 Z-L-Glutamic acid α-t-butyl ester η -methylamide (13)

Methylamine (24% aqueous, 50 mL, 0.37 mmol) was added to a stirred solution of the t-butyl ester (3) (89 mg, 0.28 mmol) in methanol (1 mL). After 0.5 h no starting material was observed by t.l.c (silica, 75% ether/petrol) .

The reaction was worked up by removal of the solvents and methylamine in vacuo to yield a mixture of two products as shown by t.l.c Flash chromatography (45% ether/petrol) of the crude product and removal of the solvents in vacuo yielded the -methylamide as a white crystalline solid (59 mg, 60%) m.p. 109-109.5°, [a_ _D ²⁰ +3.9 (c = 1.1, chloro¬ form) . A significant by-product of the reaction was the ₇ - ethyl ester (34 mg, 35%), formed by the competitive ring opening with methoxide. Repetition of the reaction in THF eliminated this side reaction and gave an improved yield of 88%.

MS M+ 350, m/e 249 (55%), 174 (27%), 98 (27%), 92 (15%), 91 (100%), 84 (22%), 73 (15%), 57 (28%), 42 (16%), 19 (13%).

^■ K NMR (CDCI3): -5 1.45 (9H s) 1.19 (1H m) 2.22 (2H m) 2.78 (3H bd J = 4.6) 4.20 (1H m) 5.10 (2H s) 5.60 (1H bd J = 7.5) 6.03 (1H bs) 7.34 (5H ) . ¹³ C NMR (CDC13): δ 26.28 (CH3) 27.90 (CH3) 29.11 (CH2) 32.44 (CH2) 54.01 (CH) 66.93 (CH2) 82.43 (C) 128.02 128.13 128.47 (aromatic CH's) 136.21 (aro¬ matic C) 156.57 (carbamate) 171.02 (amide) 172.02 (amide) 172.57 (ester). Example 15

Z-L-Glutamic acid ₇ -methylamide (14) (see: Anderson, G. W. & Callahan . F.M. (1962) J. Am. Chem. Soc , _82, 3359-3363. ) The ₇ -Methylamide (13) (33 mg, 0.10 mmol) was taken up in cold trifluoroacetic acid (TFA) (see for example: Bryan, D.B., Hall, R.F., Holden, K.G., Huffman, W.F. & Gleason, J.G. (1977), J. Am. Chem. Soc., 9J9., 2353-2355.) and the mixture stirred in an ice bath for 0.5 h. After this time no starting ester remained by t.l.c. (silica, 5% methanol/chloroform) . The reaction was worked up by removal of the TFA in vacuo to yield the crude product as a pale yellow oil. Purification of this product was effected by flash chromatography (5% methanol/chloroform) to yield a white crystalline solid (24 mg, 85%) on removal of the sol¬ vents in vacuo. Recrystallisation (chloroform/petrol) of the

SUBSTITUTE SHEET

acid (14) gave a pure sample for spectroscopic analysis, m.p. 117-118°, _e*_ _D ²⁰ + 16.2 (c = 1.0, chloroform).

^• R NMR (CDC1 ₃ ): δ 2.00 (1H m) 2.17 (1H m) 2.30 (2H ) 2.71 (3H bd J = 4.3) 4.29 (1H ) 5.06 (2H s ₎ 6.12 (1H bd J = 7.2) 6.67 (1H bs) 7.30 (5H m) 9.36 (1H bs) .

Example 16

L-Glutamic acid ₇ -methylamide (15) (see: Lichtenstein, N. (1942) J. Am. Chem. Soc , 64, 1021-1022. )

Z-L-Glutamic acid ₇ -methylamide (14) (2.40 g, 8.15 mmol) was taken up in absolute ethanol (50 mL) to which 10% palladium on charcoal catalyst (260 mg) was added. This mix¬ ture was hydrogenated in the standard manner and monitored by t.l.c. for the disappearance of UV activity. After 4 h no starting acid remained and the reaction was worked up by fil¬ tration of the mixture. The catalyst was washed with water (2 x 10 mL) and the solvents removed in vacuo from the com¬ bined washes and filtrate. Recrystallization of the residue (80% ethanol/water) yielded the pure amino acid (15) (1.01 g, 79%) as a white crystalline solid, m.p. 193-194° (lit. Lichtenstein, N. (1942) J. Am. Chem. Soc, .64., 1021-1022. m.p. 192°), [α] _D ²⁰ +4.1 (c = 1.23, water) (lit.[α] _D ²⁰ +6.5). The purity of this material was confirmed by HPLC on an amino acid analyzer. H NMR (D ₂ 0): δ 2.12 (2H m) 2.39 (2H m) 2.71 (3H s) 3.75 (1H bt J = 6.7) .

¹³ C NMR (D ₂ 0): δ 26.49 (CH3) 27.24 (CH2) 32.12 (CH2) 58.88 (CH) 174.81 (amide) 175.78 (acid). The ring opening of (3) was attempted with a series of other nucleophilles as outlined below. In each case The experimental procedure was identical to that illustrated above. Example 17

Z-L-Glutamic acid α-t-butyl ester η-hydrazide (16) (see for example: Klieger, E. & Gibian. H. (1962) Justus Liebig' s Ann. Chem. , 655, 195-211. and Tashner, E. , Wasielewski , C. ,

^S okolows ^k a, ^T . ^Bi ernat . J. F. (1961 ) Justus _L ie _b ig' s _A nn . Chem. , 646. 127-133. )

Ester (3) (118 mg, 0.37 mmol) was reacted with hydrazine hydrate ₍ 16μL, 0.32 mmol) to yield the ₇ -hydra- zide (16) as a white crystalline solid ₍ 124 mg, 95% ₎ , m.p. 95-100° ⁽ lit.Shin, C.G., Yonezawa, Y. & Wanatabe.E. (1985) Tetrahedron Lett., .26., 85-88. m.p. 112-113° ₎ .

- K NMR (CDC1 ₃ ): * 1.45 (9H s) 1.90 (1H m ₎ 2.22 (3H m) 3.87 (1H bs) 4.22 (1H bt J = 7.4) 5.11 (2H s) 5.54 (1H bd J = 7.4) 7.37 (5H m) . Example 18

^Z - ^L - ^Gl u ^t amic ac ^{id α} - ^t -butyl ester η-N' -methy _lh y _d r azi _d e _{(1 7)}

Ester (3) (288 mg, 0.9 mmol) was reacted with methylhydrazine (60 μL, 1.13 mmol) to yield the ₇ -N'- methylhydrazide (17) as a glass in low yield ₍ 103 mg, 31% ₎ . MS M+ 365, m/e 320 (2%) 309 (71%) 292 (16%) 136 (32%) 130 (59%) 113 (76%) 91 (59%) 84 (71%) 57 (100%) 46 (95%) 42 (63%) 29 (51%) 19 (63%) 18 (57%).

^X H NMR (CDCI3): δ 1.43 (9H s) 1.92 (1H m) 2.18 (3H ) 2.57 (3H m) 3.57 (1H bs) 4.20 (1H bt J = 8.0 ⁾ 5.08 (2H s) 5.80 (1H d J = 8.0) 7.32 ₍ 5H m ₎ 7.87 (1H bs).

¹³ C MR (CDC1 ₃ ): δ 27.90 (CH3) 28.91 (CH2) 30.66 (CH2) 39.16 (CH3) 53.93 (CH) 67.01 (CH2) 82.49 (C) 128.06 (aromatic CH) 128.15 (aromatic CH) 128.47 (aromatic CH) 136.17 (aromatic C ₎ 156.37 (carbamate) 170.30 (amide) 171.93 (acid).

E_-_aτnp1<*- 1 Q

^Z -L- ^G lutamic acid η-N' -methylhydr azide ₍₁ 8 ₎

The 7- ^N '-methylhydrazide (17) (77 mg, 0.21 mmol) was reacted with TFA to. yield the free acid (18) (28 mg, 90% ₎ as a clear glass, [α] _D ²⁰ -10.0 (c = 1.0, water). The ^l H NMR spectrum confirmed the removal of the t-butyl group. Example 20

^L -Glutamic acid η-N' -methylhydrazide ₍₁ 9 ₎

Hydrogentation of the acid (18) (23 mg, 0.07 mmol) in the presence of 10% palladium on charcoal (3 mg) yielded the amino acid (19) as a clear glass (10 mg, 83%).

^~ H NMR (D ₂ 0): δ 2.09 (2H m) 2.33 (2H ) 2.53 (3H bs) 3.65 (1H m) 7.43 (1H s) . Example 21

Z-L-Glutamic acid α-t-butyl ester η-2' - hydroxyethylhydr azide (20)

The protected pyroglutamate (3) (4.0 g, 12.5 mmol) was reacted with hydroxyethylhydrazine (2.09 g, 27.6 mmol) to yield the hydrazide (20) (4.90 g, 99%) as a hygroscopic white solid which was pure by t.l.c (silica, 10% methanol/chloro¬ form, Rf 0.53), [α] ²⁰ +3.3 (c = 2.8, chloroform).

-B. NMR (CDCls): δ 1.45 (9H s) 1.90 (1H m) 2.23 (3H m) 2.90 (2H m) 3.55 (2H m) 4.20 (1H m) 5.09 (2H s) 5.70 (1H bd J = 7.9) 7.34 (5H m) 8.02 (1H bs) . ¹³ C NMR (CDC1 ₃ ): δ 27.93 (CH3) 29.31 (CH2) 30.66 (CH2) 53.51 (CH) 58.75 (CH2) 60.05 (CH2) 67.16 (CH2) 82.76 (C) 128.12 (aromatic CH) 128.27 (aromatic CH) 128.53 (aromatic CH) 136.07 (aromatic C) 154.54 (carbamate) 170.84 (amide) 173.53 (ester) . Example 22 Z-L-Glutamic acid η-2' -hydroxyethylhydr azide (21 )

De-esterification of the hydrazide (20) (4.7 g) in excess TFA yielded the acid (20) which was purified by flash chromatography (10% methanol/chloroform) to give pure (21) (3.8 g, 94%), m.p. 88-95°dec

^■ E NMR (D ₂ 0): δ 1.93 (lH m) 2.14 (1H m) 2.30 (2H m) 2.94 (2H m) 3.63 (2H m) 4.03 (1H m) 5.13 (2H s) 7.43 (5 H m) . Example 23 L-Glutamic acid η-2' -hydroxyethylhydr azide (22)

Hydrogenation of acid (21) (2.1 g) in the standard manner yielded the amino acid (22) (1.24 g, 99%) as a hygro¬ scopic white solid.

^: H NMR (D ₂ 0): δ 2.00 (2H m) 2.36 (2H ) 3.18 (2H t J = 4.9) 3.58 3.60 (2H d AB J = 12.66) 3.87 (1H t

J = 7.7) .

¹³ C NMR (D ₂ 0): δ 27.10 (CH2) 29.56 (CH2) 53.03

(CH) 53.83 (CH2) 57.05 (CH2) 171.23 (amide) 175.43

(acid) . Example 24 Z-L-Glutamic acid α-t-butyl ester η-hydroxamate (23)

The protected pyroglutamic acid (3) (5.0 g, 15.6 mmol) was dissolved in HPLC grade methanol (20 L) . To this solution was added hydroxylamine hydrochloride (27 mL, 2.4 M, 62.5 mmol) which had been adjusted to pH 8.0 with sodium hydroxide. The reaction mixture was stirred for 24 h after which time no starting material was detected by t.l.c (silica 80% ether/petrol). The mixture was worked up by adjusting the pH to 2.0 and extracting the mixture with ether (3 x 100 mL). Evaporation of the dried ether extract yielded a vis¬ cous oil (4.80 g, 87%) which was a single compound as shown by t.l.c (silica, 80% ether/petrol, Rf 0.21).

