FIELD OF THE INVENTION

[0002] The present invention provides a method and composition for administration of a biologically active moiety by the use of mammalian cells that are naturally immune privileged, and that have been isolated and genetically modified so as to express the biologically active moiety in pharmacologically effective amounts. The biologically active moiety is not naturally expressed by the cells or is not expressed in pharmacologically effective amounts. More specifically, the invention employs in vitro genetic engineering of allogeneic and xenogeneic donor cells that are naturally immune privileged and then administering of the genetically modified cells to host mammals for sustained delivery of the biologically active moiety in vivo.

DESCRIPTION OF THE RELATED ART

[0003] Immune privilege: Naturally occurring sites or tissues, such as the eye, testis, and the brain, that are immune-privileged were first described as such more than a century ago. Immune-privileged sites are regions of the body where grafts of foreign tissue survive for extended periods relative to sites that are not privileged. Grafts of immune-privileged tissues are more resistant to immune rejection than tissues that are not privileged. Expression of various molecules and multiple features has been found to contribute to maintenance of immune privilege (Streilein, 1995). Among those sites and tissues identified as immune-privileged are the anterior chamber of the eye, the cornea, retina, brain and peripheral nervous system, hair follicles, cartilage, liver, adrenal cortex, pregnant uterus, placenta, ovary, testis, prostate (Streilein, 1995). This invention provides a method and a composition for providing a biologically active moiety in vivo by administering cells from the tissues and sites that are naturally immune privileged, and that have been genetically modified to express the biologically active moiety in pharmacologically effective amounts in vivo.

[0004] Maintenance of immune privilege in these tissues has been thought to be variously associated with expression or secretion of a number of molecules, including immunosuppressive cytokines, corticosteroids, and Fas ligand, and the reduced or absent expression of class I and class II major histocompatibility complex molecules.

[0005] Different types of immune-privileged cells are biologically unique. Immune privilege is a complex property that is possessed by various naturally occurring cells and tissues that express multiple molecules that mediate the phenomenon. All of the naturally occurring immune-privileged cells express a multitude of molecules that are immunosuppressive as shown in Table 1. All immune-privileged cells apparently express Fas ligand. Nonetheless, there is significant variability in the expression of purported molecular mediators from one cell type to another (see Table 1). Thus, the different types of cells are biologically unique in the way that they create their immune-privileged status. This implies that the ability of the different types of cells to survive allogeneic implantation will vary. In addition, these cells are derived from various tissues of the body and, therefore, vary in their endogenous expression of other molecules and in their normal functional roles. The genetic modification of different cell types also varies and this is very significant for delivery of recombinant proteins in vivo.

[0006] Multiple molecular mediators produce immune privileged status. One likely molecular mediator of immune privilege is Fas ligand. Fas ligand, the naturally occurring ligand of Fas, was purified, cloned and identified as a protein of approximately 40,000 M _r homologous to members of the tumor necrosis factor (TNF) family (Suda et al., 1993). Expression of 1

[0007] recombinant Fas ligand on the surface of COS cells (a fibroblast-like kidney derived cell line) induced apoptosis in Fas-expressing cells within a few hours (Suda et al., 1993). Fas ligand does not have a signal sequence, but has a domain of hydrophobic amino acids in the middle and no signal sequence at the NH ₂ -terminus indicating that it is a type II transmembrane protein with the COOH terminal region outside the cell. Human and mouse Fas ligand have 76.9% homology and are functionally interchangeable (Takahashi et al., 1994). Human Fas ligand has been termed Apo-1 ligand while both Fas ligand and Apo-1 ligand are also referred to as CD95 ligand. The human Fas ligand contains 281 amino acids, and consists of a 79 amino acid cytoplasmic domain, a 23 amino acid transmembrane domain and a 179 amino acid extracellular domain.

[0008] Expression of Fas ligand in immune-privileged tissues or sites has been shown in tissues such as the testes, eye, spleen and thymus (Griffith et al., 1995). Cells that naturally express Fas ligand appear to possess immune privilege or produce specific immunological unresponsiveness. This was first demonstrated by Bellgrau et al. (Bellgrau et al., 1995; Selawry and Cameron, 1993). Sertoli cells from the testis of gld mice, a mutant strain of mice expressing a non-functional Fas ligand due to a point mutation, were rejected upon transplantation. Sertoli cells from the testis of normal and lpr mice, a mutant strain of mice that lack functional Fas but express normal Fas ligand, when transplanted under the kidney capsule of allogeneic animals were not rejected. Expression of Fas ligand by the testicular Sertoli cells was demonstrated by reverse transcription-polymerase chain reaction (RT-PCR).

[0009] Induction of apoptosis via the Fas-Fas ligand interaction also is a potent mechanism of immune privilege in the eye (Griffith et al., 1995). The results of Griffith and co-workers indicate that expression of Fas ligand triggers apoptosis in antigen-activated T cells that express Fas, and that constitutive expression of Fas ligand may be essential to maintenance of immune-privileged sites and tissues.

