WO1999060105A1

WO1999060105A1 - Flat sheet cell encapsulation device

Info

Publication number: WO1999060105A1
Application number: PCT/IB1999/000889
Authority: WO
Inventors: Markus Borkenhagen
Original assignee: Modex Therapeutiques, S.A.
Priority date: 1998-05-15
Filing date: 1999-05-14
Publication date: 1999-11-25
Also published as: AU3623199A

Abstract

The present invention provides a cell implantation device comprised of a cell matrix system which is fixedly coupled to a flat sheet supporting structure, preferably ring-shaped. Cells are inoculated onto the matrix, which may be a cell supporting fiber or a hydrogel. The proliferation, viability, and functionality of the cells are characterized prior to encapsulation in an immunoisolatory semi-permeable membrane. Cell growth under controlled conditions allows the encapsulated device to provide a relatively constant dose of secreted protein.

Description

FLAT SHEET CELL ENCAPSULATIONDEVICE

RELATED APPLICATIONS

This application claims priority under 35 U.S.C. § 119(e) to U.S. Provisional Patent Application 60/085,648, filed May 15, 1998.

FIELD OF THE INVENTION

The present invention relates generally to cell encapsulation devices, and more particularly to devices for cell screening and implantation into a mammalian host.

BACKGROUND OF THE INVENTION

In encapsulated cell therapy, xenogenic or allogenic cells are isolated from the host's immune system by surrounding the cells in a semi-permeable membrane prior to implantation within the host. This semi -permeable membrane jacket is relatively impermeable to large molecules, such as components of the host's immune system, but is permeable to small molecules. In one embodiment, the semi-permeable membrane jacket has a molecular weight cutoff of between 50 and 2000 kilodaltons (kDa). Thus, the semi-permeable membrane allows the implanted cells to receive nutrients necessary for viability and allows metabolic waste to be removed. The membrane also allows therapeutic molecules produced by the implanted cells to diffuse to host cells. For example, endogenous proteins or those cloned into the cell are delivered to the host. The use of an immuno-protective, semi-permeable membrane now allows transplantation of encapsulated cells from one species into a host from a different species without the risk of immune rejection or use of immunosuppressive drugs. Applications of encapsulated cell therapy include, for example, treatments for diabetes, hemophilia, anemia, β-thalassemia, Parkinson's disease, and amyotropic lateral sclerosis.

The use of biologically compatible polymeric materials in construction of an encapsulation device is critical to a successful cell encapsulation therapy. Important components of the encapsulation device include the surrounding semi-permeable membrane and the internal cell-supporting matrix, or scaffold. The scaffold defines the microenvironment for the encapsulated cells and keeps the cells well distributed within the intracapsular compartment. The optimal internal scaffold for a particular cell encapsulation device is highly dependent on the cell type. For example, while adherent cells often prefer a solid surface on which to lie, suspension cells may prefer a hydrophilic lightly cross-linked hydrogel as a matrix material.

In the absence of a scaffold, adherent cells aggregate to form clusters. When the clusters grow too large, they typically develop a central necrotic core. Dying cells accumulate around the core and, upon lysing, release factors detrimental to the health of neighboring cells. The lysed cell fragments are also transported to the host environment, there eliciting an antigenic response.

The prior art discloses that the position of the cells may be stabilized by an internal scaffold composed of an immobilizing matrix formed from a hydrogel. A "hydrogel" is a 3 -dimensional network of cross-linked hydrophilic polymers, in the form of a gel and substantially (e.g., greater than 90%) composed of water. In typical prior art systems, however, cells are encapsulated within an immunoisolatory membrane and then allowed to continue to proliferate. Once the cells are placed inside the encapsulating membrane, it is extremely difficult to characterize their morphology. Moreover, it is nearly impossible to control cell proliferation and functionality. As a result, devices made using methods of the prior art exhibit wide ranges of total expression of the desirable secreted protein. Consequently, it is difficult to control the therapeutic dosage received by the host.

Thus a need remains to develop an encapsulation device in which the cells to be encapsulated are selected for a desired morphology prior to encapsulation, resulting in better control of expression of the secreted protein after encapsulation.

