Inhibitors of progastrin-induced repression of ICAT for treating and/or preventing colorectal cancer, adenomatous polyposis or metastasis displaying progastrin- secreting cells and cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active.

FIELD OF THE INVENTION

The present invention relates to methods and compositions for the treatment and the prevention of colorectal cancer, adenomatous polyposis and metastases.

BACKGROUND OF THE INVENTION

Tumorigenesis of the human colon involves in 66 % of the cases somatic mutations of the tumor-supressor gene adenomatous polyposis coli (APC) or in the beta-catenin gene (mean value calculated from the following list of references: (Conlin et al., 2005; De Filippo et al., 2002; Huang et al., 1996; Johnson et al., 2005; Kim et al., 2003; Luchtenborg et al., 2005; Mikami et al., 2006; Morin et al., 1997; Powell et al., 1992; Rowan et al., 2000; Segditsas and Tomlinson, 2006; Shitoh et al., 2004; Smith et al., 2002; Sparks et al., 1998; Suraweera et al., 2006; Takayama et al., 2001)). These mutations are considered as an early event of the colorectal carcinogenesis occurring in patients with sporadic colorectal cancer. Germinal mutations in the ape gene are also responsible for familial polyposis coli, a hereditary syndrome associated with a high risk of intestinal cancer. These mutations result in defectuous regulation of the cytoplasmic pool of the adherens junction protein beta-catenin, resulting in a constitutive activation of the Tcf-4-mediated transcriptional pathway. This constitutive activation results in a high level of transcription of Tcf4 target genes, such as c-myc or cyclin Dl. Another potential target of this pathway is the GAST gene encoding the progastrin prohormone.

Transcription of this gene was shown to be increased after activation of the beta-catenin / Tcf4 complex in vitro (Koh et al., 2000).

A role for progastrin in colon carcinogenesis was first suggested for around 15 years ago, when progastrin was detected in colorectal tumor extracts (Finley et al., 1993;

Kochman et al., 1992; Nemeth et al., 1993), and when plasma levels of progastrin were shown to be elevated in around 75 % of the patients bearing a colorectal tumor, while being undetectable in controls (Ciccotosto et al., 1995; Konturek et al., 2002; Siddheshwar et al., 2001; Van Solinge et al., 1993). Based on these observations, the theory that progastrin could be involved in the colon tumorigenesis has been investigated both in vitro and in vivo. Studies first started by showing that glycine- extended gastrin (one of the products of progastrin maturation) had a positive effect on proliferation (Hollande et al., 1997; Koh et al., 1999; Litvak et al., 1999; Singh et al., 1994; Stepan et al., 1999). Meanwhile, it was shown that overexpression of progastrin in the liver, where its maturation into smaller peptides cannot occur, induces mitogenesis in the intestinal epithelium (Wang et al., 1996). This same transgenic mice were used to demonstrate that progastrin overexpression induces a susceptibility to azoxymethane- induced carcinogenesis (Singh et al., 2000). However, this observation was challenged when it was shown that knocking-down the gastrin gene also induced the same susceptibility to carcinogenesis (Cobb et al., 2002). The proliferative effects of progastrin were however further demonstrated in vitro on the rat intestinal cell line IEC6 (Brown et al., 2003), and in vivo on a transgenic mouse model displaying intestinal overexpression of progastrin under the control of the fabp promoter (Cobb et al., 2004). However, a major disadvantage of this mouse model is that progastrin is overexpressed in the differentiated cells of the intestine epithelium, and not in the proliferative ones. In 2003 and 2005, Ottewell et al provided data on the fact that progastrin stimulates murine colonic epithelial mitosis after DNA damage and on the fact that the COOH-terminal 26- amino acid residues of progastrin are sufficient for stimulation of mitosis in murine colonic epithelium in vivo (Ottewell et al., 2005; Ottewell et al., 2003).

Based on these data coming mostly from overexpression experiments, progastrin was accepted as a « growth factor » on intestinal epithelial cells. In contrast, only three studies have tried to deplete cells in vitro and one in vivo. In vitro, 2 progastrin producing cell lines were transfected with a gastrin gene antisens and it was shown that colony forming was diminished as well as tumor engrafting in nude mice (Singh et al.,

1996). Importantly however, it was not demonstrated to which gastrin gene product these blocking effects were attributed. A more recent study demonstrated that progastrin depletion allowed the restoration of stable adherens and tight junctions in intestinal epithelial cells, linking progastrin with the activation of migration (Hollande et al., 2003). Finally, the last in vitro study was focused on the role of glycine-extended gastrin 17, a maturation product of progastrin, since the cells used in that work secrete large amounts of this peptide but very little progastrin (Hollande et al., 1997). In vivo, the KO mice, as nientionned above, were more suceptible to tumorigenesis.

Importantly, the only two studies reporting a connection between the beta-catenin / Tcf4 complex and the gastrin family of peptides demonstrated in vitro only that activation of the Tcf4 transcriptional pathway (Koh et al., 2000), particularly in correlation with Ras (Chakladar et al., 2005), leads to activation of the gastrin gene promoter and to increased transcription of the gene. In contrast, prior to the data provided here, nothing was known concerning the actual nature of the gastrin gene-derived peptide secreted in the colon upon abnormal activation of the beta-catenin / Tcf4 pathway, no report mentioned the possibility to reverse the activation of this pathway through a blockage of progastrin function, and no correlation had been established in human colorectal cancer patients between those displaying an activation of this pathway within their tumor and those who have elevated expression of the gastrin gene. Consequently, it appears that no prior data provided the sufficient scientific evidence to show that depletion of progastrin could reverse colorectal tumorigenesis induced by a constitutive beta-catenin / Tcf4 activation. Such data are provided below in the description of the invention.

Gastrin and Progastrin

Gastrin is a classical gut peptide hormone, which was identified originally as a stimulant of gastric acid secretion. It is produced principally by the G cells of the Gastric antrum and to a variable extent in the upper small intestine, with much lower amounts in the colon and pancreas. Over recent years, there has been increasing interest in the role of the gastrin family of peptides in colorectal carcinogenesis. In particular, evidence is accumulating that the precursor forms of gastrin (progastrin and glycine-extended

gastrin), which were previously thought to be inactive, play a role in the development of colorectal cancer (see above comments). Gastrin occurs in various molecular forms. The human COOH terminally amidated gastrins, Gl 7 and G34, are generated from a 101- amino acid precursor molecule, preprogastrin, by posttranslational modifications. Preprogastrin is cotranslationally translocated into the endoplasmic reticulum where the signal peptide is rapidly cleaved to give rise to progastrin. Progastrin is subsequently cleaved by prohormone convertases and carboxypeptidase E to give rise to a peptide with a COOH-terminal glycine residue, namely G34-Gly, and an additional cleavage generates the peptide G17-Gly. G34-Gly can be converted to the COOH terminally amidated peptide G34 by peptidyl alpha-amidating monooxygenase and can similarly be cleaved to generate Gl 7. Gl 7 or Gamide is the predominant antral form of gastrin, with nonamidated precursors (progastrin, glycine-extended gastrin) generally comprising less than 10 % of the total secreted peptide in humans. However, in certain clinical situations where processing is impaired, a greater proportion of nonamidated gastrin is secreted. For example, tissue and plasma concentrations of progastrin are elevated in some patients with colorectal carcinoma as mentioned above (Ciccotosto et al., 1995; Konturek et al., 2002; Siddheshwar et al., 2001; Van Solinge et al., 1993).

In WO9919353, a method of treatment of a condition associated with hyperactivity of autocrine stimulation of proliferation of cells of the gastrointestinal mucosa is suggested.

The method comprises the step of administering an effective amount of an antagonist of binding of progastrin to intracellular gastrin/CCK-C receptors or other progastrin receptors to a mammal in need of such treatment.

US6165990 discloses methods for the treatment of colon cancer. The expression of gastrin by colon cancers is inhibited by the use of antisense gastrin expression.

In WO 2004/089976 A, methods and compositions for the treatment of conditions associated with elevated levels of non-amidated gastrin are disclosed. In one aspect the method comprises the step of administering, to a mammal in need of such treatment, an effective amount of a compound which has the ability to inhibit the binding of ferric ions to any one or more of glycine-extended gastrin 17, progastrin or progastrin-derived

peptides, but which does not inhibit the activity of amidated gastrin, and thereby to inhibit the activity of non-amidated gastrins. Data is provided to support the claim that the capacity to inhibit the binding of ferric ions to glycine-extended gastrinl7 does reduce its biological activity, but no data is provided concerning the effects on the biological activity of progastrin.

WO 2006/032980 A aims to protect progastrin-binding molecules that selectively binds progastrin, wherein the molecule does not bind gastrin-17(G17), gastrin-34(G34), glycine-extended gastrin-17(G17-Gly), or glycine-extended gastrin-34(G34-Gly), in particularly monoclonal antibodies selective for progastrin and hybridomas that produce them. A method is disclosed to quantify progastrin levels in biological fluids using the antibodies mentioned above, but this claim is not supported by any data concerning the ability of these antibodies to selectively detect progastrin in any biological fluid. A method of prevention or treatment of a gastrin-promoted disease or condition comprising the administration of antibodies is also disclosed, without data to support the claim that these antibodies demonstrate any selective ability to alter the biological activity of progastrin. This document provides only methods for establishing the specificity of such antibodies.

To date, effective and specific ways of treating and/or preventing colorectal cancer, metastases and adenomatous polyposis are scarce. Therefore, it is an object of the present invention to provide an improved method for treating and/or preventing colorectal cancer, metastasis or adenomatous polyposis.

SUMMARY OF THE INVENTION

In fulfilling this object, there is provided a method for treating and/or preventing colorectal cancer, adenomatous polyposis or metastasis stemming from progastrin- secreting cells, in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active, comprising the step of administering an effective amount of an

inhibitor of progastrin-induced repression of the Inhibitor of beta-catenin and Tcf-4 binding (ICAT) to an individual in need thereof.

The present invention also provides a method for treating and/or preventing colorectal cancer, adenomatous polyposis or metastasis displaying progastrin-secreting cells and cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active, comprising the step of administering an effective amount of an inhibitor of progastrin-induced repression of the Inhibitor of beta-catenin and Tcf-4 binding (ICAT) to an individual in need thereof.

The present invention describes for the very first time that depletion of progastrin, and not any of the other products of the GAST gene, is able to reverse tumorigenesis by acting directly on the constitutive activation of the beta-catenin/Tcf4 activity. This reversal involves the regulation of ICAT expression, low in a context of high progastrin, high in the context of progastrin depletion. When ICAT is high, the constitutive activation of beta-catenin/Tcf4 activity is profoundly decreased due to the fact that ICAT hijacks beta-catenin away from Tcf4.

Therefore, for the first time, it is demonstrated that depletion of progastrin is a very efficient therapy for colorectal cancer, adenomatous polyposis or metastasis stemming from progastrin secreting cells, in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active and for colorectal cancer, adenomatous polyposis or metastasis displaying progastrin-secreting cells and cells in which the beta-catenin/Tcf- 4-mediated transcriptional pathway is constitutively active. The population of those human colorectal tumors, adenomatous polyposis or metastasis having a low ICAT expression, a constitutive beta-catenin/Tcf-4-mediated activity, and a high progastrin expression can thus be selected for anti progastrin therapy.

The present invention also relates to a method for determining if a patient suffering from colorectal cancer, adenomatous polyposis or metastasis is responsive to a therapeutic treatment with an inhibitor of progastrin-induced repression of the Inhibitor of beta-

catenin and Tcf-4 binding (ICAT) comprising the step of determining whether the colorectal cancer, adenomatous polyposis or metastasis stems from progastrin secreting cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active.

The present invention also provides a method which combines the selection of a specific population of patients that are likely to be responsive to the treatment regimen, with the treatment and/or prevention of colorectal cancer, adenomatous polyposis or metastasis.

Also provided is the use of an inhibitor of progastrin-induced repression of the Inhibitor of beta-catenin and Tcf-4 binding (ICAT) in the manufacture of a medicament for treating and/or preventing colorectal cancer, adenomatous polyposis or metastasis stemming from progastrin secreting cells, in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active.

Also provided is the use of an inhibitor of progastrin-induced repression of the Inhibitor of beta-catenin and Tcf-4 binding (ICAT) in the manufacture of a medicament for treating and/or preventing colorectal cancer, adenomatous polyposis or metastasis displaying progastrin-secreting cells and cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active.

In another embodiment of the present invention, there is provided a method for screening for compounds for treating and/or preventing colorectal cancer, adenomatous polyposis or metastasis stemming from progastrin secreting cells, in which the beta- catenin/Tcf-4-mediated transcriptional pathway is constitutively active.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides a method for treating and/or preventing colorectal cancer, adenomatous polyposis or metastasis displaying progastrin-secreting cells and cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active, comprising the step of administering an effective amount of an inhibitor of

progastrin-induced repression of the Inhibitor of beta-catenin and Tcf-4 binding (ICAT) to an individual in need thereof.

Also provided is the use of an inhibitor of progastrin-induced repression of the Inhibitor of beta-catenin and Tcf-4 binding (ICAT) in the manufacture of a medicament for treating and/or preventing colorectal cancer, adenomatous polyposis or metastasis displaying progastrin-secreting cells and cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active.

Typically the progastrin secreting cells and the cells in which the beta-catenin/Tcf-4- mediated transcriptional pathway are colonic cells.

Typically progastrin-secreting cells and cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active may be the same.

The present invention also provides a method for treating and/or preventing colorectal cancer, adenomatous polyposis or metastasis stemming from progastrin secreting cells, in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active, comprising the step of administering an effective amount of an inhibitor of progastrin-induced repression of the Inhibitor of beta-catenin and Tcf-4 binding (ICAT) to an individual in need thereof.

Typically the progastrin secreting cells are colonic cells.

Also provided is the use of an inhibitor of progastrin-induced repression of the Inhibitor of beta-catenin and Tcf-4 binding (ICAT) in the manufacture of a medicament for treating and/or preventing colorectal cancer, adenomatous polyposis or metastasis stemming from progastrin secreting cells, in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active.

Metastases from colorectal cancer are an important cause of death for patients, and they are rarely operated, either because they are difficult to access or because surgical removal would present critical risks for the immediate patient survival. This is notably the case when the metastasic mass has grown in close proximity to a vital artery. The

treatment according to the present invention induces a regression of these metastases and is therefore extremely useful, either by completely ridding the body of the metastases or at least by allowing surgical removal by forcing the metastases to shrink away from the artery. Furthermore down-regulation of progastrin synthesis in colorectal cancer cells restores the adhesive capacity of colorectal tumor cells and leads to a reduction of their migration potential (Hollande et al., 2003). Typically the metastases treated may not be surgically removable. The metastasis may originate from a primary tumor for which the activation status of the beta-catenin/Tcf-4-mediated transcriptional pathway is unknown. Typically the primary tumor from which the metastasis originates is a colon tumor.