- H NMR (CDCI3): -5 1.43 (9H s) 1.90 (1H m) 2.15

(1H m) 2.21 (2H m) 4.18 (1H m) 5.09 (2H s) 5.78 (1H bd) 7.30 (5H m) . Example 25 Z-L-Glutamic acid α-t-butyl ester η-anilide (24)

Acid (6) (3.61 g, 10.7 mmol) was dissolved in anhydrous THF (25 mL) and cooled to 5°. Triethylamine (1.63 mL, 11.7 mmol) was added and the mixture stirred for ca. 2 min. After this time ethyl chloroformate (1.13 mL, 11.7 mmol) was added and the mixture stirred for a further 3 min. Aniline (2.92 mL, 32 mmol) in THF (5 mL) was then added and the mixture allowed to warm to room temperature over 0.5 h. After 2 h little starting acid was observed by t.l.c (silica, 50% ether/petrol). Removal of the solvents in vacuo yielded a crude mixture which was dissolved in DCM (50 mL) and washed with water (20 mL), 0.1 M hydrochloric acid (10 mL) and satu¬ rated sodium chloride (20 mL) . Evaporation of the solvent from the dried solution yielded the crude anilide (24) which

was purified by flash chromatography (70% ether/petrol) to yield a pale yellow solid (2.58 g, 59%). Recrystallisation from ether/petrol yielded a white solid m.p. 136-137°, [α] _D ²⁰ -18.5 (c = 11.25, chloroform).

^L H NMR (CDC1 ₃ ): δ 1.44 (9H s) 1.97 (1H ) 2.35 (1H m) 2.43 (2H m) 2.80 (1H bs) 4.30 (1H ) 5.11 (2H s) 5.66 (1H d J = 7.6) 7.10 (1H m) 7.34 (7H m) 7.60 (2H m) . Example 26 Z-L-Glutamic acid η-anilide (25)

The ester (24) (206 mg, 0.5 mmol) was treated with excess TFA (1 mL) at 0°. The reaction was monitored by t.l.c (70% ether/petrol). After 2 h a significant amount of the starting ester remained so the mixture was allowed to warm to room temperature. After a further 2 h no starting ester could be detected by t.l.c, so the TFA was removed in vacuo to yield crude (25) . Trituration with carbon tetrachloride yielded the pure acid (25) (158 mg, 89%) as white powder, m.p. 155-158°, [α] _D ²⁰ -9.03 (c = 10.73, methanol).

^• H NMR (CH ₃ OD): δ 1.99 (1H m) 2.24 (lH m) 2.49 (2H m) 4.23 (1H m) 5.06 (2H s) 7.06 (1H m) 7.29 (8H m) 7.51 (1H m) . Example 27 L-Glutamic acid η-anilide (26)

To a solution of the acid (25) (1.02 g, 2.87 mmol) in anhydrous methanol (16 mL) was added 5% palladium on char¬ coal (203 mg) . This mixture was hydrogenated for 2 h after which time no evidence of the starting acid could be detected by t.l.c (25% methanol/chloroform) . The mixture was worked up in the standard manner to yield pure (26) (636 mg, 99%) as a white crystalline solid m.p. 210-210.5°, [α] ⁰ +8.9

(c=8.90, water) .

^~ E NMR (D ₂ 0): δ 2.20 2.24 (2H dAB J = 7.7) 2.59 (2H dt J = 7.7, 5.0) 3.82 (1H t J = 5.0) 7.25 (1H m) 7.42 (4H m) .

Example 28

Z-L-Glutamic acid α-benzyl ester η -2' -hydroxyethy _la mi _de (27)

The α-benzyl ester (7) (1.27 g, 3.42 mmol ₎ was dissolved in THF (6 mL) and cooled to 5°. To this solution triethylamine (0.48 mL, 3.5 mmol) was added and the mixture

^■ v stirred for ca. 2 min. After this time ethyl chloroformate (0.38 mL, 3.5 mmol) was added and the mixture stirred for a further 3 min. Ethanolamine (0.41 mL) in THF (1.5 mL) was slowly added to the reaction mixture and the mixture allowed to warm to room temperature. After 10 min no starting ester could be detected by t.l.c (silica, 80% ether/petrol). The reaction mixture was worked up by removal of the solvents in vacuo and dissolution of the resulting oil in ethyl acetate (50 mL). The ethyl acetate was washed with 5% sodium bicar¬ bonate (10 mL) and water (10 mL) then dried over sodium sul¬ phate. Removal of the solvent in vacuo yielded (27) (1.10 g, 75%) as a colourless oil which was pure by t.l.c (silica, 80% ether/petrol, Rf 0.10 and silica 10% methanol/chloroform, Rf 0.60) .

^• H NMR (CDC1 ₃ ): δ 1.93 (2H m) 2.20 (2H m) 3.28 (1H m) 3.43 (1H m) 3.65 (2H m) 4.37 (1H mt J = 7.9) 5.08 (2H s) 5.08, 5.11 (2H AB) 5.88 (1H d J = 7.9) 6.38 (1H bs) 7.33 (10H m) . Example 29 L-Glutamic acid η-2' -hydroxyethylamide (12)

Hydrogenation of the amide (27) (600 mg) in meth¬ anol (5 mL) with 10% palladium on charcoal (60 mg) yielded the amino acid (11) (210 mg, 76%), which was identical by ~ E NMR with that prepared by the ring opening of Z-L-pyroglu- tamic acid, m.p. 210-211° (c.f. example 13 m.p. 210-210.5°), [α] _D ⁰ +6.88 (c = 9.39, water) (c.f. example 13, [α] _D ²⁰ +7.3 (c = 10.96, water)). Example 30 Z-L-Glutamic acid α-benzyl-η- 2' -aminoethyl ester (28)

A solution of the α-benzyl ester (7) (2.0 g, 5.39 mmol) in THF (9 mL) was treated with triethylamine (0.75 mL,

SUBSTITUTE SHEET

5.40 mmol) at 5° for 2 min. After this time ethyl chloro- formate (0.52 mL, 5.4 mmol) was added and the mixture stirred for ca. 3 min. A solution of Z-ethanolamine (1.16 g, 6 mmol) and triethylamine (1.5 mL, 10.8 mmol) in THF ^" (5 mL) was slowly added to the above mixture and the reaction mixture stirred for 0.5 h during which time it was allowed to warm to room temperature. The reaction mixture was allowed to stir for a further 24 h after which time no starting ester (7) could be detected by t.l.c (silica 0.1% acetic acid/ether). The reaction mixture was worked up in the manner described above to yield a mixture of Z-ethanolamine and the diester (28) (1:4) (2.8 g, 91%) as confirmed by H NMR. Attempts to separate the two components by flash chromatography (t.l.c, silica, 0.1 acetic acid/ether, Rf 0.517, 0.510) were unsuc¬ cessful using a variety of solvents (80% ether/petrol; 60% ether/1% aceton in petrol) so hydrogenation was carried out on the mixture.

Η KHR (CDCL ₃ ): δ 1.93 (1H m) 2.10 (1H m) 2.34 (2H m) 2.52 (2H m) 3.73 (1H m) 4.12 (2H m) 4.51 (4H s) 5.03 (1H bd) ^' 7.24 (lOH m). Example 31

Attempted synthesis of L-glutamic acid η-2' -aminoethyl ester (29)

Hydrogenation of the diester (28) (2.0 g) in meth¬ anol (10 mL) with 10% palladium on charcoal (200 mg) as catalyst yielded L-glutamic acid as the sole product (460 mg, 88% corrected yield). No evidence for the η - 2 '-aminoethyl ester could be found by t.l.c (silica, 10% methanol/chloro- form) or H NMR of the crude product. Example 32

Z-L-Glutamic acid α-benzyl ester η -tris (hydroxymethyl) methylamide (30)

The α-benzyl ester (7) (1.4 g, 3.77 mmol) was dissolved in THF (5 mL) and reacted with triethylamine (600 mL, 4.5 mmol), ethyl chloroformate (410 mL, 4.5 mmol), and tris(hydroxymethylJmethylamine (2 g, 18.9 mmol) in THF (5 mL) and water (4 mL) . Flash chromatography (100% chloroform) of

the crude product yielded pure (30) (1.46 g, 79%) as a hygro¬ scopic white glass, [α] _D ²⁰ -12.0 (c = 12.27, methanol).

^■ H NMR (CDCL ₃ ): δ 2.16 (lH m) 2.27 (3H m) 3.54 3.63 (6H AB) 4.41 (1H m) 4.70 (3H bs) 5.10 (2H s) 5.10 5.14 (2H AB) 5.93 (1H d J = 8.3) 6.66 (1H bs) 7.31 (10H ) .

¹³ C NMR (CDC1 ₃ ): δ 28.03 (CH2) 32.12 (CH2) 53.21 (CH) 61.71 (C) 62.82 (CH2) 67.15 (CH2) 67.32 (CH2) 127.9 (aromatic CH) 128.20 (aromatic CH) 128.47 (aromatic CH) 128.56 (aromatic CH) 135.03 (aromatic C) 135.89 (aromatic C) 156.35 (carbamate) 171.85 (amide) 173.62 (ester). Example 33

Attempted synthesis of L-Glutamic acid η-tris (hydroxy¬ methyl) methylamide (31 )

Hydrogenation of the amide (30) (1.10 g) in meth¬ anol (20 mL) with 10% palladium on charcoal catalyst fur¬ nished a single compound in low yield upon filtration of the mixture through a celite bed and removal of the solvents in vacuo. Washing the celite bed with 0.IM hydrochloric acid reclaimed the majority of the product. Removal of the sol¬ vents in vacuo yielded the cyclised amino acid (32) (520 mg, 92%) as a hygroscopic white solid. Repetition of the above experiment in the absence of hydrochloric acid again yielded the cyclised amino acid (32) (70%) as hygroscopic white powder.

-Η. NMR (32)(D ₂ 0): 51.87 (2H ddAB J = 6.6, 7.2,

7.4) 2.36 (2H dd J = 7.2, 7.4) 3.35 (4H s) 3.74 (1H t J = 6.6) 4.39 (2H s) upon addition of lithium hydroxide in D20 the peak at 4.39 disappears and the peak at 3.35 increases to 6H.

¹³ C NMR (D20): δ 24.3 (CH2) 28.8 (CH2) 5.17

SUBSTITUTE SHEET

(CH) 59.0 (CH2) 59.6 (C) 61.9 (CH2) 171.1 (amide) 173.0 (ester). Example 34

Z-L-Glutamic acid α-benzyl ester η -bis (hydroxyethyl) amide (33)

A solution of acid (7) (3.76 g, 10.1 mmol) in THF (10 mL) was reacted with triethylamine (1.55 mL, 11.1 mmol ₎ , ethyl chloroformate (1.06 mL, 11.1 mmol) and bis ₍ hydroxy _¬ ethyl)amine (3.4 mL, 30.3 mmol) at 15° for 1 h. Standard workup of this reaction yielded the bis(hydroxyethyl)amide (33) (2.81 g, 59%) as a pure compound as shown by t.l.c (10% methanol/chloroform, Rf 0.63, Rf starting acid 0.48), [α] _D ²⁰ -9.98 (c = 16.08, methanol).

-H NMR (CDC1 ₃ ): δ 1.75 (1H m) 1.90 (1H m) 2.08 (2H m) 3.00 (2H m) 3.07 (2H m) 3.55 (4H m) 3.96 (1H m) 4.62 (2H s) 4.62 4.75 (2H AB) 7.20 (10H m) . Example 35 L-Gluta ic acid η-bis (hydroxyethyl) amide (34)

The bis(hydroxyethyl)amide (33) (2.15 g) was hydrogenated in methanol (20 mL), in the presence of 10% palladium on charcoal (250 mg) as catalyst. Standard workup of this mixture yielded the amino acid (34) (1.08 g, 92%) as a hygroscopic white powder.

