WO2004022104A2 - Foxp2 and uses thereof - Google Patents

Abstract

The present invention relates to FOXP2 and in particular to the use of FOXP2 in treating or identifying patients with cancer. More particularly the invention discloses a method of treating a disorder in a patient characterised by reduced expression of FOXP2 protein, which method comprises, administering to a patient in need thereof a therapeutic amount of any of a nucleic acid molecule encoding said FOXP2 protein, FOXP2 protein or a composition capable of stimulating an increase in FOXP2 protein expression in the tissues of said patient. Also covered by the application are further methods, compositions, vectors, kits and assays relating to FOXP2.

Description

FOXP2 and Uses Thereof

The present invention is concerned with FOXP2 and in particular with the use of FOXP2 in treating or identifying patients with cancer.

Transcription factors are critically important for the regulation of gene expression at the transcriptional level . The forkhead DNA binding domain that defines the jori ea /winged helix transcription factor family was identified in 1990 (Weigel and Jackie, 1990) in a Drosophila homeotic gene. This family has numerous members with diverse roles including the control of cellular differentiation and proliferation, pattern formation, signal transduction and DNA repair.

Genes in this family have previously been implicated in mammalian oncogenesis . For example Qin is a retrovirally transduced murine oncogene (Li and Vogt , 1993). Studies of Foxll mutant mice have established that beta-catenin is an indirect target of this gene which may therefore be linked to colorectal cancer (Perreault et al . , 2001). AFX, AF6q21 and FKHR are involved in chromosome translocations in human" malignancies (Galili et al . , 1993; Davis et al . , 1994; Parry et al., 1994; Borkhardt et al . , 1997) and have been identified as targets ^"of the P(I)3 kinase/protein kinase B signalling pathway which has been implicated in tumourigenicity (Kops and Burgering, 1999) . FKHR or its related family members have been reported to be critical proteins downstream of the PTEN tumour suppressor gene and restoration of their function might be able to suppress tumorigenesis in PTEN-deficient tumour cells (Nakamura et al . , 2000). Recently inactivation of AFX and FKHRL1 forkhead proteins by protein kinase B signalling in quiescent cells has been shown to be critical for cell cycle reentry thus potentially contributing to transformation (Kops et al . , 2002 ) .

Previously the inventors have identified a novel winged helix transcription factor, FOXP1 , as a candidate tumour suppressor gene mapping to a tumour suppressor locus on chromosome 3p (Banham et al . , 2001). Deletions at chromosome 3p have been reported to be very common in human tumours (Sezinger et al . , 1991) and differential F0XP1 protein and mRNA expression were identified in several tumour types (Banham et al . , 2001) . The FOXP4 gene is a new member of this subfamily, whose mRNA expression has been shown to be significantly reduced in patients with kidney tumours (Teufel et al. , 2003) .

There are currently four members of the FOXP branch of the forkhead family, FOXP1 , F0XP2, FOXP3 and FOXP4. These forkhead genes are distinct from other members of the family in that they have an atypically C-terminal forkhead domain and an N-terminal C2H2 zinc finger motif.

Mutations in both FOXP2 and FOXP3 genes have associated these .molecules with human disease. Mutation of F0XP2 has been linked to an inherited speech and language disorder (Lai et al . , 2.001) . -This mutation/disruption of FOXP2 is linked to only rare severe forms of speech and language disorder and a comprehensive search for mutations in specific language impairment and autism (also linked to 7q31) did not identify any changes in the FOXP2 gene (Newbury et al . , 2002) . However, additional coding and non-coding exons incorporated into alternatively spliced forms of- FOXP2 that have not been screened for mutation, have recently been identified (Bruce and Margolis, 2002) . Mutations in FOXP3 have been linked to the immune dysregulation diseases such as X-linked neonatal diabetes mellitus, enteropathy and endocrinopathy syndrome (IPEX) and X-linked autoimmunity-allergic disregulation syndrome (XLAAD) (Chatila et al . , 2000; Bennett et al . , 2001; Bennett et al . , 2001; Brunkow et al . , 2001; ildin et al . , 2001) .

A study of urine Foxpl and Foxp2 demonstrated that these proteins act as transcriptional repressors that regulate gene transcription in the lung (Shu et al . , 2001) . This does not mean that FOXP2 always acts as a transcriptional repressor and this protein may act as an activator of gene expression in other tissues or on different promoters within the same tissue. In this study Northern blotting analysis of adult tissues demonstrated that Foxpl was most highly expressed in lung, brain and spleen with lower expression in other tissues including heart, skeletal muscle, kidney, small intestine and liver. Foxp2 expression was, however, highest in the lung with lower expression levels observed in spleen, small intestine, skeletal muscle, brain and kidney. In si tu hybridization was used to characterise the expression of Foxpl and Foxp2 during murine embryonic development . The highest levels of Foxpl and Foxp2 expression were observed in the developing lung with expression aiso being observed in neural, gastrointestinal and cardiovascular tissues. The localisation of embryonic mRNA expression within tissues indicated that in addition to their proposed role in lung development Foxpl and Foxp2 might play important roles in developing neural, gastrointestinal and cardiovascular tissues. The expression of Foxpl in both the mesodermal and endodermal cells of the gut, along with the expression of Foxp2 in the mesodermal layer of the intestine suggested that they might also- regulate intestinal development (Shu et al. , 2001) . The murine Foxpl and Foxp2 proteins were both able to repress the transcription of a reporter gene under control of either the mouse CC10 or human SP-C promoter (Shu et al , 2001) . Interestingly two f orkhead _^proteins, Fkhlp and Fkh2p, have been shown to share promoter targets and overlapping biological roles in cell cycle progression and differentiation in the yeast Saccharomyces cerevisiae (reviewed in Breeden, (2000)) . Loss of both FKH1 and FKH2 is required for the pseudohyphal growth phenotype indicating that these proteins have redundant roles. However there is data indicating that despite their similarities there are differences in their effects on transcription which suggest that Fkhlp and Fkh2p may have different inherent DNA-binding properties or mediate different protein-protein interactions. This has been shown to be the case as purified Fkh2p, but not Fkhlp could bind cooperatively with itself, and Fkh2p, but not Fkhlp, could bind cooperatively with Mc lp (Hollenhorst et al . , 2001) . Analysis of Fkhlp and Fkh2p binding to promoter targets in vivo using mutant strains indicated that the two proteins compete for promoter-occupancy at a number of target promoters. In addition bona fide Fkh target promoters contained two or more Fkh-binding sites that allowed the Fkhlp , and Fkh2p proteins to form multiple protein-DNA complexes in vi tro (Hollenhorst et al . , 2001) . While the yeast genes described above are not direct homologues of the human FOXPl and F0XP2 genes, the co-expression of both FOXPl and F0XP2 in some tissues together with the ability of the murine FOXPl and FOXP2 proteins to bind the same lung specific promoters raises the possibility that the human genes may also compete for promoter occupancy at target promoters. The heterodimerisation of Foxpl splice variants, and of Foxpl with Foχp3 , has now recently been reported for the murine Foxp proteins (Wang et al. , 2003) . The present inventors have now identified that differential expression of F0XP2 mRNA in an individual is associated with different human disorders, particularly cancer.

Therefore, in a first embodiment of the invention there is provided a method of treating a disorder in a patient characterised by a reduced level of FOXP2 mRNA or protein, which method comprises administering to a patient in need thereof a therapeutic amount of any of a nucleic acid molecule encoding said FOXP2 protein, FOXP2 protein or a composition capable of stimulating an increase in FOXP2 protein in the tissues of said patient.

Thus, advantageously, by identifying gene copy number or the level of FOXP2 mRNA or protein expression associated with a particular condition it is possible to remedy or treat that condition by providing to the individual an enhancer of the FOXP2 protein as appropriate. In preferred embodiments of the invention the condition is cancer.

It has been found that in cancers, such as kidney, colon, uterus i prostate, breast or stomach cancer the levels of the F0XP2 mRNA were found to be markedly diminished thus leading to the. conclusion that F0XP2 expression levels are critical determinants or factors in the development of certain human cancers .

Accordingly, in another embodiment of the invention there is provided a method of treating cancer in a patient which method comprises administering- to a patient in need thereof an enhancer of F0XP2 expression as determined by the level of expression of FOXP2 associated with said cancer. As would be apparent to those of skill in the art, particularly those skilled in the art of molecular biology, numerous enhancers/inhibitors are available. In one embodiment an enhancer which can increase expression of FOXP2, may be a nucleic acid molecule encoding the FOXP2 protein. The sequence of the FOXP2 protein and variants generated by alternative splicing are known in the art and are provided in Figures 9 and 10. Alternatively, the nucleic acid molecule may include for example a sequence of nucleotides including an appropriate regulatory region for expressing said FOXP2 and regions of homology with the DNA upstream of the native F0XP2 open reading frame such that it is capable of inserting into the genome by homologous recombination in such a manner so as to strongly express FOXP2.

The nucleic acid molecules described herein may also, advantageously, be included in a suitable expression vector to express the proteins encoded therefrom in a suitable host . Incorporation of cloned DNA into a suitable expression vector for subsequent transformation of said cell and subsequent selection of the transformed cells is well known to those skilled in the art as provided in Sambrook et al . (1989), Molecular - cloning: A Laboratory Manual, Cold Spring Harbour Laboratory.

Such an expression vector includes a vector having a nucleic acid according to the invention operably linked to regulatory sequences, such as promoter regions, that are capable of effecting expression of said DNA fragments. The term "operably linked" refers to a juxtaposition wherein the components described are in a relationship permitting them to function in their intended manner. Such vectors may be transformed into a suitable host cell to provide for the expression of a protein according to the invention. The nucleic acid molecule may encode a mature protein or a protein having a prosequence, including that encoding a leader sequence on the preprotein which is then cleaved by the host cell to form a mature protein.

The vectors may be, for example, plasmid, virus or phage vectors provided with an origin of replication, and optionally a promoter for the expression of said nucleotide and optionally a regulator of the promoter. The vectors may contain one or more selectable markers, such as, for example, an antibiotic resistance gene.

Regulatory elements required for expression include promoter sequences to bind RNA polymerase and to direct an appropriate level of transcription initiation and also translation initiation sequences for ribosome binding. For example, a bacterial expression vector may include a promoter such as the lac promoter and for translation initiation the Shine-Dalgarno sequence and the start codon AUG. Similarly, a eukaryotic expression vector may include a heterologous or homologous promoter for RNA polymerase II, a downstream polyadenylation signal, the start codon AUG, and a termination codon for detachment of the ribosome . Such vectors may be obtained: commercially or be assembled from the sequences described by methods well known in the art .

Transcription of DNA encoding the FOXP2 protein by higher eukaryotes may be optimised by including an enhancer sequence in the vector. Enhancers are cis-acting elements of DNA that act on a promoter to increase the level of transcription. Vectors will also generally include origins of replication in addition to the selectable markers. In accordance with the present invention, a defined nucleic acid includes not only the identical nucleic acid but also any minor base variations including, in particular, substitutions in cases which result in a synonymous codon (a - different codon specifying the same amino acid residue) due to the degenerate code in conservative amino acid substitutions.- The term "nucleic acid sequence" also includes the complementary sequence to any single stranded sequence given regarding base variations .

The present invention also advantageously provides oligonucleotides comprising at least 10 consecutive nucleotides of a nucleic acid as shown in Figure 10A/B and preferably from 10 to 40 consecutive nucleotides of a nucleic acid according to the invention. These oligonucleotides may, advantageously be used as probes or primers to initiate replication, or the like. Oligonucleotides having a defined sequence may be produced according to techniques well known in the art, such as by recombinant or synthetic means. They may also be used in diagnostic kits or the like for detecting the presence of a nucleic acid capable of expressing FOXP2. These tests generally comprise contacting the probe with the sample under .hybridising conditions and detecting for the presence ^'of any duplex or triplex formation between the probe and any nucleic ^""acid in the sample.

