WO2009015491A1

WO2009015491A1 - Umps isoforms as novel predictive markers and therapeutic targets for chemotherapeutics

Info

Publication number: WO2009015491A1
Application number: PCT/CA2008/001427
Authority: WO
Inventors: Marco A. Marra; Isabella T. Tai; Malachi Griffith
Original assignee: Bc Cancer Agency
Priority date: 2007-08-02
Filing date: 2008-08-01
Publication date: 2009-02-05

Abstract

a method of predicting a subject's response to a chemotherapeutic treatment comprising the steps of: obtaining a sample from the subject, detecting a target gene in the sample, and assessing sample to detect at least one isoform of the target gene. if an isoform is detected, then identifying and quantifying isoform. the quantified isoform is compared to a suitable control. a change in expression of the quantified isoform relative to the control indicates the subject's responsiveness to said chemotherapeutic treatment. one embodiment relates to assessing a subject's responsiveness to chemotherapeutic treatments comprising 5-fluorouracil, by detecting, assessing and quantifying selected isoforms of uridine 5'-monophosphate synthase gene. another embodiment relates to assessing a subject's responsiveness to chemotherapeutic treatments comprising 5-fluorouracil, by detecting, assessing and quantifying isoforms a and b, and determining the ratio of expression of isoforms a to b compared to the ratio of expression of a suitable control.

Description

UMPS ISOFORMS AS NOVEL PREDICTIVE MARKERS AND THERAPEUTIC TARGETS FOR CHEMOTHERAPEUTICS

TECHNICAL FIELD

The present invention relates generally to the prediction and modulation of a subject's response to a chemotherapeutic treatment. More particularly, this invention relates to the use of UMPS gene expression as a predictor of the response in a subject to a particular chemotherapeutic treatment.

BACKGROUND ART

Cancer is often treated by a combination of surgery, radiation and chemotherapy. Unfortunately, some patients' cancers are resistant to traditional chemotherapeutics due to differences in their genes. The drug 5-fluorouracil (5-FU) is commonly used in chemotherapy for the treatment of many cancers, particularly in the treatment of breast, head and neck, anal, stomach, bowel, gullet, colon and some skin cancers. Response to this drug among cancers is variable and is thought to be mediated in part by differences in metabolism. In addition, the ability of 5 -FU to destroy tumour cells is influenced by many genes which may vary from patient to patient. A number of genes such as uridine 5 '-monophosphate synthase (UMPS)/orotate phosphoribosyl-transferase (OPRT), dihydropyrimidine dehydrogenase (DPYD), Endothelial cell growth factor 1 (ECGFl), Uridine phosphorylase 1 (UPPl), Thymidylate synthetase (TS) and others are known to be involved in the metabolism of 5-FU. Differential expression and RNA processing of these genes may contribute to 5-FU resistance and poor clinical outcome.

One of the most studied genes related to 5-FU action is dihydropyrimidine dehydrogenase (DPYD), for which a clinical test related to genetic predisposition to 5-FU toxicity now exists (Myriad Genetics, Inc). Although DPYD is believed to account for a large percentage of the catabolism of 5-FU in the liver, additional genes are known for their role in anti-cancer activity of 5-FU. Any drug which is not catabolized in the liver and reaches the tumour cells depends on a variety of other enzymes for conversion to its active form (Maring et al., 2005, Pharmacogenomics J, 5:226-243; Tokunaga et al., 2007, Surgery, have examined the potential role of uridine 5'- monophosphate synthase (UMPS), also known as OPRT, in mediating 5-FU resistance as well as its potential utility as a clinical biomarker of resistance in colorectal cancer patients. However, these studies did not consider the diversity of UMPS transcripts as a whole and particularly the alternative splicing mechanism (AS) which may account for the potential variation when measuring UMPS expression. UMPS has recently been cited as one of the most critical enzymes involved in the metabolism of 5-FU (Sakamoto et al., 2007, Biochem Biophys Res Commun, 363: 216-222) although the relative importance of genes predicted to be involved in 5-FU action is still unknown. Once UMPS is activated as FdUMP, 5-FU is thought to exert an anti-cancer effect via several possible mechanisms: (1) by inhibiting thymidylate synthase (TS or TYMS) which results in depletion of thymidine levels and inhibition of DNA synthesis, (2) by incorporating into DNA or RNA which damages their ability to function properly and ultimately leads to cell death, and (3) by directly inducing programmed cell death.

The chemotherapeutic drug 5-FU has widespread use in treating a variety of human diseases, more particularly cancers, and even more specifically colorectal cancer (CRC). A query completed in July 2008 of CRC samples collected by the Ontario Tumour Bank (OTB request #1125) identified 150 cases, of which 72% received 5-FU as part of their systemic therapy. Furthermore, new combination therapy 5-FU-based drugs such as S-I and Eniluracil have recently been developed. However, despite the common usage of 5-FU and 5-FU based drugs, the response rate to these has been identified as only 20% when used as a single agent. (Imyanitov et. al., 2007 Clin Chim Acta, 379: 1-13; Sakamoto et al., 2007). While these drugs are highly effective for some patients, for others they either have no favourable response or result in adverse reactions. The factors that determine the efficacy and effectiveness of chemotherapeutic treatment with 5-FU are complex. It is likely that any clinical test of 5-FU response that is based on the characterization of only one or two of the many genes involved in its metabolism will fall short of being applicable to the general population.

Currently, 5-FU is administered to a number of patients undergoing chemotherapeutic treatment. In particular, 5-FU is administered to the majority of CRC patients, often as a part of a drug cocktail. Although 5-FU is highly effective in a number of patients, many do not

^{^}lowever, due to the complex metabolic pathway and netabolism of 5-FU, it has been difficult to identify which patients would benefit from this drug's administration. A better understanding of what causes resistance to 5 -FU, and furthermore what causes resistance to drugs in general, is needed to better design new drugs to combat or avoid this resistance. Furthermore, a method to identify a positive response to a particular drug therapy by a patient prior to its administration is needed to better tailor cancer treatments to the needs of individual patients and ultimately lessen the incidence, morbidity and mortality from cancer.

Although previous studies have examined the expression of single genes (or a small number of genes) with suspected relevance to 5-FU action (Kai et al., 2007, Cancer Lett, 258: 45-54; Kidd et al., 2005, Clin Cancer Res, 11: 2612-2619; Kodera et al., 2007, Cancer Lett, 252: 307-313; Matsusaka et al., 2007, Chemotherapy, 53: 36-41 ; Sakamoto et al. 2007; Maring et al., 2005; Tokunaga et al. 2007), none of these studies employed a genome-wide approach to interrogate all potential members of the 5-FU pathway. The use of genomic methods, such as microarrays to identify gene variants associated with diseases such as cancer is an area of rapid development (Griffith et al., 2007, Genes Genomes & Genomics, pp. 201-242). Eukaryotic genomes are generally known to include between about 7,000 to 29,000 genes. The majority of microarray gene expression experiments to date have operated under the assumption that each of the gene loci generates a single mRNA transcript and protein product. However, analysis of expressed sequence tags (EST), full length cDNAs, and other data suggest that at least about 75% of all genes are alternatively transcribed to produce distinct mRNA sequences, otherwise known as splice variants or isoforms, from a single locus by alternative transcript initiation, alternative splicing or alternative poly-adenylation, collectively referred to as alternative expression. The number of unique transcripts in the transcriptome is estimated to be at least 2-3 times the number of transcribed loci. Given the size and apparent complexity of the transcriptome, an increase in the scope and complexity of microarray expression analysis including the ability to measure alternative expression (AE) events is desirable. Furthermore, none of these previous studies employed a technology capable of profiling specific mRNA isoforms or identifying polymorphisms across an entire transcriptome.

In addition, there is a further need to characterize of the occurrence of alternative expression (AE) events, the production of splice variants or isoforms, and the underlying s them. Moreover, there exists a need to better understand the UMPS gene, its transcript diversity and, variations that occur when quantifying UMPS expression.

DISCLOSURE OF THE INVENTION

Exemplary embodiments of the present invention are directed to identifying isoforms of target gene and utilizing these isoforms as novel predictive markers and therapeutic targets for chemotherapeutic resistance. More particularly the present invention is directed toward UMPS isoforms and the amino acid sequence characterizing these isoforms as novel predictive markers and therapeutic targets for chemotherapeutic resistance, particularly 5-FU resistance.

One exemplary embodiment of the present invention is directed to a series of nucleotide coding sequences identified as SEQ ID NO:1, SEQ ID NO:2, SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:9, SEQ ID NO: 10 for expression of isoforms of the UMPS gene comprising an amino acid sequences identified as SEQ ID NO:11, SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO:14, SEQ ID NO:15, SEQ ID NO:16, SEQ ID NO:17, SEQ ID NO:18, SEQ ID NO:19, SEQ ID NO:20 respectively.