SUMMARY OF THE INVENTION

The invention provides a cell implantation device. The device has a cell matrix, or scaffold, affixed to a supporting frame. The supporting frame is preferably a flat, ring-shaped element. Cells are inoculated onto the matrix. Once the cells exhibit the desired morphology, the cell-containing matrix is encapsulated in a semi-permeable immunoisolatory membrane. The cell matrix, or scaffold, in the device of the invention advantageously provides cells with a template for cellular organization in a three-dimensional orientation resembling their typical physiological shape. The cell matrix is particularly useful for the use of suspension cells for encapsulated cell therapy. In one embodiment of the present invention, the matrix is composed of a hydrogel material selected from agarose, alginate, chitosan, collagen, Matrigel®, Vitrogen®, or PVA foam. In a preferred embodiment, the hydrogel is agarose. In this embodiment, the selected cells are mixed with a hydrogel solution at a desired cell density and a desired concentration of gel polymer. After mixing, the cell/hydrogel suspension is cast into the structural support which serves as a mold while the suspension solidifies into the cell-supporting matrix. Once the matrix solidifies, mechanical forces, surface tension, and capillary forces hold the matrix in the structural support. Upon curing, the hydrogel solution becomes gelled and acts as a supporting matrix for the cells. The matrix/support can then be encapsulated in the semi-permeable membrane.

In another embodiment, the cell matrix is composed of fibers selected from the group consisting of silk, cotton, chitin, for example, acrylic, polyester, polyethylene, polypropylene, polyacrylonitrile, polyethylene terephthalate, nylon, polyamides, polyurethanes, or polybutester. In a preferred embodiment, the matrix is polyethyleneterephthalate (PET) fiber. The fibrous material can be coated with a protective coating of hydrogel prior to encapsulation in the membrane. In this embodiment, the matrix is a cell-supporting fiber wound or woven around the structural support. PET fibers can be wound around a ring shaped structural support. Alternatively, the structural support is a polyvinyl alcohol sponge incorporated into the structural support. A desired cell type is then inoculated onto the selected solid matrix and cultured in the appropriate medium.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows three top views of components of the device. FIG. 1 A is a representation of two structural support components. FIG. IB is a representation of two structural support components, each fabricated with a cell/agarose component. FIG. 1C is a representation of two structural support components fabricated with PET fibers.

FIG. 2 is a schematic representation of two preferred device fabrication processes.

FIG. 3 is a graph showing mEPO output over time for devices with cell captured in an agarose matrix and cultured in various media. FIG. 4 is a graph showing mEPO output over time for devices with cell captured in a PET fiber matrix and cultured in various media.

FIG. 5 is a graph showing mEPO output over time for pre- and post-transplant devices cultured in various media. FIG. 6 is a graph showing the diffusive properties of a dip-coated membrane fabricated with 8 % PAN/PVC + 20 % PVP (10⁶ daltons [Da]).

FIG. 7 is a graph showing hematocrit levels over time in rats which were implanted with the device subcutaneously.

FIG. 8 is a graph showing hematocrit levels over time in rats which were implanted with the device interperitoneally.

FIG. 9 shows the viability scores of devices which were implanted syngenically and allogenically.

DETAILED DESCRIPTION OF THE INVENTION

The invention provides a cell encapsulation device configured as a flat sheet. The device has (1) a structural support or supporting frame element (e.g., in the shape of a ring); (2) matrices to support the cells to be transplanted; and (3) a semi-permeable membrane that permits diffusion of nutrients, waste, and other small molecules to and from the transplanted cells, while isolating those cells from the host.

The details of one or more embodiments of the invention are set forth in the accompanying description below. 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. Matrices

The invention contemplates the use of various matrices, or matrix systems, to support the cells to be encapsulated. The selection of a matrix may depend on the cell type used, with a particular cell type preferentially adhering and growing onto one matrix over another. In one embodiment, the matrix is a hydrogel. A "hydrogel" is a substance formed when a polymer is cross linked to create a 3-dimensional structure, entrapping water molecules to form a gel.