According to the present invention these probes may be anchored to a solid support. Preferably, they are present on an array so that multiple probes can simultaneously hybridize to a single biological sample. The probes can be spotted onto the array or synthesised in si tu on the array. (See Lockhart et al . , Nature Biotechnology, vol. 14, December 1996 "Expression monitoring by hybridisation to high density oligonucleotide arrays" . The nucleic acid sequences described herein may be produced using recombinant or synthetic techniques, such as for example using PCR which generally involves making a pair of primers, which may be from approximately 10 to 50 nucleotides to a region of the gene which is desired to be cloned, bringing the primers into contact with cDNA, or genomic DNA from a human cell, performing a polymerase chain reaction under conditions which brings about amplification of the desired region, isolating the amplified region or fragment and recovering the amplified DNA. Generally, such techniques are well known in the art, such as described in Sambrook et al . (Molecular Cloning: a Laboratory Manual, 1989) .

The nucleic acids or oligonucleotides may carry a revealing label. Suitable labels include radioisotopes such as ³²P or ³⁵S, enzyme labels or other protein labels such as biotin or fluorescent markers. Such labels may be added to the nucleic acids or oligonucleotides of the invention and may be detected using known techniques per se . Such visual markers are also particularly advantageous for providing a visual indication of the level of F0XP2 expression in a cell.

.Advantageously, human allelic variants or polymorphisms of the. FOXP2 protein may be identified by, for example, probing cDNA or genomic libraries from a range of individuals, for example, from different populations. Furthermore, nucleic acids and probes according to the invention may be used to sequence genomic DNA from patients using techniques well known in the art, such as the Sanger dideoxy chain termination method, which may, advantageously, ascertain any predisposition of a patient to disorders associated with variants of the F0XP2 protein. The nucleotide sequences identified herein can be used in numerous ways as a reagent. The following description should be considered exemplary and utilizes known techniques. The nucleotide sequence of the subject, invention may be used to detect differences in the chromosomal location due to translocation, inversion, etc. among normal, carrier or affected individuals with a disease resulting from an alteration in this gene. If no structural alterations exist, the presence of point mutations are ascertained. Mutations observed in some or all affected individuals, but not in normal individuals, indicates that the mutation may be associated with the disease. In the very least, the nucleotide sequences can be used as diagnostic probes for the presence of a specific mRNA in a particular cell type, as a probe to "subtract-out" known sequences in the process of discovering novel nucleotide sequences, for selecting and making oligomers for attachment to a "gene chip" or other support, to raise anti-DNA antibodies using DNA immunization techniques, and as an antigen to elicit an immune response.

The FOXP2 protein described herein which may be used in the method of the invention includes all possible amino acid variants.-encoded by the nucleic acid molecule of Figure 10A/B including a protein encoded by said molecule and having conservative amino acid changes in addition to splice variants of the FOXP2 protein. Such proteins or polypeptides further include variants of such sequences, including naturally occurring allelic variants which are substantially homologous to said proteins or polypeptides. In this context, substantial homology is regarded as a sequence which has at least 70%, preferably 80% or 90% and more preferably 95% amino acid homology with the proteins or polypeptides encoded by the nucleic acid molecules according to the invention. The protein according to the invention may be recombinant, synthetic or naturally occurring, but is preferably recombinant .

Nuclear receptors are hormone,-dependent transcription factors that play essential roles in development, differentiation, cell proliferation and metabolism. They are under the control of a wide variety of hormones and ligands, such as steroids, retinoids, thyroid hormone, 1,25-dihydroxy- vitamin D3 , and prostanoids . Ligand binding promotes a conformational change in receptor structure that promotes interaction between activated receptors and coactivators. Many of these receptors are therapeutic targets for disease treatment: antagonists of estrogen receptor-α (e.g. tamoxifen) are clinically used to treat breast cancer (Dees and Kennedy, 1998) whereas retinoic acid receptor (RAR) agonists and antagonists block the growth of a number of neoplastic cells including breast tumour cells (Fanjul et al . , 1998; Shiohara et al . , 1999). Agonists against retinoid X receptors (RXR) and peroxisome proliferator-activated receptor-γ (PPARγ) are potential candidates for treating cancer and diabetes (Murkerjee et al . , 1997; Bischoff et al . , 1998; Elstner et al . , 1998; Spiegelman, 1998). Thus, the identification of molecules that selectively activate or inhibit specific nuclear receptors is of considerable biological significance and these have the potential for clinical applications.

Ligand-dependent activation of gene transcription by nuclear receptors such as RAR, thyroid hormone receptor (TR) , estrogen receptor (ER) and PPARγ are dependent on the recruitment of coactivators. The alpha-helical LXXLL motif found in some coactivators (where L is leucine and X is any amino acid) is sufficient for ligand-dependent interaction with nuclear receptors and this signature motif is commonly referred to as the nuclear receptor (NR) box. A large number of co-regulatory factors have been described for some of these receptors, for example the estrogen receptor recruits both coactivators, co-repressors or proteins that remodel chromatin, reviewed by (Sommer and Fuqua, 2001) .

The present inventors have now identified the presence of an LXXLL motif in the terminus of F0XP2 indicating that this protein has the potential to bind and to coregulate nuclear receptors. Therefore, according to a further aspect of the present invention, there is provided a method of treating diseases or conditions mediated by FOXP2 co-regulation of a nuclear receptor, such as the estrogen receptor, which method comprises administering to an individual in need thereof, an inhibitor or enhancer of the interaction of FOXPl and said receptor.

Thus, advantageously, it is possible by virtue of the present study to identify potential therapeutic agents that can function in the prevention or treatment of those diseases or conditions that are mediated by FOXP2 co-regulation of a particular nuclear receptor.

In one embodiment, an inhibitor or antagonist of the interaction of said nuclear receptor and the FOXP2 may be provided^"- to -said individual, which inhibitor or antagonist comprises any of an antisense molecule of FOXP2 or a nucleic acid molecule capable of hybridising to nucleic acid encoding FOXP2 to inhibit expression thereof, an antibody capable of binding to FOXP2 in a cell, or any other molecule capable of antagonising the interaction of FOXP2 and said estrogen receptor. Thus advantageously, and as outlined in more detail in the examples below, for those conditions where there is a characteristic decrease of the F0XP2 protein, a molecule that has been found to co-localise with nuclear receptors, an appropriate enhancer can be provided to an individual in need thereof to treat said condition.

In one embodiment, an inhibitor may comprise any of an antisense molecule of F0XP2 or a nucleic acid molecule capable of hybridising to nucleic acid encoding FOXPl to inhibit expression thereof, an antibody capable of binding to F0XP2 in a cell, or any other molecule capable of antagonising the interaction of FOXP2 and said receptor.

As would be apparent to those of skill in the art, particularly those skilled in the art of molecular biology, numerous enhancers/inhibitors are available. In one embodiment an enhancer which can increase expression of FOXP2 , may be a nucleic acid molecule encoding the FOXPl protein. The sequence of the FOXP2 protein as well as variants generated by alternative splicing are known in the art. Alternatively, the nucleic acid molecule may include for example a sequence of nucleotides including an appropriate regulatory region for expressing said FOXP2 and regions of homology with the DNA upstream of the native F0XP2 open reading frame such that it is capable of inserting into the genome by homologous recombination in such a manner so as to strongly express FOXP2.

In the embodiment of the invention that requires an inhibitor, many such appropriate inhibitors of FOXP2 expression would be known to those of skill in the art. An inhibitor of expression of the F0XP2 protein may include for example the use of antisense technology. Antisense technology can be used to control gene expression through triple-helix formation of antisense DNA or RNA, both of which methods are based on binding of a polynucleotide to DNA or RNA. For example, the 5' coding portion or the mature protein sequence, which encodes for the protein of the present invention, is used to design an antisense RNA oligonucleotide of from 10 to 40 base pairs in length. The antisense RNA oligonucleotide hybridises to the mRNA in vivo and- blocks translation of an mRNA molecule into the protein (anti-sense - Okano, J. Neurochem. , 56:560 (1991); Oligodeoxynucleotides as Antisense Inhibitors of Gene Expression, CRC Press, Boca Raton, FL (1988)) . A DNA oligonucleotide is designed to be complementary to a region of the gene involved in transcription (triple- helix - see Lee et al . Nucl . Acids Res., 6:3073 (1979); Cooney et al . , Science, 241:456 (1988); and Dervan et al . , Science, 251: 1360 (1991), thereby preventing transcription and the production of the protein.

The antisense oligonucleotide described above can be delivered to cells by procedures in the art such that the anti-sense RNA and DNA may be expressed in vivo to inhibit production of the protein in the manner described above.

For example, nucleic acid molecules encoding FOXP2 may be inserted into the vectors described in an antisense orientation in order to provide for the production of antisense RNA. Antisense RNA or other antisense nucleic acids-, including antisense peptide nucleic acid (PNA) , may be produced by -synthetic means.

Another method of inhibiting FOXP2 includes utilising RNA interference. RNA interference (RNAi) is a sequence-specific posttranscriptional gene silencing mechanism, which is triggered by double-stranded DNA and causes degradation of mRNAs homologous in sequence to the dsRNA (Cogoni and Macino, 2000; Guru, 2000; Hammond et al . , 2001) . In mammalian cells this work was originally hampered by a number of dsDNA- triggered pathways that non-specifically suppress gene expression. However the use of short dsRNAs of approximately 21-nt has overcome this problem and has been shown to be an effective way of specifically silencing human genes in cultured somatic cells (Elbashir e.t al . , 2001). Ambion have an excellent website including a tool for predicting target siRNA sequences within a specific gene and for designing DNA oligonucleotides for use with their siRNA construction kit.

An inhibitor of FOXP2 protein also includes an antibody which is capable of binding to FOXP2 or an epitope thereof. Such an antibody may be raised according to standard techniques well known to those skilled in the art by using the protein of the invention or a fragment or single epitope thereof as the challenging antigen.

Reference to such an antibody as described above includes not only complete antibody molecules but fragments thereof. Antibody fragments which contain the idiotype of the molecule can be generated by known techniques, for example, such fragments include but are not limited to the F(ab')₂ fragment which can be produced by pepsin digestion of the antibody molecule; the Fab' fragments which can be generated by reducing „_:the disulfide bridges of the F(ab')a fragments and the Fab fragments which can be generated by treating the antibody: molecule with papain and a reducing agent. Chimeric humanized and fully humanized mAb can now be made by recombinant engineering. By addition of the human constant chain to F(ab^')₂ fragments it is possible to create a humanized monoclonal antibody which is useful in immunotherapeutic applications where patients making antibodies against the mouse Ig would otherwise be at- a disadvantage. Breedveld F.C. Therapeutic Monoclonal Antibodies. Lancet 2000 Feb 26; 335, P735-40. A further aspect of the present invention also provides a method of identifying a protein of the invention in a sample, which method comprises contacting said sample with an antibody as described herein and monitoring_^ for any specific binding of any proteins to said antibody. A kit for identifying the presence of such proteins in a sample is also provided comprising an antibody according to the invention and means for contacting said antibody with said sample.

In a further aspect the invention provides an in vitro method of detecting expression of a FOXP2 protein in cells and tissues permitting the association of a particular condition with the level of F0XP2 expression in a mammalian subject. This method comprises contacting a sample of tissue or cells removed from the mammalian subject with an antibody which is capable of binding to the FOXP2 protein of the invention or an epitope thereof detecting specific binding of the antibody to its target protein in the said tissue and monitoring the level of expression of said F0XP2 protein.

Preferably the method described is performed on cells or tissues removed from a human subject. However, it is also within- the scope of the invention to perform the method on cells or tissues removed from non-human mammals such as mouse or monkey by using an antibody which is cross-reactive against a homologous protein expressed in the non-human mammalian species .

Immunostaining with an antibody immunologically specific for the F0XP2 protein, can be used to detect expression of the protein in samples of normal, preneoplastic and neoplastic human tissues. In a preferred embodiment an antibody according to the ■ invention is that which is produced by the 2Fx2AD2 hybridoma deposited in the European Collection of Cell Cultures (Provisional Accession number 0308,2101) .

