WO2005014637A1 - Apoptosis inhibitors - Google Patents

Abstract

Ezrin mutated at the CD95 binding site reduces CD95-mediated apoptosis and controls metastasis and aggressiveness in tumor cells.

Description

APOPTOSIS INHIBITORS

The present invention relates to inhibitors of apoptosis, uses thereof and formulations and medicaments comprising them.

Ezrin is a member of the ERM family, comprising ezrin, radixin and moesin, which serve to provide a regulated linkage between the actin cytoskeleton and the plasma membrane. In this respect, it is known that ezrin plays a part in CD95-mediated apoptosis. Its role remains a mystery, however.

The CD95 (Fas/APO-1) linkage to the actin cytoskeleton through ezrin is a major requirement for susceptibility to the CD95-mediated apoptosis in CD4+ T cells. However, there is no evidence for a direct interaction between these two proteins.

A major requirement for susceptibility to CD95(APO-l/Fas)-mediated apoptosis in CD4+ T cells is CD95 polarisation on cell uropods, due to the CD95 linkage to the actin cytoskeleton through ezrin [1]. Ezrin, Radixin and Moesin (ERM) are closely related proteins involved in cellular polarisation and in various cellular functions [2-4]. ERM are found in micro villi, filopodia, membrane ruffles, and cell-to-cell contact sites, where they co-localise and associate with F-actin. [5-10]. ERM are members of the erythrocyte protein 4.1 super-family characterised by a ~300-residue globular N- terminal domain highly conserved in the ERM family (FERM domain) [3, 11-14]. ERM interact directly with various membrane proteins, such as CD43, CD44, ICAM-1, -2, and 3, through their FERM amino-terminal domain (reviewed in [3]). These proteins can also indirectly bind to H⁺/Na⁺ exchanger 3 (NHE3) via the cytoplasmic protein EBP50 [15-16]. ERM are present both in an inactive/closed and in an active/opened form [9, 14] and a direct consequence of ERM activation is their recruitment to the plasma membrane allowing the ERM linkage to the membrane molecules [17-18]. The ERM/membrane protein interaction is stabilised at the plasma membrane by the ERM association to phosphatidylinositol (4,5)-bisphosphate [19-20] and regulated by Tyrosine phosphorylation in Y145 and Y353 [21-26]. However, while the ERM domains implicated in actin binding are known [27- 30], the specific sites involved in the binding to membrane proteins are much less well characterised. Particularly, the specific amino acid sequence included in the N-terminal FERM domain of ERM involved in binding to the various membrane molecules is not known. Recently, the three-dimensional structure of the ezrin FERM domain has been described [31-32]. The fold consists of three lobes in a trefoil, globally similar to structures reported previously for ERM proteins, with the PIP₂ binding domain located on the first and the third lobe [33-35].

We have previously shown (i) ezrin/CD95 association in CD95-expressing cells (i.e. lymphoblastoid CD4+ T cells and day-6 activated lymphocytes); (ii) that moesin is not involved in the association with Fas and (iii) the biological in vivo relevance of this association through treatment with ezrin antisense oligonucleotides that markedly inhibited susceptibility to CD95-mediated apoptosis of human lymphoid cells [1]. Thus, the previous data, although providing evidence of the biological relevance of the CD95/ezrin interaction, provided no information on the specificity of ezrin in the binding to CD95 (i.e. ezrin vs. radixin), nor on the epitopes involved in the ezrin binding to CD95, nor any evidence for a direct interaction between ezrin and CD95, nor anything on the in vivo relevance of this direct interaction.

The role of the membrane protein interaction with actin filaments in conferring polarised behaviour on cells is important in the understanding of cell biology [39, 41]. The ERM family appears to be the most involved protein family in this function [3, 42- 44]. We have previously demonstrated that the association between CD95 and ezrin is an important feature of susceptibility to CD95-mediated apoptosis, and that this association has relevant biological significance inasmuch as treatment with ezrin antisense oligonucleotides strongly inhibited the susceptibility to CD95 mediated apoptosis of live human lymphoid cells [1].

We have now surprisingly found that cells transformed with an ezrin mutant show resistance to CD95-mediated apoptosis and that cancer cells transformed with the mutant show a substantially reduced ability to metastasise. Thus, in a first aspect, the present invention provides a nucleic acid sequence substantially encoding ezrin, wherein the portion encoding the CD95 binding site carries one or more substitutions sufficient to inhibit or prevent CD95 binding to the translation product of the sequence.

We have shown that CD95 binds directly to the ezrin FERM domain, while it is unable to bind radixin. Moreover, GST pull-down assays showed that the region responsible for CD95 binding is located in the middle lobe of ezrin FERM domain. Transfection with a recombinant GFP-ezrin, mutated in CD95-binding epitope, induced in transfected cells both inhibition of Fas-mediated apoptosis and the suppression of ezrin/CD95 co-localisation. Our data provide evidence for a direct ezrin/CD95 association and permitted mapping of the ezrin CD95-binding site between amino acids 149 and 168. In this region ERM display 60-65% identity versus 86% of the whole FERM domain.

As stated above, we have established that some or all of residues 149-168 inclusive are essential for effective binding to CD95. Replacing all or part of this region with the corresponding sequence in moesin or radixin eliminates the ability of the hybrid molecule to bind CD95. In addition, cells transformed or transfected with DNA or RNA suitable to express this hybrid molecule show a substantially reduced tendency to respond to CD95-mediated apoptosis. In addition, the aggressiveness of transformed tumour cells is substantially reduced, both in terms of tumour growth and metastasis.

The residues in the sequence 149-168 which vary between ezrin and moesin are residues 149-152, 154, 159, 160, 164 and 165. Although it may be sufficient to substitute any single one of these residues, it is generally preferred to substitute a plurality of the residues, and preferably all of the residues. It is preferred that the substitutions correspond to the residues in either radixin or moesin, and it is particularly preferred that the substitutions, insofar as any are made, correspond to either radixin or moesin. It is not necessary that the substitutions correspond to either moesin or radixin directly, provided that the translation product substantially retains the three-dimensional conformation of the ERM family. In general, it is preferred that the substitutions either conform to the residues present in moesin or radixin, or are conservative replacement therefor.

Likewise, while it is preferred to retain the overall ezrin sequence outside of the 149-168 sequence, it is possible to vary the remainder of the ezrin sequence by deletion, insertion, addition, substitution or inversion. It is preferred to minimise any such mutation, restricting to those that might be useful or that occur in nature, for example. Where possible, it is preferred to keep mutations as close to the original sequence as possible, and/or to use conservative substitution.

The present invention extends to vectors and, especially, expression vectors comprising the nucleic acid sequences defined above. It will be appreciated that expression vectors comprising DNA of the present invention will comprise appropriate control sequences, such as promoters, initiation sequences, termination sequences and enhancers, for example.

The extension further envisages cells transformed by the sequences of the invention and especially expression vectors of the invention. Such cells may be obtained by any suitable transformation means, such as by a retrovirus or the calcium phosphate protocol, for example.

More particularly, the sequences of the present invention may be used to transform or transfect cells in vivo in order to treat a medical condition, wherein the condition is characterised by unbalanced Fas-mediated apoptosis. Such conditions include, for example, Lyell's syndrome, GVHD, multiple sclerosis, viral hepatitis and AIDS. In addition, cancerous tissue may also be transformed or transfected to help prevent spread of the condition.

In order to treat a patient, conventional methods of gene therapy may be employed, and it is also sufficient to transform the target cells simply with an expression plasmid. Nevertheless, a retroviral vector, such as the adenovirus vector, is generally preferred, and may be targeted, or administered systemically.

