CA2421585A1 - Method and composition for treating tumors by selective induction of apoptosis - Google Patents

Abstract

Compositions and methods are provided for inducing death of cancer cells through apotposis mediated by the death ligand, TRAIL. The method comprising:
introducing an expression vector into a group of cells comprising cells that express a receptor for TRAIL. The expression vector comprises a polynucleotide sequence encoding TRAIL whose expression is preferably regulated by a conditional promoter in the vector. The cells into which the expression vector is introduced express TRAIL when conditions are suitable to activate the conditional promoter. The expressed TRAIL induces cell death in those cells which express the TRAIL receptor through interaction between TRAIL and the receptor such as DR4 and DR5. The method can be used for treating tumors site-specifically and in a dose-adjustable manner.

Description

METHOD COMPOSITION FOR TREATING TUMORS
BY SELECTIVE INDUCTION OF APOPTOSIS
BACKGROUND OF THE INVENTION
FIELD OF THE INVENTION
The present invention relates to compositions and methods for inducing programmed cell death (apoptosis) in cancer cells, and more particularly, relates to compositions and methods for treating tumors by using expression vectors that expresses an apoptosis-signaling ligand such as Fas ligand (Apo-1 ligand) and TRAIL (Apo-2 ligand).
Expression of the apoptosis-signaling iigand induces apoptosis in cells expressing an apoptosis-mediating receptor such as Fas (receptor for Fas ligand), and DR4 or DR5 (receptor for TRAIL).
BACKGROUND OF THE INVENTION
Currently, a major treatment for cancerous tumors is surgical removal of the affected areas of the tissue, organ, or gland. For example, the current treatment for breast cancer is focused on removal of the diseased mammary gland, followed by combination of chemo-and radiation therapy. However, high recurrence rates are a major obstacle to the complete eradication of cancerous cells. It is believed that although the cancer cells in the malignant tumors can be removed surgically, cancerous cells that have invaded the surrounding tissue or lymph nodes frequently cause tumor recurrence. One reason for frequent tumor recurrence may be that during the development of the primary cancer, complete removal of all the cancer cells by surgical procedures is extremely difficult. The remaining cancer cells often remain quiescent for extended periods of time, which is termed tumor dormancy. Meltzer et al. (1990) "Dormancy and breast cancer" J. Surg.
Oncol. 43:181-188. Once the primary tissue is surgically removed, the surgical injury can stimulate rapid tissue and blood vessel regeneration at the wound. These regeneration processes send out positive signals to the surrounding tissue, for example by tissue and vessel growth factors. These factors and the rapid proliferative environment induce the transition of the remaining tumor cells from dormancy to rapid proliferation, and thereby cause reoccurrence of the cancer.
Two basic features are shared by all cancer cells: the uncontrolled cell cycling; and the inability to enter the pathway of programmed cell death, apoptosis. Apoptosis, or programmed cell death (PCD), is a genetically controlled response for cells to commit suicide. The symptoms of apoptosis are viability loss accompanied by cytotoxic boiling, chromatin condensation, and DNA fragmentation.
Wyllie et al. (1980) "Cell death: the significance of apoptosis" Int. Rev.
Cytol. 68:251-306. The apoptotic process has important roles in regulating the development of tissues, the sizes and shapes of organs, and the life span of cells. In the process of tissue and organ development apoptosis accounts for most or all of the PCD responsible for tissue modeling in vertebrate development for the physiological cell death in the course of normal tissue turn over. Apoptosis is also responsible for the extensive elimination of cells of the B and T cell lineages during negative selection in the immune response.
Apoptosis acts as a safeguard to prevent overgrowth of cells and tissues. The development of defects in PCD mechanisms can extend the life span of a cell and can contribute to neoplastic cell expransion.
Also, defects in PCD can contribute to carcinogenesis by permitting genetic instability and accumulation of gene mutations promoting resistance to immune-based destruction and conferring resistance to cytotoxic drugs and radiation. These manifestations indeed are seen in malignant cells not responding to these therapies. Although irradiation, chemotherapy and the appropriate hormone therapy all induce apoptosis to some extent in tumor cells, higher doses of the drugs or radiation may be required for suppressing the growth of cancer cells, which, in turn, can cause severe side effects on patients.
SUMMARY OF THE INVENTION
The present invention provides novel methods and compositions for treating cancer, in particular, solid tumors, by expressing apoptosis signaling ligands such as Fast and TRAIL in a site-specific and controlled manner.
In one aspect, the present invention provides a method for inducing death in cells that express an apoptosis-mediating receptor.
In one, the method comprises: introducing an expression vector into a group of cells comprising cells that express an apoptosis-mediating receptor. The expression vector comprises a polynucleotide sequence encoding an apoptosis-signaling ligand whose expression is preferably regulated by a conditional promoter in the vector. The cells into which the expression vector is introduced express the apoptosis-signaling ligand when conditions are suitable to activate the conditional promoter. The expressed apoptosis-signaling ligand induces cell death in those cells which express the apoptosis-mediating receptor through interaction between the apoptosis-signaling ligand and the apoptosis-mediating receptor.
According to the embodiment, the apoptosis-mediating receptor may be a membrane-bound receptor such as the receptor for Fas ligand, Fas, and the receptors of TRAIL, DR4 and DRS. Optionally, the apoptosis-mediating receptor may be a receptor for tumor necrosis factor (TNF) although TNF may have higher systemic toxicity than Fas and TRAIL.

Also according to the embodiment, the apoptosis-signaling ligand can be any protein that is capable of binding to the apoptosis-mediating receptor. For example, the apoptosis-signaling ligand is an antibody that is capable of binding to Fas (or DR4/DR5) and signals Fas (or DR4/DR5)-mediated apoptosis in cells expressing Fas (or DR4/DR5).
The antibody may be expressed as a single-chain antibody by an expression vector of the present invention and binds to its cognate antigen on the apoptosis-mediating receptor.
Preferably, the apoptosis-signaling ligand is a membrane protein such as Fast and TRAIL. Optionally, the apoptosis-signaling ligand may be TNF although TNF may have higher systemic toxicity than Fas and TRAIL.
Also optionally, the apoptosis-signaling ligand may be a non-membrane-bound protein that can induce apoptosis when expressed intracellularly. Examples of such an intracellular apoptosis-signaling ligand include, but are not limited to, Bax, Bad, Bak, and Bik.
Also according to the embodiment, the expression vector may be a plasmid. The plasmid can be transfected into cancer cells via liposome-mediated delivery or other methods of transfection.
Preferably, the expression vector is a viral vector. The viral vector may be an adenovirus, adeno-associated virus, vaccinia, retrovirus, or herpes simplex virus vector.
Most preferably, the expression vector is an adenoviral vector.
The adenoviral vector may be replication competent or replication incompetent, depending on the dosage of the apoptosis-signaling ligand to be administered into the tumor site.
The expression of the apoptosis-signaling ligand is regulated by a conditional promoter in the expression vector. The conditional promoter may be a tissue-specific promoter such as a prostate-specific promoter, a breast-specific promoter, a pancreas-specific promoter, a colon-specific promoter, a brain-specific promoter, a kidney-specific promoter, a bladder-specific promoter, a lung-specific promoter, a liver-specific promoter, a thyroid-specific promoter, a stomach-specific promoter, an ovary-specific promoter, and a cervix-specific promoter.
Examples of the prostate-specific promoter include, but are not limited to, prostate specific antigen (PSA) promoter and its mutants ~PSA, ARR2PB and probasin (PB) promoters, gp91-phox gene promoter, and prostate-specific kallikrein (hKLK2) promoter.
Examples of the liver-specific promoter include, but are not limited to, liver albumin promoter, alpha-fetoprotein promoter, a,-antitrypsin promoter, and transferrin transthyretin promoter.
Examples of the colon-specific promoter include, but are not limited to, carbonic anhydrase I promoter and carcinoembrogen's antigen promoter.
Examples of the ovary- or placenta-specific promoter include, but are not limited to, estrogen-responsive promoter, aromatase cytochrome P450 promoter, cholesterol side chain cleavage P450 promoter, 17 alpha-hydroxylase P450 promoter.
Examples of the breast-specific promoter include, but are not limited to, G.I. erb-B2 promoter, erb-B3 promoter, ~-casein, ~-lacto-globulin, and WAB (whey acidic protein) promoter.
Examples of the lung-specific promoter include, but are not limited to, surfactant protein C Uroglobin (cc-10, Cllacell 10 kd protein) promoter.
Examples of the skin-specific promoter include, but are not limited to, K-14-keratin promoter, human keratin 1 or 6 promoter, and loicrin promoter.
Examples of the brain-specific promoter include, but are not limited to, glial fibrillary acidic protein promoter, mature astrocyte specific protein promoter, myelin promoter, and tyrosine hydroxylase promoter.
Examples of the pancreas-specific promoter include, but are not limited villin promoter, glucagon promoter, and Insulin Islet amyloid polypeptide (amylin) promoter.
Examples of the thyroid-specific promoter include, but are not limited to, thyroglobulin promoter, and calcitonin promoter.
Examples of the bone-specific promoter include, but are not limited to, Alpha 1 (I) collagen promoter, osteocalcin promoter, and bone sialoglycoprotein promoter.
Examples of the kidney-specific promoter include, but are not limited to, renin promoter, liver/bone/kidney alkaline phosphatase promoter, and erythropoietin (epo) promoter.
Alternatively, the conditional promoter may be an inducible promoter which is activated or suppressed in the presence of an inducing agent, such as tetracycline and its derivatives or analogs (e.g.
doxycycline), steroid such as glucocorticoid, estrogen, androgen, and progestrone.
Also according to embodiment, the method further comprises creating the conditions suitable to activate the conditional promoter, such as delivering to the group of cells tetracycline or deoxycycline, and delivering to the group of cells a steroid selected from the group consisting of glucocorticoid, estrogen, androgen, and progestrone .
Also according to embodiment the expression vector further comprises a reporter gene. The expression vector may express the reporter gene as a fusion protein with the apoptosis-signaling ligand.
Alternatively, the expression vector may express the reporter gene as a single protein bicistronically with the apoptosis-signaling ligand via a mechanism of internal ribosome entry site (IRES) or splicing donor/acceptor sites.