Other inhibitor molecules can also include for example small molecules. Such molecules include, for example, non- biological molecules, such as organic compounds, organometallic compounds, salts of organic and organometallic compounds, saccharides, amino acids. Other small molecules can include biological molecules of low molecular weight including lipids, oligosaccharides, oligopeptides or their derivatives.

As described more fully in the examples provided, F0XP2 has been mapped to a tumour suppressor gene locus on chromosome 7q31, a region which is well known to be associated with genetic abnormalities causing a wide range of human cancers or conditions. Thus, FOXP2 may be a critical factor in the progression of such conditions and F0XP2 gene dosage may be associated with a number of these conditions. Thus, according to a further aspect of the invention there is provided a method of assessing the prognosis of a patient suffering from any of myeloproliferative disorders, acute myeloid leukaemia, splenic marginal zone B-cell lymphoma, breast, cancer, prostate cancer and head and neck cancers, comprising monitoring the level of FOXP2 expression or gene loss, wherein loss of expression is clinically significant. Also provided by the present invention is a method of subtyping splenic marginal zone B-cell lymphomas, which method comprises testing the level of FOXP2 expression in a lymphoma cell of an individual wherein reduced F0XP2 expression or F0XP2 gene copy number is characteristic of a poor prognosis. Furthermore, there is also provided a method of screening pre- neoplastic lesions in a patient having an increased propensity to progress to a neoplastic state, which method comprises testing the level of FOXP2 expression in a pre-neoplastic cell of said individual wherein altered expression is indicative of the likelihood of said pre-neoplasm progressing to a neoplastic state.

The nucleic acid molecules and the amino acid or protein sequences described herein, may advantageously be used in the treatment of the human or animal body or alternatively in the manufacture of a medicament for treating cancer. They may also be included in a pharmaceutical composition together with any suitable pharmaceutically acceptable carrier diluent or excipient therefor. The nucleic acid molecule or the amino acid sequence or protein may be encapsulated and/or combined with suitable carriers in solid dosage forms for oral administration which would be well known to those of skill in the art or alternatively with suitable carriers for administration in an aerosol spray.

In the pharmaceutical composition of the invention, preferred compositions include pharmaceutically acceptable carriers including, for example, non-toxic salts, sterile water or,.the like. A suitable buffer may also be present allowing the compositions to be lyophilized and stored in sterile ^'"conditions prior to reconstitution by the addition of sterile water for subsequent administration. The carrier can also contain other pharmaceutically acceptable excipients for modifying other conditions such as pH, osmolarity, viscosity, sterility, lipophilicity, somobility or the like. Pharmaceutical compositions which permit sustained or delayed release following administration may also be used.

Furthermore, as would be appreciated by the skilled practitioner, the specific dosage regime may be calculated according to the body surface area of the patient or the volume of body space to be occupied, dependent on the particular route of administration to be used. The amount of the composition actually administered will, however, be determined by a medical practitioner based on the circumstances pertaining to the disorder to be treated, such as the severity of the symptoms, the age, weight and response of the individual .

The invention will be further understood with reference to the following experimental Examples, together with the accompanying Figures in which:

Table 1 summarises the changes in FOXP2 expression as a result of different treatments applied to cell lines on the BD Clontech Cancer Cell Line Profiling Array. A + indicates up- regulated expression compared to the control (12) and a - indicates down-regulated expression compared to the untreated control (12) . Reference was made to the results from the ubiquitin loading control when assessing whether differences were due to the treatment or to sample loading.

Figure 1A illustrates the results obtained from hybridizing the F0XP2 cDNA to Clontech' s Multiple Tissue Expression Array. The identity and position of tissues on the array is shown in the lower panel .

Figure IB illustrates the results obtained after stripping the MTE array described above and re-probing the array with the manufacturer's ubiquitin probe as a sample loading control .

Figure 2A illustrates the results obtained from hybridizing the FOXP2 cDNA to Clontech' s Matched tumor/normal expression array. Tissue sources for cDNAs on the array are as follows: normals row A/ tumours row B, kidney 1-14; normals row D/ tumours row E, breast 1-9, prostate 11-13; normals row G/ tumours row H, uterus 1-7, kidney 8, ovary 10-12, cervix 14; normals row J/ tumours row K, colon 1-11, lung 13-15; normals row M/ tumours row N, stomach 1-8, rectum 10-16, small intestine 18; Human cancer cell lines row P, 1 HeLa, 2 Daudi , 3 K562, 4 HL-60, 5 G361, 6 A549, 7 MOLT-4, 8 SW480, and 9 Raji.

Figure 2B illustrates the results obtained after stripping the MTN array described above and re-probing the same array with the manufacturer' s ubiquitin probe as a sample loading control .

Figure 3A illustrates the results obtained from hybridizing the FOXP2 cDNA to BD Clontech' s Blood Disease Profiling Array. Each row represents a fractionated blood sample from an individual patient (numbers indicate different patients) . Each column represents a different blood fraction: A= CD14 antigen-expressing cells; B= CD19 antigen-expressing cells; C= CD3 antigen-expressing cells; D= mononuclear cells; E= polymorphonuclear cells; F= total leukocytes; G= lymph node. Each A-F set corresponds to a different disease as labelled: on .the figure. Additional lymph node samples are spotted for Hodgkin's disease and non-Hodgkin' s lymphoma.

Figure 3B illustrates the results obtained after stripping the blood disease array described above and re-probing the array with the manufacturer's ubiquitin probe as a sample loading control .

Figure 4A illustrates the results .obtained from hybridizing the ³P labelled FOXP2 cDNA probe to BD Clontech' s Cancer Cell Line Profiling Array. The top panel shows the array hybridised with the FOXP2 probe. The lower panel shows the same array after stripping and re-probing with the manufacturer's ubiquitin probe.

Figure 4B provides further information on the BD Clontech Cancer Cell Line Profiling Array.

Figure 5 illustrates indirect immunoperoxidase labelling of COS-1 cells transfected with either pcDNA4HisMax/FOXP2 (A) or pcDNA4HisMax (vector alone) (B) . Western blot analysis of lysates of COS-1 cells transfected with either pcDNA4HisMax/FOXP2 or pcDNA4HisMax (vector alone) (C) .

Figure 6 illustrates an alignment of the N-terminal LXXLL motifs identified in the FOXP proteins.

Figure 7 illustrates the transient expression of the Xpress/His tagged FOXP2 protein in the MCF-7 breast cancer cell line 72 hours post transfection visualised by immunoperoxidase staining with either the anti-Xpress or the anti-His antibodies. Black arrows indicate pcDNA4HisMax/FOXP2 transfected cells. The empty pcDNA4HisMax vector was used as a control .

Figure- 8^'- illustrates pcDNA4HisMax/F0XP2 transfected MCF-7 cells that were double immunofluorescently labelled with antibodies recognising the F0XP2 protein (anti-His, dilution 1/400, green) and the ER (ID5, dilution 1/50, red). Panel a) shows a low power confocal microscope image of MCF-7 cells labelled with both antibodies and panel b) and c) show single labelling of the FOXP2 and ER proteins respectively. Panel d) shows a high power confocal microscope image of transfected cells double labelled with both antibodies, with e) and f) showing the single labelling for FOXP2 and ER respectively. White arrows indicate FOXP2 transfected cells.

Figure 9 shows an amino acid sequence alignment of F0XP2 proteins.

Figure 10A provides nucleic acid sequences of each the known FOXP2 exons (including untranslated exons) and 10B illustrates cDNA sequences of the alternatively spliced forms of FOXP2 encoding the amino acid sequences aligned in Figure 9.

Figure 11 illustrates immunoperoxidase staining of FOXP2 or vector control (as indicated) transfected COS-1 cells with the FOXP2 monoclonal antibody 2Fx2AD2 and the Everest FOXP2 polyclonal antibody together with the Xpress antibody (Invitrogen) against the epitope tag on the recombinant FOXP2 protein. Panels a-c show staining of cytospin preparations of cells, while d-f show staining of formalin fixed paraffin embedded COS transfectants .

Figure 12 illustrates the results obtained from immunoperoxidase staining with the FOXP2 monoclonal antibody 2Fx2AD2 verifying its reactivity with the FOXP2 protein (a) and not with the FOXPl (c) or F0XP4 proteins (e) . The presence of the FOXPl and FOXP4 proteins was demonstrated using the JC12 (d) and Xpress (f) antibodies respectively.

Figure 13 illustrates the results from immunoperoxidase staining of breast tumours with the FOXPl monoclonal antibody 2Fx2AD2. Tumour cells with nuclear positivity are labelled with short arrows while normal cells with nuclear positivty are labelled with long arrows. The tumour in panel f) is cytoplasmically stained while those shown in g) and h) are negative . Expression of the FOXP2 gene in normal and neoplastic human tissues.

The inventors have characterised the expression of the FOXP2 mRNA in both normal human tissues and in matched normal and tumour tissues from cancer patients to investigate whether differential expression of the F0XP2 mRNA occurs in human cancer.

Multiple Tissue Expression (MTE) , Matched Tumour/Normal (MTN) , Blood Disease Profiling and Cancer Cell Line Profiling arrays (BD Clontech) were pre-hybridised according to the manufacturer's instructions. A 1. Ikb fragment of FOXP 2 , encoding the last 60 amino acid residues followed by 3' UTR, was radiolabelled using the High Prime DNA Labelling Kit (Roche Molecular Biochemicals) . The probe was denatured and incubated with Cot-1 and sheared salmon sperm DNA as described (MTE/MTN Array User Manual, Clontech), before being added to the arrays and incubated overnight with gentle rotation at 65 °C. The probe was then decanted to waste and the arrays washed for 4x 20min in lx SSC, 1% SDS at 65°C followed by 2x 20min in O.lxSSC, 0.5% SDS at 65°C. The washed membranes were exposed to film "at -70°C. Additional information about the tissues and cases on these arrays can be obtained from the Clontech website. Loading of the cDNAs on both arrays is normalised for three housekeeping genes to enable quantitative comparisons between gene expression in different tissues. The ubiquitin probe provided by the manufacturer was also hybridised to the stripped arrays (after FOXP2 probing) as an additional control for sample loading. The results from the ubiquitin probed array were used to confirm that differential expression was not the result of unequal sample loading. Normal tissues

The normal tissues on the MTE array are from non-diseased victims of sudden death/trauma and are pooled from a number of individuals. The expression of the FOXP2 mRNA in normal human tissues showed a more restricted expression pattern than that observed for the FOXPl mRNA (Banham et al . , 2001) (Figure 1A) . Lai et al . (2003) have recently examined human F0XP2 and murine Foxp2 mRNA expression in the developing CNS . They identified expression in the cortical plate, thalamus, striatum, cerebellar Purkinje cells and inferior olives of the medulla, among other regions, and suggested that the gene may be involved in development of corticostriatal and olivocerebellar circuitry. The MTE array data indicates that expression in these regions persists into adulthood. Of note, these new data also identify expression in the hippocamupus, which is a surprising finding, given that FOXP2/Foxp2 are not expressed in this region during embryonic development or in newborn. The strongest expression was observed in gastric tissues, bladder and uterus which had not been investigated in the previous F0XP2 study. While FOXPl mRNA was widely, expressed in lymphoid tissues this was not found to be ^"the case -for FOXP 2. Although strong expression of the murine Foxp2 mRNA had- been reported in spleen (Shu et al . , 2001) only weak expression of F0XP2 was observed in spleen, however this expression was higher than in other lymphoid tissue. Interestingly the strongest expression of murine Foxp2 was reported to be in lung tissue. This was not the case for the human F0XP2 gene, which showed relatively low expression in lung.

In contrast to FOXPl , which was strongly expressed in all foetal tissues on the array, the FOXP2 mRNA expression was more restricted in embryonic tissues being expressed primarily in the brain, liver and lung indicating that these molecules may have different roles during embryonic development. While the expression of F0XP2 had been previously reported in these tissues (Lai et al . , 2001), its absence or very low level expression in foetal heart, spleen and thymus had not.