According to one aspect of the present invention, there is provided a method of predicting a subject's response to a chemotherapeutic treatment. The method generally comprises; first obtaining a sample from a subject and second, detecting a target gene in the sample. Third, assessing the sample to detect at least one isoform of a target gene. Fourth, where an isoform of the target gene is detected, identifying and quantifying the isoform. Fifth, comparing the quantified isoform with a suitable control. A change in expression of the quantified isoform compared to the control indicates the subject's responsiveness to the chemotherapeutic treatment to be received.

Another exemplary embodiment of the present invention is directed to a method of predicting a subject's response to a chemotherapeutic treatment comprising 5-fluorouracil (5- FU). The method generally comprises; first obtaining a sample from the subject and second, assessing the sample to detect at least one isoform of a target gene uridine 5'-monophosphate cΛmrtiαc,- cττiMP^ς^ TWΛ wViPr^p _an isoform of the target gene is detected, identifying and comparing the quantified isoform with a suitable control. A change in expression of the quantified isoform compared to the control indicates a clinical responsiveness to the chemotherapeutic treatment.

According to another aspect, there is provided a method for predicting a subject's response to a chemotherapeutic treatment comprising 5-fluorouracil (5-FU). The method generally comprises; first obtaining a sample from the subject and second, assessing the sample to detect isoform B of a target gene uridine 5'-monophosphate synthase (UMPS) identified as SEQ ID NO:2. Third, where isoform B of the target gene is detected, quantifying the isoform. The sample indicates a negative clinical responsiveness to said chemotherapeutic treatment with 5-FU when isoform B is detected.

According to yet another aspect, there is provided a method for predicting a subject's response to a chemotherapeutic treatment comprising 5-fluorouracil (5-FU). The method generally comprises; first obtaining a sample from the subject. Second, assessing the sample to detect isoform A and isoform B of a target gene uridine 5 '-monophosphate synthase (UMPS) identified as SEQ ID NO:1 and SEQ ID NO:2 respectively. Third, where isoform A is detected quantifying isoform A. Fourth, where isoform B is detected, quantifying isoform B. Fifth, quantifying the ratio of expression of quantified isoform A and isoform B. Sixth, quantifying the ratio of expression of a suitable control known to be responsive to a treatment with 5-FU and seventh comparing the ratio of expression of the sample to the ratio of expression of the control. A clinical responsiveness to the chemotherapeutic treatment is indicated by a change in the ratio of expression of the sample compared to the ratio of expression of the control.

According to a further aspect, there is provided a method for predicting a subject's response to a chemotherapeutic treatment comprising 5-fluorouracil (5-FU). The method generally comprises; first obtaining a sample from the subject. Second, assessing the sample to detect isoform A and isoform B of a target gene uridine 5 '-monophosphate synthase

(UMPS) identified as SEQ ID NO:1 and SEQ ID NO:2 respectively. Third, where isoform A is detected quantifying isoform A. Fourth, where isoform B is detected, quantifying isoform B. Fifth, quantifying the ratio of expression of quantified isoform A and isoform B. A degree of clinical responsiveness to the chemotherapeutic treatment is indicated by the ratio of expression of the samule. According to another aspect, there is provided a method of sensitizing a tumour to administration of a chemotherapeutic treatment comprising 5 -FU by enhancing the expression of isoform A identified as SEQ ID NO:1.

According to yet another aspect, there is provided a method of sensitizing a tumour to administration of a chemotherapeutic treatment comprising 5 -FU by increasing the ratio of expression of isoform A identified as SEQ ID NO:1 to isoform B identified as SEQ ID NO:2, where the ratio of expression is at least greater than about 0.1.

According to another aspect, there is provided a method of sensitizing a tumour to administration of a chemotherapeutic treatment comprising 5-FU by reducing the expression of isoform B identified as SEQ ID NO:2.

Another exemplary embodiment of the present invention is direct to identifying at least one isoform of a target gene resulting from differential gene expression and alternative splicing expression events. The method generally comprises; first obtaining a sample responsive to a chemotherapeutic compound and a sample resistant to the chemotherapeutic compound. Second, applying a microarray analysis to the responsive cells and resistant cells. Third, detecting and identifying differential gene expression and alternative splicing expression events associated with a resistance to the chemotherapeutic compound. Fourth, detecting and quantifying the differential gene expression and alternative splicing expression events. At least one isoform is identified when at least a single genomic change is identified in the resistant sample that is not present in the responsive sample.

Another embodiment is directed to a kit for detecting expression of an UMPS isoform in a physiological sample collected from a subject. The kit comprising at least some of suitably selected reagents and receptacles for producing a reaction therein indicative of the subject's responsiveness to a chemotherapeutic treatment.

Further aspects of the invention will become apparent from consideration of the ensuing description of preferred embodiments of the invention. A person skilled in the art will realize that other embodiments of the invention are possible and that the details of the invention can be modified in a number of respects, all without departing from the inventive vings, descriptions and examples are to be regarded as ictive. BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the present invention will now be described, by way of example only, with reference to the attached Figures, wherein:

Fig. 1 is a chart illustrating metabolic pathway options for 5-FU ;

Fig 2a is an amino acid sequence SEQ ID NO: 11 comprising an isoform of the

UMPS gene;

Fig. 2b is an amino acid sequence SEQ ID NO: 12 comprising a second isoform of the UMPS gene;

Fig. 3a is a block diagram of UMPS locus on chromosome 3;

Fig .3b is a block diagram of UMPS isoform A and isoform Bs;

Fig. 3c is the expression level of UMPS isoform A and isoform B in responsive and resistant cells;

Fig.4 is the RT-PCR data generated from polyA+RNA isolated from responsive and resistant cells;

DETAILED DESCRIPTION OF THE INVENTION

Definitions:

Alternative expression - the expression of mRNA isoforms with different exon content;

Alternative expression analysis - measurement of the expression of specific mRNA isoforms;

Differential expression - measurement of the difference in expression level between two or more states for example responsive cells and resistant cells, stage 1 and stage 3 cancers;

Differential isoform expression- differential expression of a specific isoform; niffm-^pntial σ^pri^{p pγ}nr^p<:si_on _ differential expression of all isoforms of a gene Isoform - is a version of a gene sequence with small differences when compared to another isoform of the gene;

Splice Variant - an mRNA isoform produced specifically by differences in RNA splicing Single Nucleotide Polymorphism (SNP) - is any change of a single base in a nucleotide sequence on comparison to a control.

Several isoforms of the UMPS gene have been previously disclosed (GenBank Accession Numbers: EU921886, EU921887, EU921888, EU921889, EU921890, EU921891, EU921892, EU921893, EU921894, EU921895). These isoforms are characterized by the nucleotide sequences set forth in SEQ ID NO: 1 - 10. These isoforms are characterized by the corresponding amino acid sequences set forth in SEQ ID NO: 11-20.

Genomic analysis of the UMPS gene sequence and its expression has surprisingly indicated that there exist novel variants of the UMPS gene which are associated with 5-FU resistance. Through the application of genomic tools known in the art, particularly microarray analysis and whole transcriptome shotgun sequencing (WTSS), a number of isoforms and sequence polymorphisms of the UMPS gene have been identified which are associated with human diseases such as cancer. While previous studies have examined the potential roles UMPS may play in mediating 5-FU resistance, none have considered the diversity of the UMPS transcripts produced by alternative expression and to sequence polymorphisms.

More generally, it has been discovered through the application of genomic analysis, specifically a microarray analysis, to samples responsive to treatment with a chemotherapeutic compound and samples resistant to treatment with a chemotherapeutic compound that isoforms of a target gene may be identified.

In addition, it has been discovered that the presence and expression of these isoforms and sequence polymorphisms may be linked to 5-FU resistance. Furthermore, it has been discovered that by detecting, identifying and quantifying the presence and expression of various UMPS isoforms and sequence polymorphisms in a sample, the clinical r^pςnnn^ςivpnpςς nf a ςnhiprt tn a rhemotherapeutic treatment comprising 5-FU may be Furthermore, it has also been shown that detecting, identifying and quantifying the presence and expression of various UMPS isoforms and sequence polymorphisms, and comparing the quantified isoform with a suitable control, enables a prediction of a subject's clinical responsiveness to chemotherapeutic treatment comprising 5-FU by the magnitude of change in the expression of the quantified isoform compared to the control. More specifically, it has been shown that by detecting and quantifying the presence of isoform B in a sample from the subject, the response of a subject to a chemotherapeutic treatment comprising 5-FU may be predicted.