-ΕL NMR (D ₂ 0/DC1): δ 2.02 (2H m) 2.17 (2H m) 3.06 (2H m) 3.65 (2H m) 4.03 (4H m) 5.02 (lH m). ^1S C NMR (D ₂ 0): δ 27.64 (CH2) 29.01 (CH2) 50.57 (CH2) 52.14 (CH2) 53.35 (CH) 60.54 (CH2) 60.74 (CH2) 172.31 (amide) 173.53 (acid). Example 36

Z-L-Glutamic acid α-benzyl ester η-1 ' , 1 ' -di.methyl-2' - hydroxyethylamide (35)

The α-benzyl ester (7) (3.57 g, 9.61 mmol) was dissolved in THF (10 mL) . To this solution was added tri¬ ethylamine (1.47 mL, 10.57 mmol), ethyl chloroformate (1 mL, 10.6 mmol) and l,l-dimethyl-2-hydroxyethylamine (2.75 mL, 29 mmol) and the mixture stirred at 15° for 1 h. Standard workup followed by chromatography of the crude product (70% r

ether/petrol) yielded (35) as a colourless oil (4.08 g, 96% ₎ , [α] _D ²⁰ -9.5 (c = 12.61, chloroform ₎ .

- H NMR (CDC1 ₃ ): 6 1.20 (6H s) 1.92 ₍ 1H m) 2.16 (3H m) 3.48 3.50 (2H AB J = 11.5) 4.36 (1H ) 5.05 (2H s) 5.05 5.12 (2H AB) 6.01 (1H d J~~ 8.0 ₎ 6.20 (1H s) 7.30 (10H m) .

¹³ C NMR (CDCI3): δ 24.03 (CH3) 24.18 ₍ CH3) 28.02 (CH2) 32.51 (CH2) 53.33 (C) 55.74 (CH) 66.87 (CH2) 67.09 (CH2) 69.61 (CH2) 127.89 (aromatic CH) 128.01 (aromatic CH) 128.09 (aromatic CH) 128.32 (aromatic CH) 128.44 (aromatic CH) 135.03 (aromatic C) 135.98 (aromatic C) 156.26 (carbamate ₎ 171.78 (amide) 172.50 (acid). Example 37

^L -Glutamic aci ^{d η} -1 ' , 1 ' -dimethyl- 2' -hydroxyethylamide (36 ₎ Hydrogenation of the amide (35) (2.20 g) in HPLC grade methanol (20 mL) with 10% palladium on charcoal (220 mg ⁾ as catalyst, yielded the amino acid (36) (1.10 g, 99%) as a hygroscopic white solid, [ ] _D ²⁰ + 6.82 (c = 11.04, H ₂ 0) . ^■ . NMR (D ₂ 0): δ 1.04 (6H s) 1.99 (2H m) 2.24 (2H m) 3.43 (2H s) 3.89 (1H t J = 6.45 ₎ . ¹³ C NMR (D ₂ 0): δ 22.93 (CH3) 25.53 (CH2) 31.66 (CH2) 52.11 (CH) 54.91 (C) 66.87 (CH2) 171.40 (amide) 173.59 (acid). Example 38

^Z -L-Glutamic acid α-benzyl ester η-ethylami _d e ₍ 3 ₇₎

A solution of acid (7) (5.27 g, 14.2 mmol) in THF (40 mL) was reacted with triethylamine (2.23 mL, 16.0 mmol), ethyl chloroformate (1.53 mL, 16.0 mmol) and anhydrous ethylamine (1.9 mL, 28.0 mmol) at 5° for 1 h. Standard

*» workup of this reaction yielded the crude ethylamide (37) which was recrystallised from chloroform/ether (1:4) to yield the ethylamide (37) (5.01 g, 88%) homogeneous by t.l.c (80% ether/petrol, Rf 0.19), m.p. 121-121.5°, [α] _D ²⁰ -6.02 (c = 13.13, chlorofor ) .

-E NMR (CDCla). δ 1.09 (3H t J - 7,3), 1.82 (1H m), 1.99 (1H m), 2.17 (2H m), 3.24 (2H dq J = 7.0,

SUBSTITUTE SHEET

- 166 -

7.3), 4.39 (1H ), 5.10 (2H s), 5.12, 5.17 (2H AB J = 12.4), 5.70 (2H bd J = 7.0), 7.33 (10H m) . ¹³ C NMR (CDC1 ₃ ): δ 14.7 (CH ₃ ), 28.4 (CH ₂ ) , 32.4 (CH ₂ ), 34.4 (CH ₂ ), 53.6 (CH), 67.0 (CH ₂ ) , 67.3 (CH ₂ ), 128.0 (aromatic CH), 128.2 (aromatic CH), 128.4 (aromatic CH), 128.5 (aromatic CH) , 128.6 (aromatic CH), 135.2 (aromatic C), 136.1 (aromatic C), 156.3 (carbamate C=0), 171.5 (C=0), 171.8 (C=0). Example 39 L-Glutamic acid η-ethylamide (38)

The ₇ -ethylamide (37) (1.13 g) was hydrogenated in methanol (20 mL), in the presence of 10% palladium on char¬ coal (140 mg) as catalyst. Standard workup of this mixture yielded the amino acid (38) (460 mg, 92%) as a crystalline white solid, m.p. 213-214.5°, [α] _D ²⁰ +7.8 (c = 12.47,

H ₂ 0), l - ²⁰ Hg(587) ^+2*8 ( ^{c = 12} ' ⁴⁷ ' ^H2 °)'

^~ H NMR (D ₂ 0)r δ 0.95 (3H t J = 7.3), 2.01 (2H m), 2.25 (2H m), 3.00 (2H q J = 7.3), 3.80 (1H t J = 6.3).

¹³ C NMR (D ₂ 0): δ 11.1 (CH ₃ ), 23.7 (CH ₂ ) , 29.1 (CH ₂ ), 32.2 (CH ₂ ), 50.6 (CH), 170.0 (C=0), 171.6 (C=0). Example 40 Z-L-Glutamic acid α-benzyl ester η-diethylamide (39)

The α-benzyl ester (7) (3.04 g, 8.19 mmol) was dissolved in THF (15 mL) . To this solution was added tri¬ ethylamine (1.25 mL, 9.0 mmol), ethyl chloroformate (0.86 mL, 9.0 mmol) and diethylamine (1.86 mL, 18 mmol) and the mixture stirred at 15° for 1 h. Standard workup afforded crude product (38) as a colourless oil (3.23 g, 92%) which was homogeneous by t.l.c (silica, 80% ether/petrol, Rf. 0.24), [αJ _D ²⁰ -6.89 (c = 10.05, chloroform).

^• R NMR (CDCI3): δ 1.05 (6H t J = 7.1), 2.20 (4H m β and 1 ₇ protons), 3.15 (2H q J = 7.1), 3.24 (2H q j = 7.1), 4.37 (1H m α proton), 5.03 5.07 (2H

- 167 -

AB), 5.14 5.19 (2H AB), 6.06 (1H bd J = 7.7 NH proton), 7.29 (10H aromatics). ¹³ C NMR (CDC1 ₃ ): δ 14.9 (CH ₃ ), 27.9 (CH ₂ ) , 28.4 (CH ₂ ), 31.9 (CH ₂ ), 34.6 (CH ₂ ) ^' , 54.9 (CH), 68.0 (CH ₂ ), 68.5 (CH ₂ ), 128.1 (aromatic CH) , 128.3 (aromatic CH) , 128.4 (aromatic CH), 128.6 (aromatic CH), 128.8 (aromatic CH), 136.2 (aromatic C), 136.9 (aromatic C), 157.1 (carbamate C=0), 171.9 (C=0), 172.6 (C=0) . Example 41 L-Glutamic acid η-diethylamide (40)

Hydrogenation of the diethylamide (39) (1.32 g) in HPLC grade methanol (20 mL) with 10% palladium on charcoal (120 mg) as catalyst, yielded tHe amino acid (40) (600 mg, 95%), as a crystalline white solid, m.p. 201-203°, [α] ²⁰ + 8.34 (c = 13.45, H ₂ 0) . ^•

- Ε. NMR (D ₂ 0): δ 1.00 (6H t J = 7.4), 2.03 (2Hm), 2.34 (2H m), 3.01 (2H q J = 7.4), 3.11 (2H q J = 7.4), 3.85 (1H t J = 6.6).

¹³ C NMR (D ₂ 0): δ 13.2 (CH ₃ ) , 26.0 (CH ₂ ) , 26.1 (CH ₂ ), 29.3 (CH ₂ ), 30.6 (CH ₂ ), 51.1 (CH), 168.9 (C=0), 171.1 (C=0). Example 42

Z-L-Glutamic acid α-benzyl ester η-N- (2' hydroxyethyl) piperazide (41) Method A: mixed anhydride.

A solution of the α-benzyl ester (7) (4.07 g, 10.97 mmol) in THF (25 mL) was reacted with triethylamine (1.68 mL, 12.0 mmol), ethyl chloroformate (1.15 mL, 12.0 mmol) and N-2'-hydroxyethylpiperazine (1.6 mL, 14.0 mmol) at 5° for 1 h. Standard workup of this reaction yielded the crude N-2'-hydroxyethylpiperazide (41). Flash chromatography (10% methanol/chloroform) followed by column chromatorgraphy (sephadex LH-20, 60% DCM/petrol) yielded (41) (4.54 g, 86%) homogeneous by t.l.c (10% methanol/chloroform, Rf 0.21, [α] _D ²⁰ -5.03 (c = 9.54, DCM).

SUBSTITUTE SHEET ⁵

- 168 -

-R NMR (CDCI3): δ 2.01 (lH m), 2.21 (lH m), 2.39 (6H m), 2.50 (2H m), 3.31 (4H bm), 3.59 (2H m), 4.40 (1H m), 5.10 (4H m), 5.68 (1H bd J = 8.0), 7.31 (10H m) . Method B: using the 2-hydroxypyridine activated ester (8).

A solution of the activated ester (8) (5.34 g, 11.91 mmol) in dichloromethane (20 mL) was added to a stire solution of l(2'-hydroxyethyl)piperazine (2.5 mL, 14.3 mmol in dichloromethane (20 mL) at 5°. The reaction mixture was stirred for 3.5 h, after which time no starting ester was detectable by t.l.c (silica; 3% methanol/chloroform) . Wash ing of the reaction solution with water (3 x 10 mL) and removal of the solvents in vacuo yielded the desired produ (41) (3.28 g, 57%) as a pale yellow oil. The optical rota¬ tion and - E NMR spectrum were identical to that determined above. Short range ^-^C correlations were determined for this compound. δ 27.2 (CH ₂ ), 28.8 (CH ₂ ), 41.5 (CH ₂ ), 45.1 (CH ₂ ), 52.3 (CH ₂ ), 52.8 (CH ₂ ) , 53.7 (CH), 57.7 (CH ₂ -N) 66.7 (CH ₂ -OH), 127.9 (aromatic CH), 128. (aromatic CH) , 128.2 (aromatic CH), 128.3 (aroma CH), 128.4 (aromatic CH), 135.2 (aromatic C), 13 (aromatic C), 156.0 (carbamate C=0) , 170.0 (amid C=0), 171.8 (ester C=0). Example 43

L-Glutamic acid η-N- (2' -hydroxyethyl) piper azide (42)

Hydrogenation of the amide (41) (0.32 g) in HPLC grade methanol (10 mL) with 5% palladium on charcoal (71 as catalyst, yielded the crude amino acid (42). Ion excha chromatography (system II, NH4+ form, water then IM ammon effected a partial purification of the crude mixture.