In another embodiment solid matrices, which may be more appropriate for certain types of cells, are used. This embodiment is shown in FIG 1. FIG. 1A shows two examples of a structural support, each having a tab for handling the device. FIG. IB and FIG. 1C show the structural support fixedly coupled to the cell matrix system. In FIG. 1C, the cell matrix system is a cell/agarose mixture. In FIG. 1C, the cell matrix system is the fibrous material, preferably PET fibers, which is wound or otherwise fixedly coupled to the structural support.

To inoculate and culture the cells, the support can be dipped into a cell/agarose suspension and allowed to attain the desired conformation (see, FIG. 2, left side of the FIG. (A)). Alternatively, a support containing the cell supporting fiber can be placed in a Petri dish on a non-adherent surface suitable to culture cells, such as, for example, non-adherent agarose The cells to be implanted are then inoculated onto the fiber (see, FIG. 2, right side of the FIG. (B)). The cells are then cultured until they attach to the fiber. An extracellular matrix may be used to coat the fiber to aid cell attachment. Examples of fibers suitable for use in the present invention include natural fibers such cotton, silk, chitin or carbon. Non-naturally occurring fibers, such as, for example, acrylic, polyester, polyethylene, polypropylene, polyacrylonitrile, polyethylene terephthalate, nylon, polyamides, polyurethanes, or polybutester are also suitable for use in the present invention. Preferably, the fiber is PET, which is typically comprised of bundles of approximately 80 PET monofilaments. The diameter of the monofilaments selected can range between 1-30 μm. Monofilaments having a range of diameters between approximately 1-10 μm are preferred, with a diameter of 5 μm being most preferred. The type of fiber selected and the dimensions of the fiber may depend on cell type as one cell type may prefer one type of fiber over another. Additionally, the fibrous matrix selected may be coated with a protective layer of hydrogel prior to encapsulation in the membrane. Structural Support

The structural component to support the cell matrix system can be formed in many different dimensions and geometric shapes. For example, the supports may be circular, elliptical, square, rectangular, triangular, or star shaped. Since devices are fabricated as flat sheets, the thickness of the ring is much less than the diameter of the ring. TABLE 1 lists the dimensions of devices that have been fabricated.

TABLE 1

Ring Inner Outer Volume Cell Number Corresponding

Thickness Diameter Diameter Hollow Fiber

[μl] (in millions)

[mm] [mm] [mm] Length [cm]

0.4 12 14 45 6 12

0.7 12 14 80 10 20

0.4 20 22 125 15 30

0.7 20 22 219 27 54

(PMMA PET) 10 13 31 8 16 0.4

(PMMA/PET) 20 23 125 32 64 0.4

The structural support can be made by methods well known to those skilled in the art. The structural support can be made, for example, from a polymeric solution cast into a mold or by stamping, cutting, and extruding. Typically, molds for casting polymeric solutions are fabricated from polytetrafluoroetheylene (PTFE), which is resistant to organic solvents.

Alternatively, other organic solvent-resistant materials can be used. After the polymeric solution is cast into the mold, the organic solvent is phase-separated from the polymer using an aqueous solution. The completed structural support is removed from the mold after the polymeric solution solidifies. The structural support may be fabricated from any biocompatible, substantially non-degradable polymer. A "biocompatible" device causes substantially no adverse consequences to the host upon transplantation and thereafter. Examples of biocompatible polymers suitable for use in the invention include acrylic, polyester, polyethylene, polypropylene, polyacetonitrile, polyethylene terephthalate, nylon, polyamides, polyurethanes, polybutester, silk, cotton, chitin, carbon, as well as other biocompatible metals. Preferably, the polymer support is cast from a solution of 20 % weight to volume (w:v) PA /PVC in dimethyl sulfoxide (DMSO).

Cells

The present invention contemplates the use of several types cells to be inoculated on the matrices. Examples of cells suitable for use in the present invention include C₂C_I2 mouse myoblast cells , LNS-1 cells, and β tet cells. Also useful are human Hs683 glial-derived cells, human A172 glial-derived cells, porcine glioblasts, chondroblasts isolated from human long bone, mice 3T3 fibroblast cells, rabbit SIRC corneal-derived cells, Chinese hamster ovary (CHO) cells, baby hamster kidney (BHK) cells, and CAC cells. Cells are also useful when endogenously producing a protein of desired interest, such as when transfected to express the protein of interest. Devices have been fabricated with C₂C₁₂ myoblasts transfected with the murine EPO (mEPO) gene. Additionally, devices have been fabricated with C₂C₁₂ cells transfected to express GLP-1.