The mutated ezrin of the invention may be administered directly to cells in need thereof, such as by liposomes, or may be produced in the target cells by expression from an encoding nucleic acid introduced into the cells, e.g. from a viral vector. The vector may be targeted to the specific cells to be treated, or it may contain regulatory elements which are activated, optionally selectively, by the target cells.

Nucleic acid encoding the substance may thus be used in methods of gene therapy, for instance in treatment of individuals, e.g. with the aim of preventing or curing (wholly or partially) a disorder.

Vectors such as viral vectors have been used in the prior art to introduce nucleic acid sequences into a wide variety of different target cells. Typically, the vectors are exposed to the target cells so that transfection can take place in a sufficient proportion of the cells to provide a useful therapeutic or prophylactic effect from the expression of the desired peptide. The transfected nucleic acid may be permanently incorporated into the genome of each of the targeted cells, providing a long lasting effect, or alternatively the treatment need to be repeated periodically.

A variety of vectors, both viral and plasmid, are known in the art, such as are exemplified in US-A-5,252,479 and WO 93/07282. In particular, a number of viruses have been used as gene transfer vectors, including papovaviruses, such as SV40, vaccinia virus, herpesviruses, including HSV and EBV, and retroviruses. In addition, many gene therapy protocols in the prior art have used disabled murine retroviruses.

As an alternative to the use of viral vectors in gene therapy, other known methods of introducing nucleic acid into cells include mechanical techniques, such as microinjection, transfer mediated by liposomes, and receptor-mediated DNA transfer. Receptor-mediated gene transfer, in which the nucleic acid is linked to a protein ligand via polylysine, with the ligand being specific for a receptor present on the surface of the target cells, is an example of a technique for specifically targeting nucleic acid to particular cells.

There is an extensive body of literature in which naked DNA or RNA encoding a therapeutic polypeptide or peptide is administered, for expression in vivo, to produce a therapeutic effect. DNA or RNA may be administered by injection and this may be with or with transfection-facilitating materials, such as liposomes or cationic lipids.

Thus, the present invention provides a method of controlling CD95-mediated apoptosis in an individual or for controlling tumour cells in an individual, comprising transforming or transfecting tumour cells in said individual with a sequence according to present invention. Preferably, the transformation or transfection is selected from the group consisting of: receptor-mediated or virus-mediated transfection; microinjection; and by means of a gene gun, wherein the nucleotides are coated onto metallic microprojectiles.

The present invention also provides a transformed or transfected host cell, transformed or transfected with the mutant ezrin sequence. Preferably, the host cell expresses the mutant ezrin protein.

Surprisingly, it is not necessary to block expression of native ezrin, as it has been found that simple transformation with the mutated ezrin of the invention is sufficient to block or repress Fas mediated apoptosis.

We have now shown that the ezrin/CD95 association occurs through direct binding between the ezrin FERM domain and the CD95 cytoplasmic domain. Notably, although a deal of ezrin partners have been identified (reviewed in 3), the mechanisms by which ezrin exerts its functions are still unclear. In fact, while ezrin binding domains to actin are well characterised [27-30], the ezrin epitopes involved in the linkage to membrane proteins are poorly known. The results of our studies not only showed that the intracellular domain of CD95 bound directly to the ezrin FERM domain but, using the GST pull-down assay, we were able to show that a small epitope contained in the ezrin FERM domain is involved in binding CD95. In vitro analysis, using various deletion mutants, revealed that determinants of the ezrin binding to CD95 were located within amino acids 149-168 in the middle lobe of ezrin FERM domain.

The Ezrin protein sequence is well known and is available from NCBI at [http://www.ncbi.nlm.nih.gov/enτjez/querv.fca?db=gene&cmd==Retrieve&dopt=^:Graphi cs&list uids=74301. The DNA sequence of the Ezrin protein is given in SEQ ID NO. 26, whilst the full 586 amino acid sequence of the Ezrin protein is given in SEQ ID NO.

27.

Ezrin CD95 direct interaction was shown by performing a GST pull-down assay using GST-ezrin deletion fragments together with in vitro translated CD95 or constitutive CD95, obtained from total extracts of human cell lines expressing high levels of a functional CD95. Moreover, together with the previous evidence that moesin is not involved in the linkage to CD95, we have newly shown that radixin, while not expressed in lymphocytes [1, 37], did not co-immunoprecipitate with CD95 in radixin- expressing cells and did not interact with CD95 in the two hybrid assay. This suggested that ezrin is the specific CD95 partner between the ERM family proteins.

These results were supported by co-immunoprecipitation data, as CD95- apoptosis induction experiments in two different human cell lines, each transfected with a GFP-ezrin construct mutated at the epitope responsible for CD95 binding. These data showed that GFP-tagged ezrin did not co-immunoprecipitate with CD95, and cells expressing the mutated ezrin were strongly and specifically protected from CD95- mediated apoptosis, while no protection was observed after treatment of transfected cells with TNF or Staurosporin.

It has been recently shown that a CD95-to-actin linkage is crucial in both conferring susceptibility to CD95-mediated apoptosis [1] and for the early steps of the CD95 signalling pathway, including DISC formation and CD95 internalisation [45]. Our experiments showed that the mutation inhibited the association of CD95 with the mutated ezrin, through inhibition of both ezrin-to-CD95 binding and co-localisation. Without being bound by theory, it may also be that, through this inhibition, CD95 no longer interacts, or is no longer capable of interacting, with actin, thus possibly resulting in the inhibition of DISC formation as well as CD95 internalisation. Recent findings show that, as an early event, disialoganglioside GD3 associates with ezrin and co-localises with ezrin in uropods of human lymphocytes, following the CD95 triggering [46], supporting this possibility. Moreover, dramatic effects on ezrin and the actin cytoskeleton occur following CD95 triggering [47] and disruption of cytoskeleton may induce apoptosis via CD95 activation [48].

Our data support a key role of ezrin in linking CD95 to actin and, as a consequence, the involvement of actin in CD95 multiple cascades.

Our results show that (i) the ezrin/CD95 linkage is ezrin-specific in that radixin does not detectably bind to CD95 and moesin is not involved in the linkage to CD95 in CD95-susceptible cells [1]; (ii) the ezrin region responsible of the CD95 binding consists of 20 amino acids included in the median lobe of the ezrin FERM domain, where ezrin shows relevant differences to both moesin and radixin, supporting ezrin specificity for CD95 binding; and (iii) ezrin mutations on the mapped CD95 binding epitope result in the loss of CD95 and ezrin co-localisation and co-immunoprecipitation, consistent with the observed inhibition of sensitivity to CD95-mediated apoptosis, in two different CD95-prone human cell lines. Altogether, the results show that direct CD95/ezrin binding is a key requirement for susceptibility to CD95-mediated apoptosis, suggesting that ezrin-mediated CD95-to-actin binding may have a role in actin- dependent DISC formation and CD95 internalisation [45].

These experiments, together with the GST pull-down experiments, allowed us to map the CD95 binding site on ezrin to between amino acids 149 and 168. Notably, in this sequence, ezrin, radixin and moesin show only a 60-65% of identity between the three proteins (Fig 6 A), while ERM proteins show a ~ 86% of identity in the whole FERM domain [14]. This epitope is located within the ezrin FERM domain, in a region that is comprised between amino acids 12-115 and 233-310, reported as the two determinants of PIP2 binding domain (Fig 6 B) [20]. Altogether the results of our study identify an epitope that confers on ezrin specificity for CD95-binding, and suggests that the involvement of actin filaments in CD95-mediated apoptosis may occur through a direct CD95/ezrin/actin linkage. This linkage may have a key role both in conferring CD95 susceptibility to the cells and in allowing the early steps of the CD95-mediated apoptosis.

