WO1992001811A1

WO1992001811A1 - Assay for ctp-synthetase mutations which cause multidrug resistance

Info

Publication number: WO1992001811A1
Application number: PCT/GB1991/001243
Authority: WO
Inventors: Mark Meuth
Original assignee: Imperial Cancer Research Technology Limited
Priority date: 1990-07-25
Filing date: 1991-07-25
Publication date: 1992-02-06

Abstract

Tumor cell mutations in exons (4, 5 or 7) of the CTP synthetase gene or in an intron governing splicing thereof are assayed by RFLP- or PCR-based methods etc to reveal the presence of the MDR (multiple drug resistance) phenotype.

Description

Assay for CTP-synthetase mutations which cause multidrug resistance

BACKGROUND AND PRIOR ART

The stringent regulation of intracellular levels of cytidine nucleotides in mammalian cells is essential for maintaining the accuracy of DNA replication and governs sensitivity to several cytotoxic drugs. These nucleotides are synthesized in both prokaryotic and eukaryotic cells by the amination of UTP in the reaction UTP+ATP+glutamine—> CTP+ADP+P_i+glutamate. The enzyme catalyzing this step, CTP synthetase, is also a key regulatory point as its activity is subject to activation by GTP and inhibition by CTP, and mutant cell strains altered in this regulation (ie insensitive to inhibition by CTP) display a complex phenotype: i) increased intracellular pools of CTP and dCTP, ii) multidrug resistance, and iii) an increased rate of spontaneous mutation.

Since many nucleotide and base analogs are widely used in the chemotherapy of leukemias and other tumors, the development of such genetically codominant mutations in target cell populations could have important consequences for the outcome of the treatment. More specifically, the multidrug resistance phenotype (MDR) is relevant in the context of treatment with a number of agents which are the mainstays of modern hematological chemotherapy, for example arabinosyl cytosine (AraC), which is used for treatment of acute myelogenous leukemia (AML) , acute ly phocytic leukemia (ALL) and non-Hodgkin's lymphomas; 5-fluorouracil (5-FU), used against several solid tumors, including colorectal carcinomas; and DNA alkylating agents such as cyclophosphamide (used against high grade lymphoma, ALL), melphalan (used against myeloma) and chlorambucil (used against low grade lymphoma, CLL) . These agents are also used singly or in combination to treat teεticular tumors, melanomas, ovarian tumors, oat cell carcinomas of the lung and stomach tumors.

OBJECTS OF THE INVENTION

It is an object of the invention to provide an assay for CTP synthetase mutations which cause MDR.

SUMMARY OF THE INVENTION

One aspect of the invention provides a genetic assay method comprising the steps of (1) obtaining a polynucleotide having the same nucleotide sequence as the CTP synthetase gene, or a relevant part thereof, of the tumor cell of an organism and (2) assaying that polynucleotide for a mutation in exons 4, 5 or 7 or for a mutation in an intron which controls the splicing of any of the said exons, the mutation being indicative of the CTP synthetase encoded by the said gene having deficient regulation by CTP.

The "relevant part" of the CTP synthetase gene is any part which contains the region(s) in which the mutation is being sought.

Preferably, the assay is capable of detecting mutations in the polynucleotide which give rise to amino acid changes in the region 110-236, especially in the sub- regions 110-120, 148-166 and 230-236 and particularly at amino acid positions 114, 152, 154, 157, 161, 164 or 235.

Assays based on mutations in the said exons are preferred to those based on mutations in the introns controlling splicing of the exons. The polynucleotide may be genomic DNA, cDNA or mRNA or it may have been cloned or amplified from any of these. (In the case of RNA, of course, the sequence will be the same as the CTP gene but with uracil bases in place of thymidine bases.) The only important criterion is that the polynucleotide being assayed should represent the region of the CTP synthetase gene in which the said mutation is being sought. DESCRIPTION OF PREFERRED EMBODIMENTS

The assay technique may be chosen from those which are available and convenient in the art at the time. Currently, these include sequencing the nucleotide sequence by techniques such as the Sanger or Maxam/Gilbert methods; differential hybridization of an oligonucleotide probe designed to hybridize at the relevant positions of either the wild-type or mutant sequence; denaturing gel electrophoresis following digestion with an appropriate restriction^" enzyme, preferably following amplification of the relevant DNA regions; SI nuclease sequence analysis; non-denaturing gel electrophoresis, preferably following amplification of the relevant DNA regions; conventional RFLP (restriction fragment length polymorphism) assays; or DNA amplification using oligomicleotides which are matched for the wild-type region and unmatched for the mutant region or vice versa.