By individual, it is meant animal or human being.

By constitutively active beta-catenin/Tcf-4-mediated transcriptional pathway, it is meant a permanent, and unregulated, activation of the pathway leading to the formation of a transcriptional complex between beta-catenin and Tcf-4, leading to an enhanced transcription of Tcf-4 target genes due to a lack of degradation of cytoplasmic beta- catenin.

Examples of cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active, are cells in which the adenomatous polyposis coli (APC) tumor- suppressor gene is mutated or cells in which the beta-catenin gene is mutated in a way which prevents a normal degradation of beta-catenin.

Typically, before applying a method of treatment according to the present invention to a patient suffering from colorectal cancer, adenomatous polyposis or metastasis, a diagnostic test may be performed in order to determine whether the colorectal cancer, adenomatous polyposis or metastasis stems from progastrin secreting cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active or whether the colorectal cancer, adenomatous polyposis or metastasis displays progastrin-secreting cells and cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active. By performing such a pre-treatment diagnostic test, it is possible to

determine whether a patient would be responsive to a method of treatment according to the invention.

It falls within the ability of the skilled artisan to carry out such a diagnostic test. Typically, plasma levels of progastrin in the blood of the patients may be determined in order to treat only those patients having an elevated progastrin level in the plasma. Plasma progastrin survey may also be used to detect an eventual recurrence of the tumor and/or of metastases. Throughout their life, colorectal tumors secrete factors promoting the genesis of new blood vessels (such as VEGF), thereby ensuring that the tumor receives a sufficient amount of nutrients to achieve maximal growth (Wray et al 2004). The presence of these new vessels also allows the tumor to release various factors into the blood stream, and this is notably the case for progastrin. Progastrin-targeted therapy may therefore include the measurement of plasma progastrin levels.

Alternatively or additionally, mutations of the APC gene or of the beta-catenin gene may be detected or the abnormal level of transcription of Tcf target genes, such as for example c-Myc and cyclin Dl, may be determined by staining for these markers on tissue slides taken from the patient or by using microarray technologies for example.

The present invention also relates to a method for determining if a patient suffering from colorectal cancer, adenomatous polyposis or metastasis is responsive to a therapeutic treatment with an inhibitor of progastrin-induced repression of the Inhibitor of beta- catenin and Tcf-4 binding (ICAT) comprising the step of determining whether the colorectal cancer, adenomatous polyposis or metastasis stems from progastrin secreting cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active or whether the colorectal cancer, adenomatous polyposis or metastasis displays progastrin-secreting cells and cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active.

Typically the step of determining whether the colorectal cancer, adenomatous polyposis or metastasis stems from progastrin secreting cells in which the beta-catenin/Tcf-4-

mediated transcriptional pathway is constitutively active or whether the colorectal cancer, adenomatous polyposis or metastasis displays progastrin-secreting cells and cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active, comprises the step of measuring plasma levels of progastrin of said patient. Additionally or alternatively the step of determining whether the colorectal cancer, adenomatous polyposis or metastasis stems from progastrin secreting cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active or whether the colorectal cancer, adenomatous polyposis or metastasis displays progastrin-secreting cells and cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active, comprises the step of detecting a mutation of the APC gene or of the beta-catenin gene or abnormal level of transcription of Tcf target genes.

Typically, the progastrin assay may be a time-resolved immunofluorimetric assay (TR- IFMA), as described for proGRP (Nordlund et al, 2007). This type of assay allows quantification in the range of 16 ng/1 to 20,000 ng/1, values in the range of progastrin plasma levels measured in patients bearing a colorectal tumor. Two monoclonal antibodies may be used, one directed against the N-terminus of progastrin and the other against the C-terminus. The N-ter antibody may be biotinylated and used as the solid- phase antibody. The C-ter antibody may be labeled with Eu ³⁺ and used as the tracer antibody.

Examples of inhibitors of progastrin-induced repression of ICAT may be selected within the group consisting of agent downregulating the expression of progastrin in colonic cells, antibody against progastrin, inhibitor of phosphatidylinositol 3 kinase (PI3K) and an inhibitor of Integrin-linked kinase (ILK).

In an embodiment of the present invention, the inhibitor of progastrin-induced repression of ICAT is an agent downregulating the expression of progastrin in colonic cells. Typically, agent downregulating the expression of progastrin in colonic cells comprises a nucleic acid which interferes with the expression of progastrin.

Examples of such agents are antisense molecules or vectors comprising said antisense molecules. Antisense molecules are complementary strands of small segments of mRNA. Methods for designing effective antisense molecules being well known (see for example US6165990), it falls within the ability of the skilled artisan to design antisense molecules able to downregulate the expression of progastrin in colonic cells. Further examples of agent downregulating the expression of progastrin are RNA interference (RNAi) molecules such as, for example, short interfering RNAs (siRNAs) and short hairpin RNAs (shRNAs). RNAi refers to the introduction of homologous double stranded RNA to specifically target a gene's product, in the present case progastrin, resulting in a null or hypomorphic phenotype. Methods for designing effective RNAi molecules being well known (see for review Harmon and Rossi Nature. 2004 Sep 16;431(7006):371-8), it falls within the ability of the skilled artisan to design RNAi molecules able to downregulate the expression of progastrin in colonic cells.

Examples of siRNAs able to downregulate the expression of progastrin in colonic cells are nucleic acid molecules which comprise one of the following sequences: siRNAs hPG:

Sense: 5'-GAAGAAGCCUAUGGAUGGATT-S ' (SEQ ID N°27) Antisense: 5'-UCCAUCCAUAGGCUUCUUCTT-S' (SEQ ID N°28) siRNAs mPG:

Sense: 5'-GAAGAGGCCUACGGAUGGTT-S' (SEQ ID N°29)

Antisense: 5'-CCAUCCGUAGGCCUCUUCTT-S' (SEQ ID N°30)

Examples of shRNAs able to downregulate the expression of progastrin in colonic cells are nucleic acid molecules which comprise one of the following sequences: shRNA hPG:

Sense:5'GAAGAAGCCTATGGATGGATTCAAGAGAAGGTAGGTATCCGAAGAA GTTTTTT3' (SEQ ID N°l) Antisense: 5'AATTAAAAAACTTCTTCGGATACCTACCTTCTCTTGAATCCATCCATAGGC TTCTTCGGCC 3 ' (SEQ ID N°2)

One embodiment of the present invention relates to a medicament comprising a siRNA comprising one of the sequences selected from the group consisting of SEQ ID N°27, SEQ ID N°28, SEQ ID N°29 and SEQ ID N°30. Typically the medicament may comprise a pair of siRNAs. Examples of pairs of siRNAs are a first nucleic acid comprising SEQ ID N°27 and second nucleic acid comprising SEQ ID N°28 or a first nucleic acid comprising SEQ ID N°29 and second nucleic acid comprising SEQ ID N°30.

One embodiment of the present invention relates to a medicament comprising a shRNA comprising one of the sequences selected from the group consisting of SEQ ID N ⁰ I and SEQ ID N°2. Typically the medicament may comprise a pair of shRNAs. Examples of pairs of shRNAs are a first nucleic acid comprising SEQ ID N ⁰ I and second nucleic acid comprising SEQ ID N°2.

In a further embodiment of the invention, the inhibitor of progastrin-induced repression of ICAT is an antibody against progastrin or a biologically active fragment or derivative thereof, which does not recognize amidated and glycine-extended forms of gastrin. The person skilled in the art will be aware of standard methods for production of such specific antibody. For example, specific antibodies or biologically active fragments or derivatives thereof may be generated by immunizing an animal with progastrin or progastrin fragments and by selecting the antibodies which bind to progastrin and which do not recognize amidated and glycine-extended forms of gastrin. In a preferred embodiment, the progastrin fragments are amino_acid sequences which are specific to progastrin. The length of these sequences may be comprised between 8 and 15 amino acids. These sequences may comprise COOH-terminal amino acid residues or NH ₂ - terminal amino acid residues of progastrin which are not present in the gastrin forms Gl 7 and G34-Gly. These sequences may comprise the cleavage site of prohormone convertase or of carboxypeptidase E. In one embodiment of the present invention the antibody or biologically active fragment or derivative thereof, binds to the COOH-terminal 10-amino acid residues of progastrin or to the COOH-terminal 15-amino acid residues of progastrin. An example of COOH-

terminal 10-amino acid residues of progastrin is FGRRSAEDEN (SEQ ID N°31). In an alternative embodiment, the antibody biologically active fragment or derivative thereof binds to the NH ₂ -terminal 40-amino acid residues of progastrin. Typically the antibody biologically active fragment or derivative thereof may bind to SWKPRSQQPD APLGT (SEQ ID N°32). These amino_acid sequences are specific to progastrin as they are not present in the gastrin forms Gl 7, G17-gly, G34 and G34-Gly.

The person skilled in the art will be aware of standard methods for production of both polyclonal and monoclonal antibodies and biologically active fragments and derivatives thereof. By biologically active fragment or derivative thereof, it is meant able to bind to the same epitope as the antibody, in the present case an epitope of progastrin. Antibody fragments, particularly Fab fragments and other fragments which retain epitope-binding capacity and specificity are also well known, as are chimeric antibodies, and "humanized" antibodies, in which structural (not determining specificity for antigen) regions of the antibody are replaced with analogous or similar regions from another species. Thus antibodies generated in mice can be "humanized" to reduce negative effects which may occur upon administration to human subjects. Chimeric antibodies are now accepted therapeutic modalities with several now on the market. The present invention therefore comprehends use of antibody specific for progastrin which include F(ab') ₂ , F(ab) ₂ , Fab, Fv and Fd antibody fragments, chimeric antibodies in which one or more regions have been replaced by homologous human or non-human portions. The person skilled in the art will also be aware that biologically-active antibody derivatives such as for example ScFv fragments and divalent ScFv-type molecules can be prepared using recombinant methods. The antibody may be labelled with a detectable marker, which is suitably a radioactive label, such as radioactive iodine, or may be a fluorescent or chemiluminescent label. A person skilled in the art will be able to select suitable radioactive, fluorescent or chemiluminescent labels.

One embodiment of the present invention relates to a medicament comprising an antibody or biologically active fragment or derivative thereof, which binds to the

COOH-terminal 15-amino acid residues, or to the NH ₂ -terminal 40-amino acid residues of progastrin.

Typically antibody or biologically active fragment or derivative thereof binds to FGRRSAEDEN (SEQ ID N°31) or to SWKPRS QQPDAPLGT (SEQ ID N°32).

In healthy humans, progastrin comprise less than 10% of circulating gastrins and no physiological role has been associated with progastrin. Consequently by specifically targeting progastrin with antibodies or biologically active fragments or derivatives thereof, side effects of the treatment are diminished if not avoided.

In a further embodiment of the present invention, the inhibitor of progastrin-induced repression of ICAT is an inhibitor of PBK. Inhibitors of PBK are well known. LY294002, wortmannin and quercetin are commonly used inhibitors of PBK. US2006058321 and WO2004052373, for example disclose families of inhibitors of PBK. Antisense, RNAi molecules, dominant-negative forms of PBK, antibodies against PBK, biologically active fragments and derivatives thereof may also be used as inhibitors of PBK.

In a further embodiment of the present invention, the inhibitor of progastrin-induced repression of ICAT is an inhibitor of ILK. Inhibitors of ILK are well known. KP-392 and QLT-0267 are commonly used inhibitors of ILK. For example, small molecule inhibitors of ILK are described in US6214813. Antisense inhibitors of ILK are described in US6177273. RNAi molecules, dominant-negative forms of ILK and antibodies against ILK, biologically active fragments and derivatives thereof may also be used as inhibitors of ILK.

Typically medicaments according to the invention comprise an inhibitor of progastrin- induced repression of the Inhibitor of beta-catenin and Tcf-4 binding (ICAT), together with a pharmaceutically-acceptable carrier. A person skilled in the art will be aware of suitable carriers. Suitable formulations for administration by any desired route may be

prepared by standard methods, for example by reference to well-known text such as Remington; The Science and Practice of Pharmacy.

In another embodiment of the present invention, there is provided a method for screening for compounds for treating and/or preventing colorectal cancer, adenomatous polyposis or metastasis stemming from progastrin secreting cells, in which the beta- catenin/Tcf-4-mediated transcriptional pathway is constitutively active, comprising the following steps of: a) providing progastrin secreting cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is constitutively active; and b) adding the compound to be screened; and c) selecting the compound which induces ICAT expression.

In an alternative embodiment of the present invention, there is provided a method for screening for compounds for treating and/or preventing colorectal cancer, adenomatous polyposis or metastasis stemming from progastrin secreting cells, in which the beta- catenin/Tcf-4-mediated transcriptional pathway is constitutively active, comprising the following steps of: a) providing progastrin sensitive cells; and b) adding progastrin to the medium containing the cells; and c) adding the compound to be screened; and d) selecting the compound which induces ICAT expression.

Typically the progastrin secreting cells or progastrin sensitive cells are colonic cells.

By progastrin sensitive cells, it is meant cells in which the beta-catenin/Tcf-4-mediated transcriptional pathway is active when progastrin is present in the culture medium.

The person skilled in the art will be aware of standard methods for implementing these methods. Typically ICAT expression may be determined by RT-quantitative PCR.

In the following, the invention will be illustrated by means of the following examples as well as the figures.

FIGURE LEGENDS Figure 1. Progastrin depletion decreases tumorigenicity of human CRC cell lines in Balbc/ nude mice and inhibits spontaneous tumor development in the intestine of APC δ14 mice. (A) Histogram shows GASTiήRNA quantification in SW480/βgal ^w and SW480/G^S7^ (clones [1] and [2], expressing or not a shRNA-insensitive preprogastrin cDNA (cPG)) cells. Results are expressed as a percentage of levels found in SW480/βgal ^(") (*, p<0.05 compared to Sψ4S0/ GAS^, n=3). Bottom table, RIA quantification (pmol/l/24h) of progastrin, glycine-extended gastrin and amidated gastrin secreted by SW480/βgal ^(") and SW480/G ^! ^5'7 ^f; cells. (B) Evolution of tumor volume over time after subcutaneous injection of DLD-l/VO, DLD-1/ASG, SW480/ GAS^ or SW480/βgal ^(") cells in BALB/c nude mice. Means +/- s.e.m of 2 clones/cell line (4 mice/clone), (*, p<0.05 compared to DLD-l/VO or SW480^gal ^w cells, Student's t- test). (C-D) 3-month old APC δ14 mice were treated for two weeks with siRNA targetting either the murine Gast or the Luciferase gene (9 mice/group). Results are expressed as a ratio between Gast gene expression in tumor/healthy mucosa in the ileum (C) or colon (D) (left panels), quantified using RT-QPCR. The number and size of adenomas were quantified after methylene blue staining in the ileum (C) and in the colon (D), and are expressed as total number of tumors for each size group (right panels). GAST siRNA failed to bring down levels of Gast gene expression in sample A (C), collected in a mouse with a high number of intestinal tumors. "B" and "C" correspond to mice that did not bear any colon or ileal tumor, respectively (D).