In culture, cellular behavior is often related to cell morphology (i.e. differentiation, growth arrested). Accordingly, cells can be cultured under selected conditions until the cells exhibit a particular morphology corresponding with the desired phenotype. For example, myoblasts in culture grow in the form of clusters. These clusters have the capacity to fuse into larger clusters, referred to in the literature as "myoballs". Myoblasts cultured in the present invention may be monitored, characterized, and controlled for the desired morphology prior to encapsulation. For instance, the myoblasts may be cultured in the present invention until all, or substantially all, of the myoblasts exist in the form of myoballs. Once the myoblasts have become myoballs, they no longer proliferate. Once the cells stop proliferating, the number of cells, and therefore the amount of protein they produce, becomes relatively constant. In this manner the dose of protein delivered to the patient can be regulated. The number of cells that each device is able to contain is dictated by many factors, including the structural parameters of the device type (i.e. geometry and size). For devices in which a hydrogel is used, the maximum number of cells per device is also a function of both the volume of the hydrogel and the cell density per unit volume of gel. For example, a ring-shaped device having a 9.7 mm thickness and an inner diameter of 12 mm has a volume of 80 μl. Assuming that in order to maintain sufficient structural integrity, 50% of the volume is occupied by the hydrogel, the remaining volume available to be occupied by the cells is 40 μl. Given that 4 μl holds approximately 1 ,000,000 cells, the device therefore holds a maximum of approximately 10,000,000 cells.

In addition to device parameters, the total length of PET fibers determines the maximum cell number for the PET fiber devices. For example, in a ring-shaped PET fiber device having a ring thickness of 0.4 mm, an inner diameter of 12 mm and an outer diameter of 13 mm, the corresponding length of PET fibers, with a diameter of approximately 5 μm, which can be wound around the device is approximately 16 cm. Given that each 1 cm of fibers supports approximately 500,000 cells, such a device holds approximately 8,000,000 cells.

Encapsulation

Once the cultured cells have reached their desired morphology on the cell matrix system/structural support, the cells on the matrix/structural support can then be encapsulated. Encapsulation involves a 2-step process. FIG. 2 shows two preferred encapsulation processes of the present invention. In FIG. 2, left side of the FIG. (A), a suspension of cells in agarose is polymerized in a PAN/PVC structural support. The cells are cultured until they exhibit the desired morphology and are then dipped in a polymer solution and subsequently dipped in an aqueous solution, preferably a physiological buffer, which phase inverts the polymer solution and forms a semi-permeable membrane around the cells. The finished devices are then placed in Petri dishes for in vitro storage. In FIG. 2, right side of the FIG. (B), cell-supporting fibers, preferably PET fibers, are fixedly coupled to a PAN/PVC structural support. A cell solution is inoculated on the fibers and the cells are cultured until the exhibit the selected morphology. The support containing the cells is then dipped in a polymer solution and then dipped in an aqueous solution, preferably a physiological buffer , which phase inverts the polymer solution and forms a semi-permeable membrane around the cells. The finished devices are then placed in Petri dishes for in vitro storage.

In both encapsulation procedures, percentages of the polymer solutions may be varied to produce membranes with different thickness. Membranes can be characterized by diffusion properties and by scanning electron microscopy (SEM). To measure diffusion properties, various molecular weight polydextrans are fluorescence labeled, incorporated into agarose, and encapsulated. The membrane capsules are then placed into fluorophore-free water. The -amount of fluorophore that has diffused into the water is measured periodically, over a 6-hr period. Fluorescence in aliquots of the water is detected with a fluorospectrometer. To study the morphological aspects of membranes made from the various polymeric solutions, the membranes are viewed using SEM techniques common in the art. Individual membranes are glycerinated, dried, and gold-plated. The membranes are then introduced into the vacuum chamber of an electron microscope, so that their surface morphology can be viewed.