Repetition of this experiment on a larger scale was again complicated by difficulties in purifying the final product. It appears that the 7-hydroxyethylpiperazide is relatively unstable at elevated temperatures, resulting in a mixture of decomposition products which were difficult to separate. - H NMR (D ₂ 0): δ 2.06 (2H ddd J = 6.6,7.0,7.1 β protons), 2.60 (8H m comprising 2 g protons, 4 piperazine ring protons and 2 N-CH ₂ protons), 3.57 (4H m comprising 4 piperazine ring protons and 2 N CH ₂ protons), 3.57 (4H m comprising 4 piperazine ring protons), 3.65 (lH t J = 6.6 α proton), 3.73 (2H t J = 6.0 CH.-PH protons) . ¹³ C NMR (D ₂ 0): δ 26.9 (CH ₂ ), 29.9 (CH ₂ ), 39.9 (2 x CH _a ), 43.5 (2 x CH ₂ ) , 52.7 (CH ₂ ), 55.3 (CH), 59.4 (CH ₂ ), 174.0 (amide C=0), 175.2 (acid C=0) . Example 44 Z-L-Glutamic acid α-benzyl ester η-morpholide (43)

The α-benzyl ester (7) (1.80 g, 4.84 mmol) was dissolved in THF (15 mL) and reacted with triethylamine (0.74) mL, 5.3 mmol), ethyl chloroformate (0.51 mL, 5.3 mmol), and morpholine (0.51 mL, 6.0 mmol) in THF (5 mL) . Standard workup and recrystallization (ether/petrol 1:1) of the crude product yielded pure (43) (1.81 g, 85%) as a white solid, m.p.62-65°, [α] _D ²⁰ -10.40 (c = 10.58, DCM).

-E NMR (CDC1 ₃ ): δ 2.04 (1H m), 2.27 (3H m) , 3.26 (2H m), 3.58 (6H m), 4.40 (lH m), 5.15 (4H m), 5.73 (1H bd J = 7.2), 7.34 (10H m) ¹³ C NMR (CDC1 ₃ ): δ 27.4 (CH ₂ ) , 28.7 (CH ₂ ), 42.0 (CH ₂ ), 45.6 (CH ₂ ), 53.7 (CH) , 66.4 (CH ₂ ) , 66.7 (CH ₂ ), 66.9 (CH ₂ ), 67.2 (CH ₃ ), 128.1 (aro¬ matic CH), 128.2 (aromatic CH), 128.4 (aromatic CH), 128.5 (aromatic CH), 128.6 (aromatic CH) , 135.3 (aromatic C), 135.4 (aromatic C), 156.0 (carbamate C=0), 170.3 (C=0), 171.8 (C=0) .

SUBSTITUTE SHEET

Example 45

L-Glutamic acid η-morpholide (44) Method A: Catalytic Hydrogenation.

Hydrogenation of the morpholide (43) (2.42 g, 5.49 mmol) in HPLC grade methanol (30 mL) using 10% palladium on charcoal (242 mg) as catalyst yielded the crude amino acid (44) as a white solid which was recrystallised from water/ acetone (1:2) (1.175 g, 92%) mp 166-169°, tic (cellulose; 70 ethanol/water) gave R _f 0.61 with a slight contamination by L glutamic acid (Rf 0.69), [αJ _D ²⁰ -0.8 (c = 10.0, methanol).

^■ R NMR (D ₂ 0): δ 2.12 (2H m), 2.60 (2H t), 3.57

(4H t), 3.74 (5H m) .

^{1 3} C NMR (D ₂ 0) : δ 28 . 24 (CH ₂ ) , 32 . 06 (CH ₂ ) ,

44 .62 ( 2 x CH ₂ ) , 48 .32 (CH ₂ ) , 56 .52 (CH) , 68 .68

(2 x CH ₂ ), 175.16 (acid C=0) , 176.35 (amide C=0) . Method B: Transfer Hydrogenation in ethanol.

The morpholide (43) (223 mg, 0.5 mmol) was taken u in absolute ethanol (10 mL) . To this solution was added 10% palladium on charcoal (24 mg) and cyclohexa-l,4-diene (480 μL, 10 eq) . The reaction mixture was stirred under dry nitrogen for 1 h at room temperature. After this time the mixture was filtered and the charcoal washed with water. Removal of the solvents in vacuo yielded the amino acid (44) as the sole product by t.l.c (cellulose, ethanol 70%/water 30%).

Example 46 Z-L-Glutamic acid α-benzyl ester η-O-methylhydroxamate (45)

O-Methylhydroxylamine hydrochloride (5 g) was tak up in distilled pyridine (10 mL) and distilled at 70°. The distillate (4 mL) collected at -78° contained a significant amount of pyridine as well as the free 0-methyl hydroxy¬ lamine.

The α-benzyl ester (7) (5.0 g, 13.4 mmol) was dissolved in THF (30 mL) and reacted with triethylamine (1. g, 1.94 mL, 14.8 mmol), ethyl chloroformate (1.52 g, 1.34 m 14.8 mmol), and the 0-methylhydroxylamine distillation product in pyridine. The mixture was allowed to stir for l

after which time the solvents were removed in vacuo to yield a mixture of three products as indicated by t.l.c ₍ 80% ether/ petrol) .Chromatography (silica, 50% ether / petrol followed _β by 80% ether / petrol) yielded the desired product (45) as a white crystalline solid (4.15 g, 77%), m.p. 143-144.5°, [α] ²⁰ - 8.01 ( c = 11.67, chloroform). The other compo- nents were identified as the starting ester (7) and the ethyl carbamate of O-methylhydroxylamine. H NMR (CDC1 ₃ )(45): δ 2.19 (4H m β and ₇ protons) ,' 3.69 (3H s CH ₃ ) , 4.36 (1H m α proton), 5.08 (2H s benzylic CH ₂ ) , 5.15 5.19 (2H AB benzylic CH ₂ ) , 7.31 (10H m aromatics) , . ¹³ C NMR (CDC1 ₃ ), (45): δ 28.6 (CH ₂ ), 29.2 (CH ₂ ), 53.4 (CH), 64.2 (CH ₃ ), 67.2 (CH ₂ ), 67.4 (CH ₂ ), 128.0 (aromatic CH) , 128.2 (aromatic CH), 128.3 (aromatic CH) , 128.5 (aromatic CH) , 128.6 (aromatic CH), 135.0 (aromatic C), 135.9 (aromatic C), 156.5 (carbamate C=0), 169.8 (C=0), 171.6 (C=0). Example 47 L-Glutamic acid η-O-methylhydroxamate (46)

The protected O-methylhydroxamate (45) (3.4 g) was hydrogenated for 2 h in methanol (50 mL) with 10% palladium on charcoal (400 mg) as catalyst. Filtration of the resultant mixture and washing of the charcoal with hot water (3 x 5 mL) returned the product (46). Removal of the solvents in vacuo yielded the amino acid (46) as a hygroscopic white solid (1.35 g, 90%), [α] _D ²⁰ +9.20 ( c = 9.87 , H ₂ 0/HC1 1 drop). -Η. NMR (D2O/DCI): δ 2.23 (2H m), 2.44 (2H m), 3.58 (3H s), 3.75 (1H t J = 7.1). ¹³ C NMR (D2O/DCI): δ 26.4 (CH ₂ ), 28.7 (CH ₂ ), 54.4 (CH), 64.5 (CH ₃ ), 171.2 (C=0), 174.1 (C=0). Example 48

Z-L-Glutamic acid α-benzyl ester η-2' thioethylamide (47) The α-benzyl ester (7) (4.67 g, 12.6 mmol) was dissolved in anhydrous THF (40 mL) and cooled to 5°. This solution was reacted with triethylamine (3.82 g, 5.30 mL, 36

E

mmol) followed by ethyl chloroformate (1.63 g, 1.44 mL, 14 mmol) and stirred for 5 min. After this time 2-thioethylamin hydrochloride (1.7 g, 14 mmol) was added as a solid and the mixture stirred for 4 h. Removal of the solvents in vacuo yielded the crude mixture which was re-dissolved in dichloro methane (50 mL) and washed with water (2 x 40 mL) . The aque¬ ous extracts were re-extracted with dichloromethane (70 mL) and the combined organic extracts were dried over sodium sul phate prior to removal of the solvents in vacuo. Chromatogra phy (silica, 95% ether/petrol 100% ether and 5% methanol/ chloroform) of the crude product yielded the desired product (47) as a waxy white solid (3.7 g, 74%), m.p. 95-97°, [α] _D ²⁰ -5.24 ( c = 12.40, dichloromethane).

^X H NMR (CDCls): δ 2.00 (2H m), 2.24 (2H m), 2.73 (2H t J = 6.3), 3.46 (2H ddd J = 6.3,6.3,6.5), 4.3 (1H m), 5.06 (2H s), 5.09 5.13 (2H AB), 5.90 (1H d J = 7.9 NH proton), 6.62 (1H t J=6.5 SH proton), 7.31 (10H m aromatics). Example 49 L-Glutamic acid η-2' thioethylamide (48)

A solution of the 2'-thioethylamide (199 mg) in anhydrous liquid ammonia (50 mL) was hydrogenated with 10% palladium on charcoal (98 mg) as catalyst for 4 h. After thi time the ammonia was allowed to evaporate and the resultant mixture suspended in methanol. The methanolic (5 mL) suspen¬ sion was filtered through a celite plug which was subse¬ quently washed with water (5 mL). Removal of the solvents in vacuo yielded a mixture of the starting amide and the desire amino acid (48). Dissolution of the crude mixture in chloro¬ form (5 mL) and extraction with water (2 x 5 mL) yielded the amino acid (48) (28 mg) on lyophilisation of the aqueous extract. Removal of the solvents in vacuo from the organic layer returned unreacted starting amide (47) (159 mg). Example 50 Z-L-Glutamic acid α-benzyl ester η-2' methoxyethylamide (49)

The α-benzyl ester (7) (4.11 g, 11.07 mmol) was dissolved in THF (30 mL). To this solution was added

triethylamine (1.80 mL, 12.17 mmol), ethyl chloroformate (1.30 mL, 12.17 mmol) and 2-methoxyethylamine ( 1.0 g, 12.4 mmol) and the mixture stirred at 15° for 3 h. After this time the reaction was complete as determined by t.l.c (silica; 5% petrol/ether). Standard workup followed by recrystallisation (50% ether/petrol) of the crude product yielded ₍ 49 ₎ as a crystalline solid (4.91 g, 99%),m.p. 80-84°, [ _β ] ²⁰ -5.5 (c

- 8.86, dichloromethane).

^■ U ^N MR (CDC1 ₃ ): 6 1.99 (2H m) 2.21 (2H m) 3.33 (3H s), 3.36 (4H bt J = 5.4), 4.40 (1H m ₎ , 5.10 5.16 (4H m), 5.75 (1H d J = 7.6), 7.34 ₍ 10H m) . ¹³ C MR (CDCl.): δ 28.3 (CH ₂ ), 32.2 (CH ₂ ),

39.2 (CH ₂ ), 53.6 (CH), 58.7 (CH ₃ ) , 67.0 (CH ₂ ),

67.3 (CH ₂ ), 70.9 (CH ₂ ), 128.1 (aromatic CH) , 128.2 (aromatic CH), 128.3 (aromatic CH), 128.5 (aromatic CH), 128.6 (aromatic CH), 135.0 (aromatic C), 135.5 (aromatic C), 156.2 (carbamate C=0 ₎ , 169.5 (C=0), 170.8 (C=0).

E_raτnp1<=» ■■__

^L - ^G lutamic acid η-2' methoxyethylamide ₍ 50 ₎

The 7-methoxyethylamide (49) (2.51 g ₎ was hydro _¬ genated in methanol (20 mL), in the presence of 10% palladium on charcoal (264 mg) as catalyst. Standard workup of this mixture yielded the amino acid (50) (1.19 g, 96%).as a crys _¬ talline white solid after recrystallisation from water/meth- ano l ⁱ l), m.p. 183-185°, [α] _D ²⁰ +1.6 ( c = 6.96, H ₂ 0 ),

^[α]2 °Hg(436) ^+17,7 ( ^{c = 6} - ⁹⁶ ' ^H 3° )• R _f 0.30 (cellulose; n-butanol/pyridine/04%acetic acid in water 22:10:10) R 0.56 (silica; 70% ethanol/water) . ^f

- ≡ NMR (D ₂ 0): δ 2.12 (2H ), 2.42 ₍ 2H m ₎ , 3.35 (3H s), 3.38 (2H t J - 6.2), 3.55 (2H t J - 6.2), 3.75 (1H t J = 6.9).

