WO2000054786A1 - Methods and compositions for tolerizing hosts for long term survival of tissue transplants - Google Patents

Abstract

The invention provides encapsulated cells and methods for the tolerization of a graft recipient host. Encapsulated cells tolerize the host. Concurrent administration of an immunosuppressive agent allows successful transplantation of a desired graft. After tolerization, a graft can be transplanted in the desired location without the sustained use of an immunosuppression or immunotherapy. Thus, the invention provides a method for transplanting genetically modified xenogeneic myoblasts in a peripheral immunoreactive site while ensuring their long-term survival. Through a combination of encapsulation techniques and the use of selected immunosuppressors, a variety of different xenogeneic cell types can now be transplanted.

Description

METHODS AND COMPOSITIONS FOR TOLERIZING HOSTS FOR LONG TERM

SURVIVAL OF TISSUE TRANSPLANTS

TECHNICAL FIELD OF THE INVENTION

This invention relates generally to tissue transplantation. More particularly, the invention relates to methods and compositions for tolerizing a host mammal for long term survival of xenogeneic or allogeneic tissue subsequently transplanted into the host.

BACKGROUND OF THE INVENTION

The transplantation of encapsulated cells that have been genetically modified to secrete therapeutic proteins has been used for the treatment of several human diseases including anemia, diabetes, and multiple neurodegenerative disorders. The long-term delivery of therapeutic molecules either locally or systemically via the transplantation of encapsulated cells circumvents many of the problems encountered with conventional techniques of delivering pharmaceuticals. Cell encapsulation allows delivery of engineered cells to isolated locations, such as the brain parenchyma, while eliminating the need for the repeated injections normally required to maintain drug concentration within its therapeutic window. Isolating cells by enclosing them within a semipermeable polymer membrane prevents the cell contact- mediated death of donor cells following transplantation into a recipient. This physical barrier does not, however, protect encapsulated cells from destruction by other immune defense mechanisms that do not require cell contact.

Transplantation of organs and tissue in an allogeneic setting has become a common medical practice with the recent developments in immunomodulatory drugs. Unfortunately, long term immunosuppression is undesirable for the host, as immunosuppression renders the host immunocompromised for extended periods of time. On the other hand, long term survival of many tissue and organ transplants in the absence of immunosuppression has been notoriously difficult to achieve, especially when the tissue is from an allogeneic source to the host. Several attempts to implant encapsulated cells into hosts without their destruction by immune defense mechanisms have been made. Encapsulated C₂C₁₂ cells secreting the neurotrophic factor CNTF showed a sustained viability in the rat following a 3 months residence in the intrathecal space (Deglon et al, 1 Hum. Gene Ther. 2135 (1996)). Also, clinical trials for the treatment of cancer pain and amyotrophic lateral sclerosis have been done using encapsulated xenogeneic tissue in the Central nervous system (CNS) (Buchser et al, 85 Anesthesiology 1005 (1996), Aebischer et al, 2 Nat. Med. 696 (1996)). Both trials demonstrated extended graft viability as long as 6 months following implantation in the absence of immunosuppression. This prolonged survival in the central nervous system is favored by its immunoprivileged status, because of the presence of the blood-brain barrier and the absence of conventional lymphatic drainage. However, the subcutaneous site (not an immunoprivileged area) has until now been a difficult challenge for transplantation.

SUMM RY OF THE INVENTION

This invention provides methods and compositions for tolerizing mammals (preferably human patients) to minimize rejection of the transplanted tissue and to enhance the long-term survival of the transplanted tissue. In one embodiment, encapsulated cells are implanted into the recipient host. The encapsulated cell implant can be retrieved when necessary or desired. The cells can be either xenogeneic or allogeneic to the host. The host is concurrently immunosuppressed transiently (for at least 1 week, preferably at least 2 weeks or 3 weeks, and most preferably for at least 4 weeks). The encapsulated cell devices is then removed, and a second transplant from the same cell or tissue source as the first encapsulated cell device is implanted into the host, according to known techniques. The second implant is preferably an encapsulated tissue or cell implant (in one or more encapsulation devices).

The invention demonstrates that host tolerance can be developed in the subcutaneous tissue to encapsulated xenogeneic cells following a transient immunosuppression. This has direct clinical relevance in the field of cell therapy, because capsules can be replaced without additional immunosuppression, thus facilitating long-term cell-based therapies. Long-term treatments are facilitated, because devices can be replaced without supplemental immunosuppression following encapsulated cell senescence. The safety aspect of encapsulation is also preserved, because a recipient host tolerized to encapsulated xenografts maintains the ability of rejecting unencapsulated xenogeneic cells in the rare event that cells escape from the device. BRIEF DESCRIPTION OF THE FIGURES FIG. 1 is a graphical representation of the hematocrit levels of C3H mice, DBA/2J mice and Fischer rats that were implanted with mEpo-secreting capsules. FIG. 2 is a comparison of the hematocrit in control Fischer rats, inmunosuppressed

Fischer rats, Rnu nude rats implanted with mEpo-secreting capsules.

FIG. 3 is a schematic diagram of the immunosuppression protocol of Fischer rats with FK506. The schematic diagram shows FK506 dosing regimes used for the induction of long- term acceptance of encapsulated C1C12 mEpo cells in xenogeneic rat recipients (a) 1 week, (b) 2 weeks, and 4 weeks immunosuppression with FK506 (hatched bar: administration of

FK506 (1 mg/kg/day); * acquired tolerance was evaluated by implanting a second capsule in a separate set of experiments).

FIG. 4 is a set of graphical representations of the long-term acceptance of encapsulated C₂C₁₂ mEpo cells in Fischer rats immunosuppressed for lweek (FIG. 4A), 2 weeks (FIG. 4B), and 4 weeks (FIG. 4C).

FIG. 5 is a graphical representation of the long-term acceptance of encapsulated C₂C_l2 mEpo cells in Fischer rats that were immunosuppressed for 2 weeks.

FIG. 6 is a graph showing that C3H mice, DB A/2J mice and Fischer rats were implanted with 280 kDa cut-off capsules containing C₂C₁₂ mEpo cells. The hematocrit was monitored for 35 days, at which point all animals were explanted. Data is presented as mean ±

SEM. (C3H mice: n=10; DBA 2J mice: n=4; Fischer rats: n=l0)

FIG. 7 is a graph showing that Rnu nude rats, Fischer rats immunosuppressed with FK506 (1 mg/kg/day), and control Fischer rats were implanted with 280 kDa cut-off capsules containing C₂C_π mEpo myoblasts. The hematocrit was followed for 4 weeks following implantation, at which point the capsules were explanted and analyzed. Data is presented as mean ± SEM. (Nude rats: n=8 from day 0 to day 14 and n=7 from day 14 to day 28; 1 month treatment with FK506: n=5; control Fischer rats: n=5)

FIG. 8 is a graph showing that immunocompetent C3H mice and Fischer rats were implanted with mEpo secreting capsules having membranes with a molecular weight cut-off of 32 kDa. The progression of the hematocrit was followed for a period of 35 days. Results previously obtained with membranes having a 280 kDa cut-off are shown for comparison. Data is presented as mean ± SEM. (C3H mice with 280 kDa membrane: n=5; C3H mice with 32 kDa membrane: n=6; Fischer rats with 280 kDa membrane: n=10; Fischer rats with 32 kDa membrane: n=5)

FIG. 9 is a graph showing that Fischer rats were transiently immunosuppressed for either 1, 2, or 4 weeks with FK506 (1 mg/kg/day) following implantation with mEpo secreting capsules (280 kDa cut-off). Animals in each group increased their hematocrits and maintained these elevated levels for a period of 13 weeks. Data is presented as mean ± SEM. (1 week treatments: n=5 from day 0 to day 63 and n=3 from day 63 to day 91 ; 2 week treatments: n=5 from day 0 to day 63 and n=3 from day 63 to day 91 ; 1 month treatments: n=5 from day 0 to day 14, n-4 from day 14 to day 63, and n=2 from day 63 to day 91). FIG. 10 is a graph showing that Fischer rats were injected intramuscularly with mEpo secreting C₂C₁₂ myoblasts and immunosuppressed for 4 weeks with FK506 (1 mg/kg/day).

Control Fischer rats did not receive immunosuppression. Data is presented as mean ± SEM.

(FK506 treated Fischer rats: n=5; control Fischer rats: n=4)

FIG. 11 is a graph showing that animals were implanted with capsules containing C₂C₁₂ mEpo cells at week 0. On week 9, animals were challenged with an additional implant containing C₂C₁₂ mEpo cells (n=6). Both implants lead to only transient elevation of animal hematocrit.

FIG. 12 is a set of graphs showing that animals were initially treated with FK506 for either (FIG. 12A) 1 week or (FIG. 12B) 4 weeks. Primary implants containing C₂C₁₂ mEpo cells were performed on week 0. On week 9, animals were challenged with a second implant containing C,C₁₂ mEpo cells. Four weeks later, all devices were removed and analyzed (n=8).

On week 19, animals from each group were intramuscularly injected with unencapsulated

C₂C₁₂ mEpo cells (n=6). Hatched bar = administration of FK506 (1 mg/kg/day).

Reimplantation of encapsulated C₂C₁₂ mEpo cells lead to sustained elevation of animal hematocrit whereas transplantation of unencapsulated cells did not.

FIG. 13 is a graph showing that long-term tolerance to encapsulated xenogeneic myoblasts. Four Fischer rats were tolerized to capsules containing C₂C₁₂ mEpo myoblasts by treating 4 weeks with FK506. On week 13, all devices were explanted, and the hematocrit eventually returned to pre-implant levels. Seven months later (week 41 ), these animals were reimplanted with encapsulated C₂C₁₂ mEpo myoblasts leading to an increase in hematocrit.

*On week 9, 2 of these animals received a second implant.

FIG. 14 is a set of graphs showing that antigens implicated in rejection of encapsulated xenogeneic mEpo secreting myoblasts appear to be molecules inherent to myoblasts. Fischer rats were implanted with capsules containing unmodified C₂C₁₂ myoblasts and either (FIG. 14A) left untreated or (FIG. 14B) immunosuppressed 4 weeks with FK506. After a 4 week residence in vivo, all devices were explanted. On week 9, animals received a second capsule containing C₂C₁₂ mEpo myoblasts, which was retrieved 4 weeks later (n=5). Only animals previously immunosuppressed showed a sustained elevated hematocrit.

DETAILED DESCRIPTION This invention provides methods and compositions for tolerizing of hosts for long term survival of allogeneic or xenogeneic tissue or cells. Methods of tolerization according to the invention demonstrate that immunomodulation can be used to prolong the survival of immunoisolated xenografts. "Immunoisolation" as used herein is a technique in which engineered cells are enclosed within implantable polymeric capsules formed by permiselective membranes. Immunoisolation prevents the cell-to-cell contact between host and implanted tissues, eliminating direct xenorecognition, as the membranes used have a pore size that permits the diffusion of nutrients and bioactive molecules, while reducing passage of antibodies and complement molecules, as described below.

The methods of the invention include methods for improving survival of a transplant in a recipient host. The methods of the invention comprise implanting one or more encapsulated cell devices into the recipient host. An immunosuppressive agent is administered to the recipient host in an amount effective to permit survival of the encapsulated tissue or cell transplant for a period of at least one week. By using the method of the invention, the transplant has improved survival in the recipient host compared to a recipient host that did not receive an administration of an immunosuppressive agent in an amount effective to permit survival of the encapsulated tissue or cell transplant for a period of at least one week. Alternatively, the transplant has improved survival compared to a recipient host that did not receive an implantation of one or more encapsulated cell devices containing tissue or cells that are allogeneic or xenogeneic to the recipient host, followed by an administration of an immunosuppressive agent in an amount effective to permit survival of the encapsulated tissue or cell transplant for a period of at least one week. In one embodiment, the encapsulated cell devices are implanted according to known techniques, and contain cells or tissue that is either allogeneic or xenogeneic to the recipient host. Concurrently with the implantation of the encapsulated cell device, the recipient host is administered an immunosuppressive agent in an amount effective to permit survival of the encapsulated tissue or cell transplant for a period of at least 1 week. The encapsulated cell devices are then removed, and a second transplant from the same cell or tissue source as the first encapsulated cell devices is implanted into the host, according to known techniques (see, or example, EXAMPLES 2, 6, and 7). Preferably, the second implant is encapsulated. In another embodiment, encapsulated xenogeneic cells stimulate an initial-immune response immediately following subcutaneous transplantation, as shown below. Transient immune blockade abrogates this indirect immune attack and allows encapsulated xenogeneic cells to survive indefinitely, or at least until senescence. Therefore, a graft of encapsulated xenogeneic or allogeneic cells, as described below, is a useful alternative to syngeneic graft transplantation.