The reporter gene preferably encodes a fluorescent protein such as green, yellow and blue fluorescent proteins, and more preferably green fluorescent protein (GFP).
Also according to the embodiment, the expression vector further comprises a polynucleotide sequence encoding a regulatory protein.
The regulatory protein may be expressed as a fusion protein with the apoptosis-signaling ligand, or expressed as a single protein from a different promoter on the expression vector. Optionally, the regulatory protein may be expressed as a single protein bicistronically with the apoptosis-signaling ligand via a mechanism of internal ribosome entry site (IRES) or splicing donor/acceptor sites.
For example, the regulatory protein may be a protein that causes tissue-specific localization of the apoptosis-signaling ligand.
The method of present invention can be used to treat tumors.
Accordingly, the group of cells to be induced to undergo apoptosis are contained in a solid tumor. Examples of solid tumors include, but are not limited to, breast, prostate, brain, bladder, pancreas, rectum, parathyroid, thyroid, adrenal, head and neck, colon, stomach, bronchi and kidney tumors.
The expression vector may be introduced into a tumor by using any pharmaceutically acceptable routes of administration. For example, the expression vector may be administered into the group of tumor cells parenterally, intraperitoneally, intravenously, intraartierally, transdermally, sublingually, intramuscularly, rectally, transbuccally, intranasally, liposomally, via inhalation, vaginally, intraoccularly, via local delivery by catheter or stent, subcutaneously, intraadiposally, intraarticularly, intrathecally, or in a slow release dosage form.
Preferably, the expression vector is introduced into the tumor by direct injection of the expression vector into the tumor loci.
Optionally, the method can be performed ex vivo where the group of cells into which the expression vector is introduced are contained in a sample taken from a patient having cancer, or contained in contained in a cell culture.
The expression vector may be introduced into a mixture of cells which express Fas and cells which do not express Fas.
Optionally, the expression vector may be introduced into cells which do not express Fas.
Also optionally, the expression vector may be introduced into cells which do express Fas.
Also optionally, the expression vector may be introduced into cells which cells which do not express Fas. By a "bystander effect", those cancer cells expressing Fas near those cells transduced by the expression vector are killed via Fas-Fast interactions.
In another aspect, the present invention provides an adenoviral expression vector that can be used to induce apoptosis of cancer cells.
The adenoviral vector comprises: a conditional promoter, and a polynucleotide sequence encoding a membrane-bound ligand whose expression is regulated by the conditional promoter in the vector, the ligand signaling apoptosis in cells that express an apoptosis-mediating receptor.
Also according to the embodiment, the membrane-bound ligand can be any protein that is capable of binding to an apoptosis-mediating receptor on the surface of cancer cells. Preferably, the membrane-bound protein is Fast or TRAIL. Optionally, the membrane-bound protein may be TNF although TNF may have higher systemic toxicity than Fas and TRAIL.
Also according to the embodiment, the adenoviral vector may be replication competent or replication incompetent, depending on the dosage of the the ligand to be administered into the tumor site.
The expression of the ligand is regulated by a conditional promoter in the adenovirai expression vector. The conditional promoter may be a tissue-specific promoter such as a prostate-specific promoter, a breast-specific promoter, a pancreas-specific promoter, a colon-specific promoter, a brain-specific promoter, a kidney-specific promoter, a bladder-specific promoter, a lung-specific promoter, a liver-specific promoter, a thyroid-specific promoter, a stomach-specific promoter, an ovary-specific promoter, and a cervix-specific promoter.
In yet another, the present invention provides an adenoviral expression vector for tight controlling expression of a target protein in response to tetracycline. The adenoviral expression vector comprises: a tetracycline-responsive element; a polynucleotide sequence encoding a transactivator protein which is capable of binding to the tetracycline-responsive element; and a polynucleotide sequence encoding a target protein whose expression is regulated by the binding of the transactivator protein to the tetracycline-responsive element.
According to this embodiment, the tetracycline-responsive element and the polynucleotide sequence encoding the transactivator protein are positioned at opposite ends of the adenoviral vector. For example, the tetracycline-responsive element is positioned in the E4 region of the adenoviral vector and the polynucleotide sequence encoding the transactivator protein is positioned in the E1 of the adenoviral vector.
Optionally, the adenoviral vector does not include the E3 region of adenovirus.
Also optionally, the adenoviral vector does not include the E4 region of adenovirus except for the Orf6 of the E4 region.
The expression of the target protein may be repressed in the presence of tetracycline or doxycycline. Alternatively, expression of the target protein may be activated in the presence of doxycycline.
Also according to the embodiment, the target protein may be membrane-bound apoptosis signaling protein such as Fast and TRAIL.
Also according to the embodiment, the viral expression vector may further comprise a polynucleotide sequence encoding a reporter protein. The reporter protein and the target protein may be encoded as a fusion protein or expressed as a single protein bicistronically with the target protein via a mechanism of internal ribosome entry site (IRES) or splicing donor/acceptor sites. .
The reporter gene preferably encodes a fluorescent protein such as green, yellow and blue fluorescent proteins, and more preferably green fluorescent protein (GFP).
Also according to the embodiment, the expression vector further comprises a polynucleotide sequence encoding a regulatory protein.
The regulatory protein may be expressed as a fusion protein with the apoptosis-signaling ligand, or expressed as a single protein from a different promoter on the expression vector. Optionally, the regulatory protein may be expressed as a single protein bicistronically with the apoptosis-signaling ligand via a mechanism of internal ribosome entry site (IRES) or splicing donor/acceptor sites.
For example, the regulatory protein may be a protein that causes tissue-specific localization of the apoptosis-signaling ligand.
Examples of the adenoviral vector according to the embodiment, include, but are not limited to, pAdTEr and Ad/FasL-GFPTET.
The expression vectors of the present invention can also be used in combination with other anti=cancer agents such as chemotherapeutics (e.g. alkylating agents, antibiotic agents, antimetabolic agents, hormonal agents and plant-derived agents) and biologic agents (e.g. cytokines, cancer vaccines, and gene therapy delivering tumor suppressing genes). For example, co-administering to the cancer patient the expression vector encoding TRAIL and an anti-cancer drug such as doxorubicin should overcome the resistance by synergistically sensitizing the cancer cells to TRAIL-mediated apoptosis through suppression of apoptosis-inhibiting molecules or upregulation of pro-apoptosis molecules by the drug. Therefore, by using the combination therapy of the present invention, cancer patients may be treated with subtoxic amount of chemotherapeutics and yet achieve a better clinical efficacy without suffering from severe side effects associated with using high dosages of chemotherapeutics.
BRIEF DESCRIPTION OF THE FIGURES
Figures 1A, 1 B, and 1 C schematically show the pLAd-C.tTA
vector, the pRAd.T.GFsL vector, and the rAd/FasL-GFPTET vector, respectively. In Figure 1A, the pLAd-C.tTA vector is shown. This plasmid contains the leftmost 450 by of Ad5 genome, followed by a strong CMVie enhancer/promoter and a tTA gene from pUHD15-1 inserted into the MCS. Adapter contains restriction sites Xba1, Avr2 and Spe1, all of which generate cohesive ends compatible with Xba1. After assembly into rAd vectors, E1A poly A is utilized for efficient tTA
expression. A similar strategy was used to construct pLAd vectors containing other transgenes. In Figure 1 B, the pRAd.T.GFsL vector is shown. This plasmid contains Ad5 (sub360) sequences from the unique EcoR1 site (27333 bp) to the right ITR (35935 bp), with E3 and E4 deletions (the Orf6 of E4 is retained). The diagram shows the structure of the regulatable Fast-GFP expression cassette, consisting of the THE
promoter, Fast-GFP fusion protein and bovine groth hormone (BGH) poly A. This cassette was inserted into a MCS at 35810 bp. In vitro assembly of the rAd/FasL-GFPTErvector is shown in Figure 1 C. The region of the junction between the GFP and Fast reading frames is expanded. Other rAd vectors were generated using a similar strategy.
Figure 2 is a graph showing a comparison of titers of rAd vectors with Fast activity in 293 and 293CrmA cells. Twelve-well plates were seeded with 104 293 or 293CrmA cells and infected with r-Ad/FasL, rAdFasL-GFPTEr, or rAd/LacZ at MOI of 5 one day later. Fourty-eight hours post-transduction, cells were collected and lysed. Lysates were titrated and PFU/ml determined on 293CrmA cells. Results represent means and average errors of 2 sets of independent experiments.
Figure 3 illustrates the construction of the TRAIL expression vector Ad.TRAIL/GFPTer which was constructed by using similar methods described in the legend of Figure 1 except that TRAIL and GFP
genes are separated by an IRES which facilitates bicistronical expression of these two genes.
Figure 4 shows different sensitivities of cancer cells to Fast- and TRAIL-induced apoptosis. Cancer cells, A459, HeLa, LnCP, and C3A, were analyzed for susceptbility to adenovirus infection and sensitivity to Fast- and TRAIL-induced apoptosis. Cells were infected at MOI 10 with AdGFP (pannels in the first and the second columns from left), Ad/FasL-GFPTEr (the third column) and Ad.TRAIL/GFPTET (the forth column). The susceptibility of adenovirus infection of the cells are represented by the number of GFP expression cells (the first collum), the morphology of the cells are shown in the bright-field view (second column). Morphology of the cells infected with Ad/FasL-GFPTET and Ad.TRAIL/GFPTET are shown in panels in the third and the forth colum, respectively.
Figure 5 shows that TRAIL expression does not induce apoptosis in untransformed fibroblasts. To determine that if TRAIL expression will induce apoptosis in normal cells, low-passage human foreskin fibroblasts were infected with AdGFP, Ad/FasL-GFPTET, and Ad.TRAIL/GFPTET at MOI about 10. The bright-field veiw shows the normal morphology of fibroblasts transduced with AdGFP ( panel GFP
hFF). Fibroblasts demonstrated poor infectability by adenovirus as shown by the low number of GFP expression cells (panel GFP).
However, these cells are highly sensitive to Fast induced apoptosis (panel FasL). In contrast, no apparent apoptosis can be observed in TRAIL transduced cells (panel TRAIL), even at five folds of the MOI
(panel TRAIL x5).
Figure 6 shows suppression of the growth of human breast turmors implanted in nude mice by injection of an adenoviral vector of the present invention (Ad/FasL-GFPTET vector) which comprises Fas ligand. Equal numbers of breast cancer cells were implanted in each side of six mice. Tumors on the right side of the mice were injected with the Ad/FasL-GFPTEr vector, and tumors on the left side of the same mice were injected with a control vector, Ad/LacZ. In four of the six mice, most of the tumor masses disappeared after one injection (indicated by yellow arrows). In two of the mice, suppression of tumor growth was greater than 80% (black arrows) in comparison to tumors on the control side of the same mice.
DETAILED DESCRIPTION OF THE INVENTION
The present invention provides novel methods and expression vectors for treating cancer, in particular, solid tumors, by expressing apoptosis-signaling ligands such as Fast and TRAIL in a site-specific and controlled manner. The controlled expression of these apoptosis-signaling ligands should significantly reduce cytotoxicity associated with uncontrolled, systemic administration of these ligands.
According to the present invention, an expression vector such as an adenoviral vector carrying genes encoding the apoptosis-signaling ligand (e.g. FasL and TRAIL) can be introduced into the tumor site via many pharmaceutically acceptable routes of administration. The cells transduced by the adenovirus expresses the ligand, preferably, as a membrane-bound protein. Through interactions between the apoptosis-signaling ligand (e.g. TRAIL) and an apoptosis-mediating receptor (e.g.
DR4 and DR5) in the cell, a cascade of signal transduction occurs. The event triggers multiple apoptosis pathways in which the apoptosis signal is amplified by expression of multiple apoptotic enzymes such as proteases and endonucleases. Since the interactions between the ligand and the receptor can occur between two cells, the tumor cells that are not transduced by adenovirus can be induced to undergo apoptosis due to a "bystander effect". This effect may be due to specific interactions between the apoptosis-signaling ligand expressed in cells transduced by the adenovirus and the apoptosis-mediating receptor expressed on the surface of the untransduced tumor cells. Therefore, by using the method of the present invention the efficiency of cell killing should be higher than those approaches involving direct injection of the ligand as a protein or cells expressing the ligand.
One important feature of the present invention is that expression of the apoptosis-signaling ligand is controlled by a conditional promoter, such as a tissue-specific or an inducible promoter. By controlling the expression of the ligand site-specifically (e.g. using a tissue-specific promoter) and/or flexible adjustment of dosage (e.g. using an inducible promoter), potential systemic toxicity of the ligand should be significantly reduced.
fn particular, the adenoviral vector encoding the ligand can be directly injected into the tumor site and locally transfers the ligand into the tumor cells. Depending on the dosage of the ligand to be delivered, the adenoviral vector can be replication competent or replication incompetent. Once injected into the tumor, the adenovirus transduces the tumor cells which, as a result, expresses high levels of the ligand locally. Through interactions between the ligand and the receptors) expressed on the surface of the tumor cells, the apoptosis signal is amplified by expression of multiple proteins and enzymes along the pathways of the ligand-induced apoptosis. Thus, massive tumor cells can be eradicated with minimum injuries to surrounding healthy tissues.
In a sense, this approach provided by the present invention is like a "molecular surgery" which is more precise and safer than conventional approaches involving undiscriminating, uncontrolled administration of cancer therapeutics.
By using the methods of the present invention, primary tumors can be eradicated and meanwhile, the reoccurrence of the cancer can be prevented by activating cancer cell apoptosis at the tumor site.
In one aspect, the present invention provides a method for .
inducing death in cells that express an apoptosis-mediating receptor.
The mode of death may be necrosis, apoptosis or combination of both.
The method comprises: introducing an expression vector into a group of cells comprising cells that express an apoptosis-mediating receptor. The expression vector comprises a polynucleotide sequence encoding an apoptosis-signaling ligand whose expression is preferably regulated by a conditional promoter in the vector. The cells into which the expression vector is introduced express the apoptosis-signaling ligand when conditions are suitable to activate the conditional promoter.
The expressed apoptosis-signaling ligand induces cell death in those cells which express the apoptosis-mediating receptor through interaction between the apoptosis-signaling ligand and the apoptosis-mediating receptor.
According to the embodiment, the apoptosis-mediating receptor may be a membrane-bound receptor such as the receptor for Fas ligand, Fas, and the receptors of TRAIL, DR4 and DRS. Optionally, the apoptosis-mediating receptor may be a receptor for tumor necrosis factor (TNF) although TNF may have higher systemic toxicity than Fas and TRAIL.
Also according to the embodiment, the apoptosis-signaling ligand can be any protein that is capable of binding to the apoptosis-mediating receptor. For example, the apoptosis-signaling ligand is an antibody that is capable of binding to Fas (or DR4/DR5) and signals Fas (or DR4/DR5)-mediated apoptosis in cells expressing Fas (or DR4/DR5).
The antibody may be expressed as a single-chain antibody by an expression vector of the present invention and binds to its cognate antigen on the apoptosis-mediating receptor.
Preferably, the apoptosis-signaling ligand is a membrane protein such as Fast and TRAIL. Optionally, the apoptosis-signaling ligand may be TNF although TNF may have higher systemic toxicity than Fas and TRAIL.
1. Apoptosis-mediating receptors and apoptosis-signaling ligands According to the present invention, the apoptosis-mediating receptor is death receptor that mediates programmed cells death upon binding with an apoptosis signaling ligand. The receptor may be a cell-surface receptor that is membrane-bound, or resides in cytoplasm or nucleus. In a preferred embodiment, the apoptosis-mediating receptor is a cell membrane-associated receptor. A prominent example of such an apoptosis-mediating receptor belongs to the tumor necrosis factor (TNF) receptor superfamily.
The TNF receptor superfamily is defined by the presence of related, cysteine-rich, extraceliular domains. Examples of TNF
receptors include, but are not limited to NTR/GFR (p75) such as NGF, BDNF, NT-3 and NT-4, TNF-R1 (CD120a), TNF-R2 (CD120b), Fas (CD5/Apo-1), DR3 (TRAMP/WSL-1), DR4 (TRAIL-R1), DR5 (TRAIL-R2), DcR1 (TRAIL-R3), DcR2 (TRAIL-R4), CD30, CD40, Cd27, 4-1 BB
(CD137), OX-40, LT-~iR, human HVEM (herpes virus early mediator), OPG (osteoprotegerin)/OC1 F, and RANK. Ashkenazi and Dixit (1999) "Apoptosis control by death and decoy receptors" Curr. Opin. Cell Biol.
11:255-260.

All of the receptors are type I transmembrane proteins with an extracellular region composed of two-six cysteine rich domains that are about 25% identity among members and contribute to ligand binding.
Fas, TNF-R1, TRAIL-DR4, DRS, TRAMP (DR3), CAR1 have similar cytoplasmic domains. Sequence comparison of the intracellular region of these receptors revealed a homologous, well-conserved region of about 80 amino acids called the death domain. Orlinck and Chao (1998) "TNF-related ligands and their receptors" Cell Signal 10:543-551.
The death domain is required for the specific recruitment of cellular signaling molecules (adaptor proteins) that are implicated in apoptosis.
Nagata (1997) "Apoptosis by death factor" Cell 88:355-365.
The ligands that bind to the receptors in the TNF receptor superfamily include, but are not limited to, neorotrophins, TNF-a, Fas ligand (FasL/CD-95L/Apo-1 L), TRAIL/Apo-2L, CD30L, CD40L, CD27L, 4-1 BBL, OX-40L, and lymphotoxin (LT) a, ~. Except for LT-oc, all ligands are synthesized as type II membrane proteins; their N-terminus is in the cytoplasm and their C-terminus extends into the extracellular region. Nagata (1997) "Apoptosis by death factor" Cell 88:355-365. A
region of about 150 amino acid residues in the extracellular domain is 20-25% homologous among the TNF family members.
A common feature of the ligands is that all active ligands are composed of three identical subunits (trimers) and activate their respective receptors by oligomerization. Schulze-Osthoff et al. (1998) "Apoptosis by death receptors" Eur. J. Biochem. 254;439-459. Although most members are found as membrane-bound molecules; specific metalloproteases are capable of generating soluble forms. The zinc-dependent metalloprotease for TNF-a called TACE is one example of such specific metalloproteases. Orlinck and Chao (1998) "TNF-related ligands and their receptors" Cell Signal 10:543-551.
2. Fas ligand-mediated apoptosis In a preferred emobodiment, the apoptosis-signaling ligand is Fas ligand. According to the present invention, controlled expression of Fas ligand by an expression vector in tumor site should induce apoptosis in cells expressing Fas through Fas-Fast interactions while minimizing side effects associated with undiscriminating attack of Fas ligand to those normal cells which also express Fas.
Fas (APO-1, CD95), or the Fas ligand receptor, is a 45 kDa type I
membrane protein and belongs to the TNF/nerve growth factor receptor superfamily. Bajorath, J. and A. Aruffo. (1997) "Prediction of the three dimensional structure of the human Fas receptor by comparative molecular modeling" J. Comput Aided Mol Des 11:3-8; and Watanabe-Fukunaga, R., C. 1. Brannan, N. Itoh, S. Yonehara, N. G. Copeland, N.
A. Jenkins and S. Nagata "The cDNA structure, expression, and chromosomal assignment of the mouse Fas antigen" J. Immunol.
148:1274-9.
The ligand of Fas, Fast, is a 40-kDa type II membrane protein belonging to the tumor necrosis factor family. Takahashi, T., M. Tanaka, J. Inazawa, T. Abe, T. Suda and S. Nagata. (1994) "Human Fas ligand:
gene structure, chromosomal location and species specificity" lnt.
Immunol. 6:1567-74. Binding of Fast (and certain anti-Fas antibodies) to Fas causes receptor oligomerization and sends a signal through a caspase pathway, resulting in rapid death of receptor-bearing cells through apoptosis. Larsen, C. P., D. Z. Alexander, R. Hendrix, S. C.
Ritchie and T. C. Pearson. (1995) "Fas-mediated cytotoxicity. An immunoeffector or immunoregulatory pathway in T cell-mediated immune responses?" Transplantation 60:221-4; Longthorne, V. L. and G. T. Williams. (1997) "Caspase activity is required for commitment to Fas-mediated apoptosis" EMBO. J. 16:3805-12; Nagata, S. and P.
Golstein. (1995) "The Fas death factor" Science 267:1449-56; and Ogasawara, J., R. Watanabe-Fukunaga, M. Adachi, A. Matsuzawa, T.