Figure 2. Progastrin stimulates β-catenin/Tcf-4 activity in human CRC cells (A)

Tcf-4 transcriptional activity was quantified using the TOP/FOP Luciferase reporter gene assay(Morin et al., 1997) in untreated SW480/βgal ^(") and Sψ^O/GASf^ cells

(clones [1] and [2]) (black bars), or in the same cells treated with 5nM recombinant progastrin for 72h (rPG, light grey bars) or transfected with a shRNA-insensitive

preprogastrin cDNA (cPG, dark grey bars). Value are means ± s.e.m. of four similar experiments (*, p<0.05 compared to SW480/GL4£T ^r~; cells). Levels of Tcf-4 and dephosphorylated β-catenin were similar in all of these cells, as shown in Figure S2. (B) Independent and overlaid confocal images of β-catenin (green), Tcf-4 (red), and DAPI (blue) in SW480/βgal ^w cells and SW480/G-4S7* ^y cells with or without 5nM progastrin treatment for 72h. Bars represent 40μm. Higher magnification is used for progastrin- treated cells to provide better visualization of the partial reversion. (C) SW480/G4>S7^ clones [1] and [2] were treated or not as in A. Expression levels of actin and proteins encoded by the Tcf-4 target gene c-Myc, cyclin-Dl, Sox-9 and CLDNl were quantified by immunoblotting, using SW480/βgal^ cells as controls (see also mRNA expressions in supplemental Fig. S3). Values (standardized using actin as a loading control) are mean ± s.e.m. from three independent experiments. Student's t-test, p<0.05 (*) compared to SW480/G_4_S'7 ^fy cells.

Figure 3. Progastrin down-regulates ICAT expression in vitro and in vivo. (A-B)

Progastrin-depleted SW480/ 'GAS^ cells (clones [1] and [2]) were transfected or not with a codon-optimized, shRNA insensitive preprogastrin cDNA (cPG). ICAT mRNA levels were then quantified by RT-QPCR (A) in comparison with SW480/βgal ^("} cells, and expression of ICAT protein was analyzed by immunoblotting (B) in SW4S0/GAS^ before (light grey bars) or after transient transfection with cPG (grey bars), in comparison with SW480/β-gal ^{(") (} black bars). (C) ICAT expression was analyzed by immunohistochemistry and immunofluorescent staining in tissue sections obtained from the colonic epithelium of mice displaying intestinal-specific overexpression of progastrin (Tg/Tg) (Cobb et al., 2004), as well as on control mice from the same genetic background (Wild-type). Bars represent 40μm.

Figure 4. De novo ICAT expression is responsible for the decreased β-catenin-Tcf-4 activity in progastrin-depleted CRC cells. (A) Top panel: dephosphorylated β-catenin immunoprecipitates from SW480/βgal ^(") and SW480/GL4>S7^ +/- recombinant progastrin (rPG, 5nM, 72h) were probed for dephosphorylated β-catenin, Tcf-4 and ICAT. Bottom

panel, quantification from 3 similar experiments performed on 2 independent clones (## p<0.01 and *p<0.05 compared to SW480/βgal ^(") and SW480/ 'GASI* ^' *, respectively, Student's t-test). (B) Independent and overlaid confocal images of β-catenin, ICAT, and DAPI in SW480/βgal ^w cells and Sψ480/GASΫ ^') cells treated or not with 5nM progastrin for 72h. Bars represent 40μm. (C-D) Tcf-4 luciferase reporter assay (TOP/FOP) (C) and quantification of c-Myc (top) and cyclin Dl (bottom) mRNA and protein (D) were performed in SW480/ GAS^ /Luc^ and SW480/G,4,S7 ^f; /ICAT ⁽ - ⁾ cells with or without re-expression of a shRNA-insensitive ICAT cDNA (+cICAT).

Figure 5. ICAT expression is inversely correlated to progastrin levels and β- catenin-Tcf-4 target gene expression in murine and human tumors. (A). 3-month old APCδ14 mice were treated for two weeks with siRNA targetting either the murine GAST gene or the Luciferase gene, as in Figure 1. Intestinal adenomas were processed for RNA extraction and ICAT mRNA expression was quantified (Ad) relative to the expression in matching epithelium (Ep) from the same animals (n=5). (B) Expression of the Tcf-4 targets cyclin Dl, CD44 and C-Myc was analyzed using western immunoblotting in adenomas (Ad) and macroscopically intact epithelium (Ep) from the same animals, in comparison with intestinal epithelium of control mice from the same genetic background (see quantification on supplementary Figure S5). (C) Adenomas from APCδ14 animals treated with Luc or G^ST-specific siRNA were paraffin- embedded and processed for immunohistochemical detection of c-Myc and ICAT. Representative images are provided from GASI*^ and Luc ^~ adenomas collected in the ileum. Bars represent 20μm. (D) GAST and ICATmKNA expressions in microdissected tumors (black bars) from 23 patients with CRC, expressed relative to the amounts found in their respective matching healthy epithelium (white bars), which were normalized to 1 for each patient. Values are means ± s.e.m of three independent experiments. Insert: log values of GAST and ICAT expression were plotted against each other, with the regression curve and its confidence interval. Spearman's correlation coefficient (r) is provided with its degree of significance (p). (E) Top panel: progastrin immunoblotting (IB) in healthy (H) or tumor (T) samples from 4 patients with CRC. Bottom panel:

representative immunohistochemistry (IHC) for ICAT, c-Myc, claudin-1, and CD44 patients with either high (patient 4) or low (patient 22) tumor progastrin expression (positive control in WB was recombinant progastrin (rPG); exposure time was 20 sec for positive control, and 10 min for human samples). Similar IB and IHC results were obtained on samples from the 11 patients tested. Bar represents 20 μm.

Figure 6. Repression of ICAT is essential to the tumor-promoting role of progastrin in human CRC cells. (A) Top panel: Down-regulation of ICAT reverses the decreased growth-rate of progastrin-depleted cells in soft agar, as shown by the marked differences in colony size between SW480/βgal ^θ , SW480/ 'GAS^, SW480/G^57 ^f"; /ICAT ^(") and SW480/ 'GAS^ '/Luc^ cells. Bottom panel: quantification of 3 similar experiments, expressed as a mean ± s.e.m. from ten randomly chosen fields per clone. (B) SW480/G^57 ^f"; /Luc ⁽ - ⁾ and SW480/GJ,S7 ^f"; /ICAT ⁽ - ⁾ cells were injected in the posterior leg of BALB/c nude mice, and tumor growth was measured regularly. Mean ± s.e.m. of 4 clones each. *, p<0.05 compared to SW480/(24,S'7 ^f; /Luc ^(") cells, Student's t-test.

Figure 7 Activation of the PI3MLK pathway is essential to the repression of ICAT by progastrin (A) PBk activation is essential for progastrin-induced ICAT repression in human CRC cells. ICAT mRNA expression was measured as described before (Left), and phosphorylation of Akt/PKB was analyzed by Western blotting (Right) in SW480/βgal ^("} and SψλSO/GAS^ cells with or without treatment with 5nM progastrin (PG) and/or the PBkinase inhibitor LY-294002 (LY), as indicated. (B) ILK expression and activity are reduced in progastrin-depleted human CRC cells. Top panel, ILK expression (Western blot) and enzymatic activity (in vitro phosphorylation of Myelin Basic Protein (MBP)) were analyzed after ILK immunoprecipitation from SW480/βgal ^(") and cell lysates, with or without treatment with recombinant progastrin (5nM, 72h). Middle panel shows the quantification of ILK activity, performed on three independent experiments after correcting for variations of ILK expression in cells described above. In the bottom panel, phosphorylation of the ILK target βl-Integrin was analyzed by sequentially probing βl-Integrin-immunoprecipitates with an anti-

phosphoserine antibody and an antibody against βl-Integrin. ILK, βl-Integrin, and MBP phosphorylation were never detected when lysates were immunoprecipitated with preimmune rabbit serum (not shown). (C) ILK activation is essential for progastrin- mediated ICAT repression in human CRC cells. Expression of ICAT mRNA (Left panel) and phosphorylation of Akt/PKB on Ser473 (Right panel) were quantified in SW480/βgal ^(") and SW480/ 'GAST* ^' * cells transfected or not with a dominant negative ILK (δN-ILK), and with or without treatment with 5nM exogenous progastrin (PG) as indicated. For all panels, p<0.05 compared to SW480/βgal ^w (#), SW480/G,4S7 ^f; (*), or progastrin-treated SW480/G.4S7 ^γy (°) cells.

Figure 8. Progastrin depletion induces differentiation and apoptosis of human CRC cells in vitro and promotes tumor differentiation in the intestine of APC δ14 mice in vivo. (A) Representative experiment showing the nuclear localization of PTEN in two clones of Sψ480/GAST ^f') but not in SW480/βgal ^(") cells. (B) Quantification of Muc-2, intestinal alkaline phosphatase (ALP), and chromogranin A (CgA) mRNA expression in SW480 cells 48h after transfection with siRNAs targetting GAST or β-galactosidase. (C) Muc-2 expression and caspase-3 activation in SW480 CRC cells after transient transfection with GvlST-specific but not β-galactosidase-specific siRNAs. Cells were transfected with rhodamine-tagged siRNAs (red) targetting GAST (arrows) or β- galactosidase (arrowheads), and treated or not with 5nM recombinant progastrin. Muc-2 expression (A488, green) and activated caspase-3 (Cy-5, shown here in red) were detected by immunofluorescent staining, and cell nuclei were stained with DAPI (blue). Similar results were obtained with DLD-I cells. Bars represent 20μm. (D). 3- month old APC δ14 mice were treated for two weeks with siRNA targetting either the murine GAST gene or the Luciferase gene, as in Figure 1. Intestinal adenomas were paraffin-embedded and processed for immunohistochemical detection of mucus- producing cells (Muc2 and Alcian blue) and cells engaged in the apoptotic pathway (activated caspase 3, black arrowheads). Representative images are provided from GAST- ^" * and Luc ^~ adenomas collected in the ileum. Bars represent 20μm.

Figure 9. Signaling pathway connecting progastrin depletion with a decreased activation of the β-catenin/Tcf-4 transcriptional complex in APC-mutated colorectal carcinoma cells. (A). CRC cells produce and secrete endogenous progastrin, which act in turn on tumor cells to activate PI3 kinase and ILK, thereby inducing a repression of ICAT and facilitating the amplification of β-catenin/Tcf-4 transcriptional activity initiated by the APC mutation. This process results in maximal transcription of target genes such as the gene encoding progastrin, thus creating a self-amplifying activation loop. (B) Inhibition of progastrin action, such as that induced by siRNA silencing of the GAST gene (©), markedly reduces the activity of PB kinase and ILK (©), resulting in a strong up-regulation of ICAT (® ). Competition of this small inhibitor with Tcf-4 for binding to β-catenin is sufficient to strongly down-regulate transcription of Tcf-4 targets in tumor cells (®).

Figure 10. Increasing concentrations of antigen was coated onto 96-well plates as indicated (for details, cf. "Methods" section), followed by incubation with selective antibodies against the N-terminal (top panel) or C-terminal (bottom panel) extremities of progastrin(l-80). Antigens tested were the original immunogens used to generate the antibodies (Top panel: SWKPRSQQPDAPLGT, bottom panel: FGRRSAEDEN), amidated gastrin 17, glycine-extended gastrin 17, the C-terminal flanking peptide of progastrin (SAEDEN), and Keyhole Limpet Hemocyanin (KLH), which was used as a carrier protein during the immunization process. Results are expressed as direct 492nm readings after incubation with secondary antibodies and OPD substrate.

Figure 11. SW480 colorectal carcinoma cells were treated for 30 hours with a control antibody or antibodies directed against the C-terminal or N-terminal extremities of progastrin (1/5,000 dilution), as described in the Methods section. After cell lysis, expression of the mRNA for ICAT, c-Myc and cylin Dl was quantified using RT-qPCR. Results are expressed as ratio between the expression in cells treated with C-terminal (grey bars) or N-terminal (white bars) progastrin antibodies and cells treated with the

control antibody, in which each gene expression level was fixed at a value of 1 (represented here by the horizontal axis).

Figure Sl. Overexpression of progastrin in the adenomas of APC. δ14 mice and lack of proinflammatory response after a two-week treatment with siRNA selective for the murine GAST gene. (A) Amidated and glycine-extended gastrin levels in the healthy intestinal mucosa and the adenomas of 3.5-months old APC δ14 mice, quantified by radioimmunoassay. (B) Representative progastrin immunoblot in healthy intestinal mucosa and adenoma protein extracts (50μg) from 3.5-month old APC δ14 mice treated with Luc or GAST gene-specific siRNA, in comparison with control mice from the same genetic background (No RIA is currently available to quantify murine progastrin). Films were exposed for 30 min. Histogram, quantification of PG in adenomas collected in the colon and ileum of 4 animals per group, expressed relative to colon and ileum of C57/BL6 animals, (for each intestinal segment, p<0.05 compared to respective values in C57/BL6 * or in APCδ14 Luc^*) (C) Colo-26 murine CRC cells and Young adult mouse colon (YAMC) cells were transfected with siRNA directed against the murine (mGAST siRNA) or the human (hGAST siRNA) GAST sequence, and expression of GASTmKNA was quantified using RT-QPCR. (D) Scatter plot summarizing the levels of plasma IL-6 (black) and TNFα (white) in mice treated with Luc or GAST siRNA. No significant differences were noticed between these two groups, where levels of both cytokines reflect the absence of inflammation.

