NOTCH mutations leading to increased receptor signaling

Abstract

The present invention is based upon the identification of regions within the NOTCH-1 receptor that, when mutated, lead to increased receptor signaling. The mutations are associated with uncontrolled cellular growth and this growth may be arrested using agents that interfere with NOTCH-1 activity, such as inhibitors of gamma-secretase. Assays for the NOTCH-1 mutations may be used diagnostically or as part of a treatment regimen for cancer patients.

Description

CROSS REFERENCE TO RELATED APPLICATIONS

The present application claims priority to, and the benefit of, U.S. provisional applications 60/598,546, filed on Aug. 4, 2004 and 60/672,053, filed on Apr. 18, 2005. The contents of these prior applications are hereby incorporated by reference in their entirety.

STATEMENT OF GOVERNMENT FUNDING

The United States Government has a paid-up license in this invention and the right in limited circumstances to require the patent owner to license others under reasonable terms as provided for by the terms of NIH Grant No. R01CA082308, R01CA92433 and P01CA68484 awarded by the Department of Health and Human Services.

FIELD OF THE INVENTION

The present invention is directed to human NOTCH receptors that have undergone mutations resulting in either increased responsiveness to biological stimuli or constitutive activation independent of the need for other stimuli. The mutations occur in specific, well-defined, regions of NOTCH receptors and are somatically acquired by cancer cells as part of the process of malignant transformation. The growth of these cancer cells can be slowed or stopped using agents that interfere with receptor signaling.

BACKGROUND OF THE INVENTION

NOTCH receptors 1-4 are transmembrane receptor proteins that signal through a pathway that relies on regulated proteolysis. Following ligand-binding, the receptor is sequentially: i) cleaved extracellularly by metalloproteases of the ADAM family (Brou, et al., Mol. Cell 5:207-216 (2000); Mumm, et al. Mol Cell 5:197-206 (2000)); ii) mono-ubiquitinated on a lysine residue lying just internal to the transmembrane domain (Gupta-Rossi, et al., J. Cell Biol. 166:73-83 (2004)); iii) endocytosed (Gupta-Rossi, et al., J. Cell Biol. 166:73-83 (2004)), and iv) proteolytically cleaved by a gamma-secretase enzyme (De Strooper, et al., Nature 398:518-522 (1999)). This final step in the activation process permits the intracellular portion of NOTCH receptors to translocate to the cell nucleus where it interacts with transcription factors to alter gene activity. NOTCH receptor signaling appears to play an important role in the differentiation and proliferation of cells and in controlling apoptosis, three processes that are important with respect to neoplastic transformation (see U.S. Pat. No. 6,703,221).

The NOTCH-1 gene was discovered through its involvement in a (7;9) chromosomal translocation found in fewer than 1% of T-cell acute lymphoblastic leukemias (T-ALLs) (Ellisen, et al., Cell 66:649-661 (1991)). NOTCH-1 is highly expressed in thymocytes (Ellisen, et al., Cell 66:649-661(1991)), where it induces common lymphoid progenitors to adopt a T cell fate (Radtke, et al., Immunity 10:547-558). Subsequently, it promotes the assembly of pre-T cell receptor complexes which play a critical role in driving a proliferative burst that accompanies maturation of CD4−/CD8−thymocytes to the CD4+/CD8+ developmental stage (Wolfer, et al., Immunity 16:869-879 (2002)).

In its resting state, mature NOTCH-1 is a heterodimeric receptor comprised of a ligand-binding extracellular subunit (NEC) and a non-covalently associated transmembrane subunit (NTM) (Rand, et al., Mol. Cell. Biol. 20:1825-1835; Logeat, et al., Proc. Natl. Acad. Sc.i USA 95:8108-8112 (1998)). NEC consists of a ligand-binding domain comprised of epidermal growth factor-like repeats, three iterated Lin12/NOTCH repeats, and a conserved 103 amino acid sequence (hereafter termed HD, for heterodimerization domain) that is sufficient for association with the extracellular portion of NTM.

Physiologic activation of NOTCH receptors occurs when a ligand of the Delta-Serrate-Lag2 (DSL) family binds to the NEC subunit and initiates a cascade of successive proteolytic cleavages in the NTM subunit. The final cleavage, which is catalyzed by γ-secretase, a multiprotein complex containing presenilin-1 or -2, nicastrin, APH-1, and PEN-2 (Francis, et al., Dev. Cell 3:85-97 (2002); Kimberly, et al., Proc. Natl. Acad. Sci. USA 100:6382-6387 (2003)) releases the intracellular part of NTM (called intracellular NOTCH, or ICN) from the membrane, permitting it to translocate to the cell nucleus. There, it associates with the DNA-binding factor CSL and co-activators of the Mastermind family to form a short-lived transcriptional activation complex (Wallberg, et al., Mol. Cell Biol. 22:7812-7819 (2002); Fryer, et al., Genes Dev. 16:1397-1411 (2002); Nam, et al., J. Biol. Chem. 278:21232-21239 (2003)) Degradation and turnover of the complex is apparently regulated by F-box factors of the SEL-10 family (Oberg, et al., J. Biol. Chem. 276:35847-35853 (2001)).

The (7;9) translocation creates a NOTCH-T cell receptor β fusion gene that encodes N-terminally-deleted, constitutively active NOTCH-1 polypeptides similar to the ICN (Ellisen, et al., Cell 66:649-661 (1991); Aster, et al., Cold Spring Harb. Symp. Quant. Biol. 59:125-136 (1994); Das, et al., J. Biol. Chem., epublished May 3, 2004)) and these truncated and constitutively active forms of NOTCH-1 induce T-ALL in mouse models (Aster, et al., Mol. Cell Biol. 20:7505-7515 (2000)). NOTCH-1 is also the site of frequent retroviral insertions that cooperate with the E2A-PBX1 and cMYC transgenes in multistep pathways leading to the development of murine T-ALL (Hoemann, et al., Mol. Cell. Biol. 20:3831-3842 (2000); Feldman, et al., Blood 96:1906-1913(2000)). Further, NOTCH inhibitors cause a G_O/G_I cell cycle arrest in cell lines derived from human and murine NOTCH-1-associated T-ALLs, indicating that NOTCH signaling is required for the sustained growth of these leukemias (Weng, et al., Mol. Cell Biol. 23:655-664 (2003)).

The (7;9) translocation has only been associated with a very small percentage of patients having T-ALL. The identification of additional mutations associated with NOTCH-1 would have important implications for the pathogenesis of this type of cancer. Although mutations in NOTCH-2, NOTCH-3, and NOTCH-4 have not been identified in human cancer, it is known that abnormal increases in function of these NOTCH receptors in other mammals can cause T-ALL (NOTCH-2 and -3, Bellavia, et al., Embo J. 19:3337-3348 (2000); Rohn, J. Virol. 70:8071-8080 (1996); Weng, et al., Mol. Cell. Biol. 23:655-664)) and breast cancer (NOTCH-4, Callahan and Rafat, J. Mammary Gland Biol Neoplasia 6:23-36 (2001)). Identification of novel mutations in human tumors should be useful diagnostically in helping to identify the presence of cancer and in identifying cancer cells that respond to inhibitors of NOTCH signaling, thereby making it possible to direct rational cancer treatment with NOTCH signaling pathway inhibitors.

SUMMARY OF THE INVENTION

The present invention is based upon the discovery that mutations in two specific regions of the NOTCH-1 receptor lead to increased receptor signaling and are associated with uncontrolled cellular growth. The first is a region within the heterodimerization domain that is characterized by the presence of many amino acids that are highly conserved between species. The amino acids (residues 1571-1622 in SEQ ID NO:1, see FIGS. 1 and 3 for sequence identification numbers) are encoded by exon 26 and are found within the NEC subunit (i.e. the extracellular portion) of the NOTCH-1 protein. In intact NOTCH-1, they are non-covalently bound to amino acids encoded by exon 27 (amino acids 1674-1730) which lie on the NTM subunit (i.e., the transmembrane portion of the receptor). It is believed that mutations, especially non-conservative mutations, in these “hotspots” of the HD domain, i.e., 1571-1622 of the NEC subunit and 1674-1730 of the NTM subunit, destabilize the NOTCH-1 heterodimer. This leads to dissociation of NEC and successive metalloprotease cleavage, ubiquination, endocytosis, and cleavage by gamma-secretase, permitting nuclear translocation of the intracellular portion of NOTCH-1 and activation of target genes.

A second region involved by mutations that lead to uncontrolled cellular growth is the C-terminal portion of NOTCH-1 (amino acid 2311 to 2556), which includes the PEST region (amino acid 2423 to amino acid 2556). These mutations appear to reduce the rate at which the receptor is degraded intracellularly after activation, i.e., after it has been released due to gamma-secretase cleavage. As a result, the transcription of NOTCH-1-responsive genes continues for an inappropriately long period of time, again, leading to increased signaling and, ultimately, uncontrolled growth.

