WO1994016057A1

WO1994016057A1 - Enhancer sequence for modulating expression in epithelial cells

Info

Publication number: WO1994016057A1
Application number: PCT/US1993/011915
Authority: WO
Inventors: Donald Kufe; Miyako Abe
Original assignee: Dana Farber Cancer Institute Inc
Current assignee: Dana Farber Cancer Institute Inc
Priority date: 1992-12-31
Filing date: 1993-12-08
Publication date: 1994-07-21
Anticipated expiration: 1995-06-30
Also published as: EP0677099B1; US5874415A; EP0677099A4; DE69333921D1; CA2153068C; EP0677099A1; US5565334A; DE69333921T2; CA2153068A1

Abstract

Isolated DNA encompassing the DF3 enhancer as well as a sequence encoding a heterologous polypeptide provides epithelial tissue-selective gene expression of the heterologous polypeptide, useful in methods of therapy.

Description

ENHANCER SEQUENCE FOR MODULATING EXPRESSION

IN EPITHELIAL CELLS Statement as to Federally Sponsored Research The work described herein was supported in part by United States Public Health Service Grant CA-38879 from the National Cancer Institute, and by a grant from the Department of Health and Human Services. The United States Government therefore has certain rights in the invention.

The invention relates to tissue-specific enhancer DNA sequences, which modulate the expression of downstream sequences in eukaryotic cells.

Background of the Invention Epithelium is the tissue that covers and lines the free surfaces of the body, and includes cells of the skin and sensory organs, as well as the specialized cells lining the blood vessels, gastrointestinal tract, air passages, ducts of the kidneys and endocrine organs. Epithelial cells maintain many diverse and essential cellular functions, including synthesizing and secreting proteins and other products, providing barriers, and proliferating to replace constantly eroded surface epithelia. Carcinomas, which are cancers derived from epithelial cells, account for the majority of malignant neoplasias, including most cancers of the breast, lung, colon, prostate, ovary, and pancreas. The so-called DF3 gene encodes a high molecular weight glycoprotein which is aberrantly expressed by malignant human epithelia of various types. The DF3 glycoprotein is a member of a family of related carcinoma-associated antigens with core proteins ranging from 160 to 230 kD (Abe et al., J. Immunol. 139:257-261, 1987; Abe et al. , Cancer Res. 49:2834-2839, 1989). This antigen is expressed on the apical borders of secretory mammary epithelium and at high levels in the cytosol of less differentiated malignant cells (Kufe et al., Hybridoma 3:223-232, 1984). DF3 antigen expression correlates with the degree of breast tumor differentiation (Lundy et al.. Breast Cancer Res. Treat. 5:269-276, 1985). Moreover, the detection of this antigen in human milk (Abe et al.. Cancer Res. 44:4574- 4577, 1984) has suggested that its expression represents a differentiated function of normal mammary cells.

Previous studies have demonstrated that the DF3 gene is highly over-expressed in human breast carcinomas and that expression of this gene is regulated at the transcriptional level (Kufe et al., Hybridoma 3:223-232, 1984; Abe et al., J. Cell. Physiol. 143:226- 231, 1990).

Summary of the Invention Applicants have determined that the regulatory sequences for the DF3 gene, which is selectively expressed only in epithelial cells, are located in the 5'-flanking region of this gene; that a novel cis-acting enhancer sequence functions to control transcription in cells expressing this gene; and that this enhancer can be linked to a heterologous coding sequence to increase tissue-specific expression of that coding sequence in epithelium, and particularly in carcinoma cells.

As disclosed herein, the DF3 enhancer has been cloned and sequenced. By DF3 enhancer is meant either the sequence of SEQ ID NO:2, or a fragment thereof with enhancer activity. In preferred embodiments, the enhancer of the invention includes a sequence substantially identical to SEQ ID N0:1. In other preferred embodiments, the sequence is substantially identical to either the first 11 base pairs (SEQ ID NO:3) or the last 21 base pairs (SEQ ID NO:4) of SEQ ID NO:l. Enhancer activity can be determined by inserting the sequence to be evaluated into a chloramphenicol acetyltransferase (CAT) expression vector, transiently transfecting appropriate cells with the vector, and measuring CAT activity in the cells. Fragments which have enhancer activity cause transfected cells to express at least 25% of the CAT activity expressed by cells transfected with the CAT expression vector containing SEQ ID NO:2 as the enhancer sequence. This procedure is described in detail below. One embodiment of the invention provides DNA, for example isolated DNA, which includes the DF3 enhancer, but does not encode DF3. "Isolated DNA" containing the DF3 enhancer refers to a segment of DNA that includes the DF3 enhancer, but which is free of the genes that, in the naturally-occurring genome of the organism from which the segment is derived, flank the DF3 gene. The isolated DNA, therefore, is not immediately contiguous with both of the coding sequences with which it is immediately contiguous (i.e., one at the 5' end and one at the 3' end) in the naturally-occurring genome of the organism from which the DNA of the invention is derived. The term includes, for example, a recombinant DNA which is incorporated into a vector; into an autonomously replicating plasmid or virus; or into the genomic DNA of a prokaryote or eukaryote, or which exists as a separate molecule (e.g., a cDNA or a genomic DNA fragment produced by PCR or restriction endonuclease treatment) independent of other DNA sequences. Another aspect of the invention involves DNA containing DF3 enhancer and a DNA sequence (e.g., an antisense oligomer) which is antisense to a given naturally occurring mRNA encoding a naturally occurring polypeptide, which polypeptide is not DF3. An antisense oligomer is a short (generally at least 14 bp, and up to 100 or more bp) nucleotide sequence formulated to be complementary to a portion of a specific mRNA sequence. It is preferably located downstream from the enhancer of the invention and a functional promoter sequence, such as that of the HSV thy idine kinase gene; a poly A tail is typically located at the end of the antisense sequence to signal the end of the sequence. Standard methods relating to antisense technology have been described (Melani et al., Cancer Res. 51:2897-2901, 1991.) Following transcription of the antisense sequence into RNA, the antisense RNA binds to its target mRNA molecules within a cell, thereby inhibiting translation of the mRNA and down-regulating expression of the protein encoded by the mRNA. If, for example, the gene product of the target sequence stimulates tumor growth, down-regulation of the gene after antisense binding will result in suppression of tumor growth. Intended antisense oligodeoxynucleotides with antiproliferative activity, suitable for treatment of carcinomatous cells, include sequences antisense to N-myc in neuroectodermal cells (Whitesell et al., Mol. Cell. Biol. 11:1360-1370, 1991), c-myb for treatment of colon adenocarcinoma (Melani et al., Cancer Res. 51:2897-2901, 1991) , type 1 regulatory subunit of the cAMP receptor protein kinase in breast, colon, and gastric carcinoma cells (Szczylik et al.. Science 253:562-568, 1991), and K-ras proto-oncogene in small cell lung cancer lines (Vleminck et al.. Cell 66:107-119, 1991).