Kasugai, Y. Kitamura, N. Itoh, T. Suda and S. Nagata. (1993) "Lethal effect of the anti-Fas antibody in mice" [published erratum appears in (1993) Nature Oct 7;365(6446):568] Nature 364:806-9.
Fas is expressed in almost all cell types. When Fas binds to Fast, it activates the genetically programmed cell death through a cascade expression of interleukin-coupled enzymes (ICE) or caspases.
Chandler et al. (1998) "Different subcellular distribution of caspase-3 and caspase-7 following Fas-induced apoptosis in mouse liver" J. Biol.
Chem. 273:10815-10818; Jones et al. (1998) "Fas-mediated apoptosis in mouse hepatocytes involves the processing and activation of caspases" Hepatology 27:1632-1642.
Since both ligand and receptor are membrane proteins, Fas-induced apoptosis is normally mediated through cell-cell contact.
However, a soluble form of Fast is also produced by some cells and has been shown to have a somewhat altered activity, depending on the target cell Tanaka. M., T. Itai, M. Adachi and S. Nagata (1998) "Downregulation of Fas ligand by shedding" [see comments]. Nat. Med.
4:31-6; and Tanaka, M., T. Suda, T. Takahashi and S. Nagata (1995) "Expression of the functional soluble form of human fas ligand in activated lymphocytes" EMBO. J. 14:1129-35.
For example, the present invention provides a method for inducing death of tumor cells expressing Fas (Fas+ cells) by a vector-mediated gene transfer of a Fas ligand to the cells. In this method, the vector-transduced cell expressing the Fas ligand induces Fas+ tumor cells to undergo apoptosis and die. The vector may be injected into the tumor with a syringe or a micropump, thus eliminating the need for conventional surgery to remove the tumor.
There may be multiple mechanism by which Fast expressed by the cancer cells transduced by the vector. The cancer cell death may be induced in several ways: 1) Fast binds to Fas on adjacent tumor cells and induces their apoptosis; 2) Fast induces apoptosis of endothelial cells and destroys the blood vessels supplying the tumor; 3) expression of Fast on tumor cells induces apoptosis of surrounding tissues and deprives tumor cells of any nursery support; and 4) apoptosis prevents the release of positive factors that may reactivate quiescent tumor cells responsible for reoccurring cancers.
A major advantage of this approach is that the Fas-Fast interaction is the major signaling event that activates several apoptosis pathways, following both p53-dependent and independent pathways.
Callers et al. (1998) "Fas-mediated apoptosis with normal expression of bcl-2 and p53 in lymphocytes from aplastic anemia" Br. J. Haematol.
100:698-703. Thus, apoptosis signaling is amplified by more than one cascade of enzyme expressions, and the apoptosis does not depend on p53 or other cell-cycle checkpoint proteins. For example, although gene therapy with the p53 gene has shown great promise in treating cancers, (8oulikas (1997) "Gene therapy of prostate cancer: p53, suicidal genes, and other targets." Anticancer Res. 17:1471-1505), p53 gene therapy may be effective in about 50-60% of the tumor cells that have a p53 mutation. Iwaya et al. (1997) "A histologic grade and p53 immunoreaction as indicators of early recurrence of node-negative breast cancer" Jpn J Clin Oncol 27:6-12.
Another advantage is that Fast is generally a membrane-bound signaling protein rather than an intracellular protein, such as p53 and caspases. Fast expression on the cell surface transmits the apoptotic signal to surrounding cancer cells by a strong "bystander effect", and does not require delivering the therapeutic gene into all cancer cells.
Therefore, the present invention fulfills the need for a non-surgical method of cancer treatment that provides significant improvement over current gene therapy methods, avoids the use of toxic drugs and helps prevent tumor recurrence.
By expressing Fas ligand in a controlled manner, e.g. via a control of a tissue-specific or an inducible promoter, growth of tumors can be suppressed by selectively promoting apoptosis in tumors and systemic toxicity of Fas Ligand can be reduced.
3. TRAIL-mediated selective apoptosis of cancer cells TRAIL, or Apo-2 ligand, is a 281 amino acid, type II
transmembrane protein and is most closely related to Fast (28% amino acid homology). Like Fast, TRAIL can kill many sensitive tumor cell lines in 4-8 h. In contrast, TNF kill tumor cell lines in more than 24 h.
Wiley et al. (1995) "Identification and characterization of a new member of the TNF family that induces apoptosis" Immunity 3:673-682. The TRAIL receptors, DR4 and DRS, like the full-length Fas receptors, contain a death domain that possibly interacts with an adaptor molecule (e.g. FADD (Fas-associated death domain)-like adaptor) in order to mediate the apoptosis signal.
The initiation of TRAIL-mediated apoptosis involves the clustering of three DR4 or DR5 on the target cell surface by cross-linking the receptors with the ligand (TRAIL). Upon oligomerization of the receptors, an adaptor molecule similar to FADD is recruited to the DR4 or DR5 receptor cluster via death domain interactions. Chinnaiyan et al.
(1996) "Signal transduction by DR3, a death-domain-containing receptor related to TNFR-1 and CD95" Science 274:990-992.
The cross-linking of agonistic receptors DR4 and DR5 to TRAIL can be inhibited by the decoy receptors (DcR1 and DcR2). Sheridan et al.
(1997) "Control of TRAIL-induced apoptosis by a family of signaling and decoy receptors" Science 277:818-821. The decoy receptor are able to inhibit TRAIL-mediated apoptosis because they lack functional death domain to mediate the death signal and they can compete with the binding to TRAIL by DR4 and DRS. Griffith et al. (1999) "Functional analysis of TRAIL receptors using monoclonal antibodies" J. Immunol.
162:2597-2605.

The TRAIL adaptor molecule similar to FADD possibly contains a death effector domain that binds to FLICE (capase-8), the aspartate-specific cysteine protease that initiates a caspase amplification cascade leading to the ultimate apoptotic phenotypes. Muzio (1998) "Singling by proteolysis: death adaptors induce apoptosis" Int. J. Clin. Lab. Res. 28:
141-147. When the adaptor is recruited to the death domain of the TRAIL receptors DR4 or DRS, FLICE zymogen is brought together in close proximity by the FADD-like adaptor and is activated by by FLICE
auto-cleavage. The FLICE activating complex that consists of TRAIL
receptor-adaptor-FLICE is named as DISC (death inducing signaling complex). Kischkel et al. (1995) "Cytotoxicity-associated dependent APO-1 (Fas/CD95)-associated protein form a death-inducing signaling complex (DISC) with the receptor" EMBO J. 14:5579-5588. The FLICE
enzyme subsequently activates caspase-3 and other caspases by cleaving their zymogen forms. Martinet-Lorenzo et al. (1998) "Involvement of Apo-2 ligand/TRAIL in activation-induced death of Jurkat and human peripheral blood T cells" Euro. J. Immunol. 28:2714-2725. Active caspase-3 can then cleave ICAD (inhibitor of caspase-activated deoxy-ribonuclease), resulting in the release of active nuclease that cleaves DNA into 180-220 by fragments, a typical hallmark of apoptosis.
TRAIL expression has been detected in a wide variety of human tissues, with highest levels found in spleen, lung and prostate. Wiley et al. (1995) "Identification and characterization of a new member of the TNF family that induces apoptosis" Immunity 3:673-82. In the present invention, it is demonstrated that compared to normal cells cancer cells have selective sensitivities to TRAIL-induced apoptosis. For example, while human cancer cells line, such as LNCAP (prostate), HeLa (cervical), A549 (lung), and C3A (liver), are sensitive to TRAIL-mediated apoptosis, primary human fibroblasts from foreskin samples are essentially unaffected when similar levels of TRAIL are expressed in the cells. Thus, compared to Fast, TRAIL induces apoptosis in a more tumor-specific manner, which, in turn, can have a less systemic toxicity when expressed in vivo.
There may be many possible reasons why tumor cells are particularly sensitive to TRAIL-mediated apoptosis. One possibility is that healthy normal cells may express intracellular regulators such as FLICE-inhibitory proteins (FLIPs) that blocks the biochemical signaling pathways that lead to cell death. Griffith et al. (1998) "Intracellular regulation of TRAIL-induced apoptosis in human melanoma cells" J.
Immunol. 161:2833-2840. It may also be possible that the lack of cytotoxic effects of TRAIL on normal cells may be due to expression of decoy receptors such as DcR1 and DcR2 which inhibit TRAIL-mediated apoptosis by competing with DR4 or DR5 for binding to TRAIL.
By using the method of present invention, TRAIL can be introduced into cancer cells by a conditional expresssion vector such as an adenoviral vector and induces apoptosis of cancer cells selectively.
Since TRAIL exerts less toxicity to normal cells and its expression can be controlled site-specifically and dose-dependently, systemic toxicity of this ligand should be reduced.
4. Expression vectors for apoptosis-signaling ligands The expression vector that can be used to practice the methods of the present invention may be any gene-transferring vector. The expression vector may be a plasmid encoding the apoptosis-signaling ligand (e.g. TRAIL). The plasmid can be transfected into cancer cells via liposome-mediated delivery or other methods of transfection.
Preferably, the expression vector is a viral vector. The viral vector may be an ~adenovirus, adeno-associated virus, vaccinia, retrovirus, or herpes simplex virus vector.
The present invention provides an adenoviral vector that is preferably used to induce death of cancer cells in a site-specific and controlled manner. The expression of the apoptosis-signaling ligand may be controlled by using a tissue-specific promoter or an inducible promoter. Alternatively, the expression of the apoptosis-signaling ligand may be constitutive in the transduced cells. The adeniviral expression vector can be used for delivering the apoptosis-signaling ligand to a wide range of cell types both in vitro and in vivo. Further, the expression of apoptosis-signaling can be tightly regulated, which not only facilitates production of adenoviral expression vectors encoding the apoptosis-signaling ligand, but also provides a means for controlling expression of the ligand in vivo to minimize systemic toxicity. In addition, the present invention also provides means for easily and reliably quantitating the levels and cellular localization of exogenous apoptosis-signaling ligands.
In a preferred embodiment, the adenoviral vector comprises: a conditional promoter, and a polynucleotide sequence encoding a membrane-bound ligand whose expression is regulated by the conditional promoter in the vector, the ligand signaling apoptosis in cells that express an apoptosis-mediating receptor. The adenoviral vector may be replication competent or replication incompetent, depending on the dosage of the apoptosis-signaling ligand to be administered into the tumor site.
The membrane-bound ligand can be any protein that is capable of binding to an apoptosis-mediating receptor on the surface of cancer cells. Preferably, the membrane-bound protein is Fast or TRAIL.
Optionally, the membrane-bound protein may be TNF although TNF
may have higher systemic toxicity than Fas and TRAIL.
Alternatively, the adenoviral vector may encode another type of apoptosis-signaling ligand such that when that the ligand is introduced into a cell, the transduced cell expresses the ligand intracellularly.
Interactions of the ligand with an apoptosis-mediating receptor causes the cell to undergo apoptosis. Examples of such intracellular apoptosis signaling molecules include, but are not limited to, Bax, Bad, Bak, and Bik. Adams et al. "Control of cell death" WEHI Annual Report 1996/1997.
In another embodiment of the present invention, the expression vector encoding the apoptosis-signaling ligand can also encode another protein such as a regulatory protein, which may be used to regulate the expression of the ligand. For example, the regulatory protein can cause the tissue-specific localization of the Fas ligand on the cell membrane, or alternatively cause the premature turn-over of the Fas ligand in non-target cells, or regulate the expression of the Fast via regulation of transcription and/or translation.
The regulatory protein can also be encoded by another expression vector that is delivered to the cell, either concurrently or consecutively with the nucleic acid encoding the protein to be expressed. In this embodiment, the two expression vectors can have different sequences, such as different promoters, such that they can be independently regulated, such as by the administration of a drug that selectively regulates the expression of one or both of the promoters, such as by the use of a steroid hormone, e.g. a glucocorticoid hormone that can regulate a promoter that is inducible by that hormone. Other steroid hormones fihat may be used include, but are not limited to, estrogen, androgen, and progestrone.
The apoptosis-signaling ligand may also be expressed as a fusion protein with another protein. This protein fused with the ligand may be used for such purposes as localization of the protein, activation or deactivation of the ligand, monitoring the location of the ligand, isolation of the ligand, and quantitating the amount of the ligand.
In one embodiment, the fusion protein comprises a Fas ligand and reporter protein such as a fluorescent protein (FP). Examples of reporter proteins include, but are not limited, the GFP (green fluorescent protein) gene, the YFP (yellow fluorescent protein) gene, BFP (blue fluorescent protein) gene, the CAT gene, the neo gene, the hygromycin gene, and so forth. An example of a Fast-GFP fusion protein-expressing construct is shown in Figure 1 and is further described herein.
Alternatively, the reporter gene may be expressed as a single protein bicistronically with the apoptosis-signaling ligand via a mechanism of internal ribosome entry site (IRES) or splicing donor/acceptor sites.
The expression vector may further encode a sequence that is capable of regulating the expression of the apoptosis-signaling ligand.
For example, the vector can contain a glucocorticoid regulatory element (GRE) such that glucocorticoid hormones can be used to regulate the expression of the Fas ligand.
Another example of a regulatory sequence that can regulate the expression of an adjacent gene is by cloning an RNA aptamer, such as H10 and H19, into the promoter region whereby administration of a drug such as Hoechst dye 33258 can block expression of the gene in vivo.
Werstuck et al. (1998) "Controlling gene expression in living cells through small molecule-RNA interactions" Science 282:296-298.
In other embodiments of the present invention, the regulatory sequence comprises the Tet-operon or the lac operon, or any other operon that can function as a regulatory sequence in a eukaryotic cell.
In a preferred embodiment, expression of apoptosis-signaling ligand is under the control of tetracycline-regulated gene expression system, wherein expression of the ligand is controlled by a tet-responsive element, wherein the ligand expression requires the interaction of the tet-responsive element and a tet transactivator.
In a more preferred embodiment, tight control of the ligand expression is achieved using an Ad vector in which the tet-responsive element arid the transactivator element are built into the opposite ends of the same vector to avoid enhancer interference. Expression can be conveniently regulated by tetracycline or any derivative thereof, which includes, but is not limited to, doxycycline, in a dose-dependent manner.
For example, the vector efficiently delivers Fast-GFP gene to cells in vitro, and the expression level of the fusion protein may be modulated by the concentration of doxycycline in culture media. An example of such a regulatory system is particularly described herein.
In one embodiment of the present invention, the promoter is a tissue-specific promoter which one skilled in the art will appreciate can confer tissue-specificity to the expression of the nucleic acid encoding the apoptosis-signaling ligand such as Fast and TRAIL.
For example, the tissue-specific promoter may be a prostate-specific, a breast tissue-specific, a colon tissue-specific, a pancreas-specific a brain-specific, a kidney-specific, a liver-specific, a bladder-specific, a bone-specific, a lung-specific, a thyroid-specific a stomach-specific, an ovary-specific, or a cervix-specific promoter.
Where the tissue-specific promoter is a prostate-specific promoter, the promoter includes, but is not limited to the PSA promoter, the ~PSA promoter, the ARR2PB promoter, the PB promoter, gp91-phox gene promoter, and prostate-specific kallikrein (hKLK2) promoter.
Where the tissue-specific promoter is a breast-specific promoter, the promoter includes, but is not limited to MMTV promoter, G.I. erb-B2 promoter, erb-B3 promoter, (i-casein, ~3-lacto-globulin, and WAB (whey acidic protein) Where the tissue-specific promoter is a liver-specific promoter, the promoter includes, but is not limited to liver albumin promoter, alpha-fetoprotein promoter, a~-antitrypsin promoter, and transferrin transthyretin promoter.
Where the tissue-specific promoter is a brain-specific promoter, the promoter includes, but is not limited to, JC virus early promoter, tyrosine hydoxylase promoter, dopamine hydroxylase promoter, neuron specific enolase promoter, glial fibrillary acidic protein promoter, mature astrocyte specific protein promoter, and myelin promoter.
Where the tissue-specific promoter is a colon-specific promoter, the promoter includes, but is not limited to, the MUC1 promoter, carbonic anhydrase I promoter and carcinoembrogen's antigen promoter.
Where the tissue-specific promoter is ovary- or placenta-specific promoter, the promoter includes, but is not limited to, estrogen-responsive promoter, aromatase cytochrome P450 promoter, cholesterol side chain cleavage P450 promoter, and 17 alpha-hydroxylase P450 promoter.
Where the tissue-specific promoter is a lung-specific promoter, the promoter includes, but is not limited to, surfactant protein C
Uroglobin (cc-10, Cllacell 10 kd protein) promoter.
Where the tissue-specific promoter is a skin-specific promoter, the promoter includes, but is not limited to, K-14-keratin promoter, human keratin 1 or 6 promoter, and loicrin promoter.
Where the tissue-specific promoter is a pancreas-specific promoter, the promoter includes, but is not limited to, villin promoter, glucagon promoter, and Insulin Islet amyloid polypeptide (amylin) promoter.
Where the tissue-specific promoter is a thyroid-specific promoter, the promoter includes, but is not limited to, thyroglobulin promoter, and calcitonin promoter.
Where the tissue-specific promoter is a bone-specific promoter, the promoter includes, but is not limited to, Alpha 1 (I) collagen promoter, osteocalcin promoter, and bone sialoglycoprotein promoter.
Where the tissue-specific promoter is a kidney-specific promoter, the promoter includes, but is not limited to, renin promoter, liver/bone/kidney alkaline phosphatase promoter, and erythropoietin (epo) promoter.
It should be noted that other tissue specific promoters will be revealed by the human genome project and other endeavors of human gene discovery. These promoters will be useable as appropriate means to direct tissue specific expression from the expression vectors of the present invention.
Furthermore, one of ordinary skill will readily know how to identify a promoter specific to a particular cell type. For example, by comparing the differential expression of genes in different tissue types, e.g., using gene chip technology, one can identify genes expressed only in one particular tissue type. These genes can then be isolated and sequenced, and their promoters may be isolated and tested in an animal model for the ability to drive tissue specific expression of a heterologous gene. Such methods are well within the ability of the one of ordinary skill in the art. An example of a method by which a tissue specific promoter may be identified may be found in Greenberg et al. (1994) Molecular Endocrinology 8: 230-239.
The tissue-specificity may also be achieved by selecting an expression vector that has a high degree of tissue specificity. For example, a vector that selectively infects mucosal cells, such as those associated with colon cancer, can be chosen, and then optionally, used in combination with a specific delivery means, such as by the use of a suppository, to selectively deliver the nucleic acid encoding the apoptosis-signaling ligand such as Fas and TRAIL to those desired cells.
One skilled in the art will recognize that various vectors have more or less applicability depending on the particular host. One example of a particular technique for introducing nucleic acids into a particular host is the use of retroviral vector systems which can package a recombinant retroviral genome. See e.g., Pastan et al. (1988) "A
retrovirus carrying an MDR1 cDNA confers multidrug resistance and polarized expression of P-glycoprotein in MDCK cells." Proc. Nat. Acad.
Sci. 85:4486; and Miller et al. (1986) "Redesign of retrovirus packaging cell lines to avoid recombination leading to helper virus production." Mol.
Cell Biol. 6:2895. The produced recombinant retrovirus can then be used to infect and thereby deliver to the infected cells a nucleic acid sequence encoding the apoptosis-signaling ligand. The exact method of introducing the nucleic acid into mammalian cells is, of course, not limited to the use of retroviral vectors. Other techniques are widely available for this procedure including the use of adenoviral vectors (Mitani et al. "Transduction of human bone marrow by adenoviral vector." Human Gene Therapy 5:941-948 (1994)), adenoassociated viral vectors (Goodman et al. "Recombinant adenoassociated virus-mediated gene transfer into hematopoietic progenitor cells." Blood 84:1492-1500 (1994)), lentiviral vectors (Naidini et al. "In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector." Science 272:263-267 (1996)), pseudotyped retroviral vectors (Agrawal et al. "Cell-cycle kinetics and VSV-G pseudotyped retrovirus mediated gene transfer in blood-derived CD34+ cells." Exp. Hematol. 24:738-747 (1996)), vaccinia vectors, and physical transfection techniques (Schwarzenberger et al.
"Targeted gene transfer to human hematopoietic progenitor cell lines through the c-kit receptor." Blood 87:472-478 (1996)). This invention can be used in conjunction with any of these or other commonly used gene transfer methods. In a preferred embodiment of the present invention, the specific vector for delivering the nucleic acid encoding a Fas ligand comprises an adenovirus vector.
5. Expression vectors encoding a trans-regulatory protein Because it is desirable to be able to regulate expression of the apoptosis-signaling ligand, the present invention also provides an expression vector for the regulatable expression for tightly controlling expression of a target protein (e.g. FasL and TRAIL).
The expression vector comprises: a transcription regulatory sequence; a polynucleotide sequence encoding a trans-acting regulator protein which is capable of binding to the transcription regulatory sequence; and a polynucleotide sequence encoding a target protein whose expression is regulated by the binding of the traps-acting regulator protein to the transcription regulatory sequence.
The transcription regulatory sequence and the polynucleotide sequence encoding the traps-acting regulator protein may be positioned at opposite ends of the adenoviral vector. For example, the transcription regulatory sequence is positioned in the E4 region of the adenoviral vector and the polynucleotide sequence encoding the traps-acting protein is positioned in the E1 of the adenoviral vector.
In this vector, the nucleic acid encoding the target protein is operatively linked to a transcription regulatory sequence. The expression of the target protein may be inducible, e.g. expression of Fast or a Fast fusion will not proceed unless the appropriate activator for the particular transcription regulatory sequence is present.
Alternatively, the expression of the target protein may be repressible, i.e., expression of Fast or a Fast fusion will proceed unless the appropriate repressor for the particular transcription regulatory sequence is present.
The traps-acting regulator protein interacts with the transcription regulatory sequence to affect transcription of the target protein. Where the transcription regulatory sequence is inducible, the traps-acting regulator protein is a traps-activator. Where the transcription regulatory sequence is repressible, the traps-acting factor is a traps-repressor.
In a more preferred embodiment, the transcription regulatory sequence is a tet responsive element (TRE), and the traps-acting factor is a tet-responsive transacting expression element (tTA).
In the most preferred embodiment, the invention utilizes the vector Ad/FasL-GFPT~T. This is a replication-deficient adenoviral vector that expresses a fusion of murine Fast and green fluorescent protein (GFP). Fast-GFP retains full activity of wild-type Fast, at the same time allowing for easy visualization and quantification in both living and fixed cells. The fusion protein is under the control of tetracycline-regulated gene expression system. A tight control is achieved by creating this novel "double recombinant" Adenoviral vector, in which the tet-responsive element and the transactivator element are built into the opposite ends of the same vector to avoid enhancer interference.
Expression of the Fast-GFP fusion can be conveniently regulated by tetracycline or any derivative thereof, which includes, but is not limited to, doxycycline, in a dose-dependent manner. The vector efficiently delivers Fast-GFP gene to cells in vivo and in vitro, and the expression level of the fusion protein may be modulated by the concentration of doxycycline added to the culture media or administered to the subject. As may be seen in the following examples, Ad/FasL-GFPTe-r, is able to deliver Fast-GFP to transformed and primary cell lines, with the expression of the fusion protein in those cells regulated by varying the level of doxycycline in the media. Amounts of Fast-GFP
can be easily detected and quantified through the fluorescence of its GFP component, and correlated with the levels of apoptosis in the target and neighboring cells.
This vector design, which delivers an entire tet-regulated gene expression system, is more efficient and economical than strategies using multiple vectors, and can be applied to any situation where regulation of protein expression is desired.
Accordingly, the present invention provides an expression vector for tightly controlling expression of a target protein in response to tetracycline or a tetracycline derivative. The expression vector comprises: a tetracycline-responsive element; a polynucleotide sequence encoding a transactivator protein which is capable of binding to the tetracycline-responsive element; and a polynucleotide sequence encoding a target protein whose expression is regulated by the binding of the transactivator protein to the tetracycline-responsive element.
In a preferred embodiment, the vector is a viral vector. In a more preferred embodiment, the viral vector is an adenoviral vector. In this adenoviral vector, the tetracycline-responsive element and the polynucleotide sequence encoding the transactivator protein are positioned at opposite ends of the adenoviral vector. For example, the tetracycline-responsive element is positioned in the E4 region of the adenoviral vector and the polynucleotide sequence encoding the transactivator protein is positioned in the E1 of the adenoviral vector.
Optionally, the adenoviral vector does not include the E3 region of adenovirus. Also optionally, the adenoviral vector does not include the E4 region of adenovirus except for the Orf6 of the E4 region.
The expression of the target protein may be repressed in the presence of tetracycline or doxycycline. Alternatively, expression of the target protein may be activated in the presence of doxycycline.
It should be noted that the vector may also be any other type of viral vector, including but not limited to an adeno-associated viral vector, a vaccinia viral vector or a retroviral vector.
The expression vector may further comprise a polynucleotide sequence encoding a reporter protein. The reporter protein and the target protein may be encoded as a fusion protein or expressed as a single protein bicistronically with the target protein via a mechanism of internal ribosome entry site (IRES) or splicing donor/acceptor sites. .
The reporter gene preferably encodes a fluorescent protein such as green, yellow and blue fluorescent proteins, and more preferably green fluorescent protein (GFP).
For example, an adenoviral vector can be constructed for expression of a fusion protein, Fast-GFP, by ligating pLAd-C.tTA and pRAd-TGFsL to a portion of the Ad5 genome (snb 360) to produce the vector Ad/FasL-GFPTET as described below and as shown in Figures 1A-C.