Tolerization. The terms "tolerance" and "tolerization" encompass the ability to endure or be less responsive to a stimulus, such as an immune response to a transplant or graft, especially over a period of continued exposure. Thus, "tolerance" refers to the inhibition of a graft recipient host's immune response that would otherwise occur, e.g., in response to the introduction of a non-self MHC antigen into the recipient host. Tolerance can involve humoral, cellular, or both humoral and cellular responses.

The immunomodulatory or immunosuppressive agent used in the methods of the invention can be any suitable immunosuppressive agent known in the art. Two preferred agents are, e.g., cyclosporin A and FK506. The dosage for effective immunosuppression will vary according to the mammalian host, but typically the effective dosage for, e.g., human patients, are those that maintain the viability of the encapsulated cells, and are well known or could be routinely ascertained by one of ordinary skill in the art. Likewise, the immunosuppressive agent may be administered according to any suitable regimen known in the art (see, for example, EXAMPLES 3, 6, and 7). The mature adult immune system can be rendered tolerant to specific alloantigens using a number of different strategies. Many of these protocols require that a recipient receives a short-term treatment with an immunomodulator while being exposed to specific transplantation antigens. In addition to accepting a primary graft, a tolerized host should remain unresponsive to subsequent grafts of identical genetic background that are transplanted in the absence of immunosuppression. Such peripheral tolerance to allografts can be established when grafts are performed along with treatments that block T cell surface molecules such as CD4 and CD8 (Qin et al, 20 Eur. J. Immunol. 2737 (1990)), B7 (Tran et al, 159 J. Immunol. 2232 (1997)), and lymphocyte function-associated, antigen (LFA-1) (Isobe et al, 96 Circulation 2247 (1997)). Similarly, the immunosuppressors cyclosporine (Nagao et al, 33 Transplantation 31 (1982), Mottram et al, 50 Transplantation 1033 (1990)) and FK506 (Ochiai et al, 44 Transplantation 734 (1987), Inamura et al, 45 Transplantation 206 (1988)) have been used to establish a state of host unresponsiveness to allografts. Antigen specific tolerance using many of these techniques relies-on a newly generated population of

CD4+ T cells that suppress rejection by T cells and eventually recruit them to become tolerant as well (Qin et al, 259 Science 974 (1993)).

Cell Encapsulation. The invention provides a composition in which cells are encapsulated in an immunoisolatory capsule. An "immunoisolatory capsule" means that the capsule upon implantation into a host minimizes the deleterious effects of the host's immune system on the cells within its core.

Encapsulated cell therapy is based on the concept of isolating cells from a host's immune system by surrounding the cells with a semipermeable biocompatible material before implantation within the host. Cells are immunoisolated from the host by enclosing them within implantable polymeric capsules formed by a microporous membrane. This approach prevents the cell-to cell contact between host and implanted tissues, eliminating antigen recognition through direct presentation. The membranes used can also be tailored to control the diffusion of molecules, such as antibody and complement, based on their molecular weight (Lysaght et al, 56 J. Cell Biochem. 196 (1996), Colton, 14 Trends Biotechnol. 158 (1996)). Using encapsulation techniques, cells can be transplanted into a host without immune rejection, either with or without use of immunosuppressive drugs. Useful biocompatible polymer capsules usually contain a core which contains a cell or cells, either suspended in a liquid medium or immobilized within an immobilizing matrix, and a surrounding or peripheral region of permselective matrix or membrane ("jacket") which does not contain isolated cells, which is biocompatible, and which is sufficient to protect isolated cells if present in the core from detrimental immunological attack. Encapsulation hinders elements of the immune system from entering the capsule, thereby protecting the encapsulated cells from immune destruction. The semipermeable nature of the capsule membrane also permits the biologically active molecule of interest to easily diffuse from the capsule into the surrounding host tissue ( ee, EXAMPLES 1,6, and 7).

The capsule is made from a biocompatible material. A "biocompatible material" is a material that, after implantation in a host, does not elicit a detrimental host response sufficient to result in the rejection of the capsule or to render it inoperable, for example through degradation. The biocompatible material is relatively impermeable to large molecules, such as components of the host's immune system, but is permeable to small molecules, such as insulin, growth factors, and nutrients, while metabolic waste to be removed. A variety of biocompatible materials are suitable for delivery of growth factors by the composition of the invention. Numerous biocompatible materials are known, having various outer surface morphologies and other mechanical and structural characteristics. Preferably the capsule of this invention will be similar to those described by PCT International patent applications WO 92/19195 or WO 95/05452, incorporated by reference; or United States patents 5,639,275; 5,653,975; 4,892,538; 5,156,844; 5,283,187; or 5,550,050, incorporated by reference. Such capsules will allow for the passage of metabolites, nutrients and therapeutic substances while minimizing the detrimental effects of the host immune system. Components of the biocompatible material may include a surrounding semipermeable membrane and the internal cell-supporting scaffolding, preferably, the transformed cells are seeded onto the scaffolding, which is encapsulated by the permselective membrane. The filamentous cell-supporting scaffold may be made from any biocompatible material selected from the group consisting of acrylic, polyester, polyethylene, polypropylene polyacetonitrile, polyethylene teraphthalate, nylon, polyamides, polyurethanes, polybutester, silk, cotton, chitin, carbon, or biocompatible metals. Also, bonded fiber structures can be used for cell implantation (United States patent 5,512,600, incorporated by reference). Further, biodegradable polymers can be use as scaffolds for hepatocytes and pancreatic cells, as reviewed by Cima et al, 38 Biotech. Bioeng. 145-58

(1991). Biodegradable polymers include those comprised of poly(lactic acid) PLA, poly(lactic- coglycolic acid) PLGA, and poly(glycolic acid) PGA and their equivalents. Foam scaffolds have been used to provide surfaces onto which transplanted cells may adhere. PCT International patent application 98/05304, incorporated by reference. Woven mesh tubes have been used as vascular grafts. PCT International patent application WO 99/52573, incorporated by reference. Additionally, the core can be composed of an immobilizing matrix formed from a hydrogel, which stabilizes the position of the cells. A hydrogel is a 3-dimensional network of cross-linked hydrophilic polymers in the form of a gel, substantially composed of water. Various polymers and polymer blends can be used to manufacture the surrounding semipermeable membrane, including polyacrylates (including acrylic copolymers), polyvinylidenes, polyvinyl chloride copolymers, polyurethanes, polystyrenes, polyamides, cellulose acetates, cellulose nitrates, polysulfones (including polyether sulfones), polyphosphazenes, polyacrylonitriles, poly(acrylonitrile/covinyl chloride), as well as derivatives, copolymers and mixtures thereof. Preferably, the surrounding semipermeable membrane is a biocompatible semipermeable hollow fiber membrane. Such membranes, and methods of making them are disclosed by United States patents 5,284,761 and 5,158,881, incorporated by reference. The surrounding semipermeable membrane is formed from a polyether sulfone hollow fiber, such as those described by United States-patent 4,976,859 or

4,968,733, incorporated by reference. An alternate surrounding semipermeable membrane material is poly(acrylonitrile/covinyl chloride).

The capsule can be any configuration appropriate for maintaining biological activity and providing access for delivery of the product or function, including for example, cylindrical, rectangular, disk-shaped, patch-shaped, ovoid, stellate, or spherical. Moreover, the capsule can be coiled or wrapped into a mesh-like or nested structure. If the capsule is to be retrieved after it is implanted, configurations which tend to lead to migration of the capsules from the site of implantation, such as spherical capsules small enough to travel in the recipient host's blood vessels, are not preferred. Certain shapes, such as rectangles, patches, disks, cylinders, and flat sheets offer greater structural integrity and are preferable where retrieval is desired.

When macrocapsules are used, preferably between 10³ and 10⁸ cells are encapsulated, most preferably 10⁵ to 10⁷ cells are encapsulated in each device. Dosage may be controlled by implanting a fewer or greater number of capsules, preferably between 1 and 10 capsules per patient.

The scaffolding may be coated with extracellular matrix (ECM) molecules. Suitable examples of ECM molecules include, for example, collagen, laminin, and fibronectin. The surface of the scaffolding may also be modified by treating with plasma irradiation to impart charge to enhance adhesion of cells. Any suitable method of sealing the capsules may be used, including the use of polymer adhesives or crimping, knotting and heat sealing. In addition, any suitable "dry" sealing method can also be used, as described, e.g., in United States patent 5,653,687.

The encapsulated cell devices are implanted according to known techniques. Many implantation sites are contemplated for the devices and methods of this invention. These implantation sites include, but are not limited to, the central nervous system, including the brain, spinal cord, and aqueous and vitreous humors of the eye.

Cell-based therapies have been developed that use cell encapsulation technology to implant primary tissue and established cell lines. Cell lines are most appropriate for use with encapsulated cell technology. Cell lines offer the advantage of unlimited supply, permitting scale-up and establishment of cell banks for potential clinical applications. Using xenogeneic cells provides the additional safety that cell rejection will occur by the host immune system in the case of capsule rupture. It is important to note that cell lines can be screened prior to use for the presence of pathogens which could pose a threat to human recipients.

The cells of the invention can be native or recombinant cells. A "recombinant " cell is a cell or progeny of a cell into which has been introduced, by means of recombinant genetic techniques, any desired polynucleotide. The terms "tissue", "cell", and "cells" also encompasses any types of transplantable or implantable tissue or cells from a donor other than the recipient host that contains antigen presenting cells (APC's). The donor tissue being used in the invention can be any one of a wide variety of tissues, for example, soft tissue such as the amniotic membrane of a newborn, bone marrow, hematopoietic precursor cells, collagen, and bone protein to stimulate cartilage growth; organs such as skin, heart, liver, spleen, pancreas, thyroid lobe, lung, kidney, tubular organs (e.g., intestine, blood vessels, or esophagus); parts of organs, such as heart valves; and isolated cells or clusters of cells, such as islet cells of the pancreas or liver cells. The donor tissue or cells can be taken from any source, whether from cadavers or living donors. Examples of suitable donors include live animals such as laboratory animals, for example, dogs, cats, mice, rats, gerbils, guinea pigs, cows, primates, or human beings. Donors are preferably mammalian, including human beings. When both the donor of the graft and the host are human, they are preferably matched for HLA Class II antigens to as to improve histocompatibility. Human donors are preferably of the same or compatible major ABO blood group. In the field of islet transplantation, encapsulation of primary allogeneic and xenogeneic islets has been pursued to replace the function of the pancreas with a bioartificial equivalent (Lacy et al, 254 Science 1782 (1991), Sullivan et al, 252 Science 718 (1991)). Encapsulated genetically engineered cell lines have been used for systemic delivery of therapeutic proteins such as erythropoietin (Rinsch et al, 8 Hum. Gene Ther. 1881 (1997), Regulier et al, 5 Gene Ther. 1014 (1998)), human growth hormone (Chang et al, 4 Hum. Gene Ther. 433 (1993)) and factor IX (Brauker et al, 9 Hum. Gene Ther. 879 (1998)), as well as for the localized release of the neurotrophic factors CNTF (Aebischer et al, 1 Hum. Gene Ther. 851 (1996)) and GDNF (Tseng et al, 17 J. Neurosci. 325 (1997)) in the central nervous system. Systemic administration of secreted molecules can be achieved by implanting devices subcutaneously or in the peritoneal cavity.