Neoplastic tissues

Analysis of F0XP2 mRNA expression in human tumour tissues showed differential F0XP2 mRNA expression in some tumours, compared to their adjacent histologically normal tissue (Figure 2A) . A similar pattern had been observed previously with FOXPl (Banham et al . , 2001) indicating that both of these forkhead proteins may play an important role in cancer. Reduced expression of the FOXP2 mRNA was frequently observed in colon tumours (Fig 2A- K3 , K4 , K5 , K6 , K8 , K10, Kll) . The FOXP2 gene is not strongly expressed in normal kidney but 4 renal cancer cases (Fig 2A- B2 , B4 , B6, B14) showed less mRNA in the tumour than in the matched normal tissue. Also 4 renal cancer cases did not express the FOXP2 mRNA in either the tumour or normal tissue (Fig 2A- 1A/B, 9A/B, 10A/B, 11A/B) , . although for 3 of these cases (Fig 2A- 1, 10, 11) this is likely to be due to insufficient sample loading on the array. The three prostate cancer cases (Fig 2A- HE, 12E, 13E) all showed reduced FOXP2 mRNA in the tumour tissue. Three breast cancer cases (Fig 2A- IE, 2E, 7E) showed reduced FOXP2 expression in the tumour tissue. One of the uterine tumours showed loss of F0XP2 expression in both tumour and normal tissue (Fig 2A- 7D/E) while two more showed decreased F0XP2 expression in the tumour tissue (Fig 2A- 4H, 5H) . One stomach tumour (Fig 2A- 2N) showed decreased FOXP2 expression in the tumour tissue while the apparent overexpression seen in other tumours was the result of unequal sample loading . An important consideration when interpreting this data is that adjacent histologically normal tissues from a cancer patient are not necessarily normal. Thus early genetic changes which do not affect the appearance of the tissue may also be present in the matched normal tissue of cancer patients. In addition, the cellular composition of normal and tumour tissues are different .

Blood disease profiling array

This array includes samples from 48 patients with one of several diseases of the blood together with normal individuals and patients with von Willebrand's disease as controls. Six different blood fractions were isolated from each patient's blood sample- total leukocytes, polvmorphonuclear cells, mononuclear cells, CD14 positive cells, CD19 positive cells and CD3 positive cells.

Hybridization of the F0XP2 cDNA probe to this array showed widespread very low level mRNA expression in blood fractions with no apparent disease specific or fraction, specific. ^• expression. The F0XP2 mRNA was more strongly expressed in lymph node indicating that there are cell populations -present in this tissue that more strongly express F0XP2 mRNA than those in the blood fractions.

Cancer Cell Line Profiling Array

The expression of FOXP2 in cancer cell lines was investigated using BD Clontech' s Cancer Cell Line Profiling array (Figure 4A) . This array enables the expression profile of a target gene to be investigated in 26 human cancer cell lines (described in figure 4B) . The samples for each cell line include an untreated reference sample together with samples from the cell line treated with 26 different agents that affect cell growth, differentiation, apoptosis and many other biological processes (described in figure 4B) .

FOXP2 was differentially expressed in the cancer cell lines (Figure 4A) . Of the three lung cancer cell lines on the array only the NCI-H1299 cell line strongly expressed F0XP2 mRNA. F0XP2 mRNA was only weakly expressed in three of the six colon cell lines while the remaining three cell lines were largely negative. This finding is consistent with the down- regulation of FOXP2 expression observed in the patients with colon cancer on the MTN array. Of the three breast cancer cell lines on the array only the estrogen receptor negative cell line (MDA-MB-231) appeared to express F0XP2 mRNA. Both ovarian and cervical cell lines were F0XP2 positive. Consistent with the MTN array, prostate cell lines were largely F0XP2 negative. The neuroblastoma cell lines were largely F0XP2- negative while the glioblastoma cell line U-87 MG did express some F0XP2 mRNA. The two renal cell adenocarcinoma cell lines showed the highest levels of F0XP2 expression.

The- differential expression of FOXP2 within a cell line in response to treatment can only be determined when the signal" from" the control untreated sample (12) is readable. In some instances, non-specific hybridisation of the F0XP2 probe to the membrane occurred and the signal is unreadable for a number of cell lines. Such results were excluded from the analyses (readable rows are indicated in bold on table 1) .

F0XP2 expression was most commonly affected by hydrogen peroxide, which induces oxidative stress, being up-regulated in five cell lines and down-regulated in two (Fig 4A) . Desferrioxamine causes hypoxia and in response to this treatment F0XP2 expression was down-regulated in two cell lines and up-regulated in one. There are a number of reports in the scientific literature indicating that other forkhead proteins are also linked to oxidative stress (Furukawa-Hibi et al . , 2002; Kops et al . , 2002) and to hypoxia (Scott et al . , 2002; Tang & Lasky, 2003) .

The second most frequent finding was the up-regulation of FOXP2 mRNA expression in response to gamma irradiation, which induces DNA damage. This was identified in four cell lines, including those derived from lung, colon and kidney tumours, indicating that this is not a cell type specific response. Mammalian cells respond to the genomic changes caused by exposure to ionizing radiation in an attempt to maintain genomic integrity. This response involves a number of pathways including DNA repair pathways and cell-cycle checkpoints (recently reviewed by Belli et al . , 2002; Bartek & Lucas, 2001) . Stabilization of the p53 tumour suppressor protein is also induced by DNA damage and an association with DNA damage may provide an insight into the biological significance of reduced FOXP2 mRNA expression in human tumours.

Results from the cell line array also demonstrated the effects of drug treatment on F0XP2 mRNA expression. Two cell lines -showed reduced F0XP2 mRNA expression after treatment with Actinomycin D (a DNA-binding, transcription inhibitor) , Cis-platin and Thiotepa (both act by acylating at N-7 position of guanine) , Taxol (a Tubulin-active drug) and N- (Phosphonacetyl) -L-aspartate (PALA) (an RNA synthesis inhibitor) .Other treatments also affected F0XP2 mRNA expression as illustrated in figure 4A and Table 1. These data suggest that F0XP2 expression is affected by a range of drugs used to treat cancer. PCR amplification of the entire POXP2 coding region from a cDNA library

A cDNA encoding the full-length F0XP2 protein was amplified by PCR in a 50 μl reaction containing 1 μl of a 1:10 dilution of human foetal brain cDNA library (Cat. No. HL5504U, Clontech), oligonucleotide primers (forward 5'-CTC GGA TTC ATG ATG CAG GAA TCT GCG ACA GAG ACA ATA AGC-3' and reverse 5'-CTC GAA TTC TCA TTC CAG ATC TTC AGA TAA AGG CTC TTC TTC-3', Invitrogen) , at 0.5 M, dNTPs at 200 mM, lx reaction buffer and 2.5 Units of Pfu Turbo™ polymerase (Stratagene) . The reaction mixture was heated to 95 °C for 3 min followed by a 'touch-down' thermal cycling protocol: 94 °C, 30 s; 62 °C, 30 s; and 72 °C, 5 min. The annealing temperature was reduced in subsequent rounds, as follows: 61, 60, 59, 58, 57, 56, 55, 55, 52, 52 °C; then maintained at 50 °C for 26 cycles, with a final 10 min extension at 72 °C. Analysis of the reaction mixture on an agarose gel showed a band of approximately 2. Ikb . Following BamHI/j-JcoRI digestion and gel-purification, the F0XP2 PCR product was ligated into pBluescript II KS+ (Stratagene) and cloned using standard molecular biology- methods .-.(Sambrook et al . , 1989) to create plasmid B032.

The cDNA insert was fully sequenced allowing confirmation that the plasmid contained the F0XP2 cDNA sequence encoding the full length protein (accession No. AF337817) . The polyglutamine region had one less Q then in the reported sequence and there was a single base pair change of T to C in the codon encoding aa 29. However, as the T to C substitution was a silent change that did not change the encoded amino acid, this cDNA was used in protein expression studies. A recent report found little polymorphism and no expansions of the CAG/CAA repeat in 142 individuals with progressive movement disorders (Bruce and Margolis, 2002) . The CAG/CAA repeat was also found to be very stable in children with specific language impairment or autism (Newbury et al . , 2002) .

Eukaryotic expression of the FOXP2 protein in COS-1 cells

The F0XP2 cDNA was excised from B032 by Ba HI/EcoRI digestion and sub-cloned into pcDNA4HisMax C (Invitrogen) to create plasmid B040. Plasmid DNA was prepared for transfection according to the manufacturer's instructions' (Plasmid Midi Kit, Qiagen) .

pcDNA4HisMax/F0XP2 or pcDNA4HisMax alone were transfected into COS-1 cells using Fugene 6 transfection reagent, following the protocol described by the manufacturer (Roche Molecular Biochemicals) . Approximately 24h post transfection, the cells were washed with sterile PBS and harvested by trypsinisation. Cell pellets were snap frozen and stored at -

70°C while cytocentrifuge preparations were made for immunocytochemical staining and stored at -20°C (Erber et al . , 1984) . COS-1 cells transfected with pcDNA4HisMax/FOXP2 , or with pcDNA4HisMax alone were indirectly immunoenzymatically labelled using anti-Xpress antibody at 1:200 and DAKO Envision™ + System, HRP (DAB) as directed by the manufacturer (DAKO) . Expression of the FOXP2 protein, with an added His tag and Xpress epitope, was confirmed by immunostaining (Figure 5 A) .

Western blotting of lysates from the transfected COS cells with the anti-Xpress antibody was used to confirm that a protein of the expected molecular weight was expressed from the F0XP2 cDNA. Lysates of cells transfected with pcDNA4HisMax-FOXP2 or pcDNA4HisMax were resolved by 12% SDS- PAGE and blotted onto a transfer membrane (Immobilon™, Millipore) . The membrane was blocked by overnight incubation in TBS/3% non-fat milk powder (blocking buffer) at 4 °C, then incubated with the anti-Xpress antibody diluted 1:5,000 in TBS/10% normal human serum, for lh at room temperature. After extensive washing in TBS/O.05% Tween, the membrane was incubated with rabbit anti-mouse immunoglobulins-HRP conjugate. (DAKO)- diluted 1:2,000 in blocking buffer, for lh at room temperature . The ^' membrane was washed thoroughly and the antibody- binding was detected by a chemiluminescence (ECL™, AP Biotech) after exposure of the blot to film.

A band of the predicted molecular weight was detected by Western blotting using the anti-Xpress antibody indicating that the full length FOXP2 protein was being expressed (Figure 5 C) .

F0XP2 may be a co-regulator of hormone-dependent nuclear receptors Nuclear receptors are hormone-dependent transcription factors that play essential roles in development, differentiation, cell proliferation and metabolism. They are under the control of a wide variety of hormones and ligands, such as steroids, retinoids, thyroid hormone, 1, 25-dihydroxy- vitamin D3 , and prostanoids . Ligand binding promotes a conformational change in receptor structure that promotes interaction between activated receptors and co-activators. Many of these receptors are therapeutic targets for disease treatment: antagonists of estrogen receptor-α (e.g. tamoxifen) are clinically used to treat breast cancer (Dees and Kennedy, 1998) whereas retinoic acid receptor (RAR) agonists and antagonists block the growth of a number of neoplastic cells including breast tumour cells (Fanjul et al . , 1998; Shiohara et al . , 1999) . Agonists against retinoid X receptors (RXR) and peroxisome proliferator-activated receptor-γ (PPARγ) are potential candidates for treating cancer and diabetes (Murkerjee et al . , 1997; Bischoff et al . , 1998; Elstner et al . , 1998; Spiegelman, 1998). Thus, the identification of molecules that selectively activate or inhibit specific nuclear receptors is of considerable biological significance^' and these have the potential for clinical applications.

Ligand-dependent activation of gene transcription by nuclear receptors such as RAR, thyroid hormone receptor (TR) , estrogen receptor (ER) and PPARγ are dependent on the recruitment of co-activators. The alpha-helical LXXLL motif found in some co-activators (where L is leucine and X is any amino acid) is sufficient for ligand-dependent interaction with nuclear receptors and this signature motif is commonly referred to as the nuclear receptor (NR) box. A large number of co-regulatory factors have been described for some of these receptors, for example the estrogen receptor recruits both co- activators, co-repressors or proteins that remodel chromatin, reviewed by (Sommer and Fuqua, 2001) .

