Compositions and methods of treating tumors

IS 2 270 586 T3

DESCRIPTION

Compounds and methods for the treatment of tumors.

Technical sector to which the invention belongs

The present invention relates to proteins that lack tyrosine kinase activity and that dimerize with members of the erbB family of receptors; to nucleic acid molecules encoding said proteins; to pharmaceutical compounds comprising said nucleic acid molecules in combination with carrier vehicles that facilitate the transfer of the nucleic acid molecule into a cell; and to the use of said pharmaceutical compound in the manufacture of a medicament for preventing tumors and treating individuals who have tumors. The present invention relates to compounds that are useful for converting tumor cells resistant to radiation and / or chemically induced cell death into cells that are sensitive to radiation.

The present invention relates to the use of said compounds in the manufacture of a drug to be administered in combination with radiation and / or chemotherapy, to treat individuals who have tumors.

Background of the invention

The erbB family of receptors comprises erbB (EGFR), erbB2 (p 185), erbB3, and erbB4. Ullrich, et al. (1984) Nature 309, 418-425, which is incorporated herein by reference, describes EGFR. Schechter, AL, et al. (1984) Nature 312, 513-516, and Yamamoto, T., et al. (1986) Nature 319, 230-234, describe p185neu / erbB2. Kraus, MH, et al. (1989) Proc. Natl. Acad. Sci. USA 86, 9193-9197 describes erbB3. Plowman, GD, (1993) Proc. Natl. Acad Sci. USA 90, 1746-1750, describes erbB4.

The rat cellular proto-oncogen c-neu and its corresponding human c-erbB2 encode 185 kDa transmembrane glycoproteins designated p185. Tyrosine kinase (tk) activity has been related to the expression of the transforming phenotype of oncogenic p185 (Bargmann et al., Proc. Natl. Acad. Sci. USA, 1988, 85, 5394; and Stem et al., Mol. Cell. Biol., 1988, 8, 3969. The oncogenic neu was initially identified in rat neuroglioblastomas (Schechtler et al., Nature, 1984, 312, 513 and was found to be activated by a carcinogen-induced point mutation generating a single amino acid substitution, a Val to Glu substitution in the position 664, in the transmembrane region of the transforming protein (Bargmann et al., Cell, 1986, 45, 649). This alteration results in a constitutive activity of its intrinsic kinase and in malignant transformation of cells (Bargmann et al., EMBO J., 1988, 7, 2043). The activation of the tyrosine kinase of the p185 protein is shown to be related to a shift in molecular equilibrium from monomeric to dimeric forms (Weiner et al., Nature, 1989, 339, 230).

Overexpression of c-neu or c-erbB2 at levels 100 times higher than normal (i.e.,> 10⁶ receptors / cell) also results in the transformation of NIH3T3 cells (Chazin et al., Oncongene, 1992, 7, 1859; DiFiore et al., Science, 1987, 237, 178; and DiMarco et al., Mol. Cell. Biol., 1990 , 10, 3247). However, NIH3T3 cells or NR6 cells expressing cellular p185 at the level of 10⁵ receptors / cell is not transformed (Hung et al., Proc. Natl. Acad. Sci. USA, 1989, 86, 2545; and Kokai et al., Cell, 1989, 58, 287), except that they are co-expressed with the factor receptor epidermal growth cell (EGFR), a homologous tyrosine kinase (Kokai et al. Cell. 1989, 58, 287). Therefore, oncogenic lap185ylap185 can both result in cell transformation.

Cellular p185 is highly homologous to EGFR (Schechter et al., Nature, 1984, 312, 513; and Yamamoto, et al., Nature, 1986, 319, 230) but it is distinct nonetheless. Numerous studies indicate that EGFR and cellular p185 are capable of interacting (Stern et al., Mol. Cell. Biol., 1988, 8, 3969; King et al., EMBO J, 1988, 7, 1647; Kokai et al., Proc. Natl. Acad. Sci. USA, 1988, 85, 5389; and Dougall et al., J. Cell. Biochem., 1993, 53, 61). The intermolecular association of EGFR and cellular p185 shows upregulation of EGFR function (Wada et al., Cell, 1990, 61, 1339). Furthermore, heterodimers that form kinase complexes active both in vivo and in vitro can be detected (Qian et al., Proc. Natl. Acad. Sci. USA, 1992, 89, 1330).

Similarly, interactions of p185 with other members of the erbB family have been reported (Carraway et al., Cell 1994, 78, 5-8; Alroy et al., FEBS Lett. 1997, 410, 83-86; Riese et al. , Mol. Cell. Biol. 1995,15, 57705776; Tzahar et al., EMBO J. 1997, 16, 4938-4950; Surden et al., Neuron 1997, 18, 847-855; Pinkas-Kramarski et al., Oncogene 1997, 15, 2803-2815). Human p185 forms heterodimers with either erbB3 or erbB4 under physiological conditions, mainly in cardiac muscles and in the nervous system, particularly in development.

Cellular p185 proteins are found in adult secretory epithelial cells of the lung, salivary gland, sinuses, pancreas, ovaries, gastrointestinal tract, and skin (Kokai et al., Proc. Natl. Acad Sci. USA, 1987, 84, 8498; Mori et al., Lab. Invest., 1989, 61, 93; and Press et al., Oncogene, 1990, 5, 953). Recent studies have shown that c-erbB2 amplification occurs very frequently in a number of human adenocarcinomas such as gastric (Akiyama et al., Science, 1986, 232, 1644), lung (Kern et al., Cancer Res. , 1990, 50, 5184) and pancreatic adenocarcinomas (Williams et al., Pathobiol. 1991, 59,46). Increased c-erbB2 expression in a subset of breast and ovarian carcinomas has also been reported to be related to a poor clinical prognosis (Slamon et al., Science, 1987, 235, 177; and Slamon et al., Science, 1989 , 244, 707).

IS 2 270 586 T3

The heterodimeric association of EGFR and p185 has also been detected in human breast cancer cell lines, such as SK-Br-3 (Goldman et al., Biochemistry, 1990, 29, 11024), and transfected cells (Spivak-Kroizman et al. , J. Biol. Chem., 1992, 267, 8056). Additionally, erbB2 and EGFR co-expression have been reported in breast and prostate cancers. Furthermore, the heterodimeric association of p185 and erbB3, as well as the heterodimeric association of p185 and erbB4 have also been detected in human cancers. The coexpression of erbB2 and erbB3 has been observed in human breast cancers. EGFR, erbB2 and erbB3 co-expression has been observed in prostate carcinoma.

Amplification and / or alteration of the EGFr gene is frequently observed in glial tumor progression (Sugawa, et al. (1990) Proc. Natl. Acad Sci. 87: 8602-8606; Ekstrand, et al., (1992) Proc. Natl Acad. Sci. 89: 4309-4313), particularly in glioblastoma, the most malignant glial tumor (Libermann, et al. Supra; Wong, et al. Supra; James, et al. (1988) Cancer Res. 48: 5546-5551; Cavenee, WK (1992) Cancer 70: 1788-93, Nishikawa, et al., (1994) Proc. Natl. Acad. Sci. 91: 7727-7731; Schlegel, et al. (1994) Int J. Cancer 56: 72-77). A significant proportion of these tumors show EGFr amplification with or without gene alteration (Ekstrand, et al. Supra; Libermann, et al. Supra; Wong, et al. (1987) Proc. Natl. Acad. Sci. 84: 6899-6903), and this has been correlated with a shorter interval of disease recurrence and lower survival (Schlegel, et al. Supra).

EGFr amplification can be associated with aberrant EGFr transcripts in conjunction with normal EGFr transcripts (Sugawa, et al. Supra). The frequent amplification and subsequent structural alteration suggests that EGFr may be important for the maintenance of the malignant glioma phenotype. A frequently observed mutant EGFr has been identified in a subset of human glioblastomas and results from an 801 bp in-frame truncating (corresponding to exons 2-7) in the extracellular domain of the receptor (Sugawa, et al. Supra; Malden et al., (1988) Cancer Res. 48: 2711-2714; Humphrey, et al. (1990) Proc. Natl. Acad. Sci. 87: 4207-4211; Wong, et al. (1992) Proc. Natl. Acad Sci. 89: 2965-2969), which is believed to result in constitutive kinase activation and may also affect ligand binding characteristics of the molecule (Nishikawa, et al. Supra; Callaghan, et al. (1993) Oncogene 8: 2939 -2948).

The observed EGFr mutations in human epithelial malignancies consist of overexpression with or without amplification and, less commonly, coding sequence alterations. Oncogenic transformation elicited by EGFr mutants appear to be tissue specific and have been observed in erythroid leukemia, fibrosarcoma, angiosarcoma, melanoma, and also glioblastoma (Carter, et al. (1994) Crit Rev Oncogenesis 5: 389-428). Overexpression of normal EGFr can cause oncogenic transformation in certain cases, probably in an EGF-dependent manner (Carter, et al. Supra; Haley, et al. (1989) Oncogene 4: 273-283). Transfection of large amounts of wild-type EGFr into NIH3T3 cells results in ligand-dependent but incomplete transformation (Yamazaki, et al. (1990) Jpn. J. Cancer Res. 81: 773-779). Overexpression can cause altered cell cycle regulation of EGFr kinase, and can contribute to the transformed state, as has been observed for oncogenic p185neu (Kiyokawa, et al. (1995) Proc. Natl. Acad Sci. 92: 1092-1096) .

There is a need for therapeutic compounds useful for treating individuals identified as having erbB-mediated tumors. There is a need to develop prophylactic compounds for individuals susceptible to developing erbB-mediated tumors. There is a need for methods for treating individuals identified as having erbB-mediated tumors. There is a need for methods of preventing individuals who are susceptible to developing erbB-mediated tumors from developing such tumors.

Characteristics of the invention

The present invention relates to the use of compounds that alter and inhibit erbB-associated kinase activity in tumor cells in the manufacture of a drug to be administered in combination with radiation of a cytotoxic agent to treat an individual affected by a tumor, as indicated in claims 1-23.

The present invention relates to the use of a compound that inhibits the formation of erbB protein dimers, which produce high tyrosine kinase activity in tumor cells in the manufacture of a drug to be administered in combination with a therapeutically amount effective anticancer radiation treatment of an individual who has an erbB protein-mediated tumor. In some embodiments, the erbB protein mediated tumor is a p185 mediated tumor. In some embodiments, the erbB protein-mediated tumor is an EGFR-mediated tumor. In some embodiments, the erbB protein mediated tumor is a glial tumor. In some embodiments, the erbB protein-mediated tumor is a glioblastoma. In some embodiments, administration of the compound occurs by intratumoral administration. In some embodiments, the individual receives surgery prior to compound administration. In some embodiments, the compound administered to a patient comprises a compound that interacts with an erbB protein in a tumor cell to alter the erbB protein sufficiently to result in a decreased propensity for it to dimerize with another erbB protein. In some embodiments, the compound that interacts with an erbB protein in a tumor cell to alter the erbB protein sufficiently to result in a decreased propensity for it to dimerize with another erbB protein is an antibody. In some embodiments, the antibody is a monoclonal antibody. In some embodiments, the compound that is administered to a patient comprises a compound that competitively interacts with an erbB protein in a tumor cell to competitively inhibit dimer formation with another erbB protein and prevent elevated tyrosine kinase activity. In some embodiments, the interacting compound

ES 2 270 586 T3 competitively with an erbB protein in a tumor cell to competitively inhibit formation with another erbB protein is a peptide. In some embodiments, the compound that competitively interacts with an erbB protein in a tumor cell to competitively inhibit dimer formation with another erbB protein is an antibody. In some embodiments, the compound that is administered to the tumor cell is a nucleic acid molecule that encodes a protein that competitively interacts with an erbB protein in a tumor cell to competitively inhibit dimer formation with another erbB protein. In some embodiments, the protein is a mutant or truncated kinase-deficient erbB protein. In some embodiments, the protein is a truncated or mutant kinase-deficient p185 protein. In some embodiments, the protein interacts with the transmembrane region of an erbB protein. In some embodiments, the protein comprises a rat neu transmembrane region with a val to glu mutation at amino acid 664. In some embodiments, the protein interacts with the ectodomain region of the first erbB protein. In some embodiments, the protein comprises a p185 ectodomain. In some embodiments, the nucleic acid molecule is administered by intratumoral administration. In some embodiments, the individual receives surgery prior to administration of the nucleic acid molecule. In some embodiments, the nucleic acid molecule is the viral genome of a recombinant adenovirus. In some embodiments, the nucleic acid molecule comprises coding sequences operably linked to regulatory elements for translation in cells of the individual, and the coding sequence comprises: a truncated rat neu gene with a stop codon at amino acid 691; a truncated rat neu gene with a stop codon at amino acid 691 and a val-> glu mutation at amino acid 664; a chimeric p185 gene comprising the rat neu transmembrane-linked human p185c-erbB2 ectodomain with a stop codon at amino acid 691; a chimeric p185 gene comprising the rat neu transmembrane-linked human p185c-erbB2 ectodomain with a stop codon at amino acid 691 and val-> glu mutation at amino acid 664; a chimeric p185 gene comprising the human EGFR ectodomain linked to the rat neu transmembrane with a stop codon at amino acid 691; a chimeric p185 gene comprising the human EGFR ectodomain linked to the rat neu transmembrane with a stop codon at amino acid 691 and a val-> glu mutation at amino acid 664; a chimeric p185 gene comprising the human erbB3 ectodomain linked to the rat neu transmembrane with a stop codon at amino acid 691; a chimeric p185 gene comprising the human erbB3 ectodomain linked to the rat neu transmembrane with a stop codon at amino acid 691 and a val-> glu mutation at amino acid 664; a chimeric p185 gene comprising the rat neu transmembrane-linked human erbB4 ectodomain with a stop codon at amino acid 691; or a chimeric p185 gene comprising the rat neu transmembrane-linked human erbB4 ectodomain with a stop codon at amino acid 691 and a val-> glu mutation at amino acid 664.

The present invention relates to the use of a compound that alters the kinase activity of a multimeric pool in the manufacture of a drug to be administered in combination with a therapeutic amount of gamma radiation to an individual who has a tumor, so that the tumor is characterized by tumor cells that have multimeric receptor sets that provide kinase activity associated with a transformed phenotype. In some embodiments, the tumor is characterized by tumor cells having multimeric receptor sets selected from the group consisting of: homodimeric erbB, heterodimeric erbB, and platelet-derived growth factor receptor multimers. In some embodiments, the tumor is characterized by tumor cells that have erbB homodimers that are mutant EGFR homodimers or p185 homodimers. In some embodiments, the tumor is characterized by tumor cells having erbB heterodimers that are p185 / EGFR heterodimers, p185 / mutant EGFR heterodimers, p185 / erbB3 heterodimers; p185 / erbB4 heterodimers or mutant EGFR / EGFR heterodimers. In some embodiments, the compound that alters kinase activity associated with the multimeric receptor set comprises an active agent selected from the group consisting of antibodies, peptides, and non-protein kinase inhibitors. In some embodiments, the compound that alters the kinase activity associated with the multimeric receptor comprises an active agent that is a nucleic acid molecule that encodes a protein or peptide that interacts with a monomeric component of the set to prevent the monomeric component from interacting with a second monomer component of the set.

The present invention relates to the use of a small peptide, non-protein compound or nucleic acid molecules that encode a non-antibody protein or peptide that alters the kinase activity associated with the multimeric receptor set in the manufacture of a drug to be administered in combination with a therapeutic amount of gamma radiation and / or a therapeutic amount of a cytotoxic chemotherapeutic agent to treat an individual who has a tumor, such that the tumor is characterized by tumor cells that have multimeric receptor sets that provide kinase activity associated with a transformed phenotype. In some embodiments, the tumor is characterized by tumor cells having multimeric receptor sets selected from the group consisting of: homodimeric erbBs, heterodimeric erbBs, and platelet-derived growth factor receptor multimers. In some embodiments, the tumor is characterized by tumor cells that have homodimers of erbB that are homodimers of mutant EGFR, or homodimers of p185. In some embodiments, the tumor is characterized by tumor cells having erbB heterodimers that are p185 / EGFR heterodimers, mutant p185 / EGFR heterodimers, p185 / erbB3 heterodimers; p185 / erbB4 heterodimers or mutant EGFR / EGFR heterodimers. In some embodiments, the compound that alters kinase activity associated with the multimeric receptor set comprises an active agent selected from the group consisting of peptides, and non-protein kinase inhibitors. In some embodiments, the compound that alters kinase activity associated with the multimeric receptor assembly comprises an active agent that is a nucleic acid molecule that encodes a protein or peptide that interacts with a monomeric component of the assembly to prevent the monomeric component from interacting. with a second monomer component of the set.

IS 2 270 586 T3

Brief description of the figures

Figures 1A, 1B, 1C and 1D show data regarding the distribution of cell cycles in human glioblastoma cells, with or without radiation treatment. The cells were applied in 60 mm dishes and allowed to stick before being irradiated by gamma rays (10Gy) (Figures 1B and 1D) or falsely irradiated (Figures 1A and 1C). After 72 hours, the cells were analyzed by flow cytometry after PI staining. Cell distributions according to DNA content are indicated in each panel. Representative experiments were carried out four independent times.

Figures 2A and 2B show data relating to the determination of apoptosis and clonogenic survival after gamma-ray irradiation of human glioblastoma cells. In Figure 2A, cells were layered and allowed to stick prior to exposure to gamma irradiation (10Gy), in 10% serum medium or serum-free medium. After 72h, apoptosis quantification was carried out by two independent observers. The apoptotic index is the percentage of apoptotic cells with morphological evidence of apoptosis, determined by staining of nuclei with DAPI. The results presented are means ± SEM of four independent experiments and the mean is indicated in parentheses. U87MG cells were cultured in 10% serum medium or serum-free medium, and U87 / T691 cells were cultured in 10% serum medium or serum-free medium. In Figure 2B, U87MG and U373MG human glioma cells were stained with DAPI and analyzed for apoptotic morphology 72h after gamma irradiation. The mean is indicated in parentheses and the index shown in this representative experiment is mean ± SD. These results were confirmed in two additional experiments. The apoptotic indices were interpreted to be underestimates, since floating cells could not be tested by this technique.

Figure 3 shows clonogenic survival after irradiation. The cells were applied in the form of a layer and subjected to gamma ray irradiation with different doses of irradiation, followed by incubation for 7-10 days at 37 ° C, with 5% CO.₂. Colonies were stained and those with more than 50 cells were counted under a dissecting microscope. The log of survival was determined by calculating the ratio of the number of colonies formed to the number of cells applied in a layer, after correction for the performance of the application in layer. Similar experiments were carried out three times.

Figure 4 is a schematic map of recombinant adenovirus H5.001 CBLacZ.

Figure 5 shows inhibition of cell proliferation of p 185^{c erbB 2} expressing human tumor cells. 5,000 cells / well were layered with an indicated amount of CDR4D5 mimetic or anti-p 185^c-erbB-2Ab (Neomarkers Inc., CA) and incubated at 37 degrees for 24h. 100 pg of MTT (in 200 pg) was added to each of the wells for 4 hours, followed by the addition of 100 µl of lysis buffer to each well. After 1224 hours, OD At 570 nm was read with an ELISA reader. The% proliferation indicates the degree of proliferation with respect to the control cells (without CDR4D5 or anti-p185 treatment^c-erbB-2 Mimetic ab).

Figures 6A and 6B show that anti-p185^c-erbB-2 CDR4D5 mimetic sensitizes human tumor cells to apoptosis induced by gamma-ray irradiation. U3, u3t, S and M indicate cell lines U373MG and U373 / T691, SKBR3 and MCF7, respectively. I go m indicate 4D5 mimetic and CD4 serine mimetic go relevant.

Description of the preferred embodiments

As used in this description, the terms "erbB-associated cancer" and "erbB-associated tumors" are intended to refer to tumor cells and neoplasms that express a member of the erbB family of genes, the expression of which results in mediated transformation. by erbB. Neu-associated tumors and EGFR-associated tumors are examples of erbB-associated tumors.

As used in this description, the terms "Neu-associated cancer," "Neu-associated tumors," and "ap 185-associated tumors" are intended to refer to tumor cells and neoplasms that express the Neu gene to produce p 185. Neu-associated cancer is an erbB-associated cancer in which cell transformation is mediated by tyrosine kinase activity, related to p 185.

As used in this description, the terms "EGFR-associated cancer" and "EGFR-associated tumors" are intended to refer to EGFR-expressing tumor cells and neoplasms. EGFR-associated cancer is erbB-associated cancer, in which cell transformation is mediated by EGFR-related tyrosine kinase activity.

As used in this description, the terms "mutant EGFR-associated cancer" and "mutant EGFR-associated tumors" are intended to refer to tumor cells and neoplasms expressing mutant forms of EGFR. A mutant EGFR-associated cancer is an EGFR-associated cancer in which cell transformation is mediated by tyrosine kinase activity, related to mutant EGFR. Receptor subunit alterations, as a result of structural changes, can be coupled to receptor oligomerization resulting in amplification of signaling. A mutant EGFR can be a constitutively activated mutant EGFR form with extracellular clearance (AEGFR) that is commonly seen in human glial tumors. An AEGFR oncoprotein commonly observed in human glial neoplasms and other epite5 malignancies

ES 2 270 586 T3 human lials (AEGFR or EGFRvIII) results from an internal truncating that involves exons 2 to 7 (amino acids 6 to 273), in the gene that encodes the extracellular region of the molecule that results in the expression of AEGFRs of 140155kDa constitutively phosphorylated and truncated. AEGFR receptors have been observed to exist spontaneously in dimer form and mediate constitutive signaling and oncogenic transformation of rodent fibroblasts independently of the ligand, while overstressed p170holo-EGFRs are only weakly transformed in the presence of EGF. . The AEGFR oncoproteins confer a remarkable growth advantage in vivo in human glioblastoma cells and in murine fibroblasts. Recent reports indicate that AEGFR receptors are present on the surface of cells and are internalized more slowly than ligand-stimulated holo-EGFRs, which can increase the efficiency of AEGFR oncoprotein transformation. Other mutations that functionally separate the extracellular domain from the transmembrane and cytoplasmic region of RTK polypeptides have also been observed to lead to spontaneous dimerization and the acquisition of transformation potential, suggesting that a part of the extracellular domain imposes a structural limitation on dimer formation. , which is presumably removed by ligand binding or mass action. The extracellular deletions observed in avian AEGFR or v-erbB oncogenes presumably facilitate dimer formation by mimicking the conformational changes that result from ligand binding. Soluble extracellular domains of EGFR have been shown to oligomerize and structural alteration in the ectodomain can induce spontaneous oligomerization of extracellular domains, cytoplasmic domains, or both. Extracellular deletion in EGFR removes most of the amino acids that comprise subdomains I and II of EGFR, including a large portion of the first (most amino-terminal) of two cysteine-rich sequences in the extracellular region of the receptor. Subdomain III, which has been reported to confer ligand-binding characteristics to EGFR, is preserved in the EGFR oncoprotein, although EGFRs do not appear to bind ligands in NIH3T3 cells. Coexpression of holo-EGFRs and AEGFRs has been observed in human glioblastoma and other tumor samples, suggesting that AEGFR / EGFR, which co-express cells, may be a close correlate of human disease.

As used in this description, the term "EGFR species" is intended to refer to wild-type EGFR and mutant forms thereof.

As used in this description, the term "erbB-mediated cellular transnphromation" is intended to refer to the cellular transformation that erbB-associated tumor cells and neoplasms undergo. Cells undergo erbB-mediated transformation related to elevated levels of tyrosine kinase activity by members of the erbB receptor family. The transformed genotype of erbB-mediated transformed cells can be disrupted and / or reversed by expression of tyrosine kinase-deficient proteins that dimerize with members of the erbB receptor family.

As used in this description, the term "p185-mediated cellular transformation" is intended to refer to the cellular transformation that p185-associated tumor cells and neoplasms undergo, and whose transformed phenotype can be disrupted and / or reversed by expression. tyrosine kinase-deficient proteins that dimerize with p185. P185-mediated cell transformation is erb-mediated cell transformation.

As used in this description, the term "EGFR-mediated cellular transformation" is intended to refer to the cellular transformation that EGFR-associated tumor cells and neoplasms undergo, and whose transformed phenotype can be disrupted and / or reversed by expression of tyrosine kinase-deficient proteins that dimerize with EGFR. EGFR-mediated cell transformation is erb-mediated cell transformation.