Preferably, a non-denaturing gel is used to detect differing lengths of fragments resulting from digestion with an appropriate restriction enzyme. The DNA is usually amplified before digestion, for example using the polymerase chain reaction (PCR) method disclosed in Saiki et al and modifications thereof. Otherwise 10-100 times more DNA would need to be obtained in the first place, and even then the assay would work only if the restriction enzyme cuts the nucleic acid infrequently.

An "appropriate restriction enzyme" is one which will recognise and cut the wild-type sequence and not the mutated sequence or vice versa. Various enzymes are disclosed below as specific examples, but any enzyme which cuts at the same place (an "isoschizomer") or which recognises the same sequence and cuts the DNA at a point within or adjacent the sequence will be suitable; more are being discovered all the time. It is convenient if the enzyme cuts nucleotides only infrequently, in other words if it recognises a sequence which occurs only rarely.

Restriction enzymes useful in connection with the mutations described above include, for example, Sphl (GCATGC) and StuI (AGGCCT) . One of the specific mutations found (G—>A at 546) removes the SpΛl site. There are numerous possible gained sites, for example Mspϊ and TaqI sites may be created in mutations.

In a PCR-based assay, the amplification or not of the DNA is used as the basis of the assay, rather than just as a means of generating enough DNA for the assay to be performed. If the primer matches the wild-type sequence, then wild-type DNA will be amplified and mutant DNA will not. If the primer matches a mutant sequence (it is convenient for the PCR primer to have the mutation as its 3'-most nucleotide), then matching mutant DNA will be amplified whereas wild-type and non-matching mutant DNA will not. The former approach is more indicative of the presence of a mutation at the relevant site but the latter approach is more informative regarding the nature of the mutation and is more sensitive. A positive reaction can be confirmed by using a primer for amplification of the opposite strand.

The nucleic acids, usually DNA rather than RNA, which are assayed may be obtained from any appropriate tumor cell of the body, for example a biopsy sample from a solid tumor or a white blood cell obtained from blood samples taken from patients at presentation or during or after chemotherapy. The mutations appear to be tumor-specific and to arise during tumor development. They may be enriched during a first round of drug therapy. Preferably, the DNA is extracted by known techniques and a specific region of the CTP synthetase sequence is amplified using the PCR. In an RFLP-based assay, it is then digested with the restriction enzyme and subjected to PAGE (polyacry1amide gel electrophoresis). The gel is stained and photographed to reveal a pattern of fragments indicative of whether the patient is MDR. The stain will usually comprise a labelled probe which will hybridise to one or more of the nucleotide fragments in question, the label being any agent which can be visualized, for example a radioactive atom. The whole procedure, using current technology, takes about 5 hours.

The phenotype may develop as a result of spontaneous mutation in a target cell or as a result of the mutagenic action of chemotherapeutic agents. Cells with the mutation have a growth advantage in the normal cellular milieu or during chemotherapy and are therefore selected for.

A kit may be provided, according to another aspect of the invention, to perform the assay. A kit in which the assay is based on selective amplification by PCR will typically contain the primer(s) needed for the PCR amplification and also control nucleic acid for both MDR and non-MDR phenotypes, so that the results of the assay can be analyzed more readily. An RFLP kit will typically contain restriction enzyme(s) and probe(s) and, optionally, PCR primers for general (non-selective) amplification of the DNA prior to RFLP analysis. The assays of the invention will be valuable in relation to human medicine and may be used prior to treatment with a drug which is not metabolised correctly in the MDR phenotype. The drug may be any of those mentioned above, especially araC. Instead of, or as well as, increasing the dose of the drug to compensate for the MDR phenotype, a different drug, the metabolism of which is unaffected or affected desirably by the MDR phenotype, may be administered.

Furthermore, the mutations may be important in tumor progression and metastasis and detecting them may therefore be useful for monitoring tumor progression. Thus, successive assays may be performed on the same tumor to monitor its progress, the treatment being adapted appropriately. The proportion of cells having the MDR phenotype may indicate the development of the tumor and its metastatic potential, hence indicating the need for an antimetastasis treatment.

