WO2009109952A2

WO2009109952A2 - Detection and treatment of an invasive cancer phenotype

Info

Publication number: WO2009109952A2
Application number: PCT/IE2009/000005
Authority: WO
Inventors: Naomi Walsh; Norma O'donovan; Paul Dowling; Martin Clynes
Original assignee: Dublin City University
Priority date: 2008-03-03
Filing date: 2009-03-03
Publication date: 2009-09-11
Also published as: WO2009109952A3

Abstract

A method for the inhibition, prevention or treatment of invasive/metastatic cancer in an individual in need thereof, comprises a step of treating the individual with an agent capable of attenuating the activity of protein selected from the group consisting of: STIPl; and ALDHlAl. Suitably, the invasive/metastatic cancer is pancreatic cancer, typically a metastases selected from the group consisting of: bone metastases; lung metastases; liver metastases; bone marrow metastases; breast metastases; and brain metastases. The agent is an agent that suppresses the expression of the protein, for example, siRNA, miRNA, shRNA, a ribozyme, or antisense oligonucelotide.

Description

DETECTION AND TREATMENT OF AN INVASIVE CANCER PHENOTYPE

Introduction

The invention relates to methods of detection, inhibition, prevention, or treatment of invasive/metastatic cancer in an individual. In particular, the invention relates to a methods of detection, inhibition, prevention, or treatment of invasive/metastatic cancer in an individual having pancreatic cancer.

Pancreatic cancer is one of the most lethal cancers and is the 8^th leading cause of cancer- related deaths in Europe (1). Pancreatic cancer is associated with poor prognosis, whereby the rate of mortality is similar to that of the rate of incidence. It is the most fatal malignancy; all-stage 5-year survival rate is less than 5% (2),(3). Conventional approaches including, surgery, radiation, chemotherapy and combination of theses therapies, has had little effect on the survival rate of patients diagnosed with pancreatic cancer. Pancreatic cancer appears to be inherently resistant to a wide variety of chemotherapeutic agents, which can differ greatly and are unrelated with respect to molecular structure and target specificity. The malignant progression of the invasiveness and metastatic potential of this cancer is complex and poorly understood. In this study, we established the proteomic profile of proteins secreted into the media from pancreatic cancer cell lines with varying invasive and malignant transformation characteristics. Theoretically, proteins secreted by tumour cells are more likely to be detected easily in bodily fluids such as urine, blood serum and pancreatic ductal juice. Therefore, secreted proteins and their metabolites found in vivo could represent a panel of potential biomarkers. As pancreatic cancer invades and metastasises at an early stage without symptoms, it is vital to develop early detection systems for the diagnosis of pancreatic cancer. Studies have reported proteomic analyses of pancreatic tissue, pancreatic juice as well as blood plasma and sera (4). Molecular markers and biomarkers constitute major targets for the early detection of cancer, identification of cancer risk and/or prediction of therapeutic response (5). Proteomics provides an excellent means for analysis of bodily fluids for classifying proteins and identifying biomarkers for early detection of cancers. The main biomarker currently available for pancreatic cancer detection, CA 19-9 has demonstrated to have sensitivity up to 90% and specificity up to 98% in the diagnosis of this malignancy (6, 7), however, this marker is not fully specific as false-positive or false- negative findings occurs in patients with other gastrointestinal malignancies and also in patients with benign disease, particularly when associated with obstructive jaundice or cirrhosis, which may contribute to late diagnosis of pancreatic cancer. Approximately 10% of the population with the Lewis negative genotype are not able to produce CA 19- 9, due to a deficiency in a fucosyltransferase specified by the Le gene that is involved in its synthesis (8). Therefore, in a sub-set of patients, CA 19-9 expression will be falsely low even in the presence of advanced pancreatic cancer (9).

Statement of Invention

The invention is based on the finding that the expression level of certain proteins is modulated in cancers according to the invasiveness/metastatic potential of the cancer. Thus, certain proteins have been found to be overexpressed in a highly invasive cancer, certain protein have been found to be underexpressed in a highly invasive cancer, and certain proteins have been found to be overexpressed in a low invasiveness cancer. Thus, the expression levels of these proteins function as biomarkers of invasiveness/metastases potential, and biomarkers in the early diagnosis of cancer. Further, it has been found that the invasiveness of a cancer may be attenuated by modulation of the expression of a subset of specific biomarkers, especially STIPl and ALDHlAl.

According to the invention, there is provided a method of assessing the status of a cancer in an individual comprising a step of assessing a biological sample from the individual for the expression level of a protein selected from the group ALDHlAl, VIM, STIPl, TPIl, KRT18, GAPDH, GSN, Integrin Bl, Integrin α5, and Integrin α6, and correlating the expression level of the protein with cancer status. Typically, the GSN protein is GSN isoform b. In one embodiment, the method comprises a step of assessing a biological sample obtained from the individual for the expression level of a protein selected from the group ALDHlAl, VIM, STIPl , TPIl.

In another embodiment, the method comprises a step of assessing a biological sample obtained from the individual for the expression level of a protein selected from the group KRTl 8, GAPDH, GSN.

In another embodiment, the method comprises a step of assessing a biological sample obtained from the individual for the expression level of a protein selected from the group Integrin Bl, Integrin α5, and Integrin α6.

The term "cancer status" should be taken to mean cancer diagnosis, especially early cancer diagnosis, invasive/metastases potential of a cancer, assessment of likely patient outcome due to the cancer, and assessment of effectiveness of a treatment for a cancer.

The term "cancer" should be taken to mean any cancer, including a cancer selected from the group consisting of: fibrosarcoma; myxosarcoma; liposarcoma; chondrosarcom; osteogenic sarcoma; chordoma; angiosarcoma; endotheliosarcoma; lymphangiosarcoma; lymphangioendotheliosarcoma; synovioma; mesothelioma; Ewing's tumor; leiomyosarcoma; rhabdomyosarcoma; colon carcinoma; pancreatic cancer; breast cancer; ovarian cancer; prostate cancer; squamous cell carcinoma; basal cell carcinoma; adenocarcinoma; sweat gland carcinoma; sebaceous gland carcinoma; papillary carcinoma; papillary adenocarcinomas; cystadenocarcinoma; medullary carcinoma; bronchogenic carcinoma; renal cell carcinoma; hepatoma; bile duct carcinoma; choriocarcinoma; seminoma; embryonal carcinoma; Wilms' tumor; cervical cancer; uterine cancer; testicular tumor; lung carcinoma; small cell lung carcinoma; bladder carcinoma; epithelial carcinoma; glioma; astrocytoma; medulloblastoma; craniopharyngioma; ependymoma; pinealoma; hemangioblastoma; acoustic neuroma; oligodendroglioma; meningioma; melanoma; retinoblastoma; and leukemias. In a preferred embodiment of the invention, the cancer is pancreatic cancer. In this specification, the term "biological sample" may be any sample obtained from an individual such as, for example, blood, serum, saliva, urine, cerebrospinal fluid, tissue, cells, etc. Suitably, when the protein being assayed is a circulating protein, the biological sample will be serum. GSN is known to be serum proteins and have been shown to be differentially expressed in the conditioning medium of cancers of differing invasiveness/metastases potential. In many cases, the individual will be a person suspected of having cancer, or pre-disposed to developing cancer as determined by other phenotypic, genotypic or hereditary traits. In other cases, the individual may be a person known to have cancer, and who is undergoing a therapeutic treatment regime, in which case the method of the invention may be employed to monitor the effectiveness of the treatment, or may be a post-operative patient being monitored for re-occurrence of the disease.

In one embodiment of the invention, the method is a method of assessing the invasive/metastatic potential of a cancer, and in which overexpression of a protein selected from the group ALDHlAl, VIM, STIPl, TPIl is associated with an invasive/metastatic potential. Typically, the group of proteins comprises STIPl and ALDHlAl .

In another embodiment of the invention, the method is a method for the early detection of a cancer, in which overexpression of a protein selected from the group ALDHlAl, VIM, STIPl, TPIl is associated with early detection of the cancer. Typically, the cancer is pancreatic cancer. Typically, the group of proteins comprises STIPl and ALDHlAl.

In another embodiment of the invention, the method is a method of monitoring the effectiveness of a treatment for a cancer, especially a treatment for reducing the invasiveness/metastates potential of a cancer, or a treatment for a metastases, in which a decrease in the expression of a protein selected from the group ALDHlAl, VIM, STIPl, TPIl is associated with effectiveness of the treatment. Typically, the group comprises STIPl and ALDHlAl . In one embodiment of the invention, the method is a method of assessing the invasive/metastatic potential of a cancer, and in which underexpression of a protein selected from the group KRT 18, GAPDH, GSN is associated with an invasive/metastatic potential.

In another embodiment of the invention, the method is a method for the early detection of a cancer, in which underexpression of a protein selected from the group KRT 18, GAPDH, GSN is associated with early detection of the cancer. Typically, the cancer is pancreatic cancer.

In another embodiment of the invention, the method is a method of monitoring the effectiveness of a treatment for a cancer, especially a treatment for reducing the invasiveness/metastates potential of a cancer, in which an increase in the expression of a protein selected from the group KRT 18, GAPDH, GSN is associated with effectiveness of the treatment.

In one embodiment of the invention, the method is a method of assessing the invasive/metastatic potential of a cancer, and in which overexpression of a protein selected from the group Integrin Bl, Integrin α5, and Integrin α6 is associated with non- invasive/non-metastatic potential.

In another embodiment of the invention, the method is a method of monitoring the effectiveness of a treatment for a cancer, especially a treatment for reducing the invasiveness/metastases potential of a cancer, in which an increase in the expression of a protein selected from the group Integrin βl, Integrin α5, and Integrin α6 is associated with effectiveness of the treatment.

In this specification, the term "overexpression" of a protein should be taken to mean a level of expression of the protein which is significantly higher than the level of expression the protein in a reference non-aggressive pancreatic cancer cell. In this specification, the term "underexpression" of a protein should be taken to mean a level of expression of the protein which is significantly lower than the level of expression the protein in a reference non-aggressive pancreatic cancer cell.

The invention also relates to a method of treating a cancer in an individual comprising a step of attenuating an activity of a protein selected from the group ALDHlAl, VIM, STIPl, TPIl in the individual. Typically, the group comprises STIPl and ALDHlAl. Typically, the method is a method of inhibiting, preventing or treating an invasive/metastatic cancer in an individual, typically in an individual with an established cancer. Suitably, the established cancer is pancreatic cancer. The term "inhibiting, preventing or treating an invasive/metastatic cancer" should be understood as including one or more of decreasing the invasiveness/metastatic potential of the cancer, inhibiting or preventing invasion of the cancer cells, and preventing or treating metastases in an individual. The term "invasive/metastatic cancer" should be understood as meaning invasive cancer, or metastatic cancer, or, in one embodiment, a cancer having both and invasive and metastatic phenotype. The invention also relates to the use of agent capable of attenuating the activity of a protein selected from the group ALDHlAl, VIM, STIPl, TPIl as a medicament. The invention also relates to a pharmaceutical composition comprising an agent capable of attenuating the activity of a protein selected from the group ALDHlAl, VIM, STIPl, TPIl.

The invention also relates to a method of treating a cancer in an individual comprising a step of increasing an activity of a protein selected from the group KRT 18, GAPDH, GSN in the individual. Typically, the method is a method of decreasing the invasiveness/metastatic potential of the cancer, or preventing or treating metastases in an individual. The invention also relates to the use of agent capable of increasing the activity of a protein selected from the group KRTl 8, GAPDH, GSN as a medicament. The invention also relates to a pharmaceutical composition comprising an agent capable of increasing the activity of a protein selected from the group KRTl 8, GAPDH, GSN. The invention also relates to a method of treating a cancer in an individual comprising a step of increasing an activity of a protein selected from the group Integrin Bl, Integrin α5, and Integrin α6 in the individual. Typically, the method is a method of decreasing the invasiveness/metastatic potential of the cancer, or preventing or treating metastases in an individual. The invention also relates to the use of agent capable of increasing the activity of a protein selected from the group Integrin Bl, Integrin α5, and Integrin α6 as a medicament. The invention also relates to a pharmaceutical composition comprising an agent capable of increasing the activity of a protein selected from the group Integrin B 1 , Integrin α5, and Integrin α6.

When the methods of the invention involve an agent that attenuates a protein activity, this should be taken to include an agent that suppresses the expression of the protein (i.e. interferes with expression of the gene encoding the protein), including suppression of transcription or translation, and an agent that directly inhibits the protein activity (antagonist). In one preferred embodiment of the invention, expression of the protein is suppressed by means of RNA interference (RNAi). RNA interference (RNAi) is an evolutionally highly conserved process of post-transcriptional gene silencing (PTGS) by which double stranded RNA (known as siRNA molecules), when introduced into a cell, causes sequence-specific degradation of mRNA sequences. The RNAi machinery, once it finds a double-stranded RNA molecule, cuts it up, separates the two strands, and then proceeds to destroy RNA molecules that are complementary to one of those segments, or prevent their translation into proteins. Thus, suppression of a proteins expression may be achieved by treating an individual with siRNA molecules designed to target mRNA for the protein. siRNA molecules designed to knockdown STIPl are provided in SEQUENCE ID NO's: 8, 9 and 10. siRNA molecules designed to knockdown ALDHlAl are provided in SEQUENCE ID NO's: 11, 12 and 13. Other types of gene knockdown tools will be well known to the person skilled in the filed of molecular biology. For example, micro RNA's (miRNAs) are small (~22nt) non-coding RNAs (ncRNAs) that regulate gene expression at the level of translation. Each miRNA apparently regulates multiple genes and hundreds of miRNA genes are predicted to be present in mammals. Recently miRNAs have been found to be critical for development, cell proliferation and cell development, apoptosis and fat metabolism, and cell differentiation. Alternatively, small hairpin RNA (shRNA) molecules are short RNA molecules having a small hairpin loop in their tertiary structure tha may be employed to silence genes. The design of miRNA or shRNA molecules capable of silencing a given protein will be apparent to those skilled in the field of miRNA or shRNA molecule design. As an alternative, the level of protein expression can be modulated using antisense or ribozyme approaches to inhibit or prevent translation of the protein mRNA transcripts or triple helix approaches to inhibit transcription of the gene for the protein. Antisense approaches involve the design of oligonucleotides (either DNA or RNA) that are complementary to mRNA for the protein. The antisense oligonucleotides will bind to the complementary mRNA transcripts and prevent translation. Ribozyme molecules designed to catalytically cleave mRNA transcripts of a given protein can also be used to prevent translation and expression of the protein. (See, e. g. , PCT International PublicationW090/l 1364, published October 4,1990 ; Sarver et al. , 1990, Science 247: 1222-1225).