As used in this description, the term "mutant EGFR-mediated cellular transformation" is intended to refer to cellular transformation that undergoes mutant EGFR-associated tumor cells and neoplasms, and whose transformed phenotype can be disrupted and / or reversed by expression of tyrosine kinase deficient proteins that dimerize with mutant EGFR. A mutant EGFR-mediated cell transformation is an erb-mediated cell transformation.

As used in this description, the term "carrier components" is intended to refer to vehicles by which nucleic acid molecules can be delivered to cells of an individual. Carrier components are intended to include viral particles, such as viral particles from gene therapy vectors, as well as other vehicles, carriers, complexes, entities, and structures that are useful in delivering a nucleic acid molecule to a cell.

As used in this description, the term "high risk individual" is intended to refer to an individual who has had an erbB-associated tumor, such as, for example, a neu-associated tumor, that has been removed or that it is in remission and that, therefore, is susceptible to relapse or recurrence. As part of a treatment regimen for a high-risk individual, the individual can be treated prophylactically against tumors that have been diagnosed in order to combat a recurrence. Therefore, once it is known that an individual has had erbB-associated cancer, the individual can be treated in accordance with the present invention to prevent normal cells from transforming into tumor cells.

The present invention is useful for therapeutically treating an individual identified as affected by erbB-associated tumors, such as neu-associated tumors, in order to reverse the transformed phenotype of tumor cells. The present invention is useful for prophylactically treating an individual who is predisposed to develop

ES 2 270 586 T3 erbB associated tumors or that has had erbB associated tumors and is therefore susceptible to relapse or recurrence.

The translation product of the neu oncogen is p185, a transmembrane glycoprotein that has tyrosine kinase activity and a molecular weight of 185,000 daltons, as determined by electrophoresing the glycoprotein and comparing its movement with marker proteins of known molecular weight. . Experiments have shown that p185 forms dimers with other p185 molecules or with epidermal growth factor receptor (EGFR) and that these dimers show high tyrosine kinase activity that contributes to the transformed phenotype in cells that have such dimers.

The p185neu mutants interfere with activated p185neu homodimers (Qian, O'Rourke, Zhao, Greene: Oncogene 13: 2149-2157, 1996). The p185neu mutants also interfere with normal EGFR homodimers, mutant EGFR and also activated EGFR homodimers in fibroblasts and in primary human cancer cells. Administration of nucleic acid molecules that encode proteins capable of forming dimers with other p185 molecules or with EGFR, but whose dimers do not show high tyrosine kinase activity, eliminates the transformed phenotype of neu-associated tumors in a population that suffers from mediated tumors. by p185. Furthermore, the administration of said nucleic acid molecules inhibits neoplastic development in animals susceptible to developing neu transformed tumors.

As explained above, interactions of p185-erbB2 with other elements of the erbB family have been reported (Carraway and others, Alroy and others, supra; Riese and others, supra; Tzahar and others, supra; and Surden and others, supra; Pinkas-Kramarski et al., supra). Accordingly, kinase deficient mutants of p185neu / erbB2 (human homologue) that require the ability to form heterodimers with EGFR, erbB3, and erbB4 can be used to form dimers with erbB3 and erbB4, as well as EGFR and modulate signaling in human tumor cells. Therefore, the present invention further relates to the administration of nucleic acid molecules that encode proteins capable of forming dimers with erbB3 and erbB4, but whose dimers do not show high tyrosine kinase activity to eliminate the transformed phenotype of tumors in a population. affected by such tumors. Furthermore, the administration of molecules of said nucleic acid inhibits neoplastic development in animals susceptible to developing tumors.

In addition to p185 as mutants that interfere with erbB-mediated cell transformation, other mutant erbB elements may be useful to dimerize with wild-type erbB proteins and inhibit high tyrosine kinase activity associated with wild-type homodimers and heterodimers.

The present invention provides nucleic acid molecules having a nucleotide sequence that encodes a protein that lacks tyrosine kinase activity and dimerizes with an element of the erbB receptor family. The protein dimerizes with an erbB protein selected from the group consisting of EGFR, p185, erbB3 and erbB4, preferably at least two elements of the erbB family selected from the group consisting of EGFR, p185, erbB3 and erbB4, more preferably at least three elements of the erbB family selected from the group consisting of EGFR, p185, erbB3 and erbB4 and more preferably the protein dimerizes with each of EGFR, p185, erbB3 and erbB4. It is preferable that the protein is a mutated or truncated form of a protein that is an element of the erbB family or a chimeric protein that includes sequences of elements of the erbB family derived from different species. In some preferred embodiments the invention provides nucleic acid molecules having a nucleotide sequence that encodes a protein that lacks tyrosine kinase activity and dimerizes with human EGFR or human p185. The nucleic acid molecules are applied in combination with transport components such that in the administration of the combination the nucleic acid molecule is delivered to cells of the individual. When provided as a pharmaceutical compound, the combination is useful for the treatment of individuals affected by erbB-mediated cellular transformations such as p185-mediated cellular transformation and EGFR-mediated cellular transformation. This pharmaceutical compound may also be useful for the prevention of erbB-mediated cellular transformation particularly in individuals susceptible to such transformation. The nucleic acid molecules of the invention can also be useful to produce specific erbB protein species in competent cells that can be subsequently isolated and used in various immunoassays to detect the presence of specific antibodies for said erbB proteins present in different body fluids.

Rat cell P185 lacking kinase activity due to a single amino acid substitution in the consensus sequence for ATP binding, N757 or due to cytoplasmic domain deletion, N691 disruption, was able to support EGF-induced heterodimerization with EGFR in living cells . EGF was also able to stimulate the trans-phosphorylation of N757 via EGFR. However composite EGFR heterodimers and certain truncated p185 proteins had kinase inactivity. (See Qian et al., Proc. Natl. Acad Sci. USA, 1994, 91, 1500). Similar results were observed using another modified construct in which the transmembrane region of the truncated p185 protein contained a unique amino acid sequence change, T691stop. Structural alterations in receptors have been shown to act as dominant negative mutations that can suppress the function of wild-type (wt) receptors, such as an insulin receptor (Chou et al., J.Biol. Chem. 1987, 262, 1842 ) or EGFR (Honegger et al., J. Cell Biol., 1990, 110, 1541; and Kashles et al., Mol. Cell. Biol., 1991, 11, 1454).

The present invention discloses a growth inhibition receptor-based strategy that targets activated oncoprotein receptors of the erbB tyrosine kinase family. Many systemic epithelial cancers express oncogenic forms of erbB receptors that can confer tumorigenic potential by overexpression,

ES 2 270 586 T3 mutation or co-expression with other elements of the erbB family. Since the enzymatic function of erbB receptor kinase is activated in dimerization or oligomerization, the present invention inhibits the catalytic activity of surface receptors by the formation of receptor complexes without kinase, thus reducing the tumorigenic effects of the translation product of erbB.

The present invention relates to amino acid molecules comprising a nucleotide sequence that encodes a protein that lacks tyrosine kinase activity and dimerizes with members of the erbB receptor family such as erbB1 (EGFR), erbB2 (p185), erbB3 and / or erbB4. The nucleic acid sequence can be DNA or RNA. The nucleic acid sequence can encode any protein that dimerizes with an erbB protein and lacks tyrosine kinase activity. The nucleic acid sequence preferably encodes rat or human erbB protein that can dimerize with erbB proteins and also lacks tyrosine kinase activity. According to one aspect of the invention, the nucleic acid molecule comprises a nucleic acid sequence that encodes a protein that lacks thyrose kinase activity and dimerizes with human EGFR or human p185. The nucleic acid sequence can be DNA or RNA. The nucleic acid sequence can encode any protein that dimerizes with human EGFR and / or p185 and that features tyrosine kinase activity. The nucleic acid sequence preferably encodes rat or human p185 species that can dimension with human p185 or human EGFR and that also lack tyrosine kinase activity.

Administration of nucleic acid molecules that encode proteins capable of forming dimers with erbB translation products but whose dimers do not show elevated tyrosine kinase activity eliminates the transformed genotype of erbB-associated tumors in an affected population of erbB-mediated tumors. Furthermore, the administration of said nucleic acid molecules inhibits neoplastic development in animals susceptible to developing erbB-associated tumors. For example, experiments have shown that p185 forms dimers with erbB translation products such as other p185 molecules or with epidermal growth factor receptor (EGFR) and that these dimers show high tyrosine kinase activity that is associated with the transformed phenotype. in cells that have such dimers. Administration of nucleic acid molecules that encode proteins capable of forming dimers with erbB translation products, such as other p185 molecules, or with EGFR but whose dimers do not show elevated tyrosine kinase activity, eliminates the transformed phenotype of neu-associated tumors. in a population affected by p185-mediated tumors. Furthermore, the administration of said nucleic acid molecules inhibits neoplastic development in animals susceptible to developing neu transformed tumors.

The appearance of mammalian tumor cells that express the translation product of an element of the erbB gene family on their surfaces and therefore have undergone erbB-mediated cellular transformation can be reversed or prevented by the administration of acid molecules. nucleic cells comprising sequences encoding proteins that form dimers with translation products of erbB genes but do not have tyrosine kinase activity. According to the invention these nucleic acid molecules are supplied in combination with delivery components, ie delivery vehicles or carriers, in order to facilitate the incorporation of said nucleic acid molecules into the cells of an animal. An effective amount of such combinations is administered to an individual identified as affected or susceptible to erbB-associated tumors.

The present invention provides nucleic acid molecules having a nucleotide sequence that encodes a protein lacking tyrosine kinase activity and that dimerizes with the translation product of a member of the erbB gene family. Nucleic acid molecules are provided in combination with delivery components such that upon administration of the combination, the nucleic acid molecule is delivered to cells of the individual. When provided as a pharmaceutical compound, the combination is useful for the treatment of individuals affected by erbB-mediated cellular transformations. This pharmaceutical compound may also be useful for the prevention of erbB-mediated cellular transformation, particularly in individuals susceptible to such transformation. The nucleic acid molecules of the invention can also be useful for producing specific translation products of an element of the erbB gene family in competent cells that can then be isolated and used in different immunoassays to detect the presence of specific antibodies for the translation product present in different body fluids.

The nucleic acid molecules of the invention are used in combination with a number of delivery components, such as recombinant viral expression vectors or other suitable delivery means, to affect their introduction and expression in compatible host cells. In general the viral vectors can be DNA viruses such as recombinant adenoviruses and recombinant vaccinia viruses or RNA viruses such as recombinant retroviruses. Other recombinant vectors comprise recombinant prokaryotes that can infect cells and express recombinant genes. In addition to recombinant vectors, other delivery components are also envisioned such as liposome encapsulation, lipofectin-mediated transfection, transferrin-mediated transfection, and other receptor mediated means. The invention is intended to include these other forms of expression vectors and other suitable delivery means that serve equivalent functions and are known in the art.

In a preferred embodiment of the present invention DNA is delivered to competent host cells by means of an adenovirus. One skilled in the art would readily understand this technique of delivering DNA to a host cell by such means. While the invention preferably includes adenovirus, the invention is intended to include any virus that serves equivalent functions. Examples of adenoviral vectors include

Recombinant ES 2 270 586 T3 those having the E1a region deleted and carrying a temperature-sensitive mutation in E2a (Engelhardt et al., Hum Gene Ther 5: 1217-1229, 1994). Other examples of recombinant adenoviral vectors useful for delivering a nucleic acid sequence according to the present invention are described in US Patent Nos. 5,756,283 and 5,707,618.

In another preferred embodiment of the present invention RNA is delivered to competent host cells by means of a retrovirus. One skilled in the art would readily understand this technique of delivering RNA to a host cell with such means. Any retrovirus that serves to express the protein encoded by the RNA is intended to be included in the present invention.

In another preferred embodiment of the present invention the nucleic acid is delivered through folate receptor means. The nucleic acid sequence to be delivered to a host cell is related to polylysine and the complex is delivered to the tumor cell via the folate receptor. US Patent 5,108,921 of April 28, 1992 to Low et al. Describes these delivery components.

In another preferred embodiment of the present invention, nucleic acid is delivered through the use of lipofectin-mediated dNa transfer. LipofectAMINE ™ Liposome Reagent (Life Technologies, Gaithersburg MD) is a commercially available liposome encapsulation reagent that can be used to encapsulate cells according to the manufacturer's instructions. Nucleic acid molecules encapsulated with lipofectAMINE ™ liposome reagent can be delivered to a host cell using liposome formulation delivery methods.

In another preferred embodiment of the present invention, nucleic acid is delivered by using lipid-mediated cationic transfer of DNA, such as that described in US Patent No. 5,703,055.

In another preferred embodiment of the present invention, the nucleic acid is delivered using liposome-mediated DNA transfer, as described in US Patent Nos. 4,235,871,4,241,046 and 4,394,448.

Pharmaceutical compounds according to the invention include delivery components in combination with nucleic acid molecules that further comprise pharmaceutically acceptable carriers or vehicles, such as, for example, saline. Any means that allows satisfactory delivery of nucleic acid can be used. Those skilled in the art will readily understand the multitude of pharmaceutically acceptable media that can be used in the present invention.

Pharmaceutical compounds can be formulated by one skilled in the art with selected compounds depending on the chosen mode of administration. Suitable pharmaceutical carriers are described in Remington's Pharmaceutical Sciences, A. Osol, a reference text in this field, which is incorporated herein by reference.

The pharmaceutical compounds of the present invention can be administered by any means that enable the active agent to reach the site of action of the agent in the body of a mammal. Pharmaceuticals can be administered parenterally, ie, intratumorally, intravenously, subcutaneously, intramuscularly. The preferred routes are intravenous and intratumoral administrations.

The dosage varies depending on known factors, such as the pharmacodynamic characteristics of the particular agent and its form and route of administration; recipient's age, health and weight; nature and extent of symptoms, type of simultaneous treatment, frequency of treatment and desired effect.

In some embodiments, the invention relates to the use of the compounds in the manufacture of a medicament for treating patients affected by human adenocarcinomas which are erbB-associated cancers, such as gastric, lung and pancreatic adenocarcinomas and human breast carcinomas and ovaries, as well as human breast and prostate cancers which are erbB-associated cancers. In some embodiments, the invention relates to the use of the compounds in the manufacture of a medicament for the prevention of these erbB-associated cancers in high-risk individuals. In some embodiments, the invention relates to the use of the compounds in the manufacture of a medicament for treating patients affected by glial tumor progression, particularly glioblastoma, the most malignant glial tumor. In some embodiments, the invention relates to methods for the prevention of these erbB-associated cancers in high-risk individuals.

In some embodiments, the invention relates to methods for treating patients afflicted with human epithelial malignancies, such as erythroid leukemia, fibrosarcoma, angiosarcoma, and melanoma. In some embodiments, the invention relates to methods for the prevention of these erbB-associated cancers in high-risk individuals.

According to some embodiments of the invention, the pharmaceuticals are administered locally to the tumor site. In some embodiments, the pharmaceuticals are administered directly to tumor cells and tissues immediately surrounding the tumor. In some embodiments, the pharmaceuticals are delivered to brain tumors, such as, for example, glioblastomas. In some realizations

ES 2 270 586 T3 nes, pharmaceutical compounds are delivered to brain tumors as part of surgical resection of the tumor. In certain embodiments, the pharmaceuticals are delivered to brain tumors using stereotactic surgical techniques.

The patient is treated by radiation or other chemotherapy in conjunction with the administration of pharmaceutical compounds according to the invention. Chemotherapy approaches include the administration of cytotoxic and / or cytostatic agents. Expression of nucleotide molecules, according to the invention, in erbB-associated tumors has been shown to render tumors radiosensitive. That is, tumors are more vulnerable to destruction by radiation during radiotherapy when the patient is treated with pharmaceutical compounds according to the invention. The use of multiple therapeutic approaches provides the patient with a more broad-based intervention. In some preferred embodiments, treatment with pharmaceuticals according to the invention is preceded by surgical intervention. In preferred embodiments, radiation therapy follows the administration of pharmaceutical compounds according to the invention. In preferred embodiments, radiation therapy using gamma ray radiation is facilitated after administration of compounds that render radiation resistant tumors radiation sensitive. Those skilled in the art can readily formulate an appropriate radiotherapeutic regimen. Carlos A Perez & Luther W Brady: Principles and Practice of Radiation Oncology, 2^to Edition JB Lippincott Co, Phila., 1992 describes radiation therapy protocols and parameters that can be used in the present invention. Regarding GBM (glioblastoma, most malignant glial brain tumor), Simpson WJ et al .: Influence of location and extent of surgical resection on survival of patients with glioblastoma multiforms: Results of three consecutive Radiation Therapy Oncology Group (RTOG) clinical trials. Int J Radiat Oncol Biol Phys 26: 239-244,1993, describes clinical protocols useful in the present invention. Similarly, Borgelt et al., The palliation of brain metastases: Final results of the first two studies of the Radiation Therapy Oncology Group. Int J Radiat Oncol Biol Phys 6: 1-9, 1980 describes clinical protocols useful in the methods of the present invention.

According to some embodiments, variants of the p185neu / erbB-2 receptor are used since this receptor has been found to be the preferred partner for heterodimer assembly for all kinases of the erbB family, including erbB 1 / Growth Factor Receptor Epidermal (EGFR), erbB3, and erbB4. According to a preferred embodiment of the invention, a kinase-deficient form of p185neu for use in treating erbB-expressing human tumors is delivered by recombinant adenovirus particles as gene therapy to treat local, residual, rather than mass disease. advanced. The application of surgical techniques will be used both for local administration and for the reduction in the characteristic of mass disease of solid tumors. Multiple human cancers expressing erbB receptor combinations may be targeted for this receptor-based growth inhibition strategy.

Inhibition mediated by the introduction of mutant p185neu receptors causes synergistic growth inhibition when combined with conventional cytotoxic agents, such as gamma irradiation. The present invention can be used to treat many solid epithelial tumors since the invention complements the use of established treatment modalities.

The kinase-deficient T691 stop form of p185neu is more effective in achieving inhibition of cell growth and transformation than the N691 stop form of p185neu in human brain tumor cells expressing high levels of EGFR. T691stop neu also inhibits constitutive signaling from a specifically expressed mutant EGFR in many epithelial tumors, including malignant human gliomas, and has been shown to reduce oncogenic full-length p185neu kinase activity in primary mammalian cancer cells. T691stop neu contains the rat neu transmembrane point mutation that changes amino acid at position 664 from valine to glutamine, resulting in a change in the tendency of this receptor to form dimer and oligomeric complexes.

The mutant T691 stop neu cDNA has been subcloned into recombinant adenoviral constructs. A vector core has the E1a region deleted and carries a temperature-sensitive mutation in E2a (Engelhardt et al., Hum. Gene Ther. 5: 1217-1229, 1994). Recombinant derivation was accomplished in two steps. First, the inventors subcloned T691stop neu into an expression plasmid (pAd.CMV.link according to Example 6) containing adenoviral sequences necessary for recombinant viral production generating pAd.CMV.T691 stop. Plaque-purified recombinant adenoviral particles were generated expressing T691 stop neu. Our laboratory confirmed functional expression of T691 stop mutant neu proteins after infection of human cancer cells with recombinant adenoviral particles by flow cytometric analysis. Therefore, a pure adenoviral recombinant expressing high levels of T691 stop neu has been isolated in human cancer cells and can be used as an anticancer reagent. Adenoviral recombinants expressing T691 stop neu with alterations in the viral nucleus have been designed, making the administration of viral recombinants more suitable for human application. Specifically, adenoviral recombinants have been made containing the E1a deletion with an additional deletion of the E4 region (See Example 7).

In accordance with certain preferred embodiments, the present invention provides an anticancer gene therapy treatment for treating local, residual disease as a therapeutic adjuvant in combination with pre-existing treatments. Supply is local at the time of surgery, most likely after resection of the entire major disease. For primary malignant brain tumors, gene therapy is provided at the time of tumor resection or in certain cases by stereotactic implantation, a precise and normal delivery method, or local resection. The expression of T691 stop neu is not cytocidal or toxic to non-dividing cells. Do not

ES 2 270 586 T3 however, inhibition of erbB receptor signaling by mutant T691 stop neu receptors renders the cell population of a dividing tumor more sensitive to radiation-induced apoptotic cell death. The T691 stop neu form also includes a higher fraction of growth arrest in tumor cells treated with gamma irradiation. Inhibition of factor-mediated growth signaling has been correlated with increased sensitivity to standard anticancer reagents in a number of systems.

Residual, local disease can be treated according to the present invention with a receptor-based growth inhibition strategy that deactivates signaling through oncoproteins of the erbB family. This gene therapy strategy is part of a combined treatment to achieve synergistic growth inhibition by direct mechanisms (inhibition of receptor signaling) and indirect mechanisms (that is, by making cancer cells more sensitive to simultaneous treatment with pre-existing agents).

Many tumors are notable for overexpression and / or mutation of erbB receptors, including EGFR (erbB1), p185neu / erbB2, erbB3, and / or erbB4. In many cases, the co-expression of elements of the erbB family results in signaling that is synergistic and contributes to cell transformation. Tumors notable for overexpression of p185neu / erbB2 include cancers of the breast, ovary, lung, and pancreas. Tumors notable for EGFR overexpression include primary glial brain tumors and prostate cancers. Gene delivery of truncated p185neu forms, ie the T691 stop neu cDNAs or proteins, provides a rational strategy for the treatment of residual cancers, of local disease, in these human cancers.

The present invention is particularly useful for the treatment of patients having glial brain tumors, ie tumors characterized by glioblastomas. These cells express a mutant form of EGFR typically associated with tumorigenicity. The present invention has been found to be particularly useful in treating such patients.

In some embodiments, the nucleic acid sequences encoding the different species of rat p185 are constructed from c-neu cDNA according to the procedure outlined in the examples. Nucleic acid sequences encoding mutated rat p185 and WT species are prepared in this way. The nucleotide sequences of the prepared p185 constructs are verified by DNA sequencing. One skilled in the art would readily find methods of constructing such nucleic acid constructs.

After preparing such constructs, they are transfected into suitable host cells within which they are expressed. One of ordinary skill in the art would readily understand the vast number of host cells that could be used. Within these suitable host cells the ability of the p185 species, produced from the prepared nucleotide construct, to dimerize with p185 or EGFR has been examined. This examination may include immunoblotting, flow cytometry, SDS-PAGE analysis, as well as other techniques that are well known to those of skill in the art. Furthermore, the tyrosine kinase activity of the p185 species can also be evaluated. It is also within the skill of one of skill in the art to assess tyrosine kinase activity by a number of different techniques.

Once the lack of tk phenotype of the p185 species has been established, as well as the ability to dimerize with EGFR or p185, the nucleic acid sequence encoding the p185 species can be subcloned into an expression vector suitable for transfection into cells. human. Alternatively, the nucleic acid sequence can be used in combination with another delivery medium, as noted above.

According to one aspect of the invention, the nucleic acid molecule comprises a nucleic acid sequence that encodes a protein that lacks tyrosine kinase activity and dimerizes with human EGFR or human p185. The nucleic acid sequence can be either DNA or RNA. The nucleic acid sequence can encode any protein that dimerizes with human EGFR and / or p185 and lacks tyrosine kinase activity. The nucleic acid sequence preferably encodes rat or human p185 species which can dimerize with human p185 or EGFR and which also lacks tyrosine kinase activity. In some embodiments, the nucleic acid sequence encodes a protein comprising the ectodomain region of rat or human p185 species. These proteins dimerize with erbB proteins, such as human p185 or EGFR, and also lack tyrosine kinase activity.

In some preferred embodiments, the constructs include the rat neu transmembrane region. In some preferred embodiments, the rat neu transmembrane region contains a val-> glu mutation at amino acid 664. The rat neu transmembrane region without the mutation at amino acid 664 is designated the "N" form and the rat neu transmembrane region with the mutation at amino acid 664 it is designated "T" form.

In some preferred embodiments of the present invention, the nucleic acid sequence encodes rat p185 truncation species. The present invention includes any truncated species of rat p185 comprising N-terminal or C-terminal deletions that dimerize with human p185 or EGFR and lacking tyrosine kinase activity. Furthermore, truncated species comprising substituted amino acids can also be effective. However, the truncate species must be able to dimerize with human p185 or EGFR. Therefore, any part of p185 capable of dimerizing with human p185 or EGFR, while simultaneously having a tk phenotype^-, is included too. Preferably, the nucleic acid sequence encodes a protein consisting of rat p185 amino acid residues from about 1-690 to about 1-740. In some preferred embodiments, they comprise the N form of the transmembrane region while in others, the T form is present.