The inventors have identified the presence of an LXXLL motif in the N-terminus of the FOXPl, FOXP2 , FOXP3 and the FOXP4 proteins (Figure 6) indicating that all the members of this subfamily have the potential to bind and co-regulate nuclear receptors. Additional amino acids flanking the NR box (e.g. positions -1 and +6) and the identity of the amino acids at +2 and +3 (XX) are important for co-activator choice by nuclear receptors (Shao et al . , 2000) . Position +3 differs in the FOXPl protein, while the F0XP3 protein differs from the other FOXP proteins at positions -1 and +6, raising the possibility that these may reflect a functional difference in receptor binding. An analysis of the expression of the FOXPl protein expression in breast cancer, by the inventors, has recently identified a statistically significant association with the expression levels of the estrogen receptor protein. Further studies will determine whether the FOXP proteins can bind to nuclear receptors and whether their expression or protein:protein interactions are hormonally regulated.

FOXP2 . contains a potential nuclear-receptor-interaction- domain.

A conserved bipartite leucine-rich helix LXXXIXXX (I/L) termed a nuclear-receptor-interaction domain (NRID) mediates the binding of the silencing mediator of the co-repressors for retinoid and thyroid hormone receptors (SMRT) and nuclear receptor co-repressor (NCOR) proteins to the ligand-free forms of nuclear receptors. This LXXXIXXX (I/L) motif is similar to the LXXLL recognition motif present in nuclear co-activators but is predicted to form an extended alpha-helix which is one helical turn longer than the co-activator motif (recently reviewed by Jepsen and Rosenfeld, 2002) .

FOXP2 has a single NRID motif near the C-terminus (aa 594-602 LVKNIPTSL) . It is possible that this motif may have a role in recruiting co-repressor proteins and that there may be alternatively spliced forms of FOXP2 that generate a protein containing more than one of these motifs. Short forms of the proteins that lack this motif have already been described.

Alternative splicing alters the number of nuclear receptor interaction domains within the FOXP2 protein.

There are a number of alternatively spliced forms of the FOXP2 protein that are predicted to change its amino-acid sequence and to either add or delete functional domains. The expression of a number of these variants has been confirmed in several human tissues using PCR (Bruce and Margolis, 2002) . However, there have been no quantitative studies investigating the expression levels of these variants, nor has their expression been characterised at the protein or functional levels. It is likely, based on the available sequence data,_, that the. proteins encoded by these variants have different biological roles.

A number of these variants (illustrated as FOXP2 V and VI in Figure 9) result in a C-terminally truncated protein that lacks the forkhead DNA binding domain and the potential NRID domain. Interestingly the inclusion of the additional exon encoding the sequence ELLPETKLCICGHSSGDGHPHNTFA in variants VI and II leads to the creation of an additional LXXLL motif, LQELL, within the N-terminus of the FOXP2 protein, aa 85-89. Alternative splicing of the MTAl gene has also been reported to generate an additional LXXLL motiE (Kumar et al . , 2002). It has been previously reported that there is a differential requirement for the number of LXXLL motifs used by different receptors. While the ER requires only a single NR box other receptors such as RAR-RXR and PPARγ-RXR heterodimers require two NR boxes for co-operative binding (Westin et al . , 1998). The protein with two NR boxes may, therefore, be able to mediate co-operative binding between FOXP2 and homodimeric or heterodimeric nuclear receptors .

The inventors have previously described an N-terminal splice variant of the FOXPl protein that lacks the LXXLL motif (encoded by cDNA in plasmid pAB196) . This raises the possibility that FOXP proteins that lack these motifs may have very different biological functions to those that retain the motif.

FOXP2 protein is found primarily localised near the plasma membrane rather than the nucleus when transfected into MCF-7 cells .

To investigate a possible FOXP2 ER interaction a tagged FOXP2 protein was transiently expressed in the MCF-7 cell line to determine whether the FOXP2 protein co-localised with the ER. ^"

Initial transfection experiments using the Fugene reagent and expressing a recombinant tagged FOXP2 protein (from plasmid B040) in COS cells demonstrated that the protein was present in both the nucleus and cytoplasm 24 hours post- transfection (Figure 5) . However, previous experiments using our DEAE Dextran transfection method indicated that the related full-length FOXPl protein was nuclear 72 hours post transfection, while both nuclear and cytoplasmic FOXPl protein was observed after 24h using the Fugene protocol . Previous time course experiments in the laboratory using the NPM-ALK protein have indicated that several days may be required for the expressed protein to translocate from the cytoplasm to the nucleus.

To investigate whether after 3 days the FOXP2 protein had a nuclear localisation and if there was any evidence for its co-localisation with the ER; pcDNA4HisMax/FOXP2 or pcDNA4HisMax were transfected into MCF-7 cells using the Fugene 6 transfection reagent according to the manufacturer's instructions (Roche Molecular Biochemicals) . Approximately 3 days post-transfection, the cells were washed with sterile PBS and harvested by trypsinisation. Cell pellets were snap frozen and stored at -70 C while cytocentrifuge preparations were made for immunocytochemical staining and stored at -20 C.

MCF-7 cells transfected with pcDNA4HisMax/FOXP2 , or with pcDNA4HisMax were indirectly immunoenzymatically labelled using anti-Xpress antibody at 1:200 and DAKO Envision + System, HRP (DAB) as directed by the manufacturer (DAKO) . Surprisingly the FOXP2 protein was not localised in the nucleus but cytoplasmic/plasma membrane staining of the transfected _ MCF-7 cells was observed (Figure 7) . Further studies will be required to determine whether this is a time course dependent, transfection protocol dependent or cell type specific phenomenon. Many of the FOXP2 transfected COS cells in a subsequent transfection did not look normal raising the possibility that over expression of the FOXP2 protein is toxic and/or drives the cells into apoptosis or senescence.

FOXP2 transfectants do not express nuclear ER and F0XP2 protein may affect the subcellular localisation of the ER. MCF-7 cells transfected with pcDNA4HisMax/F0XP2 were double immunofluorescently labelled using isotype-specific secondary antibodies recognising either the His tag primary antibody (green) to detect the tagged F0XP2 protein or the ER antibody ID5 (red) .

The results demonstrating the cell membrane localisation of the FOXP2 protein (Figure 8) were the same as obtained in the previous experiment (Figure 7) . However, the immunolabelling results obtained with the anti-ER antibody, ID5, were unexpected. The FOXP2 transfected MCF-7 cells did not show nuclear ER staining, instead there was evidence for the ER protein localising at the cell membrane (Figure 8f, indicated with white arrow) . Immunostaining human tumours with the anti-FOXPl antibody JC12 has identified a cell membrane staining pattern in cases of diffuse large B-cell lymphoma (DLBCL) and bladder cancer indicating that this finding may be biologically relevant in human cancers .

Further studies will be required to investigate this finding and to determine the localisation of the F0XP2 protein and ER _. in other transfected cell lines. Hopefully the production of an antibody against the FOXP2 protein will enable its, study in vivo under physiological conditions. Interestingly a very recent report in Nature has described a naturally occuring short form of the metastatic tumour antigen 1 that sequesters ER in the cytoplasm, and enhances non- genomic responses of ER (Kumar et al . , 2002), while a short form of F0XP2 that lacks the forkhead domain has also been reported to aggregate in the cytoplasm (Bruce and Margolis, 2002) . Membrane Estrogen receptors

Many of the long term actions of steroids are generally termed "genomic actions" (reviewed in (Evans, 1988; Mangelsdorf et al . , 1995)) . These are considered to be largely mediated by nuclear transcription factors that recognise specific recognition sites in the promoters of steroid- responsive genes. However there are some rapid responses that occur within seconds or minutes of steroid application and membrane-associated versions of steroid receptors have thus been proposed to account for these actions (Pietras and Szego, 1977) . For example estrogen can induce extremely rapid increases in the levels of intracellular second messengers such as calcium and cAMP, as well as activation of mitogen- activated protein kinase and phospholipase (Pietras and Szego, 1975; Pietras and Szego, 1977; Wehling, 1997; Collins and Webb, 1999) . There are three groups of structures proposed for the plasma membrane estrogen receptor. These include, (1) a protein identical to the nuclear receptor; (2) a protein sharing some of the domains present in the nuclear receptor; and (3) a protein unrelated to the nuclear receptor (reviewed in (Nadal et al . , 2001)) . Recently there has been renewed_, interest _, in this area as better understanding of the rapid actions of estrogen may lead to new therapeutic strategies for the treatment of cell proliferative, neurodegenerative and cardiovascular defects .

Production of a monoclonal antibody, 2FX2AD2, against the FOXP2 protein.

A region of the F0XP2 cDNA encoding the C-terminus of the

FOXP2 protein from amino acids SPTLVK to the stop codon

SEDLE*, was amplified by PCR using primers incorporating BamHI

(5') and EcoRI (3') restriction sites, plasmid B032 as template and Pfu Turbo Polymerase (Stratagene) .This BamRI/EcoRI fragment was then ligated into the same sites in the bacterial expression vector pGEX-4T-2. The expression plasmid was sequenced to confirm PCR fidelity. The pGEX/F0XP2 plasmid was then transformed into E. coli strain BL21 and recombinant GST-FOXP2 fusion protein expression was induced with ltnM IPTG and the protein purified on glutathione sepharose beads according to standard protocols (AP Biotech) .

Mouse monoclonal antibody 2Fx2AD2 (isotype IgGl) was raised as described previously (Mason et al . , 1983) Hybridoraa supernatants were screened for reactivity with the GST-F0XP2 protein and GST alone by ELISA and positive supernatants were then tested on transfectants . The one hybridoma cell line that secreted antibodies recognising both recombinant protein and F0XP2 transfectants was then cloned to create the cell line 2FX2AD2.

The 2Fx2AD2 FOXP2 monoclonal antibody recognises the FOXP2 protein in both frozen and routinely fixed COS-1 cell transfectants .

The _ immunohistochemical staining of cytospin preparations of COS-1 cells transfected with the FOXP2 cDNA (plasmid pcDNA4HisMax(C) /FOXP2, B040) using the FOXP2 monoclonal antibody (1:10 dilution), 2FX2AD2, is illustrated in Figure lib. A commercial goat polyclonal antibody (Everest Biotech) against FOXP2 (1:20 dilution) also immunostained the transfected cells (Fig lie) . The Xpress antibody against the epitope tag fused to the FOXP2 protein illustrates the expression of the recombinant FOXP2 protein in the transfected COS cells (Fig 11a) . Interestingly, our 2Fx2AD2 FOXP2 monoclonal antibody strongly recognises the nuclear FOXP2 protein but, unlike the goat polyclonal antibody and the Xpress antibody, does not give the expected amount of cytoplasmic staining. It is possible that the epitope recognised by our antibody may not be present in the cytoplasmic protein or may be modified, eg by post translational modification such as phosphorylation, preventing antibody binding. The subcellular localisation of other forkhead transcription factors is reported to be regulated by phosphorylation.

COS-1 cells expressing FOXP2 were also fixed in buffered formalin and then embedded in paraffin wax. The 2Fx2AD2 antibody continues to recognise the F0XP2 protein in sections cut from these formalin fixed paraffin embedded cell pellets demonstrating its utility to detect the F0XP2 protein in routinely fixed clinical material (Fig He) . We have also shown that the Everest FOXP2 polyclonal antibody detects the FOXP2 protein in routinely fixed material (Fig llf) .

For staining paraffin embedded tissues with antibody 2Fx2AD2 sections were subjected to antigen retrieval in a microwave pressure cooker (2 mins at pressure) in 50 mM Tris, 2mM ETDA pH9. Primary antibody 2Fx2AD2 was used at a dilution of between 1:10 or 1:20 and the secondary antibody and HRP detection system was the DAKO Envision™ kit. The Everest goat antibody was used at a 1:100 dilution on sections subjected to the antigen retrieval system described above. The secondary antibody, rabbit anti-goat Ig-HRP conjugate (DAKO) , was diluted 1:200 and staining was visualised with the DAKO Envision™ substrate.