In the central nervous system, encapsulated xenogeneic cells have displayed extended viability, surviving at least 6 months (Aebischer et al, 1 Hum Gene Ther 851 (1996)). Encapsulated xenogeneic primary islets have displayed long-term survival in the absence of immunosuppression when transplanted intraperitoneally across a variety of species combinations. Encapsulated xenogeneic primary islets implanted intraperitoneally have successfully provided long-term correction of glucose levels in various animal recipients rendered diabetic (Lacy et al, 254 Science 1782 (1991), Lanza et al, 88 Proc. Natl. Acad. Sci. USA 1100 (1991), Sun et al, 98 J. Clin. Invest. 1417 (1996)).

On the other hand, other cell types are more prone to immune attack when transplanted across xenogeneic barriers, despite the presence of a protective polymer membrane. When immunoisolated xenogeneic cells are implanted subcutaneously or intraperitoneally, they may be rejected by the immune system via a mechanism not requiring direct cell contact (Loudovaris et al, 24 Transplant Proc 2291 (1992), Loudovaris et al, 24 Transplant Proc.

2938 (1992), Loudovaris et al, 61 Transplantation 1678 (1996)). Studies have shown that encapsulated cells, including COS cells, primary fetal lung tissue (and even NIT-1 islet β- cells) are rejected by certain xenogeneic hosts following implantation either intraperitoneally or in the subcutaneous site (Loudovaris et al, 24 Transplant Proc 2291 (1992), Loudovaris et al, 24 Transplant Proc. 2938 (1992), Loudovaris et al, 61 Transplantation 1678 (1996),

Brauker et al, 61 Transplantation 1671 (1996)). In one of these models, the African green monkey kidney COS-7 cell line was encapsulated and transplanted in the ovarian fat pad of mice recipients (Loudovaris et al, 24 Transplant Proc 2291 (1992)). Depleting CD4+ T cells but not CD8+ T cells prevented cell rejection by these mice during the 3 weeks considered, indicating that encapsulated xenograft destruction is mediated by a process reliant on CD4+ T cells. Xenoantigens either shed or secreted by the encapsulated xenografts stimulated a host immune response by an indirect antigen presentation (Loudovaris et al, 24 Transplant Proc 2291 (1992)).

In only one publication have encapsulated xenogeneic cells demonstrated prolonged survival in the subcutaneous site, and this involved primary islets using a concordant rat to mice model (Lacy et al, 254 Science 1782 (1991)). This disparity between the survival of islets and other cell types following encapsulation and xenotransplantation is likely due partly to the low antigenicity of islets. By contrast, the present invention can use cells that are highly antigenic.

The xenotransplantation of cells that are highly antigenic is useful, because xenotransplantation now becomes a better therapeutic option. Xenotransplantation offers a method for overcoming the limitations imposed by an insufficient supply of human tissues and organs for transplantation. The terms "transplant" and variations thereof refers to the insertion of a graft into a recipient host, whether the transplantation is syngeneic (where the donor and recipient host are genetically identical), allogeneic (where the donor and recipient host are of different genetic origins but of the same species), or xenogeneic (where the donor and recipient host are from different species).

The term "graft" as used herein refers to biological material derived from a donor for transplantation into a recipient host. The term "recipient host" as used herein refers to any compatible transplant host. By "compatible" is meant a host that will accept the donated graft according to the present invention. Examples of potentially useful recipient hosts includes animals, preferably mammals such as farm animals, for example, horses, cows or sheep; household pets, for example, dogs or cats; laboratory animals, such as mice, rats, gerbils or guinea pigs; or primates, for example, apes or human beings. Useful recipient hosts can also include aquatic animals, who often live under conditions of lower oxygen tensions than terrestrial animals. Preferably, the recipient host is a human being.

In EXAMPLES 6 and 7, murine C,C_]2 myoblasts were used for encapsulation and subcutaneous implantation. C₂C_I2 myoblasts engineered to secrete murine erythropoietin (Epo; the primary regulator of erythrocyte homeostasis) were used to enable in vivo monitoring of xenograft survival via fluctuations in the hematocrit. Using microporous membranes, C₂C₁₂ mouse Epo cells were encapsulated and implanted subcutaneously to compare the cell viability in syngeneic. allogeneic and xenogeneic models. To determine if a tolerance to encapsulated syngeneic, allogeneic, or xenogeneic grafts could be induced following short-term immunosuppressive therapy, Fischer rats were administered FK506 for periods of 1, 2 and 4 weeks after which their hematocrits were monitored until 3 months post-implantation. Animals increased their hematocrits over 70% and sustained these levels for the 3 months, independent of the duration of treatment with FK506.

Immune response. The survival of a xenograft depends on the embodiment of the graft and the phylogenetic distance between donor and recipient host. Xenogeneic grafts that provide their own vasculature are immediately subjected to "hyperacute rejection" upon their reperfusion (see e.g., Platt, 16 Current Reviews in Immunology 331-358 (1996). See also, Platt, 8 Current Opinion in Immunology 721-28 (1996)). The binding of xenoreactive antibodies to Gala 1-3 Gal antigens on the endothelial surface of xenograft vessels activates the complement cascade, resulting in the destruction of the transplanted organ (id. See also,

Cooper et al, 1 Transplantation Immunology 198-205 (1993)). As xenografts consisting of free tissues or isolated cells do not possess their own functional vasculature, their outcome depends on a second type of response, which involves cellular rejection. Two separate mechanisms based on direct versus indirect T cell recognition have been considered. Direct xenoantigen recognition requires physical contact between helper T cells and xenogeneic antigen-presenting cells (APC), while indirect xenoantigen recognition occurs when helper T cells respond to xenogeneic peptides presented on host APC. In both cases, the activation of cytotoxic T lymphocytes (CTL) by helper T cells induces a CTL-mediated mechanism of rejection. In an alternative scenario, graft rejection can be mediated by a non-classical CTL- independent pathway involving either natural killer cells or macrophages.

Encapsulated xenografts elicit an immune response by an indirect presentation of shed or secreted antigens to host T cells. A developed tolerance to xenoantigens must therefore occur through the indirect pathway. Tolerance to minor antigens can be established through indirect presentation alone (Davies et al, 157 J. Immunol. 529 (1996)), suggesting the same mechanism may apply to other types of antigens, including xenoantigens. Antigens can either shed or secreted by the encapsulated xenografts diffuse through the immunoisolating membrane, leading to activation of the host immune system (Loudovaris et al, 24 Transplant Proc. 2291 (1992), Brauker et al, 61 Transplantation 1671 (1996), Weber et al., 49 Transplantation 396 (1990)). The xenoantigens released by encapsulated xenografts may be proteins naturally produced by these cells, but which show a sufficient difference with their corresponding homologue in the host to initiate an immune response. Once they diffuse outside the capsule membrane, xenoantigens are taken up by antigen presenting cells which in turn stimulate CD4+ T cells to mount an immune attack (Loudovaris et al, 24 Transplant Proc 2291 (1992)). A localized migration of lymphocytes, macrophages, granulocytes and multinucleate giant cells develops around the device, leading to the destruction of the encapsulated xenografts (Loudovaris et al, 24 Transplant Proc. 2291 (1992), Brauker et al, 61 Transplantation 1671 (1996), Weber et al, 49 Transplantation 396 (1990)). The death of the enclosed xenogeneic cells is likely due to the combined effect of locally released immune effectors as well as metabolic stress.

Although the immune effectors involved in the destruction of encapsulated cells have yet to be identified, several test provide conclusive evidence that antibody/complement- mediated lysis is not implicated in this process. Decreasing the molecular weight cut-off of the membrane from 280 kDa to 32 kDa, so as to reduce the entry of antibody and complement, showed no improvement in the survival of the encapsulated C₂C₁₂ mEpo cells transplanted in immunocompetent Fischer rats, indicating that released xenoantigens must be small in size. In vitro tests have shown that antibodies and complement are ineffective in killing a non- vascularized 3-dimensional tissue extruded out of capsules, presumably because of their low capacity to diffuse into such a 3-dimensional structure. Loudovaris et al. (24 Transplant Proc. 2938 (1992)) have also reported that mice depleted for the complement components C3 and C5 are able to reject encapsulated xenogeneic cells. Taken together, these results show that smaller cytotoxic molecules capable of diffusing into the capsule are responsible for the death of the xenogeneic cells.

The activation of immune response is clinically important, because even a local inflammation caused by an immune response to implanted encapsulated cells can be serious for the host. For example, when animals are immunosuppressed only 1 week, the second capsule provokes the rejection of cells in the first implant (see, EXAMPLE 7). As both devices are positioned at different locations, with inflammation localized around each capsule exterior, the immune reaction induced appears to be rather specific against the xenogeneic cells. If the immune reaction was a general inflammatory response following transplantation, the survival of encapsulated cells in the first implant should not be affected. This shows that the induced tolerance threshold to xenoantigens is dependent on the persistence of xenoantigens in an immunosuppressed background. Studies in allotransplantation have highlighted the importance of antigen persistence on the efficacy of T cell tolerization (Scully et al, 24 Eur. J. Immunol. 2383 (1994), Ehl et al, 4 Nature Med. 1015 (1998)), suggesting that this may similarly apply to xenotransplantation.

While T cells may remain unresponsive to indirect xenoantigen presentation, other immune pathways which function through direct contact, including natural killer cells, macrophages and complement can efficiently act to eliminate unencapsulated xenografts. This is important from a safety point of view, for in the unlikely event of capsule rupture, xenogeneic cells lines would be quickly rejected by the hosts immune system (see, EXAMPLE 6).

Equivalents. The details of one or more embodiments of the invention have been set forth in the accompanying description above. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, the preferred methods and materials are now described. Other features, objects, and advantages of the invention will be apparent from the description and from the claims. In the specification and the appended claims, the singular forms include plural referents unless the context clearly dictates otherwise. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. All patents and publications cited in this specification are incorporated by reference.

The following EXAMPLES are presented in order to more fully illustrate the preferred embodiments of the invention. These EXAMPLES should in no way be construed as limiting the scope of the invention, as defined by the appended claims.

EXAMPLE 1 GENERATION OF THE C₂C₁₂ mEpo SECRETING CELL LINE

The murine C₂C₁₂ myoblast cell line was used to examine the criteria for the survival of encapsulated xenogeneic cells in the subcutaneous site of a rat recipient host. Mouse C₂C_]2 myoblasts obtained from the American Type Culture Collection (ATCC; CRL 1772, Rockville, MD), were transfected with the pPI-mEpo-ND plasmid (Regulier et al., 5 Gene Ther. 1014 (1998)) using calcium phosphate precipitation (mammalian transfection kit,

Stratagene, La Jolla, CA, USA). The murine C₂C₁₂myoblast cell line is able to secrete high levels of recombinant proteins over prolonged periods (Regulier et al., 5 Gene Ther. 1014 (1998)) and can be induced to differentiate into a post-mitotic state when exposed to low- serum containing medium (Yaffe & Saxel, 270 Nature 725 (1977)). Stably transfected C₂C₁₂ cells are capable of secreting therapeutic levels of recombinant proteins, such as erythropoietin

(Epo), in both a constitutive and regulated manner (Regulier et al., 5 Gene Ther. 1014 (1998), Rinsch et al., 8 Hum. Gene Ther. 1881 (1997)). In vivo, encapsulated C₂C₁₂ myoblasts engineered to secrete mEpo have shown long-term viability subcutaneously in both syngeneic and allogeneic animal models (Regulier et al., 5 Gene Ther. 1014 (1998)). The cells were selected for 2 weeks in 0.8 mg/ml G418. Subsequently, selected cells were incubated with increasing concentrations of methotrexate (1-200 M) over 6 weeks to amplify the copy number of the integrated plasmids. Cells were grown in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal calf serum (FCS), 2 mM L-glutamine, 4.5 g/1 glucose, 100 U/ml penicillin and 100 U/ml streptomycin.