[0010] The eye is a privileged site that cannot tolerate destructive inflammatory responses or vision is destroyed. Activated neutrophils and lymphocytes entering the anterior chamber of the eye in response to a viral infection underwent apoptosis mediated by the Fas-Fas ligand interaction and did not cause tissue damage (Griffith et al., 1995). In contrast, viral infection in gld mice, a mutant strain of mice lacking functional Fas ligand due to a point mutation (Takahasi et al., 1994) resulted in inflammation and invasion of the eye by inflammatory cells that did not undergo apoptosis. Thus, immune privilege in the eye is maintained not through a passive process, but is an active process that induces cell death in potentially dangerous infiltrating cells by the Fas-Fas ligand interaction.

[0011] Griffith et aL using Northern (messenger RNA) blot and reverse transcriptasepolymerase chain reaction (RT-PCR) analysis of total RNA determined the expression of Fas ligand by the testis, thymus, eye and the spleen (Griffith et al., 1995). The location of Fas ligand in these tissues was also established immunohistochemically. Addition of competing Fas ligand peptide inhibited the binding of the antibody and verified the specificity of the antibody reaction within the tissue.

[0012] Another likely mediator of immune-privileged status is galectin-3. Galectin-3 is found on cytotrophoblasts and retinal pigment epithelial cells, but not on Sertoli cells. Galectin-3 is not a member of the Bcl-2 family, but at residues 180-183 it contains the four amino acid motif (NWGR) conserved in the BH1 domain of the bcl-2 family, and it has 48% sequence similarity with Bcl-2 (Yang et al., 1996). Consistent with this, galectin-3 has been found to be a novel antiapoptotic protein (Akahani et al., 1997; Yang et al., 1996), improving cellular adhesion and preventing apoptosis induced by loss of cell anchorage (anoikis) (Kim et al., 1999; Matarrese et al., 2000a), and influencing mitochondrial homeostatis (Matarrese et al., 2000b). Substitution of the Gly182 residue with Ala in the NWGR motif abrogates its antiapoptotic activity (Akahani et al., 1997; Kim et al., 1999). In T cells galectin-3 interacts with Bcl-2 in a lactose inhibitable manner and confers resistance to apoptosis induced by anti-Fas antibody and staurosporine (Yang et al., 1996). In the BT549 breast carcinoma cell line expression of galectin-3 inhibits cisplatin-induced apoptosis (Akahani et al., 1997).

[0013] Despite genetic differences mothers do not reject their partially allogeneic embryos. Since the advent of modern methods for in vitro fertilization and embryo implantation it has become clear that mothers are no more likely to reject fully allogeneic embryos. Cytotrophoblasts are of fetal origin but they normally exist within semi-allogeneic maternal tissue at the maternal-fetal interface. Thus, among those cells reported to possess immune privilege, cytotrophoblasts should possess the most developed ability to rebuff immune attack.

[0014] Expression of the products of the highly polymorphic class I HLA-A, -B and -C genes that stimulate graft rejection is blocked in the placental trophoblast cells that instead express HLA-G, a nonpolymorphic gene (Hammer et al., 1997; Hunt and Orr, 1992). This is a mechanism of inducing maternal tolerance. Expression of HLA-G has been cited as a possible mechanism for immune privilege of placental cells (Ellis et al., 1986; Kovats et al., 1990; McMaster et al., 1995), and recent in vitro studies support this theory (Carosella, 2000; Carosella et al., 1999; Lefebvre et al., 1999; Moreau et al., 1998; Rouas-Freiss et al., 1999; Rouas-Freiss et al., 2000; Sasaki et al., 1999a; Sasaki et al., 1999b). A number of laboratories have reported survival of implanted allogeneic trophoblasts or ectoplacental cones that contain trophoblastic cells without immunosuppression (Bevilacqua et al., 1991; Bowen and Hunt, 1999; Croy et al., 1984).

[0015] Fas ligand seems to play a role in the immune privilege of cytotrophoblasts, since cytotrophoblast-induced cell death of lymphoid cells in culture was partially inhibited by anti-FasL antibodies (Coumans et al., 1999). However, apparently normal pregnancies ensued in gld mutant mice that fail to produce functional Fas ligand and that were carrying trophoblast transgenic pups that abnormally expressed MHC class I in giant trophoblast cells (Rogers et al., 1998). Cytotrophoblast production of the cytokine IL-10 was found to be an important factor in maternal tolerance, and was immunoinhibitory in in vitro tests (Roth et al., 1996).

[0016] The ability of retinal pigment epithelium and corneal endothelium from mice to survive allogeneic implantation in a non-immune-privileged site was recently demonstrated (Hori et al., 2000; Wenkel and Streilein, 2000). RPE grafts from neonatal C57BL/6 or C57BL/6 gid mutant mice (deficient in functional CD95 ligand expression) were transplanted into the anterior chamber, the subretinal space, the subconjunctival space, and underneath the kidney capsule of BALB/c mice. The grafts from the normal C57BL/6 donors showed significantly enhanced survival at all sites compared with conjunctival grafts but the allogeneic gld grafts were rapidly rejected after transplantation beneath the kidney capsule (Wenkel and Streilein, 2000). When deprived of their epithelia, syngeneic corneas and allogeneic C57BL/6 corneas survived almost indefinitely beneath the kidney capsule (Hori et al., 2000).