IS 2 270 586 T3

In another preferred embodiment of the present invention, the nucleic acid sequence encodes rat p185 species that lack tyrosine kinase activity by substituting or deleting parts of amino acids, specifically those of the region of the molecule responsible for the tyrosine kinase activity. The present invention includes any rat p185 tk species, comprising substitution or deletion of amino acids responsible for tk activity, wherein the species also dimerizes with human p185 or EGFR. Furthermore, these species comprising amino acids substituted outside of the associated tk sequences may also be effective. In some preferred embodiments, they comprise the N-form of the transmembrane region, while in others the T-form is present.

Rat p185 positions 753-758 comprise the critical lysine residue that directly binds the ATP molecule that is the phosphate donor in the tyrosine kinase reaction (Moller et al., FEBS Lett., 1985,186,1; and Sternberg et al., FEBS Lett, 1984, 175, 387). Lys⁷⁵⁷ is 15 amino acid residues downstream of a conserved motif that is also found in nucleotide-binding proteins without kinase activity (Wierenga et al., Nature, 1983, 302, 842). Glycine residues are believed to form a hydrophobic pocket around the critical lysine residue that directly binds the ATP molecule (Moller et al., FEBS Lett, 1985, 186, 1; and Sternberg et al., FEBS Lett 1984, 175, 387 ). Therefore, any p185 species that comprises a break in the ATP-binding domain or surrounding region, where ATP no longer binds to the critical Lys residue, is also included. However, these species must also dimerize with human p185 or EGFR. Preferably, the nucleic acid sequence encodes a protein having the amino acid sequence of rat p185, defined in GENEBANK Acession No. X03362, and Bargmann et al., (1986) Nature 319, 226-230, MEDLINE Identifier: 86118662; and Lofts, et al. (1993) Oncogene 8, 2813-2820; wherein this amino acid sequence contains a substitution or deletion, or any combination thereof, from about position 753 to about position 758, such that said substitution does not comprise a lysine residue. In some preferred embodiments they comprise the N form of the transmembrane region, while in others the T form is present.

In another preferred embodiment of the present invention, the nucleic acid sequence encodes rat p185, while the amino acid sequence contains a substitution or deletion at position 757. This substitution or deletion specifically removes the critical Lys residue at this position. Therefore, ATP can no longer bind this molecule resulting in a phenophype tk^-: In some preferred embodiments, it comprises the N form of the transmembrane region, while in others the T form is present.

In another preferred embodiment of the present invention, the nucleic acid sequence encodes human p185 truncation species. The present invention includes any species of human p185 truncate comprising N-terminal or C-terminal deletions that dimerize with human p185 or EGFR and lacking tyrosine kinase activity. Furthermore, truncation species comprising substituted amino acids can also be effective. However, the truncation species must be capable of dimerizing with human p185 or EGFR. Therefore, any part of human p185 capable of dimerizing with human p185 or EGFR, simultaneously having a tk phenotype^-, is included in this invention. Preferably, the nucleic acid sequence encodes a protein consisting of human p185 amino acid residues from about 1-646 to about 1-704. In some embodiments, the nucleic acid sequence encodes a protein consisting of human p185 amino acid residues from about 1-653.

In another preferred embodiment of the present invention, the nucleic acid sequence encodes human p185 species that lack tyrosine kinase activity by substituting or deleting parts of amino acids, specifically those found within the region of the molecule responsible. of tyrosine kinase activity. The present invention includes any species of tk^- human p185, comprising substitution or deletion of amino acids responsible for tk activity, while the species also dimerizes with human p185 or EGFR. Furthermore, such species that comprise amino acids substituted outside of the tk-associated sequences may also be effective.

Positions 749-754 of human p185 comprise the critical lysine residue that binds directly to the ATP molecule that is a phosphate donor in the tyrosine kinase reaction. Any p185 species that comprises a break in the ATP binding domain or surrounding region, in which ATP no longer binds to the critical Lys residue, is included in the invention. However, these species must also dimerize with human p185 or EGFR. Preferably, the nucleic acid sequence encodes a protein having the amino acid sequence of human p185 as defined in the accession number GENEBANK X03363 which is incorporated into the current disclosure by reference and the publication Yamamoto et al., (1986) Nature 319, 230-234, MEDLINE identifier: 86118663, and Papewalls et al., (1991) Nucleic acids Res. 19, 5452-5452, MEDLINE identifier: 92020265, in which this amino acid sequence contains a substitution or deletion or any combination thereof, from position 749 to position 754 approximately, such that said substitution does not comprise a residue of lysine.

In another preferred embodiment of the present invention, the nucleic acid sequence encodes human p185 in which the amino acid sequence contains a substitution or deletion at position 753. This substitution or deletion specifically removes the critical lys residue at this position. Therefore ATP can no longer bind this molecule resulting in a tk phenotype.^-.

In some embodiments, the nucleic acid encodes a human EGFR protein, a human p185 protein, a human erbB3-derived protein, or a human erbB4-derived protein. In some embodiments, the acid

Nucleic ES 2 270 586 T3 encodes a fusion protein. The fusion protein is encoded by chimeric sequences derived from human and non-human sequences, particularly rat ones.

In some embodiments, the nucleic acid is selected from the group consisting of:

- a truncated rat neu with a stop codon at amino acid 691 (stop construct N691);

- a truncated rat neu with a stop codon at amino acid 691 and a val-> glu mutation at amino acid 664 (stop construct T691);

- a chimeric p185 gene comprising the human p185c-erbB2 ectodomain linked to rat neu transmenbrane with a stop codon at amino acid 691 (stop N691 construct);

- a chimeric p185 gene comprising human p185c-erbB2 ectodomain linked to rat neu transmenbrane with a stop codon at amino acid 691 and a val-> glu mutation at amino acid 664 (stop construct T691);

- a chimeric p185 gene comprising human EGFR ectodomain linked to rat neu transmenbrane with a stop codon at amino acid 691 (stop construct N691);

- a chimeric p185 gene comprising human EGFR ectodomain linked to rat neu transmenbrane with a stop codon at amino acid 691 and val-> glu mutation at amino acid 664 (stop construct T691);

- a chimeric p185 gene comprising erbB3 ectodomain linked to rat neu transmembrane with a stop codon at amino acid 691 (stop construct N691);

- a chimeric p185 gene comprising human erbB3 ectodomain linked to rat neu transmenbrane with a stop codon at amino acid 691 and val-> glu mutation at amino acid 664 (stop construct T691);

- a chimeric p185 gene comprising erbB4 ectodomain linked to rat neu transmenbrane with a stop codon at amino acid 691 (stop construct N691); Y

a chimeric p185 gene comprising human erbB4 ectodomain linked to rat neu transmembrane with a stop codon at amino acid 691 and val-> glu mutation at amino acid 664 (stop construct T691).

Unlike replicating non-transformed cells that can be killed by exposure to therapeutic radiation, tumor cells are resistant to the induction of cell death by radiation. It has now been discovered that by altering the multimeric pools that produce high kinase activity associated with the transformed phenotype of a tumor cell, this tumor cell, which is ordinarily resistant to radiation-induced cell death, becomes sensitive to radiation. Accordingly, one aspect of the present invention provides methods for rendering radiation resistant tumor cells sensitive to radiation. The present invention relates to the use of a compound that alters the kinase activity associated with the multimeric receptor set in the manufacture of a medicament to be administered in combination with gamma radiation to treat an individual having tumor cells that has multimeric receptor sets that they provide kinase activity associated with a transformed phenotype.

There are several known receptor assemblages that in tumor cells show elevated kinase activity that is associated with the transformed phonotype. Members of the erbB family of receptors are known to form multimeric clusters that result in elevated tyrosine kinase activity in tumor cells. Multimeric pools comprising members of the erbB family include erbB homodimers, as well as erbB heterodimers comprising monomer components of different members of the erbB family. Platelet-derived growth factor receptor (PDGFR) multimeric receptor assemblies also show elevated kinase activity that is associated with the transformed phenotype.

In accordance with another aspect of the present invention, erbB-mediated dimer formation in tumor cells is disrupted to make such cells more susceptible to cell destruction using radiation. Accordingly, combination therapies are envisioned comprising first administering to an individual a compound comprising an active agent that results in erbB dimerization interference followed by exposure of the patient to therapeutic amounts of radiation. In accordance with these aspects of the invention, compounds are envisioned for use in treating individuals affected by a tumor mediated by the erbB protein. A compound that inhibits elevated tyrosine kinase activity that results from dimerization of erbB proteins in a tumor cell is administered to the individual, followed by, after a time sufficient for the compound to inhibit tyrosine kinase activity associated with dimerization of tumor cell erbB proteins, exposing the individual to a therapeutically effective amount of anticancer radiation.

IS 2 270 586 T3

In some tumor cells, the p185 translation product of the c-erbB2 gene is overexpressed and forms homodimers and heterodimers with other members of the erbB family. This dimerization of overexpressed p185 leads to elevated tyrosine kinase activities which is associated with the transformed phenotype. Disruption of tyrosine kinase activity, such as by inhibiting dimer formation between monomer components, results in a cytostatic effect on tumor cells. It has now been discovered that the alteration also transforms previously radiation resistant tumor cells into radiation sensitive ones.

Similarly, a mutant form of EGFR (AEGFR) that is ligand-independent is expressed in some tumor cells. AEGFR forms homodimers and heterodimers with wild-type EGFRs and other elements of the erbB family. This dimerization of AEGFR leads to elevated tyrosine kinase activities that is associated with the transformed phenotype. Alteration of tyrosine kinase activity, such as inhibiting dimer formation between monomer components, results in a cytostatic effect on tumor cells. It has now been discovered that this alteration also makes previously radiation resistant tumor cells sensitive to radiation.

In some embodiments, the erbB protein-mediated tumor is a brain cancer tumor. In some preferred embodiments, the erbB protein mediated tumor is a glial tumor. In some preferred embodiments, the erbB protein-mediated tumor is a glioblastoma. In some embodiments, the erbB protein-mediated tumor is a breast cancer tumor. In some embodiments, the erbB protein mediated tumor is an ovarian cancer tumor. In some embodiments, the erbB protein-mediated tumor is a pancreatic cancer tumor.

In some embodiments, the kinase activity associated with the multimeric receptor set is altered by administering to the individual a compound comprising an active agent that interacts with a monomer component of the set, and in so doing prevents dimerization by physically altering the monomer so that it is less thermodynamically arranged to form the whole. These physical alterations can be, for example, conformational, steric and / or electrostatic changes that transform the monomer into a state less favorable for dimer formation. Examples of active agents that physically alter monomer, antibodies, proteins, peptides, and non-protein molecules are included.

As used in this description, the term "antibody" is intended to designate antibodies and also antibody fragments, such as FAb and F (Ab) fragments.₂. The antibodies can be, in some preferred embodiments, monoclonal antibodies or humanized antibodies. Antibodies against p185 are described in US Patent Nos. 5,677,171 and 5,705,157 which also describe antibodies against EGFR. US Patent No. 5,470,571 also describes antibodies against EGFR.

In some embodiments, antibody-mimicking peptides are arranged to inhibit the formation of the multimeric pool and the high kinase activity associated with such formation. For example, peptides are designed that have sequences corresponding to CDR regions from antibodies. Methods for the preparation of such peptides are also described in WO 95/34312. Imitations of antibody peptides against p185 are described in US Patent No. 5,663,144.

According to certain embodiments of the invention, the compound that is administered to the individual to alter the kinase activity associated with the multimeric receptor set comprises an active agent that is a nucleic acid molecule that encodes an interacting kinase-deficient protein or peptide. with a monomer component of said assembly to prevent them from interacting with another component of the assembly. That is, the nucleic acid molecule encodes a kinase-deficient protein that competes with endogenous cell proteins to form multimeric complexes. The complexes formed between the kinase-deficient protein and the endogenous proteins of the cell do not provide high kinase activity. Therefore, kinase-deficient proteins act as adjuvants to bind endogenous proteins and thus prevent the formation of multimeric complexes with kinase activity. Examples of such nucleic acid molecules are described in US Patent No. 5,837,523.

According to certain methods, the compound that is administered to a patient comprising a compound that competitively interacts with an erbB protein in a tumor cell to competitively inhibit dimer formation with another erbB protein for more reduced dimerization of erbB proteins is a nucleic acid molecule that encodes a protein. The protein blocks dimer formation by competitively interacting with the erbB protein. In some embodiments, the protein interacts with the transmembrane region of said erbB protein. In some embodiments, the protein that interacts with the transmembrane region of an erbB protein comprises a rat neu transmembrane region with a val to glu mutation at amino acid 664. In some embodiments, the protein interacts with the ectodomain region of an erbB protein. In some such embodiments, the protein that interacts with the ectodomain region of an erbB protein comprises a p185 ectodomain. In some preferred embodiments, the nucleic acid molecule encoding the proteins is a viral genome. In some preferred embodiments, it is a recombinant adenovirus genome. According to some embodiments, the nucleic acid molecule comprises coding sequences operably linked to regulatory elements for translation in cells of said individual. In some embodiments, the coding sequence comprises sequences selected from the group consisting of: truncated rat neu gene with a stop codon at amino acid 691; a truncated rat neu gene with a stop codon at amino acid 691 and a val-> glu mutation at amino acid 664; a chimeric p185 gene comprising human ectodomain p185c-human erbB2 linked to rat neu transmembrane with a stop codon at the amino acid

IS 2 270 586 T3

691; a chimeric p185 gene comprising p185 c-erbB2 ectodomain linked to rat neu transmembrane with a stop codon at amino acid 691 and val-> glu mutation at amino acid 664; a chimeric p185 gene comprising rat neu transmembrane-linked human EGFR ectodomain with a stop codon at amino acid 691; a chimeric p185 gene comprising rat neu transmembrane-linked human EGFR ectodomain with a stop codon at amino acid 691 and val-> glu mutation at amino acid 664; a chimeric p185 gene comprising human erbB3 ectodomain, linked to rat neu transmembrane with a stop codon at amino acid 691; a chimeric p185 gene comprising rat neu transmembrane-linked human erbB3 ectodomain with a stop codon at amino acid 691 and val-> glu mutation at amino acid 664; a chimeric p185 gene comprising a rat neu transmembrane-linked human erbB4 ectodomain with a stop codon at amino acid 691; and a chimeric p185 gene comprising rat neu transmembrane-linked human erbB4 ectodomain with a stop codon at amino acid 691 and a val-> glu mutation at amino acid 664.

In some embodiments, the nucleic acid molecules encode peptides that interact with transmembrane regions of erbB proteins and thereby prevent dimerization of erbB proteins. Examples of such nucleic acid molecules are disclosed in Lofts, et al. 1993 Oncogene 8: 2813-2820.

According to certain preferred embodiments, the compound comprising an active agent that causes disruption of kinase activity associated with the multimeric receptor is administered by any route of administration that can be used to deliver the agent to the tumor. In some embodiments, the compound is administered by intravenous, intraarterial, intramuscular, intradermal, subcutaneous, parenteral, or intratumoral administration. According to certain preferred embodiments, the individual has undergone surgery to remove tumor masses prior to compound administration.

In accordance with aspects of the present invention, after administering the compound comprising an active agent that causes alteration of kinase activity associated with the multimeric receptor set, the individual is exposed to a therapeutic amount of gamma radiation. Radiation therapy can begin at any time, after a certain period of time has elapsed for the active agent to cause the disruption of kinase activity associated with the multimeric receptor set. In general, the individual is exposed to radiation in some cases 1-10 minutes later, in some cases 1-10 hours later, and in other cases 24-72 hours after administration of the active agent. In some cases, radiation is delivered in a single dose, while in other embodiments multiple doses are administered over several hours, days, and / or weeks. The active agent makes radiation resistant tumor cells sensitive to radiation. Therefore, once the active agent inhibits kinase activity, radiation exposure can take place. Gamma ray radiation is delivered according to standard radiotherapeutic protocols using normal dosages and regimens. Administration of the active agent makes radiation more effective in eradicating tumor cells.

In accordance with aspects of the present invention, after administration of the compound comprising an active agent that causes disruption of kinase activity associated with the multimeric receptor set, the individual receives administration of a cytotoxic chemotherapeutic agent in addition to, or instead, exposure to a therapeutic amount of gamma radiation. As in the case of radiation therapy, chemotherapy can begin at any time after a sufficient period of time has elapsed for the active agent to cause the disruption of kinase activity associated with the multimeric receptor set. In general, the individual receives chemotherapeutic administration in some cases 1-10 minutes, in others 1-10 hours, and in other cases 24-72 hours after administration of the kinase inhibitor active agent. In some cases, the chemotherapeutic agent is used in a single dose, while in other embodiments, multiple doses are administered over several hours, days, and / or weeks. The active agent makes tumor cells more sensitive to cytotoxic agents. Therefore, once the active agent inhibits kinase activity, administration of chemotherapeutic agents can take place. Chemotherapeutic agents are delivered according to standard radiotherapeutic protocols using standard agents, dosages, and regimens. In some embodiments, the chemotherapeutic agent is selected from the group consisting of: cisplatin, doxirubicin, danurubicin, tamoxifen, taxol, and methotrexate. In some embodiments, both chemotherapy and radiation treatments are used after administration of the active agent. In these embodiments, standard combinations of the two therapeutic modalities are used in conjunction with administration of the kinase inhibitor active agent.

The present invention is further illustrated by the following examples.

Examples Example 1

Construction of Mutants, Expression Vectors and Creation of Cell Lines

Detailed methods for the construction of mutant p185 species, expression vectors, and cell lines have been previously described (Qian et al., Proc. Natl. Acad Sci. USA, 1994, 91, 1500; and Weiner et al., Oncogene, 1989, 4, 1175).

IS 2 270 586 T3

Construction of mutant N757

The ATP-binding mutant Nneu K757M (N757) was derived from pSV2TneuK757M (Weiner et al., Oncogene, 1989, 4, 1175) by subcloning techniques. This construct was prepared by site-directed mutagenesis to substitute a Met for Lys.⁷⁵⁷. Those skilled in the art will readily understand the preparation of this mutant by site-directed mutagenesis. Briefly, an Xba 1 fragment of pSV2neuT corresponding to a 1.2 kb band spanning the probable ATP binding site of the published nucleotide sequence was cloned into M13Mp18 and transfected into E. Coli strain CJ236 (dot^-, ung) pUC13 so that instead of the HindIII of the polylinker it passed to the 5 'end of the inserted sequences. Mutagenesis was carried out as described using a primer in which the AAG codon, which codes for Lys, was replaced by the Aug codon corresponding to Met (Bargmann et al., Nature, 1986, 319, 226). Point mutations created in this way were verified by DNA sequencing. The plasmid carrying the new mutation was fractionated with Xba1 which released the original fragment. This fragment was isolated by standard techniques, known to those skilled in the art, and ligated again into pSV2neu to regenerate the oncogenic expression vector p185neu except that the vector contained the substitution of Met for Lys at amino acid position 757 (clone M757).

Construction of the mutant N691stop

The carboxy terminal 591 of the stop amino acid deletion mutant N691 was derived from pSV2Nneu (Bargmann et al., Nature, 1986, 319, 226) by substituting a stop codon for the normal codon Thr⁶⁹¹ by site-directed mutagenesis.

Ndx construction

The carboxy terminal 541 amino acid deletion mutant Ndx was derived from c-neu cDNA by deletion of an XbaI fragment and insertion of a stop codon for the normal codon at position 741 via site-directed mutagenesis.

Construction of expression vectors

For expression vectors, fragments containing mouse dihydrofolate reductase (DHFR) cDNA from pSV2DHFR and bacterial neomycin phosphotransferase resistant gene (neo ') from pSV2NEO (Southern et al., J. Mol. Appl. Genet., 1982, 1, 327 ) were subcloned into pSV2Nneu so that a combined 14.8 kb DHFR, neo^r, and Nneu cDNA. The wt or mutated neu fragments were isolated and added back into an expression vector pSV2 neo^r/ dhfr / Nneu. All these CDNAs were under the control of the simian virus 40 (SV40) pre-promoter. A unit of gene encoding the bacterial hygromycin resistance gene (Hyg^r) under the control of herpes simplex virus thymidine kinase promoter was isolated from pHyg and replaced by a neo gene fragment^r in pEGFR1 (Gorman et al., J. Cell. Biochem., 1988, 12A, Suppl., C219) to generate another combined expression vector, pEGFR / Hyg^r. The human EGFR cDNA was under the control of the Sra promoter, an efficient transcriptional control element containing the SV40 pre-promoter and the R-U5 segment of human T-cell leukemia virus type 1 with long terminal repeat (Takebe and others, Mol. Cell. Biol., 1988, 8, 466). Transfection and maintenance of cell lines

The pEGFR / Hyg construct^r it was first transfected into NR6 cells (Pruss et al., Proc. Natl. Acad. Sci. USA, 1977, 74, 3918, which is incorporated into the current disclosure by reference) by calcium phosphate precipitation. After 3 weeks of hygromycin selection (35 pg / ml), the EGFR expression of the resulting colonies was identified by anti-EGFR immunoblotting. Cells expressing EGFR were further cloned by limiting dilution prior to second-pass transfection with neu cDNA expression vectors. EGFR expression cells, designated NE91, together with NR6 cells were transfected with pSV2neo^r/ dhfr / neu encoding neu wt or mutant proteins and selected with G418. Clones expressing Neu in NR6 cells and NE91 cells were selected by flow cytometric assay with staining of the 7.16.4 anti-neu monoclonal antibody (Drebin et al., Cell, 1985, 41, 695) and were designated NR6 Neu and NE Neu respectively. . These double transfected clones (expressing Neu and EGFR) single (expressing only Neu) containing DHFR were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 5% fetal bovine serum, G418 (0.3 mg / ml) and hygromycin ( 15 pg / ml). Neu amplification was achieved by progressively increasing dosages (0.3-1.0 pM) of methotrexate for a few months in order to raise the level of receptor expression.

Flow cytometry

Cells are removed from EDTA tissue culture dishes (Versene, MA Bioproducts) with buffer and washed twice in FACS medium (Hank's balanced salt solution (Gibco) supplemented with 2% fetal calf serum, 0.2 % sodium azide and 10 mM HEPES). Incubated 1 x 10⁶ cells in 0.1 ml FACS medium with monoclonal anti-neu 7.16.4 antibody (Drebin et al., Cell, 1985, 41, 695) or isotype-compensated irrelevant control antibody for 1 hour at 4 ° C. The cells were washed twice with 2.5 ml of FACS medium. The cell pellet was resuspended and the cells were incubated with 0.1 ml of rabbit, goat anti-mouse IgG, FITC conjugate (antibody heavy and light chain reactive, Tago) diluted 1:50 in FACS medium for 1 hour at 4 ° C. Cells were washed twice and analyzed on a Becton Dickenson FACS IV apparatus.

IS 2 270 586 T3

Example 2

Aci / v / tyrosine kinase

Membrane purification

Cells were lysed by the combination of freeze-thaw and Doutice homogenization, as described in Gaulton et al., J. Immunol., 1986,7, 2470. The nuclear fraction was removed by centrifugation at 2000 xg for 5 minutes. The 2000 xg supernatant fraction was then centrifuged at 25,000 xg for 30 minutes at 4 ° C and the 25,000 xg supernatant was retained as a cytosol fraction. The pellet was redissolved in 1.5 ml of membrane buffer (40 mM NaCl, 0.1 mM EDTA, 20 mM HEPES (pH 6.8), 2 mM PMSF, and 5 mM Na pyrophosphate), then applied in layers on a sucrose solution (20% -37%) in membrane buffer and centrifuged at 22,000 rpm for 18 hours for 2 ° C using a Beckman SW50.1 rotor. The membrane rich interface was removed in a total volume of 1 ml, diluted with 10 ml of membrane buffer and recentrifuged at 40,000 rpm for 60 minutes using a SW40.1 rotor exactly the same as described in Zick et al., Biochem. Biophys. Res. Commun., 1984, 119, 6. The resulting pellet containing purified membrane fragments was redissolved in 100 µl of kinase buffer (see below) for every 10⁷ original cells. Membrane proteins were quantified using a BioRad protein assay kit and were stored at -80 ° C until assay.

Tyrosine kinase activity in membranes

Membrane concentrations were determined by the Bradford method described in Gaulton et al., J. Immunol., 1986, 7, 2470.