Figure S2. Progastrin, but not amidated or glycine-extended gastrin, stimulates β- catenin/Tcf-4 transcriptional activity in CRC cells. (A) (Top panel) Tcf-4 transcriptional activity was quantified using the TOP/FOP Luciferase reporter gene assay (31) in DLD-l/VO, DLD-1/ASG (clones [1] and [2]), SW480/βgal ^w and SW48O/GAST ⁰⁰ 'clones [1] and [2]) cells. Value are means ± s.e.m. of 4 similar experiments. Levels of dephosphorylated β-catenin (Middle panel) and Tcf-4 (Lower panel) were similar in all of these cell lines, as shown by immunoblotting (using actin as a loading control). (B) Human amidated gastrin (CCK-B) receptor mRNA was detected

by RT-PCR in samples from COS-7 cells transfected with the CCK-B receptor construct, but not in DLD-I and SW480 cells or in negative control (COS-7/CCK-B without reverse transcription).

Figure S3. mRNA expression of the Tcf-4 target genes c-Myc, cyclin Dl, Sox-9 and CLDNl is regulated by progastrin in human CRC cells. SW480/GJSI ⁰ clones [1] and [2] were treated or not with 5 nM progastrin (rPG) for 72h or transfected with a codon-optimized, shRNA insensitive preprogastrin cDNA (cPG). mRNA expression of the Tcf-4 targets c-Myc, cyclin Dl, Sox9 and claudin-1 (a list of Tcf-4 target genes is available in (9)) were quantified by RT-QPCR, using SW480/βgal ^(") cells as controls. Values are mean ± s.e.m. from three independent experiments. Student's t-test, p<0.05 (*) compared to SW480/ ¹ GAS^ cells.

Figure S4. Experimental modulation of ICAT expression in CRC cells. ICAT mRNA (A) and protein levels (B) were quantified in two clones each of progastrin- depleted CRC cells (SW480/ GAS^, white bars), transfected or not with control shRNA

(SW480/O4S7 ^< - ⁾ /Luc ⁽ - ⁾ , grey bars) or shRNA selective for ICAT (SW480/GAST ^{ -

VlCAT ⁰⁰ cells, black bars), as well as in SW480/O4S7 ^< - ⁾ /ICAT ⁽ - ⁾ cells transiently transfected with a codon-optimized, shRNA insensitive ICAT cDNA (+C.ICAT) (dark- grey bars). (p<0.05 compared to SW480/ 'GAS^ '/Luc ^{w (} *\ or compared to

SW480/G^(^7 ⁽ - ⁾ /ICAT ⁽ - ^{) (#)} , student's t-test, n=3)

Figure S5. Elevated expression of Tcf-4 target genes in intestinal adenomas from APCδ14 mice is reduced after progastrin-targeting siRNA treatment. 3-month old APCδ14 mice were treated for two weeks with siRNA targetting either the murine GAST gene or the Luciferase gene, as in Figure 1. Intestinal adenomas were processed for protein extraction, and expression of the Tcf-4 targets cyclin Dl (Top), c-Myc (Middle) and CD44 (Bottom) was quantified after Western immunoblotting in adenomas (Ad) and macroscopically intact epithelium (Ep) from the same animals, in comparison with intestinal epithelium of control mice from the same genetic background (Representative immunoblot is shown on Figure 5B). Values represent

means ± sem of adenomas from 4 animals/group. For each intestinal segment, p<0.05 compared to respective values in C57/BL6 ^# or in APCδ14 Luc^*).

Figure S6. Patient population used for Laser capture microdissection of colon samples. (A) Representative photomicrographs of typical healthy and tumoral colon tissues prior to (Original) and after Laser Capture Microdissection (Microdissected). (B)

Characteristics of the patients population used to determine GAST and ICAT gene expression levels, including the pathology scoring of their tumor according to the TNM classification. Tumor localization is represented as PC (proximal colon), MC (Middle colon), DC (distal colon), Sigmoidal, or Rectum. Metastasis at the time of surgery (MS), metastasis at a later stage (M), and Deceased patients (DC) are provided, with the following code (0: No; 1: Yes)

Figure S7. Pharmacological inhibition of PI3k inhibits Tcf-4 transcriptional activity in SW480/βgal ^(") cells. SW480/βgal ^(") cells were treated or not with 10 μM LY294002 and Tcf-4 activity was quantified as described above. Values are mean ± s.e.m. from three independent experiments. Student's t-test, p<0.05 (*) compared to untreated SW480/βgal ^w cells.

EXAMPLES Example 1

In the following description, all molecular biology experiments for which no detailed protocol is given are performed according to standard protocol.

SUMMARY

Background and aims: Aberrant activation of the. β-Catenin/Tcf-4 transcriptional complex represents an initiating event for colorectal carcinogenesis, shifting the balance from differentiation towards proliferation in colonic crypts. Here, we assessed whether endogenous progastrin, encoded by a target gene of this complex, was in turn able to regulate β-Catenin/Tcf-4 activity in APC-mutated cells, and we analyzed the impact of topical progastrin depletion on intestinal tumor growth in vivo.

Methods: Stable or transient RNA silencing of the GAST gene was induced in human tumor cells and in mice carrying a heterozygous Ape mutation (APCδ14), which overexpress progastrin but not amidated or glycine-extended gastrin.

Results: Depletion of endogenous progastrin production strongly decreased intestinal tumor growth in vivo, through a marked inhibition of constitutive, β-catenin/Tcf-4 activity in tumor cells. This effect was mediated by the de novo expression of the Inhibitor of β-Catenin and Tcf-4 (ICAT), resulting from a down-regulation of Integrin- linked kinase (ILK) in progastrin-depleted cells. Accordingly, ICAT down-regulation was correlated with progastrin overexpression and Tcf-4 target gene activation in human colorectal tumors, and ICAT repression was detected in the colon epithelium of tumor- prone, progastrin-overexpressing mice. In APCδ14 mice, siRNA-mediated progastrin depletion not only reduced intestinal tumor size and numbers, but also increased goblet cell lineage differentiation and cell apoptosis in the remaining adenomas.

Conclusions: Thus, depletion of endogenous progastrin inhibits the tumorigenicity of

APC-mutated CRC cells in vivo by promoting ICAT expression, thereby counteracting Tcf-4 activity. Progastrin targetting strategies should provide an exciting prospect for the differentiation therapy of colorectal cancer.

INTRODUCTION

Various strategies aiming to reduce tumor development through the induction of tumor cell differentiation have been developed in recent years, providing very promising results in animal models and in human patients. In particular, the use of all-trans-retinoic acid has proven extremely useful in the treatment of acute promyelocytic leukaemia (Lallemand-Breitenbach et al., 2005; Wang and Chen, 2000) , and is considered as a promising approach to force tumor differentiation in advanced thyroid cancer (Coelho et al., 2005). In the intestine, great hopes have been placed on agents modulating two major pathways involved in the control of cell proliferation and differentiation, namely the Notch cascade and the Wnt/β-catenin/Tcf-4 pathway (Radtke and Clevers, 2005; van Es and Clevers, 2005).

However, since a fine-tuned interplay between Notch and Wnt activation is essential for the homeostasis of the intestinal crypts, approaches aimed at directly targeting these signaling pathways in colon tumors are likely to have- adverse effects on overall intestinal physiology. Ideally, such strategies would greatly benefit from the capacity to identify targets that are selectively activated, or strongly stimulated in tumor cells, while being absent or inactivated in the surrounding tissues. Among the targets specifically activated in intestinal tumor cells, partially processed Gastrin-gene (GAST) products, such as glycine-extended gastrin and progastrin, offer an interesting prospect for the selective targeting of tumor growth. Indeed, the GAST gene itself is a target of both Tcf- 4 and K-Ras (Chakladar et al., 2005; Koh et al., 2000), two pathways that are frequently activated and act synergistically in colorectal cancers (Janssen et al., 2006), and these G/4ST-derived peptides seem to stimulate proliferation via an autocrine/paracrine loop on tumor cells (Hollande et al., 1997; Hollande et al., 2003). The proliferative effect of GgIy was discovered during the last few years (Seva et al., 1994) and more recently, progastrin was shown to stimulate proliferation (Ottewell et al., 2005; Singh et al., 2000) and to modulate epithelial cell/cell adhesion and migration (Hollande et al., 2003; Hollande et al., 2005). In addition, conflicting results emerged from experiments using animals with a targeted deletion of the GAST gene. Indeed, this deletion strongly enhanced tumorigenesis induced by a chemical carcinogen, azoxymethane(Cobb et al., 2002), whereas it resulted in a decreased number of polyps in the genetic bakground of APC ^+/min mice (Koh et al., 2000). Yet, they provided little information concerning the potential for treatments impairing the function of these peptides to slow-down or revert the growth of preexisting tumors. Critically, progastrin is produced and secreted in significant quantities by almost 80% of human colorectal tumors and tumor cell lines, while its secretion falls mostly below detection levels under physiological conditions (Ciccotosto et al., 1995; Konturek et al., 2002; Siddheshwar et al., 2001; Van Solinge et al., 1993). Therefore, since the constitutive activation of the Wnt signaling pathway represents a hallmark of sporadic CRC, as well as of the hereditary condition of adenomatous polyposis (Fodde et al., 2001), and since the GAST gsne is itself a target of Tcf-4, we investigated whether progastrin was able, in turn, to modulate APC mutation-

driven tumorigenesis, and whether targetting progastrin could represent a relevant strategy to reduce the growth of intestinal tumors.

We show that RNA interference-mediated progastrin depletion was capable of inhibiting tumor growth induced by a mutation of the APC gene in vivo. We then demonstrate that this inhibition reflected the capacity of progastrin to modulate the level of β-catenin/Tcf- 4 transcriptional activity in tumor cells, in spite of its constitutive activation triggered by the APC mutation. De novo expression of the inhibitor of β-catenin and Tcf-4 binding (ICAT) was instrumental to inhibit the β-catenin/Tcf-4 transcriptional pathway and to decrease tumorigenicity in progastrin-depleted tumor cells, while repression of ICAT was detected in progastrin-overexpressing mouse colonic mucosa, and in human and mouse intestinal tumors. Finally, treatment with GL45T-specifϊc siRNA induced a terminal differentiation towards the goblet cell lineage, not only of human CRC cells in vitro, but also of the progastrin-secreting intestinal adenomas of mice carrying a heterozygous mutation of the Ape gene.

RESULTS

Progastrin depletion reverses tumor growth induced by APC mutations in vivo

To assess whether a strategy aimed at selectively targeting endogenous progastrin in preexisting tumors would be able to antagonize Tcf-4-promoted tumor growth in vivo, we used Balbc/nude mice xenografts from the colorectal cell line SW480 (Morin et al.,

1997), as well as a mouse model of spontaneous intestinal tumorigenesis, APCδ14

(Colnot et al., 2004). Similar to what is found in a majority of human colorectal tumors

(Korinek et al., 1997), these two experimental models bear mutations of APC and consequently display a constitutive activation of the β-catenin/Tcf-4 pathway.

Control SW480/βgal ^H and DLD-l/VO cells were shown to express high progastrin levels but little of the amidated and glycine-extended forms of gastrin (Table 1). Within

7 weeks of subcutaneous injection in Balbc/nude mice, these cells produced large tumors (Figure IB). In contrast, the ability to form tumors within the same timespan was greatly

reduced when progastrin secretion was down-regulated by stable expression of short hairpin RNA (shRNA), directed against the GAST gene (SW480/ GAST^ cells, Figure IA-B). Similar results were obtained after injection of previously characterized (Hollande et al., 2003) DLD-I colorectal cancer cells expressing antisense GAST gene cDNA (Figure IB).

Table 1:

Next, we demonstrated that progastrin inhibition decreased the growth of preexisting intestinal tumors in a more pathologically relevant context, using APCMA mice. These animals have a similar truncating mutation of the Ape gene to that found in human familial adenomatous polyposis and sporadic colorectal cancer patients. They also partially recapitulate the intestinal epithelium phenotype found in these patients, since they develop spontaneous tumors in the ileum but also more colon tumors than the APCmin ^+/" model (Colnot et al., 2004). Three month-old APCMA mice always display multiple adenomas in the ileum and colon (data not shown), and progastrin levels, but not those of amidated or glycine-extended gastrin, were found to be elevated in these adenomas (Figure Sl). These animals therefore provided us with an ideal in vivo model to specifically study the role of progastrin modulation on Tcf-4-promoted tumor growth.

Daily treatment for two weeks with siRNA directed against the murine GAST gene (APCδIAI ^■ GAST^), previously shown to be effective in mouse cells (Figure Sl), induced a reduction in GAST gene and progastrin expression in intestinal adenomas, compared to that found after treatment with a Luciferase-specific siRNA (APCδ14/ Luc ^(-) ) (Figure IC-D, and Sl). In correlation with the reduced GAST expression, a significant reduction in tumor size was detected in the intestine of APCδ14/ GAST ^H animals, compared to controls (Chi-square, p<0.0001, n=9 per group). This reduction was particularly marked in the ileum (Figure 1C), but a similar trend was visible in the colon, although not reaching significance due to the smaller number of tumors located in this area (Fischer's exact test, p=0.228) (Figure ID). In addition, a 20% decrease in the total number of intestinal tumors was detected overall in APCδ14/ GAST ¹ ^ mice compared to control animals. In particular, this reduction was drastic in two animals of the APCδ14/ GAS^ group, where the ileum or colon were found to be tumor-free. In contrast, the G/(<ST-specifϊc siRNA treatment failed to down-regulate GAST gene levels in one mouse from the same group, and this animal displayed similar number and size of tumors to those found in controls (Figure 1C). Therefore, although the efficiency of the siRNA-mediated down-regulation of progastrin production varied among APCδ14/ GAS^ animals, tumor sizes and numbers were consistently decreased in mice displaying reduced progastrin levels.

Taken together, these results show that progastrin secretion is essential for the tumorigenicity of v4PC-mutated human CRC cells, and that a treatment inhibiting progastrin production in vivo strongly decreased tumor growth in a mouse model recapitulating the early stages of adenomatous polyposis and human colorectal tumorigenesis.

Progastrin modulates the level of β-catenin/Tcf-4 transcriptional activity in APC mutated intestinal tumor cells.