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 Encapsulated mEPO Transfected Mouse C₂C_U Cells in an Agarose Matrix

Approximately 1,000,000, murine C₂C₁₂ cells transfected with the murine EPO gene, either as single cells or cell aggregates, were trapped within a 1% low-melting agarose matrix. The matrix formed as a semi-solid gel in a ring-shaped support. The volume of the ring support totaled 80 μl. The cell aggregates were formed under rotation at 60 rpm for three days in surface treated non-adherent Petri dishes in regular growth medium (DMEM, 10 % fetal calf serum [FCS], penicillin streptomycin, and MTX). The cell containing matrix and the support were encapsulated in a 100 μm thick membrane by dip-coating in an 8 % (w:v) PAN/PVC solution in DMSO.

To evaluate the possibility of using low serum media to slow or halt proliferation, three groups of devices containing either single cells or cell clusters were grown in different medium conditions. Group 1 was cultured in regular growth medium; Group 2 was cultured in Prolifix, a culture medium containing vegetable proteins instead of animal serum proteins, and Group 3 was cultured in medium containing 2 % FCS instead of 10% FCS. Cell viability was assayed by histology. Cell functionality was assayed by measuring mEPO output in the media over time. The release of mEPO was determined by techniques common in the art, such as ELISA. EPO release was compared to EPO production under standard culture conditions. In culture, 1,000,000 mEPO transfected C₂C₁₂ cells release approximately 50 IU mEPO in 24 hr.

FIG. 3 demonstrates that only single cell suspensions cultured in regular growth medium showed a significant increase in output of mEPO over time, evidence of continuous proliferation of cells within the device. The remaining devices produced a relatively constant output of mEPO over time. However, the overall output in these devices was relatively low. This finding was explained by poor viability of the cells in the form of cell clusters and the even worse viability of the single cells. Histological analysis revealed evenly distributed dead cells and dead cell clusters within each of the respective devices, suggesting that the dimensions of the device have no effect on viability. Finally, devices cultured in Prolifix displayed the worst survival compared to devices cultured in 2 % FCS.

Example 2 β TC tet Cells and INS-1 Cells in Agarose

Two groups of devices were separately fabricated in agarose and encapsulated as described in the above examples. One group contained β TC tet cells. The second group contained INS-1 cell clusters. After encapsulation, the devices were maintained in culture. At various time points, the devices were harvested and examined by histology and by light microscope. The histological and light microscopy analysis revealed that the INS-1 cell clusters showed good viability and proliferation. The β TC tet cells, however, did not survive the cell culture period post encapsulation.

Example 3 Encapsulated mEPO Transfected C₂C₁₂ Cells on PET Matrix

An analysis was performed to determine whether culturing devices in low serum containing medium slowed mitosis or altogether arrested proliferation. PET devices were inoculated with approximately 300,000 C₂C₁₂ cells transfected with the mEPO gene and were cultured for four days in regular growth medium. After day four, the devices were split into three groups. The first group was encapsulated and put directly into medium containing 2% FCS. The second group was encapsulated and put back into regular growth medium for three days. After three days, the medium was changed to low serum growth medium to compensate for cell loss during the stress of encapsulation. The third group was not encapsulated and kept in 10% FCS growth medium. Samples were taken regularly from the culture medium and the amount of mEPO in the medium was measured by ELISA. The quantity of DNA present was measured to determine the number of cells. The devices were also evaluated by histology. FIG. 4 shows the amount of mEPO released over time for the above-described three groups. These data establish that devices cultured in low serum containing media produced much less protein than devices cultured in regular growth medium. Moreover, as demonstrated by comparing TABLE 2 with FIG. 4, the amount of mEPO produced corresponds with the determined cell number at the end of the experiment. These data lead to the conclusion that culturing devices in low serum containing media significantly slowed the rate of proliferation in the cells. Additionally, the amount of mEPO in the media of non-encapsulated cells cultured in growth medium showed a plateau between days 8 and 10. This suggests that the maximum capacity of cells was reached for the given device size, geometry and type.