Membrane dilutions were incubated in quadruple in the presence or absence of tyrosine, containing synthetic polypeptide, as a specific indicator of tyrosine phosphorylation. Kinase reaction buffer (50 μl of 0.1 M Hepes pH 7.3, 10 mM MgCl₂, 5 mM MnCl₂, 50 pM Na₃VO₄ was incubated in the presence of ATP (1 pCi of gamma rays [³²P] ATP; Amersham) for 5 minutes at room temperature. Reactions were stopped by addition of 5 mM EDTA (final concentration) followed immediately by TCA immunoprecipitation on glass fiber filters (Whatman GF / A). Filters were washed thoroughly with TCA followed by ether, air dried, dipped in scintillation cocktail (Biofluor) and beta emissions were determined. Quadruple wells tested in the absence of chrosine-containing substrate were subtracted from wells containing tyrosine substrates.

The membrane proteins were incubated with the glutamic acid-tyrosine random polymer (4: 1) polyglu: tyr, PGT) as a substrate for tyrosine phosphorylation, as described in ZicK et al., Biophys. Res. Commun., 1984, 119, 6. Briefly, the membrane proteins were incubated in 50 µl of 10 mM HEPES pH 7.2, containing 10 mM MgCl₂, 100 pM Na₃ VO₄, and 150 pM (10 pCi) [³²P] ATP for 15 minutes at room temperature in the presence (specific) or absence (background) of polyglu: tyr substrate at 2.5 mg / ml. Reactions were stopped by the addition of EDTA to 50 mM final concentration and excess ATP cold and samples were applied on Whatman glass fiber filter paper. Filters were washed 3 times with ice-cold 10% TCA, containing 10 mM pyrophosphate and 1 mM ATP followed by acetate 1 time. The samples were then dried and counted in BioFlur (NEN). For immunoprecipitation of phosphotyrosine-containing membrane proteins, 50 pG of purified membranes were incubated in kinase buffer, as described above, for 15 minutes. After labeling, the samples were solubilized in lysis buffer supplemented with 5 mM EDTA, pre-cleared and immune precipitated with 2 µl of MA-2G8A6 + protein A agarose ascites. The MA-2G8 antibody specifically precipitates phosphotyrosine-labeled polypeptides, as described in Daniel et al., Proc. Natl. Acad. Sci. USA, 1985, 82, 2084.

Example 3

Dimerization with p185 or EGFR

EGFR and p185 heterodimers were detected by immunoprecipitation of anti-specific receptor antibodies and immunoblotting after EGF and chemical cross-linker treatment. The physical association of EGFR and kinase-deficient p185 proteins was examined in this way.

Chemical crosslinking test

Cells were grown overnight in 10 cm Petri dishes, incubated with or without EGF (GIBCO / BRL) at 37 ° C for 10-15 minutes, and washed twice with cold phosphate buffered saline (PBS). . Three ml of PBS containing 2 mM bis (sulfosuccinimidyl) suberate (BS³) or 3,3'-dithiobis (sulfosuccinimidylpropionate) (DTSSP) (Pierce) were added and incubated at 18 ° C for 30 minutes with occasional rocking of the plates. After tempering the crosslinking reaction mixture with buffer containing 10 mM Tris-HCl, 0.9% NaCl, and 0.1 M glycine, the cells were washed twice with cold PBS and solubilized with PI / RIPA buffer (Wada et al. Cell, 1990, 61, 1339).

IS 2 270 586 T3

Marking and immunoprecipitation

All reagents were obtained from Sigma if not indicated otherwise. For marking [³²P] were applied in the form of 1 x 10 layers⁶ cells and cultured for 24 hours and then incubated with [³²Inorganic P] (Amersham) at 0.5 mCi / ml in 5% phosphate-free FCS / RPMI for 6 hours. After labeling, cells were washed with cold phosphate buffered saline containing 400 pM EDTA, 10 mM sodium fluoride, 10 mM sodium pyrophosphate, and 400 pM sodium orthovanadate and lysed in lysis buffer (1 % NP40, 0.1% deoxycholate, 0.1% SdS, 0.15 M NaCl, 0.01 M sodium phosphate pH 7.4, 1% Trasylol, 1 mM PMSF, 2 mM EDTA, 10 mM sodium fluoride , 10 mM sodium pyrophosphate, 400 μ / M Na₃ VO₄, 10 mM iodine acetamide and 1 mMATP) for 30 minutes. The preclear leftovers were subjected to immunoprecipitation with monoclonal antibody 7.16.4, or rabbit antisera, recognizing human and rat DBW-2 neu proteins (Kokai et al., Proc. Natl. Acad Sci, USA, 1988, 84 8498). Immunoprecipitates were boiled in Laemmli sample buffer and analyzed on 8% SDS-PAGE (Laemmli, Nature, 1970, 227, 680). The dried gels were exposed to pre-treated film ("prefogged") at -70 ° C. Gel densitometer scans were carried out on a Hoefer GS300 scanning densitometer. Relative densities were determined by clipping the relevant scanned peaks in secondary experiments and weighing them on an analytical balance. The incorporation of proto-oncogenic and oncogenic p185neu was directly compared.

Focus formation and tumorigenicity tests

Cells (10⁴) in Petri dishes and cultured in DMEM containing 2% FBS. The medium was changed every 3-4 days. After 3 weeks of culture, the cells were fixed with 10% formalin and subjected to hematoxylin staining to observe morphologically transformed foci. To analyze tumor growth in nude nude mice, cells (10⁶) of each of the lines in 0.1 ml of PBS and were injected, intradermally, in the middle part of the back of single mice in NCR. PBS alone was also injected as a control. Tumor growth was monitored every 4-5 days up to 10-12 weeks.

Results

NE91 is a transfected cell line that expresses EGFR in NR6 cells (Pruss et al., Proc. Natl. Acad Sci, USA, 1977, 74, 3918), an endogenous EGFR-free mouse fibroblast cell line. P185 (Nneu) wild-type (WT) cell or Neu kinase deficient (i.e., N757 and N691stop, carrying a K757M mutation site at the ATP binding site and cytoplasmic domain deletion, respectively) were expressed in NR6 cells and NE91. The resulting transfected clones were designated NR6 Neu or NE Neu, respectively.

Mutant kinase-deficient neu proteins suppressed EGFR function in cellular transformation and prevented transformation synergy with EGFR.

It has previously been indicated that the co-expression of increased levels of EGFR and cellular p185, but not separately, transformed murine fibroblast cells completely, as has been demonstrated with the M1 cell line (Kokai et al., Cell, 1989, 58, 287) . In the present study the transformed phenotypes of these transfected cells expressing WT or Neu kinase deficient proteins in the presence or absence of EGF were analyzed.

NE91 cells expressing only EGFR formed a monolayer in the absence of EGF and foci in the presence of EGF. The incomplete transformation observed (i.e., EGF-dependent mode) is in agreement with previous reports (DiFiore et al., Cell, 1987, 51, 1063; Dobashi et al., Proc. Natl. Acad Sci, USA, 1991, 88, 8582). However, similarly to M1 cells, co-expression of WT cellular p185 and EGFR in NE NneuB2 cells resulted in complete transformation, ie, focus formation was independent of EGF. Cell lines co-expressing EGFR with any form of kinase-deficient Neu (NE N757 and NE N691 stop cells) did not form foci even in the presence of EGF. Similar results were observed when anchor-independent colony growth was tested on soft agar.

Tumor growth in single mice was used as a criterion for complete transformation in vivo. B104-1-1 cells expressing encogenic p185 were used as a positive control and tumors caused by these cells appeared rapidly (with a latency of 5 days). Cell lines expressing equivalent measures of EGFR (NE91) or cellular p185 (NR6 Nneu) alone did not give rise to tumors. However, injection of cells co-expressing EGFR and cellular p185 (M 1 and NE NneuB2) caused tumors (latency of 2-3 weeks). The results were consistent with a previous report (Kokai et al. Cell, 1989, 58, 287).

However, no tumors (> 10 weeks) were observed after injection of cell lines expressing Neu kinase deficient alone or co-expressed with EGFR. These data suggest that normal cellular kinase activity of p185 and EGFR function are required for synergistic transformation and tumor formation. Co-expression of kinase-deficient Neu proteins with EGFR not only canceled this type of synergy, but also suppressed the EGF-dependent transformation potential of EGFR. Therefore, the function of the EGF receptor mediated by ligand stimulation was further analyzed in the following studies.

IS 2 270 586 T3

EGF-induced receptor down-regulation was less efficient in Neu kinase-deficient mutant cells.

The inventors then examined whether normal receptor downregulation was affected by co-expression with kinase deficient Neu. Cells were incubated with EGF for various periods of time before surface staining of the cells with anti-neu mAb 7.16.4 or anti-EGFR mAb 425 followed by staining with F1TC conjugated anti-mouse IgG. The cell surface expression of any of the receptors was analyzed using flow cytometric analysis. Cell surface expression of EGFR in NE91 cells was reduced after 15 minutes of EGF treatment and more than 60% of the receptors disappeared from the cell surface after 1 hour of treatment. The efficiency of EGFR down regulation in M1 cells (co-expressing WT Neu and EGFR) was similar to that observed in NE91 cells. Approximately 20% of cellular p185 was "co-downregulated" along with EGFR in M1 cells. Similar results were observed in NE Nneu B2 cells. However, cell lines expressing cellular p185 alone did not respond to EGF. In cell lines in which EGFR was co-expressed with kinase-deficient mutant Neu proteins, EGFR down-regulation was less efficient (maximum reduction of approximately 20-25%). Furthermore, the surface expression of any of the mutant Neu proteins was not significantly altered in these cells by EGF treatment.

Increased receptor half-lives observed in cells co-expressed with kinase-deficient mutant neu

To determine if the receptors that were downregulated from the cell surface underwent degradation, pulse-chase labeling of receptor proteins was carried out, as described in materials and methods, and the proteins Neu and EGFR immunoprecipitated were analyzed by SDS-PAGE. EGF treatment caused rapid EGFR degradation in NE91 cells (expressing EGFR only). A similar degradation rate of EGFR was observed in M1 cells after EGF treatment. However, EGF-induced degradation of EGFR was slowed in cells by co-expressing EGFR with any of the kinase-deficient mutant Neu forms (NE N757 or NE N691 stop).

The degradation patterns of Neu WT or mutant proteins in response to EGF treatment were also investigated. WT cell p185 labeled in M1 cells and NE NneuB2 cells disappeared proportionally to the time of EGF treatment, indicating that WT cell p185 is efficiently codegraded with EGFR. There was only a slight reduction in N757 protein levels and no noticeable change in redundancy of the truncated N691 stop protein after EGF treatment up to 6 hours. The suggested normal half-life of c-erbB2 in mammalian epithelial cells is 11-13 hours (Kornilova et al., Oncogene, 1992, 7, 511). Densitometric tests of the inventors' autoradiograms confirmed that the half-life of WT cellular p185 was reduced to 3-4 hours after EGF treatment, while the levels of mutant Neu did not change significantly over the time of the examination.

EGF binding affinity in cells expressed with wt neu or mutant protein

Our experiments have shown that Neu kinase-deficient mutants suppress EGFR functions, such as kinase activity (Qian et al., Proc. Natl. Acad. Sci. USA, 1994, 91, 1500), EGF-mediated transformation , receptor down regulation and degradation. Since these effects could be interpreted in part by altered EGF binding affinities, the inventors have analyzed the binding parameters [¹²⁵I] -EGF by Scatchard analysis.

The mean dissociation constants (Kd) of [¹²⁵I] -EGF binding to these cell lines were determined by three individual experiments. EGFR in NE91 cells showed two binding components representing high binding affinities (7.5 x 10^-11 M) and low (4.4 x 10^-9 M), and the fraction of high affinity receptors was 5.4% of the total receptors. Co-expression of EGFR with WT Neu in NE NneuB2 cells resulted in a slight increase in EGF binding affinities (3.2 x 10^-11 M) and (2.0 x 10^-9 M) for both high and low affinity subclasses respectively and the fraction of high affinity receptors was 5.7%. The increased affinities for M1 cells were reproducible and the Kd values ​​(1.3 x 10^-11 M and 1.8 x 10^-9 M) were in agreement with previous reports by the inventors, Kokai et al., Cell, 1989, 58, 287; and Wada et al. Cell, 1990, 61, 1339). However, EGFR in cells co-expressing kinase-deficient Neu showed predominantly low EGF binding affinity, 4.9 x 10^-9 M and 5.2 x 10^-9 M in NE N691 and NE N757 cells, respectively, although a rare high affinity subclass of EGFR was detectable in some cases, i.e. 7.2 x 10^-11 M (0.5%) in NE N691 and 6.6 x 10 stop cells^-11 M (<1%) in NE N757 cells. These rare species may represent a pool of EGFR homodimers still observed when co-expressed with inactive Neu kinase proteins (Qian et al., Proc. Natl. Acad. Sci. USA, 1994, 91, 1500). It is evident from Scatchard analysis that EGFR in cells co-expressing kinase-active Neu WT show the normal percentage of high affinity EGF receptors with slightly increased affinity for EGF compared to NE91 cells. However, co-expression of kinase-deficient Neu protein markedly reduced EGF binding affinities in correlation with reduced heterodimeric kinase activities.

Comments

In current studies, receptor functions and cell phenotypes have been analyzed using stably transfected cell lines co-expressing EGFR with Neu WT proteins or mutants. Unlike Neu

IS 2 270 586 T3

WT, the kinase-deficient Neu did not cooperate with EGFR to mediate cell transformation; furthermore, the inventors have demonstrated new aspects of dominant negative receptor functions resulting from the interaction of mutant Neu with EGFR.

The intermolecular association and resulting tyrosine kinase activation between EGFR and WT (Qian et al., Proc. Natl. Acad. Sci. USA, 1992, 89, 1330) or mutant Neu proteins (Qian et al., Proc. Natl. Acad. Sci. USA, 1994, 91, 1500) have been well characterized. The inventors' studies have shown that the heterodimerization of EGFR and c-neu products can be detected even in the absence of EGF, and are favored over any other form of homodimerization. However, homodimerization and codimerization of WT EGFR and EGFR with deleted cytoplasmic domain were equally efficient and dependent on EGF (Kashles et al., Mol. Cell Biol., 1991,11, 1454). The predominance of heterodimers may help explain the resulting cellular phenotypes, and the dominant negative effect that can be induced by kinase-deficient neu or suppression of EGFR function that occurred significantly even with a 1: 1 ratio of EGFR and Neu proteins. mutants.

Receptor interaction with resulting activation of tyrosine kinase occurs by an intermolecular mechanism and is frequently followed by rapid transformation events, as observed in pp60.^c-src (Cooper et al., Proc. Natl. Acad. Sci. USA, 1988, 85,4232), insulin receptor (Boni-Schnetzler et al., J.Biol. Chem., 1988, 263, 6822) and EGFR (Honegger and others, Mol. Cell. Biol., 1990, 10, 4035). Transphosphorylation also occurs between heteroreceptor species, EGFR and Neu / c-erbB2 (Connelly et al., Proc. Natl. Acad Sci. USA, 1990, 87, 6054; Spivak-Kroizman et al., J. Biol. Chem., 1992 , 267, 8056; and Qian et al., Proc. Natl. Acad Sci. USA 1994, 91, 1500). Preferential heterodimerization of EGFR and Neu receptor (Qian et al., Proc. Natl. Acad Sci. USA, 1994, 91, 1500) can facilitate the transphosphorylation of N757 by EGFR. Currently, the specific substrates for EGFR and Neu kinase have not been well characterized. In in vitro binding assays it has been shown that phosphorylated kinase-deficient N757 was still capable of associating with SH2-containing recombinant protein after EGF treatment. However, unlike heterodimers active in M1 and NE NneuB2 cells, the loss of Neu kinase activity of the mutant heterodimer of NE N757 cells can prevent phosphorylation of certain cell substrates. Furthermore, the predominant transphosphorylation of N757 b and EGFR and the occupancy of cellular substrates in non-functional N757 can compete with EGFR to signal molecules leading to qualitative and quantitative reductions of EGFR function. Therefore, the defective heterodimer may not transmit signals as effectively as the heterodimer with kinase activity and the EGFR homodimer, thereby hindering the synergistic signaling that leads to the transformation of cells seen in M1 and NE NneuB2 cells and inhibiting the EGFR function. EGFR heterodimerization studies with N691 stop with deletion of the cytoplasmic domain demonstrated that the heterodimeric form was inactive due to the failure of the protein-protein interaction between the cytoplasmic domains, indicating that Neu / c-erbB2 is not simply a substrate for EGFR, it is also a transactivator for EGFR (Qian et al., Proc. Natl. Acad. Sci. USA, 1994, 91, 1500). Thus, the reduced amounts of the homodimeric form of normal EGFR and the preponderance of non-productive heterodimers resulted in the suppression of the resulting dominant normal and negative EGFR function phenotype. The observation is comparable to the effects of dimers formed between EGFR WT and EGFR with deletion of the cytoplasmic domain (Kashles et al., Mol. Cell. Biol., 1991, 11, 1454).

Kinase-active receptors have been reported to have been targeted to lysosomes for degradation with ligand binding (Chen et al., Cell, 1989, 59, 33; Felder et al., Cell, 1990, 61, 623). Previous studies using kinase-deficient insulin receptors (McClain et al., J. Biol Chem, 1987,262,14663; and Russell et al., J. Biol Chem, 1987, 262, 11833) and EGFR (Honegger et al., Cell, 1987, 51, 199) suggest that kinase-active domains are essential for ligand-induced normal targeting of receptors. The inventors used EGF-treated cell lines to study how kinase receptor complexes and activities correlate with receptor endocytosis and destruction. The inventors' work demonstrates that EGFR in co-expressed Neu WT cells (M1 or NE NneuB2) undergoes rapid down-regulation and degradation on EGF stimulation. This process was significantly delayed in mutant cells compared to NE91 cells expressing EGFR alone. Only cellular WT p185, but not the kinase-deficient mutant Neu protein, showed co-down regulation and co-degradation with EGFR. Similarly, EGF treatment of cells from the human mammary HC11 cell line affected the surface expression of the c-erbB2 protein and protein production: a 3-4-fold increase in the lysosomal protein c-erbB2 and the half-life of c-erbB2 proteins was reduced from 11 hours (without treatment) to 3.5 hours (with EGF treatment) (Kornilova et al., Oncogene, 1992, 7, 511). Together with the inventors' observations these results suggest that WT Neu / C-erbB2, (but not Neu kinase deficient), associates with EGFR and an active receptor tyrosine kinase complex and undergoes normal receptor targeting.

In conclusion, our results provide experimental evidence that defective or inactive EGFR heterodimers and kinase-deficient Neu proteins hinder synergistic hetero-5 receptor signaling, suppress normal EGFR function, and suppress the transformed phenotype in cells. alive. Our experimental model suggests a causal relationship between heterodimeric kinase activities and the malignant character of cells that may have clinical implications. A recent report has shown that a truncated ectodomain of the c-erbB2 protein produced by an alternative RNA process in human carcinoma cells overexpressed the results of the p185 receptor.^c-erbB2 in resistance to the growth inhibitory effects of the anti-c-erbB2 monoclonal antibody (Scott et al., Mol. Cell. Biol., 1993, 13, 2247). It has been speculated that direct transfer of the kinase-deficient Neu cDNA gene to tumor cell lines with co-overexpression of EGFR and Neu / c-erbB2 may release the malignant phenotype, since mutant Neu proteins may suppress function. of normal EGFR or c-erbB2 receptors by homo or hetero receptor interactions.

IS 2 270 586 T3

Example 4

Inhibition of a natural EGFR oncoprotein by the p185neu ectodomain: implications for sub-domain contributions to the receptor set

Introduction

Activation of the erbB family receptor involves formation of a homodimer and a heterodimer. In many cases, the formation of heterodimers between elements of the erbB family increases the ligand binding affinity and results in the formation of a more active signaling complex that influences the phenotype of the cell. Using p185neu and EGF receptor mutants, the ectodomain of these erbB receptors alone has been shown to be sufficient to allow a thermodynamically preferred heteromeric physical association and that cytoplasmic contacts in the resulting dimer affect ligand activity, signaling, and phenotype. Biochemical analyzes of p185neu and EGFR suggest that the consequences of dimer formation between extracellular domains alone are distinct from the signaling resulting from endodomain dimer formation. Ectodomain-derived p185neu mutants are capable of specific transinhibitions of EGF receptor signaling in murine fibroblasts and primarily transformed human cells that overexpress EGFR. The active receptor complex for Neu differentiation factor (NDF / heregulin) appears to be an erbB2-erbB3 or erbB2-erbB4 heterodimer, suggesting that p185neu / erbB2 function, in part, as a trans-regulator or other receptor kinases of the erbB family.

To further examine the receptor subdomains responsible for extracellular domain-mediated trans regulatory interactions in the erbB family, the interaction between EGFR and p185neu / c-erbB2 in transformed cells was analyzed. An EGFR oncoprotein, commonly observed in human glial neoplasms and other human epithelial malignancies (AEGFR or EGFRvIII), results from internal truncating (“in-frame”) involving exons 2 to 7 (amino acids 6 to 273) in the gene that encodes the extracellular region of the molecule that results in the expression of AEGFRs or 140-155kDa, truncated, constitutively phosphorylated. AEGF receptors have been found to exist spontaneously in dimer form and mediate constitutive signaling and oncogenic transformation of rodent fibroblasts in a ligand-independent manner, while overexpressed holo-EGFR p170 are transformed, only weakly, in the presence of EGF. The AEGFR oncoproteins confer a remarkable growth advantage in vivo in human glioblastoma cells and in murine fibroblasts.

Recent reports indicate that AEGF receptors are present on the cell surface and internalize more slowly than ligand-stimulated holo-EGFRs, which can increase the transformation efficiency of AEGFR oncoproteins. Other mutations have also been observed that functionally separate the extracellular domain with respect to the transmembrane and cytoplasmic region of RTK polypeptides, leading to spontaneous dimerization and the acquisition of transformation potential, suggesting that a part of the extracellular domain imposes a structural limitation on the dimer formation which is presumably eliminated by ligand binding or mass action. The extracellular deletions observed in AEGFRs or avian v-erbB oncogenes presumably facilitate dimer formation by mimicking the conformational changes resulting from ligand binding. It has been observed that the soluble extracellular domains of EGFR oligomerize and that structural alteration in the ectodomain can induce spontaneous oligomerization of extracellular domains, cytoplasmic domain, or both.

Extracellular deletion in AEGFR removes most of the amino acids that comprise subdomains I and II of EGFR, which includes a significant part of the first (most amino-terminal) sequence of two cysteine-rich sequences, in the extracellular region of the receptor. Subdomain III, which has been reported to confer ligand-binding characteristics to EGFR, is preserved in the AEGFR oncoprotein, although AEGFRs do not appear to bind ligand in NIH3T3 cells. Coexpression of holo-EGFR and AEGFR has been observed in human glioblastoma and other tumor samples, suggesting that EGFR / DEGFR co-expressing cells may be a close correlate of human disease.

A carboxyl-terminal deletion mutant, derived from the ectodomain of the oncogene p185neu (T691stop neu), lacking the complete kinase domain and carboxyl-terminal autophosphorylation sites, was expressed in human glioblastoma cells that co-express full-length EGFR and AEGFR for examine whether the p185neu ectodomain could associate with truncated AEGFRs, with deletion of ectodomain, modulating AEGFR-mediated signaling.

Results

Expression of EGFR and p185neu mutants in human glioblastoma cells

U87MG human glioblastoma cells express elevated levels (10⁵ receptors / cell) of wild-type endogenous EGFR. Three clonal derivatives of parenteral human glioblastoma cells U87MG were used for these studies: U87 / T691-1 cells contain T691stop neu in the background U87MG; U87MG.AeGfR cells express high levels (10⁶ receptors / cell) of human EGFR proteins in parenteral U87MG cells and U87MG.AEGFR / T691 doubly transfected cells contain EGFR, AEGFR and neu mutant T691stop proteins. EGFR and neu truncated protein expression levels were compared in U87MG-derived human glioblastoma cells after metabolic tagging. Subclones derived from human glioblastoma cells

ES 2 270 586 T3 parenteral U87MG, notable for the expression of AEGFR and / or mutant T691stop neu receptors were labeled with ³⁵S-cysteine ​​for 15h and the cell lysates were immunoprecipitated with anti-EGFR mAb 528 reactive with both EGFR and AEGFR ectodomains or with anti-neu mAb 7.16.4 that recognizes the p185neu ectodomain. Immune complexes were resolved and separated by 8% SDS-page. Protein signals representing EGFR (170kDa), AEGFR (140-155kDa) and truncated T691stop neu proteins (115kDa) were observed. U87MG cells express only full-length endogenous EGFR; U87 / T691-1 cells express endogenous EGFR and T691stop neu proteins; U87MG.AEGFR cells express endogenous EGFR and transfected AEGFR; and U87MG.AEGFR / T691 cells express EGFR, AEGFR, and T691stop neu proteins. All signals were observed after autoradiography (24h exposure). Immunoprecipitation with mAb 528 reagent (Oncogene Science) with EGFR and AEGFR reagent demonstrated all forms of EGFR expressed in U87MG-derived cell lines. EGFRs were identified in U87MG.AEGFR and U87MG.AEGFR / T691 cells only. Metabolic labeling and immunoprecipitation with mAb 7.16.4 reagent with the p185neu ectodomain allowed the identification of 115kDa neu mutant T691stop receptors in U87 / T691-1 cells and in U87MG.AEGFR / T691 cells. Flow cytometric analysis of the U87 / T691-1 and U87MG.AEGFR / T691 subclones with mAb 7.16.4 confirmed the surface localization of the T691stop neu proteins. Flow cytometric analysis also confirmed the surface localization of AEGFR cells in both U87MG.AEGFR and U87MG.AEGFR / T691 subclones. Glioblastoma U87MG cells contain negligible levels of erbB-2 or erbB-3.