A suitable control for the purposes of this invention may be for example a clinically selected control, a stored dataset of results generated from studies of the presence and expression of various UMPS isoforms and sequence polymorphisms in one or more population(s) of cancer- free subjects, a stored dataset of results generated from studies of the presence and expression of various UMPS isoforms and sequence polymorphisms in one or more population(s) of subjects responsive to a chemotherapeutic treatment, a stored dataset of results generated from studies of the presence and expression of various UMPS isoforms and sequence polymorphisms in samples collected from a single subject over an extended period of time, or combinations thereof. A clinically selected control may be for example a physiological specimen responsive to a chemotherapeutic treatment or a cancer-free physiological specimen. A sample for the purposes of this invention may be for example non-cancerous cells, cancerous cells, tumour cells, tissues, fluids (blood, plasma, urine), DNA, RNA and combinations thereof.

5-FU and the like are exemplified by 5-FU analogs, 5-FU prodrugs, and 5-FU mimetics. The clinical responsiveness of a subject to a chemotherapeutic treatment comprising

5-FU is typically expressed as positive or negative. It was discovered by the present invention that a positive clinical responsiveness may indicate for example, an increase in the expression of isoform A identified in SEQ ID NO:1, a decrease in the expression of isoform B identified in SEQ ID NO:2, and combinations thereof. It may further indicate, an increase in the expression of isoform A compared to a suitable control, a decrease in the expression of ~ control, an increase in the ratio of expression of isoform A o of expression of a suitable control, and combinations thereof . A negative clinical responsiveness may indicate for example, a decrease in the expression of isoform A identified in SEQ ID NO:1, an increase in the expression of isoform B identified in SEQ ID NO:2, and combinations thereof. It may further indicate, a decrease in the expression of isoform A compared to a suitable control, an increase in the expression of isoform B compared to a suitable control, a decrease in the ratio of expression of isoform A to isoform B compared to the ratio of expression of a suitable control, and combinations thereof.

Furthermore, it was discovered that by detecting and quantifying in a sample the expression of isoform A and isoform B of UMPS gene identified in SEQ ID NO:1 and SEQ ID NO:2 respectively, generating ratio of expression of isoform A to isoform B as disclosed in example 4 herein, and comparing the sample ratio of expression ratio to the ratio of expression of a suitable control, the responsiveness of a subject to chemotherapeutic treatment with 5-FU may be predicted. An increase in the change in ratio of expression of the sample compared to the ratio of expression of the control indicates a positive clinical responsiveness. A decrease in the change in ratio of expression of the sample compared to the ratio of expression of the control indicates a negative clinical responsiveness.

Moreover, by detecting and quantifying in a sample the expression of isoform A and isoform B of the UMPS gene identified in SEQ ID NO:1 and SEQ ID NO:2 respectively and generating the ratio of expression for the quantified isoforms; the degree of clinical responsiveness of a subject to chemotherapeutic treatment with 5-FU may be predicted. The degree of clinical responsiveness to chemotherapeutic treatment with 5-FU indicates a high responsiveness where the ratio of expression is greater than about 10; a partial or slight responsiveness where the ratio of expression is greater than about 0.1 and less than about 10; and a high resistance when the ratio of expression is less than about 0.1. More generally, it has been discovered that the response of a subject to a treatment with a chemotherapeutic compound may be predicted by detecting the presence of a target gene, detecting and identifying the presence of at least one isoform of that target gene in a sample obtained from the subject, quantifying the isoform expression, and comparing the quantified isoform with a suitable control. A change in the expression of the isoform compared to the control indicates the subject's responsiveness to the chemotherapeutic tr^pfltm^pnt An inrr^pn^ςpH r^pcnnndyeness may be determined by either an increase or decrease i to the control. A decreased responsiveness may be determined by either an increase or decrease in quantified isoform as compared to the control.

Moreover, it can be inferred that by detecting, identifying and quantifying various isoforms, sequence polymorphisms and variants of target genes of interest, particular compounds which target these isoforms may be identified and by extension, predict a subject's responsiveness to treatment with that compound.

Another aspect of the invention provides a method for sensitizing tumors to the chemotherapeutic treatment with 5-FU. It has been discovered that by restoring or enhancing the expression of the UMPS isoform A, reducing the expression of UMPS isoform B, and combinations thereof, a tumour may be sensitized to administration of chemotherapeutic treatment comprising to 5-FU. Moreover, a tumour may be sensitized to administration of chemotherapeutic treatment comprising to 5-FU by increasing the ratio of expression of UMPS isoform A to isoform B, identified in SEQ ID NO:1 and SEQ ID NO:2 respectively, to at least greater than about 0.1. Enhancement of expression of isoform A may include for example: (a) introducing a gene(s) into a tumor cell(s) which preferentially expresses isoform A using methods known in the art; and (b) modulating the regulation of the splicing of the UMPS gene using methods known in the art. Reducing the levels of expression of isoform B may be done by selectively targeting isoform B using methods known in the art such as siRNA. It has further been discovered by the present invention, that once a subject has been identified as a responder to a chemotherapeutic treatment, i.e. more specifically to a chemotherapeutic treatment comprising 5-FU, the effective amount of chemotherapeutic required for treatment to achieve a favorable outcome may be determined.

Furthermore, through the microarray and genomic analysis, such as WTSS, of the 5- FU responsive and resistant samples, additional genes known to be involved in the 5-FU metabolic pathway, may be identified and classified according to their importance in predicting responsiveness to 5-FU. In addition, this same analysis may also be used to potentially identify additional genes which are involved in 5-FU metabolic pathway.

Microarray analysis

ranscriptional diversity of genomes have focused on the use logies and the application of sequencing platforms known in the art. In the past, cDNA and oligonucleotide microarrays have been used as a means of studying gene expression patterns across tissues. The idea of using splicing microarrays consisting of exon-junction and other probe configurations to detect alternative expression events was first suggested by Douglas Black (Black, 2000). Current microarray design platforms for 'alternative expression analysis' facilitate the design of gene expression arrays for analysis of mRNA isoforms generated from a single locus by the use of alternative transcription initiation, splicing and polyadenylation sites.

For the purposes of the examples detailed below, the ALEXA Platform (www.ALEXAPlatforni.org) microarray was used however, other systems such as GeneChip® developed by Affymetrix would also be suitable.

The microarray platform takes as input, a complete set of gene models for a particular species and generates a complete set of probes corresponding to at least one of the exons, introns, exon-j unctions and exon-boundaries of these genes. The target gene is interrogated by these probe sets. The results illustrate alternative expression (AE) events such as exon skipping, alternative exon boundary usage and, intron retention identified in the target gene on comparison to the known gene sequence.

Using microarray analysis, specifically the ALEXA platform (Griffith et. al., 2008, Nat Methods, 5: 118), a series of custom splicing microarrays were designed that are capable of profiling the expression of genes at the level of individual isoforms and sequence polymorphisms. The microarray outputs expression level estimates for entire genes as well as for individual exons, exon-j unctions and exon-boundaries and therefore can be used to profile AE events such as alternatively expressed transcripts generated from a single locus by alternative transcript initiation, splicing and poly-adenylation.

In addition to using custom splicing microarrays to profile the expression level of transcripts in RNA samples isolated from 5-FU responsive or resistant cell lines and patient samples, a parallel sequencing technology has been applied in order to sequence these transcripts using WTSS. Sequencing the whole transcriptome using the WTSS technology (Morin et al., 2008, Biotechniques, 45: 1-10) involves the generation of random fragments from RNA followed by direct sequencing of clonal derivatives of these fragments on an zer. Large groupings of exons present in sufficient quantities can be 're-sequenced' to reveal any exonic single base differences, including single nucleotide polymorphisms (SNP) and novel mutations.

Identification of Genes in the 5-FU metabolic pathway

In order to identify new genes involved in the 5-FU metabolic pathway illustrated in Figure 1 , a database comprising information related to isolated total RNA, polyA+ RNA and genomic DNA from 5-FU responsive cells lines, 5-FU resistant cell lines as well as clinical CRC samples from subjects that received 5-FU as part of their chemotherapeutic treatment is compiled. The database is divided into responders and non-responders to chemotherapeutic treatment with 5-FU. Responders are prioritized according to length of disease- free survival period and non-responders. Through application of microarray analysis and WTSS to this database, a list of candidate genes identified as differentially expressed and/or mutated in chemotherapeutic treatment with 5-FU may be generated. Better understanding of the major genes involved in 5-FU metabolism may prove useful in developing clinical tests to predict 5-FU response and assist in developing therapeutic regimens for the treatment of CRC and other cancers. This approach is generally applicable in the identification of genomic related events that differentiate responders and non-responders for many chemotherapeutic treatments and cancer types. Non-responders are generally classified as those subjects with adverse response to drug or whose tumour rapidly returns, progresses, metastasizes, or leads to death.

Combination Therapy with 5-FU

For patients receiving chemotherapeutic treatment the regimen for that treatment often includes the use of drug cocktails where combinations of two or more drugs are administered to a patient as opposed to using a single chemotherapeutic agent. Combination therapies further include the use 5-FU alternatives for example, 5-FU mimetics, a 5-FU analogs, 5-FU pro-drugs or combinations thereof.