In one embodiment, attenuation of a proteins activity is achieved using an agent that directly inhibits the proteins activity, such as for example an antagonist or inhibitor of the protein or an antibody specific to the protein. Where the target protein is STIPl, the inhibitor may be a HSP90 inhibitor. Examples of such inhibitors include geldanamycin (17-AAG), retaspimycin, and small molecule inhibitors of HSP90. Other inhibitors will be well known to those skilled in the art. Thus, the invention also relates to the use of a HSP90 inhibitor in the inhibition, prevention, and or treatment of invasive/metastasic cancer, especially pancreatic cancer. When the target protein is ALDHlAl, the inhibitor is, for example, Disulfiram, 4-(N, N-dipropylamino)benzaldehyde (DPAB) or 4-(N, N- diethylamino) benzaldehyde (DEAB). Other examples of ALDHlAl will be known to the skilled person.

Thus, the invention also relates to the use of a ALDHlAl inhibitor in the inhibition, prevention, and or treatment of invasive/metastasic cancer, especially pancreatic cancer. When the methods of the invention involve an agent that increases a protein activity, this should be taken to include the administration of the protein itself, or a biologically active fragment or variant of the protein, or the administration of an agonist of the protein.

In this specification, the term "biologically active" should be taken to mean that the fragment retains all or part of the biological functionality of the parent protein. In the case of KRTl 8, GAPDH, GSN, Integrin Bl, Integrin α5, and Integrin α6, this means that the fragment of one of these proteins will be biologically active if it retains the ability to inhibit an invasive phenotype in a cancer cell.

A "fragment" of a protein means a contiguous stretch of amino acid residues of at least 5 amino acids, preferably at least 6 amino acids. Typically, the "fragment" will comprise at least 10, preferably at least 20, more preferably at least 30, and ideally at least 40 contiguous amino acids. In this regard, it would be a relatively straightforward task to make fragments of the protein and assess the biological activity of such fragments using the in-vitro models described below.

A "variant" of a protein shall be taken to mean proteins having amino acid sequences which are substantially identical to the wild-type protein, especially the human wild-type protein. Thus, for example, the term should be taken to include proteins or polypeptides that are altered in respect of one or more amino acid residues. Preferably such alterations involve the insertion, addition, deletion and/or substitution of 5 or fewer amino acids, more preferably of 4 or fewer, even more preferably of 3 or fewer, most preferably of 1 or 2 amino acids only. Insertion, addition and substitution with natural and modified amino acids is envisaged. The variant may have conservative amino acid changes, wherein the amino acid being introduced is similar structurally, chemically, or functionally to that being substituted. Typically, proteins which have been altered by substitution or deletion of catalytically-important residues will be excluded from the term "variant". Generally, the variant will have at least 70% amino acid sequence homology, preferably at least 80% sequence homology, more preferably at least 90% sequence homology, and ideally at least 95%, 96%, 97%, 98% or 99% sequence homology with wild-type human protein. In this context, sequence homology comprises both sequence identity and similarity, i.e. a polypeptide sequence that shares 70% amino acid homology with wild-type human protein is one in which any 70% of aligned residues are either identical to, or conservative substitutions of, the corresponding residues in wild-type human protein.

The term "variant" is also intended to include chemical derivatives of the protein, i.e. where one or more residues of the protein is chemically derivatized by reaction of a functional side group. Also included within the term variant are molecules in which naturally occurring amino acid residues are replaced with amino acid analogues.

Proteins and polypeptides (including variants and fragments thereof) of and for use in the invention may be generated wholly or partly by chemical synthesis or by expression from nucleic acid. The proteins and peptides of and for use in the present invention can be readily prepared according to well-established, standard liquid or, preferably, solid-phase peptide synthesis methods known in the art (see, for example, J. M. Stewart and J. D. Young, Solid Phase Peptide Synthesis, 2nd edition, Pierce Chemical Company, Rockford, Illinois (1984), in M. Bodanzsky and A. Bodanzsky, The Practice of Peptide Synthesis, Springer Verlag, New York (1984).

The invention also relates to a kit for assessing cancer status in an individual, comprising components for detecting and/or measuring the level of a protein selected from the group comprising ALDHlAl, VIM, STIPl, TPIl, KRT18, GAPDH, GSN, Integrin Bl, Integrin α5, and Integrin α6. Typically, the group comprises STIPl and ALDHlAl. In one embodiment, the kit comprises a support having an antibody specific to at least one protein selected from the above group anchored thereon. Preferably, the support comprises a plurality of antibodies specific to at least two proteins anchored thereon. More preferably, the kit comprises a support carrying a repertoire of antibodies suitable for detecting three, four, five, six, seven, eight, nine, or ten proteins anchored thereon. In one embodiment, the support is selected from the group comprising: a microtitre plate; a glass slide; a polymer membrane; and an affinity column. In one particularly preferred embodiment of the invention, the kit comprises an ELIS A™ kit adapted to detect one or more of the above group of proteins.

The invention also relates to a pharmaceutical composition comprising an agent that attenuates the activity of a protein selected from the group consisting of STIPl and ALDHlAl, and a suitable carrier or pharmaceutical excipient. Suitably, the agent is an oligonucelotide capable of knocking down the protein. Typically, the oligonucleotide is selected from the group consisting of: siRNA; miRNA; shRNA; a ribozyme; and an antisense oligonucelotide. In one embodiment, the agent is selected from the group consisting of: an inhibitor of the protein; and an antibody that specifically binds to the protein. Where the protein is STIPl, and the agent is ideally is a HSP90 inhibitor. Where the protein is ALDHlAl, and the agent is ideally a ALDHlAl inhibitor. Details of suitable inhibitors are provided above. In one embodiment, the composition includes an effective amount of a cytotoxic agent.

Where the agent is an antibody, it is suitably a blocking antibody, and ideally a humanised, or fully human, antibodies. Techniques for generating such antibodies are well known to the person skilled in the art.

The invention also relates to the use of a HSP90 inhibitor for the inhibition, prevention or treatment of an invasive/metastatic cancer. Typically, the invasive/metastatic cancer is pancreatic cancer.

The invention also relates to methods of identifying compounds useful in the inhibition, prevention or treatment of an invasive/metastatic cancer, comprising determining a reference level of activity of a protein, contacting the protein with a candidate compound, and determining the level of activity of the contacted protein, wherein a decrease in the level of activity of the contacted protein relative to the reference level of protein activity is an indication that the candidate compound is useful in the inhibition, prevention or treatment of an invasive/metastatic cancer, wherein the protein is selected from the group consisting of: STIP 1; and ALDHlAl . Typically, the protein is provided in the form of protein expressing cells, and in which the level of activity is determined by assaying for a level of expression of protein in the cells. Where the protein is STIP, the STIP-expressing cells are, for example, pancreatic tumour cells. Suitably, the invasive metastatic cancer is pancreatic cancer.

The invention also relates to a method of identifying an agent that suppresses expression of STIPl or ALDHlAl protein comprising the steps of providing a source of STIPl or ALDHlAl expressing cells, treating the cells with a candidate agent, and assaying the cells for expression of STIPl or ALDHlAl, wherein a decrease in the level of expression of STIPl or ALDHlAl protein in the treated cells relative to untreated cells is an indication that the candidate agent is useful in suppressing expression of STIPl or ALDHlAl protein.

The invention also relates to a method of identifying an agent useful in the inhibition, prevention or treatment of an invasive/metastatic cancer, comprising a step of providing a sample of cells that express HSP90, treating the cells with a candidate agent, and assaying the cells for expression of HSP90, wherein a decrease in the level of expression of HSP90 in the treated cells relative to untreated cells is an indication that the candidate agent is useful in the inhibition, prevention or treatment of an invasive/metastatic cancer.

Generally, with the screening assays of the invention, a sample of cells will be chosen that express the target protein of interest. Many cancer cell lines will be useful in this regard, including (for ALDHlAl activity or espression assays), many known lung cancer cell lines.

The invention also relates to a method of detecting a cancer cell having an invasive/metastatic phenotype, comprising a step of assaying a biological sample from an individual for a level of a biomarker, and correlating the level with invasive/metastatic potential, wherein the biomarker is selected from the group consisting: STIPl ; and ALDHlAl. Suitably, the step of correlating the level with invasive/metastatic potential involves comparing the level of the protein with a reference level from a cell line having a reference invasiveness/agressiveness.

In one embodiment of the invention, the biological sample is a sample of cells, and wherein the cells are stained for the biomarker, and in which an invasive/metastatic phenotype is correlated with the level of the staining of the or each of the biomarkers. The Allred system may be employed in this regard, in which an Allred score of 0 correlates with no cells staining, 1 correlates with less than 1%, 2 correlates with 1% to 10%, 3 correlates with 11% to 33%, 4 correlates with 34% to 67%, and 5 correlates with more than 67%. This scoring system is the sum of a proportion score and an intensity score. The proportion score is an estimate of the proportion of positive cells on the entire slide and is divided into the 5 above categories. The intensity score estimates the average staining intensity of positive tumor cells: 0, no staining; 1, weak positive membrane staining; 2, moderate; and 3, strong staining. The 2 scores are added together to give a final numerical score ranging from 0 to 8.

An alternative method in Table 1 outlines the percentage and intensity grade of staining routinely used in pancreatic cancer scoring.

Table 1 Percentage and intensity grade of staining

Percentage grade of staining Intensity grade of staining

In one embodiment, the biological sample is a biological fluid, the method comprising a step of determining a level of the biomarker, and comparing the measured level of biomarker with a reference level, wherein a measured level greater than the reference level correlates with the cancer cells having an invasive/metastatic potential, and wherein a measured level less than the reference level correlates with the cancer cells not having an invasive/metastatic potential.

ALDHlAl (aldehyde dehydrogenase) is an enzyme involved in the conversion of aldehydes to their corresponding acids by NAD(P)+ dependent reactions (10). It has been found to be approximately 9-fold up-regulated in Clone #3 compared to Clone #8, and to be more highly expressed in pancreatic tumour tissue compared to normal pancreatic tissue. The amino acid sequence of ALDHlAl is provided below (SEQUENCE ID NO: 1):

MSSSGTPDLPVLLTDLKIQYTKIFINNEWHDSVSGKKFPVFNPATEEELC QVEEGDKEDVDKAVKAARQAFQIGSPWRTMDASERGRLLYKLADLIERDR LLLATMESMNGGKLYSNAYLSDLAGCIKTLRYCAGWADKIOGRTIPIDGN FFTYTRHEPIGVCGOHPWNFPLVMLIWKIGPALSCGNTVVVKPAEOTPL TALHVASLIKEAGFPPGVVNIVPGYGPTAGAAISSHMDIDKVAFTGSTEV GKLIKEAAGKSNLKRVTLELGGKSPCIVLADADLDNAVEFAHHGVFYHQG QCCIAASRIFVEESIYDEFVRRS VERAKKYILGNPLTPGVTQGPQIDKEQ YDKILDLIESGKKEGAKLECGGGPWGNKGYFVQPTVFSNVTDEMRIAKEE IFGPVOOIMKFKSLDDVIKRANNTFYGLSAGVFTKDIDKAITISSALOAG TVWVNCYGVVSAQCPFGGFKMSGNGRELGEYGFHEYTEVKTVTVKISQKN S Underlined sequences are some of the peptide identification by MALDI-TOF MS

Overexpression of the protein in a cancer therefore functions as a biomarker of invasiveness/metastasis potential. Further, attenuation of expression of the protein has been shown to decrease invasiveness, and increase adhesion, in a highly invasive pancreatic cancer cell model relative to untreated cells., Direct inhibition of the protei is shown to decrease invasiveness and % survival in pancreatic cancer cell lines compared to lung and breast cancer cell lines. Thus, attenuation of ALDHlAl activity is a prophylactic or therapeutic treatment for cancer invasiveness/metastasis, especially in pancreatic cancer. VIM (vimentin) is a cytoskeletal protein (11). It has been found to be approximately 5.5- fold up-regulated in Clone #3 compared to Clone #8. The amino acid sequence of VIM is provided below (SEQUENCE ID NO: 2):

MSTRSVSSSSYRRMFGGPGTASRPSSSRSYVTTSTRTYSLGSALRPSTSR SLYASSPGGVYATRSSAVRLRSSVPGVRLLQDSVDFSLADAINTEFKNTR TNEKVELQELNDRFANYIDKVRFLEOONKILLAELEOLKGOGKSRLGDLY EEEMRELRRQVDQLTNDKARVEVERDNLAEDIMRLREKLQEEMLQREEAE NTLQSFRQDVDNASLARLDLERKVESLQEEIAFLKKLHEEEIQELQAQIQ EQHVQIDVDVSKPDLTAALRDVROOYESVAAKNLOEAEEWYKSKFADLSE AANRNNDALRQ AKQESTEYRRQ VQSLTCEVDALKGTNESLERQMREMEEN FAVEAANYQDTIGRLQDEIONMKEEMARHLREYODLLNVK MALDIEIATY RKLLEGEESRISLPLPNFSSLNLRETNLDSLPLVDTHSKRTFLIKTVETR DGQVINETSQHHDDLE

Underlined sequence is minimum of four peptide identification by MALDI-TOF MS Overexpression of the protein in a cancer therefore functions as a biomarker of invasiveness/metastasis potential. Further, attenuation of expression of the protein has been shown to decrease invasiveness, and increase adhesion, in a highly invasive pancreatic cancer cell model relative to untreated cells. Thus, attenuation of VIM activity is a prophylactic or therapeutic treatment for cancer invasiveness/metastasis.