It has been discovered that mutations in the HD domain and in the PEST domain are often found in leukemic cell lines and in biological samples obtained from patients with T cell acute lymphoblastic leukemia (T-ALL). Uncontrolled growth in these cells may be arrested using agents that interfere with NOTCH signaling, such as gamma-secretase inhibitors. Overall, the experiments performed have led to the conclusion that there is a group of cancer cells that depend upon abnormal NOTCH activity to maintain uncontrolled growth and that mutations in the HD domain and in the PEST domain of NOTCH-1 may be used diagnostically to identify cancer cells likely to respond to gamma-secretase inhibitors or other factors that diminish NOTCH activity.

In its first aspect, the invention is directed to a substantially purified mutant NOTCH-1 receptor which has the amino acid sequence of wild type NOTCH-1, shown herein as SEQ ID NO:1, except for mutations within the hotspots of the HD domain. These mutations lead to increased receptor signaling relative to unmutated NOTCH-1, due to increased susceptibility to cleavage by gamma-secretase. The specific hotspot regions of SEQ ID NO:1 in which the mutations occur are amino acids 1571-1622 and/or amino acids 1674-1730.

As used herein, the term “substantially purified” means that the receptor protein (or the polynucleotide encoding the protein) has been separated from the biological components associated with NOTCH-1 as found in nature. Typically, a substantially pure protein (or polynucleotide) will comprise at least 85% of a sample, with greater percentages being preferred. Many means are available for assessing the purity of protein or nucleic acid samples, including analysis by electrophoresis, chromatography and analytical centrifugation. The term “unmutated NOTCH-1” refers to NOTCH-1 receptors having the sequence of SEQ ID NO:1 and to polynucleotides that encode receptors having this sequence. The term also includes allelic variations in these sequences that are normally found in cells and which are not associated with abnormal NOTCH-1 signaling or with abnormal, uncontrolled cellular growth.

In order to determine whether a mutation leads to increased gamma-secretase signaling, the assays described herein in the Examples section may be used. The most preferred method involves the use of cells transfected with a reporter gene, e.g., a polynucleotide encoding luciferase, that is under the control of a regulatory element turned on in response to NOTCH-1 signaling. Activation of the reporter gene upon exposure of cells to gamma-secretase can be readily measured and reflects NOTCH activity (see, e.g., Hsieh, et al., Mol. Cell. Biol. 16:952-959 (1996)).

Comparison of HD amino acid sequences from a variety of species has resulted in the identification of certain specific amino acid positions that are highly conserved in hotspots. Mutations in these positions, particularly non-conservative mutations, are particularly likely to destabilize the interaction between the NEC and NTM subunits of NOTCH-1. The term “non-conservative mutation” refers to a mutation in which wild type amino acids are replaced with amino acids having different chemical characteristics. Examples of a non-conservative substitution include the substitution of a proline for other amino acids within a well-structured protein domain (i.e., the HD domain), the substitution of a basic amino acid for a non-basic amino acid, the substitution of an acidic amino acid for a non-acidic amino acid, the substitution of a polar amino acid for a non-polar amino acid, etc. The distinction between conservative and non-conservative mutations is well recognized by those of skill in the art. Conserved amino acids in the HD domain where mutations are associated with increased NOTCH-1 signaling are found at positions: 1571; 1573; 1575; 1576; 1578; 1579; 1583; 1586; 1588; 1590; 1593; 1594; 1595; 1598; 1600-1607; 1610-1618; 1674; 1675; 1677; 1679-1684; 1686; 1695; 1696; 1703; 1706; 1715; 1720; 1725; and 1728. All of these positions are identified with reference to SEQ ID NO:1.

The invention includes specific mutations that are associated with the uncontrolled growth of cancer cells. With respect to the HD domain, this occurs in cases where amino acids 1571-1618 are replaced with a sequence selected from the group consisting of SEQ ID NO:12-SEQ ID NO:22, SEQ ID NO:102-118 or SEQ ID NO:145. Other mutations associated with the growth of cancer cells include the replacement of amino acids 1571-1622 with a sequence selected from the group consisting of SEQ ID NO:23-SEQ ID NO:29 or SEQ ID NO:119-127. Similarly, the sequence from amino acid 1674 to amino acid 1730 may be replaced with a sequence selected from the group consisting of SEQ ID NO:70-SEQ ID NO:75 or SEQ ID NO:128-137.

The inventors have also found that mutations in SEQ ID NO:1 occurring in the C-terminal portion of NOTCH-1, i.e., from amino acid 2311 to amino acid 2556, lead to increased NOTCH-1 signaling in response to exposure of cells to gamma-secretase. All of these mutations cause the appearance of a premature stop codon or a shift in the translational reading frame, and thus lead to the loss of some or all of the sequences corresponding to the PEST region (amino acid 2423 to 2556). It is believed that these mutations lead to a reduced rate at which receptor is degraded. Specific mutations include the replacement of the portion of SEQ ID NO:1 corresponding to amino acids 2311-2556 with a sequence selected from the group consisting of SEQ ID NO:30-SEQ ID NO:59 or SEQ ID NO:139-144.

Mutations in the HD domain and the PEST domain may occur independently of one another or may be present in a single NOTCH-1 receptor. For example, a substantially purified mutant NOTCH-1 receptor, may comprise or consist of the amino acid sequence of SEQ ID NO:1, but wherein:

• • a) the receptor has at least one of the following mutations:
    • i) the portion of said amino acid sequence of SEQ ID NO:1 from amino acid 1571 to amino acid 1618 is replaced with a sequence selected from the group consisting of SEQ ID NO:12-SEQ ID NO:22, SEQ ID NO:102-118, or SEQ ID NO:145;
    • ii) the portion of the amino acid sequence of SEQ ID NO:1 from amino acid 1571 to amino acid 1622 is replaced with a sequence selected from the group consisting of SEQ ID NO:23-SEQ ID NO:29, SEQ ID NO:119-127;
    • iii) the portion of the amino acid sequence of SEQ ID NO:1from amino acid 1674 to amino acid 1730 is replaced with a sequence selected from the group consisting of SEQ ID NO:70-SEQ ID NO:75 or SEQ ID NO:128-137; and
  • b) and, in addition, the receptor is further mutated such that the portion of the amino acid sequence of SEQ ID NO:1 from amino acid 2311 to amino acid 2556 is replaced with a sequence selected from the group consisting of SEQ ID NO:30-SEQ ID NO:59 or SEQ ID NO:138-144.

Alternatively, a substantially purified mutant NOTCH-1 receptor, may comprise or consist of the amino acid sequence of SEQ ID NO:1, but wherein:

• • a) the amino acid sequence of SEQ ID NO:1 is mutated in one or more amino acid positions selected from the group consisting of: 1571; 1573; 1575; 1576; 1578; 1579; 1583; 1586; 1588; 1590; 1593; 1594; 1595; 1598; 1600-1607; 1610-1618; 1674; 1675; 1677; 1679-1684; 1686; 1695; 1696; 1703; 1706; 1715; 1720; 1725; and 1728; and
  • b) the amino acid sequence of SEQ ID NO:1 is further mutated such that the portion of said amino acid sequence from amino acid 2311 to amino acid 2556 is replaced with a sequence selected from the group consisting of SEQ ID NO:30-SEQ ID NO:59.
    When mutations are present in both the HD and PEST regions, this leads to receptors in which signaling in response to gamma-secretase is especially enhanced.

In another aspect, the invention includes polynucleotides that encode any of the mutant NOTCH-1 receptors described above. Nucleic acid sequences coding for the mutant receptors may be incorporated into a vector in which they are operably linked to a promoter. The term “operably linked” refers to genetic elements that are joined in a manner that enables them to carry out their normal functions. For example, a gene is operably linked to a promoter when its transcription is under the control of the promoter and the protein normally encoded by the gene is ultimately produced. The invention includes the vectors coding for the mutant receptors as well as host cells transformed with the vectors.

In another aspect, the invention is directed to methods of diagnostically testing a biological sample for the presence of cancer cells by assaying the sample for any of the mutant NOTCH-1 receptors described above. The term biological sample, as used herein includes all of the different types of samples commonly used in the art of medicine in making diagnoses. Thus, the term encompasses, biopsy samples of tissue or fluid, blood, plasma, serum, lymph, urine, etc. Depending on the type of cancer being considered, certain types of samples will be the ones predominantly used. For example, for leukemia, samples of blood, plasma, or serum would tend to be favored. The choice of appropriate samples for testing is routine in clinical medicine

Also included are the same methods applied to the NOTCH-2, NOTCH-3 and NOTCH-4 receptors, which contain regions with high levels of identity to the regions with mutations described for NOTCH-1. The NOTCH-1, -2, -3 and -4 receptors are very highly related and likely perform very similar or identical functions in a variety of cell types, differing mainly in that different members of this receptor family are expressed and used preferentially by different tissues and cell types. Cancers arising in different tissues may thus acquire very similar types of mutations that affect any of the NOTCH-1, -2, -3 or -4 receptors with identical consequences in terms of promoting tumor cell growth, depending on which of the receptor subtypes are expressed by the tissues from which these cancers arise. Thus, in this application, unless otherwise indicated, the term “NOTCH” receptor will be used to refer to diagnostic applications applied collectively to each of the NOTCH-1, -2, -3, and -4 receptors.