The specific mutations which have so far been identified have arisen from work on the hamster gene in CHO (Chinese Hamster Ovary) cells, but the area concerned is highly conserved and has a high degree of homology with humans (100% amino acid homology) and, as far as can be determined, with other organisms, including E. coll . SEQl in the Sequence Listing is the wild-type CTP synthetase and SEQ2 is the corresponding amino acid sequence. Specific mutations include the following (referring to the CHO gene), individually or in combination:

Thr Val Gly Asp lie Glu Ser Met Pro Phe lie 555

I

ACA GTG GGA GAC ATT GAA AGC ATG CCC TTC ATT

I I I

GAA ACT ATA

Glu Thr lie

Glu Ala Phe Arg Gin (SEQ4)

602

GAG GCC TTC CGC CAG (SEQ3)

I I

AAG CTC

Lys Leu Val Gin Val Val Pro His He Thr (SEQ6) 431 461

GTC CAA GTT GTC CCT CAC ATC ACA (SEQ5)

TTC Phe

Met Phe Cys His Val (SEQ8)

801 815

I I

ATG TTC TGC CAT GTG (SEQ7)

TAT Lys

where the numbering refers to the hamster sequence.

The corresponding human and E. coll sequences are:

* * * * * * ********* ** ** **

124 KERVL EG GE GHDV V LVEIGGTVGDIESLPFLEAIRQMAVE

122 QEWVMRQALIPVDEDGLEPQVCVIELGGTVGDIESMPFIEAFRQFQFK

QEWVMRQALIPVDEDGLEPQVCVIELGGTVGDIESMPFIEAFRQFQFK

E T I K L 2 2 ** _t . * .. IGREHTLFMHLT 175 E. coll (SEQ9) VKRENFCNIHVS 181 human (SEQ10) VKRENFCNIHVS CHO (SEQ10) mutants

**** ****

112 TVQVIPHIT 120 E. coll (SEQ11)

110 TVQWPHIT 118 human (SEQ6)

TVQWPHIT hamster (SEQ6)

K mutant

** .. ** . *

223 KIALFCNV 230 E. coll (SEQ12)

229 KISMFCHV 236 human (SEQ13)

KISMFCHV hamster (SEQ13)

K mutant

Thus, in the human gene examples of specific mutations include those affecting amino acid residues 114, 152, 154, 157, 161, 164 and 235. In the coding sequence, the only mutations which will be of any significance will be those which result in a change to the amino acid sequence of the CTP synthetase ("significant mutations"). Thus, those assays of the invention which are directed to the coding sequence either will be such as have been demonstrated to reveal significant mutations (because the absence or presence of a given restriction site is known to affect the amino acid sequence) or will be such as to allow the nature of the mutation, rather than just its presence, to be identified.

In the case of mutations in an intron, the assay will reveal whether the mutation is one which affects splicing of the said exon or exons.

In both cases, mutations of interest are those which do not reduce the CTP synthetase activity to a fatally low level but which reduce the inhibition by CTP.

It is within the scope of this invention to identify mutant CTP synthetases themselves (rather than mutations in the gene) by sequencing the said regions of the protein. However, at least with 1991 technology, this is not nearly as feasible as the assays described above based on nucleotide sequences. A more feasible approach, but still one which is less desirable than the genetic assays, is to use monoclonal antibodies specific for the mutant or for the wild-type enzyme. Such monoclonal antibodies can be prepared by known techniques, for example those disclosed in H. Zola "Monoclonal Antibodies" (CRC Press 1988) or J.G.R Hurrell: "Monoclonal Hybridoma Antibodies" (CRC Press 1982), which are incorporated herein by reference.

A further aspect of the invention provides a method of treating a patient with a drug, resistance to which is affected by the presence or absence of the MDR phenotype, the method comprising (1) determining whether a tumor cell of the patient has a significant mutation in exons 4, 5 or 7 encoding amino acid residues 110 to 236 of CTP synthetase or in an intron which controls splicing of any of the said exons and (2) devising a dosage regimen of the said drug for the patient according to whether there is such a mutation.

The said dosage regimen may be zero; an alternative drug or radiotherapy may be used instead.

Some aspects and embodiments of the invention will now be described by way of example and with reference to the accompanying drawings in which: Figure 1 shows the wild- type human CTP synthetase cDNA sequence; Figure 2 shows Chinese Hamster Ovary CTP synthetase cDNA; Figure 3 shows a first peptide sequence from Chinese Hamster Ovary CTP synthetase; Figure 4 shows a second peptide from Chinese Hamster Ovary CTP synthetase; and Figure 5 shows a third peptide sequence from Chinese Hamster Ovary CTP synthetase.

EXAMPLE 1: SEQUENCE OF WILD-TYPE GENE

The elucidation of the wild-type sequence is described in Ya auchi et al (1990) EMBO J. 9(7), 2095-99. We reproduce relevant parts of it below for completeness.