Immunoprecipitation and immunoblotting of EGFR into U87MG.AEGFR cells revealed the presence of endogenous EGFR (170kDa) and transfected AEGFRs as doublet species of Mr140kDa and 155kDa. U87MG.AEGFR cell lysates were immunoprecipitated with mAb Δ124 reagent with AEGFR alone or mAb 528 reagent with the extracellular domain of EGFR and AEGFR. Equal amounts of protein, as determined by the Bio-Rad Protein Assay Kit (Bio-Rad Laboratories), were immunoprecipitated and immune complexes were separated by SDS / 8% PAGE under reducing conditions. Immunoprecipitated EGFRs were detected by immunoblotting with Ab-4, a polyclonal antibody against human EGFR. Antibody Δ124 precipitated two 140-155 kDa AEGFR species. Reagent mAb 528 precipitated EGFR (M_r= 170kDa) endogenous, as well as AEGFR (140-155 kDa). The two AEGFR species in U87MG.AEGFR cells were more clearly resolved by mAb A124. All protein signals were visualized by the enhanced chemiluminescence (ECL) system (Amersham). Scanning densitometry of AEGFR receptors immunoprecipitated with mAb A124 revealed that the ratio of AEGFR forms from 155kDa to 140kDa was 2.3 in U87MG.AEGFR cells. Additionally, densitometric analysis of mAb 528 immune complexes demonstrated that the ratio of AEGFR / EGFR in U87MG.AEGFR cells was approximately 10: 1. This model was also demonstrated in U87MG.AEGFR / T691 double transfectants. Scanning densitometry was used to confirm a stoichiometric ratio of AEGFR / AGFR of 10: 1 by immunoprecipitation of cell lysates using reagent mAb 528 with the extracellular domains of EGFR and AEGFR, followed by immunoblotting with an EGFR polyclonal antiserum reagent on cells. U87MG.AEGFR and U87MG.AEGFR / T691.

The T691 stop neu ectodomain was shown to efficiently form full-length heterodimers, wild-type EGFRs on the surface of parental U87MG cells and in rodent fibroblasts using the waterproof membrane linker DTSSP (3,3'-dithiobis (Sulfosuccinimidylpropionate) The p185neu ectodomain inhibited the EGF-induced down-regulation of endogenous EGFR in U87MG-derived cells, as determined by immunoblotting. Flow cytometric analysis indicated that receptor association, which occurred on the surface of cells, mediates EGFR inhibition rather than endocytosis and degradation. In particular, experiments were carried out demonstrating the association between endogenous EGFR and truncated neu receptors in vivo and the inhibition of EGF-induced downward modulation of EGFR in glioma cells expressing T691 stop neu receptors. Parental U87MG and U87 / T691-1 cells (U87MG cells expressing T691 stop neu) were precipitated with anti-EGFR mAb 528 (Oncogene Science) or anti-neu mAb 7.16.4 after crosslinking with DTSSP (3,3'-dithiobis (sulfosuccinimidylpropionate) (2 Mm) (Pierce) with or without EGF treatment (100 ng / ml at 37 ° C for 10-15 min.) Immunocomplexes were analyzed by SDS / 8% PAGE under reducing conditions. EGFR (Mr = 170kDa) was detected by immunoblotting with Ab-4 (Oncogene Science), a polyclonal antibody against human EGFR. Coprecipitated EGFR proteins were detected in U87 / T691-1 cells immunoprecipitated with antineu mAb 7.16.4. EGF treatment resulted in more efficient downward modulation of EGFR in U87MG cells than in U87 / T691-1 cells. These data are in agreement with previous studies carried out in rodent fibroblasts. The T691 stop neu ectodomain inhibited EGF-induced phosphorylation of endogenous wild-type EGFR from U87MG-derived cells. AEGFR is constitutively phosphorylated in U87MG.AEGFR cells, while p170 EGFR is phosphorylated in U87MG parental cells and U87MG.EGFR cells only after addition of EGF. Blotting with an antiphosphotyrosine antibody revealed that the lower molecular weight species (p140) of AEGFR is subphosphorylated relative to the p155 species in U87MG.AEGFR cells. Ligand-dependent activation of EGFRs in U87MG-derived human glioglastoma cells was determined as follows. The phosphotyrosine content of anti-EGFR immune complexes was determined in U87MG cells, U87 / T6911 cells (contain full-length endogenous EGFR and mutant-neu T691stop) and U87MG.AEGFR cells (contain endogenous EGFR and AEGFR). Equal number of cells were layered and left without feeding in serum-free medium for 24 h, after fixation in 10 cm dishes. Cells ± EGF (100 ng / ml at 37 ° C for 10-15 min) were treated, washed twice with cold PBS and solubilized with PI / RIPA buffer. Lysates of equal protein concentrations were immunoprecipitated, as determined, by the Bio-Rad assay (Bio-Rad Laboratories) with anti-EGFR mAb 528 and the immune complexes were analyzed by SDS / 8% PAGe under reducing conditions. Phosphorylated EGFRs were detected in parental U87MG and U87MG.AEGFR cells, but not in U87 / T691-1 cells. Phosphorylation of endogenous, full-length EGFR (170kDa) was dependent on EGF in U87MG cells and in

ES 2 270 586 T3 cells U87MG.AEGFR. However, phosphorylation of AEGFRs did not depend on EGF treatment in U87MG.AEGFR cells. The blot was separated and tested again with the polyclonal anti-EGFR antibody Ab-4. The presence of all EGFRs was confirmed in cell types treated as above. EGFR proteins appear as a 140-155kDa doublet, with the higher molecular weight species most significantly phosphorylated. All protein signals were visualized by the enhanced chemiluminescence (ECL) system (Amersham).

AEGFR-mediated cell growth modulation and transformation by the mutant T691stop neu

Cell proliferation and transformation performance of human glioblastoma cell lines, derived from U87MG, were evaluated in vitro and in vivo to determine the modulation of the signaling receptor AEGF by mutant ectodomain-derived p185neu proteins. Inhibition of cell growth under conditions of whole or reduced serum was studied in the following experiments. 2 x 10⁴ cells from each cell line were plated in 6-well plates and allowed to dock in full growth medium. The next day, cells were kept in full growth medium (10% FBS) or changed to 2% serum FBS. Cells were allowed to grow for four days, then trypsinized and counted. Parental U87MG cells were used for normalization (growth ratio = 1.0 for all experiments). The growth of all derived cell lines was expressed as a fraction of the parental cell line for comparison purposes. U87MG.AAEGFR cells express endogenous EGFR and AEGFRs, U87MG.AEGFR / T691 cells express EGFR, AEGFR, and T691 stop neu, and U87 / T691-1 cells express endogenous EGFR and T691 proteins. Expression of AEGFR (U87MG.AeGfR cells) increased cell proliferation under reduced serum conditions on parental U87MG cells, which is consistent with ligand-independent activation of AEGFRs. Expression of the neu mutant T691 stop proteins inhibited cell growth in reduced serum medium and markedly in full growth media both in glioblastoma cells that co-express EGFR / DEGFR and in parental U87MG cells that contain only endogenous EGFR. It should be noted that the U87MG.AEGFR / T691 subclones showed less cell proliferation than the parental U87MG cells lacking EGFR in complete growth medium and to a greater extent, under reduced serum conditions.

AEGFR did not increase the in vitro transformation efficiency in anchor independent growth assays relative to parental U87MG cells. The independent growth of the anchor was studied in the following experiments. 1000 to 3000 cells from each cell line were seeded on soft agar dishes and cultured for 21-28 days. Colonies were then visualized and counted after staining. U87MG cells are primarily transformed human cells that contain multiple somatic genetic alterations, including deletions of p16 and in the putative protein tyrosine phosphatase gene, PTEN. Soft agar growth of U87MG.AEGFR / T691 cells was reduced by 41.3% and 45% compared to parental U87MG and U87MG.AEGFR cells, respectively. The anchor-independent growth inhibition by T691 stop neu proteins was more significant in parental U87MG cells lacking AEGFR (mean inhibition 75.2% relative to parental U87MG cells in three independent experiments).

The AEGF receptor confers a selective growth advantage in vivo on the background of U87MG cells, while many studies have shown that holo-EGFRs are not transformed in vivo, except under defined conditions where p185neu receptors are coexpressed. Expression of the neu mutant T691 stop ectodomain in U87MG cells has been shown to preferentially inhibit the U87MG oncogenic phenotype when compared to a truncated form of p185 neu (N691 stop) that differs from T691 stop by containing the proto-onchogenic transmembrane region. Comparison of tumor growth in nude mice between U87MG derived cell lines was as follows. 10⁶ cells from each cell line were injected intradermally on day 0 and tumor volume was recorded 1-2 x / week. U87MG cells were injected on one side and the transfected cell line was injected on the contralateral side of the animal. Expression of the T691 stop neu protein suppressed the AEGFR-mediated in vivo selective growth advantage in U87MG cells. This result has been confirmed by an analysis of three additional u87MG.AEGFR / T691 subclones. The U87MG.AEGFR / T691 subclone showed growth kinetics similar to the parental U87MG cells, although it was more inhibited than the U87MG cells in vivo. The inhibition observed in vivo for all UB7MG.AEGFR / T691 subclones was directly related to the stoichiometry of the neu mutant T691 stop expression.

Mutant T691stop neu forms heterodimers with AEGFR in vivo.

Due to the genetic complexity of the U87MG.AEGFR transfectants and doubly transfected U87MG.AEGFR / T691 subclones and the multiple homodimeric and heterodimeric complexes that migrate to similar molecular weights in SDS-pAge gradient analysis, the DTSSP cross-linker impermeable to the thiodivisible membrane to examine individual components of putative surface-located heteromeric complexes. MAb 528 was used to immunoprecipitate all EGFRs (wild type and AEGFR) that can form heteromers with mutant neu proteins. Coprecipitated AEGFR monomers were removed from anti-neu immune complexes by in vivo crosslinking experiments using DTSSP. Crosslinking of U87MG.AeGfR / T691 cells and separation of immune complexes by SDS / 6-8% PAGE under reduction conditions showed evidence of heterodimer formation between T691stop neu proteins and the p140AEGFR, p155AEFGR and p-170EGFR forms. Most of the mutant T691 stop neu receptors were associated with the p140AEGFR form using these methods, although faint bands were repeatedly observed identifying the heterodimerized proteins p155AEGFR and p170EGFR. Identification of AEGFR proteins on the cell surface of U87MG.AeGfR cells and double transfectants

IS 2 270 586 T3

U87MG.AEGFR / T691 by flow cytometry supports observations made with the membrane-impermeable linker. Mutant T691 stop neu receptors form heterodimers with AEGF receptors on the cell surface. Experiments were carried out to detect AEGFR co-precipitates of antineu immune complexes by in vivo crosslinking. 1 mg of parental lysed U87MG cells (lanes 1, 2), U87MG.EGFR cells and U87MG.EGFR / T691 cells were immunoprecipitated with anti-EGFR mAb 528 or anti-neu mAb 7.16.4 after crosslinking with DTSSP (3.3 ' -dithiobis (sulfosuccinimidylpropylonate) (2mM) (Pierce) following EGF treatment (37 ° C for 15 minutes). Immune complexes were analyzed by SDS / 8% PAGE under reducing conditions and the nylon membrane was transferred with Ab-4, a polyclonal Ab against human EGFR. EGFRs were identified in U87MG cells (endogenous EGFR only, Mr = 170kDa); U87MG.AEGFR cells (EGFR and endogenous AEGFRs, Mr 140155kDa); and U87MG.AEGFR / T691 cells (EGFR and AEGFR, endogenous Mr 140-155kDa). EGFR proteins co-precipitated were detected in cells at U87MG.AEGFR / T691 immunoprecipitated with anti-neu mAb 7.16.4. In U87MG.AEGFR / T691 cells, T691 stop neu was found to co-precipitate with the low molecular weight form of AEGFR (140 kDa, strongest signal), the least migrant form of AEGFR (p155), and holo-EGFR ( p 170) endogenous. EGF treatment resulted in better visualization in immunocomplexed monomers but co-precipitated EGFRs could be identified in anti-neu immune complexes of U87MG-derived cell lines without EGF treatment.

The phosphotyrosine content of EGFR cells in immunoprecipitated U87MG.AEGFR / T691 was determined using the blot described above. The blot was separated and retested with the antiphosphotyrosine antibody, mAb PY-20 (Santa Cruz Biotechnology, Santa Cruz, CA.). EGFR proteins detected in anti-neu immune complexes had negligible phosphotyrosine content after EGF treatment in U87MG.AeGfR / T691 cells compared to anti-EGFR immune complexes in parental U87MG cells and in U87MG.AEGFR and U87MG.AEGFR / T691 cells. The phosphotyrosine content of anti-EGFR immune complexes in U87MG.AEGFR and U87MG.AEGFR / T691 cells was not appreciably different for these protein lysate concentrations. Blotting with an antiphosphotyrosine antibody confirmed that both AEGFR species associated with T691 stop neu are subphosphorylated. Negligible phosphotyrosine content was demonstrated in all experiments for EGFR monomers immunocomplexed to T691 stop neu receptors. EGF has been shown to increase the efficiency of heterodimer formation between p170EGFR and the neu ectodomain, although this association is ligand-independent. EGF minimally increases the formation of EGFR-p185neu ectodomain immune complexes, suggesting that EGF may stabilize the heteromeric formation, or perhaps the homodimeric EGFR formation. Reduction of phosphotyrosine content of EGFR monomers in vivo by co-expression of T691stop neu

The phosphotyrosine content for AEGFR monomers immunocomplexed to T691 stop neu receptors was negligible in all experiments. The results also revealed that the lower molecular weight of AEGFR (140kDa) was relatively subphosphorylated compared to the p155AEGFR form in both U87MG.AEGFR cells and U87MG.AEGFR / T691 double transfectants.

No differences in phosphotyrosine content of AEGFR monomers were observed between U87MG.AEGFR cells and U87MG.AEGFR / T691 cells in immunoprecipitations of larger cell lysates using mAb 528, which reacts with all forms of EGFR, for cross-linking experiments in alive. Therefore, in order to specifically examine the in vivo phosphotyrosine content of AEGFRs in cells expressing T691 stop neu, an antibody reactive with AEGFR only was used to precipitate AEGFRs from cell lysates containing low protein concentrations, than those required to detect heterodimeric complexes. The phosphotyrosine content of AEGFR in vivo in U87MG.AEGFR cells with or without co-expression of neu mutant T691 stop was determined as follows. Lysates (200 μ-g) of U87MG.AEGFR cells and U87MG.AEGFR / T691 cells were immunoprecipitated with mAb Δ124 reagent with AEGFR alone and with antiphosphotyrosine mAb PY-20 or the polyclonal anti-EGFR Ab-4 antibody. PY-20 blotting after immunoprecipitation with mAb A124 reagent with AEGFR revealed only several phosphoproteins in U87MG.AEGFR and U87MG.AEGFR / T691 cells. The slower migrating form of AEGFR (155kDa) was detected while the faster migrating form of AEGFR (140kDa) was not detected by the PY-20 antibody, indicating a relatively lower phosphotyrosine content than p155kDa. After removing the membrane and retesting with Ab-4 reagent with all EGFRs, both forms of AEGFR were visualized in both U87MG.AEGFR and U87MG.AEGFR / T691 double transfectants. Scanning densitometric analysis of phosphorylation content of immunoprecipitated p155AEGFR monomers in these cell lines revealed a 33.7% decrease in U87MG.AEGFR / T691 cells compared to U87MG.AEGFR cells, under full growth conditions. The constitutive differences observed in AEGFR phosphotyrosine content were not, therefore, outweighed by the serum-containing factors. The PTyr / AEGFR ratio in U87MG.AEGFR cells was 1.57; this ratio in U87MG.AEGFR / T691 cells was found to be 1.04, by determination by scanning densitometry. The difference was observed in two additional experiments.

The phosphotyrosine content in living cells of immunoprecipitated AEGFR monomers, not only AEGFR monomers immunocomplexed to T691 stop neu proteins, was analyzed. In addition to observing that AEGFR immunocomplexed to mutant T691 stop neu receptors have negligible phosphotyrosine content, these data indicate that the surface expression of T691 stop neu alone is sufficient to reduce the content of monomeric phosphotyrosine AEGFR in trans. In cells containing T691 stop neu, the observed 33.7% reduction in phosphotyrosine content of immunoprecipitated AEGFR monomers may decrease the signaling of the activated AEGF receptor complex, since the signaling complex may be a high-order multimer and AEGFR is has indicated that it has a lower stoichiometry of phosphotyrosine content than wild-type EGFR stimulated by EGF.

IS 2 270 586 T3

Substrate binding and / or catalytic activity of the AEGFR receptor kinase could be altered by a reduction in the content of monomeric AEGFR phosphotyrosine. The lower level of constitutive phosphotyrosine content of AEGFR relative to ligand-stimulated wild-type EGFR may be responsible for the inactivating effect of in vivo growth behavior of single point mutations at AEGFR terminal autophosphorylation sites. Hetero-oligomers formed by the association between AEGFR dimers and T691 stop neu dimers may be a mechanism for the reduction of AEGFR phosphotyrosine content and the resulting phenotypic inhibition of T691 stop neu expression and surface localization.

Reduction of the in vitro kinase activity of AEGFR by the expression of T691 stop neu

Since a reduction in the content of monomeric phosphotyrosine AEGFR was observed in cells expressing mutant T691 stop neu receptors, it was investigated whether the catalytic activity of the EGFR receptor kinase could be altered by the expression of the T691 stop neu protein. Using conditions identical to those that confirmed the presence of AEGFR in heterodimers associated with cross-linked T691 stop neu, experiments were carried out studying the reduction of EGFR kinase activity in vitro by the expression of T691 stop neu. 200 pg of lysates were obtained from U87MG.AEGFR and U87MG.AEGFR / T691 cells with and without pretreatment with the DTSSP membrane-impermeable linker (2 mM). Anti-EGFR immune complexes (mAb Δ124) from these cells were suspended in 50 µl of kinase reaction buffer, containing 0.2 mCi [³²P] -y-ATP at room temperature for 30 min. Protein samples were separated by 10% SDS-PAGE and analyzed by autoradiography. Immune antiEGFR complexes were shown to have increased kinase activity in vitro in U87MG ^ EGFR cells pretated with a membrane impermeable linker (DTSSP), but not in doubly transfected cells expressing mutant T691 stop neu receptors. The expression of T691 stop neu resulted in a remarkable inhibition of the transphosphorylation of the more slowly migrating form (155kDa) of ΔEGFR, due to the heterodimer formation confirmed by the use of DTSSP. These results were confirmed on three independent occasions. Since 155kDa ΔEGFR is exhibited more abundantly in both U87MGAEGFR and U87MGAEGFR / T691 cells, the T691 stop neu-mediated reduction of ΔEGFR receptor kinase catalytic activity may explain the phenotypic inhibition observed in double transfected human glioglastoma cells. No significant differences in phosphorylation of the p140ΔEGFR form were observed in these in vitro experiments; however, since this species was phosphorylated in vitro, a fraction can be phosphorylated in vivo. Inhibition of the catalytic activity of the ΔEGFR receptor was observed continuously for the transphosphorylation of the receptor. At the lowest protein concentrations used in in vitro kinase experiments there was minimal phosphorylation of the exogenous Histone III substrate in all anti-EGFR immune complexes.

The phenotypic inhibition of ΔEGFR signaling mediated by mutant T691 stop neu receptors in human glioblastoma cells therefore appears to result from: (1) heterodimer formation between T691 stop neu proteins and both forms of ΔEGFR, although a increased heterodimer formation with p140ΔEGFR; (2) transinhibition of the monomeric phosphotyrosine content of 155ΔEGFR in in vivo expression by T691 stop neu and (3) inhibition of p155ΔEGFR kinase transphosphorylation as a consequence of T691 stop neu expression and heterodimer formation.

Commentary

U87MG cells in which active ΔEGF receptors are constitutively co-expressed with endogenous EGFR represent an intimate approximation of an especially aggressive subset of human glioblastomas, that is, tumors in which p16 deletion, allelic loss on chromosome 10q, and EGFR activation, whereas p53 has not been mutated. U87MG-derived human glioblastoma cells expressing endogenous EGFR, elevated amounts of ΔEGFR oncoproteins, and kinase-deficient T691 stop mutant receptors (U87MGAEGr'R / T691 doubly transfected subclones) were inhibited in all in vitro and in vivo assays to a greater extent. as parental cells U87MG. This represented a significant reduction in the phenotype observed with expression of ΔEGFR oncoproteins alone in the background of U87MG (U87MGAEGr'R cells), particularly in vivo. Since the overexpression of wild-type EGFR alone is non-oncogenic in vivo, the observed formation of ΔEGFR-neu ectodomain heterodimers and the ratio of ΔEGFR: EGFR proteins in these cells is demonstrated that the observed growth inhibition conferred by T691 stop neu and shown by U87MGAEGFR / T691 cells was mediated by deactivation of signaling, via ΔEGFR receptors instead of endogenous p170 EGFR.

Mutant T691 stop neu receptors may inactivate a ΔEGFR-EGFR heterodimeric complex, although the stoichiometric ratio of ΔEGFR: EGFR is approximately 10: 1 in U87MGAEGFR cells and in U87MG.EGFR / T691 cells. Unlike endogenous wild-type EGFR, ΔEGF receptor dimer formation and autophosphorylation in glioma cells occur independently of ligand, and NIH3T3 cells expressing ΔEGFR show ligand-independent growth and transformation characteristics, suggesting that oncogenic signaling results of constitutively phosphorylated ΔEGFR dimers. Others have not identified EGFR ^ EGFR heterodimers in U87MGAEGFR cells. Additionally, tyrosine phosphorylation of ΔEGFR kinase-deficient mutants, expressed in U87MG cells, cannot be restored by activating wild-type EGFR with ligand treatment suggesting the lack of substantial transphosphorylation between EGFR and ΔEGFR. Given the thermodynamic preference of p185neu erbB2 proteins to heterodimerize with EGFR and other erbB receptors, ΔEGFRs can form dimers with proteins derived from the p185neu ectodomain more easily than with holo-EGFR.

IS 2 270 586 T3

T691stop inhibits 170EGFR phosphorylation in U87MG cells, and EGFR and AEGFR monomers immunocomplexed to T691 stop neu proteins have negligible phosphotyrosine content. The demonstration of association between T691stop and AEGFR does not necessarily indicate a preferential association over p170 EGFR because AEGFR and EGFR are not expressed at comparable levels in these cells. Flow cytometric analysis of all EGFRs in subclones expressing U87MG.AEGFR and U87MG.AEGFR / T691 indicated that expression of the p185neu ectodomain did not alter total EGFR, wild-type EGFR, or cell surface AEGFR populations, in U87MG cells containing endogenous EGFR alone or in U87MG.AEGFR transfectants containing EGFR and AEGFR. This corresponds to the observation that the p185neu ectodomain deactivates EGFR signaling by the formation of defective heteromeric or oligomeric receptor arrays located on the cell surface, rather than by inducing AEGFR receptor internalization and downward modulation.