C₂C₁₂ cells were harvested using a 0.125% trypsin-EDTA solution prepared in modified Puck's saline and were diluted with DMEM in order to achieve a suspension of 1 x 10⁵ cells/μl. The suspension was then injected into microporous polyethersulfone (PES) hollow fibers (OD = 720 μm; ID = 524 μm; cutoff = 32 kDa and 280 kDa) (Akzo Nobel Faser AG, Wupperthal, GERMANY). One end of the capsule was sealed through photopolymerisation

(460 nm) of acrylate based glue (Luxtrak LCM 23, Ablestik, Electronic Materials & Adhesives, Rancho Dominguez, CA, USA) while the other end was heat-sealed after cell injection. One cm long capsules loaded with approximately 6 x 10⁵ C₂C₁₂mEpo cells were implanted subcutaneously on the dorsal flank of mice. Capsules implanted into rats contained 1 x 10⁶ cells and measured 2 cm. Their structure was reinforced with an internal titanium coil to prevent kinking. Encapsulated cells were differentiated for 3 days in DMEM containing 10%) Prolifix (Bio Media, Boussens, FRANCE) and subsequently returned to DMEM containing 10% FCS for 4 days in an incubator at 37°C and 5% CO₂.

The secretion of mEpo from encapsulated mEpo cells was measured both pre- implantation and following explant by incubating the capsules for 1 hr in 1 ml DMEM 10%)

FCS. Epo levels in the conditioned media were measured using an enzyme linked immunosorbent assay (ELISA) (Quantikine IVD, R&D systems, Minneapolis, MI, USA). Cross-reaction of the kit allowed detection of mEpo in culture supernatants. The mEpo secretion was corrected using the Epo-dependent murine DaE7 cell line as a test of hormone bioactivity (Sakaguchi et al, 15(10) Exp. Hematol. 1028-34 (1987)). The correlation function was 7.637x - 0.005 (x = ELISA measured value).

The average secretion of the resulting C₂C₁₂ mEpo cell line was 51 IU Epo/10⁶ cells/day, as determined using a human Epo ELISA test which cross-reacted with mEpo. This relative secretion value was corrected according to bioactivity tests performed on the murine DaE7 cell line (Sakaguchi et al, 15(10) Exp Hematol. 1028-34 (1987)). The normalized value of mEpo secretion was calculated to be 390 IU Epo/10⁶ cells/day. EXAMPLE 2 CAPSULE IMPLANTATION

C3H and DBA 2J mice (Iffa Credo, Saint-Germain sur l'Abresle, FRANCE) were chosen for syngeneic and allogeneic transplantation. Fischer rats (Iffa Credo, FRANCE) and

Rnu rats (Charles River, Sulzfeld, GERMANY) were used as xenogeneic transplant recipient hosts. For surgical implantation animals were anesthetized by inhalation of isoflurane (Forene, Abbott Laboratories, Cham, SWITZERLAND). Capsules were implanted subcutaneously in the dorsal flank of the animals by means of a trocar (Abbocath-T 16 G, Abbott Laboratories, Cham, SWITZERLAND). The entry site in the skin was closed using a nonresorbable suture

(Dermalon 5-0). Upon recovery, the animals were returned to the animal care facility, where they had access to food and water ad libitum.

Immunocompetent Fischer rats consistently rejected encapsulated C₂C₁₂ mEpo secreting cells implanted subcutaneously. Nude rats subcutaneously implanted with encapsulated C₂C_]2 mEpo cells maintained elevated hematocrits and capsules explanted after 1 month continued to secrete high levels of erythropoietin. These observations indicated that the immune system is responsible for the destruction of encapsulated xenogeneic cells at the subcutaneous site.

For intramuscular injections of mEpo secreting myoblasts, C₂C_]2 mEpo cells were harvested from confluent cultures by trypsinization. Subsequently, they were washed 3x in

Hank's balanced salt solution (HBSS) and resuspended at a concentration of 10⁵ cells/μl HBSS. Under general anesthesia, a small incision was made in the hind limb skin to expose the underlying muscles. Using a Hamilton syringe with a 26S-gauge needle, 5 million myoblasts were injected in the tibialis cranialis and gastrocnemius muscles. The incision was then closed with Dermalon 5-0.

To compare the survival in syngeneic, allogeneic, and xenogeneic models, encapsulated C₂C,₂ cells secreting mEpo were transplanted subcutaneously in C3H mice, DBA/2J mice and Fischer rats. C3H and DBA/2J mice were each implanted with a single 1 cm long capsule. Prior to implantation, encapsulated cells secreted 18.1 ± 9.62 IU/24hrs of mEpo (C3H mice) and 35.3 ±. 8.42 IU/day (DBA/2J mice) (TABLE 1). Both mice strains experienced a significant increase in their hematocrit and maintained these elevated levels for the 5 week trial period. At this point, C3H and DBA/2J mice had attained hematocrits of 79.7 ± 4.18%o and 89.5 ± 3.7%>, respectively (FIG. 1). At explant, mEpo secretion had decreased to 24%o for C3H mice and 32% for DBA/2J mice, relative to pre-implantation levels. Histological analysis revealed living myoblasts in most explanted devices, with the occasional presence of multinucleated myotubes. In a few instances, necrosis was observed at the capsule core. The biocompatibility of the capsules appeared to be excellent in the subcutaneous site, as they were surrounded by an extensive neovascular network with only a thin layer of fibroblasts adhering to membrane.

Fischer rats were implanted with reinforced capsules 2 cm long containing C₂C₁₂ mEpo cells. The mean secretion pre-implantation was 23.3 ± 9.07 IU mEpo/day. (TABLE 1). In vivo, the delivered Epo induced a significant, but only a transient increase in the hematocrit. Two weeks following implantation, the hematocrits rose to a high of 65.9 ± 2.38%_> and then progressively decreased to pre-implantation levels (FIG. 1). At explant, histological examination of the devices revealed that none of the capsules contained viable myoblasts. The pericapsular tissue was composed of a thick fibroblast layer infiltrated principally by lymphocytes and neutrophil granulocytes. This suggested that the decline in the hematocrit levels was due to a gradual, immune mediated destruction of the encapsulated cells following implantation.

Consequently, immunocompetent rats mounted a lethal immune response to encapsulated xenogeneic C₂C_I2 mEpo cells. Pericapsular tissue at explant was characterized by a thick, vascular fibrotic layer having a massive infiltration of lymphocytes. This localized presence of activated lymphocytes most likely induced the death of the encapsulated cells by the release of high concentrations of cytokines into the area.

EXAMPLE 3 IMMUNOSUPPRESSION

To examine the role of the immune system in the rejection of encapsulated xenogeneic myoblasts following transplantation, the response of nude rats to capsules containing C₂C₁₂ mEpo cells was tested. Nude rats possess a rnu autosomal recessive locus which provokes hairlessness and thymic aplasia, rendering them severely immune deficient (Hougen HS). Selected rats were pretreated for 3 days with FK506 (lmg/kg BW) (Prograf, Fujisawa GmbH, Mϋnchen, GERMANY). After transplantation, these animals were treated daily (5 days out of 7) for 1, 2 or 4 weeks at a dose of lmg/kg body weight. FK506 doses were injected i.m. into the quadriceps muscle, alternating daily between the left and right leg. Blood was drawn weekly from the tail vein into heparinized capillary tubes. The hematocrit was then measured by a standard microhematocrit method. At the end of the test period, the capsules were carefully explanted, fixed in Lang's fixative for 3 hours and dehydrated using alcohol in preparation for glycol-me hacrylate embedding (Leica Instruments GmbH, Nussloch, GERMANY). After retrieval of the titanium reinforcement, the capsules were cut at 5 μm thickness and stained with cresyl violet and hematoxylin eosin.

Eight Rnu rats were implanted following the protocol previously used with the Fisher rats. Three Fischer rats were implanted in parallel to serve as internal controls. Pre- implantation, cell containing devices secreted 29.9 ± 7.09 UI mEpo/day (TABLE 2). During the first week post-implantation, the hematocrit of nude rats increased in a manner similar to that of control rats, reaching 53.4 ± 3.2%. Beginning on day 14, a progressive and significant increase of the hematocrit was observed in nude rats (78.4 ± 3.69%> on day 28), whereas the hematocrit in control rats had already decreased to basal levels (54 ± 1.73 on day 28) (FIG. 2). On day 28, animals were sacrificed and capsules were explanted. C₂C₁₂mEpo capsules retrieved from Rnu rats secreted 20.8%> of their preimplantation levels or 6.9 ± 7.1 IU mEpo/day (TABLE 2). As expected, no detectable mEpo secretion was measured in capsules explanted from control Fischer rats.

Histological analysis of the devices retrieved from nude rats showed the presence of viable myoblasts. On the other hand, devices retrieved from control Fischer rats were filled with cellular debris. The morphology of the adherent tissue surrounding the explanted polyethersulfone membrane revealed extensive differences between the explants from both rat strains. In control immunocompetent rats, this tissue was poorly vascularized and consisted of several fibroblast layers profusely infiltrated with lymphocytes and neutrophil granulocytes. In contrast, the pericapsular tissue in nude rats was thinner, better vascularized and lymphocyte infiltration was rare. These results demonstrate a clear involvement of the immune system in the destruction of the xenografted encapsulated myoblasts.

Observing that the encapsulated xenogeneic C₂C₁₂ cells survived in the immune deficient Rnu rats, the immunosuppressor FK506 was administered to Fischer rats to evaluate if it could similarly protect the encapsulated C₂C₁₂ cells against immune mediated destruction. Seven Fischer rats were each implanted following the protocol previously described. The mean capsule secretion before implantation was 34.4 ± 7.99 UI mEpo/day (TABLE 3). Beginning 3 days prior to transplantation, 5 of Fischer rats were immunosuppressed with FK506 for the test period of 4 weeks. The other 2 non-immunosuppressed rats served as internal controls.

During the first 2 weeks, both the immunosuppressed and the control rats showed an upward trend in their hematocrits, as previously observed. A difference between the progression of the 2 groups became evident beginning on the third week post-implantation. At this point, immunosuppressed rats continued to increase their hematocrits, while the levels of the control rats began to decline. On the fourth week, rats treated with FK506 continued to maintain hematocrits above 64%, while untreated animals had decreased to pre-implantation levels (47.5±2.12%) (FIG. 2). Subsequently, capsules were explanted from the animals and assessed for their secretion of mEpo. Capsules retrieved from FK506 treated rats secreted on average 10.6 ± 5.59 UI mEpo/24 hr or 31% of pre-implantation levels. Epo secretion by capsules retrieved from untreated rats was undetectable, as expected. Immunosuppression by FK506 significantly prevented the death of encapsulated xenogeneic cells, with cell viability, as measured by Epo secretion, being equivalent to that previously observed in syngeneic and allogeneic transplant models. Predictably, histology revealed an excellent cellular morphology and viability within the capsules explanted from the immunosuppressed rats. The tissue reaction to capsules implanted in immunosuppressed rats resembled that observed in nude rats.

EXAMPLE 4

LONG-TERM ACCEPTANCE OF ENCAPSULATED C₂C₁₂mEpo CELLS BY FISCHER

RATS FOLLOWING TRANSIENT FK506 THERAPY

As continuous administration of FK506 permitted the survival of encapsulated cells in xenogeneic recipient hosts, short-term dosing regimens were evaluated for their efficacy on long-term encapsulated cell survival. After 3 days pretreatment with FK506, Fischer rats were implanted and subsequently immunosuppressed for 1, 2 or 4 weeks (FIG. 3) In the 3 treatment groups, a gradual increase in the hematocrit was observed, with levels reaching over 71% on week 9 (FIG. 4). At this time point a second mEpo capsule measuring 1 cm was implanted on the contralateral side of 2 rats in each group. Animals were followed by weekly hematocrit measurements for an additional 4 weeks at which point all capsules were explanted for analysis. Animal groups implanted with only 1 capsule and immunosuppressed with FK506 for

1 or 2 weeks showed a long-term tolerance to encapsulated C₂C₁₂ mEpo cells. These rats maintained an elevated hematocrit exceeding 65% throughout the 3 month trial period. (FIG. 4A, FIG. 4B) Following a residence time of 13 weeks, explanted devices continued to secrete a significant amount of mEpo (TABLE 4). The level of secretion at explant versus implant correlated with the duration of the FK506 therapy. A 2 week long treatment with FK506 improved the survival of the encapsulated C₂C₁₂ mEpo cells to 1 1.9%o, up from the 8.8% observed in rats receiving FK506 for only 1 week. Rats receiving a second capsule, long after the initial 1 or 2 week FK506 treatment had ended, were inclined to reject both implants (TABLE 4). In these cases, both the first and the second capsule, having an in vivo residence time of 4 and 13 weeks respectively, showed minimal to zero secretion at explant.