Generally, the treatment regimen may include slowing the catabolism of 5-FU in the liver such that the concentration of drug that actually makes it to the tumor is higher, reducing side effects, and facilitating the use of a lower effective dose. Since UMPS fimrtinnc in th^p tnmrmr r^piic increasing the concentration of drug at the tumour site is of Combination therapy for a subject receiving a chemo therapeutic treatment comprising 5-FU may include the use of 5-FU mimetics, 5-FU analogs, 5-FU prodrugs as an alternative to 5-FU. Typically, combination therapies that are intended to improve the efficacy of 5-FU or the like.

The following are examples of drugs that may be used as an alternative 5-FU.

5-FU Analogs

Fluorodeoxyuridine (FUDR): Is an analog of 5-FU or 2'-Deoxy-5-fluorouridine.

5-FU Prodrugs

Tegafur (FT): Is an oral flouropyrimidine.

Capecitabine (Xeloda): Is an orally administered pro-drug of 5-FU.

The following are examples of drugs that may be used in combination with 5-FU.

5-FU Combinations

Tegafur-uracil (UFT or Uftoral® or UFUR): The uracil component acts as a competitive inhibitor of dihydropyrimidine dehydrogenase (DPD) enzyme and slows the catabolism of 5-FU. The uracil component generally reduces the side effects of Tegafur without reducing its effectiveness.

S-I : Is an oral fluorouracil anti -tumor drug that combines three pharmacological agents; tegafur (FT), 5-chloro-2,4-dihydroxypyridine (CDHP), and potassium oxonate (Oxo). The 5-FU component acts as a neo-plastic agent, the CDHP component acts to improve efficacy and reduce metabolism of the 5-FU in the liver, and the Oxo component acts to inhibits UMPS/OPRT and reduce toxicity by reducing 5-FU activation preferentially in the small intestine while still allowing activation in the bone marrow and tumour regions.

Eniluracil (Ethynyluracil)-5-FU: Eniluracil is an inactivator of DPD. The Eniluracil component reduces the probability of an adverse reaction to 5-FU, improves the effectiveness of 5-FU, and lowers the effective dose of 5-FU given.

cule which 5-FU was originally designed to mimic. Other clinically used cancer chemotherapeutics which may be used in combination with 5-FU are for example: Oxaliplatin, Leucovorin, Bleomycin, Cisplatin, Cyclophosphamide, Doxorubicin, Fludarabine, Furosemide, Gemcitabine, Irinotecan, Etoposide, analogs of cytosine instead of uridine.

EXAMPLES

CELL LINES, SEQUENCING, EXPRESSION AND ISOLATION CONDITIONS

Cell lines: All cell lines (MIPlOl, HCTl 16 and RKO; responsive and resistant) were maintained in Dulbecco's Modified Eagle Medium (DMEM) media supplemented with 1% penicillin-streptomycin, 1% kanamycin (Invitrogen) and 10% newborn calf serum at 37°C and 5% CO2 (T ai et al. 2005). For resistant cell lines, media were also supplemented as follows: MIPlOl cells resistant to 5-FU (MIP101/5FU), 50OuM 5-FU; HCTl 16 cells resistant to 5-FU (HCTl 16/5FU), lOuM 5-FU; RKO cells resistant to 5-FU (RKO/5FU), 25uM 5-FU.

RNA Isolation: Total RNA was isolated from cells cultured to about 75% confluence using RNeasy Columns (Qiagen). RNA was DNAseI treated using an RNAse free DNAseI kit from Invitrogen. RNA was quantified and tested for degradation using an Agilent 2100 Bioanalyzer. PoIyA+ RNA was purified from total RNA using an oligoTex kit (Qiagen).

PCR and RT-PCR validation of UMPS isoform expression: Single stranded cDNA was generated from polyA+ RNA isolated from each cell line using Superscript III reverse transcriptase and random hexamer primer (Invitrogen). PCR primers were designed to flank exon 2, forward primer set forth in SEQ ID NO:29 and reverse primer set forth in SEQ ID NO:30. PCR was performed with Invitrogen's Platinum Pfx enzyme.

Cloning & sequence validation of UMPS mRNA isoforms: Clones representing the full UMPS open reading frame and most of the untranslated regions (UTR) were generated by TOPO® TA cloning system (Invitrogen) using primers designed against the UMPS reference sequence (Genbank #NM_000373 ) set forth in SEQ ID 79 and SEQ ID NO:80. PCR was performed with Invitrogen's Platinum Taq, High Fidelity enzyme. Clones were screened for correct insert size and forward orientation relative to the M13F site of the me digestion with EcoRI and Notl/Xhol (double digest) respectively. 96 clones were full-length sequenced by Sanger sequencing with an ABI 3730 device using M13F and M13R primers set forth in SEQ ID NO: 81 and SEQ ID NO:82 as well as custom primers set forth in SEQ ID. NO.84 and SEQ ID NO:91. Clone sequences were assembled by Phred/Phrap and manually finished using Consed as previously described (Baross et. al., 2004, Genome Res, 14: 2083-2092). Vector sequence was masked except for a short linker sequence set forth in SEQ ID NO:93 at each end of each clone.

PCR & sequence validation of UMPS genomic DNA: The genomic region of the UMPS locus was sequenced by generating 22 amplicons overlapping the region from lkb upstream of UMPS exon 1 to the end of exon 3. Each primer contained either an M13F (SEQ ID NO:81) or M13R (SEQ ID NO:82) linker which were used for direct sequencing of PCR products. Genomic DNA was isolated from cells grown to about 75% confluence using a Gentra PureGene kit (Qiagen). PCR was performed with Platinum Taq, High Fidelity (Invitrogen) enzyme and each amplicon was column or gel purified and Sanger sequenced with an ABI 3730 device using M13F and M13R primers. Reaction conditions were optimized for each primer pair. Table 1 contains the primer sequences, reaction conditions and additional details for all primer pairs used. Sequencing of the target amplicons was carried out by the BC Cancer Agency Genome Sciences Centre production sequencing group using previously published reaction chemistries (Pugh et al., 2007, BMC Cancer, 7:128).

*PCR conditions are as specified in the Invitrogen Platinum Taq HiFi manual except as indicated

Sequence analysis for identification of mutations was conducted with Mutation Surveyor™ (SoftGenetics).

Example 1 - Use of ALEXA Platform microarray analysis to identify significant DE events and expressed isoforms related to various genes for 5-FU responsive/resistant cell lines

The following example details the use of a prototype array to profile the expression of 5-FU responsive colorectal cancer cell line MIPlOl and its drug resistant derivative MIP101/5FU. On comparison of the responsive and resistant cells lines, a number of differentially expressed (DE) mRNA isoforms associated with acquired 5-FU resistance were discovered.

Probe extraction and filtering for array designs

Gene models, their corresponding genomic sequence and related information were downloaded and imported into the ALEXA database for each species of interest. Probes were extracted from a genomic sequence of each gene model. Exon and intron probes were extracted at 5 base pair (bp) intervals. Exon-exon junction probes were extracted to represent every possible valid combination of two exons for each gene. Exon-intron junction probes were extracted to span every unique exon boundary in the gene. Exon-exon and exon-intron probes were extracted such that the sequence was centered on the junction. Finally, 1.5 million random probe sequences (negative controls) were generated to uniformly represent the range of probes.

Probe sequences were scored according to a number of variables such as thermodynamic properties and specificity including melting temperature (Tm). Sequence specificity was determined by comparison of probe sequences against databases containing all sequence transcripts, mRNAs, ESTs, all probe sequences, and the entire genome. The filtered to ensure probes met a minimum set of φtimized to ensure the ideal selection parameters. Creation of a validation array design

The genes of interest for the array design were selected by identifying all genes with at least about 2-fold or greater DE of one or more of their exons according to exon microarrays. Approximately 100 genes defined as housekeeping controls on the exon array were also selected. Unlike most genes on the array, these were targeted by intron as well as exon probes. Approximately, about an additional 400 genes were selected for their potential relevance to cancer biology or drug resistance. These included genes of the ABC drug transport family, genes with known cancer-related isoforms previously identified in technical literature, genes from the cancer gene census, and genes associated with specific the gene "ontology" terms.

The ALEXA validation design was generated by selecting probes corresponding to genes as described above. The final prototype design consisted of 385,000 probes of about 26-46 bp in length corresponding to about 2,511 genes. Each exon, intron or junction was represented by about 2 to 4 probes. Exon-exon junction probes were excluded if they represented an event where more than 3 exons would be skipped. The array was composed of probes representing about 31,000 exons, about 93,000 exon-exon junctions, about 50,000 exon-intron junctions, about 500 introns and about 4,500 random sequences. Random probes were used to estimate false positives and for background correction. This array design was submitted to third party vendor NimbleGen Systems Inc. for synthesis. Tissue culture

The colorectal cancer cell line, MIPlOl and a previously generated 5-FU resistant derivative, MIP101/5FU were maintained as described above.