STIPl (stress induced phosphoprotein 1) mediates the association of the molecular chaperones Hsp70 and Hsp90 (12). It has been found to be approximately 2.6-fold up- regulated in Clone #3 compared to Clone #8. The amino acid sequence of STIPl is provided below (SEQUENCE ID NO: 3): MEQVNELKEKGNKALSVGNIDDALQCYSEAIKLDPHNHVLYSNRSAAYAK KGDYOKAYEDGCKTVDLKPDWGKGYSRKAAALEFLNRFEEAKRTYEEGLK HEANNPQLKEGLQNMEARLAERKFMNPFNMPNLYQKLESDPRTRTLLSDP TYRELIEOLRNKPSDLGTKLODPRIMTTLSVLLGVDLGSMDEEEEIATPP PPPPPKKETKPEPMEEDLPENKKOALKEKELGNDAYKKKDFDTALKHYDK AKELDPTNMTYITNQAA VYFEKGDYNKCRELCEKAIEVGRENREDYRQIA KAYARIGNSYFKEEKYKDAIHFYNKSLAEHRTPDVLKKCQQAEKILKEQE RLAYINPDLALEEKNKGNECFOKGDYPQAMKHYTEAIKRNPKDAKLYSNR AACYTKLLEFQLALKDCEECIQLEPTFIKGYTRKAAALEAMKD YTKAMDV YQKALDLDSSCKEAADGYQRCMMAQYNRHDSPEDVKRRAMADPEVQQIMS DPAMRLILEQMQKDPQALSEHLKNPVIAQKIQKLMDVGLIAIR Underlined sequence is minimum of four peptide identification by MALDI-TOF MS

Overexpression of the protein in a cancer therefore functions as a biomarker of invasiveness/metastasis potential. Further, attenuation of expression, or activity, of the protein has been shown to decrease invasiveness, and increase adhesion, in a highly invasive pancreatic cancer cell model relative to untreated cells, and in other cell lines of pancreatic cancer. Thus, attenuation of STIPl activity is a prophylactic or therapeutic treatment for cancer invasiveness/metastasis, especially in pancreatic cancer.

Glycolytic protein TPIl (triphosphate isomerise a) has been found to be approximately 4- fold up-regulated in Clone #3 compared to Clone #8. The amino acid sequence of TPIl is provided below (SEQUENCE ID NO: 4):

APSRKFFVGGNWKMNGRKQSLGELIGTLNAAKVPADTEVVCAPPTAYIDFARQK LDPKIAVAAONCYKVTNGAFTGEISPGMIKDCGATWVVLGHSERRHVFGESDEL IGQKVAHALAEGLGVIACIGEKLDEREAGITEKVVFEQTKVIADNVKDWSKVVL AYEPVWAIGTGKTATPOOAOEVHEKLRGWLKSNVSDAVAOSTR IYGGSVTGATCKELASOPDVDGFLVGGASLKPEFVDIINAKO

Underlined sequence is minimum of four peptide identification by MALDI-TOF MS Overexpression of the protein in a cancer therefore functions as a biomarker of invasiveness/metastasis potential.

Cytoskeletal protein KRTl 8 (keratin 18) has been found to be approximately 3-fold down-regulated in Clone #3 compared to Clone #8. The amino acid sequence of KRTl 8 is provided below (SEQUENCE ID NO: 5):

MSFTTRSTFSTNYRSLGSVQAPSYGARPVSSAASVYAGAGGSGSRISVSR STSFRGGMGSGGLATGIAGGLAGMGGIQNEKETMQSLNDRLASYLDRVRS LETENRRLESKIREHLEKKGPQVRDWSHYFKIIEDLRAQIFANTVDNARI VLQIDNARLAADDFRVKHETELAMROSVENDIHGLRKVIDDTNITRLOLE TEIEALKEELLFMKKNHEEEVKGLQAQIASSGLTVEVDAPKSQDLAKIMA DIRAOYDELARKNREELDKYWSOOIEESTTVVTTOSAEVGAAETTLTELR RTVQSLEIDLDSMRNLKASLENSLREVEARYALOMEOLNGILLHLESELA QTRAEGQRQAQEYEALLNIKVKLEAEIATYRRLLEDGEDFNLGDALDSSN SMQTIQKTTTRRIVDGKVVSETNDTKVLRH

Underexpression of the protein in a cancer therefore functions as a biomarker of invasiveness/metastasis potential.

Glycolytic protein GAPDH (glyceraldehyde 3-phosphate dehydrogenase) has been found to be approximately 2.6-fold down-regulated in Clone #3 compared to Clone #8. The amino acid and nucleic acid sequence of GAPDH is provided below (SEQUENCE ID

NO: 6):

MGKVKVGVNGFGRIGRLVTRAAFNSGKVDIVAINDPFIDLNYMVYMFQYD

STHGKFHGTVKAENGKLVINGNPITIFQERDPSKIKWGDAGAEYVVESTG VFTTMEKAGAHLOGGAKRVIISAPSADAPMFVMGVNHEKYDNSLKIISNA

SCTTNCLAPLAKVIHDNFGIVEGLMTTVHAITATQKTVDGPSGKL WRDGR

GALONIIPASTGAAKAVGKVIPELNGK LTGMAFRVPTANVSVVDLTCRLE

KPAKYDDIKKVVKQASEGPLKGILGYTEHQVVSSDFNSDTHSSTFDAGAG

IALNDHFVKLISWYDNEFGYSNRVVDLMAHMASKE Underexpression of the protein in a cancer therefore functions as a biomarker of invasiveness/metastasis potential .

GSN (gelsolin) is a calcium-binding protein which binds to and regulates actin filaments (13). Isoform b of the protein has been found to be approximately 21 -fold down-regulated in Clone #3 compared to Clone #8. The amino acid sequence of GSN isoform b is provided below (SEQUENCE ID No: 7):

MVVEHPEFLKAGKEPGLQIWRVEKFDLVPVPTNLYGDFFTGDAYVILKTV QLRNGNLQYDLHYWLGNECSQDESGAAAIFTVQLDDYLNGRAVQHREVQG FESATFLGYFKSGLKYKKGGVASGFKHVVPNEVVVQRLFQVKGRRVVRAT EVPVSWESFNNGDCFILDLGNNIHQWCGSNSNRYERLKATQVSKGIRDNE RSGRARVHVSEEGTEPEAMLQVLGPKPALPAGTEDTAKEDAANRKLAKLY KVSNGAGTMSVSLVADENPFAQGALKSEDCFILDHGKDGKIFVWKGKQAN TEERKAALKTASDFITKMDYPKQTQVSVLPEGGETPLFKQFFKNWRDPDQ TDGLGLSYLSSHIANVERVPFDAATLHTSTAMAAQHGMDDDGTGQKQIWR IEGSNKVPVDPATYGQFYGGDSYIILYNYRHGGRQGQIIYNWQGAQSTQD EVAASAILTAQLDEELGGTPVQSRVVQGKEPAHLMSLFGGKPMIIYKGGT SREGGOTAPASTRLFOVRANSAGATRAVEVLPKAGALNSNDAFVLKTPSA AYLWVGTGASEAEKTGAQELLRVLRAOPVOVAEGSEPDGFWEALGGKAAY RTSPRLKDKKMDAHPPRLFACSNKIGRFVIEEVPGELMQEDLATDDVMLL DTWDOVFVWVGKDSOEEEKTEALTSAKRYIETDPANRDRRTPITVVKOGF EPPSFVGWFLGWDDDYWSVDPLDRAMAELAA

Underexpression of the protein in a cancer therefore functions as a biomarker of invasiveness/metastasis potential. Further, attenuation of expression of the cytoplasmic form of the protein has been shown to increase invasiveness in a pancreatic cancer cell model of low invasiveness relative to untreated cells.

Integrin Bl, Integrin α5, and Integrin α6 are cell surface receptors known to be associated with receptors of fibronectin and laminin. The proteins are overexpressed in a low invasiveness cell line compared with normal and high invasiveness cell models of pancreatic cancer. Attenuation of the protein in a low invasiveness cell model of pancreatic cancer by siRNA has been shown to increase the invasiveness of the treated cells.

Brief Description of the Figures

Figure 1 : A. Invasion assay of MiaPaCa-2, Clone #3 and Clone #8 through different extracellular matrix (ECM) proteins: matrigel, laminin, fibronectin, collagen type IV and I. Motility assay refers to invasion assay without the presence of extracellular matrices. Results are displayed as the total mean number of cells invading at 20Ox magnification (n = 3). B. Adhesion of MiaPaCa-2, Clone #3 and Clone #8 to ECM proteins: matrigel, laminin, fibronectin, collagen type IV and type I. Results are expressed as absorbance at 405 nm with a reference wavelength of 620 nm. Data shown is mean ± standard deviation (n = 3). Student's t-test; p < 0.05*, 0.01**, 0.005*** .

Figure 2: A. Percentage survival of MiaPaCa-2, Clones #3 and Clone #8 in suspension compared to adherent cells. B.Percentage colony formation efficiency (%CFE) of MiaPaCa-2, Clone #3 and Clone #8 under anchorage independent growth conditions. Data shown is mean ± standard deviation (n=3). Student's t-test; p < 0.05*, 0.01**, 0.005*** .

Figure 3 A. Bar graph displays the total number of cells invading under control conditions and also after 24 hr incubation on matrigel, scatter graph displays the total number of superinvading cells counted after 24 hrs incubation on matrigel of Clone #8, MiaPaCa-2 and Clone #3. B. Images showing the morphology of (i) MiaPaCa-2, (ii) Clone #3 and (iii) Clone #8. Magnification, 20Ox, scale bar, 200μm

Figure 4 2D DIGE expression map of Cy2, Cy3 and Cy5 labelled Clone #8 compared to Clone #3 proteins. Six proteins, ALDHlAl, STIPl, TPIl, VIM, KRT 18 and GAPDH spots were further identified as potentially important in the invasion status of the cell lines.

Figure 5 Immunofluorescence, 3D spot images and Western blot images of A vimentin and B cytokeratin 18 protein expression in MiaPaCa-2, Clone #3 and Clone #8.

Figure 6 3D spot images and Western blot images of A. ALDHlAl, B. STIPl C. TPIl and D. GAPDH expression in MiaPaCa-2, Clone #3 and Clone #8. BiP used as loading control.

Figure 7: Western blot of A. Integrin βl B. Integrin α2 C. Integrin α5 D. Integrin α6 and β-actin (below) used as loading control in (1) MiaPaCa-2, (2) Clone #3 and (3) Clone #8. E. Knockdown of integrin βl in Clone #8 cells 48 hours post transfection (siRNAs ITGβl #1 and #3). F. Knockdown of integrin α5 in Clone #8 cells 48 hours post transfection (siRNAs ITGα5 #1, #2 and #3). G. Knockdown of integrin α6 in Clone #8 cells 48 hours post transfection (siRNAs ITGα6 #1 and #2).

Figure 8: A. Invasion of Clone #8 through matrigel, laminin and fibronectin and motility assay. B. Adhesion assay of Clone #8 to matrigel, laminin and fibronectin. C. Anoikis assay. Experiments were performed 48 hours post-transfection with two different exon targeted siRNA integrin Beta 1. Untransfected- and scrambled siRNA transfected- cell lines were the controls for this experiment. Student's t-test;p < 0.05*, 0.01 **, 0.005*** .

Figure 9: A. Invasion through matrigel, laminin and fibronectin. B. Motility assay. C. Adhesion assay to matrigel, laminin and fibronectin. D. Anoikis assay of Clone #8 control, treated with scrambled siRNA and three independent integrin ITGα5 siRNA targets and two integrin ITGαό target siRNAs.. Experiments were performed 48 hours post-transfection with different exon targeted siRNAs. Untransfected- and scrambled siRNA transfected- cell lines were the controls for this experiment. Statistics: Student's t- test; p < 0.05*, 0.01**, 0.005*** .

Figure 10: Western blot of siRNA ALDHlAl knockdown in Clone #3. Three independent target siRNA of ALDHl Al were transfected into Clone #3 cells. Protein was harvested 48 hrs post-transfection and used to determine an ALDHl Al- siRNA specific decrease at protein level in response to siRNA transfection by Western blot, α-tubulin antibody was used to demonstrate even loading between the samples.

Figure 1 1 : (A) Invasion assays of Clone #3 (i) under control conditions (ii) transfected with scrambled siRNA (iii) transfected with ALDHlAl siRNA (1) (iv) transfected with ALDHlAl siRNA (2) (v) transfected with ALDHlAl siRNA (3), 48 hrs post transfection. Magnification, 20Ox. Scale bar, 200μm. (B) Total number of Clone #3 cells invading post ALDHlAl siRNA transfection. Statistics: p < 0.05*, 0.01 **, 0.005*** (unpaired t-test) to scrambled controls

Figure 12: Percentage adhesion of Clone #3 untreated, scrambled and treated with three target ALDHlAl siRNAs to matrigel 48 hrs after transfection. Results are expressed as % adhesion relative to untreated control cells. Data shown is mean ± standard deviation

(n = 3).