The presence of mutant receptor may be detected either by assays directed at the protein or by assays directed at the gene encoding the protein. With respect to assays directed at genes, the preferred method of testing is to amplify either the full-length polynucleotide encoding a mutant NOTCH receptor, or a portion of this polynucleotide, using the polymerase chain reaction (PCR). Oligonucleotide primers that can be used for amplification are described in the Examples section, but one of skill in the art can readily identify other appropriate primers based upon sequence information included herewith and known in the art.

Alternatively, gene mutations may be detected by performing hybridizations using probes under conditions that permit hybrids to form with mutant, but not wild type, nucleic acid. Both the assays and the oligonucleotides used in the assays are encompassed by the invention. The oligonucleotides should be at least 15 bases in length and hybridization assays should be carried out under stringent conditions. As used herein, the term “stringent conditions” refers to hybridizations carried out at approximately 60-70° C. and at a low salt concentration (equivalent to, for example, 0.02-0.15 M NaCl).

Assays designed to detect the mutant protein per se may be carried out using antibodies that bind to mutant NOTCH receptors with at least a hundredfold greater affinity than to unmutated NOTCH receptors. These antibodies may be either monoclonal or polyclonal in nature and can be made by injecting an antibody-producing animal with peptides derived from regions of NOTCH receptor containing mutations. Antibodies made in this manner and which preferentially bind to mutant receptor are also a part of the invention.

The invention is also directed to methods for slowing or stopping the growth of cells with the mutant receptors described above by exposing the cells to an effective amount of an agent, such as a gamma-secretase inhibitor, that interferes with NOTCH receptor signaling. Examples of appropriate inhibitors include III-31-C, N-[N-(3,5-5 difluorophenacetyl)-L-alanyl]S-phenylglycine t-butyl ester) (DAPT), compound E, D-helical peptide 294, isocoumarins, BOC-Lys(Cbz)Ile-Leu-epoxide, and (Z-LL)₂-ketone (see Kornilova, et al., J. Biol. Chem. 278:16479-16473 (2003)). The term “effective amount” indicates a sufficient concentration of inhibitor or other agent to significantly reduce NOTCH-1 activity, for example, a sufficient amount to reduce activity by at least 20%. The presence of an effective amount of an agent that inhibits Notch receptor activity should also be reflected clinically. For example, the rate growth of cancer cells should be slowed to a significant extent (e.g., by at least 20%) relative to the rate of growth occurring prior to treatment.

Although gamma-secretase inhibitors have been found to be effective as therapeutics, other agents that inhibit NOTCH activity may also be used. The mutations described herein should cause several effects, including: i) increased metalloprotease cleavage; ii) increased mono-ubiquitination (an enzymatic step carried out by an E3 ligase); iii) increased endocytosis, iv) increased gamma-secretase cleavage (as already discussed); v) increased nuclear translocation; and vi) increased formation of a nuclear NOTCH transcriptional activation complex. Drugs that act at any of these steps should stop the growth of cancer cells. One agent, an inhibitor of the nuclear complex, a dominant negative form of a protein called mastermind-like-1, has been previously described and inhibits all four NOTCH receptor subtypes (Weng, et al., Mol Cell Biol 23:655-664 (2003)).

Also encompassed by the invention are methods of assaying a test compound for its ability to inhibit abnormal cellular growth induced by the presence of a mutated NOTCH receptor. This involves incubating the test compound with cells that express any of the mutant NOTCH-1 receptors described above. Such cells may either be isolated from a biological sample or created by transfecting a cell with an expression vector encoding the mutant receptor. The incubation is carried out in the presence of gamma-secretase and the amount of NOTCH receptor activity that occurs is determined. Any method for measuring receptor activity may be used, but the luciferase reporter gene assay described in Hsieh, et al. (Mol. Cell. Biol. 16:952-959 (1996)) is preferred. The results obtained are then compared with those from assays carried out under essentially the same conditions but in the absence of the test compound. A reduction in activity caused by the test compound suggests that it is a NOTCH receptor inhibitor and may be useful in limiting the growth of cells expressing the receptor. This can be confirmed directly by comparing the growth of such cells in the presence and the absence of increasing concentrations of the test compound.

In addition to being of value in assays designed to evaluate the ability of a test compound to inhibit NOTCH receptor activity, polynucleotides encoding mutant NOTCH receptors may serve as a source for probes that can be used in the hybridization assays discussed above and as controls in diagnostic assays. The mutant receptor protein can be used to generate antibodies specific for mutated forms of NOTCH and can serve as a control in immunoassays designed to either quantitate the amount of NOTCH receptor present in a sample or to distinguish between normal and mutated forms of receptor. Both the mutant proteins and the polynucleotides encoding the mutant proteins should also be of interest to researchers studying biological processes such as differentiation and apoptosis.

As discussed above the invention includes isolated nucleic acid molecules encoding mutant Notch-1 receptors. Such nucleic acid molecules may comprise a nucleotide sequence selected from the group consisting of:

• • a) a sequence of nucleotides that encodes a mutant Notch-1 receptor comprising any of the mutant amino acid sequences described herein and wherein, except for the codons encoding the mutant amino acid(s), the nucleotide sequence is that of SEQ ID NO:76;
  • b) a sequence of nucleotides that encodes a mutant Notch-1 receptor having any of the amino acid sequences described herein;
  • c) a sequence of nucleotides degenerate with the mutant Notch-1 encoding sequence of (a) or (b).

In another aspect, the invention includes a method of determining the oncogenic potential of a biological sample by assaying the sample for the presence any of one or more mutant Notch-1 receptors as described herein. The presence of mutant receptors indicates increased oncogenic potential relative to a sample in which such mutant NOTCH-1 receptors are not present.

In addition, the invention includes a method for following the progress of a therapeutic regime designed to alleviate a condition characterized by abnormal expression of a mutant NOTCH-1 receptor comprising:

• • a) assaying a sample from a subject to determine the level of mutant NOTCH-1 receptor (or nucleic acid encoding a mutant Notch-1 receptor) at a first time point;
  • b) assaying the level of the same receptor (or nucleic acid) at a second time point; and
  • c) comparing the level at the second time point to the level determined at the first time point. Decreasing levels of mutant protein or nucleic acid is an indication that a therapy is working. Increasing levels are an indication that a therapy is not working.

The invention also encompasses a method for determining regression, progression or onset of a pathological disorder characterized by dysfunctional NOTCH-1 signal transduction. The method involves incubating a sample obtained from a patient having or suspected of having the disorder with a complementary nucleic acid hybridization probe having a sequence of nucleotides that is substantially homologous to that encoding one or more mutant NOTCH-1 receptors as described herein. The probe should generally be designed to interact with nucleic acid encoding the mutant form of the receptor and not the normal form as shown in SEQ ID NO:1. Binding between the probe and any complimentary MRNA that may be present in the sample is used to indicate the regression, progression or onset of the pathological disorder in said patient. Alternatively, assays may be performed using antibody (preferably monoclonal antibody) specific for a mutant form of the receptor. As the level of mutant receptor or nucleic acid encoding mutant receptor increases the likelihood that the disorder is present or progressing also increases. Reduced or absent levels of mutant receptor or nucleic acid is an indication of disorder regression or absence.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1-1 to 1-7: Aligned Full length amino acid sequences of human NOTCH receptors. The sequences shown are: NOTCH-1 (SEQ ID NO:1); NOTCH-2 (SEQ ID NO:2); NOTCH-3 (SEQ ID NO:3); and NOTCH-4 (SEQ ID NO:4).

FIG. 2A-2C. Identification of T-ALL cell lines that require NOTCH signals for growth. (2A) Effect of the γ-secretase inhibitor compound E (GSI, 1 μM) or vehicle alone (mock) on the cell cycle distribution of sensitive cell lines. Cells were treated for 4-8 days, fixed in cold 70% ethanol, stained with propidium iodide, and analyzed by flow cytometry. Live cells were gated by forward/side scatter criteria. (2B) Rescue of GSI-induced cell cycle arrest by MSCV-GFP-ICN1 retrovirus. (2C) Induction of cell cycle arrest by dominant negative MSCV-GFP-Mastermind-like-1 retrovirus. In (2B) and (2C), control cultures were transduced with control empty MSCV-GFP retrovirus or the indicated test virus. Two days post-transduction, cultures were split in two and treated with either compound E (GSI) or DMSO vehicle (mock) for 7-10 days. Cultures were then harvested and stained with the DNA dye DRAQ5, which permits flow cytometric measurement of DNA content in unsorted GFP− and GFP+ subpopulations. A minimum of 15,000 events was collected to create each DNA content histogram.