¹³ C NMR (D ₂ 0): δ 27.8 (CH ₂ ), 32.9 (CH ₂ ), 40.3 (CH ₂ ), 55.6 (CH), 59.4 (CH ₃ ), 71.7 (CH ₂ ), 175.4 (C=0), 176.2 (C-O). Example 52

^Z - ^L - ^G lutamic acid α-benzyl ester η-2' chloroethylami _d e ₍ 5 _{1 )}

The 2'-hydroxyethylamide (27) (3.89 g, 9.39 mmol) was dissolved in dichloromethane (30 mL). To this solution was added triethylamine (1.85 mL, 12.5 mmol), dimethylamino- pyridine (DMAP) (132 mg, 11 mol% ) and methanesulphonyl chloride ( 1.08 mL, 12.5 mmol) and the mixture stirred at 5° for 3 h. The reaction mixture was allowed to warm to room temperature after this time, and stirred for 2 days. Standard workup followed by chromatography (silica; 10% petrol/ether , 10% petrol/ethyl acetate, 100% ethyl acetate) yielded a white solid which was further purified by recrystallisation (50% chloroform/carbon tetrachloride) to afford (51) (2.05 g, 56%) as a crystalline solid.

-ϋ NMR (CDC1 ₃ ): δ 1.98 (lH m), 2.19 (3H m), 3.55 (4H m), 4.42 (1H m) , 5.06 5.16 (4H 2 x AB's), 5.65 (1H d J = 7.8), 7.34 (10H m) .

¹³ C NMR (CDCI3): δ 28.5 (CH2), 32.2 (CH2), 41.3 (CH2), 43.7 (CH2), 53.5 (CH) , 67.1 (benzylic CH2), 67.3 (benzylic CH2), 128.1 (aromatic CH), 128.2 (aromatic CH), 128.4 (aromatic CH) , 128.5 (aromatic CH), 128.6 (aromatic CH) , 131.5 (aromatic C), 136.0 (aromatic C), 153.2 (carbamate C=0), 171.7 (amide C=0), 173.1 (ester C=0). Example 53 L-Glutamic acid η-2' chloroethylamide (52)

The 2'-chloroethylamide (51) (2.04 g) was hydroge¬ nated in methanol (30 mL), in the presence of 10% palladium on charcoal (226 mg) as catalyst. Standard workup of this mixture yielded the amino acid (52) (863g, 100%). H NMR (D ₂ 0): δ 2.16 (2H m), 2.46 (2H m), 3.54 (2H t J = 5.6 ), 3.66 (2H t J = 5.6), 3.84 (lH t J = 6.2).

¹³ C NMR (D ₂ 0): δ 27.7 (CH ₂ ) , 32.9 (CH ₂ ), 42.6 (CH ₂ ), 44.6 (CH ₂ ), 55.1 (CH), 174.7 (C=0), 176.2 (C=0). Example 54 Z-L-Gluta ic acid α-benzyl ester η-2' -hydroxyani lide (53)

The α-benzyl ester (7) (3.95 g, 10.64 mmol) was dissolved in THF (30 mL) . To this solution was added tri¬ ethylamine (1.63 mL, 12 mmol), ethyl chloroformate (1.11 mL, 12 mmol) and 2-hydroxyaniline ( 1.74 g, 12 mmol) and the mix¬ ture stirred at 15° for 3 h. After this time the reaction was complete as determined by t.l.c (silica; 10% methanol). Stan¬ dard workup followed by chromatography (silica; 100% chloro¬ form) of the crude product yielded (53) as a white solid (2.29 g, 51%) .

Η HMR (CDCls): δ 1 . 91 (1H m), 2.34 (lH m), 2.46

(2H m), 4.48 (1H m), 5.10 5.16 (4H 2 x AB's), 5.73

(1H d J = 6.2), 6.8 - 7.3 (14H m) . Example 55 L-Glutamic acid η-2' -hydroxyani lide (54)

The 2'-hydroxyanilide (53) (1.54 g) was hydroge¬ nated in methanol (30 mL), in the presence of 10% palladium on charcoal (300 mg) as catalyst. Standard workup of this mixture followed by extraction of the aqueous solution with chloroform yielded the amino acid (54) (0.73 g, 93 %) on lyophilization.

-B. NMR (D ₂ 0): δ 2.24 (2H m), 2.67 (2H m), 3.93

(1H t J = 6.3), 6.97 (2H m) , 7.18 (1H m) , 7.36

(1H m) . Example 56 Z-L-Gluta ic acid α-benzyl ester η-3' -chloroanilide (55)

The α-benzyl ester (7) (3.81 g, 10.25 mmol) was dissolved in THF (45 mL) . To this solution was added tri¬ ethylamine (1.57 mL, 12 mmol), ethyl chloroformate (1.11 mL, 12 mmol) and 3-chloroaniline ( 1.29 mL, 12 mmol) and the mixture stirred at 15° for 1 h. After this time the reaction was complete as determined by t.l.c (silica; 10% methanol). Standard workup yielded (55) as a pale yellow oil (5.20 g, 96%), [α] _D ²⁰ -6.2 ( c = 8.54, chloroform).

*H NMR (CDCI3): δ 1.85-2.02 (2H m), 2.33 (2H m) ,

4.43 (1H m), 5.10 5.16 (4H 2 x AB's), 5.69 (1H d J

= 8.0), 7.04-7.66 (14H m) . ⁱ³ C NMR (CDCls): δ 29.3 (CH*), 33.4 (CH*),

53.4 (CH), 67.4 (CH ₂ ), 67.6 (CH ₂ ) , 117.6 (aromatic CH) , 119.7 (aromatic CH) , 124.1 (aromatic CH), 128.1 (aromatic CH), 128.3 (aromatic CH) , 128.4 (aromatic CH), 128.5 (aromatic CH), 128.6 (aromatic CH) , 129.8 (aromatic CH) , 134.1 (aromatic C), 134.3 (aromatic C), 135.8 (aromatic C) , 138.2 (aromatic C), 156.2 (carbamate C=0), 170.1 (amide C=0), 172.9 (ester C=0). Example 57 L-Gluta ic acid η-3' chloroanilide (56)

The protected ₇ -3'-chloroanilide (55) (1.0 g) was hydrogenated for 1.5 h in methanol (20 mL) with 10% palladium on charcoal (101 mg) as catalyst. Filtration of the resultant mixture and washing of the charcoal with hot water (3 x 5 mL) returned the product (56). Removal of the solvents in vacuo yielded the amino acid (56) (280 mg, 52%) as a hygroscopic white solid, [α] ⁰ +7.10 ( c = 10.2 , water). A small amount of glutamic acid ( approx 5% ) was also present as determined by t.l.c (cellulose; n- butanol/pyridine/0.4% acetic acid in water 22:10:10, R _f product 0.62, R _f Glutamic acid 0.20), which was removed by washing with a small amount (ca. 500 uL) of warm water.

^~ H NMR (D ₂ 0/DC1): 6 2.19 (2H m), 2.61 (2H m), 3.99 (1H t J = 6.7), 7.02 - 7.17 (3H m) , 7.66 (1H s) . T.-»raτnp1 e. «_B

Z-L-Glutamic acid α-benzyl ester ₇ -4' -chloroanilide (57)

The α-benzyl ester (7) (4.16 g, 11.20 mmol) was dissolved in THF (30 mL) . To this solution was added tri- ethylamine (1.72 mL, 12.3 mmol), ethyl chloroformate (1.18 mL, 12.3 mmol) and 4-chloroaniline (1.79 g, 14 mmol) and the mixture stirred at 15° for 1 h. After this time the reaction was complete as determined by t.l.c (silica; 80% ether/ petrol) . Standard workup followed by removal of the solvents in vacuo yielded (57) (5.21 g, 93%) as a crystalline solid m.p. ,[α]D ²⁰ -5.21 ( c = 12.0 , chloroform).

^■ H NMR (CDCI3): δ 1.99 (lH m), 2.30 (3H m), 4.43 (1H m), 5.11 5.16 (4H 2 x AB's), 5.63 (1H d J = 8.0) , 7.22-7.48 (14H m) .

¹³ C NMR (CDCI3): δ 29.4 (CH ₂ ), 33.6 ^'" (CH ₂ ), 53.4 (CH), 67.3 (CH ₂ ), 67.6 (CH ₂ ), 121.0 (aro¬ matic CH), 128.1 (aromatic CH), 128.3 (aromatic CH), 128.4 (aromatic CH), 128.5 (aromatic CH), 128.7 (aromatic CH), 128.9 (aromatic CH), 129.0 (aromatic CH), 134.9 (aromatic C), 135.8 (2 x aromatic C), 136.6 (aromatic C), 156.7 (carbamate C=0), 170.2 (amide C=0), 171.6 (ester C=0). Example 59 L-Glutamic acid ₇ -4 ' -chloroanilide (58)

The protected 4'-chloroanilide (57) (2.00 g) was hydrogenated in methanol (10 mL) and ethyl acetate (5 mL) in the presence of 10% palladium on charcoal (186 mg) as cata¬ lyst. Standard workup of this mixture yielded the amino acid (58) (1.01 g, 94 %) as a colourless oil which was pure by t.l.c (silica; 70% ethanol/water) . [α] _D ²⁰ -7.1 ( c = 9.10 , water) .

-H NMR (D ₂ 0): δ 2.22 (2H m) , 2.65 (2H m), 3.83 (1H t J - 6.2), 7.25 (2H m) , 7.43 (2H m) . Example 60 Z-L-Glutamic acid α-benzyl ester η-piperidide (59)

The α-benzyl ester (7) (5.30 g, 14.27 mmol) was dissolved in THF (30 mL). To this solution was added tri¬ ethylamine (2.30 mL, 15 mmol), ethyl chloroformate (1.60 mL, 15 mmol) and piperidine ( 1.60 mL, 15 mmol) and the mixture stirred at 15° for 3 h. After this time the reaction was complete as determined by t.l.c (silica; 100% ether). Stan¬ dard workup followed by chromatography (silica; 100% chloro¬ form) of the crude product yielded (59) as a yellow oil (5.37 g, 86%) homogeneous by t.l.c (silica; 100% chloroform, R _f 0.16: 25% ethyl acetate/petrol, R _f 0.50) . [α] _D ² °-8.46 ( c = 13.5, chloroform). ^l H NMR (CDCI3): δ 1.41-1.54 (6H bm), 2.04 (1H m), 2.17 (1H m), 2.29 (2H m), 3.18 (2H m), 3.44 (2H

EET

- 178 -

m), 4.36 (1H m), 5.06 5.14 (4H 2AB's), 6.14 (1H d J = 7.5), 7.29 (10H m) .

¹³ C NMR (CDC1 ₃ ): . 24.1 (CH ₂ ), 25.1 (CH ₂ ) , 26.9 (CH ₃ ), 28.7 (CH ₂ ), 42.4 (CH ₂ ), 46.0 (CH ₂ ), 53.6 (CH), 66.3 (benzylic CH ₂ ), 66.7 (benzylic CH ₂ ) , 127.6 (aromatic CH) , 127.8 (aromatic CH) , 127.9 (aromatic CH) , 128.0 (aromatic CH) , 128.2 (aromatic CH), 135.1 (aromatic C), 136.1 (aromatic C), 155.9 (carbamate C=0), 169.5 (amide C=0), 171.7 (ester C=0) . Example 61 L-Glutamic acid η-piperidide (60)

The ₇ -piperidide (59) (4.03 g) was hydrogenated in methanol (20 mL), in the presence of 10% palladium on char¬ coal (430 mg) as catalyst. After 4 h the reaction mixture was worked up in the standard manner to yield the amino acid (60) (1.68 g, 85 %) homogeneous by t.l.c (silica; 70% ethanol/ water, R _f 0.65: cellulose;n-butanol/pyridine/0.4%acetic acid in water 22:10:10, R _f 0.50),m.p. 162-162.5°, [α] _D ²⁰ -2.24 ( c = 15.6, methanol) [α] ²⁰ +1.92 ( c = 11.5, water). ^X H NMR (D ₂ 0): δ 1.33 (6H m) , 1.91 (2H dt J = 7.4,6.1), 2.40 (2H t J = 7.4), 3.29 (4H m) , 3.57 (1H t J = 6.1).