Autophosphorylation of tyrosine residues at EGF receptors activates binding sites to target molecules and can also regulate the catalytic activity of the EGF receptor. The constitutive phosphotyrosine content of AEGFR monomers in vivo and the kinase activity of AEGFR in vitro are reduced in trans as a consequence of T691 stop neu expression. The phenotypic inhibition observed in vitro and in vivo in U87MG cells that co-express AEGFR and t691 stop neu proteins (U87MG.AEGFR / T691 cells) relative to AEGFR alone (U87MG.AEGFR cells) could be due in part to reduced binding sites in EGFRs to target molecules as a result of heterodimer formation with T691 stop neu. Additionally, the kinase activity for transphosphorylation of the AEGF receptor or other substrates can also be reduced by a conformational change induced by association with the mutant T691 stop neu receptor. The data support the argument that a decrease in receptor transphosphorylation kinase activity contributes to a reduction in transformation. In vitro kinase activity for an exogenous Histone substrate was observed to a much lesser extent than for receptor transphosphorylation and was not appreciably altered by T691 stop neu expression. The in vitro kinase activity of AeGfr for exogenous substrate is only minimally altered by substitutions of the carboxyl terminal autophosphorylation sites; the reduction of phosphotyrosine content in vivo as a consequence of carboxyl end point mutations in AEGFR appeared to correlate more reliably with phenotypic inhibition. The level of inhibition achieved in vivo by the T691 stop neu mutant, expressed in U87MG.AEGFR cells, was similar to that shown by an ATP-binding site mutant in AEGFR or by a tyrosine-substituted AEGFR mutant 1068, 1148, and 1173.

Studies using transdominant p185neu mutants have indicated that the ectodomains of p185neu and EGFR are sufficient for physical association and that EGFR signaling can be modulated with these kinase negative p185neu mutants. Cytoplasmic domain receptor interactions determine productive signaling for both p185neu and EGF receptors. In the absence of crystallographic data from the extracellular regions of p185neu and EGFR, the structural characteristics of ectodomain interactions between these two receptors are undefined. AEGF receptors have been found to exist in dimerized form in the absence of ligand. Soluble extracellular regions of EGFR have been observed to oligomerize in response to subsequent EGF cross-linking, although proteolytic fragments derived from subdomain III alone did not oligomerize, suggesting that other subdomains contribute to dimer formation. Subdomain III has been reported to confer ligand binding characteristics to EGFR. Activated avian erbB oncogenes form homodimers in the absence of ligand with deletions of the entire extracellular region other than a part of subdomain IV (second cysteine-rich domain). However, the physiological significance of the observed formation of this homodimer was unclear as it did not correlate with the tissue-specific transformation characteristics of these mutants.

Physical association can take place between the extracellular regions p185neu and EGFR in transformed cells, despite a deletion that includes most of the two independent subdomains (I, II) in the extracellular region of the EGF receptor. Extracellular mutants of p185neu deleted from either subdomain I or II still retain the ability to form heterodimers with full-length EGFR, confirming that these sequences are not critical for physical, heteromeric p185neu / EGFR associations. The phenotypic inhibition of EGFR oncoproteins by T691 stop neu proteins supports an argument that a physical association mainly governed by subdomains III and IV is sufficient to modulate signaling. Based on an analysis of the transformation efficiency of fibroblasts expressing wild-type human EGFR and extracellular subdomain deletion mutants of p185neu, subdomain III in p185neu appears to be the least relevant extracellular subdomain for the formation of a transforming p185neu / transforming heterodimeric signaling complex. EGFR, suggesting that the interactions mediated by subdomain IV may be the most important for the initiation and / or stabilization of homodimeric and heterodimeric receptor complexes.

Each of the two cysteine-rich extracellular domains of subdomains II and IV of p185neu and EGFR may possess a unique fold known as an "EGF fold" or "cysteine ​​knot." The motif is characterized by a repeat of six cysteine ​​residues and at least two interchain disulfide bonds. Although a similar motif has been observed in other proteins, its presence is highly conserved in cytokines and transmembrane receptors, including the structurally resolved tumor necrosis factor (TNF) receptor. Several receptor tyrosine kinase have been shown to contain these cysteine-rich domains and are presumed to adopt conformations similar to that of the TNF receptor. Tumor necrosis factor (TNF) receptors have been observed as dimers in uncomplexed crystal structures. In this form, the last cysteine-rich extracellular domain forms the major dimeric contacts. In these studies, the domain near the membrane is disordered, perhaps due to the lack of a transmembrane region that maintains this domain in a stable state. Therefore, it has been hypothesized that in the receptor as a whole, the last cysteine-rich domain, amino-terminal with respect to the

ES 2 270 586 T3 transmembrane sequence, can be stabilized by the transmembrane sequence and possibly involved in the formation of functional dimers. A high degree of sequence homology between the second cysteine-rich domain (subdomain IV) of p185neu and EGFR and the folding of the cysteine ​​knot in the TNF receptor have been identified. A simple sequence comparison in the near-transmembrane domains of the TNF and p185neu receptors shows that at least four of the six cysteines have been conserved.

Regulation by trans-receptor interactions has been observed for all members of the erbB family, many of which show altered expression or regulation in cases of malignant disease of the human epithelium. The physiological tendency of this family of receptors to form heteromeric associations suggests that targeting of human erbB oncoproteins, with the structure or pharmaceutical agents that limit certain subdomains, in the p185neu ectodome may be available in certain cases of human malignancy. Alternatively, the ectodomain p185neu cDNA could be delivered to erbB receptor positive tumor cells in human neoplasia gene therapy. AEGFR oncoproteins are differentially expressed in many human epithelial neoplasms and may represent a specific tumor target, however, these receptors are not ligand-binding, constitutively phosphorylated, and poorly internalized. These characteristics may limit efforts to inhibit AEGF receptor signaling in human tumors. The preferred thermodynamic tendency for erbB heterodimer formation in the absence of ligand suggests that targeting grbB oncoproteins with the p185neu ectodomain, novel pharmaceutical agents or peptide limitations revealing dimer formation would be more effective in achieving growth inhibition than inhibition of the induced activation of ligand with monoclonal antibodies or ligand-binding antagonists.

Materials and method

Vector construction

The T691 stop neu deletion mutant was derived from the oncogenic rat neu cDNA, pSV2Tneu, containing a single point mutation in the transmembrane region. Site-directed mutagenesis was used to introduce a stop codon at position Thr-691 in the endodomain. This cytoplasmically deleted form of p185neu, lacking the catalytic kinase domain, and carboxyl terminal autophosphorylation sites, was then inserted into a mammalian expression vector. A fragment containing the hygromicin gene^r pHYG was subcloned into APtag-1, an alkaline phosphatase (AP) expression vector under the control of the Maloney leukemia virus promoter and LTR (MuLVLTR). The AP gene was then replaced by the mutant T691 stop neu cDNA. Therefore, the T691 stop was expressed in the expression vector pMuLVLTR / T691 1stop / Hyg^r. Cell maintenance and development of stably transfected cell lines

The parental cell human glioblastoma cell line U87MG and the previously reported human glioblastoma subclone U87MG.AEGFR containing human AEGFR receptors (Nishikawa et al., 1994) were obtained from Dr. Webster Cavenee (Ludwing Cancer Institute, San Diego, California). For stable cell transfections, ten micrograms of the pMuLVL-TR / T691stop / Hyg construct were transfected.^r in U87MG.AEGFR cells via lipofectamine reagent (GIBCO / BRL, Gaithersberg, MD) under conditions determined by transfections using the pCMV-β (bacterial β-galactosidase) reporter construct (Clontech). The optimal efficiency of chemiluminescence transfection detected by a luminometer (Tropix) was determined. All cells were cultured in Dulbecco's modified Eagle's medium (DMEM, Bio-Whittaker, Walkersville, MD) with 10% fetal bovine serum ((Hyclone, Ogden, UT), 100U penicillin, 50 pg / ml streptomycin, and 2 mM Lglutamine (GIBCO BRL) Cells cultured at 37 ° C were maintained in 95% air / 5% CO2 U87MG.AEGFR cells were supplemented with 4 mg / ml G418 (Geneticin, GIBCO BRL) for maintenance of transgene expression of AEGFR.

For the development of double transfectants derived from U87MG expressing AEGFR and T691 stop neu proteins, the media were supplemented with G418 sulfate and hygromycin B. After 2-3 weeks in selection media containing 70 pg / ml hygromycin B (Boehringer Manheim), and 4 mg / ml G418 sulfate (Geneticin, GIBCO BRL), established clones U87MG.AEGFR expressing T691 stop neu (designated U87MG.AEGFR / t691 cells) and were screened by flow cytometric analysis, using mAb 7.16.4 against the neu ectodomain. The medium for stably transfected subclones was supplemented with 4 mg / ml G418 sulfate and 35-70 ug / ml hygromycin B for maintenance of transgene expression. Stably transfected cell lines were checked periodically by flow cytometric analysis with mAb 7.16.4 to document stable levels of T691 stop neu transgene expression.

Metabolic marking of cells followed by immunoprecipitation

Subconfluent cells are seeded (1 x 10⁶) overnight, in 10 cm dishes, in complete growth medium (10% FBS-DMEM). The next day the cells are subjected to DMEM without cysteine ​​for one hour and then pulsed with³⁵S-cysteine ​​(50 pCi / ml) (Amersham) for 15h in 3% dialyzed FBS / cysteine-free DMEM. The lysates are collected, after two washes in PBS, using a PI / RIPA buffer. Immunoprecipitations are carried out on ice for 60 minutes and complexes are separated by binding to protein A-sepharose before separation by 8% SDS-PAGE, drying and exposure to film. Monoclonal antibody 7.16.4 against the p185neu ectodomain is described in US Patent No. 5,677,171. Monoclonal antibody 528

ES 2 270 586 T3 against EGFR and EGFR extracellular domains was obtained from Oncogene Science. 5 pg of antibody was used for immunoprecipitation of proteins from lysates recovered from the 10 cm dishes.

Crosslinking, Immunoprecipitation and Western Blot Studies

Equal numbers of cells were applied and grown overnight in 10 cm dishes. The cells were subjected to serum-free medium for 24 h and were treated with EGF (100 mg / ml at 37 ° C for 10-15 min) and then washed twice with cold phosphate-buffered saline (PBS). For cross-linking, PBS containing 2 mM DTSSP (3,3'-dithiobis (sulfosuccinimidylpropionate) (Pierce) was added and the cells were incubated at 23 ° C for 30 min with occasional rocking of the plates. The cross-linking reaction was stopped with buffer. containing 10 mM Tris HCl, 0.9% NaCl, and 0.1 M glycine Cells were then washed twice with cold PBS and solubilized with PI / RIPA buffer. Cell lysates were immunoprecipitated with anti-neu mAb 7.16.4, anti-EGFR mAB 528, or anti-AEGFR mAb Λ124. Immune complexes of neu proteins or EGFR proteins were solubilized and separated by SDS-PAGE gels (6-8%) and transferred onto nitrocellulose before immunoblotting with the polyclonal anti-EGFR antibody, Ab-4 (Oncogene Science) or anti-phosphotyrosine. mAb PY-20 (Santa Cruz Biotechnology, Santa Cruz, CA).

In vitro kinase assay

Cells were applied to 100 mm culture dishes and the next day they were washed twice in cold PBS and lysed in 1 ml of lysis buffer (50 mM Hepes, pH 7.5, 150mM NaCl, 3% Brj- 35.2 mM EDTA, 0.02mg / ml Aprotinin, 10% gricerol, 1.5 mM MgCl₂). Cell lysates were centrifuged at 20,000 g for 15 min. Cell lysate protein concentrations were measured with Dc protein Assay (Bio-Rad). 40 microliters of Protein A-Sepharose 50% (vol / vol) were used to collect the immune complexes which were then washed three times with a wash buffer (50 mM Hepes, 150 mM NaCl, 0.1% Brij-35 , 2 mM EDTA, 0.01 mg / ml Aprotinin 0.03 mM NA₃Vo₄). The tablets were suspended in 20 microliters of 20 mM HEPES (pH 7.4, 5 mM MnCl₂, 0.1% Brij-35, 0.03 mMNa₃Vo₄, 0.02 mg / ml aprotinin) containing 5 pCi or ³²py-ATP, and incubated at room temperature for 30 min. The reaction was terminated by the addition of 3x electrophoresis sample plug containing 2 mM ATP. After incubation of the cells at 100 ° C for 3 min, the samples were analyzed by SDS-PAGE.

In vitro and in vivo tumorigenicity assays

Cell growth was assessed under whole or reduced serum conditions as follows: 2x10⁴ cells from each cell line were applied in 6-well plates and allowed to stick in complete growth medium. The next day cells were either maintained in complete growth medium (10% -FBS) or switched to 2% serum -FBS. Cells were grown for four days and then trypsinized and counted.

Anchor independent growth was determined through the efficiency of colony formation of cells suspended in soft agar. 1-3 X 10 were suspended³ cells in 1 ml of top layer (0.18% agarose / 10% FBSDMEM), in 6cm culture dishes, containing a 3 ml cell-free feed layer consisting of 0.25% agarose in DMEM supplemented with 10% FBS and 20 mM Hepes (pH 7.5). Colonies (> 0.3 mm) were visualized and counted on days 21-28 for all cell lines after staining with p-iodonitrotetrazolium violet (1 mg / ml). Each of the cell lines was examined in triplicate for the three separate experiments.

Homozygous 6-8 week old NCr plain mice were purchased from the National Cancer Institute. Cells were suspended (1 X 10⁶) in 0.1 ml of PBS and injected intradermally into the mid-back of each animal. U87MG cells were injected on one side of individual animals and stably transfected cell lines were injected on the contralateral side to make direct comparisons of growth within each animal. PSB alone was also injected into each animal as an additional control. The animals were maintained in accordance with the standards of the Committee on Animals of the University of Pennsylvania and those prepared by the Committee on Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resource. Tumor growth is monitored twice a week for a period of 10-12 weeks. Tumor dimensions were calculated by measuring tumor volume (length x width x thickness).

Antibodies

Monoclonal antibody (mAb) 7.16.4 reactive with the ectodomain of p185neu has been previously described. The antiphosphotyrosine antibody mAB PY-20 was obtained from Santa Cruz Biotechnology (Santa Cruz CA). MAb 528 (Ab-1) reactive with the extracellular region of EGFR and AEGFR was purchased from Oncogene Science (Uniondale, NY). AEGFR reactive mAb Δ124 was obtained from Dr. Webster K. Cavenee, Ludwig Cancer Institute, San Diego, CA. The Ab-4 polyclonal antibody, reactive with EGFR and used for immunoblots, was obtained from Oncogene Science (Uniondale, NY).

IS 2 270 586 T3

Example 5

Conversion of a radioresistant phenotype to a more sensitive one by deactivating ERBB receptor signaling in human cancer cells

Introduction

The molecular parameters that determine how a cell becomes more or less sensitive to DNA damage induced by radiation or chemotherapeutic agents are not well understood. The status of signaling pathways at critical points in the cell cycle has been suggested to be an important determinant of the response to DNA damage, and mutations in the components of the critical point ("checkpoint") are prevalent in human cancers. A recently introduced hypothesis suggests that tumor cells show growth arrest or apoptosis in response to cytotoxic therapies that depend on the functional status of critical point pathways and that radiation-induced apoptosis may result from growth disruption pathways. Similarly, in other systems using non-transformed cells, incomplete DNA repair mechanisms, which occur during the delay of the critical point phase, increase the tendency for apoptosis.

Human glioblastomas show many genetic alterations including amplification and / or mutation of the gene encoding Epidermal Growth Factor Receptor (EGFR) resulting in some cases in the expression of a constitutively activated EGF receptor kinase.

Expression of an EGFR transreceptor inhibitor, derived from the ectodomain of the oncogen p185neu (T691stop neu), forms heterodimers with both full-length EGFRs and a constitutively activated, extracellular deletion EGFR (AEGFR) mutant, this form being commonly observed in glial tumors. humans, particularly those with a higher degree of pathology. The growth and transformation of EGFR-positive cells or EGFR / AEGFR co-expressing human glioma cells are inhibited by p185neu kinase-deficient deletion mutants. The surface-localized heterodimeric receptor complex of mutant T695 stop neu / EGFR has decreased affinity for the EGF ligand, altered internalization kinetics, reduced phosphotyrosine content, and reduced enzyme kinase activity relative to full-length homodimeric complexes of EGFR and EGFR.

The specific pathways that mediate oncogenic transformation in EGFR-positive transformed human cells have not been fully characterized. Natural AEGFR oncoproteins can increase the constitutive activity of the Grb2 / Shc / Ras pathway and signal via phosphatidyl inositol-3 (Pl-3) kinases, presumably by binding to defined adapter proteins. Specific protein mitogen-derived kinases (MAP), such as those of the c-jun amino terminal kinase (JNK) family, can be constitutively activated by ligand-independent oncogenic AEGF receptors. While holo-EGFRs have been found to be weakly transformative, only in a ligand-dependent manner at high levels of receptor expression in fibroblasts, many human tumors show elevated levels of EGFR, and this may contribute to unregulated kinase activity in transformed cells.

Experiments were designed to determine whether the specific inhibition of signaling through EGFR overexpressed in radioresistant human glioma cells, would alter the physiological response of these cells with respect to the induction of genomic damage. Gamma-ray irradiation combined with inhibition of the erbB receptor resulted in a greater degree of radiation-induced growth arrest and apoptosis in cells normally resistant to ionizing radiation. Increased apoptosis occurred in transformed human glioma cells containing a wild-type or mutated p53 protein, and suggested that both p53-dependent and p53-independent mechanisms mediated such physiological outcome. Distal pathways with respect to the specific inhibitory interaction between the neu mutant T691 stop protein and the EGF receptor determine the tumor's responsiveness to genomic damage, and these pathways may be modulated by close receptor associations. Specific inhibitory pathways initiated at the cell membrane level and associated with growth arrest and / or apoptosis can modulate subsequent tipping points in response to DNA damage. These results have implication for the design of receptor-specific agents capable of sensitizing cells to cytotoxic therapies and suggest that erbB receptor-specific inhibition combined with cytotoxic treatments may enhance the response to anticancer agents.

Material and methods

Vector Construction

The derivation of the mutant T691 stop neu receptor construct has been set forth in detail above.

Cell maintenance and development of stably transfected cell lines

The human glioblastoma cell line U87MG was obtained from Dr. Webster Cavenee (Ludwig Cancer Institute, San Diego, California). U373MG human glioma cells, originally isolated from a human anaplastic astrocytoma, were obtained from the American Type Tissue Collection (ATCC) (Rockville, MD).

IS 2 270 586 T3

Flow cytometric analysis of cell cycle distribution

Cells were stained for flow cytometry by sequential treatment with 0.003% trypsin solution, followed by 0.05% trypsin inhibitor, 0.01% RNase A solution, and then 0.0416 propidium iodide. % (PI) and 5 mM spermine tetrachloride solution. Each of the treatments was carried out for 10 minutes with continued stirring at room temperature. All reagents were obtained from Sigma. Cell cycle analysis was carried out within two hours of staining on a Becton-Dickinson FACScan flow cytometer. Ten thousand events were collected for each sample and the data was analyzed using the ModFIT cell cycle analysis program (Becton-Dickinson, version 2.0).

Nuclei staining and morphological analysis of apoptosis

Cells were plated for a minimum of 12 hours prior to irradiation. Irradiation was carried out under conditions identical to the colony formation tests. The plates were washed twice with PBS at the indicated times, and fixed in a 50:50 mixture of ice cold methanol / acetone for ten minutes. Plates were subsequently stained with hydrated 4 ', 6-Diamidino2-phenylindole dihydrochloride (DAPI) (Sigma, St. Louis, MO) at a concentration of 0.1 pg / ml in PBS. Consistency between observations was confirmed in apoptosis counts with dUTP notch end marking (TUNEL) -terminal deoxynucleotidyl transferase mediated staining and three independent observers.

Cell counts were carried out within 30 minutes of staining and photographs were taken on a Zeiss Axioplan epifluorescence microscope. At least three independent fields of 100 cells were counted for each sample.

Colony Formation Assay

Cell survival after irradiation was evaluated by colony formation assay. The number of cells to be applied was calculated to form 20 to 200 colonies per dish at each radiation dose and they were applied in 10 cm culture dishes (Fisher Scientific, Pittsburg, pA). Cells were irradiated using a JL irradiator Cesium-137 Shepherd model 30 Mark I delivering 12.8 Gy / min with cells on a rotating platform to ensure uniform irradiation. Cells were incubated after irradiation at 37 ° C with 5% CO2 for 7-10 days and then subjected to crystal violet staining. Colonies containing more than 50 cells were counted with a dissecting microscope. The surviving fraction is the ratio of the number of colonies formed to the number of applied cells, and was corrected for application efficiency. At least three different cell concentrations were used for each radiation dose.

Western transfer

For each time point, 10⁵ cells per 6 cm plate by lysis in 400 µl of sample buffer (10% glycerol, 2% SDS, 100 mM DTT, 50 mM Tris, pH 6.8). 30 µl of each lysate were loaded per lane and separated by electrophoresis on a 15% SDS-polyacrylamide gel before overnight transfer to a nitrocellulose membrane (Bio-Rad, Hercules, CA). The membranes were tested with mouse anti-human p53 monoclonal antibody (NeoMarkers, Femont, CA.), followed by goat anti-mouse secondary antibody coupled to horseradish peroxidase (Amersham, Arlington Heights, IL). In order to reduce background antibody binding, incubation with secondary antibody was carried out in powdered milk at 2.5% in PBS. Detection was carried out by chemiluminescence (ECL, Amersham, Arlington Heights, IL). Relative levels of p53 expression were determined by scanning the blots using a scanning densitometer (Molecular Dynamics).

Antibodies

Monoclonal antibody 7.16.4 reactive against the p185neu ectodomain has been described above. Anti-ERK and anti-JNK antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Polyclonal antibodies reactive with p53 and p21 were obtained from NeoMarkers (Fremont, CA.). Antibodies reactive with bcl-2, bax, and bcl-x_L were obtained from Oncogene Science (Uniondale, NY).

Results

Cell cycle distribution of human glioblastoma cells in cycles with gamma irradiation: effects of erbB signaling deactivation on growth arrest

For both U87MG and U87 / T691 cells, prolonged treatment with serum alone (72-100h) led to increased accumulation of cells in G0 / G1, with modest reductions in both the S and G2 / M populations. U87 / T691 cells showed a higher G0 / G1 fraction than parental U87MG cells in the presence of serum or after prolonged serum starvation, indicating that the increased growth arrest relatively induced by the expression of the mutant T691 stop neu receptor could not overcome by growth in whole serum.

IS 2 270 586 T3

The induction of growth arrest by exposure of asynchronously cycling transformed human glial cell populations to gamma irradiation was superior to that induced by prolonged serum suppression alone. In both U87MG and U87GM / T691 cells, irradiation of cells grown under whole serum growth conditions caused strong increases in G0 / G1 and G2 / M, and a decrease in the percentage of cells in S phase, determined by flow cytometric staining. for DNA content (Figure 1B and 1D). Reduced S phase and accumulation of cells in G2 are characteristic of cells with DNA damage. The data in Figures 1A, 1B, 1C and 1D show a representative experiment of cells analyzed 72h after gamma irradiation. The above time points indicated similar trends, but the analysis 72h after irradiation was chosen to be consistent with subsequent experiments. An analysis of three independent experiments revealed the following changes in the cell cycle distribution (mean cell percentage ± SEM; ± radiation treatment [RT]:

1. ) U87MG parental cells:

G0 / G1: 26 ± 2.8, + RT 51.5 ± 2.1;

S: 66 ± 4.2, + RT 21 ± 2.8;

G2 / M: 8 ± 1.4, + RT 28.5 ± 0.7

2. ) U87 / T691 cells:

G0 / G1: 34.5 ± 4.9, + RT 71 ± 7.1;

S: 57.5 ± 4.9, + RT 16 ± 4.2;

G2 / M: 7.5 ± 0.7, + RT 12.5 ± 3.5

U87 / T691 cells showed a higher G0 / G1 fraction and reduced S and G2 / M populations compared to parental glioblastoma cells when cultured asynchronously in cultures with or without radiation treatment and the greatest difference occurred in the population G0-G1. Radiation-induced increases in the G2 / M fraction were seen in both U87MG and U87 / T691 cells, albeit to a greater extent, in parental U87MG cells. The combination of serum suppression and radiation treatment in these cell populations was not additive and did not appreciably alter the cell cycle distributions in any cell line from that seen with whole serum radiation treatment. Therefore, it is observed that the deactivation of EGFR-mediated signaling induces a growth arrest by a different mechanism than that observed with prolonged serum suppression.

Transreceptor inhibition sensitizes human glioblastoma cells to radiation-induced apoptosis.