[0017] Cytotrophoblasts, retinal pigment epithelial cells, and Sertoli cells all express indoleamine 2,3-dioxygenase (IDO), a tryptophan catabolizing enzyme, that appears to be critical in maintenance of maternal tolerance (Munn et al., 1998). In pregnant mice treated with 1-methyl-tryptophan, an inhibitor of IDO, rapid T-cell mediated rejection of all allogeneic pregnancies occurred. Syngeneic pregnancies of mice treated with the same inhibitor were not affected (Munn et al., 1998). The expression of IDO is regulated in human cells by interferons (Malina and Martin, 1996); the most efficient of these is interferon-□IFN-□. Interferons have anti-cancer activity and can inhibit tumor cell growth in culture (Taylor and Feng, 1991). It has been shown in vitro that a primary mechanism of the cytotoxicity of IFN-□ is the induction of IDO. IDO uses two superoxide radicals to cleave the pyrrole ring of tryptophan, an essential amino acid, in the first, and rate limiting step of tryptophan catabolism, and is an antioxidant enzyme (Malina and Martin, 1996). It is now well established that tryptophan starvation resulting from IFN-□treatment is the mechanism of the antiproliferative activity of IFN-□ on many cell lines and intracellular parasites (Taylor and Feng, 1991). Tryptophan starvation can lead to apoptosis of cells. Within 48 h of treatment with IFN-□ ME180 human epidermoid carcinoma cells underwent apoptosis that could be prevented by adding tryptophan and induced by removing it in the absence of IFN-{tilde over (□)} Replication of the parasite Toxoplasma gondii was inhibited by treatment of infected RPE cells in culture with IFN-{tilde over (□)} The inhibition could be reversed by addition of tryptophan (Nagineni et al., 1996).

[0018] The number of different immunosuppressive molecules immune-privileged cells express suggests that modifying non-immune-privileged cells to express one single molecular mediator could be insufficient to achieve allogeneic survival in vivo without immunosuppressive drugs. Modification of cells with genes encoding individual molecules mediating immune privilege to artificially transfer the property to non-immune privileged cells does appear to be an approach with serious limitations.

[0019] Reports of the role of Fas ligand in maintenance of immune privilege stimulated research in the transgenic expression of FasL on the allogeneic cells to prevent rejection. Fas ligand induces apoptosis of cells, including activated lymphocytes, that express its receptor, Fas (CD95/APO-1), and prevents inflammatory reactions at immune privileged sites by triggering Fas-mediated apoptosis of infiltrating proinflammatory cells. The initially promising reports (Bellgrau et al., 1995; Griffith et al., 1995; Lau et al., 1996) were followed by a number of reports of failures to achieve tolerance using transgenic Fas ligand (Allison et al., 1997; Chervonsky et al., 1997; Kang et al., 1997). An increasing number of studies have shown that Fas ligand can induce potent inflammatory responses that appear to limit its ability to inhibiting graft rejection (Kang et al., 1997; O'Connell, 2000; Ottonello et al., 1999; Turvey et al., 2000).

[0020] Chervonsky and coworkers demonstrated that expression of transgenic Fas ligand on allogeneic β-islet cells caused rejection due to Fas-mediated destruction of the islet cells themselves (Chervonsky et al., 1997). Normally islet cells do not express Fas, but contact with cells expressing Fas ligand can lead to upregulation of Fas on islet cells, and the capacity for Fas upregulation increases with age. Chervonsky et aL concluded that Fas-mediated apoptosis of islet cells may play a major role in development of Type 1 (autoimmune) diabetes.

[0021] In a review of this area, Green and Ware (Green and Ware, 1997) discuss several other possible explanations for the variation in the results with transgenic Fas ligand; such as the age of donor animals, or factors in the host at the specific transplantation site such as interferon γ or IL-8. The amount of soluble Fas ligand secreted may vary among Fas ligand-expressing cells. The secreted form of Fas ligand is a monomer, unlike the membrane bound protein, that is trimerized when functional. Perhaps varying amounts of recombinant Fas ligand on the cell membranes is significant in terms of tipping a balance between local inflammation and immunoprotection. Recently human recombinant soluble Fas ligand has been found to be endowed with potent chemotactic properties toward human neutrophilic polymorphonuclear leukocytes (neutrophils) (Ottonello et al., 1999). Also, naturally immune-privileged cells may express other molecules that are critical for induction of immune privilege.