RNA isolation, labeling and hybridizations

Total RNA was extracted from cultures grown to 75% confluence using Trizol reagent (Invitrogen). Total RNA was DNAseI treated with an RNAse free DNase set followed by cleanup on RNeasy columns (Qiagen Inc. Mississauga, ON). RNA was quantified and tested for degradation using an Agilent 2100 Bioanalyzer. To prepare samples for hybridizations to the validation ALEXA arrays, polyA+ RNA was isolated from total RNA with a μMACS mRNA isolation kit (Miltenyi Biotec, Gladbach, Germany) followed by double stranded cDNA synthesis with a 'Superscript Choice System' for cDNA Synthesis en). 5 μg of each cDNA sample was shipped to NimbleGen. Labeling, hybridization and scanning was conducted by NimbleGen using their 'ChIP-chip' protocol (optimized for 50-mers) and raw data was returned to us.

Data processing

Raw probe values for ALEXA arrays were provided directly by NimbleGen. The ALEXA design contained about 4,300 randomly generated probe sequences. These probes were selected to uniformly represent all experimental probes. For each array hybridization, a loess model was fit to a plot of probe intensity versus Tm for all random probes. A Tm- specific estimate of background hybridization was then estimated for every probe on the array by interpolating from the loess model fit. The data was then normalized across the arrays by quantiles normalization. Differential expression (DE) or 'fold-change' values for probesets (each corresponding to an exon, intron or junction) were calculated by taking the mean of individual probe intensities for each probeset, taking the mean of the probeset means across biological triplicates, transforming to a Iog2 scale and calculating the Iog2 difference between 5-FU responsive and resistant cells (responsive minus resistant). DE values for entire genes were calculated in a similar fashion by combining the probe intensities for all exons of each gene. Both exon and canonical junction probes were considered when estimating expression of the entire gene.

Visualization

To facilitate manual examination of expression and DE, values for all probes were divided into 20 quantile bins and plotted as individual custom UCSC genome browser tracks (data not shown). Expression or differential expression values were calculated on a Iog2 scale and shaded according to their magnitude. The genomic position of predicted ORFs and protein features were also calculated for reference. Custom tracks representing replicate and mean data were generated for every gene profiled (data not shown). Identification of significant DE events

Significant DE events were identified by comparing the population of individual probes of each probeset (exon, intron or junction) between biological triplicates of 5-FU responsive and resistant cells (typically 3 probes x 3 replicates or 9 values for each condition). P-values were calculated by a Wilcoxen rank sum test. The probability of an event being DE between conditions was expected to depend on the probe type. For example, are only predicted sequences which may or may not occur in the transcriptome and therefore may not be expressed in either condition. For this reason, DE events were summarized separately for each probe type. At least 40 differential gene expression events associated with 5-FU resistance were identified. Probes without evidence of expression in one or both conditions were filtered before statistical testing. Specifically, a probeset was required to have a mean Iog2 expression value greater than the 97.5th percentile of all negative control probes (about 8) in either 5-FU responsive or resistant cells. Multiple testing problem (MTP) correction was applied to the filtered list. Events with a fold-change of two or greater and a MTP corrected p-value < 0.05 were considered significant as described below. Gene ontology analysis

Tests for statistical enrichment of particular gene ontology terms associated with the list of genes identified as DE were conducted with GOstat as known in the art (Beissbarth et. al, 2004, Bioinformatics, 20(9): 1464-1465).

Identification of putative alternative transcription events To identify events potentially indicating differential expression of specific isoforms, probesets were filtered to eliminate those with low expression. A splicing index value was then calculated to estimate the differential expression of each probeset after normalization to account for DE of the entire gene. The splicing index was calculated as:

SIi = lo_g2((eSi/gSj)/(eRi/gRj)) (1) for the i-th exon (e) of the j-th gene (g) in 5-FU responsive (S) and resistant (R) samples. A Wilcoxen test was then applied to test for differences in the SI values for a particular probeset between responsive and resistant cells. Probesets with a significant Wilcoxen p-value (< 0.05), an SI value > 1 and an absolute difference between their SI and gene level DE > 1.56 (abs(SI - DE)) were selected as putative differential alternative expression events. The resulting list was ranked according to abs(SI-DE) to identify possible cases of reciprocally expressed isoforms. The top candidate isoforms shown in Table 2 were selected from this list by manual examination of data displayed in custom UCSC tracks corresponding to the genomic loci implicated. Each event was classified as 'alternative TSS/polyA', 'alternative exon boundary', 'intron retention', 'exon skipping' and 'complex' (a combination of the other classes). EST and mRNA support was determined by BLAST of

ESTs and mRNAs that map within the target locus of the probe sequence according to UCSC. Hits of 95% of the length of the probe or greater were considered to be a supporting match. EST and mRNA support was also visually confirmed using custom tracks of expression data in the UCSC browser. Table 2 below illustrates a series of genes involved in the 5-FU pathway and a respective pair of isoforms, generically labeled as isoform 1 and isoform 2, of each gene that were identified . The fold-change values of putative alternate isoforms of a variety of genes involved in the 5-FU pathway were determined by manually grouping all probes which correspond to each putative isoform. Positive fold change values indicate over-expression in 5-FU responsive cells. Negative fold change values indicate over-expression in 5-FU resistant cells. Where the expression of one or more isoforms has changed between responsive and resistant cells, there is an indication that this isoform may be involved in resistance to 5-FU.

Table 2 - Differential isoform expression events associated with 5-FU resistance

Statistical analysis

AU statistical tests were two-tailed. When comparing differential expression values, Pearson correlation coefficients were reported. Comparisons of population means were conducted by a Student's t-test only when the assumption of normality could be satisfied by visual examination of Q-Q plots and the Anderson-Darling test for normality otherwise a non-parametric Wilcoxen rank sum test was used. When identifying significant differential expression events, the sample size for each event often consisted of about 6 to 12 observations for each condition (responsive and resistant). Reliable demonstration of normality was not possible with samples of this size and therefore the Wilcoxen test was always used for these comparisons. MTP correction was accomplished by Benjamini and Hochberg's step-up false discovery rate controlling procedure using the 'multtest' package of R. An MTP corrected p-value < 0.05 was considered significant. To test for enrichment of particular types of annotations (protein features, etc.) in the group of events identified as significantly DE compared to the total population of events, a Chi-Squared test was used if the assumption of normality could be verified, otherwise a Fisher's Exact test was used.

Example 2 - Identification of AE events associated with 5-FU resistance

To identify AE events associated with 5-FU a panel of CRC cell lines, specifically

HCTl 16 cell lines and their resistant derivatives, were studied. Cell lines were received from Dr. Isabella Tai at the BC Cancer Agency (Tai et al. 2005). Using the ALEXA system as described in Example 1, the arrays were generated and output expression data analyzed. Figure 3 A illustrates the UMPS gene locus on chromosome 3 having 6 exons of specified length. Figure 3B illustrates the positions of ALEXA probesets consisting of 2-4 oligonucleotide probes specific to each of two UMPS iso forms. The probes were labeled according to the exons or junctions that they profile (E1-E3 detects the connection of exon 1 to exon 3). The black arrows indicate the predicted Open Reading Frame (ORP) of each isoform and the position of protein domains is indicated beneath each isoform. The array data predicted that one isoform, isoform A set forth in SEQ ID NO:1, of the 5-FU metabolism gene UMPS was down-regulated in resistant cells and a second short isoform, isoform B set forth in SEQ ID NO:2, missing exon 2 was up-regulated. Figure 3C illustrates the Iog2 expression values for the probes specific to each isoform from triplicate samples of each cell line. The median Iog2 expression value of all exons are shown by the dotted line and all negative controls are shown by the solid line. Isoform A was about 5 -fold over-expressed in 5-FU responsive cells relative to resistant cells and isoform B was about 6- fold over-expressed in 5-FU resistant cells relative to responsive cells as shown in Figure 3C. These results were verified by RT-PCR using a mixture of plasmid DNA clones which represented UMPS isoform A and isoform B set forth in SEQ ID NO:1 and SEQ ID NO:2. These plasmids were mixed in molar ratios of 1 :0, 100:1, 25:1, 10:1, 5:1, 2:1, 1 :1, 1:2, 1 :5, 1 : 10, 1 :25 and 0: 1 and used as templates for PCR reactions. The same PCR primers used in the PCR reactions, shown above in Table 1, were used to amplify UMPS isoforms from single strand-cDNA generated from polyA+ RNA extracted from 5-FU responsive and resistant cell lines. Products from the RT-PCR analysis were quantified using an Agilent 2100 bioanalyzer and results are illustrated in Figure 4. Both the MIPlOl and HCTl 16 cell lines showed an increase in the presence of UMPS isoform B set forth in SEQ ID NO:2 in the 5-FU resistant derivative line as compared to the responsive line. Quantified values for the increase and decrease in this expression are shown in Table 3. A negative value indicated a decrease in the quantity of an isoform and a positive value indicates an increase in the same.