TABLE 2

Number of cells cultured in: Days Growth medium 2% FCS post-enc. 2% FCS directly

0 300,000 300,000 300,000

13 6.51 x lO⁶ 3.2 x lO⁶ 3.0 x lO⁶

In a second study, 18 devices were inoculated with 300,000 C₂C₁₂ cells transfected with the mEPO gene. The cell-seeded devices were kept in regular growth medium for seven days before encapsulation. During the course of the experiment, the number of cells present in each device was monitored. On the day of encapsulation and on day four after encapsulation, both the amount of mEPO and the number of cells were quantified. At the day of encapsulation, the devices were divided into three groups. The first group was left unencapsulated and continued to be cultured in regular growth media. The second group of devices was encapsulated and immediately put into culture medium containing 2% FCS. The third group was left unencapsulated and was also transferred into medium containing 2% FCS.

FIG. 5 compares cell number and amount of mEPO in the cell media for both pre- and post-encapsulation devices. Additionally, all unencapsulated devices were monitored by microscopy. At day -three post-encapsulation (day 10 overall) none of the devices showed any proliferation. This result is consistent with data shown in FIG. 5. The graph of proliferation as measured by mEPO release and cell number plateaus for devices of this size around day eight. This result was further confirmed by histological analysis. Prior to sacrificing devices for DNA analysis, unencapsulated devices were analyzed by light microscope at a 32x magnification. At this magnification, it was observed that most visible cells displayed myoball morphology. Because myoblasts stop proliferating when they are formed in the myoball morphology, this result is consistent the plateau of cell growth in FIG. 6. It should be noted, however, that some cells in the device were not visible because they had colonized the interior of the PET bundles.

Example 4 Rate of Colonization of Device with C₂C₁₂ Cells Transfected with GLP-1

To determine whether the rate of colonization of C₂C₁₂ cells vary with different clones, murine C₂C₁₂ cells transfected with the GP-1 gene were inoculated onto a PET fiber device at an initial loading density of 300,000 cells per device. After 3 days, the number of cells increased to 4,700,000 cells per device +/- 500,000 cells. After 7 days, the number of cells increased to 5,800,000 cells per device +/- 400,000 cells. Examination of the cells by light microscope revealed a very homogeneous, rapid colonization of fibers.

The apparent maximum number of cells that can be maintained within this given device configuration is between 5-6,000,000 cells, somewhat lower than the theoretical value of 8,000,000 cells. This finding emphasizes that in order to achieve the high cell densities described in TABLE 1, the cells must be loaded under compressive forces. One may obtain better seeding and survival of the cells within the devices by allowing the cells to attain a natural density upon and within the matrix by cell-matrix interaction, rather than by profusion,

Example 5 Membrane Characterization

Membranes were fabricated as described above and membrane mass transport properties were determined by measuring the diffusive flux of fluorescenfly labeled dextran molecules of different molecular weights. Mass transfer of the membrane was studied by immobilizing a membrane between two chambers of equal volume. One chamber was filled with the fluorescently labeled dextran solution and the other chamber was filled with the same volume of fluorophore-free solution. Over time, fluorescently labeled dextran molecules that are small enough to pass through the membrane will diffuse into the fluorophore-free chamber, changing the fluorescence of the recipient chamber. This change in fluorescence was measured by analyzing samples of the solution in a fluorimeter (Perkin Elmer). In addition to studying the diffusion properties, membranes were also characterized by scanning electron microscopy (SEM).

By SEM analysis, membrane thickness varied between about 100-300 μm. This variation was caused by the fabrication process in which the device is dip-coated to form the membrane. As the device is removed from the polymer solution, gravity causes the polymer solution to collect disproportionately on the bottom of the device. When the device is then phase inverted in the precipitation bath, the membrane forms with the thicker portion on the bottom and the thinner portion on the top. TABLE 3 summarizes the diffusive properties and the physical properties of the membrane as determined by SEM. The diffusive properties of the dip-coated membrane fabricated with 8% PAN/PVC + 20% PVP (10⁶ Da) are also illustrated in FIG. 6.