[0022] A colon carcinoma cell line, CT26, was stably transfected with Fas ligand (CT26-CD95L). When injected in syngeneic Balb/c mice subcutaneously, the CT26-CD95L cells were rejected by neutrophils activated by Fas ligand. However, CT26-CD95L survived in the intraocular space because of the presence of transforming growth factor-beta (TGF-beta) that inhibited neutrophil activation. Importantly, providing TGF-beta to the subcutaneous sites prevented rejection of the tumor at those sites. Thus, Fas ligand together with TGF-beta was able to promote immunologic tolerance of the tumor cells but expression of Fas ligand alone was not able to do so suggesting that together these cytokines generate a microenvironment that promotes immune tolerance that could prevent allograft rejection (Chen et al., 1998).

[0023] Similar conclusions can be drawn from recent work with Sertoli cells from non-obese diabetic (NOD) mice, a model for autoimmune Type 1 diabetes. The NOD Sertoli cells were implanted under the right renal capsule of diabetic NOD mice, whereas NOD islets alone were implanted under the left renal capsule (Suarez-Pinzon et al., 2000). After 60 days 9 of 14 mice that received islet and Sertoli cells grafts were normoglycemic compared to 0 of the 6 mice that received islet grafts alone. Immunohistochemistry revealed that TGF-beta expression by the grafted Sertoli cells was high, but the expression of Fas ligand decreased after transplantation. Administration of anti-TGF-beta antibody completely abrogated the protective effect of Sertoli cells on islet graft survival, whereas anti-Fas ligand antibody did not. (Suarez-Pinzon et al., 2000). The expression of TGF-beta was critical in the ability of the Sertoli cells to prevent death of the islets due to the autoimmune condition of the NOD mice. Together these data demonstrate the complexity of the phenomenon termed immune privilege, and the fact that it could be difficult to recreate it by recombinant expression of a single molecule mediator. In addition, the data suggest that assays that measure the amount of TGF-beta secreted by immune-privileged cells in vitro and in vivo could be useful to determine the population of cells that would be most effective in warding off rejection by the host immune system in allogeneic implants. We plan to compare the allogeneic survival of various types of immune privileged cells with the levels of TGF-beta they secrete. The complexity and variety of the means that nature has used to create the immune privileged status of particular cells are indications of the difficulty of achieving this status.

[0024] The use of the genetically modified cells that are naturally immune privileged is a practical and novel method for ex vivo gene therapy and protein drug delivery. In this method we do not attempt to dissect and then reassemble what nature has provided, but rather to make use of it in a novel manner. Our in vitro study of the immunosuppressive effects of naturally occurring murine immune privileged cells has revealed characteristics that make trophoblasts more suitable for delivery in some parts of the body Sertoli cells. We postulate that this type of assay and other in vitro assays such as determination of TGF-beta secretion will reveal measurable differences between the cell types that will aid in identification and characterization of cells that will be significantly more useful as in vivo drug delivery vehicles than other types of immune-privileged cells.

[0025] In in vivo studies we are comparing the survival of different types of allogeneic immune-privileged cells and the immune response of the host animal to the cells. We postulate based on our in vitro studies that some of these types, such as trophoblast cells, will be better able to survive and cause less immune and inflammatory reactions from the host.

[0026] Transplant Rejection: Transplantation of healthy organs or cells into a mammal suffering from a disease that affects the organ or cells may be necessary to save the mammal's life. A major problem in transplantation or implantation of any foreign tissue or cell is immune-mediated graft rejection in which the recipient's T-lymphocytes recognize donor histocompatibility antigens as foreign. Thus, use of non-autologous human (allografts) and or mammalian (xenografts) cells requires preventing immune rejection by the host. The donor and recipient are matched as closely as possible to prevent rejection in transplantation of humans. Survival of even well-matched grafts often necessitates high dose chronic treatment with nonspecific immunosuppressive drugs that can result in opportunistic infections and in the majority of transplant patients, long term complications (Gjertson, 1991; Manninen et al., 1991). Rejection is still a leading cause of graft failure, despite progress in immunosuppressive therapy.

[0027] Both TGF-beta and Fas ligand along with CD8 accessory molecule, and the major histocompatibility class I gene products, have been directly implicated in the mechanism of the “veto effect”, that is deletion of graft reactive T cells by administration of low doses of donor bone marrow cells. This method to induce transplantation tolerance for allografts without chronic immunosuppression is close to clinical use. It involves transient peritransplant depletion of host T cells followed by intravenous administration of a low dose of donor bone marrow cells (George and Thomas, 1999).

[0028] Delivery of Therapeutic Proteins: Over the last 15 years the application of recombinant DNA technology in the pharmaceutical field gave rise to the entire biotechnology industry. A distinct advantage of biotechnology-derived proteins over those isolated from natural sources is enhanced purity. Obstacles to the use of proteins as therapeutic agents include the propensity to aggregate, adhere to surfaces, become denatured and rapidly metabolized. These have been at least partially overcome, and increasing numbers of protein products are on the market. New protein-based pharmaceuticals that have and are arising from biotechnology processes include a wide spectrum of pharmacologically active substances, such as hormones, hormone-like regulatory compounds, enzyme inhibitors, vaccines, and antibodies.