FIG. 3A-3G2. HD and PEST domain mutations in γ-secretase inhibitor-sensitive T-ALL cell lines and in primary human T-ALL samples. (3A) sequence conservation of the exon 26 HD domain mutational “hotspot” region among vertebrate NOTCH-1 receptors. Positions of recurrent mutations are in bold and italics. (3B) missense mutations and deletions identified in cell lines (underlined) and primary T-ALLs. One mutation SEQ ID NO:145, was associated with acute myelogenous leukemia. Mutation sites are in bold and italics. (3C) in-frame insertions in the HD domain identified in primary T-ALLs. (3D) PEST domain mutations. The amino acid sequences of mutated NOTCH-1 alleles in cell lines (underlined) and primary T-ALLs are shown. Stop codons are indicated by asterisks. (3E) Schematic representation of PEST domain mutations. The circles (white, primary T-ALLs; black, cell lines) correspond to the positions at which the NOTCH-1 polypeptide sequence terminates. Key: NEC, NOTCH-1 extracellular domain; LNR, Lin/NOTCH repeats; HD, heterodimerization domain; NTM, NOTCH transmem-brane subunit; TM, transmembrane domain; ANK, ankyrin repeat domain; TAD, transcriptional activation domain. In (A-D), amino acid residues correspond to those of human pro-NOTCH-1. (3F) Western blot analysis of lysates from NOTCH-dependent T-ALL cell lines. Whole cell extracts were analyzed with a NOTCH-1 antibody raised against the intracellular transcriptional activation domain (Aster, et al., Mol. Cell. Biol. 20:7505-15). Positions of pro-NOTCH-1 and the normal NTM subunit are denoted. The lower panel is a longer exposure of the same blot that accentuates the presence of additional NOTCH-1 polypeptides of smaller size than the normal NTM polypeptide. Each lane was loaded with 25 μg of total protein. (3G) Sequence conservation and mutations in exon 27, encoding the HD domain of the NOTCH-1 TMT subunit. This subunit of NOTCH is noncovalently bound to exon 26, the HD domain of the NEC subunit. Mutation sites are in bold and italics. For TALL 04.36 the i in the sequence indicates the site of insertion of the 14 amino acids shown immediately below.

FIG. 4A-4B. HD and PEST domain mutations activate NOTCH-1 signaling synergistically. (4A) Reporter gene assays. U2OS cells were transiently co-transfected in 24-well format with the indicated pcDNA3 plasmids, a NOTCH-responsive luciferase reporter gene, and an internal Renilla luciferase internal control plasmid, as described previously (Aster, et al., Mol. Cell. Biol. 20:7505-15). Twenty-five ng of pcDNA3 plasmid was used per well, except for experiments with pcDNA3-ICN1, in which 5 ng of plasmid per well were used. Normalized luciferase activities in whole cell lysates were determined in triplicate and expressed relative to the activity in lysates prepared from cells transfected with the empty vector control. (4B) Model for synergistic NOTCH-1 activation by dual mutations involving both the HD and PEST domains.

FIG. 5-1 to 5-11: Full length human NOTCH-1 gene sequence (SEQ ID NO:76) and encoded amino acid sequence (SEQ ID NO:1).

FIG. 6-1 to 6-3: Compilation of HD and PEST domain mutations.

DETAILED DESCRIPTION INVENTION

The present invention is concerned with the identification of mutations in the human NOTCH-1 receptor that lead to an increase in receptor signaling. The mutations are associated with cancer cells that respond positively to treatment with agents that interfere with NOTCH-1 activation, such as gamma-secretase inhibitors and are located within two specific regions of the NOTCH-1 protein, the HD domain and the PEST domain. It is expected that analogous mutations in corresponding regions of NOTCH-2, NOTCH-3 and NOTCH-4 receptor subtypes should also lead to increased signaling and abnormal growth.

Assays designed to detect the presence of NOTCH-1 mutations, e.g., assays based upon PCR amplification, may be used diagnostically to identify tissue samples with cells in which abnormal NOTCH-1 receptor signaling is occurring and which are, as a result, more likely to exhibit uncontrolled growth. Cancers characterized by this particular abnormality should be amenable to treatment with agents that reduce NOTCH-1 signaling, such as gamma-secretase inhibitors. Any of the inhibitors described herein or which have been described in the art (for example, in connection with potential use in the treatment of Alzheimer's disease) may be used in treatment regimens. The diagnostic assays and therapeutic methods may be applied to other types of cancer besides T-ALL. Such cancers include but are not limited to: carcinomas of the breast, prostate, lung, pancreas, ovary, colon, bladder, kidney and GI tract, B cell leukemias, B and T cell lymphomas, Hodgkin's lymphoma, acute myelogenous leukemias, melanomas, neuroblastomas, medulloblastomas, glioblastoma multiforme and other glial cancers of the brain, and mesotheliomas.

I. Production of Polynucleotides Encoding Mutant NOTCH-1 Receptors

The full length amino acid sequence of human NOTCH-1 is shown in FIG. 1 and has been given the sequence identification number SEQ ID NO:1. This is the normnal, unmutated form of receptor. Mutations in the HD domain and PEST domain that have been associated with samples from cancer patients or cancer cell lines are shown in FIG. 3. It should be noted that certain positions in the hotspots of the HD domain are highly conserved among different species (see FIG. 3A) and are especially sensitive to mutations that lead to increased gamma-secretase-generated activity.

Many methods are available for producing polynucleotides that encode NOTCH-1 receptors having the mutated sequences shown herein (see, e.g., Sambrook, et al., Molecular Cloning: A Laboratory Manual, 2^nd ed., Cold Spring Harbor Press (1989)). For example, a cDNA library prepared by reverse transcribing MRNA isolated from cells that express the mutant NOTCH gene may be screened using probes synthesized based upon the sequences shown in FIG. 1. Alternatively, the desired sequences may be obtained by PCR amplifying reverse transcribed RNA. Primers for PCR may be constructed using the sequences shown herein and specific oligonucleotides that have been found to be suitable for amplification are described in the Examples section. Confirmation that the correct sequence has been amplified and identification of the presence of specific mutations may be obtained by sequencing amplification products.

Alternatively, nucleic acids may be produced using standard chemical synthetic methods well known in the art. Once obtained, the polynucleotides can be used in recombinant methods for making receptor protein, expressed in cells for the purpose of screening for potential therapeutic agents, or used as controls in diagnostic assays.

II. Production of Mutant Receptor Protein

NOTCH-1 receptor protein can be synthesized using chemical methods or produced using the polynucleotides described above. In the latter case, the DNA sequence encoding the mutant receptor may be placed in a vector containing transcriptional and translational signals recognizable by an appropriate host. The cloned receptor sequences, in double-stranded form, are inserted into the expression vector in an operable linkage. Appropriate vectors and techniques for cloning and expressing receptor are well known in the art of molecular biology.

Expression vectors may be introduced into host cells, preferably mammalian host cells, by methods such as calcium phosphate precipitation, microinjection, electroporation or viral transfer. Cells expressing the recombinant mutant receptor can be selected using standard, well established methods. One simple method for confirming the presence of the mutant receptor nucleic acid in cells is to perform PCR amplification using primers known to flank the receptor-encoding nucleotide sequence. The presence of receptor may also be confirmed using functional assays of NOTCH-1 such as the luciferase assay described in the Examples section and in the art (see, Hsieh, et al., Mol. Cell. Biol. 16:952-959 (1996)).

Once cells producing recombinant mutant NOTCH-1 receptor have been identified, they may be used in assays designed to identify agents capable of reducing signaling, for example, by incubation with gamma-secretase, or in assays that measure the ability of NOTCH-1 inhibitors to arrest uncontrolled cellular growth. The NOTCH-1 receptor itself may be used in diagnostic assays or to generate antibodies specific for mutant receptor that can be used in such assays.

III. Antibodies to Mutant NOTCH Receptor

The present invention encompasses antibodies that bind specifically to mutant forms of NOTCH receptors and processes for producing such antibodies. Antibodies that “bind specifically to mutant NOTCH receptors” are defined as those that have at least a hundredfold greater affinity for the mutant form of the receptor than for the unmutated form. The process for producing such antibodies may involve either injecting the full length mutant receptor protein into an appropriate animal or, preferably, injecting short peptides that include regions where mutations occur. The peptides should be at least five amino acids in length and may be injected either individually or in combinations.