The 4 week treatment regime with FK506 proved to be the most effective in assuring the long-term survival of encapsulated C₂C₁₂ mEpo cells following xenotransplantation. As with the 1 and 2 week treatment groups, implanted rats administered FK506 for 4 weeks sustained elevated hematocrits exceeding 70% throughout the study period (FIG. 4C) After 13 weeks in vivo, capsules retrieved from rats implanted with only one device continued to secrete 37% of their original, pre-implantation levels (TABLE 4). Histology of devices at explant revealed a clear improvement of cell survival in animals receiving FK506 for 4 weeks versus 1 week. Rats which received a second device in the absence of immunosuppression showed less of a tendency to reject the second implant as compared to the previous groups treated with FK506 for 1 and 2 weeks. In these 2 animals, the initial capsules continued to secrete high levels of mEpo at explant, with 1 capsule showing no change in its secretion from implant to explant (TABLE 4). The second implants made in the absence of immuosuppression showed significant mEpo secretion following 4 weeks in vivo. An average cell survival of 27% was observed in these 2 cases as compared the 4.6% and 2.3% viability observed in animals immunosuppressed for 1 and 2, respectively.

Thus, in the context of an immunoisolated xenograft, extended survival of encapsulated C₂C₁₂ mEpo cells in rats is possible following an initial, short course treatment with FK506. Rats immunosuppressed for 1 week following implantation maintained elevated hematocrits for a period lasting 13 weeks. At explant, capsules showed areas of viable cells in the interior and tissue outside the capsule appeared free of lymphocytes. While capsules continued to secrete mEpo, the levels were only 8.8% of those measured pre-implantation.

A longer initial treatment with FK506 improved the long-term survival of the encapsulated C₂C₁₂ mEpo cells. Devices explanted at 13 weeks from animals administered FK506 for 4 weeks secreted 37% of their original pre-implant levels. The choice of the dose delivered, as measured by treatment duration, appears to be critical in determining the success of inducing unresponsiveness to the xenografted capsules.

Rejection of encapsulated xenografts by immunocompetent rats is rapid and after 4 weeks explanted capsules are completely void of cells. On the other hand, devices retrieved from immunosuppressed animals continue to secrete detectable levels of mEpo following 13 weeks. The fact that a longer treatment with FK506 extends the life span of the C₂C₁₂ mEpo ells shows that a process of delayed graft rejection is taking place. Given an adequate dose of

FK506, a long-term tolerance can be developed. Xenografted capsules implanted for 13 weeks (37%) for rats treated with FK506 for 4 weeks) had a superior survival to allografted capsules implanted for only 5 weeks (32% for DBA/2J mice) (TABLE 1 and TABLE 4).

Thus, a threshold dose of immunosuppression is needed immediately following xenograft implant for its development. During this period, the unstimulated host immune system is exposed to xenoantigens continually shed by the encapsulated cells. This prolonged exposure to xenoantigens in an immunosuppressed background leads to the tolerization of the host.

EXAMPLE 5

TOLERIZATION USING UNENCAPSULATED XENOTRANSPLANTATION

To evaluate the protective role of FK506 in a model of unencapsulated xenotransplantation, C₂C,₂ mEpo cells were injected directly into the tibialis cranialis and gastrocnemius muscles of Fischer rats. Following injection, animals were treated daily for 2 weeks with FK506. During this period the hematocrit increased to a high of 68%, following a trend identical to that observed in FK506 treated rats implanted with encapsulated cells. After withdrawal of the FK506, the hematocrits eventually declined to baseline levels. The kinetics of cell rejection varied from animal to animal, as seen in the heterogeneity of the hematocrit profiles (FIG. 5) At 6 weeks post-injection, the animals began to return to their normal hematocrit levels. One injected rat maintained a hematocrit of 65%> as long as 1 1 weeks before eventually rejecting the implanted cells.

Control, immunocompetent rats injected with C₂C₁₂ mEpo cells transiently increased their hematocrit during the first week but subsequently returned to their basal levels 3 weeks later (FIG. 5). This illustrates that a transient treatment with FK506 can significantly prolong the life span of transplanted xenogeneic myoblasts. However, even with the use of F506, the long-term survival of injected cells, as evaluated by hematocrit levels, was clearly inferior to that observed with encapsulated cells. This highlights the importance of combining the technique of immunoisolation with transient immunosuppression when transplanting xenogeneic cells into a site visible to the immune system.

EXAMPLE 6 LONG-TERM HOST UNRESPONSIVENESS TO ENCAPSULATED XENOGENEIC

MYOBLASTS FOLLOWING TRANSIENT IMMUNOSUPPRESSION

Murine C₂C_I2 myoblasts engineered to secrete murine erythropoietin (Epo) were used to enable in vivo monitoring of xenograft survival by fluctuations in the hematocrit. These C₂C,₂mEpo cells were encapsulated in a semipermeable membrane and subsequently implanted in the subcutaneous site of xenogeneic rat recipients. The C₂C₁₂ myoblasts were then used to evaluate the response of control versus FK506 treated xenogeneic recipients (Fischer rats) to encapsulated myoblasts implanted in the subcutaneous site. Encapsulated C₂C₁₂ mEpo cells were rapidly eliminated in immunocompetent Fischer rats. Devices transplanted into nude rats induced a sustained increase in the hematocrit, associated with an extended viability of the encapsulated cells. Short-term immunosuppression with FK506, for periods lasting either 1 , 2, or 4 weeks following implantation, permitted the long-term survival of encapsulated C₂C₁₂ mEpo cells in Fischer rats. Animals increased their hematocrits to over 70%) and maintained these levels for 13 weeks, independent of the duration of FK506 treatment. Unencapsulated C₂C,₂ mEpo cells injected intramusculary in immunosuppressed animals were rejected over this same period.

This EXAMPLE shows the importance of combining the technique of cell encapsulation with transient immunosuppression to achieve long-term survival of xenografted myoblasts in a peripheral inimunoreactive site. Encapsulation alone cannot protect xenogeneic myoblasts from immune destruction in the subcutaneous site.

Transplantation of encapsulated C2C12 mEpo myoblasts in C3H mice, DBA/2 J mice and Fischer rats. C₂C₁₂ myoblasts stably transfected with the pPI-mEpo-ND expression vector released 163 IU Epo/1 06 cells/day, as determined using a human Epo ELISA test which cross- reacted with mEpo (Regulier et al, 5 Gene Ther. 1014 (1998)). For cell culture, mouse C₂C₁₂ myoblasts obtained from the American Type Culture Collection (ATCC; CRL 1772,

Rockville, MD) were transfected with the pPI-mEpo-ND plasmid (Regulier et al, 5 Gene Ther. 1014 (1998)) using calcium phosphate precipitation (mammalian transfection kit, Stratagen, Basel, SWITZERLAND). The cells were selected for 2 weeks in 0.8 mg/ml G418. Subsequently, selected cells were incubated with increasing concentrations of methotrexate (1 to 200 M) over 6 weeks to amplify the copy number of the integrated plasmid. Selected pools of high expressing cells were obtained by diluting the total population to a final concentration of 10 cells per well. The C₂C₁₂ mEpo myoblasts used in this EXAMPLE were derived from the original pool previously described (Regulier et al, 5 Gene Ther. 1014 (1998)). Cells were grown in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal calf serum (FC S), 2 mM L- glutamine, 4.5 g/1 glucose, 100 U/ml penicillin and 100 U/ml streptomycin.

For cell encapsulation, C₂C₁₂ cells were harvested using a 0.125%> trypsin-EDTA solution prepared in modified Puck's saline and were diluted with DMEM in order to achieve a suspension of lxl 0⁵ cells/μl. The suspension was then injected into semipermeable polyethersulfone (PES) hollow fibers (OD = 720 μm; ID = 524 μm; molecular weight cutoff = 32 kDa and 280 kDa) (Akzo Nobel Faser AG, Wupperthal, GERMANY). Both ends of the capsule were sealed by photopolymerization at a wavelength of 460 nm of an acrylate based glue (Luxtrak LCM 23, Ablestik, Electronic Materials & Adhesives, Rancho Dominguez, CA,

USA).

Encapsulated cells were differentiated for 3 days in DMEM containing 10%_> Prolifix (Bio Media, Boussens, FRANCE), a serum free medium capable of inducing myoblast differentiation. The encapsulated cells were subsequently returned to DMEM containing 10%_> FCS for 4 days.

These encapsulated cells were transplanted subcutaneously in C3H mice, DBA/2J mice and Fischer rats in order to compare their survival in syngeneic, allogeneic and xenogeneic recipients. C3H and DBA/2J mice were each implanted with 1 capsule containing C,C_I2 mEpo myoblasts. Capsule implantation was performed as follows: C3H and DBA/2J mice (Iffa Credo, Saint-German sur l'Abresle, FRANCE) were chosen for syngeneic and allogeneic transplantation, respectively. Fischer rats (Iffa Credo, FRANCE) and Rnu rats (Charles River, Sulzfeld, GERMANY) were used as xenogeneic transplant recipients. For implantation, animals were anesthetized by inhalation of isoflurane (Forene, Abbott Laboratories, Cham, SWITZERLAND). Capsules were implanted subcutaneously in the dorsal flank of the animals using a trocar (Abbocath-T 16 G, Abbott Laboratories, Chain, SWITZERLAND). The entry site in the skin was closed using a nonresorbable suture (Prolene 6-0). Upon recovery, the animals were returned to the animal care facility, where they had access to food and water ad libitum. Capsules implanted into the mice were 1 cm long capsules loaded with approximately 6xl0⁵ C₂C₁₂ mEpo cells and implanted subcutaneously on the dorsal flank of the mice. Capsules implanted into the rats contained lxl 0⁶ cells and measured 2 cm long. Their structure was reinforced with an internal titanium coil to prevent kinking. The membrane used for immunoisolation had a mean molecular weight cut-off of 280 kDa. Immunosuppression was performed as follows: Selected rats were pretreated for 3 days with FK506 (1 mg/kg BW) (Fujisawa GmbH, Munchen, GERMANY). After transplantation, these animals were treated daily (5 days out of 7) for 1 , 2 or 4 weeks at a dose of 1 mg/kg body weight. FK506 doses were injected intramuscularly into the quadriceps muscle, alternating daily between the left and right legs. Intramuscular injections of cells was performed as follows: C₂C₁₂ mEpo cells were harvested from confluent cultures by trypsinization.

Subsequently, they were washed 3x in Hank's balanced salt solution (HBSS) and resuspended at a concentration of lxlO⁵ cells/μl HBSS. Under general anesthesia, a small incision was made in the hind limb skin to expose the underlying muscles. Using a Hamilton syringe with a 26S-gauge needle, 5 million myoblasts were injected in the tibialis cranialis and gastrocnemius muscles. The incision was then closed with a Dermalon 5-0 suture.

The secretion of mEpo from encapsulated mEpo cells was measured both preimplantation and following explant by incubating the capsules for 1 hr in 1 ml DMEM containing 10%> FCS. Epo levels in the conditioned media were measured using an enzyme linked immunosorbent assay (ELISA) (Quantikine IVD, R&D systems, Minneapolis). Cross- reaction of the kit allowed detection of mEpo in culture supernatants (Regulier et al, 5 Gene

Ther. 1014 (1998)).