[0029] Few protein biopharmaceuticals can be successfully administered orally because of their instability in the acidic environment of the stomach and the barrier to absorption presented by the gastrointestinal tract (Hudson and Black, 1993). Rapid metabolism by a multitude of enzymes and nonlinear pharmacokinetics are other challenges in the delivery of protein and peptide drugs (Wearley, 1991).

[0030] Optimally a drug or biologic substance is delivered to the site of pharmacologic action, is able to penetrate the biologic barriers, and access the site of action, either intra- or extracellular, in therapeutically effective doses (Bruck, 1991). The field of controlled drug delivery may be divided in four categories as follows: (1) non-specific or non-targeted drug release systems such as polymeric diffusion systems and infusion pumps; (2) pharmaceutical formulations including various coatings to sustain action of drugs; (3) prodrugs that can undergo transformation in the body before eliciting their pharmacologic effects; and (4) targeted delivery of drugs and biologicals via carriers, such as liposomes, biodegradable polymers, antibodies, and genetically engineered cells.

[0031] Most protein products are delivered by invasive routes such as intravenous (i.v.), intramuscular (i.m.), or subcutaneous (s.c.). This is an obvious disadvantagetheir delivery can be associated with some risk and cause minor discomfort. Until new dosage forms are developed, the availability of proteins in the ambulatory setting is limited. However, some methods to minimizing the remaining obstacles of non-invasive protein/peptide drug delivery have been found.

[0032] The methods for enhancing protein delivery include increasing the absorption, minimizing metabolism and prolonging the half-life of the protein (Wearley, 1991). Administration of either enzyme inhibitors or protective polymers and permeation enhancers can improve the bioavailability of proteins and peptides delivered by noninvasive means. However, the bioavailability may still remain fairly low.

[0033] Nasal administration of nonpeptides and peptides of ten residues or less has been quite successful. Examples include oxytocin, vasopressin and desmopressin acetate and luteinizing hormone-release hormone and its superanalogs buserelin, leuprolide and nafarelin. However, when the number of amino acids is increased to 20 or greater, as in insulin, glucagon, or growth hormone releasing hormone, low bioavailability is the result, except when delivered with a penetration enhancer. Other drug delivery routes for proteins studied include transdermal, buccal, rectal, respiratory and ocular (Wearley, 1991).

[0034] Most regulatory proteins, such as insulin and growth hormone, must reach distant organs or tissue without being extensively metabolized. Unique ways to achieve this goal include encapsulating the protein-based compounds in lipid complexes such as liposomes, protecting proteins in a sheath composed of poorly soluble biopolymers such as polyethylene glycol, fusing the protein with antibodies that can be directed to distinct tissue, and using the body's own cells as carriers. The use of gene-carrying cells as “factories” that produce the desired protein in a targeted tissue is very promising (Hudson and Black, 1993). Cells themselves have been used to deliver protein-based toxins to malignant cancer cells. For example, in the first clinical trial of gene therapy, tumor-invading T-lymphocytes were engineered to secrete tumor necrosis factor, which is cytotoxic (Ledley, 1989; Rosenberg et al., 1990).

[0035] Gene Therapy: Human gene therapy began in the 1950s and 1960s when successful renal transplantation lead to the concept that injecting healthy cells into patients with genetic diseases might be therapeutic (Brady, 1966). Initial clinical studies were undertaken in the 1970s with transplantation to treat Gaucher disease (Groth et al., 1972) and Hunter syndrome (Dean et al., 1975) and the term ‘gene therapy’ was coined (Friedmann and Toblin, 1972).

[0036] Gene therapy is an approach to human disease based on the transfer of genetic material (DNA) into an individual. This can be achieved by direct administration of DNA or DNA-containing viruses to blood or tissues (in vivo), or indirectly through the introduction of cells engineered to contain foreign DNA (ex vivo) (Orkin and Motulsky, 1995). Only the somatic cells and not the germ cells (eggs and sperm) are the targets of gene therapy efforts. Until recently, most of the work in human gene therapy centered on rare genetic diseases. However, gene therapy may be appropriate in a variety of clinical settings, such as:

[0037] 1) single-gene inherited disorders such as delivery of normal factor VIII genes to patients with hemophilia;

[0038] 2) common, multifactorial disorders such as coronary heart disease;

[0039] 3) cancer by correction of mutations in tumor suppressor genes, e.g. p53, or approaches such as delivery of genes encoding enzymes involved in conversion of prodrugs to active form, and

[0040] 4) infectious diseases such as HIV.

[0041] Diseases that are currently treated by the administration of proteins may be amenable to treatment by gene therapy, and in these cases gene therapy can be thought of as an in vivo protein production and delivery system.

[0042] The first human patients received gene therapy at the NIH in 1991. As of June 1995 there have been 106 clinical protocols involving gene transfer in humans approved by the NIH Recombinant DNA Advisory Committee (RAC), and a total of 597 human subjects have undergone gene transfer experiments. Estimated expenditures on gene therapy research by the NIH and biopharmaceutical industry are over $400 million a year (Hanania et al., 1995; Orkin and Motulsky, 1995). However, despite promising results in animals, clinical efficacy has not been definitively shown in a gene therapy protocol in humans.