Human glioblastoma cells have been shown to be especially resistant to radiation treatment experimentally and clinically. EGFR overexpression and / or mutation has been correlated with especially aggressive human glial tumors and it has been suggested that oncogenicity was due to reduced apoptosis in vitro and in vivo. It was examined whether inhibition of EGFR-mediated signaling in human glioblastoma cells by the mutant T691 stop neu receptor could sensitize cells to apoptotic cell death.

With prolonged serum suppression only 0-1% apoptosis was observed in parental U87MG cells by staining with 4'-6-diamidino-2-phenylindole (DAPI) or TUNEL staining, which was less than observed in other studies. Cells derived from U87MG were found not to show a sub-G0 peak by flow cytometric analysis after PI staining under apoptosis-inducing conditions, which is in agreement with other investigators. The expression of the inhibitor T691 stop neu in U87MG cells resulted in only 0-2% apoptosis with prolonged serum suppression, with determination by immunohistochemical identification of apoptotic nuclei with DAPI.

Apoptosis was maximal in studies repeated at 72h and this time was selected for all additional experiments. The expression of the T691 stop neu protein in the background of U87MG cells increased the level of radiation-induced apoptosis to 23 ± 7.9% (mean ± SEM) at 72h in four independent experiments in complete growth medium (Figure 2A). . Prolonged serum suppression combined with radiation resulted in 33 ± 10.6% apoptosis in U87 / T691 cells and 11 ± 1.5% apoptosis in parental U87MG cells, a comparable increase in both populations over that seen with cell radiation. in full growth medium. Experiments were performed including morphological evaluation of apoptosis in human glioma cells after irradiation with gamma rays. All cells were stained with DAPI at 72h after being exposed to gamma radiation. Nuclei showing apoptotic morphology were observed. The morphological changes of nuclear exudates ("blebbing") and fragmentation, which are characteristic of apoptosis, are shown by immunohistochemical analysis of culture cells derived from U87MG with DAPI staining. Apoptotic indices represent an underrepresentation of total cell death after radiation in U87MG / T691 cells since the inventors were unable to examine floating cells immunohistochemically.

IS 2 270 586 T3

Clonogenic survival of human glioblastoma cells subjected to radiation. The inventors measured the number of cells that escape growth arrest or death and are capable of continuing to form a colony, an assay commonly used to determine radiosensitivity. In certain cases, clonogenic growth assays have not correlated with sensitivity to chemotherapy radiation, presumably since the fate of dead or stably disrupted cells is not determined in this assay. As shown in Figure 3, U87 / T691 cells showed increased sensitivity to radiation over a range of radiation concentrations (2-10Gy). U87 / T691 cells were approximately one-half log more sensitive to radiation than their non-transfected parental counterparts for all radiation doses tested. These results were confirmed with additional subclones expressing T691 stop neu. U87MG cells and their derivatives contain wild-type p53 and p21 protein.

Relationship of radiation sensitivity of human glioblastoma cells to p53 status

The p53 state has been shown to influence the response to ionizing radiation in a number of transformed and non-transformed cell types. A p53 induction assay was carried out in human glioblastoma cells after gamma irradiation. 10 were applied⁵ U87MG and U87 / T691 cells containing a wild-type p53 gene product and gamma-irradiated (10Gy) after overnight docking. Lysates were collected at the indicated times after radiation, subjected to SDS-PAGE and immunoblotted with an antibody reactive with p53. Control cells were MCF7 breast cancer cells containing immunoreactive p53 protein. A more energetic induction of the p53 protein was observed continuously at 12h after gamma-ray radiation in U87 / T691 subclones. Westerm analysis of cell lysates obtained at different time points after radiation treatment indicated persistent increases in p53 protein levels detected at all times between 6-72h after radiation, in U87MG cells and in their corresponding T691 stop neu transfected derivatives. . The zero time point indicates cells that were gamma irradiated and immediately lysed for analysis. The densities of p53 were comparable, at this time, to the falsely irradiated cycling cells. A 10-fold increase in p53 density was observed in U87 / T691 cells 12h after radiation, compared to only 1.5 to 3-fold increases in both U87MG and U87 / T691 cells at any other time points examined. This trend was observed continuously (four experiments) and was observed in U87 / T691 cells early, even at 6h after radiation in some experiments, and suggests that p53-dependent signaling pathways may be more efficiently activated. by deactivation of EGFR in the presence of genomic damage. Alterations in critical point proteins regulated by p53 have been observed 12h after induction of genomic damage by gamma irradiation. Growth inhibition and differentiation of human breast cancer cells, after erbB receptor binding, have been associated with activation of a p53-dependent pathway.

P21 was induced in both U87MG and U87 / T691 cells after radiation treatment, with the highest levels observed 24h after radiation exposure in both cell lines. In both U87MG and U87 / T691 cells the density of the p21 protein 6-24h after radiation was comparable. While other researchers have suggested that upregulation of bcl-x_L is associated with reduced apoptosis in human glioma cells, the inventors did not detect changes in the expression of the bcl-x protein_L after radiation in U87MG and U87 / T691 cells. Both constitutive and radiation-induced levels of bcl-x_L they were equivalent in U87MG and U87 / T691 cells. Examination of bax and bcl-2 protein levels revealed no differences between glioblastoma cells and their inhibited subclones.

Apoptosis in human glioblastoma cells with p53 mutation

U373MG human glioma cells contain a mutated p53 gene product, are deficient in p21 expression, and show comparable elevation of EGFR to U87MG surface cells by flow cytometric analysis. These cells were used to determine whether the apoptosis observed after inhibition of EGFR-mediated signaling and gamma irradiation was dependent on a wild-type p53 protein. U373MG cells showed increases in the levels of a mutated p53 protein following gamma ray irradiation, but did not express p21 constitutively or after radiation treatment.

The mutant T691 stop neu receptor was expressed in U373MG glioma cells and confirmed expression comparable to U87 / T691 cells in four U373 / T691 subclones by metabolic labeling and flow cytometric analysis. Flow cytometry indicated that the surface levels of the mutant T691 stop neu receptor were equivalent in subclones U87 / T691, U373 / t691 cl 1 and U373 / T691 cl 12, and two derivatives of U373MG expressing T691 stop neu. U373MG derivatives expressing the mutant T691 stop neu receptor were capable of growth disruption in low serum content and showed an interrupted meadow of confluent cells without the development of morphologically transformed foci in vitro, indicating that the wild-type proteins p53 and p21 they were not required to disrupt growth or inhibit transformation of glioma cells in which erbB signaling had been disabled. Subclones U373 / T691 cl 1 and U373 / T691 cl 12 were then irradiated together with U373MG cells and showed increased levels of apoptosis relative to their parental equivalents (Figure 2B). In the representative experiment shown, two U373 / T691 subclones showed respectively 32% and 59% apoptosis 72h after gamma ray radiation, compared to 2% apoptosis in U373MG parental cells and 20% apoptotic index in U87 / T691 cells. Deactivation of EGFR signaling by expression of T691 stop neu in two different human glioma cell lines, containing differences in p53 and p21 states, resulted in increased radiation-induced apoptosis. Glio32 cell sensitization

ES 2 270 586 T3 human blastoma to genomic damage can therefore occur in the absence of wild-type p53 and p21 proteins. Taken together, these data suggest that both p53-dependent and p53-independent pathways may mediate sensitization to cell death induced by a combination of transreceptor inhibition and genomic damage. It should be noted that human glioblastoma cells in which EGFR signaling is disabled do not appear to be more sensitive to prolonged serum suppression or tumor necrosis factor α-mediated cell death than parental cells.

Commentary

Specific inhibition of EGFR signaling that inhibits cell growth and transformation also sensitized radio-resistant human glioma cells to radiation-induced genomic damage. Glioblastoma cells expressing a transdominant mutant receptor, derived from p185 neu, showed greater G1 phase disruption and higher levels of apoptosis after radiation than their parental counterparts. In mammalian fibroblasts and specialized neuronal cells, suppression of serum or growth factor can lead to apoptosis under certain conditions. Prolonged serum suppression alone did not induce apoptosis in human glioblastoma cells in these studies. Inducing apoptosis required DNA damage combined with inactivation of erbB receptor signaling or serum suppression. Apoptosis was induced by radiation in 23% of U87MG derivatives and in 32-59% of U373MG-derived subclones in which EGFR had been inactivated (compared to only 1-2% in parental cells) in complete growth medium, indicating that inhibition of EGFR signaling by transreceptor inhibition could not be overcome by growth in serum. Serum suppression combined with radiation damage increased the observed levels of apoptosis in U87MG cells and in human glioblastoma derivatives expressing T691stop neu to the same degree. Notably, after DNA damage the apoptosis observed by inactivation of erbB receptor signaling on the cell surface was greater than that observed with serum suppression.

Monitoring systems or checkpoints have been developed to interrupt cell cycling when genome or mitotic spindle damage has occurred. The DNA damage checkpoint functions differentially at different stages of the cell cycle and requires the coordinated action of multiple pleiotropic gene products involved in growth disruption, DNA repair, activation of transcription, and apoptosis. DNA damage checkpoints are signal transduction pathways that communicate information from damaged DNA to components of the cell cycle. The data presented in these studies show that receptor tyrosine kinase (RTK) mediated signaling events can influence DNA damage checkpoint signaling pathways. In particular, EGFR inhibition in malignant human glioma cells can increase the degree of growth arrest and apoptosis observed after X-ray induced DNA damage.

The resistance of gamma-irradiated cells is affected by the functional status of different oncogens. Oncogenic Ras or Raf expression decreases radiosensitivity in NIH3T3 cells and Ras expression^H plus either c- or v-myc oncogenes conferred resistance to rat embryo fibroblasts exposed to gamma ray radiation. It is also true that the expression of different oncogenes can sensitize cells to apoptosis on exposure to low serum or anticancer agents. The division delay that occurs in both Gland G2 phases of the cell cycle is influenced by the expression of dominant oncoproteins, such as H-ras. Expression of the wild-type protein p53 has been associated with decreased survival after gamma-ray irradiation due to the induction of a higher fraction of apoptosis on cells containing a mutated p53 protein. However, tumor cells containing a mutated p53 protein and proliferating lymphoid cells derived from p53 - / - mice have been shown to undergo apoptosis after radiation, suggesting p53-independent mechanisms of cell death after genomic damage.

The factors that mediate the degree of growth arrest with respect to apoptosis observed after DNA damage in a specific cell type have not been defined, and the cell-specific factors that influence the detection of DNA damage, cell recovery and the decision to apoptosis are not fully understood. P53-dependent mechanisms may influence the response of inhibited glioma cells to undergo relative growth arrest and / or apoptosis. The results in U373MG-derived cells also indicate that apoptotic cell death that occurs after genomic damage in transformed human cells in which EGFR signaling is inhibited involves mechanisms that are distinct from the p53 and p21 proteins. The p21 - / - mice develop normally and do not appear to have defects in programmed cell death required for normal organ development, indicating that p21 is not likely to be necessary for apoptosis. The p53 - / - mice show genetic instability and contain high levels of c-myc. These mice undergo high levels of apoptosis in vivo, indicating that p53-independent apoptosis mechanisms are functional in both normal tissues and transformed cells.

Interestingly, recent work demonstrates that the absence of p21 in isogenically balanced colorectal carcinoma cells resulted in reduced growth arrest compared to positive p21 derivatives of the same cell line, and this has been correlated to growth tumor more inhibited in vivo. These observations were attributed to increased apoptosis due to defects in p21-mediated checkpoint growth disruption, although the increased tendency to apoptosis by p21 - / - cells was not directly demonstrated in this work. It was suggested that apoptosis induction was preferable to growth arrest in anticancer therapy in vivo. In the inventors' studies, unlike those of

IS 2 270 586 T3

Waldmann et al. (1997), there was a correlation between apoptosis, increased growth arrest, and reduced clonogenic survival after radiation.

In certain circumstances, particularly in cancer cells, apoptosis after genomic damage can be promoted if there are defects in pathways that mediate growth disruption. Additionally, even where cells are capable of undergoing growth arrest and apoptosis, such as in p21-deficient human U373MG cells in which EGFR signaling has been turned off, the cells can be induced to apoptosis afterwards. of certain signals, such as radiation, presumably by activation of specific pathways. Our data indicates that the relative proportion of growth disruption or apoptosis induced by genomic damage is influenced both by the integrity of specific checkpoints and by alterations in erbB signaling pathways. Notably, modulation of RTK signaling pathways can influence checkpoint results after DNA damage in transformed cells. Other investigators have shown that activation of erbB signaling pathways in breast cancer cells contributes to radio resistance, suggesting that erbB family signaling pathways influence the response to DNA damage in many tumor types. By combining biological signaling inhibition with agents capable of specifically inhibiting receptor oncoproteins of the tyrosine kinase family, one may be able to influence the kinetics of tumor cell response to standard cytotoxic agents. The timing of the administration of cytotoxic therapies can be optimized in such combination therapies and these data require that the selective antitumor effects of currently existing anticancer treatments could be improved, even in the treatment of advanced human malignancies, containing alterations in multiple pathways. checkpoint signal transduction.

Example 6

Recombinant adenovirus

Recombinant adenoviruses according to some embodiments of the present invention are inactivated by elimination of Ela and Elb, which are normally required to activate the expression of other viral genes. These recombinant adenoviruses, which are based on human Ad5, are capable of efficient gene transduction in human hepatocytes without apparent cytopathic effects or expression of adenoviral proteins. The recombinant adenovirus construct that can be adapted for use in the present invention is described in Kozarsky et al., 1993 Somatic Cell and Molecular Genetics 19 (5): 449-458, which is incorporated herein by reference. Said reference shows recombinant adenoviruses with an IacZ insert or plug. The IacZ insert or dummy may be substituted by sequences encoding erbB dimerizing proteins, deficient in tyrosine kinase, according to the invention. The gene construct of the invention is inserted into a linker sequence at the site of the IacZ insert or plug.

Materials and methods

Recombinant Adenoviruses. The plasmids used to generate the recombinant, adenovirus Ad.CBlacZ and Ad.CBhLDLR with E1 deletion were constructed as follows. The plasmid CMVj6AIacZ (10) was digested with SnaB1 and Nhel, and gagJAhLDLR (16) was digested with NheI and then partially digested with XhoI to isolate a fragment containing the promoter / J-aclin and the gene IacZ, or the human LDL receptor cDNA. These fragments were blunted at their ends by Klenow. The plasmid pAdCMV-lacZ (17) was digested with SnaBI and NotI to eliminate the CMV promoter and IacZ gene (retaining the CMV promoter), blunt-ended by Klenow, and ligated with inserts containing the / l- promoter. actin fused to either the IacZ or LDLR genes. The resulting vectors were designated respectively pAdCBlacZ and pAd-CBhLDLR.

Plasmids were linearized with NheI and co-transfected into 293 cells with wild-type adenoviral DNA (sub 360 (18) strain containing partial E3 deletion) that had been digested with XbaI and GlaI to remove 5'ITR. Recombinant adenoviruses were isolated after transfection (19), subjected to two rounds of plaque purification, and lysates were purified by cesium chloride centrifugation (20). Viral materials were evaluated for concentrations with dilution-limiting plaque assay in 293 cells and stored at 20 ° C after four-fold dilution with 10 mM Tris Cl, pH 8.1, 100 mM NaCl, 0.1 bovine serum albumin. %, and glycerol 50%. The concentrations of the glycerol materials were: Ad.CBlacZ, 2.4 x 10⁹ plaque forming units (PFU) / ml; Ad, CvhLDLR, 4x10⁹ PFU / ml; wild type Ad, 8x10⁹ PFU / ml.

Gene constructs according to the invention are inserted in place of the IacZ sequences in a linker sequence in the plasmid.

Example 7

Recombinant adenovirus

Recombinant adenoviruses according to some embodiments of the present invention are inactivated by deletion of the E1 and E4 genes, which are normally required to activate the expression of other viral genes. These recombinant adenoviruses, which are based on human Ad5, are capable of highly efficient transduction of genes into

ES 2 270 586 T3 human hepatocytes with no apparent cytopathic effects or expression of adenoviral proteins. The construction of recombinant adenoviruses that can be adapted for use in the present invention is described in WO96 / 3953. This reference indicates recombinant adenoviruses with an IacZ insert. The IacZ insert may be substituted for sequences encoding tyrosine kinase deficient erbB dimerizing proteins, in accordance with the present invention. The gene construct of the invention is inserted into a linker sequence at the site of the IacZ insert.

Figure 4 is a schematic map of recombinant adenoviruses H5.001CBLacZ, with indicated restriction enzyme endonuclease sites. The streaked bars represent the CBLacZ minigene; the black bar represents the Ad5 viral nucleus, the hatched bar represents Ad E4 deletion.

The new envelope cell lines allow the production of recombinant adenoviruses with functional deletion in the E1 and E4 genes.

Primitive region 4 (E4) of adenovirus serotype 5 consists of 7 open reading frames (ORFs) that are believed to be involved in viral DNA replication, host cell shut-off, and accumulation. late mRNA. To generate recombinant adenoviruses (Ad) deleted at E4, the function of the E4 region must be delivered to the recombinant virus by a helper virus or an envelope cell line.

To avoid this problem, the envelope cell line contains the Ad5 E1 gene rather than only the ORF6 of the Ad5 E4 gene. The E4 ORF6 alone can satisfy the requirements for E4 in the viral life cycle. ORF6 is further preferable in the transcriptional control of an inducible promoter, such as the zinc-inducible sheep metallothionin promoter or the glucocorticoid-inducible mouse mammary tumor virus (MMTV) promoter, particularly, dexamethasone.

After the desired translation vector "shuttle" containing the adenoviral sequences has been transfected into the cell line, the expression of E4 ORF 6 can be induced by the appropriate inducer.

In a preferred form, the envelope cell line is a 293 E1 human embryonic kidney (HEK) expressing the cell line into which the E4 ORF6 sequence is introduced under the control of an inducible promoter. The glucocorticoid inducer MMTV promoter is preferable at present because the zinc sulfate inducer or the MT promoter can be toxic to cells.

The specific indication of the construction of envelope cell lines containing only the ORF6 region of Ad5 E4 or, for the purpose of functional comparisons, the entire E4 region is explained below. Briefly described, the entire E4 region and an ORF6 sequence of the Ad E4 gene are obtained by known techniques (see, for example, Sambrook et al., "Molecular Cloning. A Laboratory Manual. ”, 2d edit., Cold Srping Harbor Laboratory, New York (1989) and references cited there). To isolate the ORF6 region, the anchored polymerase chain reaction technique was used to amplify the ORF6 sequence from its start codon to the terminal codon. Primers selected from the published ORF6 sequence are used to amplify the ORF sequence and insert restriction sites at the end of the sequence. The entire E4 gene sequence including the E4 ORF6 sequence is published in the Ad5 Genbank sequence (Genback Accession No. M73260).

A minigene has been constructed that has placed the ORF6 sequence under the control of a selected promoter. The ORF6 sequence gene is operably linked to regulatory components in a manner that allows its transcription. These components include conventional regulatory elements, such as a promoter to activate ORF6 expression. An inducible promoter was an inducible sheep metallothionin (MT) Zn + promoter.² (MG Peterson et al., Eur J. Biochem., 174: 417-424 (1988)). The second promoter is the dexamethasone-inducible mouse mammary tumor virus (MMTV) promoter.

The polyA sequence used in the MMTV-ORF6 minigene is supplied by the growth hormone gene terminator and an SV40 origin of replication.

The ORF6-containing minigene is subcloned into a pBR322-based displacement plasmid containing a neomycin resistance gene, resulting in a shuttle vector.

E1 / E4 ORF6s expressing envelope cell lines are useful in the generation of recombinant E1 / E4 deleting adenoviruses.

Recombinant adenoviruses

The new E1 / E4 expression cell line is useful in the further construction of recombinant adenoviruses with E1 / E4 deletion containing any selected transgene. Recombinant adenoviruses are capable of delivering a suitable gene to mammalian cells and tissues. These recombinant adenoviruses are functionally killed at least in the E1a, E1b and E4 Ad gene regions. By the term "functionally deleted" it is meant that a sufficient amount of the gene region is deleted or otherwise damaged, for example by mutation or modification, such that the gene region is no longer capable of producing the products of gene expression. If desired, the entire gene region can be removed.

IS 2 270 586 T3

Similarly, the methods used for the selection of useful viral sequences in a vector, the cloning and construction of the "minigene" and its insertion into a desired viral displacement vector and the production of a recombinant infectious virus, are within the scope of the technical capability provided by the subject matter disclosed in this description.

Construction of the "minigene" containing the transgene

A minigene in this context is defined as indicated above, except that the components of this minigene are designed to express the product gene in vivo. These components include standard regulatory elements necessary to activate expression of the transgene in a cell transfected with the recombinant virus. For this minigene, a selected promoter is operably linked to the transgene and located, with other regulatory elements, within the selected viral sequences of the recombinant vector. Promoter selection is a routine matter and is not a limitation of the present invention. Useful promoters can be constitutive promoters or regulated (inducible) promoters that will allow control of the amount of transgene to be expressed. For example, a desirable promoter is that of the cytomegalovirus (CMV) immediate promoter / amplifier (see, eg, Boshart et al., Cell, 41: 521-530 (1985)).

Another desirable promoter includes the Rous sarcoma virus LTR promoter / amplifier. Another promoter / amplifier sequence is the chicken / t-actin (CB) cytoplasmic promoter (TA Kost et al., Nucl. Acids Res., 11 (23): 8287 (1983)). Other suitable promoters can be selected by one of ordinary skill in the art.

Recombinant adenovirus production

Adenovirus sequences useful in this invention can include DNA sequences from a number of types of adenovirus available from Genbank, including the Ad5 type (Genbank Accession No. M73260). Adenovirus sequences can be obtained from any known adenovirus serotype, such as serotypes 2, 3, 4, 7, 12, and 40, and including any of the 41 currently identified human types.

Similarly, known adenoviruses can also be used in the vector constructs of the present invention to infect other animals. Selection of the adenovirus type is not anticipated to limit the following invention. A number of adenovirus strains are available from the American Type Culture Collection, Rockville, Maryland, or are available on request from a number of commercial and institutional sources. In the following exemplary embodiment, an adenovirus type 5 (Ad5) was used for convenience.

An adenovirus according to the present invention contains the functional deletion of the prior E1a gene (spanning mu 1.3 to 4.5) immediately prior to adenoviral and the lagging prior gene E1b (spanning mu 4.6 to 11.2). Similarly, adenovirus has a functional deletion of the E4 region (spanning mu 92 to 97.2), or at least ORF6 of the E4 region.

Exemplary recombinant adenoviruses for use in this invention can be obtained, for example, by homologous recombination of desired fragments of various recombinant adenoviruses, a technique that has been commonly used to generate other recombinant adenoviruses for use in gene therapy. The recombinant adenovirus, H5.001 CBLacZ, is constructed by homologous recombination between the viral core of adenovirus dl1004 (also H5dl1004) and the DNA of the pAdCBLacZ minigene. H5dl1004 is an Ad5 virus deleted from approximately map unit 92.1 to map unit 98, that is, substantially all of the E4 gene. The d11004 virus is described in Bridge and Ketner, J. Virol., 632 (2): 631-638 (February 1989).

The vector pAdCBLacZ is a cDNA plasmid containing Ad mu 0-1, an E1 deletion in which a bacterial / l-galaclosidase gene is inserted under the control of a chicken / t-aclin promoter, with other regulatory elements such as as described below, and flanked by Ad mu 9-16 and a plasmid sequence.

Novel E1a / E1b and E4 expressing envelope cell lines

Construction of plasmids pMTE4ORF6 expressing E4 ORF6

An example of a useful plasmid for the construction of an envelope cell line of the present invention is pMTE4ORF6, which contains a sheep metallothionin promoter (MT promoter) in control of the transcription of a human E4 ORF6 gene sequence, a terminator of growth hormone (GH), an SV40 origin of replication, plasmid sequences from plasmids based on pBR322 including the neomycin resistance gene, an SV40 polyadenylation site and an ampicillin resistance gene.

The different functional fragments of this plasmid can be easily substituted with other conventionally used sequences and are not critical for the design of the plasmid pMMTVE4ORF6.

Another example of a useful plasmid for the construction of an envelope cell line of the present invention is pMMTVE4ORF6, which contains a mouse mammary tumor virus (MMTV) promoter in transcriptional control of a human E4 ORF6 gene sequence, a terminator of growth hormone (GH), an SV40 origin of replication, plasmid sequences from plasmid pBR322, including a neomycin resistance gene, and

ES 2 270 586 T3 an ampicillin resistance gene. The different functional fragments of this plasmid can easily be substituted by other conventionally used sequences and are not critical to the design of the plasmid.