In the following Example, the accompanying drawings and representations are referred to, in which:

Figure 1 shows CD95 association with ERM.

A) Detection of ERM on CD95 immunoprecipitates. Western blotting for radixin (Lanes 1-2) on: total extracts (lane 1), and CD95 immunoprecipitate (lane 2) from CEM-VBL100 cells. Western blotting for ezrin on CD95 immunoprecipitates from CEM-VBL100 total extracts (lane 3). Western blotting for ezrin on A+G utilised for immunoprecipitation plus Mouse

(as isotype control antibody) (lane 4).

B) Ezrin, Radixin and CD95 functional domains. The drawing shows ezrin (representative of ERM proteins) tyrosine phosphorylation sites (Y145 and Y353 for ezrin, and Y145 for radixin) placed respectively in the FERM domain (oval) and α- helical region (white bar), and the carboxy-terminal threonine phosphorylation site (T567) in the ezrin molecules (upper panel). Moreover, the CD95 extracellular domain (striped bar), intracytoplasmic domain (black box) and transmembrane region (white box) are shown (lower panel).

C) Ezrin, radixin and CD95 fragments used for the two-hybrid assay. The drawing shows the radixin, ezrin and CD95 fragments used to perform the two hybrid assay. Particularly, the whole radixin FERM domain and Ezrin (1-362) including the whole ezrin FERM domain plus the α-helical fragment with the second tyrosine phosphorylation site (Y353) lacking in radixin (upper panel), and the intracytoplasmic region of CD95 (185-335) (lower panel) are shown.

Figure 2 shows GST-ezrin fragments used to test the ezrin/CD95 interaction. The picture shows the fusion proteins constructs used in the GST pull-down assay containing the GST epitope tag. A schematic representation, of both the three GST- ezrin fragments obtained from the N-terminal domain (GST-ezrin 149-242, GST-ezrin 149-200, GST-ezrin 149-168, GST-ezrinl69-242) and the C-terminal domain (GST- ezrin 508-586), is shown.

Figure 3 shows the binding of recombinant ezrin constructs.

A) CD95 immunoblot (Recombinant CD95). Incubation of the GST-ezrin fusion proteins with the CD95 Death Domain recombinant protein. Lane 1 Jurkat total extracts alone; lane 2 CD95 cytoplasmic domain recombinant protein; lane 3 GST-ezrin 149- 242; lane 4 GST-ezrin 149-200; lane 5 GST-ezrin 149-168; lane 6 GST-ezrin 169-242; lane 7 GST-ezrin 508-586; lane 8 GST alone. Bound proteins were separated by 12% SDS-PAGE. The presence of CD95 was detected by immuno-blotting with an anti CD95 antibody. The position of the molecular size markers (kDa) are indicated by lines and numbers.

B) CD95 immunoblot (Jurkat cells total extracts). Incubation of the GST fusion proteins with the Jurkat total extracts. Lane 1 Jurkat total extracts alone; lane 2 GST- ezrin 149-242; lane 3 GST-ezrin 149-200; lane 4 GST-ezrin 149-168; lane 5 GST-ezrin 169-242; lane 6 GST-ezrin 508-586; lane 7 GST alone. Bound proteins were separated on 12% SDS-PAGE. The presence of CD95 was detected by immunoblotting with an anti CD95 antibody. The position of the molecular size markers (kDa) are indicated by lines and numbers.

C) Actin immunoblot (Jurkat cells total extracts). Incubation of the GST fusion proteins with the Jurkat total extracts. Lane 1 Jurkat total extracts alone; lane 2 GST- ezrin 149-242; lane 3 GST-ezrin 149-200; lane 4 GST-ezrin 149-168; lane 5 GST-ezrin 169-242; lane 6 GST-ezrin 508-586; lane 7 GST alone. Bound proteins were separated on 12% SDS-PAGE. The presence of actin was detected by immuno-blotting with an anti-actin antibody. The position of the molecular size markers (kDa) are indicated by lines and numbers. Figure 4 shows susceptibility to apoptosis of transfected cells.

A) Schematic representation of GFP-Ezrin/moesin/ezrin (GFP-ez/moe) chimeric fusion protein. Ezrin aa 147-173 (SEQ ID NO. 28), (i.e. containing CD95 binding epitope) and moesin aa 147-173 (SEQ ID NO. 29), replacing the same ezrin . amino acids in the chimeric protein, are compared. Differences are framed.

B) Biparametric analysis of GFP-transfected HeLa cells after staining with annexin V-alexa 568 performed in living cells. First column: untransfected cells; second column: GFP alone transfected cells; third column: WT GFP-ezrin transfected cells; fourth column: GFP-ez/moe transfected cells. In the first row: control untreated cells; in the second row: α-Fas-treated cells; in the third and fourth rows: cells treated with TNF-α or STS, respectively. In the quadrants I and IV non-apoptotic cells are shown: transfected cells (single GFP positive, quadrant I) or non-transfected cells (GFP/annexin V double negative, quadrant IV). In the quadrants II and III apoptotic cells are shown: events in the quadrant II of the dot plots correspond to transfected cells undergoing apoptosis (double GFP/annexin V positive). In the quadrant III non- transfected apoptotic cells (GFP negative/annexin V positive) are included. Numbers in the quadrants represent the percentage of apoptotic cells obtained in one experiment representative of four.

C) Biparametric analysis of GFP-transfected Hut78 cells after staining with annexin V-alexa 568 performed in living cells. First column: untransfected cells; second column: GFP-only transfected cells; third column: WT GFP-ezrin transfected cells; fourth column: GFP-ez/moe transfected cells. In the first row: control untreated cells; in the second row: α-Fas-treated cells; in the third and fourth rows: cells treated with TNF-α or STS, respectively. In the quadrants I and IV non-apoptotic cells are shown: transfected cells (single GFP positive, quadrant I) or non-transfected cells (GFP/annexin V double negative, quadrant W). In quadrants II and III apoptotic cells are shown: events in the quadrant II of the dot plots correspond to transfected cells undergoing apoptosis (double GFP/annexin V positive). In quadrant III non-transfected apoptotic cells (GFP negative/annexin V positive) are included. Numbers in the quadrants represent the percentage obtained in one experiment representative of four. Figure 5. Co-immunoprecipitation and co-localization of GFP -tagged-ezrin and CD95 .

A) Western blotting for GFP on CD95 immunoprecipitates from: Hut78 untransfected cells (lane 1), GFP-ezrin Hut78 cells (lane 2), GFP-ez/moe Hut78 cells, (lane 3) (upper panel); Western blotting for CD95 on GFP immunoprecipitates from: Hut78 untransfected cells (lane 1), GFP-ezrin Hut78 cells (lane 2), GFP-ez/moe Hut78 cells (lane 3) (lower panel).

B) Confocal analysis of CD95 (red) and GFP-ezrin (green) co-localisation in HeLa cells. Upper panels: GFP-ez/moe transfected cells; Central panels: wilde-type GFP- ezrin transfected cells; Lower panels: untransfected cells (ezrin-FITC). Data are representative of three independent experiments.

C) IVM analysis of CD95 (red) and GFP-ezrin (green) co-localisation in Hut78 cells. Left panel, GFP-ez/moe transfected cells; Central panel, wilde-type GFP-ezrin transfected cells; Right panel untransfected cells (ezrin-FITC). Yellow/orange areas indicate ezrin and CD95 co-localisation. Only double staining merge images of one representative experiment of three are shown.

Figure 6 shows the Ezrin FERM Domain.