[0043] Major problems in this field include the following:

[0044] 1) inability to achieve efficient gene transfer;

[0045] 2) lack of persistence in gene maintenance and expression;

[0046] 3) inability to achieve expression in appropriate tissues and cells;

[0047] 4) immunorejection after introduction of genetically modified allogeneic or xenogeneic cells (Tai and Sun, 1993);

[0048] 5) inadequate understanding of the interactions of the vectors with the host, and

[0049] 6) lack of understanding of the results of gene therapy protocols, which are hindered by a low frequency of gene transfer, reliance on qualitative assessments of transfer and expression, lack of suitable controls and rigorously defined endpoints (Orkin and Motulsky, 1995).

[0050] Vector systems that currently have been used or are under consideration for use in gene therapy include retrovirus, adenovirus, adeno-associated virus, herpes virus, pox virus, naked DNA and facilitated DNA (Orkin and Motulsky, 1995). Methods being explored to deliver DNA include particle bombardment (also known as ballistic, microprojectile or gene gun method), electrically-induced DNA transfer, calcium phosphate-mediated DNA transfection, liposomal and receptor-mediated gene delivery (Bennett et al., 1994; Wolff, 1994).

[0051] Many protocols approved for somatic-cell gene therapy do not involve direct administration of the genetic vector, but rather are ex vivo strategies that require the isolation of somatic cells from a patient, the stable introduction of a gene of therapeutic interest into the cells, the isolation and clonal propagation of a single engineered cell, and finally, the reintroduction of the cells into the patient (Heartlein et al., 1994; Kessler et al., 1993). An ex vivo strategy ensures that the genes are delivered to the cells of the right tissues or organs.

[0052] Human clinical trials of gene therapy for specific diseases have been performed or have commenced in areas including cancer vaccines, genetic sensitization trials in which sensitization to ganciclovir is conferred by transfection with herpes virus thymidine kinase, gene replacement trials with the adenosine deaminase gene for severe combined immmunodeficiency disease, (SCID), for Gaucher's disease in which the glucocerebrosidase gene is dysfunctional, and in cystic fibrosis with cDNA to replace the cystic fibrosis transmembrane conductance regulator (CFTR) to replace the gene that is missing in these patients (Hanania et al., 1995). Another strategy being explored with gene therapy is that of chemoprotection for autologous bone marrow transplantation and chemotherapy sensitization with anti-oncogenes such as administration of a vector containing a functional wild-type p53 transcription unit.

[0053] Transkaryotic implantation is a term for isolation of somatic cells from a patient, introduction of a gene of therapeutic interest, isolation and clonal propagation of a single engineered cell, and finally, reintroduction of the cells into the patient. The use of nonimmortalized clonal strains of secondary and primary cells may offer significant advantages, besides lacking tumorigenicity, these cells are more likely to maintain differentiated functions than immortalized cells. Transkaryotic implantation circumvents the problem of immunorejection by the transfection of autologous cells for reintroduction into the original donor. Transkaryotic implantation will tend to be costly and labor intensive.

[0054] Another approach, termed microencapsulation, has been designed to circumvent the problem of immunorejection of genetically engineered allogeneic or xenogeneic cells from “universal” cell lines (Al-Hendy et al., 1995; Hughes et al., 1994; Squinto et al., 1994; Tai and Sun, 1993; Uludag and Sefton, 1992; Wang et al., 1991). A feature of this method is the prevention of immunorejection by physical isolation of the implanted cells from the host (recipient) immune system by enclosure within microcapsules. The membranes are designed to provide free passage for the recombinant protein products. Mice transplanted with encapsulated transformed fibroblasts secreting human growth hormone (hGH) had detectable levels of hGH over 115 days, the course of a recent study (Tai and Sun, 1993). Approximately 60% viability was observed among encapsulated myoblasts retrieved after secreting mouse growth hormone for six months in Snell dwarf mice, that had enhanced growth (Al-Hendy et al., 1995). Potential problems of this approach include the eventual breakdown of the capsule, the need for high-level product secretion in some cases, and difficulty in achieving long-term survival of encapsulated cells. The pores in the capsules allow for diffusion of recombinant protein products out, but also allow antigenic proteins from dying cells out, and allow the diffusion in of host proteins and molecules, such as IL-1 (17 kDa), TNF-□ (17-51 kDa), IL-6 (26 kDa), oxygen radical, and nitric oxide (Babensee et al., 1998; Hagihara et al., 1997; Rihova, 2000). Immune responses to encapsulated cells are well documented and are greater for xenogeneic cells (Babensee et al., 1998; Rihova, 2000).