Methods for making and selecting antibodies are well known to those of skill in the art as evidenced by standard reference works such as Harlow, et al., Antibodies, A Laboratory Manual, Cold Spring Harbor Laboratory, N.Y. (1988); Klein, Immunology: The Science of Self-Nonself Discrimination (1982); Kennett, et al., Monoclonal Antibodies and Hybridomas: A New Dimension in Biological Analyses (1980); and Campbell, “Monoclonal Antibody Technology,” in: Laboratory Techniques in Biochemistry and Molecular Biology (1984).

The term “antibody,” as used herein, is meant to include intact molecules as well as fragments that retain their ability to bind antigen such as Fab and F(ab′)₂ fragments. The term “antibody” is also defined as referring to both monoclonal antibodies and polyclonal antibodies. Polyclonal antibodies are derived from the sera of animals immunized with an appropriate antigen. Monoclonal antibodies can be prepared using hybridoma technology as taught by references such as: Hammerling, et al., in Monoclonal Antibodies and T-Cell Hybridomas, Elsevier, N.Y., pp. 563-681 (1981). In general, this technology involves immunizing an immunocompetent animal, typically a mouse, with either intact protein or a fragment derived therefrom. Splenocytes are then extracted from the immunized animal and are fused with suitable myeloma cells, such as SP₂O cells. Thereafter, the resulting hybridoma cells are selectively maintained in HAT medium and then cloned by limiting dilution (Wands, et al., Gastroenterology 80:225-232 (1981)). The cells obtained through such selection may then be assayed to identify clones that secrete antibodies binding preferentially to mutant forms of NOTCH-1 or mutant forms of other NOTCH receptors.

The antibodies, or fragments of antibodies, of the present invention may be used to detect the presence of mutant NOTCH receptor proteins using a variety of immunoassays. For example, the antibodies may be used in radioimmunoassays or immunometric assays, also known as “two site” or “sandwich” assays (see Chard, “An Introduction to Radioimmune Assay and Related Techniques” in: Laboratory Techniques in Biochemistry and Molecular Biology, North Holland Publishing Co., N.Y. (1978)). In a typical immunometric assay, a quantity of unlabeled antibody is bound to a solid support that is insoluble in the fluid being tested, e.g., blood, lymph, cellular extracts, etc. After the initial binding of antigen to immobilized antibody, a quantity of detectably labeled second antibody (which may or may not be the same as the first) is added to permit detection and/or quantitation of bound antigen (see, e.g., Kirkham, et al., Radioimmune Assay Methods, pp. 199-206 (1970)). Many variations of these types of assays are known in the art and may be employed for the detection of mutant NOTCH receptors.

Antibodies to mutant NOTCH receptors may also be used in the purification of either the intact receptor or fragments of these receptors (see generally, Dean, et al., Affinity Chromatograph A Practical Approach, IRLP Press (1986)). Typically, antibody is immobilized on a chromatographic matrix such as Sepharose 4B. The matrix is then packed into a column and a preparation containing mutant receptor is passed through the column under conditions that promote binding, e.g., under conditions of low salt. The column is then washed and bound receptor is eluted using a buffer that promotes dissociation from antibody (e.g., a buffer having an altered pH or salt concentration). The eluted receptor protein may be transferred into a buffer of choice, e.g., by dialysis and either stored or used directly. Purified receptor may be used in the immunoassays described above or for the generation of antibodies for use in assays.

IV. Assay Methods

Diagnostic assays for determining whether a biological sample contains cancer cells in which the mutations described herein contribute to uncontrolled growth may be performed either using the immunoassays described above to analyze protein or, preferably, by assaying a sample for the presence of nucleic acid sequences encoding a mutant form of receptor. The preferred method for nucleic acid analysis is by PCR amplifying polynucleic acid that has been extracted from cells using methods well known in the art. Amplification 5 may be performed using the oligonucleotide primers described in the Examples section below or using other oligonucleotides based upon the sequence of NOTCH-1 shown as SEQ ID NO:1. The amplified NOTCH receptor product may then be sequenced to determine whether it carries a mutation leading to increased signaling.

Alternatively, hybridizations may be performed under stringent conditions using probes that bind only to mutant sequences. The probes should be a minimum of 15 nucleotides in length and must cover the portion of the polynucleotide sequence that encodes the mutated region of NOTCH-1. Stringent conditions would typically involve hybridization at a temperature of 60-70° C. with a low salt concentration (e.g., 0.02 M to 0.15 M NaCl). Procedures for identifying and labeling probes as well as for carrying out hybridizations and analyzing results are well known in the art of molecular biology.

Host cells transformed with vectors encoding recombinant mutant NOTCH receptors may also be used in assays designed to identify test compounds that reduce inappropriate signaling by the receptor. The assays will typically involve incubating cells in the presence of gamma-secretase and test compound. NOTCH receptor signaling can then be quantitated using assays such as those described herein or using other assays that are known in the art (see also, Hsieh, et al., Mol. Cell. Biol. 16:952-959 (1996)). As a simple alternative, cells known to be expressing mutated forms of NOTCH-1 that lead to uncontrolled growth may be cultured in the presence and absence of gamma-secretase and test compound to determine the extent to which the compound is capable of arresting growth.

V. Treatment Methods

Gamma-secretase inhibitors have been developed by pharmaceutical companies interested in their potential use as a treatment for Alzheimer's disease. These same inhibitors can be used to treat cancer cells in cases in which uncontrolled growth is associated with the NOTCH receptor mutations described herein.

The therapeutic regimen involves first testing a biological sample containing cancer cells removed from a patient to determine whether a mutated form of a NOTCH receptor is present. Patients whose samples evidence the presence of mutations would then be treated using a gamma-secretase inhibitor or another agent that interferes with NOTCH receptor activity. The dosage administered will depend upon the particular condition being treated, the route of administration and clinical considerations that are well known in the art. Dosages can be gradually increased until a beneficial effect, e.g., a slowing of tumor growth, is detected. Agents may be provided in either single or multiple dosage regimens and may be given either alone or in conjunction with other therapeutic agents.

Treatment of mutant NOTCH receptor-associated cancers is compatible with any route of administration and dosage form. Depending upon the particular condition being treated, certain dosage forms will tend to be more convenient or effective than others. For example, topical administration may be preferred in treating skin cancers, whereas parenteral administration might be preferred for leukemias. Apart from parenteral and topical preparations, agents may be administered orally, perorally, internally, intranasally, rectally, vaginally, lingually and transdermally. Specific dosage forms include tablets, pills, capsules, powders, aerosols, suppositories, skin patches, parenterals and oral liquids including suspensions, solutions and emulsions. Sustained release dosage forms may also be used. All dosage forms may be prepared using methods that are standard in the art (see, e.g., Remington's Pharmaceutical Sciences, 16^th ed., Easton, Pa. (1980)).

EXAMPLES

The present example provides results suggesting that approximately 65% of human T cell acute lymphoblastic leukemias (T-ALLs), including tumors from all major molecular oncogenic subtypes and those occurring in children and adults, have NOTCH-1 mutations.

Most of these mutations occur in regions encoding an extracellular heterodimerization domain and as truncations removing a C-terminal PEST destruction box. The mutations cause increased NOTCH-1 signaling, and T-ALL cell lines bearing such mutations are growth arrested by NOTCH-1 pathway inhibitors.

Introduction

T cell acute lymphoblastic leukemia (T-ALL) is an aggressive neoplasm that preferentially affects children and adolescents. It is commonly associated with acquired chromosomal translocations and other genetic or epigenetic abnormalities, which lead to aberrant expression of a select group of transcription factors, including the helix-loop-helix proteins TAL1, TAL2, LYL1, and BHLHB1, the orphan homeobox proteins HOX 11 and HOX11L2, and the fusion proteins MLL-ENL and CALM-AF10. Mis-expression of these proteins, alone or in combination, perturbs normal pathways of T cell development, growth, and survival, and defines subgroups of T-ALL with distinct gene expression profiles. Nevertheless, enforced expression of any one of these proteins alone in murine models is not sufficient to cause T-ALL, emphasizing the multistep molecular pathogenesis of this disease.

Materials and Methods

Cell Lines

All T-ALL cell lines were cultured in RPMI (Invitrogen) supplemented with 10% fetal bovine serum (Hyclone), 1 mM sodium pyruvate, 2 mM L-glutamine, 100 U/ml penicillin G, and 100 μ/ml streptomycin at 37° C. under 5% CO₂. U2OS cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (Hyclone), 2 mM L-glutamine, 100 U/ml penicillin G, and 100 μ/ml streptomycin at 37° C. under 5% CO₂.