TABLE 3

Composition of Polymer MW Cut-off Diffusion Coefficient Surface Membrane

Solution [w/v] [daltons (Da)] [cmVsec] Smoothness Thickness [μm]

8% PAN/PVC 12.00 1 0 + 80-10

6% PAN/PVC +30% w/w ++ 100-200 PVP 10⁶Da

6% PAN/PVC +50% w/w +++ 200-300 PVP 10⁶ Da

Physical properties of the membrane and a control PES-5 membrane are summarized in TABLE 4. The molecular weight cut-off for the 8% PAN/PVC + 20% PVP flat sheet membrane was determined to be approximately 50,000 Da. This value is significantly lower than the molecular weight cut-off determined for the PES-5 membrane, which was approximately 240,000 Da. In terms of overall transport properties, however, the absolute values for the diffusion coefficients of the individual dextran molecules (TABLE 4, Column 2) do not differ significantly. In order to provide sufficient immunoisolation from the host, the diffusion coefficients may need to be at least two orders of magnitude lower.

TABLE 4

Membrane Type Diff. Coeff. Cut-off [kDa] Wall thickness [μm] ID [μm]

[cm²/s] T70

PES 5 0.5 - l^e - 7 240 100 500

PAN/PVC 9° - 8 50 100-300 400 thick

Example 6 In Vivo Data in Mice and Rats

In vivo functionality and biocompatibility of the devices were studied in mice and rats. Eighteen devices were prepared and divided into three groups of six as summarized in TABLE 5. For each device, 40 cm of PET yarn was wound onto PMMA rings. The devices were inoculated with approximately 300,000 cells (day zero) and incubated for seven days in culture. After seven days, devices were first coated in a layer of agarose and then encapsulated by dip-coating into a polymer solution of 8% weight to volume of PAN/PVC + 20% weight to weight PVP (10⁶ Da) followed by phase inversion in a physiological solution. Immediately after encapsulation, all implants were directly transferred into cell culture medium containing 2% serum.

Group 1 devices containing mouse C₂C₁₂ cells transfected with GLP-1 were implanted syngenically into mice. Group 2 devices containing mouse C₂C,₂ cells transfected with GLP-1 were implanted allogenically into mice. Lastly, Group 3 devices containing mouse C₂C₁₂ transfected with rat EPO were xenogenically transplanted into rats. In all three groups, three devices were implanted subcutuaneously (SC) and three devices were transplanted intraperitoneally (IP). TABLE 5 summarizes the experimental methodology.

TABLE 5

Group 1 Group 2 Group 3 mouse C₂C₁₂ #149 GLP1 mouse C₂C₁₂ #149 GLP1 mouse C₂C₁₂rat EPO (pool) n=6 n=6 n=6 syngenic allogenic xenogenic (rat)

SC: n=3 SC: n=3 SC: n=3 IP: n=3 IP: n=3 IP: n=3

Before and after explant, the release of rat EPO was determined for xenogenically implanted devices. Devices which were syngenically and allogenically implanted into mice were evaluated for GLP-1 expression, both prior to implant and post explant. All devices were examined histologically for cell survival and biocompatibility. Additionally, the hematocrit of all EPO-device-implanted rats was monitored by a standard hematocrit method (Koepke, 1991).

The EPO output of devices pre-implant was quantified and is summarized in TABLE 6. As shown in the TABLE, the amount of EPO released from the capsules prior to implantation is relatively uniform. Additionally, the EPO of one capsule post explanation was determined to be undetectable.

TABLE 6

Capsule number pre-implant post-explant UI/24 hrs/caps UI/24 hrs/caps

1 SC 223.5 -

2 SC 230 0

3 SC 265 -

4 IP 310 -

5 IP 237.5 -

6 IP 260 -

Rat Data.

FIG. 7 and FIG. 8 illustrate the hematocrit of individual animals over time for the subcutaneous and intraperitoneal implants, respectively. As shown in FIG. 7 and FIG. 8, the hematocrit rises dramatically between day 0 and day 10. Thereafter, the hematocrit level plateaus until approximately day fifteen. After approximately day fifteen, the hematocrit level for four of the six animals begins to decline. Two of 6 animals continued to show increased hematocrit for significantly longer as shown in the figures. Mouse Data.

In order to obtain a two week time point of cell viability within the capsule, two devices were explanted from mice after two weeks in vivo. The remaining capsules were explanted after three weeks. All capsules were analyzed for GLP-1 output and by histology for cell survival. TABLE 7 illustrates the output GLP-1 from the capsules post-explantation.