Since we showed that targeting progastrin inhibited tumor growth in two different APC- mutated intestinal tumor models, we postulated that this effect was due to the capacity of

progastrin to modulate the activity levels of the β-catenin/Tcf-4 transcriptional complex, known to be constitutively activated by the APC mutation. This activity, quantified by transcription of a luciferase reporter gene (TOP/FOP) (Korinek et al., 1997), was indeed elevated in control cells (SW480/βgal ^H ), but was significantly inhibited in progastrin- depleted clones (Figure 2A), without detectable modulation of β-catenin or Tcf-4 levels (Figure S2). A similar result was obtained in previously established (Hollande et al., 2003) DLD-I cells expressing an antisense GASTcDNA (Figure S2). Treatment with 5nM recombinant progastrin, but also endogenous progastrin produced by expression of a codon-optimized, shRNA insensitive, preprogastrin construct in SW480/ GAS^ clones, were found to restore a strong level of transcription of the luciferase reporter gene, confirming that progastrin is able to stimulate this transcriptional pathway (Figure 2A). In addition, immunofluorescent staining indicated that the reduction of β-catenin/Tcf-4 activity correlated with a marked reduction in the nuclear amounts of Tcf-4 and dephosphorylated β-catenin in Sψ4S0/GAST ^{') cells, and treatment of these cells with 5 nM recombinant progastrin partially restored the nuclear compartmentation of β-catenin and Tcf-4 (Figure 2B). Similarly, the expression of several Tcf-4 target genes (c-myc, cyclin Dl, Sox-9 and claudin-l)(Blache et al., 2004; van de Wetering et al., 2002) was strongly down-regulated after progastrin-depletion, and was significantly stimulated by reexpression of progastrin or treatment with the recombinant peptide (Figures 2C and S3). In contrast, treatment with amidated or glycine-extended gastrinl7, or with the amidated gastrin (CCK-B) receptor antagonist L365, 260, did not affect β- catenin/Tcf-4 activity in these cells (Figure S2), indicating that short processed forms of gastrin are unable to mimick the regulation of β-catenin/Tcf-4 activity by progastrin in these cells.

The above data clearly demonstrates that progastrin produced by SW480 CRC cells stimulates the activity of the β-catenin/Tcf-4 complex. It also shows that blocking progastrin production results in a strong inhibition of this transcriptional pathway, in spite of its constitutive activation in ^PC-mutated cells.

The inhibitor of β-catenin and Tcf-4 (ICAT) is the molecular switch mediating the modulation of β-catenin/Tcf-4 activity by progastrin.

In view of the mutated status of the APC gene in the CRC cells used, and since the nuclear compartmentation of Tcf-4 and β-catenin was significantly altered in progastrin- depleted cells (Figure 2A), we hypothesized that the decreased β-catenin/Tcf-4 activity could result from an increased interaction of β-catenin with another of its partners. One of these partners, ICAT (Inhibitor of β-catenin and Tcf-4), was recently identified as a direct inhibitor of the association between these two proteins (Tago et al., 2000), and is responsible for reduced proliferation and increased cell death when overexpressed in DLD-I and SW480 cells (Sekiya et al., 2002). Progastrin depletion induced a strong expression of ICAT mRNA and protein in SW480/ 'GAST^ cells (Figure 3A and B), while the expression of another binding partner of β-catenin, E-cadherin, was unaffected (data not shown). De novo repression of ICAT was induced by reexpression of the shRNA-insensitive preprogastrin cDNA (Figure 3A and B), as well as after treatment with 5 nM recombinant progastrin (data not shown).

In addition, ICAT expression was much weaker in the colonic epithelium of transgenic mice displaying tissue-specific intestinal overexpression of human progastrin (Tg/Tg) (Cobb et al., 2004) than that detected in their wild-type littermates (Figure 3C). Since the progastrin expressed by these mice is mutated and therefore unsensitive to processing enzymes(Cobb et al., 2004), this result demonstrates that full-length progastrin is also able to down-regulate ICAT expression in vivo.

Thus, endogenous progastrin is capable of repressing ICAT expression in vitro and in vivo, and progastrin depletion in CRC cells is sufficient to induce a strong de novo expression of this inhibitor.

Therefore, we investigated whether ICAT was indeed the molecular target involved in the regulation of β-catenin/Tcf-4 by progastrin in CRC cells. Using co- immunoprecipitation, we detected a significant increase in the association between

dephosphorylated β-catenin and ICAT in progastrin-depleted SW480/G,4_S7 ^(") cells (Figure 4A), paralleled by the appearance of cytoplasmic colocalization of these two proteins (Figure 4B). In contrast, the amount of Tcf-4 co-immunoprecipitating with dephosphorylated β-catenin was significantly decreased compared to SW480/β-ga^ (Figure 4A). In turn, β-catenin was again preferentially bound to Tcf-4 after treatment of SW480/ 'GASI^ cells with recombinant progastrin, while binding to ICAT and colocalization of the two proteins were reduced (Figure 4A-B).

To confirm that ICAT re-expression in progastrin-depleted cells was functionally responsible for the inhibition of the β-cat/Tcf-4 transcriptional activity, we experimentally prevented the de novo expression of ICAT mRNA and protein in

SW480/ GASI^ cells, using retrovirus-driven shRNA expression (Figure S4). We found that the activation of the β-catenin/Tcf-4 transcriptional complex was restored in the resulting SW480/O4,ST ^(") /ICAT ^(") cells (Figure 4C), which also displayed increased expression of Tcf-4 target genes (Figure 4D). In turn, this increase was prevented by overexpression of a codon-optimized version of ICAT cDNA in SW480/G ^! ^5'7 ^<") /ICAT ^(") cells (Figure 4C, 4D and S4), demonstrating the specificity of ICAT in this process.

These results indicate that the modulation of ICAT expression levels is the major molecular event underlying the regulation βcat/Tcf-4 activity by progastrin.

In vivo, low ICAT levels (Figure 5A and 5C) and high expression of Tcf-4 target genes (Figure 5B-C, and Figure S5) were detected in the progastrin-overexpressing intestinal tumors of APCδ14 animals. In contrast, ICAT expression was significantly increased in adenomas collected from mice treated with G^T^-specific siRNAs (Figure 5A-C), concomitant with a strong repression of Tcf-4 target genes. Interestingly, expression of cyclin Dl, c-Myc and CD44 in the macroscopically healthy epithelium of APCδ14//-W ^" * mice was found to be elevated when compared to GAST^ animals, suggesting that mutation of a single Ape allele is sufficient to partially increase target gene expression in

APCδ14 intestines, and that progastrin depletion is already effective to reduce Tcf-4 activity in this precancerous setting (Figure 5B and Figure S5).

In addition, levels of GAST mRNA in microdissected human colon tumors and matching healthy samples obtained from 23 patients with CRC (Figure S 6) were increased to various degrees in the tumors of 78% of patients studied (18/23) (Figure 5D, top graph), while ICAT expression was down-regulated in the corresponding tumor samples (Figure 5D, bottom graph). A significant statistical correlation was found overall between increased GAST gene expression and repression of ICAT (Spearman's correlation coefficient: r = -0.46, p = 0.027), corroborating the relevance of the progastrin-induced repression of ICAT in vivo (Figure 5D, insert). This correlation was confirmed by immunohistochemistry on samples from patients with low (eg. patient 22) or high (eg. patient 4) progastrin expression, showing that ICAT was repressed and immunoreactivity for Tcf-4 target genes was strong in tumors expressing high progastrin levels, while these expression patterns were reversed in tumors with low progastrin (Figure 5E). Similar results were obtained on CRC samples from ail 11 patients tested by immunohistochemistry, and this correlation mirrors the inverse correlation between ICAT levels and β-catenin/Tcf-4 activity observed in vitro.

Overall, these results demonstrate that, despite the constitutive activation of the β- catenin/Tcf-4 complex in .^PC-mutated cells, the activity of this pathway can be inhibited by progastrin depletion through a re-expression of the inhibitor ICAT, in vivo and in vitro. To our knowledge, these results also provide the first demonstration of an hormonal regulation of ICAT, a small inhibitor of β-catenin binding to Tcf-4 (Tago et al., 2000) previously described as a tumor suppressor (Daniels and Weis, 2002).

Restoration of ICAT expression is responsible for the decreased tiimorigenicity induced by progastrin depletion in CRC cells.

In view of the essential role played by the constitutive activation of β-catenin/Tcf-4 transcriptional activity in colon tumorigenesis, and since our results indicated that ICAT reexpression was responsible for the decreased Tcf-4-mediated transcription in

progastrin-depleted tumor cells, we further investigated in vitro and in vivo the role of ICAT repression as an essential component of the tumor-promoting activity of progastrin.

In vitro, inhibition of ICAT re-expression in SW480/GL45'I ^f; /ICAT ^(") cells was found to enhance their capacity for anchorage-independent growth in a soft agar growth assay, up to a similar level to that found for SW480/β-Gal ^(") cells (Figure 6A). In vivo, the ability of SW480/GλSI ^f; /ICAT ^(") cells to form tumors after injection in BALB/c nude mice was much higher than that of SW480/Gv4 ₁ S7 ^f; /Luc ^(") cells (Figure 6B), and similar to SW480/β-Gal ^('} cells (see Figure IB).

Overall, these results clearly demonstrate that repression of ICAT by endogenous progastrin in CRC cells is instrumental to the tumor-promoting activity of this prohormone. Critically, they also show that the de novo expression of ICAT induced by progastrin depletion is strong enough to induce a significant inhibition of β-catenin/Tcf- 4 activity and thereby decrease tumor growth in vitro and in vivo.

Phosphatidylinositol 3-kinase-mediated activation of Integrin-Linked Kinase is responsible for ICAT repression and modulation of the β-catenin/Tcf-4 pathway by progastrin.

Since progastrin was recently shown to activate the phosphatidylinositol 3-kinase (PDk) pathway in intestinal cells in vitro (Hollande et al., 2003) and in vivo (Ferrand et al., 2005), we determined whether activation of this enzyme was necessary for progastrin- induced ICAT repression and stimulation of β-catenin/Tcf-4 activity in SW480 cells. First, we found that incubation of SW480/βgal ^(") cells with the selective PI3k inhibitor LY294002 was sufficient to induce the expression of ICAT (Figure 7A), and to reduce β-catenin/Tcf-4 transcriptional activity (Figure S7), thus unravelling a previously unknown link between these two pathways. In contrast, pharmacological inhibition of the Src or ERK1/2 pathway using 10 μM PP2 or 1 μM PD98059, respectively, did not affect ICAT expression or β-catenin/Tcf-4 activity (data not shown). In addition,

LY294002 abolished the down-regulation of ICAT induced by a treatment of clones with recombinant progastrin (Figure 7A), showing that PDk activation is essential to the repression of ICAT by progastrin. Accordingly, serine 473- phosphorylation of Akt/PKB, a downstream target of PDk, decreased markedly in progastrin-depleted SW480/G _^ 457^ cells, but was reinstated after incubation with 5nM recombinant progastrin (Figure 7A).

Ser473 of Akt/PKB is known to be a direct phosphorylation target of integrin-linked kinase (ILK) (Delcommenne et al., 1998). To determine whether ILK activation was the missing molecular link between progastrin and the repression of ICAT, we first assessed whether the expression and/or activity of ILK were regulated by progastrin in SW480 cells. Both the expression and activity of ILK were significantly decreased in SW480/ GAS^ cells, and this inhibition was partially reversed after treatment with 5nM recombinant progastrin (Figure 7B). These results are corroborated by data showing a drastic reduction in the serine phosphorylation of integrinβl, a downstream target of ILK (Mulrooney et al., 2000) in SW480/G,4 ₎ S7 ^{f j} clones (Figure 7B).

In addition, we found that transient expression of dominant-negative ILK (DN-ILK), which was confirmed to repress Akt/PKB phosphorylation, was able to induce ICAT expression in SW480/βgal ^(") cells (Figure 7C), indicating that this enzyme is involved in a signalling pathway controlling the expression of the ICAT gene. Pharmacological inhibition of PDk in these SW480/βgal ^("} /DN-ILK cells did not increase ICAT expression any further, indicating that the two enzymes are involved in the same pathway, or in redundant pathways, in the regulation of ICAT gene expression (data not shown).

Furthermore, the repression of ICAT mRNA and protein and the increased phosphorylation of Akt/PKB, induced by treatment of progastrin-depleted S W48 OIGAS^ cells with recombinant progastrin, was prevented by dominant-negative- mediated inhibition of ILK (Figure 7C). In contrast, shRNA-mediated repression of ICAT in SW480/GAST ^{') cells (as shown in Figure S4) did not alter ILK expression or

activity (data not shown), confirming that ICAT is located downstream of ILK in , progastrin-mediated signalling.

These results clearly demonstrate that the activations of PI3k and ILK are instrumental to the progastrin-mediated repression of ICAT in CRC cells.

Depletion of progastrin in vitro and in vivo induces the terminal differentiation of intestinal tumor cells along the mucus-producing goblet cell lineage.

Since the Wnt/β-catenin/Tcf-4 pathway is deemed to be essential for the maintenance of the stem/progenitor phenotype of cells at the bottom of the intestinal crypt, and is widely seen as a molecular switch between cell proliferation and differentiation (Radtke and Clevers, 2005; van de Wetering et al., 2002), we next determined whether its down- regulation in progastrin-depleted CRC cells was able to drive these cells towards differentiation.

In view of its recently described role as an early promoter of cell differentiation (Chung and Eng, 2005), we first analyzed the localization of PTEN (phosphatase and tensin homologue deleted from chromosome 10) in stable SW480/GJ)S ^r 7^ ⁾ clones. In striking contrast with its predominantly cytoplasmic localization in SW480/βgal^ cells, PTEN was mostly found in the nucleus of SW480/GAS^ clones (Figure 8A), a localization thought to be associated with its role in cell differentiation (Lian and Di Cristofano, 2005).

Since the process of developing stable progastrin-depleted clones is expected to select the least differentiated and non-apoptotic cells, the expression of genes associated with terminally differentiated intestinal cell lineages was quantified in SW480 cells, after transient transfection with rhodamine-tagged siRNA oligonucleotides directed against the GAST or the β-galactosidase gene. A significant induction in the expression of the goblet cell-specific gene Muc-2 was detected in SW480/O4 _i S'7 ^r"; but not in SW480/β- gal^ cells, without detectable alterations of the genes encoding chromogranin A

(expressed in enteroendocrine cells) or enterocyte-specific alkaline phosphatase (Figure

8B). Finally, the detection of caspase-3 activation in GAS^ cells, and the finding that cells positive for activated caspase-3 were also Muc2-positive (Figure 8C), indicated that progastrin depletion was capable of differentiating CRC cells towards a secretory lineage, before promoting their apoptosis in vitro. Treatment with recombinant progastrin for 48h reversed these phenotypes (Figure 8C), while amidated and glycine- extended gastrins had no effect (not shown). Similar results were obtained with DLD-I cells transfected with G<4>ST-specific siRNA oligonucleotides (not shown). Thus progastrin-depletion was capable of driving CRC cells towards a terminal differentiation and apoptosis in vitro.