C3H and DBA 2J mice were each implanted with 1 capsule containing C₂C₁₂ mEpo myoblasts. Prior to implantation, capsules containing C₂C₁₂ mEpo myoblasts were measured for their secretion of Epo (TABLE 5). Both mice strains experienced a significant increase in their hematocrit and maintained these elevated levels for the 5 week trial period (FIG. 6). For hematocrit measurement, blood was drawn weekly from the tail vein into heparinized capillary tubes. The hematocrit was then measured by a standard microhematocrit method (Koepke, Microhematocrit method. In Koepke JA, ed., Practical Laboratory Hematology. 112 (New York, Churchill Livingstone, 1991)). By the test endpoint, C3H and DB A 2 J mice had attained hematocrits of 79.7 ± 4.18% and 89.5 ± 3.7%, respectively (FIG. 6). At explant, devices retrieved from both C3H and DBA/2J mice continued to secrete significant quantities of mEpo (TABLE 5). Histological analysis was performed as follows: At the test end point, the capsules were carefully explanted, fixed in Lang's fixative for 3 hours and dehydrated using alcohol in preparation for glycol-methacrylate embedding (Leica Instruments GmbH, Nussloch, GERMANY). After removing the titanium reinforcement, the capsules were cut at 5 μm thickness nd stained with hematoxylin and eosin. For analysis of injected muscles, animals were sacrificed by pentobarbital overdose and perfused transcardially with 4%> paraformaldehyde. Muscles were dissected and embedded in paraffin for histology. Sections were cut at 7 μm thickness and stained with hematoxylin and eosin.

Histological analysis revealed living myoblasts in explanted devices. In a few instances, a necrotic core was observed at the capsules' center. The biocompatibility of the system was characterized by an extensive neovascular network surrounding the capsules, with only a thin fibrotic reaction adhering to the membrane.

Fischer rats were each implanted with one 2 cm long capsule releasing 23.3 ± 2.9 IU mEpo/day (TABLE 5). In vivo, the delivered mEpo induced a significant, but only transient increase in the hematocrit. Two weeks following implantation, the hematocrit levels had risen to a high of 65.9 + 24% which then progressively decreased to pre-implantation levels (FIG. 6). At explant, histology of the devices revealed that the encapsulated cells had all died. The capsules were surrounded by a pronounced pericapsular tissue reaction characterized by an extensive infiltration of neutrophils and lymphocytes, suggesting an immune-mediated cell destruction.

Transplantation of encapsulated C2C12 mEpo myoblasts in Rnu nude rats. To examine the role of the immune system in the rejection of encapsulated xenogeneic myoblasts following transplantation, the response of nude rats to capsules containing C₂C₁₂ mEpo myoblasts was tested. Nude rats possess a rnu/rnu mutation which provokes hairlessness and thymic aplasia, rendering them severely immunodeficient (Festing et al, 274 Nature 365 (1978)).

Eight nude rats were implanted for 28 days following the protocol previously used. Fischer rats were implanted in parallel to serve as controls. Pre-implantation, devices implanted in immunodeficient rats secreted 33.0 ± 1.9 IU mEpo/day (TABLE 5). During the first week post-implantation, the hematocrit of nude rats increased in a manner similar to that of control rats, reaching 53.4 ± 3.2% (FIG. 7). On day 14, a continued increase of the hematocrit was observed in nude rats (78.4 ± 3.7% on day 28) (FIG. 7), whereas the hematocrit in control rats had already substantially declined (54 ± 1.7 on day 28) at that time.

On day 28, animals were sacrificed and capsules were explanted. C₂C,₂mEpo capsules retrieved from Rnu rats secreted 6.9 ± 1.2 IU mEpo/day (TABLE 5). As expected, no detectable mEpo secretion was measured in capsules explanted from control Fischer rats. Histological analysis of the devices retrieved from nude rats showed the presence of viable myoblasts. On the other hand, devices retrieved from control Fischer rats were filled with cellular debris. The morphology of the adherent tissue surrounding the explanted polyethersulfone membrane revealed extensive differences between the explants from both rat strains. In control immunocompetent rats, this tissue was poorly vascularized and profusely infiltrated with lymphocytes and neutrophils as observed previously. In contrast, the pericapsular tissue in nude rats was thinner and better vascularized, with a small infiltration of lymphocytes.

Effect of membrane molecular weight cut-off on the survival of encapsulated C2C12 mEpo myoblasts in C3H mice and Fischer rats. Using membranes with a molecular weight cut-off smaller than 280 kDa may protect transplanted tissues from immune rejection both by reducing the outward flow of immunogenic antigens from the capsule and by preventing the entry of antibody and cytolytic complement proteins inside. To test this, C₂C_]2 mEpo myoblasts were encapsulated in microporous membranes having a mean molecular weight cut- off of 32 kDa. Encapsulated cells were first implanted in syngeneic C3H mice to verify that the transport properties permitted sufficient access to nutrients. Following transplantation, C3H mice quickly increased their hematocrit and maintained levels above 70%> throughout the test period (FIG. 8). Histology of capsules explanted after 5 weeks revealed a cell viability that was comparable to that observed with 280 kDa membranes. This demonstrated that the pore size of the membrane did not hinder the exchange of essential nutrients.

Subsequently, Fischer rats were implanted with C₂C₁₂ mEpo myoblasts encapsulated in identical membranes. These capsules induced only a transient increase in the hematocrit before returning to baseline levels 5 weeks later (FIG. 8). No difference was observed between the hematocrit profiles of Fischer rats receiving capsules having a 280 kDa versus 32 kDa cutoff (FIG. 8). At explant, devices contained no viable cells and secretion of mEpo was undetectable. Histology of capsules having a 32 kDa membrane was comparable to those with a 280 kDa cut-off, with numerous lymphocytes observed around the capsules.

Survival of encapsulated C2C12 mEpo myoblasts in Fischer rats immunosupressed with FK506. Observing that the encapsulated xenogeneic C₂C,₂ cells survived in the immune deficient nude rats, the immunosuppressor FK506 was administered to Fischer rats to evaluate if it could similarly protect the encapsulated C₂C_I2 cells against immune mediated destruction. Five Fischer rats were each implanted. The mean capsule secretion before implantation was 34.4 ± 3.3 IU mEpo/day (TABLE 5). Beginning 3 days prior to transplantation, animals were immunosuppressed with FK506 for the test period of 4 weeks.

During the first 2 weeks, immunosuppressed animals showed an upward trend in their hematocrits as observed in immunocompetent control animals previously described (FIG. 7). A difference between the progression of the 2 groups became evident beginning on the third week post-implantation. At this point, immunosuppressed rats continued to increase their hematocrits, while the levels of the control rats began to decline. On the fourth week, rats treated with FK506 continued to maintain hematocrits above 64%>, while untreated animals had decreased to pre-implantation levels (47.5± 2.1%>) (FIG. 7). Capsules retrieved at this time from FK506 treated rats secreted on average 10.6 ± 2.5 LU mEpo/24 hr (TABLE 5). Clusters of cells were distributed throughout the capsules, with occasional regions containing densely packed cells. The pericapsular tissue contained many neovessels in proximity to the capsule membrane, while infiltrating lymphocytes were scarce. Differences in absolute hematocrit elevation obtained by immunosuppressed rats and nude rats in response to mEpo is likely due to side effects of FK506 on red blood cell production (Ichihashi et al, 341 Lancet 1035

(1993)).

Long-term acceptance of encapsulated C2C12 mEpo myoblasts by Fischer rats following transient FK506 therapy. As continuous administration of FK506 permitted the survival of encapsulated cells in xenogeneic recipients for 4 weeks, short-term dosing regimens were evaluated for their efficacy on long-term encapsulated xenograft survival.

Fischer rats were implanted with encapsulated C₂Cι₂ mEpo myoblasts and immunosuppressed for either 1 , 2, or 4 weeks. In the 3 treatment groups, a gradual increase in the hematocrit was observed, with levels reaching over 70% on week 13 (FIG. 9). A long-term unresponsiveness to encapsulated C₂C₁₂ mEpo myoblasts developed following only a 1 week administration of FK506. No difference was observed in the progression of hematocrit in the 3 animal groups during the EXAMPLE. All animals maintained an elevated hematocrit exceeding 65% throughout the 3 month trial period (FIG. 9). Devices explanted at 13 weeks continued to secrete measurable quantities of mEpo (TABLE 6). TABLE 6 ^'

Secretion of mEpo pre-implantation versus post-explantation in rats transiently immunosuppressed with FK506 for either 1, 2, or 4 weeks.

Mouse Epo release Expl/impl (%)

(IU/24 hr) implant explant

1 week FK506

Rat #l 34 5 14.7%

Rat #2 46 7.5 16.3%

Rat #3 38 3.7 9.7% mean 39.3 5.4 13.6 %

2 weeks FK506

Rat #l 39 4.7 12.1%

Rat #2 29 4.8 16.6%

Rat #3 36 13 36.1% mean 34.7 7.5 21.6 %

4 weeks FK506

Rat #l 36 10.6 29.4%

Rat #2 36 12.3 34.2% mean 36 11.4 31.7

Fischer rats were immunosuppressed with FK506 (1 mg/kg) ( m a daily basis, 5 out of 7 days, for either 1 , 2, or 4 weeks. Capsules were retrieved and measured for mEpo secretion following 91 days residence in vivo.

The 4 week treatment regimen with FK506 appeared to be slightly more effective than 1 and 2 week regimens in assuring the long-term survival of encapsulated C₂C₁₂ mEpo cells (TABLE 6). After 13 weeks in vivo, the capsules retrieved from implanted rats treated 4 weeks secreted slightly higher levels of mEpo than rats administered FK506 for 1 and 2 weeks (TABLE 6). Histology of devices at explant revealed viable cells with some central necrosis in animals from all 1, 2, and 4 week FK506 treatment groups. Outside the capsules, an extensive network of blood vessels was observed close to the membrane, with few lymphocytes in the vicinity. Injection ofC2Cf2 mEpo myoblasts in immuocompetent and immunosuppressed Fischer rats .The protective role of the capsule in achieving long-term xenograft acceptance following a transient immunosuppression was evaluated in a model of unencapsulated xenotransplantation. C₂C₁₂ mEpo myoblasts were injected directly into the tibialis cranialis and gastrocnemius muscles of either immunocompetent Fischer rats or those administered FK506 for 4 weeks. Control, untreated rats injected with C₂C; ^mEpo myoblasts transiently increased their hematocrit during the first week, but on the second week hematocrits began to decline, returning to their basal levels 3 weeks later (FIG. 10). Animals treated daily with FK506 for 4 weeks increased their hematocrit to a high of 64.2% during this period, following a trend resembling that observed in FK506 treated rats implanted with encapsulated C₂C₁₂ mEpo myoblasts. In one animal sacrificed at the end of the fourth week of FK506 treatment, histology of the injected leg muscle showed viable C₂Cj2 mEpo myoblasts, with the absence of infiltrating lymphocytes. After withdrawal of the FK506, the hematocrit levels remained elevated for the 4 weeks that followed. On the ninth week post-injection, the hematocrit levels began to decrease, with all of the animals returning to baseline levels by week 13. The kinetics of cell rejection varied slightly from animal to animal, as seen in the error bars of the hematocrit profile (FIG. 10A). Similar results were observed in rats that were injected with C₂C₁₂ mEpo cells and treated only 2 weeks with FK506.

Discussion. In this EXAMPLE, murine C₂C₁₂ myoblasts engineered to secrete mEpo were used to examine the criteria for the survival of encapsulated xenogeneic myoblasts in the subcutaneous site of a rat recipient. Immunocompetent Fischer rats consistently rejected encapsulated C₂C₁₂ mEpo secreting cells implanted subcutaneously and capsules retrieved following 1 month in vivo were found to contain only cellular debris. The pericapsular tissue was characterized by a massive lymphocytic infiltration, suggesting an immune mediated destruction. It is important to note that unmodified C₂C₁₂ myoblasts and C₂C₁₂ cell lines engineered to secrete either rat erythropoietin or the glucagon-like peptide 1 were similarly rejected, showing an identical tissue response when encapsulated and implanted subcutaneously in Fischer rats.