Expression a target protein other than the Fast-GFP fusion can be regulated by using a similar adenoviral vector (designated pAdTeT) with the Fast-GFP fusion sequence replaced by the polynucleotide encoding the target sequence. The vector pAdTET can be constructed by removing the Fast-GFP fusion sequence from vector pRAd-TGFsL, inserting the target sequence into this site, and ligating the resulting vector to pLAd-C.tTA, in the same way as described for the production of the vector Ad/FasL-GFPTErin Figure 1A-C. The vector pAdTETcan be utilized to express an unlimited variety of heterologous proteins for which tight regulation is desired.
The expression vector may further comprise a selectable marker which can be used to screen for those cells which contain the vector and which express the selectable marker. In this manner, one can readily separate those cells containing the nucleic acid or the vector and expressing the selectable marker from those cells either containing the nucleic acid or the vector but not expressing the selectable marker, and from those cells not containing the nucleic acid or the vector. The specific selectable marker used can of course be any selectable marker which can be used to select against eukaryotic cells not containing and expressing the selectable marker. The selection can be based on the death of cells not containing and expressing the selectable marker, such as where the selectable marker is a gene encoding a drug resistance protein. An example of such a drug resistance gene for eukaryotic cells is a neomycin resistance gene. Cells expressing a neomycin resistance gene are able to survive in the presence of the antibiotic 6418, or Geneticin7, whereas those eukaryotic cells not containing or not expressing a neomycin resistance gene are selected against in the presence of 6418. One skilled in the art will appreciate that there are other examples of selectable markers, such as the hph gene which can be selected for with the antibiotic Hygromycin B, or the E, coli Ecogpt gene which can be selected for with the antibiotic Mycophenolic acid.

The specific selectable marker used is therefore variable.
The selectable marker can also be a marker that can be used to isolate those cells containing and expressing the selectable marker gene from those not containing and/or not expressing the selectable marker gene by a means other than the ability to grow in the presence of an antibiotic. For example, the selectable marker can encode a protein which, when expressed, allows those cells expressing the selectable marker encoding the marker to be identified. For example, the selectable marker can encode a luminescent protein, such as a luciferase protein or a green fluorescent protein, and the cells expressing the selectable marker encoding the luminescent protein can be identified from those cells not containing or not expressing the selectable marker encoding a luminescent protein. Alternatively, the selectable marker can be a sequence encoding a protein such as chloramphenicol acetyl transferase (CAT). By methods well known in the art, those cells producing CAT can readily be identified and distinguished from those cells not producing CAT.
6. Construction of the expression vectors of the present invention The expression vectors of the present invention can be constructed by using recombinant DNA technologies. For example, the regulatable adenoviral vector described above may be derived from adenvirus type 5 and modified to include heterologous sequences encoding the apoptosis-signaling ligand (e.g. Fas and TRAIL) and the transcription regulatory sequence.
One skilled in the art will appreciate that there are numerous techniques available by which one can obtain a nucleic acid sequence encoding an apoptosis-signaling ligand, and optionally, additional sequences such as one or more transcrition regulatory sequence. One method of obtaining the nucleic acid is by constructing the nucleic acid by synthesizing a recombinant DNA molecule. For example, oligonucleotide synthesis procedures are routine in the art and oligonucleotides coding for a particular protein or regulatory region are readily obtainable through automated DNA synthesis. A nucleic acid for one strand of a double-stranded molecule can be synthesized and hybridized to its complementary strand. One can design these oligonucleotides such that the resulting double-stranded molecule has either internal restriction sites or appropriate 5' or 3' overhangs at the termini for cloning into an appropriate vector. Double-stranded molecules coding for relatively large proteins or regulatory regions can be synthesized by first constructing several different double-stranded molecules that code for particular regions of the protein or regulatory region, followed by ligating these DNA molecules together. For example, Cunningham, et al. (1989) "Receptor and Antibody Epitopes in Human Growth Hormone Identified by Homolog-Scanning Mutagenesis"
Science, Vol. 243, pp. 1330-1336, have constructed a synthetic gene encoding the human growth hormone gene by first constructing overlapping and complementary synthetic oligonucleotides and ligating these fragments together. See also, Ferretti et al. (1986) Proc. Nat.
Acad. Sci. 82:599-603, wherein synthesis of a 1057 base pair synthetic bovine rhodopsin gene from synthetic oligonucleotides is disclosed.
Once the appropriate DNA molecule is synthesized, this DNA can be cloned downstream of an appropriate promoter. Techniques such as this are routine in the art and are well documented.
An example of another method of obtaining a nucleic acid encoding an apoptosis-signaling ligand is to isolate the corresponding wild-type nucleic acid from the organism in which it is found and clone it in an appropriate vector. For example, a DNA or cDNA library can be constructed and screened for the presence of the nucleic acid of interest. Methods of constructing and screening such libraries are well known in the art and kits for performing the construction and screening steps are commercially available (for example, Stratagene Cloning Systems, La Jolla, CA). Once isolated, the nucleic acid can be directly cloned into an appropriate vector, or if necessary, be modified to facilitate the subsequent cloning steps. Such modification steps are routine, an example of which is the addition of oligonucleotide linkers which contain restriction sites to the termini of the nucleic acid. General methods are set forth in Sambrook et al., "Molecular Cloning, a Laboratory Manual" Cold Spring Harbor Laboratory Press (1989). Once isolated, one can alter selected codons using standard laboratory techniques, PCR for example.
Yet another example of a method of obtaining a nucleic acid encoding an apoptosis-signaling ligand is to amplify the corresponding wild-type nucleic acid from the nucleic acids found within a host organism containing the wild-type nucleic acid and clone the amplified nucleic acid in an appropriate vector. One skilled in the art will appreciate that the amplification step may be combined with a mutation step, using primers not completely homologous to the target nucleic acid for example, to simultaneously amplify the nucleic acid and alter specific positions of the nucleic acid.
By using these recombinant DNA techniques, a replication-incompetent adenoviral vector encoding an apoptosis-signaling ligand can be constructed. For example, a complex adenoviral vector encoding TRAIL can be constructed and used to infect tumor cells. The vector that further comprises GFP which is expressed bicistronically with TRAIL is designated Ad.TRAIL/GFPTEr.
The vector, Ad.TRAIL/GFPTET is a complex adenoviral vector that expresses multiple genes and regulatory mechanisms. Construction of the adenoviral vectors is diagramed in Figure 3. The sequence encoding TRAIL and GFP separated by an IRES is cloned into the right-end (E4 region) of the type 5 adenovirus genome using a shuttle vector, resulting in a shuttle vector pRAdTRE-TRAIL/GFP. The pRAdTRE-TRAIL/GFP shuttle vector contains the right end of the adenoviral genome including the right long terminal repeats R-TR.
Another shuttle vector, pLAd-C.tTA, contains a tetracycline transactivator gene tTA in the E1 region of the type 5 adenovirus genome. The vector pLAd-C.tTA also contains the left end of the adenoviral genome including the left long terminal repeats L-TR and the adenoviral packaging signal yr. The vectors pRAdTRE-TRAIL/GFP and pLAd-C.tTA are both linearized and ligated to the backbone of the adenovirus to form the recombinant adenoviral vector, Ad.TRAIL/GFPTer.
7. Routes of.administration and formulations The expression vector encoding the apoptosis-signaling ligand may be introduced into a tumor by using any pharmaceutically acceptable routes of administration. For example, the expression vector may be administered into a group of tumor cells parenterally, intraperitoneally, intravenously, intraartierally, transdermally, sublingually, intramuscularly, rectally, transbuccally, intranasally, liposomally, via inhalation, vaginally, intraoccularly, via local delivery by catheter or stent, subcutaneously, intraadiposally, intraarticularly, intrathecally, or in a slow release dosage form.
One skilled in the art will recognize that this aspect of the methods can comprise either a stable or a transient introduction of the sequence encoding the apoptosis-signaling ligand (e.g. FasL and TRAIL) into the cell. Additionally, the stably or the transiently introduced ligand-encoding sequence may or may not become integrated into the genome of the host.
One skilled in the art will also recognize that the precise procedure for introducing the expression vector into the cell may, of course, vary and may depend on the specific type or identity of the cell.
Examples of methods for introducing an expression vector into a cell include, but are not limited to electroporation, cell fusion, DEAE-dextran mediated transfection, calcium phosphate-mediated transfection, infection with a viral vector, microinjection, lipofectin-mediated transfection, liposome delivery, and particle bombardment techniques, including various procedures for "naked DNA" delivery.
Optionally, the method can be performed ex vivo where the group of cells into which the expression vector is introduced are contained in a sample taken from a patient having cancer, or contained in contained in a cell culture.
For example, the expression vector may be introduced into a mixture of cells which express Fas and cells which do not express Fas.
Optionally, the expression vector may be introduced into cells which do not express Fas. Also optionally, the expression vector may be introduced into cells which do express Fas. Also optionally, the expression vector may be introduced into cells which cells which do not express Fas. By a "bystander effect", those cancer cells expressing Fas near those cells transduced by the expression vector are killed via Fas-Fast interactions.
The various vectors and hosts used to express the apoptosis-signaling ligand may be used to express the ligand in cell culture or in vitro. For example, an expression vector encoding a Fas ligand may be introduced into a tissue culture cell line, such as COS cells, and expressed in the cell culture. In this manner, one skilled in the art can select a cell type that may have a limited life in the host organism such that the host can effectively clear the cell expressing the the apoptosis-signaling ligand in a period of time such that any possible apoptotic effects on non-target surrounding cells or tissues can be minimized.
Alternatively, cells from a subject may be removed from the subject, administered the expression vector encoding the apoptosis-signaling ligand, and then replaced into the subject. In this ex vivo treatment procedure, the cells can be manipulated to facilitate the uptake of the nucleic acid encoding the apoptosis-signaling ligand without unnecessary adverse effects on the subject.
The various vectors and hosts used to express the apoptosis-signaling ligand may be used to express the nucleic acids in vivo. For example, an expression vector encoding Fast may be introduced into cells of a eukaryotic host, preferably tumor cells, to treat Fas+ tumor cells in situ.
As briefly discussed above, one skilled in the art will appreciate that specific tissues can be treated by selectively administering the vector to the host. For example, administering an adenovirus vector via an aerosol such as through the use of an inhaler can selectively administer the vector to the lungs. Optionally, the use of a suppository can be used to selectively administer the vector to cells of the colon.
Also optionally, delivering the vector topically such as in a cream can selectively deliver the vector or nucleic acid to skin cells or the cervix.
One skilled in the art will recognize the various methods that can routinely be used to selectively deliver the expression vector to specific organs or cells. For example, delivery of the expression vector can be manually facilitated through such methods as injection of the vector into the selected site. For example, direct injection can be used to deliver the vector to specific brain and/or breast location. In one embodiment of the present invention, direct injection of the vector encoding a Fas ligand or TRAIL is used for delivery into breast tumor masses.
It is contemplated that using the methods and vectors of the present invention, apoptosis-signaling ligand can be administered to a cell or to a subject, most preferably, humans, to treat disease states, preferably cancer. The present vector, whether alone, in combination with another compound or composition (e.g., a chemotherapy agent), or as part of a vector-based delivery system, may be administered parenterally (e.g., intravenously), by intramuscular injection, by intraperitoneal injection, topically, transdermally, or the like, although topical administration is typically preferred.
The exact amount of such nucleic acids, compositions, vectors, etc., required may vary from subject to subject, depending on the species, age, weight and general condition of the subject, the severity of the disease or condition that is being treated, the particular compound or composition used, its mode of administration, and the like. Thus, it is not possible or necessary to specify an exact amount. However, an appropriate amount may be determined by one of ordinary skill in the art using methods well known in the art (see, e.g., Martin et al., 1989).
For topical administration, the composition of the present invention may be in pharmaceutical compositions in the form of solid, semi-solid or liquid dosage forms, such as, for example powders, liquids, suspension, lotions, creams, gels or the like, preferably in unit dosage form suitable for single administration of a precise dosage. The compositions can typically include an effective amount of the selected nucleic acid, composition, or vector in combination with a pharmaceutically acceptable carrier and, in addition, may include other medicinal agents, pharmaceutical agents, carriers, adjuvants, diluents, etc.
By "pharmaceutically acceptable" is meant a material that is not biologically or otherwise undesirable, i.e., the material may be administered to an individual along with the selected nucleic acid, composition thereof, or vector without causing any undesirable biological effects or interacting in a deleterious manner with any of the other components of the pharmaceutical composition in which it is contained.
Alternatively or additionally, parenteral administration, if used, is generally characterized by injection e.g., by intravenous injection including regional perfusion through a blood vessel supplying the tissues(s) or organs) having the target cell(s). Injectables can be, prepared in conventional forms, either as liquid solutions or suspensions, solid forms suitable for solution or suspension in liquid prior to injection, or as emulsions. Parenteral administration can also employ the use of a slow release or sustained release system, such that a constant level of dosage is maintained (See, for example, U.S. Patent No. 3,710,795). The compound can be injected directly to the site of cells or tissues expressing a Fas+ phenotype, or they can be injected such that they diffuse or circulate to the site of the Fas+ phenotypic cells.
Dosages will depend upon the mode of administration, the disease or condition to be treated, and the individual subject's condition.
Dosages will also depend upon the material being administered, e.g., a nucleic acid, a vector comprising a nucleic acid, or another type of compound or composition. Such dosages are known in the art.
Furthermore, the dosage can be adjusted according to the typical dosage for the specific disease or condition to be treated.
Furthermore, culture cells of the target cell type can be used to optimize the dosage for the target cells in vivo, and transformation from varying dosages achieved in culture cells of the same type as the target cell type can be monitored. Often a single dose can be sufficient;
however, the dose can be repeated if desirable. The dosage should not be so large as to cause adverse side effects. Generally, the dosage will vary with the age, condition, sex and extent of the disease in the patient and can be determined by one of skill in the art. The dosage can also be adjusted by the individual physician in the event of any complication.
Examples of effective doses in non-human animals are provided in the Examples. Based on art accepted formulas, effective doses in humans can be routinely calculated from the doses provided and shown to be effective.