TABLE 3 - Quantified RT-PCR product in responsive vs. resistant cell lines

f indicates a significant fold change difference between responsive and resistant cells (p < 0 05 by two-tailed Student's t-test).

Example 3 - Cloning and Sequencing of UMPS isoforms

Approximately 300 UMPS cDNA clones were generated by RT-PCR of polyA+ RNA using an oligo-dT primer set forth in SEQ ID NO:77 for ss-cDNA synthesis followed by amplification with a primer designed to flank the UMPS ORF. PCR products were gel purified and cloned by Topo-TA cloning. Clones were verified for size and orientation by restriction enzyme digestion and 96 clones were selected for full-length sequencing. Clone sequences were assembled and analyzed by BLAT to the human genome (NCBI reference: hgl8). Of the clones mapped to the UMPS locus, 95 of 96 clones corresponded to known or novel isoforms. One clone appeared to represent a contaminating genomic sequence from Ralstonia pickettii. Clones representing several UMPS isoforms were rescued, particularly isoform A set forth in SEQ ID NO: 1 as well as isoform B set forth in SEQ ID NO:2. Figure 2 A and 2B illustrate the amino acid sequences corresponding to each of isoform A and isoform B. Of those isoforms identified, at least 10 distinct spliced UMPS isoforms were discovered, these are detailed in Table 4. Representatives of each of these isoforms were examined to determine if they were supported by existing mRNA or EST sequences.

TABLE 4 - Alternative isoforms

The observed exon-exon boundaries in the sequenced clones were used to predict the effect of the ORF in each distinct isoform. The genome coordinates or exon boundaries from BLAT alignments of each clone to the human genome were used to extract a hypothetical full length sequence for each isoform using the EnsEMBL API. These sequences were then used to predict open reading frames for each clone and each sequence was examined to see if it was predicted to result in nonsense mediated decay (NMD). A sequence was tagged as a NMD using a rule adopted by the VEGA transcript (Wilming et al., 2008, Nucleic Acids Res., 36: 753-760) classification project which defines NMD as being probably if the predicted coding sequence terminated more than 50 bp upstream of a splice site.

^~ iression ratios to predict 5-FU resistance Twenty- two PCR amplicons generated from the genomic region of UMPS starting from the region 1 -kilobyte upstream of exon 1 and ending at the terminus of exon 3 were generated. The DNA for these PCR amplicons was isolated from six 5 -FU responsive/resistant cell lines of MIPlOl, RKO and HCTl 16 and one genomic DNA control line from ClonTech (Mountain View, CA. CAT#636401).). Sequencing was performed directly from the PCR amplicons using M13F and M13R linkers (SEQ ID NO:81 & SEQ ID NO:82 respectively) included in the primer sequences which are identified in Table 1. These amplicons captured a portion of the sequence including all four splice sites involved in the splicing of exon2. The region covered by these amplicons more generally included the UMPS exon2 and a portion of the flanking intron on either side.

Sequence discrepancies relative to a reference human genome (NCBI human genome reference sequence, version 36) within these amplicon sequences and control sequence were identified by using the Mutation Surveyor™ tool by Software Genetics ® and classified as known SNPs or novel mutations by comparison to dbSNP, a database of known human polymorphisms (Sherry et. al., 2001, Nucleic Acids Res, 29(1): 308-311). In two of the amplicons that overlap the acceptor splice site of exon 2, namely the 5' end of exon2, the Mutation Surveyor™ identified a splice site mutation in both the forward and reverse reads for the MIP101/5FU cell line but not the MIPlOl cell line. The mutation was thought to be acquired by MIP101/5FU in the process of becoming resistant, and was classified as a heterozygous mutation.

In the MIPlOl cells it was found that both alleles of this locus produced pre-mRNAs with a canonical splice site at exon2. The splicing machinery recognized this splice site and included exon2, which generally resulted in expression of isoform A set forth in SEQ ID NO:1. In the MIP101/5FU cell lines, the first allele of this locus produced a pre-mRNA having the canonical splice site however, the second allele produced a pre-mRNA with a mutation at the acceptor splice site for exon2. Generally for all pre-mRNAs generated from this second allele, the splicing machinery failed to recognize the mutated exon2 acceptor site and therefore skipped exon2, resulting in expression of the isoform B set forth in SEQ ID NO:2. The mutation in the second allele was classified as either a loss-of-function mutation which generally confers resistance to 5 -FU. This resistance is generally classified as either a) a gene dosage effect because the resistant cells have one functional copy of the gene instead ive mutation where isoform B actively antagonizes isoform A. The two alleles of the MEP 101/5FU cell line which shows resistance to chemotherapeutic treatment with 5 -FU, when taken together, create about approximately an equal mixture of each of the two isoforms resulting in an isoform expression ratio of 1 : 1. Using the above technique, the ratio of expression calculated for a sample taken from a subject's tumor may be compared to a control ratio of expression and the responsiveness of the subject to chemotherapeutic treatment with 5 -FU predicted. The control may be for example derived from a sample known to be responsive to 5-FU , a sample known to be resistant to 5-FU, or a cancer- free sample. An increase or a decrease in the ratio of expression of the tumour sample compared to the control, depending on the type of control selected, indicates the potential responsiveness to chemotherapeutic treatment with 5-FU.

Example 5- Functional assay to test for conferred 5-FU resistance

Cell viability assay: Cells were seeded generally in 24-well plates to achieve about 60% confluence 24 h after seeding. The cells were transiently transfected with the UMPS isoform B set forth in SEQ ID NO:2 for 36-48 h before incubation with 0-2500 uM 5-FU or 100 uM irinotecan (CPT-11) for 36-72 h. Cell viability was assessed by MTS assay (Promega) at 490 nm. (assay detailed in Tai et. Al., 2007)

Isoform B was cloned and sequenced from resistant cell line MEP101/5-FU as outlined above. Plasmid pcDNA3.1 (Invitrogen) was used to create an expression vector for isoform

B. The expression vector was then transfected into cells from the MIPlOl cell line which are responsive to 5-FU. Using a cell viability assay, the cells response to 5-FU is then quantified.

Example 6 - Whole Transcriptome shotgun sequence to identify SNP

Whole transcriptome shotgun sequencing (WTSS) generates random fragments from RNA, followed by direct sequencing of clonal derivatives of these fragments on an Illumina/Solexa IG genetic analyzer. Exons present in sufficient quantities can be 're- sequenced' to reveal any exonic single base differences, including single nucleotide polymorphisms (SNP) and novel mutations.

The publicly available human reference genome (UCSC version hgl8) sequence tie 3 billion positions in the human genome spread across X and Y chromosomes 1 through 22. However, a real genome in a cell is diploid, meaning that every chromosome is present in two copies. At many positions in the genome some people's chromosomal sequences match the reference genome while others exhibit some variation on either one or both of their chromosome copies. Each these variations, denoted as mutations, are recorded in dbSNP.

The UMPS genomic locus was sequenced as described in Example 4 and was then compared against the UMPS gene extracted from the human genome reference sequence (UCSC hgl8). Using WTSS as outlined in manuscript by Morin et al. (Morin et. al., 2008) data from a whole transcriptome shotgun sequencing of the cell line, HeLa S3 was conducted for a single genomic locus of UMPS. Reads of 36bp in length were mapped to the human genome (hgl8) using Eland. The coverage of each base position was calculated and displayed as a custom UCSC 'wiggle' track. For this locus, the base coverage ranges from 1 to 55. The average base coverage of this gene was 7.2 reads per base, the majority of exonic sequence is covered by at least one read. Using this data in combination with the amplicon sequencing data described above, at least four novel SNPs were identified as detailed in Table 5 below, and one novel coding SNP which affects the open reading frame of the UMPS gene.

The reference base denoted in Table 5 indicated the nucleotides found on each copy of chromosome 3 on which the UMPS gene resides in the reference genome. These nucleotides are denoted as for example C/C to indicate the presence of cytosine at a certain position on both copies of the chromosome. Each SNP identified is linked to a particular consequence to the gene. These consequences may for example be intronic (affecting the intron), splice site (affecting the intron or exon splice site), coding (coding of an amino acid).