A) A schematic representation of the CD95 binding epitope within the ezrin FERM domain. The fragment responsible of the ezrin binding to CD95 is shown and the amino acid sequence has been compared to those of moesin and radixin in the same fragment. In Bold Italic the amino-acids that differ between ezrin and moesin, and those underlined are the amino-acids that differ between ezrin and radixin. The analysis has shown only a 60-65% identity between the three proteins in the epitope responsible of the ezrin binding to CD95.

B) Ezrin FERM domain structure where the CD95 and PIP2 binding domains are highlighted. Notably, the ezrin fragment that binds to CD95 is included between the two accounting for the ezrin binding to P_P₂. EXAMPLE 1

Western blotting and immunoprecipitation

Whole extracts from Hela, Jurkat or CEM-VBLIOO were resuspended in SDS sample buffer, denatured by boiling, separated by 12% SDS-PAGE and analysed by Western Blot. CD95, ezrin, radixin and actin were detected with an anti-CD95 pAb (C20, Santa Cruz USA), an anti-ezrin mAb (Transduction Laboratories, USA), an anti- radixin pAb (C15, Santa Cruz), and an anti-actin mAb (Chemicon, USA), and visualised with peroxidase anti-Ig followed by ECL detection (Pierce, USA). CD95 protein was immunoprecipitated from pre-cleared cell lysates with an anti-CD95 antibody (clone DX2, Calbiochem, USA) overnight at 4°C in the presence of protein A+G Sepharose beads (Sigma- Aldrich, USA). Immunoprecipitated beads were washed four times in Lysis buffer, resuspended in SDS sample buffer and resolved by 10% SDS-PAGE. Then, proteins were transferred to nitrocellulose membrane and analysed by Western blotting.

Mammalian Two-Hybrid assay cDNA fragments for human ezrin and CD95 were amplified by RT-PCR, using total RNA extracted from Jurkat cells as template. cDNA for human radixin was obtained from CEM-VBL100 cells as template. The primers utilised were the following: CD95_185-335, GTCGAGCCACTAATTGTTTGGGT (SEQ ID NO 1) and CTGCAGCTAGACCAAGCTTTGG (SEQ ID NO 2); ezri_nι.₃₆₂: GTCGACACTCACCAGAAACCGA (SEQ ID NO 3) and TCTAGACTCTGCCTTCTTTGTC (SEQ ID NO 4). Primers for radixin FERM domain were the following: GGATCCAATTCGGCACACGAGACA (SEQ ID NO 5) and AAGCTTAGGCTTCCTTCTTCGCAT (SEQ ID NO 6). PCR products were cloned in the pGEM-T vector (Promega, USA), and then CD95 fragment was digested with Sail- Pstl and cloned into the Sail and Pstl sites of pM GAL4-BD cloning vector (Clontech, USA). Ezrin and radixin PCR products were digested with Sall-Xbal and BamHl-Hindlll and cloned into the same sites of pVP16-AD cloning vector (Clontech, USA). The resulting plasmids were designed as pM-CD95, pVP16-ezrin and pVP16- radixin. pG5CAT was the reporter vector containing the CAT gene downstream of five consensus GAL4 binding sites. Both the cloning vectors and pG5CAT were supplied by Mammalian MATCHMAKER two-Hybrid Assay Kit (Clontech, USA). The interaction between CD95 and ezrin was assayed by measuring CAT gene expression. As control plasmids pM53 (vector expressing a fusion of the GAL4 DNA-BD to the mouse protein pM53), pVP16T (vector expressing a fusion of the VP16 AD to the SV40 large T-antigen), pVP16-CP (vector expressing a fusion of the VP 16 AD to a viral coat protein) or pM and pVP16, were utilised. All the control plasmids were supplied by the Kit manifacturer. Plasmids are detailed in Table 1. The correctness of the ORFs was confirmed in all cases by sequencing.

Cell culture, transfection and CAT Assay

24 h before transfection, 5X10⁵ Hela cells were seeded onto 100 mm plates and cultured in RPMI 1640 10% FCS. Cells were transfected with 4μg of each plasmid by the calcium phosphate protocol, as previously described (Ausbel et al., 1994). 72 hours later cells were harvested, lysed and cell extracts were processed to evaluate protein- protein interaction, measured by a CAT Elisa Kit (Roche, Germany) according to the manufacturer's instructions. Briefly antibodies to CAT are prebound to the surface of a microtitre plate and lysates from transfected cells were added to the wells, allowing expressed CAT binding to the anti-C AT covered plates and the presence of CAT was revealed by a sandwich ELISA. The absorbance of the sample is determined using an ELISA reader.

Production and purification of GST fusion proteins

A series of deleted N-terminal ezrin fusion proteins were expressed and purified from JM109 bacterial strain (Promega, USA). Ezrin 1-362, ezrin 149-242, ezrin 149- 200, ezrin 149-168 and ezrin 508-586 were expressed as glutathione S-transferase (GST) fusion proteins in pGEX-6P vector (Amersham Pharmacia Biotech UK). The following primers were used to direct the synthesis of ezrin truncated forms, adding Bam and EcoRI cloning sites in frame with and directly apposed respectively to the 5' and 3' ends of each coding sequence: 1) GGATCCACTCACCAGAAACCGA (SEQ TD NO 7) and 2) GAATTCCCTCTGCCTTCTTTGTC (SEQ ID NO 8), (residues 1- 362); 3) GGATCCCAGGACCTGGAAATG (SEQ ID NO 9) and 4) GAATTCCCAAGGAAAGCCAAT (SEQ ID NO 10) (residues 149-242); the third and 5) GAATTCCATTTCCAGGTCCTG (SEQ ED NO 11) (residues 149-200); the third and 6) GAATTCCCACTGGTCCCTGGTA (SEQ ID NO 12) (residues 149-168); 7) GGATCCGAGGACCGGATCCAGG (SEQ ID NO 13) and the forth (residues 169- 242); 8) GGATCCATCCGGGATGACCGCAAT (SEQ ID NO 14) and 9) GAATTCTTACAGGGCCTCGAACTCG (SEQ ID NO 15) (residues 508-586). PCR products were cloned into p-GEM-T (Promega, USA) then excised with the appropriate restriction enzimes and subcloned into pGEX-6P vector (Amersham-Pharmacia UK) at the BamHl and EcoRI sites to produce GST-ezrin fragments fusion proteins. Synthesis of GST fusion proteins was induced by incubating transformed bacteria with 0,1 mM isopropyl-β thiogalactopyranoside at 37°C for 3 hours. Cells were harvested, and the fusion protein was purified using Glutathione-agarose beads (Sigma- Aldrich, USA), in NΕTN buffer (lOOmM NaCl, lmM ΕDTA, 20mM Tris ph 8, 0.2% NP40, 3μl ml aprotinin, O.lmM PMSF).

Human CD95 cytoplasmic domain was obtained by PCR amplification directed by the following primers: GGATCCCCACTAATTGTTTGGGT (SΕQ ID NO 16) and GAATTCCTAGACCAAGCTTTGG (SΕQ ID NO 17) and produced as described for GST-ezrin fusion proteins. The recombinant protein was purified on sepharose columns (Qiagen, USA) after the cleavage of the GST tag utilising the PreScission Protease (Amersham Pharmacia, UK), according to the manufacturer 's instructions. The obtained protein was of about 19 kDa.

In vitro binding assay

50 μl of Gluthatione sepharose beads slurry containing 30μg GST-ezrin fusion proteins were separately mixed with 15μg of purified CD95 Death Domain in NΕTN buffer for 16 hours at 4° C, or mixed with 1 mg of Jurkat cells total extract in NETN buffer for 16 hours at 4°C. Then these preparations were subjected to Western immunoblot assay with anti CD95 pAb (C-20 Santa Cruz, USA) or anti GST pAb (Amersham Pharmacia, UK), or anti actin mAb (Chemicon, USA) and visualised with peroxidase anti-Ig followed by ECL detection (Pierce,USA).