In addition, we sought to demonstrate that reduction of progastrin levels in vivo would also be sufficient to induce differentiation of intestinal tumors induced by a constitutive activation of the β-catenin/Tcf-4 pathway. Immunohistochemistry on samples collected from APCδ14 mice treated with luciferase or G4<ST-specific siRNAs (see Figure 1), demonstrated that the remaining small intestinal and colonic adenomas of APC δIAIGAST" ^' * animals displayed greatly increased numbers of Muc-2 and Alcian-blue positive cells compared to controls (Figure 8D), in line with the increased expression of ICAT and the reduced expression of c-Myc in the same samples (see Figure 5A-C). Finally, while cells staining for activated caspase 3 were very scarce in control animals, their numbers were found to be strongly elevated in adenomas from APC δIAIGAST* ^' * animals (Figure 8D). These results demonstrate that siRNA-mediated down-regulation of the GAST gene is able to drive tumor cells towards terminal differentation and apoptosis in vivo. Intestinal adenomas developed by APCδ14 mice do not express glycine-extended or amidated forms of gastrin. Since high molecular weight processing intermediates were never detected by immunoblotting using polyclonal antibodies raised against recombinant progastrin, and since full-length recombinant progastrin was shown to regulate CRC cell differentiation in vitro, we conclude that tumor cell differentiation and reduced tumor growth after siRNA treatment is mostly due to reduced concentrations of full-length progastrin.

DISCUSSION

The present work provides the demonstration that blocking the production of progastrin significantly inhibits "constitutive)' beta-catenin/Tcf-4 transcriptional activity in human CRC cells, reduces their anchorage-independent growth in vitro as well as their ability to form tumors in nude mice, and promotes their differentiation and apoptosis. Progastrin depletion resulted in re-expression of the "Inhibitor of beta-catenin And Tcf-4" (ICAT), a small peptide originally identified in Xenopus and shown to inhibit the interaction of beta-catenin with Tcf-4 (Gottardi and Gumbiner, 2004; Tago et al., 2000). In addition, treatment with recombinant progastrin down-regulated ICAT and increased beta- catenin/Tcf-4 activity in cells where GAS gene expression had previously been switched off, confirming that chronic progastrin secretion was involved in maintaining a high level of activation for this pathway in CRC cell lines. The physiological relevance of the link unraveled here between chronic progastrin secretion and ICAT repression was emphasized in vivo by results showing a tight correlation between these two events within colorectal tumors in a panel of 16 patients with CRC, and by the demonstration that ICAT expression was down-regulated in the colonic epithelium of tumor- susceptible, progastrin-overexpressing mice (Cobb et al., 2004).

This data is consistent with the previous observation that progastrin promotes the development of colorectal tumors, and provides a molecular explanation for the strong reduction in the number and size of adenomas observed in gastrin-deficient APCmin ^+/~ mice(Koh et al., 2000). Furthermore, expression profiling of target genes in gastric parietal cells from mice with a null-mutation in the GAS gene compared to wild-type mice from a similar background, showed that 20% of differentially expressed genes were also known to be target genes for Wnt and Myc, emphasizing the link between these signalling pathways (Jain et al., 2006).

It is of note that, while increased progastrin expression in CRC is well documented

(Ciccotosto et al., 1995; Konturek et al., 2002; Siddheshwar et al., 2001; Van Solinge et al., 1993), the repression of ICAT described in the present work conflicts with the only previous report concerning this field (Koyama et al., 2002). This discrepancy could be

due to the fact that the increased expression of ICAT reported by Koyama et al. was measured using semi-quantitative RT-PCR on non-microdissected tissue. Our results, allowing for a direct comparison of ICAT expression by cells of epithelial origin in the healthy mucosa and in tumors, appear to be more consistent with the described role for ICAT as a tumor suppressor. Indeed, progastrin-mediated ICAT repression appears essential for tumor growth, since newly synthesized ICAT in progastrin-depleted CRC cells was shown to compete with Tcf-4 for binding to beta-catenin, resulting in decreased transcriptional activity of Tcf-4 and decreased tumorigenicity, in agreement with previously reported data concerning this peptide (Sekiya et al., 2002).

The capacity to reinstate, albeit incompletely, the repression of ICAT and the activation of the beta-catenin/Tcf-4 pathway by treatment with exogenous progastrin, argues in favor of the existence of an autocrine/paracrine loop mediated by a membrane receptor. The analysis of signaling events involved in ICAT-mediated modulation of beta- catenin/Tcf-4 activity by progastrin allowed us to identify a novel link between the beta- catenin/Tcf-4 and PI3kinase/ILK pathways in CRC cells. Indeed, pharmacological inhibition of PDk and dominant-negative induced disruption of ILK signaling increased ICAT expression and prevented progastrin-mediated repression of this peptide. Previous results by Tan et al. (Tan et al., 2001) demonstrated that inhibition of ILK suppressed beta-catenin-Tcf/Lef-dependent transcription through repression of snail, thereby increasing E-cadherin expression and recruitment of beta-catenin at the membrane in APC-mutated human colon carcinoma cells. Results from the present work indicate that even in non-confluent cells, where membrane-recruitment of beta-catenin is low, progastrin-mediated ILK signaling is able to modulate beta-catenin/Tcf-4 activity through a control of ICAT expression.

Finally, our results also show that agents antagonizing progastrin functions provide new therapeutic options to support the present surgical approach in colorectal cancer. Results presented here show that inhibition of progastrin signaling provides a specific therapeutic alternative for colon cancer management. Since progastrin seems to be secreted only by the tumor and not by the healthy epithelium, such a strategy should

counteract the effect of the frequent APC mutations and decrease the activation of beta- catenin/Tcf-4 signaling closer to physiological levels, thereby promoting the preservation of normal crypt integrity.

METHODS

Cells and cell culture conditions

Colorectal cancer cell lines DLDl and SW480 were maintained at 37 ⁰ C in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (Eurobio, Les Ulis, France), 1% L-glutamine and penicillin/streptomycin.

Materials

The following primary antibodies were used: mouse anti-dephosphorylated beta-catenin (clone 8E4; AG. Scientific), mouse anti beta-catenin and anti-E-cadherin (Transduction Laboratories), goat anti-Tcf-4 (Santa Cruz Biotechnology), rabbit anti-ICAT (a generous gift from Dr C. Gottardi; (Gottardi and Gumbiner, 2004)), polyclonal anti-PTEN (Upstate), rabbit anti-phosphoserin (Zymed), rabbit anti-ILK for western blot (Cell Signaling) and for immunoprecipitation (Upstate), mouse anti-integrin βl (Chemicon), rabbit anti-Akt, anti-Ser473-phosphorylated Akt, anti-activated caspase-3 (Cell Signaling), Anti-Muc2 monoclonal antibody (Neomarkers), Human tissue collection

Specimens of colon tumors and histologically normal epithelium (taken at a significant distance from the tumor) from 16 patients, were obtained from the pathologist after resection according to French government regulations and local committee guidelines. Informed consent was obtained from all patients. Tissue samples were stored in liquid nitrogen until further use.

Tissue sections, microdissection and RNA preparation

Tissue sections were prepared from liquid nitrogen-frozen tumor samples, and hematoxylin/eosin staining was performed to assess their quality, orientation within the tissue, and epithelial content. Immunofluorescent detection of beta-catenin, Tcf-4, c-myc and ICAT was performed on sections adjacent to those used for microdissection.

Laser Capture Microdissection (LCM) was performed on four to six sections per sample using a PixCell® He microdissector from Arcturus (Alphelys, Plaisir, France). The following laser beam settings were used: 265mV, 45mWh, 15μm diameter, 18ms. Microdissected tissues were then directly collected in RNA lysis buffer and RNA was prepared using the RNAeasy Microkit (Qiagen, Courtaboeuf, France), according to the manufacturer's instructions. The quality and amount of RNA recovered were assessed using RNA pico Labchips (Agilent Technologies, Palo Alto, California).

Stable transfection

Transfection of gastrin gene antisense and control cDNA into DLDl cell lines was described in (Hollande et al., 2003). In order to generate progastrin shRNA, the following oligonucleotides were cloned into pSilencer vector (Ambion) and transfected in SW480 cells: sense: 5'-

GAAGAAGCCTATGGATGGATTCAAGAGAAGGTAGGTATCCGAAGAAGTTTT TT-3' (SEQ ID N°l) and antisense: 5'- AATTAAAAAACTTCTTCGGATACCTACCTTCTCTTGAATCCATCCATAGGCTT CTTCGGCC 3' (SEQ ID N°2). β-galactosidase oligonucleotides used to generate control shRNA were: sense: 5'

GATCCCAAGGCCAGACGCGAATTATTTTTCAAGAGAAAATAATTCGCGTCTG GCCTTTTTTTGGAAA 3' (SEQ ID N°3); antisense 5'AGCTTTTCCAAAAAAAGGCCAGACGCGAATTATTTTCTCTTGAAAAATAAT TCGCGTCTGGCCTTGG 3' (SEQ ID N°4). 5 X 10 ⁴ cells/well were seeded and transfected 24h later with 500 ng of pSilencer/progastrin-shRNA or pSilencer/β-galactosidase-shRNA, 50 ng of pCDNA-3, and 5 μl/well of Exgen 500 (Euromedex). Selection was performed with Neomycin (500ng/μl). Retrovirus-mediated transfection of ICAT or Luciferase shRNA in SW48O/GAS ⁰⁰ cells shRNA oligonucleotides for ICAT (sense 5'

GATCCGGATGGGATCAAACCTGACATTTTCAAGAGAAATGTCAGGTTTGATC CCATCTTTTTTG V (SEQ ID N°5) and antisense 5' AATTCAAAAAAGATGGGATCAAACCTGACATTTCTCTTGAAAATGTCAGGTT

TGATCCCATCCG 3' (SEQ ID N°6)). and luciferase (sense 5'GATCCGACATTAAGAAGGGCCCAGCTTTTCAAGAGAGCTGGGCCTTAATC TTTTTT 3' (SEQ ID N°7), and antisense

5'AATTGATTAAGGCCCAGCTCTCTTGAAAAGCTGGGCCCTTCTTAATGTCG 3' (SEQ ID N°8)) were cloned into pSIREN-retroQ (Clontech). The amphotropic packaging cell line (φNX ; Dr Garry Nolan, Stanford University, Stanford, California) was transfected as described by Pear et al (Pear et al., 1993). The medium containing the virus was removed 48 hr after transfection and centrifuged at 1500 rpm for 5 min to pellet cell debris. Target cells (2.10 ⁵ /well in 6 well-plates) were infected in the presence of polybrene (8 μg/ml) with 2 ml of the centrifuged medium containing the virus. Infected cells were treated with puromycin (5 μg/ml) in order to select for positive clones.

Transient transfections

The day before transfection, SW480 cells were seeded in 24-well plates (5x 10 ⁴ cells/well). Cells were transfected with 100 pmol of rhodamine tagged GAS-speciύc or β-galactosidase-specific shRNA, and 5 μl/well of Exgen 500 (Euromedex), according to the manufacturer's instructions. 48 hr or 72 hr after transfection, cells were lysed (for immunoblotting and PCR) or fixed in 1 % paraformaldehyde for immunostaining.

Alcian blue staining 72 hr after transfection, cells on coverslips were washed with PBS and incubated for 30 min with alcian blue (lOg/1 in 3% acetic acid). After rinsing cells in water, coverslips were mounted on glass slides in mowiol and stored at 4°C until use.

RT-Quantitative PCR

2.5 μg of total RNA was pretreated with DNAse RQl (Promega) for 30 min at 37 ⁰ C and used for reverse transcription with M-MLV reverse transcriptase (In vitrogea). Quantitative PCR was carried out using the LightCycler FastStart DNA MasterPlus SYBR Green I kit (Roche Diagnostics, Meylan, France), under the following conditions:

Actin: denaturation during 10 min at 95 ⁰ C, amplification for 50 cycles: 10 sec at 95°C, 6 sec at 58°C, 11 sec at 72°C. Melting curve: 0 sec at 95°C, 30 sec at 68°C, 0 sec at 95 ⁰ C and cooling was for 2 min at 4O ⁰ C: c-myc, ILK, chromogranin A and GAPDH amplification: denaturation during 10 min at 95 ⁰ C, amplification for 50 cycles: 10 sec at 7O ⁰ C, 6 sec at 58°C, 13 sec at 72°C. The melting curve was 0 sec at 95 ⁰ C, 30 sec at 8O ⁰ C, 0 sec at 95°C and cooling was for 2 min at 40 ⁰ C. Progastrin, Muc-2, ALP and ICAT amplification: denaturation during 10 min at 95°C, amplification during 50 cycles with: 10 sec at 7O ⁰ C, 6 sec at 95 ⁰ C, 11 sec at 72 ⁰ C. The melting curve was 0 sec at 95°C, 30 sec at 80 ⁰ C, 0 sec at 7O ⁰ C and cooling was for 2 min at 4O ⁰ C. Primers used for amplification of selected genes were: GAPDH-sense:

5'GGTGGTCTCCTCTGACTTCAACAS ' (SEQ ID N°9) antisense: 5'GTTGCTGTAGCCAAATTCGTTGTS' (SEQ ID N ⁰ IO); Actin-sense: 5'CGGGAATCTGCGTGACATS' (SEQ ID N ⁰ I l); antisense:

5 'AAGGAAGGCTGGAAGAGTGCS' (SEQ ID N°12); ILK-sense: 5' ATGACTGCCCGAATT AGC ATG3' (SEQ ID N°13); antisense:

5'GCTACCCAGGCAGGTGCATAS' (SEQ ID N°14); ICAT- sense: 5'GCTCTGGTGCTTTAGTTAGGS' (SEQ ID N°15); antisense:

5 'GCACTTGGTTTCTTTCTTTTCS' (SEQ ID N°16); c-myc-sense: 5'CGTCTCCACACATCAGAGCACAAS' (SEQ ID N°17); antisense: 5'TCTTGGCAGCAGGATAGTCCTTS' (SEQ ID N ⁰ 18); human progastrin-sense: 5'AGAGGATCCAAATGCAGCGACTATGTGTGTATGS' (SEQ ID N°19); antisense: 5'GGCGAATTCCTAGGATTGTTAGTTCTCATCCTCAGS' (SEQ ID N°20); MUC-2- sense: 5'TGGGTGTCCCTCGTCTCCTACAS' (SEQ ID N°21); antisense: 5'TGTTGCCAAACCGGTGGTAS' (SEQ ID N°22); Alkaline Phosphatase-sense: 5'CTCCAACATGGACATTGACGS' (SEQ ID N°23); antisense: 5 'CAGTGCGGTTCCACACATACS' (SEQ ID N°24); Chromogranin A-sense: 5αTCACCGCCACTGCCACCACCA3' (SEQ ID N°25); antisense: 5'CACCTTAGTGTCCCCTTTTGTCATAGGGCTS' (SEQ ID N°26).