[0055] Transgenic Animals. Many human therapeutic proteins are currently produced on a large scale with the aid of recombinant DNA technology in microbial bioreactors and a few in animal cell cultures. A disadvantage of the microbial production of therapeutic proteins is that while microbes such as bacteria and yeast do translate the genetic code into the correct amino acid sequence, they do not necessarily add the correct post-translational modifications such as glycosylation which takes place in the Golgi apparatus of eukaryotic cells or fold the protein to yield the ultimately biologically active product. While the actual production of proteins from microbial bioreactors may be inexpensive, purification and processing of the proteins tends to be costly. Animal cell culture can circumvent some of these problems, but it tends to be prohibitively expensive due to long generation times and requirement for rich media. Systems that have been used to produce recombinant proteins include bacteria, yeast, fungi, plants, baculovirus, mammalian cells and transgenic animals.

[0056] Another possible alternative is the manufacture of proteins in animals, which requires transferring foreign genes into the animals' embryos. If the foreign gene is introduced into the one-cell embryo (fertilized oocyte), and integrated, the transgene becomes a dominant Mendelian genetic characteristic that is inherited by the progeny of the founder animals. The ability to genetically manipulate mammals has opened an immense potential with almost unlimited applications in basic and applied research, and the production of human pharmaceuticals in transgenic animals has become more attractive. With targeted gene transfer, the expression of the transgene of interest can be directed, for example, to the mammary gland so that the protein is secreted into the milk (Janne et al., 1992).

[0057] Considerable progress has been made in targeting tissue-specific expression to the mammary gland and the blood of animals. For example, human proteins have been produced in the milk or blood of transgenic mice, rabbits, sheep, pigs and goats. These proteins include factor IX, alpha-1 antitrypsin, t-PA, antithrombin III, protein C, and human growth hormone (Hoyer et al., 1994). This approach has great potential productivity. If similar yields of Factor IX could be obtained in pigs as were produced of human protein C, a vitamin K-dependent plasma protein, then twenty pigs transgenic for Factor IX could easily produce the two kg of protein that is used each year in the US (Hoyer et al., 1994).

[0058] No recombinant proteins extracted from transgenic animals are yet on the market (Houdebine, 1994), however, there is relatively slow but real progress being made in improving the efficiency of this process. Predictive reports suggest that 10% of the recombinant proteins, corresponding to a $100 million annual market, will be prepared from the milk of transgenic animals by the end of the century.

[0059] Deficiency of prior art. The prior art is deficient in a simple, reliable method for cell-based “gene therapy” that would enable sustained, systemic delivery of proteins and peptides in vivo with little or no need for chronic immunosuppression to prevent rejection. This type of approach has been termed nonautologous somatic gene therapy (Al-Hendy et al., 1995). The present invention will lead to the development of a convenient method for the sustained, systemic delivery of proteins, glycoproteins, and peptides by genetically modified cells that are naturally immune privileged to fulfill a long-standing need ( FIG. 1 ).

[0060] Research and development of gene therapy is a very active field and includes numerous clinical trials in human beings. Nonetheless, difficulties have been encountered and despite promising results in animals, clinical efficacy has not been definitively shown in a gene therapy protocol in humans. The invention described here would enable development of “universal” cell lines that could be thoroughly characterized for safety and quality assurance before implantation. In comparison, thorough characterization of transfected autologous cells for somatic cell gene therapy would be costly and time-consuming. This method obviates the need for patient specific genetic manipulation and is amenable to industrial scale quality control. Thus, accurate analysis of the efficiency of the gene transfer, and the persistence of gene maintenance and expression will be possible. The invention would help ensure the clinical success of cell-based gene therapy and, therefore, greater reflection of the promising results obtained in animal studies (Friedmann et al., 1994).

[0061] Prior art related to this invention includes a large number of patents for transfected cell lines, transgenic animals, and human gene therapy, including a broad-based patent issued by the U.S. Patent and Trademark Office PTO covering ex vivo gene therapy (Anderson et al., 1995).

[0062] Additional prior art to the current invention includes foreign patent document WO9528167 entitled “Methods of Treating Disease Using Sertoli Cells and Allografts or Xenografts,” invented by Helen P. Selawry published on Oct. 26, 1995 which describes a method to create an immune-privileged site in a recipient mammal using Sertoli cells. The method relies on cotransplantation of allogeneic or xenogeneic cells that produce a desirable biological factor together with immune-privileged Sertoli cells to prevent rejection. The use of the genetically-modified Sertoli cells or other immune-privileged cells that are naturally immune privileged to produce and deliver peptides and proteins was not envisioned or described in U.S. Pat. No. 5,579,534. One drawback to the method described in U.S. Pat. No. WO9528167 is that more than one cell type must be administered for therapy.