Transfection assay

The green fluorescent protein (GFP) Ezrin/Moesin/Ezrin (GFP-ez/moe) fusion protein was obtained starting from three separate fragments: ezrin j.ι₄₆, moesin ₁47-173, ezrin _{1 4-586} and the primers that were used to direct their synthesis were the following: ezrin ₁.₁₄₆ : CTG CAGACTCACCAGAAACCGA (SEQ ID NO 18) and GAT ATC GCT GAG GTA CCC AGAC(SEQ ID NO 19); moesin_147-]73 GG TAC CTG G CC GGAGACAAG (SEQ ID NO 20) and ATCCGCTCCTCCCACTGGTCC (SEQ ID NO 21); ezrini₇₄.₅₈6 GGATCCAGGTGTGGCATGCGGAA(SEQ ID NO 22) and GAATTCTTACAGGGCCTCGAAC (SEQ ID NO 23). PCR products were cloned into pTopo vector (Invitrogen, USA) then excised with the appropriate couple of restriction enzymes (respectively XhoIKpnl; Kpnl BamHl; BamHl EcoRI) and ligated to acquire a single fragment that subsequently was ligated in the pEGFPNl vector (Clontech). at the Xhol and EcoRI sites to produce The GFP-ez/moe fusion protein. Primers utilised to produce the full length ezrin were the following : CTGCAGACTCACCAGAAACCGA (SEQ ID NO 24) and GAATTCTTACAGGGCCTCGAAC (SEQ ID NO 25). And the GFP-ezrin fusion protein was obtained as described above. Plasmids encoding the GFP-ez/moe or GFP- ezrin fusion proteins were transfected into Hela cells growing on coverslips using the Calcium Phosphate protocol [36], thereby obtaining GFP-ez/moe or GFP-ezrin Hela cells, respectively. Hut78 cells were transfected with the same plasmids using Lipofectamine 2000 Transfection kit (Invitrogen USA), thus obtaining GFP-ez/moe or GFP-ezrin Hut78 cells. The percentage of transfected cells was evaluated by FACS analysis. Western blotting and immunoprecipitation

Whole extracts from Hela, Hut78, Jurkat or CEM-VBL100, GFP-ezrin and GFP- Ez/Moe Hela or Hut78 cells were resuspended in SDS sample buffer, denaturated by boiling, separated by 12% SDS-PAGE and analysed by Western Blot. CD95, ezrin, radixin, actin and GFP-tagged proteins were detected with an anti-CD95 pAb (C20, Santa Cruz USA), an anti-ezrin mAb (Transduction Laboratories, USA), an anti-radixin pAb (C15, Santa Cruz), an anti-actin mAb (Chemicon, USA), and anti-GFP tag mAb (clone 1E4, MBL, Japan), and visualised with peroxidase anti-Ig followed by ECL detection (Pierce, USA). CD95, radixin, ezrin GFP-tagged proteins were immunoprecipitated overnight at 4°C in the presence of protein A+G Sepharose beads (protein G plus, Pierce) from pre-cleared cell lysates respectively with an anti-CD95 antibody (clone DX2, Calbiochem, USA), an anti-ezrin antibody (clone 3C12 Sigma USA), an anti-radixin pAb (C15, Santa Cruz) and GFP mAb (clone 1E4, MBL, Japan). Mouse IgGi (Santa Cruz, USA), was used as control isotype. Immunoprecipitated beads were washed four times in Lysis buffer, resuspended in SDS sample buffer and resolved by 10% SDS-PAGE. Then, proteins were transferred to nitrocellulose membrane and analysed by western blotting.

Cell Death Assays.

HeLa and Hut78 cells, either untransfected or GFP-transfected, were cultivated in 6 cm plates. Fortyeight hours after transfection, two colour flow cytometric analysis was performed after the following treatments: i) 24h after CD95 triggering (500 ng/ml of IgM anti-CD95 antibody, clone CHI 1, upstate Biotechnology, Lake Placid, NY); ii) 6h TNF-α exposure (50 IU/ml Sigma) or iii) 6h staurosporin (STS, 1 μM, Sigma). Untreated cells were considered as controls. At the end of treatments cells were washed and stained by using annexin V-alexa 568 (Molecular Probes). By using this technique, transfected cells showed green fluorescence emission due to GFP and then they were easily distinguishable from non-transfected cells. Apoptotic cells showed red fluorescence emission due to the annexin V binding. Green/red double positive events thus corresponded to apoptosis of transfected cells. The samples were analysed with a FACScan cytometer (Becton Dickinson) equipped with a 488 argon laser. At least 50,000 events have been acquired. Data were recorded and statistically analysed by a Macintosh computer using CellQuest Software. Statistical analysis of apoptosis data was performed by using Student's t-test. Data reported are the mean of 4 separate experiments ± standard deviation (S.D.). Only/? values of less than 0.01 were considered as significant.

Confocal Microscopy and Intensified Video Microscopy analyses

For microscopy analyses, Hela and Hut78 cells were seeded on cover glass placed in 60mm Petri dishes. 48 hours after transfection with the various GFP plasmids described above, cells were fixed (paraformaldehyde 3%, 30 min, +4°C), and permeabilised (TritonX-100 0.5%, 10 minutes at room temperature). For localisation of CD95, and the various GFP-tagged ezrin fusion proteins, samples were incubated at 37°C for 30 min with polyclonal antibodies to CD95 (Santa Cruz Biotechnology) then incubated with anti-rabbit IgG TRITC-conjugate (Sigma Chemical Co., St Louis, MO). For detection of the naϊve ezrin on untransfected cells monoclonal antibody to ezrin (Biogenesis, UK) was used. Then, samples were, washed in PBS and treated with fluorescent secondary antibody Rodamine Red (R-6393, goat anti mouse IgG (H+L) Molecular Probes Europe BV, PoortGebouw, Leiden Netherlands). Cells were observed and analysed by confocal laser-scanning microscope (FV500, Olympus Optical CO.Europa GMBH, Hamburg, Germany) or by Intensified Video Microscopy (IVM). For IVM analyses, images were captured by a colour chilled 3CCD camera (DELTA Systems, Italy) and figures were obtained by adding CD95-TRITC (red) and ezrin (FITC or GFP, green) images by the OPTILAB (Graftek, France) software for image analysis.

Growth of HeLa Cells in Mice

Two in vivo experiments were performed in SCJD mice, where the in vivo s.c. growth and metastatic behaviour of HeLa cells transfected with the ezrin mutant (in a similar fashion to that described above) with untransfected HeLa cells were performed. The number of mice were 5 for each treatment in each experiment (10 animals for each treatment as a whole in the two experiments). The results showed that the mean tumour diameter in the animals transplanted with the untransfected HeLa cells was 2.5±0.2 (mean ±SEM on 10 animals) and 0.9+0.3 in the animals transplanted with the transfected HeLa cells (pθ.001). Moreover, while HeLa tumour metastases were clearly detectable in the local lymph nodes, spleen and liver in the animals transplanted with the untransfected HeLa cells, metastases were undetectable in the SCID mice transplanted with the HeLa cells transfected with the ezrin mutant. These results clearly show that the ezrin mutants inhibit tumour growth and prevent metastasis in cancer cells.