Soft agar colony formation assay

1 X 10 ⁵ cells were resuspended in 0.2% agar in DMEM-F 12 supplemented with 10% fetal bovine serum, 1% glutamine and penicillin/streptomycin, and plated onto a 6-cm petri dish containing a solidified bottom layer (0.5% agar in growth medium). 8 days after plating, cells were stained in 0.02% crystal violet and the size and number of colonies was quantified under a microscope. AU clones were seeded in triplicates and ten fields were counted for each.

In vivo studies In vivo experiments were performed following the French guidelines for experimental animal studies (Direction des Services Veterinaires, Ministere de l'Agriculture, Agreement N°B34-172-27) and fulfill the UKCCCR guidelines for the welfare of animals in experimental neoplasia. 2 X 10 ⁶ cells were injected subcutaneously into 6- weeks-old athymic BALB/c-nu/nu {nude) mice. Tumor growth was followed regularly thereafter (Estimated tumor volume = (Length x width x thickness)/2).

APCδ14 mice were housed in conventional conditions, and randomized in two groups. The first group (n=9), was treated once daily by intraperitoneal administration of 250 μg/kg siRNA against the murine GAST mRNA. Control mice (n=9) were treated with a similar dose of siRNA against Luciferase. (siRNA sequence on Figure S7 online). SiRNA design incorporated 3' overhangs and did not contain a 5'UGUGU3' sequence, to minimize proinflammatory response and degradation in vivo (Judge et al., 2005; Marques et al., 2006). After a 10-day treatment, mice were culled, their blood collected and their intestine dissected out. Adenoma scoring in the intestine and colon was described in (Colnot et al., 2004). Samples of healthy intestinal mucosa, as well as intestinal and colon adenomas, were fixed in 4% PFA and paraffin-embedded, or frozen in liquid nitrogen, for Western blotting, IHC or RNA extraction. Plasma was prepared and assayed for IL-6 and TNFα using the FACSarray system (Becton Dickinson).

Luciferase reporter gene assay

The protocol used to measure the transcriptional activity of TCF/LEF-1 was described before (Morin et al., 1997). Briefly, cells were transfected with 500 ng of TCF/LEF-1 reporter (pTOP-FLASH), or control vector (pFOP-FLASH) and with 25 ng pCMV- Renilla. Luciferase activity was quantifiied using the Dual luciferase reporter assay system (Promega), and was normalized relative to the activity of Renilla.

Immunofluorescent staining

Immunofluorescent staining was performed as described previously (Hollande et al., 2003) Slides were then observed using a laser scanning LEICA sp2 confocal microscope with a x40 oil immersion lens.

Immunochemistry on paraffin tissue sections

Paraffin tissue sections were prepared from FVB/N mice that were either transgenic for the human GAS gene (Tg/Tg) or wild type littermates(Cobb et al., 2004). Following dewaxing and hydration, sections were pretreated with peroxidase for 20 min at room temperature. Antigen was retrieved by boiling samples for 20 min in 1OmM Tris -1 mM EDTA, pH 9. Slides were allowed to cool down to room temperature, and incubated overnight at 4°C with antibodies in PBS +0.05% BSA. In all cases, the Envision+ kit (DAKO) was used as a secondary reagent. Stainings were developed using DAB (Sigma) and slides were counterstained with either hematoxylin or Nuclear Fast Red and mounted.

Protein lysates, immunoprecipitation, SDS-PAGE and Western immunoblotting

Protein lysates, immunoprecipitations and immunoblotting were performed as described before (Hollande et al., 2003) Proteins were visualized using ECL Plus (Amersham Biosciences) and the bands were quantified by densitometry using ImageJ 1.32 J (NIH, Bethesda, MD).

Immune complex kinase assays

The reaction was started by the addition of 25 μl of kinase buffer (50 mM Hepes pH 7.0, ImM MnCl ₂ , 1 mM Na orthovanadate, 2 mM NaF, 5 μg Myelin Basic Protein (Sigma)) with 0.5 μg of ATP (250 mM ATP, lμCi[γ- ³² P]ATP) per tube, and incubated at 3O ⁰ C for

20 min. The reaction was terminated with the addition of 10 μl of 4X sample buffer. The tubes were centrifuged at 10,000g for 1 min and the proteins were resolved on a 14% SDS-page gel. Phosphorylation of the substrate was visualized by autoradiography (Marotta et al, 2003). Radioimmunoassay

GAS gene products progastrin, glycine-extended gastrin (G-gly) and amidated gastrin (G-NH2) in cell supernatants were quantitated by radioimmunoassay, as described before (HoUande et al., 1997).

Statistical Analysis All statistical analyses were performed using SAS version 9.1 for Windows (Cary, NC, USA). Student's t-test was used in order to determine the statistical significance of differences between control and progastrin-depleted cells, with or without treatment with recombinant progastrin. In order to determine the association between GAS and ICAT gene expressions in human tumors, the variables were normalized using their natural log values before determining Pearson's correlation coefficient (r).

Example 2

Selective antibodies against progastrin reverse the repression of progastrin on ICAT expression and induce a decrease in c-myc expression.

INTRODUCTION

The present example describes the characterisation of two independently generated polyclonal antibodies against progastrin, aiming to demonstrate that they selectively recognise the full-length progastrin(l-80) peptide but not the processed peptides potentially present within human blood, such as amidated gastrin, glycine-extended gastrin, and the six amino-acid C-terminal flanking peptide (CFTP). This characterisation was performed in vitro using an ELISA assay.

The second step described hereunder was the validation of the proof of concept that the effects of progastrin on ICAT expression and on beta-catenin/Tcf4 activity can be

blocked by a selective antibody. In order to demonstrate this so-called neutralising activity of the progastrin antibodies, we used human SW480 colorectal tumor cells, which bear a mutation in the ape gene and display endogenous progastrin secretion due to the constitutive beta-catenin / Tcf4 activity.The capacity of these cells to secrete high amounts of progastrin and very low levels of amidated and glycine-extended gastrin is described in Table I.

These cells were treated independently with two different polyclonal antibodies, one directed against the N-terminal sequence of progastrin, the other against the C-Terminal sequence.

The N-Terminal sequence was: SWKPRSQQPDAPLGT (SEQ ID N°32) The C-terminal sequence was: FGRRS AEDEN (SEQ ID N°31)

RESULTS -Antibody selectivity: Selectivity of antibodies was assessed by an ELISA test described in the Methods section. Results (Figure 10) show that both polyclonal antobodies selectively recognize progastrin and the respective peptides used as immunogens for their production. They do not bind to Glycine-extended Gastrin, to amidated gastrin or to the C-terminus flanking peptide (CTFP) (sequence : SAEDEN), another peptide derived from the maturation process of progastrin and detected in the human serum (Smith KA, Gastroenterology 2006).

- Neutralizing activity of progastrin antibodies: The ability of progastrin antibodies to inhibit the repression of ICAT by progastrin and to induce a resulting decrease in the transcriptional activity of the beta-catenin/Tcf4 complex was addressed in SW480 colorectal carcinoma cells. When cells were incubated for 30 hours in the presence of rabbit polyclonal antibodies directed against the N-terminal or C-terminal extremity of progastrin^ -80), both at a 1/5000 dilution, ICAT mRNA expression was strongly increased while those of c-Myc and cyclin Dl were markedly decreased, in comparison with cells treated with the same concentration of a non-specific rabbit polyclonal

antibody. C-myc and cyclin Dl are recognised targets of beta-catenin / Tcf4 transcriptional activity, and the decrease in their expression is considered as a consequence of the decrease in beta-catenin/Tcf4 activity (Figure 11).

CONCLUSION

These results demonstrate in vitro that progastrin inhibition of ICAT expression can be reversed by its binding to a selective antibody, resulting in an inhibition of the constitutive beta-catenin / Tcf4 activity. The present invention thus provides for the first time the proof of concept that a progastrin antibody can be used to block the protumorigenic effects of progastrin occuring via ICAT and beta-catenin / Tcf4 activity.

METHODS

ELISA

Antigens were diluted at the indicated concentration in PBS, and 100 μl were coated in multisorb 96 well plates at 4°C overnight. The next day, wells are washed 3 times with 200 μl of PBS/Tween 1% and 100 μl of the blocking solution (PBS/Tween 1%/BSA 0.1%) was added for 2 hours at 22 ⁰ C. After another 3 washes with PBS/1% Tween, primary antibodies were diluted in the blocking solution and 100 μl was added to the wells for 2 hours at 22 ⁰ C. Antibody concentration is indicated in the Figures. Secondary antibodies are diluted in the same blocking buffer and after 3 washes, 100 μl are added to the wells for 2 hours at 22 ⁰ C. Finally, wells were washed again with PBS/Tween 1% and 100 μl of OPD solution substrate was added for 20 minutes at 22°C. The reaction was stopped with 50 μl OfH ₂ SO ₄ 4N and reading was performed at 492 nm.

Cell culture and treatment with progastrin-selective antibodies

The CRC cell line SW480 was maintained at 37°C in Dulbecco's Modified Eagle

Medium (DMEM) supplemented with 10 % fetal bovine serum (Eurobio, Les UHs,

France), 1 % L-glutamine and penicillin/streptomycin. SW480 cells were plated in 6 well plates (200,000 cells/well) in DMEM containing 10% FBS, 1% antibiotics and 1%

glutamine, left to grow overnight at 37°C, 5%CO ₂ , then serum starved for 24 hours. The next morning, the medium was replaced with DMEM without FBS containing a 1/5,000 dilution of control or progastrin-selective polyclonal antibodies, followed by incubation at 37 ⁰ C, 5%CO ₂ . The medium containing antibodies was renewed 12 hours later. After 30 hours, cells were washed with PBS and lysed with the RNA extraction kit lysis buffer.

RNA preparation and RT-Quantitative PCR

RNA was prepared from SW480 cells using the RNeasy Protect Minikit (Qiagen France SA, 91974 Courtaboeuf, France). The quality and amount of RNA recovered were assessed using RNA pico Labchips (Agilent Technologies, Palo Alto, California). For reverse transcription, 2.5 μg of total RNA from each sample was pretreated with DNAse

RQl (Promega) for 30 min at 37°C and incubated with M-MLV (InVitrogen).

Quantitative PCR was carried out from 2μl of cDNA per sample, using the LightCycler FastStart DNA MasterPlus SYBR Green I kit (Roche Diagnostics). Expression of

GAPDH mRNA was used to calibrate RNA loading.

Cycle conditions for Quantitative PCR

- For c-myc and GAPDH amplification : denaturation during 10 min at 95°C, amplification for 50 cycles: 10 sec at 95 ⁰ C, 6 sec at 58°C, 13 sec at 72°C. The melting curve was 0 sec at 95 ⁰ C, 30 sec at 80 ⁰ C, 0 sec at 95°C and cooling was for 2 min at 40C°.

- For ICAT amplification : denaturation during 10 min at 95 ⁰ C, amplification during 50 cycles with: 10 sec at 95°C, 6 sec at 60 ⁰ C, 11 sec at 72°C. The melting curve was 0 sec at 95 ⁰ C, 30 sec at 8O ⁰ C, 0 sec at 7O ⁰ C and cooling was for 2 min at 40 ⁰ C.

- For cyclin Dl amplification: 10 min at 95°C, amplification during 50 cycles with: 10 sec at 95 ⁰ C, 6 sec at 7O ⁰ C, 11 sec at 72 ⁰ C. The melting curve was 0 sec at 95 ⁰ C, 30 sec at 8O ⁰ C, 0 sec at 95 ⁰ C and cooling was for 2 min at 4O ⁰ C.

Primers for Quantitative PCR on human sequences: GAPDH-sense: 5'GGTGGTCTCCTCTGACTTCAACAS' (SEQ ID N°33) GAPDH-antisense: 5'GTTGCTGTAGCCAAATTCGTTGTS' (SEQ ID N°34) ICAT-sense: 5'GCTCTGGTGCTTTAGTTAGGS' (SEQ ID N°35) ICAT-antisense: 5 'GCACTTGGTTTCTTTCTTTTC3 ' (SEQ ID N°36) c-Myc-sense: 5'CGTCTCCACACATCAGAGCACAAS' (SEQ ID N°37) c-Myc-antisense: 5'TCTTGGCAGCAGGATAGTCCTTS' (SEQ ID N°38) Cyclin Dl -sense: 5'-CCGTCCATGGGGAAGATC-S ' (SEQ ID N°39)

Cyclin Dl-antisense: 5'-ATGGCCAGCGGGAAGAC-S' (SEQ ID N°40)

GENERAL CONCLUSIONS ON EXAMPLES 1 AND 2

For the first time, the sufficient and necessary data are provided demonstrating that progastrin depletion of human colorectal tumor cells reverses tumorigenesis, as it induces differenciation and apoptosis of cells that have a constitutive beta-catenin / Tcf4 activity. These effects are specific of progastrin. This is shown in vitro on two cell lines (DLD-I and SW480) as well as in vivo with these same cell lines engrafted in nude mice and in a mouse model recapitulating the human tumorigenesis. Indeed, in this mouse model (APCδ14), the ape gene bears a mutation leading to activation of the beta-catenin / Tcf4 pathway and mice spontaneously develop adenomas and adenocarcinomas. These mice were treated in vivo with a siRNA targetting the peprogastrin mRNA. This resulted in a reversal of tumorigenesis, with differentiation and apoptosis leading to tumor shrinkage. It is shown that this treatment profoundly inhibits the constitutive beta- catenin / Tcf4 activity. Therefore, it is the very first time that tumorigenesis initiated by an ape gene mutation is shown to be reversed. At the molecular level, it is shown that depletion of progastrin has anti-tumor effects because it induces the re-expression of ICAT, an endogenous inhibitor of beta-catenin/Tcf4 transcriptional activity.

In human colorectal cancer, we demonstrate for the first time the occurrence of a correlation between hyperactivation of the beta-catenin / Tcf4 pathway, high levels of gastrin gene expression and low levels of ICAT expression. In addition, over expression of progastrin itself was shown in tumor samples where Tcf4 target gene levels are elavated. In spontaneously arising intestinal adenomas of APCδ14 mice, we also detected high levels of progastrin (and not of glycine-extended gastrin and amidated gastrin) and low ICAT expression. These parameters were reversed after treatment with gastrin gene selective siRNA.