Statistics: p < 0.05*, 0.01 **, 0.005*** (unpaired t-test) to scrambled controls

Figure 13: Percentage survival of Clone #3, untreated, scrambled and transfected with three independent ALDHlAl siRNA targets in suspension compared to adherent cells. Data shown is mean ± standard deviation (n = 3). Statistics: p < 0.05*, 0.01**, 0.005*** (unpaired t-test).

Figure 14: Western blot analysis of ALDHlAl cDNA transient transfection in Clone #8. Two time points of 48 hrs and 72 hrs post cDNA transfection were used and β-actin was used as loading control.

Figure 15: Invasion assays of (A) (i) Clone #8 under control conditions (ii) Clone #8 transfected with empty vector (EV) (iii) Clone #8 transfected with ALDHlAl cDNA. Magnification, 20Ox. Scale bar, 200μm. (B) Invasion assay of Clone #8 of total number of cells invading 48 hrs post ALDHlAl cDNA transfection. Statistics: p < 0.05*, 0.01 **, 0.005*** (unpaired t-test) to empty vector control

Figure 16: Percentage survival of Clone #3, untreated, scrambled and transfected with three independent ALDHlAl siRNA targets in 4-HC chemosensitivity assay. Data shown is mean ± standard deviation (n = 3).

Figure 17: Western blot of Clone #3 control, treated with 5 μM ATRA for 48 hrs and after continuous ATRA treatment, β-actin was used as loading control.

Figure 18: (A) Invasion of (i) Clone #3 and (ii) Clone #8 after 8 days continuous exposure to 5 μM ATRA. (B) Total number of cells invading. Statistics: p < 0.05*, 0.01**, 0.005*** (unpaired t-test) to control Figure 19: Morphology of Clone #3 (i) under normal culture conditions (ii) 5 μM ATRA treatment (iii) Clone #8 under normal culture conditioned and (iv) 5μM ATRA treatment. Magnification, 10Ox. Scale bar, 200 μm.

Figure 20: Western blot of siRNA VIM knockdown in Clone #3. Three independent target siRNA of VIM were transfected into Clone #3 cells.

Protein was harvested 48 hrs post-transfection and used to determine a VIM-siRNA specific decrease at protein level in response to siRNA transfection by Western blot, α- tubulin antibody was used to demonstrate even loading between the samples. This is a representative picture of at least 3 independent analyses.

Figure 21: (A) Invasion assays of Clone #3 (i) under control conditions (ii) transfected with scrambled siRNA (iii) transfected with VIM siRNA (1) (iv) transfected with VIM siRNA (2) (v) transfected with VIM siRNA (3). Magnification, 20Ox. Scale bar, 200μm. (B) Invasion assay of Clone #3 of total number of cells invading post siRNA vimentin transfection. Statistics: p < 0.05*, 0.01**, 0.005*** (unpaired t-test) to scrambled control

Figure 22: Percentage adhesion of Clone #3 control, scrambled and treated with three target VIM siRNAs to matrigel. Results are expressed as % adhesion relative to scrambled cells. Data shown is mean ± standard deviation (n = 3). Statistics: p < 0.05*, 0.01**, 0.005*** (unpaired t-test) compared to scrambled control.

Figure 23: Percentage survival of Clone #3, untreated, scrambled and transfected with three independent VIM siRNA targets in suspension compared to adherent cells. Data shown is mean ± standard deviation (n = 3). Statistics: p < 0.05*, 0.01**, 0.005*** (unpaired t-test) compared to scrambled control.

Figure 24: Proliferation assay of Clone #3 control, scrambled and transfected with siRNA VIM (1), (2) and (3). Results graphed as % survival relative to non-treated control (n=3). Figure 25: Morphology of (i) MiaPaCa-2, (ii) Clone #3 and (iii) Clone #8 and after transfection (iv) Clone #3 control (v) Clone #3 scrambled (vi) Clone #3 siRNA kinesin (vii) Clone #3 transfected with VIM (1) (viii) Clone #3 transfected with VIM (2) and (ix) Clone #3 transfected with VIM (3) 48 hours post-transfection. Magnification at 2Ox, scale bar, 200μm.

Figure 26: Western blot of siRNA STIPl knockdown in Clone #3. Three independent target siRNA of STIPl were transfected into Clone #3 cells.

Protein was harvested 48 hrs post-transfection and used to determine a STIPl -siRNA specific decrease at protein level, α-tubulin antibody was used to demonstrate even loading between the samples. This is a representative picture of at least 2 independent analyses.

Figure 27: (A) Invasion assays of Clone #3 (i) under control conditions (ii) transfected with scrambled siRNA (iii) transfected with STIPl siRNA (1) (iv) transfected with STIPl siRNA (2) (v) transfected with STIPl siRNA (3). Magnification, 20Ox. Scale bar,

200μm. (B) Invasion assay of Clone #3 of total number of cells invading post siRNA

STIPl transfection.

Statistics: p < 0.05*, 0.01 **, 0.005*** (unpaired t-test) to scrambled control.

Figure 28: Percentage adhesion of Clone #3 control, scrambled and treated with three target siRNA STIPl to matrigel. Results are expressed as % adhesion relative to untreated cells. Data shown is mean ± standard deviation (n = 3).

Statistics: p < 0.05*, 0.01**, 0.005*** (unpaired t-test) compared to scrambled control.

Figure 29: Percentage survival of Clone #3, untreated, scrambled and transfected with three independent STIPl siRNA targets in suspension compared to adherent cells. Data shown is mean ± standard deviation (n = 3).

Statistics: p < 0.05*, 0.01 **, 0.005*** (unpaired t-test) compared to scrambled control. Figure 30: Proliferation assay of Clone #3 control, scrambled and transfected with siRNA STIPl (1), (2) and (3). Results graphed as percentage survival relative to non-treated control (n=3). Statistics: p < 0.05*, 0.01**, 0.005*** (unpaired t-test) compared to scrambled control.

Figure 31. Western blot validation and 3D spot images of A. GSN, B. NDPK, C. LGALSl, D. ALDHlAl, E. BiP (loading control) in CM#3 and CM#8.

Figure 32 A. Western blot of two independent target siRNA-GSN knockdown in CM#8. Bip antibody was used to demonstrate even loading between the samples. B. Invasion assay pictures and graph of Clone #8 cells with the addition of (i) untreated control CM#8, (ii) scrambled transfected CM#8, (iii) GSN siRNA (1) CM#8 and (iv) GSN siRNA (2) CM#8 («=3). Magnification, 20Ox. Scale bar, 200 μm. Statistics; * < 0.05, ** < 0.01, *** < 0.005.

Figure 33 A. Western blot of GSN expression in MiaPaCa-2, Clone #3 and Clone #8 cells and after 24 hrs grown on matrigel. β-actin, used as loading control. B. Invasion assay of Clone #8 untreated control, scrambled transfected, GSN siRNA (1) and GSN siRNA (2) transfected cells(«=3). Magnification, 20Ox. Scale bar, 200 μm. Statistics; * < 0.05, ** < 0.01, *** < 0.005.

Figure 34 A. Western blot of ALDHlAl knockdown in CM#3 untreated control, scrambled, siRNA ALDHlAl (1), siRNA ALDHlAl (2) and siRNA ALDHlAl (3). B. Invasion assay of Clone #3 with addition of CM#3 media (control), CM#3 media with scrambled, CM#3 treated with ALDH 1 A 1 siRNA ( 1 ), CM#3 treated with ALDH 1 A 1 siRNA (2), CM#3 treated with ALDHlAl siRNA (3) («=3). Magnification, 20Ox. Scale bar, 200μm. Statistics; * < 0.05, ** < 0.01, *** < 0.005.

Figure 35: A. Proteomics analysis (2D DIGE MALDI-TOF MS) of STIPl up-regulation in Clone #3 compared to Clone #8 as shown by 3D spot image and protein expression map (PEM). B. Immunoblot of STIPl expression levels in a panel of pancreatic cancer cell lines corresponding to invasive abilities of cells.

Figure 36: Immunoblotting for STIPl, HSP70, HSP90, AbI, HER2 and AKT after transfection with STIP 1 -siRNA in BxPc-3, Panc-1 and Clone #3.

Figure 37: Invasion assays of (A.) Panc-1 (B.) BxPc-3 48 hrs post transfection with scrambled siRNA and three independent siRNA sequences against STIPl. The total number of invading cells was determined by counting the number of cells per field in 10 random fields, at 200* magnification. The average number of cells per field was then multiplied by a factor of 140 (growth area of membrane/field area viewed at 200* magnification (calibrated using a microscope graticule)).

Figure 38: Immunoblotting for MMP2 after transfection with STIPl-siRNA in, Panc-1 and BxPc-3.

Figure 39: (A.) Invasion assay picture representations of invasion under control conditions and 24 hr treatment with 17 AAG in Panc-1 and BxPc-3 cells. Invasion is significantly reduced. (B.) Percentage survival of panel of 8 pancreatic cancer cell lines treated with 17 AAG chemosensitivity assay.

Figure 40: IHC detection of STIPl (A-D) in pancreatic cancer and normal pancreas tissues. (A) Strong STIPl cytoplasmic staining in PC tumour ducts. (B-C) Strong STIPl expression in poorly differentiated PC tumours. (D) Moderate staining of normal pancreas ducts and acinar cells. Original magnification 200^χ.

Figure 41: Toxicity profiles of pancreatic cancer cell lines (MiaPaCa-2, Panc-1, BxPc-3, Clone#3 and Clone #8), breast cancer cell lines (SKBR3 and T47D) and lung cancer cell line (DLKP) to disulfiram. Figure 42: IHC detection of ALDHl Al (D-H) in pancreatic cancer and normal pancreas tissues. (D) Moderate staining of normal pancreas ducts and acinar cells. (E) ALDHlAl highly expressed in well differentiated PC tumour. (F-G) Weak ALDHlAl staining observed in <10% of poorly differentiated PC tumours. (H) Positive staining in epithelial cells of normal pancreas. Original magnification 200^χ.

Detailed Description of the Invention

Materials and Methods

Cell lines

The human pancreatic cell line MiaPaCa-2 was obtained from the European Collection of

Cell Cultures (ECACC, UK). Clone #3 and Clone #8 were obtained by single cell dilution in this laboratory. Briefly the parental cell line was diluted to a concentration of 3 cells/ml and 100 μl plated onto each well of a 96-well plate. After 24 hours each well was studied for single cells, and allowed to grow into colonies. The colonies were then screened by invasion assay to assess their invasive abilities. Cells were maintained in a humidified atmosphere containing 5 % CO₂ at 37 ⁰C in DMEM supplemented with 5 % FCS (Sigma-Aldrich). Antibiotics were not used in the growth media. All cell lines were free from Mycoplasma as tested with the indirect Hoechst staining method.

Preincubation of cells with matrigel coated flasks

Matrigel (Sigma-Aldrich, UK) was coated onto flasks (1 ml/25 cm²) at a concentration of 1 mg/ml. The coated flasks were then placed at 4 ⁰C overnight. The flasks were placed into an incubator at 37 ⁰C for approximately 2 hrs to allow the matrigel polymerise. The excess media in the flasks was then removed and fresh complete media containing the cell suspension was added. Cells attached to the matrigel on the bottom of the flask and after 24 hrs were removed with 0.5 ml/T25 cm² dispase (BD Biosciences). Dispase is a bacillus derived neutral metaloprotease that recovers cells cultured on matrigel. Collection of pancreatic cancer cell line-conditioned media Clone #3 and Clone #8 monolayers were cultured in Tl 75 cm³ flasks until approximately 60% confluent in culture medium. Cells were then washed X3 with serum free (SF) DMEM and incubated for 1 hr with SF DMEM. Cells were washed X3 again in SF DMEM, then placed in SF DMEM for 72 hrs. At the time of collection, cellular debris was removed by centrifugation and filtration through 0.22 μm filter; aliquots were frozen at -80 °C until analysed. Invasion assays Invasion assays were performed using an adapted method (14). Matrigel was diluted to 1 mg/ml in serum free DMEM. 100 μl of matrigel was placed into each insert (Falcon) (8.0 μm pore size) in a 24 well plate (Costar). The coated inserts were incubated overnight at 4 ⁰C. The following day, the matrigel was allowed polymerise at 37 ⁰C for 1 hr. The inserts were then washed with DMEM, 100 μl of lxl0⁵/100 μl cells in complete DMEM and 100 μl of CM supplemented with 5 % serum was added onto the insert. 250 μl of total DMEM: 250 μl CM supplemented with 5% serum was added to the 24-well. After 24 hours, the inside of the insert was wiped with a wet cotton swab. The under surface was gently rinsed with PBS and stained with 0.25% crystal violet for 10 minutes, rinsed again with sterile water and allowed to dry. Total number of invading cells was calculated by counting 10 random grid fields of insert at 200 x magnification. Experiments were performed in triplicate.