Results

1. CD95 ERM association

Experiments aimed at investigating a possible direct interaction between CD95 and ERM proteins were performed. Our previous results in human lymphocytes showed that only ezrin interacts with CD95, while radixin was not expressed in lymphocytes [1, 37] and moesin was not implied in the binding to CD95 [1]. Therefore, we preliminarily evaluated the radixin binding to CD95 in cellular systems expressing radixin. To this purpose we carried out co-immunoprecipitation experiments using a clone of CEM cells (VBL-100), known to fully express radixin [38]. As shown in Figure 1, consistently with a previous report [1], ezrin and CD95 co- immunoprecipitated, while radixin was undetectable in CD95 immunoprecipitates (Figure IA). Notably, in the same cell extracts p-glycoprotein (pi 70) proved to co- immunoprecipitate with radixin (data not shown), further supporting our previous findings [38]. This first set of results showed that radixin/CD95 interaction did not occur in cells expressing radixin, while in the same cells CD95 interacted with ezrin, further supporting a specific role of ezrin in connecting CD95 to actin. 2. Binding of ezrin FERM domain to CD95

2.1 Two hybrid assay.

To evaluate a possible direct interaction between ezrin and CD95, two hybrid assay experiments were performed. The mammalian two-hybrid system was used in order both to allow the occurrence of post-transductional changes, such as Tyrosine phosphorylation, and to avoid possible false positive results. By analogy with other ERM binding membrane proteins, the two-hybrid system was performed using either the ezrin or the radixin FERM domain, together with the whole CD95 cytoplasmic domain. Particularly, the ezrin N terminal 392 aa, including the FERM domain and both Tyrosine phosphorylation sites (i.e. Y145 and Y353) [39], or the whole radixin FERM domain (known to lack Y353), were fused to the GAL4 transcriptional activation domain (AD) of the pVP16-AD, and CD95 cytoplasmic domain (aa 185-335) was fused to the GAL4 DNA binding domain (DNA-BD) of the pM GAL4-BD cloning vector (Figures IB and C). pG5CAT containing CAT gene was used as reporter plasmid. Plasmids containing the CD95 and ezrin or radixin fragments and the plasmid containing CAT, were transfected in HeLa cells. As control experiments, the interaction of mouse p53 with either SV40 large T antigen (as positive control), or with polioma virus coat protein (as negative control) was evaluated (Table 1). The two- hybrid interaction between CD95 and ezrin constructs was detected by expression of CAT assessed by a CAT ELISA colorimetric assay (detailed in Experimental procedures). In 7/7 repeated experiments the results clearly showed that the ezrin N- terminal domain interacted directly with CD95 cytoplasmic domain, while no detectable interaction was observed between CD95 and radixin. The absorbance values obtained from CD95/ezrin interaction were comparable to positive controls, while no CAT expression was detected in all the negative controls (Table 1). This set of experiments showed that CD95 and ezrin interact directly and that this interaction is specific for the ezrin FERM domain. Table 1

CAT expression of the various samples of Hela transfected cells in the Mammalian two hybrid assay

The Mammalian two hybrid assay for the interaction between Ezrin or Radixin and CD95, was analysed by measurement of expression of the reporter gene CAT trough an ELISA colorimetric analysis. The presence of ezrin CD95 interaction was evidenced by comparing the absorbance values of the various conditions with the positive controls.

(A) Positive controls: (1) CAT (CAT enzyme standard solution). (2) pM53/pVP16T/pG5CAT: pM-53 (vector expressing a fusion of the GAL4 DNA-BD to the mouse protein p53)/ pVP16-T (vector expressing a fusion of the VP16 AD to the SV40 large T-antigen)/pG5CAT (mammalian reporter vector).

(B) Negative controls: (1) pM53/pVP16-CP/ pG5CAT: pM-53 (vector expressing a fusion of the GAL4 DNA-BD to the mouse protein p53)/ PVP16-CP (vector expressing a fusion of the VP 16 AD to a viral coat protein, which does not interact with p53)/ pG5CAT (mammalian reporter vector); (2) pM/pVP16/ pG5CAT: pM (GAL4 DNA-binding domain cloning vector)/pVP16 (activation domain cloning vector)/ pG5CAT (mammalian reporter vector); (3) ρM-CD95/pVP16/pG5CAT: pM-CD95 (vector expressing a fusion of the GAL4 DNA-BD to the CD95 cytoplasmic domain)/pVP16 (activation domain cloning vector)/ pG5CAT (mammalian reporter vector); (4) pM/pVP16-ezrin/pG5CAT: pM (GAL4 DNA-binding domain cloning vector)/pVP16-ezrin (vector expressing a fusion of the VP16 AD to ezrini. 36₂)/pG5CAT (mammalian reporter vector).

(C) CD95/ERM plasmids: (1) PM-CD95/pVP16-ezrin pG5CAT: pM-CD95 (vector expressing a fusion of the GAL4 DNA-BD to the CD95 cytoplasmic domain)/pVP16-ezrin (vector expressing a fusion of the VP16 AD to ezrinι_-362)/pG5CAT (mammalian reporter vector); (2) PM-CD95/pVP16-radixin/pG5CAT: pM-CD95 (vector expressing a fusion of the GAL4 DNA-BD to the CD95 cytoplasmic domain)/ pVP16-radixin (vector expressing a fusion of the VP16 AD to radixin FERM domain) /pG5CAT (mammalian reporter vector).

2.2. GST in vitro binding assay

In order to identify the specific ezrin epitopes involved in the ezrin CD95 interaction highlighted by the two-hybrids experiments, a series of GST pull-down assays were performed. To the purpose of mapping the region responsible for CD95/ezrin association, a series of recombinant truncations of the ezrin FERM domain were constructed (Figure 2). Thus, we investigated the CD95 binding to a series of ezrin fragments produced as GST fusion proteins in a prokaryotic expression system. In order to confirm and extend the two hybrid assay results, we first explored the ability of the GST-ezrin variants to interact with the recombinant CD95. To this purpose, we utilised the whole CD95 cytoplasmic domain purified protein, in GST-pull down experiments. The results first showed that GST-ezrin 1-362 bound to CD95, confirming the results of the two-hybrid assay (not shown). We thus performed experiments using a series of recombinant truncations of the ezrin FERM domain in order to isolate the epitope accounting for the CD95 binding. The chosen regions were ezrin 149-242, ezrin 149-200, ezrin 149-168, and then 169-242. We preliminarily evaluated the specificity of our GST-ezrin fusion proteins, blotting with an anti-GST antibody the bacterial lysates containing the various fusion proteins (not shown). Thus, we evaluated the binding of the various GST-ezrin fusion proteins to CD95. The results showed CD95-binding activity for GST-ezrin 149-242, GST-ezrin 149-200, and GST-ezrin 149- 168, as assessed by WB for CD95 (Figure 3 A). However, GST-ezrin 169-242 showed no detectable CD95 -binding activity, and neither did the carboxy-terminal truncation (ezrin 508-586), which comprises the consensus sequence motif for binding to actin [27] (Figure 3A).