¹³ C NMR (D ₂ 0): δ 23.6 (CH ₂ ) , 25.2 (CH ₂ ), 25.8 (CH ₂ ), 26.0 (CH ₂ ), 28.9 (CH ₂ ), 43.4 (CH ₂ ), 47.1 (CH ₃ ), 54.1 (CH), 171.9 (C=0) , 173.9 (C=0) . Example 62 Z-L-Glutamic acid α-benzyl ester η-3' -hydroxypiperidide

(61)

The α-benzyl ester (7) (8.53 g, 22.96 mmol) was dissolved in THF (30 mL) . To this solution was added tri¬ ethylamine (3.60 mL, 25.5 mmol), ethyl chloroformate (2.40 mL, 25.5 mmol) and 3-hydroxypiperidine ( 2.65 g, 26 mmol) and the mixture stirred at 15° for 5 h. After this time the reaction was complete as determined by t.l.c (silica; 100% ether). Standard workup followed by chromatography (silica; 100% ether, 100% ethyl acetate) of the crude product yielded

(61) as a colourless oil (8.70 g, 83%) homogeneous by t.l.c (silica; 100% ethyl acetate, R _f 0.11) . [α]_. ²⁰ -3.24 ₍ c = 10.5, chloroform). The ^~ E NMR spectrum indicated the product (61) was a mixture of two diastereoiso ers; attempts to separate the two isomers by HPLC (silica; 10% petrol/ethyl acetate) were unsuccessful.

^■ E NMR (CDC1 ₃ ): δ 1.3-2.2 (8H complex m ₎ , 3.1- 3.7 (5H complex m) , 4.42 (1H m), 5.08 5.16 ₍ 4H 2 x AB's), 5.8 (1H bs), 7.34 (lOH m). ¹³ C NMR (CDCls): δ 21.5 and 22.7 (CH ₂ ratio 1:1), 27.5 (CH ₂ β carbon), 28.9 (CH ₂ carbon),

31.8 and 32.3 (CH ₂ ratio 1:1), 42.3 and 45.7 (CH ₂ ratio 1:1), 48.7 and 52.1 (CH ₂ ratio 1:1), 53.7 (CH α carbon), 65.6 and 65.9 (CH-OH ratio 1:1 ₎ ,

66.9 (benzylic CH ₂ ) , 67.2 (benzylic CH ₂ ) , 128.1 (aromatic CH), 128.3 (aromatic CH), 128.4 (aromatic CH), 128.5 (aromatic CH), 128.6 (aromatic CH), 135.3 (aromatic C), 136.2 (aromatic C), 156.8 (carbamate C=0), 170.7 (amide C=0), 172.0 (ester C=0) .

Example 63

L-Glutamic acid η-3' -hydroxypiperidide (62)

The 7-3'-hydroxypiperidide (61) (7.30 g) was hydrogenated in methanol (20 mL), in the presence of 10% palladium on charcoal (800 mg) as catalyst. After 5 h the reaction mixture was worked up in the standard manner to yield the amino acid (62) (3.49 g, 95 %) homogeneous by t.l.c (silica; 70% ethanol/water, R _f 0.62: cellulose;n- butanol/ pyridine/0.4%acetic acid in water 22:10:10, R _f 0.26),m.p. 151-152°, [α] _D ²⁰ -4.00 ( c = 10.96, water). The chemical shift assignments in the proton spectra were confirmed by selective decouplings.

*H NMR (D ₂ 0): δ 1.4 - 2.0 (4H complex m), 2.11 (2H dt J = 6.7,6.6), 2.61 (2H t J = 6.7), 3.2 - 3.95 (6H complex m) .

¹³ C NMR (D _a O): δ 22.5 and 23.5 (CH ₃ ) , 27.4 (CH ³ ), 30.2 (CH2), 32.2 and 32.5 (CH*),, 44.0 and

SUBSTITUTE SHEET

- 180 -

47.4 (CHa)/ 49.4 and 53.3 (CH ₂ ), 55.4 (CH), 66.9 and 67.0 (CH-OH), 174.1 (amide C=0), 175.2 (acid C=0) . Example 64

Z-L-Glutamic acid α-benzyl ester η- [ 1- ( 4-methylpipera- zino) Jamide (63)

The α-benzyl ester (7) (5.80 g, 15.60 mmol) was dissolved in THF (40 mL) . To this solution was added tri¬ ethylamine (2.45 mL, 17.1 mmol), ethyl chloroformate (1.70 mL, 17.1 mmol) and l-amino-4-methylpiperazine ( 2.14 mL , 17.1 mmol) and the mixture stirred at 15° for 16 h. Over this time a precipitate formed in the reaction vessel. Standard workup of the reaction mixture yielded crude (63) which was further purified by recrystallisation (50% chloroform/ carbon tetrachloride) to furnish (63) as a crystalline solid (6.96 g, 95%) homogeneous by t.l.c (silica; 5% methanol/chloroform, R _f 0.30), m.p. 174°,[α] _D ²⁰ -6.00 ( c = 11.0, chloroform). Both ^~ E and ¹³ C NMR experiments in chloroform and pyridine indicate that two major conformations of the piperazine ring are present in solution. -

^~ E NMR (CDCls): 5 2.21 and 2.25 (3H CH ₃ N- methyl ratio 2:1), 1.92 - 2.79 (12H complex m), 4.39 (1H m α proton), 5.09 5.14 (4H 2 x AB's benzylic protons), 5.84 (1H d J = 7.3 NH proton), 7.33 (10H m aromatic protons). ¹³ C NMR (CDCls): δ 26.7 and 28.6 (β CH ₂ ratio 2:1), 28.1 and 30.9 ( g CH. ratio 2:1), 45.6 (CH ₃ N-methyl), 53.5 and 53.9 (α CH ratio 2:1), 54.1 (CH _S ), 54.3 (CH ₂ ), 55.3 (CH ₂ ), 56.2 (CH ₂ ), 66.8 and 67.3 (benzylic CHa ) , 67.0 (benzylic CH ₂ ) , 128.0 (aromatic CH) , 128.2 (aromatic CH) , 128.3 (aromatic CH), 128.4 (aromatic CH), 128.5 (aromatic CH) , 128.6 (aromatic CH), 135.1 (aromatic C) 136.2 (aromatic C), 156.1 and 156.3 (carbamate C=0 ratio 2:1), 169.1 and 171.7 (amide C=0 ratio 2:1), 171.9 and 174.9 (ester C=0 ratio 2:1). -H NMR (D5-Pyridine): δ 2.10 and 2.12 (2 s, total

- 181 -

3H, ratio 1:2, methyl protons), 2.42 - 3.04 (12H ), 4.90 and 4.95 (2 m, total 1H, ratio 1:2, alpha protons), 5.26 (4H m, benzylic protons), 7.28 (10H , aromatic protons), 8.43 (1H t J = 6.8, alpha NH proton) .

¹³ C NMR (D5-Pyridine): δ 27.41 and 28.09 ( β CH ₂ ratio 2:1), 28.86 and 31.30 (7 CH ₂ ratio 2:1), 45.65 and 45.77 ( N-methyl CH ₃ ratio 2:1), 54.99 (CHa), 55.00 (CH ₂ ), 55.07 (2 x CH ₂ ) , 56.32 (α CH), 66.55 (benzylic CH ₂ ) , 66.83 (benzylic CH ₂ ) , 128.1 (aromatic CH) , 128.3 (aromatic CH) , 128.4 (aromatic CH), 128.7 (aromatic CH), 128.8 (aromatic CH), 136.6 (aromatic C), 137.6 (aromatic C), 157.4 (carbamate C=0), 169.5 and 172.9 (amide C=0 ratio 2:1), 173.1 and 174.8 (ester C=0 ratio 2:1) .

Example 65

L-Glutamic acid η-[l- (4-methylpiperazino) Jamide (64)

The ₇ -[l-(4-methylpiperazino) ]amide (63) (3.04 g) was hydrogenated in methanol (20 mL), in the presence of 10% palladium on charcoal (322 mg) as catalyst. After 3 h the reaction mixture was worked up in the standard manner to yield the amino acid (64) (1.56 g, 98 %) homogeneous by t.l.< (silica; 70% ethanol/water, R _f 0.10: cellulose;n- butanol/ pyridine/0.4%acetic acid in water 22:10:10, R _f 0.05),m.p. 190-192°, [α] _D ²⁰ -5.45°( c = 8.2, water).

-R NMR (D ₂ 0): δ 2.09 (2H m β protons), 2.33 (2H m g protons), 2.38 (3H s N-methyl), 2.75 (4H m), 2.87 (4H m), 3.70 (1H m α proton). ¹³ C NMR (D ₂ 0): δ 27.9 (CH ₂ ), 31.2 (CH ₂ ), 45.2 (CH ₃ N-methyl), 54.3 (2 x CH ₂ ), 54.6 (2 x CHa), 55.5 (CH), 173.2 (amide C=0), 175.8 (acid C=0) .

S B TIT TE HE ^"

Example 66

Z-L-Glutamic acid α-benzyl ester η- (4-morpholino) amide (65)

The α-benzyl ester (7) (4.04 g, 10.87 mmol) was dissolved in THF (30 mL) . To this solution was added tri¬ ethylamine (1.70 !_!,_ 12.0 mmol), ethyl chloroformate (1.13 mL, 12.0 mmol) and 4-aminomorpholine ( 1.19 mL , 12.0 mmol) and the mixture stirred at 15° for 2 h. Over this time a precipitate formed in the reaction vessel. Standard workup followed by chromatography (silica; 15% ethyl acetate/petrol) of the crude product yielded (65) which was further purified by recrystallisation (50% chloroform/ carbon tetrachloride) to furnish (65) as a crystalline solid (3.16 g, 64%) homoge _¬ neous by t.l.c (silica; 10% petrol/ethyl acetate + 1 drop of acetic acid, R _f 0.22), m.p. 150-151°,[α] _D ²⁰ -3.80 (c = 10.4, chloroform). Both ^l E and ¹³ C NMR experiments at 300 K and 325 K indicate that two major conformations of the morpholine ring are present in solution.

-E NMR (CDC1 ₃ ): δ 2.02 ( m) , 2.13 (3H m), 2.52 (m), 2.75 (5H m), 3.63 (bs), 3.73 (4H t J = ), 4.39 (1H ddd J =), 5.08 5.14 (4H AB's), 5.91 (1H dd J =), 6.91 (1H bs), 7.04 (lH bs), 7.32 (lOH m). ¹³ C NMR (CDCls): δ 26.5 and 28.3 (CH ₂ ratio 3:2), 28.0 and 30.8 (CH ₂ ratio 3:2), 53.4 and 53.8 (CH ratio 2:3), 55.2 and 56.4 (CH ₂ ratio 2:3), 66.1 (benzylic CH ₂ ), 66.2 (benzylic CH ₂ ), 66.7 and 66.9 (CH ₂ ratio 2:3), 67.0 and 67.2 (CH ₂ ratio 3:2), 127.9 (aromatic CH), 128.0 (aromatic CH), 128.1 (aromatic CH), 128.2 (aromatic CH), 128.3 (aromatic CH), 128.4 (aromatic CH) , 128.5 (aromatic CH), 135.0 and 135.2 (aromatic C ratio 2:3), 135.9 and 136.1 (aromatic C ratio 2:3), 155.9 and 156.3 (carbamate C=0 ratio 3:2), 169.2 and 171.6 (amide C=0 ratio 3:2), 171.9 and 175.1 (ester C=0 ratio 3:2). Example 67 L-Glutamic acid η-(4-morpho lino) amide (66)

The 7-(4-morpholino)amide (65) (1.97 g) was hydrogenated in methanol (20 mL), in the presence of 10% palladium on charcoal (222 mg) as catalyst. After 2.5 h the reaction mixture was worked up in the standard manner to yield the amino acid (66) (991 mg, 99 %) homogeneous by t.l.c (silica; 70% ethanol'/water, R _f 0.63: cellulose;n- butanol/ pyridine/0.4%acetic acid in water 22:10:10, R _f 0.15),m.p. 200°, [α] _D ²⁰ -2.31 ( c = 10.2, water). H NMR (D ₂ 0): δ 2.11 (2H ddd J = 6.9,7.1,7.2 protons), 2.33 (2H m g protons), 2.83 (4H t J = 3.8 ), 3.75 (1H t J = 6.9 α proton), 3.80 (4H t J = 3.8) .