The 2FX2AD2 antibody is specific for FOXP2 and does not recognise the closely related proteins FOXPl and FOXP4. There is significant sequence similarity between the closely related FOXPl, F0XP2 and F0XP4 proteins. Therefore the specificity of the FOXP2 monoclonal antibody 2FX2AD2 was tested by immunostaining COS-l cells transfected with the FOXPl cDNA (pAB195, Banham et al . , 2001) (Fig 12c) and with the FOXP4 cDNA, B059 (Fig 12e) . To generate the FOXP4 expression plasmid, pcDNA4HisMax (C) /FOXP4 (B059) , a FOXP4 cDNA clone with accession no. BC040962, IMAGE : 5527139 was used as PCR template. PCR, primers incorporated BarriΑI (5') and EcoRI (3') restriction sites. The FOXP4 PCR product was cut with these enzymes and ligated into the BamHI and EcoRI sites of the vector pcDNA4HisMax(C) and the FOXP4 coding region was verified by sequencing. This experiment confirmed that the 2Fx2AD2 monoclonal antibody specifically recognised the FOXP2 and not the FOXPl or F0XP4 proteins.

The FOXP2 protein is differentially expressed in breast tumours .

Preliminary data obtained from immunohistochemical staining studies with the 2Fx2AD2 antibody on routinely fixed sections from breast tumours has detected both the nuclear expression (Fig 13a-d) , cytoplasmic expression (Fig 13f) and the absence of this protein (Fig 13e,g,h) in a number of cases, . with, the majority of cases appearing to lack FOXP2 expression. Strong FOXP2 expression was not observed in the majority of the tumour cells, in any of the cases, while nuclear staining was more common than cytoplasmic. As the Xpress antibody gives stronger staining of FOXP2 transfectants than the FOXP2 monoclonal antibody (indicating that stronger labelling of the FOXP2 protein can be obtained) it is possible that further optimisation of the immunostaining with 2Fx2AD2 may increase the amount of staining observed. These results are particularly significant when considering the potential association between the FOXP2 protein and the estrogen recepto .

FOXP2 maps to a tumour suppressor gene locus on chromosome

7q31

Consistent deletions and loss of heterozygosity (LOH) in polymorphic markers in a chromosomal fragment are known to indicate genomic regions containing tumour suppressor genes. The FOXP2 gene has been mapped to 7q31 with a possible map location of 7q31.33 from the locus link data. There is an extensive literature reporting genomic abnormalities, both losses and gains, within the 7q31 region for a wide range of human cancers .

Gastric cancer

The MTN array hybridisation data indicate that changes in FOXP2 expression may be a frequent event in colon cancer. Importantly there are also genetic studies implicating copy number changes in the 7q31 region in gastric cancer. A study of 33 • primary gastric carcinomas reported that increases ^" at 7q31-q32 were observed in early stage tumours (Koizumi et al . , 1997) .. "iHigh-level amplifications of 7q31-q32 were also identified in gastric adenocarcinomas (Nessling et al . , 1998) and 7q31 in 7-10% of primary gastric cancers (Sakakura et al . , 1999; Nakanishi et al . , 2000). A study of gastric carcinoma cell lines identified the most frequent marker as a partial gain of chromosome 7 with breakpoints on 7q22 and 7q31 (Chun et al . , 2000) . One recent study reported 30% of primary colorectal cancers as having gains of 7q31-36 (Aragane et al . , 2001) . These studies indicate that gains of genomic DNA in the region containing FOXP2 are important in this malignancy. B-cell lymphoma

Losses of 7q are usually associated with myeloid disorders or myelodysplastic syndromes. However an association has been identified between del(7q) and small cell non- Hodgkin's lymphoma (NHL) (Offit et al . , 1995) or splenic lymphoma with villous lymphocytes (Oscier et al . , 1993).

In small cell NHL this region has been reduced to a commor deleted region in 6 patients between loci D7S685 and D7S514 betweer bands 7q31 and 7q32 (J.M. et al . , 1997).

In splenic marginal zone B-cell lymphoma (SMZBCL) 7q31-32 allelic loss is a frequent finding that may be useful as a genetic marker of this neoplasia, in conjunction with other morphologic, phenotypic, and clinical features. There was also a nearly significant correlation between 7q allelic loss and tumoral progression (Mateo et al . , 1999). Another study also reported 7q31 deletion in mantle cell lymphoma (MZL) (Cuneo et al . , 2001) . In a study of 47 cases of SMZBCL, 12 cases showed a deletion of 7q. The findings from this study supported the possibility that there are two forms of SMZBCL, one with gain of 3q and the other with deletion at 7q (Sole et al . , 2001) . Recently an -analysis of the IgV(H) somatic mutations in SMZBCL has defined a group of unmutated cases with frequent 7q deletion and adverse clinical course (Algara et al . , 2002).

Low-grade marginal zone B-cell lymphomas (MALT) can transform into aggressive DLBCL. Allelic imbalances at 7q31 have been detected in both MALT lymphoma and DLBCL (Starostik et al . , 2002) . The finding that the FOXP2 mRNA is more strongly expressed in spleen when compared to other lymphoid tissues is potentially significant as secondary lymphoid follicles in the spleen contain marginal zones while those from other sites, such as tonsil, do not.

Myeloproliferative disease

Deletion of part of chromosome 7q is a consistent aberration in primary and therapy-induced human acute leukaemias and myelodysplasias (MDS) . Especially in patients that have been exposed to genotoxic agents. In most studies this has invariably been associated with a poor prognosis. One study found that when the 7q31 band was present patients survived longer suggesting that genetic information at 7q31 may delay cytogenetic and clinical progression of myeloid disease with 7q loss (Pedersen and Ellegaard, 1994) . A study of 25 cases of refractory acute myeloid leukaemia (AML) found an overlapping minimal region of loss between 7q31.2-q32 (El- Rifai et al . , 1997) while a separate study of de novo AML found smallest commonly deleted regions at 7q31.1 and 7q33-34 (Koike et al . , 1999). A more recent study of deletion and monosomy,- of chromosome 7 in 150 patients with myeloproliferative diseases found 8/150 patients with monosomy 7 and .4/150 with deletions of 7q, with 7/45 patients having LOH at 7q31.1 (Tripputi et al . , 2001).

Breast cancer

LOH at 7q31 was found in 40% of informative patients tudied and these had a significantly shorter metastasis-free survival and overall survival indicating that 7q31 may contain a breast tumour or metastasis suppressor gene (Bieche et al . ,

1992) . A study comparing genetic changes in primary cancer compared to relapse concluded that inactivation of a putative tumour suppressor gene at 7q31 was a very early event in breast tumourigenesis (Champeme et al . , 1995) a finding which was also reported in another study (Sztan et al . , 1996) . Interestingly LOH at the FOXPl locus is an early event in cancer development with changes in protein expression having been detected by the inventors in pre-neoplastic lesions (Banham et al . , 2001). A study looking for metastasis-related genes identified 7q31 as a candidate region (Driouch et al . , 1998) .

A large study of LOH at 7q31-q32 in 683 breast tumours found varying rates of LOH with a maximum of 40% and did not find evidence that this genetic alteration was a strong determinant of disease (Devilee et al . , 1997).

Prostate cancer

Cytogenetic studies have demonstrated that 7q22-32 is commonly altered in prostate adenocarcinomas . A study of 43 cases found that 47% had LOH at 7q with 7q31. l-7q31.2 being the most frequently deleted region. This change was reported to be ..an early event in prostate cancer (Latil et al . , 1995). The most common site of allelic loss included loci D7S523 and D7S486. at 7q31.1. LOH at 7q31.1 correlated with higher tumour grade and lymph node metastasis indicating that a tumour suppressor gene may be present in this region (Takahashi et al . , 1995). A study using microcell-mediated chromosome transfer demonstrated that a metastasis suppressor gene mapped to 7q31.2-32 (Ichikawa et al . , 2000). One study found both losses and gains of 7q31 and concluded that the 7q31 region is genetically unstable in prostate cancer, possibly due to fragility at FRA7G (Jenkins et al . , 1998). In contrast 7q31 gains have also been reported in prostate cancer (Sattler et al . , 1999; Alers et al . , 2000). Gains of 7q, particularly 7q31, were found to be one of the most common genetic alterations in PIN, which is often a precursor of prostatic cancer (Qian et al . , 1999).

Head and neck cancer

LOH at 7q31, being found in recurrent tumours, is a novel prognostic factor independent of other clinical factors for head and neck squamous cell carcinoma (Matsuura et al . , 1998). A comparison of grade 2 and 3 tumours also showed a highly changed frequency of 7q31 LOH (Fiedler et al . , 2002).

Other tumour types

7q abnormalities have also been detected in non-small cell lung cancer (Luk et al . , 2001), pancreatic cancer

(Achille et al . , 1996), renal cell carcinoma (Shridhar et al . , 1997; Glukhova et al . , 1998), ovarian cancer (Huang et al . , 1999), hepatocellular carcinoma (Rao et al . , 2001), osteosarcoma (Stock et al . , 2000), biliary tract carcinoma

(Shiraishi et al . , 2001), thyroid neoplasms (Oriola et al . , . 2001) and nasopharyngeal carcinoma (Tan et al . , 2002).

FOXP2 expression on cancer cell line profiling array

References

Achille, A., M. 0. Biasi, G. Za boni, et al . (1996). Chromosome 7q allelic losses in pancreatic carcinoma. Cancer Res. 56: 3808-3813.

Alers, J. C, J. Rochat, P. J. Krijtenburg, et al . (2000) . Identification of genetic markers for prostatic cancer progression. Lab. Invest. 80: 931-942.

Algara, P., M. S. Mateo, M. Sanchez-Beato, et al . (2002). Analysis of the IgV(H) somatic mutations in splenic marginal zone lymphomas defines a group of unmutated cases with frequent 7q deletion and adverse clinical course. Blood 99: 1299-1304.

Aragane, H. , C. Sakakura, M. Nakanishi, et al . (2001). Chromosomal aberrations in colorectal cancers and liver metastases analyzed by comparative genomic hybridization. Int. J. Cancer 94: 623-629.

Banham, A. H., N. Beasley, E. Campo, et al . (2001). The FOXPl winged helix transcription factor in a novel candidate tumor suppressor gene on chromosome 3p. Cancer Res. 61: 8820- 8829.

Bartek, J. , J. Lukas (2001). Mammalian Gl- and S-phase checkpoints in response to DNA damage. Curr. Opinion Cell Biol. 13: 738-747.

Belli. M. , 0. Sapora, M.A. Tabocchini (2002). Molecular targets in cellular response to ionizing radiation and implications in space radiation protection. J. Radiat. Res. 43: S13-S19. Bennett, C. L., M. E. Brunkow, F. Ramsdell, et al .

(2001) . A rare polyadenylation signal mutation of the FOXP3 gene (AAUAAA-AAUGAA) leads to the IPEX syndrome. Immunogenet . 53: 435-439.

Bennett, C. L., J. Christie, F. Ramsdell, et al . (2001). The immune dysregulation polyendocrinopathy, enteropathy, X- linked syndrome (IPEX) is caused by mutations of FOXP3. Nature Genet. 27: 20-21.

Bieche, I., M. H. Champeme, F. Matifas, et al . (1992). Loss of heterozygosity on chromosome 7q and aggressive primary breast cancer. Lancet 339: 139-143.

Bischoff, E. D., T. E. Moon, R. A. Heyman, et al . (1998). Beyond tamoxifen: the retinoid X receptor-selective ligand LGD1069 (TARGRETIN) causes complete regression of mammary carcinoma. Cancer Res. 58: 479-484.

Borkhardt, A., R. Repp, 0. A. Haas, et al . (1997). Cloning and characterization of AFX, the gene that fuses to MLL in acute leukemias with translocation t(X;ll) (ql3;q23). 0ncogene.-14 : 195-202.

Breeden, L. L. (2000) . Cyclin transcription: Timing is everything. Curr. Biol . 10: R586-R588.

Bruce, H. A. and R. L. Margolis (2002). FOXP2 : novel exons, splice variants, and CAG repeat length stability. Human Genetics 111: 136-144.