In order to verify the possible occurrence of CD95/ezrin fragments binding in a more physiological condition we performed additional experiments using a constitutively expressed CD95. To this purpose we performed the GST assay pulling down the total extracts of human T cells expressing high level of constitutive CD95 (i.e. Jurkat cells) on the same ezrin fragments used in the experiments with the recombinant CD95. The results were highly consistent with those obtained with the recombinant CD95, in showing the binding of GST-ezrin 149-242, GST-ezrin 149-200, and GST-ezrin 149-168 to the Jurkat cells CD95, as detected by western blot with an anti-CD95 antibody (Figure 3 B). Again the GST-ezrin 169-242 and the GST-ezrin 508-586 did not show any detectable CD95-binding activity (Figure 3B). As a further control, we also evaluated the interaction of the native actin with the ezrin fragments. The results showed that actin was exclusively detected in Jurkat total extracts and in the GST-ezrin 508-586 pulled down Jurkat total extracts (Figure 3C), confirming the specificity of our ezrin fragments for the binding to either CD95 or actin, as appropriate 3. In vivo characterisation of CD95 binding domain

Two human cell lines of different histotypes, Hut78 (a lymphoblastoid CD4+ T cell line) and HeLa (a cervix adenocarcinoma cell line) and both prone to CD95 mediated apoptosis, were used in order to investigate the biological relevance of the identified ezrin region in susceptibility to CD95-mediated apoptosis. Cells were transiently transfected with different expression vectors encoding GFP alone, GFP- tagged full length ezrin (GFP-ezrin), and a mutated GFP-ezrin (GFP-ez/moe). In order to avoid the occurrence of ezrin 4.1 domain mutations interfering with globular folding, a chimeric GFP-ezrin/moesin ezrin (GFP-ez/moe) fusion protein was obtained replacing ezrin aa 148-166 with the corresponding moesin amino acid sequence, displaying a similar globular folding [32, 35] (Figure 4A). Seventytwo hours after transfection, cells were lysed and analysed to verify the presence of the tagged protein. Western Blot analysis showed up the expression of GFP-ezrin fusion proteins in total lysates of transfected cells (not shown). Thus, the functional role of the ezrin epitope in CD95- mediated apoptosis in intact cells was verified.

HeLa and Hut78 cells were transfected with plasmids containing the various GFP-ezrins, as well as control plasmids, and 48 hours after transfection, cells were triggered with α-Fas, TNF-α, (which receptors do not interact with ezrin [1]) or Staurosporin (STS) (a stimulator of the intrinsic/mitochondrial pathway [40]). The transfection efficiency in our experimental system ranged from 18.6 to 47.9 %. Biparametric FACS analysis of GFP/annexin V double-positive cells revealed that: i) only GFP ez/moe cells were significantly (pO.Ol) protected from CD95-induced apoptosis (Figure 4B and 4C, second row, quadrant II); ii) GFP-ezrin did not confer any significant protective effect; and, importantly, iii) under the same experimental conditions, no protective effect against TNF-α or STS induced apoptosis was observed (Figure 4B and 4C, third and fourth rows, respectively); iv) the percentage of apoptotic cells in untransfected cells as well as in cells transfected with GFP-vector alone was not significantly different from that found in GFP-ezrin cells (Figure 4B and 4C, first and second column, respectively). To investigate whether interfering with CD95/ezrin binding was important in protecting against CD95-mediated apoptosis in cells transfected with mutated ezrin, co- immunoprecipitation experiments were performed. The experiments blotting with anti- GFP tag mAb on CD95 immunoprecipitates and with anti-CD95 mAb on GFP-tag immunoprecipitates in untransfected Hut78 cells, or in Hut78 cells transfected with the GST-WT or the GST-ez/moe (Figure 5 A, lower panel), showed that a specific band corresponding to CD95 was clearly detectable only in GFP-ezrin cells. Consistently with these results, GFP-tagged ezrin was detectable only in CD95 immunoprecipitates from GFP-ezrin cells, while CD95 was undetectable in the GFP immunoprecipitates of GFP-ez/moe transfected cells (Figure 5 A, upper panel). Comparable results were obtained with HeLa cells (data not shown). These results showed that the inserted mutation on the CD95 binding domain of ezrin accounted for protection against CD95- mediated apoptosis shown in the same cells.

Finally, we analysed by laser scanning confocal microscopy (LSCM) or IVM the relative distribution of ezrin, CD95 and GFP-ezrin or GFP-Ez/moe fusion proteins in HeLa or Hut 78 transfected and untransfected cells. The result showed that the GFP- ez/moe and CD95, while localising in similar subcellular compartments, mostly at the plasma membrane, did not co-localise, as analysed by LSCM (Fig 6B upper panels). On the other hand, ezrin and CD95 were shown to polarise and co-localise both in GFP- ezrin transfected and untransfected cells (Fig 6B central and lower panels, respectively). Consistent results were obtained in Hut 78 cells. In fact, figure 5 C shows that CD95 and ezrin did not colocalise in GFP-ez/moe transfected cells (left panel), while ezrin and CD95 colocalised both in GFP-ezrin transfected cells (central panel) and untransfected cells (right panel). This set of results showed that the inhibition of the susceptibility of human cells to CD95-mediated apoptosis, as well as the absence of ezrin-to-CD95 binding, due to the substitution of the ezrin CD95 membrane binding domain with the homologous region of moesin (GFP-ez/moe), was consistent with the inhibition of ezrin /CD95 colocalisation in the transfected cells. EXAMPLE 2

To investigate the mechanism underlying the inhibition of human tumour metastasis observed transplanting SCED mice with human tumour cells transfected with the mutant ezrin, we performed further experiments based on two driving hypothesis: l.The mutant ezrin did not allow linkage to CD44, in turn inhibiting the ability of tumour cells to move into tissues, migrate through vessels and to adhere to metastatic organs.

2.The mutant ezrin did not allow linkage to lysosomal antigens, thus inhibiting acidic vacuoles traffic and, in turn, the capacity of tumour cells to feed on dead cells or debris.

To verify these hypothesis we performed molecular and functional experiments. The molecular experiments showed that human tumour cells transfected with the mutant ezrin did not co-immunoprecipitate with both CD44 and lysosomal antigens (i.e. LAMP1 and CD63). These results were much more clear when the experiments were performed in cell fractions, such as the cytoskeletal or endolysosomal fractions.

The functional experiments showed that in human tumour cells transfected with the mutant ezrin both the ability to phagocytose dead cells and yeasts, and to migrate through transwells, was inhibited.

These results suggest that ezrin may have a dual role in tumour metastasis in favouring both migration and feeding of malignant cells through connection with key proteins in the development of these tumour functions. References

1 Parlato, S., Giammarioli, A.M., Logozzi, M., Lozupone, F., Matarrese, P., Luciani, F., Falchi, M., Malorni, W., and Fais S. (2000) EMBO J. 19, 5123-34.

2 Tsukita, S., Yonemura, S., and Tsukita, S. (1997) Trends Biochem. Sci.. 22, 53-8.

3 Louvet-Vallee S. (2000) Biol. Cell. 92, 305-16.

4 Gautreau, A., Louvard, D., and Arpin, M. (2002) Curr. Opin. Cell. Biol. 14, 104-9.

5 Bretscher, A. (1983) J. Cell. Biol. 91, 425-32.

6 Franck, Z., Gary, R., and Bretscher, A. (1993) J. Cell. Sci. 105, 219-231.

7 Berryman, M., Franck, Z., and Bretscher, A. (1993) J. Cell Sci. 105, 1025-1043.

8 Yonemura, S., Nagafuchi, A., Sato, N., and Tsukita S. (1993) J. Cell Biol 120, 437-49.

9 Bretscher, A., Reczek, D., and Berryman, M. (1997) J. Cell. Sci. 110, 3011 -8.

10 Fais, S., Luciani, F., Logozzi, M., Parlato, S., and Lozupone, F. (2000) Histol Histopathol 15, 539-49.

11 Takeuchi, K., Kawashima, A., Nagafuchi, A., and Tsukita, S. (1994) J. Cell. Sci. 107, 1921-8.

12 Tsukita, S., and Yonemura, S. (1997). Curr. Opin. Cell Biol. 9, 70-75.

13 Chishti, A.H., Kim, A.C., Marfatia, S.M., Lutchman, M., Hanspal, M., Jindal, H., Liu, S.C, Low, P.S., Rouleau, G.A., Mohandas, N., Chasis, J.A., Conboy, J.G., Gascard, P., Takakuwa, Y., Huang, S.C, Benz, E.J. Jr., Bretscher, A., Fehon, R.G., Gusella, J.F., Ramesh, V., Solomon, F., Marchesi, V.T., Tsukita, S., Tsukita, S., Hoover,K.B., et al (1998) Trends. Biochem. Sci. 23, 281-2.