¹³ C NMR (D ₂ 0): δ 27.6 (CH ₂ ), 31.3 (CHa), 55.5 (CH), 56.2 (2 x CH ₂ ), 67.4 (2 x CH ₂ ) , 173.3 (amide C=0), 175.3 (acid C=0) .

Z-L-Glutamic acid α-benzyl ester η-2' fluoroethylamide (67)

The α-benzyl ester (7) (2.50 g, 6.70 mmol) was dissolved in THF (30 mL) . To this solution was added tri¬ ethylamine (1.03 mL, 7.40 mmol), ethyl chloroformate (0.65 mL, 7.40 mmol) and 2-fluoroethylamine hydrochloride ( 0.72 g, 7.4 mmol) in triethylamine (1 mL) and the mixture stirred at 15° overnight. After this time the reaction was complete as determined by t.l.c (silica; 70% ether/petrol). Standard * workup followed by chromatography (silica; 70% ether/petrol) of the crude product yielded (67) as a crystalline solid (2.71 g, 91%) homogeneous by t.l.c (silica; 70% ether/ petrol),[α] _D ²⁰ -9.20 ( c = 13.0, chloroform).

-R NMR (CDCls): -5 1.70 (lH m), 2.00 (lH m), 2.22 (2H m), 3.46 (lH m), 3.55 (lH m), 4.37 (2H m CH ₂ - F), 4.53 (1H m), 5.09 5.17 (4H 2 x AB's benzylic protons), 5.69 (1H bs NH proton), 6.11 (1H bs NH proton), 7.34 (10H m aromatic protons) •» .

SUBSTITUTE SHEET

¹³ C NMR (CDCI3): δ 28.4 (CH ₂ ), 32.2 (CHa), 39.8 (CHa), 40.1 (CHa), 53.5 (CH) , 67.1 (benzylic CH ₂ ), 67.3 (benzylic CH ₂ ) , 128.1 (aromatic CH) , 128.2 (aromatic CH), 128.4 (aromatic CH), 128.5 (aromatic CH) , 128.6 aromatic CH) , 135.1 (aromatic C), 136.1 (aromatic C), 156.3 (carbamate C=0), 171.7 (amide C=0 ) , 172.0 (ester C=0). Example 69

Z-L-Glutamic acid α-benzyl ester η-2' , 2' , 2' - trifluoroethylamide (68)

The α-benzyl ester (7) ( 6.10g, 16.43 mmol) was dissolved in THF (50 mL) . To this solution was added tri¬ ethylamine (2.60 mL, 17.8 mmol), ethyl chloroformate (1.8 mL, 17.8 mmol) and 2,2,2-trifluoroethylamine ( 1.45 ml, 17.8 mmol) and the mixture stirred at 15° overnight. After this time the reaction was complete as determined by t.l.c (silica; 50% ethyl acetate/petrol) . Standard workup followed by chromatography (silica; 50% ethyl acetate/petrol) of the crude product yielded (68) as a crystalline solid (6.81 g, %) homogeneous by t.l.c (silica; 50% ethyl acetate/petrol).

^~ E NMR (CDCI3): δ 1.96 (1H m), 2.25 (3H m), 3.81 (2H m), 5.10 (s), 5.17 (4H m) , 5.65 (1H d, J = 8.0), 6.43 (1H b), 7.27 (lOH m). ¹³ C NMR (CDCI3): δ 28.6 (CH ₂ ), 32.1 (CH ₂ ),

40.3 (CH ₂ q, J = 35.0), 53.4 (CH) , 67.2 (CHa),

67.4 (CH ₃ ), 124.0 (C q, J = 293), 128.1 (aromatic CH), 128.3 (aromatic CH), 128.4 (aromatic H) 128.6 (aromatic CH) , 128.7 (aromatic CH) , 135.1 (aromatic C), 136.0 (aromatic C), 156.5 (carbamate C=0), 171.6 amide C=0), 172.1 (ester C=0) .

Example 70

L-Glutamic acid η-2' , 2', 2' -trifluoroethylamide (69)

The trifluoroethylamide (68) (3.26 g) was hydroge¬ nated in methanol (50 mL), in the presence of 10% palladium on charcoal (478 mg) as catalyst. After 1.5 h the reaction mixture was worked up in the standard manner to yield the amino acid (69) (1.65 g, 99 %) homogeneous by t.l.c (silica;

70 ^% ethanol/water, R _f 0.70: cellulose;n- butanol/pyridine/ 0.4 ^% acetic acid in water 22:10:10, R _f 0.57), .p. 215°d. ^~ E NMR (D ₂ 0): δ 2.13 (2H m) , 2.48 (2H dt, J =

* * 7.7,3.3), 3.75 (1H t, J = 6.2), 3.93 (2H q, J = 9.3).

¹³ C NMR (D ₂ 0): δ 27.6 (CH ₂ ) , 32.7 (CH ₂ ) , 41.8 (CH ₂ q, J = 34), 55.6 (CH), 125.7 (C q, J = 277), 175.3 (amide C=0), 176.8 (acid C=0). Example 71

Z-L-Glutamic acid α-t-butyl ester η-O-methylhydroximate ( 70)

Diazomethane was prepared (see : Hudlicky, M.(1980) J. Org. Chem., 45., 5377- 5378.) by dropping a solution of dizald (9 g) in ether (60 mL) into a stirred solution of potassium hydroxide (2.8 g) in water (5 mL), diethylene glycol monomethyl ether (15 mL) and ether (10 mL) in an ice bath. This mixture was allowed to warm to room temperature, then heated in a water bath to 50° and the diazomethane dis _¬ tilled in the usual manner. To the distilled ethereal diazo _¬ methane was added a solution of hydroxamate (23) (4.0 g ₎ in ether (20 mL) and the mixture cooled in an ice bath. The mixture was allowed to stand for 3 h after which time a crystalline compound had precipitated. Excess diazomethane was destroyed with acetic acid and the crystalline solid reclaimed by filtration. T.l.c analysis of this material (silica 80% ether / petrol) showed it to be starting hydrox _¬ amate (23), so the whole mixture was resubmitted to fresh diazomethane and allowed to stand at room temperature over _¬ night. After this time no diazomethane remained and a mixture of three products was observed by t.l.c (silica 45% ether / petrol, Rf 0.36, 0.23, 0.10). Removal of the solvents yielded the crude mixture as a slightly green oil (3.72 g, 93%). Chromatography of the mixture (silica 45% ether / petrol) yielded pure samples of the three compounds (Rf 0.36, 2.35 g, 59% ; Rf 0.23, 600 mg, 15% ; Rf 0.10, 350 mg, 8%). The highest Rf component was found to be the O- methylhydroximate (70) by NMR spectroscopy while the lower Rf components were

SUBSTITUTE SHEET

found to be dimethylation products. Synthesis of the 0- methylhydroxa ate by a different method (see example 46) indicated that methylation had not occurred at the hydroxa _¬ mate oxygen. Analysis of the spectra suggested the isolated compoun ^d was the 0- methylhydroximate. The product ₍ 70 ₎ was isolated as a viscous oil, [α] _D ² °-7.0 ( c = 10.19, meth _¬ anol). Subsequent repetition of this experiment furnished low y ⁱ elds of (23) and higher yields of the dimethylation products. ^l E NMR (CDC1 ₃ ): δ 1.38 (9H s) 1.88 (IH ddd J = 8.1, 14.2, 14.2) 2.10 (IH ddd J = 6.6, 13.2, 13.3 ₎ 2.29 (2H d AB J = 6.7, 8.1) 3.57 (3H s) 4.20 (IH dd J - 7.7, 12.9) 5.02 (2H s) 5.42 (IH bd J = 6) 7.28 (5H m) .

13,

C NMR (CDCI3): δ 27.64 (CH2) 27.73 (CH3) 29.78 (CH2) 51.51 (CH3) 53.63 (CH) 66.68 (CH2) 82.11 (C) 127.89 (aromatic CH) 128.27 (aromatic CH) 136.13 (aromatic C) 155.78 (carbamate) 170.78 (imide) 172.98 (ester). Minor component (Rf 0.23): 0 ,O-dimethylated.

-E NMR (CDCls): δ 1.46 (9H s) 1.97 (IH m) 2.06 (lH m) 2.38 (2H m) 3.62 (3H s) 3.71 (3H s) 4.29 (IH dd J = 7.4, 12.8) 5.01 (2H s) 5.69 (IH bd J = 6) 7.33 (5H m) .

¹³ C NMR (CDCI3): δ 22.93 (CH2) 27.61 (CH3) 27.72 (CH2) 53.76 (CH3) 53.85 (CH) 61.20 (CH3 ₎ 66.46 (CH2) 81.74 (C) 127.75 (aromatic CH) 128.17 (aromatic CH) 136.21 (aromatic C) 155.64 (car _¬ bamate) 163.65 (imide) 170.80 (ester). Minor component (Rf 0.10): 0,N-dimethylated. ^l E NMR (CDCls): δ 1.43 (9H s) 1.96 (IH m) 2.15 (lH m) 2.47 (2H m) 3.12 (3H s) 3.59 (3H s) 4.26 (IH m) 5.06 5.09 (2H AB) 5.65 (IH bd J = 6.5) 7.31 (5H m) .

¹³ C NMR (CDCI3): δ 22.10 (CH2) 27.33 (CH3) 31.63 (CH2) 53.41 (CH) 53.62 (CH3 ₎ 60.52 ₍ CH3 ₎ 66.16 (CH2) 81.49 (C) 127.43 (aromatic CH) 327.84

(aromatic CH) 135.85 (aromatic C ₎ 155.45 ⁽ carbamate) 170.64 (imide) 172.82 ₍ ester ₎ . Example 72

^Z - ^L -Glutamic acid i-O-methylhydroximate ₍ 7 _{1 )}

The fully protected hydroximate (70) (2.10 g 5 73 mmol) was dissolved in TFA (9 mL) at 0°. The reaction mixture as stirred for 1 h and monitored by t.l.c (silica 80% ether / petrol). After this time the TFA was removed in vacuo at ca. 20° and the resulting solid recrystallised from ether / petrol (60 : 40) to yield pure (71) (1.73 g, 98%).

^: H NMR (CDC1 ₃ ): δ 2.04 (IH dddd J = 7.0, 7.4, 8.0, 14.5 ⁾ 2.25 (IH dddd J = 5.0, 7.0, 7.7, 14 5 ₎ 2.46 (2H ddddd J = 1.0, 7.0, 7.4, 7.7, 8.0) 3.66 (3H ^β ) 4.43 (iH ddd J = 1.0, 5.0, 7.0) 5.11 (2H s) 5.53 (IH d J = 7.7) 5.94 (IH b) 7.30 ₍ 5H m ₎ . All couplings were confirmed by decoupling experiments.

C NMR (CDCI3): δ 27.28 (CH2) 30.03 (CH2) 51.96 (CH3) 53.19 (CH) 67.27 (CH2) 128.13 (aromatic CH ⁾ 128.28 (aromatic CH) 128.56 ₍ aromatic CH ₎ 135.96 (aromatic C) 156.19 (carbamate) 173.53 (imide) 175.60 (acid). Example 73

^L -Glutamic acid i-O-methylhydroximate ₍ 72 ₎

The free acid (71) (1.45 g, 4.69 mmol) was taken up xn methanol (18 mL) and 10% palladium on charcoal (145 mg) added. This mixture was hydrogenated in the standard manner for 1 h at room temperature, after which time no starting material remained. The hydrogenation mixture was filtered and the solids washed with water (2 x 10 mL). The solvents were removed from the combined washes and filtrate to yield the crude product. Recrystallisation from boiling methanol yielded the pure amino acid (72) (670 mg, 82%) as a white

T EET

crystalline solid, m.p. 186-187°, [α] _D ²⁰ +9.82 (c = 12.46, H20) .