For administration to a cell in a subject, the compound or composition, once in the subject, will of course adjust to the subjects body temperature. For ex vivo administration, the compound or composition can be administered by any standard methods that would maintain viability of the cells, such as by adding it to culture medium (appropriate for the target cells) and adding this medium directly to the cells. As is known in the art, any medium used in this method can be aqueous and non-toxic so as not to render the cells non-viable. In addition, it can contain standard nutrients for maintaining viability of cells, if desired.
For in vivo administration, the complex can be added to, for example, a blood sample or a tissue sample from the patient, or to a pharmaceutically acceptable carrier, e.g., saline and buffered saline, and administered by any of several means known in the art.
Other examples of administration include inhalation of an aerosol, subcutaneous or intramuscular injection, direct transfection of a nucleic acid sequence encoding the compound where the compound is a nucleic acid or a protein into, e.g., bone marrow cells prepared for transplantation and subsequent transplantation into the subject, and direct transfection into an organ that is subsequently transplanted into the subject.
Further administration methods include oral administration, particularly when the composition is encapsulated, or rectal administration, particularly when the composition is in suppository form.
A pharmaceutically acceptable carrier includes any material that is not biologically or otherwise undesirable, i.e., the material may be administered to an individual along with the selected complex without causing any undesirable biological effects or interacting in a deleterious manner with any of the other components of the pharmaceutical composition in which it is contained.
Specifically, if a particular cell type in vivo is to be targeted, for example, by regional perfusion of an organ or tumor, cells from the target tissue can be biopsied and optimal dosages for import of the complex into that tissue can be determined in vitro, as described herein and as known in the art, to optimize the in vivo dosage, including concentration and time length.
Alternatively, culture cells of the same cell type can also be used to optimize the dosage for the target cells in vivo. For example, intratumoral injection amounts and rates can be controlled using a controllable pump, such as a computer controlled pump or a micro-,thermal pump, to control the rate and distribution of the nucleic acid or vector in the tumor or tissue. Example 4 demonstrates effective dosages of Ad/FasL-GFPrer used for in vivo treatment of both breast and brain tumors in mice. One of ordinary skill will readily know how to extrapolate these figures to determine effective human dosages.
For either ex vivo or in vivo use, the nucleic acid, vector, or composition can be administered at any effective concentration. An effective concentration is that amount that results in killing, reduction, inhibition, or prevention of a transformed phenotype of the cells.
The expression vector of the present invention may be administered in a composition. For example, the composition may further comprise other medicinal agents, pharmaceutical agents, carriers, adjuvants, diluents, etc. Furthermore, the composition can comprise, in addition to the nucleic acid or vector, lipids such as liposomes, such as cationic liposomes (e.g., DOTMA, DOPE, DC-cholesterol) or anionic liposomes. Liposomes may further comprise proteins to facilitate targeting a particular cell, if desired. Administration of a composition comprising a nucleic acid or a vector and a cationic liposome can be administered to the blood afferent to a target organ or inhaled into the respiratory tract to target cells of the respiratory tract.
Regarding liposomes, see, e.g., Brigham et al. Am. J. Resp. Cell. Mol.
Biol. 1:95-100 (199); Felgner et al. Proc. Natl. Acad. Sci USA

84:7413-7417 (1987); U.S. Pat. No.4,897,355. Furthermore, the nucleic acid or a vector can be administered as a component of a microcapsule that can be targeted to specific cell types, such as macrophages, or where the diffusion of the compound or delivery of the compound from the microcapsule is designed for a specific rate or dosage.
Any cell, specifically a tumor cell, which expresses an apoptosis-mediating receptor can be treated by the methods of the present invention. For example, Fas is primarily a surface protein and a cell expressing Fast can be used to treat the Fas-expressing cell by the Fas-Fast induction of apoptosis. The cell expressing the Fast can interact with the Fas-expressing cell via interactions of the Fas and the Fast on the surface of the cells, and therefore treat Fas-expressing cells that the Fast-expressing cells can make contact with. Additionally, the Fast-producing cells may also regulate the Fas-expressing cell by producing soluble Fast which then interacts with Fas and also induces apoptosis of the Fas-expressing cells.
The interaction of the Fas and the Fast is typically a ligand-receptor binding, although the interaction may not have to be binding per se, but includes any cellular reaction which results from any interaction of the Fas and the Fast. Therefore any cellular apoptosis via Fas that results from the expression of a Fast by that same cell or a second cell which expresses a Fas ligand is hereby contemplated.
Although any cell expressing Fas can be induced to undergo apoptosis using the methods of the present invention, a preferred embodiment is inducing Fas+ tumor cells to undergo apoptosis using these methods. In this embodiment, these tumor cells can selectively be induced to undergo apoptosis and then die, thereby treating a tumor.
In another preferred embodiment, the tumor is a solid tumor and the tumor is injected with a recombinant virus which can infect the cells of the tumor and thereby cause them to express Fast, and whereby the interaction of the Fast-expressing cells with the Fas-expressing cells causes the Fas+ cells to undergo apoptosis.
The Fas-expressing cells which are affected by the FasL-expressing cells are typically cells adjacent to the Fast-expressing cells since typically a cell-to-cell contact is necessary for the apoptotic signal be effectuated. The affected Fas cells can be removed from the immediate surroundings of the Fast-expressing cell, however, such as where the Fast-expressing cell has mobilized and/or where the FasL-expressing cell produces soluble Fast.
The Fast-expressing cells can also cause their own death if those cells also are Fas+ cells. In this approach, the methods of the present invention can cause Fas+ cells to die, but the tumor cells that now express the Fast also will die, thereby eliminating those tumor cells that might otherwise cause regression of the tumor.
8. Combination therapy The present invention also provides a method which utilizes a combination therapy that combines expression of the apoptosis-signaling ligand with administration of anti-cancer agents. It is believed that by co-administering anti-cancer drugs, apoptosis of cancer cells can be enhanced or sensitized, especially in those cancer cells are resistant to Fast- or TRAIL-mediated apoptosis.
A major hurdle in treating cancer is the development of resistant tumor cells to drugs and the development of anti-apoptotic machinery which can spell over Fast or TRAIL sensitivity to apoptosis. It is desirable to administer subtoxic concentration of chemotherapeutic drugs and TRAIL (or Fas) on TRAIL (or Fas)-resistant tumor cells, which should result in maximum tumor suppression and minimum side effects associated with administration of high dosage of chemotherapeutics.
The combination therapy of the present invention may overcome tumor resistance to Fas- or TRAIL-mediated apoptosis by multiple mechanisms of actions. Anticancer agents such as chemotherapeutic agents or cytokines may sensitize Fas- or TRAIL-mediated apoptosis by 1 ) suppression of anti-apoptotic molecules, and/or 2) upregulation of pro-apoptotic molecules. For example, Bcl-x~ and Bcl-2, major inhibitors of the mitochondria) apoptotic pathway, can be regulated by anti-cancer drugs. Paclitaxel, a plant-derived anti-cancer drug that at low levels can reduce the activity of Bcl-2 by inducing phosphorylation of Bcl-2.
In addition, drugs and cytokines can also upregulate the expression of pro-apoptotic molecules to lower the signaling threshold required for the induction of TRAIL-mediated apoptosis. For example, expression of DRS, one of the death-inducing TRAIL receptors, can be induced by genotoxic drugs and TNF-a. The induction of DR5 appears to be regulated by both p53-dependent and p53-independent mechanims. Sheikh et al. (1998) "p53-dependent and independent regulation of the death receptor KILLER/DR5 gene expression in response to genotoxic stress and tumor necrosis factor alpha" Cancer Res. 58:1593-1598. Moreover, the mRNA of caspases (caspases-1, -2, -6, -8, and -9) can be upregulated by y-interferon. The upregulation of these caspases should enhance the sensitivity to apoptosis induced by expression of apoptosis-signaling ligand according to the present invention.
A wide variety of anti-cancer agents may be co-administered with the expression vectors of the present invention. Examples of the anti-cancer agent include, but are not limited to, alkylating agents, antibiotic agents, antimetabolic agents, hormonal agents, plant-derived agents, and biologic agents.
Examples of alkylating agents include, but are not limited to, bischloroethylamines (nitrogen mustards, e.g. chlorambucil, cyclophosphamide, ifosfamide, mechlorethamine, melphalan, uracil mustard), aziridines (e.g. thiotepa), alkyl alkone sulfonates (e.g.
busulfan), nitrosoureas (e.g. carmustine, lomustine, streptozocin), nonclassic alkylating agents (altretamine, dacarbazine, and procarbazine), platinum compounds (carboplastin and cisplatin).
Examples of antibiotic agents include, but are not limited to, anthracyclines (e.g. doxorubicin, daunorubicin, epirubicin, idarubicin and anthracenedione), mitomycin C, bleomycin, dactinomycin, plicatomycin.
Examples of antimetabolic agents include, but are not limited to, fluorouracil (5-FU), floxuridine (5-FUdR), methotrexate, leucovorin, hydroxyurea, thioguanine (6-TG), mercaptopurine (6-MP), cytarabine, pentostatin, fludarabine phosphate, cladribine (2-CDA), asparaginase, and gemcitabine.
Examples of such hormonal agents are synthetic estrogens (e.g.
diethylstibestrol), antiestrogens (e.g. tamoxifen, toremifene, fluoxymesterol and raloxifene), antiandrogens (bicalutamide, nilutamide, flutamide), aromatase inhibitors (e.g., aminoglutethimide, anastrozole and tetrazole), ketoconazole, goserelin acetate, leuprolide, megestrol acetate and mifepristone.
Examples of plant-derived agents include, but are not limited to, vinca alkaloids (e.g., vincristine, vinblastine, vindesine, vinzolidine and vinorelbine), podophyllotoxins (e.g., etoposide (VP-16) and teniposide (VM-26)), camptothecin and its derivatives (e.g., 9-nitro-camptothecin and 9-amino-camptothecin), and taxanes (e.g., paclitaxel and docetaxel).
Examples of biologic agents include, but are not limited to, immuno-modulating proteins such as cytokines, monoclonal antibodies against tumor antigens, tumor suppressor genes, and cancer vaccines.
Examples of interleukins that may be used in conjunction with the composition of the present invention include, but are not limited to, interleukin 2 (IL-2), and interleukin 4 (IL-4), interleukin 12 (IL-12).