Gamma-Secretase Inhibitor Treatment of Cell Lines

Gamma-secretase activity was inhibited by addition of compound E to growing cells to a final concentration of 1 μ/M. Mock-treated cultures were exposed to vehicle only, DMSO at a final concentration of 0.01%. The following human T-ALL cell lines were tested for sensitivity to compound E: T-ALL-1, ALL-SIL, KOPT-K1, HPB-ALL, DND-41, CCRF-HSB2, RPMI8402, THP-6, BE-13, JM-Jurkat, Jurkat-E6, REX, CEM, LOUCY, DU-528, KE-37, SKW-3, MKB-1, MOLT-3, MOLT-4, MOLT-13, MOLT-15, MOLT-16, SUPT-7, SUPT-11, SUPT-13, P12-Ichikawa, KARPAS-45, DSMZ-PEER and PF-382. The murine cell line T-6E, which was derived from a T-ALL induced with a membrane-tethered form of truncated NOTCH-1 that requires gamma-secretase cleavage for activation, was used as a positive control.

Retroviruses

Construction of retroviral expression plasmids (MSCV-GFP-ICN1 and MSCV-dominant negative Mastermind-like-1-GFP), production of high-titer pseudotyped, amphotropic, replication-defective retroviruses and spinoculation procedure were performed as described previously (Weng, et al., Mol. Cell Biol. 23:655-664).

Cell Cycle Analysis

Cells were stained with either propidium iodide after overnight fixation in cold 70% ethanol, or DRAQ5 (Biostatus, Leicester, UK) in complete medium (as suggested by the manufacturer) in instances where GFP fluorescence and DNA content measurement were measured simultaneously. DNA content was ascertained by flow cytometry and cell cycle factors determined using multicycle software (Phoenix Flow Systems, San Diego).

NOTCH-1 Expression Plasmids

NOTCH-1 expression plasmids in pcDNA3 have been described previously. To create a generic PEST deletion, the insertion mutation in the PEST domain of the T-ALL cell line ALL-SIL was amplified by PCR, identified by sequencing, and then sub-cloned into pcDNA3-NOTCH-1. Site-directed mutagenesis of the HD domain was conducted using the QuikChange kit (Stratagene). Mutagenic PCR primers were designed using the Stratagene web tool.

Receptor Gene Assays

NOTCH-1 expression plasmids (pcDNA3) were transiently transfected with U2OS cells with (i) an artificial luciferase reporter gene under the control of a promoter containing CSL binding sites and (ii) an internal Renilla luciferase control gene, as described previously (Aster, et al., Mol. Cell Biol. 20:7505-7515 (2000)), using Lipofectarnine Plus (Invitrogen). Total DNA was kept constant by adding empty vector as needed. All transfections were carried out in triplicate, and each experiment was repeated at least three times. Cell lysates were harvested 44-48 hours post-transfection and luciferase assays were carried out using the Dual Luciferase Assay System (Promega) on a Turner Systems luminometer, as per the manufacturer's suggestions.

PCR Amplification of NOTCH-1 cDNAs

RNA isolated from T-ALL cell lines (1)μ/g was reverse transcribed into cDNA using oligo-dT priming and Superscript reverse transcriptase (Invitrogen) in 20 μl reactions per the manufacturer's suggested conditions. PCR amplifications were performed using 1 μl of cDNA as template, 20 pMol of 5′-TTTGAATTCGGGCTGGACTGTGCGGAGCA TGTACCCGA (SEQ ID NO:78) and 5′-TTTGGATCCTCCGGAATGCGGGCGATCTG GGACTGCA (SEQ ID NO:79) NOTCH-1-specific primers, 0.25 μl of Ex-Taq Takara, Japan), 1× reaction buffer, 1×dNTPs and 7.5% DMSO in a 50 μl reaction volume. Cycle parameters were: 94° C. for three minutes; 30 cycles of 94° C. for one minute; 60° C. for thirty seconds; 72° C. for two minutes; and 72° C. for eight minutes. PCR products were purified on Qiagen columns, digested with Bam HI and Eco RI, ligated into pSP72 and used to transform DH5 alpha cells. Miniprep DNA from individual colonies was sequenced using T7 and SP6 primers.

Patient Material

Samples of cryopreserved lymphoblasts from 96 children and young adults with T-ALL treated at St. Jude Children's Research Hospital and Dana Farber Cancer Institute were obtained with informed consent at the time of diagnosis. Genomic DNA from each sample was extracted with a commercial kit (GENTRA) following the manufacturer's instructions. RNA was prepared from cryopreserved lymphoblasts with RNAaqueous reagents (Ambion) according to the manufacturer's instructions.

Quantitative RT-PCR

Quantitative RT-PCR analysis of HOX11, HOX11L2, TAL1, LYL1, TAL2 and BHLHB1 was performed as previously described (Ferrando, et al., Cancer Cell 1:75-87; Ferrando, et al., Lancet 363:535-536) using an ABI PRISM 7700 Sequence Detection System instrument (Perkin Elmer Applied Biosystems). RT-PCR-based detection of MLL-ENL and CALM-AF10 fusion transcripts was also performed as previously described (Ferrando, et al., Lancet 363:535-536; Asnafi, et al., Blood 102:1000-1006).

Genomic DNA Preparation

Genomic DNA was prepared from primary T-ALLs by extraction with a commercial kit (GENTRA) following the manufacturer's instructions. DNA was prepared from cell lines by SDS lysis/proteinase K digestion, followed by phenol:chloroform extraction and ethanol precipitation. DNAs were resuspended in TE buffer, spectrophotometrically quantified and stored at −20° C.

Mutation Detection

Mutation detection in exons 26, 27, and 34 of NOTCH-1 was performed via PCR-based denaturing HPLC using a WAVE DNA fragment analysis system (Transgenomic) equipped with a DNASep HT cartridge. Amplicons for mutation analysis were prepared with the following primer combinations. The N-terminal region of the HD domain of NOTCH-1, encoded in exon 26 was divided into two amplicons. Amplicon HD-N1 was amplified by PCR using primers HD-NLFW1: 5′-AGCCCCCTGTACGACCAGTA (SEQ ID NO:80), HD-N1FW1: 5′-CTTGCGCAGCTCCTCCTC (SEQ ID NO:81), and by nested PCR using primers HD-N1FW2: 5′-GACCAGTACTGCAAGGACCA (SEQ ID NO:82), HD-N1RV2: 5′-TCCTCGCGGCCGTAGTAG (SEQ ID NO:83). Amplicon HD2 was amplified by PCR using primers HD-N2FW1: 5′-GTGCTGCACACCAACGTG (SEQ ID NO:84), HD-N2RV1: 5′-GAGGGCCCAGGAGAGTTG (SEQ ID NO:85), and by nested PCR using primers HD-N2FW2: 5′-GCACACGGCCAGCAGATGAT (SEQ ID NO:86), and HD-N2RV2: 5′-CGCCGGGTCTCACTCAC (SEQ ID NO:87).

The C terminal region of the HD domain of NOTCH 1, encoded in exon 27, was amplified by PCR using primers HD-CFW1: GTGGCGTCATGGGCCTCA (SEQ ID NO:88) and HD-CRV1: TAGCAACTGGCACAAACAGC (SEQ ID NO:89), and by nested PCR using primers HD-CFW2: CATGGGCCTCAGTGTCCT (SEQ ID NO:90) and HD-CRV2: GCACAAACAGCCAGCGTGTC (SEQ ID NO:91).

Sequences of exon 34 encoding the PEST domain and the continuous N-region containing the TAD of NOTCH-1 were divided into three amplicons. Amplicon PEST1 and PEST2 were amplified by PCR using primers PESTFW1: 5′-GCAGCATGGCATGGTAGG (SEQ ID NO:92), and PESTRV1: 5′-AACATGTGTTTTAAAAAGGCTCCTC (SEQ ID NO:93), and by nested PCR using primers PEST1FW: 5′-AAACATCCAGCAGCAGC AAA (SEQ ID NO:94), PEST1RV: 5′-CACAGGCGAGGAGTAGCTGTG (SEQ ID NO:95), and PEST2FW: 5′-GTGACCGCAGCCCAGTTC (SEQ ID NO:96), and PEST2RV: 5′-AAAGGAAGCCGGGGTCTC (SEQ ID NO:97). Amplicon TAD1 was amplified by PCR using primers TAD1FW1: 5′-AGACTGGCCCACCTCGTCTCT (SEQ ID NO:98), TAD1RV1: 5′-GCTCTCCACTCAGGAAGCTC (SEQ ID NO:99), and by nested PCR using primers TAD1FW2: 5′-CGTCTCTCCCACCTGCCTGT (SEQ ID NO:100), and TAD1RV2: 5′-CTGAGCTCACGCCAAGGT (SEQ ID NO:101). Before DHPLC analysis, nested PCR products were denatured at 95° C. for five minutes and slowly ramped to 40° C. using a Peltier Thermal Cycler (MJ Research, PTC-225) in order to allow the formation of homo- and hetero-duplexes in samples heterozygous for a mutant allele.