Materials and methods Cells

The cytidine auxotrophic strain CR"2, originally isolated from the Toronto Pro" CHO cell line (Kelsall and Meuth, 1988),, was grown in α-minimal essential medium ( -MEM, GIBCO) supplemented with 7.5% fetal calf serum, 10 μM thymidine and 20 μm cytidine. When CTP synthetase proficient transfectants were selected, 5% dialysed fetal calf serum was used for the cytidine deficient medium. Chromosome and DNA mediated gene transfer

Metaphase chromosomes were prepared essentially as described by Lewis et al , (1980). Colcemid was added to a 11 exponentially growing culture of HeLa cells to a final concentration of 50 ng/ml. After 18 h cells were harvested and resuspended in 100 ml of ice cold 75 mM KC1 for 30 min. After recentrifugation, the cells were then resuspended in 2 ml ice cold 15 mM HEPES(Na) pH 7.0, 3mM CaCl₂, and 1% NP-40 for disruption by a Dounce homogenizer (10 strokes, on ice). The homogenate was centrifuged at 200 g to remove debris and the resulting supernatant was then centrifuged at 1500 g for 20 min at 4°C. The pellet obtained was washed once with ice cold homogenization buffer without detergent. This chromosome preparation was then resuspended in 10 ml of gene transfer buffer (25 mM HEPES(Na) pH 7.1, 140 mM NaCl, and 0.75 mM Na₂HP0₄) and 0.5 ml of 2.5 M CaCl₂ was added. The resultant co-precipitate was immediately added to recipient (CR"2) cells (2-3 x 10⁶/10 cm dish) in growth medium supplemented with antibiotics (0.05 mg/ml gentamycin and 1 μg/ml amphotericin B, Sigma), 15 mM HEPES(Na) pH 7.1, and 0.1% polyethylene glycol (MW1500). For DNA mediated gene transfer, DNA was resuspended in the same gene transfer buffer and added to cells as described above.

Recipient cells were exposed to chromosome or DNA-calcium phosphate co-precipitate for 8 - 16 h, treated with 10% DMSO in the last 30 min, and then allowed to recover in nonselective growth medium for 24 h. Treated cells were trypsinized, washed in PBS twice and plated at a density of 2 - 5 x 10⁵/10 cm dish in the absence of cytidine. Colonies appearing after 2 - 4 weeks were picked for further analysis.

Genomic and cDNA cloning

.EcoRI-digested DNA, purified from tertiary transfectants, was fractionated by electrophoresis on agarose gels followed by electroelution of DNA of the desired size as described previously (Nalbantoglu et al . , 1986). This size-fractionated DNA was then ligated with ScoRI-cut DNA from the lambda insertion vector NM1149. These reactions were then packaged and plated on the strain NM5I4 recA hfl giving efficiencies of about 1 x 10⁶ recombinants/μg vector. These libraries were screened with the human Alu repeat BLUR8 (Amersham) and the positives were picked for further screening against a panel of transfectants. _?<3u3A partial digests of HeLa cell DNA were ligated with BaiuΗl-digested DNA purified from the lambda Dash vector (Stratagene) . Libraries of recombinant phage were obtained by plating the packaged DNA on E. coll strain P2392 and screened using a 1.9 kb fragment present in all the transfectants (isolated in the screens described above) .

A human lambda gtll cDNA library, prepared with RNA isolated from human testis (Clontech), was screened with the conserved 3.3 kb unique human DNA fragment common to all our transfectants. Positives obtained were then used for DNA sequence analysis.

RACE PCR

Total RNA prepared from HeLa cells for RACE PCR (rapid amplification of cDNA end polymerase chain reaction: Frohman et al . , 1988) was reverse transcribed using an oligonucleotide primer (nucleotides 784—> 765 or 5'- CAACATGGCAGAACATTGAT-3' , SEQ14) and super RT (Anglian Biotec Ltd. UK) . The product was A tailed by nucleotidyl transferase (BRL) and the opposite DNA strand was synthesized using a (dT)17 adaptor and the Tag polymerase (Perkin Elmer Cetus). The resultant double-stranded DNA was then amplified using the adaptor and a nested primer (nucleotides 758 --> 739 or 5'-TCCTTCACTGATGTGTCAAG-3' , SEQl5) by PCR. The product of this reaction was digested with Sail, and Bglll and cloned into the M13 vectors mpl8 and 19. The clone with the longest insert (700bp) was sequenced.

DNA sequencing

Fragments purified from the lambda gtll isolates were subcloned into the M13 vectors mpl8 or 19. Sequencing was performed by the dideoxy chain termination method using [ -³⁵S]dATP. Polymerization reactions using Sequenase (USB) were primed by either the 17mer universal primer or appropriate internal primers. Sequencing reactions were electrophores_.ed on 5% polyacrylamide denaturing gels.