TABLE 7

Capsule # syngenic allogenic GLP-1 [ng] gLP-1 [ng]

1 0 2 weeks: 229

2 7070 7650

3 0 7830

4 2 weeks: 217 8045

5 7485 7750

6 7885 7770

As TABLE 7 illustrates, there was no significant difference in GLP-1 output between the devices explanted at two weeks. With the exception of two capsules in which GLP-1 production was undetectable, the remaining capsules all produced the same relative amount of GLP-1 output. Note, however, that the values of GLP-1 on the three-week time point were very high compared to the two week time point. This difference was a result of an error in the cAMP assay for the three-week time point. Thus, only relative comparisons may be made.

Histology of the individual devices revealed excellent viability of cells inside capsules that had been implanted into an allogenic environment, with approximately 67% viability. Capsules that had been syngenically implanted fared less well, but still had a viability of approximately 41%. FIG. 9 shows the viability scores of the syngenic and allogenic capsules.

EQUIVALENTS

From the foregoing detailed description of the specific embodiments of the invention, it should be apparent that a novel flat sheet cell implantation device and method has been described. Although particular embodiments have been disclosed herein in detail, this has been done by way of example for purposes of illustration only, and is not intended to be limiting with respect to the scope of the appended claims which follow. In particular, it is contemplated by the inventor that various substitutions, alterations, and modifications may be made to the invention without departing from the spirit and scope of the invention as defined by the claims.

Claims

What is claimed is:

1. A method of making a cell implantation device comprising the steps of:

(a) affixing a supporting frame element to a cell matrix system;

(b) inoculating cells onto the cell matrix system;

(c) culturing the cells until the cells exhibit a selected morphology having a corresponding selected phenotype; and

(d) encapsulating at least the cell matrix and cells within a semi-permeable immunoisolatory membrane.

2. The method of claim 1 , wherein the semi-permeable immunoisolatory membrane is formed from a polymer solution capable of forming a semi-permeable immunoisolatory membrane upon phase inversion in an aqueous medium; and

3. The method of claim 2, wherein said polymer solution is selected from the group consisting essentially of PAN, PVC, PVP, and combinations thereof.

4. The method of claim 1 wherein the cells are myoblasts, the selected morphology is cells in the form of myoballs, and the selected phenotype is the substantial halting of proliferation of the cells upon forming into myoballs.

5. The method of claim 1 wherein the cell matrix system is a network of fibers selected from the group consisting of silk, cotton, chitin, silk, cotton, chitin, carbon, acrylic, polyester, polyethylene, polypropylene, polyacrylonitrile, polyethylene terephthalate, nylon, polyamides, polyurethanes, polybutester, and polyethyleneterephthalate (PET).

7. A cell implantation device, comprising: a cell matrix system affixed to a supporting frame element, wherein the cell matrix system can be inoculated with cells having a selected moφhology in response to the cell matrix system, and wherein the frame element and cell matrix are encapsulated in a semi-permeable membrane jacket.

8. The device of claim 7, wherein the cells are myoblasts.

9. The device of claim 7, wherein semi-permeable membrane jacket is formed from a polymer solution selected from the group consisting essentially of PAN, PVC, PVP, and combinations thereof.

10. The device of claim 7, wherein the cell matrix system is a network of fibers selected from the group consisting of silk, cotton, chitin, silk, cotton, chitin, carbon, acrylic, polyester, polyethylene, polypropylene, polyacrylonitrile, polyethylene terephthalate, nylon, polyamides, polyurethanes, polybutester, and PET.

12. The device of claim 7, wherein the cell matrix and cells are coated with a hydrogel prior to encapsulation.

13. The device of claim 7, wherein the device is configured as a flat sheet.

14. The device of claim 13, wherein the device contains between 5 million and 500 million cells.

15. The device of claim 7, wherein the semi-permeable membrane jacket has a molecular weight cutoff of between 50 and 2000 kilodaltons (kDa).

16. A method for treating a subject, comprising: implanting into the subject a cell matrix system affixed to a supporting frame element, wherein the cell matrix system can be inoculated with cells having a selected moφhology in response to the cell matrix system, and wherein the frame element and cell matrix are encapsulated in a semi-permeable immunoisolatory membrane.