Table 5 - SNPs observed by genomic sequencing of the UMPS locus

Reference Location of Genomic

SNP Type Genotype Consequence Base Consequence Position

MIPlOl MIP/5FU RKO RKO/5FU HCT116 HCT/5FU

C/C C/C C/T C/T C/C C/C Between

W Substitution C/C Intronic 125936387 exon 1 and 2

A/A A/A A/G A/G A/A A/A

Between

X Substitution A/A Intronic 125936406 exon 1 and 2

G/G G/T G/G G/G G/G G/G

5' Splice Site

Y Substitution G/G Splice Site 125936629 of Exon 2

C/C C/C C/C C/T C/C C/C Exonic Within Exon

Z Substitution C/C 12593671 1 Coding 2

DM VAN/274450-00002/6891735 4

A SNP which is identified as affecting a splice site or the exonic coding may affect the expression of an isoform of the UMPS gene and affect the responsiveness to chemotherapeutic treatment with 5-FU. Of particular relevance are SNP Y and SNP Z, each of which contain a single base substitution within the UMPS gene on chromosome 3. SNP Y demonstrates a mutation in coding in the 5' end of the splice site for exon 2 and may affect the ability of the splice machinery to recognize the exon 2 acceptor site. SNP Z demonstrates a mutation within exon 2 causing a change in the coding of an amino acid from proline to serine which may prevent the UMPS gene from functioning properly and thus confer resistance.

Example 7 - Method of identifying genes involved in 5-FU resistance

Many genes have been identified in the human genome but to date have unknown functions. Using a microarray analysis such as ALEXA and/or WTSS and applying this analysis to these genes of unknown function, the presence of the related gene sequences in a population of subject's is determined. Once the presence of at least one of these genes is detected, using the methods disclosed above in Example 1 to identify significant DE events in combination with WTSS and applying these technologies to a comparison of a responsive and resistant samples genes involved in 5-FU metabolism may be identified. More specifically, differences in sequencing and expression between responsive and resistant samples i.e. MIPlOl (5-FU responsive) and MIP/5FU (5-FU resistant) of various genes indicate whether that gene is involved in mediating a 5-FU response. Table 2 shown above illustrates a number of genes exhibiting differences in expression on comparison of responsive and resistant cells.

Example 8 - Method of determining an effective dosage for treatment with 5-FU

Where a subject has been identified as responsive to a chemotherapeutic treatment, more specifically a treatment comprising 5-FU, the effective amount of chemotherapeutic for that treatment to achieve a favorable outcome may be determined.

Using the methods disclosed above to detect and identify isoforms of the UMPS gene, and specifically identifying isoforms related to resistance to 5-FU, the level of UMPS activity in a sample may be predicted. In general, a low level of UMPS activity is related to resistance to 5 -FU.

On analysis of a sample obtained from a subject through application of the above- noted techniques, the expression of isoforms of the UMPS gene are detected and quantified. At least one of an increased expression of isoform A identified in SEQ ID 1 in the sample on comparison to a suitable control, a decreased expression of isoform B identified in SEQ ID 2 in the sample on comparison to a suitable control, and an increase in the ratio of expression of isoform A and isoform B in the compared to the ratio of expression for a suitable control indicates a high level UMPS expression and in turn indicates a positive clinical responsiveness to chemotherapeutic treatment comprising 5-FU. Subjects demonstrating a high level of UMPS expression may be given a lower dose of 5-FU.

At least one of a decreased expression of isoform A identified in SEQ ID 1 in the sample on comparison to a suitable control, an increased expression of isoform B identified in SEQ ID 2 in the sample on comparison to a suitable control, and a decrease in the ratio of expression of isoform A and isoform B in the compared to the ratio of expression for a suitable control indicates a low level UMPS expression and in turn indicates a negative clinical responsiveness to chemotherapeutic treatment comprising 5-FU. Subjects demonstrating a low level of UMPS expression may be given a higher dose of 5-FU.

The above-described embodiments have been provided as examples, for clarity in understanding the invention. A person of skill in the art will recognize that alterations, modifications and variations may be effected to the embodiments described above while remaining within the scope of the invention as defined by the claims appended hereto.

Claims

1. An isolated nucleic acid molecule comprising a nucleotide sequence selected from the group consisting of nucleotide sequences set forth in SEQ ID NO:1, SEQ ID NO:2, SEQ ID NO:3, SEQ ID NO:4, SEQ ID NO:5, SEQ ID NO: 6, SEQ ID NO:7, SEQ ID NO:8, SEQ ID NO:9, SEQ ID NO: 10, said nucleotide sequence encoding at least a corresponding amino acid sequence set forth in SEQ ID NO: 11, SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO:20.

2. An isolated nucleic acid molecule according to claim 1, wherein said nucleotide sequence is set forth in SEQ ID NO:1

3. An isolated nucleic acid molecule according to claim 1, wherein said nucleotide sequence is set forth in SEQ ID NO:2.

4. A method of predicting a subject's response to a chemotherapeutic treatment, said method comprising the steps of: obtaining a sample from said subject; detecting a target gene in said sample; assessing said sample to detect at least one isoform of the target gene; if said isoform of the target gene is detected, identifying and quantifying said isoform; comparing the quantified isoform with a suitable control; wherein a change in expression of said quantified isoform compared to said control indicates the subject's responsiveness to said chemotherapeutic treatment.

5. A method according to claim 4, wherein said sample is a sample selected from the group consisting of: non-cancerous cells, cancerous cells, tumour cells, tissues, fluids, DNA, and RNA.

6. A method according to claim 5, wherein said fluid is selected from the group consisting of: blood, plasma, and urine.

7. A method according to claim 4, wherein said suitable control is selected from the group consisting of a clinically selected control, a stored dataset comprising results generated from studies of one or more population(s) of cancer- free subjects, a stored dataset comprising results generated from studies of one or more population(s) of subjects responsive to said chemotherapeutic treatment, a stored dataset of results generated from studies of samples collected from a single subject over extended time periods, or combinations thereof.

8. A method according to claim 7, wherein said clinically selected control is a physiological specimen responsive to said chemotherapeutic treatment.

9. A method according to claim 7, wherein said clinically selected control is a cancer-free physiological specimen.

10. A method according to claim 4, wherein said change in expression is an increase in said quantified isoform compared to said control and indicates an increased responsiveness to said chemotherapeutic treatment.

11. A method according to claim 4, wherein said change in expression is an increase in said quantified isoform compared to said control and indicates a decreased responsiveness to said chemotherapeutic treatment.

12. A method according to claim 4, wherein said change in expression is a decrease in said quantified isoform compared to said control and indicates a decreased responsiveness to said chemotherapeutic treatment.

13. A method according to claim 4, wherein said change in expression is a decrease in said quantified isoform compared to said control and indicates an increased responsiveness to said chemotherapeutic treatment.

14. A method of predicting a subject's response to a chemotherapeutic treatment comprising 5-fluorouracil (5-FU); said method comprising the steps of: obtaining a sample from said subject; assessing said sample to detect at least one isoform of a target gene uridine 5'- monophosphate synthase (UMPS); if said isoform of said target gene is detected, identifying and quantifying said isoform; comparing said quantified isoform with a suitable control; wherein a change in expression of said quantified isoform compared to said control indicates a clinical responsiveness to said chemotherapeutic treatment.

15. A method according to claim 14, wherein said isoform is isoform B having a nucleotide sequence set forth in SEQ ID NO:2, and wherein said change in expression is an increase in the expression of isoform B compared to said control indicates a negative clinical responsiveness.

16. A method according to claim 15, wherein said negative clinical responsiveness indicates a decrease in a responsiveness to said chemotherapeutic treatment comprising 5-FU.

17. A method according to claim 14, wherein said isoform is isoform B having a nucleotide sequence set forth in SEQ ID NO:2, and wherein said change in expression is a decrease in the expression of isoform B compared to said control indicates a positive clinical responsiveness.

18. A method according to claim 17, wherein said positive clinical responsiveness indicates an increase in a responsiveness to said chemotherapeutic treatment comprising 5-FU.

19. A method according to claim 14, wherein said isoform is isoform A having a nucleotide sequence set forth in SEQ ID NO:1 and wherein said change in expression is an increase in the expression of isoform A compared to said control indicates a positive clinical responsiveness.

20. A method according to claim 19, wherein said positive clinical responsiveness is an increase in a responsiveness to said chemotherapeutic treatment comprising 5-FU.

21. A method according to claim 14, wherein said sample is a sample selected from the group consisting of: non-cancerous cells, cancerous cells, tumour cells, tissues, fluids, DNA, and RNA.

22. A method according to claim 21, wherein said fluid is selected from the group consisting of: blood, plasma, and urine.

23. A method according to claim 14, wherein said suitable control is selected from the group consisting of a clinically selected control, a stored dataset comprising results generated from studies of one or more population(s) of cancer- free subjects, a stored dataset comprising results generated from studies of one or more population(s) of subjects responsive to said chemotherapeutic treatment, a stored dataset of results generated from studies of samples collected from a single subject over extended time periods, or combinations thereof.

24. A method according to claim 23, wherein said clinically selected control is a physiological specimen responsive to treatment with said chemotherapeutic treatment.