Also provided is isolated DNA which contains the DF3 enhancer and a sequence encoding a heterologous polypeptide. A heterologous polypeptide is herein defined as any polypeptide other than DF3. The isolated DNA may include a nucleotide sequence substantially identical to that of SEQ ID NO:l, as shown in Figure 6. In other embodiments, the isolated DNA contains either the first 11 base pairs or the last 21 base pairs of SEQ ID NO:l, or both. In another embodiment of the invention, the isolated DNA includes a nucleotide sequence substantially identical to that of SEQ ID NO:2, as shown in Figure 7. The invention includes variations on the enhancer sequences shown which are at least 80% identical, and preferably at least 90% identical, to the naturally-occurring sequences given in the figures, and which function as enhancers of expression in carcinoma cells.

The isolated DNA of the invention may also include a radiation-inducible element. Preferably, the radiation-inducible element comprises a radiation- responsive enhancer-promoter, such as a CArG domain of an Egr-1 promoter, a TNF-α promoter, or a c-Jun promoter. These elements have been described in eichselbaum et al., U.S.S.N. 07/633,626, filed December 20, 1990, the specification of which is herein incorporated by reference.

The heterologous polypeptide encoded by the DNA of the invention may be a cytotoxic polypeptide. Intended polypeptides include, but are not limited to, an enzymatically active subunit of a toxin such as Pseudomonas exotoxin A, diphtheria toxin, Shiga toxin, Shiga-like toxin, E. coli LT, C3 toxin, pertussis toxin, tetanus toxin, cholera toxin, or botulism toxin, or another cytotoxic polypeptide such as gelonin, ricin, or tumor necrosis factor. Other possible heterologous polypeptides include but are not limited to cAMP receptor polypeptide kinase, platelet factor 4, monocyte chemoattractants, herpes virus thymidine kinase, cytidine dea inase, WTp53, retinoblastoma protein, E-cadherin, fibronectin receptor, monocyte chemoattractant protein-I, interleukin-2, and interleukin-4.

The DNA or isolated DNA of the invention may be introduced into target cells by standard vectors and/or gene delivery systems. Suitable gene delivery systems may include liposomes, receptor-mediated delivery systems, naked DNA, and viral vectors such as herpes viruses, retroviruses, and adenoviruses, among others. Standard methods for transfecting cells with isolated DNA are well known to those skilled in the art of molecular biology.

The invention also includes cells transfected with the DNA of the invention. Preferably, the cells are epithelial cells (e.g., carcinoma cells), and they express a heterologous polypeptide encoded by the DNA of the invention.

A therapeutic composition is provided which includes a pharmaceutically acceptable carrier and a therapeutically effective amount of a nucleic acid, wherein the nucleic acid includes the DF3 enhancer operatively linked to a sequence encoding a heterologous polypeptide, to generate high-level expression of the polypeptide in epithelial cells transfected with the nucleic acid. By "operatively linked" it is meant that the enhancer is located sufficiently close to the coding sequence that it functions to enhance expression in an epithelial cell. For example, the enhancer may be located immediately upstream of the coding sequence start site, or up to approximately 500 bp upstream of the start site. It may alternatively be located downstream from the 3' end of the coding sequence, or within an intron between two exons of the coding sequence. The enhancer sequence may be inverse to the orientation observed in the naturally-occurring DF3 gene. Two or more copies of the DF3 enhancer may be arranged in tandem to increase expression of the heterologous protein to even higher levels. When the DF3 enhancer is arranged in tandem with a different tissue-specific enhancer, expression in multiple tissue types can be simultaneously enhanced. Likewise, two or more antisense or coding sequences can be arranged in tandem in the DF3-enhancer-containing DNA of the invention; expression of both of these sequences would be increased in epithelial cells, and especially in carcinoma cells. The therapeutic composition may also include a gene delivery system as described above. Pharmaceutically acceptable carriers are biologically compatible vehicles which are suitable or administration to an animal: e.g., physiological saline. A therapeutically effective amount is an amount of the DNA of the invention which is capable of producing a medically desirable result in a treated animal.

The invention provides a method for treating an animal with a carcinoma, by identifying the carcinoma- bearing animal and treating the animal with the therapeutic composition of the invention. This method may be used to treat any type of carcinoma, including but not limited to adenocarcinoma, adrenal carcinoma, basal cell carcinoma, brain carcinoma, ovarian carcinoma, breast carcinoma, colon carcinoma, pancreatic carcinoma, gastric carcinoma, hepatic carcinoma, biliary system carcinoma, skin cell carcinoma, reproductive tract carcinoma, urinary organ carcinoma, small cell carcinoma, squamous-cell carcinoma, undifferentiated carcinoma, and pulmonary carcinoma. In this list, where a neoplasm is designated by organ association, as in "ovarian carcinoma" or "pulmonary carcinoma", any carcinoma located in or immediately adjacent to the organ is included, without regard to whether the carcinoma is primary or is a metastasis from another location.

In addition, the invention provides a method of treating an animal with a metabolic disorder, by administering the therapeutic composition of the invention, where the heterologous polypeptide encoded by the DNA alleviates symptoms of the metabolic disorder. This method would be useful for treatment of various metabolic diseases characterized by the absence or reduced amount of a gene product (such as phenylalanine hydroxylase resulting in classic phenylketonuria, cystic fibrosis transmembrane regulator in cystic fibrosis, or Factor VIII or IX in hemophilia) , or by structurally altered genes which are functionally inactive, as with alpha_x antitrypsin in emphysema. The heterologous polypeptide encoded by the DNA in the therapeutic composition would correspond to the missing or malfunctioning gene product, e.g. phenylalanine hydroxylase, the CF gene product, Factor VIII, or alpha_χ antitrypsin. Animals intended to be treated with the methods of the invention include any mammals to which the compounds of the invention may be administered, and in which the enhancer of the invention is functional. Animals specifically intended for treatment with the compounds and methods of the invention include humans, nonhuman primates, sheep, horses, cattle, pigs, goats, dogs, cats, rabbits, guinea pigs, hamsters, gerbils, rats and mice, as well as organs, tumors and cells derived or originating from these hosts.