DNA sequence analysis

Sequences were analysed using Intelligenetics software. Protein databases were screened using the program Prosrch (Coulson et al . , 1987) on an AMT DAP (Distributed Array Processor) 610. Results

Complementation of CTP synthetase deficient mutants by CMGT

The Chinese hamster (CHO) CTP synthetase deficient cytidine auxotroph CR"² was treated with DNA or mitotic chromosomes co-precipitated with calcium phosphate. After recovery and expression, the cells were plated in selective medium (lacking cytidine) to allow formation of colonies acquiring the human CTP synthetase gene. The frequency of such colonies in the first round of transfer was ~10⁶. When purified DNA was used instead of isolated chromosomes, the frequency of colony formation was about 10"⁷. In control cultures (no added DNA or chromosomes) no colonies were observed (frequency <10~⁷).

To determine if the colonies survived as a result of the uptake and expression of the human CTP synthetase gene, DNAs purified from several of the isolates were digested with the restriction endonuclease EcόRl. Digests were fractionated by electrophoresis on agarose gels and transferred to nitrocellulose filters. These filters were then hybridized with labelled human Alu repeat BLUR 8. DNA purified from the colonies recovered after CMGT contained a smear of fragments by hybridizing with the BLUR 8 probe. Colonies derived from DNA transfers, on the other hand, showed no human Alu containing DNA (data not shown) .

To reduce the number of human DNA fragments, chromosomes isolated from one of the transfectants (apparently containing the least number of Alu⁺ fragments, PC91) were used to transfect CR^~2 for a second round of transfection and chromosomes purified from a secondary transfectant (PC91C) were subsequently used in a third round. The frequencies of colonies growing in the absence of cytidine in the latter two rounds were somewhat lower than the first (1 - 2 x 10"⁷), but DNA purified from these survivors clearly contained human Alu sequences. It also appeared that the colonies obtained from these further rounds .of CMGT had progressively fewer human Alu containing fragments.

Isolation of human Alu containing fragments linked to CTP synthetase

As common bands were evident among the tertiary transformants, DNA obtained from one of the isolates (91C12) was digested with EcoRI , size fractionated and ligated to .EcoRI cleaved DNA from the lambda vector NM1149. The recombinant phage libraries derived were screened for human Λiu-bearing fragments: 120 independent isolates were eventually picked and cloned inserts from these recombinants were used to screen a panel of primary, secondary and tertiary transfectants. Presumably any fragments retained by all the strains would be closely linked with the CTP synthetase structural gene. Of the 120 fragments screened in this manner only two were common to the entire panel.

To determine the proximity of these two fragments, one of them (1.9 kb in size) was used to probe genomic libraries prepared from partially digested human DNA ligated into the lambda DASH vector. One isolate having a 16 kb human DNA insert was obtained in the screen. To our surprise, the insert also contained the second (4.9 kb) "linked" fragment only about 6 kb away from the first. Furthermore, a unique 3.3 kb fragment subcloned from the 16 kb insert gave faint hamster specific signals when it was used to probe blots containing digests of human, hamster and transfectant DNA, indicating that these fragments carried sequences conserved in both hamsters and humans.

Given the conserved sequences present on this fragment, RNA prepared from human cells was fractionated and probed on Northern blots with the labelled 3.3 kb fragment. A clear band of about 3 kb hybridized with this probe. In contrast RNA prepared from the CTP synthetase deficient hamster strain showed no band.

Isolation of human cDNAs homologous to the conserved sequences

The conserved 3.3 kb fragment was used to screen human cDNA libraries. Of several screened, only one (prepared from human testis DNA) yielded positives. DNA purified from all the positive clones had common 1.1 and 0.5 kb .EcσRI fragments, but one (clone 14) had a larger, 2.3 kb, insert. All of the clones hybridized with the original 4. kb fragment as well.

Sequence analysis of the cDNA clone and isolation of the complete coding sequence

To determine whether the 2.3 kb cDNA encoded the human CTP synthetase, the fragment was sequenced and the open reading frames found were used to search protein databases. This analysis revealed a long open reading frame which had a strong homology with only one protein, that of the Escherlchla coll CTP synthetase (Weng et al . , 1986). The match was extensive, >40% identical amino acids and >70% similarity when conservative substitutions were considered, and it included the entire open reading frame.

However, it only covered about two thirds of the E. coll gene (the carboxyl portion up to amino acid 146) indicating that the amino end was still to be cloned.