Sample preparation and protein labelling

Three 50 ml of CM#3 and CM#8 (3 biological replicates and two technical replicates/CM of cell line) were concentrated using 10,000 molecular weight cut-off (Millipore); samples were cleaned-up using ready-prep 2D clean-up kit (BioRad). Protein concentration was determined using the BSA protein assay kit (Bio-Rad). CM samples were labelled with N-hydroxy succinimidyl ester-derivatives of the cyanine dyes Cy2, Cy3 and Cy5 (15). Typically, 50 μg of the CM was minimally labelled with 200 pmol of either Cy3 or Cy5 for comparison on the same 2-D gel. Labelling reactions were performed on ice in the dark for 30 min and then quenched with a 50-fold molar excess of free lysine to dye for 10 min on ice. A pool containing equal amounts of all samples was also prepared and labelled with Cy2 to be used as a standard on all gels to aid image matching and cross-gel statistical analysis. The Cy3 and Cy5 reverse labelling reactions (50 μg of each) from each CM sample were mixed and run on the same gels with an equal amount (50 μg) of Cy2-labelled standard

Protein separation by 2-DE and gel imaging

Immobilised 24 cm linear pH gradient (IPG) strips, pH 3-11, were rehydrated in rehydration buffer (7 M urea, 2 M thiourea, 4% CHAPS, 0.5% IPG buffer, 50 mM DTT) overnight, according to manufactures guidelines. IEF was performed using as IPGphor apparatus (GE Healthcare) for 40 kV/h at 20 ⁰C with resistance set at 50 mA. Strips were equilibrated for 20 min in 50 mM Tris-HCL, pH 8.8, 6 M urea, 30% v/v glycerol, 1% w/v SDS containing 65 mM DTT and the for 20 min in the same buffer containing 240 mM iodoacetamide. Equilibrated IPG strips were transferred onto 18 x 20 cm 12.5% uniform polyacrylamide gels poured between low fluorescence glass plates. Strips were overlaid with 0.5% w/v low melting point agarose in running buffer containing bromophenol blue. Gels were run at 2.5 W/gel for 30 min and then 100 W total at 10 ⁰C. All the images were collected on a Typhoon 9400 Variable Mode Imager (GE Healthcare). Statistics and quantification of protein expression were carried out in DeCyder software (GE Healthcare).

Spot digestion and MALDI-TOF analysis

Excision of protein spots, trypsin digestion and protein identification by MS analysis using an Ettan MALDI-TOF Pro (GE Healthcare) was performed. Preparative gels containing 300 μg of protein were fixed in 30% v/v methanol, 7.5% v/v acetic acid overnight and washed in water, and total protein was detected by post-staining with CBB and Deep purple stain (Molecular Probes) for 3 hrs at room temperature. Excess dye was removed by washing twice in water, and gels were imaged using a Typhoon 9400 Variable Mode Imager (GE Healthcare) at the appropriate excitation and emission wavelengths for the stain. The subsequent gel image was imported into the BVA module of DeCyder software and was matched to images generated from DIGE analysis. Spots of interest were selected and confirmed using this software for subsequent picking using an Ettan Spot Picker. Gel plugs were placed into a presil iconised 1.5 ml plastic tube for destaining, desalting and washing steps. The remaining liquid above the gel plugs was removed and sufficient ACN was added in order to cover the gel plugs. Following shrinkage of the gel plugs, ACN was removed and the protein containing gel pieces were rehydrated for 5 min with a minimal volume of 100 mM ammonium bicarbonate. An equal volume of ACN was added, and after 15 min of incubation the solution was removed from the gel plugs and the samples were dried for 30 min using a vacuum centrifuge. Individual gel pieces were then rehydrated in digestion buffer (12.5 ng trypsin per μl of 10% ACN, 40 mM ammonium bicarbonate) to cover the gel pieces. Exhaustive digestion was carried out overnight at 37 ⁰C. After digestion, the samples were centrifuged at 12000 x g for 10 min using a bench top centrifuge. The supernatant was carefully removed from each sample and placed into clean plastic tubes. Samples were stored at -80 ⁰C until analysed by M.S. For spectrometric analysis, mixtures of tryptic peptides from individual samples were desalted using Millipore C-18 Zip-Tips (Millipore) and eluted onto the sample plate with the matrix solution (5 mg/ml CHCA in 50% ACN/0.1% TFA v/v). Mass spectra were recorded using the MALDI-TOF instrument operating in the positive reflectron mode at the following parameters: accelerating voltage 20 kV; and pulsed extraction; on (focus mass 2500). Internal calibration was performed using anti-analysis peaks at m/z 842.50, m/z 221 1.104 and external calibration was performed using Pep4 mix. The mass spectra were analysed using MALDI evaluation software (GE Healthcare), and protein identification was achieved with the PMF Pro-Found search engine.

Western blot

Protein concentrations were determined using the Bio-Rad protein assay (Bio-Rad). 35 μg of protein was separated by 7.5% and 15% SDS-PAGE under reducing conditions. Proteins were transferred to nitrocellulose membrane, efficiency and equal loading of protein was evaluated by Ponceau S staining. Membranes were blocked at 4 ⁰C overnight in TBS (25mM Tris-HCl, pH 7.4, 15OmM NaCl, 2.7mM KCl) containing 5% (w/v) lowfat milk powder. Membranes were probed with monoclonal antibodies, anti-gelsolin (Sigma), anti-nucleotide diphosphate kinase (Abeam), anti-galectin-1 (Abeam) and anti- aldehyde dehydrogenase (Calbiochem) anti-vimentin (Sigma), anti-stress-induced phosphoprotein 1 (Santa Cruz), anti-triosephosphate isomerase (Abeam), anti-GAPDH (Ambion), anti-cytokeratin 18 (Santa Cruz). Integrin anti-βl, anti-α5 and anti-α6 monoclonal antibodies were obtained from Becton Dickinson (BD Biosciences UK) and Chemicon (Europe, UK) respectively. Secondary antibodies, anti-mouse and anti-rabbit were obtained from Sigma. Protein bands were detected with Luminol reagent (Santa Cruz Biotechnology).

Conditioned media siRNA transfection

GSN and ALDHlAl

Cells were transfected with two different GSN siRNA targets (Ambion, #8127, 8031) and three different ALDHlAl siRNA targets (Ambion, #106197, #106196, and #106195) and scrambled siRNA (Ambion, # 17010), using NeoFx transfection reagent (Ambion, AM4511), according to the manufacturer's instructions. Briefly, 3 x 10⁵ pancreatic cancer cells were seeded in six-well plates in complete medium at 70% confluence and after 24 hrs transfected with the indicated siRNA, scrambled siRNA and control. After 32 hrs, the media was removed, washed 3x in SF DMEM and 1 ml of SF DMEM was added onto the cells. The effects of siRNA silencing were analysed on the SF CM after 48 and 72 hrs. SiRNA transfected SF CM was collected, centrifuged and filtered through a 0.22 μm filter. SiRNA transfected SF CM was concentrated using 10,000 molecular weight cut-off concentrators (Millipore); samples were cleaned-up using ready-prep 2D clean-up kit (BioRad) and protein concentration was determined using the BSA protein assay kit (Bio- Rad). Western blot analysis was then carried out to assess efficient transfection. All experiments were repeated in triplicate.

Cell lysate siRNA transfection ALDHlAl, VIM and STIPl Three pre-designed siRNAs were chosen for each of the protein/gene targets and transfected into cells, ALDHlAl (Ambiom, #106197, #106196, and #106195) VIM (Ambion, #1311 1, #138994, #138993) and STIPl (Ambion, #136103, #136104, #18523). For each set of siRNA transfections carried out, a control, non-transfected (NT) cell line and a scrambled (SCR) siRNA transfected control were used. siRNA experiments in 6- well plates were set up using 2 μl NeoFx to transfect 30 nM siRNA in a cell density of 3x10⁵ per well of a 6-well plate. Transfection medium, optimem (Ambion) was removed after 24 hours and replaced with fresh growth medium. The transfected cells were collected for Western blot and assayed for changes in invasion capacity at 48 hours using the in vitro invasion assay.

cDNA trasnfection

The target cell line was trypsinised in 6-well plates and set up at 50-70% confluency. Following incubation overnight at 37⁰C, transfection mixtures of 5μl transfection reagent (lipofectamine) in 125 μl optimem (Ambion) were prepared and combined with 2μg cDNA in 125 μl optimem and incubated for 15 minutes at RT. Control wells were untreated and empty vector controls contained transfection mixture without cDNA in the presence of empty plasmid, pCMV6-XL5 (Origene). The transfection mixture was added directly onto cells in 6-well plate and incubated for a further 48 hrs.

Integrins Two integrin βl (ITGBl) target siRNAs (#109877, #109878 (validated) Ambion Inc.) were used to silence integrin βl expression. Three integrin α5 (ITGA5) target siRNAs (#106728, #1 1 1 1 13, #106729 Ambion Inc.) and two integrin cc6 (ITGA6) target siRNAs (#8146, #103827 (validated) Ambion Inc.) were used to silence the respective target genes. Solutions of siRNA at a final concentration of 30 nM were prepared in OptiMEM (Gibco™). NeoFX solution was prepared in OptiMEM and incubated at RT for 10 minutes. After incubation, an equal volume of neoFX solution was added to each siRNA solution, mixed well and incubated for a further 10 minutes. 100 μl of neoFX/OptiMEM solutions were added into a 6 well plate in duplicate. Mia clone #8 (3 x 10⁵) cells were added onto the siRNA solution. The plates were gently mixed and incubated for 24 hours. The transfection mixture was removed and replaced with fresh medium. Positive control kinesin (Ambion Inc.) was included in each triplicate experiment. Invasion, adhesion and anoikis assays were then carried out 48 hours after transfection, as previously described.

Invasion assays 100 μl of matrigel (1 mg/ml) was placed into each invasion insert (Falcon) (8.0 μm pore size) in a 24 well plate (Costar). The coated inserts were incubated overnight at 4 ⁰C. Matrigel was allowed polymerize at 37 ⁰C for 1 hr, then washed with serum-free DMEM. 100 μl of fresh DMEM containing 5% serum was added to the wells and lxl0⁵/100 μl cells were seeded onto the insert. 500 μl of fresh DMEM with 5% serum was added to the well. After 24 hour incubation, the inside of the insert was wiped with a wet cotton swab. The under surface was gently rinsed with PBS and stained with 0.25% crystal violet for 10 minutes, rinsed again with sterile water and allowed to dry. To determine total number of invading cells, the inserts were then viewed under the microscope and the number of cells per field in 10 random fields were counted at 200^χ magnification. The average number of cells per field was then multiplied by a factor of 140 (growth area of membrane/field area viewed at 200^χ magnification (calibrated using a microscope graticule)). The mean values were obtained from a minimum of three individual experiments and were subjected to /-tests.

SiRNA transfections (STIPl and ALDHlAl)

For each set of siRNA transfections carried out, a non-treated control and a scrambled (SCR) transfection control (Ambion, # 17010) were used. SiRNA experiments were set up using 2 μl NeoFx (Ambion, AM4511), to transfect 30 nM siRNA at a cell density of 3x10⁵ /well/ml of a 6-well plate. Cells were transfected with with three different ALDHlAl siRNAs targeting the ALDHlAl isoform (NM 000689 Exon 11, 12: Sequence GGAACAGUGUGGGUGAAUUtt (sense), Ambion, # 106197 - SEQUENCE ID NO: 11), (NM 000689 Exon 9: Sequence GGAGUGUUGAGCGGGCUAAtt (sense), Ambion, #106196 - SEQUENCE ID NO: 12), and (NM 000689 Exon 4: Sequence GGGCCGU ACAAUACCAAUUtt (sense), Ambion, #106195 - SEQUENCE ID NO: 13) and three different STIPl siRNA sequences (NM 006819 Exon 3, Sequence GGAGGGCUU AAAACACGAGtt (sense), Ambion, #136103 - SEQUENCE ID NO: 8), (NM_006819 Exon 2: Sequence GGAGACUACCAGAAGGCUUtt (sense) Ambion, #18532 - SEQUENCE ID NO: 9) and (NM_006819 Exon7, 8: Sequence GCAUAUGCUCGAAUUGGCTAtt (sense) Ambion, #136104 - SEQUENCE ID NO: 10). Transfection medium was removed after 24 hours and replaced with fresh growth medium. The transfected cells were collected for immunoblot and assayed for changes in invasion capacity at 48 hours using the in vitro invasion assay (as previously described).

Immunoblotting Whole protein was extracted from cell lysates using Ix lysis buffer (50 mM Tris-Cl, 150 mM NaCl, and 0.5% NP-40). Lysates were centrifuged for 10 min at 14,000 rpm at 4° C. Protein concentrations were determined using the Bio-Rad protein assay (Bio-Rad). 35 μg of protein was separated by 7.5% and 15% SDS-PAGE under reducing conditions. Proteins were transferred to nitrocellulose membrane, efficiency and equal loading of protein was visualised by Ponceau S staining. Membranes were blocked at 4 ⁰C overnight in TBS (25mM Tris-HCl, pH 7.4, 15OmM NaCl, 2.7mM KCl) containing 5% (w/v) low fat milk powder. Membranes were probed with monoclonal antibodies, anti-aldehyde dehydrogenase (Abeam), anti-stress-induced phosphoprotein 1 (Santa Cruz) (Abeam), HSP90 (Cell signalling), HSP70 AND AbI (Abeam), HER2, AKT (Calbiochem) and β- actin (Sigma-Aldrich) (loading control). Secondary antibodies, anti-mouse, anti-rabbit and anti-goat were obtained from Sigma. Protein bands were detected with Luminol reagent (Santa Cruz Biotechnology).

IHC Analysis Patients The patient group consisted of 5 consenting patients diagnosed with primary tumours of the pancreas. All patients were treated at St. Vincent's University Hospital (SVUH), Dublin in 2005. IHC studies on tumour-free pancreatic tissue were performed using corresponding non-cancerous tissue. Pathological material was examined on each case by SK. Formalin-fixed paraffin-embedded pancreatic tumour tissue and corresponding normal pancreas was available for all patients. Representative 4-μm sections of tissue block were cut using a microtome, mounted onto poly-1-lysine coated slides and dried overnight at 37 ⁰C. Slides were stored at room temperature until required.

Immunohistochemistry Briefly the slides were immunohistochemically stained using primary antibodies specific for ALDHlAl and STIPl from Abeam. The staining procedure includes an antigen retrieval step consisting of 20-minute incubation in pH 9.0 buffer (TARGET Retrieval, Dako) in a 95°C water bath followed by cooling to room temperature. Staining was performed using an automated staining apparatus for IHC (Autostainer, Dako) according to the manufacturer's guidelines. The slides were counterstained with haematoxylin.