The invention also includes a method of producing a polypeptide in vitro by culturing a cell of epithelial origin which has been transfected with the DNA of the invention, under conditions which facilitate the expression of the heterologous polypeptides encoded by the transgene. The polypeptides can then be isolated and purified by standard means.

Selective killing of cells and tumors is also encompassed by the invention. In this case, cytotoxic polypeptides expressed by the transfected cells act to kill the cells, or cells in the immediate vicinity of the transfected cells. In another embodiment, the transgene also encompasses, besides the DF3 enhancer and a sequence encoding the cytotoxic polypeptide, a radiation-inducible element. In this case, targeted radiation directed at a tumor, either in vitro or in situ in an animal, induces expression of a cytotoxic polypeptide only in those transfected cells which are of epithelial origin and which are irradiated. The heterologous polypeptide may also be one that, when secreted by the transfected cells, is capable of recruiting the organism's natural immunological defenses to attack cells in the vicinity of the transfected cells. The invention further provides a vaccine, and a method of vaccinating an animal, wherein the heterologous polypeptide associated with the DF3 enhancer includes an antigenic determinant of an infectious organism. Appropriate organisms might include bacteria, viruses, protozoa, or helminths. For example, antigenic determinants of those pathogens which cause acquired immune deficiency syndrome (AIDS) , polio, typhoid, cholera, and various enteric diarrheal diseases could be expressed and secreted by transfected epithelial cells lining the gastrointestinal tract, and thereby confer immunity to infection by both local and systemic immunological mechanisms. Useful antigens include the B subunit of cholera toxin (Mekalanos et al.. Nature 306:551-557, 1983); the B subunit of Shiga toxin/Shiga- like toxin I (Calderwood et al., Proc. Natl. Acad. Sci. USA 84:4364-4368, 1987) or an antigenic peptide thereof (Harari et al.. Infect. Immun. 56:1618-1624, 1988; Harari et al., Mol. Immunol. 27:613-621, 1990); an antigenic, enzymatically inactive mutant of diphtheria toxin (e.g., a deletion mutant lacking the A subunit portion of the toxin, or having an inactivating mutation in the A subunit) ; an antigenic, enzymatically inactive mutant of pertussis toxin, tetanus toxin, or E. coli LT; or a viral capsid polypeptide. Immunity to respiratory infections would result from transfection of lung epithelial cells with the appropriate DNA of the invention. Similar immunization would be useful in vaccinating animals against various diseases of veterinary importance, including many parasitic diseases such as toxocariasis, filariasis, coccidiosis, strongyloidiasis, and cestodiasis. The specialized ability of epithelial cells to produce and secrete proteins makes them particularly suitable targets for DNA-based vaccines. In a further embodiment of the invention, epithelial cells within an animal are transfected with DNA which includes the DF3 enhancer and a sequence encoding a polypeptide which confers resistance to the effects of a chemotherapeutic drug. In a preferred example, the polypeptide encodes P-glycoprotein, the product of the multidrug resistance (MDR-1) gene. Cells which express this glycoprotein have a highly enhanced membrane transport pump, which quickly clears the cells of many common chemotherapeutic drugs, including vinca alkaloids, anthracyclines, epipodophyllotoxins, L- asparaginase, colchicine, and other antineoplastic antibiotics. Another polypeptide which might be beneficially expressed by transfected epithelial cells is dihydrofolate reductase, the expression of which would protect transfected cells from the toxic effects of methotrexate. Such expression in epithelial cells of a cancer patient would protect the transgenic cells from the effects of chemotherapeutics administered to treat non-carcinomatous neoplasms in the patient, without affecting the efficacy of the chemotherapeutics against the non-carcinomatous neoplasm itself. The patient would then be spared the often life-threatening side effects of chemotherapy, such as gastrointestinal injury, which result from epithelial cell damage. In addition, protection of the epithelial cells would allow higher doses of chemotherapeutic drugs to be administered, with potentially better clinical outcomes. In another embodiment of this method, cells of the gastrointestinal tract could be protected by transfecting these cells with an orally-administered vector; nongastrointestinal malignancies, including carcinomatous and noncarcinomatous neoplasms, could then be treated with systemic chemotherapeutics with minimal effect on the gastrointestinal epithelial cells.

Another aspect of the invention includes a transgenic animal, the genome of which contains a transgene including DF3 enhancer operatively linked to a sequence encoding a heterologous polypeptide. Detailed methods for producing these animals are presented in U.S. Patent 4,736,866, the specification of which is hereby incorporated by reference. The intended transgenic animals may be of any nonhuman mammalian or avian species, but preferably include nonhuman primates, sheep, horses, cattle, pigs, goats, dogs, cats, rabbits, chickens, and rodents such as guinea pigs, hamsters, gerbils, rats, and mice.

The compounds and compositions of the invention may be administered by any medically acceptable method. Suitable methods include injections, by parenteral routes such as intravenous, subcutaneous, intramuscular, intratumor, intraperitoneal, intraventricular, intraepidural, or others, as well as oral, nasal, transdermal, transmucosal, ophthalmic, rectal, or topical routes. The compounds may also be applied directly to exposed tissue surfaces, such as during a surgical procedure. In addition, the compounds of the invention are suitable for inhalant administration, for example as a nebulized or aerosolized preparation.

As is well known in the medical arts, dosages for any one patient depends upon many factors, including the patient's size, body surface area, age, the particular compound to be administered, sex, time and route of administration, general health, and other drugs being administered concurrently. Dosages for the compounds of the invention will vary, but a preferred dosage for oral or inhaled administration to a patient is from 10⁵ to 10²⁰ copies of a DNA molecule of the invention, as specified above. A patient receiving intravenous vector preparations may be infused with from approximately 10⁶ to 10²² copies of the vector.