To obtain upstream portions of the CTP synthetase message, the RACE polymerase chain reaction (RACE PCR) technique (Frohman et al . , 1988) was employed. An oligonucleotide complementary to the sequence just downstream of the Bglll site of our cDNA was used as primer for reverse transcription of the CTP synthetase message in RNA prepared from HeLa cells. Transcripts were then A tailed and a second (nested) oligonucleotide complementary to the cDNA together with a tail-specific oligonucleotide (also bearing a Sail site for subsequent cloning) were used to amplify the products generated from reverse transcription. A portion of the products of the amplification were fractioned by electrophoresis on agarose gels and blotted onto nitrocellulose for probing with an appropriated DNA fragment (upstream of the Bglll site) . This revealed that the extension was successful and a further 527 bp upstream was sequenced from Bglll - Sail digested products subcloned into M13. From this determination a potential ATG start codon (consistent with the criteria of Kozak, 1986) was found in a position which would correspond to the start of the E. coll enzyme and an open reading frame of 1773 nucleotides or 591 amino acids was evident. A polyadenylation signal was present 905 bp downstream from the stop codon but there was no poly(A)tail. This sequence has been deposited into the EMBL database, accession number X52142.

The sequence of the CTP synthetase cDNA is included as SEQl in the Sequence Listing.

EXAMPLE 2; ASSAY FOR MUTATED GENE

DNA or RNA is purified by standard techniques from white blood cells of patients with acute leukemia or from solid tumor cells. DNA can be used directly in polymerase chain reaction techniques to detect the mutant gene but RNA is first reverse transcribed using a suitable oligonucleotide primer (for example TCAGCCATCTCTTTCCATTT, SEQ16) 3' of the critical region and reverse transcriptase (super RT, Anglian Biotec Ltd., UK) as recommended by the manufacturer. The region of interest is then specifically amplified using a suitable 5' oligo- nucleotide (for example CACAGATGCAATCCAGGAGT, SEQ17) in combination with nested 3' oligos, for example as follows:

mutation site RNA based assays:

exon 4 5' primer: GTAACTATGAGCGGTTCCTT (SEQl8) 3' primer: AGTTCTCTCTTTTGACCTTG (SEQl9)

product: 309 bp

exon 5 5' primer: CACAGATGCAATCCAGGAGT (SEQ20) 3' primer: GGAACTAGACTGACGTGGAT (SEQ21)

product: 200 bp

exon 7 5' primer: CAAGGTCAAAAGAGAGAACT (SEQ22) 3' primer: CTCTAACAACAAGGGGACTC (SEQ23)

product: 227 bp DNA based assays:

exon 4 5' primer: GCAACAACTTAGACATCTGC (SEQ24) 3' primer: ACTCCTGGATTGCATCTGTG (SEQ25)

product: 236 bp

exon 5 primers as above

product: ~1400 bp

exon 7 5' primer: GTTCAACAGGGGAACAGAAG (SEQ26) 3' primer: ATTTACTCACTTGTTCAGGC (SEQ27)

product: ~500 bp

Tag polymerase (Perkin Elmer Cetus) or one of the new, more accurate thermostable polymerases (Vent TM DNA polymerase, New England Biolabs) is used to amplify the DNA through 30 cycles of denaturation (95°, 15 sec), reannealing (52°, 15 sec), and elongation (72°, 1 min.). The critical mutation can then be detected in the resulting product (200-300 bp in the reactions templated by the cDNA) by any of the following techniques: 1) The double-stranded product is first digested with one or more suitable restriction endonucleases (e.g. Sph I and Stu I) having sites within this region of the wild- tyP*³ gene. Mutations falling in the recognition sequences will eliminate the sites and the reactions will retain a full-length fragment after fractionation on agarose gels.

2) The double-stranded product may be bound to nitrocellulose in dot or slot blots and mutation-specific oligonucleotides (see below) end-labelled with ³²P or fluorescently-labelled nucleotides may be hybridized with the bound DNA followed by a stringent washing protocol (twice for 10 min in 2 x SSC/0.2% SDS followed by 30-60 min in 3M tetramethyl-ammonium chloride, 50 mM Tris HC1 (pH 7.5), 2 mM EDTA, 0.3% SDS at 61°C) . DNA samples having the mutation will be labelled while those lacking the mutations will be negative. The wild-type oligonucleotide may be hybridized to a similar filter as a positive control. This approach is preferred as, at least in our hands, it is quick and sensitive.