Statistical analysis

Student's t-test was used to identify the difference in mean values between treated and non-treated samples. In siRNA experiments, siRNA scrambled transfected cells were used as control compared to siRNA treated samples. This was to ensure no 'off-target' effects of the transfection procedure. Non-treated controls were used to ensure scrambled siRNA was having no effects and to normalise data. A p value of < 0.05 * was deemed significant, p value < 0.01 ** was deemed more significant, p value < 0.005 *** was deemed highly significant.

Results

Invasion through ECM proteins:

Figure 1 A. illustrates the invasion of MiaPaCa-2 and sub-clones Clone #3 and Clone #8 through matrigel, laminin, fibronectin and collagens type IV and I. MiaPaCa-2 displays moderate invasion through matrigel, while Clone #3 is 2.5 times more invasive (p= 0.001) and Clone #8 is 5 fold less invasive (p=0.00001) than MiaPaCa-2 through matrigel. The invasion of MiaPaCa-2 and Clone #3 is comparable through laminin and fibronectin although Clone #3 displays a slight increase whereas Clone #8 shows a significant decrease of 6.25 and 4 fold (p=0.002, p= 0.008) through laminin and fibronectin, respectively. Clone #3 shows a slight increase in invasion through the collagens type I and IV; Clone #8 shows significantly decreased invasion through the collagens (1.6 and 1.6 fold (p=0.03, p=0.02)). The highest level of invasion was observed through fibronectin for each of the cell lines, while, the lowest level of invasion displayed by the cell lines was through the collagens, type IV and I. Clone #3 also displays significantly increased motility (p=0.00005) whereas the motility of Clone #8 is similar to that of MiaPaCa-2 (Figure 1). Therefore, Clone #3 is a more motile and slightly more invasive cell line than the parent whereas Clone #8 shows similar motility but significantly decreased invasion compared to the parent, MiaPaCa-2.

Adhesion to ECM proteins:

The more invasive cell line, Clone #3, displays significantly decreased adhesion to matrigel (p=0.01), laminin (p=0.02), fibronectin (p=0.01) and collagen type IV (p=0.01) compared to parental line (Figure 1 B.). In contrast a significant increase in adhesion was observed to collagen type I (p=0.003), although the level of adhesion to the collagens was lower than that to fibronectin or laminin.

The less invasive cell line, Clone #8, shows significantly increased adhesion to matrigel (p=0.04) and laminin (p=0.002). Adhesion to fibronectin and collagen type I were also increased, but not significantly and adhesion to collagen type IV was decreased significantly (p=0.001).

Invasion and superinvasion assays

Under control conditions, cells were grown on plastic flasks prior to in vitro invasion assaying in the boyden chamber. However, when grown on matrigel for 24 hours, the invasive status of the cells increased further and a new phenomenon known as superinvasion was observed (9) (Fig. 3 A.). The morphology of the parental cell line, MiaPaCa-2 and its two sub-populations Clone #3 and Clone #8 are different and show diverse growth patterns (Fig 3 B. (i-iii)). Clone #3 appears more disperse and has a more spindle-like phenotype, while Clone #8 appears rounded and grows in colonies. The parent has a mix of these two populations.

Identification of proteins by DIGE analysis MiaPaCa-2, Clone #3 and Clone #8 were grown on matrigel 24 hours prior to protein extraction (triplicate biological replicates), then compared using 2D-DIGE MALDl-TOF and differentially abundant proteins of interest were detected using an internal Cy2 labelled standard (pool of all samples) and DeCyder software. In order to compare and contrast the differential protein expression between the sub-populations, Clone #3 and Clone #8 to MiaPaCa-2 and each other, three comparative analyses were carried out. A total number of 113 protein spots were identified in each comparison in the pH range 3- 11. Biological variation analysis of these spots, showing >1.2-fold change in expression with a t-test score of < 0.05, revealed 56 proteins significantly changed between Clone #8 and MiaPaCa-2, 37 proteins between MiaPaCa-2 and Clone #3 and 60 proteins were changed between Clone #3 and Clone #8.

The same procedure as outlined above was performed on the conditioned media of MaiPaCa-2, Clone #3 and Clone #8. Biological variation analysis of these spots showing greater that 1.2-fold change in expression with a t-test score of < 0.05, revealed 37 proteins which are significantly changed between Clone #8 versus MiaPaCa-2, while 32 proteins changed between MiaPaCa-2 versus Clone #3 and 41 proteins were significantly differentially regulated between Clone #8 versus Clone #3. Protein identification by MALDI-TOF MS had an expectation value of <0.005; all reported identifications in this study had an expectation value of 0.000, indicating a perfect match. Table 2 outlines the most significant proteins identified in cell lysates and conditioned media.

Table 2: List of differentially regulated proteins

Protein name Protein AC

M aPaCa-2 C one #3

Clone #8 MiaPaCa-2 Clone #8

Validation of identified proteins by Western blot analysis and immunofluorescence Clone #3 over-expresses VIM, while Clone #8 shows low expression compared to MiaPaCa-2, as verified by Western blot and immunofluorescence (Figure 5 A.). KRT 18 was highly expressed in Clone #8 compared to MiaPaCa-2 and Clone #3. The protein is consistently down-regulated as the invasion increases in our pancreatic cancer cell line model (Fig 5 B.). Expression of ALDHlAl is higher in Clone #3 than in MiaPaCa-2 and Clone #8 (Fig 6 A.). Figure 6 B. shows STIPl expression is increased in Clone #3 compared to Clone #8 or the parental cell line, which is consistent with proteomic analysis. TPIl was also investigated by Western blot (Fig 6 C); however, the results were not consistent with the 2D-DIGE analysis. This may be due to a number of reasons, possibly due to the presence of multiple isoforms and variants of TPIl. GAPDH is also higher in the less invasive cell line, Clone #8 compared to Clone #3 or the parental cell line, MiaPaCa-2. Expression of GAPDH decreases as the invasion status of the cells increases.

Role for Integrins in invasion and adhesion of pancreatic cancer Table 3:

Protein name Gene Protein AC Expression in Expression Expression in Clone #3 in Clone #8 symbol number MiaPaCa-2

Integrin expression: Expression of integrins βl, α2, α5 and α6, which are associated with adhesion to laminin and fibronectin were examined in the cell lines, by western blotting (Figure 7 A-D). Compared to MiaPaCa-2, Clone #8 showed a higher expression of integrins βl, α2, α5 and integrin α6. Expression of integrins βl, α2, α5 and α6 were lower in Clone #3. Integrin βl knockdown:

The role of integrin βl in the differing invasive and adhesive abilities in the low invasive, high adhesive cell line, Clone #8 was investigated using RNAi. Clone #8 was chosen as it expresses high levels of integrin βl compared to Clone #3 and MiaPaCa-2 (Figure 7 A.). Forty-eight hours following transfection of two independent siRNAs, cells were subjected to invasion, motility, adhesion and anoikis assays as described previously. Protein expression of integrin βl was knocked down by approx. 80% (Figure 7 E.). Integrin βl siRNA transfected into clonal cell line, Clone #8 resulted in significant increase in invasion through matrigel (p=0.005 and p=0.04), although invasion through laminin was not significantly altered. Invasion through fibronectin significantly increased (p=0.04 and p=0.02). Motility of Clone #8 after siRNA βl transfection was also significantly increased (p=0.01 and p=0.03) compared to the scrambled control (Figure 8 A.). A decrease in adhesion to matrigel was observed, but this was not statistically significantly, while adhesion to fibronectin (p=0.03) was also decreased with the integrin βl siRNA treatment (Figure 8 B.). Adhesion to laminin was not altered after transfection with integrin βl siRNA. Anoikis assays were also carried out to investigate whether the knockdown of integrin βl had any effect on the survival of Clone #8 in suspension. Figure 5 C. shows a significant increase in the percentage of cells surviving in suspension, treated with integrin βl siRNA compared to cells treated with scrambled control relative to adherent cells (p=0.01 , p=0.003).

Integrin α5 and α6 knockdown:

Integrin α5 and α6 expression was assessed in Figure 7 (C-D) in the model of pancreatic cancer cell lines. To further evaluate the role of integrins in invasion, adhesion and anoikis, siRNA experiments targeting these integrins were also carried out in Clone #8 cells. Expression of integrin α5 and α6 was reduced (Figure 7 F-G). Transfection of integrin α5 siRNA into Clone #8 resulted in significant increase in invasion through matrigel (p=0.0003, p=0.005 and p=0.005), laminin (p=0.07, p=0.008 and p=0.0002) and fibronectin (p=0.0002, p=0.0001 and p=0.008) compared to the scrambled control. Transfection of siRNA cc6 into Clone #8 resulted in a significant increase in invasion through matrigel (p=0.00009 and p=0.02), with no significant increase in invasion through laminin, but significant increased invasion through fibronectin (p=0.004 and p=0.04) (Figure 9 A.).

A slight decrease in adhesion to matrigel and laminin was observed although not significantly, while significant reduction in adhesion to fibronectin was seen after integrin α5 siRNA treatment (p=0.02, p=0.03, p=0.02). Adhesion to matrigel and fibronectin was not altered with integrin α6 siRNA treatment; however adhesion to laminin was significantly reduced (p=0.08 and p=0.01) (Figure 9 B.).

Anoikis assays were also carried out to investigate the role of integrins α5 and α6 on the survival of Clone #8 in suspension. No significant increase in resistance to anoikis was observed after either integrin α5 and α6 siRNA transfection compared to cells treated with scrambled control relative to adherent cells (Figure 9 C).

Investigation into the role of ALDHlAl in pancreatic cancer cell invasion

Aldehyde dehydrogenase was identified as a protein potentially involved in invasion in our in vitro pancreatic cancer cell line model. The analysis showed that ALDHlAl was 9-fold up-regulated in the more invasive sub-population, Clone #3 compared to the low invasive cell line, Clone #8 (Table 1).

Effect of siRNA transfection of ALDHl Al into Clone #3

The protein aldehyde dehydrogenase IAl was shown to be up-regulated as the invasion status of the cell lines increased, therefore the more invasive cell line, Clone #3 expressed an abundance of the protein. Clone #3 was used for siRNA knockdown and further functional analysis. Figure 10 shows by Western blot, the efficient knockdown of ALDHlAl in three siRNA treated Clone #3 cells compared to non-treated control and siRNA scrambled transfected cells.

Effect of ALDHl Al siRNAon invasion in Clone #3 48 hours post-transfection with ALDHlAl siRNA, invasion assays were performed. The total number of cells invading was reduced in Clone #3 cells transfected with ALDHlAl siRNA. Figure 11 (A) shows representative pictures of invading cells, and (B) highlights the total number of cells invading. ALDHlAl siRNA transfection reduces the invasive capabilities of the cells. Invasion was significantly reduced by 2.2-fold (p=0.0013) with siRNA ALDHlAl (1), 1.8-fold (p=0.00035) with siRNA ALDHlAl (2) and 2.8-fold (p=0.0018) with ALDH 1 A 1 (3) transfection.

Effect of ALDH 1 A 1 siRNA on adhesion in Clone #3

Adhesion assays were also carried out to assess the involvement of ALDHlAl in adhesion to matrigel. Figure 12 shows that ALDHlAl siRNA transfection in Clone #3 cells increased the percentage of adhesion relative to untreated control cells. Adhesion was increased by all three ALDHlAl siRNA targets. Adhesion was significantly increased 27% with siRNA ALDHlAl (1) (p=0.04), 23% with siRNA ALDHlAl (2) (p=0.07) and 32% with siRNA ALDHlAl (3) (p=0.001) to matrigel, compared to scrambled control cells.

Effect of ALDHl Al siRNA on anoikis in Clone #3

To determine the effect of ALDHlAl knockdown on suspension survival in Clone #3, anoikis assays were performed with ALDHlAl siRNAs. Figure 13 shows that anoikis is modestly induced in siRNA ALDHlAl (1) and (2) transfected cells compared to scrambled treated cells. Treatment with ALDHlAl siRNA (3) statistically induces anoikis (p=0.03) compared to the scrambled control. However, no significant difference is observed in anoikis of cells transfected with ALDHlAl siRNA (1) and (2) compared to the scrambled controls.

Effect of ALDHlAl siRNA on proliferation in Clone #3

Proliferation assays were carried out over 5 days after transfection of ALDHlAl siRNAs into Clone #3 cells. Figure 14 displays the percentage survival of transfected cells relative to untreated control. Kinesin was used as a control for efficient transfection. There was no significant difference in proliferation of siRNA ALDHlAl treated cells compared to control cells, therefore loss of ALDHlAl did not affect proliferation in Clone #3 cells. ALDHlAl cDNA transfection in the low invasive cell line, Clone #8 Proteomic data and Western blot validation confirmed that ALDHlAl expression was highest in the invasive sub-population, Clone #3 and lowest in the lesser invasive sub- population, Clone #8. To determine whether increasing ALDHlAl expression in Clone #8 would have an effect on invasion, ALDHlAl cDNA was transfected into Clone #8. Figure 15 displays ALDHlAl expression in Clone #8 control, empty vector and ALDHlAl cDNA transfected cells. Western blot analysis clearly shows that ALDHlAl expression was induced in the ALDHlAl cDNA transfected Clone #8 cells confirming efficient transfection. Two time points, 48 and 72 hrs post transfection were used to assess optimal length of transient transfection.

Evaluation of ALDHl Al cDNA transient transfection of Clone #8 on invasion

The effect of increasing ALDHlAl expression in the poorly invasive Clone #8 was examined. Invasion assay analysis was performed 48 hrs post transfection. Figure 16 (A) shows representative pictures of invading cells, and (B) highlights the total number of invading cells of Clone #8 control (untreated), Clone #8 transfected with empty vector and Clone #8 transfected with ALDHlAl cDNA. ALDHlAl transfection of Clone #8 cells increased the invasive capacities of the cells, compared to empty vector control. Invasion was significantly increased 2.1 -fold (p=0.01) in Clone #8 ALDHlAl cDNA transfected cells compared to Clone #8 cells transfected with empty vector.