Other features and advantages of the invention will be apparent from the following description and from the claims.

Brief Description of the Drawings Figure 1A (SEQ ID NO:5) is a representation of the nucleotide sequence of the 5'-flanking region of the DF3 gene, plus the first 70 nucleotides of coding sequence. The transcription start site is indicated by the arrow. Potential binding sites with complete homology to the indicated cis-elements are highlighted by boxes. Figure IB is a transient expression analysis of DF3 promoter deletion constructs: MCF-7 cells were transfected with deleted promoter fragments linked to CAT. Relative CAT activities are expressed as values compared to that obtained with p-686CAT (assigned a value of 100) . The results represent the mean±SD of the indicated number (n) of experiments.

Figure 2A is a transient expression analysis of p- 686CAT internal deletion vectors: p-686CAT was digested with BstXI and Smal to create a construct with deletion of the region from positions -603 to -410. Smaller fragments of f(-603/-410) were religated into the deleted region. Relative CAT activities are expressed as values compared to that obtained with p-686CAT. The results represent the mean±SD of the indicated number (n) of independent experiments.

Figure 2B is a transient expression analysis of heterologous promoter constructs: Fragments of the DF3 promoter were inserted upstream to the TK promoter in pBLCAT2. MCF-7 cells were transfected with the indicated constructs. The results (mean ±SD) are expressed as reactive CAT activity compared to that obtained with TK- CAT (assigned a value of 100) .

Figures 3A and B are gel retardation analyses of nucleoprotein complexes found with f(-507/-483) : (A) f(-507/-483) (CCGGGAAGTGGTGGGGGGAGGGAGC) (SEQ ID NO:6) was end-labeled and incubated with 10 μg MCF-7 nuclear proteins. Unlabeled f(-507/-483) and f(-611/- 579) were added at a 50- or 250-fold molar excess compared to the labeled probe. The unlabeled oligonucleotide corresponding to the consensus sequence for SP1 was added at a 50-fold molar excess.

(B) f(-507/-483) and a mutated fragment at positions -496 and -495 (CCGGGAAGTGGCAGGGGGAGGGAGC) (SEQ ID NO:7) were end-labeled (both lxlO⁵ CPM, 5xl0⁴ CPM/ng) and incubated with MCF-7 nuclear proteins.

Figure 4 is a DNase I analysis of protein binding to f(-618/-410) . f(-618/-410) was end-labeled (lxlO⁴ CPM/0.1 ng) and incubated with the indicated amounts of MCF-7 nuclear proteins. The samples were subjected to DNase I digestion and analyzed in 12% polyacrylamide/urea gels. C represents analysis of cytosines as determined by Maxam-Gilbert sequencing. The hatched boxes reflect regions protected from DNase I digestion.

Figure 5 is an autoradiogram of a gel illustrating f(-507/-483) interacting with a 45 kD nuclear protein in MCF-7 cells. Nuclear proteins (200 μg) from MCF-7 and HL-60 cells were separated by polyacrylamide gel electrophoresis and transferred to a nitrocellulose sheet. Lanes were incubated with ³²P-f(-507/-483) in the absence or presence of a 100-fold molar excess of unlabeled fragment. Figure 6 is a representation of the nucleotide sequence of a functional region of the DF3 enhancer (SEQ ID NO: 1) , which represents nucleotides -598 to -485 of the 5'-flanking region of the DF3 gene shown in Figure 1A (SEQ ID NO: 2) .

Figure 7 (SEQ ID NO:2) is a representation of the nucleotide sequence of the DF3 5'-flanking sequence, as shown in Figure 1A (SEQ ID NO:5) , except without any of the DF3 coding sequence shown in Figure 1A.

Detailed Description

Cell culture. Human MCF-7 breast carcinoma cells (Michigan Cancer Foundation, Detroit, MI; American Type Culture Collection) were grown as a monolayer in Dulbecco's modified Eagle's medium (DMEM) with 10% heat- inactivated fetal bovine serum, 100 units/ml penicillin, 100 μg/ml streptomycin, 2 mM L-glutamine and 0.25 IU/ml insulin.

Plasmid construction. A pΔ vector was generated by excising the SV40 promoter and enhancer regions from pSVT7 (Bird et al., J. Cell. Biol. 105:2905-2914, 1987). Chloramphenicol acetyltransferase (CAT) expression vectors, pSVT7CAT and pΔCAT, were produced by inserting the CAT gene in the ulticloning Hindlll site of pSVT7 and pΔ, respectively. A p-1656CAT vector was constructed by inserting the Xmnl fragment (positions -1656 to +31) of the DF3 gene into pΔCAT at the Pstl site by blunt- ended DNA ligation. The p-725CAT vector was prepared by inserting the blunt-ended Sstl/X nl fragment (positions - 725 to +31) into pΔCAT. A series of deletion vectors was generated from p-725CAT after treatment with Sstl or Xbal and subsequent digestion with exonuclease III and SI nuclease. Sequencing of the constructs was performed by dideoxy termination (Sanger et al., Proc. Natl. Acad. Sci. USA 74:5463-5467, 1977).

Reporter assays. All vectors used in the reporter assays were purified twice by cesium chloride gradient centrifugation and phenol-chloroform extraction. Twenty μg of each plasmid were transfected into 3xl0⁵ MCF-7 cells by the Ca₂P0₄ procedure (Davis et al., Basic Methods in Molecular Biology (Elsevier, NY) pp. 285-289, 1986). The cells were incubated in complete medium for 48 hours after transfection, harvested and lysed in 0.25 M Tris- HCl (pH 8.0) by three cycles of freeze-thawing. CAT activity was assayed in 125 μl reactions containing 10 to 50 μl of cell extract, 125 nCi [¹⁴C]chloramphenicol (DuPont Co., Boston, MA; 57 mCi/mmol) , 250 mM Tris-HCl, pH 8.0, and 25 μg n-butyryl coenzyme A (Sigma) for 1 hour at 37°C. The reaction was terminated by adding 300 μl xylene. The non-butyrated chloramphenicol was removed by washing twice with 100 μl of 0.25 M Tris-HCl (pH 8.0). The xylene layer containing the butyrated chloramphenicol was assayed by scintillation counting (Seed et al., Gene 67:271-277, 1988).