Endogenous E4 promoter pLTR.E4 (-)

A plasmid used as a control for the construction of an envelope cell line of the present invention is pLTR.E4 (-). This plasmid contains the constitutive retroviral MLV LTR and most of the Ad E4 gene region except that the endogenous E4 promoter and a part of E4 ORF1 are missing. The other plasmid sequences remain the same as described.

Endogenous E4 promoter pLTR.E4 (+)

Yet another plasmid is pLTR.E4, which contains the MLV LTR and the endogenous E4 promoter and an intact E4 gene. The other plasmid sequences remain the same as previously described.

Clone transfections and selections

Each of the plasmids described above was transfected by the calcium phosphate precipitation technique in a human kidney embryonic cell line 293 (ATCC CRL 1573) that expresses the product of the adenovirus E1 genes, seeded in 100 mm plates (10 pg of plasmid / plate). 24 hours after transfection the cells were harvested and seeded with varying dilutions (1:10 - 1: 100) in 100 mm dishes for about 10 days. The seeding media contains G418 (Geneticin, BRL) at a concentration of 1 mg / ml. Resistant colonies that developed were selected using the following assays and expanded. Preliminary clone analyzes were based on the improved transduction efficiency of a recombinant adeno-associated virus, AV.CMV-LacZ, and immunofluorescence localization of the Ad Er protein as indicated below. Assay of amplification of transduction of AV.CMBLacZ.

The E1 and E4 Ad gene products are required for recombinant adeno-associated virus (AAV) function. This primary assay involves seeding the envelope cell lines of Example 1 into 35 mm 96-well culture plates (2x10⁶ cells / well) and infecting cells with AV.CMVLacZ, heat treated and purified with a MOI of 1000 virus particles / cell.

AV.CMVLacZ Preparation

A recombinant AAV virus was prepared by conventional genetic engineering techniques for the purpose of this experiment. Recombinant AAV is generated by plasmid transfections in the presence of helper adenovirus (Samulski et al., J. Virol., 63: 3822-3828 (1989)). A cis-acting plasmid pAV.CMVLacZ is derived from psub201 (Samulski et al., J. Virol. 61: 3096-3101 (1987)) and contains E. coli β galactosidase minigene in place of AAV Rep and Cap genes. The 5 'to 3' organization of the recombinant AV.CMVLacZ genome (4.9 kb) includes (a) the 5'AAV ITR (bp 1-173) was obtained by PCR using pAV2 (CA Laughlin et al., Gene, 23: 65-73 (1983)) as a template;

(b) a prior immediate CMB amplifier / promoter (Boshart et al., Cell, 41: 521-530 (1985));

(c) an SV40 intron;

(d) E. coli beta-galactosidase cDNA;

(e) an SV40 polyadenylation signal (a 237 Bam Hi-Bc1I restriction fragment containing cleavage / poly-A signals from both pre and post transcription units; and (f) 3'AAV ITR, obtained from pAV2 as a fragment SnaBI-Bg1II Rep and Cap genes are arranged by the trans-acting plasmid pAAV / Ad.

Monolayers of 293 cells grown to 90% confluence are infected in 150 mm culture dishes (5x10⁷cells / dish) with H5.CBALP at an MOI of 10. H5.CBALP (also called H5.010ALP) is a recombinant adenovirus that contains an alkaline phosphotase minigene in place of adenovirus E1A and E1b gene sequences (units of map 1-9.2 of the GenBank Ad5 sequence (Accession No.M73260)). The alkaline phosphotase cDNA is under transcriptional control of the CMV-amplified ^ -actin promoter in this virus. This helper virus is described in Goldman et al., Hum. Gene Ther., 6: 839-851 (July, 1995); Engelhardt and others, Hum. Gene Ther., 5: 1217-1229 (October, 1994); and references cited in this description.

Infections are performed in Dulbecco's Modified Eagles medium (DMEM) supplemented with 2% fetal bovine serum (FBS) at 20 ml of medium / 150 mm dish. Two hours after infection, 50 pg of plasmid DNA (37.5 pg transactional and 12.5 pg cis-acting) in 2.5 ml of transfection cocktail is added to each dish and distributed evenly. Transfections are based on calcium phosphate, as described (B. Cullen, Meth. Enzymol., 152: 684-704 (1987)). The cells are left in this state for 10-14 hours, after which

ES 2 270 586 T3 the infection / transfection medium is replaced with 20 ml of fresh DMEM / 2% FBS. Forty to fifty hours after transfection, cells suspended in 10 mM Tris-Cl buffer (pH 8.0) (0.5 ml / 150 mm dish) are collected and a lysate is prepared by sonication. The lysate is brought to 10 mM manganese chloride, after which bovine pancreatic Dnase I (20,000 units) and Rnase (final concentration 0.2 mg / ml) are added, and the reaction is incubated at 37 ° C for 30 minutes. Sodium deoxycholate is added to a final concentration of 1% and incubated at 37 ° C for an additional 10 minutes.

The treated lysate is chilled on ice for 10 minutes and solid CsCl is added to a final density of 1.3 g / mL. The lysate is brought to a final volume of 60 ml with a solution of 1.3 g / ml CsCl in 10 mM Tris-Cl (pH 8.0) and divided into three equal aliquots. Each of the 20 ml samples is layered over a stepped CsCl gradient composed of two 9.0 ml fractions with densities of 1.45 g / ml and 1.60 g / ml.

Centrifugation is carried out at 25,000 rpm in a Beckman SW-28 rotor apparatus for 24 hours at 4 ° C.

Fractions containing maximum concentrations of functional AV.CMVLacZ virus are pooled and subjected to three sequential rounds of equilibrium sedimentation in CsCl. The rotor selection includes a Beckman NVT-90 (80,000 rpm for 4 hours) and SW-41 (35,000 rpm for 20 hours). At equilibrium, AV.CMVLacZ appears as an opalescent band at 1.40-1.41 g / ml CsCl. Densities are calculated from refractive index measurements. The purified vector is exchanged to 20 mM HEPES buffer (pH 7.8) containing 150 mM NaCl (HBS) by dialysis and stored frozen at -80 ° C in the presence of 10% glycerol or as liquid material at -20 ° C. in HBS / glycerol 40%.

Purified virus is checked for contamination of H5.CBALP helper and AV.CMVLacZ virus concentrations. Helper virus is checked by histochemical staining for reporter alkaline phosphatase activity. A purified virus sample representing 1.0% of the final product is added to a growing layer of 293 cells seeded in a 60 mm plate. Forty-eight hours later, cells are fixed in 0.5% glutaraldehyde / phosphate buffered saline (PBS) for 10 minutes at room temperature, washed in PBS (3x10 minutes), and incubated at 65 ° C for 40 minutes to inactivate endogenous alkaline phosphatase activity. The monolayer is allowed to cool to room temperature, washed once briefly in 100 mM Tris-Cl (pH9.5) / 100 mM NaCl / 5 mM MgCl, and incubated at 37 ° C for 30 minutes in the same buffer containing 0 , 33 mg / ml of nitro blue tetrazolium chloride (NBT) and 0.165 mg / ml of 5-bromo-4-chloro-3-indolyphosphate p-toluidine salt (BCIP). Color formation is stopped by washing the monolayer in 10 mM Tris-Cl (pH 8.0) / 5 mM EDTA. Routinely, the purification scheme described above removes all detectable H5.CBALP helper virus in the third round of increasing density ultracentrifugation.

AV.CMVLacZ concentrations are measured according to genome copy number (virus particles / ml), absorbance at 260 nm (A₂₆₀ particles / ml) and Formation of LacZ Units (LFU / ml). Virus particle concentrations are based on Southern blotting. Briefly, a purified AV.CMVLacZ sample is treated with capsid digestion buffer (50 mM Tris-Cl, pH 8.0 / 1.0 mM EDTA, pH 8.0 / 0.5% SDS / Proteinase K 1.0 mg / ml) at 50 ° C for one hour to release virus DNA. Reactions are allowed to cool to room temperature, loading dye added and electrophoresed with 1.2% agarose gel intermediate. Standard amounts of the AV.CMVLaCZ genome were resolved on the gel.

DNAs are electroblotted onto a nylon membrane, hybridized with random primer-labeled restriction fragment ³²P, and the resulting blot scanned on a Phosphorlmager 445 SI (Molecular Dynamics). A standard curve is generated from the duplex forms and is used to extrapolate the number of virus genomes in the sample. LFU concentrations are generated by infecting indicator cells with limiting dilutions of virus samples. Indicator cells include HeLa and 293. Twenty-four hours later, the cells are fixed in glutaraldehyde and the cells are subjected to histochemical staining for E. coli jd-galactosidase (LacZ) activity, as described in JM Wilson et al., Proc. Natl. Acad. Sci. USA, 85: 3014-3018 (1988). An LFU is described as the amount of virus sufficient to cause visually detectable expression of jd-galactosidase in a cell 24 hours after infection.

ORF6 Expression Induction

The induction of expression of ORF6 with 10 pM dexamethesone or 150 pM zinc sulfate (for negative control, without using inducer) is started 2 hours before the addition of virus and is continued throughout the experiment. Twenty-four hours after virus addition, cells are harvested, lysates generated by sonication, and analyzed for expression of jd-galactosidase (i.e., jd-galactosidase activity) and virus DNA, as described. previously. Hirt extracts are prepared for low molecular weight DNA from cell extracts. The preparation of the Hirt extracts and subsequent Southern hybridization analysis are carried out using standard procedures known to those skilled in the art.

In the absence of inducers, envelope cell lines generate low levels of jd-galactosidase in rAAV-infected cells. Induction of ORF6 expression with the inducer dexamethasone results in a concomitant increase in transduction of AV.CMVLacZ cells to a level well above that of the parental 293 line. E1 expression alone is insufficient to have an effect on the increase of rAAV transduction, mediated by adenovirus. Immunofluorescence localization of Ad5 Posterior Protein.

IS 2 270 586 T3

Positive clones from the assays are infected with the E4 deleting recombinant adenovirus H5DI1004 and screened for E4 complementation using an immunofluorescence assay for subsequent gene expression. The H511004 virus was obtained from John Hopkins University Dr. Ketner and is described in Bridge and Ketner, J. Virol., 632 (2): 631-638 (Feb. 1989), which is incorporated into this description. as reference. Since E4 ORF6 complements the expression of the subsequent Ad gene, specifically the formation of adenovirus hexon and penton fibers, ORF6-containing cell lines are capable of antibody binding against these proteins.

Each of the cell lines is infected with the H5d11004 virus with elimination of E4 at a MOI of 0.1. Cells are treated with mouse anti-adenovirus FITC-labeled monoclonal antibody to hexon or penton fibers at a 1:10 dilution (Chemicon International Inc., Temecula, CA). Positive clones are identified by reaction with the antibody.

Relative Plating Efficiency

Cell lines demonstrating strong complementation ability are screened for the relative plating efficiency of HSd11004 compared to W 162 cells (Vero cell line complementing E4 that does not express E1) (again Weinberg and Ketner, Proc. Natl. Acad Sci., USA, 80 (17): 5383-5386 (1983)). RPE%, ie the relative plaque formation efficiency, is determined by plotting the concentration of H5dl1004 in cell lines / concentration of H5dl1004 on W 162 cells. For example, the RPE of 293 cells is 0.

The positive cell lines selected by all the criteria are identified in the following Table I, with the results of the tests.

TABLE I

<td colspan="6">Double Complementary Cell Lines E1 / E4</td>

<td>Cellphone line</td><td>Transgen</td><td>Promoter</td><td>IF / LP</td><td>AV.CMV LacZ</td><td>RPE%</td>

<td> 293-10-3</td><td>ORF6</td><td>MT</td><td> ++++</td><td> ++++</td><td> 246</td>

<td> 293-39-11</td><td>ORF6</td><td>LTR</td><td> ++++</td><td> +++</td><td> 52</td>

<td> 293-84-31</td><td>E4-</td><td>LTR</td><td> ++++</td><td> ++++</td><td> 179</td>

<td> 293-12-31</td><td>E4 full</td><td>LTR + E4</td><td> ++++</td><td> ++++</td><td> 174</td>

<td> 293-27-6</td><td>ORF6</td><td>MMTV</td><td></td><td> +++++</td><td> 327</td>

<td> 293-27-17</td><td>ORF6</td><td>MMTV</td><td></td><td> ++++</td><td> 313</td>

<td> 293-27-18</td><td>ORF6</td><td>MMTV</td><td></td><td> ++++</td><td> 339</td>

<td> 293-27-28</td><td>ORF6</td><td>MMTV</td><td></td><td> ++++</td><td> 261</td>

Construction and purification of H5.001 CBLacZ

The plasmid pAd.CBLa6Z is constructed as described in detail in K. Kozarsky et al., Som. Cell Mol. Genet., 19 (5): 449-458 (1993), which is incorporated herein by reference. This plasmid contains a minigene comprising a 5 'flank NheI restriction site, followed by Ad5 mu 0-1 sequence, followed by E1 deletion into which is inserted the CMV amplifier / chicken ^ -actin promoter sequence (TA Kost et al., Nucl. Acids Res., 11 (23): 8287 (1983)), which controls the transcription of the next bacterial ^ -galactosidase, followed by a poly A sequence and with 3 'flank Ad mu 9-16 and another NheI site. In the plasmid, the minigene is flanked on both sides by plasmid sequence containing drug resistance markers.

The plasmid pAd.CBLacZ is linearized with NheI and co-transfected by the calcium phosphate co-transfection method in the new envelope cell line with H5dl1004 digested by Clal (Ad5 sequence deleted from map unit 92.1 to map unit 98, corresponding substantially the entire E4 gene).

Homologous recombination takes place in the cell line between these two viral constructs between Ad 9-16 map units, resulting in recombinant adenovirus, designated H5.001 CBLacZ. This recombinant adenovirus contains the sequence from approximately nucleotide 1 to 4628 of pAd.CBLacZ and Ad5 map units 9-92.1 and 97.3 to 100 of H5d11004. This recombinant adenovirus is therefore subjected to functional deletion and substantial structural deletion of the Ad E1 and E4 genes.

IS 2 270 586 T3

Viral plaques are selected and screened by the S-galactosidase assay, and H5.001CBLacZ was isolated after three rounds of plaque purification. The purified virus is also subjected to cesium chloride density centrifugation and mass production. For the following mouse experiments, the virus is used after column purification, and glycerol is added to a final concentration of 10% (v / v). The virus is stored at -70 ° C until use. Growth kinetics of H5.001 CBLacZ in envelope cell lines.

The cell lines described above are infected with recombinant H5.001CBLacZ at a MOI of 0.5. The maximum viral yield is indicated in the following table II in the form of LFU / ml.

TABLE II

<td>Cellphone line</td><td>Maximum viral yield</td>

<td> 293-10-3</td><td>2.8 x 10¹⁰</td>

<td> 293-39-11</td><td>9.5 x 10⁸</td>

<td> 293-84-31</td><td>1.1 x 10⁹</td>

<td> 293-12-31</td><td>4.5 x 10⁸</td>

<td> 293-27-6</td><td>2.8 x 10¹⁰</td>

<td> 293-27-17</td><td>2.5 x 10¹⁰</td>

<td> 293-27-18</td><td>2.9 x 10¹⁰</td>

<td> 293-27-28</td><td>1.2 x 10¹⁰</td>

When cultured in 293-27-18 cells (E4 ORF6 cell line with dexamethasone-inducible MMTV promoter), the maximum yield of this virus is 2.9 x 10¹⁰ LFU / ml. Several of the cell lines are passed between 5 and 20 times, and the viral production of the passages remained stable. However, the RPE decreased after repeated passages of cells.

The gene constructs, according to the invention, are inserted in place of the lacZ sequences in a linker sequence in the plasmid.

Example 8

Background

Monoclonal anti-p185 antibodies have been shown to^neu (mAbs) inhibit the growth of tumors expressing p185^neu in a dose-dependent manner in vitro and in vivo. Anti-p185 combinations^neu Reactive mAbs with distinctive epitope domains revealed a synergistic inhibitory effect on neu overexpression of tumors in vivo. These studies have demonstrated the potential of mAb-based specific oncoprotein therapies.

The three-dimensional structure of the antigen-antibody complexes reveals that the binding site is defined by 6 hypervariable complementarity determining regions (CDR) with a loop structure (Peterson and Greene, 1994), but the specificity of the interactions is conferred by the loops. CDR3. The prerequisite for the conformation and structure of CDR loops can be mimicked by small peptides when the sequence and structure of anti-receptor antibodies are known.

Inhibition of cell growth and transformation can be achieved in transformed glial cells by modulating erbB receptor signaling. Recent studies by the inventors indicate that the induction of apoptosis may be the basis of a successful therapy for human cancer. Radiation resistant human glioblastoma cells, in which erbB receptor signaling was inhibited by transfection of truncated neu T691, exhibited increased growth disruption and apoptosis in response to DNA damage. Inhibition of erbB signaling is a potent stimulus for the induction of apoptosis. Proximate receptor interactions between erbB receptor members therefore influence activated cell cycle checkpoint pathways in response to DNA damage. Therefore, erbB receptor inactivation may improve the response to gamma radiation and other cytotoxic therapies.

The data suggest that radioresistant human tumor cells, which require a complete erbB signaling pathway, can be rendered radiation sensitive and may be an apoptotic pathway to any DNA damage by inhibition of the erbB signaling pathway.

IS 2 270 586 T3

Anti-p185 mimetic CDR4D5 designed^c-erbB-2, and was developed and used to investigate whether anti-p185-derived mimetic CDR4D5^c-erbB-2 Ab can inhibit the proliferation of human tumor cells and increase apoptosis after gamma-ray irradiation. The experiments carried out are described below.

Material and method

1. Peptidomimetic design

An anti-erbB2 antibody, 4D5 has been shown to be effective in demodulating the erbB2 receptor. The crystal structure of the humanized antibody (1FVD) is analyzed. The CDR3 of 4D5 was used as a matrix. Several cyclic peptide analogs were generated. Peptides that can be used include:

SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO: SEQ ID NO:

FCGDGFYACYMDV-CONH2

FCDGFYACYMDV-CONH2

FCDPFYACYMDV-CONH2

FCPDGFYACYMDV-CONH2

FCDPPFYACYMDV-CONH2

FCDGFYACYMDV-CONH2

FCDPFYACYMDV-CONH2

FCDPPFYACYMDV-CONH2

FCGDGFYACYMDV-COOH

FCDGFYACYMDV-COOH

FCDPFYACYMDV-COOH

FCPDGFYACYMDV-COOH

FCDPPFYACYMDV-COOH

FCDGFYACYMDV-COOH

FCDPFYACYMDV-COOH

FCDPPFYACYMDV-COOH

2. Cell lines

The following human tumor lines expressing varying levels of the p185 receptor were used^{c erbB 2}: a) U87MG (p185^c-erbB-2 not detectable) b) parental U373MG and U373 / T691 transfected T691 expressing p185^c-erbB-2low-moderate, c) MCF7 expressing p185^c-erbB-2 moderate, and SKBR3 (high level of p185^c-erbB-2).

3. Surface flow cytometric analysis of c-erbB-2 receptor expression

Subconfluent cells were harvested by rapid trypsin treatment (&lt;3 min), and kept on ice. The cells were washed and resuspended in FACS buffer (PBS with 0.5% BSA and 0.1% NaAzide) at an approximate concentration of 2 x 10⁶ cells / ml, then incubated with the primary reagents (anti-p185^c-erbB-2 Ab) and secondary (anti-IgG-FITC) for 30 min, each at 4 ° C, with two washes between each step. After staining, cells were resuspended in FACS buffer and analyzed immediately. Flow cytometric analysis was carried out on a Becton-Dickinson FACScan. P185 cell lines^c-erbB-2 Positive were determined by the difference in mean channel fluorescence between cell lines stained with anti-receptor antibody and the corresponding cell lines stained with the secondary reagents alone (anti-IgG-FITC). The homogeneity of the p185 clones^c-erbB-2 Positive staining was determined by the peak of the cell population with positive staining around its axis. On FACS histograms, increased fluorescence is indicated by a rightward shift of the positive peak, away from the background peak. The highest expression of the receptor on the surface was correlated with the degree of displacement. Relative receptor numbers are estimated in each cell type by comparing the mean fluorescence intensity with that of cells with known receptor copy numbers.

Four. Cell proliferation assay

The proliferation assay was measured by incorporation of MTT [3- (4,5-Dimethylthiazol-2-yl) -2,5-diphenyltetrazole or bromide]. Cell lines were applied in 96-well plates (5,000 cells / well) in 10% DMEM with indicated amount of CDR4D5 mimetic and incubated for 48 hours. The MTT was delivered to the cells for 4 hours. Cells were lysed in 50% SDS / 20% dimethyl sulfoxide and kept overnight at 37 ° C. Proliferation was evaluated by taking optical density readings at 570 nm, using an ELISA reader. The number of cells used in this assay was determined to be within the linear range for this cell type. 5. Radio-sensitization effect of CDR4D5 anti-p185^c-erbB-2 mimetic, determined by morphological analysis of apoptosis.

30,000 cells were allowed to bind to cover plates overnight in 6-well plates. Cells were incubated with 50 pg / ml CDR4D5 mimetic for 48 hours before irradiation. 10 Gy of irradiation was supplied and the cells were incubated at 37 ° C. Nuclear morphology was evaluated at the following points of

ES 2 270 586 T3 time: 12, 24, 48 and 72 hours after irradiation. The cover plates were washed twice with PBS at the indicated times and fixed in a 50:50 mixture of ice cold nietanol / acetone for 1 minute. The fixed cells were then stained with hydrated 4 ', 6'-Diamidino-2-phenylindole dihydrochloride (DAPI) (Sigma, ST. Louis, MO) with a concentration of 0.25 ng / ml in PBS and the morphological evaluation of apoptotic nuclei was determined using direct counting. Inter-observer correspondence in apoptosis counts was confirmed by deoxynucleotidyl transferase mediated deoxynucleotidyl transferase-mediated marking (TUNEL) -dUTP notch end staining and by analysis of three independent observers.

Cell counts were carried out within 30 minutes of staining and photographs were taken on a Zeiss Axioplan epifluorescence microscope. At least three independent fields of 100 cells were counted for each sample.

Results

1. Expression of the surface c-erbB-2 receptor

Flow cytometric analysis was used to determine the expression of the p185 receptor.^{c- erbB -2} surface area in human tumor cells. Surface c-erbB-2 receptor expression was highest in SkBR3, moderate in MCF7, low to moderate in U373MG, and not detectable in U87MG. The mean fluorescence of SKBR3 was 50 times that of the control and that of MCF7, U373MG, and U87MG were 6.5, 2, and 1 times the control, respectively.

2. Inhibition of proliferation

Treatment with CDR4D5 inhibited tumor cell proliferation in a dose-dependent manner and inversely proportional to the density of the p185 receptor.^c-erbB-2 Of surface. CDR4D5 did not inhibit the proliferation of parental U373MG cells that do not express c-erbB-2 and U373 / T691 cells were 62% inhibited with 1pg of CDR4D5 mimetic. The proliferation of MCF7 and SKBR3 cells was inhibited 43% -53%, and 39% -49%, respectively, in a dose-dependent manner (Figure 5).

3. Radiosensitizing effect of the anti-p185 mimetic CDR4D5^c-esr&gt;^^-1

Apoptosis was maximum at 72 hours after radiation in all cells. Treatment with CDR4D5 mimetic in U373MG cells resulted in a 20-28% increase in apoptosis than for untreated U373MG cells at 48h and 72h after radiation. The effect of CFR4D5 treatment on apoptosis in U373MG cells was comparable to the result with truncated neu, a mutant inhibitory receptor that deactivates erbB signaling and induces increased apoptosis in response to radiation (Figure 6A). A significant radiosensitizing effect of CDR4D5 was observed at 72 hours after radiation in MCF7 and SKBR3 cell lines (Figure 6B). Since the sensitivity to apoptotic cell death was inversely correlated with the levels of the surface receptor p185^c-erbB-2, by increasing the amount of CDR4D5, in accordance with the amount of surface cerbB-2 receptor expression, the effect should be enhanced.

This 4D5 mimetic has a small size, approximately 1.5KD, a protease resistant peptide specific for the human p185 receptor.^c-erbB-2 and it is less immunogenic than full-length antibodies. The 4D5 mimetic shows the use of anti-receptor mimetics in cancer diagnosis and treatment, leading to synergistic effects that are combined with cytotoxic therapeutics, such as gamma-ray irradiation.