Brunkow, M. E., E. W. Jeffery, K. A. Hjerrild, et al . (2001) . Disruption of a new forkhead/ inged-helix protein, scurfin, results in the fatal lymphoproliferative disorder of the scurfy mouse. Nature Genet. 27: 68-73.

Champeme, M. H. , I. Bieche, M. Beuzelin, et al . (1995). Loss of heterozygoisty on 7q31 occurs early during breast tumorigenesis . Genes Chrm. Cancer 12: 304-306.

Chatila, T. A., F. Blaeser, H. Nga, et al . (2000). JM2 , encoding a forkhead-related protein, is mutated in X-linked autoimmunity-allergic disregulation syndrome. J. Clin. Invest. 106: R75-R81.

Chun, Y. H., J. I. Kil, Y. S. Suh, et al . (2000). Characterization of chromosomal aberrations in human gastric carcinoma cell lines using chromosome painting. Cancer Genet. Cytogenet. 119: 18-25.

Cogoni, C. and G. Macino (2000) . Post-transcriptional gene silencing across kingdoms. Genes Dev. 10: 638-643.

Collins, P. and C. Webb (1999) . Estrogen hits the surface. Nature Med. 5: 1130-1131.

Cuneo, A., R. Bigoni, M. G. Roberti, et al . (2001). Molecular cytogenetic characterization of marginal zone B-cell lymphoma: correlation with clinicopathologic findings in 14 cases. Haematologica 86: 64-70.

Davis, R. J., C. M. D'Cruz, M. A. Lovell, et al . (1994). Fusion of PAX7 to FHKR by the variant t(l;3) (p36;ql4) translocation in alveolar rhabdomyosarcoma . Cancer Res. 54: 2869-2872. Dees, E. C. and M. J. Kennedy (1998) . Recent advances in systemic therapy for breast cancer. Curr. Opin. Oncol. 10: 517-522.

Devilee, P., J. Hermans, J. Eyfjord, et al . (1997). Loss of heterozygosity at 7q31 in breast cancer: results from an International Collaborative Study Group. The Breast Cancer Somatic Genetics Consortium. Genes Chrm. Cancer 18: 193-199.

Driouch, K. , M. Briffod, I. Bieche, et al . (1998). Location of several putative genes possibly involved in human breast cancer progression. Cancer Res. 58: 2081-2086.

Elbashir, S. M. , J. Harborth, W. Lendeckel, et al . (2001) . Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 411: 494-498.

El-Rifai, W. , E. Elonen, M. Larramendy, et al . (1997). Chromosomal breakpoints and changes in DNA copy number in refractory acute myeloid leukaemia. Leukemia 11: 958-963.

Elstner, E. , C. Muller, K. Koshizuka, et al . (1998). Ligands for peroxisome proliferator-activated receptorgamma and retinoic acid receptor inhibit growth and induce apoptosis of human" breast cancer cells in vitro and in BNX mice. Proc. Natl. Acad. Sci USA 95: 8806-8811.

Erber, W. N., A. J. Pinching and D. Y. Mason (1984). Immunocytochemical detection of T and B cell populations in routine blood smears. Lancet 1: 1042-1046.

Evans, R. M. (1988) . The steroid and thyroid hormone receptor super-family. Science 240: 889-895. Fanjul, A. N. , F. J. Piedrafita, H. Al-Shamma, et al .

(1998) . Apoptosis induction and potent antiestrogen receptor- negative breast cancer activity in vivo by a retinoid antagonist. Cancer Res. 58: 4607-4_.610.

Fiedler, W. , C. Hoppe, B. Schimmel, et al . (2002). Molecular characterization of head and neck tumors by analysis of telomerase activity and a panel of microsatellite markers. Int. J. Mol. Med. 9: 417-423.

Furukawa-Hibi, Y. , K. Yoshida-Araki, T. Ohta, K. Ikeda, N. Motoyama (2002) . FOXO forkhead transcription factors induce G(2)-M checkpoint in response to oxidative stress. J. Biol . Chem. 277:26729-26732.

Galili, N. , R. J. Davis, W. J. Fredericks, et al . (1993). Fusion of a fork head domain to PAX-3 in the solid tumor alveolar rhabdomyosarcoma . Nat. Genet. 5: 230-235.

Glukhova, L., A. F. Goguel, I. Chudoba, et al . (1998). Overrepresentation of 7q31 and 17q in renal cell carcinomas. Genes Chrm. Cancer 22: 171-178.

Guru, T. (2000). A silence that speaks volumes. Nature 404: 8.04^"--808.

Hammond, S. M. , A. A. Caudy and G. J. Hannon (2001). Post-transcriptional gene silencing by double-stranded RNA. Nature Rev. Gen. 2: 110-119.

Hollenhorst, P. C, G. Pietz and C. A. Fox (2001). Mechanisms controlling differential promoter-occupancy by the yeast forkhead proteins Fkhlp and Fkh2p: implications for regulating the cell cycle and differentiation. Genes & Dev. 15: 2445-2456.

Huang, H. , C. P. Reed, A. Mordi, et al . (1999). Frequent deletions within FRA7G at 7q31.2 in invasive epithelial ovarian cancer. Genes Chrm. Cancer 24: 48-55.

Hernandez; J.M., H., E. F. P. M. Schoenmakers, P. D. Cin, et al . (1997) . Molecular delineation of the commonly deleted segment in mature B-cell lymphoid neoplasias with deletion of 7q. Genes Chrm. Cancer 18: 147-150.

Ichikawa, T., S. Hosoki, H. Suzuki, et al . (2000). Mapping of metastasis suppressor genes for prostate cancer by microcell-mediated chromosome transfer. Asioan J. Androl . 2: 167-171.

Jepsen, K. and M. G. Rosenfeld (2002) . Biological roles and mechanistic actions of co-repressor complexes. J. Cell Sci. 115: 698-698.

Jenkins, R. B., J. Qian, H. K. Lee, et al . (1998). A molecular • cytogenetic analysis of 7q31 in prostate cancer. Cancer Res. 58: 759-766.

Koike, M., T. Tasaka, S. Spira, et al . (1999). Allelotyping of scute myelogenous leukemia: loss of heterozygosity at 7q31.1 (D7S486) and q33-34 (D7S498, D7S505) . Leuk. Res. 23: 307-310.

Koizumi, Y. , S. Tanaka, R-. Mou, et al . (1997). Changes in DNA copy number in primary gastric carcinomas by comparative genomic hybridization. Clin. Cancer Res. 3: 1067-1076. Kops, G. J. and B. M. Burgering (1999) . Forkhead transcription factors: new insights into protein kinase B (c- akt) signalling. J. Mol. Med. 77: 656-665.

Kops, G.J., T.B. Dansen et al . (2002). Forkhead transcription factor FOX03a protects quiescent cells from oxidative stress. Nature 419:316-321.

Kops, G. J., R. H. Medema, J. Glassford, et al . (2002). Control of cell cycle exit and entry by protein kinase B- regulated forkhead transcription factors. Mol. Cell. Biol. 22: 2025-2036.

Kumar, R. , R.-A. Wang, A. Mazumdar, et al . (2002). A naturally occuring MTAl variant sequesters estrogen receptor- alpha in the cytoplasm. Nature 418: 654-657.

Lai, C. S., S. E. Fisher, J. A. Hurst, et al . (2001). A forkhead-domain gene is mutated in a severe speech and language disorder. Nature 413: 465-466.

Lai, C. S., D. Gerrelli, A. P. Monaco, S. E. Fisher, A. J. Copp (200.3) . FOXP2 expression during brain development coincides with adult sites of pathology in a severe speech and language"- disorder . Brain 126: Epub ahead of print.

Latil, A., 0. Cussenot, G. Fournier, et al . (1995). Loss of heterozygosity at 7q31 is a frequent and early event in prostate cancer. Clin. Cancer Res. 1: 1385-1389.

Li, J. and P. K. Vogt (1993) . The retroviral oncogene gin belongs to the transcription factor family that includes the homeotic gene fork head. Proc . Natl . Acad. Sci. U.S.A. 90: 4490-4494. Luk, C, M. S. Tsao, J. Bayani, et al . (2001). Molecular cytogenetic analysis of non-small cell lung carcinoma by spectral karyotyping and comparative genomic hybridization. Cancer Genet. Cytogenet . 125: 87-99.

Mangelsdorf, D. J. , C. Thummel, M. Beato, et al . (1995). The nuclear receptor family - the second decade. Cell 83: 835- 839.

Mason, D.Y. , Cordell, J.L., Pulford, K.A.F. (1983). Production of monoclonal antibodies for immunocytochemical use. In, G.R. Bullock and P. Petrusz (eds) , Techniques in Immunocytochemistry, pp 175-216. London: Acaddemic Press, Ltd.

Mateo, M. , M. Mollejo, R. Villuendas, et al . (1999). 7q31-32 allelic loss is a frequent finding in splenic marginal zone lymphoma. Am. J. Pathol . 154: 1583-1589.

Matsuura, K. , K. Shiga, J. Yokoyama, et al . (1998). Loss of heterozygosity of chromosome 9p21 and 7q31 is correlated with high incidence of recurrent tumor in head and neck squamous .cell carcinoma. Anticancer Res. 18: 453-458.

Murkerj-ee, R., P. J. Davies, J. R. J. Paterniti, et al . (1997) . Sensitization of diabetic and obese mice to insulin by retinoid X receptor agonists. Nature 386: 407-410.

Nadal, A., A. B. Ropero, E. Fuentes, et al . (2001). The plasma membrane estrogen receptor: nuclear or unclear? Trends Pharm. Sci. 22: 597-599.

Nakamura, N. , S. Ramaswamy, F. Vazquez, et al . (2000). Forkhead transcription factors are critical effectors of cell eath and cell cycle arrest downstream of PTEN. Mol. Cell.

Biol. 20: 8969-8982.

Nakanishi, M. , C. Sakakura, Y. Fujita, et al . (2000). Genomic alterations in primary gastric cancers analyzed by comparative genomic hybridization and clinicopathological factors. Hepatogastroenterol . 47: 658-662.

Nessling, M., S. Solinas-Toldo, K. K. Wilgenbus, et al . (1998) . Mapping of chromosomal imbalances in gastric adenocarcinoma revealed amplified protooncogenes MYCN, MET, WNT2, and ERBB2. Genes Chrm. Cancer 23: 307-316.

Newbury, D. F. , E. Bonora, J. A. Lamb, et al . (2002). FOXP2 is not a major susceptibility gene for autism or specific language impairment. Am. J. Hum. Genet. 70: 1318- 1327.

Offit, K. , D. C. Louie, A. Parsa, et al . (1995). Del (7) (q32) is associated with a subset of small lymphocytic lymphoma with plasmacytoid features. Blood 86: 2365-2370.

Oripla, J., I. Halperin, C. Mallofre, et al . (2001). Screening of selected genomic areas potentially involved in thyroid ^"neoplasms . Eur. J. Cancer 37: 2470-2474.

Oscier, D. G. , E. Matutes, A. Gardiner, et al . (1993). Cytogenetic studies in splenic lymphoma with villous lymphocytes. Br. J. Haematol . 85: 487-491.

Parry, P., Y. Wei and G. Evans (1994). Cloning and characterization of the t(X;ll) breakpoint from a leukemic cell line identify a new member of the forkhead gene family. Genes Chrm. Cancer 11: 79-84. Pedersen, B. and J. Ellegaard (1994) . A factor encoded by 7q31 suppresses expansion of the 7q- clone and delays cytogenetic progression. Cancer Genet. Cytogenet. 78: 181-188.

Perreault, N. , J. P. Katz, S. D. Sackett, et al . (2001). Foxll controls the Wnt/beta-catenin pathway by modulating the expression of proteoglycans in the gut. J. Biol. Chem. 276: 43328-43333.

Pietras, R. and C. Szego (1975) . Endometrial cell calcium and oestrogen action. Nature 253: 357-359.

Pietras, R. J. and C. M. Szego (1977) . Specific binding sites for oestrogen at the outer surfaces of isolated endometrial cells. Nature 265: 69-72.