Electrophoretic mobility shift assays (EMSAs) . Nuclear proteins were prepared according to previously described methods (Dignam et al., Nucl. Acids. Res. 11:1475-1489, 1983). Synthetic oligonucleotides were end-labeled with the appropriate [α-³²P]dNTPs using DNA polymerase I (Klenow fragment) and purified by Nuc Trap columns (Stratagene, LaJolla, CA) . Binding assays were performed as described (Henninghausen et al., Guide to Molecular Cloning Techniques, (Academic Press, NY) pp. 721-735, 1987). Ten μg of nuclear proteins were incubated with 10 μg poly(dl-dC) in 25 μl of 10 mM Tris- HCl buffer (pH 7.5) containing 50 or 150 mM KC1, 5 mM MgCl₂, 1 mM DTT, 1 mM EDTA, 12.5% glycerol and 0.1% Triton X-100 for 30 min at room temperature. After adding the labeled probe (lxlO⁵ CPM, 1 ng) and incubating for an additional 30 min, the samples were separated in a low ionic strength 4.5% polyacrylamide gel containing 1 mM EDTA, 3.3 mM sodium acetate and 6.7 mM Tris- HC1 (pH 7.5) or in a high ionic strength gel containing 50 mM Tris-HCl (pH 8.5), 380 mM glycine and 2 mM EDTA (Henninghausen et al. , Guide to Molecular Cloning Techniques (Academic Press NY) pp. 721-735, 1987; Staudt et al., Nature 323:640-643, 1986). Unlabeled oligonucleotides as competitors, including the consensus sequences of SP1, NF-l, AP-1, AP-2 and AP-3 (Stratagene) , were added at the same time as the labeled fragment.

DNase I footprint analysis. Fragment f(-618/-410) (SEQ ID NO:8) was ligated into pGEM3 at the Smal site. The vector was digested with EcoRI and Accl, end-labeled with ³²P-dATP and dTTP at the EcoRI site and purified in a 5% native polyacrylamide gel. The specific activity of the labeled fragment was approximately 3xl0⁵ CPM/ng. The assay was performed as described (Jones et al.. Cell 42:559-572, 1985). The reaction products were analyzed by 12% polyacrylamide/urea gel electrophoresis and subjected to autoradiography. Sequences were monitored by the Maxam-Gilbert procedure (Maxam et al.. Methods in Enzymol. (Academic Press, NY) 65:499-560, 1980).

Southwestern analysis. Nuclear proteins (200 μg) were prepared as described (Dignam et al., Nucl. Acids. Res. 11:1475-1489, 1983), separated by 3-15% SDS- polyacrylamide gel electrophoresis and transferred to a nitrocellulose sheet (Lammnli, Nature 227:680-695, 1970; Towbin et al., Proc. Natl. Acad. Sci. USA 76:4350-4355, 1979) . The sheet was incubated in 10 mM Tris-HCl (pH 7.5) containing 5% skimmed milk and then in 10 mM Tris- HCl (pH 7.9) containing ³²P-labeled oligonucleotide probe (10⁶ CPM/ml), 50 mM NaCl, 10 mM MgCl₂, 0.1 mM EDTA, 1 mM DTT and 10 μg/ml poly(dl-dC) for 60 min at room temperature. The sheet was washed with the same buffer containing 200 mM NaCl and exposed to x-ray film (Patel et al., EMBO J. 2:137-142, 1983; Miskimins et al., Proc. Natl. Acad. Sci. USA 82:6741-6744, 1985). RESULTS

Sequence of the DF3 5'-flanking region. A genomic library from MCF-7 cells was previously used to isolate a clone containing over 2 kb of sequences upstream to the DF3 transcription start site (Abe et al., Biochem. Biophys. Res. Commun. 165:644-649, 1989, herein incorporated by reference) . Digestion of this clone with Xmnl resulted in a fragment extending from positions - 1656 to +31. The nucleotide sequence to position -1656 includes a potential TATA box (TATAAA) 25 bases upstream to the transcription start site (Figure 1A) . The 5' flanking region also includes potential binding sites with complete homology to the consensus sequences for SP1, AP-1, AP-2, AP-3, NF-1 and the estrogen receptor half-site (Figure 1A) .

Functional analysis of the putative DF3 promoter region. A series of experiments confirmed that DF3 gene expression is regulated at the transcriptional level in MCF-7 cells. A series of 5' deleted DF3 promoter-CAT constructs were transfected into these cells to identify the functional cis-element(s) . Transfection of plasmid P-1656CAT resulted in CAT activities that were approximately 25-fold higher than that obtained with pΔCAT (Figure IB) . Similar levels of CAT activity were obtained with p-725CAT, p-686CAT and p-618CAT (Figure IB) . In contrast, transfection of p-549CAT was associated with a decrease in transcriptional activity to approximately 15% of that found with p-686CAT. Moreover, transfection of the more extensively deleted constructs (p-448CAT, p-324CAT, p-167CAT) yielded levels of relative activity that were similar to those obtained with constructs containing a minimal promoter region (p-19CAT) (Figure IB) . Taken together, these findings indicated that the sequence between positions -618 and -549 includes one or more elements involved in the control of DF3 gene transcription.

In order to further define the functional element(s) , a series of internal deletions was prepared using p-686CAT. Deletion of the region from positions - 603 to -410 decreased (>80%) transcription of p-686CAT to a level approaching that obtained with p-549CAT (Figure 2A) . Similar findings were obtained with reinsertion of f(-410/-505) or f(-481/-410) (Figure 2A) . These results were thus in concert with the likelihood that the functional enhancer sequences were upstream to position - 549 (Figure IB). However, ligation of f(-505/-588) into this cassette had little if any effect, while reinsertion of f(-600/-505) or f(-505/-600) resulted in only partial (54-60%) recovery of CAT activity (Figure 2A) . In contrast, f(-48S/-598) conferred complete restitution of the activity obtained with p-686CAT (Figure 2A) . These results indicated that sequences within f(-598/-4S5) were responsible for control of DF3 gene transcription.