Comparable results have been reported in a model of encapsulated xenotransplantation in which the African green monkey kidney cell line, COS-7, was encapsulated and implanted in the ovarian fat pad of mice recipients (Loudovaris et al, 61 Transplantation 1678 (1996)). In addition to showing that encapsulated xenografts die in immunocompetent mice and survive in immunodeficient animals, Loudovaris et al demonstrated that this immune destruction was mediated by CD4+T cells (Loudovaris et al, 61 Transplantation 1678 (1996)). By administering an anti-CD4 depleting monoclonal antibody, Loudovaris et al showed encapsulated graft survival as long as 3 weeks, while depleting the CD8+ T cells showed no protective effect. This shows that encapsulated xenograft destruction is mediated by CD4⁺ T cells.

This EXAMPLE documents transient immunosuppression as a method of achieving long-term host unresponsiveness to encapsulated xenogeneic myoblasts grafted outside the central nervous system. In the concordant mouse-to-rat model presented, an initial treatment with FK506 permitted encapsulated C₂C₁₂mEpo myoblasts to survive at least 3 months when transplanted into the subcutaneous site of Fischer rats. No differences were observed in the hematocrit level attained at 13 weeks in animals given longer initial treatments with FK506. However, slightly higher levels of mEpo were produced by devices retrieved from animals immunosuppressed 4 weeks versus only 1 week. It is important to note that capsules secreting as little as 2 IU mEpo/day are able to maintain rats at their threshold hematocrit levels, making device secretion at explant a more quantitative means of evaluating the survival of xenogeneic myoblasts.

EXAMPLE 7 TOLERIZATION OF A HOST TO ENCAPSULATED XENOGENEIC MYOBLASTS

In this EXAMPLE, Fischer rats were rendered unresponsive to encapsulated murine C₂C, ₂ myoblasts secreting mouse erythropoietin by either a 1 or 4 week initial treatment of FK506. To examine if a tolerance to xenografts had been established, animal were challenged with a second implant 9 weeks after the initial implantation. Challenging animals treated only 1 week with FK506 led to rejection of both primary and secondary implants. Administering

FK506 for 4 weeks tolerized animals and both implants remained viable over the test period. Tolerized animals rejected unencapsulated xenogeneic cells injected at a later time, highlighting the requirement of the polymer membrane for immune protection. Developed tolerance to encapsulated xenogeneic myoblasts lasted over extended periods (at least 7 months), in the absence of both immunosuppression and stimulating xenoantigens.

This EXAMPLE shows that a host tolerance can be established to xenoantigens released by encapsulated xenogeneic cells by using a short-term immunomodulation. Murine erythropoietin (mEpo) was used as a reporter gene to permit monitoring of xenograft viability by fluctuations in animal hematocrit and by comparing mEpo secretion from devices at explant versus implant. In a concordant mouse-to-rat model of xenotransplantation, encapsulated murine C₂C_]2 mEpo myoblasts are rejected by Fischer rats following subcutaneous implantation, however, transient immunosuppression with FK506 for either 1, 2 or 4 weeks at implant permits the long-term survival of these cells. We then hypothesized that by exposing arLimmunosuppressed host to continuously shed or secreted xenoantigens, a prolonged tolerance to these antigens can be developed.

This EXAMPLE further examined the nature of the host's acceptance of encapsulated xenografts following FK506 treatment. Fischer rats rendered unresponsive to encapsulated C₂C₁₂mEpo cells by transient immunosuppression were challenged with a secondary implant containing identical cells, in the absence of immunosuppression, to establish if animals had been tolerized. In particular, the role of the length of initial immunosuppression on the survival of cells within the 2 implants was tested. In addition, the extent of host acceptance was considered by tolerizing animals to encapsulated naive myoblasts and challenging animals with encapsulated genetically modified myoblasts. Challenging host unresponsiveness to encapsulated xenografts. Mouse C₂C₁₂ myoblasts obtained from the American Type Culture Collection (ATCC; CRL 1772, Rockville, MD) were stably transfected with the pPI-mEpo-ND plasmid (Regulier et al., 5 Gene Ther. 1014 (1998)) using calcium phosphate precipitation as previously described. The C₂C₁₂ mEpo secreting cells used in this EXAMPLE were derived from this original pool . Cells were cultured in Dulbecco's modified Eagle's medium (DMEM) containing 10%> fetal calf serum

(FCS), 2 mM L-glutamine, 4.5 g/1 glucose, 100 U/ml penicillin and 100 U/ml streptomycin.

C₂C₁₂ cells were encapsulated as described by Regulier et al. (5 Gene Ther. 1014 (1998)). Briefly, cells were harvested and diluted in DMEM to obtain a concentration of lxl 0⁵ cells/μl. The suspension was then injected into semipermeable polyethersulfone (PES) hollow fibers (GD = 720 μm; ID = 524 μm; molecular weight cutoff = 32 kDa and 280 kDa) (Akzo

Nobel Faser AG, Wupperthal, GERMANY), which were subsequently sealed at both ends. In total, 1.2xl0⁶ C₂C₁₂ mEpo cells were encapsulated for transplantation.

The encapsulated cells were differentiated for 3 days in DMEM containing 10% Prolifix (Bio Media, Boussens, FRANCE), a serum free medium capable of inducing myoblast differentiation (Regulier et al, 5 Gene Ther. 1014 (1998)). The encapsulated cells were subsequently returned to DMEM containing 10% FCS for 4 days.

Capsules were implanted subcutaneously on the dorsal flank of Fischer rats. Capsule implantation was performed as follows: Fischer rats (Iffa Credo, FRANCE) were used as concordant xenogeneic transplant recipients. Animals were anesthetized using isoflurane (Forene, Abbott Laboratories, SWITZERLAND) and implanted using a trocar (Abbocath-T 16 G, Abbott Laboratories, SWITZERLAND). One or two cm long capsules were subcutaneously implanted in the dorsal flank. Upon recovery, the animals were returned to the animal care facility, where they had access to food and water ad libitum.

The subcutaneous transplantation at week 9 of an additional capsule containing C₂C₁₂ myoblasts releasing mEpo led to a brief increase of hematocrit in control animals, although to lower levels than those observed following initial implants (FIG. 1 1). For hematocrit measurement, blood was drawn weekly from the tail vein into heparinized capillary tubes. The hematocrit was then measured by a standard microhematocrit method (Koepke,

Microhematocrit method. In Koepke, ed., Practical Laboratory Hematology. 112 (New York, Churchill Livingstone, 1991)).

For immunosuppression, Fischer rats were pretreated for 3 days with FK506 (1 mg/kg BW) (Fujisawa GmbH, Mϋnchen, GERMANY). Following transplantation, these animals were treated daily (5 days out of 7) for either 1 or 4 weeks at a dose of 1 mg/kg BW. FK506 doses were injected intramuscularly into the quadriceps muscle, alternating daily between the left and right legs.

Animal groups originally treated with FK506 for either 1 or 4 weeks responded differently to the second device, as viewed by subsequent changes in the hematocrit. (FIG. 12A, FIG 12B). Four weeks after the second implantation (week 13), the animals given FK506 for 1 week showed a heterogeneity in their hematocrit, with values ranging from 55% to 80%. (FIG. 12 A). At this time, 5 out of the 8 animals in this group had decreased their hematocrit relative to week 1 1 (FIG. 12A). On the other hand, the group administered FK506 for 4 weeks continued to maintain elevated hematocrits, with a narrow distribution of values, spanning from 72% to 80% at week 13 (FIG. 12B).

All implants were removed on week 13 and analyzed for the survival of the xeno transplanted myoblasts. Cell viability within capsules was quantitatively measured by comparing mEpo secretion of devices at explant and implant (TABLE 7) and qualitatively evaluated by histology. The secretion of Epo from encapsulated C₂C₁₂ mEpo myoblasts was measured both pre-implantation and following explant by incubating the capsules for 1 hr in 1 ml DMEM containing 10% FCS. Epo levels in the conditioned media were measured using an enzyme linked immunosorbent assay (ELISA) (Quantikine IVD, R&D systems, Minneapolis, USA). Cross-reaction of the kit allowed detection of mEpo in culture supernatants (Rinsch et al, 8 Hum. Gene Ther. 1881 (1997); Regulier et al, 5 Gene Ther. 1014 (1998)). At the test end point, the capsules were carefully explanted, fixed in Lang's fixative for 3 hr and dehydrated using alcohol in preparation for glycol-methacrylate embedding (Leica Instruments GmbH, Nussloch, GERMANY). Capsules were cut at 5 μm thickness and stained with hematoxylin and eosin.

A clear difference was observed in the viability of myoblasts within devices residing in animals originally treated 4 weeks versus 1 week with FK506. The first and second implants retrieved from animals immunosuppressed 1 week secreted only 3.8±2.4%_> and 2.8±1.4%_> of pre-implantation levels, respectively (TABLE 7). Histological examination of devices revealed mostly empty capsules containing few to no cells. Both the first and second implants were surrounded by a pronounced tissue reaction which was poorly vascularized near to the membrane and contained many infiltrating lymphocytes.

TABLE 7 Secretion of mEpo by capsules implanted in rats transiently immunosuppressed with

FK506 for either 1 or 4 weeks.

Epo release (IU/24 hrs) Expl/impl (%) Viable Capsules Implant Explant at Explant*

Untreated

First implant (n=6) 17.6 ± 1.2 0 0/6

Second implant (n=6) 14.5 + 0.9 0 0/6

1 week FK506

First implant (n=8) 28.3 ± 2.9 0.9 ± 0.5 3.8 ± 2.4 3/8 Second implant (n=8 10.2 ± 1.1 0.2 ± 0.1 2.8 ± 1.4 2/8

4 weeks FK506

First implant (n=8) 28.4 ± 3.1 5.3 ± 1.7 26.0 ± 12.1 ** 8/8

Second implant (n=8) 9.8 ± 1.0 1.8 ±_.0.4 19.7 ± 4.9** 8/8

Fischer rats were immunosuppressed with FK506(1 mg/kg) on a daily basis, 5 out of 7 days, for either 1 or 4 weeks.

*Individual capsules were counted as viable if they secreted greater than 0.3 IU/24 hr.

**Relatιve secretion (explant/implant) of implants was significantly higher (n=8, pO.OOOl) following a

4 week versus 1 week treatment with FK506. Data shown as the mean + SEM.

On the contrary, a 4 week initial immunosuppression markedly improved the survival of encapsulated xenogeneic myoblasts. Upon retrieval, the first implants produced mEpo at

26.0±12.1 % of pre-implantation levels, while the second implants secreted 19.7±4.9%_> of their original levels (TABLE 7). This represents a significant increase (pO.Ol) in mEpo secretion levels (explant/implant), and hence xenogeneic myoblast survival, in comparison to animals immunosuppressed only 1 week. Analysis of capsules by histology confirmed the presence of large viable cell clusters inside both the first and second implants, with a characteristic necrosis at the center of the capsule. The pericapsular tissue reaction was moderate, containing fewer lymphocytes than observed around devices retrieved from animals immunosuppressed only 1 week. Implants also appeared highly vascularized, with blood vessels close to the -membrane surface, an effect thought to be due to the reported angiogenic activity of mEpo

(Anagnostou et al, 87 Proc. Natl. Acad. Sci. USA 5978 (1990)).