Examples of interferons that may be used in conjunction with CPT
include, but are not limited to, interferon a, interferon ~i (fibroblast interferon) and interferon y (fibroblast interferon). Examples of such cytokines include, but are not limited to erythropoietin (epoietin a), granulocyte-CSF (filgrastin), and granulocyte, macrophage-CSF
(sargramostim). Other immuno-modulating agents other than cytokines include, but are not limited to bacillus Calmette-Guerin, levamisole, and octreotide.
Example of monoclonal antibodies against tumor antigens that can be used in conjunction with CPT include, but are not limited to, HERCEPTINO (Trastruzumab) and RITUXANO (Rituximab).
Examples of the tumor suppressor genes include, but are not limited to, DPC-4, NF-1, NF-2, R8, p53, V1/T1, BRCA1 and BRCA2.
Example of cancer vaccines include, but are not limited to gangliosides (GM2), prostate specific antigen (PSA), a-fetoprotein (AFP), carcinoembryonic antigen (CEA) (produced by colon cancers and other adenocarcinomas, e.g. breast, lung, gastric, and pancreas cancers), melanoma associated antigens (MART-1, gp100, MACE 1,3 tyrosinase), papillomavirus E6 and E7 fragments, whole cells or portions/lysates of antologous tumor cells and allogeneic tumor cells.
An adjuvant may be used to augment the immune response to TAAs. Examples of adjuvants include, but are not limited to, bacillus Calmette-Guerin (BCG), endotoxin lipopolysaccharides, keyhole limpet hemocyanin (GKLH), interleukin-2 (IL-2), granulocyte-macrophage colony-stimulating factor (GM-CSF) and cytoxan, a chemotherapeutic agent which is believed to reduce tumor-induced suppression when given in low doses.
In an embodiment, actinomycin D, a drug that inhibits RNA
synthesis and decreases expression of Bcl-x~ may be used to sensitize TRAIL-mediated apoptosis in cancer cells, for example, in cancer cells of Kaposi's sarcoma (KS) that is associated with AIDS. KS is the most common malignancy arising in persons with HIV infection (AIDS-KS).
Although a number of modalities have been used for 15 years, cure or long-term complete remission from KS is unlikely with the currently available therapeutic modalities. Lee and Mitsuyasu (1996) "Chemotherapy of AIDS-related Kaposi's sarcoma" Hematol. Oncol.
Clin. North Am. 10:1051-1068. High levels of Bcl-xand Bcl-x~ are detected in AIDS-KS lesions, which may be attributed to resistance of KS cells to killing by chemotherapeutic drugs and NK cells. Thus, co-administering to KS patients the expression vector encoding TRAIL and one or more genotoxic drugs should over the resistance by synergistically sensitizing the cancer cells to TRAIL-mediated apoptosis through suppression of Bcl-x and Bcl-x~ levels by the genotoxic drugs.
In another emobodiment, doxorubicin may be used in combination of the expression vector expressing TRAIL to treat patients with prostate cancer. Prostate cancer is one of the most prevalent cancers in American men and the survival rate of patients with advanced prostate cancer is currently low. Landis et al. "Cancer statitics" CA Cancer J. Clin. 49: 8-31. While surgery, hormone therapy, and chemotherapy can eradicate the majority of prostate cancer, relapse of advanced cancer metastasis can occur. Since the prostate cells that are hormone refractory are also insensitive to radiation therapy and chemotherapy, these cells possibly develop resistance to all apoptotic programs induced by various stimuli as they progress to become more malignant. However, co-administrating a genotoxic drug with the expression vector encoding TRAIL should overcome the resistance by sensitizing prostate cancer cells to TRAIL-mediated apoptosis through suppression of apoptosis-inhibiting molecules or upregulation of pro-apoptotic molecules.
The expression vectors of the present invention, when expressing the apoptosis-signaling ligand for treating cancer (or other diseases), may be administered in conjunction with other therapeutic agents against the cancer (or the other diseases to be treated) before, during, or after the administration of the other therapeutic agent. These therapeutic agents can be administered at doses either known or determined to be effective and may be administered at reduced doses due to the presence of the apoptosis-signaling ligand expressed by the vector of the present invention.
The present invention is more particularly described in the following examples which is intended as illustrative only since numerous modifications and variations therein will be apparent to those skilled in the art.
EXAMPLE
Example 1 Expression vector for Fas Ligand In an example of the methods described above and depicted in Figure 1, a recombinant adenovirus containing a nucleic acid encoding a murine Fas ligand was constructed. Additionally, a recombinant adenovirus was constructed containing a nucleic acid encoding a murine Fas ligand and also encoding the jellyfish green fluorescent protein (GFP) such that a fusion protein was ultimately translated. This fusion protein was used to monitor the expression and localization of the protein in cultured cells and in animal tissues following transduction with the adenovirus vector.
Three different tumor cell lines were isolated from breast cancer patients, all of which exhibited a high degree of sensitivity to the Fas ligand treatment via the adenovirus vector. This demonstrates the tumor cells could be effectively treated, or killed, using these methods.
Parallel experiments also demonstrated several prostate cancer cell lines are extremely sensitive to Fas-mediated apoptosis since complete killing of these cells was obtained using Adenovirus-mediated introduction of a nucleic acid encoding Fas ligand into these cells.
Example 2: Controlled Delivery of a Fast-GFP Fusion Protein with a Complex Adenoviral Vector Fas ligand (Fast) induces apoptosis in cells that express Fas receptor and plays important roles in immune response, degenerative and lymphoproliferative diseases and tumorigenesis. It is also involved in generation of immune privilege sites and is therefore of interest to the field of gene therapy. We describe the construction and characterization of replication-deficient adenoviral vectors that express a fusion of murine Fast and green fluorescent protein (GFP). Fast-GFP retains full activity of wild-type Fast, at the same time allowing for easy visualization and quantification in both living and fixed cells. The fusion protein is under the control of tetracycline-regulated gene expression system. A tight control is achieved by creating a novel A double recombinant Ad vector, in which the tet-responsive element and the transactivator element are built into the opposite ends of the same vector to avoid enhancer interference. Expression can be conveniently regulated by tetracycline or its derivatives in a dose-dependent manner.
The vector was able to efficiently deliver Fast-GFP gene to cells in vitro, and the expression level of the fusion protein was modulated by the concentration of doxycycline in culture media. This regulation allows us to produce high titers of the vector by inhibiting Fast expression in a CrmA-expressing cell line. induction of apoptosis was demonstrated in all cell lines tested. These results indicate that our vector is a potentially valuable tool for Fast-based gene therapy of cancer and for the study of FasL/Fas-mediated apoptosis and immune privilege.

Materials and Methods Cells: HeLa and 293 cells were obtained from the American Type Culture Collection (ATCC CCL-2.1 and ATCC CRL-1573, respectively) and maintained as monolayers at 37 C under 5% C02 in Dulbecco's modified Eagle's medium (DMEM; Gibco BRL) supplemented with 10% bovine calf serum (BCS; HyClone) and 1 penicillin/streptomycin (Cellgro). Cultured rat myoblasts were maintained in H-21 (Cellgro) media supplemented with 20% Fetal Bovine Serum (FBS; HyClone) and 1 % each of penicillin/streptomycin and fungizone.
For DNA transfections, 5x105 cells per well were seeded on 6-well plates (Greiner) and transfected 24 hours later using LipofectAMINE (Gibco BRL) according to manufacturer=s instructions.
To produce a cytokine response modifier A (CrmA)-expressing 293 cell line, pCrmA-I-Neo was transfected into HEK293 cells. Neo-positive clones were selected by adding 6418 to the media at 0.4 g/L for 4 weeks, at the end of which time individual clones were picked up, propagated and assayed for CrmA expression by their resistance to Fast-induced apoptosis.
Construction of plasmids and recombinant adenoviral vectors:
Vectors pEGFP-1 and pEGFP1-C1 were obtained from Clontech. They contain a red-shifted variant of wild type green fluorescent protein (wt GFP) gene, with brighter fluorescence and "humanized" codon usage.
(Zhang, G., V. Gurtu and S. R. Kain. 1996. "An enhanced green fluorescent protein allows sensitive detection of gene transfer in mammalian. cells." (Biochem Biophys Res Commun 227:707-11.) This protein will be referred to as "GFP" in this Example. The mouse Fast cDNA sequence, available in Genbank, was in a Bluescript (Invitrogen) vector. Vectors pUHDlO-3 and pUHD15-1 (Gossen, M. and H. Bujard, "Tight control of gene expression in mammalian cells by tetracycline-responsive promoters" Proc Natl Acad Sci U S A 89:5547-51, 1992) are available from Clontech. GFP-Fast fusion gene was constructed by inserting DNA coding for as 11 to as 279 of the murine Fas ligand in-frame downstream of the GFP sequence in pEGFP-C1, to generate pC.GFsI. The fusion gene from pC.GFsI was inserted into pUHD10-3 to produce p10-3.GFsl. Cowpox virus (Chordopoxvirinae) cytokine response modifier A (crmA; CPV-W2) cDNA in pcDNA3 vector is available from Genentech. The CrmA gene was excised from pcDNA3 and inserted into pIRES-Neo vector (Clontech) to generate pCrmA-I-Neo.
GFP, Fast, Fast-GFP and LacZ genes were cloned into the E1 shuttle vector, pLAd-CMV to generate pLAd-C.Gf, pLAd-C.FsI, pLAd-C.GFsI and pLAd-C.Lz constructs, respectively (Fig. 1A). The Tet-OFF
fusion activator protein expression cassette was extracted from pUHD15-1 and inserted into pLAd-CMVie to generate pLAd-C.tTA. The GFP-Fast fusion gene expression cassette was excised from p10-3.GFsl and inserted into pRAd.mcs, a shuttle vector for transgene insertion between E4 and right ITR of AdS. The resulting construct was called pRAd-T.GFsI (Fig. 1 B).
The assembly of Ad/FasL-GFPTEr vector is shown in Figure 1 C.
Other rAd genomes used in this study were constructed using a similar strategy. All vectors were based on Ad5sub360 (0E3) with additional deletion of all E4 ORFs with the exception of ORF6. (Huang, M. M. and P. Hearing. 1989) The adenovirus early region 4 open reading frame 6/7 protein regulates the DNA binding activity of the cellular transcription factor, E2F, through a direct complex. (Genes Dev 3:1699-710).

Propagation of viral vectors: The 293 cells, which provide Ad5 E1a and E1b functions in trans (Graham, F. L., J. Smiley, W. C. Russell and R. Nairn; "Characteristics of a human cell line transformed by DNA
from human adenovirus type 5" (J Gen Virol 36:59-74,1977), were transfected with the ligation mixture containing the rAd vector DNA using LipofectAMINE method. Transfected cells were maintained until adenovirus-related cytopathic effects (CPE) were observed (typically between seven and 14 days), at which point the cells were collected.
Vector propagation and amplification was then achieved by standard techniques. The stocks were titrated on 293 or 293CrmA cells and plaques were scored to determine vector yields as PFU/ml. Vectors were also titrated using GFP fluorescence or X-gal staining, as appropriate. In both cases, titer estimates were in good agreement with PFU/ml.
IlIlestern blot analysis: 10 cm plates (Greiner) were seeded with 106 cells of primary rat myoblasts. After 24 hours, plates were infected with Ad/FasL-GFPTET or control vector at multiplicity of infection (M01) of 2. At 24 hours postinfection, the plates were washed twice with PBS.
The cells were collected and lysed in 200 p,1 of cell lysis buffer containing 50 mM Tris-HCI (pH 7.8), 1 mM EDTA, 2% SDS, 0.1 % Bromophenol Blue, 1 mM PMSF (Sigma), 50 p,g/ml leupeptin (Sigma), 2 p,g/ml aprotinin (Sigma) and 1 ng/ml pepstatin (Sigma). The samples were boiled for 5 minutes and 1/10 of the original amount (106 cells) was loaded per lane of an 8% SDS-PAGE minigel (BioRad), which was run at 20 mA for 3 hours. Human recombinant Fast (C-terminal) was obtained from Santa Cruz Laboratories. The proteins were transferred to a nitrocellulose membrane (Pharmacia Biotech) using a semi-dry gel transfer apparatus (BioRad). The membrane was blocked by incubation (2 hours at 37°C) in a solution containing 10 mM Tris-HCI (pH 7.5), 140 mM NaCI, 3% (w/v) BSA, 5% (w/v) powdered milk, 0.2% (v/v) Tween-20 (Amresco, Solon, OH) and 0.02% (w/v) sodium azide (Sigma). The polyclonal rabbit anti-Fast antibody (Santa Cruz) was diluted 1:100 with blocking solution and incubated with the membrane for 2 hours at ambient temperature. The blot was washed with 10 mM Tris-HCI (pH
7.5) and 140 mM NaCI solution twice, then incubated with goat anti-rabbit IgG conjugated with HRPO (Caltag, Burlingame, CA) diluted 1:10000. The blot was developed in ECL reagent (Amersham Life Science) overnight and visualized with Kodak X-ray film.
Detection of apoptosis: Early detection of apoptosis in cultured adherent cells was accomplished by utilizing the In Situ Cell Death Detection Kit, AP (Boehringer Mannheim) according to manufacturers instructions. This kit utilizes the terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) process to incorporate fluorescein at free 3'-OH DNA ends and detect it with anti-fluorescein antibody conjugated to alkaline phosphatase. After substrate reaction, stained cells can be visualized using light microscopy.
Results:
Functional analysis of Fast and Fast-GFP proteins: In order to demonstrate that the Fas ligand-GFP (Fast-GFP) fusion protein retains full Fast activity, we have analyzed and compared the function of the Fast and Fast-GFP proteins by using transient DNA transfections into cells susceptible to Fas-mediated apoptosis. Triplicates of wells of HeLa cells were transfected with vectors expressing Fast, GFP-Fast or -galactosidase as a control. At 24 hours post-transfection, cells were fixed and analyzed for apoptosis by using the TUNEL kit. Typically, transfection efficiencies between 10 and 25% were achieved as determined by X-Gal staining of cells transfected with pcDNA3-LacZ.

Large numbers of HeLa cells transfected with vectors expressing either Fast or Fast-GFP showed typical apoptotic morphology (such as membrane blebbing and loss of adherence) and stained positive in the TuNE~ assay. Very few cells transfected with a control plasmid underwent apoptosis. The numbers of apoptotic cells in wells transfected with Fast-GFP vector were reproducibly similar to those transfected with Fast vector, suggesting that the wild-type and fusion proteins have comparable activity.
Construction and characterization of adenoviral vectors: Our goal was to produce large amounts of adenoviral vectors in which the Fast expression could be regulated. This regulation allows control of the levels of Fast expression in target cells and thus facilitates the study of its biological effects. In addition, amplification of rAd vectors constitutively expressing Fast or Fast-GFP in 293 cells would be expected to produce low titers because Fast expression causes apoptosis of the virus-producing cells. Muruve, D. A., A. G. Nicolson, R.
C. Manfro, T. B. Strom, V. P. Sukhatme and T. A. Libermann. (1997) "Adenovirus-mediated expression of Fas ligand induces hepatic apoptosis after Systemic administration and apoptosis of ex vivo-infected pancreatic islet allografts and isografts" Hum Gene Ther 8:955-63. To achieve the controlled Fast-GFP expression, we designed the Ad/FasL-GFPTETvector in which the Fast-GFP is expressed from a THE
promoter. Gossen, M. and H. Bujard. (1992) "Tight control of gene expression in mammalian cells by tetracycline- responsive promoters"
Proc Natl Acad Sci U S A 89:5547-51. We inserted CMVie promoter-driven tTA gene (the "tet-ofP' element) into the Ad5 E1 region and the TRE-controlled Fast-GFP fusion gene near the right ITR.
This strategy was based on the following considerations. First, this strategy delivers the entire tet-regulated expression system using a single vector, rather than using two Ad vectors as have been described previously. Harding, T. C., B. J. Geddes, D. Murphy, D. Knight and J. B.
Uney. (1998) "Switching transgene expression in the brain using an adenoviral tetracycline-regulatable system" see comments, Nat Biotechnol 16:553-5. Use of a single vector allows a more efficient delivery to target cells as well as a more uniform regulation of protein expression. This strategy also achieves maximum possible separation between the enhancer elements of the CMVie promoter and the THE
promoter, in order to minimize background (unregulated) expression of Fast-GFP protein (Fig. 1 B and 1 C). By placing the THE promoter at the right end of the Ad5 genome, a similar result was obtained with respect to the E1A enhancer elements, which are located within the Ad5 packaging signals Hearing, P. and T. Shenk. 1983. The adenovirus type 5 E1A transcriptional control region contains a duplicated enhancer element. Cell 33:695-703. These elements have been reported to interact with some promoters cloned into the E1 region Shi, Q., Y. Wang and R. Worton. (1997) "Modulation of the specificity and activity of a cellular promoter in an adenoviral vector" Hum Gene Ther 8:403-10.
The genomes of recombinant adenoviral vectors used in the present invention were assembled in vitro in large-scale ligation reactions as schematically diagrammed in Figure 1C. These genomes were then gel-purified and transfected into 293 cells and the resulting cultures were propagated until virus-induced CPE was observed. In the case of vectors expressing ~3-galactosidase or GFP, CPE occurred at significantly earlier time points than for vectors expressing Fast or FasL-GFP, indicating that adenoviral vector replication was likely deleteriously affected by Fast activity. Primary vector stocks were amplified according to established techniques, and recombinant adenoviral DNA
was extracted and examined for structural integrity by restriction enzyme digests.