Mutation specific oligonucleotide probes: (mutation in bold and underlined)

tttcccagTTTTCCCTCATATC (DNA-based)* (SEQ28) TGTCCAAGTTTTCCCTCATATC (RNA-based) (SEQ29)

GGAACCGTGGAGGACATAGAAA (SEQ30)

GTGGGGGACACAGAAAGCATGC (SEQ31)

C

TAGAAAGCATACCCTTTATTGA** (SEQ32)

T

GCCCTTTATTAAGGCCTTCCGT (SEQ33)

GAGGCCTTCCTTCAGTTCCAAT (SEQ34)

AATGTTCTGCTATGTTGAGCCT (SEQ35)

*This mutation falls near the splice acceptor site; intron sequence shown in lower case letters. **Mutations to C, A or T as shown will cause an amino acid change.

3) The double-stranded PCR product may be produced using end-labelled primers. A formamide-dye mixture is then added to an aliquot of the product before being fractionated on a non-denaturing polyacrylamide gel in tris-borate buffer containing 10% glycerol as described by Orita et al (1989). Under such conditions mutant molecules run aberrantly and are easily detectable when compared to the wild-type control. Alternative systems of electrophoresis have been described (e.g. Kogan &

Gitschier, 1990) and the denatured-reannealed product may also be subjected to chemical treatment that results in cleavage at mismatches (Montandon et al (1989). In the latter case the reaction products are then fractionated by standard polyacrylamide gels and any digested products are apparent as smaller fragments.

DNA can also be used as a starting material in which case the regions amplified from the above priming oligonucleotides are larger as they include introns. The PCR reaction is identical except that the elongation time will be extended to 2 min. The mutation in exon 4 is such that intron sequences must be used as 5' primer. The products of the reaction are analysed for the critical mutations as described above except in the case of the non-denaturing polyacrylamide gel. For this technique it is important that the DNA be small so oligonucleotides amplifying only the exon bearing the mutations (eg AGCTTGGTGGAACCGTG (SEQ36) and CTGGGGAACTAGACTGAC (SEQ37) producing a fragment of 119 bp for exon 5, for example) have to be used. In all cases the critical region may be sequenced (as described previously, Phear et al 1989) using an ³²P end-labelled primer to detect and identify the mutations. An alternative approach is to use mutant specific oligonucleotides, which bear the mutant specific nucleotide at the very 3' end of the primer, in the polymerase chain reaction together with an upstream or downstream oligonucleotide. This technique will give a specific signal only from those samples bearing the mutant gene. DNA or RNA samples processed as described above are incubated with mutant specific oligonucleo¬ tides, an upstream primer and Tag polymerase together with "Perfect Match" (Stratagene) to increase the specificity of the extension from the correctly matched oligonucleotides. In samples having an exon 5 mutant gene, a 105 to 131 bp fragment should be detectable after fractionation of the product on agarose gels when the starting material is RNA or a 1.4 kb fragment if DNA is used. Exon 4 mutations will yield a 145 bp fragment from RNA and a 225 bp fragment from DNA. For exon 7, 219 bp and cθ.5 kbp fragments are generated, respectively. To confirm any positive signals, mutant oligonucleotides from the opposite strand may be used together with a downstream primer yielding 202 to 228 bp fragments in the case of RNA. DNA-based confirmation assays can be devised using intron sequences. The oligonucleotides to be used are: Reaction 1 (mutant nucleotide in bold and underlined)

RNA-based assays

Exon 4

upstream: GTAACTATGAGCGGTTCCTT (SEQ38) mutant: GCATCTGTCATATGAGGGAA (SEQ39)

Exon 5

upstream (wild-type) : CACAGATGCAATCCAGGAGTGG (SEQ40)

mutant: GGGCATGCTTTCTATGTCCT (SEQ41)

AATAAAGGGCATGCTTTCTG (SEQ42)

GGAAGGCCTCAATAAAGGGT** (SEQ43)

TGGAACTGACGGAAGGCCTT (SEQ44)

GACCTTGAATTGGAACTGAA (SEQ45)

(** a terminal G or A will also lead to an altered amino acid sequence and can thus be used in alternative probes) Exon 7

upstream: CAAGGTCAAAAGAGAGAACT (SEQ46)

mutant: ACTTGTTCAGGCTCAACATA (SEQ47)

DNA-based assays

Exon 4

upstream: GCAACAACTTAGACATCTGC ^" (SEQ48)

Exon 5

upstream: same as in RNA-based assay

Exon 7

upstream: GTTCAACAGGGGAACAGAAG (SEQ49)

The same downstream/mutant primers are used as in the DNA-based assay.