14 Turunen, O., Sainio, M., Jaaskelainen, J., Carpen, O., and Vaheri, A. (1998). Biochim. Biophys. Ada 1387, 1-16.

15 Reczek, D., Berryman, M., and Bretscher, A. (1997) J Cell Biol. 139, 169-79.

16 Kurashima, K., D'Souza, S., Szaszi, K., Ramjeesingh, R., Orlowski, J., and Grinstein, S. (1999) J. Biol. Chem. 11 , 29843-9.

17 Bretscher, A. (1999) Curr. Opin. Cell Biol. 11, 109-16. 18 Gautreau, A, .Louvard, D., and Arpin, M. (2000) J. Cell Biol 150, 193-203.

19 Hirao, M., Sato, N., Kondo, T., Yonemura, S., Monden, M., Sasaki, T., Takai, Y., Tsukita, S., and Tsukita, S. (1996) J Cell. Biol. 135, 37-51.

20 Barret, C, Roy, C, Montcourrier, P., Mangeat, P., and Niggli, V. (2000) J. Cell. Biol 151, 1067-80.

21 Bretscher, A. (1989) J. Cell Biol 108, 921 -30.

22 Krieg, J., and Hunter T,. (1992) J. Biol. Chem. 267, 19258-19265.

23 Thuillier, L., Hivroz, C, Fagard, R., Andreoli, C, and Mangeat, P. (1994) Cell. Immunol. 156, 322-31.

24 Crepaldi, T., Gautreau, A., Comoglio, P.M., Louvard, D., and Arpin, M. (1997) J Cell Biol. 138, 423-34.

25 Perez, O.D., Kinoshita, S., Hitoshi, Y., Payan, D.G., Kitamura, T., Nolan, G.P., and Lorens, J.B. (2002) Immunity 16, 51-65.

26 Gautreau, A., Poullet, P., Louvard, D., and Arpin M. (1999) Proc. Natl. Acad. Sci. USA. 96, 7300-5.

27 Turunen, O., Wahlstrom, T., and Vaheri, (1994). J. Cell Biol. 126, 1445-53.

28 Pestonjamasp, K., Amieva, M.R., Strassel, C.P., Nauseef, W.M., Furthmayr, H., and Luna E,J. (1995) Mol Biol. Cell. 6, 247-59.

29 Yao, X., Cheng, L., and Forte, J.G. (1996) J. Biol. Chem. 271, 7224-9.

30 Roy, C, Martin, M., and Mangeat, P. (1997) JBiol Chem. 212, 20088-95.

31 Smith, W.J., Cerione, R.A. (2002) Acta Crystallogr. D. Biol. Crystallogr. 58, 1359-61.

32 Smith, W.J., Nassar, N., Bretscher, A., Cerione, R.A., and Karplus PA. (2003) J Biol. Chem. 278, 4949-56.

33 Hamada,. K, Shimizu, T., Matsui, T., Tsukita, S., and Hakoshima, T. (2000) EMBOJ. 19, 4449-62.

34 Pearson, M.A., Reczek, D., Bretscher, A., and Karplus, P.A. (2000) Cell 101, 259-70.

35 Edwards, S.D., and Keep, N.H. (2001) Biochemistry 40, 7061-8. 36 Ausbel, F. M., Brent, R., Kingdom, R. E., Moore, D. M., Seidman, J. G., Smith, J. A. and Struhl, K. (1994). Current Protocols in Molecular Biology. (Greene Publishing Associated Inc., & John Wiley & Sons, inc.).

37 Ramoni, C, Luciani, F., Spadaro, F., Lugini, L., Lozupone, F., and Fais, S. Eur. J. Immunol. (2002) 32: 3059-65.

38 Luciani, F., Molinari, A., Lozupone, F., Calcabrini, A., Lugini, L., Stringaro, A., Puddu, P., Arancia, G., Cianfriglia, M., and Fais, S. (2002) Blood 99: 641-8.

39 Jiang, W.G., Hiscox, S., Singhrao, S.K., Puntis, M.C, Nakamura, T., Mansel, R.E., and Hallett, M.B. (1995) Biochem. Biophys. Res.Commun. 217, 1062-9.

40 Wang, G.Q., Gastman, B.R., Wieckowski, E., Goldstein, L.A., Rabinovitz, A., Yin, X.M., and Rabinowich, H., (2001) J. Biol Chem. 276:3610-9.

41 del Pozo, M.A., Nieto, M., Serrador, J.M., Sancho, D., Vicente-Manzanares, M., Martinez, C, and Sanchez-Madrid, F. (1998) CellAdhes Commun Cell. Adhes. Commun. 6, 125-33.

42 Serrador, J.M., Alonso-Lebrero, J.L., del Pozo, M.A., Furthmayr, H., Schwartz- Albiez, R., Calvo, J., Lozano, F., and Sanchez-Madrid, F. (1997) J. Cell. Biol. 138, 1409-23.

43 Serrador, J.M., Nieto, M., Alonso-Lebrero, J.L., del Pozo, M.A., Calvo, J., Furthmayr, H., Schwartz-Albiez, R., Lozano, F., Gonzalez-Amaro, R., Sanchez-Mateo, P., and Sanchez-Madrid F,. (1998) Blood 91, 4632-44.

44 Vicente-Manzanares, M., and Sanchez-Madrid, F. (2000) Dev. Immunol. 7, 51- 65.

45 Algeciras-Schimnich, A., Shen, L., Barnhart, B.C., Murmann, A.E., Burkhardt, J.K., and Peter, M.E., (2002) Mol. Cell Biol. 22, 207-20.

46 Giammarioli, A.M., Garofalo, T., Sorice, M., Misasi, R., Gambardella, L., Gradini, R., Fais, S., Pavan. A., and Malorni W. (2001) FEBSLett. 506, 45-50.

47 Kondo, T., Takeuchi, K., Doi, Y., Yonemura, S., Nagata, S., and Tsukita, S. (1997) J. Cell. Biol. 139, 749-58.

48 Kulms, D., Dussmann, H., and Poppelmann, B., Stander, S., Schwarz, A., Schwarz, T. (2002) Cell Death Differ. 9, 598-608.

Claims

Claims:

1. A nucleic acid sequence substantially encoding ezrin [SEQ ID NO. 27], wherein the portion encoding the CD95 binding site carries one or more substitutions sufficient to inhibit or prevent CD95 binding to the translation product of the sequence.

2. A sequence according to claim 1, wherein some or all of encoded ezrin residues 149-168 of SEQ ID NO. 27 inclusive are substituted.

3. A sequence according to claim 2, wherein one or more of residues 149-152, 154, 159, 160, 164 and 165 is substituted.

4. A sequence according to any preceding claim, wherein the substitutions are for the corresponding residues in either radixin or moesin.

5. A vector comprising a sequence according to any preceding claim.

6. A vector according to claim 5, which is an expression vector.

7. A method for controlling CD95-mediated apoptosis in an individual, comprising transforming or transfecting cells in said individual with a sequence according to any preceding claim.

8. A method according to claim 7, wherein the condition to be treated is Lyell's syndrome, GVHD, multiple sclerosis, viral hepatitis or AIDS.

9. A method for controlling tumour cells in an individual, comprising transforming or transfecting tumour cells in said individual with a sequence according to any of claims 1 to 6.

10. A method according to any of claims 7-9, wherein the transformation or transfection is selected from the group consisting of: receptor-mediated or virus- mediated transfection; microinjection; and by means of a gene gun, wherein the nucleotides are coated onto metallic microprojectiles.

1. A transformed or transfected host cell, as defined in any of claims 7-9.