A heterologous promoter was used to determine the enhancing activities of sequences upstream to position - 485. Single copies of fragments f(-753/-505) , f(-485/- 598), f(-505/-588) and f(-505/-614) were inserted upstream to the herpes simplex virus minimal thymidine kinase (TK) promoter in pBLCAT2 (Figure 2B) . CAT activity of PBLCAT2 was normalized to 100. Insertion of f(-753/-505) into this vector had little if any effect, while f(-48S/-598) enhanced TK promoter activity by over 10-fold (Figure 2B) . These results provided further support for the presence of functional elements between positions -598 and -485. Moreover, the finding that both f(-505/-588) and f(-505/-614) decreased the activity of pBLCAT2 (Figure 2B) emphasized the importance of the sequence between -505 and -485.

Interaction of nuclear proteins with the DF3 promoter. The results of the reporter assays prompted further studies on the interaction of nuclear proteins with the putative regulatory regions of the DF3 promoter. EMSAs were performed with f(-507/-483) . Incubation of end-labeled f(-507/-483) with nuclear proteins from MCF-7 cells resulted in a clearly detectable retarded fragment (Figure 3A) . In order to determine whether this complex reflected specific DNA-protein binding, unlabeled f(- 507/-483) was added to the reaction. In these experiments, a 50-fold excess of unlabeled probe partially inhibited formation of the complex, while a 250-fold excess resulted in complete inhibition (Figure 3A) . In contrast, use of unlabeled f(-611/-579) as a competitor had no detectable effect (Figure 3A) . There was also no detectable inhibition of this complex when using unrelated oligonucleotides containing SP1, AP-1, AP-2, AP-3, or NF-1 sequences (Figure 3A and data not shown) . These findings, taken together with the absence of a detectable band when using nuclear proteins from HL- 60 (DF3 antigen negative cells, data not shown) , supported the formation of a specific f(-507/483)-MCF-7 nuclear protein complex. There was also no detectable complex formation when incubating labeled f(-515/-496) or f(-529/-510) with the MCF-7 nuclear proteins (data not shown). Finally, alterations in the sequence of f(-507/- 483) at positions -496/-495 resulted in nearly complete inhibition of complex formation (Figure 3B) .

The interaction of nuclear proteins with the DF3 promoter was further addressed by DNasel footprinting. End-labeled f(-618/-410) was incubated with increasing amounts of MCF-7 nuclear proteins and then subjected to digestion with DNase I. The demonstration that one protected region extended from positions -485 to -505 (Figure 4) is in concert with the sequences identified by reporter and gel retardation assays. Another region extending from positions -513 to -525 was also protected by MCF-7 nuclear proteins (Figure 4) . Moreover, there was enhancement of sequences adjacent to the protected regions (at positions -506 to -510 and upstream to position -530) (Figure 4) .

Southwestern analysis was performed to identify the size of the protein(s) that interact with f(-507/- 483) . This fragment interacted with a species with an apparent molecular weight of 45 kd (Figure 5) . This interaction was inhibited by a 100-fold excess of unlabeled to labeled fragment. In contrast, there was no detectable interaction of f(-507/-483) with nuclear proteins from HL-60 cells (Figure 5) . Moreover, incubation of labeled f(-611/-579) with the MCF-7 nuclear proteins failed to demonstrate DNA-protein interaction (data not shown) . Taken together, these results indicated that f(-507/-483) specifically interacts with a nuclear 45 kD protein. Example A patient is diagnosed with a malignant inoperable pulmonary carcinoma. Using standard methods, a genetic construct is prepared bearing the DF3 enhancer operably linked to the gene for herpes simplex-thymidine kinase (HS-tk) . The HS-tk gene, when introduced into tumor cells, confers sensitivity to the anti-herpes drug gancyclovir. The construct is inserted into a retroviral vector, and 10⁸ copies of the vector are injected into the tumor. Of all of the cells which may take up the vector, only the transfected carcinoma cells which naturally overexpress DF3 will highly overexpress the HS- tk transgene. The patient is then treated with gancyclovir, which acts primarily on those cells which overexpress HS-tk, causing the tumor to regress.

Other Embodiments The invention includes the use of the DF3 enhancer to modulate expression of any protein which is either partially or wholly encoded in a location downstream to the DF3 enhancer. Also included are allelic variants, natural mutants, and induced mutants of the DF3 enhancer, which variants and mutants retain the ability to enhance expression in epithelial cells comparable to the level of enhancement produced by the sequence shown in SEQ ID N0:1.

Similar enhancers derived from corresponding genes in other mammalian species are included within the scope of the invention. Such enhancers may be identified by, for example, identifying the enhancer associated with a coding sequence from a particular mammalian species, which coding sequence hybridizes under stringent conditions with a probe consisting of a 20-nucleotide segment of the human DF3 coding sequence, and which encodes a polypeptide that is expressed predominantly in epithelial cells of that species. Alternatively, a probe consisting of a segment of the human DF3 enhancer can be used as a probe for a similar sequence in the genome or cDNA of another species. A sequence which functions as an epithelium-specific enhancer and which hybridizes with a probe containing the sequence of SEQ ID NO:l or the sequence of SEQ ID NO:4, under the following hybridization conditions, would be within the invention: prehybridization in 50% formamide, 5X SSC, 25 mM potassium phosphate buffer (pH 7.4), 5X Denhardt's, and 50 μg/ml denatured salmon sperm DNA for 4-12 hours at 20°C; hybridization for 12-24 hours at 20°C; washing in 5X SSC containing 0.1% SDS, at 20°C.

SEQUENCE LISTING

(1) GENERAL INFORMATION: (i) APPLICANT: Kufe, Donald Abe, Miyako

(ii) TITLE OF INVENTION: ENHANCER SEQUENCE FOR MODULATING EXPRESSION IN EPITHELIAL CELLS

(iii) NUMBER OF SEQUENCES: 8 (iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Fish & Richardson

(B) STREET: 225 Franklin Street

(C) CITY: Boston

(D) STATE: Massachusetts

(E) COUNTRY: U.S.A.

(F) ZIP: 02110-2804

(v) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: 3.5" Diskette, 1.44 Mb

(B) COMPUTER: IBM PS/2 Model 50Z or 55SX

(C) OPERATING SYSTEM: MS-DOS (Version 5.0)

(D) SOFTWARE: WordPerfect (Version 5.1)

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER:

(B) FILING DATE:

(C) CLASSIFICATION:

(vii) PRIOR APPLICATION DATA:

(A) APPLICATION NUMBER: 07/999,742

(B) FILING DATE: December 31, 1992

(viii) ATTORNEY/AGENT INFORMATION:

(A) NAME: Fraser, Janis K.