25. A method according to claim 23, wherein said clinically selected control is a cancer- free physiological specimen.

26. A method of predicting a subject's response to a chemotherapeutic treatment comprising 5-fluorouracil (5-FU); said method comprising the steps of: obtaining a sample from said subject; assessing said sample to detect isoform B of a target gene uridine 5 '-monophosphate synthase (UMPS), said isoform B having a nucleotide sequence set forth in SEQ JD NO:2; if said isoform B of said target gene is detected, quantifying said isoform B; wherein detection of said isoform B indicates a negative clinical responsiveness to said chemotherapeutic treatment with 5-FU.

27. A method according to claim 26, wherein said sample is a sample selected from the group consisting of: non-cancerous cells, cancerous cells, tumour cells, tissues, fluids, DNA, and RNA.

28. A method according to claim 27, wherein said fluid is selected from the group consisting of: blood, plasma, and urine.

29. A method according to claim 14, wherein said chemotherapeutic treatment with 5-FU is provided in combination with at least one compound selected from the group consisting of: fluorodeoxyuridine, tegafur, uracil, Capecitabine, Uracil/tegafur, S-I, eniluracil, Oxaliplatin, Leucovorin, Bleomycin, Cisplatin, Cyclophosphamide, Doxorubicin, Fludarabine, Furosemide, Gemcitabine, Irinotecan, Etoposide, analogs of cytosine, and combinations thereof.

30. A method according to claim 26, wherein said chemotherapeutic treatment with 5-FU is provided in combination with at least one compound selected from the group consisting of: fluorodeoxyuridine, tegafur, uracil, Capecitabine, Uracil/tegafur, S-I, eniluracil, Oxaliplatin, Leucovorin, Bleomycin, Cisplatin, Cyclophosphamide, Doxorubicin, Fludarabine, Furosemide, Gemcitabine, Irinotecan, Etoposide, analogs of cytosine, and combinations thereof.

31. A method of predicting a subject's response to a chemotherapeutic treatment comprising 5-FU; said method comprising the steps of: obtaining a sample from a subject; assessing said sample to detect isoform A of a target gene undine 5 '-monophosphate synthase (UMPS), said isoform A having a nucleotide sequence set forth in SEQ ID NO:1; assessing said sample to detect isoform B of a target gene undine 5'-monophosphate synthase (UMPS), said isoform B having a nucleotide sequence set forth in SEQ ID NO:2; if said isoform A of said target gene is detected, quantifying said isoform A; if said isoform B of said target gene is detected, quantifying said isoform B; quantifying the ratio of expression of said quantified isoform A and isoform B; quantifying the ratio of expression of a suitable control known to be responsive to a treatment with 5-FU; comparing the ratio of expression of said sample to the ratio of expression of said control; wherein a change in said ratio of expression of said sample compared to said ratio of expression of said control indicates a clinical responsiveness to said chemotherapeutic treatment.

32. A method according to claim 31, wherein an increase in the ratio of expression for said sample compared to the ratio of expression for said control indicates a positive clinical responsiveness.

33. A method according to claim 32, wherein said positive clinical responsiveness indicates an increase in a responsiveness to said chemotherapeutic treatment comprising 5-FU.

34. A method according to claim 31 , wherein a decrease in the ratio of expression for said sample compared to the ratio of expression for said control indicates a negative clinical responsiveness.

35. A method according to claim 34, wherein said negative clinical responsiveness indicates a decreased responsiveness to said chemotherapeutic treatment comprising 5 -FU.

36. A method of predicting a subject's response to a chemotherapeutic treatment comprising 5-FU; said method comprising the steps of: obtaining a sample from a subject; assessing said sample to detect isoform A of a target gene uridine 5 '-monophosphate synthase (UMPS), said isoform A having a nucleotide sequence set forth in SEQ ID NO:1; assessing said sample to detect isoform B of a target gene uridine 5 '-monophosphate synthase (UMPS), said isoform B having a nucleotide sequence set forth in SEQ ID NO:2; if said isoform A of said target gene is detected, quantifying said isoform A; if said isoform B of said target gene is detected, quantifying said isoform B; quantifying the ratio of expression of said quantified isoform A and isoform B; wherein said ratio of expression of said sample indicates a degree of clinical responsiveness to said chemotherapeutic treatment with 5-FU.

37. A method according to claim 36, wherein said ratio of expression of said sample is greater than about 10, said degree indicates a high responsiveness to said chemotherapeutic treatment comprising 5-FU.

38. A method according to claim 36, wherein when said ratio of expression of said sample is less than about 0.1, said degree indicates a high resistance to said chemotherapeutic treatment comprising 5-FU.

39. A method according to claim 36, wherein when said ratio of expression of said sample is greater than about 0.1 and less than about 10, said degree indicates at least a partial responsiveness to said chemotherapeutic treatment comprising 5 -FU.

40. A method of sensitizing a tumour to administration of a chemotherapeutic treatment comprising 5-FU by enhancing the expression of isoform A having a nucleotide sequence set forth in SEQ ID NO: 1.

41. A method of sensitizing a tumour to administration of a chemotherapeutic treatment comprising 5-FU by reducing the expression of isoform B having a nucleotide sequence set forth in SEQ ID NO:2.

42. A method of sensitizing a tumour to a chemotherapeutic treatment comprising 5-FU by increasing the ratio of expression of isoform A having a nucleotide sequence set forth in SEQ ID NO:1 to isoform B having a nucleotide sequence set forth in SEQ ID NO:2 , wherein said ratio of expression is at least greater than about 0.1.

43. The method according to claim 40, wherein expression of isoform B having a nucleotide sequence set forth in SEQ ID NO:2 is suppressed.

44. The method according to claim 40, wherein said enhancement of the expression of isoform A having a nucleotide sequence set forth in SEQ ID NO:1 includes introducing a gene into said tumour wherein said gene regulates the splicing of the UMPS gene to preferentially express isoform A.

45. The method according to claim 20, wherein said suppression of the expression of isoform B having a nucleotide sequence as set forth SEQ ID NO:2 includes targeting said isoform B using siRNA.

46. A method according to claim 31 , wherein said sample is a sample selected from the group consisting of: non-cancerous cells, cancerous cells, tumour cells, tissues, fluids, DNA, and RNA.

47. A method according to claim 46, wherein said fluid is selected from the group consisting of: blood, plasma, and urine.

48. A method according to claim 36, wherein said sample is a sample selected from the group consisting of: non-cancerous cells, cancerous cells, tumour cells, tissues, fluids, DNA, and RNA.

49. A method according to claim 48, wherein said fluid is selected from the group consisting of: blood, plasma, and urine.

50. A method to identify at least one isoform of a target gene resulting from differential gene expression and alternative splicing expression events, wherein said method comprising the steps of: obtaining a sample responsive to a chemotherapeutic compound; obtaining a sample resistant to said chemotherapeutic compound; applying a microarray analysis to said responsive cells, said resistant cells, and said tumour cells; detecting and identifying differential gene expression and alternative splicing expression events associated with a resistance to said chemotherapeutic compound; quantifying the differential gene expression and alternative splicing expression events; wherein said at least one isoform is identified when at least a single genomic change is identified in said resistant cells that is not present in said responsive cells.

51. The method according to claim 50, wherein the method further comprises the step of: comparing said differential gene expression and alternative splicing expression events with the sequenced target gene within the human genome and detecting at least one mutation in said gene; wherein said at least one isoform is identified when at least a single mutation is identified in said resistant cells that is not present in said responsive cells.

52. The method according to claim 50, wherein the method further comprises the steps of: obtaining tumour cells from a tumour cell bank; applying a microarray analysis said tumour cells; wherein said at least one isoform is identified when at least a single genomic change is identified in said resistant cells and/or said tumour cells that is not present in said responsive cells.

53. The method according to claim 52, wherein the method further comprises the step of: comparing said differential gene expression and alternative splicing expression events with the sequenced target gene within the human genome and detecting at least one mutation in said gene; wherein said at least one isoform is identified when at least a single mutation is identified in said resistant cells that is not present in said responsive cells.

54. The method according to claim 50, wherein said chemotherapeutic compound is 5-FU.

55. A method of sensitizing a tumour to administration of a chemotherapeutic treatment comprising 5-FU by enhancing the protein production of isoform A having an amino acid sequence set forth in SEQ ID NO: 11.

56. A method of sensitizing a tumour to administration of a chemotherapeutic treatment comprising 5-FU by inhibiting the protein production of isoform B having an amino acid sequence set forth in SEQ ID NO: 12.

57. A method according to claim 14, wherein said quantified isoform is sequenced.

58. A method according to claim 57, wherein a change in sequence of said quantified isoform compared to said control indicates a clinical responsiveness to said chemotherapeutic treatment.

59. A kit for detecting expression of an UMPS isoform in a physiological sample collected from a subject, the kit comprising at least some of suitably selected reagents and receptacles for producing a reaction therein indicative of the subject's responsiveness to a chemotherapeutic treatment.