[0063] Similarly, Gage et al. in U.S. Pat. No. 5,082,670 entitled “Method of Grafting Genetically Modified Cells to Treat Defects, Disease or Damage Of the Central Nervous System” published on Jan. 21, 1992 in claim 24 describe the coadministration of cells (along with the genetically modified donor cells of claim 1) as a therapeutic agent for treating disease or damage to the central nervous system, said therapeutic agent consisting of cellular matter, including homogenate of placenta. Also, in claim 26, Gage et al. describe implantation of cellular material into the central nervous system (along with the genetically modified donor cells of claim 1) to facilitate reconnection or ameliorative interaction of injured neurons, said cellular material including homogenate of placenta (Gage et al., 1997). A continuation application for U.S. Pat. No. 5,082,670 was published on Jul. 22, 1997 as U.S. Pat. No. 5,650,148. U.S. Pat. No. 5,082,670 or the continuation U.S. Pat. No. 5,650,148 does not anticipate or describe genetic modification of the co-implanted cellular material or matter described in claims 24 or 26, such as placental cells. Likewise, the use of unmodified immune-privileged cells that are naturally immune privileged and naturally express Fas ligand, such as placental cells, for therapy or implantation in vivo is not described in the present invention.

[0064] Another patent related to the current invention is U.S. Pat. No. 5,759,536 entitled “Use of Fas ligand to suppress T-lymphocyte-mediated immune responses,” invented by Donald Bellgrau and Richard C. Duke and published on Jan. 7, 1995. U.S. Pat. No. 5,759,536 describes the use of cells or tissues that have been genetically modified to express recombinant Fas ligand, or therapy with recombinant Fas ligand protein itself. The use of genetically-modified immune-privileged cells that naturally express Fas ligand to deliver peptidic biomolecules was not envisioned or described in U.S. Pat. No. 5,759,536.

[0065] U.S. Pat. No. 5,702,700 entitled “Sertoli cells as neurorecovery inducing cells for Parkinson's Disease” dated Dec. 30, 1997 by the inventors Paul R. Sanberg, Don F. Cameron, and Cesario V. Borlongan was not uncovered in the searches performed prior to the Oct. 7, 1996 filing of the original patent application Ser. No. 08/726,531 describing the present invention. U.S. Pat. No. 5,702,700 describes the therapeutic use of trophic factors that are naturally secreted by Sertoli cells by implantation into the central nervous system of mammals. Testis-derived Sertoli cells have been shown to have a trophic effect on dopamine neurons and alleviate hemiparkinsonism in rats (Sanberg et al., 1997). Recent work in this area demonstrated survival of rat Sertoli cells allografts and porcine Sertoli cell xenografts for at least two months in the rat brain without cyclosporin A immunosuppression (Saporta et al., 1997). The use of genetically-modified cells that are naturally immune privileged, such as Sertoli cells, to deliver desired peptidic biomolecules, either in vitro or in vivo, was not envisioned or described by U.S. Pat. No. 5,702,700. Likewise, the use of unmodified immune-privileged cells that naturally express Fas ligand, such as Sertoli cells, for therapy or implantation in vivo is not described in the present invention. The survival of allogeneic and xenogeneic Sertoli cells in rat brain without systemic immunosuppression provides dramatic evidence of the ability of naturally immune-privileged cells to prevent immunorejection in a mammalian central nervous system (Saporta et al., 1997).

SUMMARY OF THE INVENTION

[0066] The present invention provides a method for delivery of a biologically active moiety by administering immune-privilege cells that have been genetically modified to express the biologically active moiety. The biologically active moiety is provided in vivo in pharmacologically effective amounts, and is either not naturally expressed by the immune-privileged cells, or is naturally expressed in amounts that are less than required for pharmacologically effectiveness. The biologically active moiety could be a protein, peptide, gene, or the product of a protein such as a neurotransmitter, and could be expressed as a pro-drug that is activated in the body. The administration of the immune-privileged cells can be performed by a variety of methods including subcutaneous, intravenous, intraperitoneal, and intramuscular infusion or injection. Additionally, the immune-privileged cells could be implanted in specific sites of the body by a number of surgical procedures. The cells could be adherent to an inert polymeric material that would keep them together at a specific location in the body. The cells could be implanted in a polymeric material that is a liquid that gels upon implantation in the body so that the cells are retained at the site of implantation. The expression of the biologically active moiety could be either intracellular, on the extracellular membrane, or secreted by the cells depending on where the biologically active moiety would be therapeutic. The immune-privileged cells could be freshly isolated cells, or cells that have been cultured, or that have been cultured and then frozen. The immune-privileged cells could be, or could be derived in culture from, progenitor stem cells of immune-privileged cells. Immune-privileged cells are those that naturally express Fas ligand and that have greater ability to survive allogeneic grafting or implanting with less immunosuppression than other non-immune privileged cells of the body. A listing of cells that express Fas ligand is given in Table 2. The genetic modification of the cells can be performed by the various methods that are known to one of skill in the art. Promoters could be incorporated into the genetic modification of the immune-privileged cells such that the biologically active moiety will be expressed when the promoter is turned-on by the administration, oral or otherwise, of a drug molecule such as tetracycline. The method used for genetic modification could be inserting a transgene with one or more viral vectors. In addition, the genetic modification could be performed by nonviral physical methods such as microinjection, electroporation, lipofection, and chemically-mediated transfection with calcium phosphate or liposomes, and other methods known to one of skill in the art. The pharmacologically effective amount is defined by therapeutic indices or responses appropriate to the disease state or condition that is being treated.