(B) REGISTRATION NUMBER: 34,819

(C) REFERENCE/DOCKET NUMBER: 00530/065WO1

(ix) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE: (617) 542-5070

(B) TELEFAX: (617) 542-8906

(C) TELEX: 200154

(2) INFORMATION FOR SEQUENCE IDENTIFICATION NUMBER: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 114

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear (xi) SEQUENCE DESCRIPTION: SEQ ID NO: 1:

CTGGAAAGTC CGGCTGGGGC GGGGACTGTG GGTTTCAGGG TAGAACTGCG TGTGGAACGG 60

GACAGGGAGC GGTTAGAAGG GTGGGGCTAT TCCGGGAAGT GGTGGGGGGA GGGA 114

(3) INFORMATION FOR SEQUENCE IDENTIFICATION NUMBER: 2: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 1656

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 2:

GCTTCCGTGC GCCTAGAGCG CAGCCTGCGA CTGCGGGACC CAACAACCAC GTGCTGCCGC 60

GGCCTGGGAT AGCTTCCTCC CCTCTGGCAC TGCTGCCGCA CACACCTCTT GGCTGTCGCG 120

CATTACGCAC CTCACGTGTG CTTTTGCCCC CGCCTACGTG CCTACCTGTC CCCAATACCA 180

CTCTGCTCCC CAAAGGATAG TTCTGTGTCC GTAAATCCCA TTCTGTCACC CCACCTACTC 240

TCTGCCCCCC CCTTTTTTGT TTTGAGACGG AGTCTTGCTC TGTCGCCCAG GCTGGAGTGC 300

AATGGCGCGA TCTCGGCTCA CTGCAACCTC CGCCTCCCGG GTTCAAGCGA TTCTCCTGCC 360

TCAGCCTCCT GAGTAGCTGG GGTTACAGCG CCCGCCACCA CGCTCGGCTA ATTTTTGTAG 420

TTTTTAGTAG AGACGAGGTT TCACCATCTT GGCCAGGCTG GTCTTGAACC CCTGACCTTG 480

TGATCCACTC GCCTCGGCCT TCCAAAGTGT TGGGATTACG GGCGTGACGA CCGTGCCACG 540

CCCGATCTGC CTCTTAAGTA CATAACGGCC CACACAGAAC GTGTCCAACT CCCCCGCCCA 600

CGTTCCAACG TCCTCTCCCA CATACCTCGG TGCCCCTTCC ACATACCTCA GGACCCCACC 660

CGCTTAGCTC CATTTCCTCC AGACGCCACC ACCACGCGTC CCGGAGTGCC CCCTCCTAAA 720

GCTCCCAGCC GTCCACCATG CTGTGCGTTC CTCCCTCCCT GGCCACGGCA GTGACCCTTC 780

TCTCCCGGGC CCTGCTTCCC TCTCGCGGGC TCTCGCTGCC TCACTTAAGC AGCGCTGCCC 840

TTACTCCTCT CCGCCCGGTC CGAGCGGCCC CTCAGCTTGC GCGGCCCAGC CCCGCAAGGC 900

TCCCGGTGAC CACTAGAGGG CGGGAGGAGC TCCTGGCCAG TGGTGGAGAG TGGCAAGGAA 960

GGACCCTAGG GTTCATCGGA GCCCAGGTTT ACTCCCTTAA GTGGAAATTT CTTCCCCCAC 1020

TCCCTCCTTG GCTTTCTCCA AGGAGGGAAC CCAGGCTGCT GGAAAGTCCG GCTGGGGCGG 1080

GGACTGTGGG TTTCAGGGTA GAACTGCGTG TGGAACGGGA CAGGGAGCGG TTAGAAGGGT 1140

GGGGCTATTC CGGGAAGTGG TGGGGGGAGG GAGCCCAAAA CTAGCACCTA GTCCACTCAT 1200

TATCCAGCCC TCTTATTTCT CGGCCCCGCT CTGCTTCAGT GGACCCGGGG AGGGCGGGGA 1260

AGTGGAGTGG GAGACCTAGG GGTGGGCTTC CCGACCTTGC TGTACAGGAC CTCGACCTAG 1320 CTGGCTTTCT TCCCCATCCC CACGTTAGTT GTTGCCCTGA GGCTAAAACT AGAGCCCAGG 1380

GGCCCCAAGT TCCAGACTGC CCCTCCCCCC TCCCCCGGAG CCAGGGAGTG GTTGGTGAAA 1440

GGGGGAGGCC AGCTGGAGAA CAAACGGGTA GTCAGGGGGT TGAGCGATTA GAGCCCTTGT 1500

ACCCTACCCA GGAATGGTTG GGGAGGAGGA GGAAGAGGTA GGAGGTAGGG GAGGGGGCGG 1560

GGTTTTGTCA CCTGTCACCT GCTCCGGCTG TGCCTAGGGC GGGCGGGCGG GGAGTGGGGG 1620

GACCGGTATA AAGCGGTAGG CGCCTGTGCC CGCTCC 1656

(4) INFORMATION FOR SEQUENCE IDENTIFICATION NUMBER: 3: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 11

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 3:

CTGGAAAGTC C 11

(5) INFORMATION FOR SEQUENCE IDENTIFICATION NUMBER: 4: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 21

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 4:

GGGAAGTGGT GGGGGGAGGG A 21

(6) INFORMATION FOR SEQUENCE IDENTIFICATION NUMBER: 5: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 1725

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 5:

GCTTCCGTGC GCCTAGAGCG CAGCCTGCGA CTGCGGGACC CAACAACCAC GTGCTGCCGC 60

GGCCTGGGAT AGCTTCCTCC CCTCTGGCAC TGCTGCCGCA CACACCTCTT GGCTGTCGCG 120

CATTACGCAC CTCACGTGTG CTTTTGCCCC CGCCTACGTG CCTACCTGTC CCCAATACCA 180

CTCTGCTCCC CAAAGGATAG TTCTGTGTCC GTAAATCCCA TTCTGTCACC CCACCTACTC 240

TCTGCCCCCC CCTTTTTTGT TTTGAGACGG AGTCTTGCTC TGTCGCCCAG GCTGGAGTGC 300 AATGGCGCGA TCTCGGCTCA CTGCAACCTC CGCCTCCCGG GTTCAAGCGA TTCTCCTGCC 360