Finally, data are provided demonstrating that blocking the effects of progastrin can also be achieved with antibodies specifically directed against progastrin, unable to bind glycine-extended gastrin or gastrin.

Proofs of concept and sufficient scientific data are thus provided for the use of either siRNA, shRNA or antibodies targetting progastrin to induce expression of ICAT and block and reverse colon tumorigenesis associated with a constitutive beta-catenin / Tcf4 activity.

References:

Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure.

Blache, P., M. van de Wetering, I. Duluc, C. Domon, P. Berta, J.N. Freund, H. Clevers, and P. Jay. 2004. SOX9 is an intestine crypt transcription factor, is regulated by the Wnt pathway, and represses the CDX2 and MUC2 genes. J Cell Biol.

166:37-41. Brown, D., U. Yallampalli, A. Owlia, and P. Singh. 2003. pp60c-Src Kinase mediates growth effects of the full-length precursor progastrinl-80 peptide on rat intestinal epithelial cells, in vitro. Endocrinology. 144:201-11. Chakladar, A., A. Dubeykovskiy, LJ. Wojtukiewicz, J. Pratap, S. Lei, and T.C. Wang. 2005. Synergistic activation of the murine gastrin promoter by oncogenic Ras

and beta-catenin involves SMAD recruitment. Biochem Biophys Res Commun.

336:190-6. Chung, J.H., and C. Eng. 2005. Nuclear-cytoplasmic partitioning of phosphatase and tensin homologue deleted on chromosome 10 (PTEN) differentially regulates the cell cycle and apoptosis. Cancer Res. 65:8096-100.

Ciccotosto, G.D., A. McLeish, KJ. Hardy, and A. Shulkes. 1995. Expression, processing, and secretion of gastrin in patients with colorectal carcinoma.

Gastroenterology. 109: 1142-53.

Cobb, S., T. Wood, J. Ceci, A. Varro, M. Velasco, and P. Singh. 2004. Intestinal expression of mutant and wild-type progastrin significantly increases colon carcinogenesis in response to azoxymethane in transgenic mice. Cancer.

100:1311-23. Cobb, S., T. Wood, L. Tessarollo, M. Velasco, R. Given, A. Varro, N. Tarasova, and P.

Singh. 2002. Deletion of functional gastrin gene markedly increases colon carcinogenesis in response to azoxymethane in mice. Gastroenterology. 123:516-

30. Coelho, S. M., M. Vaisman, and D.P. Carvalho. 2005. Tumour re-differentiation effect of retinoic acid: a novel therapeutic approach for advanced thyroid cancer. Curr

Pharm Des. 11:2525-31. Colnot, S., M. Niwa-Kawakita, G. Hamard, C. Godard, S. Le Plenier, C. Houbron, B.

Romagnolo, D. Berrebi, M. Giovannini, and C. Perret. 2004. Colorectal cancers in a new mouse model of familial adenomatous polyposis: influence of genetic and environmental modifiers. Lab Invest. 84:1619-30.

Daniels, D.L., and W.I. Weis. 2002. ICAT inhibits beta-catenin binding to Tcf/Lef- family transcription factors and the general coactivator p300 using independent structural modules. MoI Cell. 10:573-84. Delcommenne, M., C. Tan, V. Gray, L. Rue, J. Woodgett, and S. Dedhar. 1998.

Phosphoinositide-3-OH kinase-dependent regulation of glycogen synthase kinase

3 and protein kinase B/AKT by the integrin-linked kinase. Proc Natl Acad Sci U SA. 95:11211-6.

Ferrand, A., C. Bertrand, G. Portolan, G. Cui, J. Carlson, L. Pradayrol, D. Fourmy, M.

Dufresne, T.C. Wang, and C. Seva. 2005. Signaling pathways associated with colonic mucosa hyperproliferation in mice overexpressing gastrin precursors.

Cancer Res. 65:2770-7. Finley, G.G., R. A. Koski, M.F. Melhem, J.M. Pipas, and A.I. Meisler. 1993. Expression of the gastrin gene in the normal human colon and colorectal adenocarcinoma.

Cancer Res. 53:2919-26. Fodde, R., R. Smits, and H. Clevers. 2001. APC, signal transduction and genetic instability in colorectal cancer. Nat Rev Cancer. 1 :55-67. Gottardi, C.J., and B.M. Gumbiner. 2004. Role for ICAT in beta-catenin-dependent nuclear signaling and cadherin functions. Am J Physiol Cell Physiol. 286:C747-

56. Hollande, F., A. Imdahl, T. Mantamadiotis, G.D. Ciccotosto, A. Shulkes, and G.S.

Baldwin. 1997. Glycine-extended gastrin acts as an autocrine growth factor in a nontransformed colon cell line. Gastroenterology. 113:1576-88.

Hollande, F., DJ. Lee, A. Choquet, S. Roche, and G.S. Baldwin. 2003. Adherens junctions and tight junctions are regulated via different pathways by progastrin in epithelial cells. J Cell Sd. 116:1187-97.

Hollande, F., A. Shulkes, and G.S. Baldwin. 2005. Signaling the junctions in gut epithelium. Sd STKE. 2005:pel3.

Jain, R.N., CS. Brunkan, CS. Chew, and L.C. Samuelson. 2006. Gene expression profiling of gastrin target genes in parietal cells. Physiol Genomics. 24: 124-32. Janssen, K.P., P. Alberici, H. Fsihi, C. Gaspar, C. Breukel, P. Franken, C. Rosty, M.

Abal, F. El Marjou, R. Smits, D. Louvard, R. Fodde, and S. Robine. 2006. APC and Oncogenic KRAS Are Synergistic in Enhancing Wnt Signaling in Intestinal

Tumor Formation and Progression. Gastroenterology. 131: 1096-109. Judge, A.D., V. Sood, J.R. Shaw, D. Fang, K. McClintock, and I. MacLachlan. 2005.

Sequence-dependent stimulation of the mammalian innate immune response by synthetic siRNA. Nat BiotechnoL 23:457-62.

Kochman, MX., J. DelValle, CJ. Dickinson, and CR. Boland. 1992. Post-translational processing of gastrin in neoplastic human colonic tissues. Biochem Biophys Res

Commun. 189:1165-9.

Koh, T. J., CJ. Bulitta, J. V. Fleming, GJ. Dockray, A. Varro, and T.C Wang. 2000. Gastrin is a target of the beta-catenin/TCF-4 growth-signaling pathway in a model of intestinal polyposis. J Clin Invest. 106:533-9. Koh, T.J., GJ. Dockray, A. Varro, RJ. Cahill, CA. Dangler, J. G. Fox, and T.C. Wang.

1999. Overexpression of glycine-extended gastrin in transgenic mice results in increased colonic proliferation. J Clin Invest. 103: 1119-26. Konturek, P. C, W. Bielanski, SJ. Konturek, A. Hartwich, P. Pierzchalski, M. Gonciarz,

K. Mariicz, T. Starzynska, M. Zuchowicz, Z. Darasz, J.P. Gotze, J.F. Rehfeld, and E.G. Hahn. 2002. Progastrin and cyclooxygenase-2 in colorectal cancer. Dig

Dis ScL 47: 1984-91.

Korinek, V., N. Barker, PJ. Morin, D. van Wichen, R. de Weger, K. W. Kinzler, B. Vogelstein, and H. Clevers. 1997. Constitutive transcriptional activation by a beta-catenin-Tcf complex in APC-/- colon carcinoma. Science. 275: 1784-7. Koyama, T., K. Tago, T. Nakamura, S. Ohwada, Y. Morishita, J. Yokota, and T.

Akiyama. 2002. Mutation and expression of the beta-catenin-interacting protein

ICAT in human colorectal tumors. Jpn JClin Oncol. 32:358-62. Lallemand-Breitenbach, V., J. Zhu, S. Kogan, Z. Chen, and H. de The. 2005. Opinion: how patients have benefited from mouse models of acute promyelocytic leukaemia. Nat Rev Cancer. 5:821-7. Lian, Z., and A. Di Cristofano. 2005. Class reunion: PTEN joins the nuclear crew.

Oncogene. 24:7394-400. Litvak, D.A., M.R. Hellmich, K. Iwase, B.M. Evers, J. Martinez, M. Amblard, and CM.

Townsend, Jr. 1999. JMVl 155: a novel inhibitor of glycine-extended progastrin- mediated growth of a human colon cancer in vivo. Anticancer Res. 19:45-9. Marotta, A., K. Parhar, D. Owen, S. Dedhar, and B. SaIh. 2003. Characterisation of integrin-linked kinase signalling in sporadic human colon cancer. Br J Cancer. 88: 1755-62.

Marques, J.T., T. Devosse, D. Wang, M. Zamanian-Daryoush, P. Serbinowski, R.

Hartmann, T. Fujita, M.A. Behlke, and B.R. Williams. 2006. A structural basis for discriminating between self and nonself double-stranded RNAs in mammalian cells. Nat Biotechnol. Morin, P.J., A.B. Sparks, V. Korinek, N. Barker, H. Clevers, B. Vogelstein, and K. W.

Kinzler. 1997. Activation of beta-catenin-Tcf signaling in colon cancer by mutations in beta-catenin or APC. Science. 275:1787-90. Mulrooney, J., K. Foley, S. Vineberg, M. Barreuther, and L. Grabel. 2000.

Phosphorylation of the betal integral cytoplasmic domain: toward an understanding of function and mechanism. Exp Cell Res. 258:332-41.

Nemeth, J., B. Taylor, S. Pauwels, A. Varro, and GJ. Dockray. 1993. Identification of progastrin derived peptides in colorectal carcinoma extracts. Gut. 34:90-5. Nordlund MS, Fermer C, Nilsson O, Warren DJ, and E. Paus. 2007. Production and

Characterization of Monoclonal Antibodies for Immunoassay of the Lung Cancer Marker proGRP. Tumour Biol. 28(2): 100-10.

Ottewell, P.D., A. Varro, G.J. Dockray, CM. Kirton, AJ. Watson, T.C. Wang, R.

Dimaline, and D.M. Pritchard. 2005. COOH-terminal 26-amino acid residues of progastrin are sufficient for stimulation of mitosis in murine colonic epithelium in vivo. Am J Physiol Gastrointest Liver Physiol. 288:G541-9. Ottewell, P.D., AJ. Watson, T.C. Wang, A. Varro, GJ. Dockray, and D.M. Pritchard.

2003. Progastrin stimulates murine colonic epithelial mitosis after DNA damage.

Gastroenterology. 124: 1348-57. Pear, W.S., G.P. Nolan, MX. Scott, and D. Baltimore. 1993. Production of high-titer helper-free retroviruses by transient transfection. Proc Natl Acad Sci USA. 90:8392-6.

Radtke, F., and H. Clevers. 2005. Self-renewal and cancer of the gut: two sides of a coin.

Science. 307:1904-9. Sekiya, T., T. Nakamura, Y. Kazuki, M. Oshimura, K. Kohu, K. Tago, S. Ohwada, and

T. Akiyama. 2002. Overexpression of Icat induces G(2) arrest and cell death in

tumor cell mutants for adenomatous polyposis coli, beta-catenin, or Axin. Cancer Res. 62:3322-6. Seva, C, CJ. Dickinson, and T. Yamada. 1994. Growth-promoting effects of glycine- extended progastrin. Science. 265:410-2. Siddheshwar, R.K., J.C. Gray, and S.B. Kelly. 2001. Plasma levels of progastrin but not amidated gastrin or glycine extended gastrin are elevated in patients with colorectal carcinoma. Gut. 48:47-52.

Singh, P., A. Owlia, A. Varro, B. Dai, S. Rajaraman, and T. Wood. 1996. Gastrin gene expression is required for the proliferation and tumorigenicity of human colon cancer cells. Cancer Res. 56:4111-5.

Singh, P., M. Velasco, R. Given, A. Varro, and T.C. Wang. 2000. Progastrin expression predisposes mice to colon carcinomas and adenomas in response to a chemical carcinogen. Gastroenterology. 119:162-71.

Singh, P., Z. Xu, B. Dai, S. Rajaraman, N. Rubin, and B. Dhruva. 1994. Incomplete processing of progastrin expressed by human colon cancer cells: role of noncarboxyamidated gastrins. Am J Physiol. 266:G459-68. Stepan, V.M., M. Sawada, A, Todisco, and CJ. Dickinson. 1999. Glycine-extended gastrin exerts growth-promoting effects on human colon cancer cells. MoI Med. 5: 147-59. Tago, K., T. Nakamura, M. Nishita, J. Hyodo, S. Nagai, Y. Murata, S. Adachi, S.

Ohwada, Y. Morishita, H. Shibuya, and T. Akiyama. 2000. Inhibition of Wnt signaling by ICAT, a novel beta-catenin-interacting protein. Genes Dev. 14:1741-9.

Tan, C, P. Costello, J. Sanghera, D. Dominguez, J. Baulida, A.G. de Herreros, and S. Dedhar. 2001. Inhibition of integrin linked kinase (ILK) suppresses beta-catenin-

Lef/Tcf-dependent transcription and expression of the E-cadherin repressor, snail, in APC-/- human colon carcinoma cells. Oncogene. 20:133-40. van de Wetering, M., E. Sancho, C. Verweij, W. de Lau, I. Oving, A. Hurlstone, K. van der Horn, E. Batlle, D. Coudreuse, A.P. Haramis, M. Tjon-Pon-Fong, P. Moerer, M. van den Born, G. Soete, S. Pals, M. Eilers, R. Medema, and H. Clevers. 2002.

The beta-catenin/TCF-4 complex imposes a crypt progenitor phenotype on colorectal cancer cells. Cell. I l l :241-50. van Es, J.H., and H. Clevers. 2005. Notch and Wnt inhibitors as potential new drugs for intestinal neoplastic disease. Trends MoI Med. 11 :496-502. Van Solinge, W. W., F.C. Nielsen, L. Friis-Hansen, U.G. Falkmer, and J.F. Rehfeld.

1993. Expression but incomplete maturation of progastrin in colorectal carcinomas. Gastroenterology. 104:1099-107. Wang, T.C., TJ. Koh, A. Varro, RJ. Cahill, CA. Dangler, J.G. Fox, and GJ. Dockray.

1996. Processing and proliferative effects of human progastrin in transgenic mice. J Clin Invest. 98: 1918-29.

Wang, Z. Y., and Z. Chen. 2000. Differentiation and apoptosis induction therapy in acute promyelocytic leukaemia. Lancet Oncol. 1:101-6. Wray CJ., RiIo H.L., and S.A. Ahmad. 2004. Colon cancer angiogenesis and antiangiogenic therapy. Expert Opin Investig Drugs. 13(6):631-41.