The titers of Ad/FasL and Ad/FasL-GFPTET in 293 cells were typically 30 to 100-fold lower then titers of Ad/LacZ or Ad/GFP.
Comparison of titers of Ad vectors with Fast activity demonstrated a substantial improvement (between 8- and 12-fold) in the yield of these vectors when they were produced in 293CrmA cells (Figure 2). As shown in Figure 2, amplification of the control vector Ad/LacZ in either 293 or 293CrmA cells resulted in essentially the same yield.
Subsequently, generation and amplification of all vectors with Fast activity was carried out in 293CrmA cells.
Induction of apoptosis by adenovirus-mediated Fast expression:
To functionally demonstrate that adenovirus-mediated Fast expression, we transduced HeLa cells with Ad/FasL-GFPTeTat different MOI. At 24 hours post-transduction, cells were analyzed for apoptosis. Cells infected with Ad/FasL-GFPTET demonstrated typical apoptotic morphology. The numbers of apoptotic cells increased with the increasing vector titers. In contrast, plates transduced with the control vector Ad/LacZ at the same MOI did not generate apoptotic cells in excess of untransduced controls. The overall efficiency of transduction was determined by X-gal staining and shows increasing numbers of ~-galactosidase-positive cells with increasing MOI. We have observed that the numbers of apoptotic cells are noticeably higher than those of the cells with detectable GFP fluorescence, or of the X-gal stained cells transduced at the same. Thus, apoptosis of cells not infected with the vector, but adjacent to the cells that are, is caused by the interactions of Fast on the surface of infected cells with Fas receptors on their neighbors.
Detection and cellular localization of Fast-GFP fusion protein:
Wild-type Fast is a type II membrane protein. To demonstrate that the Fast-GFP fusion protein is also targeted to cellular membrane, we took advantage of the fluorescence of its GFP component, which can be detected in living cells using a fluorescent microscope with a FITC filter set. We have used this technique to observe the expression and cellular localization of our Fast-GFP fusion protein when expressed from rAd vector. In HeLa cells, expression of Fast-GFP causes apoptosis at protein levels close to the detection threshold of GFP.
Therefore, the expression of Fast-GFP was analyzed in primary rat myoblasts, which we found to be relatively resistant to Fast-induced IO apoptosis. High levels of Fast-GFP expression can be detected in myoblasts at 24 hours post-infection with Ad/FasL-GFPTET at MOI of 10.
Membrane-associated expression of Fast-GFP is evident in the majority of the transduced cells. In contrast, the fluorescence pattern of GFP itself is evenly distributed in the cytoplasm of the cells, while often I5 being excluded from the nucleus. These localization differences are also apparent in transduced 293CrmA cells at higher magnification.
These results indicate that the Fast-GFP fusion protein is directed to the cell surface, where it can interact with the Fas receptor in a manner analogous to that of wildtype Fast.
Regulation of Fast-GFP expression from rAd vector. To show that the present vector has the ability to regulate the amount of Fast activity produced by our rAd vector in target cells, we have performed experiments to establish the levels of Fast expression under induced or uninduced conditions at both the levels of protein synthesis and function. In Ad/FasL-GFPTET vector, expression of Fast-GFP fusion protein is designed to be activated by the binding of the tetR-VP16 fusion protein (constituatively expressed from the same vector; see Fig.
1 C) to the heptamer of tet-operators upstream of a minimal CMVie promoter. Gossen, M. and H. Bujard. (1992) "Tight control of gene expression in mammalian cells by tetracycline- responsive promoters"

Proc Natl Acad Sci U S A 89:5547-51. Presence of doxycycline in the cell should inhibit this binding and therefore the expression of Fast-GFP
in a concentration-dependent manner.
First, we determined the amounts of Fast-GFP produced in transduced cells by using Western blot analysis. We infected primary rat myoblasts with Ad/FasL-GFPTE-rat an MOI of 2 and cultured these cells in the absence or presence of doxycycline, a tetracycline derivative. Low MOI was chosen to maximize number of cells transduced with a single copy of the vector. After 48 hours, cells were lysed and the lysates analyzed by Western blotting using a polyclonal antibody against the extracellular domain of Fast. A single specific band larger than the predicted size of wt Fast was detected. The intensity of the band decreased with the increasing concentration of doxycycline, and no band could be detected in the cell lysates that have been cultured in the presence of 0.5 mg/L or higher concentration of doxycycline. No Fast-specific band was observed in cells transduced with a control vector. No bands of lesser size, corresponding to the breakdown or cleavage products, were detected either in the cell lysates or in the media supernatant. These results indicate that the amount of GFP-Fast protein produced in the cell from the Ad/FasL-GFPTeTvector can be regulated by the concentration of doxycycline in culture medium, and that this protein is stable and does not undergo appreciable cleavage once on the cell surface.
We have also analyzed the regulation of Fast activity, i.e. the induction of apoptosis in Fas-positive target cells. Wells of HeLa cells were transduced with Ad/FasL-GFPTeT at an MOI of 2 and cultured in the presence of various concentrations of doxycycline. At 24 hours post-transduction, cells were analyzed for apoptotic phenotype. The results confirm that the induction of apoptosis in cells transduced with Ad/FasL-GFPrEr can be regulated by doxycycline.
In the regulated protein expression system that we chose, presence of doxycycline inhibits the binding of tTA to THE and turns off Fast-GFP transcription in a dose-dependent manner. We elected to insert the constitutively expressed activator into the E1 region and the Fast-GFP expression cassette into a novel cloning site between the E4 promoter and the right ITR of Ad5, reasoning that this arrangement would minimize the effect of the E1A enhancer present within the packaging region of adenovirus and the CMVie enhancer within the tTA
promoter on the TRE, and thus reduce background expression of the fusion protein in the presence of inhibitor. This system performed successfully in the context of adenoviral vector, such that the expression of Fast-GFP could be efficiently regulated by varying the doxycycline concentrations in cell culture medium.
In the course of our experiments, we have observed that 293 cells are susceptible to Fast-induced apoptosis. This effect acts to significantly limit the titers of rAd vectors expressing Fast. This is true even if regulated or tissue-specific promoters are used to express Fast protein, since high levels of protein expression are unavoidable in the course of vector replication in 293 cells. In order to overcome this problem, we have generated a 293 cell line which constitutively expresses CrmA. This protein acts specifically to inhibit the activity of regulatory caspases, which are integral to the Fas apoptosis pathway.
By producing our Fast-containing vectors in these cells, we have obtained significant improvements in the vector titers.
In summary, we have developed and tested a rAd vector that expresses a novel Fast-GFP fusion protein under the control of tetracycline-regulated gene expression system. This vector combines high titers and efficient transgene delivery to multiple types of dividing and non-dividing cells with convenient regulation of protein expression and easy detection of the fusion protein in both living and fixed cells.
This vector is a valuable tool for treating disease through immunology, transplantation and cancer therapy.
Example 3: Bystander Gene Therapy Using Adenoviral Delivery of a Fas Ligand Fusion Gene This example describes a type of bystander gene therapy utilizing a Fas Ligand-fusion gene approach that induces prostatic adenocarcinoma to undergo apoptosis (programmed cell death) through a paracrine/autocrine mechanism. This work provides a novel and potent therapy for treatment of prostate cancer (PCa). Furthermore, specificity for the prostate or any other tissue may be achieved using tissue-specific promoters to allow parenteral delivery of virus for treatment of metastatic disease.
Our therapeutic approach is to deliver and express a Fas Ligand (CD95L-fusion gene) with a second generation adenovirus deleted for E1A, E3 and E4. CD95L expression is controlled by a Tet operator allowing for doxycycline regulation in vitro and in vivo. The CD95L
used in this proposal is the mouse CD95L cDNA truncated by 10 amino acids at its N terminus and fused in frame with a four-amino acid linker to the C terminus of an enhanced GFP.
Table 1 presents our data using five PCa cell lines and generally confirms literature reports (Hedlund et al. The Prostate 36:92-101, 1998; and Rokhlin et al. Can. Res. 57:1758-1768, 1997) that demonstrate PCa cell lines are resistant to CH-11 agonist activity. In contrast, we now demonstrate sensitivity to AdGFP-Fast and C2-ceramide in all five PCa cell lines tested to date.
Percent cytotoxicity was determined using the MTS assay. In brief, cells were seeded in a 12-well plate with 1 ml of media. Prior to treatments, cells were grown to 75% confluency and treated with either 500ng/ml CH-11 anti-Fas antibody, 500ng/ml Normal Mouse Serum or 30~,M C2-ceramide. For adenoviral transduction, approximately 1x105 cells/well were treated with either Ad/CMVGFP or Ad/GFP-FasLTer at an MOI between 10-1000. For each cell line, positive controls were left untreated, and 1 ml of media was used as a negative control. The cells were incubated for 48 hours at 37 °C for maximal cell killing. Media was aspirated and replaced with 0.5m1 fresh media + 100 ~,I of Cell Titer 967 Aqueous One Solution Reagent per well. Cells were incubated for an additional 1-3 hours at 37°C. After incubation, 120 ~,I of media was placed into a 96-well plate and absorbance readings were taken using a Vmax kinetic microplate reader at 490nm. Percent cytotoxicity was calculated as follows: % cytotoxicity = [1-(OD of experimental well/ OD
of positive control well)] x 100. For ceramide assays, 1x104 cells/well were seeded in a 96-well plate. The following morning cells were washed and incubated with 100p1 of 30~,M Dihydro- or C2-ceramide (diluted from a 10mM stock in ethanol) in serum-free RPMI 1640. After 24 hours, 20 ~,I Celltiter 967 Aqueous One Solution Reagent was added to each well and plates were incubated an additional 1-4 hours.
Absorbance and % cytotoxicity were determined as above. In each experiment, data points were run in triplicate.
Results:
Clearly, the five PCa cell lines analyzed in Table 1 are largely insensitive to CH-11. Sensitivity to C2-ceramide is relatively uniform at the 30 ~,M dose suggesting that the apoptosis pathway is intact. Most importantly, all the cell lines are responsive to AdGFP-Fast administration with DU145 being the least sensitive.
Several important points are made by these experiments. First, S we show using FACS analysis that CD95 (Fas receptor) was expressed on all candidate PCa cell lines, for all lines we used. Second, we show that the fas receptor blocking antibody (ZB4) does not prevent induction of apoptosis by AdGFP-Fast. We have performed this experiment several times with different doses of ZB4, always with the result that the virus induced the same extent of apoptosis in the presence or absence of the antibody. This suggests that newly synthesized CD95-CD95L may interact perhaps in the golgi (Bennett et al. Science 282:290-293, 1998), on the way to the plasma membrane, or on arrival at the cell surface as a preformed and functional apoptotic signaling complex.
Third, our results show that there is no intrinsic property of the adenovirus that facilitated induction of apoptosis in PCa. This was demonstrated by infecting PCa with control virus (AdCMVGFP) plus CH-11 at 500ng/ml. The result was that CH-11 still failed to induce apoptosis. These results show that apoptosis only occurs in CD95+-CH-11 resistant PCa cell lines when viral directed intracellular expression of CD95L occurs and this was not virus-dependent.
The final and most relevant piece of information pertains to whether we can administer AdGFP-FasLTEr without lethality to the subject. This is critically important because a dose as low as 2x10$ pfu of virus kills the mouse when administered parenterally. To address this issue, xenografts of PPC1 were developed in Balbc nu/nu mice and treated with various doses of AdCMVGFP control or AdGFP-Fast virus.
From these single dose studies, we have evidence that tumor cell growth is retarded or stopped. Further, out of 14 animals treated with virus, none have died from the virus. In summary, we conclude that the GFP-Fast fusion protein in our Ad5 delivery system has strong therapeutic potential for treating PCa.
Development of a version of AdGFP-Fast that is up-regulated by doxycycline.
Our present virus is designed to be administered orthotopically to PCa. If the virus escapes the tumor and enters the body it could be lethal if sufficient virus reaches~the reticuloendothelial system (mostly the liver). By administration of doxycycline (dox), expression of CD95L
from AdGFP-Fast can be down-regulated, and this danger avoided. A
viral vector induced by doxycycline that exhibits "very low" basal activity is constructed by using the Tet regulatable elements set forth in Example 1. This vector is completely repressed relative to GFP-Fast expression in the absence of dox and induced starting at 10ng/ml with maximal induction between 100-500ng/ml. These are easily achievable doses in humans (1-3 p,g/ml at typical dosage levels). Should adverse effects be observed, dox administration is terminated. However, doxycycline has a serum half-life of 16 hours which we believe argues that the addition of dox to down-regulate expression of Fas Ligand may be better for treating adverse effects in patients since we can rapidly achieve effective doxycycline doses within minutes by parenteral administration. If necessary, addition of a PEST signal can speed degradation (see Clontech catalogue).
Methods:
We replace our current Tet repressor and operator system with the rTSk'd B/C and rtTA system (Freundlieb et al. J. Gene Med. 1:4-12, 1999). It has already been pointed out that we can place our prostate specific promoters (PSA, PSADBam, PB and ARRPB2, Appendix) into the virus (replacing CMVie) to achieve tissue specificity where only prostate epithelial cells will be able to regulate rtTA. All viruses are grown by standard techniques from 3X plaque-purified samples assessed to be negative for wild-type adenovirus by PCR. All viruses are grown in the presence of l~,g/ml doxycycline in the HEK 293 packaging cell line that constitutively expresses the cowpox virus cytokine response modifier, crmA Rubinchik et al. This is necessary to prevent GFP-Fast induced apoptosis in the packaging cell line. Virus is always purified by isopycnic centrifugation on CsCI, desalted by chromatography, concentrated by filtration and stored frozen in PBS

10% glycerol in small aliquots at -80°C. Virus is thawed only once and administered to the animals under anesthesia, by infusion as described above, at 15 ~.I/min or via the tail vein with a tuberculin syringe. Tumor and animal tissues are collected for frozen sections or, fixed and embedded, where appropriate, and analyzed by H & E, by tunel assays for apoptosis, and by immunostaining to determine neutrophil infiltration and GFP expression where relevant.
Testing the original AdGFP-FasLTeta (dox dov~n-regulated) on prostate cancerxenografts in Balbc nulnu mice. These experiments are carried out to establish both toxicological and efficacy parameters.
Specifically, we infuse increasing doses 1x109 - 5x10'° pfu AdGFP-FasLTeta into 75 to 100 mm3 tumors to determine: A) lowest successful dose required to decrease tumor volume by 75% or more following orthotopic administration of virus with one dose and with three doses administered every four days. Tumors are developed from CD95L
sensitive PPC1, intermediately sensitive LnCAP C2-4, and more resistant Du145 cell lines. Other parameters of administration are developed based on results with the endpoint always being tumor remission. B) Highest tolerated viral dose following orthotopic administration (up to 5 x 10'° pfu). C) Determine if tumor will reoccur at a later time (6-12 months) in the same or distant site (C4-2). D) Highest dose administered i.v. (tail vein) that 50% of mice survive. E) Using data from D, test the effect of doxycycline administration on the animal survival curve and duration of doxycycline protection (Balbc nu/nu mice have no CTL response so adenovirus may survive for a long time).
Statistical analysis using a one sided t-test is employed. F) Determine the half-life of GFP-Fast in K562 cells (CD95L resistant, see Tabfe 1) by monitoring GFP (as the GFP-Fast fusion) over time in the presence of 1 ~,g/ml dox using FACS analysis.
The same set of experiments as in B1 above is carried out with the Tet inducible virus constructed as described above.
Toxicology testing Of AdGFP-FaSL,.etu (upregulated) and AdGFP-FasL,-era (down-regulated) administered to normal laboratory beagles.
Although there are a number of animal models for PCa, none but the dog model well-represent human disease in pathology and anatomy. It has recently been shown that human AdRSVbgal (serotype 5) adenovirus will infect dog epithelial cells, including prostate tumor cells, both in vitro and in vivo Andrawiss et al. Prostatic Can. Prostatic Dis.
2:25-35, 1999. Comparison of the present Ad/GFP-FasTET in dogs (immunocompetent) verses immunocompromised mice (Balbc nu/nu) provides additional support for a human phase I trial of this gene therapy approach.
In the following section, experiments are carried out on sexually mature normal dogs to see if orthotopic delivery Ad/GFP-FasLTer to hormonal prostate is safe with minimal or no collateral damage.
Purified concentrated adenovirus (Ad/GFP-FasLTer both up- and down-regulated and a reporter virus Ad/CMV-LacZ all serotype 5) is injected via an abdominal surgical approach into one lobe of the dog prostate. This approach is preferable to transrectal introduction because it is believed that direct visualization of the prostate provide for a more accurate introduction of virus in these first series of experiments.
Second, because of the highly vascular nature of the dog prostate direct visualization allows us to seal the needle track with topical tissue glue and digital pressure to prevent viral leakage from the injection site.
Based on these results, a 3D ultrasound guided transrectal introduction is used to mimic one of the proposed human approaches.
Virus dosages of 5x109, 1x10'°, and 5x10'° in a constant 400u1 volume are used: one set of 2 dogs receives AdICMV-LacZ at 5x10'° pfu to allow histochemical monitoring of viral spread. Dogs are monitored closely the first 72 hours for any signs of distress. Feces is collected and analyzed for viral shedding by PCR. Urine is also collected by foley catheter and assayed on 293 cells for shed virus and by PCR. At day 7 (2 dogs per viral dose) are euthanized with sodium pentobarbital and processed as described. (Andrawiss et al. Prostatic Can. Prostatic Dis.
2:25-35, 1999). Samples of all tissues are frozen in OCT while the remainder are either fixed and processed for histology (tunel, immunohistochemistry), or stored frozen at -80°C for DNA extraction and PCR using viral-specific primers. Expression of LacZ is examined in the Ad/CMV-LacZ group to monitor systemic viral spread.
Example 4: Intratumoral Introduction of Ad/GFP-FasLTEr Suppresses Breast Tumor and Brain Tumor Growth in Mice In this experiment, we implanted 106 MCF-7 cells bilaterally into Balbc nu/nu mice (Figure 6). When tumor sizes reached 5 mm in diameter, we infused at 15 ~,I per minute, 2 X 109 pfu Ad/GFP-FasLTeT
into the tumors on the right side of the mouse or 2 X 109 pfu Ad/LacZ

into the left side, over a period of 10 minutes using a Harvard infusion pump. At three weeks post-injection, all tumors injected with Ad/FasL-GFPTEr exhibited about 80-100% regression of the tumor in comparison with the control-treated tumor. In particular, in four of the six mice treated, most of the tumor masses disappeared after one injection (indicated by yellow arrows). In the other two of the six mice, suppression of tumor growth wad greater than 80% (indicated by black arrows) in comparison to tumors on the control side of the same mice.
In contrast, all tumors injected with Ad/LacZ grew to about 2 cm in diameter at three weeks after implantation. Histological analysis of the residual tumors in some of the mice showed only infiltrating immune cells and fibroblasts with no apparent cancer cells remaining. This demonstrates that Fast-induced apoptosis may be used as a novel treatment for breast cancer.
Similarly, we implanted 106 SF767 cells bilaterally into Balbc nu/nu mice. When tumor sizes reached 5 mm in diameter, we infused at 15 ~,I per minute, 2 X 109 pfu Ad/GFP-FasLTer into the tumors on the right side of the mouse or 2 X 109 pfu Ad/LacZ into the left side, over a period of 10 minutes using a Harvard infusion pump. Tumor suppression was about 80-100% in treated tumors as compared to untreated tumors. In contrast, all tumors injected with Ad/LacZ grew to about 2 cm in diameter at three weeks after implantation. This demonstrates that Fast-induced apoptosis may be used as a novel treatment for brain cancer.
Example 5: Comparison of sensitivities of cancer cells to Fast- and TRAIL-induced apoptosis in vitro In this example, sensitivities of different cancer cell lines (derived from prostate, cervical and liver cancers) to Fast- and TRAIL-mediated apoptosis were compared in vitro.