Rejection of Intramuscularly Injected C2C12 iEpo cells by Tolerized Animals. Fischer rats previously tolerized to encapsulated C₂C₁₂ mEpo myoblasts by a 1 or 4 week treatment with FK506, as described above, were challenged a second time with unencapsulated C₂C₁₂ mEpo myoblasts (FIG. 12A, FIG 12B). Intramuscular injections of xenogeneic myoblasts was performed as follows: C₂Cι₂ mEpo myoblasts were harvested from confluent cultures, washed 3x in Hank's balanced salt solution (HBSS) and resuspended at a final concentration of lxlO⁵ cells/ 1 in HBSS. Under general anesthesia, a small incision was made in the hind limb skin to expose the underlying muscles. Using a Hamilton syringe with a 26S- gauge needle, 5xl0⁶ were injected in the tibialis cranialis and gastrocnemius muscles.

Six weeks after all implanted capsules were removed (week 19) and when hematocrits had stabilized at their baseline values, animals were injected intramuscularly with 5xl0⁶ C₂C₁₂ mEpo cells. In response, animals from both groups increased their hematocrits, but only transiently, before rejecting the implanted myoblasts and returning to normal hematocrit levels (FIG. 12A, FIG 12B). No difference in the kinetics of myoblast rejection, as evaluated by changes in the hematocrit, was apparent between the groups that previously rejected (1 week FK506) or accepted (4 weeks FK506) a second capsule containing C₂C₁₂ mEpo myoblasts.

Long-term unresponsiveness of tolerized animals to encapsulated xenogeneic myoblasts. Four Fischer rats were tolerized to encapsulated C₂C₁₂ mEpo myoblasts by 4 weeks of initial immunosuppression. Two of these animals were challenged with a second capsule on week 9, and all of the animals were explanted on week 13 (FIG. 13). Afterwards, a period of 7 months lapsed, during which animals neither received capsule implants nor were administered immunosuppressors. These animals were then reimplanted subcutaneously with a capsule containing C₂C₁₂mEpo myoblasts (FIG. 13). On average, the animals hematocrit increased and remained elevated for the 4 weeks that followed (FIG. 13). At explant, capsules retrieved from 3 of the 4 animals continued to secrete high levels of mEpo, while one device produced mEpo near the limit of detection for the ELISA used (TABLE 8).

TABLE 8

Comparison of mEpo secretion by capsules pre-implant versus post-explant.

Long-term tolerance Epo release (IU/24 hr) Expl impl (%)

Implant Explant

Animal #

1 39.0 0.1 0.3

2 39.3 2.3 5.9

3 39.0 2.4 6.1

4 44.4 2.8 6.2

Fischer rats were immunosuppressed with FK506 (1 mg/kg) on a daily basis, 5 out of 7 days, for 4 weeks.

Xenoantigens Implicated In Rejection Of Encapsulated mEpo Secreting Myoblasts Appear To Be Molecules Inherent To Myoblasts. Tests were devised to gain further insight as to which molecules contribute to the rejection of encapsulated xenogeneic myoblasts. Capsules containing unmodified C₂C₁₂ cells were subcutaneously implanted into Fischer rats. Animals were either tolerized to encapsulated xenogeneic myoblasts by 4 weeks of FK506 immunosuppression or left untreated (FIG. 14A, FIG 14B). Four weeks after implantation, all devices were removed and analyzed for cell viability. Capsules retrieved from immunosuppressed animals contained viable cells while implants removed from untreated animals were empty and showed an extensive lymphocytic reaction around the capsule membrane. Five weeks later (week 9), these animals were challenged with a second implant, this time containing C₂C₎₂ mEpo cells (FIG. 14A, FIG 14B). In response to the new devices, the tolerized animals progressively increased their hematocrit over the 4 weeks following implantation (FIG. 14B). Control rats showed only a transient rise in hematocrit that began to decrease by the second week and returned to base line levels at week 4 (FIG. 14A). C₂C₁₂ mEpo capsules retrieved after a 4 week residence (week 13) from animals tolerized to the unmodified C₂Cι₂ cell line continued to secrete mEpo at 20.0±2.2%_> of pre- implantation levels (TABLE 9). This was comparable to secretion of secondary implants made following the initial transplant of encapsulated C₂C₁₂ mEpo myoblasts (TABLE 7). Devices removed from control, Jion-immunosuppressed animals showed no detectable mEpo secretion at this time (TABLE 9).

TABLE 9

Tolerance to native C₂C₁₂ mybolasts is conferred to C₂C₁₂mEpo secreting myoblasts.

Secretion of mEpo by secondary implants in rats initially implanted with capsules containing unmodified C₂C₁₂ myoblasts.

Epo release (IU/24 hr) Expl/impl (%) Viable Capsules Implant Explant at Explant*

Untreated Second implant 42.2 + 3.1 0 0 0/5 (n=5)

4 weeks FK506 Second implant 37.7 + 2.5 7.4 + 0.5 20.0 + 2.2 5/5 (n=5)

Fischer rats were immunosuppressed with FK506 (1 mg/kg) on a daily basis, 5 out of 7 days, for 4 weeks were indicated.

*Individual capsules were counted as viable if they secreted greater than 0.3 R7/24 hr. Data shown as the man + SEM.

Conclusion. Encapsulated C₂C₁₂ mEpo secreting myoblasts are rejected within 4 weeks when they are subcutaneously implanted in Fischer rats. Transient administration of FK506 lasting either 1, 2 or 3 weeks at the time of transplantation enabled encapsulated C₂C₁₂ mEpo cells to survive as long as 13 weeks afterwards, without showing any immediate signs of rejection (see, EXAMPLE 6). This EXAMPLE indicates that the length of initial immunosuppression plays an important role in determining the survival of secondary implants performed after immunosuppression has ended. A 4 week administration of FK506 was more efficacious than treatments lasting 1 week in inducing a tolerance to encapsulated C₂C₁₂ myoblasts. This provides evidence that tolerization occurs following a prolonged concomitant exposure to both the encapsulated xenograft and to FK506.

In those animals immunosuppressed only 1 week, the second capsule provokes the rejection of cells in the first implant. As both devices are positioned at different locations, with inflammation localized around each capsule exterior, the immune reaction induced appears to be rather specific against the xenogeneic cells. If the immune reaction was a general inflammatory response following transplantation, the survival of encapsulated cells in the first implant should not be affected. This shows that the induced tolerance threshold to xenoantigens is dependent on the persistence of xenoantigens in an immunosuppressed background. Studies in allotransplantation have highlighted the importance of antigen persistence on the efficacy of T cell tolerization (Scully et al, 24 Eur. J. Immunol. 2383 (1994), Ehl et al, 4 Nature Med. 1015 (1998)), suggesting that this may similarly apply to xenotransplantation.

In this EXAMPLE, animals tolerized to shed xenoantigens maintain this state of unresponsiveness over an extended period in the absence of transplant specific antigens. Most tolerized hosts accepted encapsulated xenografts implanted 9 months after the initial treatment with FK506 had ended and 7 months after the last host exposure to donor xenoantigens. This implies that an important modification in the host immune response to shed or secreted xenoantigens has occurred. A lack of continued xenoantigen presence may moderate host tolerance as xenogeneic myoblast survival in this case was reduced in comparison to challenges given at shorter intervals between the last host exposure to xenoantigens.

Interestingly, animals tolerized to encapsulated C₂C₁₂ mEpo cells will reject these cells when they are later transplanted unencapsulated into the recipients' muscle. It is not understood why xenogeneic cells become resistant to indirect immune attacks while remaining vulnerable to direct cell-contact mediated destruction. This may be due to the presentation of additional effector xenoantigens when xenogeneic myoblasts are directly injected intramuscularly. Different effector mechanisms may also play a role in the rejection of injected xenogeneic myoblasts including, direct attack by natural killer cells (Manilay & Sykes, 10

Current Opinion in Immunology 532 (1998)), macrophage cytotoxicity by recognition of bound preexisting xeno-antibodies, and T cell cytotoxicity.

Animals tolerized to xenoantigens originating from encapsulated unmodified C₂C₁₂ myoblasts remain unresponsive to encapsulated C₂C₁₂ mEpo secreting cells implanted at a later date. Thus antigens specific to C₂C₁₂ myoblasts define the rejection mechanism. Once tolerance to a baseline xenogeneic cell is established, certain additional genetic variations between batches of cells will not be sufficient to induce rejection.

The finding that transient immunosuppression can not only extend survival of encapsulated xenogeneic myoblasts, but more importantly, tolerize the host to subsequent implants performed in the absence of immunosuppression has important consequences in the field of xenotransplantation. In this EXAMPLE, both primary and secondary capsule implants retrieved after 13 and 4 weeks, respectively, from animals tolerized 4 weeks, contained numerous viable xenogeneic myoblasts, suggesting that these myoblasts could have survived for much longer periods in vivo. In other xenotransplantation models, short-term immunosuppression with FK506 has been shown to extend survival of primary and secondary xenogeneic skin grafts following withdrawal of FK506, although a rejection endpoint was clearly seen (Sakamoto et al, 21 Transplant Proc 527 (1989)). Successful protocols permitting indefinite- host tolerance to whole organ xenotransplants require a initial treatment with lefiunomide, to prevent T-independent rejection, combined with a continuous treatment of cyclosporin A to keep T-dependent rejection under control (Yin et al, 161 J. Immunol. 2044 (1998)). Halting treatment with immunosuppressors ultimately leads to xenograft failure. This EXAMPLE demonstrates the development of a long-term peripheral xenograft tolerance, in the absence of continuous immunosuppression, with hosts showing no indication of total graft rejection.

The foregoing description has been presented only for the purposes of illustration and is not intended to limit the invention to the precise form disclosed, but by the claims appended hereto.

Claims

CLAIMSWE CLAIM:

1. A method for improving survival of a transplant in a recipient host comprising:

(a) implanting into the recipient host one or more encapsulated cell devices containing tissue or cells that are allogeneic or xenogeneic to the recipient host;

(b) administering an immunosuppressive agent to the recipient host in an amount effective to permit survival of the encapsulated tissue or cell transplant for a period of at least one week;

(c) removing the encapsulated cell devices; and

(d) subsequently implanting into the recipient host a transplant from the same tissue source as the tissue or cells in the encapsulated cell devices; wherein the transplant has improved survival in the recipient host compared to either (i) a recipient host that did not receive an administration of an immunosuppressive agent in an amount effective to permit survival of the encapsulated tissue or cell transplant for a period of at least one week; or (ii) a recipient host that did not receive an implantation of one or more encapsulated cell devices containing tissue or cells that are allogeneic or xenogeneic to the recipient host, followed by an administration of an immunosuppressive agent in an amount effective to permit survival of the encapsulated tissue or cell transplant for a period of at least one week.

2. The method according to claim 1 , wherein the transplant is allogeneic to the recipient host.

3. The method according to claim 1 , wherein the transplant is xenogeneic to the recipient host.

4. The method according to claim 1, wherein the immunosuppressive agent is administered for between one to four weeks.

5. The method according to claim 1 , wherein the immunosuppressive agent is administered for two weeks.

6. The method according to claim 1, wherein the immunosuppressive agent is administered for three weeks.

7. The method according to claim 1, wherein the immunosuppressive agent is administered for at least four weeks.

8. The method according to claim 1 , wherein the recipient is mammal.

9. The method according to claim 1, wherein the recipient is a human.

10. The method according to claim 1 , wherein the transplant is selected from the group consisting of: at least one second encapsulated cell device; cells; tissue; and a whole organ.

1 1. The method according to claim 10, wherein the cells are genetically modified to produce a therapeutic polypeptide.

12. The method according to claim 1 , wherein the immunosuppressive agent is selected from the group consisting of FK506 and cyclosporin A.

3. A method for tolerizing a recipient host for long-term survival of a transplant comprising:

(a) implanting into the recipient host one or more encapsulated cell devices containing tissue or cells that are allogeneic or xenogeneic to the recipient host;

(b) administering an immunosuppressive agent to the recipient host in an amount effective to permit survival of the encapsulated tissue or cell transplant for a period of at least one week;

(c) removing the encapsulated cell devices; and (d) subsequently implanting into the recipient host a transplant from the same tissue source as the tissue or cells in the encapsulated cell devices; wherein the recipient host is tolerized for long-term survival of a transplant.