^L H NMR (D ₂ 0/DC1): δ 2.01 (2H ddddd J = 1.5, 2.0, 6.3, 7.2, 7.9) 2.43 (2H dddd J = 1.5, 2.0, 7.2, 7.9) 3.56 (3H s) 3.62 (IH t J = 6.3). All couplings were confirmed by decoupling experiments. ¹³ C NMR (D ₂ 0): δ 24.30 (CH2) 28.56 (CH2) 51.21 (CH3) 52.81 (CH) 172.71 (imide) 174.18 (acid). Example 74

Z-L-Glutamic acid α-benzyl ester η- ( 2' -pyridyl) methylamide ( 73)

The α-benzyl ester (7) (7.87g, 21.19 mmol) was dissolved in THF (100 mL) . To this solution was added tri¬ ethylamine (5.91 mL, 42.4 mmol), ethyl chloroformate (2.22 mL, 23.4 mmol) and 2-aminomethylpyridine (2-picolylamine) (2.40 ml, 23.4 mmol) and the mixture stirred at 15° over¬ night. After this time the reaction was complete as deter¬ mined by t.l.c (silica; ethyl acetate). Standard workup followed by chromatography (silica; ether 400 mL, ethyl acetate 500 mL, 10% methanol/chloroform 500 mL) of the crude product yielded (73) as a crystalline solid (9.57 98%) homogeneous by t.l.c (silica; 100% ethyl acetate) m.p 132- 134°C. Attempts to record NMR spectra in CDC1 ₃ resulted in decomposition of this compound.

^: H NHR (CDgOD): δ 2.0 - 2.35 (2H m), 2.39 (2H m), 4.35 (IH dd J = 4.2, 4.3), 4.43 (2H s), 5.02 -

5.09 (4H AB system), 7.36 (12H m), 7.74 (lH m),

8.42 (IH d J = 7.4).

¹³ C NMR (CDsOD): δ 27.9 (CH ₂ ), 32.0 (CH ₂ ),

43.4 (CH ₂ ), 53.6 (CH), 66.9 (CH ₂ ), 67.2 (CH ₂ ),

123.0 (hetero aromatic CH) , 123.3 (hetero aromatic CH), 128.1 (aromatic CH), 128.3 (aromatic CH) 128.4 (aromatic CH), 128.6 (aromatic CH), 139.0 (aromatic C), 146.7 (hetero aromatic CH), 156.5 (carbamate C=0), 171.6 amide C=0) , 172.1 (ester C=0).

Example 75

L-Glutamic acid ₇ - ( 2' -pyridyl) methylamide ( 74)

The 2-aminomethylpyridine amide (73) (4.76 g) was hydrogenated in methanol (50 mL), in the presence of 10% palladium on charcoal (650 mg) as catalyst. After 24 h the reaction mixture was worked up in the standard manner to yield the amino acid (74) (2.43 g, 99 %) homogeneous by t.l.c (silica; 70% ethanol/water, R _f 0.60: cellulose;n- butanol/ pyridine/0.4%acetic acid in water 22:10:10, R _f 0.38) ,m.p. 185-189°.

^~ E NMR (D ₂ 0): δ 2.17 (2H m), 2.53 (2H m), 3.80 (IH t, J = 6.2), 4.42 (2H d J = 7.4), 7.31 (2H m ), 7.77 (IH m), 8.39 (IH d J = 7.3). ¹³ C NMR (D ₂ 0): δ 27.8 (CH ₂ ), 32.9 (CH ₂ ) , 45.7 (CH ₂ ), 55.6 (CH), 123.2 (hetero aromatic CH) ,124.5 (hetero aromatic CH), 139.8 (hetero aromatic CH), 149.8 (hetero aromatic CH), 157.8 (hetero aromatic C), 175.4 (amide C=0), 176.3 (acid C=0) . Example 76

Z-L-Glutamic acid α-benzyl ester η-prop-2-enamide (allyl amide) ( 75)

The α-benzyl ester (7) ( 3.33g, 8.90 mmol) was dissolved in THF (50 mL). To this solution was added tri¬ ethylamine (1.24 mL, 9.0 mmol), ethyl chloroformate (0.86 mL, 9.0 mmol) and allyl amine ( 0.68 ml, 9.0 mmol) and the _* mixture stirred at 15° overnight. After this time the reac- tion was complete as determined by t.l.c (silica; ethyl ^x _ acetate). Standard workup followed by chromatography (silica;

80% ether/petrol) yielded (75) as a crystalline solid (3.31 gm, 85%) homogeneous by t.l.c (silica; 100% ether ).

^X H NMR (CDCls): δ 2.0 - 2.18 (4H m) , 3.80 (2H ddd J = 5.7, 5.5, 2.7), 4.38 (IH m) , 5.08 - 5.20 (6H m), 5.77 (IH dddd J = 16.1, 10.6, 5.7, 5.5) . 5.89 (IH bd J ~5.8), 6.06 (IH bd J ~ 5.5), 7.32 (10H m) .

¹³ C NMR (CDCls): δ 28.2 (CHa), 32.1 (CH ₂ ), 41.9 (CH ₂ ), 53.6 (CH), 66.9 (CHa), 67.2 (CHa), 116.3 (olefinic CH2), 128.0 (aromatic CH) , 128.1 (aromatic CH) , 128.4 (aromatic CH) 128:5 (aromatic CH), 133.9 (olefinic CH) , 135.1 (aromatic C), 136.1 (aromatic C), 156.3 (carbamate C=0), 171.7 amide C=0), 172.1 (ester C=0) . Example 77 L-Glutamic acid η-propylamide ( 76)

The allyl amide (75) (3.00 g) was hydrogenated in methanol (50 mL), in the presence of 10% palladium on char¬ coal (300 mg) as catalyst. After 24 h the reaction mixture was worked up in the standard manner to yield the amino acid (76) (1.06 g) homogeneous by t.l.c (silica; 70% ethanol/ water, cellulose;n-butanol/pyridine/0.4%acetic acid in water 22:10:10),m.p. 205-207°.

^~ E NMR (D ₂ 0): δ 0.87 (3H t J = 7.3), 1.49 (2H qt J = 7.3, 7.0), 2.11 (2H dt J = 7.4, 6.1), 2.38 (2H m), 3.12 (2H t, J = 7.0), 3.74 (IH t J = 6.1). ¹³ C NMR (D ₂ 0): 5 12.1 (CH3)., 23.2 (CH2), 28.1 (CH ₂ ), 33.1 (CH ), 42.7 (CH ₂ ), 55.6 (CH) , 175.4 (amide C=0), 175.9 (acid C=0). Example 78

Z-L-Glutamic acid α-benzyl ester η- (2' -mesyloxyethyl ) amide ( 77)

The 2'-hydroxyethylamide (27) ( 4.45g, 10.8 mmol) was dissolved in DCM (150 mL) . To this solution was added triethylamine (3.03 mL, 21.6 mmol) and methanesulphonyl chloride (mesyl chloride) ( 1.26 ml, 15.8 mmol) and the mixture stirred at 25°. Immediately upon addition of the mesyl chloride the solution became quite yellow. After 15 min the reaction was complete as determined by t.l.c (silica; 85% ethyl acetate/petrol). The DCM was removed in vacuo and the resultant oil taken up in ethyl acetate. This solution was washed with saturated sodium chloride and the organic layer dried over sodium sulphate. Removal of the solvents from the organic layer yielded crude (77) as a yellow oil.

Chromatography of the crude product (silica; 85% ethyl acetate/petrol) yielded (77) as a white solid (2.67 g, 62%) homogeneous by t.l.c (silica; 85% ethyl acetate/petrol). ^■ E NMR (CDC1 ₃ ): δ 1.98-2.23 (4H m) , ^' 2.71 (3H s), 3.50 (2H t J = 7.2), 4.22 (2H t J = 7.2), 4.31 (IH m), 5.08-5.15 (4H AB system), 5.89 (IH bd J = 6.0) 7.33 (10H m) . Example 79 Dihydrooxazole analogue (

The 2'-mesyloxyethylamide (77) (872 mg, 1.77 mmol) was dissolved in anhydrous DCM (5.0 mL) . To this solution wa added triethylamine (0.5 mL, 4 mmol) and the solution stirre at room temperature for 16h. After this time the solvents were removed in vacuo and the resultant oil was dissolved in ethyl acetate (10 mL) . The ethyl acetate solution was washed with saturated sodium chloride and dried over sodium sul¬ phate. Removal of the solvents from this solution yielded a white solid (689 mg, 98%), homogenious by t.l.c (silica; 85% ethyl acetate/petrol).

-E NMR (CDCI3): δ 2.05-2.30 (4H m) , 3.76 (2H t J

= 9.3), 4.17 (2H t J = 9.3), 4.47 (IH m) , 5.05-5.18

(4H AB system), 7.30 (10H m) .

Examples 80-81 illustrate, but do not limit in any manner, construction of the multi-well plate lids of this invention. Example 80

This Example illustrates the construction of a 6 well plate lid in accordance with this invention. A tissue culture multi-well plate (Linbro) which contained 6 flat bottom round wells (3.5 cm diameter, 1 cm depth) was used as the basis for the transport assay plate. A mark was made on the top of the lids which corresponded to the centre of each well when the plate was assembled. Six holes (1 cm diameter)

SUBSTITUTE SHEET

were drilled through the cover of the plate assembly cor _¬ responding to each of the marks using a hole saw with a 14 mm attachment. Care was necessary during drilling to avoid fracturing the plastic plate cover. Scintillation vials (Kartell, 4 ml capacity) were placed through each hole with the opening of each vial facing downwards and the body of the vial protruding out of the top surface of the plate cover. Each vial projected approximately 0.5 cm through the under _¬ side of the cover and was glued into place using a cyanoacry- late adhesive (Netweork, 800E100, rubber and plastic grade). The glue was left to dry naturally for approximately 24 hours. FIG. 68 illustrates the 6 well plate lid, and FIGS. 69 and 70 illustrate the lid in relationship with a 6 well plate. Example 81

This Example illustrates the construction of a 96 well plate lid in accordance with this invention. A tissue culture multi-well plate (Linbro) which contained 96 flat bottomed wells was used as the bottom half of the transport assay plate. For transport experiments, the cover of the 96- well plate was replaced with a modified cover made from the inner rack of a pipette tip rack holder (Flow) . The inner rack contained 96 channels (approximately 3 cm in length and open at either end) which corresponded perfectly with the 96 wells of the tissue culture plate. The pipette rack also contained legs at the bottom of each corner, which by coinci _¬ dence, firmly clipped the 96-well plate into place under each of the channels. The upper end of the pipette rack was sealed off by glueing clear perspex onto it in the following way. A piece of perspex (0.55 cm thick) was cut to the same dimensions as the pipette rack and was placed over the open channels on the upper surface of the rack. A grid was marked on the perspex which corresponded to parts of the perspex which made direct contact with the plastic of the pipette rack (i.e. around the opening of each of the channels). Raised numbering and lettering on the top of the pipette rack was filed away to provide a smooth surface for glueing. Glue

*

- 193 -

( _A crifix 92, Rohm) was applied onto the grid marks on the perspex and then the pipette rack and the perspex were squeezed together, with the glue on the perspex fusing with the plastic around the channels at the top of the pipette rack. The assembly was left for two hours before exposing to UV light for two hours to fix the glue. FIGS. 71 and 72 illustrate a 96 well plate lid, and FIGS. 73 and 74 illustrate the lid in relationship with a 96 well plate.

It will be apparent to those skilled in the art that various modifications and variations can be made to the compositions of matter and methods of this invention. Thus, it is intended that the present invention cover the modifi¬ cations and variations of this invention provided they come within the scope of the appended claims and their equivalents