An expression vector for TRAIL, Ad.TRAIL/GFPTET, was used to express TRAIL in these cancer cells. The construction of Ad.TRAIL/GFPTeT is illustrated in Figure 3. Similar to Ad/FasL-GFP'~ET, this vector contains a transactivator driven by the CMV promoter in the E1 region, and the TRAIL-IRES-GFP expression cassette under the control of the THE promoter in the E4 region, so that the expression of both TRAIL and GFP can be regulated by the addition of doxicycline to the culture media. The internal ribosome entry site (IRES) of the encephalomyocarditis virus allows expression of two genes from the same mRNA transcript. Although the GFP is not fused to the apoptotic protein TRAIL, its expression is correlated with that of TRAIL. Since TRAIL is in front of the GFP and the IRES sequence, the level of its expression should be several folds higher than GFP. Liu et al. (2000) "Generation of mammalian cells stably expressing multiple genes at predetermined levels" Anal. Biochem. 280:20-28. This will assure high levels of TRAIL expression in cells that GFP expression can be observed with UV microscope.
To determine if TRAIL expression can induce apoptosis of cancer cells, we transduced TRAIL into four different cancer cell lines: LNCaP
(prostate), HeLa (cervical), A549 (lung), and C3A (liver). We infected these cells with Ad.TRAIL/GFPTer at the same MOI of 10. All these cells demonstrated sensitivity to TRAIL induced apoptosis at different levels of sensitivities, and the sensitivities appeared lower than those of the cancer cells to Fast. To confirm this observation, we have compared the efficacy of the Fast-GFP and TRAIL in inducing apoptosis in parallel experiments. We infected the cancer cells with Ad/GFP, Ad/FasL-GFPTET and Ad.TRAIL/GFPTET at comparable MOI. Similar to the FasL-sensitivity studies (described earlier), the susceptibility of the these cells was analyzed by the number of GFP-expressing cells and the sensitivities to Fast and TRAIL induced apoptosis are determined based on cell morphology in these initial experiments. As shown in Figure 4, panels "TRAIL", cells showed different levels of apoptosis. In all the cells tested, less cells underwent apoptosis in Ad.TRAIL/GFPTEr infected wells than those infected with Ad/FasL-GFPTer (panels labeled with "Fast") suggesting that LNCaP, HeLa, A549 and C3A cells are more sensitive to Fast than TRAIL-induced apoptosis.
Example 6. Adenovirus-mediated TRAIL expression induces apoptosis in cancer cells, but not in normal fibroblasts.
It is believed that one of the major advantages of TRAIL tumor therapy is that TRAIL expression is supposed to be much less toxic to normal cells than that of Fast, while still inducing apoptosis in tumor cell. To test this "tumor specificity", we transduced normal human fibroblasts with Ad.TRAIUGFPTET at MOI about 10. We obtained primary early passage human fibroblasts from foreskin samples and tested them for their sensitivity to apoptosis induced by our Ad/FasL-GFPTer and Ad.TRAIL/GFPTEr vectors. We have found that primary human fibroblasts were quite sensitive to Fast-GFP induced apoptosis, so that even at a low transduction efficiency, most of them displayed standard apoptotic morphology (Figure 5, panel FasL). In contrast, primary fibroblasts transduced with adenovirus vector expressing TRAIL
were essentially unaffected, even at MOI five-fold higher than those used to deliver Fast-GFP (Figure 5, panel TRAILXS). We therefore confirm the findings that TRAIL does not induce significant apoptosis in normal cells, even at high expression levels from our adenovirus vector.
These results suggest that vector-mediated intratumoral delivery of TRAIL can be even safer than Fast.
Although the present invention has been described with reference to specific details of certain embodiments thereof, it is not intended that such details should be regarded as limitations upon the scope of the invention except as and to the extent that they are included in the accompanying claims.
Throughout this application, various publications are referenced.
The disclosures of these publications in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art to which this invention pertains.

TABLE 1: Fas-Mediated Cytotoxicity in Prostate Cancer Cell Lines Treated for 48 hours with either Anti-Fas Antibody, C2-Ceramide (22 hours) or AdIGFP-FasLTeT (expressed as % cytotoxicity _ SD) Cell Normal Mouse C2-Ceramide Anti-Fas IgM Ad/CMV-GFP Ad~GFP
FasLrer Line Serum (30~M (CH-11)(500 ng/ml) (M01 100) (M01 100) (500 nglml) for 22 hrs) DU145 0.87.1 61 5 6.010.4 1.92.8 69.6 4.5 PC-3 1.35.8 769 1.32.2 0.95.0 84.8 1S 1.1 PPC-1 2.30.3 587 29.22.3 2.36.1* 98.0 7.1 LNCaP 7.514.2 ND 11.613.7 1.63.2 96.4 4.3 TSU-Pr1 -3.52.3 729 -1.92.8 11.67.0 81.3 5.0 Jurkat(+ctrl) 98 2 72.3 0.9 -19.5 22.5"93.0 2.1 5.3 3.4"

K-562(-ctrl) - - -1.3 5.5" -114 2S . 8.1"

*MOI 10, "M01 1000. In all experiments N=3 (except N=2 for ceramide experiments using TSU and PC-3). Percent cytotoxicity was determined using the MTS assay. In brief, cells were seeded in a 12-well plate with one ml of media. Prior to treatments, cells were grown to 75%
confluency and treated with either 500 mg/ml CH-11 anti-Fas antibody, 500 ng/ml Normal Mouse Serum or 30 p,M C2-ceramide. For adenoviral 3S transduction, approximately 1x 105 cells/well were treated with either AdCMVGFP or AdGFPFasLTEZ at an MOI between 10-1000. For each cell line, positive controls were left untreated, and 1 ml of media was used as a negative control. The cells incubated for 48 hours at 37°C
for maximal cell killing. Media was aspirated and replaced with 0.5m1 fresh media + 100,1 of CeIITiter 96 AQueous One Solution Reagent per well.
Cells were incubated for an additional 1-3 hours at 37°C. After incubation 120~,i of media was placed into a 96 wet( plate and absorbance readings were taken using a Vmax kinetic microplate reader at 490nm. Percent cytotoxicity was calculated as follows: % cytotoxicity = [1-(OD of experimental well/ OD of positive control well)] x 100. For ceramide assays, 1x 104 cells/well were seeded in a 96-well plate. The following morning cells were washed and incubated with 100.1 of 30~.M
Dihydro- or C2-ceramide (diluted from a 10mM stock in ethanol) in serum-free RPMI 1640. After 24 hours 20,1 Celltiter 96 AQueous One Solution Reagent was added each well and plates were incubated for an additional 1-4 hours. Absorbance and % cytotoxicity were determined as above. In each experiment, data points were run in triplicate.

Claims (58)

What is claimed is:

1. A method for inducing TRAIL-mediated death of cancer cells, the method comprising:
introducing an expression vector into a group of cells comprising cells that express a receptor for TRAIL, the expression vector comprising a polynucleotide sequence encoding TRAIL, the expressed TRAIL inducing cell death in those cells which express the TRAIL
receptor through interaction between TRAIL and the receptor.

2. The method of claim 1, wherein the TRAIL receptor is a membrane-bound receptor.

3. The method of claim 2, wherein the TRAIL receptor is DR4 or DR5.

4. The method of claim 1, wherein the group of cells into which the expression vector is introduced comprises a mixture of cells which express the TRAIL receptor and cells which do not express the TRAIL
receptor.

5. The method of claim 4, wherein the expression vector is introduced into cells which do not express the TRAIL receptor.

6. The method of claim 4, wherein the expression vector is introduced into cells which do express the TRAIL receptor.

7. The method of claim 4, wherein the expression vector is introduced into cells which do not express the TRAIL receptor and cells which do express the TRAIL receptor.

8. The method of claim 1, wherein the group of cells are contained in a solid tumor.

9. The method of claim 8, wherein the solid tumor is selected from the group consisting of breast, prostate, brain, bladder, pancreas, rectum, parathyroid, thyroid, adrenal, head and neck, colon, stomach, bronchi and kidney tumors.

10. The method of claim 1, wherein introducing an expression vector into the group of cells is performed parenterally, intraperitoneally, intravenously, intraartierally, transdermally, sublingually, intramuscularly, rectally, transbuccally, intranasally, liposomally, via inhalation, vaginally, intraoccularly, via local delivery by catheter or stent, subcutaneously, intraadiposally, intraarticularly, intrathecally, or in a slow release dosage form.

11. The method of claim 1, wherein introducing the expression vector is performed by direct injection of the expression vector among the group of cells.

12. The method of claim 1, wherein the expression vector is a plasmid.

13. The method of claim 1, wherein the expression vector is a viral vector.

14. The method of claim 13, wherein the viral vector is selected from the group consisting of adenovirus, adeno-associated virus, vaccinia, retrovirus, and herpes simplex virus vectors.

15. The method of claim 13, wherein the expression vector is an adenoviral vector which is replication-competent or replication-incompetent.

16. The method of claim 1, wherein the expression of TRAIL is regulated by a conditional promoter in the vector, the cells into which the expression vector is introduced expressing TRAIL when conditions are suitable to activate the conditional promoter.

17. The method of claim 16, wherein the conditional promoter is a tissue-specific promoter.

18. The method of claim 17, wherein the tissue-specific promoter is selected from the group consisting of a prostate-specific promoter, a breast-specific promoter, a pancreas-specific promoter, a colon-specific promoter, a brain-specific promoter, a kidney-specific promoter, a bladder-specific promoter, a lung-specific promoter, a liver-specific promoter, a thyroid-specific promoter, a stomach-specific promoter, an ovary-specific promoter, and a cervix-specific promoter.

19. The method of claim 16, wherein the group of cells are prostate cancer cells and the conditional promoter of the expression vector is a prostate-specific promoter.

20. The method of claim 19, wherein the prostate-specific promoter is selected from the group consisting of PSA promoter, .DELTA.PSA promoter, ARR2PB promoter, probasin promoter, gp91-phox gene promoter, and prostate-specific kallikrein promoter.

21. The method of claim 16, wherein the conditional promoter of the expression vector is a liver-specific promoter.

22. The method of claim 21, wherein the liver-specific promoter is selected from the group consisting of liver albumin promoter, alpha-fetoprotein promoter, .alpha.,-antitrypsin promoter, and transferrin transthyretin promoter.

23. The method of claim 16, wherein the conditional promoter of the expression vector is a colon-specific promoter.

24. The method of claim 23, wherein the colon-specific promoter is selected from the group consisting of carbonic anhydrase I promoter and carcinoembrogen's antigen promoter.

25. The method of claim 16, wherein the conditional promoter of the expression vector is a ovary- or placenta-specific promoter.

26. The method of claim 25, wherein the ovary- or placenta-specific promoter is selected from the group consisting of estrogen-responsive promoter, aromatase cytochrome P450 promoter, cholesterol side chain cleavage P450 promoter, and 17 alpha-hydroxylase P450 promoter.

27. The method of claim 16, wherein the conditional promoter of the expression vector is a breast-specific promoter.

28. The method of claim 27, wherein the breast-specific promoter is selected from the group consisting of G.l. erb-B2 promoter, erb-B3 promoter, .beta.-casein, .beta.-lacto-globulin, and whey acidic protein promoter.

29. The method of claim 16, wherein the conditional promoter of the expression vector is a lung-specific promoter.

30. The method of claim 29, wherein the lung-specific promoter is surfactant protein C Uroglobin promoter.

31. The method of claim 16, wherein the conditional promoter of the expression vector is a skin-specific promoter.

32. The method of claim 31, wherein the skin-specific promoter is selected from the group consisting of K-14-keratin promoter, human keratin 1 promoter, human keratin 6 promoter, and loicrin promoter.

33. The method of claim 16, wherein the conditional promoter of the expression vector is a brain-specific promoter.

34. The method of claim 33, wherein the brain-specific promoter is selected from the group consisting of glial fibrillary acidic protein promoter, mature astrocyte specific protein promoter, myelin promoter, and tyrosine hydroxylase promoter.

35. The method of claim 16, wherein the conditional promoter of the expression vector is a pancreas-specific promoter.

36. The method of claim 35, wherein the pancreas-specific promoter is selected from the group consisting of villin promoter, glucagon promoter, and Insulin Islet amyloid polypeptide promoter.

37. The method of claim 16, wherein the conditional promoter of the expression vector is a thyroid-specific promoter.

38. The method of claim 37, wherein the thyroid-specific promoter is selected from the group consisting of thyroglobulin promoter and calcitonin promoter.

39. The method of claim 16, wherein the conditional promoter of the expression vector is a bone-specific promoter.

40. The method of claim 39, wherein the bone-specific promoter is selected from the group consisting of Alpha 1 collagen promoter, osteocalcin promoter, and bone sialoglycoprotein promoter.

41. The method of claim 16, wherein the conditional promoter of the expression vector is a kidney-specific promoter.

42. The method of claim 41, wherein the kidney-specific promoter is selected from the group consisting of renin promoter, liver/bone/kidney alkaline phosphatase promoter, and erythropoietin promoter.

43. The method of claim 16, wherein the conditional promoter is an inducible promoter.

44. The method of claim 43, wherein the inducible promoter is a promoter inducible by tetracycline or doxycycline.

45. The method of claim 43, wherein the inducible promoter is a promoter inducible by steroid.

46. The method of claim 45, wherein the steroid is selected from the group consisting of glucocorticoid, estrogen, androgen, and progestrone.

47. The method of claim 16, the method further comprising creating the conditions suitable to activate the conditional promoter.

48. The method of claim 47, wherein creating the conditions suitable to activate the conditional promoter comprises delivering to the group of cells tetracycline or deoxycycline.

49. The method of claim 47, wherein creating the conditions suitable to activate the conditional promoter comprises delivering to the group of cells a steroid selected from the group consisting of glucocorticoid, estrogen, androgen, and progestrone .

50. The method of claim 1, wherein the expression vector further comprises a reporter gene.

51. The method of claim 50, wherein the expression vector expresses the reporter gene as a fusion protein with TRAIL.

52. The method of claim 50, wherein the expression vector expresses the reporter gene as a single protein bicistronically with the TRAIL via a mechanism of internal ribosome entry site (IRES) or splicing donor/acceptor sites.

53. The method of claim 50, wherein the reporter gene encodes green fluorescent protein.

54. The method of claim 1, wherein the expression vector further comprises a polynucleotide sequence encoding a regulatory protein.

55. The method of claim 54, wherein the expression vector expresses the regulatory protein as a fusion protein with TRAIL.

56. The method of claim 55, wherein the regulatory protein in the fusion protein is a protein that causes tissue-specific localization of the apoptosis-signaling ligand.

57. The method of claim 1, wherein the method is performed ex vivo where the group of cells into which the expression vector is introduced are contained in a sample taken from a patient having cancer.

58. The method of claim 1, wherein the method is performed in vitro where the group of cells into which the expression vector is introduced are contained in a cell culture.