TCAGCCTCCT GAGTAGCTGG GGTTACAGCG CCCGCCACCA CGCTCGGCTA ATTTTTGTAG 420

TTTTTAGTAG AGACGAGGTT TCACCATCTT GGCCAGGCTG GTCTTGAACC CCTGACCTTG 480

TGATCCACTC GCCTCGGCCT TCCAAAGTGT TGGGATTACG GGCGTGACGA CCGTGCCACG 540

CCCGATCTGC CTCTTAAGTA CATAACGGCC CACACAGAAC GTGTCCAACT CCCCCGCCCA 600

CGTTCCAACG TCCTCTCCCA CATACCTCGG TGCCCCTTCC ACATACCTCA GGACCCCACC 660

CGCTTAGCTC CATTTCCTCC AGACGCCACC ACCACGCGTC CCGGAGTGCC CCCTCCTAAA 720

GCTCCCAGCC GTCCACCATG CTGTGCGTTC CTCCCTCCCT GGCCACGGCA GTGACCCTTC 780

TCTCCCGGGC CCTGCTTCCC TCTCGCGGGC TCTCGCTGCC TCACTTAAGC AGCGCTGCCC 840

TTACTCCTCT CCGCCCGGTC CGAGCGGCCC CTCAGCTTGC GCGGCCCAGC CCCGCAAGGC 900

TCCCGGTG C CACTAGAGGG CGGGAGGAGC TCCTGGCCAG TGGTGGAGAG TGGCAAGGAA 960

GGACCCTAGG GTTCATCGGA GCCCAGGTTT ACTCCCTTAA GTGGAAATTT CTTCCCCCAC 1020

TCCCTCCTTG GCTTTCTCCA AGGAGGGAAC CCAGGCTGCT GGAAAGTCCG GCTGGGGCGG 1080

GGACTGTGGG TTTCAGGGTA GAACTGCGTG TGGAACGGGA CAGGGAGCGG TTAGAAGGGT 1140

GGGGCTATTC CGGGAAGTGG TGGGGGGAGG GAGCCCAAAA CTAGCACCTA GTCCACTCAT 1200

TATCCAGCCC TCTTATTTCT CGGCCCCGCT CTGCTTCAGT GGACCCGGGG AGGGCGGGGA 1260

AGTGGAGTGG GAGACCTAGG GGTGGGCTTC CCGACCTTGC TGTACAGGAC CTCGACCTAG 1320

CTGGCTTTCT TCCCCATCCC CACGTTAGTT GTTGCCCTGA GGCTAAAACT AGAGCCCAGG 1380

GGCCCCAAGT TCCAGACTGC CCCTCCCCCC TCCCCCGGAG CCAGGGAGTG GTTGGTGAAA 1440

GGGGGAGGCC AGCTGGAGAA CAAACGGGTA GTCAGGGGGT TGAGCGATTA GAGCCCTTGT 1500

ACCCTACCCA GGAATGGTTG GGGAGGAGGA GGAAGAGGTA GGAGGTAGGG GAGGGGGCGG 1560

GGTTTTGTCA CCTGTCACCT GCTCCGGCTG TGCCTAGGGC GGGCGGGCGG GGAGTGGGGG 1620

GACCGGTATA AAGCGGTAGG CGCCTGTGCC CGCTCCACCT CTCAAGCAGC CAGCGCCTGC 1680

CTGAATCTGT TCTGCCCCCT CCCCACCCAT TTCACCACCA CCATG 1725

(7) INFORMATION FOR SEQUENCE IDENTIFICATION NUMBER: 6: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 25

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 6: CCGGGAAGTG GTGGGGGGAG GGAGC 25 (8) INFORMATION FOR SEQUENCE IDENTIFICATION NUMBER: 7: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 25

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 7: CCGGGAAGTG GCAGGGGGAG GGAGC 25

(9) INFORMATION FOR SEQUENCE IDENTIFICATION NUMBER: 8: (i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 209

(B) TYPE: nucleic acid

(C) STRANDEDNESS: double

(D) TOPOLOGY: linear

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 8:

CAAGGAGGGA ACCCAGGCTG CTGGAAAGTC CGGCTGGGGC GGGGACTGTG GGTTTCAGGG 60

TAGAACTGCG TGTGGAACGG GACAGGGAGC GGTTAGAAGG GTGGGGCTAT TCCGGGAAGT 120

GGTGGGGGGA GGGAGCCCAA AACTAGCACC TAGTCCACTC ATTATCCAGC CCTCTTATTT 180

CTCGGCCCCG CTCTGCTTCA GTGGACCCG 209

Other embodiments are within the following claims.

What is claimed is:

Claims

1. DNA comprising DF3 enhancer but not encoding DF3.

2. The DNA of claim 1, further comprising a sequence encoding a heterologous polypeptide.

3. The DNA of claim 1, further comprising a sequence antisense to a naturally occurring mRNA, which mRNA is not DF3 mRNA.

4. Isolated DNA comprising DF3 enhancer and a sequence encoding a heterologous polypeptide.

5. The DNA of claim 4, wherein said DNA comprises a nucleotide sequence substantially identical to SEQ ID NO:3 and a nucleotide sequence substantially identical to SEQ ID NO:4.

6. The DNA of claim 4, wherein the DNA further comprises a radiation-inducible element.

7. The DNA of claim 4, wherein the heterologous polypeptide is a cytotoxic polypeptide.

8. A vector comprising the DNA of claim 4.

9. A cell containing the isolated DNA of claim 4.

10. The cell of claim 9, wherein the cell is an carcinoma cell.

11. The cell of claim 10, wherein said heterologous polypeptide is a cytotoxic polypeptide.

12. A therapeutic composition comprising a pharmaceutically acceptable carrier and a therapeutically-effective amount of a nucleic acid, wherein the nucleic acid comprises both DF3 enhancer and a sequence encoding a heterologous polypeptide.

13. A transgenic, non-human animal bearing a transgene comprising DF3 enhancer and a sequence encoding a heterologous polypeptide.