Fifteen to 30 microliters of annealed PCR fragments per sample were injected on to the DNASep HT cartridge for analysis. Products were eluted at a constant flow rate of 1.5 ml/min with a linear acetonitrile gradient determined by Navigator software (Transgenomic) based on the size and GC-content of the amplicon. The gradient was produced by combining 0.1 m trimethylammonium acetate (TEAA) buffer (pH 7) (Transgenomic) and Buffer B (0.1 m TEAA with 25% acetonitrile). The elution profiles of DNA fragments, monitored by the system's UV detector, were used to produce chromatographs. The analysis sequence for each sample was as follows: a 1.9-minute time lapse for the detector, a 0.1-minute loading stage with a 3% decrease in Buffer B, a 2.0-minute linear gradient with an increasing slope of 2% Buffer B per minute, a 0.1-minute cleaning stage using 75% acetonitrile, and a 0.1-minute equilibration before the next injection. Homo- and hetero-duplex peaks were detected between the initial injection peak, produced by residual nucleotides and primers in the reaction, and the wash stage.

Melting profiles for the nested PCR products, ranging from 276 to 452 based pairs, were constructed using the Navigator software. Appropriate partial denaturing temperatures for mutation detection, which produced percent helicities between 30% and 98% at each base contained in regions of interest, were predicted for all amplicons: PEST1, 66.5° C.; PEST2, 65.0° C.; HD1, 67.0° C. and 64.5° C.; HD2, 67.0° C. and 64.0° C.; and TADI, 64.5° C. and 63.9° C. Less stable heteroduplexes denature earlier than homoduplexes and, thus, appear first in elution profiles. Therefore, direct sequencing was performed on samples with chromatographs displaying species that eluted before normal homoduplexes.

Results

To evaluate the possible role of NOTCH-1 signaling in human T-ALL beyond the rare cases associated with the t(7;9), we first tested T-ALL cell lines lacking the t(7;9) for NOTCH dependency by treating these cells with a γ-secretase inhibitor (Weng, et al., Mol. Cell. Biol. 23:655-64). We observed that five of thirty human T-ALL cell lines tested showed a G₀/G₁ cell cycle arrest, which equaled or exceeded that of the T6E cell line, a reference NOTCH-1-dependent murine T-ALL (FIG. 2A). The growth suppression caused by gamma-secretase inhibitor was abrogated by retroviral expression of ICN1 (FIG. 2B), which is the fragment released after γ-secretase cleavage, and reproduced by retroviral expression of dominant negative Mastermind-like-1 (FIG. 2C). Taken together, these results indicated that the growth of these five cell lines depends on NOTCH-mediated signals.

Because NOTCH receptors can be activated by physical dissociation of the NOTCH extracellular domain (Rand et al., Mol. Cell. Biol. 20:1825-35; Kramer, Sci. STKE 2000, PE1), we reasoned that a newly identified HD domain of NOTCH-1 could be the site of gain-of-function mutations. One class of retroviral insertion observed in murine T-ALL causes deletion of the negative regulatory C-terminal PEST sequences (Hoemann, et al., Mol. Cell. Biol. 20:3831-42; Feldman, et al., Blood 96:1906-13), making this a second candidate region for oncogenic NOTCH-1 mutations. Remarkably, DNA sequencing of these genomic regions revealed both HD domain and PEST domain mutations in 4 of the 5 NOTCH-dependent cell lines. Missense mutations affecting the HD domain (encoded by exon 26 of NOTCH-1) involved residues that are invariant in vertebrate NOTCH-1 receptors, and caused non-conservative changes in amino acid residues (FIG. 3A-C). One cell line, DND-41, had two different HD domain mutations within the same NOTCH-1 allele.

The PEST mutations (found in exon 34) were short insertions or deletions causing shifts in reading frame that are predicted to result in partial or complete deletion of the negative regulatory, C-terminal PEST domain (FIGS. 3D and 3E). PCR amplification, cloning, and sequencing of cDNAs revealed that the HD domain and PEST domain mutations lay in cis in the same NOTCH-1 allele in each of the 4 cell lines tested. Normal NOTCH-1 cDNA clones were also identified in each cell line, indicating that both alleles are expressed. This is consistent with the results of Western blot analysis (FIG. 3F), which revealed that cell lines with exon 26 and 34 mutations contained a polypeptide of the expected size of NTM and additional aberrant polypeptides of slightly smaller size than NTM. In contrast, the cell line TALL-1, which lacks mutations in exons 26 and 34, contained only NOTCH-1 polypeptides of the expected sizes.

We extended our mutational analysis of the HD and PEST domains of NOTCH-1to cryopreserved primary T-ALL samples obtained from the bone marrow of ninety-six children and adolescents at the time of diagnosis. At least one mutation was identified in 49 tumors (51%); 20 tumors (20.8%) had HD domain mutations only, 15 tumors (15.6%) had PEST domain mutations only, and 14 tumors (14.6%) had mutations in both the HD and PEST domains. Mutations were seen in tumors associated with mis-expression of HOX11 (2 of 3 cases), HOX11L2 (10 of 13, or 77%), TAL1 (11 of 31, or 35%), LYL1 (7 of 14, or 50%), MLL-ENL (1 of 3), or CALM-AF10 (1 of 2), which together define the major molecular subtypes of T-ALL. The HD domain mutations in primary tumors were clustered in a “hotspot” spanning residues 1574-1622, and included each of the three L to P missense mutations originally identified in the NOTCH-dependent T-ALL cell lines, as well as deletions of 1-2 residues and short “in-frame” insertions (FIG. 3A-B). PEST domain mutations included insertions or deletions that induced a shift in reading frame, as well as point mutations that created premature stop codons (FIGS. 3D-E, and FIG. 6).

In contrast to T-ALLs, no mutations were observed in 89 primary B-ALL samples. Mutations were also absent from four remission bone marrow samples obtained from patients whose T-ALLs harbored NOTCH-1 mutations, indicating that these mutations are not present in the germline, but rather acquired within the malignant clones. Polymorphisms have not been detected within the HD domain region of NOTCH-1 (NCBI SNP database), leading us to further suspect that the mutations in conserved residues we detected in T-ALLs were likely to have functional consequences.

To address this issue directly, we studied the effects of common HD domain mutations, with and without a PEST domain deletion, on activation of a NOTCH-sensitive luciferase reporter gene in transient expression assays (FIG. 4A). Single L to P mutations within the HD domain at residues 1575, 1594, or 1601 caused a 3-9-fold stimulation of luciferase activity, while an isolated PEST deletion (corresponding to the mutation found in the ALL-SIL cell line that produces a truncation at amino acid 2471) resulted in an approximate 1.5-2-fold stimulation. More strikingly, each HD mutation and the same PEST domain truncation in cis resulted in 20-40-fold stimulations of transcriptional activation. In contrast, the same mutations in trans produced lower levels of stimulation that were close to the average of each mutation acting alone.

Additional reporter gene assays carried out as described in FIG. 4 have proven that all HD mutations tested to date convey increased function on NOTCH-1. The mutations tested in these functional assays are:

V1577E              Sequence ID NO: 102
L1586Q              Sequence ID NO: 106
F1593S              Sequence ID NO: 117
L1597H              Sequence ID NO: 109
R1599P              Sequence ID NO: 15
I1617N              Sequence ID NO: 115
I1617T              Sequence ID NO: 116
V1677D              Sequence ID NO: 135
L1679P              Sequence ID NO: 71
I1681N              Sequence ID NO: 73
A1702P              Sequence ID NO: 129
I1719T              Sequence ID NO: 128
ins(RLGSLNIPYKIEAV) Sequence ID NO: 137

These data suggest that increased function will be a general feature of mutations found in the HD domain of NOTCH1.

The synergistic interaction of HD and PEST domain mutations in cis is consistent with a model in which (i) HD domain mutations enhance γ-secretase cleavage and increase the rate of production of ICN1, and (ii) truncations removing the PEST domain of ICN1 increase the half-life of this activated form of NOTCH (FIG. 4B). The intermediate levels of activation produced by these mutations in trans presumably reflect competition between relatively weak and strong gain-of-function NOTCH-1 polypeptides for factors required for processing and signaling. The stimulatory effects of mutated transmembrane NOTCH-1 polypeptides were completely abrogated by a γ-secretase inhibitor, indicating a requirement for proteolysis at the juxtamembrane site of mutated NTM subunits for signal transduction (FIG. 4A). In contrast, the stimulation produced by ICN1, which is constitutively nuclear, was unaffected by γ-secretase inhibition (FIG. 4A).