Qian, J. , R. B. Jenkins and D. G. Bostwick (1999). Genetic and chromosomal alterations in prostatic intraepithelial neoplasia and carcinoma detected by fluorescence in situ hybridization. Eur. Urol . 35: 479-483.

Rao,- : N., S. M. Gollin, S. Beaves, et al . (2001). Comparative genomic hybridization of hepatocellular carcinoma: correlation -with fluorescence in si tu hybridization in paraffin-embedded tissue. Mol. Diagn. 6: 27-37.

Sakakura, C, T. Mori, T. Sakabe, et al . (1999). Gains, losses, and amplifications of genomic materials in primary gastric cancers analyzed by comparative genomic hybridization. Genes Chrm. Cancer 24: 299-305^'.

Sattler, H. P., V. Rohde, H. Bonkhoff, et al . (1999). Comparative genomic hybridization reveals DNA copy number gains to frequently occur in human prostate cancer. Prostate 39: 79-86.

Scott, B.A., M.S. Avidan, CM. Crowder (2002). Regulation of hypoxic death in C. elegans by the insulin/IGF receptor homolog DAF-2. Science 296:2388-2391.

Sezinger, B. R. , H. P. Klinger, C. Junien, et al . (1991). Report of the committee on chromosome and gene loss in human neoplasia. Cytogenet. Cell. Genet. 58: 1080-1096.

Shao, G., R. A. Heyman and I. G. Schulman (2000). Three amino acids specify coactivator choice by retinoid X receptors. Mol. Endocrinol . 14: 1198-1209.

Shiohara, M. , M. I. Dawson, P. D. Hobbs, et al . (1999). Effects of novel RAR- and RXR-selective retinoids on myeloid leukemic proliferation and differentiation in vitro. Blood 93: 2057-2066.

Shiraishi, K. , K. Okita, T. Harada, et al . (2001). Comparative genomic hybridization analysis of genetic aberrations associated with development and progression of biliary tract carcinomas. Cancer 91: 570-577.

Shridhar, V., Q. C. Sun, 0. J. Miller, et al . (1997). Loss of heterozygosity on the long arm of human chromosome 7 in sporadic renal cell carcinomas. Oncogene 15: 2727-2733.

Shu, W., H. Yang, L. Zhang, et al . (2001). Characterization of a new subfamily of winged-helix/forkhead (Fox) genes that are expressed in the lung and act as transcriptional repressors . J. Biol. Chem. 276: 488-497. Sole, F., M. Salido, B. Espinet, et al . (2001). Splenic marginal zone B-cell lymphomas: two cytogenetic subtypes, one with gain of 3q and the other with loss of 7q. Haematologica 86: 71-77.

Sommer, S. and S. A. W. Fuqua (2001). Estrogen receptor and breast cancer. Cancer Biol. 11: 339-352.

Spiegelman, B. M. (1998) . PPAR-gamma: adipogenic regulator and thiazolidinedione receptor. Diabetes 47: 507- 514.

Starostik, P., J. Patzner, A. Greiner, et al . (2002). Gastric marginal zone B-cell lymphomas of MALT type develop along 2 distinct pathogenetic pathways. Blood 99: 3-9.

Stock, C, L. Kager, F. M. Fink, et al . (2000). Chromosomal regions involved in the pathogenesis of osteosarcomas . Genes Chrm. Cancer 28: 329-336.

Szalai, C, G. T. Kozma, A. Nagy, et al . (2002). Genomic approach to allergic diseases: present and future perspectives. Orv. Hetil. 143: 381-390.

Sztan, M., I. I. Besznyak, T. Kovacs, et al . (1996). Lack of correlation between survival and allele loss on chromsome 7q31-32 in primary breast cancer. Pathol Oncol. Res. 2: 48-51.

Takahashi, S., A. L. Shan, S. R. Ritland, et al . (1995). Frequent loss of heterozygosity at 7q31.1 in primary prostate cancer is associated with tumor aggressiveness and progression. Cancer Res. 55: 4114-4119. Tan, G., J. Xiao, Y. Tian, et al . (2002). Microsatellite analyses of loci at 7q31.3-q36 reveal a minimum of two common regions of deletion in nasopharyngeal carcinoma. Otolaryngol . Head Neck Surg. 126: 296-300.

Tang, T.T., L.A. Lasky (2003). The forkhead transcription factor FOX04 indices the down-regulation of hypoxia-inducible factor l{alpha} by a von Hippel-Lindau protein-independent mechanism. J. Biol. Chem. 278:30125-30135.

Teufel, A., E.A. Wong, M. Mukhopadhyay, N. Malik, H. Westphal . (2003). FoxP4, a novel forkhead transcription factor. Biochim. Biophys . Acta 1627:147-152.

Tripputi, P., B. Cassani, R. Alfano, et al . (2001). Chromosome 7 monosomy and deletions in myeloproliferative diseases. Leuk. Res. 25: 735-739.

Wang, B., D. Lin, C. Li and P. Tucker (2003). Multiple domains define the expression and regulatory properties of Foxpl forkhead transcriptional repressors. J. Biol. Chem. 278: 24259-24268.

Wehling, M. (1997) . Specific, nongenomic actions of steroid hormones. Annu. Rev. Physiol . 59: 365-393.

Weigel, D. and H. Jackie (1990) . The fork head domain: A novel DNA binding motif of eukaryotic transcription factors. Cell 63: 455-456.

Westin, S., R. Kurokawa, R. T. Nolte, et al . (1998). Allosteric interactions governing the assembly, structure and function of nuclear receptor heuerodimer-coactivator complexes. Nature 395: 199-203.

Wildin, R. S., F. Ramsdell, J. Peake, et al . (2001). X- linked neonatal diabetes mellitus, enteropathy and endocrinopathy syndrome is the human equivalent of mouse scurfy. Nature Genet. 27: 18-20.

Claims

m Clai ■ ms 62

1. A method of treating a disorder in a patient characterised by reduced expression of FOXP2 protein, which method comprises administering to a patient in need thereof a therapeutic amount of any of a nucleic acid molecule encoding said F0XP2 protein F0XP2 protein; or a composition capable of stimulating an increase in FOXP2 protein expression in the tissues of said patient.

2. A method according to claim 1 wherein said disorder is cancer .

3. A method according to claim 2 wherein said disorder is any of colon, breast, prostate, stomach, uterus or renal cancer.

4. A pharmaceutical composition comprising any of a nucleic acid molecule encoding FOXP2

FOXP2 protein; or a composition capable of increasing F0XP2 expression in the tissues of a human cell, together with a pharmaceutically acceptable carrier, diluent or excipient therefor.

5. Use of any of a nucleic acid molecule encoding F0XP2

FOXP2 protein; or a composition capable of increasing FOXP2 expression in the tissues of a human cell, in the manufacture of a medicament for treating a disorder characterised by a reduction of the level of FOXP2 in a patient.

6. Use according to claim 5 wherein the disorder is any of colon, breast, prostate, kidney, uterus or stomach cancer.

7. A method of identifying a compound capable of modulating expression of FOXP2, comprising contacting said compound with a test cell expressing FOXP2 and monitoring for altered expression of FOXP2 in said cell compared to a cell which has not been contacted with said compound.

8. A method according to claim 7 wherein said compound is capable of increasing expression of said FOXP2 protein.

9. A method of assessing the prognosis of a patient suffering from any of myeloproliferative disorders, acute myeloid leukaemia, splenic marginal zone B-cell lymphoma, breast cancer, prostate cancer, renal cancer, colon cancer, uterus cancer, stomach cancer and head and neck cancers, comprising monitoring the level of FOXP2 expression or gene loss, wherein loss of expression is characteristic of a poor prognosis .

10. A compound identified according to the method of claim 7 or 8.

11. Use of a compound according to claim 10 in the manufacture of a medicament for treating any of colon, prostate, breast, stomach, uterus or kidney cancer.

12. A method of diagnosing cancer in a patient which method comprises testing an isolated cell from an individual suspected of having cancer, wherein altered expression of F0XP2 is indicative of the presence of a cancer.

13. A method according to claim 12 wherein reduction of expression of said FOXP2 is indicative of the presence of cancer.

14. A method of subtyping splenic marginal zone

B-cell lymphomas, which method comprises testing the level of FOXP2 expression or gene copy number in a lymphoma cell of an individual wherein reduced F0XP2 expression associated with a mutation in IgV(H) is characteristic of a poor prognosis.

15. A method of screening preneoplastic lesions in an individual having an increased propensity to progress to a neoplastic state, which method comprises testing the level of FOXP2 expression in a preneoplastic cell of said individual wherein altered expression is indicative of the likelihood of said preneoplasm progressing to a neoplastic state.

16. A method according to claim 15, wherein said preneoplastic cell is prognostic of progression to a neoplastic state in breast or prostate cancer when a reduced level of FOXP2 expression is present .

17. A. kit for identifying the presence of F0XP2 in a sample comprising an antibody or antibody fragment capable of binding to F0XP2- and. means for contacting said antibody or fragment with said sample.

18. A kit according to claim 17 wherein said antibody is a monoclonal antibody produced by the hybridoma 2Fx2AD2 deposited under Accession No. 03082101.

19. A kit for detecting cancer in a patient comprising any of a labelled nucleic acid molecule capable of binding to FOXP2 nucleic acid; ^'or an antibody or fragment thereof capable of binding to F0XP2 protein to determine the level of expression of F0XP2 in a sample, and means for contacting the labelled nucleic acid molecule or antibody with said sample.

20. A kit according to claim 19 wherein said antibody is produced by the hybridoma 2Fx2AD2 deposited under Accession No. 03082101.

21. Use of a nucleic acid molecule encoding a FOXP2 protein, the FOXP2 protein or antibodies or inhibitors of expression of said FOXP2 protein in the manufacture of a medicament for modulating the effect of FOXP2 mediated steroid action on a cell. ,

22. A method of modulating FOXP2 mediated steroid action on a cell comprising administering to a patient in need thereof, an effective amount of a nucleic acid molecule encoding a FOXP2 protein, the F0XP2 protein or antibodies or inhibitors of expression of said FOXP2 protein.

23. A. method of treating a disease or condition mediated ^" by FOXP2 co-regulation of a nuclear receptor, which method comprises administering to an individual in need thereof, an inhibitor or enhancer of the interaction of F0XP2 and said nuclear receptor.

24. A method according to claim 23, comprising administering an inhibitor to said individual which comprises any of an antisense molecule of FOXP2 or a nucleic acid molecule capable of hybridising to nucleic acid encoding FOXP2 to inhibit expression thereof, an antibody capable of binding to F0XP2 in a cell, or any other molecule capable of antagonising the interaction of F0XP2 and said nuclear receptor.

25. A method according to claim 23 or 24 wherein said nuclear receptor is the estrogen receptor.

26. Use of an inhibitor or an enhancer of the interaction of FOXP2 and the estrogen receptor, in the manufacture of a medicament for treating a disorder characterised by the interaction of F0XP2 and the estrogen receptor in a patient .

27. Use according to claim 26, wherein said inhibitor comprises any of a nucleic acid molecule capable of hybridising to nucleic acid encoding said FOXP2 to inhibit expression thereof, an antibody or fragment thereof specific for F0XP2 protein and capable of preventing or reducing function thereof, or a composition capable of inhibiting F0XP2 expression in the tissues of a human cell or the interaction of F0XP2 with said estrogen receptor.

28. Use according to claim 26, wherein said enhancer comprises any of a nucleic acid molecule encoding said F0XP2 protein, _.the F0XP2 protein or a functional fragment thereof, or a composition capable of stimulating an increase of FOXP2 expression.

29. An assay to identify compounds capable of modulating the interaction of F0XP2 and the estrogen receptor in a cell, comprising contacting said compound with a reaction mixture or cell comprising said F0XP2 and said estrogen receptor, and monitoring for any alteration in said interaction compared to a mixture or cell that has not been contacted with said compound.

30. An assay according to claim 29, wherein said compound is capable of functioning as an agonist of the interaction between said F0XP2 and said estrogen receptor.

31. An assay according to claim 29, wherein said compound is capable of inhibiting the interaction between said F0XP2 and said estrogen receptor.