Reaction 2 (mutant nucleotide in bold and underlined) Exon 4

downstream: AAGGGCATGCTTTCTATGTC (SEQ50)

mutant: GGGGAAAACTGTCCAAGTTT (SEQ51)

Exon 5

downstream (wild-type): TCCTTCACTGATGTGTCAAG (SEQ52)

mutant: GAGCTTGGTGGAACCGTGGA (SEQ53)

GGTGGAACCGTGGGGGACA C (SEQ54)

TGGGGGACATAGAAAGCATA** (SEQ55)

AGAAAGCATGCCCTTTATTA (SEQ56)

CCCTTTATTGAGGCCTTCCT (SEQ57)

Exon 7

downstream: TCAGCCATCTCTTTCCATTT (SEQ58)

mutant: GAAAATATCAATGTTCTGCT (SEQ59)

** A terminal C or T also causes an amino acid change and can therefore be used as the basis of the primer. REFERENCES (all incorporated by reference)

Coulson, A.F.W., Collins, J.F. and Lyall, A. (1987) Computer J. , 30, 420 - 424.

Frohman, M.A. , Dush, M.K. , and Martin, G.T., (1988) Proc. Natl . Acad. Scl . USA 85, 8998 - 9002.

Kelsall, A., and Meuth, M. (1988) Somat . Cell Mol . Genet. , 14, 149-154.

Kogan, S. and Gitschier, J. (1990) Proc. Natl . Acad. Scl . USA, 87, 2092-2096.

Kozak, M. (1986) Cell , 44, 283 - 292.

Lewis, W.H. , Srinivasan, P.R., Stokoe, N. and Siminovitch, L. (1980), Somat. Cell Genet. 6, 333 - 347.

Montandon, A.J., Green, P.M., Giannelli,F. and Bentley, D.R. (1989) Nucleic Acids Res. 17, 3347-3358.

Nalbantoglu, J., Hartley, D. , Phear, G. , Tear, G. and Meuth, M. (1986) EMBO J. , 5, 1199 - 1204. Orita, M., Suzuki, Y. , Sekiya, T. and Hayashi, K. (1989) Genomlcs, 5, 874-879

Phear, G., Armstrong, W. and Meuth, M. (1989) J. Mol . Biol . 209, 577-582.

Saiki, R.K., et al (1988) Science 239, 487-91.

Weng, M. , Makaroff, CA. and Zalkin, H. (1986) J. Biol . Chem. 261, 5568 - 5574.

Claims

1. An assay method for the MDR phenotype in a tumor cell of a human, the assay comprising determining whether the cell has a mutated CTP synthetase enzyme with reduced inhibition by CTP but adequate CTP synthetase activity for viability of the cell.

2. A genetic assay method comprising the steps of (1) obtaining a polynucleotide corresponding to the CTP synthetase gene, or a relevant part thereof, of a tumor cell of a human and (2) assaying the polynucleotide for a mutation in exons 4, 5 or 7 for a mutation in an intron which controls the splicing of any one of the said exons, the mutation being -indicative of the CTP synthetase enzyme encoded by the said gene having deficient regulation by CTP.

3. A method according to Claim 2 wherein the nucleotide sequence is genomic DNA.

4. An assay method according to Claim 2 wherein the mutation sought is in one of the said exons.

5. An assay method according to Claim 2 wherein the mutation sought is or causes an amino acid mutation in a residue selected from 110-236.

6. An assay method according to Claim 4 wherein the mutation sought is or causes an amino acid mutation in a residue selected from 110-120, 148-166 and 230-236.

7. An assay method according to Claim 6 wherein the mutation sought is or causes an amino acid mutation in a residue selected from 114, 152, 154, 157, 161, 164 or 235.

8. A method according to Claim 7 wherein the mutation is or causes an alteration of the amino acid sequence of the CTP synthetase in one or more of the following ways: 114 Val to Phe; 152 Gly to Glu; 154 He to Thr; 157 Met to He; 161 Glu to Lys; 164 Arg to Leu; or 235 His to Lys.

9. An assay method according to Claim 2 wherein in step (2) a PCR primer specific for a said mutation or for the absence of a said mutation at the position of a said mutation is applied to the polynucleotide to selectively generate amplified DNA according to whether a said mutation is present.

10. A kit for performing an assay method according to Claim 1 or 2 comprising at least one of the following: a restriction enzyme capable of revealing a said mutation; a probe capable of hybridizing to a region of wild-type or mutant CTP synthetase gene associated with said mutation; a PCR primer capable of selectively priming the amplification either of a wild-type or of a mutan - nucleotide sequence associated with a said mutation.