Correlation of ALDHlAl expression on drug resistance

Chemosensitivity assays were carried out in the three cell lines, MiaPaCa-2, Clone #3 and

Clone #8. 4-hydroxycyclophosphamide (4-HC) and mafosfamide are known to be detoxified by ALDHlAl . 4-diethylaminobenzaldehyde (DEAB) is a specific inhibitor of ALDHlAl and ALDHlAl converts retinal to retinoic acid. The high invasive cell line, Clone #3 is more resistant to the cytotoxic effects of cyclophosphamide metabolite, 4-HC and analogue, mafosfamide. However, the parental cell line, MiaPaCa-2 and the low invasive cell line, Clone #8 are more sensitive to the drugs (Table 4).

Table 4 IC₅₀ of MiaPaCa-2, Clone #3 and Clone #8

IC₅₀S calculated represent half maximal inhibitory concentration of each drug in MiaPaCa-2, Clone #3 and Clone #8.

Effect of ALDHl Al siRNA on drug resistance in Clone#3

Chemosensitivity assays were performed on Clone #3 cells transfected with ALDHlAl siRNA to determine whether ALDHlAl silencing sensitised the cells to the toxic effects of 4-HC. Figure 17 shows that ALDHlAl siRNA slightly increased sensitivity to 4-HC and may be associated with 4-HC resistance in pancreatic cancer cells. Table 5 outlines the IC₅oS of Clone #3 untreated control, scrambled and transfected with three independent siRNA ALDHlAl .

Table 5 IC₅₀ of Clone #3 control, scrambled and ALDHlAl siRNA treated cells to 4-HC μg/ml Control Scrambled SiRNA KtIMl-I-.! KfKIfEl

ALDHl Al (I) ALDH lAl (2) ALDHlAl (3)

4-HC 1.5 ± 0.5 1.6 ± 0.5 1.1 ± 0.1 1.4 ± 0.2 1.1 ± 0.1

Effect of ATRA treatment in Clone #3 and Clone #8

ALDHlAl acts as a catalyst irreversibly converting retinaldehyde to retinoic acid (RA). Analyses were performed to investigate whether accumulation of intracellular retinoic acid may lead to the suppression of ALDH expression (Moreb et al., 2005). Clone #3 and Clone #8 were incubated with 1.5 μg/ml (5 μM) of ATRA to assess a possible feedback loop of high levels of retinoic acid on ALDHlAl expression. This effect was measured by Western blot, invasion assays and morphological changes in the cells.

Western blot analysis of ALDHlAl expression in Clone #3 and Clone #8 cells incubated with ATRA

ALDHlAl expression was determined by Western blot in Clone #3 cells treated with 5 μM ATRA for 48 hrs and continuous 5 μM ATRA, treatment. Figure 18 shows by Western blot that ALDHlAl expression is not altered in Clone #3 after ATRA treatment.

Assays of Clone #3 and Clone #8 cells incubated with ATRA

Invasion assays were performed in Clone #3 and Clone #8 cells after 48 hrs treatment with 5 μM ATRA. Initial experiments were negative as invasion was not altered in either cell line after ATRA treatment. However, after long-term continuous treatment of 5 μM ATRA for 8 days, Clone #3 displayed unchanged invasion levels, yet invasion of Clone #8 was significantly increased (p=0.03) (Figure 19 (A) and (B)).

Morphology of Clone #3 and Clone #8 cells incubated with ATRA

Figure 20 (i) displays the morphology of control Clone-#3 cells, while (ii) highlights the morphology of these cells after continuous exposure to ATRA. These results show that the morphology is unchanged.

Clone #8 under control conditions is depicted in (iii), while (iv) represents the morphology of the cells after long-term treatment with ATRA. The typical phenotype of the cells is obviously different. The cells are spread out, spindle-shaped and elongated compared to colony-type rounded phenotype of the control cells.

Inhibition of ALDHlAl by Disulfϊram

Disulfiram, an inhibitor of ALDHlAl, decreases invasion and is more toxic to pancreatic cancer cell lines compared to ALDHlAl expressing breast (SKBR3 and T47D) and lung cancer (DLKP) cell lines (Figure 41). In addition, recent. studies have indicated ALDHlAl as a marker for stem cells. These findings demonstrate that ALDHlAl may be involved in important pathways critical for invasion and metastasis, with particular relevance to pancreatic cancer.

Expression of ALDHl Al in pancreatic cancer IHC analysis was performed on pancreatic cancer (PC) tissue (n=5) and corresponding normal pancreas (NP) tissue specimens (n=5). IHC was used to validate the expression patterns of ALDHlAl. Strong ALDHlAl expression was observed in 2/5 PC (Fig 42 E). This may be associated with differentiation status, as strong ALDHlAl expression was exclusive to well differentiated tumours (Table 1). Three moderate-poorly differentiated pancreatic cancer samples exhibited lower levels (<10%) of ALDHlAl (Fig 42 F-G). ALDHlAl staining was also observed in normal pancreas (Fig 42 H), including islet cells.

Role of vimentin in pancreatic cancer cell invasion and epithelial to mesenchymal transition (EMT)

Vimentin (VIM) was identified as a protein involved in invasion in our in vitro pancreatic cancer cell line model. The analysis showed that VIM was 5.5-fold up-regulated in the more invasive sub-clone, Clone #3 compared to the low invasive cell line, Clone #8 (Table 1).

Effect of siRNA transfection of VIM into Clone #3

The protein vimentin was shown to be up-regulated as the invasion status of the cell lines increased, therefore the more invasive cell line, Clone #3 expressed an high levels of the protein. Clone #3 was used for siRNA knockdown and further functional analysis.

Figure 21 shows by Western blot the efficient knockdown of VIM in Clone #3 cells transfected with three VIM siRNAs compared to non-treated control and siRNA scrambled transfected cells.

Effect of VIM siRNA on invasion in Clone #3 48 hours post-transfection with VIM siRNA, invasion assays were performed. The total number of cells invading was reduced in Clone #3 cells transfected with VIM siRNA. Figure 22 (A) shows representative pictures of invading cells, and (B) highlights the total number of cells invading. VIM siRNA transfection reduces the invasive capabilities of the cells, by 4-fold (p=0.00036) with VIM siRNA (1), 6-fold (p=0.00031) with VIM siRNA (2) and 6-fold (p=0.0004) with VIM siRNA (3) transfection. The loss of VIM expression through siRNA knockdown in Clone #3 cells reduces invasion and confirms proteomics results.

Effect of VIM siRNA on adhesion in Clone #3

Adhesion assays were also carried out to assess the involvement of vimentin in adhesion of Clone #3 to matrigel. Figure 23 shows the % adhesion relative to untreated control cells. siRNA VIM (1) transfection increased the percentage of adhesion (35%) relative to cells treated with scrambled control (p=0.08). Percentage adhesion was significantly increased by 25% with VIM (2) and 32% with VIM (3) siRNAs (p=0.02 and p=0.02) compared to scrambled siRNA transfected control cells. Therefore, these results suggest that siRNA VIM (2) and (3) are more efficient inhibitors of VIM protein translation and that VIM expression is involved in adhesion to matrigel in Clone #3 cells.

Effect of VIM siRNA on anoikis in Clone #3

Anoikis assays showed that VIM siRNA (3) caused a slight but statistically significant decrease in anoikis resistance compared to scrambled control (p=0.02) (Figure 24). Sensitivity to anoikis was slightly induced in VIM siRNA (1) and (2) transfected cells but only significant in siRNA VIM (3) compared to the scrambled control, indicating that role of VIM in anoikis is unclear.

Effect of VIM siRNA on proliferation in Clone #3

The effect of silencing VIM protein expression on proliferation was studied. Figure 25 displays the percentage survival of Clone #3 cells transfected with VIM siRNAs. Loss of VIM did not affect proliferation in Clone #3 cells, and VIM is not essential for proliferation of these cells. Effect of VIM siRNA on epithelial to mesenchymal transition (EMT) in Clone #3 cells EMT is characterised by morphological and behavioural changes in cells (Maeda et al., 2005). Investigations into the involvement of VIM, a mesenchymal marker (Leader et al., 1987) were determined. Figure 26 (i-iii) shows the morphological changes of Clone #3 and Clone #8 compared to the parental cell line, MiaPaCa-2 under normal culture conditions. Clone #3 exhibits a more fibroblast phenotype with spindle shaped elongated cells. Clone #8 cells appear rounded and grow in clusters.

Clone #3 transfected with siRNA VIM exhibits a rounded phenotype as observed in Clone #8 cells. The morphological changes of Clone #3 after loss of VIM expression may implicate the role of vimentin in the epithelial to mesenchymal transition.

Role of stress-induced phosphoprotein 1 in pancreatic cancer cell invasion

STIPl was identified as a protein 2-fold up-regulated in the high invasive sub-population Clone #3, compared to the low invasive sub-population, Clone #8 (Table 2). The expression of STIPl increased as the invasion status of the cells increased suggesting it may correlate to invasion in pancreatic cancer.

Effect of siRNA transfection of STIPl in Clone #3

The protein stress induced phosphoprotein 1 was shown to be up-regulated as the invasion status of the cell lines increased, therefore the more invasive cell line, Clone #3 expressed high levels of the protein. Clone #3 was used for siRNA knockdown and further functional analysis.

Figure 27 showed by Western blot the efficient knockdown of STIPl in siRNA treated Clone #3 cells compared to non-treated and scrambled controls.

Effect of siRNA STIPl on invasion in Clone #3 Invasion assays were carried out on Clone #3 cells untreated, treated with scrambled siRNA and three independent siRNA targets for STIPl. Figure 28 displays (A) representative pictures of the level of invasion and (B) the total number of cells invading. STIPl siRNA transfection reduced the invasive capabilities of Clone #3 cells. Invasion was reduced 3-fold (p=0.0002) with STIPl siRNA (1), 2-fold (p=0.0002) with STIPl siRNA (2) and 2-fold (p=0.0003) with STIPl siRNA (3) transfection. These results confirm proteomic analysis suggesting STIPl is involved in invasion of Clone #3 cells.

Effect of siRNA STIPl on adhesion in Clone #3

Figure 29 shows the percentage adhesion to matrigel of STIPl siRNA transfected cells relative to untreated cells. Adhesion was increased by 31% compared to scrambled controls with all STIPl targeted siRNAs. Adhesion was significantly increased with siRNA STIPl (2) and STIPl (3) (p=0.02 and p=0.03). These results suggest that STIPl has a role in adhesion of Clone #3 cells to matrigel.

Effect of siRNA STIPl on anoikis in Clone #3 The percentage survival relative to adherent cells of Clone #3 treated with STIPl siRNAs compared to scrambled controls is displayed in Figure 30. Results show a modest decrease in anoikis resistance with STIPl siRNA transfection; however, this decrease in survival was not significant, relative to scrambled control cells. Therefore, STIPl does not appear to play a role in anoikis in these cells.

Effect of STIPl siRNA on proliferation in Clone #3

Proliferation assays were carried out to assess the role of STIPl on proliferation in Clone #3 cells. Figure 31 shows the percentage survival of Clone #3 cells, scrambled control and transfected with STIPl siRNAs (1), (2) and (3). The percentage survival is statistically significantly reduced in STIPl siRNA (1) and (2) transfected cells. Inhibition of growth is reduced 13% (p=0.04) and 27% (p=0.003). Proliferation was not reduced with siRNA STIPl (3). These results suggest that siRNA (1) and (2) are more efficient at STIPl knockdown and that STIPl may play a role in proliferation of Clone #3 cells.

STIPl is associated with invasive pancreatic carcinoma STIPl is identified by 2D DIGE followed by MALDI-TOF MS as 2-fold up-regulated in Clone #3, a highly invasive sub-population of the human pancreatic cancer cell line MiaPaCa-2, compared to the low invasive Clone #8 (Fig 35 A). STIPl expression in a panel of pancreatic cancer cell lines was also examined. Expression of STIPl corresponded with the invasive status of the cell lines (Fig 35 B).

Role of STIPl as a HSP90/HSP70 chaperone protein

STIPl-siRNA was used to knock down STIPl expression in three invasive human pancreatic cancer cell lines: BxPc-3, Panc-1 and Clone #3. 48 hours post transfection, STIPl expression was reduced in BxPc-3, Panc-1 and Clone #3 (Fig 36). STIPl siRNA did not alter levels of HSP70 or HSP90 expression in BxPc-3, Panc-1 cell lines. The expression of the HSP90 client proteins, HER2 and AbI were significantly reduced with STIPl-siRNA transfection. However, AKT expression was not altered after STIPl- siRNA transfection (Fig 36)

STIPl, as a novel inducer of pancreatic cancer cell invasion

To determine whether loss of STIPl affects the ability of cells to invade in vitro, we compared the ability of these cells to invade through matrigel-coated Boyden chambers post transfection. As siRNA-STIPl (2) and (3) were more effective at STIPl knockdown, we used these siRNAs for further studies. STIPl siRNA transfection reduced the invasion of Panc-1 cells by 50% STIPl-siRNA (2) 0=0-0005) and 50% STIPl-siRNA (3) 0=0.003) (Fig 37 A). Invasion of BxPc-3 transfected cells were reduced by 51% STIPl- siRNA (2) 0=0.002) and 46% STIPl-siRNA (3) 0=0-002) (Fig 37 B).