More recently we have observed two mutations in tumors other than T-ALL. An insertional mutation in exon 26 of NOTCH-1 was found in a case of acute myelogenous leukemia that causes amino acid residues 1593-1594 to be changed from FL to LSL. This mutation stimulates the ability of NOTCH-1 to activate a CSL-sensitive luciferase gene in assays analogous to those shown in FIG. 4A, indicating that it increases NOTCH-1 function. A novel NOTCH-1 mutation in exon 27 was found in a case of Langerhans cell histiocytosis, in which a highly conserved residue, amino acid 1693, was changed from a C to an R. Together, these data support the prediction that mutations identical or functionally similar to those discovered in T-ALL will be found in other forms of human neoplasia.

Discussion

The prevalence of gain-of-function NOTCH-1 mutations, which are found in every major molecular oncogenic subtype of T-ALL, strongly supports a pre-eminent role for upregulated NOTCH signaling in the pathogenesis of this tumor. The existence of at least one cell line, TALL-1, that apparently requires NOTCH signals for growth, yet lacks NOTCH-1 HD domain and PEST region mutations, suggests that other mechanisms exist that account for pre-T cell transformation involving the NOTCH pathway. It is conceivable that activating mutations affecting other regions of NOTCH-1 may be identified in human T-ALL cases.

Prior work has shown that activation of NOTCH receptors is achieved primarily through release of the NEC subunit (Rand, et al., Mol. Cell. Biol. 20:1825-35; Kramer, Sci. STKE 2000, PE1), which normally represses proteolysis. Recombinant “mini-receptors” consisting of just the NEC HD domain and the extracellular portion of NTM form stable heterodimers after furin cleavage, indicating that critical inter-subunit contacts lie in these regions. Hence, HD domain mutations may act by destabilizing NEC/NTM heterodimers. It follows that similar effects might be imparted by mutations occurring in the analogous heterodimerization region of the NTM subunit.

Several factors may explain the high frequency of NOTCH-1 mutations in T-ALL. The absolute requirement for NOTCH-1 signals during several stages of normal early T cell development provides a functional basis for the frequent involvement of this pathway. Further, unlike the t(7;9), which is created by aberrant recombination during attempted V-D-Jβ rearrangement in committed T cell progenitors, the more common point mutations and insertions described here could occur in multipotent hematopoietic progenitors, which also normally express NOTCH-1 (Calvi, et al., Nature 425,841-6). If so, such mutations would be predicted to induce daughter cells to adopt a T cell fate (Allman, et al., J. Exp. Med. 194:99-106) and thereby increase the pool of cells that are at risk for additional leukemogenic events, which could include additional mutations affecting the same NOTCH-1 allele, as well as the chromosomal, genetic and epigenetic events that lead to the mis-expression of other critical transcription factors. Analogous NOTCH mutations might be involved in the molecular pathogenesis of certain more common types of human cancer, as NOTCH receptors are expressed in many normal tissues and a variety of malignancies (Allenspach, et al., Cancer Biol. Ther. 1:466-76).

Our findings identify the NOTCH pathway as a rational target for molecular therapy in T-ALL. Although approximately 75% of patients are currently cured of this disease with very intensive and cytotoxic chemotherapy regimens (Pui, et al., N. Engl. J. Med. 350:1535-48), new therapies are needed for patients with refractory disease, and less toxic, more efficacious combinations of drugs would be beneficial to all T-ALL patients. The development of potent, specific inhibitors of γ-secretase (Wolfe, Nat. Rev. Drug Discov. 1:859-66), due to the involvement of this protease in the pathogenesis of Alzheimer's disease, should expedite clinical trials of cancer therapies aimed at blocking NOTCH-1 activity.

All references cited herein are fully incorporated by reference. Having now fully described the invention, it will be understood by those of skill in the art that the invention may be practiced within a wide and equivalent range of conditions, parameters and the like, without affecting the spirit or scope of the invention or any embodiment thereof.

Claims

1. A substantially purified mutant NOTCH-1 receptor, comprising the amino acid sequence of SEQ ID NO:1 except that said amino acid sequence has one or more mutations in amino acids 1571-1622 and/or 1674-1730 and/or amino acid 2311 to amino acid 2556 such that, relative to unmutated NOTCH-1, said mutant NOTCH-1 receptor exhibits increased signaling in response to incubation with gamma-secretase.

2. The substantially purified mutant receptor of claim 1, wherein, except for said one or more mutations, said mutant receptor consists of the amino sequence of SEQ ID NO:1.

3. The substantially purified mutant NOTCH-1 receptor of claim 1, wherein said one or more mutations include one or more of the following:

a) the portion of the amino acid sequence of SEQ ID NO:1 from amino acid 1571 to amino acid 1618 is replaced with a sequence selected from the group consisting of SEQ ID NO:12-SEQ ID NO:22, SEQ ID NO:102-118, and SEQ ID NO:145;

b) the portion of the amino acid sequence of SEQ ID NO:1 from amino acid 1571 to amino acid 1622 is replaced with a sequence selected from the group consisting of SEQ ID NO:23- SEQ ID NO:29, or SEQ ID NO:119-127;

c) the portion of the amino acid sequence of SEQ ID NO:1 from amino acid 1674 to amino acid 1730 is replaced with a sequence selected from the group consisting of SEQ ID NO:70- SEQ ID NO:75, or SEQ ID NO:128-137;

d) the portion of the amino acid sequence of SEQ ID NO:1 from amino acid 2311 to amino acid 2556 is replaced with a sequence selected from the group consisting of SEQ ID NO:30- SEQ ID NO:59 or SEQ ID NO:138-144.

4. The substantially purified mutant receptor of claim 2, wherein, except for said one or more mutations, said mutant receptor consists of the amino sequence of SEQ ID NO:1.

5. The substantially purified mutant NOTCH-1 receptor of claim 1, wherein said sequence is mutated at one or more amino acid positions selected from the group consisting of: 1571; 1573; 1575; 1576; 1578; 1579; 1583; 1586; 1588; 1590; 1593; 1594; 1595; 1598; 1600-1607; 1610-1618; 1674; 1675; 1677; 1679-1684; 1686; 1695; 1696; 1703; 1706; 1715; 1720; 1725; and 1728.

6. The substantially purified receptor of claim 5, wherein, except for said one or more mutated amino acids, said receptor consists of the amino acid sequence of SEQ ID NO:1.

7. A substantially purified polynucleotide comprising a nucleotide sequence encoding the mutant NOTCH-1 receptor of claim 1.

8. The substantially purified polynucleotide of claim 7, wherein said polynucleotide consists of a nucleotide sequence encoding the mutant receptor of claim 1.

9. A substantially purified polynucleotide, at least 15 nucleotides in length which hybridizes to the polynucleotide of claim 8 under stringent conditions but which does not hybridize to a polynucleotide encoding unmutated NOTCH-1 under the same conditions.

10. A vector comprising a promoter operably linked to a polynucleotide encoding the mutant NOTCH-1 receptor of claim 1.

11. A host cell transformed with the vector of claim 10.

12. An antibody that binds to the mutant NOTCH-1 receptor of claim 1 with at least a hundredfold greater affinity than to the unmutated NOTCH-1 receptor of SEQ ID NO:1.

13. A method of diagnostically testing a biological sample for the presence of cancer cells, comprising assaying said biological sample for the mutant NOTCH-1 receptor of claim 1.

14. A method of diagnostically testing a biological sample for the presence of cancer cells, comprising assaying said biological sample for cells expressing the polynucleotide of claim 7.

15. A method of diagnostically testing a biological sample for the presence of cancer cells, comprising assaying said biological sample for cells expressing the polynucleotide of claim 8.

16. The method of claim 15, wherein said method is performed by PCR amplifying either the full length polynucleotide encoding said mutant NOTCH-1 receptor or a portion of the polynucleotide encoding said mutant NOTCH-1 receptor.

17. A method of slowing or stopping the growth of cells identified as having the mutant NOTCH-1 receptor of claim 1, comprising contacting said cells with an effective amount of a gamma-secretase inhibitor.

18. A method of treating a patient for cancer comprising:

a) assaying a biological sample from said patient for the presence of cancer cells having the mutant Notch-1 receptor of claim 1;

b) treating said patient with an effective amount of an agent that inhibits Notch-1 receptor signaling if said biological sample is found to have said mutant Notch-1 receptor.

19. The method of claim 18, wherein said cancer is T-cell acute lymphoblastic leukemia.

20. The method of claim 18, wherein said agent is a gamma-secretase inhibitor.

21. A method of assaying a test compound for its ability to inhibit abnormal cellular growth induced by the presence of a mutated NOTCH-1 receptor, comprising:

a) incubating said test compound with cells that express the mutant NOTCH-1 receptor of claim 1, said incubation being carried out in the presence of gamma-secretase;

b) comparing the amount of receptor activity that occurs as the result of the incubation of step a) with the amount of activity that occurs in a similar incubation carried out in the absence of said test compound; and

c) concluding that said test compound should inhibit the growth of cells expressing mutant NOTCH-1 receptor if the amount of activity observed in the presence of said test compound is less than the amount of activity observed in the absence of said test compound.