MMP2 down regulated as a result of STIPl knockdown

We investigated how knockdown of STIPl using siRNA could affect invasion. Previously, secretion of HSP90α into the extracellular matrix surrounding tumour cells was reported to assist the activation of MMP2 as well as contributing to tumour cell invasiveness (Eustace et al., 2004). We show that knockdown of STIPl in both Panc-1 and BxPc-3 cells reduces the expression of MMP2 (Fig 38). Direct targeting of HSP90 inhibitor, 17 AAG

Using the HSP90 inhibitor, (17-AAG) geldanamycin, we also show that that direct inhibition of HSP90 also reduces the invasion of Pane- 1 and BxPc-3. Using a panel of 8 cell lines, we show a broad sensitivity to 17 AAG, a specific inhibitor of HSP90 (Fig 39).

Expression of STIPl in pancreatic cancer tissue

The expression of STIPl in a small subset of pancreatic cancer specimens compared to matched normal pancreas has been identified («=6) (Fig 40).

Analysis of GSN and ALDHlA invasive factors in conditioned media of pancreatic cancer cells

Validation of secreted proteins Two down-regulated and two up-regulated secreted proteins identified in CM#3 compared to CM#8 were validated by Western blot. Down-regulated secreted proteins, GSN (-21 fold) and NDPK (-1.7 fold) were validated by Western blot in our model (Figure 4 A. and B.) and results were consistent with proteomic data. LGALS-I (1.3 fold) and ALDHlAl (21 fold) were up-regulated as observed in fig. 4 C. and D. ALDHlAl secretion was consistent with proteomic data, however, LGALS-I validation was unclear as the small difference in secretion abundance between CM#3 and CM#8. Bip secreted protein was used as loading control.

Invasion inhibitory role of GSN by siRNA in CM#8 Plasma GSN, a down-regulated secreted protein in CM#3 compared to CM#8 was analysed to assess its functional involvement in pancreatic cancer cell invasion. Figure 5 A. highlights the successful knockdown of GSN secretion in CM#8 by two independent siRNA targets relative to siRNA scrambled and controls (untreated) CM#8. The addition of GSN-siRNA treated CM#8 #1 onto Clone #8 significantly increased the invasive abilities of the cells 1.3 fold (p=0.01). GSN-siRNA treated CM#8 #2 onto Clone #8 increased the invasiveness of the cells 1.5 fold, although not significantly (p=0.2) (Figure 5 B.). As the fold change of the invasion of the cells after GSN knockdown was minor, GSN siRNA analysis was carried out on Clone #8 cells and added into the invasion assay. Figure 6 A. shows by Western blot that cytoplasmic GSN is expressed only in the lysates of Clone #8 cells, however after 24 hrs incubation on matrigel (simulating the in vitro invasion assay) the cytoplasmic GSN is further enhanced after cell-matrigel contact. Figure 6 B. shows the total number of cells invading after functional knockdown of cytoplasmic GSN in Clone #8 cells. Invasion was significantly increased 2.6 and 2.4 fold (p=0.02, p=0.05) in both GSN-siRNA #1 and #2 Clone #8 cells compared to scrambled control.

Invasion enhancement role of ALDHlAl by siRNA in CM#3

ALDHlAl, an up-regulated secreted protein in CM#3 compared to CM#8 was knocked down in CM#3 to assess its functional enhancing role in pancreatic cancer cell invasion. Figure 7 A. shows the efficient knock down of ALDHlAl in CM#3 by three independent targets by Western Blot. ALDHl Al -siRNA treated CM#3 was added into the invasion assay of Clone #3. Figure 7 B. highlights representative photographs of invasion inserts and the total number of invading cells, whereby reduction of ALDHlAl expression resulted in a significant decrease in invasive abilities of Clone #3 cells. ALDHlAl siRNA (1) in CM#3 reduced invasion 4.2-fold (p=0.0\), ALDHlAl siRNA (2) in CM#3 decreased invasion 2.7-fold (p=0.003) and ALDHlAl siRNA (3) in CM#3 also significantly reduced the invasive abilities of Clone #3 2.5-fold (p=0.02), compared to the scrambled control.

The invention is not limited to the embodiments hereinbefore described which may be varied in construction and detail without departing from the spirit of the invention.

References

1. Ferlay, J., Autier, P., Boniol, M., Heanue, M., Colombet, M., and Boyle, P. (2007) Ann. Oncol. 18, 581-592 2. Jemal, A., Siegel, R., Ward, E., Murray, T., Xu, J., and Thun, M. J. (2007) CA Cancer. J. Clin. 57, 43-66 3. Faivre, J., Forman, D., Esteve, J., Obradovic, M., and Sant, M. (1998) Eur. J. Cancer

34, 2184-2190

4. Negm, R. S., Verma, M., and Srivastava, S. (2002) Trends MoI. Med. 8, 288-293

5. Kuramitsu, Y., and Nakamura, K. (2006) Proteomics 6, 5650-5661 6. Lundin, J., Roberts, P. J., Kuusela, P., and Haglund, C. (1995) Anticancer Res. 15, 2181-2186

7. Safi, F., Schlosser, W., Falkenreck, S., and Beger, H. G. (1998)

Hepatogastroenterology 45, 253-259

8. Tempero, M. A., Uchida, E., Takasaki, H., Burnett, D. A., Steplewski, Z., and Pour, P. M. (1987) Cancer Res. 47, 5501-5503

9. Goggins, M. (2005) J. Clin. Oncol. 23, 4524-4531

10. Yoshida, A., Rzhetsky, A., Hsu, L. C, and Chang, C. (1998) Eur. J. Biochem. 251,

549-557

11. Goldman, R. D., Khuon, S., Chou, Y. H., Opal, P., and Steinert, P. M. (1996) J. Cell Biol. 134, 971-983

12. Carrigan, P. E., Sikkink, L. A., Smith, D. F., and Ramirez-Alvarado, M. (2006)

Protein ScL 15, 522-532

13. Kumar, N., Tomar, A., Parrill, A. L., and Khurana, S. (2004) J. Biol. Chem. 279,

45036-45046 14. Albini, A., Iwamoto, Y., Kleinman, H. K., Martin, G. R., Aaronson, S. A., Kozlowski, J. M., and McEwan, R. N. (1987) Cancer Res. 47, 3239-3245

15. Alban, A., David, S. O., Bjorkesten, L., Andersson, C, Sloge, E., Lewis, S., and Currie, I. (2003) Proteomics 3, 36-44

Claims

1. A method for the inhibition, prevention or treatment of invasive/metastatic cancer in an individual in need thereof, comprising a step of treating the individual with an agent capable of attenuating the activity of protein selected from the group consisting of: STIPl; and ALDHlAl.

2. A method as claimed in Claim 1 in which the invasive/metastatic cancer is pancreatic cancer.

3. A method as claimed in Claim 1 in which the invasive/metastatic cancer is a metastases selected from the group consisting of: bone metastases; lung metastases; liver metastases; bone marrow metastases; breast metastases; and brain metastases.

4. A method as claimed in any preceding Claim in which the agent is an agent that suppresses the expression of the protein.

5. A method as claimed in any preceding Claim in which the agent is selected from the group consisting of: siRNA; miRNA; shRNA; a ribozyme; and an antisense oligonucelotide.

6. A method as claimed in Claim 3 or 4 in which the protein is STIPl, and the agent is an siRNA molecule selected from the group consisting of: SEQUENCE ID NO:8; SEQUENCE ID NO: 9; and SEQUENCE ID NO: 10.

7. A method as claimed in Claim 3 or 4 in which the protein is ALDHlAl, and the agent is a siRNA molecule selected from the group consisting of: SEQUENCE ID NO: 11; SEQUENCE ID NO: 12; and SEQUENCE ID NO: 13.

8. A method as claimed in any preceding Claim in which the protein is STIPl, and the agent is a HSP90 inhibitor.

9. A method as claimed in Claim 8 in which the HSP90 inhibitor is selected from the group consisting of: geldanamycin; retaspimycin; and small molecule inhibitors of

HSP90.

10. A method as claimed in any preceding Claim in which the protein is ALDHlAl, and the agent is an ALDHlAl inhibitor.

11. A method as claimed in Claim 10 in which the ALDHlAl inhibitor is selected from the group consiting of: Disulfiram; 4-(N, N-dipropylamino)benzaldehyde (DPAB); and 4-(N, N-diethylamino) benzaldehyde (DEAB).

12. A method as claimed in any preceding Claim in which the agent is an antibody that specifically binds to a protein selected from the group consisting of: STIPl; and ALDHlAl.

13. Use of an agent that is capable of attenuating the activity of a protein selected from the group consisting of STIPl and ALDHlAl, as a medicament.

14. Use as claimed in Claim 13 in which the agent is capable of suppressing the expression of the protein, or directly inhibiting the protein.

15. Use of an agent that is capable of attenuating the activity of STIPl or ALDHlAl to inhibit, prevent or treat metastatic cancer.

16. Use as claimed in Claim 15 in which the agent is capable of suppressing the expression of the protein, or directly inhibiting the protein.

17. Use as claimed in Claim 16 in which the agent is a gene knockdown tool selected from the group of: siRNA; miRNA; shRNA; a ribozyme; and an antisense oligonucelotide.

18. Use of an oligonucleotide capable of knocking-down a protein selected from the group consisting of STIPl and ALDHlAl as a medicament.

19. Use of an oligonucleotide capable of knocking-down a protein selected from the group consisting of STIPl and ALDHlAl for the inhibition, prevention or treatment of metastatic cancer.

20. Use as claimed in Claim 18 or 19 in which the oligonucleotide is selected from the group consisting of: siRNA; miRNA; shRNA; a ribozyme; and an antisense oligonucelotide.

21. A pharmaceutical composition comprising an agent that attenuates the activity of a protein selected from the group consisting of STIPl and ALDHlAl , and a suitable carrier or pharmaceutical excipient.

22. A pharmaceutical composition as claimed in Claim 23 in which the agent is an oligonucelotide capable of knocking down the protein.

23. A pharmaceutical composition as claimed in Claim 25 in which the oligonucleotide is selected from the group consisting of: siRNA; miRNA; shRNA; a ribozyme; and an antisense oligonucelotide.

24. A pharmaceutical composition as claimed in Claim 23 in which the agent is selected from the group consisting of: an inhibitor of the protein; and an antibody that specifically binds to the protein.

25. A pharmaceutical composition as claimed in Claim 24 in which the protein is STIPl, and the agent is a HSP90 inhibitor.

26. A pharmaceutical composition as claimed in Claim 24 in which the protein is ALDH 1 A 1 , and the agent is a ALDH 1 A 1 inhibitor.

27. A pharmaceutical composition as claimed in any of Claims 41 to 44, and further including an effective amount of a cytotoxic agent.

28. A pharmaceutical composition as claimed in Claim 24 in which the antibody is a blocking antibody.

29. A pharmaceutical composition as claimed in Claim 24 or 28 in which the antibody is selected from the group consisting of: a humanised antibody; and a fully human antibody.

30. Use of a HSP90 inhibitor for the inhibition, prevention or treatment of an invasive/metastatic cancer.

31. Use as claimed in Claim 30 in which the invasive/metastatic cancer is pancreatic cancer.

32. A method of identifying compounds useful in the inhibition, prevention or treatment of an invasive/metastatic cancer, comprising determining a reference level of activity of a protein, contacting the protein with a candidate compound, and determining the level of activity of the contacted protein, wherein a decrease in the level of activity of the contacted protein relative to the reference level of protein activity is an indication that the candidate compound is useful in the inhibition, prevention or treatment of an invasive/metastatic cancer, wherein the protein is selected from the group consisting of: STIP 1 ; and ALDHlAl.

33. A method as claimed in Claim 32 in which the protein is provided in the form of protein expressing cells, and in which the level of activity is determined by assaying for a level of expression of protein in the cells.

34. A method as claimed in Claim 33 in which the protein is STIP, and the STIP- expressing cells are pancreatic tumour cells.

35. A method as claimed in any of Claims 32 or 33, wherein the invasive metastatic cancer is pancreatic cancer.

36. A method of identifying an agent that suppresses expression of STIPl or ALDHlAl protein comprising the steps of providing a source of STIPl or ALDHlAl expressing cells, treating the cells with a candidate agent, and assaying the cells for expression of STIPl or ALDHlAl, wherein a decrease in the level of expression of STIPl or ALDHlAl protein in the treated cells relative to untreated cells is an indication that the candidate agent is useful in suppressing expression of STIPl or ALDHlAl protein.

37. A method of identifying an agent useful in the inhibition, prevention or treatment of an invasive/metastatic cancer, comprising a step of providing a sample of cells that express HSP90, treating the cells with a candidate agent, and assaying the cells for expression of HSP90, wherein a decrease in the level of expression of HSP90 in the treated cells relative to untreated cells is an indication that the candidate agent is useful in the inhibition, prevention or treatment of an invasive/metastatic cancer.

38. A method of detecting a cancer cell having an invasive/metastatic phenotype, comprising a step of assaying a biological sample from an individual for a level of a biomarker, and correlating the level with invasive/metastatic potential, wherein the biomarker is selected from the group consisting: STIPl ; and ALDHlAl.

39. A method as claimed in Claim 32 in which the step of correlating the level with invasive/metastatic potential involves comparing the level of the protein with a reference level from a cell line having a reference invasiveness/agressiveness.

40. A method as claimed in Claim 32 in which the biological sample is a sample of cells, and wherein the cells are stained the or each biomarker, and wherein the level of staining is correlated with invasive/metastatic phenotype.

41. A method as claimed in Claim 34 in which the cancer cells are pancreatic tumour cells, or cells from a person having pancreatic cancer.

42. A method as claimed in Claim 32 in which the biological sample is a biological fluid, the method comprising a step of determining a level of the biomarker, and comparing the measured level of biomarker with a reference level, wherein a measured level greater than the reference level correlates with the cancer cells having an invasive/metastatic potential, and wherein a measured level less than the reference level correlates with the cancer cells not having an invasive/metastatic potential.